Test12-Copy10 (1).ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "1a5ae323",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2023-09-03 02:00:37.026483: I tensorflow/core/util/port.cc:110] oneDNN custom operations are on. You may see slightly different numerical results due to floating-point round-off errors from different computation orders. To turn them off, set the environment variable `TF_ENABLE_ONEDNN_OPTS=0`.\n",
      "2023-09-03 02:00:37.027481: I tensorflow/tsl/cuda/cudart_stub.cc:28] Could not find cuda drivers on your machine, GPU will not be used.\n",
      "2023-09-03 02:00:37.051242: I tensorflow/tsl/cuda/cudart_stub.cc:28] Could not find cuda drivers on your machine, GPU will not be used.\n",
      "2023-09-03 02:00:37.051815: I tensorflow/core/platform/cpu_feature_guard.cc:182] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.\n",
      "To enable the following instructions: AVX2 AVX_VNNI FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.\n",
      "2023-09-03 02:00:37.523137: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Could not find TensorRT\n",
      "[nltk_data] Downloading package stopwords to /home/phyu/nltk_data...\n",
      "[nltk_data]   Package stopwords is already up-to-date!\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 1,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#This one is the clean version LSTM for Text Fake New Detection\n",
    "\n",
    "#!pip install tensorflow\n",
    "\n",
    "import tensorflow as tf\n",
    "\n",
    "from tensorflow.keras.preprocessing.text import Tokenizer\n",
    "import pandas as pd \n",
    "import numpy as np\n",
    "from tensorflow.keras.preprocessing.sequence import pad_sequences\n",
    "from nltk.corpus import stopwords\n",
    "from nltk.stem.porter import PorterStemmer\n",
    "import nltk\n",
    "nltk.download('stopwords')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "007e8e25",
   "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>Unnamed: 0</th>\n",
       "      <th>title</th>\n",
       "      <th>text</th>\n",
       "      <th>label</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>LAW ENFORCEMENT ON HIGH ALERT Following Threat...</td>\n",
       "      <td>No comment is expected from Barack Obama Membe...</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Did they post their votes for Hillary already?</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>UNBELIEVABLE! OBAMA’S ATTORNEY GENERAL SAYS MO...</td>\n",
       "      <td>Now, most of the demonstrators gathered last ...</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>Bobby Jindal, raised Hindu, uses story of Chri...</td>\n",
       "      <td>A dozen politically active pastors came here f...</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>SATAN 2: Russia unvelis an image of its terrif...</td>\n",
       "      <td>The RS-28 Sarmat missile, dubbed Satan 2, will...</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Unnamed: 0                                              title  \\\n",
       "0           0  LAW ENFORCEMENT ON HIGH ALERT Following Threat...   \n",
       "1           1                                                NaN   \n",
       "2           2  UNBELIEVABLE! OBAMA’S ATTORNEY GENERAL SAYS MO...   \n",
       "3           3  Bobby Jindal, raised Hindu, uses story of Chri...   \n",
       "4           4  SATAN 2: Russia unvelis an image of its terrif...   \n",
       "\n",
       "                                                text  label  \n",
       "0  No comment is expected from Barack Obama Membe...      1  \n",
       "1     Did they post their votes for Hillary already?      1  \n",
       "2   Now, most of the demonstrators gathered last ...      1  \n",
       "3  A dozen politically active pastors came here f...      0  \n",
       "4  The RS-28 Sarmat missile, dubbed Satan 2, will...      1  "
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = pd.read_csv(r\"/home/phyu/Eimon/Pioneer/Data/archive_3/WELFake_Dataset.csv\")\n",
    "df.head()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "723014c6",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(72134, 4)"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.shape(df)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "b0b760a7",
   "metadata": {},
   "outputs": [],
   "source": [
    "x = list(range(10000,72133))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "90cba368",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(10001, 4)"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df1 = df.drop(df.index[x])\n",
    "np.shape(df1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "417ba326",
   "metadata": {},
   "outputs": [],
   "source": [
    "x2 = list(range(0,64133))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "3bc054cc",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(8001, 4)"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df2 = df.drop(df.index[x2])\n",
    "np.shape(df2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "d2bf2507",
   "metadata": {},
   "outputs": [],
   "source": [
    "x3 = list(range(0,42133)) + list(range(50133,72133))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "4beb171c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(8001, 4)"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df3 = df.drop(df.index[x3])\n",
    "np.shape(df3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "bd38da26",
   "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>Unnamed: 0</th>\n",
       "      <th>title</th>\n",
       "      <th>text</th>\n",
       "      <th>label</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>LAW ENFORCEMENT ON HIGH ALERT Following Threat...</td>\n",
       "      <td>No comment is expected from Barack Obama Membe...</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>UNBELIEVABLE! OBAMA’S ATTORNEY GENERAL SAYS MO...</td>\n",
       "      <td>Now, most of the demonstrators gathered last ...</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>Bobby Jindal, raised Hindu, uses story of Chri...</td>\n",
       "      <td>A dozen politically active pastors came here f...</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>SATAN 2: Russia unvelis an image of its terrif...</td>\n",
       "      <td>The RS-28 Sarmat missile, dubbed Satan 2, will...</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>5</td>\n",
       "      <td>About Time! Christian Group Sues Amazon and SP...</td>\n",
       "      <td>All we can say on this one is it s about time ...</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Unnamed: 0                                              title  \\\n",
       "0           0  LAW ENFORCEMENT ON HIGH ALERT Following Threat...   \n",
       "2           2  UNBELIEVABLE! OBAMA’S ATTORNEY GENERAL SAYS MO...   \n",
       "3           3  Bobby Jindal, raised Hindu, uses story of Chri...   \n",
       "4           4  SATAN 2: Russia unvelis an image of its terrif...   \n",
       "5           5  About Time! Christian Group Sues Amazon and SP...   \n",
       "\n",
       "                                                text  label  \n",
       "0  No comment is expected from Barack Obama Membe...      1  \n",
       "2   Now, most of the demonstrators gathered last ...      1  \n",
       "3  A dozen politically active pastors came here f...      0  \n",
       "4  The RS-28 Sarmat missile, dubbed Satan 2, will...      1  \n",
       "5  All we can say on this one is it s about time ...      1  "
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df1.dropna(inplace=True)\n",
    "df1.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "82f5683e",
   "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>Unnamed: 0</th>\n",
       "      <th>title</th>\n",
       "      <th>text</th>\n",
       "      <th>label</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>64133</th>\n",
       "      <td>64133</td>\n",
       "      <td>Over 20 Injured, Almost 40 Detained in Venezue...</td>\n",
       "      <td>— Alfredo Romero (@alfredoromero) October 26, ...</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>64134</th>\n",
       "      <td>64134</td>\n",
       "      <td>From Monster to Mr. President-Elect: Democrats...</td>\n",
       "      <td>From Monster to Mr. President-Elect: Democrat...</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>64135</th>\n",
       "      <td>64135</td>\n",
       "      <td>Boy wearing a ‘My dad is an ATM’ T-shirt chase...</td>\n",
       "      <td>Boy wearing a ‘My dad is an ATM’ T-shirt chase...</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>64136</th>\n",
       "      <td>64136</td>\n",
       "      <td>How to Survive Being an Airbnb Host - The New ...</td>\n",
       "      <td>After my first Airbnb guests checked out of ou...</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>64137</th>\n",
       "      <td>64137</td>\n",
       "      <td>Fighting in Ukraine May Complicate U.S. Thaw W...</td>\n",
       "      <td>MOSCOW  —   The war in eastern Ukraine that si...</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       Unnamed: 0                                              title  \\\n",
       "64133       64133  Over 20 Injured, Almost 40 Detained in Venezue...   \n",
       "64134       64134  From Monster to Mr. President-Elect: Democrats...   \n",
       "64135       64135  Boy wearing a ‘My dad is an ATM’ T-shirt chase...   \n",
       "64136       64136  How to Survive Being an Airbnb Host - The New ...   \n",
       "64137       64137  Fighting in Ukraine May Complicate U.S. Thaw W...   \n",
       "\n",
       "                                                    text  label  \n",
       "64133  — Alfredo Romero (@alfredoromero) October 26, ...      1  \n",
       "64134   From Monster to Mr. President-Elect: Democrat...      1  \n",
       "64135  Boy wearing a ‘My dad is an ATM’ T-shirt chase...      1  \n",
       "64136  After my first Airbnb guests checked out of ou...      0  \n",
       "64137  MOSCOW  —   The war in eastern Ukraine that si...      0  "
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df2.dropna(inplace=True)\n",
    "df2.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "76186a18",
   "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>Unnamed: 0</th>\n",
       "      <th>title</th>\n",
       "      <th>text</th>\n",
       "      <th>label</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>42133</th>\n",
       "      <td>42133</td>\n",
       "      <td>Want a Zika Test? It’s Not Easy - The New York...</td>\n",
       "      <td>MIAMI  —   After returning from a honeymoon in...</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>42134</th>\n",
       "      <td>42134</td>\n",
       "      <td>Rising Tumult Over Principal at Elite Queens H...</td>\n",
       "      <td>There has been a   and a rally, an allegation ...</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>42135</th>\n",
       "      <td>42135</td>\n",
       "      <td>APER Releasing the Arrivals</td>\n",
       "      <td>Support Us APER Releasing the Arrivals \\nMail ...</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>42136</th>\n",
       "      <td>42136</td>\n",
       "      <td>Hacker Guccifer 2.0 Says Election Could Be Rig...</td>\n",
       "      <td>Pin \\n( ZHE ) Before the recent torrent of dai...</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>42137</th>\n",
       "      <td>42137</td>\n",
       "      <td>Senate intelligence panel to probe Russia hacking</td>\n",
       "      <td>WASHINGTON (Reuters) - The leaders of the Sena...</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       Unnamed: 0                                              title  \\\n",
       "42133       42133  Want a Zika Test? It’s Not Easy - The New York...   \n",
       "42134       42134  Rising Tumult Over Principal at Elite Queens H...   \n",
       "42135       42135                        APER Releasing the Arrivals   \n",
       "42136       42136  Hacker Guccifer 2.0 Says Election Could Be Rig...   \n",
       "42137       42137  Senate intelligence panel to probe Russia hacking   \n",
       "\n",
       "                                                    text  label  \n",
       "42133  MIAMI  —   After returning from a honeymoon in...      0  \n",
       "42134  There has been a   and a rally, an allegation ...      0  \n",
       "42135  Support Us APER Releasing the Arrivals \\nMail ...      1  \n",
       "42136  Pin \\n( ZHE ) Before the recent torrent of dai...      1  \n",
       "42137  WASHINGTON (Reuters) - The leaders of the Sena...      0  "
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df3.dropna(inplace=True)\n",
    "df3.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "69e66f6e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['Unnamed: 0', 'title', 'text', 'label'], dtype='object')"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "d67e5a1f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(72134, 4)"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.shape(df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "d211787b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['Unnamed: 0', 'title', 'text', 'label'], dtype='object')"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df1.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "f558ad3f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(9928, 4)"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.shape(df1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "6eacaddc",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[Text(0.5, 0, 'Fake = 1, Real News = 0'),\n",
       " Text(0, 0.5, 'Number of Fake/True'),\n",
       " Text(0.5, 1.0, 'Count of Fake and True News Dataset #1')]"
      ]
     },
     "execution_count": 17,
     "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": [
    "import seaborn as sns\n",
    "df1_plt = sns.countplot(x = df1[\"label\"])\n",
    "df1_plt.set(xlabel='Fake = 1, Real News = 0', ylabel='Number of Fake/True',title='Count of Fake and True News Dataset #1')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "19633665",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[Text(0.5, 0, 'Fake = 1, Real News = 0'),\n",
       " Text(0, 0.5, 'Number of Fake/True'),\n",
       " Text(0.5, 1.0, 'Count of Fake and True News Dataset #2')]"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAkQAAAHHCAYAAABeLEexAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjYuMCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy89olMNAAAACXBIWXMAAA9hAAAPYQGoP6dpAABU4ElEQVR4nO3de1zO9/8/8MdV6urAVUKnSSKHohyycY0IEYvNsDGsnMcylNP6zJy3sDnNqRkTxsdpbBMr0copTNPQMCyyD5VTpdDx9fvDt/fPpaIrXV3xftxvt/ft5v16v67X+/m+XOXh/X6935dCCCFAREREJGMG+i6AiIiISN8YiIiIiEj2GIiIiIhI9hiIiIiISPYYiIiIiEj2GIiIiIhI9hiIiIiISPYYiIiIiEj2GIiIiIhI9hiIiCpBfn4+pk6dCgcHBxgYGKBPnz4636eXlxeaN2+u8/1UhlmzZkGhUOi7DCJ6hTEQUaW5cuUKPvroIzRo0AAmJiZQqVRo3749li1bhocPH+q7PADAqlWrEBYWVuHjfv/99/jqq6/Qv39/bNiwAYGBgaX29fLygkKhKHG5cOFChdf2KoiJiSn1PXt60beiOhYtWlRsW1hYGBQKBU6dOqWHyrQzdOhQjfe1evXqaNCgAfr3748ff/wRhYWF5R57y5YtWLp0acUV+wIePHiAWbNmISYmplyvT09Ph4GBASIjIwEAu3fvhlKpRE5Ojka/33//HePGjUOzZs1gbm6OevXq4f3338fff//9oodAZVRN3wWQPOzduxfvvfcelEol/Pz80Lx5c+Tm5uLIkSOYMmUKEhMTsWbNGn2XiVWrVqF27doYOnRohY4bHR2N1157DUuWLClT/7p16yIkJKRYu729fYXW9apwcXHBpk2bNNqCg4NRvXp1fPbZZ3qq6tm++uorjB07FmZmZvoupdyUSiXWrl0LAHj48CGuXbuGPXv2oH///vDy8sLPP/8MlUql9bhbtmzBuXPnMHHixAquWHsPHjzA7NmzATz+z4q2Tp48CQBo27YtACAuLg6tWrWCUqnU6LdgwQIcPXoU7733Htzd3ZGSkoIVK1agdevWOH78+CtztrcqYyAinUtKSsLAgQPh6OiI6Oho2NnZSdsCAgJw+fJl7N27V48V6l5aWhosLS3L3N/CwgJDhgzRXUGvGBsbm2Lv1/z581G7du1nvo+FhYXIzc2FiYmJrkvU0LJlSyQkJCA0NBRBQUGVuu+KVK1atWLv77x58zB//nwEBwdj1KhR2LZtm56qqxpOnjyJJk2aSD//cXFxUjh6UlBQELZs2QJjY2OpbcCAAXBzc8P8+fPxww8/VFbJssVLZqRzCxcuRFZWFtatW6cRhoo4OztjwoQJ0np+fj7mzp2Lhg0bQqlUon79+vjPf/5T7BSzQqHArFmzio1Xv359jTM8RZchjh49iqCgINSpUwfm5uZ49913cevWLY3XJSYmIjY2VroM8Lz/EWZnZ2PSpElwcHCAUqlEkyZN8PXXX0MIAQC4evUqFAoFfvvtNyQmJkrjlvf0OwD8/PPP8PX1hb29PZRKJRo2bIi5c+eioKDgua/dv38/zMzM8MEHHyA/Px8AcOHCBfTv3x9WVlYwMTFBmzZt8Msvv5Splq+//hpvvvkmatWqBVNTU3h4eGDnzp3F+ikUCowbNw4//fQTmjdvDqVSiWbNmiEiIqJY3yNHjuD111+HiYkJGjZsiG+//bZMtZRFUR2bN29Gs2bNoFQqERERIV1ye/rvpejv7+nLqC/yngFA+/bt0aVLFyxcuLBMl4uft7/09HQYGhrim2++kdpu374NAwMD1KpVS/o8AsDYsWNha2srrV+6dAn9+vWDra0tTExMULduXQwcOBAZGRllPp6nffrpp+jevTt27NihccmnLJ9dLy8v7N27F9euXZN+XurXrw8AyM3NxYwZM+Dh4QELCwuYm5vD09MTv/32W7Eatm7dCg8PD9SoUQMqlQpubm5YtmyZRp/09HRMnDhR+vl1dnbGggULpMt9V69eRZ06dQAAs2fPluop6ffOkzIyMnD79m3cvn0bcXFxaNGiBW7fvo3U1FTEx8ejadOmuH37tsZ7/Oabb2qEIQBo1KgRmjVrhvPnzz//TacXJ4h07LXXXhMNGjQoc39/f38BQPTv31+sXLlS+Pn5CQCiT58+Gv0AiJkzZxZ7vaOjo/D395fW169fLwCIVq1aiS5duojly5eLSZMmCUNDQ/H+++9L/Xbv3i3q1q0rmjZtKjZt2iQ2bdok9u/fX2qdhYWFokuXLkKhUIiRI0eKFStWiN69ewsAYuLEiUIIIbKyssSmTZtE06ZNRd26daVxU1JSSh23U6dOomnTpuLWrVsay/3794UQQvTp00e8//774quvvhKrV68W7733ngAgJk+eXGycZs2aSet79uwRSqVS+Pn5ifz8fCGEEOfOnRMWFhbC1dVVLFiwQKxYsUJ07NhRKBQKsWvXrlJrLFK3bl3x8ccfixUrVojFixeLN954QwAQ4eHhGv0AiBYtWgg7Ozsxd+5csXTpUtGgQQNhZmYmbt++LfU7c+aMMDU1FfXq1RMhISFi7ty5wsbGRri7uwttf101a9ZMdOrUqVgdLi4uok6dOmL27Nli5cqV4vTp0+K3334TAMRvv/2m0T8pKUkAEOvXr5faXvQ9AyACAgLEoUOHBACxaNEiaVvRZ/X333/Xen/u7u6iX79+0vru3buFgYGBACDOnTun8b70799fCCFETk6OcHJyEvb29mLevHli7dq1Yvbs2eL1118XV69efeZx+Pv7C3Nz81K3b9q0SQAQK1askNrK8tndv3+/aNmypahdu7b087J7924hhBC3bt0SdnZ2IigoSKxevVosXLhQNGnSRBgZGYnTp09rjAFAdO3aVaxcuVKsXLlSjBs3Trz33ntSn+zsbOHu7i5q1aol/vOf/4jQ0FDh5+cnFAqFmDBhghDi8c/v6tWrBQDx7rvvSvX8+eefz3xvOnXqJAA8d3n68/m0wsJC8dprr4nu3bs/sx9VDAYi0qmMjAwBQLzzzjtl6p+QkCAAiJEjR2q0T548WQAQ0dHRUpu2gcjb21sUFhZK7YGBgcLQ0FCkp6dLbSX9I1qan376SQAQ8+bN02jv37+/UCgU4vLly1Lb0+HkWUr7ZVp0TA8ePCj2mo8++kiYmZmJR48elbjPH3/8URgZGYlRo0aJgoICqU/Xrl2Fm5ubxusKCwvFm2++KRo1avTcWp+uJTc3VzRv3lx06dJFox2AMDY21nhP/vzzTwFALF++XGrr06ePMDExEdeuXZPa/vrrL2FoaFhhgcjAwEAkJiZqtGsTiF70PSsKREII0blzZ2Frayu9jyUForLuLyAgQNjY2EjrQUFBomPHjsLa2lqsXr1aCCHEnTt3hEKhEMuWLRNCCHH69GkBQOzYseO5dT/teYGoaOzAwECprayfXV9fX+Ho6Fisb35+vsjJydFou3fvnrCxsRHDhw+X2iZMmCBUKpUU/Esyd+5cYW5uLv7++2+N9k8//VQYGhqK5ORkIcTjEFba75rSnDp1SkRFRYmVK1cKAOKHH34QUVFRYujQocLBwUFERUWJqKgocerUqWeOUxQq161bV+Z9U/nxkhnpVGZmJgCgRo0aZeq/b98+ACg2r2LSpEkA8EJzjUaPHq1xl5GnpycKCgpw7dq1co23b98+GBoaYvz48cVqFULg119/LXet9evXR1RUlMYydepUAICpqanU7/79+7h9+zY8PT3x4MGDEu9C++9//4sBAwbgo48+wrfffgsDg8c/9nfv3kV0dDTef/99aZzbt2/jzp078PHxwaVLl/C///3vmXU+Wcu9e/eQkZEBT09P/PHHH8X6ent7o2HDhtK6u7s7VCoV/vnnHwBAQUEBIiMj0adPH9SrV0/q5+LiAh8fn7K8bWXSqVMnuLq6luu1FfGePWnWrFlISUlBaGjoC+/P09MTqampuHjxIgDg8OHD6NixIzw9PXH48GEAjy9HCiHg6ekJ4PFcNQCIjIzEgwcPyvWelKZ69eoAHn9Gi2j72X2aoaGhdFmpsLAQd+/eRX5+Ptq0aaPxmbO0tER2djaioqJKHWvHjh3w9PREzZo1pff19u3b8Pb2RkFBAQ4dOqT1MRfx8PCAt7c38vPzYW9vj8GDB8Pb2xu3bt1C165d4e3tDW9vb3h4eJQ6xoULFxAQEAC1Wg1/f/9y10Jlx0nVpFNFd5g8+UvxWa5duwYDAwM4OztrtNva2sLS0rLc4QWAxj+yAFCzZk0Aj/8hL49r167B3t6+WNhzcXGRtpeXubk5vL29S9yWmJiI6dOnIzo6WgqcRZ6e95GUlIQhQ4bgvffew/LlyzW2Xb58GUIIfP755/j8889L3FdaWhpee+21UusMDw/HvHnzkJCQoDHHq6Tb259+/4HHfwdF7/+tW7fw8OFDNGrUqFi/Jk2aSGH5RTk5OZX7tRXxnj2pY8eO6Ny5MxYuXIgxY8a80P6KQs7hw4dRt25dnD59GvPmzUOdOnXw9ddfS9tUKhVatGgB4PF7ERQUhMWLF2Pz5s3w9PTE22+/jSFDhkhhqbyysrIAaP5nSJvPbmk2bNiARYsW4cKFC8jLy5Pan/x7/fjjj7F9+3b07NkTr732Grp37473338fPXr0kPpcunQJZ86ckeYIPS0tLa1M9TwtKysLjx49AgBERUWhXbt2uH37NgoLC3H48GHMmzcPt2/fhqGhofQ76GkpKSnw9fWFhYUFdu7cCUNDw3LVQtphICKdUqlUsLe3x7lz57R63Ys8L6a0ycWl/VIRT0w4rerS09PRqVMnqFQqzJkzBw0bNoSJiQn++OMPTJs2rdizX+zs7GBnZ4d9+/bh1KlTaNOmjbStqO/kyZNLPQPzdDB90uHDh/H222+jY8eOWLVqFezs7GBkZIT169djy5YtxfpXlff/ybMURUr7vD39WXrR96wkM2fOhJeXF7799ttidyJqsz97e3s4OTnh0KFDqF+/PoQQUKvVqFOnDiZMmIBr167h8OHDePPNN6WzhACwaNEiDB06FD///DP279+P8ePHIyQkBMePH0fdunW1OpYnFf3MF9Wn7We3JD/88AOGDh2KPn36YMqUKbC2toahoSFCQkJw5coVqZ+1tTUSEhIQGRmJX3/9Fb/++ivWr18PPz8/bNiwAcDj97Zbt27SmdenNW7cuFzHPW7cOGkfRXbt2iX9efz48Rg/fjwcHR1x9erVYq/PyMhAz549kZ6ejsOHD/NRG5WIgYh0rlevXlizZg3i4uKgVquf2dfR0RGFhYW4dOmSdKYFAFJTU5Geng5HR0eprWbNmkhPT9d4fW5uLm7evFnuWrUJYo6Ojjhw4ADu37+v8b/golP/T9ZaUWJiYnDnzh3s2rULHTt2lNqTkpJK7G9iYoLw8HB06dIFPXr0QGxsLJo1awYAaNCgAQDAyMio1LNRz/Ljjz/CxMQEkZGRGs9UWb9+vdZjAUCdOnVgamqKS5cuFdtWdBlIV4r+p/705+nps3wv+p6VpFOnTvDy8sKCBQswY8aMF9qfp6cnDh06BCcnJ7Rs2RI1atRAixYtYGFhgYiICPzxxx/SM3We5ObmBjc3N0yfPh3Hjh1D+/btERoainnz5pX7uDZt2gSFQoFu3boB0O6zW9rP4c6dO9GgQQPs2rVLo8/MmTOL9TU2Nkbv3r3Ru3dvFBYW4uOPP8a3336Lzz//HM7OzmjYsCGysrKe+75q+5+zqVOnYsiQIUhKSsLo0aOxceNG2NnZYfv27di3b590x2JJwfzRo0fo3bs3/v77bxw4cKDcl3apfDiHiHRu6tSpMDc3x8iRI5Gamlps+5UrV6TbYd966y0AKPaU2sWLFwMAfH19pbaGDRsWu86/Zs2aMt1+Xhpzc/Ni/yiW5q233kJBQQFWrFih0b5kyRIoFAr07Nmz3HWUpugsy5NnVXJzc7Fq1apSX2NhYYHIyEhYW1ujW7du0v+kra2tpTMTJYXIJx9JUFotCoVC4/2+evUqfvrpJ20OSWM8Hx8f/PTTT0hOTpbaz58/Lz3lV1ccHR1haGhY7PP09Pv6ou9ZaYrmEj39cFJt9+fp6YmrV69i27Zt0iU0AwMDvPnmm1i8eDHy8vKkduDxHL+ixy8UcXNzg4GBQbHHXGhj/vz52L9/PwYMGCBdAtXms2tubl7iJbSSxjhx4gTi4uI0+t25c0dj3cDAAO7u7gAgHdf777+PuLi4Ej9b6enp0vtS9ODMsv5ecHV1hbe3N6pVq4aaNWtiyJAh8Pb2RmZmJjp06CDNH2rfvr3G6woKCjBgwADExcVhx44dz/3PI1U8niEinWvYsCG2bNmCAQMGwMXFReNJ1ceOHcOOHTuk5wa1aNEC/v7+WLNmjXSK/eTJk9iwYQP69OmDzp07S+OOHDkSY8aMQb9+/dCtWzf8+eefiIyMRO3atctdq4eHB1avXo158+bB2dkZ1tbW6NKlS4l9e/fujc6dO+Ozzz7D1atX0aJFC+zfvx8///wzJk6cqDGBuKK8+eabqFmzJvz9/TF+/HgoFAps2rTpuZedateujaioKOkX8pEjR/Daa69h5cqV6NChA9zc3DBq1Cg0aNAAqampiIuLw7///os///yz1DF9fX2xePFi9OjRA4MGDUJaWhpWrlwJZ2dnnDlzplzHN3v2bERERMDT0xMff/wx8vPzsXz5cjRr1qzcY5aFhYWFNM9KoVCgYcOGCA8PL3EeyYu8Z6Xp1KkTOnXqhNjY2BfaX1HYuXjxIr788kupvWPHjvj111+hVCrx+uuvS+3R0dEYN24c3nvvPTRu3Bj5+fnYtGkTDA0N0a9fv+fWnZ+fLz0w8NGjR7h27Rp++eUXnDlzBp07d9YIeNp8dj08PLBt2zYEBQXh9ddfR/Xq1dG7d2/06tULu3btwrvvvgtfX18kJSUhNDQUrq6u0pwl4PHvhrt376JLly6oW7curl27huXLl6Nly5bSmecpU6bgl19+Qa9evTB06FB4eHggOzsbZ8+exc6dO3H16lXUrl0bpqamcHV1xbZt29C4cWNYWVmhefPmz31y9NGjR9GuXTvpDNOxY8cwefLkUvtPmjQJv/zyC3r37o27d+8WexAjH9RaCfRybxvJ0t9//y1GjRol6tevL4yNjUWNGjVE+/btxfLlyzVuuc3LyxOzZ88WTk5OwsjISDg4OIjg4GCNPkIIUVBQIKZNmyZq164tzMzMhI+Pj7h8+XKpt90/eSuzECXfap2SkiJ8fX1FjRo1yvSckPv374vAwEBhb28vjIyMRKNGjcRXX32lcXu/ENrfdv+svkePHhXt2rUTpqamwt7eXkydOlVERkYWO5aSxrl8+bKws7MTLi4u4tatW0IIIa5cuSL8/PyEra2tMDIyEq+99pro1auX2Llz53NrXbdunWjUqJFQKpWiadOmYv369WLmzJnFbpHHE7eaP+npvyshhIiNjRUeHh7C2NhYNGjQQISGhpY45vOUdtt9SXUI8fj26n79+gkzMzNRs2ZN8dFHH4lz584Vu+1eiBd7z0qroejzWNJnVZv9WVtbCwAiNTVVajty5IgAIDw9PTX6/vPPP2L48OGiYcOGwsTERFhZWYnOnTuLAwcOPPc4ip4XVrSYmZmJ+vXri379+omdO3dqPN6hSFk/u1lZWWLQoEHC0tJSAJBuwS8sLBRffvmlcHR0FEqlUrRq1UqEh4cLf39/jdv0d+7cKbp37y6sra2FsbGxqFevnvjoo4/EzZs3Neq5f/++CA4OFs7OzsLY2FjUrl1bvPnmm+Lrr78Wubm5Ur9jx45Jn0mU8Rb8pk2birlz5wohhPj3339L/Ht90vOeXUS6pxDiJZpRSkRERKQDnENEREREssdARERERLLHQERERESyx0BEREREssdARERERLLHQERERESyxwczlkFhYSFu3LiBGjVqvNB3bBEREVHlEULg/v37sLe31/gOv5IwEJXBjRs34ODgoO8yiIiIqByuX7/+3C8rZiAqg6Iv7rx+/TpUKpWeqyEiIqKyyMzMhIODg8YXcJeGgagMii6TqVQqBiIiIqKXTFmmu3BSNREREckeAxERERHJHgMRERERyR4DEREREckeAxERERHJHgMRERERyR4DEREREckeAxERERHJHgMRERERyR4DEREREckeAxERERHJHgMRERERyR4DEREREckeAxERERHJHgMRERERyV41fRdQZP78+QgODsaECROwdOlSAMCjR48wadIkbN26FTk5OfDx8cGqVatgY2MjvS45ORljx47Fb7/9hurVq8Pf3x8hISGoVu3/H1pMTAyCgoKQmJgIBwcHTJ8+HUOHDq3kIyQiOUue46bvEoiqpHozzuq7BABV5AzR77//jm+//Rbu7u4a7YGBgdizZw927NiB2NhY3LhxA3379pW2FxQUwNfXF7m5uTh27Bg2bNiAsLAwzJgxQ+qTlJQEX19fdO7cGQkJCZg4cSJGjhyJyMjISjs+IiIiqtr0HoiysrIwePBgfPfdd6hZs6bUnpGRgXXr1mHx4sXo0qULPDw8sH79ehw7dgzHjx8HAOzfvx9//fUXfvjhB7Rs2RI9e/bE3LlzsXLlSuTm5gIAQkND4eTkhEWLFsHFxQXjxo1D//79sWTJEr0cLxEREVU9eg9EAQEB8PX1hbe3t0Z7fHw88vLyNNqbNm2KevXqIS4uDgAQFxcHNzc3jUtoPj4+yMzMRGJiotTn6bF9fHykMUqSk5ODzMxMjYWIiIheXXqdQ7R161b88ccf+P3334ttS0lJgbGxMSwtLTXabWxskJKSIvV5MgwVbS/a9qw+mZmZePjwIUxNTYvtOyQkBLNnzy73cREREdHLRW9niK5fv44JEyZg8+bNMDEx0VcZJQoODkZGRoa0XL9+Xd8lERERkQ7pLRDFx8cjLS0NrVu3RrVq1VCtWjXExsbim2++QbVq1WBjY4Pc3Fykp6drvC41NRW2trYAAFtbW6SmphbbXrTtWX1UKlWJZ4cAQKlUQqVSaSxERET06tJbIOratSvOnj2LhIQEaWnTpg0GDx4s/dnIyAgHDx6UXnPx4kUkJydDrVYDANRqNc6ePYu0tDSpT1RUFFQqFVxdXaU+T45R1KdoDCIiIiK9zSGqUaMGmjdvrtFmbm6OWrVqSe0jRoxAUFAQrKysoFKp8Mknn0CtVqNdu3YAgO7du8PV1RUffvghFi5ciJSUFEyfPh0BAQFQKpUAgDFjxmDFihWYOnUqhg8fjujoaGzfvh179+6t3AMmIiKiKqvKPJixJEuWLIGBgQH69eun8WDGIoaGhggPD8fYsWOhVqthbm4Of39/zJkzR+rj5OSEvXv3IjAwEMuWLUPdunWxdu1a+Pj46OOQiIiIqApSCCGEvouo6jIzM2FhYYGMjAzOJyKicuGTqolKpssnVWvz77fen0NEREREpG8MRERERCR7DEREREQkewxEREREJHtV+i4zufGYslHfJRBVSfFf+em7BCJ6xfEMEREREckeAxERERHJHgMRERERyR4DEREREckeAxERERHJHgMRERERyR4DEREREckeAxERERHJHgMRERERyR4DEREREckeAxERERHJHgMRERERyR4DEREREckeAxERERHJHgMRERERyR4DEREREckeAxERERHJHgMRERERyR4DEREREckeAxERERHJHgMRERERyR4DEREREckeAxERERHJHgMRERERyR4DEREREckeAxERERHJHgMRERERyZ5eA9Hq1avh7u4OlUoFlUoFtVqNX3/9Vdru5eUFhUKhsYwZM0ZjjOTkZPj6+sLMzAzW1taYMmUK8vPzNfrExMSgdevWUCqVcHZ2RlhYWGUcHhEREb0kqulz53Xr1sX8+fPRqFEjCCGwYcMGvPPOOzh9+jSaNWsGABg1ahTmzJkjvcbMzEz6c0FBAXx9fWFra4tjx47h5s2b8PPzg5GREb788ksAQFJSEnx9fTFmzBhs3rwZBw8exMiRI2FnZwcfH5/KPWAiIiKqkvQaiHr37q2x/sUXX2D16tU4fvy4FIjMzMxga2tb4uv379+Pv/76CwcOHICNjQ1atmyJuXPnYtq0aZg1axaMjY0RGhoKJycnLFq0CADg4uKCI0eOYMmSJQxEREREBKAKzSEqKCjA1q1bkZ2dDbVaLbVv3rwZtWvXRvPmzREcHIwHDx5I2+Li4uDm5gYbGxupzcfHB5mZmUhMTJT6eHt7a+zLx8cHcXFxOj4iIiIielno9QwRAJw9exZqtRqPHj1C9erVsXv3bri6ugIABg0aBEdHR9jb2+PMmTOYNm0aLl68iF27dgEAUlJSNMIQAGk9JSXlmX0yMzPx8OFDmJqaFqspJycHOTk50npmZmbFHTARERFVOXoPRE2aNEFCQgIyMjKwc+dO+Pv7IzY2Fq6urhg9erTUz83NDXZ2dujatSuuXLmChg0b6qymkJAQzJ49W2fjExERUdWi90tmxsbGcHZ2hoeHB0JCQtCiRQssW7asxL5t27YFAFy+fBkAYGtri9TUVI0+RetF845K66NSqUo8OwQAwcHByMjIkJbr16+X/wCJiIioytN7IHpaYWGhxuWqJyUkJAAA7OzsAABqtRpnz55FWlqa1CcqKgoqlUq67KZWq3Hw4EGNcaKiojTmKT1NqVRKjwIoWoiIiOjVpddLZsHBwejZsyfq1auH+/fvY8uWLYiJiUFkZCSuXLmCLVu24K233kKtWrVw5swZBAYGomPHjnB3dwcAdO/eHa6urvjwww+xcOFCpKSkYPr06QgICIBSqQQAjBkzBitWrMDUqVMxfPhwREdHY/v27di7d68+D52IiIiqEL0GorS0NPj5+eHmzZuwsLCAu7s7IiMj0a1bN1y/fh0HDhzA0qVLkZ2dDQcHB/Tr1w/Tp0+XXm9oaIjw8HCMHTsWarUa5ubm8Pf313hukZOTE/bu3YvAwEAsW7YMdevWxdq1a3nLPREREUn0GojWrVtX6jYHBwfExsY+dwxHR0fs27fvmX28vLxw+vRpresjIiIieahyc4iIiIiIKhsDEREREckeAxERERHJHgMRERERyR4DEREREckeAxERERHJHgMRERERyR4DEREREckeAxERERHJHgMRERERyR4DEREREckeAxERERHJHgMRERERyR4DEREREckeAxERERHJHgMRERERyR4DEREREckeAxERERHJHgMRERERyR4DEREREckeAxERERHJHgMRERERyR4DEREREckeAxERERHJHgMRERERyR4DEREREckeAxERERHJHgMRERERyR4DEREREckeAxERERHJHgMRERERyR4DEREREckeAxERERHJnl4D0erVq+Hu7g6VSgWVSgW1Wo1ff/1V2v7o0SMEBASgVq1aqF69Ovr164fU1FSNMZKTk+Hr6wszMzNYW1tjypQpyM/P1+gTExOD1q1bQ6lUwtnZGWFhYZVxeERERPSS0Gsgqlu3LubPn4/4+HicOnUKXbp0wTvvvIPExEQAQGBgIPbs2YMdO3YgNjYWN27cQN++faXXFxQUwNfXF7m5uTh27Bg2bNiAsLAwzJgxQ+qTlJQEX19fdO7cGQkJCZg4cSJGjhyJyMjISj9eIiIiqpoUQgih7yKeZGVlha+++gr9+/dHnTp1sGXLFvTv3x8AcOHCBbi4uCAuLg7t2rXDr7/+il69euHGjRuwsbEBAISGhmLatGm4desWjI2NMW3aNOzduxfnzp2T9jFw4ECkp6cjIiKiTDVlZmbCwsICGRkZUKlUFX/Q/8djykadjU30Mov/yk/fJbyw5Dlu+i6BqEqqN+OszsbW5t/vKjOHqKCgAFu3bkV2djbUajXi4+ORl5cHb29vqU/Tpk1Rr149xMXFAQDi4uLg5uYmhSEA8PHxQWZmpnSWKS4uTmOMoj5FY5QkJycHmZmZGgsRERG9uvQeiM6ePYvq1atDqVRizJgx2L17N1xdXZGSkgJjY2NYWlpq9LexsUFKSgoAICUlRSMMFW0v2vasPpmZmXj48GGJNYWEhMDCwkJaHBwcKuJQiYiIqIrSeyBq0qQJEhIScOLECYwdOxb+/v7466+/9FpTcHAwMjIypOX69et6rYeIiIh0q5q+CzA2NoazszMAwMPDA7///juWLVuGAQMGIDc3F+np6RpniVJTU2FrawsAsLW1xcmTJzXGK7oL7ck+T9+ZlpqaCpVKBVNT0xJrUiqVUCqVFXJ8REREVPXp/QzR0woLC5GTkwMPDw8YGRnh4MGD0raLFy8iOTkZarUaAKBWq3H27FmkpaVJfaKioqBSqeDq6ir1eXKMoj5FYxARERHp9QxRcHAwevbsiXr16uH+/fvYsmULYmJiEBkZCQsLC4wYMQJBQUGwsrKCSqXCJ598ArVajXbt2gEAunfvDldXV3z44YdYuHAhUlJSMH36dAQEBEhneMaMGYMVK1Zg6tSpGD58OKKjo7F9+3bs3btXn4dOREREVYheA1FaWhr8/Pxw8+ZNWFhYwN3dHZGRkejWrRsAYMmSJTAwMEC/fv2Qk5MDHx8frFq1Snq9oaEhwsPDMXbsWKjVapibm8Pf3x9z5syR+jg5OWHv3r0IDAzEsmXLULduXaxduxY+Pj6VfrxERERUNZXrOUSbNm1CaGgokpKSEBcXB0dHRyxduhROTk545513dFGnXvE5RET6xecQEb26XtrnEK1evRpBQUF46623kJ6ejoKCAgCApaUlli5dWq6CiYiIiPRJ60C0fPlyfPfdd/jss89gaGgotbdp0wZnz+ou5RERERHpitaBKCkpCa1atSrWrlQqkZ2dXSFFEREREVUmrQORk5MTEhISirVHRETAxcWlImoiIiIiqlRa32UWFBSEgIAAPHr0CEIInDx5Ev/9738REhKCtWvX6qJGIiIiIp3SOhCNHDkSpqammD59Oh48eIBBgwbB3t4ey5Ytw8CBA3VRIxEREZFOles5RIMHD8bgwYPx4MEDZGVlwdrauqLrIiIiIqo0L/RgRjMzM5iZmVVULURERER6oXUgcnJygkKhKHX7P//880IFEREREVU2rQPRxIkTNdbz8vJw+vRpREREYMqUKRVVFxEREVGl0ToQTZgwocT2lStX4tSpUy9cEBEREVFl0/o5RKXp2bMnfvzxx4oajoiIiKjSVFgg2rlzJ6ysrCpqOCIiIqJKo/Uls1atWmlMqhZCICUlBbdu3cKqVasqtDgiIiKiyqB1IOrTp4/GuoGBAerUqQMvLy80bdq0ouoiIiIiqjRaBaL8/Hw4OTnBx8cHNjY2uqqJiIiIqFJpNYeoWrVqGDNmDB49eqSreoiIiIgqndaTqt944w2cPn1aF7UQERER6YXWc4g+/vhjTJo0Cf/++y88PDxgbm6usd3d3b3CiiMiIiKqDGUORMOHD8fSpUulb7QfP368tE2hUEAIAYVCgYKCgoqvkoiIiEiHyhyINmzYgPnz5yMpKUmX9RARERFVujIHIiEEAMDR0VFnxRARERHpg1ZziO7fvw8TE5Nn9lGpVC9UEBEREVFl0yoQNW7cuNRtnENERERELyutAhG/r4yIiIheRVoFovbt28Pa2lpXtRARERHpRYV92z0RERHRy6rMgah27dqoVk3r5zgSERERVXllDkRZWVnw8/PDmjVrkJKSosuaiIiIiCpVmQPRhQsX4OPjg+3bt6N+/fpo27YtvvjiC5w9e1aX9RERERHpXJkDUb169fDJJ5/gwIEDSE1NxcSJE3H27Fl4enqiQYMGmDhxIqKjo3nbPREREb10yjWp2sLCAh988AG2bt2KW7du4dtvv0VBQQGGDRuGOnXqYPPmzRVdJxEREZHOvPBdZkZGRujWrRuWL1+Oa9eu4eDBg898gOOTQkJC8Prrr6NGjRqwtrZGnz59cPHiRY0+Xl5eUCgUGsuYMWM0+iQnJ8PX1xdmZmawtrbGlClTkJ+fr9EnJiYGrVu3hlKphLOzM8LCwl7ouImIiOjVUa5AdPjwYQwZMgRqtRr/+9//AACbNm3CkSNH0KpVK7z++utlGic2NhYBAQE4fvw4oqKikJeXh+7duyM7O1uj36hRo3Dz5k1pWbhwobStoKAAvr6+yM3NxbFjx7BhwwaEhYVhxowZUp+kpCT4+vqic+fOSEhIwMSJEzFy5EhERkaW5/CJiIjoFaP1ffQ//vgjPvzwQwwePBinT59GTk4OACAjIwNffvkl9u3bV+axIiIiNNbDwsJgbW2N+Ph4dOzYUWo3MzODra1tiWPs378ff/31Fw4cOAAbGxu0bNkSc+fOxbRp0zBr1iwYGxsjNDQUTk5OWLRoEQDAxcUFR44cwZIlS+Dj46PtW0BERESvGK3PEM2bNw+hoaH47rvvYGRkJLW3b98ef/zxxwsVk5GRAQDFvh5k8+bNqF27Npo3b47g4GA8ePBA2hYXFwc3NzfY2NhIbT4+PsjMzERiYqLUx9vbW2NMHx8fxMXFvVC9RERE9GrQ+gzRxYsXNc7eFLGwsEB6enq5CyksLMTEiRPRvn17NG/eXGofNGgQHB0dYW9vjzNnzmDatGm4ePEidu3aBQBISUnRCEMApPWi5yWV1iczMxMPHz6EqampxracnBzpzBcAZGZmlvu4iIiIqOrTOhDZ2tri8uXLqF+/vkb7kSNH0KBBg3IXEhAQgHPnzuHIkSMa7aNHj5b+7ObmBjs7O3Tt2hVXrlxBw4YNy72/ZwkJCcHs2bN1MjYRERFVPVpfMhs1ahQmTJiAEydOQKFQ4MaNG9i8eTMmT56MsWPHlquIcePGITw8HL/99hvq1q37zL5t27YFAFy+fBnA44CWmpqq0adovWjeUWl9VCpVsbNDABAcHIyMjAxpuX79ermOi4iIiF4OWp8h+vTTT1FYWIiuXbviwYMH6NixI5RKJSZPnoxPPvlEq7GEEPjkk0+we/duxMTEwMnJ6bmvSUhIAADY2dkBANRqNb744gukpaXB2toaABAVFQWVSgVXV1epz9OTvaOioqBWq0vch1KphFKp1OpYiIiI6OWldSDKz8/HZ599hilTpuDy5cvIysqCq6srqlevjtu3b6N27dplHisgIABbtmzBzz//jBo1akhzfiwsLGBqaoorV65gy5YteOutt1CrVi2cOXMGgYGB6NixI9zd3QEA3bt3h6urKz788EMsXLgQKSkpmD59OgICAqRQM2bMGKxYsQJTp07F8OHDER0dje3bt2Pv3r3aHj4RERG9grS+ZDZw4EAIIWBsbAxXV1e88cYbqF69OlJTU+Hl5aXVWKtXr0ZGRga8vLxgZ2cnLdu2bQMAGBsb48CBA+jevTuaNm2KSZMmoV+/ftizZ480hqGhIcLDw2FoaAi1Wo0hQ4bAz88Pc+bMkfo4OTlh7969iIqKQosWLbBo0SKsXbuWt9wTERERgHKcIUpOTsbIkSOxbt06qe3mzZvo0qULmjVrptVYQohnbndwcEBsbOxzx3F0dHzu84+8vLxw+vRpreojIiIiedD6DNG+fftw7NgxBAUFAQBu3LgBLy8vuLm5Yfv27RVeIBEREZGuaX2GqE6dOti/fz86dOgAAAgPD0fr1q2xefNmGBi88FejEREREVU6rQMR8PhSVlRUFDw9PdGtWzds2rQJCoWiomsjIiIiqhRlCkQ1a9YsMfA8ePAAe/bsQa1ataS2u3fvVlx1RERERJWgTIFo6dKlOi6DiIiISH/KFIj8/f11XQcRERGR3pRrDlGRR48eITc3V6NNpVK9UEFERERElU3r28Kys7Mxbtw4WFtbw9zcHDVr1tRYiIiIiF42WgeiqVOnIjo6GqtXr4ZSqcTatWsxe/Zs2NvbY+PGjbqokYiIiEintL5ktmfPHmzcuBFeXl4YNmwYPD094ezsDEdHR2zevBmDBw/WRZ1EREREOqP1GaK7d++iQYMGAB7PFyq6zb5Dhw44dOhQxVZHREREVAm0DkQNGjRAUlISAKBp06bS13Xs2bMHlpaWFVocERERUWXQOhANGzYMf/75JwDg008/xcqVK2FiYoLAwEBMmTKlwgskIiIi0rUyzyH6559/4OTkhMDAQKnN29sbFy5cQHx8PJydneHu7q6TIomIiIh0qcxniBo1aoRbt25J6wMGDEBqaiocHR3Rt29fhiEiIiJ6aZU5EAkhNNb37duH7OzsCi+IiIiIqLJpPYeIiIiI6FVT5kCkUCiKfeP90+tEREREL6MyT6oWQmDo0KFQKpUAHn+P2ZgxY2Bubq7Rb9euXRVbIREREZGOlTkQPf2N90OGDKnwYoiIiIj0ocyBaP369bqsg4iIiEhvOKmaiIiIZI+BiIiIiGSPgYiIiIhkj4GIiIiIZK9Mgah169a4d+8eAGDOnDl48OCBTosiIiIiqkxlCkTnz5+XvqZj9uzZyMrK0mlRRERERJWpTLfdt2zZEsOGDUOHDh0ghMDXX3+N6tWrl9h3xowZFVogERERka6VKRCFhYVh5syZCA8Ph0KhwK+//opq1Yq/VKFQMBARERHRS6dMgahJkybYunUrAMDAwAAHDx6EtbW1TgsjIiIiqixlflJ1kcLCQl3UQURERKQ3WgciALhy5QqWLl2K8+fPAwBcXV0xYcIENGzYsEKLIyIiIqoMWj+HKDIyEq6urjh58iTc3d3h7u6OEydOoFmzZoiKitJFjUREREQ6pXUg+vTTTxEYGIgTJ05g8eLFWLx4MU6cOIGJEydi2rRpWo0VEhKC119/HTVq1IC1tTX69OmDixcvavR59OgRAgICUKtWLVSvXh39+vVDamqqRp/k5GT4+vrCzMwM1tbWmDJlCvLz8zX6xMTEoHXr1lAqlXB2dkZYWJi2h05ERESvKK0D0fnz5zFixIhi7cOHD8dff/2l1VixsbEICAjA8ePHERUVhby8PHTv3l165hEABAYGYs+ePdixYwdiY2Nx48YN9O3bV9peUFAAX19f5Obm4tixY9iwYQPCwsI07nZLSkqCr68vOnfujISEBEycOBEjR45EZGSktodPREREryCt5xDVqVMHCQkJaNSokUZ7QkKC1neeRUREaKyHhYXB2toa8fHx6NixIzIyMrBu3Tps2bIFXbp0AQCsX78eLi4uOH78ONq1a4f9+/fjr7/+woEDB2BjY4OWLVti7ty5mDZtGmbNmgVjY2OEhobCyckJixYtAgC4uLjgyJEjWLJkCXx8fLR9C4iIiOgVo/UZolGjRmH06NFYsGABDh8+jMOHD2P+/Pn46KOPMGrUqBcqJiMjAwBgZWUFAIiPj0deXh68vb2lPk2bNkW9evUQFxcHAIiLi4ObmxtsbGykPj4+PsjMzERiYqLU58kxivoUjfG0nJwcZGZmaixERET06tL6DNHnn3+OGjVqYNGiRQgODgYA2NvbY9asWRg/fny5CyksLMTEiRPRvn17NG/eHACQkpICY2NjWFpaavS1sbFBSkqK1OfJMFS0vWjbs/pkZmbi4cOHMDU11dgWEhKC2bNnl/tYiIiI6OWidSBSKBQIDAxEYGAg7t+/DwCoUaPGCxcSEBCAc+fO4ciRIy881osKDg5GUFCQtJ6ZmQkHBwc9VkRERES6VK7nEBWpiCAEAOPGjUN4eDgOHTqEunXrSu22trbIzc1Fenq6xlmi1NRU2NraSn1OnjypMV7RXWhP9nn6zrTU1FSoVKpiZ4cAQKlUQqlUVsixERERUdWn9RyiiiSEwLhx47B7925ER0fDyclJY7uHhweMjIxw8OBBqe3ixYtITk6GWq0GAKjVapw9exZpaWlSn6ioKKhUKri6ukp9nhyjqE/RGERERCRvL3SG6EUFBARgy5Yt+Pnnn1GjRg1pzo+FhQVMTU1hYWGBESNGICgoCFZWVlCpVPjkk0+gVqvRrl07AED37t3h6uqKDz/8EAsXLkRKSgqmT5+OgIAA6SzPmDFjsGLFCkydOhXDhw9HdHQ0tm/fjr179+rt2ImIiKjq0OsZotWrVyMjIwNeXl6ws7OTlm3btkl9lixZgl69eqFfv37o2LEjbG1tsWvXLmm7oaEhwsPDYWhoCLVajSFDhsDPzw9z5syR+jg5OWHv3r2IiopCixYtsGjRIqxdu5a33BMREREAQCGEEGXtnJeXhx49eiA0NLTYc4heZZmZmbCwsEBGRgZUKpXO9uMxZaPOxiZ6mcV/5afvEl5Y8hw3fZdAVCXVm3FWZ2Nr8++3VmeIjIyMcObMmRcqjoiIiKiq0fqS2ZAhQ7Bu3Tpd1EJERESkF1pPqs7Pz8f333+PAwcOwMPDA+bm5hrbFy9eXGHFEREREVUGrQPRuXPn0Lp1awDA33//rbFNoVBUTFVERERElUjrQPTbb7/pog4iIiIivSn3bfeXL19GZGQkHj58CODxQxaJiIiIXkZaB6I7d+6ga9euaNy4Md566y3cvHkTADBixAhMmjSpwgskIiIi0jWtA1FgYCCMjIyQnJwMMzMzqX3AgAGIiIio0OKIiIiIKoPWc4j279+PyMhIjS9hBYBGjRrh2rVrFVYYERERUWXR+gxRdna2xpmhInfv3uU3xBMREdFLSetA5OnpiY0b//9XTCgUChQWFmLhwoXo3LlzhRZHREREVBm0vmS2cOFCdO3aFadOnUJubi6mTp2KxMRE3L17F0ePHtVFjUREREQ6pfUZoubNm+Pvv/9Ghw4d8M477yA7Oxt9+/bF6dOn0bBhQ13USERERKRTWp8hAgALCwt89tlnFV0LERERkV6UKxDdu3cP69atw/nz5wEArq6uGDZsGKysrCq0OCIiIqLKoPUls0OHDqF+/fr45ptvcO/ePdy7dw/ffPMNnJyccOjQIV3USERERKRTWp8hCggIwIABA7B69WoYGhoCAAoKCvDxxx8jICAAZ8+erfAiiYiIiHRJ6zNEly9fxqRJk6QwBACGhoYICgrC5cuXK7Q4IiIiosqgdSBq3bq1NHfoSefPn0eLFi0qpCgiIiKiylSmS2ZnzpyR/jx+/HhMmDABly9fRrt27QAAx48fx8qVKzF//nzdVElERESkQ2UKRC1btoRCoYAQQmqbOnVqsX6DBg3CgAEDKq46IiIiokpQpkCUlJSk6zqIiIiI9KZMgcjR0VHXdRARERHpTbkezHjjxg0cOXIEaWlpKCws1Ng2fvz4CimMiIiIqLJoHYjCwsLw0UcfwdjYGLVq1YJCoZC2KRQKBiIiIiJ66WgdiD7//HPMmDEDwcHBMDDQ+q59IiIioipH60Tz4MEDDBw4kGGIiIiIXhlap5oRI0Zgx44duqiFiIiISC+0vmQWEhKCXr16ISIiAm5ubjAyMtLYvnjx4gorjoiIiKgylCsQRUZGokmTJgBQbFI1ERER0ctG60C0aNEifP/99xg6dKgOyiEiIiKqfFrPIVIqlWjfvr0uaiEiIiLSC60D0YQJE7B8+fIK2fmhQ4fQu3dv2NvbQ6FQ4KefftLYPnToUCgUCo2lR48eGn3u3r2LwYMHQ6VSwdLSEiNGjEBWVpZGnzNnzsDT0xMmJiZwcHDAwoULK6R+IiIiejVofcns5MmTiI6ORnh4OJo1a1ZsUvWuXbvKPFZ2djZatGiB4cOHo2/fviX26dGjB9avXy+tK5VKje2DBw/GzZs3ERUVhby8PAwbNgyjR4/Gli1bAACZmZno3r07vL29ERoairNnz2L48OGwtLTE6NGjy1wrERERvbq0DkSWlpalhhdt9ezZEz179nxmH6VSCVtb2xK3nT9/HhEREfj999/Rpk0bAMDy5cvx1ltv4euvv4a9vT02b96M3NxcfP/99zA2NkazZs2QkJCAxYsXMxARERERgHIEoifP1lSGmJgYWFtbo2bNmujSpQvmzZuHWrVqAQDi4uJgaWkphSEA8Pb2hoGBAU6cOIF3330XcXFx6NixI4yNjaU+Pj4+WLBgAe7du4eaNWsW22dOTg5ycnKk9czMTB0eIREREelblX7cdI8ePbBx40YcPHgQCxYsQGxsLHr27ImCggIAQEpKCqytrTVeU61aNVhZWSElJUXqY2Njo9GnaL2oz9NCQkJgYWEhLQ4ODhV9aERERFSFaH2GyMnJ6ZnPG/rnn39eqKAnDRw4UPqzm5sb3N3d0bBhQ8TExKBr164Vtp+nBQcHIygoSFrPzMxkKCIiInqFaR2IJk6cqLGel5eH06dPIyIiAlOmTKmoukrUoEED1K5dG5cvX0bXrl1ha2uLtLQ0jT75+fm4e/euNO/I1tYWqampGn2K1kubm6RUKotN3iYiIqJXl9aBaMKECSW2r1y5EqdOnXrhgp7l33//xZ07d2BnZwcAUKvVSE9PR3x8PDw8PAAA0dHRKCwsRNu2baU+n332GfLy8qQ74qKiotCkSZMS5w8RERGR/FTYHKKePXvixx9/1Oo1WVlZSEhIQEJCAgAgKSkJCQkJSE5ORlZWFqZMmYLjx4/j6tWrOHjwIN555x04OzvDx8cHAODi4oIePXpg1KhROHnyJI4ePYpx48Zh4MCBsLe3BwAMGjQIxsbGGDFiBBITE7Ft2zYsW7ZM45IYERERyVuFBaKdO3fCyspKq9ecOnUKrVq1QqtWrQAAQUFBaNWqFWbMmAFDQ0OcOXMGb7/9Nho3bowRI0bAw8MDhw8f1rictXnzZjRt2hRdu3bFW2+9hQ4dOmDNmjXSdgsLC+zfvx9JSUnw8PDApEmTMGPGDN5yT0RERBKtL5m1atVKY1K1EAIpKSm4desWVq1apdVYXl5eEEKUuj0yMvK5Y1hZWUkPYSyNu7s7Dh8+rFVtREREJB9aB6I+ffporBsYGKBOnTrw8vJC06ZNK6ouIiIiokqjdSCaOXOmLuogIiIi0psq/WBGIiIiospQ5jNEBgYGz3wgIwAoFArk5+e/cFFERERElanMgWj37t2lbouLi8M333yDwsLCCimKiIiIqDKVORC98847xdouXryITz/9FHv27MHgwYMxZ86cCi2OiIiIqDKUaw7RjRs3MGrUKLi5uSE/Px8JCQnYsGEDHB0dK7o+IiIiIp3TKhBlZGRg2rRpcHZ2RmJiIg4ePIg9e/agefPmuqqPiIiISOfKfMls4cKFWLBgAWxtbfHf//63xEtoRERERC+jMgeiTz/9FKampnB2dsaGDRuwYcOGEvvt2rWrwoojIiIiqgxlDkR+fn7Pve2eiIiI6GVU5kAUFhamwzKIiIiI9IdPqiYiIiLZYyAiIiIi2WMgIiIiItljICIiIiLZYyAiIiIi2WMgIiIiItljICIiIiLZYyAiIiIi2WMgIiIiItljICIiIiLZYyAiIiIi2WMgIiIiItljICIiIiLZYyAiIiIi2WMgIiIiItljICIiIiLZYyAiIiIi2WMgIiIiItljICIiIiLZYyAiIiIi2WMgIiIiItnTayA6dOgQevfuDXt7eygUCvz0008a24UQmDFjBuzs7GBqagpvb29cunRJo8/du3cxePBgqFQqWFpaYsSIEcjKytLoc+bMGXh6esLExAQODg5YuHChrg+NiIiIXiJ6DUTZ2dlo0aIFVq5cWeL2hQsX4ptvvkFoaChOnDgBc3Nz+Pj44NGjR1KfwYMHIzExEVFRUQgPD8ehQ4cwevRoaXtmZia6d+8OR0dHxMfH46uvvsKsWbOwZs0anR8fERERvRyq6XPnPXv2RM+ePUvcJoTA0qVLMX36dLzzzjsAgI0bN8LGxgY//fQTBg4ciPPnzyMiIgK///472rRpAwBYvnw53nrrLXz99dewt7fH5s2bkZubi++//x7GxsZo1qwZEhISsHjxYo3gRERERPJVZecQJSUlISUlBd7e3lKbhYUF2rZti7i4OABAXFwcLC0tpTAEAN7e3jAwMMCJEyekPh07doSxsbHUx8fHBxcvXsS9e/dK3HdOTg4yMzM1FiIiInp1VdlAlJKSAgCwsbHRaLexsZG2paSkwNraWmN7tWrVYGVlpdGnpDGe3MfTQkJCYGFhIS0ODg4vfkBERERUZVXZQKRPwcHByMjIkJbr16/ruyQiIiLSoSobiGxtbQEAqampGu2pqanSNltbW6SlpWlsz8/Px927dzX6lDTGk/t4mlKphEql0liIiIjo1VVlA5GTkxNsbW1x8OBBqS0zMxMnTpyAWq0GAKjVaqSnpyM+Pl7qEx0djcLCQrRt21bqc+jQIeTl5Ul9oqKi0KRJE9SsWbOSjoaIiIiqMr0GoqysLCQkJCAhIQHA44nUCQkJSE5OhkKhwMSJEzFv3jz88ssvOHv2LPz8/GBvb48+ffoAAFxcXNCjRw+MGjUKJ0+exNGjRzFu3DgMHDgQ9vb2AIBBgwbB2NgYI0aMQGJiIrZt24Zly5YhKChIT0dNREREVY1eb7s/deoUOnfuLK0XhRR/f3+EhYVh6tSpyM7OxujRo5Geno4OHTogIiICJiYm0ms2b96McePGoWvXrjAwMEC/fv3wzTffSNstLCywf/9+BAQEwMPDA7Vr18aMGTN4yz0RERFJFEIIoe8iqrrMzExYWFggIyNDp/OJPKZs1NnYRC+z+K/89F3CC0ue46bvEoiqpHozzupsbG3+/a6yc4iIiIiIKgsDEREREckeAxERERHJHgMRERERyR4DEREREckeAxERERHJHgMRERERyR4DEREREckeAxERERHJHgMRERERyR4DEREREckeAxERERHJHgMRERERyR4DEREREckeAxERERHJHgMRERERyR4DEREREckeAxERERHJHgMRERERyR4DEREREckeAxERERHJHgMRERERyR4DEREREckeAxERERHJHgMRERERyR4DEREREckeAxERERHJHgMRERERyR4DEREREckeAxERERHJHgMRERERyR4DEREREclelQ5Es2bNgkKh0FiaNm0qbX/06BECAgJQq1YtVK9eHf369UNqaqrGGMnJyfD19YWZmRmsra0xZcoU5OfnV/ahEBERURVWTd8FPE+zZs1w4MABab1atf9fcmBgIPbu3YsdO3bAwsIC48aNQ9++fXH06FEAQEFBAXx9fWFra4tjx47h5s2b8PPzg5GREb788stKPxYiIiKqmqp8IKpWrRpsbW2LtWdkZGDdunXYsmULunTpAgBYv349XFxccPz4cbRr1w779+/HX3/9hQMHDsDGxgYtW7bE3LlzMW3aNMyaNQvGxsaVfThERERUBVXpS2YAcOnSJdjb26NBgwYYPHgwkpOTAQDx8fHIy8uDt7e31Ldp06aoV68e4uLiAABxcXFwc3ODjY2N1MfHxweZmZlITEwsdZ85OTnIzMzUWIiIiOjVVaUDUdu2bREWFoaIiAisXr0aSUlJ8PT0xP3795GSkgJjY2NYWlpqvMbGxgYpKSkAgJSUFI0wVLS9aFtpQkJCYGFhIS0ODg4Ve2BERERUpVTpS2Y9e/aU/uzu7o62bdvC0dER27dvh6mpqc72GxwcjKCgIGk9MzOToYiIiOgVVqXPED3N0tISjRs3xuXLl2Fra4vc3Fykp6dr9ElNTZXmHNna2ha766xovaR5SUWUSiVUKpXGQkRERK+ulyoQZWVl4cqVK7Czs4OHhweMjIxw8OBBafvFixeRnJwMtVoNAFCr1Th79izS0tKkPlFRUVCpVHB1da30+omIiKhqqtKXzCZPnozevXvD0dERN27cwMyZM2FoaIgPPvgAFhYWGDFiBIKCgmBlZQWVSoVPPvkEarUa7dq1AwB0794drq6u+PDDD7Fw4UKkpKRg+vTpCAgIgFKp1PPRERERUVVRpQPRv//+iw8++AB37txBnTp10KFDBxw/fhx16tQBACxZsgQGBgbo168fcnJy4OPjg1WrVkmvNzQ0RHh4OMaOHQu1Wg1zc3P4+/tjzpw5+jokIiIiqoKqdCDaunXrM7ebmJhg5cqVWLlyZal9HB0dsW/fvooujYiIiF4hL9UcIiIiIiJdYCAiIiIi2WMgIiIiItljICIiIiLZYyAiIiIi2WMgIiIiItljICIiIiLZYyAiIiIi2WMgIiIiItljICIiIiLZYyAiIiIi2WMgIiIiItljICIiIiLZYyAiIiIi2WMgIiIiItljICIiIiLZYyAiIiIi2WMgIiIiItljICIiIiLZYyAiIiIi2WMgIiIiItljICIiIiLZYyAiIiIi2WMgIiIiItljICIiIiLZYyAiIiIi2WMgIiIiItljICIiIiLZYyAiIiIi2WMgIiIiItljICIiIiLZYyAiIiIi2ZNVIFq5ciXq168PExMTtG3bFidPntR3SURERFQFyCYQbdu2DUFBQZg5cyb++OMPtGjRAj4+PkhLS9N3aURERKRnsglEixcvxqhRozBs2DC4uroiNDQUZmZm+P777/VdGhEREemZLAJRbm4u4uPj4e3tLbUZGBjA29sbcXFxeqyMiIiIqoJq+i6gMty+fRsFBQWwsbHRaLexscGFCxeK9c/JyUFOTo60npGRAQDIzMzUaZ0FOQ91Oj7Ry0rXP3uV4f6jAn2XQFQl6fLnu2hsIcRz+8oiEGkrJCQEs2fPLtbu4OCgh2qIyGL5GH2XQES6EmKh813cv38fFhbP3o8sAlHt2rVhaGiI1NRUjfbU1FTY2toW6x8cHIygoCBpvbCwEHfv3kWtWrWgUCh0Xi/pV2ZmJhwcHHD9+nWoVCp9l0NEFYg/3/IihMD9+/dhb2//3L6yCETGxsbw8PDAwYMH0adPHwCPQ87Bgwcxbty4Yv2VSiWUSqVGm6WlZSVUSlWJSqXiL0yiVxR/vuXjeWeGisgiEAFAUFAQ/P390aZNG7zxxhtYunQpsrOzMWzYMH2XRkRERHomm0A0YMAA3Lp1CzNmzEBKSgpatmyJiIiIYhOtiYiISH5kE4gAYNy4cSVeIiN6klKpxMyZM4tdNiWilx9/vqk0ClGWe9GIiIiIXmGyeDAjERER0bMwEBEREZHsMRARERGR7DEQERERkewxEBE9ZeXKlahfvz5MTEzQtm1bnDx5Ut8lEVEFOHToEHr37g17e3soFAr89NNP+i6JqhAGIqInbNu2DUFBQZg5cyb++OMPtGjRAj4+PkhLS9N3aUT0grKzs9GiRQusXLlS36VQFcTb7ome0LZtW7z++utYsWIFgMdf8eLg4IBPPvkEn376qZ6rI6KKolAosHv3bunrnIh4hojo/+Tm5iI+Ph7e3t5Sm4GBAby9vREXF6fHyoiISNcYiIj+z+3bt1FQUFDs61xsbGyQkpKip6qIiKgyMBARERGR7DEQEf2f2rVrw9DQEKmpqRrtqampsLW11VNVRERUGRiIiP6PsbExPDw8cPDgQamtsLAQBw8ehFqt1mNlRESka7L6tnui5wkKCoK/vz/atGmDN954A0uXLkV2djaGDRum79KI6AVlZWXh8uXL0npSUhISEhJgZWWFevXq6bEyqgp42z3RU1asWIGvvvoKKSkpaNmyJb755hu0bdtW32UR0QuKiYlB586di7X7+/sjLCys8guiKoWBiIiIiGSPc4iIiIhI9hiIiIiISPYYiIiIiEj2GIiIiIhI9hiIiIiISPYYiIiIiEj2GIiIiIhI9hiIiGQmLCwMlpaW+i7jpXX16lUoFAokJCTouxQiqkAMREQvoaFDh0KhUBRbnvxagpfJrl270L17d9SqVavcYaMoqBQtVlZW6NSpEw4fPlzxBT+Hl5cXFAoFtm7dqtG+dOlS1K9fv9Lr0ZW7d+9i8ODBUKlUsLS0xIgRI5CVlaXvsojKhYGI6CXVo0cP3Lx5U2NxcnLSd1nlkp2djQ4dOmDBggUvPNaBAwdw8+ZNHDp0CPb29ujVqxdSU1MroErtmJiYYPr06cjLy6v0fVeWwYMHIzExEVFRUQgPD8ehQ4cwevRofZdFVC4MREQvKaVSCVtbW43F0NAQixcvhpubG8zNzeHg4ICPP/74mf9rv3XrFtq0aYN3330XOTk5KCwsREhICJycnGBqaooWLVpg586dOj2WDz/8EDNmzIC3t/cLj1WrVi3Y2tqiefPm+M9//oPMzEycOHFC2n7u3Dn07NkT1atXh42NDT788EPcvn1b2h4REYEOHTrA0tIStWrVQq9evXDlyhWt6/jggw+Qnp6O77777pn9fv75Z7Ru3RomJiZo0KABZs+ejfz8fADA5MmT0atXL6nv0qVLoVAoEBERIbU5Oztj7dq1AB5/V9cbb7wBc3NzWFpaon379rh27ZrWtZfF+fPnERERgbVr16Jt27bo0KEDli9fjq1bt+LGjRs62SeRLjEQEb1iDAwM8M033yAxMREbNmxAdHQ0pk6dWmLf69evw9PTE82bN8fOnTuhVCoREhKCjRs3IjQ0FImJiQgMDMSQIUMQGxtb6j7HjBmD6tWrP3OpbA8fPsTGjRsBAMbGxgCA9PR0dOnSBa1atcKpU6cQERGB1NRUvP/++9LrsrOzERQUhFOnTuHgwYMwMDDAu+++i8LCQq32r1Kp8Nlnn2HOnDnIzs4usc/hw4fh5+eHCRMm4K+//sK3336LsLAwfPHFFwCATp064ciRIygoKAAAxMbGonbt2oiJiQEA/O9//8OVK1fg5eWF/Px89OnTB506dcKZM2cQFxeH0aNHQ6FQlFpjs2bNnvl31rNnz1JfGxcXB0tLS7Rp00Zq8/b2hoGBgUYAJXppCCJ66fj7+wtDQ0Nhbm4uLf379y+x744dO0StWrWk9fXr1wsLCwtx4cIF4eDgIMaPHy8KCwuFEEI8evRImJmZiWPHjmmMMWLECPHBBx+UWk9qaqq4dOnSM5eySEpKEgDE6dOny9S/pNeampoKc3NzoVAoBADh4eEhcnNzhRBCzJ07V3Tv3l3jddevXxcAxMWLF0sc99atWwKAOHv2bJlr7NSpk5gwYYJ49OiRcHR0FHPmzBFCCLFkyRLh6Ogo9evatav48ssvNV67adMmYWdnJ4QQ4t69e8LAwED8/vvvorCwUFhZWYmQkBDRtm1bIYQQP/zwg3jttdeEEELcuXNHABAxMTFlfMeEuHr16jP/zv79999SX/vFF1+Ixo0bF2uvU6eOWLVqVZlrIKoqqukxixHRC+jcuTNWr14trZubmwN4PIcmJCQEFy5cQGZmJvLz8/Ho0SM8ePAAZmZmAB6fPfH09MSgQYOwdOlSaYzLly/jwYMH6Natm8a+cnNz0apVq1Jrsba2hrW1dQUeXflt27YNTZs2xblz5zB16lSEhYXByMgIAPDnn3/it99+K/GM1ZUrV9C4cWNcunQJM2bMwIkTJ3D79m3pzFBycjKaN2+uVS1KpRJz5szBJ598grFjxxbb/ueff+Lo0aPSGSEAKCgokP6+LC0t0aJFC8TExMDY2BjGxsYYPXo0Zs6ciaysLMTGxqJTp04AACsrKwwdOhQ+Pj7o1q0bvL298f7778POzq7U+hwdHbU6HqJXGQMR0UvK3Nwczs7OGm1Xr15Fr169MHbsWHzxxRewsrLCkSNHMGLECOTm5kqBSKlUwtvbG+Hh4ZgyZQpee+01AJDmGu3du1dqK6JUKkutZcyYMfjhhx+eWW9l3X3k4OCARo0aoVGjRsjPz8e7776Lc+fOQalUIisrC7179y5x8nZRcOjduzccHR3x3Xffwd7eHoWFhWjevDlyc3PLVc+QIUPw9ddfY968ecXuMMvKysLs2bPRt2/fYq8zMTEB8PiOtZiYGCiVSnTq1AlWVlZwcXHBkSNHEBsbi0mTJkmvWb9+PcaPH4+IiAhs27YN06dPR1RUFNq1a1dibc2aNXvmHCNPT0/8+uuvJW6ztbVFWlqaRlt+fj7u3r0LW1vbUsckqqoYiIheIfHx8SgsLMSiRYtgYPB4iuD27duL9TMwMMCmTZswaNAgdO7cGTExMbC3t4erqyuUSiWSk5OlMw9lMWfOHEyePLnCjqOi9O/fHzNmzMCqVasQGBiI1q1b48cff0T9+vVRrVrxX3937tzBxYsX8d1338HT0xMAcOTIkReqwcDAACEhIejbt2+xs0StW7fGxYsXiwXbJ3Xq1Anff/89qlWrhh49egB4HJL++9//4u+//4aXl5dG/1atWqFVq1YIDg6GWq3Gli1bSg1E+/bte+ZdcKampqVuU6vVSE9PR3x8PDw8PAAA0dHRKCwsRNu2bUt9HVFVxUBE9ApxdnZGXl4eli9fjt69e+Po0aMIDQ0tsa+hoSE2b96MDz74AF26dEFMTAxsbW0xefJkBAYGorCwEB06dEBGRgaOHj0KlUoFf3//Esd60Utmd+/eRXJysnR30sWLFwFAunuuvBQKBcaPH49Zs2bho48+QkBAAL777jt88MEHmDp1KqysrHD58mVs3boVa9euRc2aNVGrVi2sWbMGdnZ2SE5Oxqefflru/Rfx9fVF27Zt8e2338LGxkZqnzFjBnr16oV69eqhf//+MDAwwJ9//olz585h3rx5AICOHTvi/v37CA8Px/z58wE8DkT9+/eHnZ0dGjduDABISkrCmjVr8Pbbb8Pe3h4XL17EpUuX4OfnV2pdL3LJzMXFBT169MCoUaMQGhqKvLw8jBs3DgMHDoS9vX25xyXSG31PYiIi7fn7+4t33nmnxG2LFy8WdnZ2wtTUVPj4+IiNGzcKAOLevXtCiP8/qbpIXl6e6Nu3r3BxcRGpqamisLBQLF26VDRp0kQYGRmJOnXqCB8fHxEbG6uz41m/fr0AUGyZOXOmxjF36tSp1DFKm+ycnZ0tatasKRYsWCCEEOLvv/8W7777rrC0tBSmpqaiadOmYuLEidLE8qioKOHi4iKUSqVwd3cXMTExAoDYvXv3M/fzpKJJ1U86duyYAKAxqVoIISIiIsSbb74pTE1NhUqlEm+88YZYs2aNRp8WLVoIW1tbaf3OnTtCoVCIgQMHSm0pKSmiT58+ws7OThgbGwtHR0cxY8YMUVBQUGqdL+rOnTvigw8+ENWrVxcqlUoMGzZM3L9/X2f7I9IlhRBC6CuMERGVVadOndC5c2fMmjVL36UQ0SuIgYiIqryMjAw0a9YMFy5c0MszjYjo1cdARERERLLHJ1UTERGR7DEQERERkewxEBEREZHsMRARERGR7DEQERERkewxEBEREZHsMRARERGR7DEQERERkewxEBEREZHsMRARERGR7P0/5ZEmfR3P3/wAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df2_plt = sns.countplot(x = df2[\"label\"])\n",
    "df2_plt.set(xlabel='Fake = 1, Real News = 0', ylabel='Number of Fake/True',title='Count of Fake and True News Dataset #2')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "a958ed36",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[Text(0.5, 0, 'Fake = 1, Real News = 0'),\n",
       " Text(0, 0.5, 'Number of Fake/True'),\n",
       " Text(0.5, 1.0, 'Count of Fake and True News Dataset #3')]"
      ]
     },
     "execution_count": 19,
     "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": [
    "df3_plt = sns.countplot(x = df3[\"label\"])\n",
    "df3_plt.set(xlabel='Fake = 1, Real News = 0', ylabel='Number of Fake/True',title='Count of Fake and True News Dataset #3')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "18998ab1",
   "metadata": {},
   "outputs": [],
   "source": [
    "import nltk\n",
    "import pandas as pd\n",
    "from nltk.corpus import stopwords\n",
    "from nltk.tokenize import word_tokenize\n",
    "import re\n",
    "import string\n",
    "\n",
    "#nltk.download('punkt')\n",
    "#nltk.download('stopwords')\n",
    "\n",
    "def preprocess_text(text):\n",
    "     \n",
    "    # remove all tokens that are not alphabetic\n",
    "    tokens = word_tokenize(text)\n",
    "    words = [word for word in tokens if word.isalpha()]\n",
    "    \n",
    "    # convert to lower case\n",
    "    tokens = [w.lower() for w in tokens]\n",
    "    \n",
    "    # remove punctuation from each word\n",
    "    table = str.maketrans('', '', string.punctuation)\n",
    "    stripped = [w.translate(table) for w in tokens]\n",
    "    \n",
    "    # remove remaining tokens that are not alphabetic\n",
    "    stripped_words = [word for word in stripped if word.isalpha()]\n",
    "    \n",
    "    # filter out stop words\n",
    "    from nltk.corpus import stopwords\n",
    "    stop_words = set(stopwords.words('english'))\n",
    "    filtered_tokens = [w for w in stripped_words if not w in stop_words]\n",
    "\n",
    "    stemmer = PorterStemmer()\n",
    "    stemmed_tokens = [stemmer.stem(word) for word in filtered_tokens]\n",
    "    return stemmed_tokens"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "86845f22",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "8306672f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "   label                                preprocessed_tokens\n",
      "0      1  [comment, expect, barack, obama, member, fukyo...\n",
      "2      1  [demonstr, gather, last, night, exercis, const...\n",
      "3      0  [dozen, polit, activ, pastor, came, privat, di...\n",
      "4      1  [sarmat, missil, dub, satan, replac, fli, mile...\n",
      "5      1  [say, one, time, someon, su, southern, poverti...\n"
     ]
    }
   ],
   "source": [
    "df1['preprocessed_tokens'] = df1['text'].apply(preprocess_text)\n",
    "\n",
    "print(df1[['label', 'preprocessed_tokens']].head())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "02f676e8",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "       label                                preprocessed_tokens\n",
      "64133      1  [alfredo, romero, alfredoromero, octob, romero...\n",
      "64134      1  [monster, mr, presidentelect, democrat, grovel...\n",
      "64135      1  [boy, wear, dad, atm, tshirt, chase, mob, fath...\n",
      "64136      0  [first, airbnb, guest, check, garag, apart, la...\n",
      "64137      0  [moscow, war, eastern, ukrain, simmer, quietli...\n"
     ]
    }
   ],
   "source": [
    "df2['preprocessed_tokens'] = df2['text'].apply(preprocess_text)\n",
    "\n",
    "print(df2[['label', 'preprocessed_tokens']].head())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "bacd07a6",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "       label                                preprocessed_tokens\n",
      "42133      0  [miami, return, honeymoon, dominican, republ, ...\n",
      "42134      0  [ralli, alleg, fake, news, petit, garner, thou...\n",
      "42135      1  [support, us, aper, releas, arriv, mail, publi...\n",
      "42136      1  [pin, zhe, recent, torrent, daili, podesta, em...\n",
      "42137      0  [washington, reuter, leader, senat, intellig, ...\n"
     ]
    }
   ],
   "source": [
    "df3['preprocessed_tokens'] = df3['text'].apply(preprocess_text)\n",
    "\n",
    "print(df3[['label', 'preprocessed_tokens']].head())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "59b17398",
   "metadata": {},
   "outputs": [],
   "source": [
    "from gensim.models import Word2Vec\n",
    "from sklearn.model_selection import train_test_split\n",
    "import seaborn as sns\n",
    "\n",
    "sentences = df1['preprocessed_tokens'].tolist()\n",
    "labels = df1['label'].tolist()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "70ede5ff",
   "metadata": {},
   "outputs": [],
   "source": [
    "sentences_2 = df2['preprocessed_tokens'].tolist()\n",
    "labels_2 = df2['label'].tolist()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "9a523f2f",
   "metadata": {},
   "outputs": [],
   "source": [
    "sentences_3 = df3['preprocessed_tokens'].tolist()\n",
    "labels_3 = df3['label'].tolist()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "a70c80fa",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(9928,)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.10/dist-packages/numpy/core/fromnumeric.py:2009: VisibleDeprecationWarning: Creating an ndarray from ragged nested sequences (which is a list-or-tuple of lists-or-tuples-or ndarrays with different lengths or shapes) is deprecated. If you meant to do this, you must specify 'dtype=object' when creating the ndarray.\n",
      "  result = asarray(a).shape\n"
     ]
    }
   ],
   "source": [
    "print(np.shape(sentences))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "af73adfc",
   "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>Unnamed: 0</th>\n",
       "      <th>title</th>\n",
       "      <th>text</th>\n",
       "      <th>label</th>\n",
       "      <th>preprocessed_tokens</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>LAW ENFORCEMENT ON HIGH ALERT Following Threat...</td>\n",
       "      <td>No comment is expected from Barack Obama Membe...</td>\n",
       "      <td>1</td>\n",
       "      <td>[comment, expect, barack, obama, member, fukyo...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>UNBELIEVABLE! OBAMA’S ATTORNEY GENERAL SAYS MO...</td>\n",
       "      <td>Now, most of the demonstrators gathered last ...</td>\n",
       "      <td>1</td>\n",
       "      <td>[demonstr, gather, last, night, exercis, const...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>Bobby Jindal, raised Hindu, uses story of Chri...</td>\n",
       "      <td>A dozen politically active pastors came here f...</td>\n",
       "      <td>0</td>\n",
       "      <td>[dozen, polit, activ, pastor, came, privat, di...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>SATAN 2: Russia unvelis an image of its terrif...</td>\n",
       "      <td>The RS-28 Sarmat missile, dubbed Satan 2, will...</td>\n",
       "      <td>1</td>\n",
       "      <td>[sarmat, missil, dub, satan, replac, fli, mile...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>5</td>\n",
       "      <td>About Time! Christian Group Sues Amazon and SP...</td>\n",
       "      <td>All we can say on this one is it s about time ...</td>\n",
       "      <td>1</td>\n",
       "      <td>[say, one, time, someon, su, southern, poverti...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Unnamed: 0                                              title  \\\n",
       "0           0  LAW ENFORCEMENT ON HIGH ALERT Following Threat...   \n",
       "2           2  UNBELIEVABLE! OBAMA’S ATTORNEY GENERAL SAYS MO...   \n",
       "3           3  Bobby Jindal, raised Hindu, uses story of Chri...   \n",
       "4           4  SATAN 2: Russia unvelis an image of its terrif...   \n",
       "5           5  About Time! Christian Group Sues Amazon and SP...   \n",
       "\n",
       "                                                text  label  \\\n",
       "0  No comment is expected from Barack Obama Membe...      1   \n",
       "2   Now, most of the demonstrators gathered last ...      1   \n",
       "3  A dozen politically active pastors came here f...      0   \n",
       "4  The RS-28 Sarmat missile, dubbed Satan 2, will...      1   \n",
       "5  All we can say on this one is it s about time ...      1   \n",
       "\n",
       "                                 preprocessed_tokens  \n",
       "0  [comment, expect, barack, obama, member, fukyo...  \n",
       "2  [demonstr, gather, last, night, exercis, const...  \n",
       "3  [dozen, polit, activ, pastor, came, privat, di...  \n",
       "4  [sarmat, missil, dub, satan, replac, fli, mile...  \n",
       "5  [say, one, time, someon, su, southern, poverti...  "
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df1.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "44f17a84",
   "metadata": {},
   "outputs": [],
   "source": [
    "y = labels\n",
    "y_2 = labels_2\n",
    "y_3 = labels_3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "815cc699",
   "metadata": {},
   "outputs": [],
   "source": [
    "X = sentences\n",
    "X_2 =sentences_2\n",
    "X_3 = sentences_3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "016f1d9d",
   "metadata": {},
   "outputs": [],
   "source": [
    "vec_size = 100\n",
    "w2v_model = Word2Vec(sentences = X, vector_size=vec_size, window=5, min_count=1, workers=4)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "2514869e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "88930"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(w2v_model.wv)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "d23e74c5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "plt.hist([len(x) for x in X], bins = 1000)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "7af152db",
   "metadata": {},
   "outputs": [],
   "source": [
    "tokenizer = Tokenizer()\n",
    "tokenizer.fit_on_texts(X)\n",
    "X = tokenizer.texts_to_sequences(X)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "14036c99",
   "metadata": {},
   "outputs": [],
   "source": [
    "tokenizer_2 = Tokenizer()\n",
    "tokenizer_2.fit_on_texts(X_2)\n",
    "X_2 = tokenizer_2.texts_to_sequences(X_2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "d9ff3ed2",
   "metadata": {},
   "outputs": [],
   "source": [
    "tokenizer_3 = Tokenizer()\n",
    "tokenizer_3.fit_on_texts(X_3)\n",
    "X_3 = tokenizer_2.texts_to_sequences(X_3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "73ab4aee",
   "metadata": {},
   "outputs": [],
   "source": [
    "import seaborn as sns\n",
    "maxlen = 1200\n",
    "X = pad_sequences(X, maxlen = maxlen)\n",
    "X_2 = pad_sequences(X_2, maxlen = maxlen)\n",
    "X_3 = pad_sequences(X_3, maxlen = maxlen)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "1d60b413",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "nos = np.array([len(x) for x in X])\n",
    "len(nos[nos>maxlen])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "c8652d7f",
   "metadata": {},
   "outputs": [],
   "source": [
    "from tensorflow.keras.layers import Input, LSTM, Embedding, RepeatVector\n",
    "\n",
    "word_embeddings = {word: w2v_model.wv[word] for word in tokenizer.word_index.keys() if word in w2v_model.wv}\n",
    "embedding_matrix = np.zeros((len(tokenizer.word_index) + 1, 100))\n",
    "for word, embedding in word_embeddings.items():\n",
    "    embedding_matrix[tokenizer.word_index[word]] = embedding\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "2f64ee49",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(88931, 100)"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "embedding_matrix.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "59947e2d",
   "metadata": {},
   "outputs": [],
   "source": [
    "from tensorflow.keras.models import Sequential\n",
    "from tensorflow.keras.layers import Dense, Dropout, Embedding, GRU, LSTM, RNN, SpatialDropout1D\n",
    "\n",
    "vocab_size = len(tokenizer.word_index) + 1\n",
    "vocab = tokenizer.word_index\n",
    "\n",
    "EMB = Embedding(vocab_size,output_dim = vec_size,weights=[embedding_matrix],input_length = maxlen, trainable = False)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "9a55967f",
   "metadata": {},
   "outputs": [],
   "source": [
    "model = Sequential()\n",
    "model.add(EMB)\n",
    "model.add(LSTM(units=128))# Adjust input_shape based on your data\n",
    "model.add(Dense(1, activation='sigmoid'))\n",
    "model.compile(optimizer='adam', loss='binary_crossentropy', metrics=['accuracy'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "cfa6eae8",
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train, X_test, y_train, y_test = train_test_split(X, np.array(y))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "6a6749a5",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/5\n",
      "163/163 [==============================] - 114s 698ms/step - loss: 0.3785 - accuracy: 0.8310 - val_loss: 0.2935 - val_accuracy: 0.8948\n",
      "Epoch 2/5\n",
      "163/163 [==============================] - 100s 614ms/step - loss: 0.3721 - accuracy: 0.8206 - val_loss: 0.3111 - val_accuracy: 0.8706\n",
      "Epoch 3/5\n",
      "163/163 [==============================] - 94s 580ms/step - loss: 0.3072 - accuracy: 0.8682 - val_loss: 0.2405 - val_accuracy: 0.9029\n",
      "Epoch 4/5\n",
      "163/163 [==============================] - 96s 588ms/step - loss: 0.2564 - accuracy: 0.8927 - val_loss: 0.2095 - val_accuracy: 0.9105\n",
      "Epoch 5/5\n",
      "163/163 [==============================] - 94s 579ms/step - loss: 0.1782 - accuracy: 0.9313 - val_loss: 0.1652 - val_accuracy: 0.9333\n"
     ]
    }
   ],
   "source": [
    "history_LSTM = model.fit(X_train, y_train, epochs=5, validation_split = 0.3).history #This is LSTM model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "id": "84a1c73d",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "78/78 [==============================] - 16s 204ms/step\n"
     ]
    }
   ],
   "source": [
    "y_pred_LSTM = (model.predict(X_test) >= 0.5).astype(int)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "id": "e7e22462",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "dict_keys(['loss', 'accuracy', 'val_loss', 'val_accuracy'])\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "print(history_LSTM.keys())\n",
    "#  \"Accuracy\"\n",
    "plt.plot(history_LSTM['accuracy'])\n",
    "plt.plot(history_LSTM['val_accuracy'])\n",
    "plt.title('LSTM model accuracy')\n",
    "plt.ylabel('accuracy')\n",
    "plt.xlabel('epoch')\n",
    "plt.legend(['train', 'validation'], loc='upper left')\n",
    "plt.show()\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "id": "eea8a3b7",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x7ff8e05c9c90>"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA0MAAADFCAYAAACWwSm3AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjYuMCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy89olMNAAAACXBIWXMAAA9hAAAPYQGoP6dpAABH+klEQVR4nO3deXxTVd4/8E+SNknTpPteCt3ZBKpAO7SOoCwFHQQHR/RhpO6PCiiDCzCOIm4F4XFQwOXHvAR1xgdEBedRZCuLSlEYoVD2ttYW7F7o3qZNcn9/3CxNm7ZJaZsun/frdV60ycntyZ07Nd+ecz5XIgiCACIiIiIiogFG6uwBEBEREREROQOLISIiIiIiGpBYDBERERER0YDEYoiIiIiIiAYkFkNERERERDQgsRgiIiIiIqIBicUQERERERENSC7OHkBXMBgMKCgogEajgUQicfZwiIiIiIjISQRBQHV1NUJCQiCVtj/30y+KoYKCAoSFhTl7GERERERE1EtcvnwZgwYNardPvyiGNBoNAPENe3h4OHk0RERERETkLFVVVQgLCzPXCO3pF8WQaWmch4cHiyEiIiIiIrJr+wwDFIiIiIiIaEBiMURERERERAMSiyEiIiIiIhqQ+sWeoV7l9GfA8X8AMnmz5mr9tYuixeOugKzlY231bd6/5WNysb/UBWDEOBEREfUCer0eTU1Nzh4G9TNyubzD2Gx7sBjqahX5wOWfnD0KG4VTy+LLWDjZKqhsFXKO9JXJAZe2isFmrQsuYCIiIuqdBEFAUVERKioqnD0U6oekUikiIiIgl8uv6zgshrraiNmA/1BA3wjom4z/NgK6xtaPWbWO+jYBem2Lx5sAnfExCNbjML2uN5PI7CycrmO2rN2+dhSInGUjIiLqFFMhFBAQAJVKZVeyF5E9DAYDCgoKUFhYiMGDB1/XtcViqKv5RYutJwkCYNB3UDg5UmS1KLSaF1/t9u3guAZdi3HrAV292HozmbydQqsrZ8scnFmzKto4y0ZERL2HXq83F0K+vr7OHg71Q/7+/igoKIBOp4Orq2unj8NiqD+QSACZi9igcvZo2mYwAIa2CqdGY/HUUUHWvNC6jr5WxWGLMbTUF2bZpC52Fk52zKy5KAD/YcCQRMCz/bs2ExER2WLaI6RS9eLPJdSnmZbH6fV6FkPUR0ilgFQhftjurQRBnMFqVTg1L6jaKd5a9W1raWQn+jYv4AS99bgNOrF19f5Ur8HAkCRg8ATxX98oLhskIiK7cWkcdZeuurZYDBE1J5EYZ0lcAbg7ezRtM+jtL5wcKbL0jUBjLVBwEig8JQaCVOQDp/5X/LnuAeKM0ZAk8d+AEVyiR0RERH0WiyGivkgqE5ursvt+hrYauHwMyD8K5KUDV/4D1JYA53aKDQCUnsZZI2OBFDzGWEgSERER9X4shojINoUGiJ4sNgBoagAKTgB5R8Ti6PIxoKESuLRbbADgqgLC4i1L6waNA1zdnPceiIiIetikSZMQFxeHdevWAQDCw8OxePFiLF68uM3XSCQS7NixA7Nnz76un91VxxlIWAwRkX1clcYZoETxe70OKDotFkZ56UB+OlB/DfjlkNgAQOoKhI61zByFxQNKD2e9AyIiojbNnDkTTU1N2L17d6vnvv/+e9xyyy04deoURo8e7dBxjx8/Dnf3rl16//LLL2Pnzp3IyMiwerywsBDe3t5d+rNa2rJlCxYvXtxv7h/FYoiIOkfmAoTeJLbEhWJaYOkFsSjKSwd+PQLUFAGXfxTbD28BEikQNMqy52jwBMDdz9nvhIiICA8//DDmzJmDK1euYNAg6zTVzZs3Y9y4cQ4XQoAYAd1TgoKCeuxn9Rfc+UxEXUMqBQJHAOMfAe7+EHjmAvDUSWDWRiBuHuAdDggGMZjhx3eBbX8G1kQBGxOA/1sMnN4OVP7m7HdBRETdQBAE1DXqnNIEQeh4gAD+8Ic/wN/fH1u2bLF6vKamBtu3b8fDDz+M8vJy3HfffQgNDYVKpcKoUaPwv//7v+0eNzw83LxkDgCysrJwyy23QKlUYsSIEdi3b1+r1yxduhSxsbFQqVSIjIzEiy++aI4r37JlC1auXIlTp05BIpFAIpGYxyyRSLBz507zcTIzM3HbbbfBzc0Nvr6+eOyxx1BTU2N+/oEHHsDs2bOxdu1aBAcHw9fXFwsWLDD/rM7Iz8/HrFmzoFar4eHhgXvuuQfFxcXm50+dOoVbb70VGo0GHh4eGDt2LP7zn/8AAPLy8jBz5kx4e3vD3d0dI0eOxK5duzo9FntwZoiIuodEAvhEiu3GP4uPVRVYltXlpQOl58XZpNILwM+bxT5eQywzR0MSxdczmpWIqE+rb9JjxEt7nPKzz72SDJW844+8Li4umD9/PrZs2YIXXnjBHN28fft26PV63HfffaipqcHYsWOxdOlSeHh44JtvvsH999+PqKgoxMfHd/gzDAYD/vjHPyIwMBA//fQTKisrbe4l0mg02LJlC0JCQpCZmYlHH30UGo0Gzz//PObOnYszZ85g9+7d2L9/PwDA09Oz1TFqa2uRnJyMCRMm4Pjx4ygpKcEjjzyChQsXWhV8Bw8eRHBwMA4ePIjs7GzMnTsXcXFxePTRRzt8P7ben6kQOnz4MHQ6HRYsWIC5c+fi0KFDAIB58+bhxhtvxHvvvQeZTIaMjAzzfYIWLFiAxsZGfPfdd3B3d8e5c+egVqsdHocjWAwRUc/xCAFG3S02AKgtF9Pq8o+KwQyFp4CKPLGd+lTsow60jvP2H844byIi6hYPPfQQ1qxZg8OHD2PSpEkAxCVyc+bMgaenJzw9PfHss8+a+y9atAh79uzBZ599ZlcxtH//fly4cAF79uxBSEgIAOCNN97AjBkzrPr97W9/M38dHh6OZ599Flu3bsXzzz8PNzc3qNVquLi4tLss7tNPP0VDQwM+/vhj856lDRs2YObMmVi9ejUCAwMBAN7e3tiwYQNkMhmGDRuGO+64A2lpaZ0qhtLS0pCZmYnc3FyEhYUBAD7++GOMHDkSx48fx/jx45Gfn4/nnnsOw4YNAwDExMSYX5+fn485c+Zg1KhRAIDIyEiHx+AoFkNE5DzuvsDwP4gNABqqgCvHjDNHR4Hf/gPUFANnd4gNAJRelv1GQ5KA4NGM8yYi6uXcXGU490qy0362vYYNG4bExER8+OGHmDRpErKzs/H999/jlVdeAQDo9Xq88cYb+Oyzz/Dbb7+hsbERWq0WKpXKruOfP38eYWFh5kIIACZMmNCq37Zt2/DOO+8gJycHNTU10Ol08PBwLIDo/PnzGDNmjFV4Q1JSEgwGAy5evGguhkaOHAmZzHKOgoODkZmZ6dDPav4zw8LCzIUQAIwYMQJeXl44f/48xo8fjyVLluCRRx7BJ598gilTpuBPf/oToqKiAABPPfUUnnjiCezduxdTpkzBnDlzOrVPyxH88yoR9R5KDyB6CjD5JeChb4Fll4EHdgG3/g2IvBVwdQcaKoCLu4B9LwL/uA1YNQT4eDZweI0Y2tDU4Ox3QURELUgkEqjkLk5pEgeXWj/88MP44osvUF1djc2bNyMqKgoTJ04EAKxZswZvv/02li5dioMHDyIjIwPJyclobGzssnN19OhRzJs3D7fffju+/vprnDx5Ei+88EKX/ozmTEvUTCQSCQwGQ7f8LEBMwjt79izuuOMOHDhwACNGjMCOHeIfPB955BH88ssvuP/++5GZmYlx48Zh/fr13TYWoJPF0MaNGxEeHg6lUomEhAQcO3aszb5ffvklxo0bBy8vL7i7uyMuLg6ffPKJVZ8HHnjAvAHM1KZPn96ZoRFRf+KqBMKTgInPAfN3AsvygEcPANNeA4beLs4SNdUCvxwEDr4GbLkdWBUGfDgdSHsFyN4v3jyWiIjITvfccw+kUik+/fRTfPzxx3jooYfMBdWRI0cwa9Ys/PnPf8aYMWMQGRmJS5cu2X3s4cOH4/LlyygsLDQ/9uOPP1r1SU9Px5AhQ/DCCy9g3LhxiImJQV5enlUfuVwOvV7f4c86deoUamtrzY8dOXIEUqkUQ4cOtXvMjjC9v8uXL5sfO3fuHCoqKjBixAjzY7GxsfjLX/6CvXv34o9//CM2b95sfi4sLAyPP/44vvzySzzzzDPYtGlTt4zVxOFlctu2bcOSJUvw/vvvIyEhAevWrUNycjIuXryIgICAVv19fHzwwgsvYNiwYZDL5fj666/x4IMPIiAgAMnJlunS6dOnW50IhULRybdERP2WzHjfotCxQOIiY5z3+WahDEfEZXWmfUjf/48Y5x08Bhic2CzO29fZ74SIiHoptVqNuXPnYvny5aiqqsIDDzxgfi4mJgaff/450tPT4e3tjbfeegvFxcVWH/TbM2XKFMTGxiIlJQVr1qxBVVUVXnjhBas+MTExyM/Px9atWzF+/Hh888035pkTk/DwcOTm5iIjIwODBg2CRqNp9dl53rx5WLFiBVJSUvDyyy+jtLQUixYtwv33329eItdZer2+1T2OFAoFpkyZglGjRmHevHlYt24ddDodnnzySUycOBHjxo1DfX09nnvuOdx9992IiIjAlStXcPz4ccyZMwcAsHjxYsyYMQOxsbG4du0aDh48iOHDh1/XWDvi8MzQW2+9hUcffRQPPvggRowYgffffx8qlQoffvihzf6TJk3CXXfdheHDhyMqKgpPP/00Ro8ejR9++MGqn0KhQFBQkLl19w2jiKgfkEqBwJFA/KPAnzYDz1wEFp0A7twAjPkvMZlOMAAFJ4EfNwLb5gFrIoGNvwO+XgJkfi4m3BERETXz8MMP49q1a0hOTrba3/O3v/0NN910E5KTkzFp0iQEBQVh9uzZdh9XKpVix44dqK+vR3x8PB555BG8/vrrVn3uvPNO/OUvf8HChQsRFxeH9PR0vPjii1Z95syZg+nTp+PWW2+Fv7+/zXhvlUqFPXv24OrVqxg/fjzuvvtuTJ48GRs2bHDsZNhQU1ODG2+80arNnDkTEokEX331Fby9vXHLLbdgypQpiIyMxLZt2wAAMpkM5eXlmD9/PmJjY3HPPfdgxowZWLlyJQCxyFqwYAGGDx+O6dOnIzY2Fu++++51j7c9EsHe8HUAjY2NUKlU+Pzzz63+h09JSUFFRQW++uqrdl8vCAIOHDiAO++8Ezt37sTUqVMBiMvkdu7cCblcDm9vb9x222147bXX4Otr+6+3Wq0WWq3W/H1VVRXCwsJQWVnp8OYyIurnKn+zpNXlpYsx3i15h1vHeXtHMM6biOg6NDQ0IDc3FxEREVAqlc4eDvVD7V1jVVVV8PT0tKs2cGiZXFlZGfR6fauptcDAQFy4YOMDhlFlZSVCQ0Oh1Wohk8nw7rvvmgshQFwi98c//hERERHIycnBX//6V8yYMQNHjx61SrcwSU1NNVeQRETt8gxtEeddZiyOjAVS0Wng2q9iy/iX2EcTbJ1Y5z+Mcd5ERET9UI9Ea2s0GmRkZKCmpgZpaWlYsmQJIiMjzfnt9957r7nvqFGjMHr0aERFReHQoUOYPHlyq+MtX74cS5YsMX9vmhkiIuqQux8wfKbYADHO+/Ixy8zRbz8D1YXAmS/EBgBu3pY9R0MmAEFjABnvTEBERNTXOfRfcz8/P8hkMhQXF1s9Xlxc3O5Nn6RSKaKjowEAcXFxOH/+PFJTU83FUEuRkZHw8/NDdna2zWJIoVAwYIGIuobSA4iZIjYAaKoXCyJTIMPlY0D9NeDiN2IDALkaCIu33Aw25CYx+Y6IiIj6FIeKIblcjrFjxyItLc28Z8hgMCAtLQ0LFy60+zgGg8Fqz09LV65cQXl5OYKDgx0ZHhHR9XN1A8JvFhsA6JuAwlOWxLr8dKChEsg5IDYAkCnEhDvTnqOweEChcd57ICIiIrs4vM5jyZIlSElJwbhx4xAfH49169ahtrYWDz74IABg/vz5CA0NRWpqKgBxf8+4ceMQFRUFrVaLXbt24ZNPPsF7770HQEyjWLlyJebMmYOgoCDk5OTg+eefR3R0tFX0NhGRU8hcgUHjxJb0lBjnXXLOMnOUlw7UlohFUn468D0AiUyM8zbNHA3+HaDycfY7ISIiohYcLobmzp2L0tJSvPTSSygqKkJcXBx2795tDlXIz8+HtNlG49raWjz55JO4cuUK3NzcMGzYMPzzn//E3LlzAYgRe6dPn8ZHH32EiooKhISEYNq0aXj11Ve5FI6Ieh+pFAi6QWwJjwGCAFz9xVIY5R0BKvKBghNiO2qMMA0YYZk5GpwIeHDmm4iIyNkcitburRyJzyMi6naVVyxpdXnpQNnF1n18IpuFMiSK8d6M8yaifoLR2tTdnBKtTUREdvAcBIz+k9gAoKbUGOdtXEpXlCnOJl39Bcj4p9hHE2IpjIYkAn5DGedNRETUzVgMERF1N7U/MOJOsQFiAINVnPcJoLoAOPO52ADAzce6OAocxThvIqI+KDw8HIsXL8bixYvt6n/o0CHceuutuHbtGry8vLp1bMRiiIio5yk9gZipYgOAxjobcd5XgQtfiw0A5BrrOO/QmwAX7qskIuoqkg6WKq9YsQIvv/yyw8c9fvw43N3d7e6fmJiIwsJCeHp6OvyzHMGiS8RiiIjI2eQqIOL3YgMAXaMxzts4c5T/I6CtBHLSxAaIcd6DxltmjgaNBxRq570HIqI+rrCw0Pz1tm3b8NJLL+HiRcueT7Xa8jtWEATo9Xq4uHT8Udrf39+hccjl8nbv30ldiwvSiYh6Gxc5EDYeuHkxMO8zYGku8PgPwIw3gRGzAHd/QK8F8n4AvnsT+GQ2sGowsOk2YO/fgIvfijeKJSIiuwUFBZmbp6cnJBKJ+fsLFy5Ao9Hg22+/xdixY6FQKPDDDz8gJycHs2bNQmBgINRqNcaPH4/9+/dbHTc8PBzr1q0zfy+RSPCPf/wDd911F1QqFWJiYvDvf//b/PyhQ4cgkUhQUVEBANiyZQu8vLywZ88eDB8+HGq1GtOnT7cq3nQ6HZ566il4eXnB19cXS5cuRUpKivm+oJ1x7do1zJ8/H97e3lCpVJgxYwaysrLMz+fl5WHmzJnw9vaGu7s7Ro4ciV27dplfO2/ePPj7+8PNzQ0xMTHYvHlzp8fSnTgzRETU20llQNAosSX8txjnXZ7TLM47HajMF5fa/fYzkL4egAQIHAkMnmCZPdLwL41E5CSCADTVOednu6q6LK1z2bJlWLt2LSIjI+Ht7Y3Lly/j9ttvx+uvvw6FQoGPP/4YM2fOxMWLFzF48OA2j7Ny5Uq8+eabWLNmDdavX4958+YhLy8PPj6270lXV1eHtWvX4pNPPoFUKsWf//xnPPvss/jXv/4FAFi9ejX+9a9/YfPmzRg+fDjefvtt7Ny5E7feemun3+sDDzyArKws/Pvf/4aHhweWLl2K22+/HefOnYOrqysWLFiAxsZGfPfdd3B3d8e5c+fMs2cvvvgizp07h2+//RZ+fn7Izs5GfX19p8fSnVgMERH1NRIJ4BcttrEp4mMV+WKcd76xOCq7BBSfEdvxTWIfnyjrUAavIYzzJqKe0VQHvBHinJ/91wJAbv+enfa88sormDp1qvl7Hx8fjBkzxvz9q6++ih07duDf//43Fi5c2OZxHnjgAdx3330AgDfeeAPvvPMOjh07hunTp9vs39TUhPfffx9RUVEAgIULF+KVV14xP79+/XosX74cd911FwBgw4YN5lmazjAVQUeOHEFiYiIA4F//+hfCwsKwc+dO/OlPf0J+fj7mzJmDUaNGAQAiIyPNr8/Pz8eNN96IcePGARBnx3orFkNERP2B12CxjRFvaI2aEkucd94RoOgMcDVHbCc/Eft4hDYrjpIAv1gWR0RE7TB9uDepqanByy+/jG+++QaFhYXQ6XSor69Hfn5+u8cZPXq0+Wt3d3d4eHigpKSkzf4qlcpcCAFAcHCwuX9lZSWKi4sRHx9vfl4mk2Hs2LEwGAwOvT+T8+fPw8XFBQkJCebHfH19MXToUJw/fx4A8NRTT+GJJ57A3r17MWXKFMyZM8f8vp544gnMmTMHJ06cwLRp0zB79mxzUdXbsBgiIuqP1AHi/qIRs8Tv6yus47wLTgBVvwGZ28UGACpf47K6JLFACholLtEjIrperipxhsZZP7uLtEyFe/bZZ7Fv3z6sXbsW0dHRcHNzw913343Gxsb2h+TqavW9RCJpt3Cx1V8QBAdH37UeeeQRJCcn45tvvsHevXuRmpqK//mf/8GiRYswY8YM5OXlYdeuXdi3bx8mT56MBQsWYO3atU4dsy0shoiIBgI3LyB2mtgAMc77ynHj7NER4PJxoK7cOs5b4QGEJVhmj0JuZJw3EXWORNJlS9V6kyNHjuCBBx4wL0+rqanBr7/+2qNj8PT0RGBgII4fP45bbrkFAKDX63HixAnExcV16pjDhw+HTqfDTz/9ZJ7RKS8vx8WLFzFixAhzv7CwMDz++ON4/PHHsXz5cmzatAmLFi0CIKbopaSkICUlBb///e/x3HPPsRgiIqJeQq4CIieKDTDGeWcYZ46OWuK8s/eJDQBclK3jvPvhhxsiInvFxMTgyy+/xMyZMyGRSPDiiy92emna9Vi0aBFSU1MRHR2NYcOGYf369bh27VqH904CgMzMTGg0GvP3EokEY8aMwaxZs/Doo4/igw8+gEajwbJlyxAaGopZs8QVB4sXL8aMGTMQGxuLa9eu4eDBgxg+fDgA4KWXXsLYsWMxcuRIaLVafP311+bnehsWQ0REZIzzjhfbzX8BDHqg+Kxlz1FeOlBXBvz6vdgAQOoCBMdZ9hwNTgDcvJ36NoiIetJbb72Fhx56CImJifDz88PSpUtRVVXV4+NYunQpioqKMH/+fMhkMjz22GNITk6GTNbxUmfTbJKJTCaDTqfD5s2b8fTTT+MPf/gDGhsbccstt2DXrl3mJXt6vR4LFizAlStX4OHhgenTp+Pvf/87APFeScuXL8evv/4KNzc3/P73v8fWrVu7/o13AYng7AWHXaCqqgqenp6orKyEh4eHs4dDRNT/CAJQliUWRvlHgV+PAFVXWnSSAIE3GIujCcDgREAT6JThEpFzNTQ0IDc3FxEREVAqlc4ezoBjMBgwfPhw3HPPPXj11VedPZxu0d415khtwJkhIiLqmEQC+MeKbdyD4mMV+c1mjo4C5VlAcabYjn0g9vGNtswcDUkUE++IiKhL5eXlYe/evZg4cSK0Wi02bNiA3Nxc/Nd//Zezh9brsRgiIqLOMcd53yt+X1NiuQlsXrp4j6PybLGd+Fjs4zGoRZx3DOO8iYiuk1QqxZYtW/Dss89CEATccMMN2L9/f6/dp9ObsBgiIqKuoQ4ARs4WG2CM8/6pWZz3SXFpXeZnYgMAlV+zmaMJ4jI7xnkTETkkLCwMR44ccfYw+iQWQ0RE1D3cvIDYZLEBQGOtGOdtmjm6clwMZTj/b7EBYpz34N9ZCqTgODHcgYiIqBuwGCIagBqa9Kisb0JVfRMqW7SaBh0i/N2RGOUHH3d+CKUuJHcHIieJDQB0WqAgwzJzlP8joK0CsvaKDQBc3IBB4yx7jgaNF2PBiYiIugCLIaI+SBAE1DbqbRYzVW0UOVUNOvPXjTr77oEwItgDN8f4ITHKF/ERPlDJ+SuDupCLQozjHpwA/H6JMc77TIs47/LWcd4hN4lL6oYkiTeFdfNy6tsgorY54547NDB0VSA2o7WJnMRgEFCt1bUqXGwVMqZipqpZH53h+v6vK5UAHm6u8GzWPJSuULrKcOa3Slwsrrbq7yqT4MbB3kiK8kNStC/GhHnBVSa9rjEQtUsQgLJLzUIZjgBVv7XoJAGCbrDMHA1OBNT+ThkuEVkYDAZkZWVBJpPB398fcrncrhuAEtlDEASUlpairq4OMTExre6n5EhtwGKI6Dro9AarGZeWxYzV1w3Nipu6JlRrdbje//e5yiRiEdOsmPFsWeC4uVj1MTV3uQuk0rb/w1RS3YCjOeU4kl2GI9nl+K2i3up5d7kMCZG+SIzyxc0xfhgaqOF/6Kh7CUKLOO904GpO635eQ4Dg0UCQqY0CPEKYWkfUwxobG1FYWIi6ujpnD4X6IYlEgkGDBkGtVrd6jsUQkQO0Ouv9M1X1OtszMza+rm3UX/fPV7pKbRYyHi3+tVXguLnKeqQAEQQBeeV1OJJThvTscqTnlOFaXZNVHz+1HBOi/HBztC8So/wQ5sN9HdQDqouB/OZx3mcB2PjPmspXLIpMBVLwaPEeSEyuI+pWgiBAp9NBr7/+/14SNefq6tpqRsik24uhjRs3Ys2aNSgqKsKYMWOwfv16xMfH2+z75Zdf4o033kB2djaampoQExODZ555Bvfff7+5jyAIWLFiBTZt2oSKigokJSXhvffeQ0xMjF3jYTE0sAmCgHpjIEBbxUzb+2ia0NB0/euZ1QoXeChdWi87s1nIuMLTzdJX4dL3PowZDALOFVYhPacMP2SX43juVdQ3Wf+HbrCPCknR4pI6hjFQj6m/BhRliq3wtPhv6QVAsPFBzMUNCBwpFkmmmaSAEQxoICLq47q1GNq2bRvmz5+P999/HwkJCVi3bh22b9+OixcvIiAgoFX/Q4cO4dq1axg2bBjkcjm+/vprPPPMM/jmm2+QnCzGra5evRqpqan46KOPEBERgRdffBGZmZk4d+4clEpll75h6p0EQdw/U1lnWU5mPRPTRoFj7Nukv74JTokE0Chc4KlqvYemZVHTep+NC1wG+N4ZrU6PjPwKcUldTjkyLldA32JP04hgDyRF+yIp2o9hDNSzmhqA0vPG4shYIBWdAZpqW/eVSAHfGGNxNMrYxgDuvj0/biIi6pRuLYYSEhIwfvx4bNiwAYC4QS4sLAyLFi3CsmXL7DrGTTfdhDvuuAOvvvoqBEFASEgInnnmGTz77LMAgMrKSgQGBmLLli249957Ozwei6HeQW8QrAqUyjaKGVv7aKrqm3CdeQCQSSVWBYuH0qXVrIzNYsbNFRpF+/tnyDHVDU04lnsVR7LFPUc2wxjCvM0zRwxjoB5n0ANXc43F0WnLTFJtie3+mpBmBZLxX+9w7kMiIuqFuq0YamxshEqlwueff47Zs2ebH09JSUFFRQW++uqrdl8vCAIOHDiAO++8Ezt37sTUqVPxyy+/ICoqCidPnkRcXJy578SJExEXF4e333671XG0Wi20Wq35+6qqKoSFhbEY6gKNOkOrYqatJWbNi5yqejEQ4HrJXaRWMy62Cpm2lp65y3tm/ww5rrRai/ScMrvCGJKi/TAsiGEM5CTVxZYCybTMzlZIAwAoPMUkO1NxFDwa8BvKm8QSETmZI8WQQ+tUysrKoNfrERgYaPV4YGAgLly40ObrKisrERoaCq1WC5lMhnfffRdTp04FABQVFZmP0fKYpudaSk1NxcqVKx0Z+oAhCAK0OoN10VLX1mxN6+VnLfd9dIZKLrMKA7BeYma7wDF9rXTte/tnqGP+GgVmxYViVlwoBEFA/tU6/JBtHcZw4EIJDlwQ/yrPMAZyGk0goJkKxEy1PKatFoMZzMvsTgMl5wFtpTHV7oilr0wO+A+zhDQEjQICbwCU/EMdEVFv1COL9jUaDTIyMlBTU4O0tDQsWbIEkZGRmDRpUqeOt3z5cixZssT8vWlmqL8w3VDTVMi0XE7W9j1pxPvQNOqvPxBAo7AVBuBis4ixKnaUrpC7cLkTtU0ikWCIrzuG+LpjXsIQqzCGI9nlOJZ7FWU1jfi/UwX4v1MFABjGQE6m0ACDfyc2E30TUHrRuP/otOXfhkpLwZTR7BjeEc2W2Y0R/9UEcZkdEZGTOVQM+fn5QSaTobi42Orx4uJiBAUFtfk6qVSK6OhoAEBcXBzOnz+P1NRUTJo0yfy64uJiBAcHWx2z+bK55hQKBRQKhSND73EGg4DqZvefaWtmxvY+Gl2rzeeOauuGmh0VOJ5urtAoXSHj/hnqIVKpBDeEeuKGUE88dksUGnUGnMy/ZhXGkH+1DvnH8vG/x/IBWMIYEqP9EB/uA3cFwxioh8lcjUvkbgBwn/iY6T5IzfcgFWUCVVeAa7liO9dsObm7v/UepKDRgG8U476JiHqQQ58g5HI5xo4di7S0NPOeIYPBgLS0NCxcuNDu4xgMBvOen4iICAQFBSEtLc1c/FRVVeGnn37CE0884cjweoVN3/2C9QeyuvyGmrZuptnWDTU93Fyh7uCGmkS9ldxFioRIXyRE+mIJrMMY0nPKcKGoGucKq3CusAqbvs81hzEkRvvi5mg/hjGQ80gkgPcQsQ2faXm8thwoblYcFZ0Gyi4BtaVAzgGxmbiqjHHfzfYhBYwAXN16/v0QEQ0ADv85dcmSJUhJScG4ceMQHx+PdevWoba2Fg8++CAAYP78+QgNDUVqaioAcX/PuHHjEBUVBa1Wi127duGTTz7Be++9B0BcMrN48WK89tpriImJMUdrh4SEWIU09CVVDZYgAaWr1GYh03YQgGWmpqduqEnUm2mUrpg8PBCTh4v7Ck1hDOnZ5fghuwy/VdTj2K9XcezXq1i3PwvuchniI3yMy+r8MDRQwz8MkHO5+wKRk8Rm0lQPFJ+zXmJXfBZoqgOuHBebiUQG+MU2ux+ScRZJ5dPT74SIqN9xuBiaO3cuSktL8dJLL6GoqAhxcXHYvXu3OQAhPz8fUqnlr7K1tbV48sknceXKFbi5uWHYsGH45z//iblz55r7PP/886itrcVjjz2GiooK3Hzzzdi9e7dd9xjqbe66KRS3DvPv0zfUJOrNbIUxmCK8TWEMBy+W4uDFUgCWMIYkY1IdwxioV3B1AwaNFZuJQQ+U57SO+64rE++TVHoeyPzM0t9jUOu4b6/B3IdEROQAh+8z1BvxPkNEBIh79c4XVZkjvI/lXm2VkCiGMYiF0YRIX/iqe/f+QxrgBAGoLmod930t13Z/paexMGoe9x0r7nEiIhoguvWmq70RiyEissUcxpAjzhxlXK5oFU4yPNhDjPBmGAP1JQ2VzeK+M4GiU0DJBcDQ1LqvTAEEDLfMIAWPFvclKTQ9P24ioh7AYoiIyIYarQ7HcsvxQ5YljKE5hjFQn6ZrBEovtIj7zgS0VTY6SwCfyGb7kIxNE2ijLxFR38JiiIjIDrbCGJpjGAP1eQYDUJHXOu67usB2f/eAFvuQRotFk5R/FCCivoPFEBGRg9oKY2jO112OCVHirBHDGKhPqy2z3oNUlAmUZwGCjZt2u7ob76nUbB+S/3DAte+FHBHRwMBiiIjoOtkTxhDm44abo/2QGOWHxCiGMVAf11gLlJwHCk9Zx33rGlr3lboAfkNbxH2PAty8e37cREQtsBgiIupizcMY0rPLcLKNMIakKF8kxTCMgfoJvQ4oz7aENJiW2tVftd3fc3DruG/PQYz7JqIexWKIiKibmcIYTMvqWoYxuEgluHGwl3m/URzDGKi/EASgqqDZPiRjkVSRZ7u/m7f1HqSgUca4b/6xgIi6B4shIqIeVlajRXpOOY5kleFIThmuXLMOY1DJZUgwhjEkRvlhWBDDGKifqa8Ais8024d0Wky3M+ha95UpgMARzfYhjRHjvuXuPT5sIup/WAwRETlZfnkdfsgWC6OjOeW4Wtto9bwpjCEp2g83M4yB+iudVtyH1DLuu7HGRmcJ4BvdYh/SGEDt3+PDJqK+jcUQEVEvYgpjSM8ux5GcMvz0C8MYaAAzGIBrua3jvmuKbPdXB7Xeh+QdwbhvImoTiyEiol6sUWdAxuUKY1JdGTIuV0DXVhhDtB/iIxjGQANATUmLuO/TQHkOABsfU+QaG3HfwwAX/hGBiFgMOXs4REQOsTeMITHKDzfHMIyBBhBtDVByrkXc9zlAr23dV+oi3v/IFPMdPBoIvAFw8+rxYRORc7EYIiLqw0xhDOnZZfgh23YYQ3yEj3lZHcMYaEDR64CyS832IRlnkxoqbPf3GmIJaTAttfMIYdw3UT/GYoiIqB/JL6/DkRyxMGIYA5ENggBUXmm9D6ky33Z/la9lBinIWCT5xQBSWc+Om4i6BYshIqJ+ymAQcKGoWtxv1E4YQ1KUnzHGm2EMNIDVXbUR930REPSt+7q4tY77DhgByPnHBaK+hsUQEdEAYU8Yw7AgDW423vyVYQw04DU1iPuQrOK+zwBNta37SqSAb0zruG93354fNxHZjcUQEdEAVaPV4XjuVfEeRx2EMSRFi2EMcheGMdAAZ9ADV3OBolPNltmdBmpLbffXhDQrjoz7kLzDuQ+JqJdgMURERAAYxkB0XaqLLDNIpqV2V3Ns91V4WBdHQaOMcd/ynh0zEbEYIiIi20xhDEeyy5BuI4zBx12ORGMYQ1KUHwb7cr8EkRVtNVB81jJ7VHQaKDkP6Btb95W6AgHDLCENprhvJT+rEHUnFkNERNShlmEMx3Kvoq7RdhhDojGMwY9hDESt6ZvEYIbm+5AKTwPaStv9vSOa7UMyNk0Ql9kRdREWQ0RE5LDmYQzpOWU4mW87jMEU4c0wBqJ2CAJQkd867rvqiu3+Kj+xOBo0HoiZBoTcBEi5n4+oM1gMERHRdTOFMRwx7jeyFcYQF+YlLqljGAORfWrLgeJM67jvskuAYLDup/IDoqcAsdOAqNsAN2/njJeoD+r2Ymjjxo1Ys2YNioqKMGbMGKxfvx7x8fE2+27atAkff/wxzpw5AwAYO3Ys3njjDav+DzzwAD766COr1yUnJ2P37t12jYfFEBFR9yur0eJoTrl5Wd3lq7bDGMRldb4YHuTBMAYiezTVA8XngMIMIPcwkHMQ0FZZnpfIgLAEsTCKmSbe/4hL6oja1K3F0LZt2zB//ny8//77SEhIwLp167B9+3ZcvHgRAQEBrfrPmzcPSUlJSExMhFKpxOrVq7Fjxw6cPXsWoaGhAMRiqLi4GJs3bza/TqFQwNvbvr+CsBgiIup59oQxTIjyRVKUuKyOYQxEdtI3Afk/All7gKx9QOkF6+c9w4CYqWJhFHELIHd3zjiJeqluLYYSEhIwfvx4bNiwAQBgMBgQFhaGRYsWYdmyZR2+Xq/Xw9vbGxs2bMD8+fMBiMVQRUUFdu7c6chQzFgMERE5lymMIT1HXFJnK4xhkLcYxpAUwzAGIodc+1UsirL2ArnfAboGy3MyBRDxe7EwipkG+EQ4bZhEvUW3FUONjY1QqVT4/PPPMXv2bPPjKSkpqKiowFdffdXhMaqrqxEQEIDt27fjD3/4AwCxGNq5cyfkcjm8vb1x22234bXXXoOvr+07PGu1Wmi1WvP3VVVVCAsLYzFERNRLNOoMOHWlAj9kdRzGkBTti/gIX6gZxkDUscY64NcfxFmjS3uBynzr5/1iLYXR4Am8zxENSN1WDBUUFCA0NBTp6emYMGGC+fHnn38ehw8fxk8//dThMZ588kns2bMHZ8+ehVKpBABs3boVKpUKERERyMnJwV//+leo1WocPXoUMpms1TFefvllrFy5stXjLIaIiHqnWq0Ox4xhDEdyynG+sMrqeVMYQ6IxqY5hDER2EAQx0ttUGOUfBYRmM7JyDRA1CYhJFpfVaYKcNlSintRri6FVq1bhzTffxKFDhzB69Og2+/3yyy+IiorC/v37MXny5FbPc2aIiKhvsyeMYXy4D26OZhgDkd3qK4BfDoqFUfY+oLbU+vngMWJhFJsMhNwISFv/wZmoP3CkGHJoTYKfnx9kMhmKi4utHi8uLkZQUPt/bVi7di1WrVqF/fv3t1sIAUBkZCT8/PyQnZ1tsxhSKBRQKLjWnIior/JTKzBzTAhmjgkBYB3GcDSnHOW1jTh8qRSHL4kf5hjGQGQHNy9g5F1iMxiAwpNiYZS1Fyg4ARSeEtt3bwIqXyB6KqO7acDrVIBCfHw81q9fD0AMUBg8eDAWLlzYZoDCm2++iddffx179uzB7373uw5/xpUrVzB48GDs3LkTd955Z4f9GaBARNR/GAwCLhZXi7NG2WX4iWEMRNevpsQSwpBzgNHd1K91e7R2SkoKPvjgA8THx2PdunX47LPPcOHCBQQGBmL+/PkIDQ1FamoqAGD16tV46aWX8OmnnyIpKcl8HLVaDbVajZqaGqxcuRJz5sxBUFAQcnJy8Pzzz6O6uhqZmZl2zQCxGCIi6r9MYQym4qijMIYbQjzhp1ZwWR1RW8zR3cZZI0Z3Uz/T7Tdd3bBhg/mmq3FxcXjnnXeQkJAAAJg0aRLCw8OxZcsWAEB4eDjy8vJaHWPFihV4+eWXUV9fj9mzZ+PkyZOoqKhASEgIpk2bhldffRWBgYF2jYfFEBHRwFGr1eHYr1dxJMt2GAMAuMokCPJUIsTTDSFebgjxUiLY0w2hXuL3wV5KeChdnTB6ol7oWp6lMGJ0N/UD3V4M9TYshoiIBq7yGi3Sc8qRniPuN8q/WgeDHf9l0yhcEOylFIsjTzeEWn3thkBPBRQu3GBOAwyju6kfYDFEREQDlk5vQHG1FgUV9cbWgIKKehRW1uO3igYUVtajoq7JrmP5axTizJKnqVBSItTLDcHG2SY/dy7Ho36M0d3UR7EYIiIiakddo85cJBVU1KOg0lIwmR7X6gwdHkcuk4rL8bwsS/JMs02hxuJJw+V41F8wupv6CBZDRERE10EQBFytbRQLo8p6Y6HUgN8q6lFonG0qrm6APf8F1ShdjIWSEsFepn1Llj1MgR5K3mCW+h5b0d3NMbqbnIjFEBERUTdr0htQXNWAAuPSO7FQarCaaaqs73g5nkQC+KsVxkJJnGEyfR1snG3ydZdzOR71bjUlQPZ+4NIeRneT07EYIiIi6gVqtTrLXqUWS/JMXzfauRwv2FwoGfctGWebTHuY1AqH7qNO1H30TcDln8TCyFZ0t8cgcY9RbDKju6lbsBgiIiLqAwRBQHltIworxCV4VvuWjMvzSqq1di3H81C6GGPExb1KzfcthXi5IchTCVcZl+ORE3QU3R1+s1gYMbqbugiLISIion6iUWdajtds31KzoIeCinpUNeg6PI5EAgRoFM3ut2RZhhdiDH3wdZdDwuVL1J0Y3U09gMUQERHRAFKj1aGwwrhvybwMz7R/SdzL1Ki3Yzmei7RZjLhx35Jxtsm0h8mdy/GoqzSP7s7aJ0Z3G5oV9ozupk5iMURERERmBoNxOV5lfatCyfR1aY19y/E83Vyb3W+p+XI8cYYp0IPL8aiTGirF8IWsfeKSuraiu2OmAaE3Mbqb2sRiiIiIiBzSfDleQYtleKbledV2LMeTSoAAjbLV/ZbEm9eKBZMPl+NRR0zR3Vn7xCCGtqK7Y6YC0ZMZ3U1WWAwRERFRl6tuaLK5DM9UMNm7HE/hIjXvVTLvWzIVTFyOR7bYE91tSqhjdPeAx2KIiIiIepzBIKCsVtvqfkumePGCinqUVmvtOpaXytWyb6lF0EOIlxsCNQq4cDnewMTobuoAiyEiIiLqlbQ6PYortVYzSuZYcWPBVK21bzleoIfSZoy4qXmrXLkcbyBgdDe1wGKIiIiI+qyqhiaxMDIHPtSb78VUWNmAwsp6NOk7/viidJUa9yk1L5QsaXkhXkqo5FyO16801QO537cd3e0bYymMGN3db7EYIiIion7LtByvechDQUWDJS2vssHu5XjexuV41oWS0ngvJjcEcDle32WO7jbOGjG6e8BgMUREREQDmlanR1Flg7lgMu1bah4vXmPncrwgD8v9lkK8lFazTaFebvDicry+gdHdAwaLISIiIqIOVDU0tViCJxZJpq+LKhvsX45nM0bceC8mTze4yfnBuldhdHe/xmKIiIiI6DoZDALKarTmvUoFFfXi1+b9TA0oq7FvOZ6Pu7xZoWRcjuclpuWJy/GUkEk5u+Q0jO7uV1gMEREREfWAhiY9iqtMiXgNKLRx09raRn2Hx5FJJQjyUCLMxw0xARrEBqoRHaBBTKAavrxJbc+yiu7eB5Set36e0d29HoshIiIiol5AEARUNeis9y1V1Fvdh6mosgE6Q9sfx7xVrogxFkYxAWrEBGoQE6CGv0bBIqknmKO79wG5h9uJ7p4K+EQ6b5xkxmKIiIiIqI/QN1uOl1tai6ySGmSXVCOrpAb5V+vQ1ic1TzdXY3FknEUKUCM2UINADxZJ3cYc3b1XjO+uaCu6eyowOJHR3U7CYoiIiIioH6hv1COntAbZJTXIKqlGVnENskpqkFdei7YmkzQKF0QbZ5FiAzWINs4mhXgqWSR1JUZ391rdXgxt3LgRa9asQVFREcaMGYP169cjPj7eZt9Nmzbh448/xpkzZwAAY8eOxRtvvGHVXxAErFixAps2bUJFRQWSkpLw3nvvISYmxq7xsBgiIiKigaShSY/cMnEWKavYVCRV49fyOujbqJLc5TJzYWSaUYoJ0CDUyw1Shjdcv4ZKIOegpTiyGd09TSyOGN3drbq1GNq2bRvmz5+P999/HwkJCVi3bh22b9+OixcvIiAgoFX/efPmISkpCYmJiVAqlVi9ejV27NiBs2fPIjQ0FACwevVqpKam4qOPPkJERARefPFFZGZm4ty5c1AqlV36homIiIj6q0adAb+W1+KSsUAyzSjlltW2GRPu5moskgLUxhklMcBhkLeKCXedZTAAhRliUdRmdPcUsThidHeX69ZiKCEhAePHj8eGDRsAAAaDAWFhYVi0aBGWLVvW4ev1ej28vb2xYcMGzJ8/H4IgICQkBM888wyeffZZAEBlZSUCAwOxZcsW3Hvvva2OodVqodVaoiyrqqoQFhbGYoiIiIjIhia9AXnlteZldpeKq5FdUoNfSmvRqDfYfI3CRYoof3Wr4IbBPiq4yKQ9/A76OKvo7oOAttLynERqjO6exujuLuJIMeTiyIEbGxvx888/Y/ny5ebHpFIppkyZgqNHj9p1jLq6OjQ1NcHHxwcAkJubi6KiIkyZMsXcx9PTEwkJCTh69KjNYig1NRUrV650ZOhEREREA5arTIroAA2iAzSY0exxnd6A/Kt1luV2JTXIKq5BTmkNtDoDzhVW4VxhldWx5DIpIv3dLcvtjEvuhvi6w5VFkm3qACDuv8Rmiu7O2gtc2itGd+cfFVvaSkZ39zCHiqGysjLo9XoEBgZaPR4YGIgLFy7YdYylS5ciJCTEXPwUFRWZj9HymKbnWlq+fDmWLFli/t40M0RERERE9nORSRHpr0akvxrJIy0b/PUGAZdNRVJJNbKNM0rZJTWob9LjQlE1LhRVWx3LVSZBhJ87YgJMoQ1igEO4rzvkLiySzGSuYhx3+M3A1FdaR3dXXQF+3iw2Rnd3O4eKoeu1atUqbN26FYcOHbJrL1BbFAoFFApFF46MiIiIiExkUgnC/dwR7ueOqSMsf7A2GAT8VlFvTra7VGyJAa9r1OOS8bFWx/JVWW4ma5xRivBzh9KVIQLwHgLEPyo2W9HdOWli+xaM7u4GDhVDfn5+kMlkKC4utnq8uLgYQUHtxwWuXbsWq1atwv79+zF69Gjz46bXFRcXIzg42OqYcXFxjgyPiIiIiLqRVCpBmI8KYT4q3DbMukgqrGoQ9yIZk+2ySmqQXVyDaq0OOaW1yCmtxe6zzY4lAYb4ulsl20UHqBEdoB64RZKrGxA7TWzCmtbR3eVZwNEs4OiGZtHd08TG6O5OcagYksvlGDt2LNLS0jB79mwAYoBCWloaFi5c2Obr3nzzTbz++uvYs2cPxo0bZ/VcREQEgoKCkJaWZi5+qqqq8NNPP+GJJ55w7N0QERERUY+TSiUI9XJDqJcbbh1qSRcWBAFFVQ3m4IbskmpcKhb3J1U16JBbVovcslrsPWf5Q7tEAgz2UYnpds1uJhsV4A6VvEcXNTmXRAIEDBNb0lMtorv3AbUlwPn/ExvA6O5O6lS0dkpKCj744APEx8dj3bp1+Oyzz3DhwgUEBgZi/vz5CA0NRWpqKgAxNvull17Cp59+iqSkJPNx1Go11Gq1uc+qVausorVPnz7NaG0iIiKifkgQBJRWa83BDZeMs0iXSqpRUdfU5usGebtZJdvFGG8qq1YMoCIJsI7uztoL/HYCQLOP9AM8urvbb7q6YcMG801X4+Li8M477yAhIQEAMGnSJISHh2PLli0AgPDwcOTl5bU6xooVK/Dyyy8DsNx09f/9v/+HiooK3HzzzXj33XcRGxtr13hYDBERERH1fYIgoLy20Rz9bbqZbFZxDcprG9t8XYin0upmstEBGsQEquGhdO3B0TtRTSmQvU8sjLIPDPjo7m4vhnobFkNERERE/Vt5jdZ4E1kx1e6SMQq8tFrb5muCPJTG4shyM9mYAA08Vf24SLIV3d2cKbo7ZhoQObFfRnezGCIiIiKiAaGirtFYHBljwI0zSkVVDW2+xl+jaHaPJMuSOx/3fpjOdi1PnDW6tBfI/Q7Q1VueM0V3xxhDG/pJdDeLISIiIiIa0KoampBliv42BjhkFVejoLLtIsnXXW5Otms+o+SnlkPSH5aWNdUDv/4AXNpjie5uzjfGUhj14ehuFkNERERERDZUNzQhp7QWWcZldqZ/r1yrb/M13ipXMfo7UG1Ot4sJUMNfo+i7RZIgAGWXjIWRMbrboLM834eju1kMERERERE5oK5Rh5ySWvNeJNPNZPOv1qGtT8seSherZXamAIcgD2XfK5JsRXc3FzTaeMPX3h/dzWKIiIiIiKgL1DfqkVNaYwxvsCy5yyuvhaGNT9FqhQuiA9TmwAbTjFKol1vfKJIcie6Oug1Q+ThrpDaxGCIiIiIi6kYNTXrkltValtoZAxx+La+Dvo0qyV0uQ3SAJfrbVCyFerlBKu3FRZK90d0x04DAkU6P7mYxRERERETkBI06A34tNy63K7bMKOWW1aJJb/tjt9JVag5rMAc4BKgR5qOCrLcVSR1Fd//5C3HWyIlYDBERERER9SJNegPyymvNy+xMN5b9pbQWjXqDzdcoXKSI8lcbCyTLjNIQHxVcZNIefgdtqMi3FEZXjgF/OQfIVU4dEoshIiIiIqI+QKc3IP9qnVWyXVZxDXJKa6DV2S6S5DIpIv3drWaTYgPVGOLrDldnFkl6HSBzcd7PN2IxRERERETUh+kNAi6biqSSamQbZ5SyS2pQ36S3+RoXqQQRfu5W90qKCdAg3E8FhUvvTX/raiyGiIiIiIj6IYNBwG8V9eZku0vFlhjwukbbRZJMKsEQXxViW9xMNtLfHUrX/lcksRgiIiIiIhpADAYBhVUN4l4kY7JdVkkNsotrUK3V2XyNVAIM8TUttxNvJhsdoEaUvxpu8r5bJLEYIiIiIiIiCIKAoqoGc3BDdkk1LhWL+5OqGmwXSRIJEOatanUz2Sh/NdwVzt8T1BEWQ0RERERE1CZBEFBarTUHN1wyziJdKqlGRV1Tm68L9XIT749knEUyFUzqXlQksRgiIiIiIiKHCYKA8tpGc/S36WayWcU1KK9tbPN1IZ5KRAdq8Lc7hiM2UNODI27Nkdqg95RwRERERETkVBKJBH5qBfzUCiRG+Vk9V16jNd5EVky1u2SMAi+t1qKgsgEFlQ1QzBrppJF3DoshIiIiIiLqkK9aAV+1AgmRvlaPV9Q1GoujGgzydu4NVx3FYoiIiIiIiDrNSyXHuHAfjAv3cfZQHObEW9QSERERERE5D4shIiIiIiIakFgMERERERHRgNQv9gyZ0sGrqqqcPBIiIiIiInImU01gzx2E+kUxVF1dDQAICwtz8kiIiIiIiKg3qK6uhqenZ7t9+sVNVw0GAwoKCqDRaCCRSJw9HFRVVSEsLAyXL1/mTWC7Ac9v9+L57V48v92L57d78fx2L57f7sXz27160/kVBAHV1dUICQmBVNr+rqB+MTMklUoxaNAgZw+jFQ8PD6dfDP0Zz2/34vntXjy/3Yvnt3vx/HYvnt/uxfPbvXrL+e1oRsiEAQpERERERDQgsRgiIiIiIqIBicVQN1AoFFixYgUUCoWzh9Iv8fx2L57f7sXz2714frsXz2/34vntXjy/3auvnt9+EaBARERERETkKM4MERERERHRgMRiiIiIiIiIBiQWQ0RERERENCCxGCIiIiIiogGJxRAREREREQ1ILIY6aePGjQgPD4dSqURCQgKOHTvWbv/t27dj2LBhUCqVGDVqFHbt2tVDI+2bHDm/W7ZsgUQisWpKpbIHR9t3fPfdd5g5cyZCQkIgkUiwc+fODl9z6NAh3HTTTVAoFIiOjsaWLVu6fZx9laPn99ChQ62uXYlEgqKiop4ZcB+TmpqK8ePHQ6PRICAgALNnz8bFixc7fB1//9qnM+eXv3/t995772H06NHw8PCAh4cHJkyYgG+//bbd1/DatZ+j55fX7vVZtWoVJBIJFi9e3G6/vnANsxjqhG3btmHJkiVYsWIFTpw4gTFjxiA5ORklJSU2+6enp+O+++7Dww8/jJMnT2L27NmYPXs2zpw508Mj7xscPb8A4OHhgcLCQnPLy8vrwRH3HbW1tRgzZgw2btxoV//c3FzccccduPXWW5GRkYHFixfjkUcewZ49e7p5pH2To+fX5OLFi1bXb0BAQDeNsG87fPgwFixYgB9//BH79u1DU1MTpk2bhtra2jZfw9+/9uvM+QX4+9degwYNwqpVq/Dzzz/jP//5D2677TbMmjULZ8+etdmf165jHD2/AK/dzjp+/Dg++OADjB49ut1+feYaFshh8fHxwoIFC8zf6/V6ISQkREhNTbXZ/5577hHuuOMOq8cSEhKE//7v/+7WcfZVjp7fzZs3C56enj00uv4DgLBjx452+zz//PPCyJEjrR6bO3eukJyc3I0j6x/sOb8HDx4UAAjXrl3rkTH1NyUlJQIA4fDhw2324e/fzrPn/PL37/Xx9vYW/vGPf9h8jtfu9Wvv/PLa7Zzq6mohJiZG2LdvnzBx4kTh6aefbrNvX7mGOTPkoMbGRvz888+YMmWK+TGpVIopU6bg6NGjNl9z9OhRq/4AkJyc3Gb/gawz5xcAampqMGTIEISFhXX4lyCyH6/dnhEXF4fg4GBMnToVR44ccfZw+ozKykoAgI+PT5t9eA13nj3nF+Dv387Q6/XYunUramtrMWHCBJt9eO12nj3nF+C12xkLFizAHXfc0eratKWvXMMshhxUVlYGvV6PwMBAq8cDAwPbXOdfVFTkUP+BrDPnd+jQofjwww/x1Vdf4Z///CcMBgMSExNx5cqVnhhyv9bWtVtVVYX6+nonjar/CA4Oxvvvv48vvvgCX3zxBcLCwjBp0iScOHHC2UPr9QwGAxYvXoykpCTccMMNbfbj79/Osff88vevYzIzM6FWq6FQKPD4449jx44dGDFihM2+vHYd58j55bXruK1bt+LEiRNITU21q39fuYZdnD0Aous1YcIEq7/8JCYmYvjw4fjggw/w6quvOnFkRO0bOnQohg4dav4+MTEROTk5+Pvf/45PPvnEiSPr/RYsWIAzZ87ghx9+cPZQ+iV7zy9//zpm6NChyMjIQGVlJT7//HOkpKTg8OHDbX5gJ8c4cn557Trm8uXLePrpp7Fv375+FzTBYshBfn5+kMlkKC4utnq8uLgYQUFBNl8TFBTkUP+BrDPntyVXV1fceOONyM7O7o4hDihtXbseHh5wc3Nz0qj6t/j4eH7A78DChQvx9ddf47vvvsOgQYPa7cvfv45z5Py2xN+/7ZPL5YiOjgYAjB07FsePH8fbb7+NDz74oFVfXruOc+T8tsRrt30///wzSkpKcNNNN5kf0+v1+O6777BhwwZotVrIZDKr1/SVa5jL5Bwkl8sxduxYpKWlmR8zGAxIS0trc13qhAkTrPoDwL59+9pdxzpQdeb8tqTX65GZmYng4ODuGuaAwWu352VkZPDabYMgCFi4cCF27NiBAwcOICIiosPX8Bq2X2fOb0v8/esYg8EArVZr8zleu9evvfPbEq/d9k2ePBmZmZnIyMgwt3HjxmHevHnIyMhoVQgBfegadnaCQ1+0detWQaFQCFu2bBHOnTsnPPbYY4KXl5dQVFQkCIIg3H///cKyZcvM/Y8cOSK4uLgIa9euFc6fPy+sWLFCcHV1FTIzM531Fno1R8/vypUrhT179gg5OTnCzz//LNx7772CUqkUzp4966y30GtVV1cLJ0+eFE6ePCkAEN566y3h5MmTQl5eniAIgrBs2TLh/vvvN/f/5ZdfBJVKJTz33HPC+fPnhY0bNwoymUzYvXu3s95Cr+bo+f373/8u7Ny5U8jKyhIyMzOFp59+WpBKpcL+/fud9RZ6tSeeeELw9PQUDh06JBQWFppbXV2duQ9//3ZeZ84vf//ab9myZcLhw4eF3Nxc4fTp08KyZcsEiUQi7N27VxAEXrvXy9Hzy2v3+rVMk+ur1zCLoU5av369MHjwYEEulwvx8fHCjz/+aH5u4sSJQkpKilX/zz77TIiNjRXkcrkwcuRI4ZtvvunhEfctjpzfxYsXm/sGBgYKt99+u3DixAknjLr3M0U5t2ym85mSkiJMnDix1Wvi4uIEuVwuREZGCps3b+7xcfcVjp7f1atXC1FRUYJSqRR8fHyESZMmCQcOHHDO4PsAW+cWgNU1yd+/ndeZ88vfv/Z76KGHhCFDhghyuVzw9/cXJk+ebP6gLgi8dq+Xo+eX1+71a1kM9dVrWCIIgtBz81BERERERES9A/cMERERERHRgMRiiIiIiIiIBiQWQ0RERERENCCxGCIiIiIiogGJxRAREREREQ1ILIaIiIiIiGhAYjFEREREREQDEoshIiIiIiIakFgMERERERHRgMRiiIiIiIiIBiQWQ0RERERENCD9f90AUwxpWRqKAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 1000x200 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "history_LSTM\n",
    "plt.figure(figsize=(10, 2))\n",
    "\n",
    "plt.plot(history_LSTM[\"val_loss\"], label=\"Validation Loss\")\n",
    "plt.plot(history_LSTM[\"loss\"], label=\"Training Loss\")\n",
    "plt.legend()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "id": "95bf3dc8",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.94      0.91      0.92      1174\n",
      "           1       0.92      0.94      0.93      1308\n",
      "\n",
      "    accuracy                           0.93      2482\n",
      "   macro avg       0.93      0.93      0.93      2482\n",
      "weighted avg       0.93      0.93      0.93      2482\n",
      "\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import confusion_matrix, classification_report, f1_score\n",
    "\n",
    "## print classification report\n",
    "print(classification_report(y_test,y_pred_LSTM))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "id": "6d36a04e",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Original Approach Accuracy: 0.9271\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import accuracy_score, precision_score, recall_score\n",
    "\n",
    "Org_accuracy = accuracy_score(y_pred_LSTM, y_test)\n",
    "print(f\"Original Approach Accuracy: {Org_accuracy:.4f}\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "id": "1ba7bf93",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(45.722222222222214, 0.5, 'Real Labels')"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 600x300 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#plot the confusion matrix\n",
    "import seaborn as sns\n",
    "from sklearn.metrics import confusion_matrix, classification_report, f1_score\n",
    "\n",
    "plt.figure(figsize = (6,3))\n",
    "\n",
    "sns.heatmap(confusion_matrix(y_test,y_pred_LSTM), annot=True, \n",
    "            fmt='', cmap='Blues')\n",
    "\n",
    "plt.xlabel('Predicted Labels')\n",
    "plt.ylabel('Real Labels')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "id": "6dea8557",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "248/248 [==============================] - 53s 212ms/step\n"
     ]
    }
   ],
   "source": [
    "y_pred_LSTM_2 = (model.predict(X_2) >= 0.5).astype(int)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "id": "64831124",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Original Approach Accuracy 2: 0.8266\n"
     ]
    }
   ],
   "source": [
    "Org_accuracy_2 = accuracy_score(y_pred_LSTM_2, y_2)\n",
    "print(f\"Original Approach Accuracy 2: {Org_accuracy_2:.4f}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "id": "c89345f7",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(45.722222222222214, 0.5, 'Real Labels')"
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 600x300 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize = (6,3))\n",
    "\n",
    "sns.heatmap(confusion_matrix(y_2,y_pred_LSTM_2), annot=True, \n",
    "            fmt='', cmap='Blues')\n",
    "\n",
    "plt.xlabel('Predicted Labels')\n",
    "plt.ylabel('Real Labels')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "id": "4e4cab42",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.82      0.83      0.82      3881\n",
      "           1       0.83      0.82      0.83      4053\n",
      "\n",
      "    accuracy                           0.83      7934\n",
      "   macro avg       0.83      0.83      0.83      7934\n",
      "weighted avg       0.83      0.83      0.83      7934\n",
      "\n"
     ]
    }
   ],
   "source": [
    "## print classification report\n",
    "print(classification_report(y_2,y_pred_LSTM_2))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "id": "a1f9e21c",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "248/248 [==============================] - 42s 171ms/step\n"
     ]
    }
   ],
   "source": [
    "y_pred_LSTM_3 = (model.predict(X_3) >= 0.5).astype(int)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "id": "394326bd",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Original Approach Accuracy 3: 0.8085\n"
     ]
    }
   ],
   "source": [
    "Org_accuracy_3 = accuracy_score(y_pred_LSTM_3, y_3)\n",
    "print(f\"Original Approach Accuracy 3: {Org_accuracy_3:.4f}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "id": "a4c229b6",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(45.722222222222214, 0.5, 'Real Labels')"
      ]
     },
     "execution_count": 57,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 600x300 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize = (6,3))\n",
    "\n",
    "sns.heatmap(confusion_matrix(y_3,y_pred_LSTM_3), annot=True, \n",
    "            fmt='', cmap='Blues')\n",
    "\n",
    "plt.xlabel('Predicted Labels')\n",
    "plt.ylabel('Real Labels')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "id": "3cb1af1a",
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train_N = np.float32(X_train)\n",
    "X_test_N = np.float32(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "id": "a928a036",
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train_N_mean = X_train_N.mean()\n",
    "X_train_N = X_train_N - X_train_N_mean\n",
    "X_train_N_std = X_train_N.std()\n",
    "X_train_N /= X_train_N_std\n",
    "\n",
    "\n",
    "X_test_N_mean = X_test_N.mean()\n",
    "X_test_N = X_test_N - X_test_N_mean\n",
    "X_test_N_std = X_test_N.std()\n",
    "X_test_N /= X_test_N_std\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "id": "6f29531c",
   "metadata": {},
   "outputs": [],
   "source": [
    "autoencoder = tf.keras.Sequential([\n",
    "    tf.keras.layers.Input(shape=(maxlen,)),\n",
    "    tf.keras.layers.Dense(128, activation='relu'),\n",
    "    tf.keras.layers.Dense(64, activation='relu'),\n",
    "    tf.keras.layers.Dense(128, activation='relu'),\n",
    "    tf.keras.layers.Dense(maxlen, activation='sigmoid')\n",
    "])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "id": "4d7a2ba1",
   "metadata": {},
   "outputs": [],
   "source": [
    "# normalize training and testing set using mean and std deviation of training set\n",
    "\n",
    "X_2_N = np.float32(X_2)\n",
    "X_3_N = np.float32(X_3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "id": "cedf4707",
   "metadata": {},
   "outputs": [],
   "source": [
    "X_2_N_mean = X_2_N.mean()\n",
    "X_3_N_mean = X_3_N.mean()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "id": "d0ac9367",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "4186.51\n"
     ]
    }
   ],
   "source": [
    "X_2_N = X_2_N - np.float32(X_2_N_mean)\n",
    "X_2_N_std = X_2_N.std()\n",
    "X_2_N /= X_2_N_std\n",
    "print(X_2_N_std)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "id": "b4fe86b6",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "3066.9062\n"
     ]
    }
   ],
   "source": [
    "X_3_N = X_3_N - np.float32(X_3_N_mean)\n",
    "X_3_N_std = X_3_N.std()\n",
    "X_3_N /= X_3_N_std\n",
    "print(X_3_N_std)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "id": "646e52b9",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/7\n",
      "233/233 [==============================] - 1s 5ms/step - loss: 0.9478 - accuracy: 0.0553 - val_loss: 0.9098 - val_accuracy: 0.1297\n",
      "Epoch 2/7\n",
      "233/233 [==============================] - 1s 3ms/step - loss: 0.9008 - accuracy: 0.2165 - val_loss: 0.8943 - val_accuracy: 0.2583\n",
      "Epoch 3/7\n",
      "233/233 [==============================] - 1s 3ms/step - loss: 0.8875 - accuracy: 0.3133 - val_loss: 0.8881 - val_accuracy: 0.3082\n",
      "Epoch 4/7\n",
      "233/233 [==============================] - 1s 3ms/step - loss: 0.8799 - accuracy: 0.3425 - val_loss: 0.8846 - val_accuracy: 0.3485\n",
      "Epoch 5/7\n",
      "233/233 [==============================] - 1s 3ms/step - loss: 0.8746 - accuracy: 0.3629 - val_loss: 0.8833 - val_accuracy: 0.3578\n",
      "Epoch 6/7\n",
      "233/233 [==============================] - 1s 3ms/step - loss: 0.8706 - accuracy: 0.3805 - val_loss: 0.8827 - val_accuracy: 0.3771\n",
      "Epoch 7/7\n",
      "233/233 [==============================] - 1s 3ms/step - loss: 0.8673 - accuracy: 0.3899 - val_loss: 0.8800 - val_accuracy: 0.3771\n"
     ]
    }
   ],
   "source": [
    "autoencoder.compile(optimizer='adam', loss='mean_squared_error',metrics=['accuracy'])\n",
    "history_auto = autoencoder.fit(X_train_N, X_train_N, epochs=7,validation_data=(X_test_N, X_test_N), shuffle=True).history"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "id": "b16a666e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x7ff8e0452080>"
      ]
     },
     "execution_count": 66,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAzoAAADICAYAAAAgA5Q7AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjYuMCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy89olMNAAAACXBIWXMAAA9hAAAPYQGoP6dpAABH40lEQVR4nO3deXhU5f3+8ffMJJN9JZAECISEVdmUJQVkUalYlYorUpXVXag05SvGBVArUUGKClWxFaSWH6IVa4uCirKIIIiNBZQghJ3sZN8mmZnfHydMCCSQQGCScL+ua66Z88w5M5/JIObm2UxOp9OJiIiIiIhIM2J2dwEiIiIiIiINTUFHRERERESaHQUdERERERFpdhR0RERERESk2VHQERERERGRZkdBR0REREREmh0FHRERERERaXYUdEREREREpNlR0BERERERkWZHQUdERERERJodj3O5aOHChcyZM4e0tDR69erF66+/Tv/+/c963fLlyxkzZgw333wzH3/8cZ3fz+FwcOzYMQICAjCZTOdSsoiIiIiINANOp5OCggJat26N2Vx7v029g877779PfHw8b775JnFxccyfP58RI0aQnJxMq1atar3uwIEDTJs2jcGDB9f3LTl27BhRUVH1vk5ERERERJqnw4cP07Zt21qfNzmdTmd9XjAuLo5+/fqxYMECwOhtiYqKYsqUKTzxxBM1XmO32xkyZAgTJ05k48aN5Obm1qtHJy8vj+DgYA4fPkxgYGB9yhURERERkWYkPz+fqKgocnNzCQoKqvW8evXo2Gw2tm/fTkJCgqvNbDYzfPhwNm/eXOt1zz33HK1atWLSpEls3LjxrO9TVlZGWVmZ67igoACAwMBABR0RERERETnrlJZ6LUaQlZWF3W4nPDy8Wnt4eDhpaWk1XvPNN9/wt7/9jbfffrvO75OYmEhQUJDrpmFrIiIiIiJSHxd01bWCggLuvfde3n77bcLCwup8XUJCAnl5ea7b4cOHL2CVIiIiIiLS3NRr6FpYWBgWi4X09PRq7enp6URERJx2/r59+zhw4AAjR450tTkcDuONPTxITk4mNjb2tOu8vLzw8vKqT2kiIiIiIiIu9Qo6VquVPn36sHbtWkaNGgUYwWXt2rVMnjz5tPO7du3Kjh07qrU9/fTTFBQU8OqrrzbokDSHw4HNZmuw1xNpLDw9PbFYLO4uQ0RERBo7pxMqyqCitOpWfvLjksrnK+9PPj75vNqu63kn9Jvk7k9ZZ/VeXjo+Pp5x48bRt29f+vfvz/z58ykqKmLChAkAjB07ljZt2pCYmIi3tzfdu3evdn1wcDDAae3nw2azsX//fldvkUhzExwcTEREhPaREhERaQqcTrCXnxIoaggQruNzDB6nXVcK1GtB5fqJ6nfhXvsCqHfQGT16NJmZmcyYMYO0tDR69+7N6tWrXQsUHDp06Iwb9zQ0p9NJamoqFouFqKioi/reIhea0+mkuLiYjIwMACIjI91ckYiISBNjLz+HnowazqtvYLmQgaNOTODpAx7exs3T+6THPuDhBR6V9ycfn3zeqdeFdXLzZ6qfeu+j4w75+fkEBQWRl5d32vLS5eXl7N27l9atW59xHW2Rpiw7O5uMjAw6d+6sYWwiItI02SsabghVfa5z2t39yesRKGo4z3VcQ/A403UWT2imI0HOlA1OVu8encbGbjf+8FqtVjdXInLh+Pr6AkawV9AREZGLxukEWxGU5p3hlnt6m63w9B4QR4W7Pw1YvM4tUJxrD4inD1iszTZwNHZNPuicoLkL0pzpz7eIiJwTpxPKi+sXUk7cSiqfuwA9InaTJ3aLF3azFxVmrxof212PrVSYvV2P7WZvo83iXXnsddLjynuLl3GN2Yrd9ZwVTOc5xcFeeSs700lOoKTylnN+79fIdG8TRN/oUHeXUWfNJuiIiIiINDtOp9ErUqeQcupx5a0BelLsWCg2+1No8iPP6Uue05fjdh9yHb7k40u+06/y3pc8/Ch2elOKlVKslOFJqbPyvvLYeWG3cgRslTdpSA8OjVHQkYtj2LBh9O7dm/nz5wMQHR3N1KlTmTp1aq3XmEwmVq5c6Voe/Fw11OuIiIg0a06nMWyrxh6TnLMMCTsRVMrPuwwHFkos/hSZ/SjEj3ynLzkOX7LtPmTbfaoFlfzK56vufSnBC6h5dIGnxUSQj5UgHw+Cfa0E+3jS0tNS2+nShHWLqH0+TGOkoOMGI0eOpLy8nNWrV5/23MaNGxkyZAg//vgjPXv2rNfrbtu2DT8/v4YqE4BZs2bx8ccfk5SUVK09NTWVkJCQBn0vERGRRqm8tH49KKfe7Offs+AwWSiz+BthxeRPAUavSo7Dh+wKHzLLvcl1+tYaWIrPEFROCPDyIMjXkyAfT4J9PYnwsRJY+TjYp6rdCDUnHnvia7VoiLU0Sgo6bjBp0iRuu+02jhw5Qtu2bas9t3jxYvr27VvvkAPQsmXLhirxrCIiIi7aezUmNptNC1+IiDQ15TX1qOTWLaSU5oH9jBMy6sSJGZtnAKWWAErMfhSa/Shw+pHr9CWnslcls9ybzHLvU4aCGfdFeFOXLpITvSsnQkiUj2dVeDmpPahaeLES6O2Bh0VbdEjzoqDjBjfddBMtW7ZkyZIlPP300672wsJCPvjgA+bMmUN2djaTJ09mw4YN5OTkEBsby5NPPsmYMWNqfd1Th6798ssvTJo0ia1btxITE8Orr7562jXTp09n5cqVHDlyhIiICO6++25mzJiBp6cnS5Ys4dlnnwWqJsMvXryY8ePHnzZ0bceOHTz22GNs3rwZX19fbrvtNubNm4e/vz8A48ePJzc3l6uuuopXXnkFm83GXXfdxfz58/H09Kzx8+zbt4/4+Hi2bNlCUVER3bp1IzExkeHDh7vOKSsrY8aMGSxbtoyMjAyioqJISEhg0iRj195du3Yxffp0NmzYgNPppHfv3ixZsoTY2NjThv4BjBo1iuDgYJYsWeL6mU6aNIlffvmFjz/+mFtvvZUlS5ac8ed2wr///W+ee+45duzYgb+/P4MHD2blypU899xzrFixgp07d1b7vL1792bkyJE8//zztX7HIiKXpIqycwspJybTN0hQMVFhDcTmEUCpxZ8isz9FGCEk1+HLcYcPWRXeZJV7k1bmxXFHDUGltO69HgHeHgT5eNLC15PYyh6UqsBS1ZtycqgJ9vXEx1O9KyInNLug43Q6KSl3z3rpdf3LxcPDg7Fjx7JkyRKeeuop1zUffPABdrudMWPGUFhYSJ8+fZg+fTqBgYGsWrWKe++9l9jYWPr373/W93A4HNx6662Eh4fz3XffkZeXV+PcnYCAAJYsWULr1q3ZsWMH999/PwEBATz++OOMHj2anTt3snr1ar788kuAGvcqKioqYsSIEQwYMIBt27aRkZHBfffdx+TJk12BAeDrr78mMjKSr7/+mr179zJ69Gh69+7N/fffX+NnKCws5IYbbuCFF17Ay8uLpUuXMnLkSJKTk2nXrh0AY8eOZfPmzbz22mv06tWL/fv3k5WVBcDRo0cZMmQIw4YN46uvviIwMJBNmzZRUVG/SZlz585lxowZzJw5s04/N4BVq1Zxyy238NRTT7F06VJsNhuffvopABMnTuTZZ59l27Zt9Otn7DD83//+l//973989NFH9apNRKTJqCgz5qQUH4eS4yc9PnWeSu7pgaWi9Lzf3okJhzWAcs9Ayk4KKwUmPwpOzFepMMJKZrk3aTYv0m3erjkqRXjjLK1fj4fVYibI15PIk8JJYGXPyolgUtVm9KwE+Xiqd0WkgTS7oFNSbueyGWvc8t4/PTcCX2vdfqQTJ05kzpw5rF+/nmHDhgFGb8ltt91GUFAQQUFBTJs2zXX+lClTWLNmDStWrKhT0Pnyyy/ZvXs3a9asoXXr1gDMnj2b3/zmN9XOO7lHKTo6mmnTprF8+XIef/xxfHx88Pf3x8PD44xD1ZYtW0ZpaSlLly51zRFasGABI0eO5KWXXiI8PByAkJAQFixYgMVioWvXrtx4442sXbu21qDTq1cvevXq5Tp+/vnnWblyJZ988gmTJ09mz549rFixgi+++MLVyxMTE+M6f+HChQQFBbF8+XJXT0vnzp3P+rM71TXXXMMf//jHam1n+rkBvPDCC9x1112uHrETnwegbdu2jBgxgsWLF7uCzuLFixk6dGi1+kVEGiV7hRFGagstJcerPy7JNY7Li87rbZ2YcHoFYrcGUu4ZQJlHACVmY3J9gdOPvMqelewKbzIrvEm3GT0rx0qt5Dh8KTyHoHJCgLcHbU8a/lVbz8rJ4SXIR70rIu7W7IJOU9G1a1cGDhzIO++8w7Bhw9i7dy8bN27kueeeA4yNUGfPns2KFSs4evQoNpuNsrIy18aRZ/Pzzz8TFRXlCjkAAwYMOO28999/n9dee419+/ZRWFhIRUXFGXeYre29evXqVW0hhEGDBuFwOEhOTnYFncsvv7zaZpeRkZHs2LGj1tctLCxk1qxZrFq1itTUVCoqKigpKeHQoUMAJCUlYbFYGDp0aI3XJyUlMXjw4FqHxtVV3759T2s7288tKSmp1gAHcP/99zNx4kTmzZuH2Wxm2bJl/PnPfz6vOkVE6sXprBzedSKo5NQcVKqFlsrel3N9S5MZu1cw5dYgyq3B2DyDKPEMosgUcNrk+ozyqrBytNSTdJv1nIMKVPWu1DapviqwVM1bCfbxJEC9KyJNVrMLOj6eFn56boTb3rs+Jk2axJQpU1i4cCGLFy8mNjbW9Uv7nDlzePXVV5k/fz49evTAz8+PqVOnYrM13Jrwmzdv5u677+bZZ59lxIgRrt6PV155pcHe42SnBg6TyYTD4aj1/GnTpvHFF18wd+5cOnbsiI+PD7fffrvrZ+Dj43PG9zvb82azGafTWa2tvPz0JTxPXcmuLj+3s733yJEj8fLyYuXKlVitVsrLy7n99tvPeI2ISI1ObAhZS++Ko+g4jmLjRnEOppLjmEpzsJTlYTqPjSCLTH4UmgPII4A8/MnFn+NOf447/MiyG7c8/Mlx+pNDALlOPwrwxVlyfqEh0NujMrBYa5hUb7QHnhJegn2seHua1bsicolpdkHHZDLVefiYu91555089thjLFu2jKVLl/Lwww+7/hLetGkTN998M/fccw9gzLnZs2cPl112WZ1eu1u3bhw+fJjU1FQiIyMB2LJlS7Vzvv32W9q3b89TTz3lajt48GC1c6xWK3b7mf9H2K1bN5YsWUJRUZErFGzatAmz2UyXLl3qVG9NNm3axPjx47nlllsAo4fnwIEDrud79OiBw+Fg/fr11RYoOKFnz568++67lJeX19ir07JlS1JTU13HdrudnTt3cvXVV5+xrrr83Hr27MnatWuZMGFCja/h4eHBuHHjWLx4MVarlbvuuuus4UhEmrZyu4OyCgel5faq+3IHpRVV97bSUpyVvSimkhzMpTlYSnOwlOXiacvFasvFqyIPn/J8fOz5+Nrz8LcX4Ent+6yYK2+1KXZ6kYM/ec4ToaTyMf7kOo0Ak+M8+XEAefhhp+7/uGc2gbfVQrCHGW9PC14n3Xt5Wozwcsqk+lN7VoJ8jLksFrPCiojUTdNIBM2Uv78/o0ePJiEhgfz8fMaPH+96rlOnTnz44Yd8++23hISEMG/ePNLT0+scdIYPH07nzp0ZN24cc+bMIT8/v9ov5ife49ChQyxfvpx+/fqxatUqVq5cWe2c6Oho9u/fT1JSEm3btiUgIAAvL69q59x9993MnDmTcePGMWvWLDIzM5kyZQr33nuva9jauejUqRMfffQRI0eOxGQy8cwzz1TrAYqOjmbcuHFMnDjRtRjBwYMHycjI4M4772Ty5Mm8/vrr3HXXXSQkJBAUFMSWLVvo378/Xbp04ZprriE+Pp5Vq1YRGxvLvHnzyM3NrVNdZ/u5zZw5k2uvvZbY2FjuuusuKioq+PTTT5k+fbrrnPvuu49u3boBRqgTkQvP6XRiszsoLXdQVhkwyirs1Y5LT713BZPKc04NKyfdl1Xel9tseFYU4F2ei689H197AUHOAoJNBQSbigjBuA+mgBBTIa0q2/xM5746mM1pIZcAI5ScCCmnBJYCUwBFliBKPAIo9Qim1CMQs9WnMniY8fKwnHYf4mkmwuOkcOJpxtvDuPfyqH7suj8RYiqv8TCb1JsiIhedgo6bTZo0ib/97W/ccMMN1ebTPP3006SkpDBixAh8fX154IEHGDVqFHl5dRsbbTabWblyJZMmTaJ///5ER0fz2muvcf3117vO+e1vf8sf/vAHJk+eTFlZGTfeeCPPPPMMs2bNcp1z22238dFHH3H11VeTm5vrWl76ZL6+vqxZs4bHHnuMfv36VVte+nzMmzePiRMnMnDgQMLCwpg+fTr5+fnVznnjjTd48skneeSRR8jOzqZdu3Y8+eSTALRo0YKvvvqK//u//2Po0KFYLBZ69+7NoEGDAGNBiB9//JGxY8fi4eHBH/7wh7P25tT15zZs2DA++OADnn/+eV588UUCAwMZMmRItdfp1KkTAwcO5Pjx48TFxZ3Xz0qkOcspsrE/u4gDWUWk55fVGi5Oaz8pwJwcUk4ZsXoWTgIpJshUSAiFhJgKCaq8D6OQYFMhIaYCgiki2FRASGVbkKm46iXO1qVyCgdmisz+FFsCKbEEUuIZRJlnEDbPyrkt3iE4vIJxeIfi9AnG5BuKyTcUD++AyiBiIcjTTKtTQovVYtZcExG5pJicp05SaITy8/MJCgoiLy/vtInypaWl7N+/nw4dOuDt7e2mCkXqz+l00qlTJx555BHi4+PPeK7+nEtzl1tsY39WEQezi9mfVcSBymBzILuYvJLah2XVnRMfylxBJLgyuIR5FNHSUkSIqYhQkxFgAikkyFlAgLMAP0cBFmqfS3g2dmsATu8QnD6h4BOM2a8F5spggk8o+ISA63Gw8dgrCMwKJCIitTlTNjiZenRE3CAzM5Ply5eTlpZW6zwekeYmr7jcCDDZRUaYqQwyB7KLyC0+c5iJDPKmfQtfWgf74G+xE2IqIogCAikkwJGPv6MAP3s+vvZ8fCry8arIxcuWh9WWh4fNmHhvrm3TSGfl7Uw8fY1QcnIgOS2onHhcdZ7Fcn6rPoqIyLlT0BFxg1atWhEWFsaiRYsICQlxdzkiDSa/tJwDWSeCTDEHs4tcw85yzhJmogJM9A4u5fKAImK9C4jyyKWlKYeg8iw8itIh/xhkZpzffixmzxrCSUjNQcX1OAQ8tViIiEhTo6Aj4gZNYMSoSK0KSss5kFXM/uwiDmZVBZkD2cUcLzp9CXwTDsLIp7vpOJ19C+nmV0QHr3zaWHIJc2YTUJ6FV0k6ppIcyMS4nZWpKoTU2KNSS2ix+oMmxYuIXBIUdERE5DSFZRWV4aWosofGGGJ2MLuIrMKqMONPMeGmHMJNOQzjOOGWXKK98oj2zCPCnEuoIxs/WzZmZ4VxgR3Ir/k9AbB4QWAkBLSGgAgIrLwPiDQe+4cbgcU7WPNYRETkjBR0REQuUUVlFZVBptg1b+ZgthFq8gqLaEWOK8REmI7za1Mu4abjhHvm0MaSSytTDj7OktNf2AGcNh3GBP6tqgJLQMRJYSbSaA+INEKMelxERKQBKOiIiDRjxbYKV5A5kF3EgcxCsjNTKTl+FGtxGhGmHMLJIdx0nK6mXCJMxwk35RDmfaZul0onRmB6BVYGmJMCi+u4Msz4h4NF/8sREZGLR//XERFp4kpsdg5kF3EkPZPMY4coyDhEWe4RTAWp+JVl0sqUQ4Qph56m47QkFy9T5TAy61le2Ox5UmCp7IE5NcwERICX/wX/jCIiIvWloCMi0tjZKyjNPUb6kQNkpx2kKOswFblHMRem4VOaQYg9mzamHLqdvEnlCbX8LV/u3QJzYGssQafOhTnp2CdU82BERKTJUtAREXEXpxNKcqAgDQqOUZ57jLyMQxRnHcaRn4pHURp+ZZkEOnLxxkF7oH1Nr3NSFikzeVPk3Qq7XySWoEj8WkThFdqm+twY/wg8Pc7WnSMiItK0Keg0I9HR0UydOpWpU6fW6fx169Zx9dVXk5OTQ3Bw8AWtTeSSU14KBanGLf9YZZhJxZ53DFvOUZwFqViL0/FwVM3a9wTCanm5CqeZLFMIeR5hlHq3whEQiTW4Nf5hUbRoHY1/WBQEROLlHYjXRfmAIiIijZuCjhuYzrKi0MyZM5k1a1a9X3fbtm34+fnV+fyBAweSmppKUFBQvd9L5JLlsENRFhQcg/zUqjBTkOo6dhakGnvC1MACnLr1ZI7TnzRnCOnOUI5bQrH5hENgJN4hbQls1Y6w1u1p26Yd4f7eRGhFMhERkTpR0HGD1NRU1+P333+fGTNmkJyc7Grz96+a2Ot0OrHb7Xh4nP2ratmyZb3qsFqtRERE1Oua5sJms2G1auiOnMTphLJ8o+fF1QNz7KTj1Mq2NHDaz/hSJ6JIqdOTNGco6YSQ7gwxHjuNXhlzUCQ+oW0JiWhHVMtQosN86dHCj1A/61n/MURERETOTkHHDU4OF0FBQZhMJlfbieFkn376KU8//TQ7duzg888/Jyoqivj4eLZs2UJRURHdunUjMTGR4cOHu17r1KFrJpOJt99+m1WrVrFmzRratGnDK6+8wm9/+9tq73Vi6NqSJUuYOnUq77//PlOnTuXw4cNcddVVLF68mMjISAAqKiqIj49n6dKlWCwW7rvvPtLS0sjLy+Pjjz+u8fNmZ2czefJkNmzYQE5ODrGxsTz55JOMGTPGdY7D4WDu3LksWrSIw4cPEx4ezoMPPshTTz0FwJEjR/i///s/1qxZQ1lZGd26dWPhwoXExcUxfvx4cnNzq73/1KlTSUpKYt26dQAMGzaM7t274+HhwXvvvUePHj34+uuvmTdvHosXLyYlJYXQ0FBGjhzJyy+/XC1sbtq0iaeeeoqtW7fi5eVF//79Wb58Of/+97/5wx/+wLFjx/DyqhosNGrUKAICAvj73/9ezz8ZcsFU2KAwrXoPzEnDyVy9MeVFdXo5O2aynIGu4HIixGQQ7Gor8AwjtEUrolv6E93Cl+gWflwZ5kd0mB8tFGZEREQuuOYXdJxOKK9h5aGLwdO3wTa6e+KJJ5g7dy4xMTGEhIRw+PBhbrjhBl544QW8vLxYunQpI0eOJDk5mXbt2tX6Os8++ywvv/wyc+bM4fXXX+fuu+/m4MGDhIaG1nh+cXExc+fO5e9//ztms5l77rmHadOm8Y9//AOAl156iX/84x8sXryYbt268eqrr/Lxxx9z9dVX11pDaWkpffr0Yfr06QQGBrJq1SruvfdeYmNj6d+/PwAJCQm8/fbb/PnPf+aqq64iNTWV3bt3A1BYWMjQoUNp06YNn3zyCREREfzwww84HI56/UzfffddHn74YTZt2uRqM5vNvPbaa3To0IGUlBQeeeQRHn/8cf7yl78AkJSUxLXXXsvEiRN59dVX8fDw4Ouvv8Zut3PHHXfw+9//nk8++YQ77rgDgIyMDFatWsXnn39er9qkAdmKIPkz+PnfcHyfEWCKs+p8eYklgOPmUI7ZgzlUHkRqZZA5uUcmiyDsWPC1WogO8yM6zAgy/cL86BDmR/sWvrT091KYERERcaPmF3TKi2F2a/e895PHwFr3OTJn8txzz/HrX//adRwaGkqvXr1cx88//zwrV67kk08+YfLkybW+zvjx4109J7Nnz+a1115j69atXH/99TWeX15ezptvvklsbCwAkydP5rnnnnM9//rrr5OQkMAtt9wCwIIFC/j000/P+FnatGnDtGnTXMdTpkxhzZo1rFixgv79+1NQUMCrr77KggULGDduHACxsbFcddVVACxbtozMzEy2bdvmCmgdO3Y843vWpFOnTrz88svV2k5euCE6Opo//elPPPTQQ66g8/LLL9O3b1/XMcDll1/uevy73/2OxYsXu4LOe++9R7t27Rg2bFi965PzUFEGe7+EHR/CntU1/mOH02yl1KcV+Z5hZBHKEXswKWWB7Cn255g9hHSCSXeGUIJ3tet8PC20D/OlQ5gfg8L86NDCCDIdwvxoGaAwIyIi0lg1v6DTTPTt27facWFhIbNmzWLVqlWkpqZSUVFBSUkJhw4dOuPr9OzZ0/XYz8+PwMBAMjIyaj3f19fXFXIAIiMjXefn5eWRnp7u6oUBsFgs9OnT54y9K3a7ndmzZ7NixQqOHj2KzWajrKwMX19fAH7++WfKysq49tpra7w+KSmJK664otZeqLrq06fPaW1ffvkliYmJ7N69m/z8fCoqKigtLaW4uBhfX1+SkpJcIaYm999/P/369ePo0aO0adOGJUuWMH78eP3yezHYK2D/etj5T/j5P1CW53qqwDeKpKBr+d7ekV0F/vyY50Omwx+Ka/5evD3NRLfwY2gLY2hZdAtfoit7Z1opzIiIiDRJ5xR0Fi5cyJw5c0hLS6NXr168/vrr1X75PdlHH33E7Nmz2bt3L+Xl5XTq1Ik//vGP3HvvvedVeK08fY2eFXfw9G2wlzp19bRp06bxxRdfMHfuXDp27IiPjw+33347NpvtzCV5elY7NplMZwwlNZ3vdDrrWX11c+bM4dVXX2X+/Pn06NEDPz8/pk6d6qrdx+fUNaiqO9vzZrP5tBrLy8tPO+/Un+mBAwe46aabePjhh3nhhRcIDQ3lm2++YdKkSdhsNnx9fc/63ldccQW9evVi6dKlXHfddezatYtVq1ad8Ro5Dw4HHN6Cc8c/cexaiaUk2/VUpqkF/yqP41/2gewo7QDHq4cTLw8jzJwYZmYEmqowYzYrzIiIiDQn9Q4677//PvHx8bz55pvExcUxf/58RowYQXJyMq1atTrt/NDQUJ566im6du2K1WrlP//5DxMmTKBVq1aMGDGiQT5ENSZTgw0fa0w2bdrE+PHjXUPGCgsLOXDgwEWtISgoiPDwcLZt28aQIUMAo7fmhx9+oHfv3rVet2nTJm6++WbuuecewFh4YM+ePVx22WWAMaTMx8eHtWvXct999512fc+ePfnrX//K8ePHa+zVadmyJTt37qzWlpSUdFpoO9X27dtxOBy88sormCt3f1+xYsVp77127VqeffbZWl/nvvvuY/78+Rw9epThw4cTFRV1xveV+skrsnF41yZMuz6izbHVBJdnYMJYpjnbGcCn9jg+sQ/ke2dnnJgJ9bMyKDKALuGBdGzl7wo2EYHeCjMiIiKXkHoHnXnz5nH//fczYcIEAN58801WrVrFO++8wxNPPHHa+afOVXjsscd49913+eabby5M0GmmOnXqxEcffcTIkSMxmUw888wz9Z6M3xCmTJlCYmIiHTt2pGvXrrz++uvk5OSccWhPp06d+PDDD/n2228JCQlh3rx5pKenu4KOt7c306dP5/HHH8dqtTJo0CAyMzPZtWsXkyZNYsyYMcyePZtRo0aRmJhIZGQk//3vf2ndujUDBgzgmmuuYc6cOSxdupQBAwbw3nvvsXPnTq644oozfpaOHTtSXl7O66+/zsiRI9m0aRNvvvlmtXMSEhLo0aMHjzzyCA899BBWq5Wvv/6aO+64g7AwY2vH3/3ud0ybNo23336bpUuXnudP+NJlq3CQklVIcloBP6cWUHDof3TMWMOw8o10N6e7zst3+rDG3o/PTIPIDPsVnSNDuC4igN9HBtAlIkCLAIiIiAhQz6Bjs9nYvn07CQkJrjaz2czw4cPZvHnzWa93Op189dVXJCcn89JLL9V6XllZGWVlVbuF5+fn16fMZmnevHlMnDiRgQMHEhYWxvTp093yc5k+fTppaWmMHTsWi8XCAw88wIgRI7BYLLVe8/TTT5OSksKIESPw9fXlgQceYNSoUeTlVc2peOaZZ/Dw8GDGjBkcO3aMyMhIHnroIcDY7+fzzz/nj3/8IzfccAMVFRVcdtllLFy4EIARI0bwzDPP8Pjjj1NaWsrEiRMZO3YsO3bsOONn6dWrF/PmzeOll14iISGBIUOGkJiYyNixY13ndO7cmc8//5wnn3yS/v374+PjQ1xcXLWlsYOCgrjttttYtWoVo0aNOpcf6yXF6XSSnl/Gz2n5JKcVsDs1n91pBezLLCTSkcZN5s2Msmymq/mwcYEZSrHyP7+BHGt7A55druOKtmHc0sIPD4vZvR9GREREGi2Tsx4TMI4dO0abNm349ttvGTBggKv98ccfZ/369Xz33Xc1XpeXl0ebNm0oKyvDYrHwl7/8hYkTJ9b6PrNmzapxqFBeXh6BgYHV2kpLS9m/fz8dOnTA29v7tGvkwnI4HHTr1o0777yT559/3t3luM21117L5ZdfzmuvvXZBXr+p/jkvKqtgT3oBu08KNLvTCsgrqZpDFc5xbrJsYaTlW3qbU1ztdpMH+W2H4XXFHfhefhN4+df0FiIiInKJyc/PJygoqMZscLKLsupaQEAASUlJFBYWsnbtWuLj44mJial1Cd6EhATi4+Ndx/n5+Zr30EgcPHiQzz//nKFDh1JWVsaCBQvYv38/v/vd79xdmlvk5OSwbt061q1bV20J6kuN3eHkYHaRMeysMtQkpxdwMLvmPa1amgu4JzCJG0zf0rHkf5gw/r3FaTJj6jAUut+GpdtNhPiEXMyPISIiIs1IvYJOWFgYFouF9PT0au3p6elERETUep3ZbHbte9K7d29+/vlnEhMTaw06Xl5e1Xaal8bDbDazZMkSpk2bhtPppHv37nz55Zd069bN3aW5xRVXXEFOTg4vvfQSXbp0cXc5F8XxIhu7U/P5Oa2A5DSjl2ZPegGl5TXPGWsV4EWXiAB6tzQzxPEdXbM+x//oN5hK7VUntRsA3W/DdNnN4H/6oiYiIiIi9VWvoGO1WunTpw9r1651zUVwOBysXbv2jJtWnsrhcFSbgyNNR1RUFJs2bXJ3GY3GxV757mIqq7CzN6OQ3akF7E6rGnaWWVDzf7venma6hBsLAnSNCKRrRABdWlhocfRr2LkYfvwc7Ccthx7ZG7rfBpffAsHqsRUREZGGVe+ha/Hx8YwbN46+ffvSv39/5s+fT1FRkWsVtrFjx9KmTRsSExMBSExMpG/fvsTGxlJWVsann37K3//+d954442G/SQick6cTidHc0vYnVpAcnoBP6caiwSkZBVhd5w+hc9kgnahvkaQiQikW0QAXSMDaRfqi8Vsgooy2PcVJH0IyZ9BeVHVxWFdoMftcPmtENbxIn5KERERudTUO+iMHj2azMxMZsyYQVpaGr1792b16tWEh4cDcOjQIdeeJABFRUU88sgjHDlyBB8fH7p27cp7773H6NGjG+5TiEid5JeWs6dyHk1yWr4RbtIKKCirqPH8YF9PuoQH0C2ysocmIoDO4QH4eZ3yV4e9Avavg50fws//htK8k16kvdFz0/02CL/cSEoiIiIiF1i9Vl1zlzOtrHBiNaro6Oiz7mIv0lQVFxdz8ODBOq+6VmF3sD+rqHK4Wb5rb5qjuSU1nu9pMRHb0p+ulb0zXSuHn4UHnmFPGocDjmyFnf+EXSuhKLPqOf8I6H6rEW7a9FG4ERERkQbTqFZdu5A8PT0xmUxkZmbSsmVLbRQozYrT6cRms5GZmYnZbMZqtZ72fGZhmatn5ufKXpq9mYXYKmpeHCAyyLtq2FmkEWg6hPlh9ajDnjROJ6T+WBVu8g5XPecTCpfdbISb9gPBXPveSiIiIiIXWpMPOhaLhbZt23LkyJFmPTFcLm2+vr60imzDzmP5lYsDVC0QcLzIVuM1flYLnSt7ZrpFBtAl3Hgc5OtZ/wIyk41ws/OfkL23qt0aAN1uMsJNzDCwnMNri4iIiFwATT7oAPj7+9OpUyfKy8vPfrJII+dwOEnNK2F/VhH7MotIzijixyPppGTvpKaBpmYTRIf5uYabnbhvG+KD2XwePZw5B2DnR0a4Sd9Z1e7hDZ2vN8JNp+vAs+lsYCoiIiKXjmYRdMDo2bFYNFRGmpbcYhu704xhZ7vT8vk51diTpthmr/H8Fn5WulYON+sSEUC3iEA6hfvj7dlAf/bzU+Gnj2HHh3D0+6p2syd0vNYIN11+A14BDfN+IiIiIhdIswk6Io2ZrcJBSpaxJ83PlYsD7E4tIC2/tMbzrR5mOrXyrxp2VtlL0zLgAmykW3wcfvqX0XNz4BugstvIZIbowUa46TYSfEMb/r1FRERELhAFHZEG5HQ6ScsvrTaPJjmtgL0ZhVTUsCcNQNsQn6ohZ5EBdI0IILqFHx6WOiwOcK5K8yH5UyPc7PsKHCctLx0VZ4Sby0ZBQPiFq0FERETkAlLQETlHRWUVJKcXVK54ll+5N00BeSU1zxUL8PKoPuws0tiTJsD7Ik3gLy+BPWuMcPPL51BxUm9SRE8j3Fx+C4S0vzj1iIiIiFxACjoiZ2F3ODmYXbknTWp+ZU9NAYeOF9d4vsVsIralH11cCwMYe9O0DvK++MufV9gg5Wtjzk3yp2ArrHquRSfocTtcfiu07Hxx6xIRERG5wBR0RE6SXVjmCjK7U/NJTjcWBygtr3lPmlYBXidtsGnMpenYyh8vDzcujOGwG3Ntdv4Tfv4ESnKqngtqV7WRZ0QPbeQpIiIizZaCjlzSSsvtfPzfo6zakcrutAIyC8pqPM/b0+zah6aLay5NIKF+1hrPv+icTjiyrWojz8L0quf8w40had1vg7b9FG5ERETkkqCgI5ekvOJy3vvuIIs3HSCrsCrcmEzQPtTXtcrZiWFn7UJ9sZzPnjQXgtMJaTsqN/L8CPIOVT3nHQyX3WyEm+irwKyl10VEROTSoqAjl5QjOcW8880Blm875NqrpnWQN2MHRhPXIZTO4QH4eTXy/yyyfqkMN/+ErD1V7VZ/6HqjEW5irgaPRtLbJCIiIuIGjfw3OpGGsetYHos2pPCf/6Vir1zmuWtEAA8OjeGmnq3xvJBLOTeE3ENGr83OD41enBMsXtB5hBFuOl0HVl/31SgiIiLSiCjoSLPldDr5Zm8WizaksPGXLFf7oI4teGBILEM6hV38VdDqoyAdfvrYWDHtyNaqdrMHxF5jhJsuN4B3oNtKFBEREWmsFHSk2Sm3O1j1v1QWbUjhp9R8wFjy+YYekTw4JIbubYLcXOEZFB+Hn/9tDEs7sBGcJ1Z7MxlzbbrfBt1+C34t3FqmiIiISGOnoCPNRlFZBcu3Headb/ZzNLcEAB9PC6P7RTHpqg5EhTbSYV1lBZD8mRFu9q4Fx0kbjrbtZ4Sby0ZBYKTbShQRERFpahR0pMnLKCjl3W8P8PfNB8kvrQAgzN/KuAHR3POr9oQ0liWgT1ZeAr98YYSbPWugoqTqufAelXvd3Aoh0W4rUURERKQpU9CRJmtfZiFvb0jhox+OYrMbQ7xiwvy4b3AMt17ZBm/PRraksr0cUtZVbuT5H7AVVD0XGgs9bofLb4VWXd1WooiIiEhzoaAjTc73B47z1oYUvvipalPMK9sF8+DQWH7dLRxzY9rvxmGHg98a4eanf0HJ8arnAttW9tzcBpG9tJGniIiISANS0JEmweFw8vlP6SzasI8fDuW62n99WTgPDomhb3So+4o7ldMJR7cb4WbXSihIrXrOryVcfosRbtr2B3MjX9ZaREREpIlS0JFGrbTczkc/HOWvG1NIySoCwGoxc+uVbbhvcAwdW/m7ucJKTiek76rayDP3YNVz3kHGSmndb4PowWDRf3YiIiIiF5p+45JGKbfYxt83H+TdzQfIKrQBEOjtwb0D2jNuYDStArzdXGGl7H1GsNnxIWQlV7V7+kHXG4xwE3sNeHi5r0YRERGRS5CCjjQqh48X87dv9rPi+8MU2+wAtAn2YeJVHRjdLwp/r0bwRzbvCOz8CHZ+CKk/VrVbrNDpOiPcdB4BVj/31SgiIiJyiWsEvzWKwM6jeby1IYVPd6RidzgB6BYZyINDYrixZySeFjfPZSnMMBYT2PlPOLS5qt1kgZhhxoppXW80hqmJiIiIiNsp6IjbOJ1ONvySxaIN+9i0N9vVPrhTGA8MieGqjmGY3LkSWUmOsQz0zn/C/vXgdFQ+YYL2Ays38rwZ/MLcV6OIiIiI1EhBRy66cruD//zvGG+tT2F3mrGXjMVs4qaekdw/OIbubdzYK1JWCHtWG+Hmly/AUV71XJs+leFmFAS1cVuJIiIiInJ2Cjpy0RSWVbB86yHe+WY/x/JKAfC1WrirXzsmXhVN2xBf9xRmL4c9a4xws2c1lBdXPdfqMiPcdL8VQmPcU5+IiIiI1JuCjlxwGfmlLP72AO9tOUhBaQUAYf5eTBgUzT1x7Qny9XRPYU4n/PI5rHkSsvdWtYd0MObcXH4rhF/mntpERERE5Lwo6MgFszejgEUbUvj4v8ew2Y35LTEt/XhgcAyjrmiDt6fFfcVl7DYCzr61xrFvC+g1xui9aX0FuHNukIiIiIicNwUdaVBOp5NtB3JYtGEfX/6c4Wrv2z6EB4bEMLxbOGazG0NE8XFY9yJs+ys47WD2hF89DEP+D7wD3VeXiIiIiDQoBR1pEHaHky9+SuPN9SkkHc4FjE6R6y4L54EhsfRpH+LmAivg+3dg3WxjNTWALjfCdc9Di1j31iYiIiIiDe6cNidZuHAh0dHReHt7ExcXx9atW2s99+2332bw4MGEhIQQEhLC8OHDz3i+NC2l5Xbe23KQa19Zx0Pv/UDS4VysHmbG9G/H2vihvHVvX/eHnL1r4c1B8Nn/GSGn1WUw9l8wZplCjoiIiEgzVe8enffff5/4+HjefPNN4uLimD9/PiNGjCA5OZlWrVqddv66desYM2YMAwcOxNvbm5deeonrrruOXbt20aaNluhtqnKKbPx9y0He/fYA2UU2AIJ8PLn3V+0ZNzCalgFebq4QyNoLnz9lrKQG4BMK1zwFV44HizozRURERJozk9PpdNbngri4OPr168eCBQsAcDgcREVFMWXKFJ544omzXm+32wkJCWHBggWMHTu2Tu+Zn59PUFAQeXl5BAZqHoU7HT5ezF83prDi+yOUlNsBaBPsw32DO3Bn3yj8vBpBgCjJhQ1z4Lu3jH1wzB7Q/wEY+jj4uLl3SURERETOS12zQb1+K7XZbGzfvp2EhARXm9lsZvjw4WzevLlOr1FcXEx5eTmhoaG1nlNWVkZZWZnrOD8/vz5lygWw40geb23Yx6c7UnFURuPLWwfywJAYbuwRiYflnEZBNiyHHX54F756AYqzjLZO18F1L0DLzu6tTUREREQuqnoFnaysLOx2O+Hh4dXaw8PD2b17d51eY/r06bRu3Zrhw4fXek5iYiLPPvtsfUqTC8DpdLJuTyaL1qewOSXb1T6kc0seHBLDwNgWmBrLMsz7N8DqBEjfaRyHdYYRidCp9j9nIiIiItJ8XdRxRi+++CLLly9n3bp1eHt713peQkIC8fHxruP8/HyioqIuRokC2Coc/PvHYyzakEJyegEAHmYTI3u15v7BMVzWuhENHzy+H754Bn7+t3HsHQTDnoR+k8Dipo1IRURERMTt6hV0wsLCsFgspKenV2tPT08nIiLijNfOnTuXF198kS+//JKePXue8VwvLy+8vBrBZPZLTEFpOcu3HuZv3+wnLb8UAD+rhTH92zHxqg60DvZxc4UnKSuAja/A5oVgt4HJDH0nGiHHr4W7qxMRERERN6tX0LFarfTp04e1a9cyatQowFiMYO3atUyePLnW615++WVeeOEF1qxZQ9++fc+rYGl46fmlvLNpP8u2HKKgrAKAlgFeTBgUzd1x7QnyaUQ9Iw4H/LgM1j4HhZWBO2aYMUwt/DK3liYiIiIijUe9h67Fx8czbtw4+vbtS//+/Zk/fz5FRUVMmDABgLFjx9KmTRsSExMBeOmll5gxYwbLli0jOjqatLQ0APz9/fH392/AjyL19Ut6AYs2pPBx0lHK7cYKA7Et/XhwSCw3X9EaLw+Lmys8xcHNsHo6pP5oHIfGwIjZ0Pl6Y3dSEREREZFK9Q46o0ePJjMzkxkzZpCWlkbv3r1ZvXq1a4GCQ4cOYTZXrcD1xhtvYLPZuP3226u9zsyZM5k1a9b5VS/15nQ6+W7/cRZtSOGr3Rmu9v7RoTwwJIZrurbCbG5koSH3MHwxA3Z9ZBx7BRpLRfd/EDys7q1NRERERBqleu+j4w7aR+f82R1O1uxK460NKfx4OBcwOkFGXBbBA0NjuLJdI9xfxlYE38yHb1+DilLABFeOhWueAf+W7q5ORERERNzgguyjI01Pic3Oh9sP89dv9nMwuxgAq4eZO/q05b7BMXQI83NzhTVwOGDHB/DlLCg4ZrS1vwquT4TIMy9kISIiIiICCjrN1vEiG0s3H2Dp5oMcL7IBEOzrydhftWfswGjC/BvpqnZHvofPpsPR743j4HZw3Z+g2281D0dERERE6kxBp5k5mF3EXzfu54PthyktdwDQNsSH+wfHcEfftvhaG+lXnn/M6MH53/vGsacfDPkj/OpR8Kx9zyURERERkZo00t96pb5+PJzLog0pfLYzFUflrKsebYJ4YEgMv+kegYfFfOYXcJfyEvj2dfjmz1BuDK2j991w7QwIOPPeTCIiIiIitVHQacIcDifr9mTw1voUvtt/3NU+tHNLHhwaw4CYFpga63Avp9NYRe2LmZB32GiL+pUxD6fNle6tTURERESaPAWdJshW4eBfSUd5e2MKe9ILAfAwm/ht79Y8MCSGrhGNfGW6Y/+F1QlwaLNxHNgWfv0sdL9N83BEREREpEEo6DQh+aXl/L/vDrF40wHS8ksB8PfyYEz/KCYM6kDrYB83V3gWBemw9jlI+gfgBA8fuOoPMHAKWH3dXZ2IiIiINCMKOk1Aal4JizcdYNl3hygsqwAgPNCLCYM68Lu4dgR6e7q5wrMoL4Utf4GNr4DN6IGix50wfBYEtXFraSIiIiLSPCnoNGLJaQUs2pDCJz8epdxurDDQqZU/9w+J4eberfHysLi5wrNwOuHnf8PnT0PuQaOtTR+4/iWI6ufe2kRERESkWVPQaWScTiebU7JZtCGFdcmZrvb+HUJ5aGgMwzq3wmxuAvNY0nYY83AObDSOAyKNHpwed4K5ka4AJyIiIiLNhoJOI1Fhd7B6VxqLNqTwvyN5AJhNcH33CB4YEkvvqGD3FlhXRVnw1fPww1JwOsDD25iDM2gqePm7uzoRERERuUQo6LhZsa2CD74/wl+/SeHw8RIAvDzM3NG3LfddFUN0mJ+bK6yjChtsfQvWvwxl+Ubb5bfAr5+D4HburU1ERERELjkKOm6SXVjGu5sP8vfNB8gpLgcgxNeTsQOiGTugPS38vdxcYR05nbBnDax5Eo7vM9oie8H1L0L7ge6tTUREREQuWQo6F9mBrCLe3pjCh9uPUFbhAKBdqC/3De7AHX2i8LE28gUGTpaxG9YkwL6vjGO/lnDtTOj9OzA3oc8hIiIiIs2Ogs5F8t9DOSzakMLqXWk4jQXU6NU2iAeGxHJ99wgsTWGBgROKj8O6RNj2N3DawWKFXz0Mg6eBdyPfrFRERERELgkKOheQw+Hk6+QM3tqQwtb9x13tV3dpyQNDYvlVTCgmUxMKOPZy+P4d+Ho2lOYabV1vguueh9AYt5YmIiIiInIyBZ0LoKzCzr/+e4xFG1PYm2FskOlpMXFz7zY8MCSGzuEBbq7wHOz9ElY/CVnJxnF4dxgxG2KGurcuEREREZEaKOg0oLyScpZ9d4jFm/aTUVAGQICXB7+La8eEQR2ICPJ2c4XnIOsXWPMU/LLGOPZtAdc8DVeO0zwcEREREWm0FHQawLHcEt75Zj/Ltx2msKwCgPBALyYO6sCYuHYEenu6ucJzUJJrLBW99S1wVIDZA/o/CEMfB59gd1cnIiIiInJGCjrn4efUfN7ekMInPx6jwmGsMNA53J8HhsTy216tsXqY3VzhObBXwA/vwtcvQHG20dZpBIx4AcI6ubc2EREREZE6UtCpJ6fTyeZ92by5IYUNezJd7b+KCeXBIbEM69KyaS0wcLKU9bA6ATJ2GcdhXeD62dBxuHvrEhERERGpJwWdekhOK+CPHySx82g+AGYT/KZHJA8OiaFn22D3Fnc+jqfA58/A7v8Yx97BcPWT0HciWJrgsDsRERERueQp6NRDywAv9mYU4u1p5s6+Udx3VQztWvi6u6xzV5oPG+fCljfAbgOTBfpNgmEJ4Bvq7upERERERM6Zgk49hPpZeeOePvRqG0yon9Xd5Zw7hx2S/gFrn4eiDKMt9hpjuehW3dxbm4iIiIhIA1DQqaeru7Rydwnn5+C38Nl0SPufcRwaawScziOgqc4tEhERERE5hYLOpSLnIHwxA3762Dj2CjKWiu7/AHg04d4pEREREZEaKOg0d2WF8M2f4dvXwV4GJrOx2ec1T4NfmLurExERERG5IBR0miuHA3asgC9nQUGq0RY9GK5/ESK6u7U0EREREZELTUGnOTq8DVZPh6PbjeOQaLjuT9D1Js3DEREREZFLgoJOc5J3FL6cCTs+MI6t/jBkGsQ9DJ7e7q1NREREROQiUtBpDmzF8O1r8M18qCgBTHDF3XDNDAgId3d1IiIiIiIXnflcLlq4cCHR0dF4e3sTFxfH1q1baz13165d3HbbbURHR2MymZg/f/651iqncjphx4ewoB+sSzRCTrsB8MDXcPNChRwRERERuWTVO+i8//77xMfHM3PmTH744Qd69erFiBEjyMjIqPH84uJiYmJiePHFF4mIiDjvgqXS0R/gnevhn5Mg/wgERcHti2HCZ9D6CndXJyIiIiLiVian0+mszwVxcXH069ePBQsWAOBwOIiKimLKlCk88cQTZ7w2OjqaqVOnMnXq1HoVmZ+fT1BQEHl5eQQGBtbr2manIA3WPgdJ/zCOPX3hqj/AwCng6ePe2kRERERELrC6ZoN6zdGx2Wxs376dhIQEV5vZbGb48OFs3rz53Ks9RVlZGWVlZa7j/Pz8BnvtJqu8FLYshI3zwFZotPW8C4bPhMDW7q1NRERERKSRqVfQycrKwm63Ex5efe5HeHg4u3fvbrCiEhMTefbZZxvs9Zo0pxN+/gQ+fwZyDxptbfsZ++G07eve2kREREREGqlzWozgQktISCAvL891O3z4sLtLco/U/8GSm2DFWCPkBLSGW9+GiZ8r5IiIiIiInEG9enTCwsKwWCykp6dXa09PT2/QhQa8vLzw8vJqsNdrcgoz4avn4YelgBM8vGHg7+GqqWD1c3d1IiIiIiKNXr16dKxWK3369GHt2rWuNofDwdq1axkwYECDF3fJqbDBptfg9Svhh3cBJ1x+K0zeBtc8pZAjIiIiIlJH9d4wND4+nnHjxtG3b1/69+/P/PnzKSoqYsKECQCMHTuWNm3akJiYCBgLGPz000+ux0ePHiUpKQl/f386duzYgB+lCXM6Ifkz+PwpOJ5itEX2NubhtFeAFBERERGpr3oHndGjR5OZmcmMGTNIS0ujd+/erF692rVAwaFDhzCbqzqKjh07xhVXVO3rMnfuXObOncvQoUNZt27d+X+Cpi79J1iTACnrjGP/cLh2BvT6HZgb5RQqEREREZFGr9776LhDs9xHpygb1s2G798BpwMsVhjwKAz+I3gFuLs6EREREZFG6YLsoyMNwF4O2/4K6xKhNM9o6zYSfv08hHZwb20iIiIiIs2Egs7F9MsXsOZJyNpjHId3N+bhdBjs3rpERERERJoZBZ2LIXOPEXD2fmEc+4bBNU/DlWPBbHFvbSIiIiIizZCCzoVUkgPrXoJtb4OjAsyeEPcgDH0cvIPcXZ2IiIiISLOloHMh2CvghyXw1QtQctxo6/wbGPECtIh1a2kiIiIiIpcCBZ2Gtu9rY5hahrF3EC27wvWJEHuNe+sSEREREbmEKOg0lOx98PnTkPypcewTAlc/BX0mgEU/ZhERERGRi0m/gZ+v0nzYMAe2vAGOcjBZoP/9MHQ6+Ia6uzoRERERkUuSgs65ctjhv+/BV89DUabR1nE4jJgNLbu4tzYRERERkUucgs65OLAJVk+HtB3GcYtORsDpfJ176xIREREREUBBp37yjhgLDfz0L+PYKwiGPQH97gMPq3trExERERERFwWd+igvgd2rwGQ2Fhm4+inwa+HuqkRERERE5BQKOvUR1glufAXa9oPwy91djYiIiIiI1EJBp776jHd3BSIiIiIichZmdxcgIiIiIiLS0BR0RERERESk2VHQERERERGRZqdJzNFxOp0A5Ofnu7kSERERERFxpxOZ4ERGqE2TCDoFBQUAREVFubkSERERERFpDAoKCggKCqr1eZPzbFGoEXA4HBw7doyAgABMJpNba8nPzycqKorDhw8TGBjo1lqkYeg7bZ70vTY/+k6bJ32vzY++0+ansX2nTqeTgoICWrdujdlc+0ycJtGjYzabadu2rbvLqCYwMLBRfNHScPSdNk/6XpsffafNk77X5kffafPTmL7TM/XknKDFCEREREREpNlR0BERERERkWZHQaeevLy8mDlzJl5eXu4uRRqIvtPmSd9r86PvtHnS99r86Dttfprqd9okFiMQERERERGpD/XoiIiIiIhIs6OgIyIiIiIizY6CjoiIiIiINDsKOiIiIiIi0uwo6NTDwoULiY6Oxtvbm7i4OLZu3erukuQ8bNiwgZEjR9K6dWtMJhMff/yxu0uS85SYmEi/fv0ICAigVatWjBo1iuTkZHeXJefpjTfeoGfPnq6N6gYMGMBnn33m7rKkAb344ouYTCamTp3q7lLkPMyaNQuTyVTt1rVrV3eXJefp6NGj3HPPPbRo0QIfHx969OjB999/7+6y6kRBp47ef/994uPjmTlzJj/88AO9evVixIgRZGRkuLs0OUdFRUX06tWLhQsXursUaSDr16/n0UcfZcuWLXzxxReUl5dz3XXXUVRU5O7S5Dy0bduWF198ke3bt/P9999zzTXXcPPNN7Nr1y53lyYNYNu2bbz11lv07NnT3aVIA7j88stJTU113b755ht3lyTnIScnh0GDBuHp6clnn33GTz/9xCuvvEJISIi7S6sTLS9dR3FxcfTr148FCxYA4HA4iIqKYsqUKTzxxBNurk7Ol8lkYuXKlYwaNcrdpUgDyszMpFWrVqxfv54hQ4a4uxxpQKGhocyZM4dJkya5uxQ5D4WFhVx55ZX85S9/4U9/+hO9e/dm/vz57i5LztGsWbP4+OOPSUpKcncp0kCeeOIJNm3axMaNG91dyjlRj04d2Gw2tm/fzvDhw11tZrOZ4cOHs3nzZjdWJiJnkpeXBxi/FEvzYLfbWb58OUVFRQwYMMDd5ch5evTRR7nxxhur/f9VmrZffvmF1q1bExMTw913382hQ4fcXZKch08++YS+fftyxx130KpVK6644grefvttd5dVZwo6dZCVlYXdbic8PLxae3h4OGlpaW6qSkTOxOFwMHXqVAYNGkT37t3dXY6cpx07duDv74+XlxcPPfQQK1eu5LLLLnN3WXIeli9fzg8//EBiYqK7S5EGEhcXx5IlS1i9ejVvvPEG+/fvZ/DgwRQUFLi7NDlHKSkpvPHGG3Tq1Ik1a9bw8MMP8/vf/553333X3aXViYe7CxARuRAeffRRdu7cqfHhzUSXLl1ISkoiLy+PDz/8kHHjxrF+/XqFnSbq8OHDPPbYY3zxxRd4e3u7uxxpIL/5zW9cj3v27ElcXBzt27dnxYoVGmbaRDkcDvr27cvs2bMBuOKKK9i5cydvvvkm48aNc3N1Z6cenToICwvDYrGQnp5erT09PZ2IiAg3VSUitZk8eTL/+c9/+Prrr2nbtq27y5EGYLVa6dixI3369CExMZFevXrx6quvurssOUfbt28nIyODK6+8Eg8PDzw8PFi/fj2vvfYaHh4e2O12d5coDSA4OJjOnTuzd+9ed5ci5ygyMvK0f1Dq1q1bkxmSqKBTB1arlT59+rB27VpXm8PhYO3atRojLtKIOJ1OJk+ezMqVK/nqq6/o0KGDu0uSC8ThcFBWVubuMuQcXXvttezYsYOkpCTXrW/fvtx9990kJSVhsVjcXaI0gMLCQvbt20dkZKS7S5FzNGjQoNO2adizZw/t27d3U0X1o6FrdRQfH8+4cePo27cv/fv3Z/78+RQVFTFhwgR3lybnqLCwsNq/Mu3fv5+kpCRCQ0Np166dGyuTc/Xoo4+ybNky/vWvfxEQEOCaQxcUFISPj4+bq5NzlZCQwG9+8xvatWtHQUEBy5YtY926daxZs8bdpck5CggIOG3unJ+fHy1atNCcuiZs2rRpjBw5kvbt23Ps2DFmzpyJxWJhzJgx7i5NztEf/vAHBg4cyOzZs7nzzjvZunUrixYtYtGiRe4urU4UdOpo9OjRZGZmMmPGDNLS0ujduzerV68+bYECaTq+//57rr76atdxfHw8AOPGjWPJkiVuqkrOxxtvvAHAsGHDqrUvXryY8ePHX/yCpEFkZGQwduxYUlNTCQoKomfPnqxZs4Zf//rX7i5NRE5y5MgRxowZQ3Z2Ni1btuSqq65iy5YttGzZ0t2lyTnq168fK1euJCEhgeeee44OHTowf/587r77bneXVifaR0dERERERJodzdEREREREZFmR0FHRERERESaHQUdERERERFpdhR0RERERESk2VHQERERERGRZkdBR0REREREmh0FHRERERERaXYUdEREREREpNlR0BERERERkWZHQUdERERERJodBR0REREREWl2FHRERERERKTZ+f+UKRg0qTQGnAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 1000x200 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(10, 2))\n",
    "plt.plot(history_auto[\"val_accuracy\"], label=\"Validation accuracy\")\n",
    "plt.plot(history_auto[\"accuracy\"], label=\"Training accuracy\")\n",
    "plt.legend()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "id": "e90e90f9",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "dict_keys(['loss', 'accuracy', 'val_loss', 'val_accuracy'])\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "print(history_auto.keys())\n",
    "#  \"Accuracy\"\n",
    "plt.plot(history_auto['accuracy'])\n",
    "plt.plot(history_auto['val_accuracy'])\n",
    "plt.title('Auto Encoder model accuracy')\n",
    "plt.ylabel('accuracy')\n",
    "plt.xlabel('epoch')\n",
    "plt.legend(['train', 'validation'], loc='upper left')\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "6b5c22a3",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "id": "550159a6",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "78/78 [==============================] - 0s 1ms/step\n"
     ]
    }
   ],
   "source": [
    "reconstructed_news_1 = autoencoder.predict(X_test_N)\n",
    "reconstruction_errors_1 = np.mean((X_test_N - reconstructed_news_1) ** 2, axis=1)\n",
    "threshold_1 = np.percentile(reconstruction_errors_1, 99)\n",
    "anomaly_indices_1 = np.where(reconstruction_errors_1 > threshold_1)[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "id": "e02e1573",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "248/248 [==============================] - 0s 1ms/step\n"
     ]
    }
   ],
   "source": [
    "reconstructed_news_2 = autoencoder.predict(X_2_N)\n",
    "reconstruction_errors_2 = np.mean((X_2_N - reconstructed_news_2) ** 2, axis=1)\n",
    "threshold_2 = np.percentile(reconstruction_errors_2, 99)\n",
    "anomaly_indices_2 = np.where(reconstruction_errors_2 > threshold_2)[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "id": "9a805d7e",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "248/248 [==============================] - 0s 1ms/step\n"
     ]
    }
   ],
   "source": [
    "reconstructed_news_3 = autoencoder.predict(X_3_N)\n",
    "reconstruction_errors_3 = np.mean((X_3_N - reconstructed_news_3) ** 2, axis=1)\n",
    "threshold_3 = np.percentile(reconstruction_errors_3,99)\n",
    "anomaly_indices_3 = np.where(reconstruction_errors_3 > threshold_3)[0]\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "id": "4cfc97ec",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "AutoEncoder Accuracy Dataset #1 = 0.4814665592264303\n",
      "AutoEncoder Accuracy Dataset #2 = 0.49798336274262667\n",
      "AutoEncoder Accuracy Dataset #3 = 0.4948271511481201\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import accuracy_score, precision_score, recall_score\n",
    "\n",
    "\n",
    "def predict(model_test, data, threshold):\n",
    "    reconstructions = model_test(data)\n",
    "    loss = tf.keras.losses.mse(reconstructions, data)\n",
    "    return tf.math.greater(loss, threshold)\n",
    "\n",
    "\n",
    "preds_auto_1 = predict(autoencoder, X_test_N, threshold_1)\n",
    "print(\"AutoEncoder Accuracy Dataset #1 = {}\".format(accuracy_score(preds_auto_1,y_test)))\n",
    "\n",
    "preds_auto_2 = predict(autoencoder, X_2_N, threshold_2)\n",
    "print(\"AutoEncoder Accuracy Dataset #2 = {}\".format(accuracy_score(preds_auto_2,y_2)))\n",
    "\n",
    "preds_auto_3 = predict(autoencoder, X_3_N, threshold_3)\n",
    "print(\"AutoEncoder Accuracy Dataset #3 = {}\".format(accuracy_score(preds_auto_3,y_3)))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "id": "7c21a308",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Combined Approach Accuracy: 0.9271\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import accuracy_score\n",
    "\n",
    "anomaly_predictions_1 = np.zeros_like(y_pred_LSTM) # return zro sanme shape as y_pred_LSTM\n",
    "aTemp = anomaly_indices_1\n",
    "anomaly_predictions_1[aTemp] = 1 \n",
    "combined_predictions_1 = y_pred_LSTM | anomaly_predictions_1\n",
    "\n",
    "\n",
    "combined_accuracy = accuracy_score(combined_predictions_1, y_test)\n",
    "print(f\"Combined Approach Accuracy: {combined_accuracy:.4f}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "id": "1f8fece4",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Combined Approach Accuracy: 0.8271\n"
     ]
    }
   ],
   "source": [
    "anomaly_predictions_2 = np.zeros_like(y_pred_LSTM_2) \n",
    "aTemp_2 = anomaly_indices_2\n",
    "anomaly_predictions_2[aTemp_2] = 1\n",
    "combined_predictions_2 = y_pred_LSTM_2 | anomaly_predictions_2\n",
    "\n",
    "\n",
    "# Calculate combined accuracy\n",
    "combined_accuracy_2 = accuracy_score(combined_predictions_2, y_2)\n",
    "print(f\"Combined Approach Accuracy: {combined_accuracy_2:.4f}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "id": "a0248939",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Combined Approach Accuracy: 0.8063\n"
     ]
    }
   ],
   "source": [
    "anomaly_predictions_3 = np.zeros_like(y_pred_LSTM_3)\n",
    "aTemp_3 = anomaly_indices_3\n",
    "anomaly_predictions_3[aTemp_3] = 1\n",
    "combined_predictions_3= y_pred_LSTM_3 | anomaly_predictions_3\n",
    "\n",
    "\n",
    "# Calculate combined accuracy\n",
    "combined_accuracy_3 = accuracy_score(combined_predictions_3, y_3)\n",
    "print(f\"Combined Approach Accuracy: {combined_accuracy_3:.4f}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "id": "52328986",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.80      0.82      0.81      3906\n",
      "           1       0.82      0.80      0.81      4020\n",
      "\n",
      "    accuracy                           0.81      7926\n",
      "   macro avg       0.81      0.81      0.81      7926\n",
      "weighted avg       0.81      0.81      0.81      7926\n",
      "\n"
     ]
    }
   ],
   "source": [
    "#Classification report for Combined -Dataset #3\n",
    "print(classification_report(y_3,combined_predictions_3))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "id": "9f9baaf1",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.79      0.82      0.81      3906\n",
      "           1       0.82      0.79      0.81      4020\n",
      "\n",
      "    accuracy                           0.81      7926\n",
      "   macro avg       0.81      0.81      0.81      7926\n",
      "weighted avg       0.81      0.81      0.81      7926\n",
      "\n"
     ]
    }
   ],
   "source": [
    "#Classification report for LSTM - Dataset #3\n",
    "print(classification_report(y_3,y_pred_LSTM_3))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "id": "211d271c",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.49      0.99      0.66      3906\n",
      "           1       0.60      0.01      0.02      4020\n",
      "\n",
      "    accuracy                           0.49      7926\n",
      "   macro avg       0.55      0.50      0.34      7926\n",
      "weighted avg       0.55      0.49      0.34      7926\n",
      "\n"
     ]
    }
   ],
   "source": [
    "#Classification report for AutoEncoder - Threshold Datast #3\n",
    "print(classification_report(y_3,preds_auto_3))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "id": "891e7736",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.49      0.99      0.66      3906\n",
      "           1       0.60      0.01      0.02      4020\n",
      "\n",
      "    accuracy                           0.49      7926\n",
      "   macro avg       0.55      0.50      0.34      7926\n",
      "weighted avg       0.55      0.49      0.34      7926\n",
      "\n"
     ]
    }
   ],
   "source": [
    "#Classification report for AutoEncoder - Anomaly Dataset #3\n",
    "print(classification_report(y_3,anomaly_predictions_3))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "id": "50a63f44",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.82      0.83      0.82      3881\n",
      "           1       0.83      0.83      0.83      4053\n",
      "\n",
      "    accuracy                           0.83      7934\n",
      "   macro avg       0.83      0.83      0.83      7934\n",
      "weighted avg       0.83      0.83      0.83      7934\n",
      "\n"
     ]
    }
   ],
   "source": [
    "#Classification report for Combined - Dataset #2\n",
    "print(classification_report(y_2,combined_predictions_2))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "id": "605a1213",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.82      0.83      0.82      3881\n",
      "           1       0.83      0.82      0.83      4053\n",
      "\n",
      "    accuracy                           0.83      7934\n",
      "   macro avg       0.83      0.83      0.83      7934\n",
      "weighted avg       0.83      0.83      0.83      7934\n",
      "\n"
     ]
    }
   ],
   "source": [
    "#Classification report for AutoEncoder - LSTM Dataset #2\n",
    "print(classification_report(y_2,y_pred_LSTM_2))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "id": "1afb80a5",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.49      1.00      0.66      3881\n",
      "           1       0.94      0.02      0.04      4053\n",
      "\n",
      "    accuracy                           0.50      7934\n",
      "   macro avg       0.72      0.51      0.35      7934\n",
      "weighted avg       0.72      0.50      0.34      7934\n",
      "\n"
     ]
    }
   ],
   "source": [
    "#Classification report for AutoEncoder - Threshold Dataset #2\n",
    "print(classification_report(y_2,preds_auto_2))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "id": "ad20663c",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.49      1.00      0.66      3881\n",
      "           1       0.94      0.02      0.04      4053\n",
      "\n",
      "    accuracy                           0.50      7934\n",
      "   macro avg       0.72      0.51      0.35      7934\n",
      "weighted avg       0.72      0.50      0.34      7934\n",
      "\n"
     ]
    }
   ],
   "source": [
    "#Classification report for AutoEncoder - Anomaly Dataset #2\n",
    "print(classification_report(y_2,anomaly_predictions_2))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "id": "4e1d38de",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.94      0.91      0.92      1174\n",
      "           1       0.92      0.94      0.93      1308\n",
      "\n",
      "    accuracy                           0.93      2482\n",
      "   macro avg       0.93      0.93      0.93      2482\n",
      "weighted avg       0.93      0.93      0.93      2482\n",
      "\n"
     ]
    }
   ],
   "source": [
    "#Classification report for Combined -Dataset # 1\n",
    "print(classification_report(y_test,combined_predictions_1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "id": "6ec86fc6",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.82      0.83      0.82      3881\n",
      "           1       0.83      0.82      0.83      4053\n",
      "\n",
      "    accuracy                           0.83      7934\n",
      "   macro avg       0.83      0.83      0.83      7934\n",
      "weighted avg       0.83      0.83      0.83      7934\n",
      "\n"
     ]
    }
   ],
   "source": [
    "#Classification report for LSTM - Dataset #2\n",
    "print(classification_report(y_2,y_pred_LSTM_2))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "id": "54057704",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.48      1.00      0.65      1174\n",
      "           1       0.92      0.02      0.03      1308\n",
      "\n",
      "    accuracy                           0.48      2482\n",
      "   macro avg       0.70      0.51      0.34      2482\n",
      "weighted avg       0.71      0.48      0.32      2482\n",
      "\n"
     ]
    }
   ],
   "source": [
    "#Classification report for AutoEncoder - Threshold Dataset #1\n",
    "print(classification_report(y_test,preds_auto_1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "id": "095bf742",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.48      1.00      0.65      1174\n",
      "           1       0.92      0.02      0.03      1308\n",
      "\n",
      "    accuracy                           0.48      2482\n",
      "   macro avg       0.70      0.51      0.34      2482\n",
      "weighted avg       0.71      0.48      0.32      2482\n",
      "\n"
     ]
    }
   ],
   "source": [
    "#Classification report for AutoEncoder - Anomaly Approach Dataset #1\n",
    "print(classification_report(y_test,anomaly_predictions_1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "id": "2683f5f3",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "AutoEncoderAnomaly Approach Accuracy for Data set # 1: 0.4815\n",
      "AurtoEncoder Anomaly Approach Accuracy for Data set # 2: 0.4980\n",
      "AutoEncoder Anomaly Approach Accuracy for Data set # 3: 0.4948\n"
     ]
    }
   ],
   "source": [
    "Anomaly_accuracy_1 = accuracy_score(anomaly_predictions_1, y_test)\n",
    "print(f\"AutoEncoderAnomaly Approach Accuracy for Data set # 1: {Anomaly_accuracy_1:.4f}\")\n",
    "Anomaly_accuracy_2 = accuracy_score(anomaly_predictions_2, y_2)\n",
    "print(f\"AurtoEncoder Anomaly Approach Accuracy for Data set # 2: {Anomaly_accuracy_2:.4f}\")\n",
    "Anomaly_accuracy_3 = accuracy_score(anomaly_predictions_3, y_3)\n",
    "print(f\"AutoEncoder Anomaly Approach Accuracy for Data set # 3: {Anomaly_accuracy_3:.4f}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "id": "dddd7891",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(45.722222222222214, 0.5, 'Real Labels')"
      ]
     },
     "execution_count": 89,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 600x300 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize = (6,3))\n",
    "\n",
    "sns.heatmap(confusion_matrix(y_test,preds_auto_1), annot=True, \n",
    "            fmt='', cmap='Blues')\n",
    "\n",
    "plt.xlabel('Predicted Labels')\n",
    "plt.ylabel('Real Labels')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "id": "5b2803c0",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(45.722222222222214, 0.5, 'Real Labels')"
      ]
     },
     "execution_count": 90,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 600x300 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize = (6,3))\n",
    "\n",
    "sns.heatmap(confusion_matrix(y_2,preds_auto_2), annot=True, \n",
    "            fmt='', cmap='Blues')\n",
    "\n",
    "plt.xlabel('Predicted Labels')\n",
    "plt.ylabel('Real Labels')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "id": "f3a9166b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(45.722222222222214, 0.5, 'Real Labels')"
      ]
     },
     "execution_count": 91,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 600x300 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize = (6,3))\n",
    "\n",
    "sns.heatmap(confusion_matrix(y_3,preds_auto_3), annot=True, \n",
    "            fmt='', cmap='Blues')\n",
    "\n",
    "plt.xlabel('Predicted Labels')\n",
    "plt.ylabel('Real Labels')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "id": "91a381b2",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(45.722222222222214, 0.5, 'Real Labels')"
      ]
     },
     "execution_count": 92,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 600x300 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize = (6,3))\n",
    "\n",
    "sns.heatmap(confusion_matrix(y_test,combined_predictions_1), annot=True, \n",
    "            fmt='', cmap='Blues')\n",
    "\n",
    "plt.xlabel('Predicted Labels')\n",
    "plt.ylabel('Real Labels')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 93,
   "id": "2f302609",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(45.722222222222214, 0.5, 'Real Labels')"
      ]
     },
     "execution_count": 93,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 600x300 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize = (6,3))\n",
    "\n",
    "sns.heatmap(confusion_matrix(y_2,combined_predictions_2), annot=True, \n",
    "            fmt='', cmap='Blues')\n",
    "\n",
    "plt.xlabel('Predicted Labels')\n",
    "plt.ylabel('Real Labels')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "id": "ce34254e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(45.722222222222214, 0.5, 'Real Labels')"
      ]
     },
     "execution_count": 94,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAgIAAAEmCAYAAAAQkPhxAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjYuMCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy89olMNAAAACXBIWXMAAA9hAAAPYQGoP6dpAAA2z0lEQVR4nO3deVhUZf8G8HsGmRGBGUT2XJNSSUVFxdFETQTNXBJLcwE3fDFwgVSierHIolczdyUrxUpLrTQVN8KATMQt3FLKRHEb1GQRhGE7vz/8MTmJOoMzjHLuT9e5Xuec55z5nrl8nXue85znSARBEEBERESiJDV3AURERGQ+DAJEREQixiBAREQkYgwCREREIsYgQEREJGIMAkRERCLGIEBERCRiDAJEREQixiBAREQkYvXMXYApWHUMM3cJRCaXe2iZuUsgMrn6Jv6WMuT7ovi3uvn/uToZBIiIiPQiYcc4gwAREYmX1MLcFZgdgwAREYmXRGLuCsyOQYCIiMSLlwYYBIiISMTYI8AgQEREIsYeAQYBIiISMfYIMAgQEZGIsUeAQYCIiESMPQIMAkREJGLsEWAQICIiEeOEQgwCREQkYuwRYBAgIiIRk3KMAKMQERGJl0Sq/2KAlStXon379lAoFFAoFFCpVNi5c6d2e0lJCUJDQ9GoUSPY2NggICAAOTk5OsfIzs7GwIED0aBBAzg5OWHWrFkoLy/XaZOcnIxOnTpBLpfD3d0d8fHxBn8EDAJERCReEon+iwEaN26Mjz76CEeOHMHhw4fxwgsvYMiQITh16hQAIDw8HNu2bcOmTZuQkpKCK1euYNiwYdr9KyoqMHDgQJSWlmL//v1Yu3Yt4uPjER0drW2TlZWFgQMHok+fPsjIyMCMGTMwadIk7N6927CPQBAEwaA9ngCGPF+a6EmVe6huPhud6G71TXwB28r3I73bFv/05iO9l729PebPn4/hw4fD0dER69evx/DhwwEAZ86cQZs2bZCWloZu3bph586deOmll3DlyhU4OzsDAOLi4hAZGYnr169DJpMhMjISCQkJOHnypPY9Ro4ciby8POzatUvvutgjQERE4mVAj4BGo0FBQYHOotFoHvoWFRUV+Pbbb1FUVASVSoUjR46grKwMvr6+2jatW7dG06ZNkZaWBgBIS0tDu3bttCEAAPz9/VFQUKDtVUhLS9M5RlWbqmPoi0GAiIjEy4AxArGxsVAqlTpLbGzsfQ994sQJ2NjYQC6XIyQkBJs3b4aHhwfUajVkMhns7Ox02js7O0OtVgMA1Gq1Tgio2l617UFtCgoKUFxcrPdHwLsGiIhIvAy49h8VFYWIiAiddXK5/L7tW7VqhYyMDOTn5+O7775DUFAQUlJSalyqqTAIEBGReBkwoZBcLn/gF/+/yWQyuLu7AwC8vLxw6NAhLF68GCNGjEBpaSny8vJ0egVycnLg4uICAHBxccHBgwd1jld1V8Hdbf59p0FOTg4UCgWsrKz0rpOXBoiISLxMdPtgdSorK6HRaODl5QVLS0skJSVpt2VmZiI7OxsqlQoAoFKpcOLECVy7dk3bJjExEQqFAh4eHto2dx+jqk3VMfTFHgEiIhIvEz10KCoqCgMGDEDTpk1x69YtrF+/HsnJydi9ezeUSiUmTpyIiIgI2NvbQ6FQYOrUqVCpVOjWrRsAwM/PDx4eHhg7dizmzZsHtVqNd955B6GhodpeiZCQECxbtgyzZ8/GhAkTsHfvXmzcuBEJCQkG1cogQERE4mWiKYavXbuGwMBAXL16FUqlEu3bt8fu3bvRr18/AMDChQshlUoREBAAjUYDf39/rFixQru/hYUFtm/fjilTpkClUsHa2hpBQUGIiYnRtmnRogUSEhIQHh6OxYsXo3Hjxvj888/h7+9vUK2cR4DoCcV5BEgMTD6PwKAVD2/0/4q3vW7CSsyHPQJERCReJro08CRhECAiIvHi0wcZBIiISMTYI8AgQEREIsYeAQYBIiISL4mUQYBBgIiIREvCSwMMAkREJGLMAQwCREQkXuwRYBAgIiIRYxBgECAiIhFjEGAQICIiEWMQYBAgIiIxYw5gECAiIvFijwCDABERiRiDAIMAERGJmJQzCzIIEBGRiLFDgEGAiIjEi5cGGASIiEjEGAQYBIiISMQYBBgEiIhIzJgDGASIiEi82CPAIEBERCLGIMAgQEREIsYgwCBAREQiJpEyCDAIEBGRaLFHgEGAiIhEjEGAQYCIiESMQYBBgP5f8CvPI3h4TzRzswcAnD6nxoerdmLPr78DACYM64ERAzqjQ+vGUNhYwaXnLOQXFusco0Prxpg7fSi8nmuKigoBW5IyELngexQVl97zfvZKaxzc8Caecm5Y7bGIasOAfi/gypXL96wfMXIU3vrvHHy3cQN27tiO07+fQlFREX5JOwSFQqFtd/nyJayKW4GD6Qfw940bcHRywsCXBiN4cggsZbLaPBWqKeYA8LFLBAC4nJOH/y79Ed1Hz0OP0fORfPAPbFo4GW2edgEANKhvicT9v2P+6j3V7u/qqERC3FT8dfE6fMZ+jCGhy+HR0gWfxYyttn3cnFE48ecVk50PkT7WbfgOScn7tMunn68BAPTz7w8AKCkpRvcePTExOKTa/c+fO4fKSgH/nRODH35MwKzZUdi08VssWbyw1s6BHo1EItF7qavYI0AAgB2pJ3Vev7t8G4JfeR5d27fA6XNqLFufDADo6fVMtfsP6NkWZeUVmBG7EYIgAACmfrABhze9haebOODcxRvatsGvPA+lbQN8uGon+j//nGlOiEgP9vb2Oq9Xf74KTZo0RecuXQEAYwLHAQAOHUyvdv8ePX3Qo6eP9nXjJk1w/nwWNm74Bm/MijRN0WRUdfkLXl9mDQI3btzA6tWrkZaWBrVaDQBwcXFB9+7dMW7cODg6OpqzPNGSSiUI6NcJ1lYypB/P0msfuaweysoqtCEAAIo1dy4JdO/QUhsEWj/tgqjgAegV+DGaP+Vg/OKJaqistBQJ27dibND4R/pyKLx1C0ql0oiVkSkxCJjx0sChQ4fw7LPPYsmSJVAqlfDx8YGPjw+USiWWLFmC1q1b4/Dhww89jkajQUFBgc4iVFbUwhnUPc+5u+H6rwuQn74IS94egRFvfIYz59R67Zt8MBPOjRQID+wLy3oWsLO1wtxpQwAALo53/lGUWdbD2thxeGvRFlxU55rsPIhqYu/en3Dr1i0MHvpyjY+RfeECvln/NYa/MtKIlZEp8dKAGXsEpk6dildeeQVxcXH3fMCCICAkJARTp05FWlraA48TGxuL9957T2edhXMXWLp2NXrNdd0f53PgPTIWShsrvOzbEZ/FjIXfpMV6hYHT59QIjv4KH70xDDFTB6OishIrvkmB+kYBhMpKAMD70wYjMysH3+44ZOpTITLY5u+/R4/nfeDk5Fyj/XNycvD6fyahn39/BLzyqpGrI1PhhEJmDALHjh1DfHx8tSlLIpEgPDwcHTt2fOhxoqKiEBERobPOqSevzdVEWXmFtgv/t9MX4fVcU4S+1htTP/hWr/037DqMDbsOw8neFkXFGggCMG3MC8i69DcAoFeXZ9HW3Q0vH+oA4J8uuUs/f4T/fbEbc+N2GP+kiPRw5cplpB/Yj08WL63R/teu5WDS+EB4duyI6HffN3J1ZEp1+Ze+vswWBFxcXHDw4EG0bt262u0HDx6Es/PDk7lcLodcLtdZJ5FaGKVGsZNKJJDLDP8rcu3mLQBA4JBuKCktQ9KBMwCA12Z+Diu5pbad13PNsOq9MfCduAjnLl43TtFENfDj5h9gb98IPX16G7xvTs6dEODh8Rxi5sZCKuXNWE8S5gAzjhGYOXMmJk+ejOnTp2Pr1q1IT09Heno6tm7diunTpyMkJASzZ882V3miEzN1MHp0aommrvZ4zt0NMVMHw6fzM/h2x51xGs6NbNH+2afQsumdAX5tn3FD+2efQkNFA+0xQkb4oEPrxnBv6oT/vOqDhZGvInrpVu0cAVmXbuD3v65ql/OX7/QUnDmnxvXcwlo+Y6I7Kisr8ePmHzBoyFDUq6cbfG9cv44zp0/jYnY2AODsn3/gzOnTyM/LA/D/IWDcWLi6uiJiViRyb97EjevXceM6g+2TwlRjBGJjY9GlSxfY2trCyckJQ4cORWZmpk6b3r173/MeISG6t6pmZ2dj4MCBaNCgAZycnDBr1iyUl5frtElOTkanTp0gl8vh7u6O+Ph4g2o1W49AaGgoHBwcsHDhQqxYsQIVFXcG+FlYWMDLywvx8fF49VVeZ6stjvY2+OL9QLg4KJBfWIKTf17GoNdXYG/6nV/zk4b3xDshL2rb/7Q6HAAQHP0Vvt5259aqzm2b4Z2QgbBpIEPm+RyEffANvkngeAB6vB1I24+rV69g6LCAe7Zt2vgt4lYs074eHzgaABAzNxZDXh6GA/t/RXb2BWRnX4DfCz46+x47pfuPPj2eTNUjkJKSgtDQUHTp0gXl5eV466234Ofnh99//x3W1tbadsHBwYiJidG+btDgnx9XFRUVGDhwIFxcXLB//35cvXoVgYGBsLS0xIcffggAyMrKwsCBAxESEoJ169YhKSkJkyZNgqurK/z9/fWqVSLcfb+XmZSVleHGjTvXph0cHGBpafmQPR7MqmOYMcoieqzlHlr28EZET7j6Jv652ipyt95tM/+n3xdrda5fvw4nJyekpKTAx+dOaOzduzc6dOiARYsWVbvPzp078dJLL+HKlSvaS+VxcXGIjIzE9evXIZPJEBkZiYSEBJw8+c9cMCNHjkReXh527dqlV22PxcUsS0tLuLq6wtXV9ZFDABERkb4kEv2XR5Gfnw/g3kms1q1bBwcHB7Rt2xZRUVG4ffu2dltaWhratWunM17O398fBQUFOHXqlLaNr6+vzjH9/f0fesfd3TizIBERiZbUgNsHNRoNNBqNzrrqBqz/W2VlJWbMmIEePXqgbdu22vWjRo1Cs2bN4ObmhuPHjyMyMhKZmZn44YcfAABqtfqeQfNVr6sm4btfm4KCAhQXF8PKyuqh58UgQEREomXIL/3q5q2ZM2cO3n333QfuFxoaipMnT2Lfvn066ydPnqz9c7t27eDq6oq+ffvir7/+QsuWLfUv7BExCBARkWgZ0iNQ3bw1D+sNCAsLw/bt25GamorGjRs/sK23tzcA4OzZs2jZsqX2Nvu75eTkALhzC37V/1atu7uNQqHQqzcAeEzGCBAREZmDIbcPyuVyKBQKneV+QUAQBISFhWHz5s3Yu3cvWrRo8dBaMjIyAACurq4AAJVKhRMnTuDatWvaNomJiVAoFPDw8NC2SUpK0jlOYmIiVCqV3p8BgwAREYmWqeYRCA0Nxddff43169fD1tYWarUaarUaxcV35lX566+/8P777+PIkSM4f/48tm7disDAQPj4+KB9+/YAAD8/P3h4eGDs2LE4duwYdu/ejXfeeQehoaHaABISEoJz585h9uzZOHPmDFasWIGNGzciPDxc71oZBIiISLRMddfAypUrkZ+fj969e2vvinN1dcWGDRsAADKZDD/99BP8/PzQunVrvPHGGwgICMC2bdu0x7CwsMD27dthYWEBlUqFMWPGIDAwUGfegRYtWiAhIQGJiYnw9PTEggUL8Pnnn+s9hwDwmMwjYGycR4DEgPMIkBiYeh6Bju/t1bvtb3NeMGEl5sPBgkREJFp81gCDABERiRifPsggQEREIsYcwCBAREQixh4BBgEiIhIx5gAGASIiEjFDZhasqwyeR6C4uFjn6UgXLlzAokWLsGfPHqMWRkREZGqmmlDoSWJwEBgyZAi+/PJLAEBeXh68vb2xYMECDBkyBCtXrjR6gURERKZSW48hfpwZHASOHj2Knj17AgC+++47ODs748KFC/jyyy+xZMkSoxdIRERkKuwRqMEYgdu3b8PW1hYAsGfPHgwbNgxSqRTdunXDhQsXjF4gERGRqdTh73e9Gdwj4O7uji1btuDixYvYvXs3/Pz8AADXrl2DQqEweoFERESmwh6BGgSB6OhozJw5E82bN4e3t7f2UYd79uxBx44djV4gERGRqTAI1ODSwPDhw/H888/j6tWr8PT01K7v27cvXn75ZaMWR0REZEp1+PtdbzWaR8DFxQUuLi4667p27WqUgoiIiGpLXf6lry+9gsCwYcP0PuAPP/xQ42KIiIhqEycU0jMIKJVKU9dBRERU69ghoGcQWLNmjanrICIiqnVSJgHD7xoAgPLycvz000/49NNPcevWLQDAlStXUFhYaNTiiIiITIkzC9ZgsOCFCxfQv39/ZGdnQ6PRoF+/frC1tcX//vc/aDQaxMXFmaJOIiIio+NgwRr0CEyfPh2dO3dGbm4urKystOtffvllJCUlGbU4IiIiU5JK9F/qKoN7BH755Rfs378fMplMZ33z5s1x+fJloxVGRERkauwRqEEQqKysREVFxT3rL126pH0GARER0ZOAOaAGlwb8/PywaNEi7WuJRILCwkLMmTMHL774ojFrIyIiMimJAf/VVQb3CCxYsAD+/v7w8PBASUkJRo0ahT///BMODg745ptvTFEjERGRSVjU5Yv/ejI4CDRu3BjHjh3Dt99+i+PHj6OwsBATJ07E6NGjdQYPEhERPe54aaCGzxqoV68exowZY+xaiIiIahUnFKphEMjMzMTSpUtx+vRpAECbNm0QFhaG1q1bG7U4IiIiU2IOqMFgwe+//x5t27bFkSNH4OnpCU9PTxw9ehTt2rXD999/b4oaiYiITEIikei91FUG9wjMnj0bUVFRiImJ0Vk/Z84czJ49GwEBAUYrjoiIyJTq8Pe73gzuEbh69SoCAwPvWT9mzBhcvXrVKEURERHVBqlEovdSVxkcBHr37o1ffvnlnvX79u1Dz549jVIUERFRbZAYsNRVel0a2Lp1q/bPgwcPRmRkJI4cOYJu3boBAA4cOIBNmzbhvffeM02VREREJlCXr/3rSyIIgvCwRlKpfh0HEomk2umHa5tVxzBzl0BkcrmHlpm7BCKTq1+je9v0N+brY3q3/XqMpwkrMR+9PuLKykpT10FERFTr2CFQw3kEiIiI6gJeGqhhECgqKkJKSgqys7NRWlqqs23atGlGKYyIiMjU+KiBGtw18Ntvv8Hd3R2vvfYawsLCMHfuXMyYMQNvvfWWzlMJiYiIHnemmlAoNjYWXbp0ga2tLZycnDB06FBkZmbqtCkpKUFoaCgaNWoEGxsbBAQEICcnR6dNdnY2Bg4ciAYNGsDJyQmzZs1CeXm5Tpvk5GR06tQJcrkc7u7uiI+PN6hWg4NAeHg4Bg0ahNzcXFhZWeHAgQO4cOECvLy88PHHHxt6OCIiIrMx1e2DKSkpCA0NxYEDB5CYmIiysjL4+fmhqKhI2yY8PBzbtm3Dpk2bkJKSgitXrmDYsGHa7RUVFRg4cCBKS0uxf/9+rF27FvHx8YiOjta2ycrKwsCBA9GnTx9kZGRgxowZmDRpEnbv3q3/Z6DPXQN3s7OzQ3p6Olq1agU7OzukpaWhTZs2SE9PR1BQEM6cOWPI4UyCdw2QGPCuARIDU981MGnDSb3bfj6ibY3f5/r163ByckJKSgp8fHyQn58PR0dHrF+/HsOHDwcAnDlzBm3atEFaWhq6deuGnTt34qWXXsKVK1fg7OwMAIiLi0NkZCSuX78OmUyGyMhIJCQk4OTJf85j5MiRyMvLw65du/SqzeAeAUtLS+3thE5OTsjOzgYAKJVKXLx40dDDERERmY1Eov+i0WhQUFCgs2g0Gr3eJz8/HwBgb28PADhy5AjKysrg6+urbdO6dWs0bdoUaWlpAIC0tDS0a9dOGwIAwN/fHwUFBTh16pS2zd3HqGpTdQx9GBwEOnbsiEOHDgEAevXqhejoaKxbtw4zZsxA27Y1T0tERES1zZAxArGxsVAqlTpLbGzsQ9+jsrISM2bMQI8ePbTfk2q1GjKZDHZ2djptnZ2doVartW3uDgFV26u2PahNQUEBiouL9foMDO50+fDDD3Hr1i0AwAcffIDAwEBMmTIFzzzzDL744gtDD0dERGQ2howBjIqKQkREhM46uVz+0P1CQ0Nx8uRJ7Nu3z9DyaoXBQaBz587aPzs5Oel9DYKIiOhxY2HA/YNyuVyvL/67hYWFYfv27UhNTUXjxo21611cXFBaWoq8vDydXoGcnBy4uLho2xw8eFDneFV3Fdzd5t93GuTk5EChUMDKykqvGg2+NHA/x48fh0wmM9bhiIiITM5Utw8KgoCwsDBs3rwZe/fuRYsWLXS2e3l5wdLSEklJSdp1mZmZyM7OhkqlAgCoVCqcOHEC165d07ZJTEyEQqGAh4eHts3dx6hqU3UMfRhtPKYgCI/FcwYA4Mqvi81dApHJNezCu2Oo7iv+zbR3xxjt1/C/hIaGYv369fjxxx9ha2urvaavVCphZWUFpVKJiRMnIiIiAvb29lAoFJg6dSpUKpX2gX5+fn7w8PDA2LFjMW/ePKjVarzzzjsIDQ3V9kyEhIRg2bJlmD17NiZMmIC9e/di48aNSEhI0LtWTjFMRESiZaophleuXAkA6N27t876NWvWYNy4cQCAhQsXQiqVIiAgABqNBv7+/lixYoW2rYWFBbZv344pU6ZApVLB2toaQUFBiImJ0bZp0aIFEhISEB4ejsWLF6Nx48b4/PPP4e/vr3etBs8jcD/Hjh1Dp06dHotegdzb5q+ByNTcekw3dwlEJmfqHoEZP+o/982iIa1NWIn56N0jUFBQ8MDtVXcSEBERPSn4rAEDgoCdnd0Du1AEQeBTnIiI6InC7y0DgsDPP/9syjqIiIhqHXsEDAgCvXr1MmUdREREtY4dArxrgIiIRKwekwCDABERiRdzAIMAERGJmJRJgEGAiIjEizmAQYCIiESMdw3oGQSGDRum9wF/+OGHGhdDRERUm3hpQM8goFQqTV0HERFRrWMO0DMIrFmzxtR1EBER1TpeGuAYASIiEjEJmARqFAS+++47bNy4EdnZ2SgtLdXZdvToUaMURkREZGr1pOauwPwM/giWLFmC8ePHw9nZGb/99hu6du2KRo0a4dy5cxgwYIApaiQiIjIJiUSi91JXGRwEVqxYgVWrVmHp0qWQyWSYPXs2EhMTMW3aNOTn55uiRiIiIpOQSvRf6iqDg0B2dja6d+8OALCyssKtW7cAAGPHjsU333xj3OqIiIhMSCLRf6mrDA4CLi4uuHnzJgCgadOmOHDgAAAgKysLgiAYtzoiIiITkkokei91lcFB4IUXXsDWrVsBAOPHj0d4eDj69euHESNG4OWXXzZ6gURERKbCSwM1uGtg1apVqKysBACEhoaiUaNG2L9/PwYPHoz//Oc/Ri+QiIjIVOrwD329GRwEpFIppNJ/OhJGjhyJkSNHGrUoIiKi2iDlPAKGXxoAgF9++QVjxoyBSqXC5cuXAQBfffUV9u3bZ9TiiIiITImDBWsQBL7//nv4+/vDysoKv/32GzQaDQAgPz8fH374odELJCIiMpV6UoneS11lcBCYO3cu4uLi8Nlnn8HS0lK7vkePHpxVkIiInijsEajBGIHMzEz4+Pjcs16pVCIvL88YNREREdWKunxboL5qNI/A2bNn71m/b98+PP3000YpioiIqDawR6AGQSA4OBjTp09Heno6JBIJrly5gnXr1mHmzJmYMmWKKWokIiIyCakBS11l8KWBN998E5WVlejbty9u374NHx8fyOVyzJw5E1OnTjVFjURERCZRlx8mpC+Dg4BEIsHbb7+NWbNm4ezZsygsLISHhwdsbGxQXFwMKysrU9RJRERkdIwBj9DbIZPJ4OHhga5du8LS0hKffPIJWrRoYczaiIiITIrPGjAgCGg0GkRFRaFz587o3r07tmzZAgBYs2YNWrRogYULFyI8PNxUdRIRERmdxIClrtL70kB0dDQ+/fRT+Pr6Yv/+/XjllVcwfvx4HDhwAJ988gleeeUVWFhYmLJWIiIio6rDP/T1pncQ2LRpE7788ksMHjwYJ0+eRPv27VFeXo5jx45xsAURET2RLPj9pX8QuHTpEry8vAAAbdu2hVwuR3h4OEMAERE9sfgdZkAQqKiogEwm+2fHevVgY2NjkqKIiIhqA2OAAYMFBUHAuHHjMGzYMAwbNgwlJSUICQnRvq5aiIiInhQSiUTvxRCpqakYNGgQ3NzcIJFItAPsq4wbN+6e4/fv31+nzc2bNzF69GgoFArY2dlh4sSJKCws1Glz/Phx9OzZE/Xr10eTJk0wb948gz8DvXsEgoKCdF6PGTPG4DcjIiJ6nJhqxsCioiJ4enpiwoQJ9/2R3L9/f6xZs0b7Wi6X62wfPXo0rl69isTERJSVlWH8+PGYPHky1q9fDwAoKCiAn58ffH19ERcXhxMnTmDChAmws7PD5MmT9a5V7yBwd7FERER1ganGCAwYMAADBgx4YBu5XA4XF5dqt50+fRq7du3CoUOH0LlzZwDA0qVL8eKLL+Ljjz+Gm5sb1q1bh9LSUqxevRoymQzPPfccMjIy8MknnxgUBOry9MlEREQPZMg8AhqNBgUFBTqLRqOp8XsnJyfDyckJrVq1wpQpU/D3339rt6WlpcHOzk4bAgDA19cXUqkU6enp2jY+Pj464/f8/f2RmZmJ3NxcvetgECAiItEy5OmDsbGxUCqVOktsbGyN3rd///748ssvkZSUhP/9739ISUnBgAEDUFFRAQBQq9VwcnLS2adevXqwt7eHWq3WtnF2dtZpU/W6qo0+DH7WABERUV0hNeC+gaioKEREROis+/d1fX2NHDlS++d27dqhffv2aNmyJZKTk9G3b98aHbOmGASIiEi0DHmGgFwur/EX/8M8/fTTcHBwwNmzZ9G3b1+4uLjg2rVrOm3Ky8tx8+ZN7bgCFxcX5OTk6LSpen2/sQfV4aUBIiISLUMuDZjSpUuX8Pfff8PV1RUAoFKpkJeXhyNHjmjb7N27F5WVlfD29ta2SU1NRVlZmbZNYmIiWrVqhYYNG+r93gwCREQkWlJI9F4MUVhYiIyMDGRkZAAAsrKykJGRgezsbBQWFmLWrFk4cOAAzp8/j6SkJAwZMgTu7u7w9/cHALRp0wb9+/dHcHAwDh48iF9//RVhYWEYOXIk3NzcAACjRo2CTCbDxIkTcerUKWzYsAGLFy++5/LFw/DSABERiZapfukfPnwYffr00b6u+nIOCgrCypUrcfz4caxduxZ5eXlwc3ODn58f3n//fZ1LD+vWrUNYWBj69u0LqVSKgIAALFmyRLtdqVRiz549CA0NhZeXFxwcHBAdHW3QrYMAIBEEQXjE833s5N6uMHcJRCbn1mO6uUsgMrni35aZ9Ph7Tl/Xu61fG0cTVmI+7BEgIiLRkvBpAwwCREQkXlLmAAYBIiISL/YIMAjQfVRUVODzuOXYtWMbbv59Aw6OThg4aCjGB4dAIpGgvKwMcSuWIG1fKi5fugQbGxt08Vbh9WkRcLxrNqyZ00Px5x+nkXvzJmwVCnTxViF02hs6bYhqS/ArzyN4eE80c7MHAJw+p8aHq3Ziz6+/AwAmDOuBEQM6o0PrxlDYWMGl5yzkFxbrHKND68aYO30ovJ5riooKAVuSMhC54HsUFZcCAMYM8sZnMWOrff+mL7yJ67mF1W4j8zD1bYFPAg4WpGrFf/Epvvl6LaJjYtGipTvOnDqJue++jf+ETseIUWNReOsWombNwJBhw/HMs61xq6AAn8z/EJUVlYhfv0l7nG++Xot27TugkYMDrl+7hqUL5wMAPlu73lynVmdwsKDhXvRpi4rKSpzNvg4JJBgzyBvhQX3RbeRHOH1OjbBRvVFfbgkAeH/akHuCgKujEoc3vYXv9hzFsnU/Q2FdH/NnBUB9owCjZn0BAKgvt4TSpr7O+656byzqyy3hH7y49k62jjD1YMHUP27q3dbnWXsTVmI+7BGgap04lgGfXi+gR89eAAA3t6ewZ9cO/H7qBADAxtYWS+O+0Nln5pvvYMKYEVBfvQIX1zv3ub425p/HV7u6PYWx4ychMmIqysvKUM/SspbOhuiOHakndV6/u3wbgl95Hl3bt8Dpc2osW58MAOjp9Uy1+w/o2RZl5RWYEbsRVb+hpn6wAYc3vYWnmzjg3MUbKNGUoUTzzwQvDg1t0Lvrswh5b51pTooeCS8NcEIhuo92nh1w6OABZF84DwD4M/MMjmUchapHz/vuU3jrFiQSCWxtFdVuz8/Pw+6d29HOsyNDAJmdVCrBK/5esLaSIf14ll77yGX1UFZWgbs7Uos1dy4JdO/Qstp9Rr/UFbdLSrH5p4xHrpmM73GZWdCc2CNA1QocH4yiwiKMeHkgpBYWqKyoQEjodPR/cVC17TUaDZYv+QT9+r8IaxsbnW3LFi/Ad9+uR0lJMdq288SCJStr4xSIqvWcuxuS176B+rJ6KCzWYMQbn+HMOf2e1JZ8MBP/ixiG8MC+WLY+GdZWMsydNgQA4OKorHafoKEqbNh5WKeXgB4fdfj7XW+PdY/AxYsXMWHChAe2MfbzoemOpD27sHvndsR8OB9r13+H6JhYrPtqDRK2brmnbXlZGd6eHQFBEBD51px7to8JnIAvv/0ei1d+DqmFBd7775uog0NT6Anxx/kceI+MhU/gx/hs0z58FjMWrZ/W7wEtp8+pERz9FaaN7YubaZ/g/E8f4vzlv6G+UQChsvKe9t7tW6DN065YuyXN2KdBRiKVSPRe6qrHOgjcvHkTa9eufWCb6p4PvfDjj2qpwrpr6aKPETh+Evr1fxHuzzyLAS8NxsjRQfhyzWc67crLyvB2ZATUV69g6cov7ukNAAC7hg3RtFlzeHfrjrkffYz9+1Jx8vix2joVIh1l5RU4d/EGfjt9EdFLt+LEH5cR+lpvvfffsOswWvR7Cy3938FTvSMxN24HHBvaIOvS3/e0HfeyChlnLuK30xeNeAZkTBIDlrrKrJcGtm7d+sDt586de+gxqns+9O0KXvF4VCUlxZBIdHOihVSKyrt+9VSFgIvZF7B8VTyUdnYPPW7V/qVlpUatl6impBIJ5DLD/824dvMWACBwSDeUlJYh6cAZne3WVjIE9OuE6KUP/neOzKwuf8PryazfmEOHDoVEInlgN7HkId0x1T0fuoK3Dz6y5336IP6LT+Hi6ooWLd3xx5nT+ObrtXhp6DAAd0JA1KwZyDxzGgsWr0BlZQX+vnFnzm6FUglLSxlOnjiG06dOwrNjJ9jaKnD50kV8umIpGjdpgnbtO5jx7EisYqYOxu5fT+Hi1VzYWtfHiAGd4dP5GQx6fQUAwLmRLZwbKdCyqQMAoO0zbrhVVIKL6lzkFtwGAISM8MGBY+dQeLsUfbu1xoczhuK/S3+8Z76B4f5eqGchxTcJh2r3JMkgvGvAzEHA1dUVK1aswJAhQ6rdnpGRAS8vr1quigDgjci3sWrFEsz/MAa5uTfh4OiEocNfxcTJUwAA165fwy8pPwMAxo4cprPv8s/i4dW5K+rXt0Ly3p/wWdwylBQXo5GDI7p1fx7jgz+BTCar9XMicrS3wRfvB8LFQYH8whKc/PMyBr2+AnvT7/yanzS8J94JeVHb/qfV4QCA4Oiv8PW2dABA57bN8E7IQNg0kCHzfA7CPvim2i/7cUNV+HHvsXsCAj1e6vClf72ZdUKhwYMHo0OHDoiJial2+7Fjx9CxY0ed7mh9cEIhEgNOKERiYOoJhQ5l5evdtkuL6u8MedKZtUdg1qxZKCoquu92d3d3/Pzzz7VYERERiQkvDZg5CPTsef/JaQDA2toavXr1qqVqiIhIbHhpgBMKERGRiDEHMAgQEZGYMQkwCBARkXhxjACDABERiRjHCDAIEBGRiDEHMAgQEZGYMQkwCBARkXhxjACDABERiZiUOYBBgIiIRIxBgEGAiIjEi5cGGASIiEjEePsggwAREYkYcwCDABERiRmTAIMAERGJF8cIMAgQEZGIcYwAgwAREYkYcwCDABERiZiEXQIMAkREJF7MAQwCREQkYswBDAJERCRmTAKQmrsAIiIic5EY8J8hUlNTMWjQILi5uUEikWDLli062wVBQHR0NFxdXWFlZQVfX1/8+eefOm1u3ryJ0aNHQ6FQwM7ODhMnTkRhYaFOm+PHj6Nnz56oX78+mjRpgnnz5hn8GTAIEBGRaEkk+i+GKCoqgqenJ5YvX17t9nnz5mHJkiWIi4tDeno6rK2t4e/vj5KSEm2b0aNH49SpU0hMTMT27duRmpqKyZMna7cXFBTAz88PzZo1w5EjRzB//ny8++67WLVqlWGfgSAIgmGn9/jLvV1h7hKITM6tx3Rzl0BkcsW/LTPp8c/fKHl4o//X3KF+jd5DIpFg8+bNGDp0KIA7vQFubm544403MHPmTABAfn4+nJ2dER8fj5EjR+L06dPw8PDAoUOH0LlzZwDArl278OKLL+LSpUtwc3PDypUr8fbbb0OtVkMmkwEA3nzzTWzZsgVnzpzRuz72CBARkXhJ9F80Gg0KCgp0Fo1GY/BbZmVlQa1Ww9fXV7tOqVTC29sbaWlpAIC0tDTY2dlpQwAA+Pr6QiqVIj09XdvGx8dHGwIAwN/fH5mZmcjNzdW7HgYBIiISLUPGCMTGxkKpVOossbGxBr+nWq0GADg7O+usd3Z21m5Tq9VwcnLS2V6vXj3Y29vrtKnuGHe/hz541wAREYmW1IBr/1FRUYiIiNBZJ5fLjVxR7WMQICIi0TJkEKBcLjfKF7+LiwsAICcnB66urtr1OTk56NChg7bNtWvXdPYrLy/HzZs3tfu7uLggJydHp03V66o2+uClASIiEjEDBgkYSYsWLeDi4oKkpCTtuoKCAqSnp0OlUgEAVCoV8vLycOTIEW2bvXv3orKyEt7e3to2qampKCsr07ZJTExEq1at0LBhQ73rYRAgIiLRMtXtg4WFhcjIyEBGRgaAOwMEMzIykJ2dDYlEghkzZmDu3LnYunUrTpw4gcDAQLi5uWnvLGjTpg369++P4OBgHDx4EL/++ivCwsIwcuRIuLm5AQBGjRoFmUyGiRMn4tSpU9iwYQMWL158z+WLh+GlASIiEi1TTSx4+PBh9OnTR/u66ss5KCgI8fHxmD17NoqKijB58mTk5eXh+eefx65du1C//j+3KK5btw5hYWHo27cvpFIpAgICsGTJEu12pVKJPXv2IDQ0FF5eXnBwcEB0dLTOXAP64DwCRE8oziNAYmDqeQSu5pfq3dZVKXt4oycQewSIiEi0DJ06uC5iECAiIvFiDmAQICIi8WIOYBAgIiIRkxp6O0AdxCBARETixRzAIEBEROLFHMAgQEREIsYrAwwCREQkYrx9kEGAiIhEjD0CfNYAERGRqLFHgIiIRIs9AgwCREQkYhwjwCBAREQixh4BBgEiIhIxBgEGASIiEjFeGmAQICIiEWOPAIMAERGJGHMAgwAREYkZkwCDABERiRfHCDAIEBGRiHGMACARBEEwdxH0ZNNoNIiNjUVUVBTkcrm5yyEyCf49p7qKQYAeWUFBAZRKJfLz86FQKMxdDpFJ8O851VV86BAREZGIMQgQERGJGIMAERGRiDEI0COTy+WYM2cOB1BRnca/51RXcbAgERGRiLFHgIiISMQYBIiIiESMQYCIiEjEGASIiIhEjEGAHtny5cvRvHlz1K9fH97e3jh48KC5SyIymtTUVAwaNAhubm6QSCTYsmWLuUsiMioGAXokGzZsQEREBObMmYOjR4/C09MT/v7+uHbtmrlLIzKKoqIieHp6Yvny5eYuhcgkePsgPRJvb2906dIFy5YtAwBUVlaiSZMmmDp1Kt58800zV0dkXBKJBJs3b8bQoUPNXQqR0bBHgGqstLQUR44cga+vr3adVCqFr68v0tLSzFgZERHpi0GAauzGjRuoqKiAs7OzznpnZ2eo1WozVUVERIZgECAiIhIxBgGqMQcHB1hYWCAnJ0dnfU5ODlxcXMxUFRERGYJBgGpMJpPBy8sLSUlJ2nWVlZVISkqCSqUyY2VERKSveuYugJ5sERERCAoKQufOndG1a1csWrQIRUVFGD9+vLlLIzKKwsJCnD17Vvs6KysLGRkZsLe3R9OmTc1YGZFx8PZBemTLli3D/PnzoVar0aFDByxZsgTe3t7mLovIKJKTk9GnT5971gcFBSE+Pr72CyIyMgYBIiIiEeMYASIiIhFjECAiIhIxBgEiIiIRYxAgIiISMQYBIiIiEWMQICIiEjEGASIiIhFjECAygnHjxuk8o753796YMWNGrdeRnJwMiUSCvLw8k73Hv8+1JmqjTiLSD4MA1Vnjxo2DRCKBRCKBTCaDu7s7YmJiUF5ebvL3/uGHH/D+++/r1ba2vxSbN2+ORYsW1cp7EdHjj88aoDqtf//+WLNmDTQaDXbs2IHQ0FBYWloiKirqnralpaWQyWRGeV97e3ujHIeIyNTYI0B1mlwuh4uLC5o1a4YpU6bA19cXW7duBfBPF/cHH3wANzc3tGrVCgBw8eJFvPrqq7Czs4O9vT2GDBmC8+fPa49ZUVGBiIgI2NnZoVGjRpg9ezb+PVP3vy8NaDQaREZGokmTJpDL5XB3d8cXX3yB8+fPa+exb9iwISQSCcaNGwfgzpMcY2Nj0aJFC1hZWcHT0xPfffedzvvs2LEDzz77LKysrNCnTx+dOmuioqICEydO1L5nq1atsHjx4mrbvvfee3B0dIRCoUBISAhKS0u12/Sp/W4XLlzAoEGD0LBhQ1hbW+O5557Djh07HulciEg/7BEgUbGyssLff/+tfZ2UlASFQoHExEQAQFlZGfz9/aFSqfDLL7+gXr16mDt3Lvr374/jx49DJpNhwYIFiI+Px+rVq9GmTRssWLAAmzdvxgsvvHDf9w0MDERaWhqWLFkCT09PZGVl4caNG2jSpAm+//57BAQEIDMzEwqFAlZWVgCA2NhYfP3114iLi8MzzzyD1NRUjBkzBo6OjujVqxcuXryIYcOGITQ0FJMnT8bhw4fxxhtvPNLnU1lZicaNG2PTpk1o1KgR9u/fj8mTJ8PV1RWvvvqqzudWv359JCcn4/z58xg/fjwaNWqEDz74QK/a/y00NBSlpaVITU2FtbU1fv/9d9jY2DzSuRCRngSiOiooKEgYMmSIIAiCUFlZKSQmJgpyuVyYOXOmdruzs7Og0Wi0+3z11VdCq1athMrKSu06jUYjWFlZCbt37xYEQRBcXV2FefPmabeXlZUJjRs31r6XIAhCr169hOnTpwuCIAiZmZkCACExMbHaOn/++WcBgJCbm6tdV1JSIjRo0EDYv3+/TtuJEycKr732miAIghAVFSV4eHjobI+MjLznWP/WrFkzYeHChffd/m+hoaFCQECA9nVQUJBgb28vFBUVadetXLlSsLGxESoqKvSq/d/n3K5dO+Hdd9/VuyYiMh72CFCdtn37dtjY2KCsrAyVlZUYNWoU3n33Xe32du3a6YwLOHbsGM6ePQtbW1ud45SUlOCvv/5Cfn4+rl69qvOY5Xr16qFz5873XB6okpGRAQsLi2p/Cd/P2bNncfv2bfTr109nfWlpKTp27AgAOH369D2Pe1apVHq/x/0sX74cq1evRnZ2NoqLi1FaWooOHTrotPH09ESDBg103rewsBAXL15EYWHhQ2v/t2nTpmHKlCnYs2cPfH19ERAQgPbt2z/yuRDRwzEIUJ3Wp08frFy5EjKZDG5ubqhXT/evvLW1tc7rwsJCeHl5Yd26dfccy9HRsUY1VHX1G6KwsBAAkJCQgKeeekpnm1wur1Ed+vj2228xc+ZMLFiwACqVCra2tpg/fz7S09P1PkZNap80aRL8/f2RkJCAPXv2IDY2FgsWLMDUqVNrfjJEpBcGAarTrK2t4e7urnf7Tp06YcOGDXBycoJCoai2jaurK9LT0+Hj4wMAKC8vx5EjR9CpU6dq27dr1w6VlZVISUmBr6/vPdureiQqKiq06zw8PCCXy5GdnX3fnoQ2bdpoBz5WOXDgwMNP8gF+/fVXdO/eHa+//rp23V9//XVPu2PHjqG4uFgbcg4cOAAbGxs0adIE9vb2D629Ok2aNEFISAhCQkIQFRWFzz77jEGAqBbwrgGiu4wePRoODg4YMmQIfvnlF2RlZSE5ORnTpk3DpUuXAADTp0/HRx99hC1btuDMmTN4/fXXHzgHQPPmzREUFIQJEyZgy5Yt2mNu3LgRANCsWTNIJBJs374d169fR2FhIWxtbTFz5kyEh4dj7dq1+Ouvv3D06FEsXboUa9euBQCEhITgzz//xKxZs5CZmYn169cjPj5er/O8fPkyMjIydJbc3Fw888wzOHz4MHbv3o0//vgD//3vf3Ho0KF79i8tLcXEiRPx+++/Y8eOHZgzZw7CwsIglUr1qv3fZsyYgd27dyMrKwtHjx7Fzz//jDZt2uh1LkT0iMw9SIHIVO4eLGjI9qtXrwqBgYGCg4ODIJfLhaeffloIDg4W8vPzBUG4Mzhw+vTpgkKhEOzs7ISIiAghMDDwvoMFBUEQiouLhfDwcMHV1VWQyWSCu7u7sHr1au32mJgYwcXFRZBIJEJQUJAgCHcGOC5atEho1aqVYGlpKTg6Ogr+/v5CSkqKdr9t27YJ7u7uglwuF3r27CmsXr1ar8GCAO5ZvvrqK6GkpEQYN26coFQqBTs7O2HKlCnCm2++KXh6et7zuUVHRwuNGjUSbGxshODgYKGkpETb5mG1/3uwYFhYmNCyZUtBLpcLjo6OwtixY4UbN27c9xyIyHgkgnCfEU5ERERU5/HSABERkYgxCBAREYkYgwAREZGIMQgQERGJGIMAERGRiDEIEBERiRiDABERkYgxCBAREYkYgwAREZGIMQgQERGJGIMAERGRiDEIEBERidj/AeRUXqi9qO7OAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 600x300 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize = (6,3))\n",
    "\n",
    "sns.heatmap(confusion_matrix(y_3,combined_predictions_3), annot=True, \n",
    "            fmt='', cmap='Blues')\n",
    "\n",
    "plt.xlabel('Predicted Labels')\n",
    "plt.ylabel('Real Labels')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "46adc3ec",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "7f491630",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}