IoT_ML_with_Bot_Dataset.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "v_JYAzabKZzM"
   },
   "source": [
    "### Downloading BoT_IoT Datasets"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "background_save": true
    },
    "id": "U7z7t_Hwx9bc"
   },
   "outputs": [],
   "source": [
    "# Download Dataset\n",
    "import requests\n",
    "\n",
    "def download_dataset(url, file_name):\n",
    "  '''function to download dataset from url\n",
    "  args\n",
    "  url: str, download link\n",
    "  file_name: str, name of the file\n",
    "  '''\n",
    "  r = requests.get(url, allow_redirects=True)\n",
    "  open(file_name, 'wb').write(r.content)\n",
    "\n",
    "\n",
    "download_dataset('http://205.174.165.80/CICDataset/DoHBrw-2020/Dataset/Total-CSVs.zip', 'Total_csv.zip')\n",
    "# download_dataset('https://cloudstor.aarnet.edu.au/plus/s/umT99TnxvbpkkoE/download?path=%2FCSV%2FTraning%20and%20Testing%20Tets%20(5%25%20of%20the%20entier%20dataset)%2FAll%20features&files=UNSW_2018_IoT_Botnet_Full5pc_1.csv&downloadStartSecret=8xn48h3djjx', 'UNSW_2018_IoT_Botnet_Full5pc_1.csv')\n",
    "# download_dataset('https://cloudstor.aarnet.edu.au/plus/s/umT99TnxvbpkkoE/download?path=%2FCSV%2FTraning%20and%20Testing%20Tets%20(5%25%20of%20the%20entier%20dataset)%2FAll%20features&files=UNSW_2018_IoT_Botnet_Full5pc_2.csv&downloadStartSecret=sm8zsfza9k', 'UNSW_2018_IoT_Botnet_Full5pc_2.csv')\n",
    "# download_dataset('https://cloudstor.aarnet.edu.au/plus/s/umT99TnxvbpkkoE/download?path=%2FCSV%2FTraning%20and%20Testing%20Tets%20(5%25%20of%20the%20entier%20dataset)%2FAll%20features&files=UNSW_2018_IoT_Botnet_Full5pc_3.csv&downloadStartSecret=kwepzb3u8qs', 'UNSW_2018_IoT_Botnet_Full5pc_3.csv')\n",
    "# download_dataset('https://cloudstor.aarnet.edu.au/plus/s/umT99TnxvbpkkoE/download?path=%2FCSV%2FTraning%20and%20Testing%20Tets%20(5%25%20of%20the%20entier%20dataset)%2FAll%20features&files=UNSW_2018_IoT_Botnet_Full5pc_4.csv&downloadStartSecret=rk1z8jsq7cj', 'UNSW_2018_IoT_Botnet_Full5pc_4.csv')\n",
    "# download_dataset('https://cloudstor.aarnet.edu.au/plus/s/umT99TnxvbpkkoE/download?path=%2FCSV%2FTraning%20and%20Testing%20Tets%20(5%25%20of%20the%20entier%20dataset)%2F10-best%20features&files=UNSW_2018_IoT_Botnet_Final_10_Best.csv&downloadStartSecret=edamjd8sjy9', 'UNSW_2018_IoT_Botnet_Full5pc_best10.csv')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "z4OGVirLKi7m"
   },
   "source": [
    "### Load Datasets"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "id": "ab9-E8kpOy4-"
   },
   "outputs": [],
   "source": [
    "# best_features_dataset = pd.read_csv('/content/UNSW_2018_IoT_Botnet_Full5pc_best10.csv', delimiter=';')\n",
    "# best_features_dataset\n",
    "# del best_features_dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 145,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 482
    },
    "id": "Eiej5_b6sDb6",
    "outputId": "da69d991-bb7b-47a8-ba5c-685eabd1d805"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/tmp/ipykernel_6291/208987755.py:7: DtypeWarning: Columns (7,9) have mixed types. Specify dtype option on import or set low_memory=False.\n",
      "  dataset2 = pd.read_csv(\"UNSW_2018_IoT_Botnet_Full5pc_2.csv\")\n",
      "/tmp/ipykernel_6291/208987755.py:9: DtypeWarning: Columns (7,9) have mixed types. Specify dtype option on import or set low_memory=False.\n",
      "  dataset4 = pd.read_csv(\"UNSW_2018_IoT_Botnet_Full5pc_4.csv\")\n"
     ]
    }
   ],
   "source": [
    "# import libraries\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "# dataset1 = pd.read_csv(\"UNSW_2018_IoT_Botnet_Full5pc_1.csv\")\n",
    "dataset2 = pd.read_csv(\"UNSW_2018_IoT_Botnet_Full5pc_2.csv\")\n",
    "# dataset3 = pd.read_csv(\"UNSW_2018_IoT_Botnet_Full5pc_3.csv\")\n",
    "dataset4 = pd.read_csv(\"UNSW_2018_IoT_Botnet_Full5pc_4.csv\")\n",
    "dataset2 = dataset2.iloc[:11000, :]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "Xu0vSAxeVtao"
   },
   "source": [
    "### Feature Extraction"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "S3NjUW-NNk3B"
   },
   "source": [
    "### Merge all 4 datasets into one Pandas DataFrame"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 146,
   "metadata": {},
   "outputs": [],
   "source": [
    "normal = dataset4[dataset4.category == 'Normal']\n",
    "theft = dataset4[dataset4.category == 'Theft']\n",
    "Reconnaissance = dataset4[dataset4.category == 'Reconnaissance'][:11000]\n",
    "DDoS = dataset4[dataset4.category == 'DDoS'][:11000]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 147,
   "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>pkSeqID</th>\n",
       "      <th>stime</th>\n",
       "      <th>flgs</th>\n",
       "      <th>flgs_number</th>\n",
       "      <th>proto</th>\n",
       "      <th>proto_number</th>\n",
       "      <th>saddr</th>\n",
       "      <th>sport</th>\n",
       "      <th>daddr</th>\n",
       "      <th>dport</th>\n",
       "      <th>...</th>\n",
       "      <th>AR_P_Proto_P_DstIP</th>\n",
       "      <th>N_IN_Conn_P_DstIP</th>\n",
       "      <th>N_IN_Conn_P_SrcIP</th>\n",
       "      <th>AR_P_Proto_P_Sport</th>\n",
       "      <th>AR_P_Proto_P_Dport</th>\n",
       "      <th>Pkts_P_State_P_Protocol_P_DestIP</th>\n",
       "      <th>Pkts_P_State_P_Protocol_P_SrcIP</th>\n",
       "      <th>attack</th>\n",
       "      <th>category</th>\n",
       "      <th>subcategory</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>3000001</td>\n",
       "      <td>1.528099e+09</td>\n",
       "      <td>e</td>\n",
       "      <td>1</td>\n",
       "      <td>udp</td>\n",
       "      <td>3</td>\n",
       "      <td>192.168.100.147</td>\n",
       "      <td>6226</td>\n",
       "      <td>192.168.100.3</td>\n",
       "      <td>80</td>\n",
       "      <td>...</td>\n",
       "      <td>1.09825</td>\n",
       "      <td>100</td>\n",
       "      <td>100</td>\n",
       "      <td>1.09827</td>\n",
       "      <td>1.09825</td>\n",
       "      <td>1500</td>\n",
       "      <td>1500</td>\n",
       "      <td>1</td>\n",
       "      <td>DDoS</td>\n",
       "      <td>UDP</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>3000002</td>\n",
       "      <td>1.528099e+09</td>\n",
       "      <td>e</td>\n",
       "      <td>1</td>\n",
       "      <td>udp</td>\n",
       "      <td>3</td>\n",
       "      <td>192.168.100.147</td>\n",
       "      <td>6227</td>\n",
       "      <td>192.168.100.3</td>\n",
       "      <td>80</td>\n",
       "      <td>...</td>\n",
       "      <td>1.09825</td>\n",
       "      <td>100</td>\n",
       "      <td>100</td>\n",
       "      <td>1.09827</td>\n",
       "      <td>1.09825</td>\n",
       "      <td>1500</td>\n",
       "      <td>1500</td>\n",
       "      <td>1</td>\n",
       "      <td>DDoS</td>\n",
       "      <td>UDP</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3000003</td>\n",
       "      <td>1.528099e+09</td>\n",
       "      <td>e</td>\n",
       "      <td>1</td>\n",
       "      <td>udp</td>\n",
       "      <td>3</td>\n",
       "      <td>192.168.100.147</td>\n",
       "      <td>6228</td>\n",
       "      <td>192.168.100.3</td>\n",
       "      <td>80</td>\n",
       "      <td>...</td>\n",
       "      <td>1.09825</td>\n",
       "      <td>100</td>\n",
       "      <td>100</td>\n",
       "      <td>1.09827</td>\n",
       "      <td>1.09825</td>\n",
       "      <td>1500</td>\n",
       "      <td>1500</td>\n",
       "      <td>1</td>\n",
       "      <td>DDoS</td>\n",
       "      <td>UDP</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3000004</td>\n",
       "      <td>1.528099e+09</td>\n",
       "      <td>e</td>\n",
       "      <td>1</td>\n",
       "      <td>udp</td>\n",
       "      <td>3</td>\n",
       "      <td>192.168.100.147</td>\n",
       "      <td>6229</td>\n",
       "      <td>192.168.100.3</td>\n",
       "      <td>80</td>\n",
       "      <td>...</td>\n",
       "      <td>1.09825</td>\n",
       "      <td>100</td>\n",
       "      <td>100</td>\n",
       "      <td>1.09827</td>\n",
       "      <td>1.09825</td>\n",
       "      <td>1500</td>\n",
       "      <td>1500</td>\n",
       "      <td>1</td>\n",
       "      <td>DDoS</td>\n",
       "      <td>UDP</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>3000005</td>\n",
       "      <td>1.528099e+09</td>\n",
       "      <td>e</td>\n",
       "      <td>1</td>\n",
       "      <td>udp</td>\n",
       "      <td>3</td>\n",
       "      <td>192.168.100.147</td>\n",
       "      <td>6230</td>\n",
       "      <td>192.168.100.3</td>\n",
       "      <td>80</td>\n",
       "      <td>...</td>\n",
       "      <td>1.09825</td>\n",
       "      <td>100</td>\n",
       "      <td>100</td>\n",
       "      <td>1.09827</td>\n",
       "      <td>1.09825</td>\n",
       "      <td>1500</td>\n",
       "      <td>1500</td>\n",
       "      <td>1</td>\n",
       "      <td>DDoS</td>\n",
       "      <td>UDP</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10995</th>\n",
       "      <td>3010996</td>\n",
       "      <td>1.528099e+09</td>\n",
       "      <td>e</td>\n",
       "      <td>1</td>\n",
       "      <td>udp</td>\n",
       "      <td>3</td>\n",
       "      <td>192.168.100.147</td>\n",
       "      <td>12055</td>\n",
       "      <td>192.168.100.3</td>\n",
       "      <td>80</td>\n",
       "      <td>...</td>\n",
       "      <td>1.09822</td>\n",
       "      <td>100</td>\n",
       "      <td>100</td>\n",
       "      <td>1.09820</td>\n",
       "      <td>1.09822</td>\n",
       "      <td>1500</td>\n",
       "      <td>1500</td>\n",
       "      <td>1</td>\n",
       "      <td>DDoS</td>\n",
       "      <td>UDP</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10996</th>\n",
       "      <td>3010997</td>\n",
       "      <td>1.528099e+09</td>\n",
       "      <td>e</td>\n",
       "      <td>1</td>\n",
       "      <td>udp</td>\n",
       "      <td>3</td>\n",
       "      <td>192.168.100.147</td>\n",
       "      <td>12056</td>\n",
       "      <td>192.168.100.3</td>\n",
       "      <td>80</td>\n",
       "      <td>...</td>\n",
       "      <td>1.09822</td>\n",
       "      <td>100</td>\n",
       "      <td>100</td>\n",
       "      <td>1.09820</td>\n",
       "      <td>1.09822</td>\n",
       "      <td>1500</td>\n",
       "      <td>1500</td>\n",
       "      <td>1</td>\n",
       "      <td>DDoS</td>\n",
       "      <td>UDP</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10997</th>\n",
       "      <td>3010998</td>\n",
       "      <td>1.528099e+09</td>\n",
       "      <td>e</td>\n",
       "      <td>1</td>\n",
       "      <td>udp</td>\n",
       "      <td>3</td>\n",
       "      <td>192.168.100.147</td>\n",
       "      <td>12057</td>\n",
       "      <td>192.168.100.3</td>\n",
       "      <td>80</td>\n",
       "      <td>...</td>\n",
       "      <td>1.09822</td>\n",
       "      <td>100</td>\n",
       "      <td>100</td>\n",
       "      <td>1.09820</td>\n",
       "      <td>1.09822</td>\n",
       "      <td>1500</td>\n",
       "      <td>1500</td>\n",
       "      <td>1</td>\n",
       "      <td>DDoS</td>\n",
       "      <td>UDP</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10998</th>\n",
       "      <td>3010999</td>\n",
       "      <td>1.528099e+09</td>\n",
       "      <td>e</td>\n",
       "      <td>1</td>\n",
       "      <td>udp</td>\n",
       "      <td>3</td>\n",
       "      <td>192.168.100.147</td>\n",
       "      <td>12058</td>\n",
       "      <td>192.168.100.3</td>\n",
       "      <td>80</td>\n",
       "      <td>...</td>\n",
       "      <td>1.09822</td>\n",
       "      <td>100</td>\n",
       "      <td>100</td>\n",
       "      <td>1.09820</td>\n",
       "      <td>1.09822</td>\n",
       "      <td>1500</td>\n",
       "      <td>1500</td>\n",
       "      <td>1</td>\n",
       "      <td>DDoS</td>\n",
       "      <td>UDP</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10999</th>\n",
       "      <td>3011000</td>\n",
       "      <td>1.528099e+09</td>\n",
       "      <td>e</td>\n",
       "      <td>1</td>\n",
       "      <td>udp</td>\n",
       "      <td>3</td>\n",
       "      <td>192.168.100.147</td>\n",
       "      <td>12059</td>\n",
       "      <td>192.168.100.3</td>\n",
       "      <td>80</td>\n",
       "      <td>...</td>\n",
       "      <td>1.09822</td>\n",
       "      <td>100</td>\n",
       "      <td>100</td>\n",
       "      <td>1.09819</td>\n",
       "      <td>1.09822</td>\n",
       "      <td>1500</td>\n",
       "      <td>1500</td>\n",
       "      <td>1</td>\n",
       "      <td>DDoS</td>\n",
       "      <td>UDP</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>11000 rows × 46 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       pkSeqID         stime flgs  flgs_number proto  proto_number  \\\n",
       "0      3000001  1.528099e+09    e            1   udp             3   \n",
       "1      3000002  1.528099e+09    e            1   udp             3   \n",
       "2      3000003  1.528099e+09    e            1   udp             3   \n",
       "3      3000004  1.528099e+09    e            1   udp             3   \n",
       "4      3000005  1.528099e+09    e            1   udp             3   \n",
       "...        ...           ...  ...          ...   ...           ...   \n",
       "10995  3010996  1.528099e+09    e            1   udp             3   \n",
       "10996  3010997  1.528099e+09    e            1   udp             3   \n",
       "10997  3010998  1.528099e+09    e            1   udp             3   \n",
       "10998  3010999  1.528099e+09    e            1   udp             3   \n",
       "10999  3011000  1.528099e+09    e            1   udp             3   \n",
       "\n",
       "                 saddr  sport          daddr dport  ...  AR_P_Proto_P_DstIP  \\\n",
       "0      192.168.100.147   6226  192.168.100.3    80  ...             1.09825   \n",
       "1      192.168.100.147   6227  192.168.100.3    80  ...             1.09825   \n",
       "2      192.168.100.147   6228  192.168.100.3    80  ...             1.09825   \n",
       "3      192.168.100.147   6229  192.168.100.3    80  ...             1.09825   \n",
       "4      192.168.100.147   6230  192.168.100.3    80  ...             1.09825   \n",
       "...                ...    ...            ...   ...  ...                 ...   \n",
       "10995  192.168.100.147  12055  192.168.100.3    80  ...             1.09822   \n",
       "10996  192.168.100.147  12056  192.168.100.3    80  ...             1.09822   \n",
       "10997  192.168.100.147  12057  192.168.100.3    80  ...             1.09822   \n",
       "10998  192.168.100.147  12058  192.168.100.3    80  ...             1.09822   \n",
       "10999  192.168.100.147  12059  192.168.100.3    80  ...             1.09822   \n",
       "\n",
       "       N_IN_Conn_P_DstIP N_IN_Conn_P_SrcIP  AR_P_Proto_P_Sport  \\\n",
       "0                    100               100             1.09827   \n",
       "1                    100               100             1.09827   \n",
       "2                    100               100             1.09827   \n",
       "3                    100               100             1.09827   \n",
       "4                    100               100             1.09827   \n",
       "...                  ...               ...                 ...   \n",
       "10995                100               100             1.09820   \n",
       "10996                100               100             1.09820   \n",
       "10997                100               100             1.09820   \n",
       "10998                100               100             1.09820   \n",
       "10999                100               100             1.09819   \n",
       "\n",
       "       AR_P_Proto_P_Dport  Pkts_P_State_P_Protocol_P_DestIP  \\\n",
       "0                 1.09825                              1500   \n",
       "1                 1.09825                              1500   \n",
       "2                 1.09825                              1500   \n",
       "3                 1.09825                              1500   \n",
       "4                 1.09825                              1500   \n",
       "...                   ...                               ...   \n",
       "10995             1.09822                              1500   \n",
       "10996             1.09822                              1500   \n",
       "10997             1.09822                              1500   \n",
       "10998             1.09822                              1500   \n",
       "10999             1.09822                              1500   \n",
       "\n",
       "       Pkts_P_State_P_Protocol_P_SrcIP  attack  category  subcategory  \n",
       "0                                 1500       1      DDoS          UDP  \n",
       "1                                 1500       1      DDoS          UDP  \n",
       "2                                 1500       1      DDoS          UDP  \n",
       "3                                 1500       1      DDoS          UDP  \n",
       "4                                 1500       1      DDoS          UDP  \n",
       "...                                ...     ...       ...          ...  \n",
       "10995                             1500       1      DDoS          UDP  \n",
       "10996                             1500       1      DDoS          UDP  \n",
       "10997                             1500       1      DDoS          UDP  \n",
       "10998                             1500       1      DDoS          UDP  \n",
       "10999                             1500       1      DDoS          UDP  \n",
       "\n",
       "[11000 rows x 46 columns]"
      ]
     },
     "execution_count": 147,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "DDoS"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 148,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 444
    },
    "id": "Cfw3iJOFNsoY",
    "outputId": "ceb4c658-a231-4771-fa0c-1fbfe1ecae85"
   },
   "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>pkSeqID</th>\n",
       "      <th>stime</th>\n",
       "      <th>flgs</th>\n",
       "      <th>flgs_number</th>\n",
       "      <th>proto</th>\n",
       "      <th>proto_number</th>\n",
       "      <th>saddr</th>\n",
       "      <th>sport</th>\n",
       "      <th>daddr</th>\n",
       "      <th>dport</th>\n",
       "      <th>...</th>\n",
       "      <th>AR_P_Proto_P_DstIP</th>\n",
       "      <th>N_IN_Conn_P_DstIP</th>\n",
       "      <th>N_IN_Conn_P_SrcIP</th>\n",
       "      <th>AR_P_Proto_P_Sport</th>\n",
       "      <th>AR_P_Proto_P_Dport</th>\n",
       "      <th>Pkts_P_State_P_Protocol_P_DestIP</th>\n",
       "      <th>Pkts_P_State_P_Protocol_P_SrcIP</th>\n",
       "      <th>attack</th>\n",
       "      <th>category</th>\n",
       "      <th>subcategory</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1000001</td>\n",
       "      <td>1.528085e+09</td>\n",
       "      <td>e</td>\n",
       "      <td>1</td>\n",
       "      <td>udp</td>\n",
       "      <td>3</td>\n",
       "      <td>192.168.100.148</td>\n",
       "      <td>37153</td>\n",
       "      <td>192.168.100.6</td>\n",
       "      <td>80</td>\n",
       "      <td>...</td>\n",
       "      <td>0.319943</td>\n",
       "      <td>100</td>\n",
       "      <td>100</td>\n",
       "      <td>0.319979</td>\n",
       "      <td>0.319943</td>\n",
       "      <td>800</td>\n",
       "      <td>800</td>\n",
       "      <td>1</td>\n",
       "      <td>DoS</td>\n",
       "      <td>UDP</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1000002</td>\n",
       "      <td>1.528085e+09</td>\n",
       "      <td>e</td>\n",
       "      <td>1</td>\n",
       "      <td>udp</td>\n",
       "      <td>3</td>\n",
       "      <td>192.168.100.148</td>\n",
       "      <td>37154</td>\n",
       "      <td>192.168.100.6</td>\n",
       "      <td>80</td>\n",
       "      <td>...</td>\n",
       "      <td>0.319943</td>\n",
       "      <td>100</td>\n",
       "      <td>100</td>\n",
       "      <td>0.319979</td>\n",
       "      <td>0.319943</td>\n",
       "      <td>800</td>\n",
       "      <td>800</td>\n",
       "      <td>1</td>\n",
       "      <td>DoS</td>\n",
       "      <td>UDP</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1000003</td>\n",
       "      <td>1.528085e+09</td>\n",
       "      <td>e</td>\n",
       "      <td>1</td>\n",
       "      <td>udp</td>\n",
       "      <td>3</td>\n",
       "      <td>192.168.100.148</td>\n",
       "      <td>37155</td>\n",
       "      <td>192.168.100.6</td>\n",
       "      <td>80</td>\n",
       "      <td>...</td>\n",
       "      <td>0.319943</td>\n",
       "      <td>100</td>\n",
       "      <td>100</td>\n",
       "      <td>0.319979</td>\n",
       "      <td>0.319943</td>\n",
       "      <td>800</td>\n",
       "      <td>800</td>\n",
       "      <td>1</td>\n",
       "      <td>DoS</td>\n",
       "      <td>UDP</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1000004</td>\n",
       "      <td>1.528085e+09</td>\n",
       "      <td>e</td>\n",
       "      <td>1</td>\n",
       "      <td>udp</td>\n",
       "      <td>3</td>\n",
       "      <td>192.168.100.148</td>\n",
       "      <td>37156</td>\n",
       "      <td>192.168.100.6</td>\n",
       "      <td>80</td>\n",
       "      <td>...</td>\n",
       "      <td>0.319943</td>\n",
       "      <td>100</td>\n",
       "      <td>100</td>\n",
       "      <td>0.319979</td>\n",
       "      <td>0.319943</td>\n",
       "      <td>800</td>\n",
       "      <td>800</td>\n",
       "      <td>1</td>\n",
       "      <td>DoS</td>\n",
       "      <td>UDP</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1000005</td>\n",
       "      <td>1.528085e+09</td>\n",
       "      <td>e</td>\n",
       "      <td>1</td>\n",
       "      <td>udp</td>\n",
       "      <td>3</td>\n",
       "      <td>192.168.100.148</td>\n",
       "      <td>37157</td>\n",
       "      <td>192.168.100.6</td>\n",
       "      <td>80</td>\n",
       "      <td>...</td>\n",
       "      <td>0.319943</td>\n",
       "      <td>100</td>\n",
       "      <td>100</td>\n",
       "      <td>0.319979</td>\n",
       "      <td>0.319943</td>\n",
       "      <td>800</td>\n",
       "      <td>800</td>\n",
       "      <td>1</td>\n",
       "      <td>DoS</td>\n",
       "      <td>UDP</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10995</th>\n",
       "      <td>3010996</td>\n",
       "      <td>1.528099e+09</td>\n",
       "      <td>e</td>\n",
       "      <td>1</td>\n",
       "      <td>udp</td>\n",
       "      <td>3</td>\n",
       "      <td>192.168.100.147</td>\n",
       "      <td>12055</td>\n",
       "      <td>192.168.100.3</td>\n",
       "      <td>80</td>\n",
       "      <td>...</td>\n",
       "      <td>1.098220</td>\n",
       "      <td>100</td>\n",
       "      <td>100</td>\n",
       "      <td>1.098200</td>\n",
       "      <td>1.098220</td>\n",
       "      <td>1500</td>\n",
       "      <td>1500</td>\n",
       "      <td>1</td>\n",
       "      <td>DDoS</td>\n",
       "      <td>UDP</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10996</th>\n",
       "      <td>3010997</td>\n",
       "      <td>1.528099e+09</td>\n",
       "      <td>e</td>\n",
       "      <td>1</td>\n",
       "      <td>udp</td>\n",
       "      <td>3</td>\n",
       "      <td>192.168.100.147</td>\n",
       "      <td>12056</td>\n",
       "      <td>192.168.100.3</td>\n",
       "      <td>80</td>\n",
       "      <td>...</td>\n",
       "      <td>1.098220</td>\n",
       "      <td>100</td>\n",
       "      <td>100</td>\n",
       "      <td>1.098200</td>\n",
       "      <td>1.098220</td>\n",
       "      <td>1500</td>\n",
       "      <td>1500</td>\n",
       "      <td>1</td>\n",
       "      <td>DDoS</td>\n",
       "      <td>UDP</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10997</th>\n",
       "      <td>3010998</td>\n",
       "      <td>1.528099e+09</td>\n",
       "      <td>e</td>\n",
       "      <td>1</td>\n",
       "      <td>udp</td>\n",
       "      <td>3</td>\n",
       "      <td>192.168.100.147</td>\n",
       "      <td>12057</td>\n",
       "      <td>192.168.100.3</td>\n",
       "      <td>80</td>\n",
       "      <td>...</td>\n",
       "      <td>1.098220</td>\n",
       "      <td>100</td>\n",
       "      <td>100</td>\n",
       "      <td>1.098200</td>\n",
       "      <td>1.098220</td>\n",
       "      <td>1500</td>\n",
       "      <td>1500</td>\n",
       "      <td>1</td>\n",
       "      <td>DDoS</td>\n",
       "      <td>UDP</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10998</th>\n",
       "      <td>3010999</td>\n",
       "      <td>1.528099e+09</td>\n",
       "      <td>e</td>\n",
       "      <td>1</td>\n",
       "      <td>udp</td>\n",
       "      <td>3</td>\n",
       "      <td>192.168.100.147</td>\n",
       "      <td>12058</td>\n",
       "      <td>192.168.100.3</td>\n",
       "      <td>80</td>\n",
       "      <td>...</td>\n",
       "      <td>1.098220</td>\n",
       "      <td>100</td>\n",
       "      <td>100</td>\n",
       "      <td>1.098200</td>\n",
       "      <td>1.098220</td>\n",
       "      <td>1500</td>\n",
       "      <td>1500</td>\n",
       "      <td>1</td>\n",
       "      <td>DDoS</td>\n",
       "      <td>UDP</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10999</th>\n",
       "      <td>3011000</td>\n",
       "      <td>1.528099e+09</td>\n",
       "      <td>e</td>\n",
       "      <td>1</td>\n",
       "      <td>udp</td>\n",
       "      <td>3</td>\n",
       "      <td>192.168.100.147</td>\n",
       "      <td>12059</td>\n",
       "      <td>192.168.100.3</td>\n",
       "      <td>80</td>\n",
       "      <td>...</td>\n",
       "      <td>1.098220</td>\n",
       "      <td>100</td>\n",
       "      <td>100</td>\n",
       "      <td>1.098190</td>\n",
       "      <td>1.098220</td>\n",
       "      <td>1500</td>\n",
       "      <td>1500</td>\n",
       "      <td>1</td>\n",
       "      <td>DDoS</td>\n",
       "      <td>UDP</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>33556 rows × 46 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       pkSeqID         stime flgs  flgs_number proto  proto_number  \\\n",
       "0      1000001  1.528085e+09    e            1   udp             3   \n",
       "1      1000002  1.528085e+09    e            1   udp             3   \n",
       "2      1000003  1.528085e+09    e            1   udp             3   \n",
       "3      1000004  1.528085e+09    e            1   udp             3   \n",
       "4      1000005  1.528085e+09    e            1   udp             3   \n",
       "...        ...           ...  ...          ...   ...           ...   \n",
       "10995  3010996  1.528099e+09    e            1   udp             3   \n",
       "10996  3010997  1.528099e+09    e            1   udp             3   \n",
       "10997  3010998  1.528099e+09    e            1   udp             3   \n",
       "10998  3010999  1.528099e+09    e            1   udp             3   \n",
       "10999  3011000  1.528099e+09    e            1   udp             3   \n",
       "\n",
       "                 saddr  sport          daddr dport  ...  AR_P_Proto_P_DstIP  \\\n",
       "0      192.168.100.148  37153  192.168.100.6    80  ...            0.319943   \n",
       "1      192.168.100.148  37154  192.168.100.6    80  ...            0.319943   \n",
       "2      192.168.100.148  37155  192.168.100.6    80  ...            0.319943   \n",
       "3      192.168.100.148  37156  192.168.100.6    80  ...            0.319943   \n",
       "4      192.168.100.148  37157  192.168.100.6    80  ...            0.319943   \n",
       "...                ...    ...            ...   ...  ...                 ...   \n",
       "10995  192.168.100.147  12055  192.168.100.3    80  ...            1.098220   \n",
       "10996  192.168.100.147  12056  192.168.100.3    80  ...            1.098220   \n",
       "10997  192.168.100.147  12057  192.168.100.3    80  ...            1.098220   \n",
       "10998  192.168.100.147  12058  192.168.100.3    80  ...            1.098220   \n",
       "10999  192.168.100.147  12059  192.168.100.3    80  ...            1.098220   \n",
       "\n",
       "       N_IN_Conn_P_DstIP N_IN_Conn_P_SrcIP  AR_P_Proto_P_Sport  \\\n",
       "0                    100               100            0.319979   \n",
       "1                    100               100            0.319979   \n",
       "2                    100               100            0.319979   \n",
       "3                    100               100            0.319979   \n",
       "4                    100               100            0.319979   \n",
       "...                  ...               ...                 ...   \n",
       "10995                100               100            1.098200   \n",
       "10996                100               100            1.098200   \n",
       "10997                100               100            1.098200   \n",
       "10998                100               100            1.098200   \n",
       "10999                100               100            1.098190   \n",
       "\n",
       "       AR_P_Proto_P_Dport  Pkts_P_State_P_Protocol_P_DestIP  \\\n",
       "0                0.319943                               800   \n",
       "1                0.319943                               800   \n",
       "2                0.319943                               800   \n",
       "3                0.319943                               800   \n",
       "4                0.319943                               800   \n",
       "...                   ...                               ...   \n",
       "10995            1.098220                              1500   \n",
       "10996            1.098220                              1500   \n",
       "10997            1.098220                              1500   \n",
       "10998            1.098220                              1500   \n",
       "10999            1.098220                              1500   \n",
       "\n",
       "       Pkts_P_State_P_Protocol_P_SrcIP  attack  category  subcategory  \n",
       "0                                  800       1       DoS          UDP  \n",
       "1                                  800       1       DoS          UDP  \n",
       "2                                  800       1       DoS          UDP  \n",
       "3                                  800       1       DoS          UDP  \n",
       "4                                  800       1       DoS          UDP  \n",
       "...                                ...     ...       ...          ...  \n",
       "10995                             1500       1      DDoS          UDP  \n",
       "10996                             1500       1      DDoS          UDP  \n",
       "10997                             1500       1      DDoS          UDP  \n",
       "10998                             1500       1      DDoS          UDP  \n",
       "10999                             1500       1      DDoS          UDP  \n",
       "\n",
       "[33556 rows x 46 columns]"
      ]
     },
     "execution_count": 148,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "full_dataset = pd.concat([dataset2, normal, theft, Reconnaissance,DDoS], axis=0)\n",
    "# full_dataset = full_dataset.sample(frac = 1)\n",
    "# full_dataset = full_dataset.iloc[:100000, :]\n",
    "full_dataset"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "TSrgHW9MV3Yx"
   },
   "source": [
    "#### For charts, visualize the no. of each unique value in the proto and category columns****"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 149,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "kfgvoDGlOeK2",
    "outputId": "9a627523-4a29-4726-abba-cf355215477a"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Full Dataset Columns\n",
      "Index(['pkSeqID', 'stime', 'flgs', 'flgs_number', 'proto', 'proto_number',\n",
      "       'saddr', 'sport', 'daddr', 'dport', 'pkts', 'bytes', 'state',\n",
      "       'state_number', 'ltime', 'seq', 'dur', 'mean', 'stddev', 'sum', 'min',\n",
      "       'max', 'spkts', 'dpkts', 'sbytes', 'dbytes', 'rate', 'srate', 'drate',\n",
      "       'TnBPSrcIP', 'TnBPDstIP', 'TnP_PSrcIP', 'TnP_PDstIP', 'TnP_PerProto',\n",
      "       'TnP_Per_Dport', 'AR_P_Proto_P_SrcIP', 'AR_P_Proto_P_DstIP',\n",
      "       'N_IN_Conn_P_DstIP', 'N_IN_Conn_P_SrcIP', 'AR_P_Proto_P_Sport',\n",
      "       'AR_P_Proto_P_Dport', 'Pkts_P_State_P_Protocol_P_DestIP',\n",
      "       'Pkts_P_State_P_Protocol_P_SrcIP', 'attack', 'category', 'subcategory'],\n",
      "      dtype='object')\n",
      "\n",
      "Unique values in the 'category' column\n",
      "['DoS' 'Normal' 'Theft' 'Reconnaissance' 'DDoS']\n"
     ]
    }
   ],
   "source": [
    "# list all dataset columns\n",
    "print(\"Full Dataset Columns\", full_dataset.columns, sep='\\n',\n",
    "      end='\\n\\n')\n",
    "print(\"Unique values in the 'category' column\",\n",
    "      full_dataset['category'].unique(), sep='\\n')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 150,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "IegxWJN7RKxi",
    "outputId": "ee06f359-1a57-473a-edb0-b38823b1ab38"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "       proto_number  flgs_number  pkts  bytes        dur  spkts  dpkts  \\\n",
      "0                 3            1     8    480  25.001644      8      0   \n",
      "1                 3            1     8    480  25.001644      8      0   \n",
      "2                 3            1     8    480  25.001644      8      0   \n",
      "3                 3            1     8    480  25.001644      8      0   \n",
      "4                 3            1     8    480  25.001644      8      0   \n",
      "...             ...          ...   ...    ...        ...    ...    ...   \n",
      "10995             3            1    15    900  13.658719     15      0   \n",
      "10996             3            1    15    900  13.658719     15      0   \n",
      "10997             3            1    15    900  13.658719     15      0   \n",
      "10998             3            1    15    900  13.658719     15      0   \n",
      "10999             3            1    15    900  13.658802     15      0   \n",
      "\n",
      "       sbytes  dbytes     seq    stddev  N_IN_Conn_P_SrcIP       min  \\\n",
      "0         480       0  120567  0.023105                100  3.539353   \n",
      "1         480       0  120568  0.023096                100  3.539352   \n",
      "2         480       0  120569  0.023096                100  3.539353   \n",
      "3         480       0  120570  0.023059                100  3.539353   \n",
      "4         480       0  120571  0.023096                100  3.539353   \n",
      "...       ...     ...     ...       ...                ...       ...   \n",
      "10995     900       0  120218  1.367880                100  1.976485   \n",
      "10996     900       0  120219  1.367880                100  1.976486   \n",
      "10997     900       0  120220  1.367880                100  1.976485   \n",
      "10998     900       0  120221  1.367879                100  1.976486   \n",
      "10999     900       0  120222  1.367841                100  1.976569   \n",
      "\n",
      "       state_number      mean  N_IN_Conn_P_DstIP  drate     srate       max  \\\n",
      "0                 4  3.565624                100    0.0  0.279982  3.601970   \n",
      "1                 4  3.565624                100    0.0  0.279982  3.601970   \n",
      "2                 4  3.565624                100    0.0  0.279982  3.601970   \n",
      "3                 4  3.565624                100    0.0  0.279982  3.601970   \n",
      "4                 4  3.565624                100    0.0  0.279982  3.601969   \n",
      "...             ...       ...                ...    ...       ...       ...   \n",
      "10995             4  3.910941                100    0.0  1.024986  4.885572   \n",
      "10996             4  3.910941                100    0.0  1.024986  4.885571   \n",
      "10997             4  3.910941                100    0.0  1.024986  4.885572   \n",
      "10998             4  3.910941                100    0.0  1.024986  4.885572   \n",
      "10999             4  3.910969                100    0.0  1.024980  4.885572   \n",
      "\n",
      "      category  \n",
      "0          DoS  \n",
      "1          DoS  \n",
      "2          DoS  \n",
      "3          DoS  \n",
      "4          DoS  \n",
      "...        ...  \n",
      "10995     DDoS  \n",
      "10996     DDoS  \n",
      "10997     DDoS  \n",
      "10998     DDoS  \n",
      "10999     DDoS  \n",
      "\n",
      "[33556 rows x 20 columns]\n"
     ]
    }
   ],
   "source": [
    "# best features according to the BoT_IoT documentation\n",
    "# best_features = ['Unnamed: 0', 'pkSeqID', 'proto', 'saddr', 'sport', 'daddr', 'dport',\n",
    "#                   'seq', 'stddev', 'N_IN_Conn_P_SrcIP', 'min', 'state_number', 'mean',\n",
    "#                   'N_IN_Conn_P_DstIP', 'drate', 'srate', 'max', 'attack', 'category',\n",
    "#                   'subcategory']\n",
    "\n",
    "# the features I want to use for prototyping\n",
    "trial_features = ['proto_number', 'flgs_number', 'pkts', 'bytes', 'dur',\n",
    "                  'spkts', 'dpkts', 'sbytes', 'dbytes',\n",
    "                  'seq', 'stddev', 'N_IN_Conn_P_SrcIP', 'min', 'state_number',\n",
    "                  'mean', 'N_IN_Conn_P_DstIP', 'drate', 'srate', 'max',\n",
    "                  'category']\n",
    "\n",
    "subset_dataset =  full_dataset[trial_features]\n",
    "print(subset_dataset)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 151,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 363
    },
    "id": "FOyvTJNweNQt",
    "outputId": "2ce9149d-3060-4c0d-a5f5-1324340eda0d"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "category\n",
      "DoS               11000\n",
      "Reconnaissance    11000\n",
      "DDoS              11000\n",
      "Normal              477\n",
      "Theft                79\n",
      "Name: count, dtype: int64\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA3kAAAMWCAYAAABMUk9aAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8qNh9FAAAACXBIWXMAAA9hAAAPYQGoP6dpAAC1wklEQVR4nOzddXzV9eLH8fc562IBCwYbG90ICgoojXkN1GsH2PpTrn1tQcUOwLrWtUAs0GsgSgiKoIKCpPTodfdO/P4YDEYO2PY553tez8djD9j31HsIc+/zKZvb7XYLAAAAAGAJdtMBAAAAAAD1h5IHAAAAABZCyQMAAAAAC6HkAQAAAICFUPIAAAAAwEIoeQAAAABgIZQ8AAAAALAQSh4AAAAAWAglDwAAAAAshJIHAAAAABZCyQMAAAAAC6HkAQAAAICFUPIAAAAAwEIoeQAAAABgIZQ8AAAAALAQSh4AAAAAWAglDwAAAAAshJIHAAAAABZCyQMAAAAAC6HkAQAAAICFUPIAAAAAwEIoeQAAAABgIZQ8AAAAALAQSh4AAAAAWAglDwAAAAAshJIHAAAAABZCyQMAAAAAC6HkAQAAAICFUPIAAAAAwEIoeQAAAABgIZQ8AAAAALAQSh4AAAAAWAglDwAAAAAshJIHAAAAABZCyQMAAAAAC6HkAQAAAICFUPIAAAAAwEIoeQAAAABgIZQ8AAAAALAQSh4AAAAAWAglDwAAAAAshJIHAAAAABZCyQMAAAAAC6HkAQAAAICFUPIAAAAAwEIoeQAAAABgIZQ8AAAAALAQSh4AAAAAWAglDwAAAAAshJIHAAAAABZCyQMAAAAAC6HkAQAAAICFUPIAAAAAwEIoeQAAAABgIZQ8AAAAALAQSh4AAAAAWAglDwAAAAAshJIHAAAAABZCyQMAAAAAC6HkAQAAAICFUPIAAAAAwEIoeQAAAABgIZQ8AAAAALAQSh4AAAAAWAglDwAAAAAshJIHAAAAABZCyQMAAAAAC6HkAQAAAICFUPIAAAAAwEIoeQAAAABgIZQ8AAAAALAQSh4AAAAAWAglDwAAAAAshJIHAAAAABZCyQMAAAAAC6HkAQAAAICFUPIAAAAAwEIoeQAAAABgIZQ8AAAAALAQSh4AAAAAWAglDwAAAAAshJIHAAAAABZCyQMAAAAAC6HkAQAAAICFUPIAAAAAwEL8TQcAAMBTlVc55XZLfnabAvxsstlspiMBAHBYlDwAgNerdLiUWVSujMJy5ZdWqbTSqbIqp8p3fZRVumo+L9t1W63bd10vr3LV/L7C4ZTLXft1bDbJ326T3WaTv90mP7tN/n52+dttCvS3KzjATyEBfgoOqP79vp+HBPopKiRQzcID1SwiSLHhQYqNCFLTsED5+zG5BgBQP2xut9t9+LsBAND43G63sosrlVFYvuujQumF5cosLFf6rs8zC8uVW1opb/6/mc0mRYUEKDYiSM3Cqz/2/J5CCAA4MpQ8AIAx6QXl2phVrJ0F5cooKldGQe0il1VcoSon/5va296FMC4iWK2ahqpNbLjaxIWrTWyYWkSFMK0UAHwcJQ8A0OCyiiq0NqNo10ex1u36fWG5w3Q0ywkJ8FNqs7Ca0tcmNlxtYsPVOjZMwQF+puMBABoBJQ8AUG9ySyq1NqNI6zKKtGavQpdXWmU6ms+z26TEqJCa0tcmbk8BjI0IMh0PAFCPKHkAgCNWUFqltZlFuwpdsdakF2ldZpGyiytNR8NRiAwJUOvYMHVq3kQ9k6LUq1W02sSGm44FADhKlDwAwCG53W6tyyzWorRcLdqUq8Wb87Qtr8x0LDSwqNCA6sKXHK1eraLVIylK4UFsyg0A3oCSBwCopdLh0vLt+VqUlqfFadWlLp/plj7PbpPax0eoV6toHb+r+KU2CzMdCwBwAJQ8APBxReVV+mNzXvVIXVqelm3LV3mVy3QseIGYsMCa6Z09k6N0XFKUQgMZ7QMA0yh5AOBjMgrL9fumXC1Oy9XvaXlak16436HfwNHws9vUIT5CvVpFqXdKjE5pF6uYsEDTsQDA51DyAMDisooq9OPfmfp1Y44Wbc7V1lzW06Fx2G1S95ZRGtIxTkM6xqlLYhPO8AOARkDJAwALWpNepFmrMzRzVYb+2pYvvtPDE8RFBGlQh1gN7hCnU9rHspELADQQSh4AWIDD6dLvm3I1c3WGZq/O1JbcUtORgEMK8LOpd0qMBneI0+COcWobx5ENDWnkyJF6//33JUn+/v6KiYlR9+7ddemll2rkyJGy2+11eh6n06lnn31W77//vjZv3qyQkBC1b99eN954o0aNGtWQXwKAI0DJAwAvVVhepblrsjRrVYbmrslUYbnDdCTgqCXHhGpwh1gN7hink1o3VXCAn+lIljJy5EhlZGTo3XffldPpVEZGhmbMmKGnnnpKp5xyir766iv5+x9+ZPXhhx/Wm2++qVdeeUUnnHCCCgsLtXjxYhUUFOjOO+9shK8EQF1Q8gDAi2zNLdWs1RmatTpDv2/KVZWTb+GwnpAAP/Vr01SDdq3laxEVYjqS1xs5cqTy8/P15Zdf1ro+Z84cDR06VG+99Zauu+46bdmyRbfddptmz54tu92u008/XS+//LLi4+MlSccdd5xGjBihRx991MBXAaCuKHkA4MHcbrf+2lagWauqi93f6UWmIwGNrmNChM7r2ULnHddCCZHBpuN4pYOVPKm6uCUmJurbb7/V8ccfr7CwMI0fP14Oh0O33HKLIiIiNHfuXEnS6aefrrKyMn3++eeKjY1t3C8CQJ2x4hkAPNCSLXma9ud2fb8yXZlFFabjAEb9nV6kp7/7W8/O+Ft92zTViJ4tdUbXBIWxcUu96Nixo5YtW6ZZs2Zp2bJl2rRpk5KSkiRJH374obp06aJFixapd+/eevHFF3XhhRcqISFBXbp0Ub9+/XTuuefqjDPOMPxVANgb3x0BwENkFpZr2pLtmvrHNq3LLDYdB/A4Lrf0y/oc/bI+Rw9/uULDO8drRK8WGtAuVn52jmY4Wm63WzabTatXr1ZSUlJNwZOkzp07KyoqSqtXr1bv3r3VuXNnrVixQn/88Yfmz5+vn376SWeffbZGjhypt99+2+BXAWBvlDwAMKjC4dSsVZn6/I+t+mldtpycSg7USVmVU1/9tUNf/bVDzcKDdE6PRJ3fq4W6tog0Hc3rrF69WqmpqTVlb1/7Xrfb7erdu7d69+6tO+64Q5MmTdKVV16pBx98UKmpqY0ZHcBBUPIAwIBl2/L1+R/b9NVfO5RfWmU6DuDVsosr9N9fNum/v2xSu7jw6vV7PVuwYUsdzJkzR8uXL9cdd9yhli1basuWLdq6dWvNaN6qVatUUFCgTp06HfQ5OnfuLEkqKSlplMwADo+NVwCgkWQVVejLJdv1+R/btCaDDVSAhmSzSSemxuj8ni11RrcERQQHmI5k1KGOUBg0aJC+/PJL2e12HX/88QoPD6+18Up4eHjNxisXXnih+vfvr379+ikhIUGbNm3S/fffr9zcXK1cubJOxzAAaHiUPABoQFVOl2avztBni7dp3tosOZiOCTS6IH+7hnWO1/k9W2hg+1j5+9Xt4G8r2fcw9OjoaPXo0UOXXXaZrr766prD0A93hMJbb72lKVOmaMWKFSooKFBCQoKGDBmiMWPGqFWrVsa+PgC1UfIAoAGs2F5QMx0zt6TSdBwAuyRGBuvKvim6rE+yIkN9e3QPgHVR8gCgnlQ4nPriz+16f+Fmrd5ZaDoOgEMICfDTiF4tdE3/VLWNCzcdBwDqFSUPAI5RXkmlPvx1sz5YuFnZxZxpB3gTm006pV2srumfooHtYw+4uyQAeBtKHgAcpbTsEr09f6Om/rFdZVVO03EAHKO2ceEa2S9FF/RqqZBAP9NxAOCoUfIA4AgtTsvVmz9t1KzVGWIfFcB6IkMCdEmfJF3dN0WJHMMAwAtR8gCgDlwut2asTNdbP2/Uki35puMAaAT+dptO65qga/qn6vhW0abjAECdUfIA4BBKKx36dNFW/feXNG3JLTUdB4AhPZKidE3/FJ3ZrbkCfPAIBgDehZIHAAeQWVSu9xekafJvW5RfWmU6DgAPkdAkWFf2baXLT0xWVGig6TgAcECUPADYy9qMIr3100b9768dqnS4TMcB4KEigvx1zcmpuu6UVEUEc94eAM9CyQMASX9sztXLc9Zr3tos8V0RQF1FhQbohgGtNapfKjtyAvAYlDwAPm3F9gI9/8MazV2TZToKAC/WLDxItwxqo8tPSlaQP2UPgFmUPAA+aV1GkV6cuVYzVqYzcgeg3jSPDNatQ9rqohOS2KAFgDGUPAA+ZXNOicbPWqf/Ld3OGXcAGkxyTKhGD22nET1byM9uMx0HgI+h5AHwCTsLyjRx9jp9tnibHLQ7AI2kTWyYbh/WXv/o3lw2G2UPQOOg5AGwtrI86ecX9MW2cN2xtpvpNAB8VKfmTXTn8PYa3jnedBQAPoCSB8CaHBXSb29IP78glefLGRavEwqfU16Vv+lkAHxYj6Qo3TW8vQa0jzUdBYCFUfIAWIvbLS37RJozTirYUuumeUk36+p1pxgKBgB79EmN0d2ndlCf1BjTUQBYECUPgHVs+FGa+YiUvuyAN7uDInVKxXhtKw9q5GAAcGCndYnXQ2d1VlJMqOkoACyEkgfA+6WvkGY+LG2Yc9i7Lkm6WiPWndYIoQCgboL87bpxQGvdMritggM4Yw/AsaPkAfBeFUXV0zJ/f1NyO+v0ELd/iP5hm6iVRWENHA4AjkyLqBDdf2ZH/aN7oukoALwcJQ+Ad1r5pTTjfqloxxE/dF3ShRq+7vz6zwQA9eCk1jEac04XdUxoYjoKAC9FyQPgXfLSpOn3SOt+OOqncNv9dXXwRP2UG1VvsQCgPvnZbbr8xGTdNbyDIkMDTMcB4GUoeQC8g7NK+mWC9NPzkqPsmJ9uR4vT1W/DVfUQDAAaTtOwQN13Rkf984Qk01EAeBFKHgDPt3mB9M0dUtbf9faUbtl0R+RL+jIjrt6eEwAaSp/UGI07r6vaxUeYjgLAC1DyAHiukpzqXTOXfiSp/r9V5SX0V8+0/6v35wWAhhDgZ9O1J7fWv4a2U0ggu3ACODhKHgDP43ZLSz6UZj4qleU26Es90ewZvb2NaVAAvEfL6BCNPaeLhnaKNx0FgIei5AHwLJmrpW/ulLYsaJSXK23WQ523/btRXgsA6tOpneM15pwuSowKMR0FgIeh5AHwDJWl0rxnpIWvSq6qRn3pN+LH6KnN7Rv1NQGgPoQG+un+MzvpypNamY4CwINQ8gCYt/Z7afrdUv4WIy9fGdVWXTLHqsplM/L6AHCsBnWI1bMXdFdck2DTUQB4AEoeAHNKc6Vv75RWfmE6iT5LvFf3bDzOdAwAOGrRoQEaN6KbzuzW3HQUAIZR8gCYsXGe9MVNUtEO00kkSY6IFuqZ95SKHP6mowDAMRnRs4XGnttFTYI5RB3wVXbTAQD4GEel9MND0gfnekzBkyT/ou16KXWR6RgAcMy+WLJdp7/0kxaszzYdBYAhjOQBaDxZa6Wp10rpy0wnOSBXcLT6l7+kneWBpqMAwDGz2aRR/VJ17+kdFBzAuXqAL2EkD0DjWPS29MYAjy14kmQvz9PEpJ9MxwCAeuF2S//9ZZPOfnm+VmwvMB0HQCNiJA9AwyrJlv53q7T2O9NJ6sQdEKbTXBO1toRzpwBYR4CfTaOHtNMtg9vKz85OwoDVUfIANJz1s6Qvb5GKM0wnOSJ/J12s09edazoGANS7XslRevGi45TSLMx0FAANiJIHoP5VlUuzHpV+e0OS932LcdsDdEngy/otv4npKABQ70ID/fTAmZ10BQeoA5ZFyQNQvzJWSlOvkzJXmU5yTLa2PEunrL/cdAwAaDCDO8TqmQu7Ky6CA9QBq6HkAagfbrf023+kWWMkR7npNMfMLZv+L2KCpmc1Mx0FABpMdGiAnr2wh4Z3jjcdBUA9ouQBOHZFGdL/bqleg2ch2c0H6oRNN5qOAQANymaTbh3cVncMay87m7IAlsARCgCOzZoZ0uv9LFfwJKnZznm6KtFzDmwHgIbgdksvz1mva95fpIKyKtNxANQDRvIAHB23W5r7tDTvGXnj5ip1VRR3grptudN0DABoFK2ahuqNK49XxwQ2ngK8GSN5AI5cRbH0yRXSvKdl5YInSRGZi3VXqw2mYwBAo9icU6rzX1ugr/9iFgPgzRjJA3BkcjdJH1/m9btnHomKmA7qvPNhOd28LwbAd1x3cqruP7MTh6cDXoifWADU3YYfpbcG+1TBk6Sg3DV6PGWl6RgA0Kjenr9JV7z9m3KKK0xHAXCEGMkDUDcLX5V+eFhyO00nMcLRJEk9cp5SiZP3xgD4lsTIYP3nyuPVvWWU6SgA6oifVgAcmqNC+uJm6fsHfLbgSZJ/4VY9n7rYdAwAaHQ7Csr1z/8s1KeLt5qOAqCOGMkDcHCFO6s3WNlOuZEkV2gznVj8grIqA0xHAQAjLj8xWY+e3UWB/owTAJ6Mf6EADmzbYunNQRS8vdhLszWh1S+mYwCAMZN/26JL3lyozMJy01EAHAIjeQD2t2Sy9M0dkpPF9vtyB4ZrqGOiNpYGm44CAMbERgTp9ct76YSUGNNRABwAI3kA9nA6pO/+Lf3vFgreQdgqizWhxWzTMQDAqKyiCl361q/6YGGa6SgADoCRPADVSnOlz0ZKm+aZTuLx3H5BusD/Zf1ZEG46CgAYd93JqXrwrE6y2ThPD/AUjOQBkDJWVZ9/R8GrE5uzQi/ETjcdAwA8wtvzN+mOT5aqyukyHQXALpQ8wNetny29M1zKSzOdxKukbP9Kw5vlmo4BAB7hy6U7dO37i1VS4TAdBYAoeYBvWzFNmnKJVFlsOonXsbldeqLJF6ZjAIDH+Gltli5961flFLOmGzCNkgf4qkXvSFOvlZyVppN4rfgds3Vx83TTMQDAYyzbVqAL/7NQW3NLTUcBfBolD/BFPz0nfXun5Gb9xLF6MPAT0xEAwKNsyi7R+a8v0ModBaajAD6Lkgf4ErdbmvGANOcJ00kso0nGb/q/pDTTMQDAo2QVVeiSN37VgvXZpqMAPomSB/gKl1P63/9Jv75qOonl3OqaLJuN02gAYG9FFQ6NfHeRvlm2w3QUwOdQ8gBfUFUufXKltHSy6SSWFJKzUo+m/G06BgB4nEqnS6OnLNF7v2wyHQXwKRyGDlhdeaH08WVS2s+mk1haVWSKumc/rjKnn+koAOCRbhnURvee3tF0DMAnMJIHWFlJtvT+Pyh4jSCgIE1Ppyw1HQMAPNZrczfons/+koND04EGx0geYFX5W6UPR0g560wn8RnOsHidUPic8qr8TUcBAI81pGOcXr2sl0ICmfkANBRG8gArylor/fd0Cl4j8yvJ0PiUhaZjAIBHm/N3pi57+1fllXBOK9BQKHmA1Wz/Q3r3dKlwm+kkPmlA5mQlh5SbjgEAHm3Jlnxd+hZFD2golDzASjbOk94/RyrNMZ3EZ9kqCjWx5VzTMQDA4/2dXqQr3vlNBaVVpqMAlkPJA6zi72+lyf+UKotNJ/F5PXZ+pm4RJaZjAIDHW7mjUFf99zcVlVP0gPpEyQOsYN1M6bORkrPCdBJIsjnK9FLCDNMxAMAr/LWtQFf/93eVVDhMRwEsg5IHeLu0+dUHnTtZ1+BJ2mz/nwbE5JuOAQBe4c8t+Rr17iKVVTpNRwEsgZIHeLNti6WPLpYcZaaTYB82l0NPR39lOgYAeI3f03J13QeLVF5F0QOOFSUP8Fbpy6VJF7AGz4M13/69zo/PNB0DALzGL+tzdMOHf6jCQdEDjgUlD/BG2euqDzovzzedBIdgk1sPh3xmOgYAeJWf1mbplkl/qsrpMh0F8FqUPMDb5G2WPjhXKskynQR1EJ3+i25oucV0DADwKrP/ztRtHy2Rg6IHHBVKHuBNCndKH5wjFW43nQRH4HZNMR0BALzOjJXpuv2TpXK63KajAF6Hkgd4i5Kc6hG8vDTTSXCEQrP/0gMpa03HAACv882ynbrns7/kougBR4SSB3iD8gLpw/Ok7DWmk+AojSz/UAF2fkgBgCM1bcl23TdtmdxuvocCdUXJAzxdZYk06UIpfZnpJDgGgfkb9FTKX6ZjAIBX+nTxNj305QrTMQCvQckDPFlVuTTlEmnb76aToB6MKJykCH+H6RgA4JUm/7ZFY75aaToG4BUoeYCnclZJn10tbfrJdBLUE7/iHXophcIOAEfrvQVpenEma5yBw6HkAZ7I5ZSmXS+tnWE6CerZkOzJah5caToGAHitibPXaeof20zHADwaJQ/wNG639NVoaeUXppOgAdjL8/Ry0jzTMQDAq90/bbl+3ZhjOgbgsSh5gKeZ84S0dJLpFGhAx6d/qo7hpaZjAIDXqnS6dNOkP7Qxq9h0FMAjUfIAT7LsM+nn502nQAOzVZVofPOZpmMAgFfLL63SNe8tUl4JU+CBfVHyAE+xdZH01a2mU6CRdNg+TSdGFZqOAQBeLS2nVDd8uFgVDqfpKIBHoeQBniB/q/TxZZKj3HQSNBKbq0rPN/vadAwA8HqL0vL07885SxbYGyUPMK2yRJpyqVSSaToJGlnLbdN1Zmy26RgA4PW+XLqDoxWAvVDyAJPcbmnq9VLGctNJYIBNbj0ePtV0DACwhImz12nanxytAEiUPMCsWWOkNd+aTgGDmu6cp6sTt5uOAQCWcN/U5fqNoxUASh5gzNIp0i/jTaeAB7jH/2PTEQDAEiqdLt046Q9tyi4xHQUwipIHmLDlV+nr0aZTwEOEZ/6hu5LXm44BAJaQX1qlUe/+ztEK8GmUPKCx5W+RPr5ccvI/H+xxg2Oy/Gwu0zEAwBLSckp144d/qNLB91X4Jkoe0JgqiqWPLpFK2VERtQXlrtHjqStNxwAAy/g9LVf3fv6X6RiAEZQ8oLG4XNLU66RMfpDHgV1U9KHC/HjXGQDqy5dLd2j8LI5WgO+h5AGNZdYj0trvTKeAB/Mv2qbnUxeZjgEAljJ+1jpNX77TdAygUVHygMawZJK04GXTKeAFTsuZrNjAKtMxAMBS/v35Mm3OYcdN+A5KHtDQNi+QvrnDdAp4CXtZtia0+sV0DACwlKIKh/7voz9V4XCajgI0Ckoe0JAKd0qfXMlOmjgifTM+UpvQMtMxAMBSVmwv1LhvV5uOATQKSh7QUFwuadr17KSJI2arLNaEFrNNxwAAy/lg4WZ9u4z1ebA+Sh7QUOa/IKX9bDoFvFSXHVPVK7LIdAwAsJz7prI+D9ZHyQMawpbfpLlPm04BL2ZzVujF2OmmYwCA5bA+D76AkgfUt7K86vPwXA7TSeDlWm3/WsOb5ZqOAQCWs2J7oZ74hvV5sC5KHlDfvrpNKthiOgUswOZ2aVyTL0zHAABL+vBX1ufBuih5QH1a9I60+mvTKWAhcTtm65Lm/BACAA2B9XmwKkoeUF8yVkrfP2A6BSzogcBPTEcAAEtifR6sipIH1IfKUunzayRHuekksKAmGb/rtqRNpmMAgCWxPg9WRMkD6sOM+6Ssv02ngIXd4vpINpvbdAwAsKQPf92sb5btMB0DqDeUPOBYrZgm/fm+6RSwuJCclXo0hTcSAKCh3D91udKyWZ8Ha6DkAccib7P09e2mU8BHXF76oUL8WDcCAA2B9XmwEkoecLScDmnqtVJFgekk8BEBBWl6JnWJ6RgAYFkrdxTq8W9WmY4BHDNKHnC0fnxC2rbIdAr4mLPyJik6wGE6BgBY1qRft2j26gzTMYBjQskDjsaGH6X5402ngA/yK8nUhJSFpmMAgKU9+MUKFZZXmY4BHDVKHnCkirOkL26UxE6HMOOUzMlKDuG4DgBoKOmF5XryW45VgPei5AFH6qvbpGKmccAcW0WhJracazoGAFjax4u26pf12aZjAEeFkgcciWWfSWu/M50CUI+dn6pbBFt9A0BDum/aMpVWsg4a3oeSB9RVSbY049+mUwCSJJujXC8lzDAdAwAsbWtumZ6dscZ0DOCIUfKAupp+j1SaYzoFUKPNti81sGme6RgAYGnvL0zT4rRc0zGAI0LJA+ri72+lldNMpwBqsbmdejrqK9MxAMDS3G7p3qnLVF7FIenwHpQ84HDK8qVv7jSdAjighO0/6Pz4TNMxAMDSNmaVaPysdaZjAHVGyQMO54eHpOJ00ymAA7LJrUdCPjUdAwAs762fN2rZtnzTMYA6oeQBh7LhR2nJh6ZTAIcUlb5AN7TcbDoGAFia0+XWvZ8vU5XTZToKcFiUPOBgKkulr/9lOgVQJ7driukIAGB5f6cX6dUf15uOARwWJQ84mLlPSfmMjsA7hGYv0wMpbPMNAA3t1R/X6+/0QtMxgEOyud1ut+kQgMdJXyG9OVBycQAqvEdlVBt1yXxMVS6b6SiwqKIl01W0ZLocBRmSpIBmyYrqd6lC2pwgt9Oh/J8/VNmGxXIUpMseFKbgVj0UNXCk/COaHvQ50z+6TxVbV+x3PaT1CYr75xhJktvlVP78j1Syaq5cJXnyC4tWWLdhiux3sWy26verS9csUNHS71SZsUGuskI1HzlRgfGtaz1n7uy3VLJitmwBIYoeNFJhnQfW3Fay+meVrJyjuAsfPdY/JviA7i0j9cUt/eVn5/stPJO/6QCAx3G7pW/uoODB6wTmb9DTqX/prg3HmY4Ci/KLaKrogVfLPzpRklS8YrYypz2h5iMnyL9JM1Wmb1Bkv0sUGJcqV3mxcme/paxpj6v51eMP+pyxIx6UnHu+3zrLCrXz3dsU2vHkmmuFv36u4qXfqelZdyiwWbIqdq5TzncTZA8KVZMTzpUkuarKFdSys0I7nqzcGS/v9zql639Tyep5irvocTnydijnuwkKTu0pv5AmcpUXK//nDxR/ybh6+pOC1S3bVqC3ft6omwa2MR0FOCCmawL7+uNdadvvplMAR+W8gg8V4c8bFGgYoW1PVEib3gqIaaGAmBaKHnCV7IHBqtixRvagMMVf8oTCOp2igKYtFdSio2KG36jK9PVyFB78mA+/kAj5hUfXfJSnLZUtIEihHfaUvIodfyuk7YkKbdNb/pHxCut4skJSeqoyfc/aqPCuQxTV/1KFpBx3wNepytmq4KRuCmreTmGdB8oWGCpHfvXOyXlz31VEz7Pk3ySufv6g4BNemrlWG7OKTccADoiSB+ytOFOaNcZ0CuCo+RXv1PhU3qRAw3O7nCpZNa96BK1FxwPex1VRKskme1B4nZ+3eNkPCus0QPbA4JprQS07q3zzX6rK3S5JqszcqPJtqxTS+oQ6P29gbKoq09fLWV6sivT1cjsq5B+dqPJtK1WZsUERx59d5+cCJKnC4dK/py4TK5/giZiuCextxv1SeYHpFMAxGZw1Wc2DT9DO8kDTUWBBlVlpSv/wbrkdlbIFhihuxIMKbJa83/3cjkrlz3tPYZ0Hyh4UWqfnrtixRlXZm9X0jNG1rjc58UK5Kkq0462bJLtdcrkUNeDKWmvqDiek9fEK6zJI6e/fIZt/oJqddYfsAUHK/f41NT3rjur1hn9+I7+QJoo57VYFxraq83PDdy1Ky9O0P7frguNbmo4C1ELJA3bbMEda8bnpFMAxs5fn6eWkebpw3XDTUWBBATEt1HzURLnKS1S69hdlf/uS4i97ulbRczsdyvrqWcntVsypt9T5uYuXzVRAs1YKSuxQ63rp6p9UsnKump19twJiW6kyY6PyZr8lv/CmCu82tM7PH3Xy5Yo6+fKaz/PnT1ZwynGy2f1UsPATJV7zqsrW/66cb19U85ET6vy88G3Pfv+3Tu+aoLAgfqyG52C6JiBJVeXSN3eaTgHUm+PTP1HH8FLTMWBBNr8ABUQnKqh5O0UPHKnAuFQVLf6q5na306Gs/z0tR3664i5+vM6jeK6qcpWs/knhPU7d77a8ue8q8qQLFdZ5oAJjUxTedYgiep+rgl8/O+qvoypnq0pWzVXUKVeofMtyBbfsKr/QSIV2PKV6h84K/v2gbjIKKzg7Dx6HkgdI0sJXpLxNplMA9cZWVaoJzX8wHQM+wS23s6r6d7sLXt4OxV8yTn4hTer8LKV/z5fbWaWwLoP3f4WqCslW+0cWm80uuV1Hl9jtVs6MVxQ9+DrZA0Mkt0vu3Tsq7/71KJ8bvunt+Zu0JYc3BuA5KHlASbb0C9NyYD3tt3+hvtGsMUX9yZv3vsq3rpCjIEOVWWnK++kDlW9ZobDOg+R2OZX15VOqTF+vZmffLblcchbnyVmcV1MCJSn7mxeUN++9/Z67eNkPCm130gGLYUjbPipY8IlKNyySoyBDpWsXqHDRlwpt37fmPs6yIlVmbFRV9hZJUlXuNlVmbJSzOG//1/rr++pRu3YnSpKCWnRS+eZlqtj+twoX/U8BTZNlD677ZjFApcOlJ75dZToGUIPJw8Dcp6WKQtMpgHpnc1Xpuabf6OS8yw9/Z6AOnCX5yv7mRTlLcmUPClNgbIri/jlWIak95SjIUNn63yRJO9+tvXFK/KVPKji5uyTJUZi136hcVe52VWxbpbiLHj/g68YMu1H5P09S7g+vyVVaIL/wGIUfd4ai+l9Sc5+y9b8pZ/r4ms+zv3pWkhTZ/9Ja6/CcJXkqWPipEq54ruZaUGIHNekzQpmfj5U9NFLNzrrjKP504Ot+WJWh+euydXK7ZqajALK52fcVvix7vfTaiRx8Dstyy6b/i5ig6Vn80AEADa19fLimjz5F/n5MloNZ/A2Eb5s9hoIHS7PJrcfDp5qOAQA+YW1GsSb9utl0DICSBx+25Vdp9demUwANrunOebo6cbvpGADgE16atU4FpVWHvyPQgCh58F0/PGw6AdBo7vGbYjoCAPiEgrIqvTxnnekY8HGUPPimVf+Ttv1uOgXQaMKz/tTdyZzjBACN4YOFm7U1lyMVYA4lD77HWSXNGmM6BdDorq+aLD8bZ38BQEOrdLr07PdrTMeAD6Pkwfcs/q+Uu9F0CqDRBeWt0eOpK03HAACf8M2yHfpra77pGPBRlDz4lvJCad4zplMAxlxU9KHC/J2mYwCA5bnd0rjpq03HgI+i5MG3zH9JKs0xnQIwxr9om15IWWw6BgD4hN835eqHlemmY8AHUfLgOwq2S7++bjoFYNypOZMVF8T23gDQGJ6e8bccTtZDo3FR8uA7fhwnOcpMpwCMs5dla0LyfNMxAMAnbMwq0ZRFW03HgI+h5ME3pK+Q/uKcMGC3kzKmqE0ob3oAQGOYOHudyqtYD43GQ8mDb5j5sORmqgSwm62yWBNazDYdAwB8QlZRhT5dzGgeGg8lD9a3YU71B4BauuyYql6RRaZjAIBPeGPeRlWxNg+NhJIH65v5qOkEgEeyOSv0Yuy3pmMAgE/Ynl+mL/7cbjoGfAQlD9a2bqaUvsx0CsBjtdr+jYY3yzUdAwB8wuvzNsjpcpuOAR9AyYO1zX/JdALAo9ncLo1rMs10DADwCZuyS/TNsh2mY8AHUPJgXVsXSZt/MZ0C8HhxO+bokuY7TccAAJ/w6o/r5XYzmoeGRcmDdTGKB9TZA4GfmI4AAD5hbUaxvl+ZYToGLI6SB2vKWiOtmW46BeA1mmT8rtHJG03HAACf8OqP601HgMVR8mBN88dLYioEcCRucX4km41/NwDQ0JZvL9CPazJNx4CFUfJgPQXbpOWfmU4BeJ3gnFUak7LadAwA8AmvzGE0Dw2HkgfrWfCK5KoynQLwSpeVTlKIn9N0DACwvD8252nBhmzTMWBRlDxYS2mu9OcHplMAXiugIE3Ppi41HQMAfAJr89BQKHmwlt/flKpKTKcAvNqZeR+qaSCj4QDQ0H5Zn6M/t+SZjgELouTBOipLpd/eMJ0C8Hp+JZka32qh6RgA4BNYm4eGQMmDdfz5vlSWazoFYAknZ36k5JBy0zEAwPLm/J2plTsKTMeAxVDyYA3OKmnhq6ZTAJZhqyjUyy1/NB0DAHwCa/NQ3yh5sIbln0kFW02nACyl+87P1C2CNa4A0NBmrEjX+swi0zFgIZQ8eD+3W/plgukUgOXYHOUanzDDdAwAsDyXW/rvL2mmY8BCKHnwfmu+k7L+Np0CsKTW277UwKbs/AYADe1/S7aruMJhOgYsgpIH7zf/JdMJAMuyuZ16Oup/pmMAgOWVVDr15ZLtpmPAIih58G5bfpO2/W46BWBpzbf/oPPjM03HAADL++i3LaYjwCIoefBuf7xnOgHgEx4J+dR0BACwvFU7CzkcHfWCkgfvVV4grfrSdArAJ0SlL9CNLXmHGQAa2uRf+V6LY0fJg/da/plUVWo6BeAz/qWPZLO5TccAAEv7dvkOFZRWmY4BL0fJg/f68wPTCQCfEpq9TA+0WmM6BgBYWnmVS5//uc10DHg5Sh68046l0s6/TKcAfM7V5ZMUZHeZjgEAlvbRb5tNR4CXo+TBOzGKBxgRmL9RT6byBgsANKQNWSVauCHHdAx4MUoevE9lqbT8c9MpAJ91XsEkRfhzYC8ANKTJjObhGFDy4H1WfSlVFJhOAfgsv+KdmpDK+ZQA0JB+WJmh7OIK0zHgpSh58D5M1QSMG5Q1SS2C+eEDABpKpdOlTxdvNR0DXoqSB++StVbastB0CsDn2cvzNTHpJ9MxAMDSpvy+RW43R9fgyFHy4F3+fN90AgC79Er/RB3DOasSABrK1twyzVubZToGvBAlD97DUSn99bHpFAB2sVWVakLzH0zHAABLm/zbFtMR4IUoefAea76VSrNNpwCwl/bbv1DfaDZCAoCGMufvTKUXlJuOAS9DyYP3YMMVwOPYXFV6LuZr0zEAwLKcLrem/M5oHo4MJQ/eIW+ztOFH0ykAHECL7d/pH7GMsgNAQ/n8j21swIIjQsmDd1gySRLf3ABPZJNbY8M/Nx0DACxre36ZlmzNNx0DXoSSB8/nckpLJ5tOAeAQmu78SSMTt5mOAQCW9e2ynaYjwItQ8uD51s+WCrebTgHgMO72m2I6AgBY1nfLdzJlE3VGyYPnW8E0MMAbhGct0T2t1pmOAQCWtKOgnCmbqDNKHjybs0paO8N0CgB1dF3lZPnZXKZjAIAlMWUTdUXJg2fb9JNUzhlcgLcIylurcakrTccAAEtiyibqipIHz7aa87cAb3Nh0YcK83eajgEAlsOUTdQVJQ+ey+WS1kw3nQLAEfIv2qYXUhabjgEAlsSUTdQFJQ+ea+tvUnGG6RQAjsKpOZMVF1RlOgYAWA5TNlEXlDx4LqZqAl7LXpatick/m44BAJbDlE3UBSUPnutvSh7gzU7MmKI2oWWmYwCA5TBlE4dDyYNn2rFUyt9iOgWAY2CrLNHEFrNMxwAAy2HKJg6HkgfPxFRNwBI675iqXpFFpmMAgKUwZROHQ8mDZ6LkAZZgc1bqpdhvTccAAMthyiYOhZIHz5O9TspeYzoFgHqSvP0bndos13QMALAUpmziUCh58DyrvzKdAEA9srldGtdkmukYAGApTNnEoVDy4HmYqglYTuyOObqsOVOLAKA+MWUTB0PJg2cp2CbtWGI6BYAGcF/Ax6YjAIClMGUTB0PJg2dhFA+wrCaZizQ6eaPpGABgGTsKyrVqZ6HpGPBAlDx4FkoeYGm3OD+Szca7zgBQXxaszzEdAR6IkgfPUZwlbVloOgWABhScs0pjU1aZjgEAljF/fbbpCPBAlDx4jrXfSW6X6RQAGtilJZMU4uc0HQMALGFRWq6qnPz8hNooefAcG+eaTgCgEQQUbtazKWywBAD1obTSqSVb8k3HgIeh5MFzpP1iOgGARnJm/iQ1DawyHQMALIEpm9gXJQ+eIXudVJxuOgWARuJXkqnxrViDCwD1YQElD/ug5MEzpP1sOgGARnZy5mSlhJSbjgEAXm/p1nwVVzhMx4AHoeTBM6TNN50AQCOzVRRpQssfTccAAK/ncLn1+yaOUsAelDx4Bkoe4JO67/xM3ZsUm44BAF7vF87Lw14oeTAva61UnGE6BQADbI5yvRQ3w3QMAPB6v7AuD3uh5ME81uMBPq319v9pYNM80zEAwKutyShSdnGF6RjwEJQ8mMdUTcCn2dxOPRP1P9MxAMCrud3Sgg1M2UQ1Sh7M28z5eICvS9j+gy6IZ9o2AByLX9YxZRPVKHkwi/V4AHZ5OORT0xEAwKv9soGSh2qUPJjFejwAu0SlL9RNLTebjgEAXmtbXpk255SYjgEPQMmDWazHA7CX0Zoim81tOgYAeC2OUoBEyYNprMcDsJfQ7GV6sNVa0zEAwGsxZROSBUvesmXLDnrbl19+2XhBcHisxwNwAFeVf6ggu8t0DADwSgs35MjtZkaEr7NcyTvttNO0cePG/a5PnTpVl19+uYFEOKi0n0wnAOCBAvM36qnUv0zHAACvlFtSqfWZxaZjwDDLlbybb75ZQ4cO1c6dO2uuffLJJ7rqqqv03nvvmQuG/bEeD8BBnFswSZEBDtMxAMArLd9eYDoCDLNcyXvkkUd0zjnnaNiwYcrNzdVHH32kUaNG6YMPPtA///lP0/GwtzTW4wE4ML/inXop5TfTMQDAK63YXmg6AgzzNx2gIUyYMEFXXnmlTjrpJG3fvl1TpkzRueeeazoW9pa1VirJNJ0CgAcblDVZLYJP0PbyINNRAMCrrNjBSJ6vs0TJ++qrr/a7dt5552nevHm69NJLZbPZau5zzjnnNHY8HMiOP00nAODh7OX5mpg0TxesO9V0FADwKqt3FMrtdstms5mOAkNsbgtsv2O3123Wqc1mk9PpbOA0qJPvH5QWvmI6BQAP5w4I1ZnuiVpdHGo6CgB4lR/vHqTUZmGmY8AQS6zJc7lcdfqg4HmQjBWmEwDwAraqUo1v/r3pGADgdVaw+YpPs0TJgxfKWGk6AQAv0X77F+ofzQ8rAHAkWJfn2yxX8kaPHq2JEyfud/2VV17R7bff3viBsL+iDKkky3QKAF7C5nLomZivTccAAK+ykh02fZrlSt7UqVPVv3///a7369dPn3/+uYFE2E/GctMJAHiZFtu/09lxvDkEAHXFSJ5vs1zJy8nJUWRk5H7XmzRpouzsbAOJsJ901uMBODI2uTU2lDfqAKCu8kurtDW31HQMGGK5kte2bVvNmDFjv+vfffedWrdubSAR9sOmKwCOQkz6z7qmxVbTMQDAa6xkNM9nWeKcvL3deeeduvXWW5WVlaUhQ4ZIkmbPnq0XXnhB48ePNxsO1RjJA3CU7rR/rP/qHtMxAMArrNheqNO7NjcdAwZYruRdc801qqio0Lhx4/T4449LklJSUvT666/rqquuMpwOclRIOetMpwDgpcKzluieVuv03OZ2pqMAgMdjXZ7vssRh6AeTlZWlkJAQhYeHm46C3Xb+Jb0xwHQKAF6sIrq9umY8qiqXzXQUAPBozcKDtPihYaZjwADLrcnbW2xsLAXP0zBVE8AxCspbqydS2KUXAA4nu7hCGYXlpmPAAMtN15Skzz//XJ9++qm2bNmiysrKWrf9+eefhlJBEpuuAKgXFxR9qMf8n1aJw890FADwaCu2Fyi+SbDpGGhklhvJmzhxokaNGqW4uDgtWbJEffr0UdOmTbVx40adccYZpuMhnXffARw7/6LtejFlkekYAODxVnAouk+yXMl77bXX9Oabb+qVV15RYGCg7r33Xs2cOVOjR49WQQGLT41jJA9APRmeM1lxQVWmYwCAR1u+nZ9/fZHlSt6WLVvUr18/SVJISIiKiookSVdeeaWmTJliMhoKtktleaZTALAIe1mOJib/bDoGAHi01TsZyfNFlit5CQkJysnJkSS1atVKv/76qyRp06ZNsvBGot4hY6XpBAAs5sSMKWoXVmY6BgB4rB0FZapwOE3HQCOzXMkbMmSIvv76a0nStddeqzvuuEPDhw/XxRdfrBEjRhhO5+MyWI8HoH7ZKks0IXGW6RgA4LHcbml7Hm+G+RrLnZPncrnkcrnk71+9ceinn36q+fPnq23btrrpppsUGBhoOKEP+2yUtHKa6RQALMbtF6h/+r+sxQURpqMAgEd6/5o+Gtg+1nQMNCLLlbwtW7YoKSlJNlvtQ3Ldbre2bt2q5ORkQ8mg1/pKmatMpwBgQZtbnqOB6y8xHQMAPNLj53XVlSe1Mh0Djchy0zVTU1OVlZW13/Xc3FylpqYaSIQa+VtNJwBgUcnbv9GpzXJNxwAAj7Qtt9R0BDQyy5U8t9u93yieJBUXFys4mIMgjSnLlyqLTKcAYFE2t0vjmjAdHAAOZGseJc/X+JsOUF/uvPNOSZLNZtPDDz+s0NDQmtucTqd+++03HXfccYbSQQXbTCcAYHGxO+bo8uZnavLORNNRAMCjbM1l4xVfY5mSt2TJEknVI3nLly+vtcFKYGCgevToobvvvttUPFDyADSC+wI+1mTdaToGAHiULUzX9DmWKXk//vijJGnUqFGaMGGCmjRpYjgRailgPR6AhheRuVj/StqoCVtbm44CAB6joKxKheVVahIcYDoKGonl1uTZbLYDrskrKSnRNddcYyARJDGSB6DR3OycLJvNUhtHA8Ax28ponk+xXMl7//33VVa2/7zjsrIyffDBBwYSQRIlD0CjCc5drbEpHNcCAHtjXZ5vsUzJKywsVEFBgdxut4qKilRYWFjzkZeXp+nTpysuLs50TN9FyQPQiC4tmaQQP6fpGADgMRjJ8y2WWZMXFRVVM1Wzffv2+91us9k0duxYA8kgiZIHoFEFFG7Wc6l/6tb1vU1HAQCPwDEKvsUyJe/HH3+U2+3WkCFDNHXqVMXExNTcFhgYqFatWikxkW21jXA5paKdplMA8DFn5k5S08DjlFPJRgMAwEieb7FMyRs4cKAkadOmTUpOTt5v85WcnByNHz9et99+u4F0Pq5wh+Rm2hSAxmUvzdKEVgt0xbqBpqMAgHFb81iT50sssyZvt1atWtUUPLfbre+//14XXXSREhMTNW7cOMPpfBRTNQEY0j/zI6WElJuOAQDGbcsrldvNzsO+wnIlT5LS0tL0yCOPqFWrVjrzzDMVFBSkb7/9Vunp6aaj+SZKHgBDbBVFmtjyR9MxAMC48iqXsooqTMdAI7FMyauoqNCUKVM0dOhQderUSStWrNCLL74ou92u+++/X8OGDZOfn5/pmL6Jg9ABGNRt52c6rkmx6RgAYBybr/gOy5S8Fi1a6PXXX9fFF1+sHTt2aNq0abrwwgtNx4LESB4Ao2yOcr0Y953pGABgHGfl+Q7LlDyn01lzhAIjdh6GkgfAsNTtX2lQTJ7pGABg1M4C1ij7CsuUvJ07d+qGG27QlClTlJCQoAsuuEBffPHFfrtswgBKHgDDbG6nno760nQMADCqoKzKdAQ0EsuUvODgYF1++eWaM2eOli9frk6dOmn06NFyOBwaN26cZs6cKaeTbfyNoOQB8AAJO2bqwoQM0zEAwBhKnu+wTMnbW5s2bfTEE09o8+bN+vbbb1VRUaF//OMfio+PNx3N95QXShUFplMAgCTp4aBPTEcAAGMKKXk+w5Ilbze73a4zzjhDn3/+ubZt26YHHnjAdCTfU5JlOgEA1IjM+FU3J6WZjgEARhSWU/J8haVL3t5iY2N155131vn+KSkpGj9+fMMF8hVl+aYTAEAtt7mnyGbjQGAAvofpmr7jiEreyJEja3aw9Pf3V3Jysm6++Wbl5Vlvx7JFixbphhtuMB3D+5Xnm04AALWEZi/XQ63WmI4BAI2O6Zq+44hH8k4//XTt3LlTaWlpevvtt/X111/rlltuaYhsRsXGxio0NNR0DO9Xzno8AJ7nyvJJCrK7TMcAgEbFSJ7vOOKSFxQUpISEBLVs2VKnnnqqLr74Yv3www81t7/77rvq1KmTgoOD1bFjR7322mu1Hr9t2zZdcskliomJUVhYmE444QT99ttvNbe//vrratOmjQIDA9WhQwd9+OGHtR5vs9n09ttva8SIEQoNDVW7du308ccf19w+d+5c2Ww2zZ49WyeccIJCQ0PVr18/rVmz513bDRs26Nxzz1V8fLzCw8PVu3dvzZo1q9br7Dtdc8yYMUpOTlZQUJASExM1evTomttee+01tWvXTsHBwYqPj691CPuMGTN08sknKyoqSk2bNtU//vEPbdiwoeb2tLQ02Ww2TZs2TYMHD1ZoaKh69OihhQsX1srzyy+/aODAgQoNDVV0dLROO+20mhFUt9utZ599Vq1bt1ZISIh69Oihzz///OD/ERsTJQ+ABwrM36inUv8yHQMAGlVRucN0BDSSY1qTt3HjRs2YMUMBAQGSpLfeeksPPvigxo0bp9WrV+vJJ5/Uww8/rPfff1+SVFxcrIEDB2rHjh366quv9Ndff+nee++Vy1X9buoXX3yhf/3rX7rrrru0YsUK3XjjjRo1apR+/PHHWq87duxYXXTRRVq2bJnOPPNMXXrppTUl7o477pAkPfjgg3rhhRe0ePFi+fv765prrql5fHFxsc4880zNmjVLS5Ys0Wmnnaazzz5bW7ZsOeDX+fnnn+ull17SG2+8oXXr1unLL79Ut27dJEmLFy/W6NGj9dhjj2nNmjWaMWOGBgwYUPPYkpIS3XnnnVq0aJFmz54tu92uESNG1HzNuz344IO6++67tXTpUrVv316XXnqpHI7qf4hLly7V0KFD1aVLFy1cuFDz58/X2WefXXMkxEMPPaR3331Xr7/+ulauXKk77rhDV1xxhebNm3cU/1XrGSUPgIc6N/9DRQbwAw8A3+FwuVVcwfc9X2Bzu911Xn0+cuRITZo0ScHBwXI6nSovL5ckvfjii7rjjjuUnJysZ555RpdeemnNY5544glNnz5dCxYs0Jtvvqm7775baWlpiomJ2e/5+/fvry5duujNN9+suXbRRReppKRE3377bXVgm00PPfSQHn/8cUnVJSo8PFxvvPGGbrjhhprDz2fNmqWhQ4dKkqZPn66zzjpLZWVlCg4OPuDX1qVLF91888269dZbJVWP5N1+++26/fbb9eKLL+qNN97QihUragrtbtOmTdOoUaO0bds2RUREHPbPMCsrS3FxcVq+fLm6du2qtLQ0paam6u2339a1114rSVq1apW6dOmi1atXq2PHjrrsssu0ZcsWzZ8/f7/nKykpUbNmzTRnzhz17du35vp1112n0tJSffTRR4fN1KBmjZHmv2Q2AwAcxI9Jt2jUupNNxwCARrPgviFKjAoxHQMN7IhH8gYPHqylS5fqt99+02233abTTjtNt912m7KysrR161Zde+21Cg8Pr/l44oknaqYnLl26VD179jxgwZOk1atXq3///rWu9e/fX6tXr651rXv37jW/DwsLk7+/v+655x4NHjy45vqjjz6qIUOGaMiQIXr00UclSUOGDJFUXYzuvfdede7cWVFRUQoPD9fff/990JG8f/7znyorK1Pr1q11/fXX64svvqgZZRs+fLhatWql1q1b68orr9TkyZNVWlpa89gNGzbosssuU+vWrdWkSROlpqZK0n6vtffX1Lx5c0lSZmZmzZ/b7sK6r1WrVqm8vFzDhw+v9ef+wQcf1JoWagwjeQA82MCsj9QiuMJ0DABoNKzL8w3+R/qAsLAwtW3bVpI0ceJEDR48WGPHjq0ZAXvrrbd04okn1nqMn5+fJCkk5PDvGuweidvN7Xbvd23f0bSwsDCde+65iouL09y5cyVJ3bp1qxm1y8zM1OLFi9WxY0dJ0j333KPvv/9ezz//vNq2bauQkBBdeOGFqqysPGCmpKQkrVmzRjNnztSsWbN0yy236LnnntO8efMUERGhP//8U3PnztUPP/ygRx55RGPGjNGiRYsUFRWls88+W0lJSXrrrbeUmJgol8ulrl277vdae39Nu7/e3VM6D/Xntvs+3377rVq0aFHrtqCgoIM+rtFQ8gB4MHt5viYmzdMF6041HQUAGgU7bPqGYz4n79FHH9Xzzz8vp9OpFi1aaOPGjWrbtm2tj92jV927d9fSpUuVm5t7wOfq1KnTflMSFyxYoE6dOh02x+DBg/Xcc8+pR48ekqSnnnpKL730kl566SXdc889kqRHHnlEkvTzzz9r5MiRGjFihLp166aEhASlpaUd8vlDQkJ0zjnnaOLEiZo7d64WLlyo5cuXS5L8/f01bNgwPfvss1q2bJnS0tI0Z84c5eTkaPXq1XrooYc0dOhQderU6aiOm+jevbtmz559wNs6d+6soKAgbdmyZb8/96SkpCN+rXpXXmg6AQAcUq/0T9UpvPTwdwQAC2Akzzcc8UjevgYNGqQuXbroySef1JgxYzR69Gg1adJEZ5xxhioqKrR48WLl5eXpzjvv1KWXXqonn3xS5513np566ik1b95cS5YsUWJiovr27at77rlHF110kXr16qWhQ4fq66+/1rRp0/bb+fJQxo8fXzNt82DLDdu2batp06bp7LPPls1m08MPP7zfRih7e++99+R0OnXiiScqNDRUH374oUJCQtSqVSt988032rhxowYMGKDo6GhNnz5dLpdLHTp0UHR0tJo2bao333xTzZs315YtW3Tfffcd2R+wpPvvv1/dunXTLbfcoptuukmBgYH68ccf9c9//lPNmjXT3XffrTvuuEMul0snn3yyCgsLtWDBAoWHh+vqq68+4terVxVFZl/fB72+qFKvL65UWn713+kucX56ZECgzmgXoCqnWw/NqdD09Q5tzHMpMsimYa399fSwICVGHPw9n0HvlWjeZud+189s569vL6s+asThcmvM3ApNXl6l9GK3mofbNPK4QD00IFD2XaPT01ZX6Y0/KvXHDpdyytxacmOYjkvwq/Wcd35frveWVio80KZnhwfrkq57Rrk/XVmlD5dV6etLOd4E9cdWVaoJSd/r1HUjTEcBgAZHyfMNx1zyJOnOO+/UqFGjtH79er399tt67rnndO+99yosLEzdunXT7bffLkkKDAzUDz/8oLvuuktnnnmmHA6HOnfurFdffVWSdN5552nChAl67rnnNHr0aKWmpurdd9/VoEGDjjhTv379tHHjRknab0TrpZde0jXXXKN+/fqpWbNm+ve//63CwoOPOEVFRenpp5/WnXfeKafTqW7duunrr79W06ZNFRUVpWnTpmnMmDEqLy9Xu3btNGXKFHXp0kWS9PHHH2v06NHq2rWrOnTooIkTJx7x19O+fXv98MMPeuCBB9SnTx+FhIToxBNPrNng5vHHH1dcXJyeeuopbdy4UVFRUerVq5ceeOCBI3qdBlHFu+ONrWUTm54eFqS2MdWl7f2lVTr34zItudGulk3s+jPdqYcHBKlHvF155W7dPqNC50wp1eIbwg/6nNMuDlWlc8+bJjmlbvX4T4n+2XnPt5Bn5lfqP4ur9P55weoS56fFO5wa9b8yRQZJ/zqpeupwSaVb/ZP89c/ONl3/dfl+r/P1mip9tLxKP1wZpnU5Lo36X5mGt/ZT01C78svdenBOhWZfRcFD/Wu3/Qv1jx6iX/IiTUcBgAZVyDEKPuGIdtf0Bi+++KIefvhh3Xrrrerfv7/cbrd++eUXvfrqq3riiSdqjlhAI3n5eClnvekUPi/mmUI9NzxY1/YK3O+2Rdud6vN2iTbfHq7kyLrN4B7/a4Ue+bFCO++KUFhg9SjdPz4qVXyYTe+cu2cN6QWflio0wKYPR9ReV5qW71LqhOL9RvKe/aVCf+506uMLq4tc/PNF+ubSUPVu4acbvi5Tp2Z23dHXA9aawpK2tzhD/TdcaToGADSo0UPb6c7h7U3HQAOrl5E8T/Lyyy/r9ddf11VXXVVz7dxzz1WXLl00ZswYSl5jqyozncCnOV1ufbbKoZIqqW+S3wHvU1Dhlk1SVLDtgLcfyDtLqnRJ14CagidJJyf76T+LK7U2x6n2Tf30V7pT87c4Nf60Ax9bciA94v305h+Vyitza2OeS2VVbrWNsWv+Fof+3OnU62fV/bmAI5W4fYbOjjtdX2fGmo4CAA2GjVd8g+VK3s6dO9WvX7/9rvfr1087d+40kMjHVZaYTuCTlmc41fedEpU7pPBA6YuLQ9Q5dv+SV+5w675Z5bqsW4CaBNWt5P2+3akVmS69c07t0bl/9w9UQblbHV8pkZ9dcrqkcUOCdGm3gIM80/5Oa+uvK7oHqPdbxQoJsOn980IUFijd/G253js3RK8vrtLLv1eqWahNb/6jelooUF9scmts6Of6WjebjgIADYaS5xuOeXdNT9O2bVt9+umn+13/5JNP1K5dOwOJfBxr8ozo0MyupTeF69frwnTzCYG6+styrcqqvXFKldOtSz4vk8stvXYEI2Tv/FmprnF29WlRu2B9stKhScur9NEFIfrzhjC9f16wnl9YqfeXHvhokoMZMyhY60dHaPnN4RrRKUBP/lyhYan+CvCTnvipQvNHheq6ngG66ktGiVH/YtJ/1jUttpqOAQANprCckucLLDeSN3bsWF188cX66aef1L9/f9lsNs2fP1+zZ88+YPlDA3I5JeeR/YCP+hHoZ1PbmOqRuRMS/bRoh1MTfq3UG2dXj75VOd266PMybcp3ac5VoXUexSutcuvjlVV6bND+6+LumVmu+/oH1eyG2S3eT5sL3HpqfqWuPm7/tYB18Xe2U5OXO7TkxjD9d0mlBrTyU2yYXRd1CdA1X5WrsMJd5+xAXd1ln6L/6l7TMQCgQZRV7b9bNqzHciN5F1xwgX777Tc1a9ZMX375paZNm6ZmzZrp999/14gRbI/dqJiq6THckip2fU/fXfDW5bg068pQNQ2t+7eBT1dWqcIhXdF9/ymYpVWSfZ++5WeTXEe5tZPb7dYNX5frhVODFB5ok9MlVe066WT3r0f73MChhGUt1b2t1pmOAQANwsn/PH2C5UbyJOn444/XpEmTTMcAm64Y8cDscp3R1l9JkXYVVbj18YoqzU1zasblQXK43LrwszL9udOpby4NldMtpRdXN6aYEJsC/apb2lVflKlFhE1PDas9jfOdJVU6r6P/AYvh2e39Ne7nCiVH2tQlzk9Ldjr14q+Vuua4PYUwt8ytLQUu7Siqfs012dW/JoTblBBe+znf+rNKcWE2ndOh+vH9k/01Zl6Fft3m0HfrHOocaz+izWKAI3Ft5WS9ZH9UVS7+jgGwlkMcDQ0LsWTJg4dgPZ4RGcVuXflFmXYWuxUZZFP3eLtmXB6q4W38lZbv0ldrqs/HOe6N2iOtP14dqkEp1d8SthS4ZLfVLl1rc6p3y/zhigOfU/fyGcF6+McK3TK9XJklbiVG2HTj8QF6ZOCeqZ1franSqP/tOR/vkqnVbwQ8OjBQYwbtKZQZxS49+XOFFlwbVnOtTws/3dU3SGd9VKa4sOpNWYCGEpS3Vk+kLNe/N3Y3HQUA6pXTWqen4SAsd04ePEjOBunlXqZTAMBRcUS0UI+8p1XiYBdXANbRMzlKX9zS33QMNDDLrcmDB7Hx1wuA9/Iv2q6XUhaZjgEA9crFmjyfwHRNNBw7734D8G7DciYrIaiX0iuObodYwCqK/vxWBb9Pk7M4V4HNkhU99HoFJ3U94H1L1yxQ0ZLpqszcKLezSgHNkhXV/zKFtD6+1n0Kfv1UVXk7JZdD/tGJatJ7hMK7Dqm5T/78ySr4ZUqt57aHRSnp1j37LhT8Nk2Fv0+TJEWedKGa9D6v5raKHWuU+8NrSrjqRdn4maQG0zUPLi0tTampqVqyZImOO+64o36e0tJSXXnllZo5c6aKioqUl5enqKioestZF5YveYWFhZozZ446dOigTp06mY7jW2x8QwXg3exlOZqQPF8Xrxty+DsDFlWy+iflzn5LMaferOAWnVW09DtlfjZGide9Jv8mcfvdv3zrCgWnHqeogVfJHhSm4uWzlDn1cTW/6gUFxreRJNlDwhXZ9yIFxCRJfv4q2/C7cqaPl19oZK0yGNAsWfEXj9vz5PY9s4Qqs9JUMH+yYi98RHK7lTX1MQWnHKfA2BS5nQ7lfP+qmp5+KwVvH7668YrNduiNtK6++mqNGTOmXl7r/fff188//6wFCxaoWbNmysvLU3R09DGXxyNhuZJ30UUXacCAAbr11ltVVlamE044QWlpaXK73fr44491wQUXmI7oO+yW++sFwAf1yZiidmF9ta6EzX7gmwoXfanw7sMV0eM0SVLMsBtUtulPFS2ZruiBI/e7f8ywG2p9Hj3wapWt+02l63+vKXnBybU3NQo44VyVrJijim2rapU82f3kFx59wFxV2VsVEJuikFY9qp8jNkVVOdsUGJuiwt+nKTipi4Katz/aL9uyXD46krdz586a33/yySd65JFHtGbNmpprISEhysvLq5fX2rBhgzp16qSuXatHu9PS0urleY+E5RZN/fTTTzrllFMkSV988YXcbrfy8/M1ceJEPfHEE4bT+RjeOQNgAbbKEk1InGU6BmCE21mlyvT1CkntWet6SGpPVWz/u27P4XbJVVkme3D4QW53qyxtqapytylonymgjrwd2vbqVdr2n2uV9b9nVJWfXnNbYGyKHHnb5SjMlKMgU47c7Qps1kpVeTtUvHyWok658gi/Wt/gq+fkJSQk1HxERkbKZrPtd223jRs3avDgwQoNDVWPHj20cOHCWs+1YMECDRgwQCEhIUpKStLo0aNVUlK9a/mgQYP0wgsv6KeffpLNZtOgQYOUmpoqSerZs2fNtYZmuaGWgoICxcTESJJmzJihCy64QKGhoTrrrLN0zz33GE7nY9h4BYBFzGtaqBahTyouNK72R8iuX8PiFB8ar5jgmP2OH6kPLpdbbpe7+ldn9a8up1tu965fd31e636uvW7b57E115yHuL/TLbdLcrlce25za8/r7PtcR/D8bOztPXILs3SD26V/JbRWx5j4mutTmyVp3ua/9MJe1w7mf/M+1BfOSj3f93xFhsfUXC8pK9aNT56pKkel7HY/3XTevzX0+NNrbv+z/YmqTD1OzZslq6A4R5/P+a92TL5XL935iSLCoqSYeH1/xq369rMxkqRrz7xNp7U/QWPfukVXnH2HnFkb9emsN+Xn569rzr5LnVuz47ckySFNfXaxkZdO7tJUvc9KNfLaR+LBBx/U888/r3bt2unBBx/UpZdeqvXr18vf31/Lly/Xaaedpscff1zvvPOOsrKydOutt+rWW2/Vu+++q2nTpum+++7TihUrNG3aNAUGBmrDhg3q06ePZs2apS5duigwsOHXeVuu5CUlJWnhwoWKiYnRjBkz9PHHH0uS8vLyFBwcfJhHo14xkgfAAlY376w3i1bL4XKosLJQ6/PXH/S+/jZ/NQ1pqvjQeMWFxik2NFZxoXG1Po8PjVdYQNhBn2NfbpdLrvx8uYqKJD8/2ex2+fn5y9/fTzY/PynQXza7XfL3l81v1zUP53a55XLvU1j3KYMHKqy7i+Tucuvaq3zWPJfLtev3OmTZPGBxPkSZrnWfOpTZgz+/9nsOT5ZfUiRJyt1RqnRXYc31otxyOR1upW8sPNhDJUmL18/RJ/Pe1A2nPaayTH+VZe65v8vt0r/Pf0MVVWVas/1PvffViwqoiFL7xOMkSYmB3aRASWVSnF+crh30mMZMuVJfzZyqod3/KUnqETdcPS4YXvOcX373qWyOQEW5U/T4ZyN1z/mvKb84S89/+IDGXjZJAX5somRSk1jvmPZ+991366yzzpIkjR07Vl26dNH69evVsWNHPffcc7rssst0++23S5LatWuniRMnauDAgXr99dcVExOj0NBQBQYGKiEhQVL1HiGS1LRp05prDc1yJe/222/X5ZdfrvDwcCUnJ9cMh/7000/q1q2b2XC+hjV5ALxclV+gHmwWI0fxljrd3+F2KKM0QxmlGYe8X1hAWO2RwAN8NAtpJn97dYHzj4mRKzxcjsxMVe3cKUdGhqoyM+XIyNz1+ww5MrPkyMiQu7KypgzWKn67f/X3k83P/7DXbP5+0sGu2e17HnPQa3u/7iGu2f32uq36mt3uJ7/dJXaf+9VcC9rnmv+uDLvuX3PtMJsteIL9iuAxlcgDjMjuM+K7d0ne9/61n1+qqIyV/SM/NUl2q33P+OqS6nTLvrZcsc3i1Kpr09oZ93r+X/76QR/99Lxuu3Ccjmvb74D5E0OS5Xa6lRzfThn5W/TDkik1JW9fQQEhSoxJVVbB9gPeXlxWoO/+mKTbz3lJaZl/Ky6yZc2Hy+VQZv42tWjaugH/S+JwvOHfoyR1775nzWjz5s0lSZmZmerYsaP++OMPrV+/XpMnT665j9td/QbTpk2bPGajR8v9FH7LLbeoT58+2rp1q4YPHy77rl2YWrduzZq8xsbumgC83OvdhmtdwfJ6f96SqhJtKtikTQWbDnofu82umOCYmtIXHxqv2JBYxTWJU3xCvOJC2ys2NFaRQZG1HufMz1dVRqYcmZlyZGaoKiOjugxmZu4phznpkq9MmbTZapdd/30K6e6Saz9Isd39GD8/yc9eu9jud636Ofe7ZvfbvwAf6JrfPsV21+f+tcquXbaaPLuuBR7g2t73O8ZR3hNePV4Zjr81fNTommt3jl+icy88V/+4tccBHzNlyhS9+fgT+uTTj3XeeefV6XUWXztJGzaU6qZXBx2w6JaXlevJr9J1/j/O1GV3nLhPSZVuuf163f6vf2nk5cP13Q8Vmr85UGfe0l1up1sPfCSdeF6qOrbtfOjSvG8Jdu4z6rx7NPkIpynvuX3XCLPTtdfz66DPbzU2u3eUvICAgJrf7y6mrl3bkrpcLt14440aPXr0fo9LTk5unIB1YLmSJ0knnHCCunfvrk2bNqlNmzby9/evGXJFI2K6JgAvtqJFN/23cJWx13e5Xcouy1Z2WbZW5Rw8R4h/SHX522s6aFxonOKS4hTXobXiQk9SXEicAvz2/NDirqqSIzu7uvTtGg10ZGXuKYQZGXJkZspVWtoYX2rDcrulqiq5q6pkvR+Zj8I+ZbdWid1ddve5doVsuuuNN5Ty5xIdHx+nKevXa/O6dTpn8xZt/b9b9cxvvyq9tFQTzjxLNj8/fbnmb/3r66/12Omnq/XCX7X890WSv59CgoLVJDxMNj9/TZg1U8elpio1IUFVLpdmrVihDz75RM9fd52KvvxSNn8/PfTe+zqj70lqmZCg7KIiPTd5sooK8nXlSb0UtHXVXiPD/prz26/amva3Jr0wTn4BRRrWt7NG/3udVqycqW3p6QoI8NPQU3soNCK8uuQHBHjFqNJRjey6DlFS933s3mtl9/p89+0HLKm1Cqmrdkndd+R4n+ePivOO6ZqH0qtXL61cuVJt27at82N2r8FzOp0NFWs/lit5paWluu222/T+++9LktauXavWrVtr9OjRSkxM1H333Wc4oQ9hJA+Al6rwD9aDMRFyFheYjnJYZY4ybSnaoi1FB59SapNN0cHRNWWwZmQwNFZxXeIU37uPYkNjFRsUXesHX2dxcXXhO8T0UEd2ttSIP7jgGDmdcjudUmVlnUvvMEn3NYvVhD8WK8vpVLvAQP2neaKi/vxTxZK279yh7VVVKvzmG0nSe1s2y+Fy6YHp0/XA9Ok1z3NekyZ6snmiJCk7K0t3TZ+uDIdDQTabWgcG6Zm4eJ06d552zp0nSdqwY7tGfv218pwOxfj7q0dwsCbHxUv3P6DNe+Urd7l0a1qaXkhMVNo/zq65/kBklEZdeaUCbTY9EZ+grbt2X6+x9yjvXtOba0ZY957yvHvK8L4jo/tOea51v0Nc87PvGQ2uy7W9pyr72atf199fdj8/+e095blmRHf3mt0DXPPz3++azd9ylaDB/Pvf/9ZJJ52k//u//9P111+vsLAwrV69WjNnztTLL798wMfExcUpJCREM2bMUMuWLRUcHFxrN8+GYLn/ovfff7/++usvzZ07V6efvmeHpmHDhunRRx+l5DUmu12STeK9UwBe5pVuQ7Uxv/6naZrillu55bnKLc/Vmrw1B71foD2wZrOY3YUwPjRecQlxim3dQvGhvRQXGqdg/z0bmbmdTjmycw4xPbT6c1dRUWN8qWggl0ZH69LoA59Xt7u47fZ+cqvDPt+/YmP1r9jYQ97nhcQWdcoWbLdreuv919pdGBWlC6OiDv5ARnlrO9BaXn//I7x26DW/4YMGqskZZ5j+So9J9+7dNW/ePD344IM65ZRT5Ha71aZNG1188cUHfYy/v78mTpyoxx57TI888ohOOeUUzZ07t0Fz2twW28e4VatW+uSTT3TSSScpIiJCf/31l1q3bq3169erV69eNbvboJE81lRyOUynAIA6W5rUU1cH5MnldpmO4rGaBDY54uMkXKWl1RvHZFSXwerf154eWpWVJVVVGfzKsK/FpaX6b26OVpZXKMvp0MTEFhoWEVGnx/5ZWqqrt25R26AgfZGyZ9v8z/Lz9b/CAq2vqJAkdQ4O1u3NYtU9ZM9UvjdzcjSruEgbKyoVbLfpuJAQ3RUbq9TAoJr7/Dc3R+/m5kqSrotpqqtj9hzP8FdZmR7PSNcnrVLk5wXTMn1F0+uvV9xdd5qO4RMsN5KXlZWluLi4/a6XlJR4xdxry7H5SaLkAfAO5QEheigySK4SCt6hFFYWHv1xEs3iFZccp9jQLvsdJ+F2u+XMzT3I9NA9nzvz8xvhq4Qklbpc6hAUrBGRUfrXjgPvankgRU6n7k/fqZNCw5TtrP1zwO+lpToroomOiwtRkM2md3Jzdf22rfoqJVXxuza8WFxaqkujotQ1OEROt1sTsrN03dat+jq1tULtdq2tKNcr2dl6rUVLuSXdsn2b+oWFqV1QkKrcbo3NSNfY+AQKnoex7bWhCRqW5Upe79699e233+q2226TtGdHnLfeekt9+/Y1Gc032f0klmoA8BITug3V5rxlpmNYwjEfJ5Eap7guHRUXOkBxu46T2M1VWVkzHbRm85h9poc6MjPl3jVShKM3IDxcA8LDj/hxYzLSdVaTJrJLml1cXOu25xJrT+98LCFBP6wv0q+lpTp31zqlN5OSat1nXEJznbxhvVaVl+uE0FBtqKhU+6AgnRRW/SZB+6AgbaisULugIP03N0cnhISqW4j3b/JhNbZASl5jsVzJe+qpp3T66adr1apVcjgcmjBhglauXKmFCxdq3rx5puP5HjZfAeAlFrc6XpPzrLMOz1s0znESB5kempkpZ06O7xwn0UimFeRra1WVnmmeqP/kZB/2/uVulxxutyIPccRD0a7t63ffp31QkNIqK7Vj13q6zZWVahcYpM2VlfqyoECfp6TUx5eCesZIXuOxXMnr16+ffvnlFz3//PNq06aNfvjhB/Xq1UsLFy7kMHQTOEYBgBcoDQzTw+F+cpfyw74navDjJLKy9qwXzMio3kCm1nTRTLmtcJxEI0irrNRLWVn6MLmV/Os4VfLFrCzF+furb2joAW93u916NjNTvUJC1C6oek1em6Ag3R4bq+u2bpUk3R4bqzZBQbpm6xbdFRun+SUlejU7W/42mx6Ii9cJB3luNC528Ww8lvyT7tatW80RCjCMkgfAC7zYZZC2WWg3TV9VX8dJxIXGKS649k6StY6TOMj0UF8/TsLpduvenTv0f82aKWXXuWCH805Ojr4tLNT7SckKstsPeJ8nMjO0pqJck/bZtfOSqGhdErXnv9MXBfkKs9t1XEiIztq0UZ+0SlGGo0p37dihma1bK/Agz49GxEheo7FEySssLFSTJk1qfn8ou++HRmK3xF8xABb2a2offZq/wnQMNJKjOU5i95ESu4+TiGvdUnGHPE7iYNNDrX2cRInLpRXl5VpdXq5xGdVrMV2qPkip25q/9VbLpJo1dFL17phv5ubonaQkdQgOPuBzPpGRrh+Li/VBUrISDlEQ8hwOvZ6Tow+SkrWsvEwpgYE1Hw65lVZVqfZBB34NNB57UNDh74R6YYmfwKOjo7Vz507FxcUpKirqgLtout1u2Wy2Rj1pHpKCIqTiQy+6BwBTSoIi9EiIU+4ypmmitkpXpbYXb9f24kPvKHmg4yTiQ+MV2y5Wccf1UHxovBIOd5zEgXYTzcr2uuMkwu12/W+voxIkaUp+nn4rLdX4xBZqsVdJeyc3R2/k5OitlknqGrz/Bilut1vjMjM0q7hY7yUlq+VhRgafyszUVdHRSggI0IryclXttc7S6XbLyT9xj2Dfq+SjYVmi5M2ZM0cxu85G+fHHHw2nQS0hMYe/DwAY8lyXAdrJZis4BkdznETNmsFdx0nEhXZVfGi8QgP2rBurdZxERoYcmVn7Tw9thOMkSlwubamsrPl8e1WVVpeXK9LPT4kBAXoxK1OZDoeebp4ou81Ws2Zutxg/PwXuc/2dnBxNzMnWc82bKzEgQFmO6iMWQu12he2aUvl4Zoa+LSzUKy1aKsxur7lPhN2u4H2mXS4oKdHmqko93by5JKlbcLA2VVbqp+JipTscsttsSq3j9FE0LHvYke/U6uvmzp2rwYMHKy8vT1FRUXV+nCVK3sCBAyVJDodDc+fO1TXXXKOkfbbehSGhTU0nAIADmt+mr6ZS8NAIjvg4iX2OlIhvHa/Yrgc5TqKiorrw7V4jeIDpocdynMTK8jKN3LW5iSQ9k5UpSTqvSRM92TxR2Q6Hdh7hiOOU/DxVud26fceOWtdvadpUtzaLlSR9vKu8Xr219vrKcQkJGhEZVfN5uculJzIy9EJidcmUpPiAAD0YF68H03cq0GbTUwnN9yuGMMP0SN7IkSP1/vvv66mnntJ9991Xc/3LL7/UiBEj5LbQTrs2t5W+GkkRERFavny5Utg61zN8+X/S0kmmUwBALYUhkRqR0kaZ5Yff3h3wJHabXU2Dm+45XH73YfMhe+0mGhanJoG19yDgOAl4gtT//U/BHdobe/2RI0fqk08+UXBwsDZu3Kjo6OqNe4615FVWViqwgUaLfXokb29Dhw7V3LlzNXLkSNNRIEmhTNcE4Hme6dRfmXlstgLv43K7lFWWpayyrDofJ1HrIylO8R3aKJbjJGCA6ZE8SRo2bJjWr1+vp556Ss8+++wB7zN16lQ98sgjWr9+vZo3b67bbrtNd911V83tKSkpuu6667R+/Xp98cUXOu+88zR48GDdfvvtmjRpku666y5t3bpVZ555pt5//319/vnnevTRR1VQUKArrrhC48ePl9+uMx8nTZqk8ePHa82aNQoLC9OQIUM0fvx4xcXFHdPXabmSd8YZZ+j+++/XihUrdPzxxytsn79M55xzjqFkPoqSB8DDzG17sr6i4MHijuQ4id0jgbVGBrvEcpwE6p1fuPmS5+fnpyeffFKXXXaZRo8erZYtW9a6/Y8//tBFF12kMWPG6OKLL9aCBQt0yy23qGnTprUGkZ577jk9/PDDeuihhyRJ8+fPV2lpqSZOnKiPP/5YRUVFOv/883X++ecrKipK06dP18aNG3XBBRfo5JNP1sUXXyypehTw8ccfV4cOHZSZmak77rhDI0eO1PTp04/p67TcdE37IeZcs7umAX9+IH11m+kUACBJKgiN1nnJrZRdkWs6CuA19j1OYr81g7vOGeQ4CRxOxxXLjR6IPnLkSOXn5+vLL79U37591blzZ73zzju1pmtefvnlysrK0g8//FDzuHvvvVfffvutVq5cKal6JK9nz5764osvau7z3nvvadSoUVq/fr3atGkjSbrpppv04YcfKiMjQ+Hh1ZvOnH766UpJSdF//vOfA2ZctGiR+vTpo6KiIoWHhzNdczeXy2U6AvbGxisAPMi4jn2VzSgecESO9DiJmsPld48Mto9TrI8dJ4H92UNDjRa8fT3zzDMaMmRIrWmYkrR69Wqde+65ta71799f48ePl9PprJlmecIJJ+z3nKGhoTUFT5Li4+OVkpJSU/B2X8vMzKz5fMmSJRozZoyWLl2q3Nzcmi6zZcsWde7c+ai/Ps/5k4Y1UfIAeIiZ7QfoOwoe0GDqepxEs9BmtUYC40LjFNcszuOPk8CxsUdFmo5Qy4ABA3TaaafpgQceqDUNc/fZ2ns70MTHfZeESVLAXmdBStWzCA90bXeRKykp0amnnqpTTz1VkyZNUmxsrLZs2aLTTjtNlXsdXXI0LFny5s2bp+eff16rV6+WzWZTp06ddM899+iUU04xHc33cE4eAA+QG9ZMT9jzTccAfJ7D7VB6SbrSS9IPeb+9j5OID43fs2awdbziunZSbOgAxYfEys/uV/OYhj5OAsfGr4lnlTxJevrpp3Xcccepffs9O3527txZ8+fPr3W/BQsWqH379jWjePXl77//VnZ2tp5++uma498WL15cL89tuZI3adIkjRo1Sueff75Gjx4tt9utBQsWaOjQoXrvvfd02WWXmY7oWxjJA+ABnujQW7l5K03HAFBHJVUl2lSwSZsKNh30PruPk6g5XH73ERKRcYpLiFNcaPvDHydxoOmhmVkcJ9EA/Jo0OfydGlm3bt10+eWX6+WXX665dtddd6l37956/PHHdfHFF2vhwoV65ZVX9Nprr9X76ycnJyswMFAvv/yybrrpJq1YsUKPP/54vTy35UreuHHj9Oyzz+qOO+6oufavf/1LL774oh5//HFKXmMLiZZsdsnNWkkAZnzXcZBmUvAAy9n7OAnlHPx++x4nUbNmMClO8R3aKi60n2JDYg94nMTe00MdmfvuJspxEkfCL9LzSp4kPf744/r0009rPu/Vq5c+/fRTPfLII3r88cfVvHlzPfbYYw1yPFtsbKzee+89PfDAA5o4caJ69eql559/vl5OA7Dc7ppBQUFauXKl2rZtW+v6+vXr1bVrV5WXlxtK5sOeSZXK2MkOQOPLjojXeYmxKqgsNB0FgAfb+ziJ/Q6X3+sjet/jJIqKDj09NCNDjpwcjpOQFHnB+UocN850DJ9huZG8pKQkzZ49e7+SN3v27Jq5rmhkoTGUPABGjG3bUwX5Bz8wGgAkyS23cstzlVueq79z/z7o/Q50nER8aLxiE2IV17ql4kOPV1xYnIL8gvY8997HSewzPXTvw+atfpyEfwxLeBqT5UreXXfdpdGjR2vp0qXq16+fbDab5s+fr/fee08TJkwwHc83hTaVcg6+0xYANISvOg3VXAoegHp0NMdJ1Foz2CFOcT2PU1xo3AGPk6iZHpp5kMPmvfg4Cf/YZqYj+BTLlbybb75ZCQkJeuGFF2rm13bq1EmffPLJfmdeoJGw+QqARpYRmainnTtNxwDgo47oOIl9Dpc/ouMkdpXAWtNDMzM98jgJ/9hY0xF8iuXW5MED/e//pCWTTKcA4ENu6nmqfsk/+JQrAPAW4QHhtQ+X32fNYGxorGIPdpzE7jWCB5ge2tjHSbSa9KFCD3CAOBqG5Uby4IE4Kw9AI5raeRgFD4BlFFcVq7ig+OiOk4iKU1zzgx8n4cjL22d66P6HzTtzc+vlOAlG8hqX5UpedHT0fqfUS9WnywcHB6tt27YaOXKkRo0aZSCdj2K6JoBGsjM6Sc87Dr1WBgCspq7HScy8cKYSwhJqPvePjpZ/dLTUof1BH1PrOIndawQz991N9PDHSVDyGpflSt4jjzyicePG6YwzzlCfPn3kdru1aNEizZgxQ//3f/+nTZs26eabb5bD4dD1119vOq5vCOMfNYCG55ZND6d0UHH+WtNRAMDj+Nn8FBty5D+T2QICFJCYqIDExEPez1lUdIDpodXrBV35BbKHhh7y8ahflit58+fP1xNPPKGbbrqp1vU33nhDP/zwg6ZOnaru3btr4sSJlLzGEp1iOgEAH/Bp1+H6jWmaAHBAsaG11+3VN7+ICPlFRChon2PMYIb98HfxLt9//72GDRu23/WhQ4fq+++/lySdeeaZ2rhxY2NH810xrU0nAGBx22KS9ULFZtMxAMBjNQ9rbjoCGpHlSl5MTIy+/vrr/a5//fXXiomp3gCkpKREERERjR3NdzVpLgUwRA+gYbhl08PJ7VTmKDMdBQA8VkJowuHvBMuw3HTNhx9+WDfffLN+/PFH9enTRzabTb///rumT5+u//znP5KkmTNnauDAgYaT+piY1lLGCtMpAFjQ5G6nanHBatMxAMCjJYRT8nyJ5Ure9ddfr86dO+uVV17RtGnT5Ha71bFjR82bN0/9+vWTJN11112GU/qgmFRKHoB6t7lZa00sO/i24gCAaozk+RbLlTxJ6t+/v/r37286BvbGujwA9cxls+uhFq1UVrjBdBQA8HisyfMtlluTJ0kbNmzQQw89pMsuu0yZmZmSpBkzZmjlypWGk/mwmDamEwCwmA+6naalFDwAqJO9z8eD9Vmu5M2bN0/dunXTb7/9pqlTp6q4uFiStGzZMj366KOG0/kwRvIA1KONce30Sul60zEAwGswkudbLFfy7rvvPj3xxBOaOXOmAgMDa64PHjxYCxcuNJjMx1HyANQTp81PDzVvoQpnhekoAOAVIoMiFRUcZToGGpHlSt7y5cs1YsSI/a7HxsYqJyfHQCJIkpokSv4hplMAsID/dj9Vyws56xQA6qpNJMtmfI3lSl5UVJR27ty53/UlS5aoRYsWBhJBkmSzVe+wCQDHYG18R71evNZ0DADwKu2i25mOgEZmuZJ32WWX6d///rfS09Nls9nkcrn0yy+/6O6779ZVV11lOp5vY8omgGPgsPvrofhYVbmqTEcBAK/SNqqt6QhoZJYreePGjVNycrJatGih4uJide7cWQMGDFC/fv300EMPmY7n2yh5AI7BW91P0+qizaZjAIDXaRPFdE1fY3O73W7TIRrCxo0b9eeff8rlcqlnz55q145hauMWvyt9c7vpFAC80OrmnXVZaLkcLofpKADgdX6++Gc2XvExlhvJe+yxx1RaWqrWrVvrwgsv1EUXXaR27dqprKxMjz32mOl4vq0p7yIBOHJVfoF6sFkMBQ8AjkLT4KYUPB9kuZI3duzYmrPx9lZaWqqxY8caSIQaTNcEcBRe7zZc64q3mI4BAF6pbTTr8XyR5Uqe2+2WzWbb7/pff/2lmJgYA4lQo0kLyT/YdAoAXmRFi276b+Eq0zEAwGu1i2LJki/yNx2gvkRHR8tms8lms6l9+/a1ip7T6VRxcbFuuukmgwkhm02KTpWyVptOAsALVPgH68GYCDmLC0xHAQCvxaYrvskyJW/8+PFyu9265pprNHbsWEVGRtbcFhgYqJSUFPXt29dgQkiqnrJJyQNQB690G6qN+ctNxwAAr8bxCb7JMiXv6quvliSlpqaqX79+CggIMJwIBxTXUVrzrekUADzc0qSe+qBgpekYAOD1KHm+yTIlb7eBAwfW/L6srExVVbUPzW3SpEljR8LeEnuZTgDAw5UHhOihyCC5SlymowCAV0sIS1B4YLjpGDDAchuvlJaW6tZbb1VcXJzCw8MVHR1d6wOGtTjedAIAHm5Ct6HaXLLDdAwA8HqM4vkuy5W8e+65R3PmzNFrr72moKAgvf322xo7dqwSExP1wQcfmI6HJs2liOamUwDwUItbHa/JeazDA4D6QMnzXZYreV9//bVee+01XXjhhfL399cpp5yihx56SE8++aQmT55sOh4kRvMAHFBpYJgeDveTW27TUQDAEih5vstyJS83N1epqamSqtff5ebmSpJOPvlk/fTTTyajYbfEnqYTAPBAL3YZpG2l6aZjAIBlcBC677JcyWvdurXS0tIkSZ07d9ann34qqXqELyoqylww7NGCzVcA1PZrah99mr/CdAwAsAy7za7Wka1Nx4Ahlit5o0aN0l9//SVJuv/++2vW5t1xxx265557DKeDpF07bNoOezcAvqEkKEKPhDiZpgkA9ahFeAuF+IeYjgFDbG6329L/V92yZYsWL16sNm3aqEePHqbjYLeJvaTcDaZTAPAAY3qdpalstgIA9Wpo8lCNHzzedAwYYrlz8vaVnJys5ORk0zGwrxaUPADS/DZ9KXgA0AB6xrEHgi+zzHTNOXPmqHPnziosLNzvtoKCAnXp0kU///yzgWQ4IHbYBHxeYUikHg2sMB0DACypVxx7IPgyy5S88ePH6/rrr1eTJk32uy0yMlI33nijXnzxRQPJcECJfOMBfN0znforszzbdAwAsJwQ/xB1atrJdAwYZJmS99dff+n0008/6O2nnnqq/vjjj0ZMhENq3l2yW362MICDmNv2ZH2Vx26aANAQujfrLn9+zvJplil5GRkZCggIOOjt/v7+ysrKasREOKSAECmOd5gAX1QQGq2x/sWmYwCAZfWKZ8aUr7NMyWvRooWWLz/44v1ly5apefPmjZgIh8WUTcAnjevYV9kVuaZjAIBlUfJgmZJ35pln6pFHHlF5efl+t5WVlenRRx/VP/7xDwPJcFBsvgL4nJntB+g7pmkCQIPxt/mre7PupmPAMMuck5eRkaFevXrJz89Pt956qzp06CCbzabVq1fr1VdfldPp1J9//qn4+HjTUbFb+nLpPyebTgGgkeSGNdOIpETlVuSbjgIAltW1aVdN+ccU0zFgmGVWZMbHx2vBggW6+eabdf/992t3d7XZbDrttNP02muvUfA8TWwnKSBUqio1nQRAI3iiQ2/l5q00HQMALI2pmpAsVPIkqVWrVpo+fbry8vK0fv16ud1utWvXTtHR0aaj4UD8/KWE7tLWX00nAdDAvus4SDMpeADQ4Ch5kCxW8naLjo5W7969TcdAXbQ8gZIHWFx2RLzGuTkPDwAaA4egQ7LQxivwUq0Hm04AoIGNbdtTBZWFpmMAgOWlRqYqOpgZbKDkwbSU/pJ/sOkUABrIV52GaG7+KtMxAMAnMIqH3Sh5MCsgREruazoFgAaQEZmop53ppmMAgM9gPR52o+TBvDZDTCcA0AAebd1VRVXFpmMAgM9gJA+7UfJgXtuhphMAqGdTOw/TL/l/m44BAD4jLjROLSNamo4BD0HJg3nxXaSI5qZTAKgnO6OT9Lxju+kYAOBTjo873nQEeBBKHjwDu2wCluCWTQ+ndFBxVYnpKADgU3rG9zQdAR6EkgfPwJRNwBI+7Tpcv+WvNR0DAHwO6/GwN0oePEPrwZKNv46AN9sWk6wXKjabjgEAPic+NF4dYjqYjgEPwk/V8AxhTaXmPUynAHCU3LLp4eR2KnOUmY4CAD5nUNIg0xHgYSh58BwcpQB4rcndTtXignWmYwCATxqcxN4GqI2SB8/RhnV5gDfa3Ky1JpZtMh0DAHxSWECY+iT0MR0DHoaSB8+R1EcKjDCdAsARcNnseqhFK5U5y01HAQCf1D+xvwL8AkzHgIeh5MFz+AVIqaeYTgHgCHzQ7TQtLdxgOgYA+CzW4+FAKHnwLKzLA7zGxrh2eqV0vekYAOCz/G3+GtBygOkY8ECUPHgWzssDvILT5qeHmrdQhbPCdBQA8Fk943sqMijSdAx4IEoePEtMayk6xXQKAIfx3+6nannhRtMxAMCnsasmDoaSB8/DLpuAR1sb31GvF681HQMAfB7r8XAwlDx4nnbDTScAcBAOu78eio9VlavKdBQA8Glto9oqKSLJdAx4KEoePE+bIRLzywGP9Fb307S6aLPpGADg85iqiUOh5MHz+AdJnc42nQLAPlY376w3i1abjgEAECUPh0bJg2fqer7pBAD2UuUXqAebxcjhcpiOAgA+LzYkVl2bdTUdAx6MkgfP1HqQFBZrOgWAXV7vNlzrireYjgEAkDQwaaBsNpvpGPBglDx4Jruf1Pk80ykASFrRopv+W7jKdAwAwC5M1cThUPLgubpdaDoB4PMq/YL0YEyEnG6n6SgAAEkh/iE6sfmJpmPAw1Hy4LmSTpQi2RoYMOmV7sO0sXib6RgAgF36JfZTkF+Q6RjwcJQ8eC6bTeoywnQKwGctTTpO7xesNB0DALCXIclDTEeAF6DkwbMxZRMwojwgRA9HBsvldpmOAgDYJcQ/RMOSh5mOAS9AyYNna95DatrOdArA50zoNlRpJTtMxwAA7GV4q+EKDQg1HQNegJIHz8doHtCoFrc6XpPzlpuOAQDYx3ltzzMdAV6CkgfP15WSBzSW0sAwPRzuJ7fcpqMAAPbSIryFTog/wXQMeAlKHjxfs7ZSQnfTKQCf8GKXQdpWmm46BgBgH+e2PZcD0FFnlDx4B6ZsAg3u19Q++jR/hekYAIB92GTTuW3ONR0DXoSSB+/Q9QJJvHsFNJSSoAg9EuJkmiYAeKA+CX2UGJ5oOga8CCUP3iGypZR8kukUgGU912WAdpZlmY4BADiAc9syiocjQ8mD9+h6gekEgCXNb9NXU9lNEwA8UnhAuIa3Gm46BrwMJQ/eo/N5ks3PdArAUgpDIvVoYIXpGACAgzgt5TQF+webjgEvQ8mD9wiPldrxThZQn57p1F+Z5dmmYwAADoKz8XA0KHnwLr2vM50AsIy5bU/WV3nspgkAniqlSYqOizvOdAx4IUoevEvbYVJMa9MpAK9XEBqtsf7FpmMAAA6BDVdwtCh58C42G6N5QD0Y17GvsityTccAAByEn81P57Q5x3QMeClKHrzPcZdLAaGmUwBea2b7AfqOaZoA4NH6JvZVXGic6RjwUpQ8eJ+QKKnbP02nALxSblgzPWHPNx0DAHAYTNXEsaDkwTv1ucF0AsArPdGht3Ir8k3HAAAcQpPAJhqSNMR0DHgxSh68U0JXKbmv6RSAV/mu4yDNzFtpOgYA4DDOTD1TgX6BpmPAi1Hy4L36XG86AeA1siPiNc7NeXgA4A3+2YFlKTg2lDx4r07nSBHNTacAvMLYtj1VUFloOgYA4DD6JfZT++j2pmPAy1Hy4L38AqTjR5pOAXi8rzoN0dz8VaZjAADq4OouV5uOAAug5MG7HT9SsgeYTgF4rIzIRD3tTDcdAwBQBx2iO6hfYj/TMWABlDx4t4gEqdPZplMAHmtM664qqio2HQMAUAeM4qG+UPLg/ThOATigaZ2HaX7+36ZjAADqID40XmeknmE6BiyCkgfv16qvFN/NdArAo+yMTtJzju2mYwAA6uiKTlfI3+5vOgYsgpIHa+hznekEgMdwy6ZHUjqquKrEdBQAQB2EB4TrwvYXmo4BC6HkwRq6XSQFR5lOAXiET7sO16/5a0zHAADU0QXtLlB4YLjpGLAQSh6sITBU6nmF6RSAcdtikvVCxWbTMQAAdeRv99cVnfkZBvWLkgfr6H2tZPMznQIwxi2bHk5upzJHmekoAIA6Oi3lNCWEJZiOAYuh5ME6YlpLXUaYTgEYM7nbqVpcsM50DADAERjZZaTpCLAgSh6sZcDdkmymUwCNbnOz1ppYtsl0DADAETix+YnqGNPRdAxYECUP1hLXSep4lukUQKNy2ex6qEUrlTnLTUcBAByBUV1GmY4Ai6LkwXoG3GM6AdCoPuh2mpYWbjAdAwBwBNpFt1P/Fv1Nx4BFUfJgPYnHSe1ONZ0CaBQb49rpldL1pmMAAI7Q1Z2vNh0BFkbJgzUNuNd0AqDBOW1+eqh5C1U4K0xHAQAcgbjQOJ3Z+kzTMWBhlDxYU1JvKXWg6RRAg/pv91O1vHCj6RgAgCN0WcfLFGAPMB0DFkbJg3WxNg8Wtja+o14vXms6BgDgCEUGReqiDheZjgGLo+TBulJPkZL7mU4B1DuH3V8PxceqylVlOgoA4Ahd3+16RQRGmI4Bi6PkwdoGP2A6AVDv3up+mlYXbTYdAwBwhBLCEnRJx0tMx4APoOTB2lJPkVIHmE4B1JvVzTvrzaLVpmMAAI7CLT1uUZBfkOkY8AGUPFjf4IdMJwDqRZVfoB5sFiOHy2E6CgDgCLWObK1z2pxjOgZ8BCUP1pd8otR2mOkUwDF7vdtwrSveYjoGAOAojO45Wn52P9Mx4CMoefANgx80nQA4JitadNN/C1eZjgEAOArdm3XX0FZDTceAD6HkwTe06CV1OMt0CuCoVPoF6cGYCDndTtNRAABH4fbjbzcdAT6GkgffMfgBSTbTKYAj9kr3YdpYvM10DADAUeif2F+9E3qbjgEfQ8mD70joKnU+13QK4IgsTTpO7xesNB0DAHAUbLIxigcjKHnwLYMflOz+plMAdVIeEKKHI4PlcrtMRwEAHIXTU09Xx5iOpmPAB1Hy4Fti20u9rzedAqiTCV2HKK1kh+kYAICj4G/3123H3WY6BnwUJQ++Z9B9Umgz0ymAQ1rc6nhNzl9hOgYA4Chd0O4CJTVJMh0DPoqSB98TEiUNfdh0CuCgSgPD9HC4n9xym44CADgKIf4huqnHTaZjwIdR8uCbel4lNe9hOgVwQC92HaRtpemmYwAAjtIVna5QsxBmDcEcSh58k90unf6M6RTAfn5N7aNP85imCQDeKiooSqO6jjIdAz6Okgff1aqv1PVC0ymAGiVBEXokxMk0TQDwYtd1u04RgRGmY8DHUfLg24Y/JgWEmk4BSJKe6zJAO8uyTMcAAByl5mHNdUnHS0zHACh58HGRLaST7zSdAtD8Nn01NW+56RgAgGNwX5/7FOQXZDoGQMkD1O82KaqV6RTwYYUhkXo0sMJ0DADAMRiUNEhDkoeYjgFIouQBUkCwdOoTplPAhz3Tqb8yy7NNxwAAHKUQ/xA90OcB0zGAGpQ8QJI6nyOlDjSdAj5obtuT9RW7aQKAV7ulxy1qHt7cdAygBiUP2O2MZyS7v+kU8CEFodEa619sOgYA4Bi0j26vKzpfYToGUAslD9gtrpN0wrWmU8CHjOvYV9kVuaZjAACOkk02PdL3EfnzJjE8DCUP2Nvg+6XQpqZTwAfMbD9A3zFNEwC82oXtL1SP2B6mYwD7sbndbk7dBfa26B3pW45VQMPJDWumEUmJyq3INx0FPiRnTo5y5+SqKrtKkhTUIkhx58YponuE3A63MqZlqGhZkSozK+UX6qfwzuGK/2e8AqIDDvqcG5/aqNI1pftdD+8erpQ7UyRJbqdbmV9mKn9hvhwFDvlH+Sv65GjFnh0rm90mSSpYXKC8uXkqSyuTs9ipNmPbKKRVSK3n3Dllp/Ln58seZFf8RfGKOimq5raC3wuU/0u+Wt3BTsloPDHBMfp6xNdqEtjEdBRgP4wtA/s6fpS0+F0pgzPL0DCe6NBbuXkrTceAjwmIDlDCPxMUGB8oScqfn68tE7aozWNtFBAdoLLNZYo7J07BScFylji186Od2jxhs9qOaXvQ50y+LVlux573ip0lTq1/eL0ie0fWXMv6Nku5P+aq5XUtFdQiSGVpZdr+znbZQ+xqdmozSZKrwqXQdqFq0ruJdry7Y7/XKVxSqIKFBUq5O0UVGRXa/s52hXcNl3+4v5wlTmVMzVDKvSn19CcF1M09ve+h4MFjMV0T2JfdLp35rCSb6SSwoO86DtJMCh4MaNKziSJ6RCgoIUhBCUGKvzBe9mC7SteXyi/UT6n3pCqyT6SCmgcptG2oml/RXOVp5arMqTzoc/qH+ysgKqDmo3hFseyBdkX22VPyyjaUKaJnhCKOi1BgbKAie0cqvEu4yjaV1dwnun+04s6NU3jn8AO+TsXOCoV1DFNIaoiiToqSPcSuyszqXOmfpitmSIwCmwbW058UcHgnNj9R/2j9D9MxgIOi5AEH0qqf1Od60ylgMdkR8Rrn5jw8mOd2uZX/a371CFrb0APex1XmkmySX6hfnZ837+c8RZ4YKXvQnh8vQtuFqmRViSrSKyRJZVvKVLKuRBE9Iur8vMFJwdVTOUucKksrk7vSraD4IJWsLVHZ5jI1Hc5aajSeQHugHjrxIdMxgENiuiZwMMPGSutmSnmbTCeBRYxt21MF+atMx4APK99aro1PbJSryiV7kF3JtyUruEXwfvdzVbqU/lm6Ik+KlF9I3Upe6cZSVWyrUItrWtS63uysZnKWObXu/nXVby27pPgLaq+pO5yIbhEq7VuqDWM3yBZoU8vrW8oWZNOOD3ao5XUtlTsnVzmzcuQf7q/EUYkH/JqA+nJtt2uVEpliOgZwSJQ84GACQ6XzXpfeO1Nyu0yngZf7qtMQzaXgwbDA5oFq81gbuUpdKlhcoG1vb1Pqfam1SpHb4dbW17dKbinxqsQ6P3feT3kKahmk0Na1RwYLfitQ/sJ8tbyxpYJbBKtsS5nSP0qv2YClruJHxCt+RHzN5xlfZCi8c7hsfjZlfZWltk+0VdFfRdr25ja1HXvwdYTAsUhpkqLrul1nOgZwWEzXBA6lVV/pxJtMp4CXy4hM1NPOdNMxANn97QqKD1JIaogS/pmg4KRg5czMqbnd7XBry2tbVJVdpZR7Uuo8iueqcKngtwJFD9i/tKV/mq7YM2MVdVKUgpOCFd0/Wk1Pa6qsb7KO+uuo2FGhgl8LFHd+nEr+LlFoh1D5N/FXZJ9IlW8ul7PMedTPDRzKgyc9qEA/1n/C81HygMMZ+ojUlHeFcfTGtO6qoqpi0zGA/bkld1X17pi7C15lRqVS7kmRf3jdJ/sU/F4gd5VbUf2i9n+JCvd+P23Y7DbpKA9wcrvd2v7ediVckiC/YD+5XW65nXu+BkkSky/QAM5qfZZOan6S6RhAnVDygMMJCKmetmnjnwuO3LTOwzQ//2/TMQClf56ukjUlqsyqVPnWcmV8nqGSv0sU1TdKbqdbW17dorK0MrW8saXcLreq8qtUlV8ll2NPY9r25jalf7b/qHTez3lq0qvJAYthxHERyvo6S0VLi1SZVanCPwqV/X22mhy/Z+t5R7FDZZvLVLGjenOWyvRKlW0uU1V+1f6vNS9P/k381aRn9eND24WqZHWJSteXKvuHbAUlBskvrO6bxQB1EREYoXtOuMd0DKDOWJMH1EVSH6nvrdKCiaaTwIvsjE7Sc47tpmMAkiRHgUPb3twmR4FD9hC7gpOClXJXisK7hqsyq1JFS4okSRse2VDrcSn/TlF4p+qjDSpzKvc7XaYivUKla0uVcnfKAV+3+RXNlTktUzs+3CFHYfVh6DGDYhR7bmzNfYqWFGn7O3v+rWx9faskKfbc2Frr8BwFDmV9naXWD7WuuRbaOlTNTm+mzS9tln8Tf7W4vvbGL0B9uPP4O9U0hF1c4T1sbrf7KCdMAD7GUSH95xQpe43pJPACbtl0Q89h+jWfvy8A4M1OaXGKXhv2mukYwBFh/hlQV/5B0ojXJRvTgHB4n3YdTsED/r+9O4+rqk78P/6+l8sOCqgsEiAoyCKo5BqFC5rmMu6W2iRlfmuatHS0TUVNU8s0M8vRmsR0qpnMzCY109LJ9Oto30xNUidBMvd9RVnu7w/H+4txV+ADl9fz8eABnnvOPe9zvdl98/mcc4AKzt/dXy+mvGg6BnDTKHnAzQi9U0p5ynQKlHN7AsI19Xyu6RgAgNs0uvloVfesbjoGcNMoecDNavm8FBhvOgXKKbssGhUerbMFZ01HAQDchi61uygtIs10DOCWUPKAm2Vzu3i1TSvXLcLl3k9sp40ndpqOAQC4DaE+oXq+6fOmYwC3jJIH3IqaDaS7h5pOgXJmd/UovX5ul+kYAIDbYLVYNeHuCfJ29TYdBbhllDzgVrV4RgpONJ0C5USRxaqRoRE6V5hnOgoA4DakJ6QrOSjZdAzgtlDygFvl4ip1/bNkdTWdBOXAe4nttOnkz9dfEQBQbsUGxOrJBk+ajgHcNkoecDuC60ktnjWdAobtCozWjLP/Nh0DAHAbPG2eeiX1Fbm68MtbVHyUPOB23TNUirjbdAoYUmhx0ciQUJ0vPG86CgDgNjzf5HlFVo00HQMoEZQ84HZZXaSe70o+QaaTwIB3k+7VlpNcbAUAKrL7at2nbtHdTMcASozFbrfbTYcAnELOt9J7v5OKCkwnQRnZERSrB3zylV+UbzoKAOAWhfqE6qPOH8nXzdd0FKDEMJIHlJRaKVJahukUKCMFVptGBtWg4AFABWaz2PRy6ssUPDgdSh5QklKekmI7mU6BMvB2UjtlndptOgYA4Db8seEfVb9GfdMxgBJHyQNKWte3pIAo0ylQirJC4jX7VJbpGACA29A0pKkeqfeI6RhAqaDkASXNo6rU+z3J5mk6CUpBvoubRlQPUAHnXgJAhRXgEaCJd0+U1cJHYTgn3tlAaQhOlDq+ajoFSsHMxLbaeTrXdAwAwC2yWW2a0mKKanjVMB0FKDWUPKC0NHxQavh70ylQgraGJurdk9tMxwAA3IbnmzyvRsGNTMcAShUlDyhNHV6VgpNMp0AJuODirhEBviq0F5qOAgC4Rb1ieql33d6mYwCljpIHlCZXD+n+eRfP00OFNiOpjXad3mM6BgDgFt0ZdKeeb/q86RhAmaDkAaXNv5bUbZYki+kkuEWbwhpo7okfTccAANyiEO8QTW05Va5WV9NRgDJByQPKQt37pLufNp0CtyDP1VOjqnqoyF5kOgoA4BZ42jw1vfV0BXgEmI4ClBlKHlBWWo+Sat1jOgVu0uv1WivnzF7TMQAAt+jFlBcVGxBrOgZQpih5QFmxukg935V8Q0wnwQ3aGHGn/np8q+kYAIBbNDBxoNrXam86BlDmKHlAWfIJlHrOkTgnoNw76+atUT4usstuOgoA4Ba0vKOlBjUcZDoGYAQlDyhrEc2lLjNMp8B1TK3XUnvO7jcdAwBwC2pXra1JqZNksXDRM1ROlDzAhPoPSK1Hmk6Bq/jfyCb6+zGmaQJARVTFrYqmt54ub1dv01EAYyh5gCmpw6Xk/qZT4L+ccffVaM8ipmkCQAXkYnHR5NTJCq8SbjoKYBQlDzCp41SpTlvTKfAbkxNStffcQdMxAAC3YMidQ3RX6F2mYwDGUfIAk1xsUq9MKaS+6SSQtKZ2c318bIvpGACAW/C72r9T/wRmyAASJQ8wz91H6vuRVJWpJSad9Kyq0W7nTccAANyCxOqJGt18tOkYQLlByQPKA98g6cEFkoef6SSV1stxKTqYd9h0DADATbrD5w5Nbz1dbi5upqMA5QYlDygvatSVHnhfcnE3naTSWVXnbi3mapoAUOHU8Kyh2ffOVnXP6qajAOUKJQ8oT2qlSF3fksR9fcrKCS9/jbWdNh0DAHCTqrhV0ay2sxTmG2Y6ClDuUPKA8iaxp9RmjOkUlcZLsc11+PxR0zEAADfB0+apN9PeVLR/tOkoQLlEyQPKo7uflho/ajqF0/syJlVLmaYJABWKq9VV01pOU4PABqajAOUWJQ8or+57RarbwXQKp3XUu7rGW4+bjgEAuAlWi1UT75nIvfCA66DkAeWV1UXq8Rcp9E7TSZzS+LqNdfT8cdMxAAA3YVSzUWpXq53pGEC5R8kDyjM3L6nP3yT/WqaTOJWlsS315bEfTccAANyEp5OfVs+YnqZjABUCJQ8o73xqSA8ulLxrmE7iFA77BuklO/fDA4CK5OF6D2tA4gDTMYAKg5IHVATVakv9/0HRKwFj6zTUiQsnTccAANygHtE9NPTOoaZjABUKJQ+oKAJjKXq3aXFca606vs10DADADWob0VYZzTNMxwAqHIvdbrebDgHgJhzMkuZ2ls4cMp2kQjlQtaa6BfnpVD43PgeAiqB5SHO9mfamXF1cTUcBKhxG8oCKJjBO6v8ZI3o3aUxUPQoeAFQQSTWSNK3VNAoecIsoeUBFRNG7KQvj22jN8Z9MxwAA3IA6fnX0Vtpb8nL1Mh0FqLAoeUBFdanoeVU3naRc2+cfpskFv5qOAQC4AWG+YZrddraqulc1HQWo0Ch5QEUWGCel/4OidxV2WZRRK1an88+YjgIAuI7aVWsrs32mangxSwW4XZQ8oKJjRO+q/l6vrf73+HbTMQAA1xEXEKc57eco0CvQdBTAKXB1TcBZHNh28aqbZ7nRtyTtCQhXj2peOltw1nQUAMA1NAxsqDfT3pSvm6/pKIDTYCQPcBZB8VL/xYzo6eI0zVHh0RQ8ACjnmoU006y2syh4QAmj5AHOJCiBoifp/cR22nhip+kYAIBraB3WWm+mvSlPm6fpKIDTYbom4IwO/PifqZtHTCcpc7urR6mXn03nCvNMRwEAXEXHqI4anzJeNqvNdBTAKTGSBzijoIT/XIylmukkZarIYtXI0AgKHgCUY71iemnC3RMoeEApouQBzupS0fMJMp2kzLyXeK82nfzZdAwAwFWkJ6Qro3mGrBY+ggKliemagLM7tlv6a0/p8A7TSUrVrsBo9a4inS88bzoKAOAKnmjwhP5Q/w+mYwCVAr9GAZydf4Q0YLkU3tx0klJTaHHRyJBQCh4AlFPDGw2n4AFliJIHVAae/tJDn0rxXU0nKRXvJt2rLSd3mY4BAPgvVotVo5uP1kMJD5mOAlQqTNcEKhO7XVo+Ulo3w3SSErMjKFYP+OQrvyjfdBQAwG/YLDa9dPdL6hDVwXQUoNJhJA+oTCwWqd1LUvuXJSc46b3AatPIoBoUPAAoZ9ysbpracioFDzCk4n/KA3Dzmj0u9Zor2TxMJ7ktbye1U9ap3aZjAAB+o6p7Vc1sM1OtwluZjgJUWkzXBCqz3PXSBw9I546aTnLTskLi1dcrTwVFBaajAAD+I7JqpGa0nqHwKuGmowCVGiN5QGUW3lQa8KXkX8t0kpuS7+KmEdUDKHgAUI7cVfMu/bXDXyl4QDlAyQMqu+p1pAErpJrJppPcsJmJbbXzdK7pGACA/+gb21dvpb0lXzdf01EAiOmaAC65cFZa8LC0Y5npJNe0NTRRD7qfVqG90HQUAKj0bBabnm/6vHrX7W06CoDfYCQPwEVuXtID70t3Pmw6yVVdcHHXiABfCh4AlANV3atqVttZFDygHKLkAfj/rC5S52lSWoYki+k0l5mR1Ea7Tu8xHQMAKr3IqpF6v8P7ahLSxHQUAFfAdE0AV7b579LiQVJBnukkkqRNYQ3U3/W4iuxFpqMAQKWWUjNFk1tM5vw7oBxjJA/AlSX1lgYsl/wiTCdRnqunRlX1oOABgGH94vrpzbQ3KXhAOcdIHoBrO3dc+uQxoxdkeblhR80/vsXY/gGgsrNZbXqh6QvqFdPLdBQAN4CSB+D67HbpmynS1xOkMr7oycaIO/WI9bDs4p8qADChqntVvdbyNTUObmw6CoAbRMkDcON2rZIWDJDOHi6T3Z1181aP6ATtObu/TPYHACguqmqUZrSeobAqYaajALgJnJMH4MZFtZQe/0YKa1omu5taryUFDwAMSQlN0fwO8yl4QAXESB6Am1eYLy0fKa3/c6nt4n8jm+h/dIBpmgBQxlwsLvpD/T9oYNJAWS2MBwAVESUPwK3b+rG0eLB04XSJPu0Zd191r11Xe88dLNHnBQBcW03vmno59WU1CGxgOgqA28CvZwDcuno9pIFfS9XrlujTTk5IpeABQBlrG9FWH/3uIwoe4AQYyQNw+86flj4bfHFk7zatqd1cfyj6tQRCAQBuhIeLh4Y3Hq7edXubjgKghFDyAJSc9bOkL0ZIRfm3tPlJz6rqVqu2DuaVzdU7AaCyq+NXR5NTJ6uOfx3TUQCUIKZrAig5TR+THl4iVQm9pc1fjrubggcAZaR3TG990PEDCh7ghBjJA1DyzhyWPn5U2vX1DW+yqs7dGlSYW4qhAACSVMWtisbeNVZtItqYjgKglFDyAJQOu13635nSyrFSQd41Vz3h5a+u4RE6fP5oGYUDgMopOTBZk+6ZpBCfENNRAJQiSh6A0nVou7Twf6R9m666yjPJHbT02NayywQAlYzVYtX/JP2PHk96XC5WF9NxAJQySh6A0ldYIP3zFembKVJRQbGHvoxJ1dD8HDO5AKASCPIK0sR7JqpxcGPTUQCUEUoegLKz5zvpk8ekIzslSUe9q6tbWE0dPX/cbC4AcFItw1pq3F3j5OfhZzoKgDJEyQNQtvLPSV9mSP96W0OT2+vLYz+aTgQATsfT5qkhdw5Rn9g+pqMAMICSB8CII7u/VZ8NY7XvzD7TUQDAqaSEpmhUs1EK9bm129kAqPgoeQCMOZN/RlM3TtVHOz6SXfxTBAC3I8AjQM82flYdojqYjgLAMEoeAOP+te9fGr12tPac3mM6CgBUSF1qd9HwxsNV1b2q6SgAygFKHoBy4VzBOU3/v+l6/6f3VWQvMh0HACqEcN9wZTTPUNOQpqajAChHKHkAypXvD36vjG8zlHMyx3QUACi3bBab0uul6/H6j8vdxd10HADlDCUPQLlzvvC83vz+Tb237T0V2gtNxwGAciWxeqJGNx+tugF1TUcBUE5R8gCUWzuO7dCkf03Shv0bTEcBAOO8bF4anDxYfWL7yGqxmo4DoByj5AEo95bnLNeUjVO098xe01EAwIjUO1I1sulIhfiEmI4CoAKg5AGoEPIK8jRn6xy9u/Vd5RXmmY4DAGWimkc1Pdf0ObWv1d50FAAVCCUPQIWy7/Q+vbrxVS3fvdx0FAAoNRZZ1D26u4bcOYTbIgC4aZQ8ABXShv0bNPFfE7Xz2E7TUQCgRCVWT9SwRsOUHJRsOgqACoqSB6DCKiwq1Ec7PtKMTTN04vwJ03EA4LaE+YZpcPJgpmYCuG2UPAAV3onzJ/TG929owY4F3HIBQIXj5+6nx5Ie0/1175eri6vpOACcACWvhBUWFio/P990DFQybm5uslq5nPb2o9s16V+TtPHARtNRAOC63F3c9WDcgxqQOEC+br6m4wBwIpS8EmK327V//34dP37cdBRUQlarVZGRkXJzczMdpVz4IucLTdk4RfvO7DMdBQAuY7VY1SmqkwY1HKRg72DTcQA4IUpeCdm3b5+OHz+uwMBAeXl5yWKxmI6ESqKoqEh79+6Vq6urwsPDee/9R15Bnt7d+q7mbJ3DLRcAlBt31bxLQ+8cqroBdU1HAeDEKHkloLCwUDt27FBgYKCqVatmOg4qoRMnTmjv3r2qU6eOXF05n+O3Dp09pHe3vquPdnyk84XnTccBUEnFBsRqyJ1DdFfNu0xHAVAJUPJKQF5enrKzs1WrVi15enqajoNK6Ny5c8rJyVFkZKQ8PDxMxymXKHsATAjxDtGTDZ9Up6hOslo4dxpA2aDklYBLJY8P2DCF9+CNo+wBKAu+rr56NOlR9YvrJ3cXd9NxAFQylLwSwAdsmMZ78OZR9gCUBncXd/WK6aXHkh6Tn4ef6TgAKilKXgngAzZM4z146yh7AEqCl81Lvev2Vv+E/qruWd10HACVHCWvBFztA3at5z4v0xw5kzre9Dbp6emaO3euJMlmsykgIEBJSUnq06eP0tPTHfdeq1Wrlnbv3i1J8vDwUFBQkJo0aaLHH39crVu3LrmDwC2h5N2+Q2cP6S9b/6IFOxZQ9gDcsCpuVdQ3rq8ejHtQVd2rmo4DAJIkzgCG2rdvr3379iknJ0dLly5Vq1at9NRTT6lTp04qKChwrPfiiy9q37592r59u9577z35+fmpTZs2eumllwymB0pGDa8aeq7Jc1rafSnn0AC4rgCPAD2d/LSW91yuPzb4IwUPQLlCyYPc3d0VHBys0NBQJScn64UXXtCnn36qpUuXKjMz07Ger6+vgoODFR4ertTUVM2ePVujRo1SRkaGtm/f7lhv9erVatKkidzd3RUSEqLnnnuuWFlcsGCBEhMT5enpqWrVqqlNmzY6c+ZMWR4ycFWUPQDXEuwdrOeaPKcvenyhAYkD5O3qbToSAFyGkocrat26terXr6+FCxdec72nnnpKdrtdn376qSTp119/VYcOHdS4cWP98MMPmjlzpv7yl79o/Pjxki7eNL5Pnz565JFHlJWVpVWrVql79+5i1jDKm/8ue542bo8CVGbR/tEanzJeS7ovUb+4fvKwMTUeQPllMx0A5VdsbKw2b958zXUCAgIUGBionJwcSdJbb72lsLAwzZgxQxaLRbGxsdq7d6+effZZZWRkaN++fSooKFD37t0VEREhSUpMTCztQwFu2aWy90SDJ7Ro5yJ9uP1D/XLqF9OxAJSRZiHNlJ6QrpTQFNNRAOCGUfJwVXa7XRaL5abWy8rKUvPmzYttl5KSotOnT2vPnj2qX7++0tLSlJiYqHbt2unee+9Vz5495e/vX2rHAZSEKm5V9FDCQ/p9/O/1za/f6IOfPtC3v34ruxiFBpyNzWpT+1rtlZ6QrroBdU3HAYCbRsnDVWVlZSkyMvKa6xw5ckSHDh1yrHelYnhpKqbFYpGLi4u+/PJLrV27VsuXL9cbb7yhESNGaP369dfdF1AeWCwWpd6RqtQ7UpV7Mlcfbv9Qi/69SKcunDIdDcBt8nH1Uc+YnuoX10/B3sGm4wDALeOcPFzRV199pS1btqhHjx7XXO/111+X1WpV165dJUnx8fFau3ZtsXPs1q5dK19fX4WGhkq6+CE5JSVFY8eO1ffffy83Nzd98sknpXYsQGkJrxKuZxo/oxU9V2hUs1GK9o82HQnALYgLiNOIpiP0Zc8v9adGf6LgAajwGMmDzp8/r/3796uwsFAHDhzQsmXLNHHiRHXq1EkPPfSQY71Tp05p//79ys/PV3Z2tubPn6933nlHEydOVJ06dSRJTzzxhKZNm6ZBgwbpySef1Pbt2zV69GgNHTpUVqtV69ev18qVK3XvvfcqMDBQ69ev16FDhxQXF2fq8IHb5uV68SbIvev21ob9G/TBTx/o69yvVWAvuP7GAIzwdfVVh6gO6h7dXfHV4k3HAYASRcmDli1bppCQENlsNvn7+6t+/fqaPn26+vfv77gZuiRlZGQoIyNDbm5uCg4OVrNmzbRy5Uq1atXKsU5oaKiWLFmi4cOHq379+goICNCAAQM0cuRISVKVKlX0z3/+U9OmTdPJkycVERGhKVOm6L777ivz4wZKQ+Pgxmoc3FgHzhzQ33f8XR/v+FhH8o6YjgXgP5IDk9UjpofujbiXK2QCcFoWO9euv215eXnKzs5WZGSkPDz4HwbKHu/B8iu/MF/Lcpbpw58+1ObD175aLYDSEeARoC61u6hbdDdFVuX8bwDOj5E8AChFri6u6ly7szrX7qwfD/+oBTsX6MvdX+rE+ROmowFOzWqx6q6ad6lHdA+1CGshV6ur6UgAUGYYySsBjKLANN6DFUt+Ub7W7V2nJdlL9HXu1zpbcNZ0JMBp1PSuqa7RXdWtTjcuoAKg0mIkDwDKmKvV1XEbhnMF57T6l9Vakr1Ea35do/yifNPxgArH1eqqVmGt1CO6h5rVbCarhYuHA6jcKHkAYJCnzVPtI9urfWR7nbxwUit2r9CS7CXasH+DiuxFpuMB5ZbVYlWDGg3UNqKtOkZ1lL+Hv+lIAFBuUPIAoJyo4lZF3aO7q3t0dx0+d1hf5HyhJbuWcMEW4D9cra5qGtJUbcLbqGVYS1XzrGY6EgCUS5Q8ACiHqntWV7+4fuoX10+/nPpFy7KXaUn2Ev37+L9NRwPKlJfNS3eH3q208DSl3pEqHzcf05EAoNzjwislgItewDTeg5XHjmM7tDR7qZZmL9Wvp381HQcoFX7ufmoZ1lJp4WlqXrO53F3cTUcCgAqFkTwAqEBi/GMU4x+jp5Kf0s5jO7V271p9++u3+u7Ad7pQdMF0POCWBXkFKS08TWnhaboz6E65WF1MRwKACouSBwAVVLR/tKL9o9U/ob/yCvK0Yf+Gi6Vv77fKPpFtOh5wXZFVIx3FLqFagiwWi+lIAOAUKHkA4AQ8bB665457dM8d90iS9p7eq2/3fqu1v67V+n3rdSr/lOGEgOTt6q0GgQ3UJLiJWt7RUlF+UaYjAYBTouSVpjFVy3h/J256k/T0dM2dO1eSZLPZFBAQoKSkJPXp00fp6emyWrnXEFAR1fSpqV4xvdQrppcKigq0+dBmR+nbdnQbt2dAmfBx9VHDwIZqHNxYjYIaKb5aPNMwAaAMUPKg9u3ba86cOSosLNSBAwe0bNkyPfXUU1qwYIEWL14sm423CVCR2aw2JQclKzkoWYMaDtKxvGNat3edvt37rdbtXadD5w6Zjggn4evqq+SgZDUKaqTGwY0VGxBLqQMAA/j0Drm7uys4OFiSFBoaquTkZDVr1kxpaWnKzMzUo48+qtzcXA0aNEgrV66U1WpV+/bt9cYbbygoKEiS9MMPP+jpp5/Wxo0bZbFYFB0drVmzZqlRo0YmDw3AFfh7+KtDVAd1iOogSdp+dLu+P/i9thzeoq2HtyrnZA4jfbghvm6+ujPwTjUKbqRGwY0UFxAnq4UZIABgGiUPV9S6dWvVr19fCxcu1IABA9S1a1d5e3tr9erVKigo0BNPPKH7779fq1atkiT169dPDRs21MyZM+Xi4qJNmzbJ1dXV7EEAuCF1A+qqbkBdPaAHJEmnL5zWtiPbHKVvy+EtOnD2gOGUKA+quFXRnUF3Okbq6gbUpdQBQDlEycNVxcbGavPmzVqxYoU2b96s7OxshYWFSZLmzZunhIQEbdiwQY0bN1Zubq6GDx+u2NhYSVJ0dLTJ6ABug4+bj5qENFGTkCaOZYfPHdaWQ1scxe/HIz/q5IWTBlOitAV4BCjGP0Z1/esqJiBGsQGxquNXh1J3G270PPhatWpp9+7dkiQPDw8FBQWpSZMmevzxx9W6deub2ufBgwc1atQoLV26VAcOHJC/v7/q16+vMWPGqHnz5iV7gADKDUoersput8tisSgrK0thYWGOgidJ8fHx8vPzU1ZWlho3bqyhQ4fq0Ucf1bx589SmTRv16tVLtWvXNpgeQEmq7lldrcJbqVV4K0kX/33IPZXrKH1bD2/VT0d/0vnC84aT4mbZrDZFVo28WOZ+U+qqe1Y3Hc0p3eh58C+++KIGDhyoCxcuKCcnR/Pnz1ebNm00btw4jRgx4ob316NHD+Xn52vu3LmKiorSgQMHtHLlSh09erS0DhFAOUDJw1VlZWUpMjLSUfb+22+XjxkzRn379tXnn3+upUuXavTo0frwww/VrVu3so4NoAxYLBZFVIlQRJUIdYrqJEnKL8rXzmM79dPRn7T75G79cuoXx/dzBecMJ4YkVfOodrHIBVwsdDH+MYryi5Krlen1ZeVGzoOXJF9fX8d64eHhSk1NVUhIiDIyMtSzZ0/VrVtXkrR69WoNHz5cP/zwgwICAtS/f3+NHz9eNptNx48f15o1a7Rq1Sq1aNFCkhQREaEmTZpcIRkAZ0LJwxV99dVX2rJli4YMGaI77rhDubm5+uWXXxyjedu2bdOJEycUFxfn2CYmJkYxMTEaMmSI+vTpozlz5lDygErE1eqq+Grxiq8WX2y53W7XwbMHlXsqV7knc7X71O6L30/u1p5Te5RXmGcosXOyWqyq7lFdwT7BivCNuFjmAi4WOkbnyqffngd/qeRdyVNPPaVx48bp008/1TPPPKNff/1VHTp0UHp6ut577z399NNPGjhwoDw8PDRmzBj5+PjIx8dHixYtUrNmzeTu7l6GRwXAJEoedP78ee3fv7/Y1JGJEyeqU6dOeuihh2S1WpWUlKR+/fpp2rRpjguvtGjRQo0aNdK5c+c0fPhw9ezZU5GRkdqzZ482bNigHj16mD40AOWAxWJRkHeQgryD1Di4cbHH7Ha7Dpw9UKz85Z7MVe6pXP1y6hemf16Br6uvgn2CFeIdomCvYIX4hCjIK0gh3iEK8QlRoFcgI3MV0KXz4K8lICBAgYGBysnJkSS99dZbCgsL04wZM2SxWBQbG6u9e/fq2WefVUZGhmw2mzIzMzVw4ED9+c9/VnJyslq0aKEHHnhASUlJZXBUAEyh5EHLli1TSEiIbDab44Ts6dOnq3///o6TwBctWqRBgwYpNTW12C0UJMnFxUVHjhzRQw89pAMHDqh69erq3r27xo4da/KwAFQAFotFwd7BCvYOLnahF+liAdx/Zr8OnD2gI+eO6EjeEcf3o3lHiy07nX/a0BGULFer68XC5hOiEO8Qx8/BXsGOEuft6m06JkrB1U6NuNZ6WVlZat68ebHtUlJSdPr0ae3Zs0fh4eHq0aOHOnbsqG+++Ubr1q3TsmXL9Morr+idd95Renp6aR0OAMMoeaVpzAnTCa4rMzNTmZmZ110vPDxcn3766RUfc3Nz0wcffFDCyQBUdhaL5WLZ8Qm57rrnC8/reN5xnbxw8uLX+ZM6ceGETp4/6Vh24vwJnbxwUvmF+Sq0F/7/r6JCFdmLVGAvUFFRUbHll34ushcVX/afbYpUJA8XD3m7esvb1Vterl4Xf7b95udLy23el6/3X+t62DzK4JVFeXTpPPhrOXLkiA4dOuRY70rF0G63S1Kx5R4eHmrbtq3atm2rjIwMPfrooxo9ejQlD3BilDwAQIXn7uLumBIKVDS/PQ/+Wl5//XVZrVZ17dpV0sUrXX/88cfFyt7atWvl6+ur0NDQqz5PfHy8Fi1aVFLxAZRDlDwAAIAycr3z4C85deqU9u/fr/z8fGVnZ2v+/Pl65513NHHiRNWpU0eS9MQTT2jatGkaNGiQnnzySW3fvl2jR4/W0KFDZbVadeTIEfXq1UuPPPKIkpKS5Ovrq40bN+qVV15Rly5dTL0EAMoAJQ8AAKCM3Mh58JKUkZGhjIwMubm5KTg4WM2aNdPKlSvVqlUrxzqhoaFasmSJhg8frvr16ysgIEADBgzQyJEjJUk+Pj5q2rSpXnvtNf3888/Kz89XWFiYBg4cqBdeeKHMjx1A2bHYL03exi3Ly8tTdna2IiMj5eHB+RQoe7wHAQAAcIn1+qsAAAAAACoKSh4AAAAAOBFKHgAAAAA4EUoeAAAAADgRSh4AAAAAOBFKHiqNWrVqadq0aaZjAAAAAKWK++SVosS5iWW6vy39t9z0Nunp6Zo7d64kycXFRTVr1lTHjh01YcIE+fv7l3REozZs2CBvb2/TMQAAAIBSRcmD2rdvrzlz5qigoEDbtm3TI488ouPHj+uDDz4wHa1E1ahRw3QEAAAAoNQxXRNyd3dXcHCw7rjjDt177726//77tXz5csfjc+bMUVxcnDw8PBQbG6u33nqr2PZ79uzRAw88oICAAHl7e6tRo0Zav3694/GZM2eqdu3acnNzU926dTVv3rxi21ssFr3zzjvq1q2bvLy8FB0drcWLFzseX7VqlSwWi1auXKlGjRrJy8tLd911l7Zv3+5Y5+eff1aXLl0UFBQkHx8fNW7cWCtWrCi2n/+erjlmzBiFh4fL3d1dNWvW1ODBgx2PvfXWW4qOjpaHh4eCgoLUs2dPx2PLli3T3XffLT8/P1WrVk2dOnXSzz//7Hg8JydHFotFCxcuVKtWreTl5aX69etr3bp1xfJ8++23atGihby8vOTv76927drp2LFjkiS73a5XXnlFUVFR8vT0VP369bVgwYKr/yUCAAAA/0HJQzG7du3SsmXL5OrqKkl6++23NWLECL300kvKysrShAkTNGrUKMcUz9OnT6tFixbau3evFi9erB9++EHPPPOMioqKJEmffPKJnnrqKf3pT3/S1q1b9dhjj+nhhx/W119/XWy/Y8eOVe/evbV582Z16NBB/fr109GjR4utM2LECE2ZMkUbN26UzWbTI4884njs9OnT6tChg1asWKHvv/9e7dq1U+fOnZWbm3vF41ywYIFee+01zZo1Szt37tSiRYuUmHhxeu3GjRs1ePBgvfjii9q+fbuWLVum1NRUx7ZnzpzR0KFDtWHDBq1cuVJWq1XdunVzHPNv8w4bNkybNm1STEyM+vTpo4KCAknSpk2blJaWpoSEBK1bt05r1qxR586dVVhYKEkaOXKk5syZo5kzZ+rHH3/UkCFD9OCDD2r16tU39xcKAACASsdit9vtpkNUdHl5ecrOzlZkZKQ8PDwcyyvKOXnz58+Xh4eHCgsLlZeXJ0maOnWqhgwZovDwcL388svq06ePY5vx48dryZIlWrt2rWbPnq1hw4YpJydHAQEBlz1/SkqKEhISNHv2bMey3r1768yZM/r8888lXRzJGzlypMaNGyfpYony9fXVkiVL1L59e61atUqtWrXSihUrlJaWJklasmSJOnbsqHPnzhV7zX8rISFBf/jDH/Tkk09KujiS9/TTT+vpp5/W1KlTNWvWLG3dutVRaC9ZuHChHn74Ye3Zs0e+vr7XfQ0PHTqkwMBAbdmyRfXq1VNOTo4iIyP1zjvvaMCAAZKkbdu2KSEhQVlZWYqNjVXfvn2Vm5urNWvWXPZ8Z86cUfXq1fXVV1+pefPmjuWPPvqozp49q/fff/+yba72HgQAAEDlw0ge1KpVK23atEnr16/XoEGD1K5dOw0aNEiHDh3SL7/8ogEDBsjHx8fxNX78eMf0xE2bNqlhw4ZXLHiSlJWVpZSUlGLLUlJSlJWVVWxZUlKS42dvb2/5+vrq4MGDV10nJCREkhzrnDlzRs8884zi4+Pl5+cnHx8f/fTTT1cdyevVq5fOnTunqKgoDRw4UJ988oljlK1t27aKiIhQVFSUfv/73+uvf/2rzp4969j2559/Vt++fRUVFaUqVaooMjJSki7b17XyXhrJu5Jt27YpLy9Pbdu2Lfa6v/fee8WmhQIAAABXwoVXIG9vb9WpU0eSNH36dLVq1Upjx451jIC9/fbbatq0abFtXFxcJEmenp7XfX6LxVLsz3a7/bJl/z2aZrFYLpv++Nt1Lm1/aZ3hw4friy++0Kuvvqo6derI09NTPXv21IULF66YKSwsTNu3b9eXX36pFStW6IknntDkyZO1evVq+fr66v/+7/+0atUqLV++XBkZGRozZow2bNggPz8/de7cWWFhYXr77bdVs2ZNFRUVqV69epft61p5r/W6XVrn888/V2hoaLHH3N3dr7odAAAAIDGShysYPXq0Xn31VRUWFio0NFS7du1SnTp1in1dGr1KSkrSpk2bLjt/7pK4uLjLpiSuXbtWcXFxJZr5m2++UXp6urp166bExEQFBwcrJyfnmtt4enrqd7/7naZPn65Vq1Zp3bp12rLl4pRXm82mNm3a6JVXXtHmzZuVk5Ojr776SkeOHFFWVpZGjhyptLQ0xcXFOS6WcjOSkpK0cuXKKz4WHx8vd3d35ebmXva6h4WF3fS+AAAAULkwkofLtGzZUgkJCZowYYLGjBmjwYMHq0qVKrrvvvt0/vx5bdy4UceOHdPQoUPVp08fTZgwQV27dtXEiRMVEhKi77//XjVr1lTz5s01fPhw9e7dW8nJyUpLS9Nnn32mhQsXXnbly9tVp04dLVy4UJ07d5bFYtGoUaMuGwn8rczMTBUWFqpp06by8vLSvHnz5OnpqYiICP3jH//Qrl27lJqaKn9/fy1ZskRFRUWqW7eu/P39Va1aNc2ePVshISHKzc3Vc889d9N5n3/+eSUmJuqJJ57Q448/Ljc3N3399dfq1auXqlevrmHDhmnIkCEqKirS3XffrZMnT2rt2rXy8fFR//79b+elAgAAgJNjJA9XNHToUL399ttq166d3nnnHWVmZioxMVEtWrRQZmamYyTPzc1Ny5cvV2BgoDp06KDExERNmjTJMZ2za9euev311zV58mQlJCRo1qxZmjNnjlq2bFmieV977TX5+/vrrrvuUufOndWuXTslJydfdX0/Pz+9/fbbSklJcYyqffbZZ6pWrZr8/Py0cOFCtW7dWnFxcfrzn/+sDz74QAkJCbJarfrwww/13XffqV69ehoyZIgmT55803ljYmK0fPly/fDDD2rSpImaN2+uTz/9VDbbxd+7jBs3ThkZGZo4caLi4uLUrl07ffbZZ47XHQAAALgarq5ZAriyIUzjPQgAAIBLGMkDAAAAACdCyQMAAAAAJ0LJAwAAAAAnQskDAAAAACdCyQMAAAAAJ0LJAwAAAAAnQskDAAAAACdCyQMAAAAAJ0LJAwAAAAAnQslDhbNq1SpZLBYdP37cdBQAAACg3LGZDuDMsmLjynR/cT9l3fQ26enpmjt3riZOnKjnnnvOsXzRokXq1q2b7HZ7SUYEAAAAUMoYyYM8PDz08ssv69ixYyX2nBcuXCix5wIAAABw4yh5UJs2bRQcHKyJEydedZ2PP/5YCQkJcnd3V61atTRlypRij9eqVUvjx49Xenq6qlatqoEDByozM1N+fn76xz/+obp168rLy0s9e/bUmTNnNHfuXNWqVUv+/v4aNGiQCgsLHc81f/58NWrUSL6+vgoODlbfvn118ODBUjt+AAAAwJlQ8iAXFxdNmDBBb7zxhvbs2XPZ499995169+6tBx54QFu2bNGYMWM0atQoZWZmFltv8uTJqlevnr777juNGjVKknT27FlNnz5dH374oZYtW6ZVq1ape/fuWrJkiZYsWaJ58+Zp9uzZWrBggeN5Lly4oHHjxumHH37QokWLlJ2drfT09NJ8CQAAAACnwTl5kCR169ZNDRo00OjRo/WXv/yl2GNTp05VWlqao7jFxMRo27Ztmjx5crHy1bp1aw0bNszx5zVr1ig/P18zZ85U7dq1JUk9e/bUvHnzdODAAfn4+Cg+Pl6tWrXS119/rfvvv1+S9MgjjzieIyoqStOnT1eTJk10+vRp+fj4lNZLAAAAADgFRvLg8PLLL2vu3Lnatm1bseVZWVlKSUkptiwlJUU7d+4sNs2yUaNGlz2nl5eXo+BJUlBQkGrVqlWsrAUFBRWbjvn999+rS5cuioiIkK+vr1q2bClJys3Nva3jAwAAACoDSh4cUlNT1a5dO73wwgvFltvtdlkslsuW/Tdvb+/Llrm6uhb7s8ViueKyoqIiSdKZM2d07733ysfHR/Pnz9eGDRv0ySefSOJiLgAAAMCNYLomipk0aZIaNGigmJgYx7L4+HitWbOm2Hpr165VTEyMXFxcSnT/P/30kw4fPqxJkyYpLCxMkrRx48YS3QcAAADgzBjJQzGJiYnq16+f3njjDceyP/3pT1q5cqXGjRunHTt2aO7cuZoxY0ax8+9KSnh4uNzc3PTGG29o165dWrx4scaNG1fi+wEAAACcFSUPlxk3blyx6ZjJycn6+9//rg8//FD16tVTRkaGXnzxxVK54mWNGjWUmZmpjz76SPHx8Zo0aZJeffXVEt8PAAAA4Kws9iudXIWbkpeXp+zsbEVGRsrDw8N0HFRCvAcBAABwCSN5AAAAAOBEKHkAAAAA4EQoeQAAAADgRCh5AAAAAOBEKHkAAAAA4EQoeQAAAADgRCh5AAAAAOBEKHkAAAAA4EQoebiqnJwcWSwWbdq06bae5+zZs+rRo4eqVKkii8Wi48ePl0g+AAAAAJej5FVSFovlml/p6ekltq+5c+fqm2++0dq1a7Vv3z4dO3asRMojAAAAgMvZTAdwZm8+/lWZ7u+Pf259w+vu27fP8fPf/vY3ZWRkaPv27Y5lnp6eOnbsWInk+vnnnxUXF6d69epJujhCCAAAAKB0MJJXSQUHBzu+qlatKovFctmyS3bt2qVWrVrJy8tL9evX17p164o919q1a5WamipPT0+FhYVp8ODBOnPmjCSpZcuWmjJliv75z3/KYrGoZcuWioyMlCQ1bNjQsQwAAABAyaDk4bpGjBihYcOGadOmTYqJiVGfPn1UUFAgSdqyZYvatWun7t27a/Pmzfrb3/6mNWvW6Mknn5QkLVy4UAMHDlTz5s21b98+LVy4UP/6178kSStWrHAsAwAAAFAymK6J6xo2bJg6duwoSRo7dqwSEhL073//W7GxsZo8ebL69u2rp59+WpIUHR2t6dOnq0WLFpo5c6YCAgLk5eUlNzc3BQcHS5JOnjwpSapWrZpjGQAAAICSwUgerispKcnxc0hIiCTp4MGDkqTvvvtOmZmZ8vHxcXy1a9dORUVFys7ONpIXAAAAqMwYycN1ubq6On62WCySpKKiIsf3xx57TIMHD75su/Dw8LIJCAAAAMCBkofbkpycrB9//FF16tS54W3c3NwkSYWFhaUVCwAAAKi0mK6J2/Lss89q3bp1+uMf/6hNmzZp586dWrx4sQYNGnTVbQIDA+Xp6ally5bpwIEDOnHiRBkmBgAAAJwbJQ+3JSkpSatXr9bOnTt1zz33qGHDhho1apTj3L0rsdlsmj59umbNmqWaNWuqS5cuZZgYAAAAcG4Wu91uNx2iosvLy1N2drYiIyPl4eHhWF6eb4YO53K19yAAAAAqH87JK0WULgAAAABljemaAAAAAOBEKHkAAAAA4EQoeQAAAADgRCh5AAAAAOBEKHkliAuVwhTeewAAALiEklcCXF1dJUlnz541nASV1YULFyRJLi4uhpMAAADANG6hUAJcXFzk5+engwcPSpK8vLxksVgMp0JlUVRUpEOHDsnLy0s2G/9JAwAAVHZ8IiwhwcHBkuQoekBZslqtCg8P55cLAAAAkMXOyTwlqrCwUPn5+aZjoJJxc3OT1crsawAAAFDyAAAAAMCp8Kt/AAAAAHAilDwAAAAAcCKUPAAAAABwIpQ8AAAAAHAilDwAAAAAcCKUPAAAAABwIpQ8AAAAAHAilDwAAAAAcCKUPAAAAABwIpQ8AAAAAHAilDwAAAAAcCKUPAAAAABwIpQ8AAAAAHAilDwAAAAAcCL/DxR/VzD6CGJDAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 1000x1000 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# check the number of unique values in the target variable\n",
    "print(subset_dataset['category'].value_counts())\n",
    "subset_dataset['category'].value_counts().plot(kind = 'pie',\n",
    "                                               explode = [0, 0, 0, 0.1, 0.1],\n",
    "                                               figsize = (10, 10),\n",
    "                                               autopct = '%1.3f%%',\n",
    "                                               shadow = False)\n",
    "plt.ylabel(\"Categories of Attack\")\n",
    "plt.legend([\"DDos\", \"Dos\", \"Reconnaissance\", \"Normal\", \"Theft\"])\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 152,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "f_4VzytahWJp",
    "outputId": "ef63a286-9f3a-4a68-8fef-0e98112323f2"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "proto_number           int64\n",
       "flgs_number            int64\n",
       "pkts                   int64\n",
       "bytes                  int64\n",
       "dur                  float64\n",
       "spkts                  int64\n",
       "dpkts                  int64\n",
       "sbytes                 int64\n",
       "dbytes                 int64\n",
       "seq                    int64\n",
       "stddev               float64\n",
       "N_IN_Conn_P_SrcIP      int64\n",
       "min                  float64\n",
       "state_number           int64\n",
       "mean                 float64\n",
       "N_IN_Conn_P_DstIP      int64\n",
       "drate                float64\n",
       "srate                float64\n",
       "max                  float64\n",
       "category              object\n",
       "dtype: object"
      ]
     },
     "execution_count": 152,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# check the data type of dataframe columns\n",
    "subset_dataset.dtypes"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "7rv4WPs_ykfy"
   },
   "source": [
    "### Since the data with `Theft` is small, we remove the rows."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 153,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 670
    },
    "id": "aAIq04e9yR-_",
    "outputId": "24ad606d-2c99-474c-f055-6f87b7fe9e84"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "category\n",
      "DoS               11000\n",
      "Reconnaissance    11000\n",
      "DDoS              11000\n",
      "Normal              477\n",
      "Name: count, dtype: int64\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1000x1000 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "dataset_without_theft = subset_dataset[subset_dataset.category != \"Theft\"]\n",
    "dataset_without_theft\n",
    "\n",
    "# check the number of unique values in the target variable\n",
    "print(dataset_without_theft['category'].value_counts())\n",
    "dataset_without_theft['category'].value_counts().plot(kind = 'pie',\n",
    "                                               explode = [0, 0, 0.1, 0.1],\n",
    "                                               figsize = (10, 10),\n",
    "                                               autopct = '%1.3f%%',\n",
    "                                               shadow = False)\n",
    "plt.ylabel(\"Categories of Attack\")\n",
    "plt.legend([\"DDos\", \"Dos\", \"Reconnaissance\", \"Normal\"])\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 154,
   "metadata": {
    "id": "JRHgmboiiadP"
   },
   "outputs": [],
   "source": [
    "# # The sport and dport columns are objects. We need to convert them to integers.\n",
    "# subset_dataset[\"dport\"] = pd.to_numeric(subset_dataset[\"dport\"])\n",
    "# subset_dataset[\"sport\"] = pd.to_numeric(subset_dataset[\"sport\"])\n",
    "# #not all the strings in these columns can be parsed. I will figure another way out latet.\n",
    "# subset_dataset.dtypes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 155,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "yVJR6NklcjVr",
    "outputId": "2538d6da-184e-46aa-d0ae-55806800d330"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[ 3.        1.        8.       ...  0.        0.279982  3.60197 ]\n",
      " [ 3.        1.        8.       ...  0.        0.279982  3.60197 ]\n",
      " [ 3.        1.        8.       ...  0.        0.279982  3.60197 ]\n",
      " ...\n",
      " [ 3.        1.       15.       ...  0.        1.024986  4.885572]\n",
      " [ 3.        1.       15.       ...  0.        1.024986  4.885572]\n",
      " [ 3.        1.       15.       ...  0.        1.02498   4.885572]]\n",
      "0         DoS\n",
      "1         DoS\n",
      "2         DoS\n",
      "3         DoS\n",
      "4         DoS\n",
      "         ... \n",
      "10995    DDoS\n",
      "10996    DDoS\n",
      "10997    DDoS\n",
      "10998    DDoS\n",
      "10999    DDoS\n",
      "Name: category, Length: 33477, dtype: object\n",
      "33477\n"
     ]
    }
   ],
   "source": [
    "X = dataset_without_theft.iloc[:, :-1].values\n",
    "y = dataset_without_theft.category\n",
    "print(X)\n",
    "print(y)\n",
    "print(len(X))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "ZtoUYKw6bdx7"
   },
   "source": [
    "### Label Encode the target variable"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 156,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "M4Sc3-4XbcjQ",
    "outputId": "92eeb41e-8e9a-4955-8762-cb446227e8e3"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Mapping of Label Encoded Classes\n",
      "{'DDoS': 0, 'DoS': 1, 'Normal': 2, 'Reconnaissance': 3}\n"
     ]
    }
   ],
   "source": [
    "from sklearn.preprocessing import LabelEncoder\n",
    "\n",
    "label_encoder = LabelEncoder()\n",
    "encoded_y = label_encoder.fit_transform(y)\n",
    "label_encoder_name_mapping = dict(zip(label_encoder.classes_,\n",
    "                                     label_encoder.transform(label_encoder.classes_)))\n",
    "print(\"Mapping of Label Encoded Classes\", label_encoder_name_mapping, sep=\"\\n\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 157,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "cVC3DT1wbcg4",
    "outputId": "a9427db2-aa6a-446f-d0cf-d91ea5333bf6"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "X Training set\n",
      "[[1.000000e+00 1.000000e+00 2.000000e+00 ... 0.000000e+00 0.000000e+00\n",
      "  4.670000e-04]\n",
      " [3.000000e+00 1.000000e+00 8.000000e+00 ... 0.000000e+00 2.800530e-01\n",
      "  3.602055e+00]\n",
      " [3.000000e+00 1.000000e+00 1.100000e+01 ... 0.000000e+00 7.293130e-01\n",
      "  4.105704e+00]\n",
      " ...\n",
      " [3.000000e+00 1.000000e+00 1.100000e+01 ... 0.000000e+00 7.293590e-01\n",
      "  4.105629e+00]\n",
      " [1.000000e+00 1.000000e+00 1.000000e+00 ... 0.000000e+00 0.000000e+00\n",
      "  0.000000e+00]\n",
      " [3.000000e+00 1.000000e+00 1.500000e+01 ... 0.000000e+00 1.025101e+00\n",
      "  4.885493e+00]]\n",
      "X Validation set\n",
      "[[1.000000e+00 1.000000e+00 2.000000e+00 ... 0.000000e+00 0.000000e+00\n",
      "  1.240000e-04]\n",
      " [3.000000e+00 1.000000e+00 8.000000e+00 ... 0.000000e+00 2.800250e-01\n",
      "  3.591519e+00]\n",
      " [3.000000e+00 1.000000e+00 1.500000e+01 ... 0.000000e+00 1.025042e+00\n",
      "  4.884496e+00]\n",
      " ...\n",
      " [3.000000e+00 1.000000e+00 1.500000e+01 ... 0.000000e+00 1.025030e+00\n",
      "  4.884442e+00]\n",
      " [3.000000e+00 1.000000e+00 1.300000e+01 ... 0.000000e+00 4.466120e-01\n",
      "  4.495078e+00]\n",
      " [3.000000e+00 1.000000e+00 1.100000e+01 ... 0.000000e+00 7.293300e-01\n",
      "  4.105610e+00]]\n",
      "y Training set\n",
      "[3 1 0 ... 0 3 0]\n",
      "y Validation set\n",
      "[3 1 0 ... 0 1 0]\n"
     ]
    }
   ],
   "source": [
    "# Splitting Dataset into Training and Validation Set\n",
    "from sklearn.model_selection import train_test_split\n",
    "X_train, X_val, y_train, y_val = train_test_split(X, encoded_y, test_size=0.25,\n",
    "                                                  random_state=0,\n",
    "                                                  stratify=encoded_y)\n",
    "\n",
    "print('X Training set', X_train, sep='\\n')\n",
    "print('X Validation set', X_val, sep='\\n')\n",
    "print('y Training set', y_train, sep='\\n')\n",
    "print('y Validation set', y_val, sep='\\n')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "CCcMGJCOgpJ_"
   },
   "source": [
    "### Feature Scaling"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 158,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "ZL_D7BDignmQ",
    "outputId": "0f3cd2a8-e563-41ae-bc67-328e87b5926d"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[-1.43248578 -0.11454024 -0.03321754 ... -0.01090576 -0.01675704\n",
      "  -1.34674091]\n",
      " [ 0.69212554 -0.11454024 -0.02864697 ... -0.01090576 -0.01671956\n",
      "   0.42061226]\n",
      " [ 0.69212554 -0.11454024 -0.02636169 ... -0.01090576 -0.01665944\n",
      "   0.66776037]\n",
      " ...\n",
      " [ 0.69212554 -0.11454024 -0.02636169 ... -0.01090576 -0.01665943\n",
      "   0.66772357]\n",
      " [-1.43248578 -0.11454024 -0.03397931 ... -0.01090576 -0.01675704\n",
      "  -1.34697007]\n",
      " [ 0.69212554 -0.11454024 -0.02331464 ... -0.01090576 -0.01661985\n",
      "   1.05041451]]\n",
      "[[-1.43248578 -0.11454024 -0.03321754 ... -0.01090576 -0.01675704\n",
      "  -1.34690923]\n",
      " [ 0.69212554 -0.11454024 -0.02864697 ... -0.01090576 -0.01671957\n",
      "   0.41544209]\n",
      " [ 0.69212554 -0.11454024 -0.02331464 ... -0.01090576 -0.01661986\n",
      "   1.04992527]\n",
      " ...\n",
      " [ 0.69212554 -0.11454024 -0.02331464 ... -0.01090576 -0.01661986\n",
      "   1.04989877]\n",
      " [ 0.69212554 -0.11454024 -0.02483816 ... -0.01090576 -0.01669727\n",
      "   0.85883202]\n",
      " [ 0.69212554 -0.11454024 -0.02636169 ... -0.01090576 -0.01665943\n",
      "   0.66771424]]\n",
      "25107\n",
      "8370\n"
     ]
    }
   ],
   "source": [
    "# Feature Scaling\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "\n",
    "scaler = StandardScaler()\n",
    "scaled_X_train = scaler.fit_transform(X_train)\n",
    "scaled_X_val = scaler.transform(X_val)\n",
    "print(scaled_X_train)\n",
    "print(scaled_X_val)\n",
    "print(len(scaled_X_train))\n",
    "print(len(scaled_X_val))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "xhRk-QQpl5E7"
   },
   "source": [
    "### Train Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 159,
   "metadata": {
    "id": "ZOqTUcVmm711"
   },
   "outputs": [],
   "source": [
    "from sklearn.metrics import confusion_matrix, accuracy_score, recall_score\n",
    "from sklearn.metrics import precision_score, f1_score, ConfusionMatrixDisplay\n",
    "from sklearn.metrics import classification_report\n",
    "\n",
    "def evaluate_model(y_val, y_pred):\n",
    "  \"\"\"Function to evaluate model and return the metric of the model\n",
    "\n",
    "  It returns a dictionary with the classification metrics.\n",
    "  \"\"\"\n",
    "  accuracy = accuracy_score(y_val, y_pred)\n",
    "  precision = precision_score(y_val, y_pred, average='macro')\n",
    "  recall = recall_score(y_val, y_pred, average='macro')\n",
    "  f1 = f1_score(y_val, y_pred, average='macro')\n",
    "  result = {\"accuracy_score\": accuracy,\n",
    "            \"precision_score\": precision,\n",
    "            \"recall_score\": recall,\n",
    "            \"f1_score\": f1}\n",
    "  return result\n",
    "\n",
    "def plot_confusion_matrix(y_val, y_pred, label):\n",
    "  '''function to plot confusion matrix\n",
    "\n",
    "  Args\n",
    "  y_val: array. The validation set of the target variable.\n",
    "  y_pred: array. Model's prediction.\n",
    "  label: list. A list containing all the classes in the target variable\n",
    "\n",
    "  Returns\n",
    "  It returns a plot of the confusion matrix\n",
    "  '''\n",
    "  cm = confusion_matrix(y_val, y_pred)\n",
    "  fig, ax = plt.subplots(figsize=(10,10))\n",
    "  ConfusionMatrixDisplay(cm, display_labels=label).plot(ax=ax, values_format='')\n",
    "\n",
    "# class labels\n",
    "label = ['DDoS', 'DoS',  'Normal', 'Reconnaissance']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "BIHPYxSGS4fI"
   },
   "source": [
    "### Logistic Regression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 160,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 387
    },
    "id": "b4taA51_kCre",
    "outputId": "03de108d-af53-4ee5-b128-ea62a7ba6417"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'accuracy_score': 0.9984468339307049, 'precision_score': 0.9849915371179302, 'recall_score': 0.9868636363636364, 'f1_score': 0.9859235062731134}\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       1.00      1.00      1.00      2750\n",
      "           1       1.00      1.00      1.00      2750\n",
      "           2       0.94      0.95      0.95       120\n",
      "           3       1.00      1.00      1.00      2750\n",
      "\n",
      "    accuracy                           1.00      8370\n",
      "   macro avg       0.98      0.99      0.99      8370\n",
      "weighted avg       1.00      1.00      1.00      8370\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1000x1000 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.linear_model import LogisticRegression\n",
    "\n",
    "logistic_regression_model = LogisticRegression(max_iter=1000) # , solver='liblinear'\n",
    "logistic_regression_model.fit(scaled_X_train, y_train)\n",
    "logistic_regression_y_pred = logistic_regression_model.predict(scaled_X_val)\n",
    "print(evaluate_model(y_val, logistic_regression_y_pred))\n",
    "plot_confusion_matrix(y_val, logistic_regression_y_pred, label)\n",
    "print(classification_report(y_val, logistic_regression_y_pred))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "ddvbaiECmyt6"
   },
   "source": [
    "### KNN Algorithm"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 161,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 784
    },
    "id": "dlpKQjfmmkPn",
    "outputId": "699f2bb8-7ae5-4f58-8d1e-4634ee203c6d"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/c4leb/anaconda3/lib/python3.9/site-packages/sklearn/neighbors/_classification.py:228: FutureWarning: Unlike other reduction functions (e.g. `skew`, `kurtosis`), the default behavior of `mode` typically preserves the axis it acts along. In SciPy 1.11.0, this behavior will change: the default value of `keepdims` will become False, the `axis` over which the statistic is taken will be eliminated, and the value None will no longer be accepted. Set `keepdims` to True or False to avoid this warning.\n",
      "  mode, _ = stats.mode(_y[neigh_ind, k], axis=1)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'accuracy_score': 0.9998805256869773, 'precision_score': 0.9999091239549255, 'recall_score': 0.9979166666666667, 'f1_score': 0.9989085286129034}\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       1.00      1.00      1.00      2750\n",
      "           1       1.00      1.00      1.00      2750\n",
      "           2       1.00      0.99      1.00       120\n",
      "           3       1.00      1.00      1.00      2750\n",
      "\n",
      "    accuracy                           1.00      8370\n",
      "   macro avg       1.00      1.00      1.00      8370\n",
      "weighted avg       1.00      1.00      1.00      8370\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1000x1000 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "knn_model = KNeighborsClassifier(n_neighbors=5, metric='minkowski', p=2)\n",
    "knn_model.fit(scaled_X_train, y_train)\n",
    "knn_y_pred = knn_model.predict(scaled_X_val)\n",
    "\n",
    "print(evaluate_model(y_val, knn_y_pred))\n",
    "plot_confusion_matrix(y_val, knn_y_pred, label)\n",
    "print(classification_report(y_val, knn_y_pred))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "miBrWQz4pi-q"
   },
   "source": [
    "### Support Vector Machine"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 162,
   "metadata": {
    "id": "j9Gp3aFipglg"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'accuracy_score': 0.9992831541218637, 'precision_score': 0.9880952380952381, 'recall_score': 0.9994545454545455, 'f1_score': 0.9936294139060793}\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       1.00      1.00      1.00      2750\n",
      "           1       1.00      1.00      1.00      2750\n",
      "           2       0.95      1.00      0.98       120\n",
      "           3       1.00      1.00      1.00      2750\n",
      "\n",
      "    accuracy                           1.00      8370\n",
      "   macro avg       0.99      1.00      0.99      8370\n",
      "weighted avg       1.00      1.00      1.00      8370\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA4cAAAMeCAYAAACuhHEdAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8qNh9FAAAACXBIWXMAAA9hAAAPYQGoP6dpAAB0JklEQVR4nO3deZyN9f//8ecZs29nZjAzRmPLOgwlu36WLINsbZQSJS2KhFRkyzKoEEqSj5GUPt9CKk3I8vkIWWosmSRLRgwTY4bBmDnn+v3h49QxlpmaM+cyHvfb7brdzHW9r+u8ruNyzOu8Xtf7shiGYQgAAAAAcFPzcHcAAAAAAAD3IzkEAAAAAJAcAgAAAABIDgEAAAAAIjkEAAAAAIjkEAAAAAAgkkMAAAAAgEgOAQAAAACSPN0dAAAAAABc7vz587pw4YK7w8jD29tbvr6+7g7DJUgOAQAAAJjK+fPnVbF8oFKP29wdSh6RkZE6cOBAsUwQSQ4BAAAAmMqFCxeUetym37ZVUHCQee6EyzxtV/k7DurChQskhwAAAABQVIKDPBQcVMLdYdw0SA4BAAAAmJJdhuyyuzsMB7sMd4fgUuap0QIAAAAA3IbkEAAAAABAWykAAAAAc7IZdtlM1MlpM8zT4uoKVA4BAAAAACSHAAAAAADaSgEAAACY1MXZSs3TV2qmWFyByiEAAAAAgOQQAAAAAEBbKQAAAACTsssuM80Paq5oCh+VQwAAAAAAySEAAAAAgLZSAAAAACZlMwzZDPPMEGqmWFyByiEAAAAAgOQQAAAAAEBbKQAAAACTsssw1YPnzRSLK1A5BAAAAACQHAIAAAAAaCsFAAAAYFJ2GbKZqJWTtlIAAAAAQLFHcggAAAAAoK0UAAAAgDkxW2nRonIIAAAAACA5BAAAAADQVgoAAADApGyGIZthnlZOM8XiClQOAQAAAAAkhwAAAAAA2koBAAAAmJT9f4tZmCkWV6ByCAAAAAAgOQQAAAAA0FYKAAAAwKRsMmQz0YPnzRSLK1A5BAAAAACQHAIAAAAAaCsFAAAAYFI24+JiFmaKxRWoHAIAAAAASA4BAAAAALSVAgAAADApu8z14HkzxeIKVA4BAAAAACSHAAAAAADaSgEAAACYlF0W2WRxdxgOdhPF4gpUDgEAAAAAJIcAAAAAANpKAQAAAJiU3bi4mIWZYnEFKocAAAAAAJJDAAAAAABtpQAAAABMymay2UrNFIsrUDkEAAAAAJAcAgAAAABoKwUAAABgUrSVFi0qhwAAAAAAkkMAAAAAAG2lAAAAAEzKblhkN8zTymmmWFyByiEAAAAAgOQQAAAAAEBbKQAAAACTYrbSokXlEAAAAABAcggAAAAAoK0UAAAAgEnZ5CGbiepZNncH4GLmeacBAAAAAG5DcggAAAAAoK0UAAAAgDkZhsVUD543TBSLK1A5BAAAAACQHAIAAAAASA4BAAAAmJRNFtMtBREfH6/69esrKChI4eHh6tq1q/bs2eM0pnfv3rJYLE5Lo0aNnMZkZ2erf//+KlWqlAICAtS5c2cdPnzYaUx6erp69uwpq9Uqq9Wqnj176tSpUwWKl+QQAAAAAFxg3bp1evbZZ7Vp0yatXLlSubm5atu2rbKyspzGtWvXTkePHnUsy5cvd9o+cOBALVmyRIsWLdL69et15swZdezYUTbbnw/X6NGjh5KSkpSYmKjExEQlJSWpZ8+eBYqXCWkAAAAAwAUSExOdfp43b57Cw8O1bds2NWvWzLHex8dHkZGRVzxGRkaG5s6dqwULFqh169aSpA8//FDR0dFatWqV4uLilJycrMTERG3atEkNGzaUJM2ZM0eNGzfWnj17VK1atXzFS+UQAAAAgCnZDA/TLZKUmZnptGRnZ+frfDIyMiRJYWFhTuvXrl2r8PBwVa1aVX379tXx48cd27Zt26acnBy1bdvWsS4qKkq1atXShg0bJEkbN26U1Wp1JIaS1KhRI1mtVseY/CA5BAAAAIACiI6OdtzbZ7VaFR8ff919DMPQoEGDdOedd6pWrVqO9e3bt9fChQu1evVqvfnmm9qyZYvuuusuR8KZmpoqb29vhYaGOh0vIiJCqampjjHh4eF5XjM8PNwxJj9oKwUAAACAAkhJSVFwcLDjZx8fn+vu89xzz2nHjh1av3690/ru3bs7/lyrVi3Vq1dP5cuX11dffaV77733qsczDEMWy58T5Pz1z1cbcz0khwAAAABMyS6L7CZqdrTLkCQFBwc7JYfX079/fy1btkz/+c9/dMstt1xzbJkyZVS+fHnt3btXkhQZGakLFy4oPT3dqXp4/PhxNWnSxDHm2LFjeY6VlpamiIiIfMdpnncaAAAAAIoRwzD03HPPafHixVq9erUqVqx43X1OnDihlJQUlSlTRpJ0xx13yMvLSytXrnSMOXr0qHbt2uVIDhs3bqyMjAxt3rzZMeb7779XRkaGY0x+UDmES9jtdh05ckRBQUEFKmUDAACgaBiGodOnTysqKkoeHtSMXOHZZ5/VRx99pM8//1xBQUGO+/+sVqv8/Px05swZjR49Wvfdd5/KlCmjgwcPatiwYSpVqpTuuecex9g+ffpo8ODBKlmypMLCwjRkyBDFxsY6Zi+tUaOG2rVrp759+2r27NmSpCeffFIdO3bM90ylEskhXOTIkSOKjo52dxgAAAC4jpSUlOu2OrrL33nwvCsVNJZZs2ZJklq0aOG0ft68eerdu7dKlCihnTt36oMPPtCpU6dUpkwZtWzZUp988omCgoIc46dOnSpPT09169ZN586dU6tWrZSQkKASJUo4xixcuFADBgxwzGrauXNnzZw5s0DxWgzDMAq0B5APGRkZCgkJ0W8/VFBwIN9EwbXuqRrr7hAAALjh5CpH67Vcp06dktVqdXc4TjIzM2W1WrVsx60KCCpx/R2KSNZpmzrX3qeMjIwC3XN4o6ByCJe41EoaHOih4CCSQ7iWp8XL3SEAAHDj+V+JiFuAcAnJIQAAAABT+uuD583AVsybLs3zTgMAAAAA3IbkEAAAAABAWykAAAAAc7LLIruJZis1UyyuQOUQAAAAAEByCAAAAACgrRQAAACASdnlIZuJ6ll2MVspAAAAAKCYIzkEAAAAANBWCgAAAMCcbIaHbIZ56lk2g7ZSAAAAAEAxR3IIAAAAAKCtFAAAAIA52eUhu4nqWcxWCgAAAAAo9kgOAQAAAAC0lQIAAAAwJ5thkc2wuDsMBzPF4gpUDgEAAAAAJIcAAAAAANpKAQAAAJiUTR6ymaieZWO2UgAAAABAcUdyCAAAAACgrRQAAACAOdkND9kN89Sz7AZtpQAAAACAYo7kEAAAAABAWykAAAAAc2K20qJlnncaAAAAAOA2JIcAAAAAANpKAQAAAJiTXZLNsLg7DAe7uwNwMSqHAAAAAACSQwAAAAAAbaUAAAAATMouD9lNVM8yUyyuULzPDgAAAACQLySHAAAAAADaSgEAAACYk83wkM0wTz3LTLG4QvE+OwAAAABAvpAcAgAAAABoKwUAAABgTnZZZJfF3WE4mCkWV6ByCAAAAAAgOQQAAAAA0FYKAAAAwKSYrbRoFe+zAwAAAADkC8khAAAAAIC2UgAAAADmZJOHbCaqZ5kpFlco3mcHAAAAAMgXkkMAAAAAAG2lAAAAAMzJblhkN8zz4HkzxeIKVA4BAAAAACSHAAAAAADaSgEAAACYlN1ks5XaTRSLKxTvswMAAAAA5AvJIQAAAACAtlIAAAAA5mQ3PGQ3zFPPMlMsrlC8zw4AAAAAkC8khwAAAAAA2koBAAAAmJNNFtlkngfPmykWV6ByCAAAAAAgOQQAAAAA0FYKAAAAwKSYrbRoFe+zAwAAAADkC8khAAAAAIC2UgAAAADmZJO5Zgi1uTsAF6NyCAAAAAAgOQQAAAAA0FYKAAAAwKSYrbRoFe+zAwAAAADkC8khAAAAAIC2UgAAAADmZDM8ZDNRK6eZYnGF4n12AAAAAIB8oXIIFKJFM8L13fIQpfzqI29fu2LqnVWf4UcUXTnbMSYu6rYr7vvEq7/rgX5pkqQX76usHRsDnbY375yuYe/+5vj59KkSmjWirDausEqSGrfNUL9xvyvQWtyfwIN/qmOvP/TAM2kKC8/Rb7/46t2RUdq1OfD6OwIFxLWGosK1BhQOkkOgEO3YGKhOvf9Q1dvOypYrJUwqo2EP3ao5636Wr79dkvRx0i6nfbasDtbUwdG68+4Mp/XtH/5Dj76Y6vjZx9futH3is+X1x1EvjV+4T5L01tBoTe5fTq99cMAVp4ZionnndD095ohmDiurnzYH6O6eJzRu4QH1bVFNab97uzs8FCNcaygqXGvFmyGL7LK4OwwHw0SxuAJtpSbXu3dvWSwWWSwWeXl5KSIiQm3atNG//vUv2e1/JgsVKlRwjPPz81OFChXUrVs3rV69usCvefz4cT311FMqV66cfHx8FBkZqbi4OG3cuLEwT61YmvDRfrXtflIVqp3XrTXPa/DUQzr+u7f27vBzjAkLz3VaNn5jVZ2mZ1Sm/AWnY/n4GU7jAoL//Ps+tNdHW9cE64U3UhRT76xi6p3VwNdT9P0qq1J+9Smy88WN594n/9A3H4cp8aOSSvnVV++OKqu0I17q+OgJd4eGYoZrDUWFaw0oPCSHN4B27drp6NGjOnjwoL7++mu1bNlSzz//vDp27Kjc3FzHuNdee01Hjx7Vnj179MEHHygkJEStW7fW+PHjC/R69913n7Zv36758+frl19+0bJly9SiRQudPHmysE+t2MvKLCFJCgq5cqtnepqnNn8brLgH8/4HtmZxqB6oWUt9W1TTe2OidPbMn/9ck7cGKCDYpup1zzrW1bjjrAKCbdq9NaCQzwLFhaeXXVVqn9W2dUFO67etC1JMvSw3RYXiiGsNRYVrDShctJXeAC5V7ySpbNmyqlu3rho1aqRWrVopISFBTzzxhCQpKCjIMa5cuXJq1qyZypQpo5EjR+r+++9XtWrVJEnr1q3Tiy++qO3btyssLEy9evXSuHHj5OnpqVOnTmn9+vVau3atmjdvLkkqX768GjRo4IYzv7EZhvTe6LKq2eCMKlQ/f8UxK/8dJr9Am+7s4NxS2vLek4qMvqCw8Fwd/NlX/4ovo/27/TTxk4stpCfTPBVSKifP8UJK5Sg9jX/WuLLgMJtKeEqn/nC+Rk6leSo0PPcqewEFx7WGosK1VvwxW2nRKt5nV4zdddddqlOnjhYvXnzNcc8//7wMw9Dnn38uSfr999/VoUMH1a9fX9u3b9esWbM0d+5cjRs3TpIUGBiowMBALV26VNnZ2dc6tJPs7GxlZmY6LTe7t4eV1YFkP73yzm9XHfPNojDddU+6vH0Np/UdHj6pus0uJpUtup7SiDkH9eN/g5zaU6/U8W4YlmLeCY/CYDhfbrJYJBlXHAr8I1xrKCpca0DhIDm8gVWvXl0HDx685piwsDCFh4c7xr3zzjuKjo7WzJkzVb16dXXt2lVjxozRm2++KbvdLk9PTyUkJGj+/PkKCQlR06ZNNWzYMO3YseOarxMfHy+r1epYoqOjC+ksb0xvD784i+jkT39V6ai8FT5J2vl9gA7v81W7Hte/J6Jy7Dl5etn1+4GL9xOGlc5V+h9eecZlnPBUSGm+KcWVZZ4sIVuuFHrZNWItlUvFGYWKaw1FhWsNKFwkhzcwwzBksVy/TvTXccnJyWrcuLHTfk2bNtWZM2d0+PBhSRfvOTxy5IiWLVumuLg4rV27VnXr1lVCQsJVX+OVV15RRkaGY0lJSflnJ3eDMgxp5rCy+u5rqyb/36+KLHfhqmO/+bikqtQ+q1trXrnl9K9+2+Or3BwPlYy4mGjWqJelrMwS+vlHf8eYn3/wV1ZmCe6xwFXl5nho7w5/1W122ml93WanuVcVhYprDUWFa634sxsW0y3FGcnhDSw5OVkVK1a85pgTJ04oLS3NMe5KCaXxv16Mv6739fVVmzZtNHLkSG3YsEG9e/fWqFGjrvo6Pj4+Cg4OdlpuRjOH3aLVi8P08tu/yS/QrpPHPXXyuKeyzzm/51mnPfSfL6xXrBoeOeitD6dE6JftfkpN8dbmb4M07skKqlzrrGLqX0z8ylXJVr2WmZr2YrSSt/kreZu/pr0YrYatM5yeqQhcbvF7pdSux0m1ffCEoiuf11Ojf1d42Rx99UFJd4eGYoZrDUWFaw0oPNTbb1CrV6/Wzp079cILL1xz3FtvvSUPDw917dpVkhQTE6PPPvvMKUncsGGDgoKCVLZs2aseJyYmRkuXLi2s8IutL+eXkiS9eF8Vp/WDpx5S2+5/zva67vNQybCoZdf0PMfw9DKUtD5IS+eW1vksD5WKylHDVpl6eFCqSpT4c9xLM3/TrBFlNeyhWyVJjdpm6Nnxv7vgrFCcrFsWqqBQmx5+4ZjCwnP12x5fvfpIRR3nWWAoZFxrKCpca0DhITm8AWRnZys1NVU2m03Hjh1TYmKi4uPj1bFjRz366KOOcadPn1ZqaqpycnJ04MABffjhh3r//fcVHx+vypUrS5L69eunadOmqX///nruuee0Z88ejRo1SoMGDZKHh4dOnDihBx54QI8//rhq166toKAgbd26VZMnT1aXLl3c9RbcML45kpSvcR0eOaEOj1z5XsPwsjl6Y/Gv1z1GcKhNL808VJDwAEkXv8S49EUG4EpcaygqXGvFl00espmo2dFMsbgCyeENIDExUWXKlJGnp6dCQ0NVp04dTZ8+Xb169ZKHx58X6MiRIzVy5Eh5e3srMjJSjRo10rfffquWLVs6xpQtW1bLly/Xiy++qDp16igsLEx9+vTRq6++KunibKUNGzbU1KlTtW/fPuXk5Cg6Olp9+/bVsGHDivzcAQAAABQNi2FcPvkv8M9lZmbKarUq/ZdKCg4q3t+wwP3iom5zdwgAANxwco0crdXnysjIMN18EZd+lxz4XWf5BOadod1dss/kaFrTZaZ8zwoDlUMAAAAApmS2GULNFIsrUNIBAAAAAJAcAgAAAABoKwUAAABgUnZ5yG6iepaZYnGF4n12AAAAAIB8ITkEAAAAANBWCgAAAMCcbIZFNhPNEGqmWFyByiEAAAAAgOQQAAAAAEBbKQAAAACTshsWUz143kyxuAKVQwAAAAAAySEAAAAAgLZSAAAAACZlGB6yG+apZxkmisUVivfZAQAAAADyheQQAAAAAEBbKQAAAABzsskim8wzQ6iZYnEFKocAAAAAAJJDAAAAAABtpQAAAABMym6Y68HzdsPdEbgWlUMAAAAAAMkhAAAAAIC2UgAAAAAmZTc8ZDfRg+fNFIsrFO+zAwAAAADkC8khAAAAAIC2UgAAAADmZJdFdhM9eN5MsbgClUMAAAAAAMkhAAAAAIC2UgAAAAAmZTMsshnmaeU0UyyuQOUQAAAAAEByCAAAAACgrRQAAACASdkND1M9eN5MsbhC8T47AAAAAEC+kBwCAAAAAGgrBQAAAGBOdllkN9EMoXaZJxZXoHIIAAAAACA5BAAAAADQVgoAAADApAxZTNXKaZgoFlegcggAAAAAIDkEAAAAANBWCgAAAMCk7IbJZis1USyuQOUQAAAAAEByCAAAAACgrRQAAACASdkND9kN89SzzBSLKxTvswMAAAAA5AvJIQAAAACAtlIAAAAA5sRspUWLyiEAAAAAgOQQAAAAAFwhPj5e9evXV1BQkMLDw9W1a1ft2bPHaYxhGBo9erSioqLk5+enFi1a6KeffnIak52drf79+6tUqVIKCAhQ586ddfjwYacx6enp6tmzp6xWq6xWq3r27KlTp04VKF6SQwAAAACmZJfFdEtBrFu3Ts8++6w2bdqklStXKjc3V23btlVWVpZjzOTJkzVlyhTNnDlTW7ZsUWRkpNq0aaPTp087xgwcOFBLlizRokWLtH79ep05c0YdO3aUzWZzjOnRo4eSkpKUmJioxMREJSUlqWfPngWK12IYhlGgPYB8yMzMlNVqVfovlRQcxHcQcK24qNvcHQIAADecXCNHa/W5MjIyFBwc7O5wnFz6XbLTij7yCvB2dzgOOVkX9EXbuX/7PUtLS1N4eLjWrVunZs2ayTAMRUVFaeDAgXrppZckXawSRkREaNKkSXrqqaeUkZGh0qVLa8GCBerevbsk6ciRI4qOjtby5csVFxen5ORkxcTEaNOmTWrYsKEkadOmTWrcuLF+/vlnVatWLV/x8Vs7AAAAABRAZmam05KdnZ2v/TIyMiRJYWFhkqQDBw4oNTVVbdu2dYzx8fFR8+bNtWHDBknStm3blJOT4zQmKipKtWrVcozZuHGjrFarIzGUpEaNGslqtTrG5AfJIQAAAABTujRbqZkWSYqOjnbc22e1WhUfH3/dczEMQ4MGDdKdd96pWrVqSZJSU1MlSREREU5jIyIiHNtSU1Pl7e2t0NDQa44JDw/P85rh4eGOMfnBoywAAAAAoABSUlKc2kp9fHyuu89zzz2nHTt2aP369Xm2WSzO9zIahpFn3eUuH3Ol8fk5zl9ROQQAAACAAggODnZarpcc9u/fX8uWLdOaNWt0yy23ONZHRkZKUp7q3vHjxx3VxMjISF24cEHp6enXHHPs2LE8r5uWlpanKnktJIcAAAAATMndLaRXayvNL8Mw9Nxzz2nx4sVavXq1Klas6LS9YsWKioyM1MqVKx3rLly4oHXr1qlJkyaSpDvuuENeXl5OY44ePapdu3Y5xjRu3FgZGRnavHmzY8z333+vjIwMx5j8oK0UAAAAAFzg2Wef1UcffaTPP/9cQUFBjgqh1WqVn5+fLBaLBg4cqAkTJqhKlSqqUqWKJkyYIH9/f/Xo0cMxtk+fPho8eLBKliypsLAwDRkyRLGxsWrdurUkqUaNGmrXrp369u2r2bNnS5KefPJJdezYMd8zlUokhwAAAADgErNmzZIktWjRwmn9vHnz1Lt3b0nS0KFDde7cOfXr10/p6elq2LChVqxYoaCgIMf4qVOnytPTU926ddO5c+fUqlUrJSQkqESJEo4xCxcu1IABAxyzmnbu3FkzZ84sULw85xAuwXMOUZR4ziEAAAV3IzznMO7rJ033nMNv2r9nyvesMPBbOwAAAACA5BAAAAAAwD2HAAAAAEzq78wQ6kpmisUVqBwCAAAAAEgOAQAAAAC0lQIAAAAwKUOSXeZp5Szuj3mgcggAAAAAIDkEAAAAANBWCgAAAMCkmK20aFE5BAAAAACQHAIAAAAAaCsFAAAAYFK0lRYtkkO41D1VY+Vp8XJ3GCjmvjmS5O4QcJOIi7rN3SEAAOAytJUCAAAAAKgcAgAAADAn2kqLFpVDAAAAAADJIQAAAACAtlIAAAAAJkVbadGicggAAAAAIDkEAAAAANBWCgAAAMCkDMMiw0StnGaKxRWoHAIAAAAASA4BAAAAALSVAgAAADApuyyyyzytnGaKxRWoHAIAAAAASA4BAAAAALSVAgAAADApu2Ex1YPnzRSLK1A5BAAAAACQHAIAAAAAaCsFAAAAYFKGYTHVg+fNFIsrUDkEAAAAAJAcAgAAAABoKwUAAABgUsxWWrSoHAIAAAAASA4BAAAAALSVAgAAADApZistWlQOAQAAAAAkhwAAAAAA2koBAAAAmJRhstlKaSsFAAAAABR7JIcAAAAAANpKAQAAAJiTIckw3B3Fn0wUiktQOQQAAAAAkBwCAAAAAGgrBQAAAGBSdllkkXlmCLWbKBZXoHIIAAAAACA5BAAAAADQVgoAAADApAzDYqoHz5spFlegcggAAAAAIDkEAAAAANBWCgAAAMCk7IZFFhO1ctpNFIsrUDkEAAAAAJAcAgAAAABoKwUAAABgUoZxcTELM8XiClQOAQAAAAAkhwAAAAAA2koBAAAAmJRhWEz14HkzxeIKVA4BAAAAACSHAAAAAADaSgEAAACYFG2lRYvKIQAAAACA5BAAAAAAQFspAAAAAJOyGxZZTNTKaTdRLK5A5RAAAAAAQHIIAAAAAKCtFAAAAIBJGcbFxSzMFIsrUDkEAAAAAJAcAgAAAABoKwUAAABgUhfbSs0zQyhtpQAAAACAYo/kEAAAAABAWykAAAAAczIMi8naSs0TiytQOQQAAAAAkBwCAAAAAGgrBQAAAGBSxv8WszBTLK5A5RAAAAAAQHIIAAAAAKCtFAAAAIBJMVtp0aJyCAAAAAAgOQQAAAAA0FYKAAAAwKyYrrRIUTkEAAAAAJAcAgAAAABoKwUAAABgViabrVRmisUFqBwCAAAAAKgcAmbRsdcfeuCZNIWF5+i3X3z17sgo7doc6O6wYFKLZoTru+UhSvnVR96+dsXUO6s+w48ounK2Y0xc1G1X3PeJV3/XA/3SJEkv3ldZOzY6X2fNO6dr2Lu/OX4+faqEZo0oq40rrJKkxm0z1G/c7wq02gr5rFDc8LmGosK1BhQOKofFRO/evWWxWGSxWOTl5aWIiAi1adNG//rXv2S32/N9HJvNpvj4eFWvXl1+fn4KCwtTo0aNNG/ePBdGj+ad0/X0mCP6eHq4+rWtql3fB2jcwgMqXfaCu0ODSe3YGKhOvf/QtC/3Kn7RPtls0rCHbtX5s39+rH+ctMtpGTTlkCwWQ3feneF0rPYP/+E07vnJKU7bJz5bXvt+8tP4hfs0fuE+7fvJT5P7lyuS88SNi881FBWuteLNMMy3FGckh8VIu3btdPToUR08eFBff/21WrZsqeeff14dO3ZUbm5uvo4xevRoTZs2TWPHjtXu3bu1Zs0a9e3bV+np6S6O/uZ275N/6JuPw5T4UUml/Oqrd0eVVdoRL3V89IS7Q4NJTfhov9p2P6kK1c7r1prnNXjqIR3/3Vt7d/g5xoSF5zotG7+xqk7TMypT3vkXJh8/w2lcQPCfXygd2uujrWuC9cIbKYqpd1Yx9c5q4Osp+n6VVSm/+hTZ+eLGw+caigrXGlB4aCstRnx8fBQZGSlJKlu2rOrWratGjRqpVatWSkhI0BNPPKFDhw6pf//++vbbb+Xh4aF27dppxowZioiIkCR98cUX6tevnx544AHHcevUqeOW87lZeHrZVaX2WX0yM9xp/bZ1QYqpl+WmqHCjycosIUkKCrlyq2d6mqc2fxusIdN+y7NtzeJQrf4sVCGlc1S/5Wk9MjhV/oEXE8TkrQEKCLapet2zjvE17jirgGCbdm8NcGpjBS7hcw1FhWsNKFxUDou5u+66S3Xq1NHixYtlGIa6du2qkydPat26dVq5cqX27dun7t27O8ZHRkZq9erVSktLK9DrZGdnKzMz02lB/gSH2VTCUzr1h/N3NafSPBUanr+KL25uhiG9N7qsajY4owrVz19xzMp/h8kv0KY7Ozi3lLa896RefuegXv/sVz088JjWL7fqtT4VHdtPpnkqpFROnuOFlMpRehrfL+LK+FxDUeFaK/6M/81WaqalOON/9ptA9erVtWPHDq1atUo7duzQgQMHFB0dLUlasGCBatasqS1btqh+/fqaMmWK7r//fkVGRqpmzZpq0qSJunTpovbt21/zNeLj4zVmzJiiOJ1i6/IedotFUjHva0fheHtYWR1I9tObS/dedcw3i8J01z3p8vZ1vqg6PHzS8ecK1c+rbKVsPdeumvbu8FOV2uckSVf6b9AwLFdcD/wVn2soKlxrQOGgcngTMAxDFotFycnJio6OdiSGkhQTE6OQkBAlJyc7ft61a5c2bdqkxx57TMeOHVOnTp30xBNPXPM1XnnlFWVkZDiWlJSUa47HnzJPlpAtVwot7fwNp7VULpUZXNfbwy/OIjr5019VOipvhU+Sdn4foMP7fNWux/Xvv6kce06eXnb9fuDi/YRhpXOV/odXnnEZJzwVUppv5XFlfK6hqHCtAYWL5PAmkJycrIoVKzqSxMtdvt7Dw0P169fXCy+8oCVLlighIUFz587VgQMHrvoaPj4+Cg4OdlqQP7k5Htq7w191m512Wl+32Wnt3hrgpqhgdoYhzRxWVt99bdXk//tVkeWuPivfNx+XVJXaZ3VrzSu3nP7Vb3t8lZvjoZIRFxPNGvWylJVZQj//6O8Y8/MP/srKLMH9PLgqPtdQVLjWbgKGxXxLMcZXKsXc6tWrtXPnTr3wwgu65ZZbdOjQIaWkpDiqh7t371ZGRoZq1Khx1WPExMRIkrKy+EXQVRa/V0ovTk/RLzv8lLw1QB0eOaHwsjn66oOS7g4NJjVz2C1asyRUo+ftl1+gXSePX/w4Dwiyycfvz16qrNMe+s8XVj056kieYxw56K3Vi0PVoFWmgsNsOvSLj94bU1aVa51VTP2L/97LVclWvZaZmvZitJ6fdLEj4K2h0WrYOoPJaHBNfK6hqHCtAYWH5LAYyc7OVmpqqmw2m44dO6bExETFx8erY8eOevTRR+Xh4aHatWvr4Ycf1rRp05Sbm6t+/fqpefPmqlevniTp/vvvV9OmTdWkSRNFRkbqwIEDeuWVV1S1alVVr17dzWdYfK1bFqqgUJsefuGYwsJz9dseX736SEUd/93b3aHBpL6cX0qS9OJ9VZzWD556SG27/3kf4brPQyXDopZd8z6OxtPLUNL6IC2dW1rnszxUKipHDVtl6uFBqSpR4s9xL838TbNGlNWwh26VJDVqm6Fnx//ugrNCccLnGooK1xpQeCyGUdwf5Xhz6N27t+bPny9J8vT0VGhoqOrUqaMePXqoV69e8vC42EF8vUdZzJkzRx9//LF27dqljIwMRUZG6q677tLo0aNVvnz5fMeTmZkpq9WqFuoiT0ve+5WAwvTNkSR3h4CbRFzUbe4OAQAKTa6Ro7X6XBkZGaa7JejS75Ll3x8hD39fd4fjYD97Xr89MdaU71lhIDmES5AcoiiRHKKokBwCKE5IDguuuCeHTEgDAAAAAOCeQwAAAAAmZchcz6w0UywuQOUQAAAAAEByCAAAAACgrRQAAACASRmGRYaJHjxvplhcgcohAAAAAIDkEAAAAABAWykAAAAAMyvmM4SaCZVDAAAAAADJIQAAAACAtlIAAAAAJsVspUWLyiEAAAAAgOQQAAAAAEBbKQAAAACzMmSu2UrNFIsLUDkEAAAAAJAcAgAAAABoKwUAAABgWpb/LWZhplgKH5VDAAAAAADJIQAAAACAtlIAAAAAZsVspUWKyiEAAAAAgOQQAAAAAEBbKQAAAACzoq20SFE5BAAAAACQHAIAAAAAaCsFAAAAYFaG5eJiFmaKxQWoHAIAAAAASA4BAAAAALSVAgAAADApw7i4mIWZYnEFKocAAAAAAJJDAAAAAABtpQAAAADMypC5HjxvplhcgMohAAAAALjAf/7zH3Xq1ElRUVGyWCxaunSp0/bevXvLYrE4LY0aNXIak52drf79+6tUqVIKCAhQ586ddfjwYacx6enp6tmzp6xWq6xWq3r27KlTp04VOF6SQwAAAABwgaysLNWpU0czZ8686ph27drp6NGjjmX58uVO2wcOHKglS5Zo0aJFWr9+vc6cOaOOHTvKZrM5xvTo0UNJSUlKTExUYmKikpKS1LNnzwLHS1spAAAAAHMyLOZ68HwBY2nfvr3at29/zTE+Pj6KjIy84raMjAzNnTtXCxYsUOvWrSVJH374oaKjo7Vq1SrFxcUpOTlZiYmJ2rRpkxo2bChJmjNnjho3bqw9e/aoWrVq+Y6XyiEAAAAAFEBmZqbTkp2d/bePtXbtWoWHh6tq1arq27evjh8/7ti2bds25eTkqG3bto51UVFRqlWrljZs2CBJ2rhxo6xWqyMxlKRGjRrJarU6xuQXySEAAAAAFEB0dLTj/j6r1ar4+Pi/dZz27dtr4cKFWr16td58801t2bJFd911lyPZTE1Nlbe3t0JDQ532i4iIUGpqqmNMeHh4nmOHh4c7xuQXbaUAAAAATMliXFzM4lIsKSkpCg4Odqz38fH5W8fr3r2748+1atVSvXr1VL58eX311Ve69957r7qfYRiyWP5scf3rn682Jj+oHAIAAABAAQQHBzstfzc5vFyZMmVUvnx57d27V5IUGRmpCxcuKD093Wnc8ePHFRER4Rhz7NixPMdKS0tzjMmvfFUOp0+fnu8DDhgwoEABAAAAAACkEydOKCUlRWXKlJEk3XHHHfLy8tLKlSvVrVs3SdLRo0e1a9cuTZ48WZLUuHFjZWRkaPPmzWrQoIEk6fvvv1dGRoaaNGlSoNfPV3I4derUfB3MYrGQHAIAAAAoHIbM9eD5AsZy5swZ/frrr46fDxw4oKSkJIWFhSksLEyjR4/WfffdpzJlyujgwYMaNmyYSpUqpXvuuUeSZLVa1adPHw0ePFglS5ZUWFiYhgwZotjYWMfspTVq1FC7du3Ut29fzZ49W5L05JNPqmPHjgWaqVTKZ3J44MCBAh0UAAAAAG52W7duVcuWLR0/Dxo0SJLUq1cvzZo1Szt37tQHH3ygU6dOqUyZMmrZsqU++eQTBQUFOfaZOnWqPD091a1bN507d06tWrVSQkKCSpQo4RizcOFCDRgwwDGraefOna/5bMWr+dsT0ly4cEEHDhzQrbfeKk9P5rUBAAAAgL9q0aKFDOPq5cZvvvnmusfw9fXVjBkzNGPGjKuOCQsL04cffvi3YvyrAk9Ic/bsWfXp00f+/v6qWbOmDh06JOnivYYTJ078xwEBAAAAgKSLD50321KMFTg5fOWVV7R9+3atXbtWvr6+jvWtW7fWJ598UqjBAQAAAACKRoH7QZcuXapPPvlEjRo1cnpuRkxMjPbt21eowQEAAAAAikaBk8O0tDSFh4fnWZ+VlVXghywCAAAAwFXd4LOV3mgK3FZav359ffXVV46fLyWEc+bMUePGjQsvMgAAAABAkSlw5TA+Pl7t2rXT7t27lZubq7feeks//fSTNm7cqHXr1rkiRgAAAACAixW4ctikSRN99913Onv2rG699VatWLFCERER2rhxo+644w5XxAgAAADgZmSYcCnG/tYDCmNjYzV//vzCjgUAAAAA4CZ/Kzm02WxasmSJkpOTZbFYVKNGDXXp0kWenn/rcAAAAAAANytwNrdr1y516dJFqampqlatmiTpl19+UenSpbVs2TLFxsYWepAAAAAAbkJma+U0UywuUOB7Dp944gnVrFlThw8f1g8//KAffvhBKSkpql27tp588klXxAgAAAAAcLECVw63b9+urVu3KjQ01LEuNDRU48ePV/369Qs1OAAAAABA0Shw5bBatWo6duxYnvXHjx9X5cqVCyUoAAAAAJBhMd9SjOUrOczMzHQsEyZM0IABA/Tpp5/q8OHDOnz4sD799FMNHDhQkyZNcnW8AAAAAAAXyFdbaUhIiCyWP7NkwzDUrVs3xzrDuHhnZqdOnWSz2VwQJgAAAADAlfKVHK5Zs8bVcQAAAACAE4txcTELM8XiCvlKDps3b+7qOAAAAAAAbvS3n1p/9uxZHTp0SBcuXHBaX7t27X8cFAAAAACgaBU4OUxLS9Njjz2mr7/++orbuecQAAAAQKEwZK4Hz5spFhco8KMsBg4cqPT0dG3atEl+fn5KTEzU/PnzVaVKFS1btswVMQIAAAAAXKzAlcPVq1fr888/V/369eXh4aHy5curTZs2Cg4OVnx8vO6++25XxAkAAAAAcKECVw6zsrIUHh4uSQoLC1NaWpokKTY2Vj/88EPhRgcAAAAAKBIFTg6rVaumPXv2SJJuu+02zZ49W7///rveffddlSlTptADBAAAAAC4XoHbSgcOHKijR49KkkaNGqW4uDgtXLhQ3t7eSkhIKOz4AAAAAABFoMDJ4cMPP+z48+23366DBw/q559/Vrly5VSqVKlCDQ4AAADAzcsicz143uLuAFzsbz/n8BJ/f3/VrVu3MGIBAAAAALhJvpLDQYMG5fuAU6ZM+dvBAAAAAADcI1/J4Y8//pivg1ksxb3QCsCM4qJuc3cIuEmUCLG6OwTcJGynMtwdAoCbUL6SwzVr1rg6DgAAAABwZlguLmZhplhcoMCPsgAAAAAAFD8khwAAAACAfz5bKQAAAAC4hPG/xSzMFIsLUDkEAAAAAJAcAgAAAAD+ZnK4YMECNW3aVFFRUfrtt98kSdOmTdPnn39eqMEBAAAAuIkZJlyKsQInh7NmzdKgQYPUoUMHnTp1SjabTZIUEhKiadOmFXZ8AAAAAIAiUODkcMaMGZozZ46GDx+uEiVKONbXq1dPO3fuLNTgAAAAAABFo8CzlR44cEC33357nvU+Pj7KysoqlKAAAAAAwGJcXMzCTLG4QoErhxUrVlRSUlKe9V9//bViYmIKIyYAAAAAQBErcOXwxRdf1LPPPqvz58/LMAxt3rxZH3/8seLj4/X++++7IkYAAAAAgIsVODl87LHHlJubq6FDh+rs2bPq0aOHypYtq7feeksPPvigK2IEAAAAcDMy2wyhZorFBQqcHEpS37591bdvX/3xxx+y2+0KDw8v7LgAAAAAAEXobyWHl5QqVaqw4gAAAAAAuFGBk8OKFSvKYrFcdfv+/fv/UUAAAAAAIIm20iJW4ORw4MCBTj/n5OToxx9/VGJiol588cXCigsAAAAAUIQKnBw+//zzV1z/9ttva+vWrf84IAAAAABA0Svwcw6vpn379vrss88K63AAAAAAbnIWw3xLcVZoyeGnn36qsLCwwjocAAAAAKAIFbit9Pbbb3eakMYwDKWmpiotLU3vvPNOoQYHAAAAACgaBU4Ou3bt6vSzh4eHSpcurRYtWqh69eqFFRcAAACAm51hubiYhZlicYECJYe5ubmqUKGC4uLiFBkZ6aqYAAAAAABFrED3HHp6euqZZ55Rdna2q+IBAAAAALhBgSekadiwoX788UdXxAIAAAAAfzJMuBRjBb7nsF+/fho8eLAOHz6sO+64QwEBAU7ba9euXWjBAQAAAACKRr6Tw8cff1zTpk1T9+7dJUkDBgxwbLNYLDIMQxaLRTabrfCjBAAAAAC4VL6Tw/nz52vixIk6cOCAK+MBAAAAAEnme/C8mWJxhXwnh4Zx8Z0oX768y4IBAAAAALhHgSaksViK93M9AAAAAOBmVaAJaapWrXrdBPHkyZP/KCAAAAAAkGS+GULNFIsLFCg5HDNmjKxWq6tiAQAAAAC4SYGSwwcffFDh4eGuigUAAAAA4Cb5Tg653xAAAABAkTLZbKXFva003xPSXJqtFAAAAABQ/OS7cmi3210ZBwAAAADAjQp0zyEAAAAAFBlmKy1SBXrOIQAAAACgeCI5BAAAAADQVgoAAADApGgrLVJUDgEAAAAAJIcAAAAAANpKAQAAAJiUxbi4mIWZYnEFKocAAAAAAJJDAAAAAADJIQAAAABAJIcAAAAAAJEcAgAAAADEbKUAAAAAzMqQuR48b6ZYXIDKIQAAAACA5BAAAAAAQFspAAAAAJOyGOZ68LyZYnEFKocAAAAAAJJDAAAAAABtpQAAAADMrJi3cpoJlUMAAAAAAMkhAAAAAIC2UgAAAABmZchcbaVmisUFqBwCAAAAAEgOAQAAAAC0lQIAAAAwKYthrgfPmykWV6ByCAAAAAAgOQQAAAAA0FYKAAAAwKyYrbRIUTkEAAAAAJAcAgAAAABoKwUAAABgUsxWWrSoHAIAAAAASA4BAAAAALSVAgAAADArZistUlQOAQAAAABUDgGz6NjrDz3wTJrCwnP02y++endklHZtDnR3WCiGuNbwT9W645Tue/ywKtc8o5LhFzS2f4w2fltKklTC065HBxxU/WYnFXnLeWWd8VTSxhDNm1JRJ9N8HMfw9LLriaH71bzDcfn42JW0KURvj62iE8d8rvaywFXxuQYUDiqHuK61a9fKYrHo1KlT7g6l2GreOV1Pjzmij6eHq1/bqtr1fYDGLTyg0mUvuDs0FDNcaygMvv52HdgToFnjKufZ5uNrV+WYM/r43fLqf39djRsQo7IVzmnU2z85jXvqlX1q0uoPTRpSQ0N63iY/f5tGz9olD49i3rOFQsfnWjFnmHApxkgOi1jv3r1lsVg0ceJEp/VLly6VxWJxU1Rwt3uf/EPffBymxI9KKuVXX707qqzSjnip46Mn3B0aihmuNRSGrf8N0wfTK2rDqlJ5tp0946nhT9TWfxNL6/eD/tqzI1izxldWlVpnVLrMeUmSf2Cu2t6XqvcnV1LSxlDtTw7U6y9VV4UqWbqtcXpRnw5ucHyuAYWH5NANfH19NWnSJKWnF95/gBcu8O3YjcrTy64qtc9q27ogp/Xb1gUppl6Wm6JCccS1BncJCMqV3S6dybx4N0uVmqfl5WXohw2hjjEn03z0294A1bg9011h4gbE5xpQuEgO3aB169aKjIxUfHz8Vcd89tlnqlmzpnx8fFShQgW9+eabTtsrVKigcePGqXfv3rJarerbt68SEhIUEhKiL7/8UtWqVZO/v7/uv/9+ZWVlaf78+apQoYJCQ0PVv39/2Ww2x7E+/PBD1atXT0FBQYqMjFSPHj10/Phxl50/nAWH2VTCUzr1h/MtwKfSPBUanuumqFAcca3BHby87XrshQNa+1W4zmVdvPZCS+Uo54JFZzK9nMaeOuGl0FJ82Yn843Ot+LMY5luKM5JDNyhRooQmTJigGTNm6PDhw3m2b9u2Td26ddODDz6onTt3avTo0RoxYoQSEhKcxr3++uuqVauWtm3bphEjRkiSzp49q+nTp2vRokVKTEzU2rVrde+992r58uVavny5FixYoPfee0+ffvqp4zgXLlzQ2LFjtX37di1dulQHDhxQ7969C3RO2dnZyszMdFpQMMZlHzYWi4p9Xzvcg2sNRaWEp10vv5ksi4f09mt570+83MVrkVssUHB8rgGFg9lK3eSee+7RbbfdplGjRmnu3LlO26ZMmaJWrVo5Er6qVatq9+7dev31152StrvuuktDhgxx/Lx+/Xrl5ORo1qxZuvXWWyVJ999/vxYsWKBjx44pMDBQMTExatmypdasWaPu3btLkh5//HHHMSpVqqTp06erQYMGOnPmjAID8zfTV3x8vMaMGfO33oubXebJErLlSqGlnb/htJbKVXoa/0RReLjWUJRKeNr1ypRkRZQ9r1ceq+2oGkpS+h9e8vI2FBic41Q9tIblaHdSsDvCxQ2KzzWgcFE5dKNJkyZp/vz52r17t9P65ORkNW3a1Gld06ZNtXfvXqd20Hr16uU5pr+/vyMxlKSIiAhVqFDBKcmLiIhwahv98ccf1aVLF5UvX15BQUFq0aKFJOnQoUP5PpdXXnlFGRkZjiUlJSXf+97scnM8tHeHv+o2O+20vm6z09q9NcBNUaE44lpDUbmUGEaVP6dhfWJ1OsO5fXTvT0HKybHo9ianHOtCS2WrfJUsJf9Icoj843PtJuDumUlvstlK+UrFjZo1a6a4uDgNGzbMqSJoGEaemUuNy/slJAUE5P3Q8/Jy/g/YYrFccZ3dbpckZWVlqW3btmrbtq0+/PBDlS5dWocOHVJcXFyBJrnx8fGRjw/Ppvq7Fr9XSi9OT9EvO/yUvDVAHR45ofCyOfrqg5LuDg3FDNcaCoOvv01R5c45fo4oe16Vqp/R6QxPnTjuo2HTklW5xmmN7ldLJUrIcR/h6QxP5eZ46OwZT634LFJPvLhPmac8dTrDS0+8uF8H9wYoaWPo1V4WuCI+14DCQ3LoZhMnTtRtt92mqlWrOtbFxMRo/fr1TuM2bNigqlWrqkSJEoX6+j///LP++OMPTZw4UdHR0ZKkrVu3Fupr4PrWLQtVUKhND79wTGHhufptj69efaSijv/u7e7QUMxwraEwVKl5WpPm73D8/OTL+yVJK5dEaOHb5dX4rouPEHh7yQ9O+73Uq7Z2bgmRJL038VbZbBa9MiVZ3j52bd8UoinDaspu555DFAyfa0DhITl0s9jYWD388MOaMWOGY93gwYNVv359jR07Vt27d9fGjRs1c+ZMvfPOO4X++uXKlZO3t7dmzJihp59+Wrt27dLYsWML/XVwfV/OL6Uv5+d9ZhhQ2LjW8E/t3BKiDjHNrrr9WtsuybngoXfHV9a7468/UQ1wPXyuFWNma+U0UywuwD2HJjB27FinttG6devq3//+txYtWqRatWpp5MiReu211wo8g2h+lC5dWgkJCfq///s/xcTEaOLEiXrjjTcK/XUAAAAAmJvFuNLNbMA/lJmZKavVqhbqIk+L1/V3AIAbQIkQq7tDwE3CdirD3SHgJpBr5GitPldGRoaCg801GdSl3yWrPT9BJXx83R2Ogy37vPa8NcyU71lhoK0UAAAAgCmZ7cHzZorFFWgrBQAAAACQHAIAAAAAaCsFAAAAYFbMVlqkqBwCAAAAAEgOAQAAAAC0lQIAAAAwKWYrLVpUDgEAAAAAJIcAAAAAANpKAQAAAJgVs5UWKSqHAAAAAACSQwAAAAAAbaUAAAAAzIq20iJF5RAAAAAAQHIIAAAAAKCtFAAAAIBJWf63mIWZYnEFKocAAAAAAJJDAAAAAHCF//znP+rUqZOioqJksVi0dOlSp+2GYWj06NGKioqSn5+fWrRooZ9++slpTHZ2tvr3769SpUopICBAnTt31uHDh53GpKenq2fPnrJarbJarerZs6dOnTpV4HhJDgEAAACYk2HCpQCysrJUp04dzZw584rbJ0+erClTpmjmzJnasmWLIiMj1aZNG50+fdoxZuDAgVqyZIkWLVqk9evX68yZM+rYsaNsNptjTI8ePZSUlKTExEQlJiYqKSlJPXv2LFiw4p5DAAAAAHCJ9u3bq3379lfcZhiGpk2bpuHDh+vee++VJM2fP18RERH66KOP9NRTTykjI0Nz587VggUL1Lp1a0nShx9+qOjoaK1atUpxcXFKTk5WYmKiNm3apIYNG0qS5syZo8aNG2vPnj2qVq1avuOlcggAAAAABZCZmem0ZGdnF/gYBw4cUGpqqtq2betY5+Pjo+bNm2vDhg2SpG3btiknJ8dpTFRUlGrVquUYs3HjRlmtVkdiKEmNGjWS1Wp1jMkvkkMAAAAApmQxzLdIUnR0tOP+PqvVqvj4+AKfW2pqqiQpIiLCaX1ERIRjW2pqqry9vRUaGnrNMeHh4XmOHx4e7hiTX7SVAgAAAEABpKSkKDg42PGzj4/P3z6WxeL8gAzDMPKsu9zlY640Pj/HuRyVQwAAAAAogODgYKfl7ySHkZGRkpSnunf8+HFHNTEyMlIXLlxQenr6NcccO3Ysz/HT0tLyVCWvh+QQAAAAgDm5e2bSfzhb6bVUrFhRkZGRWrlypWPdhQsXtG7dOjVp0kSSdMcdd8jLy8tpzNGjR7Vr1y7HmMaNGysjI0ObN292jPn++++VkZHhGJNftJUCAAAAgAucOXNGv/76q+PnAwcOKCkpSWFhYSpXrpwGDhyoCRMmqEqVKqpSpYomTJggf39/9ejRQ5JktVrVp08fDR48WCVLllRYWJiGDBmi2NhYx+ylNWrUULt27dS3b1/Nnj1bkvTkk0+qY8eOBZqpVCI5BAAAAACX2Lp1q1q2bOn4edCgQZKkXr16KSEhQUOHDtW5c+fUr18/paenq2HDhlqxYoWCgoIc+0ydOlWenp7q1q2bzp07p1atWikhIUElSpRwjFm4cKEGDBjgmNW0c+fOV3224rVYDMMoxOIocFFmZqasVqtaqIs8LV7uDgcACkWJEKu7Q8BNwnYqw90h4CaQa+RorT5XRkaG0+QqZnDpd8maT01QCW9fd4fjYLtwXj/NHmbK96wwcM8hAAAAAIDkEAAAAADAPYcAAAAATOqvD543AzPF4gpUDgEAAAAAJIcAAAAAANpKAQAAAJhVIT94/h8zUywuQOUQAAAAAEByCAAAAACgrRQAAACASTFbadGicggAAAAAIDkEAAAAANBWCgAAAMCsmK20SFE5BAAAAACQHAIAAAAAaCsFAAAAYFLMVlq0qBwCAAAAAEgOAQAAAAC0lQIAAAAwK2YrLVJUDgEAAAAAJIcAAAAAANpKAQAAAJgVbaVFisohAAAAAIDkEAAAAABAWykAAAAAk7IY5nrwvJlicQUqhwAAAAAAkkMAAAAAAG2lAAAAAMyK2UqLFJVDAAAAAADJIQAAAACAtlIAAAAAJmUxDFkM8/RymikWV6ByCAAAAAAgOQQAAAAA0FYKAAAAwKyYrbRIUTkEAAAAAJAcAgAAAABoKwUAAABgUhbj4mIWZorFFagcAgAAAABIDgEAAAAAtJUCAAAAMCtmKy1SVA4BAAAAACSHAAAAAADaSgEAAACYFLOVFi2SQwAA8sl2KsPdIeAm8c2RJHeHgJtA5mm7Qqu6OwqYCW2lAAAAAAAqhwAAAABMitlKixSVQwAAAAAAySEAAAAAgLZSAAAAACbFbKVFi8ohAAAAAIDkEAAAAABAWykAAAAAs2K20iJF5RAAAAAAQHIIAAAAAKCtFAAAAICJFfcZQs2EyiEAAAAAgOQQAAAAAEBbKQAAAACzMoyLi1mYKRYXoHIIAAAAACA5BAAAAADQVgoAAADApCyGuWYrNVMsrkDlEAAAAABAcggAAAAAoK0UAAAAgFkZ/1vMwkyxuACVQwAAAAAAySEAAAAAgLZSAAAAACZlsV9czMJMsbgClUMAAAAAAMkhAAAAAIC2UgAAAABmxWylRYrKIQAAAACA5BAAAAAAQFspAAAAAJOyGBcXszBTLK5A5RAAAAAAQHIIAAAAAKCtFAAAAIBZGcbFxSzMFIsLUDkEAAAAAJAcAgAAAABoKwUAAABgUsxWWrSoHAIAAAAASA4BAAAAALSVAgAAADAr43+LWZgpFhegcggAAAAAIDkEAAAAANBWCgAAAMCkmK20aFE5BAAAAACQHAIAAAAAaCsFAAAAYFaGcXExCzPF4gJUDgEAAAAAJIcAAAAAANpKAQAAAJgUs5UWLSqHAAAAAACSQwAAAAAAbaUAAAAAzMr432IWZorFBagcAgAAAABIDgEAAAAAtJUCAAAAMClmKy1aVA4BAAAAACSHAAAAAADaSgEAAACYld24uJiFmWJxASqHAAAAAACSQwAAAAAAbaUAAAAAzMqQuR48b6ZYXIDKIQAAAACA5BAAAAAAQFspAAAAAJOyyFwPnre4OwAXo3IIAAAAACA5BAAAAADQVgoAAADArAzj4mIWZorFBagcAgAAAABIDgEAAAAAtJUCAAAAMCmLYbLZSk0UiytQOQQAAAAAkBwCAAAAAGgrBQAAAGBWxv8WszBTLC5A5RAAAAAAQHJYVCpUqKBp06a5OwyYWMdef2j+pmR9sX+HZib+oloNzrg7JBRTXGsoCiUjczR0xm/6v1279Pm+HXpn5R5Vjj3r7rBgYotmhKt/+6rqWiVW3WJravRjFZXyq4/TmLio2664/N87pfMczzCk4Q9XUlzUbdrwtfWKr3kh26JnWldTXNRt2rfLzyXnBdxI3Joc9u7dWxaLRRaLRZ6enipXrpyeeeYZpaenuzMsl9iyZYuefPJJd4cBk2reOV1Pjzmij6eHq1/bqtr1fYDGLTyg0mUvuDs0FDNcaygKgdZcTfl8r2y5Fr36SCU92by63hsTpazMEu4ODSa2Y2OgOvX+Q9O+3Kv4Rftks0nDHrpV58/++evqx0m7nJZBUw7JYjF0590ZeY63ZE5pWSzXfs2546JUMjKnsE8FhchiGKZbijO3Vw7btWuno0eP6uDBg3r//ff1xRdfqF+/fu4Oq9CVLl1a/v7+7g4DJnXvk3/om4/DlPhRSaX86qt3R5VV2hEvdXz0hLtDQzHDtYai0O3Z4/rjiLfefKGc9iT569hhbyWtD9LR33yuvzNuWhM+2q+23U+qQrXzurXmeQ2eekjHf/fW3h1/VvTCwnOdlo3fWFWn6RmVKe/8Bde+n3z12ezSGjTl0FVfb8vqIG1bF6S+I3932TkBNxq3J4c+Pj6KjIzULbfcorZt26p79+5asWKFY/u8efNUo0YN+fr6qnr16nrnnXec9j98+LAefPBBhYWFKSAgQPXq1dP333/v2D5r1izdeuut8vb2VrVq1bRgwQKn/S0Wi95//33dc8898vf3V5UqVbRs2TLH9rVr18pisejbb79VvXr15O/vryZNmmjPnj2OMfv27VOXLl0UERGhwMBA1a9fX6tWrXJ6ncvbSkePHq1y5crJx8dHUVFRGjBggGPbO++8oypVqsjX11cRERG6//77HdsSExN15513KiQkRCVLllTHjh21b98+x/aDBw/KYrFo8eLFatmypfz9/VWnTh1t3LjRKZ7vvvtOzZs3l7+/v0JDQxUXF+eo2BqGocmTJ6tSpUry8/NTnTp19Omnn179LxH/iKeXXVVqn9W2dUFO67etC1JMvSw3RYXiiGsNRaVR20z9st1Pw2cf1Cc7ftLbK/aofQ++gEDBXKo0B4XYrrg9Pc1Tm78NVtyDztfW+bMWTexXQc+OP6yw8Nyr7jvtxWgNnfGbfPyKdyUIKAi3J4d/tX//fiUmJsrLy0uSNGfOHA0fPlzjx49XcnKyJkyYoBEjRmj+/PmSpDNnzqh58+Y6cuSIli1bpu3bt2vo0KGy2+2SpCVLluj555/X4MGDtWvXLj311FN67LHHtGbNGqfXHTNmjLp166YdO3aoQ4cOevjhh3Xy5EmnMcOHD9ebb76prVu3ytPTU48//rhj25kzZ9ShQwetWrVKP/74o+Li4tSpUycdOnTlb6s+/fRTTZ06VbNnz9bevXu1dOlSxcbGSpK2bt2qAQMG6LXXXtOePXuUmJioZs2aOfbNysrSoEGDtGXLFn377bfy8PDQPffc4zjnv8Y7ZMgQJSUlqWrVqnrooYeUm3vxAzIpKUmtWrVSzZo1tXHjRq1fv16dOnWSzXbxw/fVV1/VvHnzNGvWLP3000964YUX9Mgjj2jdunVX/bvLzs5WZmam04L8CQ6zqYSndOoP58mDT6V5KvQq/6kBfwfXGopKmXIX1PHREzpywEfDelTUVx+U0jNjf1fr+09ef2dAF+8XfG90WdVscEYVqp+/4piV/w6TX6BNd3ZwbimdPbqsYuplqUm7K/8uYhjSGwPL6e6eJ1S1zrlCjx2FzG7CpRhz+6MsvvzySwUGBspms+n8+Yv/+KdMmSJJGjt2rN58803de++9kqSKFStq9+7dmj17tnr16qWPPvpIaWlp2rJli8LCwiRJlStXdhz7jTfeUO/evR1tqoMGDdKmTZv0xhtvqGXLlo5xvXv31kMPPSRJmjBhgmbMmKHNmzerXbt2jjHjx49X8+bNJUkvv/yy7r77bp0/f16+vr6qU6eO6tSp4xg7btw4LVmyRMuWLdNzzz2X55wPHTqkyMhItW7dWl5eXipXrpwaNGjg2BYQEKCOHTsqKChI5cuX1+233+7Y97777nM61ty5cxUeHq7du3erVq1ajvVDhgzR3XffLeli8luzZk39+uuvql69uiZPnqx69eo5VWFr1qwp6WLyOWXKFK1evVqNGzeWJFWqVEnr16/X7NmzHe/B5eLj4zVmzJgrbkP+XN7CbrGo2E+XDPfgWoOrWTykvTv8NG9iGUnSvl3+Kl/tvO5+9IRWfRrm5uhwI3h7WFkdSPbTm0v3XnXMN4vCdNc96fL2/fMDbOM3wUr6LkjvrNhz1f0+n1tKZ097qHv/Y4UaM1AcuL1y2LJlSyUlJen7779X//79FRcXp/79+ystLU0pKSnq06ePAgMDHcu4ceMcbZRJSUm6/fbbHYnh5ZKTk9W0aVOndU2bNlVycrLTutq1azv+HBAQoKCgIB0/fvyqY8qUufif3aUxWVlZGjp0qGJiYhQSEqLAwED9/PPPV60cPvDAAzp37pwqVaqkvn37asmSJY6qXps2bVS+fHlVqlRJPXv21MKFC3X27J+zu+3bt089evRQpUqVFBwcrIoVK0pSnte6VryXKodXsnv3bp0/f15t2rRxet8/+OADp/bVy73yyivKyMhwLCkpKVcdC2eZJ0vIliuFlnau3FhL5So9ze3f36AY4VpDUTl53FO//eLrtC5lr4/CmfgI+fD28LLauMKqyZ/+qtJRV54sZuf3ATq8z1ftLmtXTvouSEcPeuve6rFqH11H7aMvfnk/tm8FvXhfZceYn38IUMcKF7c/1qSGJOm59lX1+vPlXHhmgPm5/beBgIAAR7Vv+vTpatmypcaMGeOouM2ZM0cNGzZ02qdEiYs96H5+159y2HLZNFWGYeRZd6mN9a/7XN6m+dcxl/a/NObFF1/UN998ozfeeEOVK1eWn5+f7r//fl24cOX/BKOjo7Vnzx6tXLlSq1atUr9+/fT6669r3bp1CgoK0g8//KC1a9dqxYoVGjlypEaPHq0tW7YoJCREnTp1UnR0tObMmaOoqCjZ7XbVqlUrz2tdK95rvW+Xxnz11VcqW7as0zYfn6tPJODj43PN7bi63BwP7d3hr7rNTmtD4p9Tbddtdlobv7ny1NvA38G1hqKye0uAom/NdlpXtlK2jv/u7aaIcCMwjIuJ4YZEq17/9FdFlrv6lwnffFxSVWqf1a01nVtOuz93LM/9rU/dVV1Pjf5djdpebDPtN/awer/058y5J1K9NKzHrRr27kFVv53HrZiN2WYINVMsruD2yuHlRo0apTfeeEM2m01ly5bV/v37VblyZaflUrWsdu3aSkpKynN/4CU1atTQ+vXrndZt2LBBNWrUKNSY//vf/6p379665557FBsbq8jISB08ePCa+/j5+alz586aPn261q5dq40bN2rnzp2SJE9PT7Vu3VqTJ0/Wjh07dPDgQa1evVonTpxQcnKyXn31VbVq1Uo1atT4W4/9qF27tr799tsrbouJiZGPj48OHTqU532Pjo4u8Gshfxa/V0rtepxU2wdPKLryeT01+neFl83RVx+UdHdoKGa41lAUFr9XWtXrZunB/scUVSFbLe9JV4dHTmrZvFLuDg0mNnPYLVq9OEwvv/2b/ALtOnncUyePeyr7nPOX+lmnPfSfL6x5qobSxdlMK1Q/77RIUnjZHEeyGX5LjtP2sv/7IiOq/IWrViqBm4XbK4eXa9GihWrWrKkJEyZo9OjRGjBggIKDg9W+fXtlZ2dr69atSk9P16BBg/TQQw9pwoQJ6tq1q+Lj41WmTBn9+OOPioqKUuPGjfXiiy+qW7duqlu3rlq1aqUvvvhCixcvzjOT6D9VuXJlLV68WJ06dZLFYtGIESPyVB7/KiEhQTabTQ0bNpS/v78WLFggPz8/lS9fXl9++aX279+vZs2aKTQ0VMuXL5fdble1atUUGhqqkiVL6r333lOZMmV06NAhvfzyywWO95VXXlFsbKz69eunp59+Wt7e3lqzZo0eeOABlSpVSkOGDNELL7wgu92uO++8U5mZmdqwYYMCAwPVq1evf/JW4SrWLQtVUKhND79wTGHhufptj69efaQi37Kj0HGtoSj8st1fr/WpqMdeOaqHXzim1BRvvTsySmuWhLo7NJjYl/Mvfnnw4n1VnNYPnnpIbbv/WQhY93moZFjUsmvxey424G6mSw6lixPHPPbYY/r111/1/vvv6/XXX9fQoUMVEBCg2NhYDRw4UJLk7e2tFStWaPDgwerQoYNyc3MVExOjt99+W5LUtWtXvfXWW3r99dc1YMAAVaxYUfPmzVOLFi0KNd6pU6fq8ccfV5MmTVSqVCm99NJL15ytMyQkRBMnTtSgQYNks9kUGxurL774QiVLllRISIgWL16s0aNH6/z586pSpYo+/vhjx4QxixYt0oABA1SrVi1Vq1ZN06dPL/D5VK1aVStWrNCwYcPUoEED+fn5qWHDho5JecaOHavw8HDFx8dr//79CgkJUd26dTVs2LC//R7h+r6cX8rxHyPgSlxrKArfrwrW96uC3R0GbiDfHEnK17gOj5xQh0fy/2iU6x03MvpCvl8bbmDIXJOmmSkWF7AYRjFvnIVbZGZmymq1qoW6yNPidf0dAACAA8kKikLmabtCq+5XRkaGgoPN9WXOpd8lm905Up6evtffoYjk5p7Xf9a/Zsr3rDCY7p5DAAAAAEDRM2VbKQAAAADIMPI+oNedzBSLC1A5BAAAAAAXGD16tCwWi9MSGRnp2G4YhkaPHq2oqCj5+fmpRYsW+umnn5yOkZ2drf79+6tUqVIKCAhQ586ddfjwYZfES3IIAAAAAC5Ss2ZNHT161LFcenydJE2ePFlTpkzRzJkztWXLFkVGRqpNmzY6ffq0Y8zAgQO1ZMkSLVq0SOvXr9eZM2fUsWNH2Wy2Qo+VtlIAAAAApmQxLi5m8Xdi8fT0dKoWXmIYhqZNm6bhw4fr3nvvlSTNnz9fERER+uijj/TUU08pIyNDc+fO1YIFC9S6dWtJ0ocffqjo6GitWrVKcXFx/+h8LkflEAAAAAAKIDMz02nJzs6+6ti9e/cqKipKFStW1IMPPqj9+/dLkg4cOKDU1FS1bdvWMdbHx0fNmzfXhg0bJEnbtm1TTk6O05ioqCjVqlXLMaYwkRwCAAAAQAFER0fLarU6lvj4+CuOa9iwoT744AN98803mjNnjlJTU9WkSROdOHFCqampkqSIiAinfSIiIhzbUlNT5e3trdDQ0KuOKUy0lQIAAAAwJ5POVpqSkuL0nEMfH58rDm/fvr3jz7GxsWrcuLFuvfVWzZ8/X40aNZIkWSyWy17CyLMubxjXH/N3UDkEAAAAgAIIDg52Wq6WHF4uICBAsbGx2rt3r+M+xMsrgMePH3dUEyMjI3XhwgWlp6dfdUxhIjkEAAAAgCKQnZ2t5ORklSlTRhUrVlRkZKRWrlzp2H7hwgWtW7dOTZo0kSTdcccd8vLychpz9OhR7dq1yzGmMNFWCgAAAMCULPaLi1kUNJYhQ4aoU6dOKleunI4fP65x48YpMzNTvXr1ksVi0cCBAzVhwgRVqVJFVapU0YQJE+Tv768ePXpIkqxWq/r06aPBgwerZMmSCgsL05AhQxQbG+uYvbQwkRwCAAAAgAscPnxYDz30kP744w+VLl1ajRo10qZNm1S+fHlJ0tChQ3Xu3Dn169dP6enpatiwoVasWKGgoCDHMaZOnSpPT09169ZN586dU6tWrZSQkKASJUoUerwWwzDTHZ4oLjIzM2W1WtVCXeRp8XJ3OAAA3FC+OZLk7hBwE8g8bVdo1f3KyMhwmlzFDBy/SzZ8VZ6evu4OxyE397zWfj/OlO9ZYaByCAAAAMCcTDpbaXHFhDQAAAAAAJJDAAAAAABtpQAAAADMyvjfYhZmisUFqBwCAAAAAEgOAQAAAAC0lQIAAAAwKYthyGKiGULNFIsrUDkEAAAAAJAcAgAAAABoKwUAAABgVoZhrgfPmykWF6ByCAAAAAAgOQQAAAAA0FYKAAAAwKwMSXZ3B/EXxburlMohAAAAAIDkEAAAAAAg2koBAAAAmJTFMEz14HkzxeIKVA4BAAAAACSHAAAAAADaSgEAAACYlSFzPXjeRKG4ApVDAAAAAADJIQAAAACAtlIAAAAAZmUYJmsrNVEsLkDlEAAAAABAcggAAAAAoK0UAAAAgFnZJVncHcRf2N0dgGtROQQAAAAAkBwCAAAAAGgrBQAAAGBSFsOQxUQzhJopFlegcggAAAAAIDkEAAAAANBWCgAAAMCsDMNcD543UywuQOUQAAAAAEByCAAAAACgrRQAAACAWdFWWqSoHAIAAAAASA4BAAAAALSVAgAAADAr2kqLFJVDAAAAAADJIQAAAACAtlIAAAAAZmWXZHF3EH9hd3cArkXlEAAAAABAcggAAAAAoK0UAAAAgElZDEMWE80QaqZYXIHKIQAAAACA5BAAAAAAQFspAAAAALMyDHM9eN5MsbgAlUMAAAAAAMkhAAAAAIC2UgAAAABmZTcki4laOe0misUFqBwCAAAAAEgOAQAAAAC0lQIAAAAwK2YrLVJUDgEAAAAAJIcAAAAAANpKAQAAAJiWydpKZaZYCh+VQwAAAAAAySEAAAAAgLZSuIjxv/L/gpSZCg4OdnM0AADcWM64OwDcFM7kZkqKdvzeZkrMVlqkSA7hEqdPn5YkRUdHuzkSAAAAXMvp06dltVrdHQZMgOQQLhEVFaWUlBQFBQXJYrG4O5wbQmZmpqKjo5WSkkK1FS7FtYaiwrWGosK19vcYhqHTp08rKirK3aHAJEgO4RIeHh665ZZb3B3GDSk4OJj/2FAkuNZQVLjWUFS41grO9BVDuyFTzRBqN1EsLsCENAAAAAAAkkMAAAAAAG2lgGn4+Pho1KhR8vHxcXcoKOa41lBUuNZQVLjWijHDfnExCzPF4gIWw9Rz1wIAAAC42WRmZspqtap1uX7y9DBP0p9rz9aqQ+8oIyOjWN7fSlspAAAAAIC2UgAAAAAmZRjmevC8mWJxASqHAAAAAACSQwAAcGNZu3atLBaLTp065e5QYHIVKlTQtGnT3B0GcMMgOQQKQe/evWWxWGSxWOTl5aWIiAi1adNG//rXv2S3/zmrVYUKFRzj/Pz8VKFCBXXr1k2rV68u8GseP35cTz31lMqVKycfHx9FRkYqLi5OGzduLMxTww0iv9fg9dhsNsXHx6t69ery8/NTWFiYGjVqpHnz5rkwerjTpWtn4sSJTuuXLl0qi8XipqjgSn/9vPD09FS5cuX0zDPPKD093d2hFbotW7boySefdHcY+CfshvmWYozkECgk7dq109GjR3Xw4EF9/fXXatmypZ5//nl17NhRubm5jnGvvfaajh49qj179uiDDz5QSEiIWrdurfHjxxfo9e677z5t375d8+fP1y+//KJly5apRYsWOnnyZGGfGm4Q+b0Gr2X06NGaNm2axo4dq927d2vNmjXq27dvsfylEX/y9fXVpEmTCvXv+cKFC4V2LBS+v35evP/++/riiy/Ur18/d4dV6EqXLi1/f393hwHcMEgOgUJyqXpXtmxZ1a1bV8OGDdPnn3+ur7/+WgkJCY5xQUFBioyMVLly5dSsWTO99957GjFihEaOHKk9e/Y4xq1bt04NGjSQj4+PypQpo5dfftnxC/6pU6e0fv16TZo0SS1btlT58uXVoEEDvfLKK7r77ruL+tRhEvm5Bg8dOqQuXbooMDBQwcHB6tatm44dO+Y4xqVfEB944AFVrFhRderUUZ8+fTRo0CA3nRWKQuvWrRUZGan4+Pirjvnss89Us2ZN+fj4qEKFCnrzzTedtleoUEHjxo1T7969ZbVa1bdvXyUkJCgkJERffvmlqlWrJn9/f91///3KysrS/PnzVaFCBYWGhqp///6y2WyOY3344YeqV6+e4/OyR48eOn78uMvO/2Z06fPilltuUdu2bdW9e3etWLHCsX3evHmqUaOGfH19Vb16db3zzjtO+x8+fFgPPvigwsLCFBAQoHr16un77793bJ81a5ZuvfVWeXt7q1q1alqwYIHT/haLRe+//77uuece+fv7q0qVKlq2bJlj+6XW4W+//Vb16tWTv7+/mjRp4vT/5L59+9SlSxdFREQoMDBQ9evX16pVq5xe5/K20tGjRzs6bqKiojRgwADHtnfeeUdVqlSRr6+vIiIidP/99zu2JSYm6s4771RISIhKliypjh07at++fY7tBw8elMVi0eLFi9WyZUv5+/urTp06ebp5vvvuOzVv3lz+/v4KDQ1VXFyc40sZwzA0efJkVapUSX5+fqpTp44+/fTTq/8lAi5Acgi40F133aU6depo8eLF1xz3/PPPyzAMff7555Kk33//XR06dFD9+vW1fft2zZo1S3PnztW4ceMkSYGBgQoMDNTSpUuVnZ3t8vPAjeuv16BhGOratatOnjypdevWaeXKldq3b5+6d+/uGB8ZGanVq1crLS3NjVGjqJUoUUITJkzQjBkzdPjw4Tzbt23bpm7duunBBx/Uzp07NXr0aI0YMcLpiy9Jev3111WrVi1t27ZNI0aMkCSdPXtW06dP16JFi5SYmKi1a9fq3nvv1fLly7V8+XItWLBA7733ntMvwRcuXNDYsWO1fft2LV26VAcOHFDv3r1d+Rbc1Pbv36/ExER5eXlJkubMmaPhw4dr/PjxSk5O1oQJEzRixAjNnz9fknTmzBk1b95cR44c0bJly7R9+3YNHTrU0cK+ZMkSPf/88xo8eLB27dqlp556So899pjWrFnj9LpjxoxRt27dtGPHDnXo0EEPP/xwnu6X4cOH680339TWrVvl6empxx9/3LHtzJkz6tChg1atWqUff/xRcXFx6tSpkw4dOnTF8/z00081depUzZ49W3v37tXSpUsVGxsrSdq6dasGDBig1157TXv27FFiYqKaNWvm2DcrK0uDBg3Sli1b9O2338rDw0P33HNPnrb94cOHa8iQIUpKSlLVqlX10EMPOb7YTUpKUqtWrVSzZk1t3LhR69evV6dOnRxfjLz66quaN2+eZs2apZ9++kkvvPCCHnnkEa1bt65gf6HFzaXZSs20FGM8ygJwserVq2vHjh3XHBMWFqbw8HAdPHhQ0sVvL6OjozVz5kxZLBZVr15dR44c0UsvvaSRI0fK09NTCQkJ6tu3r959913VrVtXzZs314MPPqjatWsXwVnhRnLpGly1apV27NihAwcOKDo6WpK0YMEC1axZU1u2bFH9+vU1ZcoU3X///YqMjFTNmjXVpEkTdenSRe3bt3fzWcDV7rnnHt12220aNWqU5s6d67RtypQpatWqlSPhq1q1qnbv3q3XX3/dKWm76667NGTIEMfP69evV05OjqOKJEn333+/FixYoGPHjikwMFAxMTFq2bKl1qxZ4/ii4q8JQKVKlTR9+nQ1aNBAZ86cUWBgoKvegpvKl19+qcDAQNlsNp0/f17Sxb9nSRo7dqzefPNN3XvvvZKkihUravfu3Zo9e7Z69eqljz76SGlpadqyZYvCwsIkSZUrV3Yc+4033lDv3r0dbaqDBg3Spk2b9MYbb6hly5aOcb1799ZDDz0kSY4vJzZv3qx27do5xowfP17NmzeXJL388su6++67df78efn6+qpOnTqqU6eOY+y4ceO0ZMkSLVu2TM8991yecz506JAiIyPVunVreXl5qVy5cmrQoIFjW0BAgDp27KigoCCVL19et99+u2Pf++67z+lYc+fOVXh4uHbv3q1atWo51g8ZMsTRwTNmzBjVrFlTv/76q6pXr67JkyerXr16TlXYmjVrSrqYfE6ZMkWrV69W48aNJV289tevX6/Zs2c73gPA1agcAi5mGEa+JnX467jk5GQ1btzYab+mTZvqzJkzjm/177vvPse3tnFxcVq7dq3q1q2b55t84NK1lZycrOjoaEdiKEkxMTEKCQlRcnKy4+ddu3Zp06ZNeuyxx3Ts2DF16tRJTzzxhLvCRxGaNGmS5s+fr927dzutT05OVtOmTZ3WNW3aVHv37nVqB61Xr16eY/r7+zsSQ0mKiIhQhQoVnJK8iIgIp7bRH3/8UV26dFH58uUVFBSkFi1aSNJVK0IouJYtWyopKUnff/+9+vfvr7i4OPXv319paWlKSUlRnz59HF0qgYGBGjdunKONMikpSbfffrsjMbzc1a6XS58zl/z1y8yAgAAFBQXlaR/+65gyZcpIkmNMVlaWhg4d6vgcCwwM1M8//3zV6+SBBx7QuXPnVKlSJfXt21dLlixxVPXatGmj8uXLq1KlSurZs6cWLlyos2fPOvbdt2+fevTooUqVKik4OFgVK1aUlPeavFa8lyqHV7J7926dP39ebdq0cXrfP/jgA6f2VcDVSA4BF0tOTnb8J3I1J06cUFpammPclRJK439tDH9d7+vrqzZt2mjkyJHasGGDevfurVGjRhXyGeBGd+kavNoXFZev9/DwUP369fXCCy9oyZIlSkhI0Ny5c3XgwIGiDBtu0KxZM8XFxWnYsGFO66/1mfRXAQEBedZdalW85NKMupevu9Sel5WVpbZt2yowMFAffvihtmzZoiVLlkhikpvCFBAQoMqVK6t27dqaPn26srOzNWbMGMffw5w5c5SUlORYLn1pJEl+fn7XPf6VrpfL113rOrjSmEv7Xxrz4osv6rPPPtP48eP13//+V0lJSYqNjb3qdRIdHa09e/bo7bfflp+fn/r166dmzZopJydHQUFB+uGHH/Txxx+rTJkyGjlypOrUqeN4XEqnTp104sQJzZkzR99//73j/srLX+ta8V7rfbs05quvvnJ633fv3s19h4bc30bqtLj7DXEtkkPAhVavXq2dO3fmaUe53FtvvSUPDw917dpV0sXqzYYNG5x++dqwYYOCgoJUtmzZqx4nJiZGWVlZhRI7ioe/XoMxMTE6dOiQUlJSHNt3796tjIwM1ahR46rHiImJkSSurZvExIkT9cUXX2jDhg2OdTExMVq/fr3TuA0bNqhq1aoqUaJEob7+zz//rD/++EMTJ07U//t//0/Vq1dnMpoiMGrUKL3xxhuy2WwqW7as9u/fr8qVKzstl77ArF27tpKSkq46O3aNGjWueL1c63Pm7/jvf/+r3r1765577lFsbKwiIyMdt2dcjZ+fnzp37qzp06dr7dq12rhxo3bu3ClJ8vT0VOvWrTV58mTt2LFDBw8e1OrVq3XixAklJyfr1VdfVatWrVSjRo2/NbNv7dq19e23315xW0xMjHx8fHTo0KE87/tfuz0AV+OeQ6CQZGdnKzU1VTabTceOHVNiYqLi4+PVsWNHPfroo45xp0+fVmpqqnJycnTgwAF9+OGHev/99xUfH++4Z6Nfv36aNm2a+vfvr+eee0579uzRqFGjNGjQIHl4eOjEiRN64IEH9Pjjj6t27doKCgrS1q1bNXnyZHXp0sVdbwHc7HrXoIeHh2rXrq2HH35Y06ZNU25urvr166fmzZs72gHvv/9+NW3aVE2aNFFkZKQOHDigV155RVWrVlX16tXdfIYoCrGxsXr44Yc1Y8YMx7rBgwerfv36Gjt2rLp3766NGzdq5syZeWawLAzlypWTt7e3ZsyYoaefflq7du3S2LFjC/114KxFixaqWbOmJkyYoNGjR2vAgAEKDg5W+/btlZ2dra1btyo9PV2DBg3SQw89pAkTJqhr166Kj49XmTJl9OOPPyoqKkqNGzfWiy++qG7duqlu3bpq1aqVvvjiCy1evDjPTKL/VOXKlbV48WJ16tRJFotFI0aMuOZzXRMSEmSz2dSwYUP5+/trwYIF8vPzU/ny5fXll19q//79atasmUJDQ7V8+XLZ7XZVq1ZNoaGhKlmypN577z2VKVNGhw4d0ssvv1zgeF955RXFxsaqX79+evrpp+Xt7a01a9bogQceUKlSpTRkyBC98MILstvtuvPOO5WZmakNGzYoMDBQvXr1+idvFZBvVA6BQpKYmKgyZcqoQoUKateundasWaPp06fr888/d/pmfeTIkSpTpowqV66snj17KiMjQ99++61eeuklx5iyZctq+fLl2rx5s+rUqaOnn35affr00auvvirp4mylDRs21NSpU9WsWTPVqlVLI0aMUN++fTVz5swiP3eYw/WuQYvFoqVLlyo0NFTNmjVT69atValSJX3yySeOY8TFxemLL75Qp06dVLVqVfXq1UvVq1fXihUr5OnJ94k3i7Fjxzp1LtStW1f//ve/tWjRItWqVUsjR47Ua6+95pIZREuXLq2EhAT93//9n2JiYjRx4kS98cYbhf46yGvQoEGaM2eO4uLi9P777yshIUGxsbFq3ry5EhISHJVDb29vrVixQuHh4erQoYNiY2M1ceJEx/91Xbt21VtvvaXXX39dNWvW1OzZszVv3jzHvaOFZerUqQoNDVWTJk3UqVMnxcXFqW7dulcdHxISojlz5qhp06aOKt4XX3yhkiVLKiQkRIsXL9Zdd92lGjVq6N1339XHH3+smjVrysPDQ4sWLdK2bdtUq1YtvfDCC3r99dcLHG/VqlW1YsUKbd++XQ0aNFDjxo31+eefOz5bx44dq5EjRyo+Pl41atRwfB5f79aUYs/tbaQ312ylFuNKNw0AAAAAgJtkZmbKarWqdeST8vTwdnc4Drn2C1qV+p4yMjIUHBzs7nAKHZVDAAAAAAD3HAIAAAAwKbtd0tXvJS1y17ivtTigcggAAAAAIDkEAAAAANBWCgAAAMCszDZDqJlicQEqhwAAAAAAkkMAAAAAAMkhAKAYGT16tG677TbHz71791bXrl2LPI6DBw/KYrEoKSnpqmMqVKigadOm5fuYCQkJCgkJ+cexWSwWLV269B8fBwCKhLsfeH+lpRgjOQQAuFTv3r1lsVhksVjk5eWlSpUqaciQIcrKynL5a7/11ltKSEjI19j8JHQAABRnTEgDAHC5du3aad68ecrJydF///tfPfHEE8rKytKsWbPyjM3JyZGXl1ehvK7Vai2U4wAAcDOgcggAcDkfHx9FRkYqOjpaPXr00MMPP+xobbzUCvqvf/1LlSpVko+PjwzDUEZGhp588kmFh4crODhYd911l7Zv3+503IkTJyoiIkJBQUHq06ePzp8/77T98rZSu92uSZMmqXLlyvLx8VG5cuU0fvx4SVLFihUlSbfffrssFotatGjh2G/evHmqUaOGfH19Vb16db3zzjtOr7N582bdfvvt8vX1Vb169fTjjz8W+D2aMmWKYmNjFRAQoOjoaPXr109nzpzJM27p0qWqWrWqfH191aZNG6WkpDht/+KLL3THHXfI19dXlSpV0pgxY5Sbm1vgeADAFOyG+ZZijOQQAFDk/Pz8lJOT4/j5119/1b///W999tlnjrbOu+++W6mpqVq+fLm2bdumunXrqlWrVjp58qQk6d///rdGjRql8ePHa+vWrSpTpkyepO1yr7zyiiZNmqQRI0Zo9+7d+uijjxQRESHpYoInSatWrdLRo0e1ePFiSdKcOXM0fPhwjR8/XsnJyZowYYJGjBih+fPnS5KysrLUsWNHVatWTdu2bdPo0aM1ZMiQAr8nHh4emj59unbt2qX58+dr9erVGjp0qNOYs2fPavz48Zo/f76+++47ZWZm6sEHH3Rs/+abb/TII49owIAB2r17t2bPnq2EhARHAgwAwLXQVgoAKFKbN2/WRx99pFatWjnWXbhwQQsWLFDp0qUlSatXr9bOnTt1/Phx+fj4SJLeeOMNLV26VJ9++qmefPJJTZs2TY8//rieeOIJSdK4ceO0atWqPNXDS06fPq233npLM2fOVK9evSRJt956q+68805Jcrx2yZIlFRkZ6dhv7NixevPNN3XvvfdKulhhvJR49erVSwsXLpTNZtO//vUv+fv7q2bNmjp8+LCeeeaZAr0vAwcOdPy5YsWKGjt2rJ555hmnhDcnJ0czZ85Uw4YNJUnz589XjRo1tHnzZjVo0EDjx4/Xyy+/7Di/SpUqaezYsRo6dKhGjRpVoHgAADcfkkMAgMt9+eWXCgwMVG5urnJyctSlSxfNmDHDsb18+fKO5EyStm3bpjNnzqhkyZJOxzl37pz27dsnSUpOTtbTTz/ttL1x48Zas2bNFWNITk5Wdna2U1J6PWlpaUpJSVGfPn3Ut29fx/rc3FzH/YzJycmqU6eO/P39neIoqDVr1mjChAnavXu3MjMzlZubq/PnzysrK0sBAQGSJE9PT9WrV8+xT/Xq1RUSEqLk5GQ1aNBA27Zt05YtW5wqhTabTefPn9fZs2edYgSAG4Fh2GUYdneH4WCmWFyB5BAA4HItW7bUrFmz5OXlpaioqDwTzlxKfi6x2+0qU6aM1q5dm+dYf/dxDn5+fgXex26/+EvAnDlzHNW6S0qUKCFJMgphWvPffvtNHTp00NNPP62xY8cqLCxM69evV58+fZzab6WLj6K43KV1drtdY8aMcVQ5/8rX1/cfxwkAKN5IDgEALhcQEKDKlSvne3zdunWVmpoqT09PVahQ4YpjatSooU2bNunRRx91rNu0adNVj1mlShX5+fnp22+/dbSi/pW3t7eki5W2SyIiIlS2bFnt379fDz/88BWPGxMTowULFujcuXOOBPRacVzJ1q1blZubqzfffFMeHhenA/j3v/+dZ1xubq62bt2qBg0aSJL27NmjU6dOqXr16pIuvm979uwp0HsNAMAlJIcAANNp3bq1GjdurK5du2rSpEmqVq2ajhw5ouXLl6tr166qV6+enn/+efXq1Uv16tXTnXfeqYULF+qnn35SpUqVrnhMX19fvfTSSxo6dKi8vb3VtGlTpaWl6aefflKfPn0UHh4uPz8/JSYm6pZbbpGvr6+sVqtGjx6tAQMGKDg4WO3bt1d2dra2bt2q9PR0DRo0SD169NDw4cPVp08fvfrqqzp48KDeeOONAp3vrbfeqtzcXM2YMUOdOnXSd999p3fffTfPOC8vL/Xv31/Tp0+Xl5eXnnvuOTVq1MiRLI4cOVIdO3ZUdHS0HnjgAXl4eGjHjh3auXOnxo0bV/C/CABwN8NkM4QWQreImTFbKQDAdCwWi5YvX65mzZrp8ccfV9WqVfXggw/q4MGDjtlFu3fvrpEjR+qll17SHXfcod9+++26k8CMGDFCgwcP1siRI1WjRg11795dx48fl3Txfr7p06dr9uzZioqKUpcuXSRJTzzxhN5//30lJCQoNjZWzZs3V0JCguPRF4GBgfriiy+0e/du3X777Ro+fLgmTZpUoPO97bbbNGXKFE2aNEm1atXSwoULFR8fn2ecv7+/XnrpJfXo0UONGzeWn5+fFi1a5NgeFxenL7/8UitXrlT9+vXVqFEjTZkyReXLly9QPACAm5PFKIybJQAAAACgkGRmZspqtapVyKPytHi7OxyHXOOCvj31gTIyMhQcHOzucAodbaUAAAAAzMkwJJmollXM62q0lQIAAAAASA4BAAAAALSVAgAAADAru12ymOjB84aJYnEBKocAAAAAAJJDAAAAAABtpQAAAADMitlKixSVQwAAAAAAySEAAAAAgLZSAAAAACZl2O0yTDRbqcFspQAAAACA4o7kEAAAAABAWykAAAAAk2K20iJF5RAAAAAAQHIIAAAAAKCtFAAAAIBZ2Q3JYqJWTtpKAQAAAADFHckhAAAAAIC2UgAAAAAmZRiSTPTgedpKAQAAAADFHckhAAAAAIC2UgAAAADmZNgNGSaardSgrRQAAAAAUNyRHAIAAAAAaCsFAAAAYFKGXeaardREsbgAlUMAAAAAAMkhAAAAAIC2UgAAAAAmxWylRYvKIQAAAACA5BAAAAAAQHIIAAAAwKwMu/mWv+Gdd95RxYoV5evrqzvuuEP//e9/C/mNKhwkhwAAAADgIp988okGDhyo4cOH68cff9T/+3//T+3bt9ehQ4fcHVoeJIcAAAAA4CJTpkxRnz599MQTT6hGjRqaNm2aoqOjNWvWLHeHlgezlQIAAAAwpVzlSCaaIDRXOZKkzMxMp/U+Pj7y8fHJM/7ChQvatm2bXn75Zaf1bdu21YYNG1wX6N9EcggAAADAVLy9vRUZGan1qcvdHUoegYGBio6Odlo3atQojR49Os/YP/74QzabTREREU7rIyIilJqa6sow/xaSQwAAAACm4uvrqwMHDujChQvuDiUPwzBksVic1l2pavhXl4+/0jHMgOQQAAAAgOn4+vrK19fX3WH8I6VKlVKJEiXyVAmPHz+ep5poBkxIAwAAAAAu4O3trTvuuEMrV650Wr9y5Uo1adLETVFdHZVDAAAAAHCRQYMGqWfPnqpXr54aN26s9957T4cOHdLTTz/t7tDyIDkEAAAAABfp3r27Tpw4oddee01Hjx5VrVq1tHz5cpUvX97doeVhMQzDRJPDAgAAAADcgXsOAQAAAAAkhwAAAAAAkkMAAAAAgEgOAQAAAAAiOQQAAAAAiOQQAAAAACCSQwAAAACASA4BAAAAACI5BAAAAACI5BAAAAAAIJJDAAAAAICk/w/+8iqmHUvv8wAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 1000x1000 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn import svm\n",
    "svm_model = svm.SVC(kernel='rbf')\n",
    "svm_model.fit(scaled_X_train, y_train)\n",
    "y_pred_svm = svm_model.predict(scaled_X_val)\n",
    "\n",
    "print(evaluate_model(y_val, y_pred_svm))\n",
    "print(classification_report(y_val, y_pred_svm))\n",
    "plot_confusion_matrix(y_val, y_pred_svm, label)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "vVOY90PsADf6"
   },
   "source": [
    "### Decision Tree Classifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 163,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 784
    },
    "id": "z2lDgxtCL-xs",
    "outputId": "c1f48090-b978-4102-8f19-7bd1e2f16afb"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'accuracy_score': 0.9998805256869773, 'precision_score': 0.9979338842975207, 'recall_score': 0.9999090909090909, 'f1_score': 0.9989171927902395}\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       1.00      1.00      1.00      2750\n",
      "           1       1.00      1.00      1.00      2750\n",
      "           2       0.99      1.00      1.00       120\n",
      "           3       1.00      1.00      1.00      2750\n",
      "\n",
      "    accuracy                           1.00      8370\n",
      "   macro avg       1.00      1.00      1.00      8370\n",
      "weighted avg       1.00      1.00      1.00      8370\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1000x1000 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.tree import DecisionTreeClassifier\n",
    "\n",
    "decision_tree_model = DecisionTreeClassifier(criterion=\"entropy\")\n",
    "decision_tree_model.fit(scaled_X_train, y_train)\n",
    "decision_tree_y_pred = decision_tree_model.predict(scaled_X_val)\n",
    "\n",
    "print(evaluate_model(y_val, decision_tree_y_pred))\n",
    "print(classification_report(y_val, decision_tree_y_pred))\n",
    "plot_confusion_matrix(y_val, decision_tree_y_pred, label)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "hXeWC7ZRAqzo"
   },
   "source": [
    "### Random Forest Classifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 164,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 784
    },
    "id": "Q5RqYClOAqO5",
    "outputId": "a1a7a7dc-ed8c-428d-8123-5b01abbbbf83"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'accuracy_score': 1.0, 'precision_score': 1.0, 'recall_score': 1.0, 'f1_score': 1.0}\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       1.00      1.00      1.00      2750\n",
      "           1       1.00      1.00      1.00      2750\n",
      "           2       1.00      1.00      1.00       120\n",
      "           3       1.00      1.00      1.00      2750\n",
      "\n",
      "    accuracy                           1.00      8370\n",
      "   macro avg       1.00      1.00      1.00      8370\n",
      "weighted avg       1.00      1.00      1.00      8370\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1000x1000 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.ensemble import RandomForestClassifier\n",
    "\n",
    "random_forest_model = RandomForestClassifier(n_estimators=200,\n",
    "                                             criterion=\"entropy\")\n",
    "random_forest_model.fit(scaled_X_train, y_train)\n",
    "random_forest_y_pred = random_forest_model.predict(scaled_X_val)\n",
    "\n",
    "print(evaluate_model(y_val, random_forest_y_pred))\n",
    "print(classification_report(y_val, random_forest_y_pred))\n",
    "plot_confusion_matrix(y_val, random_forest_y_pred, label)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "3l2gTVzN6NQM"
   },
   "source": [
    "### Naive Bayes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 165,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 804
    },
    "id": "gDbMPzIp6Nvk",
    "outputId": "f366d3e7-2d74-4d79-e0cf-37e37bc0ed9c"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'accuracy_score': 0.9826762246117084, 'precision_score': 0.835946072052171, 'recall_score': 0.8214469696969697, 'f1_score': 0.8279090463229231}\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       1.00      1.00      1.00      2750\n",
      "           1       1.00      1.00      1.00      2750\n",
      "           2       0.37      0.31      0.34       120\n",
      "           3       0.97      0.98      0.97      2750\n",
      "\n",
      "    accuracy                           0.98      8370\n",
      "   macro avg       0.84      0.82      0.83      8370\n",
      "weighted avg       0.98      0.98      0.98      8370\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1000x1000 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.naive_bayes import GaussianNB\n",
    "naive_bayes_model = GaussianNB()\n",
    "naive_bayes_model.fit(scaled_X_train, y_train)\n",
    "y_pred_nb = naive_bayes_model.predict(scaled_X_val)\n",
    "\n",
    "print(evaluate_model(y_val, y_pred_nb))\n",
    "print(classification_report(y_val, y_pred_nb))\n",
    "plot_confusion_matrix(y_val, y_pred_nb, label)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "IDdyg9rG1Pmd"
   },
   "source": [
    "### XGBOOST"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 166,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 472
    },
    "id": "GdC4r06S1Vpb",
    "outputId": "ea8ba652-add5-4414-bf28-abce497cc890"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'accuracy_score': 1.0, 'precision_score': 1.0, 'recall_score': 1.0, 'f1_score': 1.0}\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       1.00      1.00      1.00      2750\n",
      "           1       1.00      1.00      1.00      2750\n",
      "           2       1.00      1.00      1.00       120\n",
      "           3       1.00      1.00      1.00      2750\n",
      "\n",
      "    accuracy                           1.00      8370\n",
      "   macro avg       1.00      1.00      1.00      8370\n",
      "weighted avg       1.00      1.00      1.00      8370\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1000x1000 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from xgboost import XGBClassifier\n",
    "\n",
    "xgboost_model = XGBClassifier()\n",
    "xgboost_model.fit(scaled_X_train, y_train)\n",
    "y_pred_xgb = xgboost_model.predict(scaled_X_val)\n",
    "\n",
    "print(evaluate_model(y_val, y_pred_xgb))\n",
    "print(classification_report(y_val, y_pred_xgb))\n",
    "plot_confusion_matrix(y_val, y_pred_xgb, label)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "_hjEAktDav8L"
   },
   "source": [
    "### Artificial Neural Network"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 137,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "yM6AOeSr2K0J",
    "outputId": "ea73ad97-8a3e-4e07-f317-af5ec75e51c9"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0., 0., 0., 1.],\n",
       "       [0., 1., 0., 0.],\n",
       "       [1., 0., 0., 0.],\n",
       "       ...,\n",
       "       [1., 0., 0., 0.],\n",
       "       [0., 0., 0., 1.],\n",
       "       [1., 0., 0., 0.]], dtype=float32)"
      ]
     },
     "execution_count": 137,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from tensorflow.keras.utils import to_categorical\n",
    "# from keras.utils import np_utils\n",
    "\n",
    "encoded_y_train = to_categorical(y_train)\n",
    "encoded_y_val = to_categorical(y_val)\n",
    "encoded_y_train"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "41FTp56n_51w"
   },
   "source": [
    "the input layer/first layer is the number of features in the dataset which is 18, second layer is 12, third is 8 and output is 4"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 167,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "dvv6-OxZQClI",
    "outputId": "67e4c9bf-7dca-4422-a72d-4d78f6c0fc46"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2.13.0\n",
      "Epoch 1/100\n",
      "589/589 [==============================] - 3s 3ms/step - loss: 0.4514 - accuracy: 0.8365 - precision_10: 0.9416 - recall_10: 0.7426 - val_loss: 0.1208 - val_accuracy: 0.9667 - val_precision_10: 0.9728 - val_recall_10: 0.9635\n",
      "Epoch 2/100\n",
      "589/589 [==============================] - 1s 3ms/step - loss: 0.0692 - accuracy: 0.9797 - precision_10: 0.9810 - recall_10: 0.9771 - val_loss: 0.0374 - val_accuracy: 0.9968 - val_precision_10: 0.9970 - val_recall_10: 0.9963\n",
      "Epoch 3/100\n",
      "589/589 [==============================] - 2s 3ms/step - loss: 0.0243 - accuracy: 0.9970 - precision_10: 0.9971 - recall_10: 0.9970 - val_loss: 0.0151 - val_accuracy: 0.9981 - val_precision_10: 0.9981 - val_recall_10: 0.9979\n",
      "Epoch 4/100\n",
      "589/589 [==============================] - 1s 2ms/step - loss: 0.0119 - accuracy: 0.9978 - precision_10: 0.9978 - recall_10: 0.9977 - val_loss: 0.0088 - val_accuracy: 0.9981 - val_precision_10: 0.9981 - val_recall_10: 0.9979\n",
      "Epoch 5/100\n",
      "589/589 [==============================] - 1s 3ms/step - loss: 0.0075 - accuracy: 0.9982 - precision_10: 0.9982 - recall_10: 0.9981 - val_loss: 0.0064 - val_accuracy: 0.9986 - val_precision_10: 0.9986 - val_recall_10: 0.9986\n",
      "Epoch 6/100\n",
      "589/589 [==============================] - 2s 3ms/step - loss: 0.0057 - accuracy: 0.9985 - precision_10: 0.9985 - recall_10: 0.9985 - val_loss: 0.0051 - val_accuracy: 0.9984 - val_precision_10: 0.9984 - val_recall_10: 0.9984\n",
      "Epoch 7/100\n",
      "589/589 [==============================] - 2s 4ms/step - loss: 0.0044 - accuracy: 0.9984 - precision_10: 0.9985 - recall_10: 0.9983 - val_loss: 0.0042 - val_accuracy: 0.9987 - val_precision_10: 0.9987 - val_recall_10: 0.9987\n",
      "Epoch 8/100\n",
      "589/589 [==============================] - 2s 3ms/step - loss: 0.0036 - accuracy: 0.9986 - precision_10: 0.9986 - recall_10: 0.9986 - val_loss: 0.0037 - val_accuracy: 0.9989 - val_precision_10: 0.9989 - val_recall_10: 0.9989\n",
      "Epoch 9/100\n",
      "589/589 [==============================] - 1s 2ms/step - loss: 0.0033 - accuracy: 0.9986 - precision_10: 0.9986 - recall_10: 0.9986 - val_loss: 0.0031 - val_accuracy: 0.9987 - val_precision_10: 0.9987 - val_recall_10: 0.9987\n",
      "Epoch 10/100\n",
      "589/589 [==============================] - 1s 2ms/step - loss: 0.0029 - accuracy: 0.9989 - precision_10: 0.9989 - recall_10: 0.9989 - val_loss: 0.0028 - val_accuracy: 0.9989 - val_precision_10: 0.9989 - val_recall_10: 0.9989\n",
      "Epoch 11/100\n",
      "589/589 [==============================] - 1s 2ms/step - loss: 0.0024 - accuracy: 0.9993 - precision_10: 0.9993 - recall_10: 0.9993 - val_loss: 0.0022 - val_accuracy: 0.9994 - val_precision_10: 0.9994 - val_recall_10: 0.9994\n",
      "Epoch 12/100\n",
      "589/589 [==============================] - 1s 2ms/step - loss: 0.0021 - accuracy: 0.9993 - precision_10: 0.9993 - recall_10: 0.9993 - val_loss: 0.0019 - val_accuracy: 0.9995 - val_precision_10: 0.9995 - val_recall_10: 0.9995\n",
      "Epoch 13/100\n",
      "589/589 [==============================] - 1s 2ms/step - loss: 0.0016 - accuracy: 0.9996 - precision_10: 0.9996 - recall_10: 0.9995 - val_loss: 0.0017 - val_accuracy: 0.9994 - val_precision_10: 0.9994 - val_recall_10: 0.9994\n",
      "Epoch 14/100\n",
      "589/589 [==============================] - 1s 2ms/step - loss: 0.0014 - accuracy: 0.9995 - precision_10: 0.9995 - recall_10: 0.9995 - val_loss: 0.0016 - val_accuracy: 0.9994 - val_precision_10: 0.9994 - val_recall_10: 0.9992\n",
      "Epoch 15/100\n",
      "589/589 [==============================] - 1s 2ms/step - loss: 0.0013 - accuracy: 0.9995 - precision_10: 0.9995 - recall_10: 0.9995 - val_loss: 0.0012 - val_accuracy: 0.9995 - val_precision_10: 0.9995 - val_recall_10: 0.9995\n",
      "Epoch 16/100\n",
      "589/589 [==============================] - 1s 2ms/step - loss: 0.0012 - accuracy: 0.9994 - precision_10: 0.9994 - recall_10: 0.9994 - val_loss: 0.0013 - val_accuracy: 0.9994 - val_precision_10: 0.9994 - val_recall_10: 0.9994\n",
      "Epoch 17/100\n",
      "589/589 [==============================] - 1s 2ms/step - loss: 0.0011 - accuracy: 0.9995 - precision_10: 0.9995 - recall_10: 0.9995 - val_loss: 9.2049e-04 - val_accuracy: 0.9997 - val_precision_10: 0.9997 - val_recall_10: 0.9997\n",
      "Epoch 18/100\n",
      "589/589 [==============================] - 1s 2ms/step - loss: 8.7245e-04 - accuracy: 0.9998 - precision_10: 0.9998 - recall_10: 0.9998 - val_loss: 8.6799e-04 - val_accuracy: 0.9997 - val_precision_10: 0.9997 - val_recall_10: 0.9997\n",
      "Epoch 19/100\n",
      "589/589 [==============================] - 1s 2ms/step - loss: 5.8801e-04 - accuracy: 0.9998 - precision_10: 0.9998 - recall_10: 0.9998 - val_loss: 0.0020 - val_accuracy: 0.9987 - val_precision_10: 0.9987 - val_recall_10: 0.9987\n",
      "Epoch 20/100\n",
      "589/589 [==============================] - 1s 2ms/step - loss: 7.8970e-04 - accuracy: 0.9997 - precision_10: 0.9997 - recall_10: 0.9997 - val_loss: 8.1442e-04 - val_accuracy: 0.9997 - val_precision_10: 0.9997 - val_recall_10: 0.9997\n",
      "Epoch 21/100\n",
      "589/589 [==============================] - 1s 2ms/step - loss: 6.7395e-04 - accuracy: 0.9998 - precision_10: 0.9998 - recall_10: 0.9998 - val_loss: 6.7011e-04 - val_accuracy: 0.9998 - val_precision_10: 0.9998 - val_recall_10: 0.9998\n",
      "Epoch 22/100\n",
      "589/589 [==============================] - 1s 2ms/step - loss: 4.6163e-04 - accuracy: 0.9998 - precision_10: 0.9998 - recall_10: 0.9998 - val_loss: 7.3090e-04 - val_accuracy: 0.9998 - val_precision_10: 0.9998 - val_recall_10: 0.9998\n",
      "Epoch 23/100\n",
      "589/589 [==============================] - 1s 2ms/step - loss: 5.7619e-04 - accuracy: 0.9998 - precision_10: 0.9998 - recall_10: 0.9998 - val_loss: 6.9942e-04 - val_accuracy: 0.9997 - val_precision_10: 0.9997 - val_recall_10: 0.9997\n",
      "Epoch 24/100\n",
      "589/589 [==============================] - 1s 2ms/step - loss: 4.2814e-04 - accuracy: 0.9999 - precision_10: 0.9999 - recall_10: 0.9999 - val_loss: 7.6095e-04 - val_accuracy: 0.9997 - val_precision_10: 0.9997 - val_recall_10: 0.9997\n",
      "Epoch 25/100\n",
      "589/589 [==============================] - 1s 2ms/step - loss: 5.0821e-04 - accuracy: 0.9997 - precision_10: 0.9997 - recall_10: 0.9997 - val_loss: 9.7435e-04 - val_accuracy: 0.9997 - val_precision_10: 0.9997 - val_recall_10: 0.9997\n",
      "Epoch 26/100\n",
      "589/589 [==============================] - 1s 2ms/step - loss: 4.6355e-04 - accuracy: 0.9998 - precision_10: 0.9998 - recall_10: 0.9998 - val_loss: 8.1432e-04 - val_accuracy: 0.9997 - val_precision_10: 0.9997 - val_recall_10: 0.9997\n",
      "Epoch 27/100\n",
      "589/589 [==============================] - 1s 2ms/step - loss: 4.4370e-04 - accuracy: 0.9998 - precision_10: 0.9998 - recall_10: 0.9998 - val_loss: 7.0623e-04 - val_accuracy: 0.9997 - val_precision_10: 0.9997 - val_recall_10: 0.9997\n",
      "Epoch 28/100\n",
      "589/589 [==============================] - 1s 2ms/step - loss: 2.9501e-04 - accuracy: 0.9999 - precision_10: 0.9999 - recall_10: 0.9999 - val_loss: 7.2764e-04 - val_accuracy: 0.9997 - val_precision_10: 0.9997 - val_recall_10: 0.9997\n",
      "Epoch 29/100\n",
      "589/589 [==============================] - 1s 2ms/step - loss: 4.6942e-04 - accuracy: 0.9998 - precision_10: 0.9998 - recall_10: 0.9998 - val_loss: 7.6776e-04 - val_accuracy: 0.9998 - val_precision_10: 0.9998 - val_recall_10: 0.9998\n",
      "Epoch 30/100\n",
      "589/589 [==============================] - 1s 2ms/step - loss: 9.2993e-04 - accuracy: 0.9998 - precision_10: 0.9998 - recall_10: 0.9998 - val_loss: 0.0018 - val_accuracy: 0.9989 - val_precision_10: 0.9989 - val_recall_10: 0.9989\n",
      "Epoch 31/100\n",
      "589/589 [==============================] - 1s 2ms/step - loss: 2.7463e-04 - accuracy: 0.9999 - precision_10: 0.9999 - recall_10: 0.9999 - val_loss: 6.0053e-04 - val_accuracy: 0.9998 - val_precision_10: 0.9998 - val_recall_10: 0.9998\n",
      "Epoch 32/100\n",
      "589/589 [==============================] - 1s 2ms/step - loss: 3.4381e-04 - accuracy: 0.9998 - precision_10: 0.9998 - recall_10: 0.9998 - val_loss: 5.8181e-04 - val_accuracy: 0.9998 - val_precision_10: 0.9998 - val_recall_10: 0.9998\n",
      "Epoch 33/100\n",
      "589/589 [==============================] - 1s 2ms/step - loss: 2.1308e-04 - accuracy: 0.9999 - precision_10: 0.9999 - recall_10: 0.9999 - val_loss: 5.5867e-04 - val_accuracy: 0.9998 - val_precision_10: 0.9998 - val_recall_10: 0.9998\n",
      "Epoch 34/100\n",
      "589/589 [==============================] - 1s 2ms/step - loss: 2.9502e-04 - accuracy: 0.9998 - precision_10: 0.9998 - recall_10: 0.9998 - val_loss: 0.0011 - val_accuracy: 0.9997 - val_precision_10: 0.9997 - val_recall_10: 0.9997\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 35/100\n",
      "589/589 [==============================] - 1s 2ms/step - loss: 3.5953e-04 - accuracy: 0.9998 - precision_10: 0.9998 - recall_10: 0.9998 - val_loss: 7.4174e-04 - val_accuracy: 0.9997 - val_precision_10: 0.9997 - val_recall_10: 0.9997\n",
      "Epoch 36/100\n",
      "589/589 [==============================] - 1s 2ms/step - loss: 1.6207e-04 - accuracy: 0.9999 - precision_10: 0.9999 - recall_10: 0.9999 - val_loss: 7.8872e-04 - val_accuracy: 0.9997 - val_precision_10: 0.9997 - val_recall_10: 0.9997\n",
      "Epoch 37/100\n",
      "589/589 [==============================] - 1s 2ms/step - loss: 3.2435e-04 - accuracy: 0.9998 - precision_10: 0.9998 - recall_10: 0.9998 - val_loss: 9.0093e-04 - val_accuracy: 0.9998 - val_precision_10: 0.9998 - val_recall_10: 0.9998\n",
      "Epoch 38/100\n",
      "589/589 [==============================] - 2s 3ms/step - loss: 2.6140e-04 - accuracy: 0.9998 - precision_10: 0.9998 - recall_10: 0.9998 - val_loss: 8.3835e-04 - val_accuracy: 0.9997 - val_precision_10: 0.9997 - val_recall_10: 0.9997\n",
      "Epoch 39/100\n",
      "589/589 [==============================] - 1s 2ms/step - loss: 5.1269e-04 - accuracy: 0.9999 - precision_10: 0.9999 - recall_10: 0.9999 - val_loss: 5.8099e-04 - val_accuracy: 0.9998 - val_precision_10: 0.9998 - val_recall_10: 0.9998\n",
      "Epoch 40/100\n",
      "589/589 [==============================] - 2s 3ms/step - loss: 1.2585e-04 - accuracy: 0.9999 - precision_10: 0.9999 - recall_10: 0.9999 - val_loss: 6.5897e-04 - val_accuracy: 0.9997 - val_precision_10: 0.9997 - val_recall_10: 0.9997\n",
      "Epoch 41/100\n",
      "589/589 [==============================] - 2s 3ms/step - loss: 2.7579e-04 - accuracy: 0.9999 - precision_10: 0.9999 - recall_10: 0.9999 - val_loss: 6.0996e-04 - val_accuracy: 0.9998 - val_precision_10: 0.9998 - val_recall_10: 0.9998\n",
      "Epoch 42/100\n",
      "589/589 [==============================] - 1s 2ms/step - loss: 1.5158e-04 - accuracy: 0.9999 - precision_10: 0.9999 - recall_10: 0.9999 - val_loss: 6.2914e-04 - val_accuracy: 0.9998 - val_precision_10: 0.9998 - val_recall_10: 0.9998\n",
      "Epoch 43/100\n",
      "589/589 [==============================] - 1s 2ms/step - loss: 4.1135e-04 - accuracy: 0.9998 - precision_10: 0.9998 - recall_10: 0.9998 - val_loss: 5.9591e-04 - val_accuracy: 0.9998 - val_precision_10: 0.9998 - val_recall_10: 0.9998\n",
      "Epoch 44/100\n",
      "589/589 [==============================] - 1s 2ms/step - loss: 1.6352e-04 - accuracy: 0.9999 - precision_10: 0.9999 - recall_10: 0.9999 - val_loss: 6.5005e-04 - val_accuracy: 0.9998 - val_precision_10: 0.9998 - val_recall_10: 0.9998\n",
      "Epoch 45/100\n",
      "589/589 [==============================] - 1s 2ms/step - loss: 9.2396e-05 - accuracy: 0.9999 - precision_10: 0.9999 - recall_10: 0.9999 - val_loss: 6.1991e-04 - val_accuracy: 0.9998 - val_precision_10: 0.9998 - val_recall_10: 0.9998\n",
      "Epoch 46/100\n",
      "589/589 [==============================] - 1s 2ms/step - loss: 9.7161e-05 - accuracy: 0.9999 - precision_10: 0.9999 - recall_10: 0.9999 - val_loss: 7.7349e-04 - val_accuracy: 0.9998 - val_precision_10: 0.9998 - val_recall_10: 0.9998\n",
      "Epoch 47/100\n",
      "589/589 [==============================] - 1s 2ms/step - loss: 2.3127e-04 - accuracy: 0.9999 - precision_10: 0.9999 - recall_10: 0.9999 - val_loss: 6.6113e-04 - val_accuracy: 0.9998 - val_precision_10: 0.9998 - val_recall_10: 0.9998\n",
      "Epoch 48/100\n",
      "589/589 [==============================] - 1s 2ms/step - loss: 9.5762e-05 - accuracy: 1.0000 - precision_10: 1.0000 - recall_10: 1.0000 - val_loss: 5.9188e-04 - val_accuracy: 0.9998 - val_precision_10: 0.9998 - val_recall_10: 0.9998\n",
      "Epoch 49/100\n",
      "589/589 [==============================] - 1s 2ms/step - loss: 1.0815e-04 - accuracy: 1.0000 - precision_10: 1.0000 - recall_10: 1.0000 - val_loss: 7.4886e-04 - val_accuracy: 0.9997 - val_precision_10: 0.9997 - val_recall_10: 0.9997\n",
      "Epoch 50/100\n",
      "589/589 [==============================] - 1s 2ms/step - loss: 2.9969e-04 - accuracy: 0.9999 - precision_10: 0.9999 - recall_10: 0.9999 - val_loss: 9.0022e-04 - val_accuracy: 0.9997 - val_precision_10: 0.9997 - val_recall_10: 0.9997\n",
      "Epoch 51/100\n",
      "589/589 [==============================] - 1s 2ms/step - loss: 7.7444e-05 - accuracy: 1.0000 - precision_10: 1.0000 - recall_10: 1.0000 - val_loss: 5.4553e-04 - val_accuracy: 0.9998 - val_precision_10: 0.9998 - val_recall_10: 0.9998\n",
      "Epoch 52/100\n",
      "589/589 [==============================] - 1s 2ms/step - loss: 7.0043e-05 - accuracy: 1.0000 - precision_10: 1.0000 - recall_10: 1.0000 - val_loss: 7.2160e-04 - val_accuracy: 0.9998 - val_precision_10: 0.9998 - val_recall_10: 0.9998\n",
      "Epoch 53/100\n",
      "589/589 [==============================] - 2s 3ms/step - loss: 8.8831e-05 - accuracy: 1.0000 - precision_10: 1.0000 - recall_10: 1.0000 - val_loss: 7.5695e-04 - val_accuracy: 0.9997 - val_precision_10: 0.9997 - val_recall_10: 0.9997\n",
      "Epoch 54/100\n",
      "589/589 [==============================] - 2s 3ms/step - loss: 1.0013e-04 - accuracy: 1.0000 - precision_10: 1.0000 - recall_10: 1.0000 - val_loss: 9.2501e-04 - val_accuracy: 0.9998 - val_precision_10: 0.9998 - val_recall_10: 0.9998\n",
      "Epoch 55/100\n",
      "589/589 [==============================] - 2s 3ms/step - loss: 5.9429e-05 - accuracy: 1.0000 - precision_10: 1.0000 - recall_10: 1.0000 - val_loss: 7.5237e-04 - val_accuracy: 0.9998 - val_precision_10: 0.9998 - val_recall_10: 0.9998\n",
      "Epoch 56/100\n",
      "589/589 [==============================] - 1s 3ms/step - loss: 2.7964e-04 - accuracy: 0.9999 - precision_10: 0.9999 - recall_10: 0.9999 - val_loss: 7.0171e-04 - val_accuracy: 0.9998 - val_precision_10: 0.9998 - val_recall_10: 0.9998\n",
      "Epoch 57/100\n",
      "589/589 [==============================] - 1s 3ms/step - loss: 6.3045e-05 - accuracy: 0.9999 - precision_10: 0.9999 - recall_10: 0.9999 - val_loss: 7.0216e-04 - val_accuracy: 0.9998 - val_precision_10: 0.9998 - val_recall_10: 0.9998\n",
      "Epoch 58/100\n",
      "589/589 [==============================] - 2s 3ms/step - loss: 3.0254e-04 - accuracy: 0.9999 - precision_10: 0.9999 - recall_10: 0.9999 - val_loss: 6.8431e-04 - val_accuracy: 0.9998 - val_precision_10: 0.9998 - val_recall_10: 0.9998\n",
      "Epoch 59/100\n",
      "589/589 [==============================] - 2s 3ms/step - loss: 4.2847e-05 - accuracy: 1.0000 - precision_10: 1.0000 - recall_10: 1.0000 - val_loss: 6.3500e-04 - val_accuracy: 0.9998 - val_precision_10: 0.9998 - val_recall_10: 0.9998\n",
      "Epoch 60/100\n",
      "589/589 [==============================] - 2s 3ms/step - loss: 3.8158e-05 - accuracy: 1.0000 - precision_10: 1.0000 - recall_10: 1.0000 - val_loss: 8.6772e-04 - val_accuracy: 0.9998 - val_precision_10: 0.9998 - val_recall_10: 0.9998\n",
      "Epoch 61/100\n",
      "589/589 [==============================] - 2s 3ms/step - loss: 4.9558e-05 - accuracy: 1.0000 - precision_10: 1.0000 - recall_10: 1.0000 - val_loss: 6.2872e-04 - val_accuracy: 0.9997 - val_precision_10: 0.9997 - val_recall_10: 0.9997\n",
      "Epoch 62/100\n",
      "589/589 [==============================] - 1s 2ms/step - loss: 1.9810e-04 - accuracy: 0.9999 - precision_10: 0.9999 - recall_10: 0.9999 - val_loss: 5.5068e-04 - val_accuracy: 0.9998 - val_precision_10: 0.9998 - val_recall_10: 0.9998\n",
      "Epoch 63/100\n",
      "589/589 [==============================] - 1s 3ms/step - loss: 2.7126e-05 - accuracy: 1.0000 - precision_10: 1.0000 - recall_10: 1.0000 - val_loss: 7.2986e-04 - val_accuracy: 0.9997 - val_precision_10: 0.9997 - val_recall_10: 0.9997\n",
      "Epoch 64/100\n",
      "589/589 [==============================] - 1s 2ms/step - loss: 2.5618e-05 - accuracy: 1.0000 - precision_10: 1.0000 - recall_10: 1.0000 - val_loss: 7.3515e-04 - val_accuracy: 0.9997 - val_precision_10: 0.9997 - val_recall_10: 0.9997\n",
      "Epoch 65/100\n",
      "589/589 [==============================] - 2s 3ms/step - loss: 2.5661e-05 - accuracy: 1.0000 - precision_10: 1.0000 - recall_10: 1.0000 - val_loss: 7.4253e-04 - val_accuracy: 0.9997 - val_precision_10: 0.9997 - val_recall_10: 0.9997\n",
      "Epoch 66/100\n",
      "589/589 [==============================] - 2s 3ms/step - loss: 2.6182e-05 - accuracy: 1.0000 - precision_10: 1.0000 - recall_10: 1.0000 - val_loss: 8.2753e-04 - val_accuracy: 0.9997 - val_precision_10: 0.9997 - val_recall_10: 0.9997\n",
      "Epoch 67/100\n",
      "589/589 [==============================] - 2s 3ms/step - loss: 2.5245e-05 - accuracy: 1.0000 - precision_10: 1.0000 - recall_10: 1.0000 - val_loss: 5.9423e-04 - val_accuracy: 0.9998 - val_precision_10: 0.9998 - val_recall_10: 0.9998\n",
      "Epoch 68/100\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "589/589 [==============================] - 1s 2ms/step - loss: 2.2438e-05 - accuracy: 1.0000 - precision_10: 1.0000 - recall_10: 1.0000 - val_loss: 6.0986e-04 - val_accuracy: 0.9998 - val_precision_10: 0.9998 - val_recall_10: 0.9998\n",
      "Epoch 69/100\n",
      "589/589 [==============================] - 1s 2ms/step - loss: 1.3689e-04 - accuracy: 0.9999 - precision_10: 0.9999 - recall_10: 0.9999 - val_loss: 8.1065e-04 - val_accuracy: 0.9997 - val_precision_10: 0.9997 - val_recall_10: 0.9997\n",
      "Epoch 70/100\n",
      "589/589 [==============================] - 1s 2ms/step - loss: 7.0102e-04 - accuracy: 0.9998 - precision_10: 0.9998 - recall_10: 0.9998 - val_loss: 0.0013 - val_accuracy: 0.9997 - val_precision_10: 0.9997 - val_recall_10: 0.9997\n",
      "Epoch 71/100\n",
      "589/589 [==============================] - 1s 2ms/step - loss: 3.2808e-05 - accuracy: 1.0000 - precision_10: 1.0000 - recall_10: 1.0000 - val_loss: 7.8643e-04 - val_accuracy: 0.9998 - val_precision_10: 0.9998 - val_recall_10: 0.9998\n",
      "Epoch 72/100\n",
      "589/589 [==============================] - 1s 2ms/step - loss: 1.6962e-05 - accuracy: 1.0000 - precision_10: 1.0000 - recall_10: 1.0000 - val_loss: 7.9230e-04 - val_accuracy: 0.9998 - val_precision_10: 0.9998 - val_recall_10: 0.9998\n",
      "Epoch 73/100\n",
      "589/589 [==============================] - 1s 3ms/step - loss: 1.7546e-05 - accuracy: 1.0000 - precision_10: 1.0000 - recall_10: 1.0000 - val_loss: 6.9242e-04 - val_accuracy: 0.9998 - val_precision_10: 0.9998 - val_recall_10: 0.9998\n",
      "Epoch 74/100\n",
      "589/589 [==============================] - 2s 3ms/step - loss: 1.4007e-05 - accuracy: 1.0000 - precision_10: 1.0000 - recall_10: 1.0000 - val_loss: 6.5633e-04 - val_accuracy: 0.9998 - val_precision_10: 0.9998 - val_recall_10: 0.9998\n",
      "Epoch 75/100\n",
      "589/589 [==============================] - 1s 3ms/step - loss: 1.5173e-05 - accuracy: 1.0000 - precision_10: 1.0000 - recall_10: 1.0000 - val_loss: 5.4975e-04 - val_accuracy: 0.9998 - val_precision_10: 0.9998 - val_recall_10: 0.9998\n",
      "Epoch 76/100\n",
      "589/589 [==============================] - 1s 2ms/step - loss: 1.6152e-05 - accuracy: 1.0000 - precision_10: 1.0000 - recall_10: 1.0000 - val_loss: 6.3602e-04 - val_accuracy: 0.9997 - val_precision_10: 0.9997 - val_recall_10: 0.9997\n",
      "Epoch 77/100\n",
      "589/589 [==============================] - 1s 2ms/step - loss: 1.4551e-05 - accuracy: 1.0000 - precision_10: 1.0000 - recall_10: 1.0000 - val_loss: 6.2954e-04 - val_accuracy: 0.9997 - val_precision_10: 0.9997 - val_recall_10: 0.9997\n",
      "Epoch 78/100\n",
      "589/589 [==============================] - 2s 3ms/step - loss: 1.7316e-05 - accuracy: 1.0000 - precision_10: 1.0000 - recall_10: 1.0000 - val_loss: 6.4827e-04 - val_accuracy: 0.9998 - val_precision_10: 0.9998 - val_recall_10: 0.9998\n",
      "Epoch 79/100\n",
      "589/589 [==============================] - 1s 2ms/step - loss: 1.2116e-05 - accuracy: 1.0000 - precision_10: 1.0000 - recall_10: 1.0000 - val_loss: 4.1276e-04 - val_accuracy: 0.9998 - val_precision_10: 0.9998 - val_recall_10: 0.9998\n",
      "Epoch 80/100\n",
      "589/589 [==============================] - 1s 2ms/step - loss: 3.1761e-04 - accuracy: 0.9998 - precision_10: 0.9998 - recall_10: 0.9998 - val_loss: 5.9242e-04 - val_accuracy: 0.9997 - val_precision_10: 0.9997 - val_recall_10: 0.9997\n",
      "Epoch 81/100\n",
      "589/589 [==============================] - 1s 3ms/step - loss: 1.2373e-05 - accuracy: 1.0000 - precision_10: 1.0000 - recall_10: 1.0000 - val_loss: 5.6580e-04 - val_accuracy: 0.9997 - val_precision_10: 0.9997 - val_recall_10: 0.9997\n",
      "Epoch 82/100\n",
      "589/589 [==============================] - 1s 2ms/step - loss: 5.5091e-04 - accuracy: 0.9998 - precision_10: 0.9998 - recall_10: 0.9998 - val_loss: 0.0017 - val_accuracy: 0.9994 - val_precision_10: 0.9994 - val_recall_10: 0.9994\n",
      "Epoch 83/100\n",
      "589/589 [==============================] - 1s 3ms/step - loss: 1.7149e-04 - accuracy: 0.9999 - precision_10: 0.9999 - recall_10: 0.9999 - val_loss: 4.8578e-04 - val_accuracy: 0.9997 - val_precision_10: 0.9997 - val_recall_10: 0.9997\n",
      "Epoch 84/100\n",
      "589/589 [==============================] - 1s 2ms/step - loss: 1.8161e-05 - accuracy: 1.0000 - precision_10: 1.0000 - recall_10: 1.0000 - val_loss: 5.0448e-04 - val_accuracy: 0.9997 - val_precision_10: 0.9997 - val_recall_10: 0.9997\n",
      "Epoch 85/100\n",
      "589/589 [==============================] - 2s 3ms/step - loss: 9.9380e-06 - accuracy: 1.0000 - precision_10: 1.0000 - recall_10: 1.0000 - val_loss: 4.8508e-04 - val_accuracy: 0.9997 - val_precision_10: 0.9997 - val_recall_10: 0.9997\n",
      "Epoch 86/100\n",
      "589/589 [==============================] - 1s 2ms/step - loss: 8.1446e-06 - accuracy: 1.0000 - precision_10: 1.0000 - recall_10: 1.0000 - val_loss: 5.2983e-04 - val_accuracy: 0.9997 - val_precision_10: 0.9997 - val_recall_10: 0.9997\n",
      "Epoch 87/100\n",
      "589/589 [==============================] - 2s 3ms/step - loss: 7.4000e-06 - accuracy: 1.0000 - precision_10: 1.0000 - recall_10: 1.0000 - val_loss: 4.2754e-04 - val_accuracy: 0.9998 - val_precision_10: 0.9998 - val_recall_10: 0.9998\n",
      "Epoch 88/100\n",
      "589/589 [==============================] - 1s 3ms/step - loss: 7.3966e-06 - accuracy: 1.0000 - precision_10: 1.0000 - recall_10: 1.0000 - val_loss: 4.8441e-04 - val_accuracy: 0.9998 - val_precision_10: 0.9998 - val_recall_10: 0.9998\n",
      "Epoch 89/100\n",
      "589/589 [==============================] - 1s 2ms/step - loss: 7.6089e-06 - accuracy: 1.0000 - precision_10: 1.0000 - recall_10: 1.0000 - val_loss: 4.0341e-04 - val_accuracy: 0.9998 - val_precision_10: 0.9998 - val_recall_10: 0.9998\n",
      "Epoch 90/100\n",
      "589/589 [==============================] - 1s 2ms/step - loss: 6.5843e-06 - accuracy: 1.0000 - precision_10: 1.0000 - recall_10: 1.0000 - val_loss: 5.5212e-04 - val_accuracy: 0.9998 - val_precision_10: 0.9998 - val_recall_10: 0.9998\n",
      "Epoch 91/100\n",
      "589/589 [==============================] - 1s 2ms/step - loss: 6.9462e-06 - accuracy: 1.0000 - precision_10: 1.0000 - recall_10: 1.0000 - val_loss: 6.0778e-04 - val_accuracy: 0.9998 - val_precision_10: 0.9998 - val_recall_10: 0.9998\n",
      "Epoch 92/100\n",
      "589/589 [==============================] - 1s 2ms/step - loss: 4.0132e-04 - accuracy: 0.9999 - precision_10: 0.9999 - recall_10: 0.9999 - val_loss: 6.0586e-04 - val_accuracy: 0.9998 - val_precision_10: 0.9998 - val_recall_10: 0.9998\n",
      "Epoch 93/100\n",
      "589/589 [==============================] - 1s 2ms/step - loss: 6.2174e-06 - accuracy: 1.0000 - precision_10: 1.0000 - recall_10: 1.0000 - val_loss: 5.2957e-04 - val_accuracy: 0.9998 - val_precision_10: 0.9998 - val_recall_10: 0.9998\n",
      "Epoch 94/100\n",
      "589/589 [==============================] - 2s 3ms/step - loss: 5.3311e-06 - accuracy: 1.0000 - precision_10: 1.0000 - recall_10: 1.0000 - val_loss: 5.2760e-04 - val_accuracy: 0.9998 - val_precision_10: 0.9998 - val_recall_10: 0.9998\n",
      "Epoch 95/100\n",
      "589/589 [==============================] - 1s 2ms/step - loss: 4.9205e-06 - accuracy: 1.0000 - precision_10: 1.0000 - recall_10: 1.0000 - val_loss: 4.6589e-04 - val_accuracy: 0.9998 - val_precision_10: 0.9998 - val_recall_10: 0.9998\n",
      "Epoch 96/100\n",
      "589/589 [==============================] - 1s 2ms/step - loss: 4.5671e-06 - accuracy: 1.0000 - precision_10: 1.0000 - recall_10: 1.0000 - val_loss: 3.8326e-04 - val_accuracy: 0.9998 - val_precision_10: 0.9998 - val_recall_10: 0.9998\n",
      "Epoch 97/100\n",
      "589/589 [==============================] - 2s 3ms/step - loss: 4.5885e-06 - accuracy: 1.0000 - precision_10: 1.0000 - recall_10: 1.0000 - val_loss: 7.0668e-04 - val_accuracy: 0.9998 - val_precision_10: 0.9998 - val_recall_10: 0.9998\n",
      "Epoch 98/100\n",
      "589/589 [==============================] - 1s 2ms/step - loss: 4.2969e-06 - accuracy: 1.0000 - precision_10: 1.0000 - recall_10: 1.0000 - val_loss: 5.3399e-04 - val_accuracy: 0.9998 - val_precision_10: 0.9998 - val_recall_10: 0.9998\n",
      "Epoch 99/100\n",
      "589/589 [==============================] - 2s 3ms/step - loss: 4.8500e-06 - accuracy: 1.0000 - precision_10: 1.0000 - recall_10: 1.0000 - val_loss: 3.9258e-04 - val_accuracy: 0.9998 - val_precision_10: 0.9998 - val_recall_10: 0.9998\n",
      "Epoch 100/100\n",
      "589/589 [==============================] - 1s 3ms/step - loss: 4.1302e-06 - accuracy: 1.0000 - precision_10: 1.0000 - recall_10: 1.0000 - val_loss: 4.9367e-04 - val_accuracy: 0.9998 - val_precision_10: 0.9998 - val_recall_10: 0.9998\n"
     ]
    }
   ],
   "source": [
    "import tensorflow as tf\n",
    "print(tf.__version__)\n",
    "\n",
    "\n",
    "\n",
    "ann_model = tf.keras.models.Sequential()\n",
    "ann_model.add(tf.keras.layers.Dense(units=12, activation='relu'))\n",
    "ann_model.add(tf.keras.layers.Dense(units=8, activation='relu'))\n",
    "ann_model.add(tf.keras.layers.Dense(units=4, activation='softmax'))\n",
    "\n",
    "ann_model.compile(optimizer = 'adam', loss = 'categorical_crossentropy', metrics = ['accuracy', tf.keras.metrics.Precision(), tf.keras.metrics.Recall()])\n",
    "history = ann_model.fit(scaled_X_train, encoded_y_train, validation_split=0.25, batch_size=32, epochs=100)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 168,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "27wFeBW6Rf2e",
    "outputId": "33294a35-f79e-4118-c4d3-bf1a4c1a2616"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "262/262 [==============================] - 0s 2ms/step\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "array([[False, False, False,  True],\n",
       "       [False,  True, False, False],\n",
       "       [ True, False, False, False],\n",
       "       ...,\n",
       "       [ True, False, False, False],\n",
       "       [False,  True, False, False],\n",
       "       [ True, False, False, False]])"
      ]
     },
     "execution_count": 168,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ann_y_pred = ann_model.predict(scaled_X_val)\n",
    "ann_y_pred = (ann_y_pred > 0.5)\n",
    "ann_y_pred"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 169,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "6hdObG1cfUUI",
    "outputId": "4715a1ba-839f-41ec-dc0c-812035a46eaf"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "262/262 [==============================] - 0s 2ms/step - loss: 0.0015 - accuracy: 0.9998 - precision_10: 0.9998 - recall_10: 0.9998\n",
      "Accuracy: 99.98\n"
     ]
    }
   ],
   "source": [
    "\n",
    "# Check the test accuracy, precision, and recall\n",
    "_, accuracy, precision, recall = ann_model.evaluate(scaled_X_val, encoded_y_val)\n",
    "print('Accuracy: %.2f' % (accuracy*100))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 170,
   "metadata": {
    "id": "e9eU18hRiWJt"
   },
   "outputs": [],
   "source": [
    "# Performance Evaluations\n",
    "# print(history.history.keys())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 171,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 314
    },
    "id": "xzd6YOuvjegy",
    "outputId": "93c98d08-024e-44a7-9f75-430d3dabf941"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 0, 'epoch')"
      ]
     },
     "execution_count": 171,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot graph between training and validation loss\n",
    "plt.plot(history.history['loss'])\n",
    "plt.plot(history.history['val_loss'])\n",
    "plt.legend(['Training', 'Validation'])\n",
    "plt.title('Loss')\n",
    "plt.xlabel('epoch')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 172,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 314
    },
    "id": "98tL3BLOi_ah",
    "outputId": "ae7628e8-951f-442e-df4e-40c62ba06fd9"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 0, 'epoch')"
      ]
     },
     "execution_count": 172,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot graph between training and validation accuracy\n",
    "plt.plot(history.history['accuracy'])\n",
    "plt.plot(history.history['val_accuracy'])\n",
    "plt.legend(['Training', 'Validation'])\n",
    "plt.title('Accuracy')\n",
    "plt.xlabel('epoch')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 173,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 314
    },
    "id": "f7pMkrWSi_T0",
    "outputId": "68dcb793-dc56-42f3-a9bb-709e31fe5261"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 0, 'epoch')"
      ]
     },
     "execution_count": 173,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot graph between training and validation precision\n",
    "plt.plot(history.history['precision_10'])\n",
    "plt.plot(history.history['val_precision_10'])\n",
    "plt.legend(['Training', 'Validation'])\n",
    "plt.title('Precision')\n",
    "plt.xlabel('epoch')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 174,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 314
    },
    "id": "57cQgQkRi_Jk",
    "outputId": "926928a8-00c5-4880-f676-c412b9cb5602"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 0, 'epoch')"
      ]
     },
     "execution_count": 174,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot graph between training and validation recall\n",
    "plt.plot(history.history['recall_10'])\n",
    "plt.plot(history.history['val_recall_10'])\n",
    "plt.legend(['Training', 'Validation'])\n",
    "plt.title('Recall')\n",
    "plt.xlabel('epoch')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 175,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 886
    },
    "id": "25HKTkziQ4iu",
    "outputId": "db9d6d52-9005-456e-e4d0-3b80fb244cc7"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'accuracy_score': 0.9997610513739545, 'precision_score': 0.999818247909851, 'recall_score': 0.9978257575757576, 'f1_score': 0.9988176195189891}\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       1.00      1.00      1.00      2750\n",
      "           1       1.00      1.00      1.00      2750\n",
      "           2       1.00      0.99      1.00       120\n",
      "           3       1.00      1.00      1.00      2750\n",
      "\n",
      "   micro avg       1.00      1.00      1.00      8370\n",
      "   macro avg       1.00      1.00      1.00      8370\n",
      "weighted avg       1.00      1.00      1.00      8370\n",
      " samples avg       1.00      1.00      1.00      8370\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1000x1000 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "print(evaluate_model(encoded_y_val, ann_y_pred))\n",
    "print(classification_report(encoded_y_val, ann_y_pred))\n",
    "plot_confusion_matrix(encoded_y_val.argmax(axis=1), ann_y_pred.argmax(axis=1), label)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# implementing the cnn model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 190,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/30\n",
      "785/785 [==============================] - 6s 6ms/step - loss: 0.0986 - accuracy: 0.9734 - precision_12: 0.9832 - recall_12: 0.9609 - val_loss: 0.0084 - val_accuracy: 0.9989 - val_precision_12: 0.9989 - val_recall_12: 0.9989\n",
      "Epoch 2/30\n",
      "785/785 [==============================] - 4s 6ms/step - loss: 0.0082 - accuracy: 0.9986 - precision_12: 0.9986 - recall_12: 0.9986 - val_loss: 0.0038 - val_accuracy: 0.9994 - val_precision_12: 0.9994 - val_recall_12: 0.9994\n",
      "Epoch 3/30\n",
      "785/785 [==============================] - 5s 6ms/step - loss: 0.0050 - accuracy: 0.9988 - precision_12: 0.9988 - recall_12: 0.9988 - val_loss: 0.0026 - val_accuracy: 0.9994 - val_precision_12: 0.9994 - val_recall_12: 0.9994\n",
      "Epoch 4/30\n",
      "785/785 [==============================] - 5s 7ms/step - loss: 0.0035 - accuracy: 0.9991 - precision_12: 0.9991 - recall_12: 0.9991 - val_loss: 0.0034 - val_accuracy: 0.9994 - val_precision_12: 0.9994 - val_recall_12: 0.9994\n",
      "Epoch 5/30\n",
      "785/785 [==============================] - 5s 6ms/step - loss: 0.0048 - accuracy: 0.9990 - precision_12: 0.9990 - recall_12: 0.9990 - val_loss: 0.0087 - val_accuracy: 0.9982 - val_precision_12: 0.9982 - val_recall_12: 0.9982\n",
      "Epoch 6/30\n",
      "785/785 [==============================] - 4s 5ms/step - loss: 0.0035 - accuracy: 0.9991 - precision_12: 0.9991 - recall_12: 0.9991 - val_loss: 0.0023 - val_accuracy: 0.9995 - val_precision_12: 0.9995 - val_recall_12: 0.9995\n",
      "Epoch 7/30\n",
      "785/785 [==============================] - 4s 6ms/step - loss: 0.0024 - accuracy: 0.9993 - precision_12: 0.9993 - recall_12: 0.9993 - val_loss: 0.0024 - val_accuracy: 0.9993 - val_precision_12: 0.9993 - val_recall_12: 0.9993\n",
      "Epoch 8/30\n",
      "785/785 [==============================] - 4s 5ms/step - loss: 0.0030 - accuracy: 0.9991 - precision_12: 0.9991 - recall_12: 0.9991 - val_loss: 0.0015 - val_accuracy: 0.9998 - val_precision_12: 0.9998 - val_recall_12: 0.9998\n",
      "Epoch 9/30\n",
      "785/785 [==============================] - 4s 5ms/step - loss: 0.0021 - accuracy: 0.9994 - precision_12: 0.9994 - recall_12: 0.9994 - val_loss: 0.0016 - val_accuracy: 0.9998 - val_precision_12: 0.9998 - val_recall_12: 0.9998\n",
      "Epoch 10/30\n",
      "785/785 [==============================] - 4s 5ms/step - loss: 0.0018 - accuracy: 0.9995 - precision_12: 0.9995 - recall_12: 0.9995 - val_loss: 0.0017 - val_accuracy: 0.9996 - val_precision_12: 0.9996 - val_recall_12: 0.9996\n",
      "Epoch 11/30\n",
      "785/785 [==============================] - 4s 5ms/step - loss: 0.0020 - accuracy: 0.9994 - precision_12: 0.9994 - recall_12: 0.9994 - val_loss: 0.0013 - val_accuracy: 0.9996 - val_precision_12: 0.9996 - val_recall_12: 0.9996\n",
      "Epoch 12/30\n",
      "785/785 [==============================] - 5s 6ms/step - loss: 0.0021 - accuracy: 0.9994 - precision_12: 0.9994 - recall_12: 0.9994 - val_loss: 0.0019 - val_accuracy: 0.9993 - val_precision_12: 0.9993 - val_recall_12: 0.9993\n",
      "Epoch 13/30\n",
      "785/785 [==============================] - 4s 5ms/step - loss: 0.0020 - accuracy: 0.9995 - precision_12: 0.9995 - recall_12: 0.9995 - val_loss: 0.0015 - val_accuracy: 0.9996 - val_precision_12: 0.9996 - val_recall_12: 0.9996\n",
      "Epoch 14/30\n",
      "785/785 [==============================] - 4s 5ms/step - loss: 0.0018 - accuracy: 0.9995 - precision_12: 0.9995 - recall_12: 0.9995 - val_loss: 0.0014 - val_accuracy: 0.9994 - val_precision_12: 0.9994 - val_recall_12: 0.9994\n",
      "Epoch 15/30\n",
      "785/785 [==============================] - 4s 5ms/step - loss: 0.0019 - accuracy: 0.9994 - precision_12: 0.9994 - recall_12: 0.9994 - val_loss: 0.0011 - val_accuracy: 0.9995 - val_precision_12: 0.9995 - val_recall_12: 0.9995\n",
      "Epoch 16/30\n",
      "785/785 [==============================] - 4s 5ms/step - loss: 0.0020 - accuracy: 0.9994 - precision_12: 0.9994 - recall_12: 0.9994 - val_loss: 0.0011 - val_accuracy: 0.9994 - val_precision_12: 0.9994 - val_recall_12: 0.9994\n",
      "Epoch 17/30\n",
      "785/785 [==============================] - 4s 5ms/step - loss: 0.0016 - accuracy: 0.9994 - precision_12: 0.9994 - recall_12: 0.9994 - val_loss: 0.0013 - val_accuracy: 0.9994 - val_precision_12: 0.9994 - val_recall_12: 0.9994\n",
      "Epoch 18/30\n",
      "785/785 [==============================] - 3s 4ms/step - loss: 0.0012 - accuracy: 0.9996 - precision_12: 0.9996 - recall_12: 0.9996 - val_loss: 0.0018 - val_accuracy: 0.9990 - val_precision_12: 0.9990 - val_recall_12: 0.9990\n",
      "Epoch 19/30\n",
      "785/785 [==============================] - 4s 5ms/step - loss: 0.0019 - accuracy: 0.9993 - precision_12: 0.9993 - recall_12: 0.9993 - val_loss: 9.1615e-04 - val_accuracy: 0.9994 - val_precision_12: 0.9994 - val_recall_12: 0.9994\n",
      "Epoch 20/30\n",
      "785/785 [==============================] - 4s 5ms/step - loss: 0.0013 - accuracy: 0.9994 - precision_12: 0.9994 - recall_12: 0.9994 - val_loss: 0.0042 - val_accuracy: 0.9984 - val_precision_12: 0.9984 - val_recall_12: 0.9984\n",
      "Epoch 21/30\n",
      "785/785 [==============================] - 3s 4ms/step - loss: 0.0021 - accuracy: 0.9994 - precision_12: 0.9994 - recall_12: 0.9994 - val_loss: 6.9713e-04 - val_accuracy: 0.9998 - val_precision_12: 0.9998 - val_recall_12: 0.9998\n",
      "Epoch 22/30\n",
      "785/785 [==============================] - 4s 5ms/step - loss: 8.8225e-04 - accuracy: 0.9998 - precision_12: 0.9998 - recall_12: 0.9998 - val_loss: 7.6096e-04 - val_accuracy: 0.9996 - val_precision_12: 0.9996 - val_recall_12: 0.9996\n",
      "Epoch 23/30\n",
      "785/785 [==============================] - 3s 4ms/step - loss: 0.0011 - accuracy: 0.9996 - precision_12: 0.9996 - recall_12: 0.9996 - val_loss: 6.7229e-04 - val_accuracy: 0.9996 - val_precision_12: 0.9996 - val_recall_12: 0.9996\n",
      "Epoch 24/30\n",
      "785/785 [==============================] - 4s 5ms/step - loss: 0.0014 - accuracy: 0.9996 - precision_12: 0.9996 - recall_12: 0.9996 - val_loss: 5.6902e-04 - val_accuracy: 0.9998 - val_precision_12: 0.9998 - val_recall_12: 0.9998\n",
      "Epoch 25/30\n",
      "785/785 [==============================] - 4s 6ms/step - loss: 0.0012 - accuracy: 0.9996 - precision_12: 0.9996 - recall_12: 0.9996 - val_loss: 0.0013 - val_accuracy: 0.9995 - val_precision_12: 0.9995 - val_recall_12: 0.9995\n",
      "Epoch 26/30\n",
      "785/785 [==============================] - 4s 5ms/step - loss: 0.0015 - accuracy: 0.9993 - precision_12: 0.9993 - recall_12: 0.9993 - val_loss: 0.0020 - val_accuracy: 0.9996 - val_precision_12: 0.9996 - val_recall_12: 0.9996\n",
      "Epoch 27/30\n",
      "785/785 [==============================] - 4s 5ms/step - loss: 6.2230e-04 - accuracy: 0.9998 - precision_12: 0.9998 - recall_12: 0.9998 - val_loss: 4.2208e-04 - val_accuracy: 0.9998 - val_precision_12: 0.9998 - val_recall_12: 0.9998\n",
      "Epoch 28/30\n",
      "785/785 [==============================] - 4s 5ms/step - loss: 3.9810e-04 - accuracy: 0.9998 - precision_12: 0.9998 - recall_12: 0.9998 - val_loss: 0.0012 - val_accuracy: 0.9996 - val_precision_12: 0.9996 - val_recall_12: 0.9996\n",
      "Epoch 29/30\n",
      "785/785 [==============================] - 4s 5ms/step - loss: 5.1282e-04 - accuracy: 0.9998 - precision_12: 0.9998 - recall_12: 0.9998 - val_loss: 2.7588e-04 - val_accuracy: 0.9998 - val_precision_12: 0.9998 - val_recall_12: 0.9998\n",
      "Epoch 30/30\n",
      "785/785 [==============================] - 4s 5ms/step - loss: 2.8708e-04 - accuracy: 0.9999 - precision_12: 0.9999 - recall_12: 0.9999 - val_loss: 4.1350e-04 - val_accuracy: 0.9998 - val_precision_12: 0.9998 - val_recall_12: 0.9998\n"
     ]
    }
   ],
   "source": [
    "import tensorflow as tf\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.preprocessing import LabelEncoder, StandardScaler\n",
    "\n",
    "cnn_model = tf.keras.models.Sequential([\n",
    "    tf.keras.layers.Reshape(target_shape=(scaled_X_train.shape[1], 1), input_shape=(scaled_X_train.shape[1],)),\n",
    "    tf.keras.layers.Conv1D(64, kernel_size=3, activation='relu'),\n",
    "    tf.keras.layers.MaxPooling1D(pool_size=2),\n",
    "    tf.keras.layers.Flatten(),\n",
    "    tf.keras.layers.Dense(64, activation='relu'),\n",
    "    tf.keras.layers.Dense(4, activation='softmax')\n",
    "])\n",
    "\n",
    "cnn_model.compile(optimizer='adam', loss='categorical_crossentropy', metrics=['accuracy', tf.keras.metrics.Precision(), tf.keras.metrics.Recall()])\n",
    "\n",
    "cnn_history = cnn_model.fit(scaled_X_train, encoded_y_train, validation_data=(scaled_X_val, encoded_y_val), batch_size=32, epochs=30)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 191,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "262/262 [==============================] - 1s 2ms/step\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "array([[False, False, False,  True],\n",
       "       [False,  True, False, False],\n",
       "       [ True, False, False, False],\n",
       "       ...,\n",
       "       [ True, False, False, False],\n",
       "       [False,  True, False, False],\n",
       "       [ True, False, False, False]])"
      ]
     },
     "execution_count": 191,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cnn_y_pred = cnn_model.predict(scaled_X_val)\n",
    "cnn_y_pred = (cnn_y_pred > 0.5)\n",
    "cnn_y_pred"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 192,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "262/262 [==============================] - 1s 3ms/step - loss: 4.1350e-04 - accuracy: 0.9998 - precision_12: 0.9998 - recall_12: 0.9998\n",
      "Accuracy: 99.98\n"
     ]
    }
   ],
   "source": [
    "# Check the validation accuracy\n",
    "_, accuracy, precision, recall = cnn_model.evaluate(scaled_X_val, encoded_y_val)\n",
    "print('Accuracy: %.2f' % (accuracy*100))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 205,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "def plot_metrics(history, metric):\n",
    "    plt.plot(history.history[metric])\n",
    "    plt.plot(history.history['val_' + metric])\n",
    "    plt.title('Model ' + metric)\n",
    "    plt.ylabel(metric)\n",
    "    plt.xlabel('Epoch')\n",
    "    plt.ylim([0.97, 1.0])  # Set the limits of the y-axis here\n",
    "    plt.legend(['Train', 'Validation'], loc='lower right', bbox_to_anchor=(1.02, 0))\n",
    "    plt.show()\n",
    "\n",
    "plot_metrics(cnn_history, 'accuracy')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 206,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_metrics(cnn_history, 'precision_12')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 207,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_metrics(cnn_history, 'recall_12')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 197,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'accuracy_score': 0.9997610513739545, 'precision_score': 0.9978257575757576, 'recall_score': 0.9978257575757576, 'f1_score': 0.9978257575757576}\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       1.00      1.00      1.00      2750\n",
      "           1       1.00      1.00      1.00      2750\n",
      "           2       0.99      0.99      0.99       120\n",
      "           3       1.00      1.00      1.00      2750\n",
      "\n",
      "    accuracy                           1.00      8370\n",
      "   macro avg       1.00      1.00      1.00      8370\n",
      "weighted avg       1.00      1.00      1.00      8370\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1000x1000 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "print(evaluate_model(encoded_y_val.argmax(axis=1), cnn_y_pred.argmax(axis=1)))\n",
    "print(classification_report(encoded_y_val.argmax(axis=1), cnn_y_pred.argmax(axis=1)))\n",
    "plot_confusion_matrix(encoded_y_val.argmax(axis=1), cnn_y_pred.argmax(axis=1), label)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Implementing RNN model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 198,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/100\n",
      "785/785 [==============================] - 18s 18ms/step - loss: 0.2429 - accuracy: 0.9047 - precision_13: 0.9147 - recall_13: 0.8860 - val_loss: 0.0349 - val_accuracy: 0.9963 - val_precision_13: 0.9963 - val_recall_13: 0.9963\n",
      "Epoch 2/100\n",
      "785/785 [==============================] - 16s 20ms/step - loss: 0.0362 - accuracy: 0.9899 - precision_13: 0.9904 - recall_13: 0.9895 - val_loss: 0.0102 - val_accuracy: 0.9970 - val_precision_13: 0.9971 - val_recall_13: 0.9970\n",
      "Epoch 3/100\n",
      "785/785 [==============================] - 13s 17ms/step - loss: 0.0072 - accuracy: 0.9984 - precision_13: 0.9984 - recall_13: 0.9982 - val_loss: 0.0130 - val_accuracy: 0.9963 - val_precision_13: 0.9964 - val_recall_13: 0.9962\n",
      "Epoch 4/100\n",
      "785/785 [==============================] - 13s 17ms/step - loss: 0.0069 - accuracy: 0.9980 - precision_13: 0.9980 - recall_13: 0.9980 - val_loss: 0.0108 - val_accuracy: 0.9973 - val_precision_13: 0.9973 - val_recall_13: 0.9973\n",
      "Epoch 5/100\n",
      "785/785 [==============================] - 12s 15ms/step - loss: 0.0055 - accuracy: 0.9982 - precision_13: 0.9982 - recall_13: 0.9982 - val_loss: 0.0051 - val_accuracy: 0.9984 - val_precision_13: 0.9984 - val_recall_13: 0.9984\n",
      "Epoch 6/100\n",
      "785/785 [==============================] - 12s 16ms/step - loss: 0.0392 - accuracy: 0.9892 - precision_13: 0.9893 - recall_13: 0.9890 - val_loss: 0.0336 - val_accuracy: 0.9821 - val_precision_13: 0.9824 - val_recall_13: 0.9821\n",
      "Epoch 7/100\n",
      "785/785 [==============================] - 14s 18ms/step - loss: 0.0060 - accuracy: 0.9982 - precision_13: 0.9984 - recall_13: 0.9982 - val_loss: 0.0047 - val_accuracy: 0.9981 - val_precision_13: 0.9982 - val_recall_13: 0.9980\n",
      "Epoch 8/100\n",
      "785/785 [==============================] - 13s 17ms/step - loss: 0.0060 - accuracy: 0.9986 - precision_13: 0.9986 - recall_13: 0.9986 - val_loss: 0.0043 - val_accuracy: 0.9984 - val_precision_13: 0.9984 - val_recall_13: 0.9984\n",
      "Epoch 9/100\n",
      "785/785 [==============================] - 14s 18ms/step - loss: 0.0025 - accuracy: 0.9988 - precision_13: 0.9988 - recall_13: 0.9988 - val_loss: 0.0061 - val_accuracy: 0.9978 - val_precision_13: 0.9980 - val_recall_13: 0.9977\n",
      "Epoch 10/100\n",
      "785/785 [==============================] - 13s 17ms/step - loss: 0.0030 - accuracy: 0.9990 - precision_13: 0.9990 - recall_13: 0.9990 - val_loss: 0.0022 - val_accuracy: 0.9994 - val_precision_13: 0.9994 - val_recall_13: 0.9994\n",
      "Epoch 11/100\n",
      "785/785 [==============================] - 13s 16ms/step - loss: 0.0021 - accuracy: 0.9991 - precision_13: 0.9991 - recall_13: 0.9991 - val_loss: 0.0022 - val_accuracy: 0.9993 - val_precision_13: 0.9993 - val_recall_13: 0.9993\n",
      "Epoch 12/100\n",
      "785/785 [==============================] - 13s 16ms/step - loss: 0.0245 - accuracy: 0.9935 - precision_13: 0.9935 - recall_13: 0.9935 - val_loss: 0.0033 - val_accuracy: 0.9992 - val_precision_13: 0.9992 - val_recall_13: 0.9992\n",
      "Epoch 13/100\n",
      "785/785 [==============================] - 14s 18ms/step - loss: 0.0020 - accuracy: 0.9991 - precision_13: 0.9992 - recall_13: 0.9991 - val_loss: 0.0017 - val_accuracy: 0.9994 - val_precision_13: 0.9994 - val_recall_13: 0.9994\n",
      "Epoch 14/100\n",
      "785/785 [==============================] - 12s 16ms/step - loss: 0.0034 - accuracy: 0.9988 - precision_13: 0.9988 - recall_13: 0.9988 - val_loss: 0.0031 - val_accuracy: 0.9988 - val_precision_13: 0.9989 - val_recall_13: 0.9988\n",
      "Epoch 15/100\n",
      "785/785 [==============================] - 17s 22ms/step - loss: 0.0017 - accuracy: 0.9993 - precision_13: 0.9993 - recall_13: 0.9993 - val_loss: 0.0029 - val_accuracy: 0.9990 - val_precision_13: 0.9990 - val_recall_13: 0.9990\n",
      "Epoch 16/100\n",
      "785/785 [==============================] - 17s 22ms/step - loss: 0.0024 - accuracy: 0.9990 - precision_13: 0.9990 - recall_13: 0.9990 - val_loss: 0.0020 - val_accuracy: 0.9993 - val_precision_13: 0.9994 - val_recall_13: 0.9993\n",
      "Epoch 17/100\n",
      "785/785 [==============================] - 15s 19ms/step - loss: 0.0016 - accuracy: 0.9993 - precision_13: 0.9993 - recall_13: 0.9993 - val_loss: 0.0016 - val_accuracy: 0.9995 - val_precision_13: 0.9995 - val_recall_13: 0.9995\n",
      "Epoch 18/100\n",
      "785/785 [==============================] - 13s 17ms/step - loss: 0.0032 - accuracy: 0.9986 - precision_13: 0.9986 - recall_13: 0.9986 - val_loss: 0.0293 - val_accuracy: 0.9955 - val_precision_13: 0.9955 - val_recall_13: 0.9955\n",
      "Epoch 19/100\n",
      "785/785 [==============================] - 14s 17ms/step - loss: 0.0029 - accuracy: 0.9991 - precision_13: 0.9991 - recall_13: 0.9991 - val_loss: 0.0015 - val_accuracy: 0.9995 - val_precision_13: 0.9995 - val_recall_13: 0.9995\n",
      "Epoch 20/100\n",
      "785/785 [==============================] - 13s 17ms/step - loss: 0.0018 - accuracy: 0.9994 - precision_13: 0.9994 - recall_13: 0.9994 - val_loss: 0.0019 - val_accuracy: 0.9996 - val_precision_13: 0.9996 - val_recall_13: 0.9996\n",
      "Epoch 21/100\n",
      "785/785 [==============================] - 14s 18ms/step - loss: 0.0022 - accuracy: 0.9992 - precision_13: 0.9992 - recall_13: 0.9992 - val_loss: 0.0016 - val_accuracy: 0.9994 - val_precision_13: 0.9994 - val_recall_13: 0.9994\n",
      "Epoch 22/100\n",
      "785/785 [==============================] - 13s 16ms/step - loss: 0.0015 - accuracy: 0.9993 - precision_13: 0.9993 - recall_13: 0.9993 - val_loss: 0.0015 - val_accuracy: 0.9994 - val_precision_13: 0.9994 - val_recall_13: 0.9994\n",
      "Epoch 23/100\n",
      "785/785 [==============================] - 14s 17ms/step - loss: 9.4249e-04 - accuracy: 0.9996 - precision_13: 0.9996 - recall_13: 0.9996 - val_loss: 0.0010 - val_accuracy: 0.9996 - val_precision_13: 0.9996 - val_recall_13: 0.9996\n",
      "Epoch 24/100\n",
      "785/785 [==============================] - 16s 21ms/step - loss: 0.0054 - accuracy: 0.9985 - precision_13: 0.9985 - recall_13: 0.9984 - val_loss: 0.0046 - val_accuracy: 0.9990 - val_precision_13: 0.9990 - val_recall_13: 0.9990\n",
      "Epoch 25/100\n",
      "785/785 [==============================] - 13s 17ms/step - loss: 0.0010 - accuracy: 0.9996 - precision_13: 0.9996 - recall_13: 0.9996 - val_loss: 0.0031 - val_accuracy: 0.9994 - val_precision_13: 0.9994 - val_recall_13: 0.9994\n",
      "Epoch 26/100\n",
      "785/785 [==============================] - 14s 17ms/step - loss: 7.4430e-04 - accuracy: 0.9997 - precision_13: 0.9997 - recall_13: 0.9997 - val_loss: 0.0031 - val_accuracy: 0.9995 - val_precision_13: 0.9995 - val_recall_13: 0.9995\n",
      "Epoch 27/100\n",
      "785/785 [==============================] - 13s 17ms/step - loss: 7.2026e-04 - accuracy: 0.9996 - precision_13: 0.9996 - recall_13: 0.9996 - val_loss: 0.0028 - val_accuracy: 0.9990 - val_precision_13: 0.9990 - val_recall_13: 0.9990\n",
      "Epoch 28/100\n",
      "785/785 [==============================] - 13s 16ms/step - loss: 0.0011 - accuracy: 0.9996 - precision_13: 0.9996 - recall_13: 0.9996 - val_loss: 0.0053 - val_accuracy: 0.9982 - val_precision_13: 0.9982 - val_recall_13: 0.9982\n",
      "Epoch 29/100\n",
      "785/785 [==============================] - 12s 16ms/step - loss: 0.0061 - accuracy: 0.9985 - precision_13: 0.9985 - recall_13: 0.9985 - val_loss: 0.0011 - val_accuracy: 0.9995 - val_precision_13: 0.9995 - val_recall_13: 0.9995\n",
      "Epoch 30/100\n",
      "785/785 [==============================] - 13s 16ms/step - loss: 0.0014 - accuracy: 0.9996 - precision_13: 0.9996 - recall_13: 0.9996 - val_loss: 0.0033 - val_accuracy: 0.9988 - val_precision_13: 0.9988 - val_recall_13: 0.9988\n",
      "Epoch 31/100\n",
      "785/785 [==============================] - 13s 16ms/step - loss: 7.1751e-04 - accuracy: 0.9998 - precision_13: 0.9998 - recall_13: 0.9998 - val_loss: 0.0023 - val_accuracy: 0.9993 - val_precision_13: 0.9993 - val_recall_13: 0.9993\n",
      "Epoch 32/100\n",
      "785/785 [==============================] - 13s 16ms/step - loss: 0.0011 - accuracy: 0.9997 - precision_13: 0.9997 - recall_13: 0.9997 - val_loss: 5.7599e-04 - val_accuracy: 0.9999 - val_precision_13: 0.9999 - val_recall_13: 0.9999\n",
      "Epoch 33/100\n",
      "785/785 [==============================] - 14s 18ms/step - loss: 0.0011 - accuracy: 0.9997 - precision_13: 0.9997 - recall_13: 0.9997 - val_loss: 7.7513e-04 - val_accuracy: 0.9998 - val_precision_13: 0.9999 - val_recall_13: 0.9998\n",
      "Epoch 34/100\n",
      "785/785 [==============================] - 13s 16ms/step - loss: 0.0031 - accuracy: 0.9993 - precision_13: 0.9993 - recall_13: 0.9993 - val_loss: 3.7987e-04 - val_accuracy: 0.9999 - val_precision_13: 0.9999 - val_recall_13: 0.9999\n",
      "Epoch 35/100\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "785/785 [==============================] - 12s 16ms/step - loss: 0.0012 - accuracy: 0.9996 - precision_13: 0.9996 - recall_13: 0.9996 - val_loss: 0.0021 - val_accuracy: 0.9992 - val_precision_13: 0.9992 - val_recall_13: 0.9992\n",
      "Epoch 36/100\n",
      "785/785 [==============================] - 13s 16ms/step - loss: 7.4607e-04 - accuracy: 0.9997 - precision_13: 0.9997 - recall_13: 0.9997 - val_loss: 8.2941e-04 - val_accuracy: 0.9998 - val_precision_13: 0.9998 - val_recall_13: 0.9998\n",
      "Epoch 37/100\n",
      "785/785 [==============================] - 13s 16ms/step - loss: 0.0013 - accuracy: 0.9997 - precision_13: 0.9997 - recall_13: 0.9997 - val_loss: 0.0013 - val_accuracy: 0.9995 - val_precision_13: 0.9995 - val_recall_13: 0.9995\n",
      "Epoch 38/100\n",
      "785/785 [==============================] - 12s 15ms/step - loss: 0.0011 - accuracy: 0.9996 - precision_13: 0.9996 - recall_13: 0.9996 - val_loss: 0.0032 - val_accuracy: 0.9989 - val_precision_13: 0.9989 - val_recall_13: 0.9989\n",
      "Epoch 39/100\n",
      "785/785 [==============================] - 13s 17ms/step - loss: 0.0083 - accuracy: 0.9985 - precision_13: 0.9985 - recall_13: 0.9985 - val_loss: 0.0016 - val_accuracy: 0.9995 - val_precision_13: 0.9996 - val_recall_13: 0.9995\n",
      "Epoch 40/100\n",
      "785/785 [==============================] - 13s 16ms/step - loss: 4.5674e-04 - accuracy: 0.9998 - precision_13: 0.9998 - recall_13: 0.9998 - val_loss: 0.0019 - val_accuracy: 0.9995 - val_precision_13: 0.9995 - val_recall_13: 0.9995\n",
      "Epoch 41/100\n",
      "785/785 [==============================] - 12s 16ms/step - loss: 0.0012 - accuracy: 0.9996 - precision_13: 0.9996 - recall_13: 0.9996 - val_loss: 9.4843e-04 - val_accuracy: 0.9996 - val_precision_13: 0.9996 - val_recall_13: 0.9996\n",
      "Epoch 42/100\n",
      "785/785 [==============================] - 14s 18ms/step - loss: 6.3320e-04 - accuracy: 0.9998 - precision_13: 0.9998 - recall_13: 0.9998 - val_loss: 9.3633e-04 - val_accuracy: 0.9996 - val_precision_13: 0.9996 - val_recall_13: 0.9996\n",
      "Epoch 43/100\n",
      "785/785 [==============================] - 13s 16ms/step - loss: 6.6068e-04 - accuracy: 0.9998 - precision_13: 0.9998 - recall_13: 0.9998 - val_loss: 0.0011 - val_accuracy: 0.9995 - val_precision_13: 0.9995 - val_recall_13: 0.9995\n",
      "Epoch 44/100\n",
      "785/785 [==============================] - 13s 17ms/step - loss: 5.1098e-04 - accuracy: 0.9997 - precision_13: 0.9997 - recall_13: 0.9997 - val_loss: 9.0728e-04 - val_accuracy: 0.9995 - val_precision_13: 0.9995 - val_recall_13: 0.9995\n",
      "Epoch 45/100\n",
      "785/785 [==============================] - 12s 15ms/step - loss: 3.8521e-04 - accuracy: 0.9999 - precision_13: 0.9999 - recall_13: 0.9999 - val_loss: 2.6252e-04 - val_accuracy: 0.9999 - val_precision_13: 0.9999 - val_recall_13: 0.9999\n",
      "Epoch 46/100\n",
      "785/785 [==============================] - 13s 17ms/step - loss: 2.6341e-05 - accuracy: 1.0000 - precision_13: 1.0000 - recall_13: 1.0000 - val_loss: 4.3383e-04 - val_accuracy: 0.9999 - val_precision_13: 0.9999 - val_recall_13: 0.9999\n",
      "Epoch 47/100\n",
      "785/785 [==============================] - 14s 17ms/step - loss: 1.4371e-05 - accuracy: 1.0000 - precision_13: 1.0000 - recall_13: 1.0000 - val_loss: 4.0068e-04 - val_accuracy: 0.9999 - val_precision_13: 0.9999 - val_recall_13: 0.9999\n",
      "Epoch 48/100\n",
      "785/785 [==============================] - 13s 16ms/step - loss: 3.1550e-04 - accuracy: 0.9999 - precision_13: 0.9999 - recall_13: 0.9999 - val_loss: 8.9408e-04 - val_accuracy: 0.9998 - val_precision_13: 0.9998 - val_recall_13: 0.9998\n",
      "Epoch 49/100\n",
      "785/785 [==============================] - 13s 16ms/step - loss: 1.1278e-04 - accuracy: 1.0000 - precision_13: 1.0000 - recall_13: 1.0000 - val_loss: 8.5069e-04 - val_accuracy: 0.9998 - val_precision_13: 0.9998 - val_recall_13: 0.9998\n",
      "Epoch 50/100\n",
      "785/785 [==============================] - 12s 16ms/step - loss: 1.7176e-04 - accuracy: 0.9999 - precision_13: 0.9999 - recall_13: 0.9999 - val_loss: 2.1920e-04 - val_accuracy: 0.9999 - val_precision_13: 0.9999 - val_recall_13: 0.9999\n",
      "Epoch 51/100\n",
      "785/785 [==============================] - 13s 16ms/step - loss: 8.1766e-06 - accuracy: 1.0000 - precision_13: 1.0000 - recall_13: 1.0000 - val_loss: 2.2906e-04 - val_accuracy: 0.9999 - val_precision_13: 0.9999 - val_recall_13: 0.9999\n",
      "Epoch 52/100\n",
      "785/785 [==============================] - 13s 16ms/step - loss: 4.9532e-06 - accuracy: 1.0000 - precision_13: 1.0000 - recall_13: 1.0000 - val_loss: 3.0666e-04 - val_accuracy: 0.9999 - val_precision_13: 0.9999 - val_recall_13: 0.9999\n",
      "Epoch 53/100\n",
      "785/785 [==============================] - 13s 16ms/step - loss: 3.4943e-06 - accuracy: 1.0000 - precision_13: 1.0000 - recall_13: 1.0000 - val_loss: 2.8318e-04 - val_accuracy: 0.9999 - val_precision_13: 0.9999 - val_recall_13: 0.9999\n",
      "Epoch 54/100\n",
      "785/785 [==============================] - 13s 16ms/step - loss: 1.9245e-06 - accuracy: 1.0000 - precision_13: 1.0000 - recall_13: 1.0000 - val_loss: 2.6241e-04 - val_accuracy: 0.9999 - val_precision_13: 0.9999 - val_recall_13: 0.9999\n",
      "Epoch 55/100\n",
      "785/785 [==============================] - 13s 17ms/step - loss: 1.0885e-06 - accuracy: 1.0000 - precision_13: 1.0000 - recall_13: 1.0000 - val_loss: 2.4994e-04 - val_accuracy: 0.9999 - val_precision_13: 0.9999 - val_recall_13: 0.9999\n",
      "Epoch 56/100\n",
      "785/785 [==============================] - 14s 17ms/step - loss: 4.0352e-07 - accuracy: 1.0000 - precision_13: 1.0000 - recall_13: 1.0000 - val_loss: 2.4253e-04 - val_accuracy: 0.9999 - val_precision_13: 0.9999 - val_recall_13: 0.9999\n",
      "Epoch 57/100\n",
      "785/785 [==============================] - 14s 18ms/step - loss: 0.0178 - accuracy: 0.9964 - precision_13: 0.9965 - recall_13: 0.9964 - val_loss: 0.0036 - val_accuracy: 0.9988 - val_precision_13: 0.9988 - val_recall_13: 0.9988\n",
      "Epoch 58/100\n",
      "785/785 [==============================] - 13s 16ms/step - loss: 0.0031 - accuracy: 0.9991 - precision_13: 0.9991 - recall_13: 0.9990 - val_loss: 0.0022 - val_accuracy: 0.9994 - val_precision_13: 0.9994 - val_recall_13: 0.9994\n",
      "Epoch 59/100\n",
      "785/785 [==============================] - 13s 16ms/step - loss: 0.0025 - accuracy: 0.9994 - precision_13: 0.9994 - recall_13: 0.9994 - val_loss: 0.0013 - val_accuracy: 0.9994 - val_precision_13: 0.9994 - val_recall_13: 0.9994\n",
      "Epoch 60/100\n",
      "785/785 [==============================] - 14s 17ms/step - loss: 0.0011 - accuracy: 0.9996 - precision_13: 0.9996 - recall_13: 0.9996 - val_loss: 7.2895e-04 - val_accuracy: 0.9999 - val_precision_13: 0.9999 - val_recall_13: 0.9999\n",
      "Epoch 61/100\n",
      "785/785 [==============================] - 13s 17ms/step - loss: 6.5274e-04 - accuracy: 0.9997 - precision_13: 0.9997 - recall_13: 0.9997 - val_loss: 6.7517e-04 - val_accuracy: 0.9998 - val_precision_13: 0.9998 - val_recall_13: 0.9998\n",
      "Epoch 62/100\n",
      "785/785 [==============================] - 12s 15ms/step - loss: 0.0012 - accuracy: 0.9996 - precision_13: 0.9996 - recall_13: 0.9996 - val_loss: 0.0024 - val_accuracy: 0.9989 - val_precision_13: 0.9989 - val_recall_13: 0.9989\n",
      "Epoch 63/100\n",
      "785/785 [==============================] - 12s 15ms/step - loss: 0.0014 - accuracy: 0.9997 - precision_13: 0.9997 - recall_13: 0.9997 - val_loss: 1.2908e-04 - val_accuracy: 1.0000 - val_precision_13: 1.0000 - val_recall_13: 1.0000\n",
      "Epoch 64/100\n",
      "785/785 [==============================] - 6833s 9s/step - loss: 5.8040e-04 - accuracy: 0.9997 - precision_13: 0.9997 - recall_13: 0.9997 - val_loss: 2.6054e-04 - val_accuracy: 1.0000 - val_precision_13: 1.0000 - val_recall_13: 1.0000\n",
      "Epoch 65/100\n",
      "785/785 [==============================] - 6s 8ms/step - loss: 5.9176e-05 - accuracy: 1.0000 - precision_13: 1.0000 - recall_13: 1.0000 - val_loss: 5.4404e-05 - val_accuracy: 1.0000 - val_precision_13: 1.0000 - val_recall_13: 1.0000\n",
      "Epoch 66/100\n",
      "785/785 [==============================] - 6s 8ms/step - loss: 2.7152e-04 - accuracy: 0.9999 - precision_13: 0.9999 - recall_13: 0.9999 - val_loss: 3.1030e-05 - val_accuracy: 1.0000 - val_precision_13: 1.0000 - val_recall_13: 1.0000\n",
      "Epoch 67/100\n",
      "785/785 [==============================] - 6s 8ms/step - loss: 2.6827e-05 - accuracy: 1.0000 - precision_13: 1.0000 - recall_13: 1.0000 - val_loss: 1.3591e-05 - val_accuracy: 1.0000 - val_precision_13: 1.0000 - val_recall_13: 1.0000\n",
      "Epoch 68/100\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "785/785 [==============================] - 6s 8ms/step - loss: 6.8123e-06 - accuracy: 1.0000 - precision_13: 1.0000 - recall_13: 1.0000 - val_loss: 1.0222e-05 - val_accuracy: 1.0000 - val_precision_13: 1.0000 - val_recall_13: 1.0000\n",
      "Epoch 69/100\n",
      "785/785 [==============================] - 6s 8ms/step - loss: 2.5378e-06 - accuracy: 1.0000 - precision_13: 1.0000 - recall_13: 1.0000 - val_loss: 1.5582e-05 - val_accuracy: 1.0000 - val_precision_13: 1.0000 - val_recall_13: 1.0000\n",
      "Epoch 70/100\n",
      "785/785 [==============================] - 6s 8ms/step - loss: 1.5734e-06 - accuracy: 1.0000 - precision_13: 1.0000 - recall_13: 1.0000 - val_loss: 1.8047e-05 - val_accuracy: 1.0000 - val_precision_13: 1.0000 - val_recall_13: 1.0000\n",
      "Epoch 71/100\n",
      "785/785 [==============================] - 6s 8ms/step - loss: 1.0530e-06 - accuracy: 1.0000 - precision_13: 1.0000 - recall_13: 1.0000 - val_loss: 1.7782e-05 - val_accuracy: 1.0000 - val_precision_13: 1.0000 - val_recall_13: 1.0000\n",
      "Epoch 72/100\n",
      "785/785 [==============================] - 6s 8ms/step - loss: 6.4477e-07 - accuracy: 1.0000 - precision_13: 1.0000 - recall_13: 1.0000 - val_loss: 1.6951e-05 - val_accuracy: 1.0000 - val_precision_13: 1.0000 - val_recall_13: 1.0000\n",
      "Epoch 73/100\n",
      "785/785 [==============================] - 6s 8ms/step - loss: 3.7148e-07 - accuracy: 1.0000 - precision_13: 1.0000 - recall_13: 1.0000 - val_loss: 2.2285e-05 - val_accuracy: 1.0000 - val_precision_13: 1.0000 - val_recall_13: 1.0000\n",
      "Epoch 74/100\n",
      "785/785 [==============================] - 6s 8ms/step - loss: 2.5812e-07 - accuracy: 1.0000 - precision_13: 1.0000 - recall_13: 1.0000 - val_loss: 1.7784e-05 - val_accuracy: 1.0000 - val_precision_13: 1.0000 - val_recall_13: 1.0000\n",
      "Epoch 75/100\n",
      "785/785 [==============================] - 6s 8ms/step - loss: 1.5221e-07 - accuracy: 1.0000 - precision_13: 1.0000 - recall_13: 1.0000 - val_loss: 2.7688e-05 - val_accuracy: 1.0000 - val_precision_13: 1.0000 - val_recall_13: 1.0000\n",
      "Epoch 76/100\n",
      "785/785 [==============================] - 6s 8ms/step - loss: 1.1282e-07 - accuracy: 1.0000 - precision_13: 1.0000 - recall_13: 1.0000 - val_loss: 2.5228e-05 - val_accuracy: 1.0000 - val_precision_13: 1.0000 - val_recall_13: 1.0000\n",
      "Epoch 77/100\n",
      "785/785 [==============================] - 6s 8ms/step - loss: 6.5274e-08 - accuracy: 1.0000 - precision_13: 1.0000 - recall_13: 1.0000 - val_loss: 2.6842e-05 - val_accuracy: 1.0000 - val_precision_13: 1.0000 - val_recall_13: 1.0000\n",
      "Epoch 78/100\n",
      "785/785 [==============================] - 6s 8ms/step - loss: 4.2618e-08 - accuracy: 1.0000 - precision_13: 1.0000 - recall_13: 1.0000 - val_loss: 3.2322e-05 - val_accuracy: 1.0000 - val_precision_13: 1.0000 - val_recall_13: 1.0000\n",
      "Epoch 79/100\n",
      "785/785 [==============================] - 6s 8ms/step - loss: 2.7386e-08 - accuracy: 1.0000 - precision_13: 1.0000 - recall_13: 1.0000 - val_loss: 3.7057e-05 - val_accuracy: 1.0000 - val_precision_13: 1.0000 - val_recall_13: 1.0000\n",
      "Epoch 80/100\n",
      "785/785 [==============================] - 6s 8ms/step - loss: 1.7838e-08 - accuracy: 1.0000 - precision_13: 1.0000 - recall_13: 1.0000 - val_loss: 4.1839e-05 - val_accuracy: 1.0000 - val_precision_13: 1.0000 - val_recall_13: 1.0000\n",
      "Epoch 81/100\n",
      "785/785 [==============================] - 6s 8ms/step - loss: 1.1542e-08 - accuracy: 1.0000 - precision_13: 1.0000 - recall_13: 1.0000 - val_loss: 3.5798e-05 - val_accuracy: 1.0000 - val_precision_13: 1.0000 - val_recall_13: 1.0000\n",
      "Epoch 82/100\n",
      "785/785 [==============================] - 6s 8ms/step - loss: 7.0034e-09 - accuracy: 1.0000 - precision_13: 1.0000 - recall_13: 1.0000 - val_loss: 5.4802e-05 - val_accuracy: 1.0000 - val_precision_13: 1.0000 - val_recall_13: 1.0000\n",
      "Epoch 83/100\n",
      "785/785 [==============================] - 7s 9ms/step - loss: 5.0899e-09 - accuracy: 1.0000 - precision_13: 1.0000 - recall_13: 1.0000 - val_loss: 6.9413e-05 - val_accuracy: 1.0000 - val_precision_13: 1.0000 - val_recall_13: 1.0000\n",
      "Epoch 84/100\n",
      "785/785 [==============================] - 7s 9ms/step - loss: 3.5990e-09 - accuracy: 1.0000 - precision_13: 1.0000 - recall_13: 1.0000 - val_loss: 7.0355e-05 - val_accuracy: 1.0000 - val_precision_13: 1.0000 - val_recall_13: 1.0000\n",
      "Epoch 85/100\n",
      "785/785 [==============================] - 7s 9ms/step - loss: 2.3313e-09 - accuracy: 1.0000 - precision_13: 1.0000 - recall_13: 1.0000 - val_loss: 9.3319e-05 - val_accuracy: 0.9999 - val_precision_13: 0.9999 - val_recall_13: 0.9999\n",
      "Epoch 86/100\n",
      "785/785 [==============================] - 8s 10ms/step - loss: 1.6761e-09 - accuracy: 1.0000 - precision_13: 1.0000 - recall_13: 1.0000 - val_loss: 8.6740e-05 - val_accuracy: 1.0000 - val_precision_13: 1.0000 - val_recall_13: 0.9999\n",
      "Epoch 87/100\n",
      "785/785 [==============================] - 8s 10ms/step - loss: 1.2250e-09 - accuracy: 1.0000 - precision_13: 1.0000 - recall_13: 1.0000 - val_loss: 1.1983e-04 - val_accuracy: 0.9999 - val_precision_13: 0.9999 - val_recall_13: 0.9999\n",
      "Epoch 88/100\n",
      "785/785 [==============================] - 7s 9ms/step - loss: 8.9738e-10 - accuracy: 1.0000 - precision_13: 1.0000 - recall_13: 1.0000 - val_loss: 1.3678e-04 - val_accuracy: 0.9999 - val_precision_13: 0.9999 - val_recall_13: 0.9999\n",
      "Epoch 89/100\n",
      "785/785 [==============================] - 7s 9ms/step - loss: 6.9796e-10 - accuracy: 1.0000 - precision_13: 1.0000 - recall_13: 1.0000 - val_loss: 1.7381e-04 - val_accuracy: 0.9999 - val_precision_13: 0.9999 - val_recall_13: 0.9999\n",
      "Epoch 90/100\n",
      "785/785 [==============================] - 7s 9ms/step - loss: 5.4128e-10 - accuracy: 1.0000 - precision_13: 1.0000 - recall_13: 1.0000 - val_loss: 1.9541e-04 - val_accuracy: 0.9999 - val_precision_13: 0.9999 - val_recall_13: 0.9999\n",
      "Epoch 91/100\n",
      "785/785 [==============================] - 7s 9ms/step - loss: 3.9884e-10 - accuracy: 1.0000 - precision_13: 1.0000 - recall_13: 1.0000 - val_loss: 2.1061e-04 - val_accuracy: 0.9999 - val_precision_13: 0.9999 - val_recall_13: 0.9999\n",
      "Epoch 92/100\n",
      "785/785 [==============================] - 7s 9ms/step - loss: 3.0388e-10 - accuracy: 1.0000 - precision_13: 1.0000 - recall_13: 1.0000 - val_loss: 2.0732e-04 - val_accuracy: 0.9999 - val_precision_13: 0.9999 - val_recall_13: 0.9999\n",
      "Epoch 93/100\n",
      "785/785 [==============================] - 7s 9ms/step - loss: 2.4215e-10 - accuracy: 1.0000 - precision_13: 1.0000 - recall_13: 1.0000 - val_loss: 2.6623e-04 - val_accuracy: 0.9999 - val_precision_13: 0.9999 - val_recall_13: 0.9999\n",
      "Epoch 94/100\n",
      "785/785 [==============================] - 7s 9ms/step - loss: 1.8043e-10 - accuracy: 1.0000 - precision_13: 1.0000 - recall_13: 1.0000 - val_loss: 2.7164e-04 - val_accuracy: 0.9999 - val_precision_13: 0.9999 - val_recall_13: 0.9999\n",
      "Epoch 95/100\n",
      "785/785 [==============================] - 7s 9ms/step - loss: 1.5194e-10 - accuracy: 1.0000 - precision_13: 1.0000 - recall_13: 1.0000 - val_loss: 3.4262e-04 - val_accuracy: 0.9999 - val_precision_13: 0.9999 - val_recall_13: 0.9999\n",
      "Epoch 96/100\n",
      "785/785 [==============================] - 7s 9ms/step - loss: 1.4719e-10 - accuracy: 1.0000 - precision_13: 1.0000 - recall_13: 1.0000 - val_loss: 3.6283e-04 - val_accuracy: 0.9999 - val_precision_13: 0.9999 - val_recall_13: 0.9999\n",
      "Epoch 97/100\n",
      "785/785 [==============================] - 7s 9ms/step - loss: 1.1870e-10 - accuracy: 1.0000 - precision_13: 1.0000 - recall_13: 1.0000 - val_loss: 4.1093e-04 - val_accuracy: 0.9999 - val_precision_13: 0.9999 - val_recall_13: 0.9999\n",
      "Epoch 98/100\n",
      "785/785 [==============================] - 7s 9ms/step - loss: 9.0213e-11 - accuracy: 1.0000 - precision_13: 1.0000 - recall_13: 1.0000 - val_loss: 4.0440e-04 - val_accuracy: 0.9999 - val_precision_13: 0.9999 - val_recall_13: 0.9999\n",
      "Epoch 99/100\n",
      "785/785 [==============================] - 7s 9ms/step - loss: 8.5465e-11 - accuracy: 1.0000 - precision_13: 1.0000 - recall_13: 1.0000 - val_loss: 4.2865e-04 - val_accuracy: 0.9999 - val_precision_13: 0.9999 - val_recall_13: 0.9999\n",
      "Epoch 100/100\n",
      "785/785 [==============================] - 7s 9ms/step - loss: 7.5969e-11 - accuracy: 1.0000 - precision_13: 1.0000 - recall_13: 1.0000 - val_loss: 4.5684e-04 - val_accuracy: 0.9999 - val_precision_13: 0.9999 - val_recall_13: 0.9999\n"
     ]
    }
   ],
   "source": [
    "from tensorflow.keras.preprocessing.sequence import pad_sequences\n",
    "from tensorflow.keras.layers import Embedding, Conv1D, MaxPooling1D, GlobalMaxPooling1D, Dense, LSTM\n",
    "\n",
    "\n",
    "rnn_model = tf.keras.models.Sequential([\n",
    "    tf.keras.layers.Reshape(target_shape=(scaled_X_train.shape[1], 1), input_shape=(scaled_X_train.shape[1],)),\n",
    "    tf.keras.layers.LSTM(64),\n",
    "    tf.keras.layers.Dense(64, activation='relu'),\n",
    "    tf.keras.layers.Dense(len(label_encoder.classes_), activation='softmax')\n",
    "])\n",
    "\n",
    "rnn_model.compile(optimizer='adam', loss='categorical_crossentropy', metrics=['accuracy', tf.keras.metrics.Precision(), tf.keras.metrics.Recall()])\n",
    "\n",
    "rnn_history = rnn_model.fit(scaled_X_train, encoded_y_train, validation_data=(scaled_X_val, encoded_y_val), batch_size=32, epochs=100)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 199,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "262/262 [==============================] - 1s 3ms/step\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "array([[False, False, False,  True],\n",
       "       [False,  True, False, False],\n",
       "       [ True, False, False, False],\n",
       "       ...,\n",
       "       [ True, False, False, False],\n",
       "       [False,  True, False, False],\n",
       "       [ True, False, False, False]])"
      ]
     },
     "execution_count": 199,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rnn_y_pred = rnn_model.predict(scaled_X_val)\n",
    "rnn_y_pred = (rnn_y_pred > 0.5)\n",
    "rnn_y_pred"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 200,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "262/262 [==============================] - 1s 4ms/step - loss: 4.5684e-04 - accuracy: 0.9999 - precision_13: 0.9999 - recall_13: 0.9999\n",
      "Accuracy: 99.99\n"
     ]
    }
   ],
   "source": [
    "# Check the validation accuracy\n",
    "_, accuracy, precision, recall = rnn_model.evaluate(scaled_X_val, encoded_y_val)\n",
    "print('Accuracy: %.2f' % (accuracy*100))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 208,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAkwAAAHFCAYAAAAAM6ZOAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8qNh9FAAAACXBIWXMAAA9hAAAPYQGoP6dpAAB2sUlEQVR4nO3deXhM1/8H8PdsyUxWJCRBVrUEsYVGLLUWsRSlRUtpdbGW6kbRolXafqn2V0up2Fuq6F4E1dIgxBr7HtJExJIgyWSW+/vjZm4yJpFkMjEz9X49T55m7py5c+am5O1zzj1HJgiCACIiIiIqltzeHSAiIiJydAxMRERERCVgYCIiIiIqAQMTERERUQkYmIiIiIhKwMBEREREVAIGJiIiIqISMDARERERlYCBiYiIiKgEDExEVKTly5dDJpNBJpNh586dFs8LgoDHHnsMMpkM7du3t+l7y2QyTJs2rcyvu3TpEmQyGZYvX27T/hARMTAR0QN5enpi6dKlFsf/+usvnD9/Hp6ennboFRHRw8XAREQPNGDAAGzYsAFZWVlmx5cuXYro6GgEBQXZqWePDp1OB71eb+9uED3SGJiI6IEGDRoEAPjuu++kY5mZmdiwYQNeeumlIl9z8+ZNjBo1CjVq1ICLiwvCwsIwefJkaLVas3ZZWVl45ZVX4OPjAw8PD3Tr1g1nzpwp8pxnz57Fc889h2rVqsHV1RXh4eGYP3++VZ8pNzcXb775Jpo0aQJvb29UqVIF0dHR+OmnnyzaGo1G/N///R+aNGkCjUaDSpUqoWXLlvj555/N2n377beIjo6Gh4cHPDw80KRJE7PKXEhICIYNG2Zx/vbt25sNae7cuRMymQyrVq3Cm2++iRo1asDV1RXnzp3D9evXMWrUKNSvXx8eHh6oVq0aOnbsiF27dlmcV6vVYsaMGQgPD4darYaPjw86dOiA+Ph4AECnTp1Qr1493L//ummotUePHmW5pET/eUp7d4CIHJuXlxf69++P2NhYvPbaawDE8CSXyzFgwADMmzfPrH1ubi46dOiA8+fPY/r06WjUqBF27dqFWbNm4fDhw/jtt98AiL+Y+/Tpg/j4eLz//vto0aIF/vnnH8TExFj04cSJE2jVqhWCgoIwZ84c+Pv7Y8uWLXj99deRkZGBDz74oEyfSavV4ubNm3jrrbdQo0YN5OXlYdu2bXj66aexbNkyvPDCC1LbYcOGYfXq1Rg+fDhmzJgBFxcXHDx4EJcuXZLavP/++/jwww/x9NNP480334S3tzeSkpJw+fLlMvWrsEmTJiE6OhqLFi2CXC5HtWrVcP36dQDABx98AH9/f9y9exebNm1C+/btsX37dil46fV6xMTEYNeuXRg/fjw6duwIvV6PvXv3Ijk5Ga1atcK4cePQu3dvbN++HZ07d5be948//sD58+fx5ZdfWt13ov8kgYioCMuWLRMACPv37xf+/PNPAYCQlJQkCIIgtGjRQhg2bJggCILQoEEDoV27dtLrFi1aJAAQvv/+e7PzffLJJwIAYevWrYIgCMIff/whABC++OILs3YzZ84UAAgffPCBdKxr165CzZo1hczMTLO2Y8aMEdRqtXDz5k1BEATh4sWLAgBh2bJlZfqser1e0Ol0wvDhw4WmTZtKx//++28BgDB58uRiX3vhwgVBoVAIzz///APfIzg4WBg6dKjF8Xbt2pldP9O1fuKJJ0rd706dOgl9+/aVjq9cuVIAICxZsqTY1xoMBiEsLEzo3bu32fGYmBihVq1agtFoLPH9iR4lHJIjohK1a9cOtWrVQmxsLI4dO4b9+/cXOxy3Y8cOuLu7o3///mbHTcNR27dvBwD8+eefAIDnn3/erN1zzz1n9jg3Nxfbt29H37594ebmBr1eL311794dubm52Lt3b5k/0/r169G6dWt4eHhAqVRCpVJh6dKlOHnypNTmjz/+AACMHj262PPExcXBYDA8sI01+vXrV+TxRYsWoVmzZlCr1VK/t2/fbtFvtVpd7M8IAORyOcaMGYNff/0VycnJAIDz589j8+bNGDVqFGQymU0/D5GzY2AiohLJZDK8+OKLWL16NRYtWoQ6deqgbdu2Rba9ceMG/P39LX7hVqtWDUqlEjdu3JDaKZVK+Pj4mLXz9/e3OJ9er8f//d//QaVSmX11794dAJCRkVGmz7Nx40Y8++yzqFGjBlavXo09e/ZIITA3N1dqd/36dSgUCos+FWYaJqtZs2aZ+lCSgIAAi2Nz587FyJEjERUVhQ0bNmDv3r3Yv38/unXrhpycHLM+Va9eHXL5g/+Kf+mll6DRaLBo0SIAwPz586HRaB4YtIgeVZzDRESlMmzYMLz//vtYtGgRZs6cWWw7Hx8f7Nu3D4IgmIWm9PR06PV6+Pr6Su30ej1u3LhhFprS0tLMzle5cmUoFAoMGTKk2CpOaGhomT7L6tWrERoainXr1pn18f5J6VWrVoXBYEBaWlqRAcbUBgCuXr2KwMDAYt9TrVZbnB8Qw57pmhRWVIVn9erVaN++PRYuXGh2/M6dOxZ92r17N4xG4wNDk7e3N4YOHYpvvvkGb731FpYtW4bnnnsOlSpVKvY1RI8qVpiIqFRq1KiBt99+G7169cLQoUOLbdepUyfcvXsXP/74o9nxlStXSs8DQIcOHQAAa9asMWv37bffmj12c3NDhw4dcOjQITRq1AjNmze3+Lq/SlUSmUwGFxcXs1CSlpZmcZecaQL6/QGlsC5dukChUDywDSDeJXf06FGzY2fOnMHp06fL1G9XV1ezY0ePHsWePXss+p2bm1uqBTxNE+f79++P27dvY8yYMaXuD9GjhBUmIiq12bNnl9jmhRdewPz58zF06FBcunQJERER2L17Nz7++GN0795duiOrS5cueOKJJ/DOO+/g3r17aN68Of755x+sWrXK4pxffPEF2rRpg7Zt22LkyJEICQnBnTt3cO7cOfzyyy/YsWNHmT5Hz549sXHjRowaNQr9+/fHlStX8OGHHyIgIABnz56V2rVt2xZDhgzBRx99hGvXrqFnz55wdXXFoUOH4ObmhrFjxyIkJATvvfcePvzwQ+Tk5GDQoEHw9vbGiRMnkJGRgenTpwMAhgwZgsGDB2PUqFHo168fLl++jE8//VSqUJW23x9++CE++OADtGvXDqdPn8aMGTMQGhpqtk7ToEGDsGzZMowYMQKnT59Ghw4dYDQasW/fPoSHh2PgwIFS2zp16qBbt274448/0KZNGzRu3LhM15LokWHvWedE5JgK3yX3IPffJScIgnDjxg1hxIgRQkBAgKBUKoXg4GBh0qRJQm5urlm727dvCy+99JJQqVIlwc3NTXjyySeFU6dOWdwlJwjiHXAvvfSSUKNGDUGlUglVq1YVWrVqJXz00UdmbVDKu+Rmz54thISECK6urkJ4eLiwZMkS4YMPPhDu/2vRYDAIn3/+udCwYUPBxcVF8Pb2FqKjo4VffvnFrN3KlSuFFi1aCGq1WvDw8BCaNm1q1g+j0Sh8+umnQlhYmKBWq4XmzZsLO3bsKPYuufXr11v0WavVCm+99ZZQo0YNQa1WC82aNRN+/PFHYejQoUJwcLBZ25ycHOH9998XateuLbi4uAg+Pj5Cx44dhfj4eIvzLl++XAAgrF27tsTrRvSokgnCfauWERHRI6Vfv37Yu3cvLl26BJVKZe/uEDkkDskRET2CtFotDh48iISEBGzatAlz585lWCJ6AFaYiIgeQZcuXUJoaCi8vLzw3HPP4auvvoJCobB3t4gcFgMTERERUQnsuqzA33//jV69eqF69eqQyWQWtyEX5a+//kJkZCTUajXCwsKkBdcK27BhA+rXrw9XV1fUr18fmzZtsmizYMEChIaGQq1WIzIyssjNK4mIiIgAOweme/fuoXHjxvjqq69K1f7ixYvo3r072rZti0OHDuG9997D66+/jg0bNkht9uzZgwEDBmDIkCE4cuQIhgwZgmeffRb79u2T2qxbtw7jx4/H5MmTcejQIbRt2xYxMTHS9gBEREREhTnMkJxMJsOmTZvQp0+fYtu8++67+Pnnn832TBoxYgSOHDkiLdw2YMAAZGVlSXtAAUC3bt1QuXJlfPfddwCAqKgoNGvWzGyhufDwcPTp0wezZs2y8ScjIiIiZ+dUd8nt2bMHXbp0MTvWtWtXLF26FDqdDiqVCnv27MEbb7xh0WbevHkAgLy8PCQmJmLixIlmbbp06YL4+Phi31ur1Zpta2A0GnHz5k34+Phwk0oiIiInIQgC7ty5U6r9FgtzqsCUlpYGPz8/s2N+fn7Q6/XIyMhAQEBAsW1M+1NlZGTAYDA8sE1RZs2aJa3YS0RERM7typUrZdo026kCE2C5IaVpRLHw8aLa3H+sNG0KmzRpEiZMmCA9zszMRFBQEK5cuQIvL6+yfQiicjiffgdHrmaiWXBlhPi4A+f/BL4fIj6pdAMe6wjU7Q7UiAQu7QZO/wFc/Bsw5gEyJfDqDqBKWIX2cc3eS1ix5zJebhuG/s1qQi43/7OVnafHVzvOYcvxNNSs7IZGNb3RqIY3Imp6w99bY3E+vcGIFXsu4dTOdfhYvggqmaFU/RAgwwXXetisa4ofsxsjFb5wUcrh7+UKPy81AhRZqHV7N5rci0dj4RSUMqNNPn9Z/B35JZ7o8rR1L9blAqv6AteOiY996gB1Y8QvvwZAcX+n7foc2D0HCG4FPPmRdFiAgC1J17Bk9wXoDEZU81BjSHQQ0u9qcTbtDs5cu4ub2XlQyWUI9XVH7WqeqOPvgSAfd6jkrLSXhcyQB7eUf+B+aSvUGUlmzxmVbrgX1B73gjtBkX0D7pe2QnMtETI4xAyahyrnsZ7QPGN5c1d5ZGVlITAwEJ6enmV6nVMFJn9/f4sqUHp6OpRKpbT5ZnFtTBUlX19fKBSKB7Ypiqurq8WmlwDg5eXFwERmBEHA8X+zkHDxJmpU1qBpYCVU81KX+7z/3s7B53FnsOHgVRgFALiIRn4uWKmdhEquMggqd8h094CLv4lfhakAKDWAPhc4uAB4erHlGxiNwMZXgOungM7TgdqdreqnIAhYcygDaTlyfLT1EuLOZmF2v0YI9XUHAPxzLgMTNx7FlZs5ABS4nqbFobR04EA6AMDPyxVNAiuhSWBlNA2qBBelHO//lISw1M2Yr1kEpcyI3IAoqKuGPrgjN84BKYloilNoqjqFSd7fQZCLCzPKBACZAIw6sW3+H21d1QjIqtWD8iGsR/TvmURUzz2H0Ovb4eU1zLqT/DIVuJ0EeLgDBh1w9yyQeBZI/BJ4rDPw3PeA/L7PIggQLvwKmasMcV5dceqSt/TU8X+zsPm4DFDUQqcG1TDn2cao5OZS6KUCMu7mwVOthFrFNZvK70kA04DbV4CTvwC3k4Gw9kBYe1RSFf47YyJwNx049RtwJQEQSvcPhv8Cr5otgAr6HVvW6TROFZiio6Pxyy+/mB3bunUrmjdvLq1QGx0djbi4OLN5TFu3bkWrVq0AAC4uLoiMjERcXBz69u0rtYmLi0Pv3r0fwqcgu8i7B5zbDuTdBRr0BVSWVYwiGQ3A8U1ApSAg8PHimxkFHLpyC38cS8Pm42m4eivH7PkalTRoElgJ9at7IcBbDX8vNfy9xS83lwf/MbydnYcFO89jefwl5OnFCkj9AC+cuXYHT95YjUrKFPwrVEGXO5/iMfm/6Cbfj26KBITIriHTqw68m/UDwp8CDHnA4nbA0e+BNhOAavWQnafHsauZOHzlNjTHv8ML6T+Ib7qmHzYY2uIj/WBkybzQqKY3ujXwR0zDAAT5uD2wvxcy7uHqrRyoFDIo5XLsu3gTL87biBnhV6C9fQ0nUrPQH4CHuxKt6lZHSuXHsSOrJg5dzcLptCxcy9Jiy/Fr2HL8mnTO/oq/8KnLYsghQGjyHNRPfWUZBIqSeRU4+Stw8mfgcjxkpoBk9sNpDtR/CgjvBVUFV94K2/3Dd3g2aQTCMnYAei2gtPwH2QMdWgMkLocAGTbUno2DhjDUTP8bje78hSjdAajObcO6NUtQq+2zaBZUGXK5DLfu5WHrn9sx4MYZaAUVJhytiTs4Y3ZahVyGt7vWxattwywqgzKZDFU9y9hPKlmlQCB61IPbeFQDmr8ofpFd2PUuubt37+LcuXMAgKZNm2Lu3Lno0KEDqlSpgqCgIEyaNAkpKSlYuXIlAHFZgYYNG+K1117DK6+8gj179mDEiBH47rvv0K9fPwBAfHw8nnjiCcycORO9e/fGTz/9hClTpmD37t2IiooCIC4rMGTIECxatAjR0dFYvHgxlixZguPHjyM4OLhUfc/KyoK3tzcyMzNZYbIHQSh+uMEkNxM4sxU4+RNwdhugzw8xVWoBT30JhLR58OuvHYfw0xjI/j0Io8IV8tF7ixzKMhgFPL3gHxy5mikd06gUiAqrgtTbuTiTfgfF/SmTyYC3utTF6A6PFfl8WmYuev7fLmTczQMARIVWwcSYemgaVBlZV5LgEdsOckGP0YYJ+E3XvNArBbgjF7lyNywaHIkn6+dXT9cNAU7+DKF+b8RWn45PNp9Cnt6IKsjCdte3UFl2FwnGumguOwO5TMB1wQvTdMPwmzEKgHi96wd4YeDjgRjSMrjgX2hGg1jhALAi/hI+/uMkegUbMaXWedzY/wNq6U4/+Fp71QDq9URu7R44JtTCkatZOHI1E0ev3kabe1sxUxUrtoscBvT4HCjDRE1JbiagvWt+TKUB3KqU/Vw28N3ei+jwRwf4y24Bg9aKw2illXYMwpLOkBlyMVffH1/qzYf0Jiq/wwjlL9hrDMfAvKmo6ilW7XadvY6xwrcYrfwZO/A4/mjwPygVBddSpZChT9MaaBZU2VYfk8jhWPv7266BaefOnejQoYPF8aFDh2L58uUYNmwYLl26hJ07d0rP/fXXX3jjjTdw/PhxVK9eHe+++y5GjBhh9voffvgBU6ZMwYULF1CrVi3MnDkTTz9t/hfKggUL8OmnnyI1NRUNGzbE559/jieeeKLUfWdgspO8bGDnLODAMqD7p0CT54pud+04sLSLWFEyqRQE6POAu/nDsZEvAk9OB9Te5q/Va4G//wdh11zIBL102BDaAYoXNlkEtT3nb2DQkr1wUcoxqK4CAzyPou6tnVBcTQA6TcWdZiNw7GomDl25jfPX7+JaVi7SMnORmpmL7DwDfNxdsO+9TuIvrs2TgBM/A0N/BnxqYf6f5/DZltMIrKLBjN4N0b5OVTGkCAKwvCdweTdQpxty+6/BHW1BmV6AgE/+OI0NB6/CRSnHihcfR3QtHyD9JIQF0ZBBQA/txzguhMDPyxXzXBcj+s4W3KscjntDt0F17Qg8t74B5Q0x6CSFvoiP8wZi38WbMIjjgVg8JBJdGvgD188AsV2BnJvF/tgEyHAQ9ZCiCkXLWj6o5pk/3HDvOnBum/nPqThRI4Bus0sOyk5ix6lruLB6HF5W/gFEPAP0+6ZUr8u8mQHj1+1QWXsVfxoa4yXd2+gU7o9mwZXzq5caVJffQODKaMgFPZ7BJ9ifG5j/agF73SbA33gNeX2XwqVx/4r7gEQOyikDkzN7pAOTQQ8kfA24VwPqdLEMHMW5mggk7wGavwS4PHhYp0gX/wZ+fh24dVF8XL8P8OyKotsmLAF+f0vsY7MXgPpPQfCLgD4nE6od04HEZWI7z+pA/d5mv4SN57ZBniEOU2wxNMdSfQxWucyGq0wH9I8FGvYze6v3f0rC3X2r8Lr3LoTkHL+vIzJgyEagVkeLLuoMRrSYuQ23s3X49uUotAp2Bz4JFSthDfsB/WPR/YtdOJGahdlPR2Dg40EFLz78LfDjSEDlBozeJ4bB++gNRoxccxBxJ67Bw1WJ715pibCq7jj65TOIvrcD2w1NceHJWLwceBWyFb3Evg6PAwJb5J9AC+yaA/z1ifjcKztws1JDfPjrCWw6lIKYhv5Y+HwzYHkP4PI/Fu8vyJWQhT4BhPcC6vWEXuMLhVxmOW9Alwtc2CkOm536Dci9bf68XAm0Hgd0nPqfCUsAkJSSiclfLcdPru8DKnfg7XOWfy5SjwJH1wGCOBR7/a4WKUm70QSncVXwxVT/hRjbo0XRFaEfhgNJP8AQMQC7Gn6EI1cy8aT3FdT/rW/++50FXNwfwiclcizW/v52qjlM5CCOrgO2vCd+r3ARJymG9wLq9gDcfSzba+8COz4E9n0NQABuXgB6zi329DqDEapCwwTIuQ3EvQ8cFMORIFdBZtQhL/MaXIo+hVi5AIDwnkCnqUjNzMH4Jftw5OptDGs1AmMG9oHH1jfEvuxbaPZSOYDrgjfe1w2DvH5v1HJTYX7iSUxQ/SBWgB7rLIVEg8GI4COfY7jLD4Bp2lJglHg90pKAo2vFX1yv/S3OUyhEpZCja31/rDtwBb8dS0UrpBcMGyZtQErDkTiRmgWFXCZWckyybwJbp4jft3u3yLAEAEqFHP83qCleXLYfey7cwNBlCfDzUiP3Zg/EuexEJ8UhdApKBX59U3xB8xcLwhIgzqnp8B5w8yJw7Hvg1/Go8vIOvNI2DJsOpWD7yXRk718Ft8v/iMHttb+x+5oSr61KhL+XGtve7gwUmrha7F82KjVQt5v4ZTQAumzz5+Uqs/P8V/h7q3FEqIVkoSqCdNeBs1vE+XUm2rvAd4OArKvSoar5X3lQIrXrYsRGdyx+4mr0KCDpByiOb0T7LjPQvm5tYHP+PxTqxjAsEZWRXbdGISd1KX/fPRcPcSLx2a3Az2OB/9UGVvQSqztZqWKbc9uBBdHAvkWA6ZbYA7HA1QNFnvrL7WfR4IMt+GTzKRiNglhxWNBSCkv3Gg3FZNXbAIDraVdQbIH0rnjHFdyrYefpdPT4cjf2XbyJXJ0Ri/46j9br8rA0YjV0XT5BbsvxOBT8ElYq+2G+/inM0g3CINUX6PP8KMwfHIkXokOwyNALF4QA4O41YEf+bdiCgGsb38VwozhR2tB6AjDhFDB8K9BqLNDrCyCgsThUtX6oWLG5T/dGAQCALcfTYDy7Lf+o+AtQu018n+gwH1Rxd5HeE3+8C2TfAKrVB6JHP+gnBbVKgSVDm6NRTW/cvJeHk6lZuOMejMw6+UMx3w4AMs4A7lWBTu8XfZKuM8WAmHoE2P8NwgM8UcfPA+6G21DETRXbtJ8I+NbGjgs5uAcNWtQNgsyakCNXAK6e5l//wbAEAFXcXKBSyPGLIVo8kLTBvMHOWWJY8qqJlIYjsVjog/n6p7DRfSD0Q/9Ai1adHnyXT41IILCleCdgwhLxLsjjG8Xn7quSElHJGJio7JLFbWjwzApgdALQYQoE/wjxVteLf4tDYXPDgfktgdVPA5nJgHcQMHgD9BEDAQjAL+PFob1CNiRexdy4M8jTG7F+50EcmtsHWPsccCcVqFIL6f02ouvZPtiXJQ4/eOhv4u+zGUX3Mb/CFJdsxLBl+3HzXh4aVPfC/55pjDp+HsjM0eHDLZfQ+s/H0PSfaPQ93Rnv3+2HJS5DIGszHhve7Imu+VWdev6eCKpWGZN1L4nnTlgiDi/+8S6qHxdvz9/kPw6KJz8AvAIK+qBSA8+uBNSVgJREYMtki262quWDSm4qZNzNQ+6preLBDu8BMjnCMnYiQnYB3SMKnTNxmVjtkSmAnvMAharEH5eHqxLLX3wckcGV0SSwEjaNag2f7lPEyk3eHbFR11mAppiJvh7VgM7TxO93fATZnVT0aVoDE5Vr4aq7DVRrALQU7/D564wYVNvVqVpivx51crkM1TzV+MUg3sGLM1uB3Czx+7RjwF6x8nm82TR0OtIOH2ufxT/Bo9Ft3Hy4hRZ/x6YZU6A+sBQ4v0P8s+TqDTzWybYfhugRwMBEZZOVCty6BMjkQODjuOUWigVCX7S6NR2dDV/gRqupQM0WAATg+kkAMnGy7qg9+PxiEFoeaI8cpZe40N6+gsXIEi7exMSNRwEImFLzMLa5voXIuzthgBw5j7+OS89uRe9fgau3cqCpLAYIb1k2vtxyrMgqkz5LnNj9w2nxzq3BLYOwYWQr9I+siT/GPYHP+jdCgLca6Xe0yNEZUNfPE7OfjsCeiZ0wMaYevDUFQUQmk6FXo+rYY2yAf9w6iZ9t5VNAwtcwQoZJuuHwbl9MpadyCPD0EvH7/UvEW/oLUSnk6FLfDzVwHW6Z58Qg9PiruFtXrAC8pVqPrg3y73BLEUMaADHABEWV8ocGVHF3wYaRrfDj6NYIrOIGVA4W7zgDgNB2QEQJk3+bDRN/rnl3gM0T0d83GQOUOwEAGR1mAwoVrtzMxvnr96CQy9DqMd9S9+1RVs3LFaeEQNz1rAUYtMDp38VK0K9vAIIBN4K6oe82D+TqjGhXpypih7UocRkKM/V6AJWCgZxbwM9jxGPhvcq+hAERcQ4TlVGyuN9erk99TPvtEjYdSoFWb1oduSr2BXRF9y5vAZkp4tBdtXBxWArAgcs3kSF44YOcAfhUtQTGP2dC3qAPLusr47VVB+BnTMfiyqtQP+MAIANOIhRvaV/BnaT6yD10GOl3tHismgdihz8O4UsXyAx5SE25grgT18zm+BiNAm6lX0VVAHeVlfFlv6Z4qnF16XmFXIZnmgeiV+Pq2H4yHT4eLogKrfLA4Y2ejQPw+bYzeOP2M9jreQBybSYEmRxva1/FVpeOmPaggFCnC/DEO8DfnwK/jANC2ppVomIiAqA8lD8/q2YLyDSV8JP3YAwQNqCd/Ahw4yAgrwt8P1QcAg3vJQ75lVeXD4GARuL5SppMLZcDPT8Hvm4HnPgJ1S6L/x98q++AO+k18Fo48PdZsarXLKiSWeCk4vl5qgHIcLZaFzS9s1AcltNlA1f3Ay4emKodgjy9EZ3Dq2H+883gqizjYpFyhfgPli2TxOoSADS0clVxokccK0xUKoIg4MS/WTi8+w8AwHdpNbB2/xVo9UY0qO6FsPxVnLNy8hcG9K4BNB4ohSXxOXEIbr2hHfYb60Cuy0baunF4Zfk+9NH+jDjXd1E/5wCgcAU6fQDVa38iq3I4km9mI/2OFvX8PbH21Zao5q2BzF0c8vGVZWJu3BlxvlO+BX+ehbvuFgDg/YHtzcJSYWqVAj0aBaBlWMkbKNeq6oEG1b2QbvTCrgbTgWr18UPIdGwwPoEnw/1K/kXWfqJYodFlAwnmq2y3ruWLTqqjAICrPuLwzNpzSnxvaCc22PGhuAJ35hVxDane821zt5hKI95BWNxQ3P38I4CWI8Xv711HrksVfKIfhE2HUgAAf50WAxOH40rP31ucn5Xokb+8yvkdQNw08fuOU5B4S3x+dIfHyh6WTJoOBlzyt4Bw8xErikRUZgxM9EA6gzhJuv3/dqL7l7vg+u8+AMBBhKNHRAB+GBGNX8e2QZPASgCATFNgKoLpuf890xSxlV6HTlDA/99tmJ85Bh+oVkGDXCC4NTAyHmg7AY8FVMamUa3RqV41tKtTFd+90hK+HvlDCfmBKcj1Dk6l3cFvx8R/Pe86ex2Lth2Fm0ycYF23Vi2bXYte+cFrYVo9GEfE438p9QHAfI5RceQKoPV48fvEZeJ6UvlcoEdrubgcwW859ZF8IxvHUjKxwNAXgsJFvGX/3DZxa5MBq0q/jENFaD8J8Bbv9jM8+SGyFZ44lXYHSSmZiD9/AwDQrk41+/XPyVTzEv9/PqHzA/wbAUY9oM0EAhojt+lLuJYl/n8c4lOOO9rUXgWrQ0c8Ayg4sEBkDf7JoWIdvXob7/xwFKfSxInBvsoc1JVfAQDMHP8qvHxrSG298odgShOYGgd6o8fY5xC/eA/aXV+D2vIUGFQeUHT9UJwrU2gVZ18PVywd1sLyZB7ivJ6+tV3w6zHg821n0CSwEsatPQwf5K+4rXIHXD2s/vz36xERgNl/nMK+izex+XgarmVp4emqRNs6pZyvUzdGnNN06xJw5DugxXDx+NUEqI3ZuCF4Iva8N4zH/gUABIbWhazmi+KaVwDQa564oao9uXoAL/4OZJyB+2Od0eHkAWw9cQ0f/Hwcd7V6+Li7oEH1R2xdsnLwz99jMD1LK965lnYUgAzo+TmSb4uru3uqlajkVs4hzk7vixXOx6zbH5CIWGGiIuTkGfDx7yfRZ/4/OJV2B5XdVPi0XyPsfs4NcghAlVpmYQkoOTAZjQKycsXnvNQqqFUKtHvlf7gV1AV3aveFYkyCuKBlabe88BArTG0CjKjspsKF6/fQZ/4/uHkvD49X1Zu1sZXAKm5oFlQJggBM/VHcXbxz/VIMx5nIFUBU/pDW3oXi5F5ArB4B2CNrjPS7OizaeR5A/pIDT7wNBLcB2k0UhzgdQaUg6Rdvn6bi/weJl8Uh0CfqVLXYf4yK55cfmNKycoGmQ8Sf9ZMzgBqRuJRxD4BYXSrrJqEWFCpxvzxrFowlIgCsMNF9TqZmYcTqRFy+IQ4ZPdW4Oj7oVR8+Hq5AXP7WDcHRFq8zTfLNytVbPAcAd/P00n5qpnAFFzdUfmm9dR11F4d9XHMzMKJdLcz64xRu3MuDl1qJd9tUAX4vaGNLvRpXx8Hk27hxT/zXf6mG4wpr+jzw50zgxlkxKNXpIgWmzOrtgQviNZTLgG4N/AEPV+DF32z8KWynY71q8HRV4o5W/Llz/lLZ+OUPyV3LyhUXfS30s06+Kf4ZLGmzYyJ6OFhhIsnRq7cxcPFeXL6RDX8vNZYObY4vBzUVwxJQsP5SUCuL13qXUGHKzBaPuyrlUKusnLxaWP6QHO6m44XoEPh7qSGTAfMGNoEvbue3sX1g6hERIM239nBVom3tMt4+7+opTrQGgL3zgTtp4po7AIIe7yE1ezy0ilPsCq9WKRATId6hKJOh7NfjEWeqMN3J1SM7z/wfG5dumCpMDExEjoCB6b/KaAB+nQDsKn4LksISL9/C80v2ISj3NL71XoBt/V3RKdyvoIEuB0g5KH5fRIXJSy0WK4sNTPnHbXa7uWm47d51aFwU2DiqFX4b2xYd6/kVbItSAYGpmpcaLUPF7V86hVezLvxFvSaut3RhJxD/f+KxgCZ4vGFdeLqK17HMlSs7GtAiEDIZ0DLUpyBcU6l4uCrh5iL+P2Sa4G1iqvIGV+EWJkSOgIHJCWn1Boz59iAW/XW++K1BLv8jru779/9KPN++CzfwwtJ9qJN3HOvUH6OVdjc8fh4O5GYWNEpJFLdY8PAHKodanEMakismMEnzl2wVmEzDbXevAQCqV9KgvmmycaFtUSrCO93qol2dqhjbsbZ1J6gUJM4nAYA9X4n/fawzXJUKTO1VHz0bBeDpZjVt09mHIDK4CjaPewILBzezd1ecjkwmkyZ+X8vKNXtOCkysMBE5BAYmJ3Q4+TZ+PZqK2X+cwvRfThQdmk79DgAQ9LmWzxWy+2wGhi5LQIT+GNaoP4GbkH+7+920gj3TAOBy/nBccHSRawB5u5UQmGxeYTINyV23fM4UmCqgwgQATYMqY8VLj+OxauW4A6/lfSuD134SAPBs80B89VwzeLg61/TCuv6eqORW7FbI9ADVCs9jyqczGJFyW9yIObg8SwoQkc0wMDmhu9qCuQ7L4y/hvU1JZgs3Zt7Lw82DmwAAMsEgDs8VwWAUMG7tIbQwHMZK10+hFnKBsA7AoLVig4QlBcNwD5i/BJjPYSoqwFXYkJw2E9DdFwrvVWxgsonAFvlbyEDc26tGc/v2h+ymqApTyq0cGIwC1Co5qjnBXDaiRwEDkxPKzhMDkLdGBbkM+C4hGW//cBQGo4A/jqVixOcrUUWXJrUXDHlFnidHZ0CjnH34RvU/uAh5QO2uYliqGwNEPAtAAH4dD+jzgCsJ4ouCWhZ5Li+1GIT0RkHqX2E2D0zqSoAiv6JhCkgmFTwkZzNt3xL/26APFxN8hElLC2QWzGG6bLpDroobl2kgchD8W9oJ5eQHkmZBldCnaQ1M+P4INhy8igOXb+LyjWyMVewBCuUSrVYLtUpjcZ7sXC3mqBbCVaaHUK8nZP2XAcr8ENJ1JnB2C5B6RNz/LO8O4OpV7MKJbi4KKOUy6I0CMnN0cL9vSMm0LYppcni5yWTiat9ZKeKwXKUg8bggFJr07eC3uNftBow7Cnj6l9yW/rOqmSpMdwoqTJfz75DjcByR42CFyQmZbj92c1Gid5Ma+GpQUyjlMly+kQ2lXIbBlZLM2mu1Rc9j0ubkoIrsLgBA1mdBQVgCxOGsztPE7498K/43MEpcfLEIMpms0FpMlvOYbF5hMvURMK8w5d0V92sDHL/CBACVg7lz/COuYLXvwoHJdIccJ3wTOQoGJieUrRMrTKbbkWMiArD8xcfxbPOa+OPFMPjdOwWgoIyfk1t0YMopHKSUlhUoNBtWMM8GKHI5gcKkeUzZxQcmm90lBxS6U65QYDJ9b+NtUYgqimnxyrSsIipMvqwwETkKBiYnlK01D0wA0Ka2Lz7t3xi1b+0SDwRGITd/XE5bTGAyO64oIsjI5UDPz8U1gwBxY9wH8HzA4pU2X1YAKBhyKxyYnGU4jiifnzTpWyvdMMEKE5Hj4RwmJ2SaVK1xKeLHdyp/a4V63aG/chSADlqt1rIdAG2eGJj0UEBZ3F5V/hFA/6XAzQvikNwDPGi174oZkstfWqDwkFz+ukzSc0QOzrSsQJ7eiNvZOnhrVNK2KCGcw0TkMFhhckI5OtMcpvvmE+XcBi7lV5jq9oA+Pw+bgtH9TBUmfUm5uUFfoO2bRa6/VNiD9pOrkMD0oCE5d1aYyDm4KhWonL+O2bU7ubh2JxdavRFKuQzVK6nt3DsiMmFgckKmCpNFYDq3DTDqAd86gO9jMMjEIJRXzKTvvDyx8qSX2SbEPGh7FJsvXAmUMCTnBBO+ifIVLC2QKw3H1aysgVLBv6KJHAX/NDqhgsB0X2VIGo4TN3EtCExFD8mZgpSpXXkVtz2KIAgFywpU+JBcuvlzRE7AT7pTTitN+A7icByRQ+EcJidUsKxAoQqTPk+sMAFA3fzAJFcBRkBXzJCcLr/CZJTbJsQUF5hydUbkGYxmbWxCGpIrtD0Kh+TICRVe7Tsn/y5YTvgmciwMTE6oYNJ3ocB0aRegzRIrKzUiAQDG/KG2vLyiV/rW55kqTLYNTPcPyZkeK+QyuN8/jFge92+PolI7x7YoRPcpvLTA7fxlObjpLpFj4ZCcE8opag6TaTiuTjdxOQAUVI70umKG5GxcYfIqITB5qZWQlTBxvEyK2h7FWbZFISrEz7tgaYHLN7nKN5EjYmByQhaTvnW5QNIG8fv6vaV2glwsIBryig5MhvwgJdh4SO7+wGRag8mmw3FA/vYohYblBKHQHCYGJnIefp4FQ3KXM0xLCrDCRORIGJicgdEA5GVLDy0mfZ/6Fci9DXjVBMLaS+0EuVh90euKnsOkzx+qs3Vgun9rFNPK3zYPTEChO+Wuidui6HPyjzMwkfMwTfo+l34Xd7TiHMVAzmEicigMTM5g02vAZ7WAfw8DKGLS98GV4n+bPm+215ugMA3JFTOHSZ8fmBQuRT5fViXNYbLpHXIm7oX2kyu8LYoLhzPIefh5i3OYTBO+A7zVUKtsON+PiMqNgcnRXUkAjq0XN5TdPh2CIEh/qWpcFMCtS8DFvwDIgCbPm782PwgZ9EUHJqNpblNR26JYwUstnidXZ4RWb5COV8i2KCYehYbkuAYTOSkfd1co5AXz+4JYXSJyOAxMjm7HRwXfn9+BvPO7kb/dlDgkd2iN+CCsnbjzfSGy/CBkLGbStylIyWwUmDzVSmkx8MJVpgpZ5dtECkzXCm2LwsBEzkUhl6Gap6v0mFuiEDkeBiZHdvFvsXokVwF1uwMA5Ds/AiAmJo0CwOH8wNR0iOXr84OQoZghOcFUebLRkJxcLoOnqzivyrRQJVDBgamoITmuwUROqJpXwTYoQZzwTeRwGJgclSAAO2aK30cOA7r/D1C4QnV1L9rIk6BWyaG4uBPIShFvr6/X0+IUcqUYhARDMUNyerHyJFPaJjABgLeb5Tymh1NhKjwkx1W+yfn4e7HCROTIGJgc1bntwJW9gFItbnzrXQNoMRwA8Jbye7ipFMCh/MnejQaIizbeRyYFJsu93YCCCpPchoHJNI+p8Grf0rYo6goMTPfSOSRHTs2vUIWJi1YSOR4GJkckCMCOD8XvW7wMeAWI37d5AwalBk3k59FfvhM49bt4vFkRw3EA5ErxX6xCMZO+TZUnUztbKOpOuQrZeNdEWocpvWCLFA7JkRPy45AckUNjYHJEp34DUg+Lt8e3eaPguEc1pNYdCgB4R78IMOqA6k0B/4giT6NQigGluCE55FeeFCobDskVsRZTxQ7JmbZHyQIyk/OPscJEzscUmKq4u1RMNZaIyoWBydEYjcCf+XOXWo4E3H3Nnj732EvIEjRQIv+2/aIme+eTq8TKkayYITlZRVaYsgtVmKRlBSpg68LC26NcPy3+l3OYyAnV8/cEAETU8LZzT4ioKAxMjub4RiD9BODqDbQaY/F0ltwTS/XiHXNQaoCI/sWeSmmqHBVRYRIEQTpuywpTUfvJVWiFqfD2KKbPySE5ckINa3jjp9Gt8cXAJvbuChEVoQL+yU/l4l4V8GsI1O8DaCpbPJ2t1WOJoQdaVspEdMdegLr4f40q8itMMOotnsszGKUqldKl4uYw6QxGaSuXCglMgDgsl3W10GMOyZFzahxYyd5dIKJiMDA5mrB2wGu7xPlJRcjOMyAbaqypMRnRkc0eeCpl/p1zCqMOBqNgtpJwTp4BKujz29kuMHndN4epcKXJs6LmZRQegnPx4LYoRERkcxySc0RyOVDMvCLTtijSPnIPoMofalNBj1ydwey5HF1BYFJU4F1ypjvkPF2VZoHNpgoPwXE4joiIKgADk5Mp2Hi35OKgaW6SSqaXhsUKzmOAyjRx3EZbowCAl1rsV2b+2ksVuvGuSeEhOA7HERFRBWBgcjKm4FOaCpPp7rciK0x5Bqhk+XObbBiYpGUFcsyH5Cps/hJgPiTHwERERBWAgcnJZGtLH5hMQUgFg0WFKVdngAtMgakC1mEyDcnl5q/yXRFLCpiYDckxMBERke0xMDmZ7PxKkaYUQ3IFgUkvzX2SzlNo0ndFVJjuaPUwGIWHVGHikBwREVUsBiYnkyPNYSpNhalg0rdp7pN0Hl3hwGT7dZgAscpUoduimBSuKnHSNxERVQAGJidTljlMpsqRi6yYOUwVEJhUCrnUt6xcnR0qTFzlm4iIbI+BycnckwJTaYbkxCCkhAE5eUazp3J0BrhUwKRvwHxpAVOFqUL3xlJ7F4Q+DskREVEFYGByMjYbkqugChNgHpikCpNbBQYmmQwI6yBWl6rWq7j3ISKiR5bdA9OCBQsQGhoKtVqNyMhI7Nq164Ht58+fj/DwcGg0GtStWxcrV640e16n02HGjBmoVasW1Go1GjdujM2bN5u1mTZtGmQymdmXv7+/zT9bRTANyWlKE5jkYhWqpIUrbR2YTNUks8BUkUNyAPDcOmB8EqD2qtj3ISKiR5Jdt0ZZt24dxo8fjwULFqB169b4+uuvERMTgxMnTiAoKMii/cKFCzFp0iQsWbIELVq0QEJCAl555RVUrlwZvXr1AgBMmTIFq1evxpIlS1CvXj1s2bIFffv2RXx8PJo2bSqdq0GDBti2bZv0WKEoRQBxADn5gcm9DENyLrBcuDInzyDtJWfrITlpe5QcvbRFSoUuXAmIVSalbYMfERGRiV0rTHPnzsXw4cPx8ssvIzw8HPPmzUNgYCAWLlxYZPtVq1bhtddew4ABAxAWFoaBAwdi+PDh+OSTT8zavPfee+jevTvCwsIwcuRIdO3aFXPmzDE7l1KphL+/v/RVtapz3F1VtknfBUNyRS0rUBHrMAFFD8lV6BwmIiKiCma3wJSXl4fExER06dLF7HiXLl0QHx9f5Gu0Wi3UarXZMY1Gg4SEBOh0uge22b17t9mxs2fPonr16ggNDcXAgQNx4cKF8n6kCmc0ClLwKdWQXH7lSCkzSJUpk4ockjMLTNkPaUiOiIioAtktMGVkZMBgMMDPz/w2cD8/P6SlpRX5mq5du+Kbb75BYmIiBEHAgQMHEBsbC51Oh4yMDKnN3LlzcfbsWRiNRsTFxeGnn35CamqqdJ6oqCisXLkSW7ZswZIlS5CWloZWrVrhxo0bxfZXq9UiKyvL7OthK1wlKm+FKSdPD5WsoobkxOHC29l5uKMVQxkDExEROTO7T/qWycx3sBcEweKYydSpUxETE4OWLVtCpVKhd+/eGDZsGICCOUhffPEFateujXr16sHFxQVjxozBiy++aDZHKSYmBv369UNERAQ6d+6M3377DQCwYsWKYvs5a9YseHt7S1+BgYHl+dhWKTwPSa0swzpM0D+4wiSvmGUFUm7nQBDEYxW6NQoREVEFs1tg8vX1hUKhsKgmpaenW1SdTDQaDWJjY5GdnY1Lly4hOTkZISEh8PT0hK+vLwCgatWq+PHHH3Hv3j1cvnwZp06dgoeHB0JDQ4vti7u7OyIiInD27Nli20yaNAmZmZnS15UrV6z41OWTU2j+klxedKg0U3gvOYu75IwVPofp6q0cAIBaJYdraQIeERGRg7JbYHJxcUFkZCTi4uLMjsfFxaFVq1YPfK1KpULNmjWhUCiwdu1a9OzZE3K5+UdRq9WoUaMG9Ho9NmzYgN69exd7Pq1Wi5MnTyIgIKDYNq6urvDy8jL7etiydWVYgwkwG5LLtbhLTl8he8kBBRO8r97KBsDhOCIicn52HSeZMGEChgwZgubNmyM6OhqLFy9GcnIyRowYAUCs6qSkpEhrLZ05cwYJCQmIiorCrVu3MHfuXCQlJZkNpe3btw8pKSlo0qQJUlJSMG3aNBiNRrzzzjtSm7feegu9evVCUFAQ0tPT8dFHHyErKwtDhw59uBegjO5pyzDhGygITDIDsrUPZy85oGCRSp1BHI9jYCIiImdn18A0YMAA3LhxAzNmzEBqaioaNmyI33//HcHBwQCA1NRUJCcnS+0NBgPmzJmD06dPQ6VSoUOHDoiPj0dISIjUJjc3F1OmTMGFCxfg4eGB7t27Y9WqVahUqZLU5urVqxg0aBAyMjJQtWpVtGzZEnv37pXe11FJQ3KqUv7Y5AXtdDqt2VO52jwoZfnbpVTQkJwJlxQgIiJnZ/eZuKNGjcKoUaOKfG758uVmj8PDw3Ho0KEHnq9du3Y4ceLEA9usXbu2TH10FKbtTcpaYQIAXZ55YNLn5RVqVzGTvot7TERE5GzsfpcclZ5paQB317IHJv19FSazxxW0NYoJAxMRETk7BiYnIu0jV+ohOQUEiHfT6XV5Zk/pdBVXYVKr5HBRFPyvVeHbohAREVUwBiYnck9bxrvkZDII+dUjQ6GKktEowKgXA5MgUwBy297yL5PJzEISAxMRETk7BiYnklOWfeRM8id+G3R5EPJXkczRVdw+cibehRaq5JAcERE5OwYmJ5Jdln3kTPIDkVzQIc8g3hWXozNAJauYNZhMCleVGJiIiMjZMTA5EVOFyd2l9Dc3ypRiYHKBHrl5Ruk8KojnklVQYCockrzUdr8Zk4iIqFwYmJxImZcVACDLrzApYZBWCn84Q3KsMBER0X8HA5MTuVeOOUyqQhvwihWmih2SMwtMbgxMRETk3BiYnIhVk77zK0guMr20jlN2XsVti2JSeC0mVpiIiMjZMTA5EdOQnFsZ5jAV3oDXFLhyzSZ9V/yQHLdGISIiZ8fA5ESsqzCJYUUF8wqTy0MaklPKZWXrLxERkQNiYHIi0krfVgzJqWCQXp+jewhDcvmByVujgkwmq5D3ICIielgYmJxItlRhKsuQXEGFKVdXEJiU+csKVFRg8vUQz+vjUTHnJyIiepi4QI4TKZjDZN2QnFRhytNX+F1yzYIqY3zn2mgeXKVCzk9ERPQwMTA5kexy3CWnkhVeVsAIlwqe9C2XyzC+c50KOTcREdHDxiE5J2EwCtDqxZW6rRuSMxRM+tbpK3wOExER0X8JA5OTMIUdwMoKU+FlBR7CwpVERET/JQxMTiJbKwYcmQxwVZbhxyYvelkBKTDJGZiIiIhKwsDkJKT5SypF2W7TL2rSt84Alwq+S46IiOi/hIHJSUiBybWM8/QLbY1iWlYgV8chOSIiorJgYHISOTorlhQApMCkhEFaliA7r+K3RiEiIvovYWByEve0+at8q8oamMSKlDiHSbzLLocVJiIiojJhYHISVq3BBBQMyUGPXGkdpsJ7ybHCREREVBIGJidRMCRn3RwmFfTIzj/Hw9gahYiI6L+EgclJWF9hKrRwZaEKE4fkiIiISo+ByUnklHNITmm2NYqhwrdGISIi+i9hYHIS0qTvsg7J5S9M6VJo4UrzSd8MTERERCVhYHIS2VYvK1AwJJedZ0Ce3gi9UeCQHBERURkwMDkJ03Cau5VDciroodUbpbWYWGEiIiIqPQYmJ2Ga9F3mIblCgQkAbt7LAwC4ykx3ybHCREREVBIGJidh/aRv08KV4usZmIiIiMqOgclJ3MsfStNYOSTnKr+vwiTnkBwREVFpMTA5ifKu9G2qKFlWmBiYiIiISsLA5CQKJn2XdQ5T/rIC+esu3cgPTC68S46IiKjUGJicRHY5h+Rc7qswqVhhIiIiKjUGJidh9aRvecE6TEChwMRlBYiIiEqNgclJ3CvnXnIu9y0rwIUriYiISo+ByUnklHMdJmWxgYkVJiIiopIwMDkBvcGIPIMRQPlW+gYKApOSgYmIiKjUGJicQHb+prmANZO+xYqUAqa75LTiY4FDckRERKXFwOQETMNxCrkMLooy/shMQ3L5ASlXZ8x/rBOflzMwERERlYSByQnc04phx02lgEwmK9uLpcCkAyDkHxQKVZg4JEdERFQSBiYnULDxbhmH4wCzITdl/tICpiUG7n+eiIiIisbA5ARy8ucwubuW8Q45wKyCVBCY9EU+T0REREVjYHICUoVJZUWFqdAcJdNaTAxMREREZcPA5ARy8rdFKfOilYDZkJtKCkymITkZILfinERERI8YBiYncE9bjjlMMlmh7VHuqzApXMTniYiI6IEYmJyAaR0m97Ku8m1iWrxSpjf7L4fjiIiISoeByQmUa0gOkIblVPdP+uYdckRERKXCwOQEyrWsAFAoMIlByYXbohAREZUJA5MTMAUm6ytM5vvJceNdIiKisrF7YFqwYAFCQ0OhVqsRGRmJXbt2PbD9/PnzER4eDo1Gg7p162LlypVmz+t0OsyYMQO1atWCWq1G48aNsXnz5nK/rz1l5w/JaayewyRWmCyWFeCQHBERUanYNTCtW7cO48ePx+TJk3Ho0CG0bdsWMTExSE5OLrL9woULMWnSJEybNg3Hjx/H9OnTMXr0aPzyyy9SmylTpuDrr7/G//3f/+HEiRMYMWIE+vbti0OHDln9vvZmqjC5l7vCJJ7HRcbAREREVBZ2DUxz587F8OHD8fLLLyM8PBzz5s1DYGAgFi5cWGT7VatW4bXXXsOAAQMQFhaGgQMHYvjw4fjkk0/M2rz33nvo3r07wsLCMHLkSHTt2hVz5syx+n3tLae8Q3L5ywooZawwERERWcNugSkvLw+JiYno0qWL2fEuXbogPj6+yNdotVqo1WqzYxqNBgkJCdDpdA9ss3v3bqvf13TerKwss6+HpWDSd/mG5ExByU1hzD/OOUxERESlYbfAlJGRAYPBAD8/P7Pjfn5+SEtLK/I1Xbt2xTfffIPExEQIgoADBw4gNjYWOp0OGRkZUpu5c+fi7NmzMBqNiIuLw08//YTU1FSr3xcAZs2aBW9vb+krMDCwPB+/TLLLvayAGIzUMjF4eSgZmIiIiMrC7pO+ZfetNC0IgsUxk6lTpyImJgYtW7aESqVC7969MWzYMACAQiGGiS+++AK1a9dGvXr14OLigjFjxuDFF1+UnrfmfQFg0qRJyMzMlL6uXLlS1o9qNa1eDDiuSit/XPnByD0/KLkphPzjHJIjIiIqDbsFJl9fXygUCouqTnp6ukX1x0Sj0SA2NhbZ2dm4dOkSkpOTERISAk9PT/j6+gIAqlatih9//BH37t3D5cuXcerUKXh4eCA0NNTq9wUAV1dXeHl5mX09LEJ+vpFbu41JfjAyDcVplPl7ybHCREREVCp2C0wuLi6IjIxEXFyc2fG4uDi0atXqga9VqVSoWbMmFAoF1q5di549e0IuN/8oarUaNWrUgF6vx4YNG9C7d+9yv6+9CBATk9XbvpkCk1I8jzvnMBEREZWJlbOIbWPChAkYMmQImjdvjujoaCxevBjJyckYMWIEAHEYLCUlRVpr6cyZM0hISEBUVBRu3bqFuXPnIikpCStWrJDOuW/fPqSkpKBJkyZISUnBtGnTYDQa8c4775T6fR2NMT/fPGjI8IHyg5GbPH/yuBSYOCRHRERUGnYNTAMGDMCNGzcwY8YMpKamomHDhvj9998RHBwMAEhNTTVbG8lgMGDOnDk4ffo0VCoVOnTogPj4eISEhEhtcnNzMWXKFFy4cAEeHh7o3r07Vq1ahUqVKpX6fR1N/ogcrC0wFVSY8ucwyVlhIiIiKgu7BiYAGDVqFEaNGlXkc8uXLzd7HB4ebrYAZVHatWuHEydOlOt9HY0glHdITgxGmvwKk1rOrVGIiIjKwu53yVHpyaytMeUvXKnOryxxSI6IiKhsGJicgOkuufJO+lYrxAqTq4x3yREREZUFA5MTkO6Ss/YE9y1cafqvqfJERERED8bA5ASE8s76vn8dJoXB7DgRERE9mN0nfVPJCvJS+ZYViAjQYERILdTLVQP/gkNyREREpcQKkxMw3SUnL2eFyVMFTIypJy0vwMBERERUOgxMTqBg0nf5tkaBIS//vzrz40RERPRADExOQBqSK+c6TFJQMgUnVpiIiIhKhYHJCUgLV1p7Aikw5Zn/lxUmIiKiUmFgcgLlrzAVNyTHChMREVFpMDA5AWlZgXKu9A1D/pYoHJIjIiIqEwYmJyAtXFnuOUwckiMiIrIGA5MTkO6Ss/YEHJIjIiIqFwYmJ1D+ZQWKu0uOFSYiIqLSYGByIuVduBJGBiYiIiJrMDA5AaO0rICtF67kkBwREVFpMDA5gYIhOStPYDEkx8BERERUFgxMTkCQVmKyksVdctwahYiIqCwYmJxA+StMpiE5bo1CRERkDQYmJyCt9F3uhSsZmIiIiKzBwOQEbDeHiUNyRERE1mBgcgrlXembFSYiIqLyYGByAgUrfZd34co88WQMTERERGXCwOQETHOYyr1wpSEPMBoKzsghOSIiolJhYHIC0sKV5Z3DBAHQ5xRxnIiIiB6EgckJCNIyTOVc6RsA8u4VfC9nhYmIiKg0rApMO3futHE36EEEm1WYYB6YOCRHRERUKlYFpm7duqFWrVr46KOPcOXKFVv3ie5TsA6TleRFVJjkqnIkMCIiokeLVYHp33//xbhx47Bx40aEhoaia9eu+P7775GXl2fr/hEgJSaZtQFHLgdkCvF7Xbb4X85fIiIiKjWrAlOVKlXw+uuv4+DBgzhw4ADq1q2L0aNHIyAgAK+//jqOHDli634+0spdYQIKAlLe3fzHHI4jIiIqrXJP+m7SpAkmTpyI0aNH4969e4iNjUVkZCTatm2L48eP26KPj7xyz2ECCgWme+aPiYiIqERWByadTocffvgB3bt3R3BwMLZs2YKvvvoK165dw8WLFxEYGIhnnnnGln19ZJV7LzmgoKLEwERERFRmSmteNHbsWHz33XcAgMGDB+PTTz9Fw4YNpefd3d0xe/ZshISE2KSTj7py7yUHFFFh4pAcERFRaVkVmE6cOIH/+7//Q79+/eDiUnSlonr16vjzzz/L1TkSlXvhSgBQ5P+oWWEiIiIqM6sC0/bt20s+sVKJdu3aWXN6uo80JFeexMQ5TERERFazag7TrFmzEBsba3E8NjYWn3zySbk7RfeRNt8tBw7JERERWc2qwPT111+jXr16FscbNGiARYsWlbtTZE6ALYbkTJO+TcsKsMJERERUWlYFprS0NAQEBFgcr1q1KlJTU8vdKTInTfouT43JtNq3tHAlK0xERESlZVVgCgwMxD///GNx/J9//kH16tXL3SkyVzCHqRwnkYbkGJiIiIjKyqpJ3y+//DLGjx8PnU6Hjh07AhAngr/zzjt48803bdpBKrRwZXlOwiE5IiIiq1kVmN555x3cvHkTo0aNkvaPU6vVePfddzFp0iSbdpAKKkzlSkyc9E1ERGQ1qwKTTCbDJ598gqlTp+LkyZPQaDSoXbs2XF1dbd0/QsEcJjmXFSAiIrILqwKTiYeHB1q0aGGrvlARTMNxQHmH5PJ/1DoGJiIiorKyOjDt378f69evR3JysjQsZ7Jx48Zyd4xEhfKSjReu5JAcERFRaVl1l9zatWvRunVrnDhxAps2bYJOp8OJEyewY8cOeHt727qPj7RCecnGC1eywkRERFRaVgWmjz/+GJ9//jl+/fVXuLi44IsvvsDJkyfx7LPPIigoyNZ9fKSZDcnZYuFKfW7+YwYmIiKi0rIqMJ0/fx49evQAALi6uuLevXuQyWR44403sHjxYpt28FFnXmGywcKVJhySIyIiKjWrAlOVKlVw584dAECNGjWQlJQEALh9+zays7Nt1zsym8Nkk2UFintMRERExbJq0nfbtm0RFxeHiIgIPPvssxg3bhx27NiBuLg4dOrUydZ9fKQJsPGQnMn9FSciIiIqllWB6auvvkJurjgXZtKkSVCpVNi9ezeefvppTJ061aYdfNSZ3SVXnhNZVJgYmIiIiEqrzIFJr9fjl19+QdeuXQEAcrkc77zzDt555x2bd47M2WThyuIeExERUbHKPIdJqVRi5MiR0Gq1NunAggULEBoaCrVajcjISOzateuB7efPn4/w8HBoNBrUrVsXK1eutGgzb9481K1bFxqNBoGBgXjjjTekihgATJs2DTKZzOzL39/fJp/H1ow2u0vuvmzMwERERFRqVg3JRUVF4dChQwgODi7Xm69btw7jx4/HggUL0Lp1a3z99deIiYnBiRMnilyeYOHChZg0aRKWLFmCFi1aICEhAa+88goqV66MXr16AQDWrFmDiRMnIjY2Fq1atcKZM2cwbNgwAMDnn38unatBgwbYtm2b9FihUJTrs1QU8yE5W1aYOCRHRERUWlYFplGjRuHNN9/E1atXERkZCXd3d7PnGzVqVKrzzJ07F8OHD8fLL78MQKwMbdmyBQsXLsSsWbMs2q9atQqvvfYaBgwYAAAICwvD3r178cknn0iBac+ePWjdujWee+45AEBISAgGDRqEhIQEs3MplUqHrSoVZnaTHO+SIyIisgurApMpsLz++uvSMZlMBkEQIJPJYDAYSjxHXl4eEhMTMXHiRLPjXbp0QXx8fJGv0Wq1UKvVZsc0Gg0SEhKg0+mgUqnQpk0brF69GgkJCXj88cdx4cIF/P777xg6dKjZ686ePYvq1avD1dUVUVFR+PjjjxEWFlZsf7VardkwZFZWVomf0RYEs3UFyuH+ihIDExERUalZFZguXrxY7jfOyMiAwWCAn5+f2XE/Pz+kpaUV+ZquXbvim2++QZ8+fdCsWTMkJiYiNjYWOp0OGRkZCAgIwMCBA3H9+nW0adMGgiBAr9dj5MiRZsEsKioKK1euRJ06dXDt2jV89NFHaNWqFY4fPw4fH58i33vWrFmYPn16uT93WVVchYlDckRERKVlVWAq79ylwu7fUNZUpSrK1KlTkZaWhpYtW0IQBPj5+WHYsGH49NNPpTlIO3fuxMyZM7FgwQJERUXh3LlzGDduHAICAqQlD2JiYqRzRkREIDo6GrVq1cKKFSswYcKEIt970qRJZs9lZWUhMDCwXJ+9NGw2h8lipW9WmIiIiErLqsBU1J1phb3wwgslnsPX1xcKhcKimpSenm5RdTLRaDSIjY3F119/jWvXriEgIACLFy+Gp6cnfH19AYihasiQIdK8qIiICNy7dw+vvvoqJk+eDLnc8sZAd3d3RERE4OzZs8X219XVFa6uriV+LpsrHJhsuXAlAxMREVGpWRWYxo0bZ/ZYp9MhOzsbLi4ucHNzK1VgcnFxQWRkJOLi4tC3b1/peFxcHHr37v3A16pUKtSsWRMAsHbtWvTs2VMKQtnZ2RahSKFQQBCEYucDabVanDx5Em3bti2x3w+b2Urf5TkRh+SIiIisZlVgunXrlsWxs2fPYuTIkXj77bdLfZ4JEyZgyJAhaN68OaKjo7F48WIkJydjxIgRAMRhsJSUFKmidebMGSQkJCAqKgq3bt3C3LlzkZSUhBUrVkjn7NWrF+bOnYumTZtKQ3JTp07FU089JQ3bvfXWW+jVqxeCgoKQnp6Ojz76CFlZWRYTwx1B4YzHhSuJiIjsw6rAVJTatWtj9uzZGDx4ME6dOlWq1wwYMAA3btzAjBkzkJqaioYNG+L333+X5kilpqYiOTlZam8wGDBnzhycPn0aKpUKHTp0QHx8PEJCQqQ2U6ZMgUwmw5QpU5CSkoKqVauiV69emDlzptTm6tWrGDRoEDIyMlC1alW0bNkSe/futencLFupuIUrWWEiIiIqLZlgs/vWgUOHDqFdu3YP7ZZ7e8rKyoK3tzcyMzPh5eVVYe+TcVeL5h+JC2xemt3D+hNdjgeWFUx2x2t/AwGNy9k7IiIi52Lt72+rKkw///yz2WNBEJCamoqvvvoKrVu3tuaUVAybxVkOyREREVnNqsDUp08fs8cymQxVq1ZFx44dMWfOHFv0i/KZJn2XazgO4F1yRERE5WBVYDIajbbuBxUnv8JU3rzEu+SIiIisZ7koETkU04hccYt5lhoXriQiIrKaVYGpf//+mD17tsXxzz77DM8880y5O0UFBJtVmBiYiIiIrGVVYPrrr7/Qo4flHVvdunXD33//Xe5OUQHbzWHikBwREZG1rApMd+/ehYuLZYVCpVI9EksKPExSham8iYl3yREREVnNqsDUsGFDrFu3zuL42rVrUb9+/XJ3igqYFq4s/5DcffP775/TRERERMWy6i65qVOnol+/fjh//jw6duwIANi+fTu+++47rF+/3qYdfNQVVJjKeaLCFSW5EihiE2IiIiIqmlWB6amnnsKPP/6Ijz/+GD/88AM0Gg0aNWqEbdu2oV27drbuIwGQlbfGVDgwcTiOiIioTKzeS65Hjx5FTvwm27JZhUmuAGRyQDByOI6IiKiMrBqX2b9/P/bt22dxfN++fThw4EC5O0UFpLvkbHEyU2WJd8gRERGViVWBafTo0bhy5YrF8ZSUFIwePbrcnaICNrtLDiioLHFIjoiIqEysCkwnTpxAs2bNLI43bdoUJ06cKHenqIC00rctTmaqLLHCREREVCZWBSZXV1dcu3bN4nhqaiqUSqunRVERBJst9Y1CQ3KsMBEREZWFVYHpySefxKRJk5CZmSkdu337Nt577z08+eSTNuscFVSY5LYYkmNgIiIisopV5aA5c+bgiSeeQHBwMJo2bQoAOHz4MPz8/LBq1SqbdvBRZ6ow2SIvSYtXckiOiIioTKwKTDVq1MDRo0exZs0aHDlyBBqNBi+++CIGDRoElYq/jG3JliNyrDARERFZx+oJR+7u7mjTpg2CgoKQl5cHAPjjjz8AiAtbkm1Ik75tMiTHu+SIiIisYVVgunDhAvr27Ytjx45BJpNBEASzX+gGg8FmHXzUVUyFiVVAIiKisrBq0ve4ceMQGhqKa9euwc3NDUlJSfjrr7/QvHlz7Ny508ZdfLRJC1fyLjkiIiK7sarCtGfPHuzYsQNVq1aFXC6HQqFAmzZtMGvWLLz++us4dOiQrfv5yDJVmGxSY5Jz0jcREZE1rKowGQwGeHh4AAB8fX3x77//AgCCg4Nx+vRp2/WObLeXHMAhOSIiIitZVWFq2LAhjh49irCwMERFReHTTz+Fi4sLFi9ejLCwMFv38ZFWMXvJcUiOiIioLKwKTFOmTMG9e/cAAB999BF69uyJtm3bwsfHB+vWrbNpBx91pgqTbRau5NYoRERE1rAqMHXt2lX6PiwsDCdOnMDNmzdRuXJl29z+TpKKGZJjhYmIiKgsbLbxW5UqVWx1KirEtkNyXIeJiIjIGlZN+qaHp6DCxCE5IiIie2FgcnBCyU1Kz1RZkjMwERERlQUDk4Oz6ea7dWOAKrWA2l1scDIiIqJHh83mMFHFKNhLzgYne6wz8PpBG5yIiIjo0cIKk4Mr2EuOdx8SERHZCwOTw7PhkBwRERFZhYHJwdl04UoiIiKyCgOTgzNKQ3JERERkLwxMDk4QmJiIiIjsjYHJwUl3ydm1F0RERI82BiYHZ9OVvomIiMgqDEwOzqZ7yREREZFVGJgcnVRhsm83iIiIHmUMTA6uYA4TExMREZG9MDA5OIEVJiIiIrtjYHJw0hwmJiYiIiK7YWBycFy4koiIyP4YmBycaeFKFpiIiIjsh4HJwUmTvhmYiIiI7IaBydFJQ3JMTERERPbCwOTgCiZ927kjREREjzAGJgfHvXeJiIjsj4HJwQmcxERERGR3dg9MCxYsQGhoKNRqNSIjI7Fr164Htp8/fz7Cw8Oh0WhQt25drFy50qLNvHnzULduXWg0GgQGBuKNN95Abm5uud7XXgpW+iYiIiJ7sWtgWrduHcaPH4/Jkyfj0KFDaNu2LWJiYpCcnFxk+4ULF2LSpEmYNm0ajh8/junTp2P06NH45ZdfpDZr1qzBxIkT8cEHH+DkyZNYunQp1q1bh0mTJln9vvZkWlZAzsRERERkNzJBkAZ9HrqoqCg0a9YMCxculI6Fh4ejT58+mDVrlkX7Vq1aoXXr1vjss8+kY+PHj8eBAwewe/duAMCYMWNw8uRJbN++XWrz5ptvIiEhQaoilfV9i5KVlQVvb29kZmbCy8urbB+8DDYnpWHE6kREBlfGhpGtKux9iIiIHgXW/v62W4UpLy8PiYmJ6NKli9nxLl26ID4+vsjXaLVaqNVqs2MajQYJCQnQ6XQAgDZt2iAxMREJCQkAgAsXLuD3339Hjx49rH5f+8q/S87OvSAiInqUKe31xhkZGTAYDPDz8zM77ufnh7S0tCJf07VrV3zzzTfo06cPmjVrhsTERMTGxkKn0yEjIwMBAQEYOHAgrl+/jjZt2kAQBOj1eowcORITJ060+n0BMaxptVrpcVZWlrUfvUy4+S4REZH92X3S9/2bygqCUOxGs1OnTkVMTAxatmwJlUqF3r17Y9iwYQAAhUIBANi5cydmzpyJBQsW4ODBg9i4cSN+/fVXfPjhh1a/LwDMmjUL3t7e0ldgYGBZP6pVCiZ9MzERERHZi90Ck6+vLxQKhUVVJz093aL6Y6LRaBAbG4vs7GxcunQJycnJCAkJgaenJ3x9fQGIoWrIkCF4+eWXERERgb59++Ljjz/GrFmzYDQarXpfAJg0aRIyMzOlrytXrpTzCpSOwNvkiIiI7M5ugcnFxQWRkZGIi4szOx4XF4dWrR48uVmlUqFmzZpQKBRYu3YtevbsCblc/CjZ2dnS9yYKhQKCIEAQBKvf19XVFV5eXmZfD4PAOUxERER2Z7c5TAAwYcIEDBkyBM2bN0d0dDQWL16M5ORkjBgxAoBY1UlJSZHWWjpz5gwSEhIQFRWFW7duYe7cuUhKSsKKFSukc/bq1Qtz585F06ZNERUVhXPnzmHq1Kl46qmnpGG7kt7XkXAOExERkf3ZNTANGDAAN27cwIwZM5CamoqGDRvi999/R3BwMAAgNTXVbG0kg8GAOXPm4PTp01CpVOjQoQPi4+MREhIitZkyZQpkMhmmTJmClJQUVK1aFb169cLMmTNL/b6OhHOYiIiI7M+u6zA5s4e1DtNPh1Mwbu1htH7MB2tebllh70NERPQocLp1mKh0CjbfZYWJiIjIXhiYHJw06Zt5iYiIyG4YmBwcB0yJiIjsj4HJwRXcJccSExERkb0wMDk4rltJRERkfwxMDs50EyMLTERERPbDwOTgWGEiIiKyPwYmR8c5TERERHbHwOTgTMsKyJmXiIiI7IaBycEZpWUFmJiIiIjshYHJwXHzXSIiIvtjYHJw0krfdu4HERHRo4yBycGxwkRERGR/DEwOrmBZASYmIiIie2FgcnRcuJKIiMjuGJgcnFRhYmAiIiKyGwYmByfNYeKQHBERkd0wMDk47iVHRERkfwxMDs7IrVGIiIjsjoHJwXHzXSIiIvtjYHJwHJIjIiKyPwYmJ8G8REREZD8MTA5O4BwmIiIiu2NgcnDcS46IiMj+GJgcnMBZ30RERHbHwOTguJccERGR/TEwOThThUnOvERERGQ3DEwOzshlBYiIiOyOgclJcEiOiIjIfhiYHBwXriQiIrI/BiYHV7AOk337QURE9ChjYHJwgvQdExMREZG9MDA5OFaYiIiI7I+BycFxpW8iIiL7Y2BycKwwERER2R8Dk4MzzWGSMzERERHZDQOTg5OWFbBzP4iIiB5lDEwOrmBIjpGJiIjIXhiYHJxQaGEBIiIisg8GJgfHSd9ERET2x8Dk4Ez1Je4lR0REZD8MTA6OFSYiIiL7Y2BycFy4koiIyP4YmBwdK0xERER2x8Dk4LhwJRERkf0xMDk4o9FUYrJvP4iIiB5lDEwOjnfJERER2R8Dk4PjXXJERET2x8Dk4HiXHBERkf0xMDk4VpiIiIjsj4HJSXAOExERkf3YPTAtWLAAoaGhUKvViIyMxK5dux7Yfv78+QgPD4dGo0HdunWxcuVKs+fbt28PmUxm8dWjRw+pzbRp0yye9/f3r5DPV15CfomJFSYiIiL7UdrzzdetW4fx48djwYIFaN26Nb7++mvExMTgxIkTCAoKsmi/cOFCTJo0CUuWLEGLFi2QkJCAV155BZUrV0avXr0AABs3bkReXp70mhs3bqBx48Z45plnzM7VoEEDbNu2TXqsUCgq6FOWT8FdckRERGQvdg1Mc+fOxfDhw/Hyyy8DAObNm4ctW7Zg4cKFmDVrlkX7VatW4bXXXsOAAQMAAGFhYdi7dy8++eQTKTBVqVLF7DVr166Fm5ubRWBSKpUOW1UqrGAOEyMTERGRvdhtSC4vLw+JiYno0qWL2fEuXbogPj6+yNdotVqo1WqzYxqNBgkJCdDpdEW+ZunSpRg4cCDc3d3Njp89exbVq1dHaGgoBg4ciAsXLjywv1qtFllZWWZfD4ORQ3JERER2Z7fAlJGRAYPBAD8/P7Pjfn5+SEtLK/I1Xbt2xTfffIPExEQIgoADBw4gNjYWOp0OGRkZFu0TEhKQlJQkVbBMoqKisHLlSmzZsgVLlixBWloaWrVqhRs3bhTb31mzZsHb21v6CgwMtOJTlx0XriQiIrI/u0/6vn+oSRCEYoefpk6dipiYGLRs2RIqlQq9e/fGsGHDABQ9B2np0qVo2LAhHn/8cbPjMTEx6NevHyIiItC5c2f89ttvAIAVK1YU289JkyYhMzNT+rpy5UpZPqbVuKwAERGR/dktMPn6+kKhUFhUk9LT0y2qTiYajQaxsbHIzs7GpUuXkJycjJCQEHh6esLX19esbXZ2NtauXWtRXSqKu7s7IiIicPbs2WLbuLq6wsvLy+zr4eDClURERPZmt8Dk4uKCyMhIxMXFmR2Pi4tDq1atHvhalUqFmjVrQqFQYO3atejZsyfkcvOP8v3330Or1WLw4MEl9kWr1eLkyZMICAgo+wepYKwwERER2Z9d75KbMGEChgwZgubNmyM6OhqLFy9GcnIyRowYAUAcBktJSZHWWjpz5gwSEhIQFRWFW7duYe7cuUhKSipyKG3p0qXo06cPfHx8LJ5766230KtXLwQFBSE9PR0fffQRsrKyMHTo0Ir9wFbgXXJERET2Z9fANGDAANy4cQMzZsxAamoqGjZsiN9//x3BwcEAgNTUVCQnJ0vtDQYD5syZg9OnT0OlUqFDhw6Ij49HSEiI2XnPnDmD3bt3Y+vWrUW+79WrVzFo0CBkZGSgatWqaNmyJfbu3Su9ryMRpGnfREREZC8ywbSUNJVJVlYWvL29kZmZWaHzmd5efwTrE6/inW51Mar9YxX2PkRERI8Ca39/2/0uOXowU5qVc0iOiIjIbuw6JEclkxautHM/iIhInBpS3ELJ5Bzy8vIQFBQEo9FYptcxMDk63iVHRGR3giAgLS0Nt2/ftndXqJyMRiMWLVqElJQU5OTkwN/fv1Q3VjEwOTiu9E1EZH+msFStWjW4ubnxzmUnZjAYkJOTg2rVqkm7hJRmWSEGJgcncC85IiK7MhgMUlgqaqkaci4GgwEAUKVKFcjlcqSnp6NatWpF7hhSGCd9OzjewkhEZF+mOUtubm527gnZmulnWpp5aQxMDo4LVxIROQb+PfzfU5afKQOTgyuYw0RERGR/7du3x/jx4+3djYeOc5gcHOcwERGRNUqqngwdOhTLly8v83k3btwIlUplZa+cFwOTg+PClUREZI3U1FTp+3Xr1uH999/H6dOnpWMajcasvU6nK1UQqlKliu066UQ4JOfgWGEiIiJr+Pv7S1/e3t6QyWTS49zcXFSqVAnff/892rdvD7VajdWrV+PGjRsYNGgQatasCTc3N0REROC7774zO+/9Q3IhISH4+OOP8dJLL8HT0xNBQUFYvHjxQ/60FY+BycFJk77t2w0iIipEEARk5+nt8mXLLWDfffddvP766zh58iS6du2K3NxcREZG4tdff0VSUhJeffVVDBkyBPv27XvgeebMmYPmzZvj0KFDGDVqFEaOHIlTp07ZrJ+OgENyDk76c8ESExGRw8jRGVD//S12ee8TM7rCzcU2v77Hjx+Pp59+2uzYW2+9JX0/duxYbN68GevXr0dUVFSx5+nevTtGjRoFQAxhn3/+OXbu3Il69erZpJ+OgIHJwQngXnJERFQxmjdvbvbYYDBg9uzZWLduHVJSUqDVaqHVauHu7v7A8zRq1Ej63jT0l56eXiF9thcGJgcncC85IiKHo1EpcGJGV7u9t63cH4TmzJmDzz//HPPmzUNERATc3d0xfvx45OXlPfA8908Wl8lkZd7c1tExMDk47iVHROR4ZDKZzYbFHMmuXbvQu3dvDB48GIC4Ue3Zs2cRHh5u557ZHyd9OzhWmIiI6GF57LHHEBcXh/j4eJw8eRKvvfYa0tLS7N0th8DA5PA4h4mIiB6OqVOnolmzZujatSvat28Pf39/9OnTx97dcgj/vXrif4ypwsSFK4mIyFrDhg3DsGHDpMchISFFLk9QpUoV/Pjjjw88186dO80eX7p0yaLN4cOHy95JB8cKk4MzciEmIiIiu2NgcnDcfJeIiMj+GJgcXMGkb0YmIiIie2FgcnCsMBEREdkfA5OD4+a7RERE9sfA5CQYmIiIiOyHgcnBFdwkx8RERERkLwxMDk7afJd5iYiIyG4YmBwc75IjIiKyPwYmB2dauJJxiYiIHrb27dtj/Pjx0uOQkBDMmzfvga+RyWQlrhZeGrY6j60wMDk4br5LRETW6NWrFzp37lzkc3v27IFMJsPBgwfLdM79+/fj1VdftUX3JNOmTUOTJk0sjqempiImJsam71UeDEwOrmAdJiYmIiIqveHDh2PHjh24fPmyxXOxsbFo0qQJmjVrVqZzVq1aFW5ubrbq4gP5+/vD1dX1obxXaTAwOTpWmIiIyAo9e/ZEtWrVsHz5crPj2dnZWLduHfr06YNBgwahZs2acHNzQ0REBL777rsHnvP+IbmzZ8/iiSeegFqtRv369REXF2fxmnfffRd16tSBm5sbwsLCMHXqVOh0OgDA8uXLMX36dBw5cgQymQwymUzq7/1DcseOHUPHjh2h0Wjg4+ODV199FXfv3pWeHzZsGPr06YP//e9/CAgIgI+PD0aPHi29V3kpbXIWqjDSXXJ27gcRERUiCIAu2z7vrXIr1b+ilUolXnjhBSxfvhzvv/++dPPQ+vXrkZeXh5dffhnfffcd3n33XXh5eeG3337DkCFDEBYWhqioqBLPbzQa8fTTT8PX1xd79+5FVlaW2XwnE09PTyxfvhzVq1fHsWPH8Morr8DT0xPvvPMOBgwYgKSkJGzevBnbtm0DAHh7e1ucIzs7G926dUPLli2xf/9+pKen4+WXX8aYMWPMAuGff/6JgIAA/Pnnnzh37hwGDBiAJk2a4JVXXinx85SEgcnBcQ4TEZED0mUDH1e3z3u/9y/g4l6qpi+99BI+++wz7Ny5Ex06dAAgDsc9/fTTqFGjBt566y2p7dixY7F582asX7++VIFp27ZtOHnyJC5duoSaNWsCAD7++GOLeUdTpkyRvg8JCcGbb76JdevW4Z133oFGo4GHhweUSiX8/f2Lfa81a9YgJycHK1euhLu7+Nm/+uor9OrVC5988gn8/PwAAJUrV8ZXX30FhUKBevXqoUePHti+fTsD06NAkL5jYiIiorKpV68eWrVqhdjYWHTo0AHnz5/Hrl27sHXrVhgMBsyePRvr1q1DSkoKtFottFqtFEhKcvLkSQQFBUlhCQCio6Mt2v3www+YN28ezp07h7t370Kv18PLy6tMn+PkyZNo3LixWd9at24No9GI06dPS4GpQYMGUCgUUpuAgAAcO3asTO9VHAYmB8e95IiIHJDKTaz02Ou9y2D48OEYM2YM5s+fj2XLliE4OBidOnXCZ599hs8//xzz5s1DREQE3N3dMX78eOTl5ZXqvKbfT4Xdv2bg3r17MXDgQEyfPh1du3aFt7c31q5dizlz5pTpMwiCUOx6hIWPq1Qqi+eMRmOZ3qs4DEwOzvS/o5yJiYjIcchkpR4Ws7dnn30W48aNw7fffosVK1bglVdegUwmw65du9C7d28MHjwYgDgn6ezZswgPDy/VeevXr4/k5GT8+++/qF5dHJ7cs2ePWZt//vkHwcHBmDx5snTs/rv2XFxcYDAYSnyvFStW4N69e1KV6Z9//oFcLkedOnVK1d/y4l1yDs4o7SVHRERUdh4eHhgwYADee+89/Pvvvxg2bBgA4LHHHkNcXBzi4+Nx8uRJvPbaa0hLSyv1eTt37oy6devihRdewJEjR7Br1y6zYGR6j+TkZKxduxbnz5/Hl19+iU2bNpm1CQkJwcWLF3H48GFkZGRAq9VavNfzzz8PtVqNoUOHIikpCX/++SfGjh2LIUOGSMNxFY2BydFxSI6IiMpp+PDhuHXrFjp37oygoCAAwNSpU9GsWTN07doV7du3h7+/P/r06VPqc8rlcmzatAlarRaPP/44Xn75ZcycOdOsTe/evfHGG29gzJgxaNKkCeLj4zF16lSzNv369UO3bt3QoUMHVK1atcilDdzc3LBlyxbcvHkTLVq0QP/+/dGpUyd89dVXZb8YVpIJRQ1CUomysrLg7e2NzMzMMk9eK4unvtqNo1czETusOTrWezgpmoiICuTm5uLixYsIDQ2FWq22d3eonAwGAw4dOoSmTZtCp9OV+mfLCpODk5YV4KAcERGR3TAwOTgBnMRERERkbwxMDk5gXiIiIrI7BiYHV7DSNyMTERGRvTAwOTjTjHzGJSIiIvthYHJwppsYuXAlEZF98aby/56y/EwZmBwcN98lIrIv03Yb2dnZdu4J2ZrpZ3r/lipF4dYoDs50lxzzEhGRfSgUClSqVAnp6ekAxEUUOa/UeZm2Ybl58yYyMjJQqVIlsw17i8PA5OAETmIiIrI7f39/AJBCEzkvo9GI69evQ6PRoEqVKtLPtiQMTA6uIC8xMRER2YtMJkNAQACqVasGnU5n7+5QOdy9exfdu3fHkSNHUKlSpVK/zu6BacGCBfjss8+QmpqKBg0aYN68eWjbtm2x7efPn4+vvvoKly5dQlBQECZPnowXXnhBer59+/b466+/LF7XvXt3/Pbbb1a/r70I3EuOiMhhKBSKUg3fkOPKy8tDcnIy5PKyTeO266TvdevWYfz48Zg8eTIOHTqEtm3bIiYmBsnJyUW2X7hwISZNmoRp06bh+PHjmD59OkaPHo1ffvlFarNx40akpqZKX0lJSVAoFHjmmWesfl974ogcERGR/dl1892oqCg0a9YMCxculI6Fh4ejT58+mDVrlkX7Vq1aoXXr1vjss8+kY+PHj8eBAwewe/fuIt9j3rx5eP/995Gamgp3d3er3rcoD2vz3Y7/24kLGffw/WvReDy0SoW9DxER0aPA2t/fdqsw5eXlITExEV26dDE73qVLF8THxxf5Gq1Wa7GbsEajQUJCQrFjykuXLsXAgQOlsGTN+9qTVGFiiYmIiMhu7DaHKSMjAwaDAX5+fmbH/fz8kJaWVuRrunbtim+++QZ9+vRBs2bNkJiYiNjYWOh0OmRkZCAgIMCsfUJCApKSkrB06dJyvS8ghjWtVis9zszMBCAm1Yqky7kLozYH9+5kISvL7lPOiIiInJrp93ZZB9js/hv4/rUsBEEodn2LqVOnIi0tDS1btoQgCPDz88OwYcPw6aefFjkJb+nSpWjYsCEef/zxcr0vAMyaNQvTp0+3OB4YGFjsa2ypw7yH8jZERESPhDt37sDb27vU7e0WmHx9faFQKCyqOunp6RbVHxONRoPY2Fh8/fXXuHbtGgICArB48WJ4enrC19fXrG12djbWrl2LGTNmlPt9AWDSpEmYMGGC9NhoNOLmzZvw8fGx+QJmWVlZCAwMxJUrVyp0fhTxWj9MvNYPD6/1w8Nr/fDY6loLgoA7d+6gevXqZXqd3QKTi4sLIiMjERcXh759+0rH4+Li0Lt37we+VqVSoWbNmgCAtWvXomfPnha3B37//ffQarUYPHiwTd7X1dUVrq6uZsfKsn6DNby8vPgH8CHhtX54eK0fHl7rh4fX+uGxxbUuS2XJxK5DchMmTMCQIUPQvHlzREdHY/HixUhOTsaIESMAiFWdlJQUrFy5EgBw5swZJCQkICoqCrdu3cLcuXORlJSEFStWWJx76dKl6NOnD3x8fMr8vkRERESF2TUwDRgwADdu3MCMGTOQmpqKhg0b4vfff0dwcDAAIDU11WxtJIPBgDlz5uD06dNQqVTo0KED4uPjERISYnbeM2fOYPfu3di6datV70tERERUmN0nfY8aNQqjRo0q8rnly5ebPQ4PD8ehQ4dKPGedOnVKnP3+oPe1N1dXV3zwwQcWQ4Bke7zWDw+v9cPDa/3w8Fo/PPa+1nZduJKIiIjIGdh1axQiIiIiZ8DARERERFQCBiYiIiKiEjAwEREREZWAgcnBLFiwAKGhoVCr1YiMjMSuXbvs3SWnN2vWLLRo0QKenp6oVq0a+vTpg9OnT5u1EQQB06ZNQ/Xq1aHRaNC+fXscP37cTj3+75g1axZkMhnGjx8vHeO1tp2UlBQMHjwYPj4+cHNzQ5MmTZCYmCg9z2ttG3q9HlOmTEFoaCg0Gg3CwsIwY8YMGI1GqQ2vtXX+/vtv9OrVC9WrV4dMJsOPP/5o9nxprqtWq8XYsWPh6+sLd3d3PPXUU7h69artOyuQw1i7dq2gUqmEJUuWCCdOnBDGjRsnuLu7C5cvX7Z315xa165dhWXLlglJSUnC4cOHhR49eghBQUHC3bt3pTazZ88WPD09hQ0bNgjHjh0TBgwYIAQEBAhZWVl27LlzS0hIEEJCQoRGjRoJ48aNk47zWtvGzZs3heDgYGHYsGHCvn37hIsXLwrbtm0Tzp07J7XhtbaNjz76SPDx8RF+/fVX4eLFi8L69esFDw8PYd68eVIbXmvr/P7778LkyZOFDRs2CACETZs2mT1fmus6YsQIoUaNGkJcXJxw8OBBoUOHDkLjxo0FvV5v074yMDmQxx9/XBgxYoTZsXr16gkTJ060U4/+m9LT0wUAwl9//SUIgiAYjUbB399fmD17ttQmNzdX8Pb2FhYtWmSvbjq1O3fuCLVr1xbi4uKEdu3aSYGJ19p23n33XaFNmzbFPs9rbTs9evQQXnrpJbNjTz/9tDB48GBBEHitbeX+wFSa63r79m1BpVIJa9euldqkpKQIcrlc2Lx5s037xyE5B5GXl4fExER06dLF7HiXLl0QHx9vp179N2VmZgIAqlSpAgC4ePEi0tLSzK69q6sr2rVrx2tvpdGjR6NHjx7o3Lmz2XFea9v5+eef0bx5czzzzDOoVq0amjZtiiVLlkjP81rbTps2bbB9+3acOXMGAHDkyBHs3r0b3bt3B8BrXVFKc10TExOh0+nM2lSvXh0NGza0+bW3+0rfJMrIyIDBYICfn5/ZcT8/P6SlpdmpV/89giBgwoQJaNOmDRo2bAgA0vUt6tpfvnz5offR2a1duxYHDx7E/v37LZ7jtbadCxcuYOHChZgwYQLee+89JCQk4PXXX4erqyteeOEFXmsbevfdd5GZmYl69epBoVDAYDBg5syZGDRoEAD+f11RSnNd09LS4OLigsqVK1u0sfXvTgYmByOTycweC4JgcYysN2bMGBw9ehS7d++2eI7XvvyuXLmCcePGYevWrVCr1cW247UuP6PRiObNm+Pjjz8GADRt2hTHjx/HwoUL8cILL0jteK3Lb926dVi9ejW+/fZbNGjQAIcPH8b48eNRvXp1DB06VGrHa10xrLmuFXHtOSTnIHx9faFQKCwScXp6ukW6JuuMHTsWP//8M/7880/UrFlTOu7v7w8AvPY2kJiYiPT0dERGRkKpVEKpVOKvv/7Cl19+CaVSKV1PXuvyCwgIQP369c2OhYeHSxuW8/9r23n77bcxceJEDBw4EBERERgyZAjeeOMNzJo1CwCvdUUpzXX19/dHXl4ebt26VWwbW2FgchAuLi6IjIxEXFyc2fG4uDi0atXKTr36bxAEAWPGjMHGjRuxY8cOhIaGmj0fGhoKf39/s2ufl5eHv/76i9e+jDp16oRjx47h8OHD0lfz5s3x/PPP4/DhwwgLC+O1tpHWrVtbLI9x5swZBAcHA+D/17aUnZ0Nudz816VCoZCWFeC1rhilua6RkZFQqVRmbVJTU5GUlGT7a2/TKeRULqZlBZYuXSqcOHFCGD9+vODu7i5cunTJ3l1zaiNHjhS8vb2FnTt3CqmpqdJXdna21Gb27NmCt7e3sHHjRuHYsWPCoEGDeEuwjRS+S04QeK1tJSEhQVAqlcLMmTOFs2fPCmvWrBHc3NyE1atXS214rW1j6NChQo0aNaRlBTZu3Cj4+voK77zzjtSG19o6d+7cEQ4dOiQcOnRIACDMnTtXOHTokLScTmmu64gRI4SaNWsK27ZtEw4ePCh07NiRywo8CubPny8EBwcLLi4uQrNmzaRb38l6AIr8WrZsmdTGaDQKH3zwgeDv7y+4uroKTzzxhHDs2DH7dfo/5P7AxGttO7/88ovQsGFDwdXVVahXr56wePFis+d5rW0jKytLGDdunBAUFCSo1WohLCxMmDx5sqDVaqU2vNbW+fPPP4v8+3no0KGCIJTuuubk5AhjxowRqlSpImg0GqFnz55CcnKyzfsqEwRBsG3NioiIiOi/hXOYiIiIiErAwERERERUAgYmIiIiohIwMBERERGVgIGJiIiIqAQMTEREREQlYGAiIiIiKgEDExGRjchkMvz444/27gYRVQAGJiL6Txg2bBhkMpnFV7du3ezdNSL6D1DauwNERLbSrVs3LFu2zOyYq6urnXpDRP8lrDAR0X+Gq6sr/P39zb4qV64MQBwuW7hwIWJiYqDRaBAaGor169ebvf7YsWPo2LEjNBoNfHx88Oqrr+Lu3btmbWJjY9GgQQO4uroiICAAY8aMMXs+IyMDffv2hZubG2rXro2ff/65Yj80ET0UDExE9MiYOnUq+vXrhyNHjmDw4MEYNGgQTp48CQDIzs5Gt27dULlyZezfvx/r16/Htm3bzALRwoULMXr0aLz66qs4duwYfv75Zzz22GNm7zF9+nQ8++yzOHr0KLp3747nn38eN2/efKifk4gqgM238yUisoOhQ4cKCoVCcHd3N/uaMWOGIAiCAEAYMWKE2WuioqKEkSNHCoIgCIsXLxYqV64s3L17V3r+t99+E+RyuZCWliYIgiBUr15dmDx5crF9ACBMmTJFenz37l1BJpMJf/zxh80+JxHZB+cwEdF/RocOHbBw4UKzY1WqVJG+j46ONnsuOjoahw8fBgCcPHkSjRs3hru7u/R869atYTQacfr0achkMvz777/o1KnTA/vQqFEj6Xt3d3d4enoiPT3d2o9ERA6CgYmI/jPc3d0thshKIpPJAACCIEjfF9VGo9GU6nwqlcritUajsUx9IiLHwzlMRPTI2Lt3r8XjevXqAQDq16+Pw4cP4969e9Lz//zzD+RyOerUqQNPT0+EhIRg+/btD7XPROQYWGEiov8MrVaLtLQ0s2NKpRK+vr4AgPXr16N58+Zo06YN1qxZg4SEBCxduhQA8Pzzz+ODDz7A0KFDMW3aNFy/fh1jx47FkCFD4OfnBwCYNm0aRowYgWrVqiEmJgZ37tzBP//8g7Fjxz7cD0pEDx0DExH9Z2zevBkBAQFmx+rWrYtTp04BEO9gW7t2LUaNGgV/f3+sWbMG9evXBwC4ublhy5YtGDduHFq0aAE3Nzf069cPc+fOlc41dOhQ5Obm4vPPP8dbb70FX19f9O/f/+F9QCKyG5kgCIK9O0FEVNFkMhk2bdqEPn362LsrROSEOIeJiIiIqAQMTEREREQl4BwmInokcPYBEZUHK0xEREREJWBgIiIiIioBAxMRERFRCRiYiIiIiErAwERERERUAgYmIiIiohIwMBERERGVgIGJiIiIqAQMTEREREQl+H8qVXsL9iK8ZgAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_metrics(rnn_history, 'accuracy')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 209,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_metrics(rnn_history, 'precision_13')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 210,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_metrics(rnn_history, 'recall_13')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 211,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'accuracy_score': 0.9998805256869773, 'precision_score': 0.9999091239549255, 'recall_score': 0.9979166666666667, 'f1_score': 0.9989085286129034}\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       1.00      1.00      1.00      2750\n",
      "           1       1.00      1.00      1.00      2750\n",
      "           2       1.00      0.99      1.00       120\n",
      "           3       1.00      1.00      1.00      2750\n",
      "\n",
      "    accuracy                           1.00      8370\n",
      "   macro avg       1.00      1.00      1.00      8370\n",
      "weighted avg       1.00      1.00      1.00      8370\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1000x1000 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "print(evaluate_model(encoded_y_val.argmax(axis=1), rnn_y_pred.argmax(axis=1)))\n",
    "print(classification_report(encoded_y_val.argmax(axis=1), rnn_y_pred.argmax(axis=1)))\n",
    "plot_confusion_matrix(encoded_y_val.argmax(axis=1), rnn_y_pred.argmax(axis=1), label)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "colab": {
   "collapsed_sections": [
    "7rv4WPs_ykfy",
    "ZtoUYKw6bdx7",
    "xhRk-QQpl5E7",
    "BIHPYxSGS4fI",
    "miBrWQz4pi-q",
    "vVOY90PsADf6",
    "hXeWC7ZRAqzo",
    "3l2gTVzN6NQM",
    "IDdyg9rG1Pmd",
    "_hjEAktDav8L"
   ],
   "provenance": []
  },
  "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": "3.9.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}