Intrusion detection System.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "f82b38b2",
   "metadata": {},
   "source": [
    "<h2 align=\"center\">  Intrusion detection System </h2>  \n",
    "<h6 align=\"right\" > Developed by Malèk YAICH </h6>\n",
    "</br>"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3c87df7f",
   "metadata": {},
   "source": [
    "# I. Importation des bibliothèques "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "f7aef4ef",
   "metadata": {},
   "outputs": [],
   "source": [
    "import warnings\n",
    "warnings.filterwarnings(\"ignore\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "ec52edae",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.metrics import classification_report, confusion_matrix,accuracy_score,precision_score,recall_score,f1_score\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "987e19bb",
   "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>Flow Duration</th>\n",
       "      <th>Total Fwd Packets</th>\n",
       "      <th>Total Backward Packets</th>\n",
       "      <th>Total Length of Fwd Packets</th>\n",
       "      <th>Total Length of Bwd Packets</th>\n",
       "      <th>Fwd Packet Length Max</th>\n",
       "      <th>Fwd Packet Length Min</th>\n",
       "      <th>Fwd Packet Length Mean</th>\n",
       "      <th>Fwd Packet Length Std</th>\n",
       "      <th>Bwd Packet Length Max</th>\n",
       "      <th>...</th>\n",
       "      <th>min_seg_size_forward</th>\n",
       "      <th>Active Mean</th>\n",
       "      <th>Active Std</th>\n",
       "      <th>Active Max</th>\n",
       "      <th>Active Min</th>\n",
       "      <th>Idle Mean</th>\n",
       "      <th>Idle Std</th>\n",
       "      <th>Idle Max</th>\n",
       "      <th>Idle Min</th>\n",
       "      <th>Label</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>5.416666e-07</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000003</td>\n",
       "      <td>4.651163e-07</td>\n",
       "      <td>9.153974e-09</td>\n",
       "      <td>0.000242</td>\n",
       "      <td>0.002581</td>\n",
       "      <td>0.00101</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000307</td>\n",
       "      <td>...</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>5.416666e-07</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000003</td>\n",
       "      <td>4.651163e-07</td>\n",
       "      <td>9.153974e-09</td>\n",
       "      <td>0.000242</td>\n",
       "      <td>0.002581</td>\n",
       "      <td>0.00101</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000307</td>\n",
       "      <td>...</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>4.416666e-07</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000003</td>\n",
       "      <td>4.651163e-07</td>\n",
       "      <td>9.153974e-09</td>\n",
       "      <td>0.000242</td>\n",
       "      <td>0.002581</td>\n",
       "      <td>0.00101</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000307</td>\n",
       "      <td>...</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>7.499999e-07</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000003</td>\n",
       "      <td>4.651163e-07</td>\n",
       "      <td>9.153974e-09</td>\n",
       "      <td>0.000242</td>\n",
       "      <td>0.002581</td>\n",
       "      <td>0.00101</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000307</td>\n",
       "      <td>...</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>7.249999e-07</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000003</td>\n",
       "      <td>4.651163e-07</td>\n",
       "      <td>9.153974e-09</td>\n",
       "      <td>0.000242</td>\n",
       "      <td>0.002581</td>\n",
       "      <td>0.00101</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000307</td>\n",
       "      <td>...</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 78 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   Flow Duration  Total Fwd Packets  Total Backward Packets  \\\n",
       "0   5.416666e-07                0.0                0.000003   \n",
       "1   5.416666e-07                0.0                0.000003   \n",
       "2   4.416666e-07                0.0                0.000003   \n",
       "3   7.499999e-07                0.0                0.000003   \n",
       "4   7.249999e-07                0.0                0.000003   \n",
       "\n",
       "   Total Length of Fwd Packets  Total Length of Bwd Packets  \\\n",
       "0                 4.651163e-07                 9.153974e-09   \n",
       "1                 4.651163e-07                 9.153974e-09   \n",
       "2                 4.651163e-07                 9.153974e-09   \n",
       "3                 4.651163e-07                 9.153974e-09   \n",
       "4                 4.651163e-07                 9.153974e-09   \n",
       "\n",
       "   Fwd Packet Length Max  Fwd Packet Length Min  Fwd Packet Length Mean  \\\n",
       "0               0.000242               0.002581                 0.00101   \n",
       "1               0.000242               0.002581                 0.00101   \n",
       "2               0.000242               0.002581                 0.00101   \n",
       "3               0.000242               0.002581                 0.00101   \n",
       "4               0.000242               0.002581                 0.00101   \n",
       "\n",
       "   Fwd Packet Length Std  Bwd Packet Length Max  ...  min_seg_size_forward  \\\n",
       "0                    0.0               0.000307  ...                   1.0   \n",
       "1                    0.0               0.000307  ...                   1.0   \n",
       "2                    0.0               0.000307  ...                   1.0   \n",
       "3                    0.0               0.000307  ...                   1.0   \n",
       "4                    0.0               0.000307  ...                   1.0   \n",
       "\n",
       "   Active Mean  Active Std  Active Max  Active Min  Idle Mean  Idle Std  \\\n",
       "0          0.0         0.0         0.0         0.0        0.0       0.0   \n",
       "1          0.0         0.0         0.0         0.0        0.0       0.0   \n",
       "2          0.0         0.0         0.0         0.0        0.0       0.0   \n",
       "3          0.0         0.0         0.0         0.0        0.0       0.0   \n",
       "4          0.0         0.0         0.0         0.0        0.0       0.0   \n",
       "\n",
       "   Idle Max  Idle Min  Label  \n",
       "0       0.0       0.0      0  \n",
       "1       0.0       0.0      0  \n",
       "2       0.0       0.0      0  \n",
       "3       0.0       0.0      0  \n",
       "4       0.0       0.0      0  \n",
       "\n",
       "[5 rows x 78 columns]"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data=pd.read_csv(\"CICIDS2017_sample_km.csv\")\n",
    "data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "7cccb404",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    18225\n",
       "3     3042\n",
       "6     2180\n",
       "1     1966\n",
       "5     1255\n",
       "2       96\n",
       "4       36\n",
       "Name: Label, dtype: int64"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.Label.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "3d061eb7",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Flow Duration                  0\n",
       "Total Fwd Packets              0\n",
       "Total Backward Packets         0\n",
       "Total Length of Fwd Packets    0\n",
       "Total Length of Bwd Packets    0\n",
       "Fwd Packet Length Max          0\n",
       "Fwd Packet Length Min          0\n",
       "Fwd Packet Length Mean         0\n",
       "Fwd Packet Length Std          0\n",
       "Bwd Packet Length Max          0\n",
       "Bwd Packet Length Min          0\n",
       "Bwd Packet Length Mean         0\n",
       "Bwd Packet Length Std          0\n",
       "Flow Bytes/s                   0\n",
       "Flow Packets/s                 0\n",
       "Flow IAT Mean                  0\n",
       "Flow IAT Std                   0\n",
       "Flow IAT Max                   0\n",
       "Flow IAT Min                   0\n",
       "Fwd IAT Total                  0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.isnull().sum().head(20)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "4a45d9d6",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(26800, 78)"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "dad51f3c",
   "metadata": {},
   "source": [
    "### Split train set and test set"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "91f828fa",
   "metadata": {},
   "outputs": [],
   "source": [
    "x=data.drop(['Label'],axis=1)\n",
    "y=data['Label']\n",
    "xtrain,xtest,ytrain,ytest= train_test_split(x,y,train_size=0.8, test_size=0.2,random_state=0)\n",
    "#shuffle= False"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "76e3354e",
   "metadata": {},
   "source": [
    "### SMOTE to solve class-imbalance"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "806382e8",
   "metadata": {},
   "outputs": [],
   "source": [
    "#pip install imblearn"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "cd5132e3",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    14569\n",
       "3     2430\n",
       "6     1728\n",
       "1     1579\n",
       "5     1024\n",
       "2       82\n",
       "4       28\n",
       "Name: Label, dtype: int64"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.Series(ytrain).value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "3eec9e2e",
   "metadata": {},
   "outputs": [],
   "source": [
    "from imblearn.over_sampling import SMOTE\n",
    "smote= SMOTE(n_jobs=-1, sampling_strategy={2:1000,4:1000})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "07322500",
   "metadata": {},
   "outputs": [],
   "source": [
    "xtrain,ytrain = smote.fit_resample(xtrain,ytrain)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "c99b7d53",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    14569\n",
       "3     2430\n",
       "6     1728\n",
       "1     1579\n",
       "5     1024\n",
       "2     1000\n",
       "4     1000\n",
       "Name: Label, dtype: int64"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.Series(ytrain).value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "190069fb",
   "metadata": {},
   "source": [
    "# II. ML model training : Ridge Classifier CV "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7168f3c5",
   "metadata": {},
   "source": [
    "RidgeClassifierCV est un modèle de classification linéaire régularisé qui utilise la régression ridge pour résoudre les problèmes de classification. Il est disponible dans la bibliothèque scikit-learn de Python.\n",
    "\n",
    "Le modèle RidgeClassifierCV est une version améliorée du RidgeClassifier standard. ils sont basé sur la régression logistique régularisée et utilise une régularisation L2 pour limiter la complexité du modèle.La différence entre les deux est que RidgeClassifierCV effectue une validation croisée pour sélectionner le meilleur paramètre de régularisation. La régularisation est utilisée pour éviter le surapprentissage et améliorer la généralisation du modèle.\n",
    "\n",
    "Le modèle RidgeClassifierCV est très utile pour les problèmes de classification linéaire, notamment lorsque le nombre de caractéristiques (ou de variables) est élevé et que le risque de surapprentissage est présent. La validation croisée automatique permet d'optimiser la performance du modèle et d'éviter les biais de sélection de paramètres."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "f053a6d3",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.linear_model import RidgeClassifierCV"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "f9cfda13",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Wall time: 331 ms\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "RC = RidgeClassifierCV().fit(xtrain, ytrain)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "19af32b4",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(5360,)"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ypred= RC.predict(xtest)\n",
    "ypred.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "bbfe3be5",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.81      0.94      0.87      3656\n",
      "           1       0.21      0.02      0.04       387\n",
      "           2       0.00      0.00      0.00        14\n",
      "           3       0.97      0.65      0.78       612\n",
      "           4       0.13      0.50      0.21         8\n",
      "           5       0.63      0.54      0.58       231\n",
      "           6       0.81      0.84      0.83       452\n",
      "\n",
      "    accuracy                           0.81      5360\n",
      "   macro avg       0.51      0.50      0.47      5360\n",
      "weighted avg       0.78      0.81      0.78      5360\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(classification_report(ytest,ypred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "9116d1e6",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy of Ridge Classifier CV : 0.8095149253731343\n",
      "\n",
      "Precision of Ridge Classifier CV : 0.7760521911382349\n",
      "\n",
      "Recall of Ridge Classifier CV : 0.8095149253731343\n",
      "\n",
      "Average F1 of Ridge Classifier CV : 0.7801699160439877\n",
      "\n",
      "F1 of Ridge Classifier CV for each type of attack: [0.87088993 0.03755869 0.         0.77603143 0.21052632 0.57808858\n",
      " 0.82532751]\n"
     ]
    }
   ],
   "source": [
    "print(\"Accuracy of Ridge Classifier CV : \"+str(accuracy_score(ytest,ypred)) +\"\\n\")\n",
    "print(\"Precision of Ridge Classifier CV : \"+str(precision_score(ytest,ypred,average=\"weighted\"))+\"\\n\")\n",
    "print(\"Recall of Ridge Classifier CV : \"+str(recall_score(ytest,ypred,average=\"weighted\"))+\"\\n\")\n",
    "print(\"Average F1 of Ridge Classifier CV : \"+str(f1_score(ytest,ypred,average=\"weighted\"))+\"\\n\")\n",
    "print(\"F1 of Ridge Classifier CV for each type of attack: \"+str(f1_score(ytest,ypred, average=None )))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9dca4ef9",
   "metadata": {},
   "source": [
    "###  plot the confusion matrix"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "d438d0bb",
   "metadata": {},
   "outputs": [
    {
     "ename": "NameError",
     "evalue": "name 'confusion_matrix' is not defined",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mNameError\u001b[0m                                 Traceback (most recent call last)",
      "\u001b[1;32m~\\AppData\\Local\\Temp\\ipykernel_15268\\1409899739.py\u001b[0m in \u001b[0;36m<module>\u001b[1;34m\u001b[0m\n\u001b[1;32m----> 1\u001b[1;33m \u001b[0mcm\u001b[0m\u001b[1;33m=\u001b[0m \u001b[0mconfusion_matrix\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mytest\u001b[0m\u001b[1;33m,\u001b[0m\u001b[0mypred\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m      2\u001b[0m \u001b[0msns\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mheatmap\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mcm\u001b[0m\u001b[1;33m,\u001b[0m\u001b[0mannot\u001b[0m\u001b[1;33m=\u001b[0m\u001b[1;32mTrue\u001b[0m \u001b[1;33m,\u001b[0m \u001b[0mlinewidths\u001b[0m\u001b[1;33m=\u001b[0m\u001b[1;36m0.5\u001b[0m\u001b[1;33m,\u001b[0m\u001b[0mlinecolor\u001b[0m\u001b[1;33m=\u001b[0m\u001b[1;34m\"red\"\u001b[0m\u001b[1;33m,\u001b[0m\u001b[0mfmt\u001b[0m\u001b[1;33m=\u001b[0m\u001b[1;34m\".0f\"\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m      3\u001b[0m \u001b[0mplt\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mxlabel\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;34m\"ypred\"\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m      4\u001b[0m \u001b[0mplt\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mylabel\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;34m\"ytrue\"\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m      5\u001b[0m \u001b[0mplt\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mtitle\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;34m\"Confusionn matrix\"\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;31mNameError\u001b[0m: name 'confusion_matrix' is not defined"
     ]
    }
   ],
   "source": [
    "cm= confusion_matrix(ytest,ypred)\n",
    "sns.heatmap(cm,annot=True , linewidths=0.5,linecolor=\"red\",fmt=\".0f\")\n",
    "plt.xlabel(\"ypred\")\n",
    "plt.ylabel(\"ytrue\")\n",
    "plt.title(\"Confusionn matrix\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fa830cab",
   "metadata": {},
   "source": [
    "# III. ML model training : VBGMM"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d6003f49",
   "metadata": {},
   "source": [
    "VBGMM signifie \"Variational Bayesian Gaussian Mixture Model\", soit en français \"Modèle de mélange de Gaussiennes bayésien variationnel\". C'est une méthode d'apprentissage non supervisé qui permet de modéliser des données multidimensionnelles à partir d'un mélange de distributions gaussiennes.\n",
    "\n",
    "Le VBGMM est une extension du GMM (Gaussian Mixture Model) qui utilise une approche bayésienne et une méthode d'optimisation variationnelle pour estimer les paramètres du modèle. Contrairement au GMM classique qui nécessite de fixer à l'avance le nombre de composantes gaussiennes du mélange, le VBGMM utilise un processus de Dirichlet pour permettre au modèle de déterminer automatiquement le nombre optimal de composantes gaussiennes nécessaires pour modéliser les données.\n",
    "\n",
    "Le VBGMM n'est pas un classificateur en tant que tel, mais il peut être utilisé pour effectuer de la classification en ajoutant une étape supplémentaire de labélisation des données.\n",
    "\n",
    "en effet on peut utiliser le modèle VBGMM comme un algorithme de clustering pour regrouper les données en clusters, puis attribuer une étiquette de classe à chaque cluster en utilisant le mode de la classe majoritaire dans chaque cluster comme étiquette de classe pour ce cluster."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "494c39cf",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.mixture import BayesianGaussianMixture"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "c1b81771",
   "metadata": {},
   "outputs": [],
   "source": [
    "VBGMM= BayesianGaussianMixture(n_components=7)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "a48c08ca",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Wall time: 52.2 s\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<style>#sk-container-id-1 {color: black;background-color: white;}#sk-container-id-1 pre{padding: 0;}#sk-container-id-1 div.sk-toggleable {background-color: white;}#sk-container-id-1 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-container-id-1 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-1 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-1 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-1 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-1 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-1 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-1 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-1 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-1 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-1 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-container-id-1 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-container-id-1 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-1 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-1 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-1 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: 0;}#sk-container-id-1 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;position: relative;}#sk-container-id-1 div.sk-item {position: relative;z-index: 1;}#sk-container-id-1 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-1 div.sk-item::before, #sk-container-id-1 div.sk-parallel-item::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: -1;}#sk-container-id-1 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-1 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-1 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-1 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-1 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;}#sk-container-id-1 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-1 div.sk-label-container {text-align: center;}#sk-container-id-1 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-container-id-1 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-1\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>BayesianGaussianMixture(n_components=7)</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-1\" type=\"checkbox\" checked><label for=\"sk-estimator-id-1\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">BayesianGaussianMixture</label><div class=\"sk-toggleable__content\"><pre>BayesianGaussianMixture(n_components=7)</pre></div></div></div></div></div>"
      ],
      "text/plain": [
       "BayesianGaussianMixture(n_components=7)"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "%%time\n",
    "VBGMM.fit(xtrain)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6660b694",
   "metadata": {},
   "source": [
    "### Obtenir les étiquettes de cluster pour les données\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "3279b07c",
   "metadata": {},
   "outputs": [],
   "source": [
    "cluster_labels = VBGMM.predict(xtrain)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "078f9cac",
   "metadata": {},
   "source": [
    "### Associer une étiquette de classe à chaque cluster\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "f5cdbf58",
   "metadata": {},
   "outputs": [],
   "source": [
    "from scipy.stats import mode\n",
    "\n",
    "labels = np.zeros_like(cluster_labels)\n",
    "for i in range(7):\n",
    "    mask = (cluster_labels == i)\n",
    "    labels[mask] = mode(ytrain[mask])[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "8b637213",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0, 3, 4], dtype=int64)"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.unique(labels)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "677c412a",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.67      0.95      0.79     14569\n",
      "           1       0.00      0.00      0.00      1579\n",
      "           2       0.00      0.00      0.00      1000\n",
      "           3       0.65      0.46      0.54      2430\n",
      "           4       0.85      0.79      0.82      1000\n",
      "           5       0.00      0.00      0.00      1024\n",
      "           6       0.00      0.00      0.00      1728\n",
      "\n",
      "    accuracy                           0.68     23330\n",
      "   macro avg       0.31      0.31      0.31     23330\n",
      "weighted avg       0.52      0.68      0.58     23330\n",
      "\n"
     ]
    }
   ],
   "source": [
    "ypred=labels\n",
    "print(classification_report(ytrain,ypred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "17592e49",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy of VBGMM : 0.676553793399057\n",
      "\n",
      "Precision of VBGMM : 0.5233980988796071\n",
      "\n",
      "Recall of VBGMM : 0.676553793399057\n",
      "\n",
      "Average F1 of VBGMM : 0.582746270929516\n",
      "\n",
      "F1 of VBGMM for each type of attack: [0.7873239  0.         0.         0.53753027 0.81875    0.\n",
      " 0.        ]\n"
     ]
    }
   ],
   "source": [
    "print(\"Accuracy of VBGMM : \"+str(accuracy_score(ytrain,ypred)) +\"\\n\")\n",
    "print(\"Precision of VBGMM : \"+str(precision_score(ytrain,ypred,average=\"weighted\"))+\"\\n\")\n",
    "print(\"Recall of VBGMM : \"+str(recall_score(ytrain,ypred,average=\"weighted\"))+\"\\n\")\n",
    "print(\"Average F1 of VBGMM : \"+str(f1_score(ytrain,ypred,average=\"weighted\"))+\"\\n\")\n",
    "print(\"F1 of VBGMM for each type of attack: \"+str(f1_score(ytrain,ypred, average=None )))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c89e0295",
   "metadata": {},
   "source": [
    "###  plot the confusion matrix"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "80c73af7",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXsAAAEWCAYAAACHVDePAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjQuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/MnkTPAAAACXBIWXMAAAsTAAALEwEAmpwYAAA+NklEQVR4nO3deZyNdfvA8c81Y0JlTRgziqKeFpXCI1pIUfYW0ZNSPz1KWtQjpZRHpbTHo5WEypYIIVuEsi/F2GUbxoy1ss9y/f4494wzjJlzxlnuM3O9ve6XM997u+aceV3zne9939dXVBVjjDEFW1S4AzDGGBN8luyNMaYQsGRvjDGFgCV7Y4wpBCzZG2NMIWDJ3hhjCgFL9uaMiUhxEZkoIn+KyLdncJz7RWRaIGMLFxG5UUTWhTsOYzKJ3WdfeIjIv4BngX8AfwMrgD6qOu8Mj/sA8CRQT1XTzjROtxMRBaqr6sZwx2KMr6xnX0iIyLPAh8AbQAXgAuBjoFUADn8hsL4wJHpfiEiRcMdgzMks2RcCIlIKeBXooqpjVfWQqqaq6kRVfc7ZpqiIfCgiO53lQxEp6qxrICKJIvIfEUkRkSQRedhZ1xt4BWgrIgdFpKOI/FdEvvY6fxUR0cwkKCIPicgfIvK3iGwWkfu92ud57VdPRBY7w0OLRaSe17rZIvKaiPziHGeaiJQ7zfefGX93r/hbi0hTEVkvIvtE5EWv7euIyHwROeBsO0BEznLWzXE2+835ftt6Hf95EdkFfJnZ5uxzsXOOa52vK4nIHhFpcCafqzH+sGRfOFwPFAPG5bLNS0Bd4BrgaqAO0NNrfUWgFBAHdAQ+EpEyqtoLz18Lo1T1XFX9IrdAROQcoD9wh6qWAOrhGU46ebuywCRn2/OA94FJInKe12b/Ah4GygNnAd1yOXVFPO9BHJ5fTgOB9sB1wI3AKyJykbNtOvAMUA7Pe9cIeBxAVW9ytrna+X5HeR2/LJ6/cjp5n1hVNwHPA9+IyNnAl8AQVZ2dS7zGBJQl+8LhPGBPHsMs9wOvqmqKqu4GegMPeK1Pddanqupk4CBwaT7jyQCuFJHiqpqkqgk5bNMM2KCqX6lqmqqOANYCLby2+VJV16vqEWA0nl9Up5OK5/pEKjASTyLvp6p/O+dPAK4CUNWlqrrAOe8W4DPgZh++p16qesyJJxtVHQhsABYCsXh+uRoTMpbsC4e9QLk8xpIrAVu9vt7qtGUd46RfFoeBc/0NRFUPAW2Bx4AkEZkkIv/wIZ7MmOK8vt7lRzx7VTXdeZ2ZjJO91h/J3F9ELhGRH0Rkl4j8hecvlxyHiLzsVtWjeWwzELgS+J+qHstjW2MCypJ94TAfOAq0zmWbnXiGIDJd4LTlxyHgbK+vK3qvVNWpqnobnh7uWjxJMK94MmPakc+Y/PEJnriqq2pJ4EVA8tgn19vaRORcPBfIvwD+6wxTGRMyluwLAVX9E8849UfOhcmzRSRGRO4QkbedzUYAPUXkfOdC5yvA16c7Zh5WADeJyAXOxeEemStEpIKItHTG7o/hGQ5Kz+EYk4FLRORfIlJERNoClwM/5DMmf5QA/gIOOn91dD5pfTJw0Sl75a4fsFRVH8FzLeLTM47SGD9Ysi8kVPV9PPfY9wR2A9uBJ4DvnU1eB5YAvwMrgWVOW37ONR0Y5RxrKdkTdBTwHzw99314xsIfz+EYe4HmzrZ7ge5Ac1Xdk5+Y/NQNz8Xfv/H81THqpPX/BYY6d+vcm9fBRKQVcDueoSvwfA7XZt6FZEwo2ENVxhhTCFjP3hhjCgFL9sYYUwhYsjfGmELAkr0xxhQC7i3Y5KksaIwxeVPN6zmIPKXu+cPnnBNT7qIzPl+ouTfZA6m7N4U7hFPEnH8xRWIq5b1hGKSl7nRlbGmpOylzzsXhDuMU+w9t4qyz4vLeMAyOH9/h2s/SlXEF4iAZOT3uUXC4OtkbY0zIaEa4IwgqS/bGGAOQYcneGGMKPLWevTHGFALpBXuiNUv2xhgDdoHWGGMKBRvGMcaYQsAu0BpjTMFnF2iNMaYwsJ69McYUAump4Y4gqCKqEFrPN97npmbtaN3+say2/30+jDsf7MzdHbrw764vkrJ7LwCpaWm8+Nq73PlAZ1r8qxMDh52YbGjy9Nnc+UBn7nywM48+25P9B/4EIGlXCg8/8Tz3PNSFOx/szJxfFwXte2nSuAEJq+awdvU8uj/XJWjn8Ve44/otYTa/LJzEnF8n8NOccQA8/+JTJKyfx5xfJzDn1wnc1vhmAIoUKcLHn73NLwsnsWDpjzzzn8dyO/QZ+fyzd0ncvoLly2Zktb35Zk9W/j6bpUum8+3oQZQqVRKAWrWuYfGiqSxeNJUli6fRquXtQYsrN+H+LE/HrXGhGb4vEci9M1WJ6Mm1cZasWMnZxYvz4mvv8v3Xnik8Dx46xLnnnAPA19+OZ9PmbfTq/iSTps1i1rwFvPtqD44cPUqr+x/lywFvU+H8ctzS6n7Gf/MZZUqX4r2PvqBYsaJ06die/77Vj39ccjHt7mzOps1b6dztFaZ9NzRbDIGojRMVFcWahLnc3vQ+EhOTWDB/Mu0feJw1azac0XHPtG5JMOPytTbObwmzaXjTnezbuz+r7fkXn+LQwUMM6P9Ftm3vadOCO5o1ouNDXSlevBgLlvxI8zvuZ/s23+Yk96c2zg03/JODBw/x5eAPqXntrQDceutNzJr1C+np6bzR50UAXnzpDYoXL8bx46mkp6dTsWJ5liyexoVVriM93fdb+860Nk6h+xk7vuOMC5MdS5jpczIsekWjiCuEFlE9+1rX1KBUyRLZ2jITPcCRI0cR5yMQEY4cPUpaWjrHjh0nJiaGc885G3X+HTl6FFXl4KHDlC9XNmufQ4cOA/D3ocOcX+68oHwfdWrXZNOmLWzevI3U1FRGjx5PyxZNgnKughDX6SjK2WefTXR0NMWcBPv33weDcq558xayf/+BbG0zZszJSuALFy4jLi4W8PwcZrYXK1aUcHSo3PpZujUuoMD37IOW7EXkHyLyvIj0F5F+zuvLgnGufp8NodGdDzBp2iyeeOQBAG5reAPFixWjYat/cdtdD/LQfXdRqmQJYooU4eVuT3DnA51p2Op+/tiyjbuae37YHv+/9vwwdRaNWrfn8W6v8OIznYMRLpXiKrI9cWfW14k7kqhUqWJQzuUPN8SlqowdP4RZc7+nw8Nts9r//egDzFvwA//7+E1KlfYMl4wf9yOHDx9m7ab5rFwzhwH9B3Fg/58hjTfTQw+1ZerUWVlf165dkxXLZ7Js6QyeeKKHX736QHDDZ5kTt8YFeC7Q+rpEoKAkexF5HhgJCLAIWOy8HiEiL+SyXycRWSIiSz7343xPP/oQM8d9RbPGDRn+3UQAVq5eR3RUFD+N/4Yfxwxh6IixbN+RRGpaGqPGTeLbLwcwa/w3XHJxVQZ9NRqAyTNm06rprcz8/ms+fvdVerz2DhlB+GBFTv0L0A3DaW6I6/Zb29Lghla0uev/eKRTe+rVr83gQd9Qs8Yt3Hh9C5KTd/P6Gz0AuK7WVaSnp3NZtXpcc2UDujzZkQurVA5pvAAvPP8kaWnpDB8xNqtt8eLlXFOzEfXqN6N79ycoWrRoSGNyw2eZE7fGBaAZqT4vkShYPfuOQG1V7auqXztLX6COsy5Hqvq5qtZS1Vqd8nHSZo0bMGP2L4DnImz9urWIKVKE88qU5pqrLidh7QbWbvBcB7ggvhIiQpNGN7Ji5WoAxk6cSpNbbgLgmisv4/jxVPb/+Vc+IsndjsQkKsefGPeMj4slKSk54Ofxlxvi2rUrBYA9u/fxw8TpXHvdVexO2UtGRgaqytAvR3FdrasBuOfelsycPpe0tDT27N7HwgVLqXltjZDG+0D7e2ja9FYe7PBEjuvXrt3IoUOHueKKS0Malxs+y5y4NS7Aevb5lAHkdBUn1lkXMFu3n7gYN2vuAqpeGO85UYXzWbT0N1SVw0eO8nvCWqpeWJkK5cqxacs29jnjr/MXLeeiKhd49qlYnoVLVgCwacs2jh07TtnSpQIZLgCLl6ygWrWqVKlSmZiYGO69txUTf5gW8PNEWlxnn12cc889J+v1LbfcwJrVG6hQ4fysbZq3aMya1esBSNy+kxtvrpu1fa06NdmwLnQT3jRu3IBu3R7nrrsf5siRo1ntVapUJjo6GoALLojjkksuYuvW7SGLC8L/WUZaXECBH7MP1n32XYGZIrIByPwpvwCoBuTcBfLBc736snj57xw48JdnXL3jA8ydv5gt2xKRKKFSxfK88tyTANx3Vwt6vvE+rds/hqK0btqYS6tVBaDzw/fToUt3ihSJplLF8vR56T+e4z/xCL3e6s+w0eMQhNdfejbHPzvPVHp6Ok937cnkScOJjopiyNBRrHYSWDiFO67zy5fj6xEfAxBdpAjfjZ7AzBlz+HTgu9S46jJUlW1bd/DMUz0BGPT51wz49C1+XTwFEWH4V2NISFgXlNi+GjaAm266nnLlyvLHpsW8+tp7nuGZs85iyuQRACxctIwnnuhB/Xp1eO65x0lNTSMjI4Onnn6JvV53F4VCuD/LSIsLKPCF0IJ266WIROEZtonDM16fCCxWVd/e0RxuvXQDm5bQfzYtof9sWkL/BOLWy6OLvvU5GRar0ybibr0M2hO06ik0sSBYxzfGmICK0LF4X1m5BGOMAZu8xBhjCgXr2RtjTMHn6+XESGXJ3hhjwHr2xhhTKETo/fO+iqhCaMYYEzQBfIJWRAaLSIqIrPJqe0dE1orI7yIyTkRKe63rISIbRWSdiDTxar9ORFY66/qL8+CPiBQVkVFO+0IRqZJXTJbsjTEGPHfj+LrkbQhw8kQG04ErVfUqYD3QA0BELgfaAVc4+3wsItHOPp8AnYDqzpJ5zI7AflWtBnwAvJVXQJbsjTEGAlouQVXnAPtOapumqpm/KRYA8c7rVsBIVT2mqpuBjUAdEYkFSqrqfPU8/ToMaO21T+ZkG2OARpLH4/6W7I0xBvwaxvGu0Oss/tZu/D9givM6jhNlZcBTbSDOWRJzaM+2j/ML5E8g1wk4XH2BNuZ89z1iD55Hxt3KrbHtP+S+0hfgKUvgVm79LN0a1xnz424cVf0c8KcSexYReQlIA77JbMrpFLm057bPabk62V9Z/p/hDuEUq1IWurI2CLi4bonF5Te3xubauAJxkBDcjSMiHYDmQCM9UZgsEfCeiCEe2Om0x+fQ7r1PoogUAUpx0rDRyWwYxxhjINAXaE8hIrcDzwMtVfWw16oJQDvnDpuqeC7ELlLVJOBvEanrjMc/CIz32qeD8/oe4CfNo6qlq3v2xhgTMgF8qEpERgANgHIikgj0wnP3TVFgunMtdYGqPqaqCSIyGliN54+ULl7VgTvjubOnOJ4x/sxx/i+Ar0RkI54efbu8YrJkb4wxENBhHFW9L4fmL3LZvg/QJ4f2JcCVObQfBdr4E5Mle2OMASuXYIwxhYIle2OMKQSCNGufW1iyN8YYgDSbvMQYYwq+Al710pK9McaAjdkbY0yhUMDH7CP2CdrXPnyJnxMmM+7nb7LaHu/2CDNXTGDMzGGMmTmMGxtdD0Czu5tktY2ZOYzfk37l0iuqA/Dl2I+Z+MuorHVly5UJSfxNGjcgYdUc1q6eR/fnuoTknL5wa1wAUVFRLF40lfHjhua9cQi59T2zuPwUwHr2bhSxPfvvR05i+BdjeGPAK9nav/psJEM+GZ6tbdJ3U5n03VQAql92Mf2Hvs26hA1Z6194vBcJv60NftCOqKgo+vfrw+1N7yMxMYkF8ycz8YdprFmzIe+dC2FcmZ568hHWrt1AyRIlwh1KFre+ZxZXPkRoEvdVxPbsly5YwZ8H/vJ7v6Z33saUcdODEJHv6tSuyaZNW9i8eRupqamMHj2eli2a5L1jIY0LIC4ulqZ3NGLw4BHhDiUbt75nFpf/ND3d5yUSRWyyP537/q8NY2d9zWsfvkTJUqf2AG9vdSuTx03L1vZav56MmTmMR595OCQxVoqryPbEE2ViE3ckUalSxZCcOzdujQvg/fd680KP18lwWe/Lre+ZxZUPBXwYJ+TJXkROm1G9JwTIT6HoUUPHcsc/7+buWx5gd/Jenuv9VLb1Na69giNHjrJx7R9Zbc8/3ou7GrTnwZaPcV3da2jZ5o58nNk/OU0ok0fBupBwa1zNmt5KSsoeli1fGe5QTuHW98ziyocAzlTlRuHo2fc+3QpV/VxVa6lqLX+nfQHYu3sfGRkZqCpjvh7PlTUvz7b+jta3njKEk7JrNwCHDx1m0thpp+wTDDsSk6gcf6ImeHxcLElJyUE/b17cGle9erVo0bwxG9cv4JuvP6Zhw/oMHdI/3GEB7n3PLK58yFDflwgUlGTvzJ6e07ISqBCMcwKUK39iVq5GTW/O1oMXERq3aMSU708k++joaEqXLQVAkSLR3Hxb/Wz7BMviJSuoVq0qVapUJiYmhnvvbcXEH6blvWMhjeulnn2pclEtql1Sl/vbP86sWb/Q4aGn8t4xBNz6nllc+VDAh3GCdTdOBaAJsP+kdgF+DcQJ3v70VWrXu5bSZUszY/kEPn5nILXrXculV1YHhR3bk+jdrW/W9rWur0lyUgqJW0+MF55VNIbPRvYjJqYIUVFRLJi7mDFfj8/pdAGVnp7O0117MnnScKKjohgydBSrV68P+nkjNS43c+t7ZnHlQ4ReePWVBGO8TES+AL5U1Xk5rBuuqv/y4SBq0xL6x7VTxllcfnNrbK6N6/iOnOZk9cvh9//tczI8+9mBZ3y+UAtKz15VO+ayLu9Eb4wxoRahY/G+itiHqowxJqAi9C4bX1myN8YYsJ69McYUBhqhd9n4ypK9McZAgb8bx5K9McZAgR/GKXC1cYwxJl8C+FCViAwWkRQRWeXVVlZEpovIBuf/Ml7reojIRhFZJyJNvNqvE5GVzrr+4tSbEJGiIjLKaV8oIlXyismSvTHGQKDLJQwBbj+p7QVgpqpWB2Y6XyMilwPtgCucfT4WkWhnn0+ATkB1Z8k8Zkdgv6pWAz4A3sorIEv2xhgDAS2EpqpzgH0nNbcCMmfeGQq09mofqarHVHUzsBGoIyKxQElVna+ep1+HnbRP5rHGAI0ye/2nY8neGGPAr569d4VeZ/GldmMFVU0CcP4v77THAdu9tkt02uKc1ye3Z9tHVdOAP4HzyIWrL9CuSlkY7hBylJa6M++NwsStsVlc/nNrbG6N60xpmu9346jq50B+KrHnJKceuebSnts+p+XqZB9f5opwh3CKxP0JrqwNAi6uW2Jx+c2tsbk2rkAcJPh34ySLSKyqJjlDNClOeyJQ2Wu7eGCn0x6fQ7v3PokiUgQoxanDRtnYMI4xxkAoJi+ZAHRwXncAxnu1t3PusKmK50LsImeo528RqeuMxz940j6Zx7oH+EnzqGrp6p69McaETAB79iIyAmgAlBORRKAX0BcYLSIdgW1AGwBVTRCR0cBqPH+kdFHVzDGlznju7CkOTHEWgC+Ar0RkI54efbu8YrJkb4wxgAYw2avqfadZ1eg02/cB+uTQvgS4Mof2ozi/LHxlyd4YYwD8uEAbiSzZG2MMFPhyCZbsjTEGLNkbY0xhEIwpWt3Ekr0xxoD17I0xplCwZG+MMQWfphXsmaoi9gnad//3GivW/cyMX8ZltZUuXZLhYwcyd/Ekho8dSKlSJbPWden6CPOWTObnhRO5+ZZ6We01rr6cGfPGMm/JZF59s0fI4m/SuAEJq+awdvU8uj/XJWTnzYtb4wL3xmZx+cetcZHhxxKBIjbZfzv8e9q3eSxbW5euj/DLzwu4sXYzfvl5AV26dgSg+qUX0equO7ilXivat3mMPu+8TFSU51t/892X6f5Mb26o1ZSqF19Aw1tvCHrsUVFR9O/Xh+Yt2lPj6oa0bduayy6rHvTzRmpc4N7YLK6CERd4HqrydYlEEZvsF85fyoH9f2Zra3xHQ74d6Skd8e3I8TRpeovTfgvjx07h+PFUtm/bwZbN27jmuhqUr1COc0ucw7LFvwEwZuSErH2CqU7tmmzatIXNm7eRmprK6NHjadmiSd47FtK4wL2xWVwFIy4g0JOXuE7Qkr2I/ENEGonIuSe1nzx7S8CUK38eKcl7AEhJ3sN555cFIDa2PEk7dmVtt2tnMrGx5akYW4GknclZ7Uk7k6kYWyFY4WWpFFeR7YknysQm7kiiUqWKQT9vXtwaF7g3NovLP26NC7BhnPwQkafwVGd7ElglIq28Vr+Ry35ZEwIEqlC0c+BTmlSVnCZ2CcW9tuE6b17cGhe4NzaLyz9ujQsK/jBOsO7G+TdwnaoedCbCHSMiVVS1HzkX3QdOmhBARF/186R7UvZSvkI5UpL3UL5COfbu9pR3TtqZTGzcid5DxUoV2LVrN0k7dxFb6URPPrZSBZJ3pZxy3EDbkZhE5fgTNcHj42JJSkrOZY/QcGtc4N7YLC7/uDUuAE2LzCTuq2AN40Sr6kEAVd2Cp9TnHSLyPrkk+zM1/cfZtGnn+SOiTbtWTJsyy2mfRau77uCss2KofEEcVS+6gBVLV5KSvIeDBw9zba2rALinXUumTZ4VrPCyLF6ygmrVqlKlSmViYmK4995WTPxhWtDPG6lxgXtjs7gKRlxAgR/GCVbPfpeIXKOqKwCcHn5zYDBQIxAnGDDwba6vX5uy55Vm8aoZvNf3YwZ8OIhPB79Hu/Z3sSMxiccefhaA9Ws3MfH7qfw0fwLpaWn07N6HjAzPJ/Zit9d4/6PXKVasGLNnzOWnGXMDEV6u0tPTebprTyZPGk50VBRDho5i9er1QT9vpMYF7o3N4ioYccGZzEkSGSQY42UiEg+kqequHNbVV9VffDiI2rSE/nHtlHEWl9/cGptr4zq+44xHDPY2u9nnZHjepJ+DNkIRLEHp2atqYi7r8k70xhgTYgW9Z2/lEowxBtCAzFruXpbsjTEG69kbY0yhYMneGGMKA424a65+sWRvjDEU/J59xBZCM8aYQNIM8XnJi4g8IyIJIrJKREaISDERKSsi00Vkg/N/Ga/te4jIRhFZJyJNvNqvE5GVzrr+klO9CR9ZsjfGGCAjXXxeciMiccBTQC1VvRKIBtoBLwAzVbU6MNP5GhG53Fl/BXA78LGIRDuH+wToBFR3lnwXkrRkb4wxeIZxfF18UAQoLiJFgLOBnUArYKizfijQ2nndChipqsdUdTOwEagjIrFASVWdr56nX4d57eM3S/bGGEPghnFUdQfwLrANSAL+VNVpQAVVTXK2SQLKO7vEAdu9DpHotMU5r09uzxdL9sYYA6j6vniXY3eWTpnHccbiWwFVgUrAOSLSPpdT5/TbQ3NpzxdX342TuD8h3CHkKC11Z94bhYlbY7O4/OfW2Nwa15ny5cJr1rbe5dhPdSuwWVV3A4jIWKAekCwisaqa5AzRZNZTTwQqe+0fj2fYJ9F5fXJ7vrg62Vcvd224QzjFhj3LXFkIClxcpCp1JxeWDUix04Daum+lK98vcPdn6cq4AnCMvC68+mEbUFdEzgaOAI2AJcAhoAPQ1/l/vLP9BGC4UwK+Ep4LsYtUNV1E/haRusBC4EHgf/kNytXJ3hhjQsWfnn2ux1FdKCJjgGV4fg8tx/NXwLnAaBHpiOcXQhtn+wQRGQ2sdrbvoqrpzuE6A0OA4sAUZ8kXS/bGGANoAJ+gVdVeQK+Tmo/h6eXntH0foE8O7UuAKwMRkyV7Y4zBnqBFRCqIyBciMsX5+nLnzxBjjCkwMlR8XiKRL7deDgGm4rlwALAe6BqkeIwxJixUxeclEvmS7Mup6micaXZVNQ1Iz30XY4yJLIEql+BWvozZHxKR83Bu5nduA/ozqFEZY0yIBepuHLfyJdk/i+c+0ItF5BfgfOCeoEZljDEhFqlj8b7KM9mr6jIRuRm4FM/ju+tUNTXokRljTAhF6li8r3y5G+dB4F/AdcC1wH1OW1i92e8VFqyezqQ5o7Laur7QmYmzRzJh1nC+HP0R5SuUy1r36NMPM2PR90yd/x03NLw+q71p69uYOHskk+eOpvsrT4Us/iaNG5Cwag5rV8+j+3NdQnbevIQ6rnf692bp2tlMmzc2q61py9uY/stYNu9eQY1rLs9qL12mFCO/H8TqrQt49a0e2Y5z5dWXMXXud/y8+Af+++bzQY154OfvsTPxN1Ysn5nVdvfdzfltxU8cP7qd6669Kqjn95X9jPnHn9o4kciXC7S1vZYbgf8CLYMYk0/GjpzI/7V7MlvboAHDaNGgHS0b/otZ0+fyRLd/A1Dtkqo0a92Ypje0oWPbJ+n91gtERUVRukwpnu/VlQ53P0bTG+/lvPLncf2NtYMee1RUFP379aF5i/bUuLohbdu25rLLqgf9vG6M69sRE+hwb+dsbevXbuTRDs+y8Nel2dqPHTvOu29+RJ9e751ynD7v9qTHM725uXZzql50IQ0a3RC0mIcNG02z5vdna0tIWEube//N3LkLgnZef9jPmP8K/a2Xqvqk1/JvoCZwVvBDy93i+cv5c3/268QHDx7Kel387OJZv4Eb3dGASd9P4/jxVBK37WTrlu1cde0VVL4wjs2btrJv7wEAfv15IU2a5/iAW0DVqV2TTZu2sHnzNlJTUxk9ejwtWzTJe8cCGNei+Us5cNLnuHH9Zv7YuOWUbY8cPsKShcs5duxYtvbyFcpxbolzWbbkdwC+GzWRxk0bBi3mufMWsm//gWxta9duZP36TUE7p7/sZ8x/GRni8xKJ8lPi+DCeQj25EpE6IlLbeX25iDwrIk3zcT6/PPPi48xZMYmWd99Ov7c+AaBC7Pkk7diVtc2unclUjC3P1s3bubh6FeIqxxIdHc1tTRsQG1ch2CFSKa4i2xNPFK9L3JFEpUoVg37evLg1rrxUiC3Prp3JWV8nOZ9vYebWz9KtcYH17BGRiSIywVl+ANZxolrb6fbpBfQHPhGRN4EBeIoAvSAiL+WyX1aN6NPVDs3LB298zE3XNGPCdz/SvmPbzOOesp2q8teff9PruTfpN7AvIyYOInFbEmlpwX+E4HTxhJtb48pLznGHIRAXcetn6da4oOA/VOXLrZfver1OA7aqauLpNnbcA1wDFAV2AfGq+peIvIOnVOcpBX/gpBrRIvqOD8GdzsTvpjBweD/6v/0Zu3amEBt3ovdQsVIFknftBuCnaXP5adpcANo+cCcZ6cFP9jsSk6gcf6JMbHxcLElJybnsERpujSsvu3YmU7HSib/IYitVIHlXSi57FHxu/SzdGhcU/Fsvc+3ZO5PevqyqPzvLLz4keoA0VU1X1cPAJlX9C0BVj+A8iRsMF150ov5/o9tvzhr3nfnjzzRr3Zizzooh/oJKVKlamd+XeSZGKVvOM8F7yVIluP//2jD66++DFV6WxUtWUK1aVapUqUxMTAz33tuKiT9MC/p5IzWuvKQk7+HQwUPUrOW5C+buti2YPmVWmKMKL7d+lm6NCzxPjfq6RKJce/ZO8fzDIlJKVf15ava4iJztJPvrMhtFpBQBSvYffNaHOvVrUaZsaeb+Npl+b39Gg1vrU/XiC8nIUHYmJvFKtzcA2LjuD6ZMmM6UeWNIS0/jvy+8RUaGJ4yX+3TjH1dcAsCAdwey5Y9tgQgvV+np6TzdtSeTJw0nOiqKIUNHsXr1+qCf141x9f/8La6vX4sy55VmwcrpfND3Yw4c+JPefXtQ9rwyfDniI1avWsuDbTx37MxbPoUSJc4lJiaGxk1v4YF7HmXDuj94qdvrvDfgdYoVK8rsmfOYNWNe0GL++quPuPmm6ylXrixb/lhC71ffZd/+A/T74HXOP78sE8YP47ffEmh60h07oWQ/Y/5LzyjYs7RKXuNlTlH9usB0PDOtAKCqp70pXUSKquqxHNrLAbGqujLvyERtpir/uHYWIZupym9u/ixdGdfxHWc8BjO34j0+d9pv3DUm4sZ8fBmzn+Qs3nJ9U3JK9E77HmCPb6EZY0zoaI7zexccviT70qraz7tBRJ4OUjzGGBMWGZE6GO8jXwapOuTQ9lCA4zDGmLDKQHxeItFpe/Yich+emjhVRWSC16oSwN5gB2aMMaFUmIdxfgWSgHrAl8A+p/1v4Pcgx2WMMSGVXliTvapuBbaKyKfAO8AyYDAwR93yyJsxxgRIAZ9v3KdCaC/hqYXzBZ6x+g0i8oaIXBzk2IwxJmQy/FgikU9PETg9+V3OkgaUAcaIyNtBjM0YY0JGEZ+XvIhIaREZIyJrRWSNiFwvImVFZLqIbHD+L+O1fQ8R2Sgi60SkiVf7dSKy0lnXX3IqLuQjXwqhPSUiS4G3gV+AGqraGc+TsXfn98TGGOMmGeL74oN+wI+q+g/gamAN8AIwU1WrAzOdrxGRy4F2wBXA7cDHTqkagE+ATnhGV6o76/PFl/vsywF3OWP4WVQ1Q0Sa5/fExhjjJoG6pVJESgI34dyirqrH8ZSQaQU0cDYbCswGngdaASOdh1E3i8hGoI6IbAFKqup857jDgNbAlPzE5cuY/SsnJ3qvdWvyc1JjjHGbdD8W73LsztLJ61AXAbuBL0VkuYgMEpFzgAqqmgTg/J856UIcsN1r/0SnLc55fXJ7vvjSsw+bDXuWhTuEHKWl7sx7ozBxa2xb9+VdDikc3Pp+gXtjc2tcZyrDj+HwbOXYT1UEz3zdT6rqQhHphzNkcxo5nVhzac8XVyf7okXjwx3CKY4dS3RlIShwcZEqF8dVoeSl4Q4jR8l/rXPte+bKuAJwjADeT54IJKrqQufrMXiSfbKIxKpqkojEAile21f22j8e2Om0x+fQni8Fu6anMcb4KFC3XqrqLmC7iGT2JBoBq4EJnCg/04ETM/5NANqJSFERqYrnQuwiZ6jnbxGp69yF8yB5zBKYG1f37I0xJlQCPI/4k8A3InIW8AfwMJ7O9WgR6QhsA9oAqGqCU0p+NZ4/UrqoauaUeZ2BIUBxPBdm83VxFizZG2MMENhyCaq6AqiVw6pGp9m+DzlM16qqS4ArAxGTJXtjjCHgPXvXsWRvjDFEbhkEX1myN8YYIncicV9ZsjfGGGwYxxhjCgUbxjHGmEIgvYD37AvEQ1WfffYu27ctZ9nSGVltNWpcxs+zv2fpkumM/W4wJUqcm22fypUrsXfPWp7p+miowwWgSeMGJKyaw9rV8+j+XJewxJATt8QVH1+JGdO+ZeXvs/ltxU88+URHAMqUKc2Pk0ewJmEeP04eQenSpUISz8XVqjJz7risZeP2JXTq/CBX1PgHk2eMZObccUydPYaa19bI2ufyKy5h0vSR/LxgIrN/nUDRomeFJNZMbvksT+bWuKyefQT46qtvadHygWxtn37yDj1f7st1tW5j/ISpPPvsY9nWv/N2L6ZOnRXKMLNERUXRv18fmrdoT42rG9K2bWsuu6x6WGJxa1xpaWk81703Na5qQP0bWtC580Ncdll1nu/ehZ9mzeOyK27gp1nzeL57aJLFpo2baXTjnTS68U5uu/lujhw5wuQfZvDKq8/xbt+PaHTjnbzdpz8vv/ocANHR0Xz0+Ts890wvbq7bgjubPUhqaiAe6veNmz7LSIgLLNlHhHnzFrJ//4FsbZdcchFz5y4AYObMOdzZ+o6sdS1bNGHz5m2sXrM+lGFmqVO7Jps2bWHz5m2kpqYyevR4WrZokveOhSiuXbtSWL5iFQAHDx5i7doNxFWqSIsWTRj21bcADPvqW1q2zHd573y7scH1bNm8ncTtO1FVSpT0/NVYsmQJknd5yp00uKU+qxPWsXrVOgD27z9ARkbo0oSbPstIiAs8d+P4ukSikCV7pxZzyCQkrKNF88YA3H1Xc+LjPcWbzj67OP/5T2de7/NBKMPJplJcRbYnnqhnlLgjiUqVKoYtnkxujevCC+O55uorWbhoORXKl2OXk1B37Uqh/PnnhTyeO+9qyrgxkwB4+YU3eOXV51iWMIter3enT+/3Abi4WhVUlZFjBzF9znd0ebpjSGN062fp1rgg4JOXuE5Qkr2ITDhpmQjclfl1Lvtl1Yg+Xe1QXz36aDcee6wD83+dxLklzuH48VQAXnn5P/T/3yAOHTp8hmfIv5xmFnPDHO5ujOucc85m9KiBPNutF3//fTCssQDExMTQuOktTPz+RwAe6ngfr7zYl2uvaMgrL77JBwNeByC6SBH+ef11PP5IN1o2uZ+mzW/jxpvrhixON36W4N64oOAP4wTrbpx4PEV9BnGiLnMt4L3cdspWI1pEnzyDANat30Sz5vcDUL1aVe643VOSonadmtx5V1PeeONFSpcqSUaGcvToUT75dOgZnM0/OxKTqBx/okxsfFwsSUnJITv/6bgtriJFivDtqIGMGDGO77/31H9KTtlDxYrl2bUrhYoVy5Oye29IY2p0242s/G01u53z3ntfa1563lPSZMK4H3m/vyfZJ+3cxa/zFrNv3wEAZkz7mRpXX87cnxeEJE63fZaZ3BoXeCYlKciCNYxTC1gKvAT8qaqzgSOq+rOq/hykc2ZzvvPnvYjwQo+nGDjoawAaNbqbSy+tx6WX1uN/A77g7bcHhDTRAyxesoJq1apSpUplYmJiuPfeVkz8YVpIY4iEuAZ+/h5r1m7kw34n/s77YeI0HnygDQAPPtCGiROnhjSmO+9pljWEA56hpHo31AHgxpvr8scfnkndZs2cx+VXXkLx4sWIjo6m3g21Wb92U8jidNtn6fa4oOAP4wSlZ6+qGcAHIvKt839ysM4FMGzYAG66sS7lypVl08ZFvPb6e5x7zjk89pindPT3309h6NBRwTq939LT03m6a08mTxpOdFQUQ4aOYvXq8Fwsdmtc9evV5oH29/D7ytUsWexJBi+/3Je33vmIkcM/5eGH7mP79h20vS90t84WL16MmxrWp1vXXllt/3nqZV5/6yWKREdz7Ngxuj39CgB/HviLTwcM4cdZ34IqM6bPYca0kPRzAHd9lpEQF0Tu8IyvJBTjZSLSDKivqi/6sZPaTFX+ce0sQi6Oy2aq8o9rP8vjO864v/3mhe19ToY9tn4dcf37kDxBq6qTgEl5bmiMMWGSEbE3VfrGyiUYYwwF/wKtJXtjjKHgj9lbsjfGGCL3LhtfWbI3xhhszN4YYwqFgp3qLdkbYwxQ8MfsC0TVS2OMOVPpqM+LL0QkWkSWi8gPztdlRWS6iGxw/i/jtW0PEdkoIutEpIlX+3UistJZ119yKi7kI0v2xhhDUAqhPQ2s8fr6BWCmqlYHZjpfIyKXA+2AK4DbgY9FJNrZ5xOgE1DdWfJd09uSvTHG4LlA6+uSFxGJB5rhKQaZqRWQWYhrKNDaq32kqh5T1c3ARqCOiMQCJVV1vnpKHQzz2sdvluyNMQb/Ji/xLsfuLJ1OOtyHQHey/yFQQVWTAJz/yzvtccB2r+0SnbY45/XJ7fni6gu0x44l5r1RGKSl7sx7ozBxa2xujSv5r3XhDuG03PqeuTWuM+XPBdps5dhPIiLNgRRVXSoiDXw4XE7j8JpLe764OtnHl7ki3CGcInF/gisLQYGLi1RZXH5za2yujSsAx/D1wqsP6gMtRaQpUAwoKSJfA8kiEquqSc4QTYqzfSJQ2Wv/eGCn0x6fQ3u+2DCOMcYQuDF7Ve2hqvGqWgXPhdefVLU9MAHo4GzWARjvvJ4AtBORoiJSFc+F2EXOUM/fIlLXuQvnQa99/Obqnr0xxoRKCB6q6guMFpGOwDagDYCqJojIaDyz+6UBXVQ1sy5bZ2AIUByY4iz5YsneGGMITrkEZ5a+2c7rvUCj02zXB+iTQ/sS4MpAxGLJ3hhjKPhP0FqyN8YYQAt4dRxL9sYYQ0DvxnElS/bGGIMN4xhjTKGQodazN8aYAq9gp/oIfqjq3f+9xop1PzPjl3FZbaVLl2T42IHMXTyJ4WMHUqpUSQBubHA9k38axYx5Y5n80yjq3Vgna58aV1/OjHljmbdkMq++2SNk8Tdp3ICEVXNYu3oe3Z/rErLz5sWtcQFERUWxeNFUxo8bmvfGIeTW98zi8k8gC6G5UcQm+2+Hf0/7No9la+vS9RF++XkBN9Zuxi8/L6BL144A7Nu7n4f/9QS33nAXz3R5if6fvJm1z5vvvkz3Z3pzQ62mVL34AhreekPQY4+KiqJ/vz40b9GeGlc3pG3b1lx2WfWgnzdS48r01JOPsHbthnCHkY1b3zOLy3/qx79IFLHJfuH8pRzY/2e2tsZ3NOTbkZ6nib8dOZ4mTW8BIGHlWpJ37QZg3ZqNFC1WlLPOiqF8hXKcW+Icli3+DYAxIydk7RNMdWrXZNOmLWzevI3U1FRGjx5PyxZN8t6xkMYFEBcXS9M7GjF48Ihwh5KNW98zi8t/aajPSyQKSbIXkRtE5FkRaRzM85Qrfx4pyXsASEnew3nnlz1lm2Ytb2PV72s4fjyVirEVSNqZnLUuaWcyFWMrBDNEACrFVWR74ol6Rok7kqhUqWLQz5sXt8YF8P57vXmhx+tkZLjrngm3vmcWl/+sZ58PIrLI6/W/gQFACaCXiLyQy35ZNaJzrB16hi75x8X06PUsLzz7aub5TtlGQ3BFPlznzYtb42rW9FZSUvawbPnKcIdyCre+ZxaX/4IwU5WrBKtnH+P1uhNwm6r2BhoD959uJ1X9XFVrqWqtk2cC8MWelL2Ur1AOgPIVyrF3976sdbGVKjBoWD+6Pv4iW7d45glI2rmL2EoVsm2TvCuFYNuRmETl+BNlYuPjYklKSs5lj9Bwa1z16tWiRfPGbFy/gG++/piGDeszdEj/cIcFuPc9s7j8p6o+L5EoWMk+SkTKiMh5gKjqbgBVPURgSk/naPqPs2nTrhUAbdq1YtqUWQCULFmCoSM/pu9rH7Jk4fKs7VOS93Dw4GGurXUVAPe0a8m0ybOCFV6WxUtWUK1aVapUqUxMTAz33tuKiT9MC/p5IzWul3r2pcpFtah2SV3ub/84s2b9QoeHngp3WIB73zOLy38F/W6cYN1nXwpYimemFRWRiqq6S0TOJefZV/w2YODbXF+/NmXPK83iVTN4r+/HDPhwEJ8Ofo927e9iR2ISjz38LAAP/fs+qlStzNPdHuPpbp47eP51dyf27tnHi91e4/2PXqdYsWLMnjGXn2bMDUR4uUpPT+fprj2ZPGk40VFRDBk6itWr1wf9vJEal5u59T2zuPxX0MslSCj/JBGRs/HMw7jZh43VZqryj2tnEbK4/ObW2Fwb1/EdZ9yJbHpBU5+T4eRtkwPSaQ2lkD5Bq6qHgbwTvTHGhFikjsX7ysolGGMMkXuXja8s2RtjDFbP3hhjCoVIvcvGV5bsjTEGSNeCPZBjyd4YY7BhHGOMKRQK+uQlEVv10hhjAkn9WHIjIpVFZJaIrBGRBBF52mkvKyLTRWSD838Zr316iMhGEVknIk282q8TkZXOuv6SU3EhH1myN8YYAlouIQ34j6peBtQFuojI5cALwExVrQ7MdL7GWdcOuAK4HfhYRKKdY32Cp75YdWe5Pb/fnyV7Y4whcMleVZNUdZnz+m9gDRAHtAIyp1kbCrR2XrcCRqrqMae6wEagjojEAiVVdb56nvga5rWP32zM3hhj8O9uHBHphKfHnelzVT2lMruIVAFqAgvxlIpJAs8vBBEp72wWByzw2i3RaUt1Xp/cni+uTvaJ+xPCHUKO0lJ35r1RmLg1NovLf26Nza1xnSl/7sZxEnuu0244hR+/A7qq6l+5DLfntEJzac8XVyf7qytcH+4QTvFb8nxXFoICFxepsrj85tbYXBtXAI4RyNo4IhKDJ9F/o6pjneZkEYl1evWxQObkGYlAZa/d44GdTnt8Du35YmP2xhhD4MbsnTtmvgDWqOr7XqsmAB2c1x2A8V7t7USkqIhUxXMhdpEz5PO3iNR1jvmg1z5+c3XP3hhjQiWAPfv6wAPAShFZ4bS9CPQFRotIR2Ab0MY5b4KIjAZW4/kjpYuqpjv7dQaGAMWBKc6SL5bsjTEGSA9Q3UtVncfpJ2lqdJp9+gB9cmhfAlwZiLgs2RtjDAX/CVpL9sYYg9XGMcaYQsF69sYYUwhYz94YYwoB69kbY0whUNAnL4nYh6p6f/Ais1ZN4rvZX2e1vf3Zq4yaMYRRM4YwefF3jJoxBIC6N9VmxNTBjJn1FSOmDqZO/etOOV6/oW9lO1awNWncgIRVc1i7eh7dn+sSsvPmxa1xAURFRbF40VTGjxua98Yh5Nb3zOLyj/rxLxJFbM9+/KjJjBg8hj7/eyWrrfujJ17/579PcvCvgwAc2PcnTz3Ynd3Je6j2j4v4ZMQH3FazVda2jZrezOFDR0IWe1RUFP379eH2pveRmJjEgvmTmfjDNNas2RCyGCIprkxPPfkIa9duoGSJEuEOJYtb3zOLy39qPXt3WrZgBX8d+Ou06xu3uIUp46YDsHbVenYn7wFg49o/OKvoWcScFQNA8bOL88Cj7Rj44ZCgx5ypTu2abNq0hc2bt5Gamsro0eNp2aJJ3jsW0rgA4uJiaXpHIwYPHhHuULJx63tmcfkvgPXsXSkoyV5E/ikiJZ3XxUWkt4hMFJG3RKRUMM7p7dq617B3zz62bU48Zd2tzRuydtV6Uo+nAtDl+X8z7NMRHD1yNNhhZakUV5HtiSfqGSXuSKJSpYohO//puDUugPff680LPV4nI8NdvS+3vmcWl/9U1eclEgWrZz8YOOy87geUAt5y2r483U4i0klElojIklxrh+bhjjtv5cdxM05pv/jSqnTt+TivPfc2AJdeUZ0Lqsbz05Q5Z3A2/+VU6tQNP0BujatZ01tJSdnDsuUrwx3KKdz6nllc/ivoPftgjdlHqWpm1dFaqnqt83qeV2GgU2SrES2iH+XjxNHR0TRq2oB2jR/O1l4+9nw+GPwmPZ98lcStOwC4qtaVXHbVpUxe/B1FoqMpW64Mg8YO4JG7nsjHmX23IzGJyvEnysTGx8WSlJQc1HP6wq1x1atXixbNG3PH7bdQrFhRSpYswdAh/enw0FPhDs2175nF5b90l/3VGGjB6tmvEpHMbPubiNQCEJFL8My+EjT/vKkWmzduJSVpd1ZbiZLnMuDrd+n3xqesWHyid/jt0HHcdk0rmta+m4daPcbWP7YHPdEDLF6ygmrVqlKlSmViYmK4995WTPxhWtDPG6lxvdSzL1UuqkW1S+pyf/vHmTXrF1ckenDve2Zx+c/uxsmfR4B+ItIT2APMF5HtwHZn3Rnr+0lvatWrSemypZm27Hs+eWcQ40b8wO2tb+VH58Jspnb/dw8XVI2n0zMP0emZhwDo3O4Z9u3ZH4hQ/Jaens7TXXsyedJwoqOiGDJ0FKtXrw9LLJEQl5u59T2zuPznluGkYJFgfoMiUgK4CM8vlURV9f3vNRG1mar849pZhCwuv7k1NtfGdXzHaef889X5pS71ORnu/nPdGZ8v1IJ6n70zs/pvwTyHMcYEQkHv2UfsQ1XGGBNIBf0CrSV7Y4yBiL2l0leW7I0xBhvGMcaYQsFKHBtjTCEQqffP+8qSvTHGYD17Y4wpFDKsxLExxhR8gax6KSK3i8g6EdkoIi+EIPw8Wc/eGGMI3N04IhINfATcBiQCi0VkgqquDsgJ8sl69sYYA6gfSx7qABtV9Q9VPQ6MBFrlsU/QubdnryqBqrMgIp2c8skBkZb3Jj5za2wWl38CHRe4Nza3xnWm/KmvIyKdgE5eTZ97fS9xeIo+ZkoE/nnmEZ6ZwtKz75T3JmHj1tgsLv+4NS5wb2xujStPqvq5qtbyWrx/aeX0SyPst/oUlmRvjDGhkghU9vo6Hth5mm1DxpK9McYE1mKguohUFZGzgHbAhDDH5OIx+8ByzbhgDtwam8XlH7fGBe6Nza1xnRFVTRORJ4CpQDQwWFUTwhxWcCcvMcYY4w42jGOMMYWAJXtjjCkECnyyd+NjywAiMlhEUkRkVbhjySQilUVkloisEZEEEXk63DFlEpFiIrJIRH5zYusd7pi8iUi0iCwXkR/CHUsmEdkiIitFZIWILAl3PJlEpLSIjBGRtc7Pmvsmmy6ACvSYvfPY8nq8HlsG7gv3Y8sAInITcBAYpqpXhjseABGJBWJVdZkzWfxSoLVL3i8BzlHVgyISA8wDnlbVBWEODQAReRaoBZRU1ebhjgc8yR6opap7wh2LNxEZCsxV1UHO3Spnq+qBMIdV4BX0nr0rH1sGUNU5wL5wx+FNVZNUdZnz+m9gDZ6nAcNOPQ46X8Y4iyt6KiISDzQDBoU7FrcTkZLATcAXAKp63BJ9aBT0ZJ/TY8uuSF5uJyJVgJrAwjCHksUZKlkBpADTVdUtsX0IdAfcViNXgWkistR5vN8NLgJ2A186w16DROSccAdVGBT0ZO/Kx5bdTkTOBb4DuqrqX+GOJ5OqpqvqNXieSKwjImEf/hKR5kCKqi4Ndyw5qK+q1wJ3AF2cocNwKwJcC3yiqjWBQ4BrrqUVZAU92bvysWU3c8bDvwO+UdWx4Y4nJ86f/bOB28MbCQD1gZbO+PhI4BYR+Tq8IXmo6k7n/xRgHJ5hzXBLBBK9/iobgyf5myAr6MnelY8tu5VzEfQLYI2qvh/ueLyJyPkiUtp5XRy4FVgb1qAAVe2hqvGqWgXPz9dPqto+zGEhIuc4F9lxhkkaA2G/80tVdwHbReRSp6kREPYbAAqDAl0uwa2PLQOIyAigAVBORBKBXqr6RXijoj7wALDSGRsHeFFVJ4cvpCyxwFDnDqsoYLSquuY2RxeqAIzz/P6mCDBcVX8Mb0hZngS+cTpgfwAPhzmeQqFA33ppjDHGo6AP4xhjjMGSvTHGFAqW7I0xphCwZG+MMYWAJXtjjCkELNmbAktEHhKRAeGOwxg3sGRvIo5zr70xxg+W7E1YiMhr3vXyRaSPiDwlInNEZJyIrBaRT0Ukyll/UEReFZGFwPUi0t6pb79CRD7L/AUgIg+LyHoR+RnPQ2LGGCzZm/D5AugA4CT0dsAOPPVb/gPUAC4G7nK2PwdYpar/BPYCbfEU+roGSAfud+rx98aT5G8DLg/VN2OM2xXocgnGvVR1i4jsFZGaeB7tX44niS9S1T8gq6TEDXiKZaXjKdAGnnoq1wGLnXIAxfGUPf4nMFtVdzv7jwIuCdk3ZYyLWbI34TQIeAioCAx22k6u35H59VFVTXdeCzBUVXt4bygirXPY3xiDDeOY8BqHp0xxbTzF6sBTp76qM7TTFs/0gyebCdwjIuUBRKSsiFyIZ6KVBiJynlOquU3QvwNjIoT17E3YqOpxEZkFHFDVdGdIZj7QF8+Y/Rw8vxBO3m+1iPTEMwtTFJAKdFHVBSLyX+cYScAyPNVOjSn0rOqlCRsnUS8D2qjqBhFpAHRzy4TdxhQkNoxjwkJELgc2AjNVdUO44zGmoLOevTHGFALWszfGmELAkr0xxhQCluyNMaYQsGRvjDGFgCV7Y4wpBP4fs/R1dC0xrb8AAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "cm= confusion_matrix(ytrain,ypred)\n",
    "sns.heatmap(cm,annot=True , linewidths=0.5,linecolor=\"red\",fmt=\".0f\")\n",
    "plt.xlabel(\"ypred\")\n",
    "plt.ylabel(\"ytrue\")\n",
    "plt.title(\"Confusion matrix\")\n",
    "plt.show()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.9.16"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}