explore.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "from collections import Counter\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "DIR = '../data/'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "# read the json file domain1_train.json in to the system\n",
    "train_domain1 = pd.read_json('data/domain1_train.json', lines=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "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>text</th>\n",
       "      <th>label</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>[70, 746, 825, 109, 2083, 0, 2, 0, 0, 0, 9, 0,...</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>[1209, 179, 1952, 4, 4959, 7, 0, 2, 978, 1522,...</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>[287, 3, 3330, 0, 23, 12, 13, 465, 74, 8, 0, 8...</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>[0, 0, 3, 592, 19, 2, 706, 1439, 2575, 7, 2, 0...</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>[9, 2, 110, 12, 42, 32, 44, 361, 9, 3860, 2358...</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>[9, 892, 3, 0, 0, 3, 2, 0, 811, 3, 26, 939, 13...</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>[0, 1115, 0, 72, 192, 64, 31, 3295, 33, 0, 1]</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>[2, 1510, 0, 9, 30, 138, 118, 2, 0, 0, 0, 5, 0...</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>[0, 1398, 324, 41, 1629, 0, 4162, 0, 4, 3020, ...</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>[9, 5, 940, 4, 394, 3, 2, 537, 8, 15, 773, 3, ...</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                text  label\n",
       "0  [70, 746, 825, 109, 2083, 0, 2, 0, 0, 0, 9, 0,...      1\n",
       "1  [1209, 179, 1952, 4, 4959, 7, 0, 2, 978, 1522,...      1\n",
       "2  [287, 3, 3330, 0, 23, 12, 13, 465, 74, 8, 0, 8...      1\n",
       "3  [0, 0, 3, 592, 19, 2, 706, 1439, 2575, 7, 2, 0...      1\n",
       "4  [9, 2, 110, 12, 42, 32, 44, 361, 9, 3860, 2358...      1\n",
       "5  [9, 892, 3, 0, 0, 3, 2, 0, 811, 3, 26, 939, 13...      1\n",
       "6      [0, 1115, 0, 72, 192, 64, 31, 3295, 33, 0, 1]      1\n",
       "7  [2, 1510, 0, 9, 30, 138, 118, 2, 0, 0, 0, 5, 0...      1\n",
       "8  [0, 1398, 324, 41, 1629, 0, 4162, 0, 4, 3020, ...      1\n",
       "9  [9, 5, 940, 4, 394, 3, 2, 537, 8, 15, 773, 3, ...      1"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_domain1.head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1    9750\n",
       "0    9750\n",
       "Name: label, dtype: int64"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_domain1['label'].value_counts(dropna=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Axes: ylabel='Frequency'>"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "train_domain1['label'].plot.hist()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count    19500.000000\n",
       "mean        39.765282\n",
       "std         31.248784\n",
       "min          1.000000\n",
       "25%         19.000000\n",
       "50%         29.000000\n",
       "75%         47.000000\n",
       "max        238.000000\n",
       "Name: text, dtype: float64"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "text1_length = train_domain1['text'].apply(len)\n",
    "text1_length.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "# read the json file domain2_train.json in to the system\n",
    "train_domain2 = pd.read_json('data/domain2_train.json', lines=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "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>text</th>\n",
       "      <th>label</th>\n",
       "      <th>model</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>[3147, 471, 4, 343, 2, 0, 1, 14, 124, 133, 436...</td>\n",
       "      <td>1</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>[10, 389, 232, 24, 2, 0, 4730, 1, 13, 10, 129,...</td>\n",
       "      <td>1</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>[861, 0, 2505, 2, 0, 1015, 2, 0, 3, 1772, 8, 3...</td>\n",
       "      <td>1</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>[325, 21, 3234, 1, 14, 187, 16, 13, 1965, 1, 1...</td>\n",
       "      <td>1</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>[5, 1091, 272, 6, 3232, 32, 2, 1012, 4, 3240, ...</td>\n",
       "      <td>1</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>[10, 410, 36, 3484, 8, 15, 0, 59, 36, 3484, 12...</td>\n",
       "      <td>1</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>[503, 7, 11, 133, 4, 1, 860, 291, 10, 5, 0, 21...</td>\n",
       "      <td>1</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>[56, 239, 4, 31, 0, 3830, 53, 90, 2, 1745, 1, ...</td>\n",
       "      <td>1</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>[117, 191, 10, 2, 117, 69, 2, 339, 505, 465, 4...</td>\n",
       "      <td>1</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>[10, 1345, 5, 853, 4, 21, 4815, 1, 53, 6, 5, 2...</td>\n",
       "      <td>1</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                text  label  model\n",
       "0  [3147, 471, 4, 343, 2, 0, 1, 14, 124, 133, 436...      1    NaN\n",
       "1  [10, 389, 232, 24, 2, 0, 4730, 1, 13, 10, 129,...      1    NaN\n",
       "2  [861, 0, 2505, 2, 0, 1015, 2, 0, 3, 1772, 8, 3...      1    NaN\n",
       "3  [325, 21, 3234, 1, 14, 187, 16, 13, 1965, 1, 1...      1    NaN\n",
       "4  [5, 1091, 272, 6, 3232, 32, 2, 1012, 4, 3240, ...      1    NaN\n",
       "5  [10, 410, 36, 3484, 8, 15, 0, 59, 36, 3484, 12...      1    NaN\n",
       "6  [503, 7, 11, 133, 4, 1, 860, 291, 10, 5, 0, 21...      1    NaN\n",
       "7  [56, 239, 4, 31, 0, 3830, 53, 90, 2, 1745, 1, ...      1    NaN\n",
       "8  [117, 191, 10, 2, 117, 69, 2, 339, 505, 465, 4...      1    NaN\n",
       "9  [10, 1345, 5, 853, 4, 21, 4815, 1, 53, 6, 5, 2...      1    NaN"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_domain2.head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    12750\n",
       "1     2150\n",
       "Name: label, dtype: int64"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_domain2['label'].value_counts(dropna=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count    14900.000000\n",
       "mean       156.528054\n",
       "std        158.577144\n",
       "min          0.000000\n",
       "25%         47.000000\n",
       "50%        122.000000\n",
       "75%        190.000000\n",
       "max       1075.000000\n",
       "Name: text, dtype: float64"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "text2_length = train_domain2['text'].apply(len)\n",
    "text2_length.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.0    2364\n",
       "1.0    2357\n",
       "2.0    2339\n",
       "3.0    2358\n",
       "4.0     789\n",
       "5.0     780\n",
       "6.0    1763\n",
       "Name: model, dtype: int64"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_domain2['model'].value_counts().sort_index()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Comparason between human and machine "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Grouping data for Domain 1\n",
    "dm1_machine, dm1_human = train_domain1[train_domain1['label'] == 0], train_domain1[train_domain1['label'] == 1]\n",
    "\n",
    "# Grouping data for Domain 2\n",
    "dm2_machine, dm2_human = train_domain2[train_domain2['label'] == 0], train_domain2[train_domain2['label'] == 1]\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Word frequency\n",
    "\n",
    "Domain1: \n",
    "\n",
    "Humen usually use first 100 word but machines use more non-conventional words. \n",
    "\n",
    "Human mean frequency: 55\n",
    "\n",
    "Machine mean frequency: 113\n",
    "\n",
    "  \n",
    "\n",
    "Domain 2:\n",
    "\n",
    "Similar distribution in frequency. \n",
    "\n",
    "Human mean frequency: 95\n",
    "\n",
    "Machine mean frequency: 377"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "def visualize_word(domain):\n",
    "    counter = Counter()\n",
    "    for text in domain['text']:\n",
    "        counter.update(text)\n",
    "    \n",
    "    token_ids = list(counter.keys())\n",
    "    token_counts = list(counter.values())\n",
    "    \n",
    "    mean_frequency = np.mean(token_counts)\n",
    "    max_frequency = max(token_counts)\n",
    "    min_frequency = min(token_counts)\n",
    "    \n",
    "    print(f\"The mean frequency of these words is {mean_frequency}\")\n",
    "    print(f\"The max frequency of these words is {max_frequency}\")\n",
    "    print(f\"The min frequency of these words is {min_frequency}\")\n",
    "    \n",
    "    # Create a bar plot\n",
    "    plt.figure(figsize=(8,6))\n",
    "    plt.bar(token_ids, token_counts)\n",
    "    plt.xlim(left=0, right=500)\n",
    "    plt.xlabel(\"Word frequency\")\n",
    "    plt.title(\"Word Frequency Distribution\")\n",
    "    plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The mean frequency of these words is 55.002274607113314\n",
      "The max frequency of these words is 44068\n",
      "The min frequency of these words is 1\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "visualize_word(dm1_human)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The mean frequency of these words is 112.85600354452814\n",
      "The max frequency of these words is 45769\n",
      "The min frequency of these words is 1\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "visualize_word(dm1_machine)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The mean frequency of these words is 94.92816442816442\n",
      "The max frequency of these words is 50768\n",
      "The min frequency of these words is 1\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "visualize_word(dm2_human)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The mean frequency of these words is 377.156054174247\n",
      "The max frequency of these words is 146042\n",
      "The min frequency of these words is 1\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "visualize_word(dm2_machine)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [],
   "source": [
    "def describe_text_length(domain):\n",
    "    description = domain['text'].str.len().describe()\n",
    "    print(description)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Text length\n",
    "\n",
    "Domain 1:\n",
    "\n",
    "Human mean text length: 27\n",
    "\n",
    "Machine mean text length: 52\n",
    "\n",
    "\n",
    "\n",
    "Domain 2:\n",
    "\n",
    "Human mean text length: 217\n",
    "\n",
    "Machine mean text length: 146"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Domain 1 human\n",
      "count    9750.000000\n",
      "mean       27.281128\n",
      "std        12.637503\n",
      "min         9.000000\n",
      "25%        18.000000\n",
      "50%        25.000000\n",
      "75%        34.000000\n",
      "max       105.000000\n",
      "Name: text, dtype: float64\n",
      "Domain 1 machine\n",
      "count    9750.000000\n",
      "mean       52.249436\n",
      "std        38.491903\n",
      "min         1.000000\n",
      "25%        22.000000\n",
      "50%        39.000000\n",
      "75%        74.000000\n",
      "max       238.000000\n",
      "Name: text, dtype: float64\n"
     ]
    }
   ],
   "source": [
    "print(\"Domain 1 human\")\n",
    "describe_text_length(dm1_human)\n",
    "print(\"Domain 1 machine\")\n",
    "describe_text_length(dm1_machine)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Domain 2 human\n",
      "count    2150.000000\n",
      "mean      216.966047\n",
      "std       212.696979\n",
      "min         0.000000\n",
      "25%        51.000000\n",
      "50%       165.000000\n",
      "75%       232.500000\n",
      "max       933.000000\n",
      "Name: text, dtype: float64\n",
      "Domain 2 machine\n",
      "count    12750.000000\n",
      "mean       146.336549\n",
      "std        145.058041\n",
      "min          2.000000\n",
      "25%         47.000000\n",
      "50%        114.000000\n",
      "75%        183.000000\n",
      "max       1075.000000\n",
      "Name: text, dtype: float64\n"
     ]
    }
   ],
   "source": [
    "print(\"Domain 2 human\")\n",
    "describe_text_length(dm2_human)\n",
    "print(\"Domain 2 machine\")\n",
    "describe_text_length(dm2_machine)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### log probability -> similar approach"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "machine的话生僻词会用的多一些，但是domain2的话情况会好一些。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [],
   "source": [
    "def sum_log_probabilities(dataframe):\n",
    "    log_probabilities = [0] * 5000\n",
    "    text = dataframe['text']\n",
    "\n",
    "    for sentence in text:\n",
    "        sentence_length = len(sentence)\n",
    "        unique_words = set(sentence)  # 获取句子中的唯一词\n",
    "        \n",
    "        for word in unique_words:\n",
    "            term_frequency = sentence.count(word)\n",
    "            log_probability = np.log(1 + (term_frequency / sentence_length))  # 使用平滑的对数概率计算方式\n",
    "            \n",
    "            log_probabilities[word] += log_probability\n",
    "    \n",
    "    plt.figure(figsize=(10,6))\n",
    "    plt.bar(range(5000), log_probabilities)\n",
    "    plt.xlabel('Word Index')\n",
    "    plt.xlim(left=0, right=500)\n",
    "    plt.ylabel('Log Probability')\n",
    "    plt.title('Log Probability vs. Word Index')\n",
    "    plt.show()\n",
    "    \n",
    "    # return log_probabilities"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sum_log_probabilities(dm1_human)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[1489.4840635928042, 307.06895798722053, 509.98601198262276, 373.5721372856443, 235.22149194720026, 189.20380428954434, 214.48636467745257, 184.23601272942838, 64.78488333366364, 166.28164267314455, 54.30747038685832, 20.35347569683774, 85.43610111429788, 48.64542835795591, 35.76701065472114, 40.3766636040112, 90.67840751695027, 76.46613824697924, 83.65227349184364, 59.529930494264214, 35.09950657510725, 31.772249639869294, 47.77210730701086, 36.86053725761223, 43.166589535563574, 0, 18.93752431554966, 11.756476064732874, 30.844496928357774, 30.441804432240396, 38.65183750326152, 39.567159573757486, 34.42640757773988, 43.47664823755673, 21.993544286566124, 42.05228113169708, 14.309405407114559, 7.066149734963957, 36.09409398319166, 7.457976412194143, 31.455023075619383, 38.023265088488195, 17.19818142551115, 19.14441929224556, 22.015106480947754, 18.303674712697404, 0.18152200450062014, 22.408101939046837, 4.405769445967583, 23.524161067272992, 31.32096978543986, 21.42084396022818, 18.70128198911583, 17.17425963869575, 16.387849726706037, 9.997085038277692, 13.657079369252852, 4.889936217000161, 18.330766024654654, 16.878769463832754, 16.331583046545077, 21.335497323024985, 32.05421953522734, 31.554026681329468, 28.824250713763313, 7.794788463757636, 11.456540534762082, 12.584928985717436, 12.99531947011415, 12.167112190764959, 19.203923485304802, 10.08082466050972, 18.6845685186486, 18.431946280986082, 23.29605594964612, 8.558854978853724, 2.7262901149323095, 17.801933372302443, 6.6907852797632685, 14.525808248098546, 15.659730168106051, 7.677684560845595, 15.940177542538303, 0.711722167571917, 11.791987463637948, 11.17193402664388, 11.982412897072567, 17.704614283854717, 5.596226195288501, 11.116504647267067, 11.574140031802257, 9.941969471683864, 8.993950701788417, 6.727695778609609, 10.897303839472164, 7.639323566567947, 6.838586744061389, 11.517001133876963, 4.412777570463791, 0.17053360104191242, 12.707377643874336, 4.307144169788576, 10.713157169702272, 3.7713896404635996, 3.6016316907780035, 3.0173216856842644, 6.81216767116627, 0.22369596523164334, 11.597212708122196, 13.35055614619538, 8.609559578211003, 0.2796680441192396, 3.956832162839996, 8.861578186594834, 6.581246019840358, 1.9840379998243576, 4.413267712571166, 10.136240448468433, 6.728728994245951, 6.387749701701167, 9.778630136413794, 4.666737196343274, 6.880049998027036, 2.0036898985594807, 6.760523069315228, 4.6083510271436765, 5.466102404638779, 5.223719783385034, 5.812516851358156, 6.192715694437863, 7.213344858042757, 2.4336704806871743, 8.124486428748337, 2.201340995488295, 5.2535034041940545, 7.340400175320552, 8.428512543118256, 6.840063670964697, 6.046278503699842, 9.808607129233135, 9.874390027922322, 5.826244370328407, 4.9929346818097144, 5.839780853995516, 3.7530911918934593, 5.120785447698499, 8.906489777799813, 6.149080731316836, 5.40164218796431, 2.066384199267255, 7.2838513296011955, 6.098712810345588, 4.149635045166092, 6.0522058318805225, 0.8314866568560174, 5.237191153232633, 9.058715067267386, 9.742286198251046, 1.2102276232773532, 5.310026181153362, 7.212627297643543, 6.5488370710714765, 1.1251477250706186, 3.3487285897878447, 6.483680210045546, 6.637363623814361, 4.545680500703636, 1.4494399378620408, 0.903159020997001, 7.893398135033382, 3.960385177411934, 3.0074484685039895, 4.02566508972429, 3.818036082540584, 6.982658577312765, 6.1206402779371345, 4.563429369422994, 5.892503284132003, 3.353769460949001, 7.281939054741844, 2.9935129929825197, 4.915513753823616, 3.2256027151124167, 1.4659696365170733, 2.8071027043229764, 6.2502789742462905, 3.615491724075722, 0.7143727800080995, 3.389568096949952, 1.281201998353365, 2.9025520588447815, 2.9365253579466297, 7.074255692241635, 2.9142207064674697, 4.734082014597085, 4.067695298837337, 7.442545938033597, 3.030459655382987, 4.673403553050038, 1.818562008447989, 5.060348950380339, 5.551379403226357, 6.603177843827426, 3.691241361586387, 3.349436621849145, 4.620616348035497, 0.7000504514317308, 1.079141000573489, 1.4804489951965762, 5.4447521256303215, 4.205353868617646, 0.4943106957991172, 3.6314190742478965, 0.23313920193321125, 4.161239898799685, 4.061100682623054, 2.2071760592244556, 2.49454793970596, 3.2947709570050328, 4.329313765480452, 1.6271432633655778, 3.6666337570456973, 3.7302125350570785, 5.349706330897828, 0.11897236543973816, 3.213868196425681, 3.809434726358192, 5.217816801600387, 1.0873413752305152, 4.519827156243202, 2.7242349262136103, 2.0359166622648672, 2.5880255787663446, 3.339707829187064, 3.437110795136931, 1.4074449911693219, 1.662392256831581, 1.4999515366448861, 0, 4.535987762391659, 1.3439854212001825, 1.3524081846490597, 4.494421546747051, 0.4812547676738042, 4.0644627873276, 0.6960101877356287, 2.2367439806139147, 4.163058493522171, 2.3862587289319874, 4.573994393014301, 3.6244717005877813, 3.7596403817409425, 4.65482945427878, 3.1317966591912785, 4.763920275415236, 2.143798996146591, 2.764761895729984, 3.5425462224093414, 1.3068442997888328, 3.026317371914861, 2.9919025009282545, 2.077112538493862, 3.823115949973497, 3.163282458936, 3.7862321314880596, 1.8942738506398142, 1.641043187179283, 5.38801628639388, 2.22967695096694, 2.083028569245928, 2.4767629528474404, 4.000188296296871, 4.126429329896096, 3.0072051187802744, 1.920658178635061, 2.4888092585253636, 2.5187206172772196, 2.910219031169918, 0.9074857220506533, 6.061056006494201, 10.917775473134194, 0.583161008319157, 2.896994107816021, 1.6628727730319461, 1.3378878255203062, 3.89815168303427, 3.5212094701678502, 3.459370514022442, 2.1688165387059355, 1.8160132342699478, 1.749859186057658, 2.0597744107316327, 3.338657578218809, 0.8377637589126276, 2.1762250877884695, 2.780207203475766, 1.452746282701479, 2.4225197039701114, 3.324759014493789, 5.745048786448768, 1.2814266350272439, 2.926380204349254, 3.7781828797353714, 3.1737625757412267, 2.620811649622907, 1.2069499010224358, 3.632118754152545, 2.7739819739023113, 2.5248777323437332, 0.7139992173559527, 0.939495135128361, 1.806910275740518, 0.012270092591814401, 1.5244658855733864, 2.9075147490890405, 3.10937235307366, 1.3560820966030065, 3.588733999326746, 2.4708063384915513, 2.728297597767873, 2.9393276911204507, 3.6632740370590935, 1.485485956092401, 2.74005019223704, 3.511292177382898, 1.8912765265022173, 3.4323303228789825, 1.9738685585694304, 1.5123572708295583, 2.293014749095242, 3.8678028009829597, 2.343119378375265, 2.2753298985231196, 1.8640463740385356, 2.4350315973923093, 3.3756097536728977, 2.4662087211836514, 1.2074909124651003, 2.71624812127045, 2.1969440848886888, 2.4609728744935477, 1.6670266756484529, 1.4460331152014083, 4.0018533596452235, 2.574228298170608, 0.9620031267432748, 2.7611990056030584, 2.87169345217708, 2.787753795585276, 3.2608827229395776, 1.6432186027997966, 0.23011840670488753, 1.0214820046188715, 0, 1.968103790673726, 0.7168860766920941, 2.469548424057748, 2.7465450280657997, 3.4611884585307053, 2.7847614849441267, 2.638307116181992, 2.546181525897256, 2.1176943322655193, 1.1690321426917065, 1.3026908819593124, 2.1400082257388218, 2.8025308556090254, 0.2748436786145007, 1.7299776318945086, 2.6087842413259135, 2.702861835653851, 2.2297335285923414, 1.8230661126981569, 0.7772020397611953, 1.8397594001637718, 2.694540890314599, 2.5540309268413304, 1.4421916577527087, 2.603295753572751, 2.2491389180042662, 1.6434338507517825, 3.624377922576848, 2.640147365654333, 2.361928188003118, 4.8558783537642265, 2.9123652121681682, 1.9317445912393387, 2.785626848268272, 2.0351427107140068, 1.822288738309642, 2.892681379650068, 1.6152188290855978, 1.2563992938807684, 2.280961054013263, 1.4635224984087791, 2.4024588629260815, 1.201594177506801, 2.257809798314922, 3.025580506891879, 3.1516328604699475, 2.2842164620597853, 3.842865709332737, 2.8580332960411883, 2.721947106791975, 2.4213668167776388, 4.5662630686338535, 2.8719096193907143, 0, 0.9485085003606751, 1.5239855298865217, 2.694710371130133, 0, 1.749013500173649, 3.730503545176048, 1.9743496937868759, 2.455327624824082, 1.217121818471935, 2.1484337923921695, 1.9123017175450163, 2.017217293637121, 1.2451783398889236, 1.9885021175963993, 2.389049003254636, 0.7665622443375922, 1.7529141217767843, 1.1143560417025913, 0.057975073746504374, 1.3952462744261886, 1.398373898554427, 1.4276387173014524, 3.017789567196336, 2.7927917827621007, 3.720288525112478, 3.890384315776678, 3.4245233888049316, 0.2541806007974391, 2.2627507182568856, 2.2488776135601274, 0.3994072162674419, 2.282619106607371, 2.7037963142632417, 2.8008206804221616, 2.654628855582292, 2.116511604025191, 2.248399130792178, 3.1042454704605085, 3.1651480117719886, 0.8634723817068638, 1.3803174195427, 2.650692495458477, 0.792313506044211, 0.7797157189397966, 1.3477226735189831, 1.0310722771160643, 1.6889222953165253, 2.0905980247137315, 2.5658834323020616, 2.461982859218721, 2.2008275515711313, 1.1629327848780602, 0.4381136577145961, 2.3212808865751176, 2.219534504790134, 0.5050942713345595, 1.474893019214395, 1.8788100484899883, 0.5375137432937149, 2.2798547596159966, 1.1614525297091742, 2.4314527165052295, 2.283029652694791, 1.9047305226166524, 3.6684573325549197, 1.4506246916160008, 3.2266867730717776, 1.477751580471964, 1.90741025220353, 2.64629368814103, 1.7233462223622171, 3.263840156326646, 2.6538254164907094, 2.784813358182803, 1.9885784780026845, 0.8240560219741253, 2.6208247100855226, 1.2491918381462694, 1.1215200666890943, 1.8174561035408265, 1.2168856627929352, 2.113122241191137, 0.7307041226066237, 2.421349213101825, 2.6235000184992043, 2.9136965725785595, 4.315486799776075, 4.419059134133724, 1.3338292921273756, 2.117195041534644, 1.6212678485866718, 5.393141053705346, 1.4390122122104767, 2.4196452872584207, 2.513575625734649, 0.11999893378198351, 0.3935526775735754, 1.1781164778041249, 1.863974044175466, 0.053174762487255986, 0.9290626626533841, 1.1709037691336273, 3.042859641134779, 3.41253594339819, 1.0070176517366922, 1.826093256190715, 1.6344399371290417, 3.9810846100222443, 1.0182237777923127, 2.3172858180902542, 1.0499600699426745, 0.6406823780992669, 2.1953765620370578, 2.765981203296387, 1.1398964377353014, 1.9293215513803543, 0.9560390772378171, 3.693290082074957, 1.1467572612015446, 0.2553465628888632, 0.38651024838510817, 2.2361480517806345, 1.7785293701999318, 2.171140880005725, 2.056590680125621, 1.3278643500334442, 0.9452929241024368, 1.2823582113248497, 1.0971585850706045, 2.1021378295354585, 0.21197281893108508, 1.1803634976058495, 1.5029943806330472, 2.1351869816999836, 0.0728316477941173, 0.8517830361514803, 0.9188266312116813, 0.1965643330109631, 0.6214213944336714, 1.784158080664446, 0.5705859760433265, 3.081802442754125, 2.0252428807761205, 0.17867992680146666, 2.258751921372826, 1.2252230396670216, 1.5696739122318932, 1.3732486610343393, 1.253322226770264, 2.214896573050476, 1.2367626432198904, 2.354688816915125, 0.36993779828674916, 0.5882042502722171, 1.9469646097304492, 1.3755639647426439, 1.0504475618406022, 2.500126869586804, 1.879737591569045, 1.4427394569466176, 1.4142486031337516, 0.9277058299660561, 0.6632668683391423, 1.035875337520942, 2.1814869448582592, 1.7165283520211057, 1.5924146958428251, 1.4375425357490577, 1.7978818099757283, 2.0542790523951098, 0.9660823775396506, 2.463375593501606, 2.122329292956364, 1.397905756857167, 0.4537122424213354, 1.18285817300061, 1.7589536459579327, 2.197941929549392, 1.8626281346388407, 0.37091732402580196, 2.715767391162022, 0.49446715294262966, 1.3348461021745468, 0.7992535681327464, 2.423735339903829, 0.6340771455120697, 1.0417805473630584, 1.355023488522056, 1.8619842882876494, 1.0441321974023776, 1.7283739921906227, 0.3092783993023429, 1.6054989615930677, 0.9531536723657564, 2.419145401182362, 2.5327337436293123, 1.656067330366302, 4.195938468978714, 2.624733908076179, 1.3288484372703073, 1.993535273019184, 1.4612919492212968, 1.5707598589694558, 1.1056978162345905, 1.3018295346001187, 0.42472907090609113, 1.6030866360158154, 1.3112992679869593, 1.9290111467644417, 3.574301011342245, 1.8569078214570152, 1.2822462253104416, 2.215667141562149, 1.5560609490179738, 2.0766568000612398, 1.1419894697290969, 1.3592218391486903, 1.3109241850286713, 1.0550502923633518, 0.7365104289845555, 3.102038297076877, 0.3399198958081419, 1.6574377417459674, 1.2173813472292416, 2.1215307944354884, 1.9185067712680057, 0.8266999414671342, 0.45564225271468217, 1.5320848343065774, 0.1330004188793501, 0.5348272487373695, 0.6371521826337472, 1.5717963948259526, 3.358921803322695, 2.2190776125073604, 0.5214432447010967, 2.265495303030284, 1.0457871046038902, 0.5621136244720175, 0.7789680584084631, 0.44121889092652944, 0.6144820748935149, 2.3778033463184145, 1.9884884290012643, 1.4951720539924314, 0.6261999962596579, 0, 2.389623749958503, 1.296017649177041, 2.0823032980653635, 0.7826022628447797, 1.1169103574376418, 1.7756619779967975, 1.8294772365369818, 2.5746301035050316, 0.9362211015758306, 0.4960067566426764, 2.113404030757411, 1.2839839268751805, 0.8742596007459886, 1.1003151282735817, 1.5534871039573197, 1.5800858874791697, 3.619786746195235, 1.5991965439971936, 0.9034212222975964, 2.239152679895319, 1.044327428253648, 1.5377601171072728, 1.1952824378144695, 1.4009592930189902, 1.7734611625551586, 1.932391213025328, 1.6236798256648222, 1.6018475521911515, 1.9349174021785478, 2.1912089565705486, 0.6306800834429955, 0.41795360772158485, 0.09131573661447107, 1.034735903621896, 1.59202486755304, 1.7159583276221941, 0.23736725988750618, 0.5894695739913557, 0.5925567817358472, 0.9090749536097165, 1.8685269336158805, 2.921664621578214, 1.2916719271347312, 0.5368234058919765, 0.7316162156601521, 1.9205229609221783, 1.6482344515035832, 0.970059124967228, 0.4914735949577976, 1.3386801837791884, 0.6199650308332413, 2.76813421495384, 1.7030521247615502, 1.1420973998233648, 2.1912955111916834, 1.127550534059508, 1.9579196580513147, 1.6471904956641854, 0.5181452526481254, 0, 0.9556091622459316, 1.662756985544601, 1.6236392943223494, 0.740344061183458, 1.0367208663286054, 1.8644439947739144, 1.354109979836084, 1.7159076559792792, 1.5924044688746963, 1.1220625265971822, 0.30328688536208215, 2.4649630243325036, 0.3638339431961506, 1.9583983018591333, 2.2403165169013777, 0.9275663103247155, 2.2570789439689554, 2.124480837950179, 0.5129554302759235, 2.0003131322925096, 1.3711642757772504, 0.34100170592627105, 0.20164133768191853, 1.2047648005796303, 0.4152157964261829, 1.3388624261711735, 0.864619230259059, 0.7496285100016286, 0.8611483907535321, 1.0330639774491281, 1.8496465899146488, 2.1969863805663743, 1.218890890203715, 1.5892054920498058, 0.18357108838224762, 0.28701106018598227, 1.3504608623150454, 0.8438708368470473, 2.1505291684454884, 0.9553194260792308, 1.1728349285229822, 1.1710510756661339, 1.9475990699228056, 0.5081129115611707, 0.22439062197283727, 0.16002802489108997, 0.03509131981127019, 1.838723511952124, 0.9999106268030751, 1.5791164712428796, 1.1321255475841936, 1.2367098314279303, 1.032646409063325, 0.6149202625356898, 1.056629706948018, 0.984538123013872, 0.3944871135584243, 1.6782721069354367, 1.7496576534718857, 1.4082496200004953, 3.433900000591305, 1.0198493432010793, 0, 0.6825161724291298, 0.5689701838500785, 0.8080266753294596, 1.344914025187278, 2.7371011960798635, 1.1418187210141837, 2.2435883830723182, 1.9754828834412432, 1.2372360947822139, 1.1511534423534047, 0.42573546226039205, 1.0252012178776237, 1.5537189915464076, 1.0289830618364757, 0.8255379056798383, 1.4787594956957901, 1.0979665697675134, 1.3290931014825422, 0.8624141883788914, 2.0478733760569696, 1.323140593826973, 0.5509717584498004, 0.6811995924062605, 1.1391293447619089, 1.8443877435734923, 7.396807953764444, 2.4010269392278776, 1.8027490811727915, 0.9503668885540749, 1.5164159461440987, 1.7855427489106157, 0.7637613320380987, 2.2906155622897075, 1.0995113579470925, 1.1580289129838202, 0.2761262391287813, 1.3961897864184354, 2.2378823830700276, 1.70676587772335, 1.7143025775911103, 1.1312474561004486, 1.403453898565613, 0.4697264517013704, 1.2657738467419115, 1.5982413869123688, 0.9521679800862156, 2.292275056510749, 2.024738418982917, 0.45805843361528653, 1.2887181837316801, 0.8819915816207385, 1.8565052692137036, 0.7922100858561639, 0.8953472102550849, 0.5403445127990446, 2.4341752895863356, 0.458371589188695, 0.8344044808707762, 2.7186485570219565, 0.473219804278852, 1.412564715703927, 1.1593232002710607, 0.4961099761372174, 0.0425596144187959, 0.24676172937156537, 1.4551235295031724, 0.724159825834328, 1.8510515621739183, 0.8821554202544304, 0.9449348801584516, 0.5999533359965223, 0.09120676821973243, 1.1897959226207402, 2.0794526474839827, 1.0673646518459943, 1.1570851790671741, 1.8530735484954246, 2.0375075990759686, 1.8813763430426096, 1.5786989381558287, 1.378032461095043, 0.4488026226593225, 1.7109085378543212, 1.5439812850542343, 11.253025923910984, 0.8866054617517126, 1.0402695156608943, 1.2138157030377776, 1.978287220763287, 0.9845508749922124, 2.143928172552091, 1.2451353766551143, 0.966267032314152, 0.3812686689237788, 0.3216864023000768, 1.4198734913284876, 2.0366084817254353, 0.8191304889327726, 1.6243045558389788, 0.7626881076147555, 0.32187472054602695, 0.35012629562496433, 0.6325722226888078, 0.7599156576950724, 0.7464039859860765, 1.1098363068708403, 0.9538421174993464, 1.546102338067541, 1.9487484407396778, 2.3487257006998896, 0.5730699043344502, 0.24431900724932493, 0.689588464802155, 0.6115510061627545, 2.0345399548802345, 0.9419057746273256, 0.49188812688488404, 1.1244615536586795, 2.103341693716763, 0.4617340177230234, 2.325138890144823, 0.9039411134901356, 2.5633587145942758, 0.4025555458480637, 0.925565803508922, 0.21695158106628856, 1.6650367814000886, 1.0251220566654156, 0.9408762467011316, 0.2159617714558384, 0.4286396456878659, 2.1704076470610922, 0, 0.7302343046504063, 0.8742749397747277, 0.6664501522883086, 0.9452695152650415, 1.3887301024063679, 1.1043627262381, 1.8073806116302003, 0.5210685265717596, 1.0405162513220458, 1.7140666848082566, 1.831500974548449, 0.18855310654509055, 0.28657826599133623, 1.7550221522762204, 0.9630665009185521, 0.6925386940148295, 1.281416610773925, 1.0415970241438914, 1.0561860051157388, 0.7456354984100261, 0.09331017913765768, 0.028987536873252187, 9.159910954764804, 1.1580857517079044, 1.154032004381605, 2.202081127379781, 1.0875342965573274, 0.2915902878815476, 1.0395537177773382, 1.1903553031639804, 2.3724855962703155, 0.3153508665379205, 0.648489373283407, 1.6259697147322782, 1.6947009001744189, 1.5575442359323595, 2.687647345890837, 1.0408152447598615, 1.621165075601636, 0.7896467646232902, 1.4150150485489215, 1.6878158709968958, 1.479424283600704, 1.2223298060397245, 1.6505318154673394, 1.2785350848685337, 1.6078690389195636, 0.10309550902697402, 1.8392594970161524, 1.291622200064728, 2.807032616626961, 0.8365595874784604, 0.8934876210289125, 1.0164846394672553, 1.602302396199479, 2.950881002589076, 1.502123288264631, 1.1534096958896967, 1.198664202136885, 0.295095146333567, 1.8874190154749415, 1.0786394368583194, 1.0061566389877197, 0.5485557788103206, 0.994132203166252, 1.1816569817020197, 0.4883750666669582, 2.2747235259457064, 1.4880424998948028, 1.9872453527036795, 0.719330056889818, 0.9805522819519237, 1.3033909321990944, 1.0125276408295418, 0.5114805502818554, 0.9251749344354316, 0.3195003868337778, 1.146914001114519, 1.5743966967556913, 1.2160815751831975, 0.9692314734162109, 0.794943730381475, 0.725504156478394, 1.8935299129744763, 0.841960132222253, 0.5724764732463652, 0.8336402209316345, 0.1513806562510682, 0.8044496817729183, 1.103765246947933, 1.3715831278488708, 1.1439950925435323, 0.2558786987769769, 1.3153929828507087, 2.8809231133669337, 1.3710152172964762, 1.8419846978269463, 0.5235719200684512, 0.8417000491021317, 1.0279554979508683, 1.3456402950987418, 0.8876437396528624, 1.5505171134455062, 1.431246857540216, 1.4769914899002756, 1.154655966563515, 1.0204372048137698, 1.0763857123344156, 1.625832834542377, 1.0296882038173698, 0.11672944802874499, 2.185676784989865, 1.291203919672172, 0.841522601728528, 0.643482458110697, 1.3740803489928375, 1.631528275161194, 0.9784238142408676, 0.16623213570245457, 1.256768313879598, 0.8374285796025506, 0.2157963021663483, 0.8564811117866231, 0.6686968482304778, 1.9837213713860895, 1.45906171134326, 1.4206222954601986, 0.7860227090063261, 1.7847075914175006, 0.6033318793231782, 2.4884797741410294, 0.062490293999384536, 1.4205720868342697, 2.5486016091599883, 0.7952228687879979, 1.3383552427891203, 1.010951313038664, 0.7558353998684844, 1.558516239000617, 1.9085425053160017, 1.4798436922512477, 0.5766244601982842, 1.3058636003878896, 0.44618830027295076, 0.9645863783794745, 1.2307326635217475, 0.47114705031195636, 0.8888693041126174, 1.0486564183239078, 1.3805640500345655, 0.3280117489246403, 0.3594682083927429, 0.705524646549982, 1.5642721423602455, 0.36681198168653734, 0.39828039873031645, 0.9875514260897481, 1.3781171649158355, 0.659385217012446, 1.579435875238793, 0.8756700356598213, 0.09836494979928195, 1.4295366149634854, 1.0721781090415126, 1.1270717268672361, 1.689001405022979, 0.42539589498098956, 0.47413383360163935, 0.6996854404119761, 1.0293714677263806, 0.6058057531443842, 0.5210668295731862, 1.4086487070404257, 0.4786630322878883, 0.5954801251405166, 0.7689432811614483, 1.469861851344547, 0.5624668319625133, 0.8439499180164969, 0.4464267564599082, 1.2084361212734183, 0.7005870302656344, 1.163714026479553, 0.9012726970325973, 1.3712295060523427, 1.265818311861342, 0.1978384620117558, 0.39394323882453003, 0.49585649836859785, 1.0007464066222125, 0.3565025617944118, 2.2984451505877126, 0, 0.971764451396366, 1.974058279081923, 2.633140011963349, 1.2760887077755823, 0.25307914411519467, 0.2041611004019197, 0.6881970404940668, 1.7386776461578932, 0.22611233557735577, 1.4886223730007027, 1.8173447126211733, 1.6604280990663962, 0.7755652678063016, 2.153985663155303, 3.28496916353248, 0.6687393420658284, 1.4508819879717918, 1.195269109944913, 0.2945423087292987, 1.1831365913088987, 1.0206806243269826, 0.29230935447673867, 0.3873500997719891, 1.219888357657332, 0.515313691710958, 0.10014344199234412, 0.48734193997470143, 0.7747724945868295, 1.2367935832657657, 0.7559995564016708, 0.9917478249383023, 0.9102812627098789, 1.246639984267347, 1.3340386354187939, 1.1939028553605084, 1.0366238814568332, 1.1720746476342983, 1.0440281660359219, 0.7075773154801339, 0.5307165575782863, 0.9699827903607242, 1.5389458503430327, 1.8870795160892937, 1.3979075430511616, 0.36291638225039063, 1.1116325176812447, 1.2218351192964256, 0.2651812957274725, 1.5146938316183882, 0.9599874249427078, 1.1247276320284882, 0.7221876777424493, 1.073112821436969, 0.8949313786035242, 0.7223750157973867, 0.3008121604162799, 1.1418565575102413, 2.873860381935701, 1.2019881506201617, 0.0532498219377879, 0.07894611096987864, 1.036714400485451, 0.5436384026793856, 1.3610073885952207, 0.7459245387167862, 0.15757411813925826, 0.9238721032136651, 0.9063264623331742, 1.0524385565105148, 1.1223963510575001, 0.3866694262070999, 1.4983992741557113, 0.5203932565162365, 0.7939440781792189, 0.5265086898869261, 1.3234446791833099, 1.566503311762432, 1.4022788499959453, 0.9810302716057591, 0.38607588709679025, 1.1255876277086059, 0.8532338203639309, 1.249636872683189, 0.4770730120202962, 2.1001214595035327, 0.20802028566005812, 0.9046566949563639, 2.1642454558445205, 1.4175453885645686, 0.7577134186933413, 0.38838819952985787, 0.7308294348592322, 1.0779103278797104, 0.6382523324168128, 1.7701523601200746, 1.2424584283689082, 1.7707711817465557, 1.221631612282467, 1.6990635508023324, 0.4048156418588077, 0.3693095645358461, 1.231863299557772, 0.14294101625391054, 0.7880316933120783, 0.6033037338266343, 2.0670345216427233, 0.4106889827117437, 1.7235634793968875, 1.4732992996846064, 1.0404605153846034, 1.0889220005419133, 0.7088215483387402, 0.6442508168988107, 2.1044434067017193, 0.9873431866058309, 0.41101689339744224, 0.2817415334134015, 0.2543169527186331, 1.1626645871134291, 0.8365487862318423, 0.35657220988804517, 1.1089601404796738, 1.1974175771031346, 1.0245751126689808, 0.6620546510337761, 0.6851826502260662, 0.7237879464709582, 0.07410797215372183, 1.5205039755530316, 1.1502770341818744, 0.595676606642706, 0.3817737999887761, 0.8098759815990662, 0.6734760613485337, 0.3758385828326696, 0.6089274856043678, 0.7861484003176599, 1.3797688416919747, 0.8566023574911532, 1.0545182847795211, 0.11491583928733101, 1.3080145719311458, 0.6804285283725967, 0.2224398640556682, 0.12258559690016313, 1.481628691434121, 1.1400037792325999, 0.4960443120146437, 0.9124571839032352, 1.130737326043242, 0.16161595354846522, 1.1210853828356062, 0.5493835203144869, 0.3813233914570485, 0.2789933730475721, 0.38556517662764667, 0.7008736200748316, 0.2684538837104264, 0.7393044023641173, 1.0926018458577695, 0.8182738329187502, 1.205652818987674, 0.8994945941001979, 0.45599545663389707, 1.5231070707736316, 0.6562962418167554, 0.42644423255481445, 2.488091493741689, 1.3819146004842522, 0.9813688957287603, 0.7667251037191386, 0.2091567047502142, 0.9809889196889012, 1.1170471441657521, 0.19831328029004647, 1.240728384384879, 0.4547558790487937, 0.8232421600160281, 1.1357495828194075, 0.5853523771650928, 0.7159160454473048, 0.9121052126908039, 0.6636637998660512, 0.13271034858938896, 1.04409024299824, 0.5282435192143586, 1.1048698266723807, 1.6593728585795318, 0.8437753880760761, 1.448941154260497, 0.7539959773123434, 0.5811331198002344, 0.7372725906311728, 0.8342214908804864, 1.50873609718037, 1.0085834024319422, 0.4616044965065871, 1.5357408244406574, 1.8088378371957188, 1.8289663275789632, 0.6225721947744451, 0.8786118483827825, 0.6179991007692404, 0.9984194109668398, 1.1731103857573406, 0.34267424265049673, 0.5881376036669067, 1.0694988702742194, 0.20045629679091453, 1.3879215828678548, 0.31561283276428476, 1.0886231556226096, 0.18270696588427698, 1.528120276126083, 1.0550753555713874, 0.720006039304713, 1.2467687992838332, 0.22035591198097415, 1.3922344267554403, 1.1469132152239563, 0.5688400113714932, 1.1335634846758174, 0.7867811505985256, 0.649003991993914, 0.9944227646742894, 0.6587752169682052, 0.7173378942481332, 1.2615333120480992, 0.5900098863221653, 1.1528176543295372, 0.20048806835152944, 0.5146365599589037, 0.03174869831458027, 0.6996879445755171, 1.4822775854293468, 0.16750647100881408, 0.5973254469078054, 0.31619365278454614, 0, 0.22935352571997034, 1.7030507248408566, 0.7323338089740186, 0.8177854840874221, 0.6265107602915491, 1.0643142114902622, 0.8570739518746039, 0.31930653924713864, 0.9901163260362618, 0.14854677193089133, 0.6333059003286908, 1.4741433298259607, 0.23111172096338645, 1.488172087712819, 0.22095139902249533, 0.2669670498570998, 1.5152005180272747, 1.3293829262274635, 0.13371635286379388, 0.7718742908227898, 0.8585171917413231, 1.1593687275737856, 0.9260498859819559, 1.110191169379655, 0.8428397056395596, 0.5862044747849459, 0.9876322721232321, 1.0042922779461896, 0.7131139359297223, 1.202353196823057, 0.07430831273337618, 1.3911903562291728, 0.720538639677923, 0.4338648800765098, 0.10114860178581384, 0.9830799095656944, 2.4951751491276815, 1.031180937002209, 0.45269466087451854, 0.7379863961564487, 0.4796831725830404, 1.6682993264600998, 0.8428728600707215, 0.49214591407109576, 0.0408219945202552, 1.1369556544213342, 2.3704539169063317, 0.7506998813706381, 1.256024864917986, 1.2581096698879655, 0.6909261311752382, 0.41136409884398883, 0.19633912169601753, 0.7693243853156676, 0.24975725585248443, 0.7035331255643861, 0.7421040422457067, 0.7173095747903415, 0.48886507186005057, 0.5793304547508874, 0.9004455894535338, 0.45894420488806553, 1.2799183888558363, 0.044451762570833796, 0.028170876966696224, 1.6750522037900433, 0.6633446207316062, 0.4503659913400335, 1.150570477753102, 0.5781122256039253, 1.061458468685919, 0.620746204189825, 1.8347341836106734, 0.9667230320317912, 0.6419111104808689, 1.0731159492137623, 0.4656893616945584, 0.10536051565782636, 0.413253795323003, 1.0892182046622811, 0.1814155270275101, 0.16949605544493346, 0.40370596763774946, 0.6325703775854989, 0.5901222222018911, 0.5014892750479035, 0.11979429080085109, 0.9898063533926802, 0.7755331526679341, 0.7073824630680741, 0.8415703609017127, 0.4790149549443691, 1.8296114658667957, 0.38919714555431134, 1.7809198718129355, 0.635906128602613, 0.7537687653227471, 0.16658418725780322, 1.2508338698801513, 0.668098039555255, 0.9062660917270092, 0.7722965162243175, 1.242028985400861, 0.8498416564326783, 1.0083882548751952, 1.4186157412560156, 0.4105459736381775, 0.5157819632482199, 0.24818234783560184, 0.5213366596265261, 1.1200902648885536, 1.9441285193529103, 0.2167765603362825, 2.4914202482095527, 0.6567030774694698, 0.8374942469507329, 1.2664085085965646, 0.12513760049605388, 0.4884674453067171, 0.1107314484011736, 0.2504858368196332, 5.690752098521655, 0.42767750695964984, 1.1951108920811815, 1.0919866405533898, 0.5943872819377296, 0.5102306633519895, 0.6876774671904194, 0.13179377272598192, 0.6811796838235467, 0.14913463982451608, 0.2450791997897924, 0.012578782206860185, 0.4954252465114691, 0.32107606368787184, 0.7886183193364542, 0.2783483644219376, 0.3646061107779967, 0.8512477938402178, 1.022527714858636, 0.3827111908099577, 0.5453137294701424, 1.20831414406162, 0.8837580029779564, 1.0250839426444038, 0.5256373006960726, 0.10536051565782621, 1.2134267481507977, 0.7359908007118331, 0.9778101615035211, 0.43979546263943003, 0.7413847484722639, 0.6283068340561194, 0.8152143646941995, 0.7381734559543323, 0.4540672339385766, 0.6042981944259459, 0.7372159838432155, 1.2213209426581766, 1.7484493268903007, 0.6289340086297286, 0.9821247880663524, 0.9851801338617209, 0.09984533496971609, 0.19999809268415186, 0.7276601056346101, 0.4626447736999166, 1.0351782162382432, 0.3549935979022495, 0.5141669783525691, 0.729321726824142, 0.8134954416178898, 0.5126695132133551, 0.9086637971342564, 0.9514566724923212, 0.08768500449910396, 0.6792360127966764, 0.3352731922035778, 0.6097569973434218, 2.634419443414199, 1.2928473580151316, 1.4248534906331833, 0.9548469526925367, 0.7669920947137617, 0.30789568886606417, 0.6013019929849424, 1.2527540037466987, 0.5760227036946102, 0.09043486544115059, 0.7627163701871569, 0.5040546680044365, 1.8626507669052477, 1.2134709553546559, 0.16448488600402378, 0.8163695594435648, 1.0004967456850848, 0.8727127113423703, 0.3650171886605824, 0.7225233833985685, 0.9814899237533793, 1.1571593941530534, 0.7320197855980842, 0.8757426216449888, 2.2624373576298007, 0.5933022146612906, 0.7225321394682945, 0.8374056865666457, 0.5282246882624302, 0.2566815040472987, 0.43706689472462484, 0.6951404944685362, 1.1452460137981455, 0.35084136739800403, 0.6926891110405151, 0.44684892851628766, 0.9578427089383158, 0, 0.6658481325261203, 1.0696792054275654, 0.6311049568543535, 0, 0.5717635211295021, 1.2307173841210162, 0.6227867495952645, 0.4856729586319426, 1.005715480529601, 0.9346610206966709, 0.7302003289471191, 0.7750829528863031, 0.3126034313290494, 1.247868678521054, 0.6003089911578701, 0.2174045554415931, 0.25270114769716506, 0.36063928218988295, 0.7373731599329608, 0.6970338199296388, 0.4501050652229555, 0.7459655494377939, 0.382618851379971, 0.3155496408203484, 1.2421374701881436, 0.5185835250826135, 1.3247949696350916, 0.3645172394142654, 1.3621827185541666, 1.406907050832896, 0.6106245966417491, 1.2929504030815548, 0.8221390391859555, 0.7889511880838829, 0.5664333362563116, 0.06371581438610785, 0.43810718811366955, 1.0997478335739024, 0.1722144518087348, 0.07457135683794586, 0.6785982611407888, 0.36057663280929086, 0.6466644175628217, 0.8567752540115844, 0.124336417116832, 0.22783532229349493, 0.030771658666753687, 0.3720520846336401, 0.1575447364429325, 0.5598768050335242, 1.1762606421495014, 0.745924602138617, 0.8156261751970162, 0.18470189702737974, 0.19448242071255886, 0.7742494197842608, 0.6058625985475518, 0.44740147580630796, 1.4532303401908626, 2.6005118633755937, 0.9723978377229063, 1.0067458525734934, 0.2573349160624888, 1.3867168756710027, 0.6068524630606812, 0.7378005492241435, 0.42604697273568415, 0.7600322635747886, 0.05991957528127649, 0.32662007800699877, 1.0684114240475107, 0, 0.8587320071165886, 0.1862843820953905, 0.873017846859436, 0.43566425844255996, 1.336194587966107, 0.6368459086107111, 0.848727167401854, 0.7125141176512325, 0.7330550946713541, 0.6762398327174901, 0, 0.6357740444502433, 1.1841527294097918, 0.736376047363958, 0.5527969347647514, 0.24869221288064086, 0.64767549047393, 0.5910274089952313, 0.48554460904317487, 0.23013272819246738, 0.6455039611386613, 1.030590494240074, 0.10701991611384343, 1.1149244293894849, 0.36119994682062667, 0.6820802831578188, 0.7515130920294628, 1.2015062846301536, 0.2866772459543456, 0.2522605477308466, 1.0327994961697349, 0.5377131352437586, 0.7669711280181302, 0.1497799243030974, 1.0742237941124375, 0.6227946641499922, 0.36522722258295903, 0.21056358454252658, 0.8024799294074826, 0.505897209098441, 1.9267860734801372, 0.7122224995827072, 0.49046624222046875, 0.48809655006849223, 0.5649313262761647, 0.6416771674246349, 0.24479368730376905, 0.27259785851469526, 0, 0.9036148000600662, 0.4407651429453925, 0.6661961432464701, 1.1476212289912215, 0.4377624561398292, 0.32404336465329825, 0.5235067421626788, 0.5361042173274218, 0.8026067323165984, 0.5373827949489647, 0.5395535993203588, 0.6956973332782839, 0.6613294159505778, 0.18319436556427587, 1.1914691923052185, 0.5209214458910543, 1.0792657729085227, 0.5326207261425661, 1.035563714108299, 0.4947607534540581, 0.46660626922666293, 0.1291655246209433, 1.1695085681734565, 0.5829473882613696, 1.077301310895584, 0.8299862302403664, 0.14483846353921415, 0.8596775268207716, 1.1700418942954462, 0.5043194932526581, 1.0892413407081354, 0.5664788406092498, 0.292128114398295, 0.6113029913091719, 0.3946734035232055, 0.809819920735012, 0.5344244246285306, 0.44638184974532846, 0.3099834458375016, 0.6188589068634771, 0.3878715692418824, 0.5581242226716837, 1.2678127077950208, 0.781346653417436, 0.32423640665463477, 0.7459620619411547, 0.32810398695069615, 0.7585724473349502, 1.33529605607125, 0.5529640656423179, 0.8329842679307473, 0.7860210606642045, 0.16000461053106182, 0.38949675844829995, 1.0079448045173154, 1.304124294576299, 0.6463963214950758, 1.223267832690184, 0.6939825734905254, 0.3716272744379525, 0.28124495112676157, 0.4569094645999007, 0.3413164953129707, 0.7083057093033044, 0.2631474648032968, 0.8262790560319718, 1.4585064411637039, 0.3484702543423082, 0, 0.6038753953690279, 0.4459534634260533, 0.6446608428299967, 1.0812729457658183, 0.38699399891693975, 0.36362468767875383, 0.6132471135396748, 0.497669703376431, 0.2718323353227552, 0.6172721778142967, 0.14416979082957831, 0.5645406471351202, 0.45246107458324913, 0.4790917186677144, 0.8610974952343016, 0.4835413589076637, 0.09739234600864946, 0.125163142954006, 1.071520517264419, 0.7178778815383435, 0.6923732934229389, 1.1252373017734274, 0.49831362442795907, 0.8711496352763477, 0.23091101995133612, 0.5224273034616655, 0.09783881014357021, 0.26071961610546923, 0.6370426343575392, 0.7336234154563291, 0.4337902487501889, 0.6835362277521114, 0.6734034449715107, 0.875587289767952, 0.22259919895634128, 0.706103541113916, 0.2648835382340424, 0.38020607051037036, 0.6884777664599228, 0.6428798866271049, 0.5151327322031377, 0.4436639544342874, 0.5540191783316227, 0.6417029552507834, 0.22333949527486405, 1.1727936870267874, 1.2810041660145488, 0.9798185708704982, 0.12615128532690004, 0.9232030041814339, 0.08004270767353636, 0.45302680777224935, 0.49571363861294804, 1.1317034890687445, 0.6122693224092172, 0.8420845839943671, 0.32355803134639477, 0.27584210785822405, 0.14099968723204848, 0.3680132083259772, 0.3987039705897163, 0.3745762657280707, 0.40879706165436785, 0.5871198363043505, 0.39666417104033536, 0.29270426476891354, 0.9654696422549237, 0.3228294602174265, 0.6276409780539345, 0.5720511786359046, 0, 1.579814667303913, 0.33397387947767754, 0.08198698338446632, 0.1941133013898803, 0.395307363230997, 0.8433472939085803, 0.6350312285890329, 0.6080060072361301, 0.12759985730440038, 0, 0.8322261293113676, 1.1104932153976268, 0.3494361324927101, 0.719433020616825, 0.4049436923340677, 0.2562084726879584, 1.0679311279988368, 0.9882649578006826, 0.533155884589547, 0.17171581218324633, 0.3087435440526125, 0.6144262527825888, 0.2913050384984892, 1.1116925244934828, 0.3569976547120049, 0.5092922158344467, 0.16872954433589973, 0.8142432846684883, 0.49394709070172343, 0.516844488069438, 0.3605303370040786, 0.5551243576227974, 0.4047248237520769, 0.6021674617211571, 0.5675330610475496, 0.322212058116264, 0.13044972608711447, 0.9392139816759059, 0.07164187965852853, 0.3707304832430949, 0.6229721064685989, 0.572013228884324, 0.2432511432608958, 0.3536637596899678, 0.5987097980703435, 0.14021434441331984, 0.5959844612643516, 0.5897905075707761, 1.100963272596718, 0.6037130420389715, 0.356403759208021, 0.7407904948782653, 0.4065195315897462, 0.5855417341360488, 0.6416136048525304, 0.39870084415626134, 0.7389417967476936, 0.521865152091221, 0.8801153720923166, 0.928221022132842, 0.17016733317091168, 0.9364394235460298, 0.9820483380060505, 0.9320617231084815, 0.7644301363835005, 0.578707635417, 0.8575885928351241, 0.42937894955594813, 0.3479820669058036, 0.3084901632529864, 0.4896747272789882, 0.6143931543841475, 1.1335890384899268, 0.23835575542584786, 0.6813042647332777, 0.8727379460130421, 0.6474302876829474, 0.73742268310049, 0.28127813309124666, 0.3500252030259162, 0.1848358757606674, 0.5124807997619253, 0.36898501532957345, 0.18301144988789236, 0.4235742888730473, 0.4110242183607634, 0, 0.06876763192413925, 0.5990173831553967, 0.3588735189421014, 0.783580215344623, 0.6803885221666138, 0.8721065245856832, 0.939185851837129, 0.21858192976525587, 0.1534343107642322, 0.3264399397356187, 0.41717625087873633, 0.3795000447996922, 0.47042898036753045, 0.35203055404255595, 0.4121389183153112, 0.5543446447972241, 0.4418631725118696, 0.045778113699440466, 0.9666970003251335, 1.4652022034824044, 0.21723571233684608, 0.4911872825256826, 0.6593091632621757, 0.7022803112362981, 0.5320190753952889, 0.657740490871785, 0.7747216417243762, 0.154798172405566, 1.3294256212770956, 1.5720862399446949, 1.1386270610142608, 0.3485465284769859, 0.5902666011657715, 0.8277918645821243, 0.31710644531451326, 0.9964364699597936, 0.5962683332965658, 0.9082208970699527, 0.33387145492115544, 0.7890816747333501, 0.8637591748189674, 0.13662105912993094, 0.6551283426217993, 0.33267212912285954, 0.519089838894882, 0.5610754280893374, 0.371163614821461, 1.3244095445309538, 0.7948978552041582, 0.4506983101911916, 0.1326295437733781, 0.294792595800313, 0.25993075031678886, 0.46058765277061586, 0.2978139363536345, 0.16930671760764415, 1.3836089770716857, 0.17395330712343804, 0.5923985741673681, 0.23103091614737586, 0.29884644918537545, 0.24275789098021383, 0.6352182512320717, 0.3485891581775612, 0.4830285053156303, 0.5853762117700444, 1.3468911396840195, 0.25743844284028966, 0.17753481987242198, 1.1421134923626521, 0.36267320842808176, 0.22160924857430375, 0.5625682124176378, 0.36039411527087084, 0.5495038163034719, 0.6542662669491934, 0.1508078484933802, 0.32166693862403495, 0.6450965596284646, 0.21845355109309747, 1.013982894738948, 0.6002194475112272, 0.8593061905674129, 0.20847805632495311, 1.1361388567759505, 0.14167362332439143, 0.19994180174138654, 0.7310773569249888, 0.27677980686535375, 0.6350257910134792, 0.5644204341199385, 0.5315082542231714, 0.03922071315328133, 0.15141995205463807, 0.32341443675012227, 0.16279848220274373, 0.4507211507684611, 0.13335281030246007, 0.2921271540685945, 0.7048030617614146, 0.049377880906889514, 0.4207687354987161, 0.5652674606476554, 0.43093728824611915, 0.6075683159036517, 0.25167451152815046, 0.018692133012152546, 0.43360611368748897, 0.32435221671431, 0.7206000021210671, 0.7048319438167499, 0.22053705265684698, 0.9159098027105835, 0.526224124782339, 0.5898362755874662, 0.8349665134733337, 0.07386984856645552, 0.816948746299379, 0.3221215425848999, 1.024610955258494, 0.42466889629301485, 0.5206144073964457, 0.7190700723344373, 0.8153478543480397, 0.03636764417087479, 0.3322991508284768, 0.1357207363430507, 0.7414138481418068, 0.21272967698974593, 0.2918801621509113, 0.18110130038872527, 0.756113810197694, 0.5968619770136343, 0.7187375259581266, 0.7267067626019074, 0.3492272952749084, 0.6636490711976902, 0.28024140663688996, 0.1994305411742272, 0.6161067564842164, 0.27470579317494215, 0.8714696638878103, 0.6128850772628021, 0.31596192342828355, 0.7940664955163249, 0.6086421704036342, 0.23179409839656642, 0.16510819323762985, 1.079337048014004, 0.11705309943046528, 1.0920114693007803, 0.6265369364846709, 0.4439837066374146, 0.17515698623411835, 0.24774792168255405, 0.8173253623795014, 0.2981741202686728, 0.5684618121400291, 0.8110077569370139, 0.274454583903639, 0.39984293326359815, 0.5919692216497224, 0.7531252926257537, 0.17068365825581525, 0.05197705509795097, 0.5457019995930225, 0.9143917026025021, 0.7180977726573474, 0.2841378253675481, 0.6616608227048786, 0.3293583437537214, 0.49391402517368393, 0.16319544732160302, 0.2892963157077133, 0.43649560813810917, 1.6368590735176163, 0.055539883474098266, 0.7252869968707825, 0.5488364222100213, 2.1360172629591587, 0.2253819343287753, 0.38921725525821776, 0.4158597978485219, 0.8109365011913455, 0.49045818057188045, 0.542093508656007, 0.14498726199198483, 0.8220787549153018, 0.2847154424062763, 0.6472880529963285, 0.3836715231525481, 0.11200528192887879, 0.3706377913531478, 0.49615045596445767, 0.33251551841302257, 1.0422846763857596, 0.6983845946091849, 0.08923113372794261, 0.5539910270620467, 0.45382758121953987, 0.9910218985629372, 0.10912209431664226, 0.7928670007194644, 0.5553275729204288, 0.21754922852641834, 0.9673445921840724, 0.4981989494659324, 0.36951091471065756, 0.14685576147422436, 0.01980262729617973, 0.5681215527124726, 0.5632743771459046, 0.5875716041202211, 1.05221081000131, 0.07856232250310231, 0.45141630839368596, 0.7526283334196838, 0.1757989533805152, 0.49803076936872026, 0.2909524029236717, 0.9260785129556333, 0.5353171354768952, 0.5239697275038657, 0.1173294180482422, 0.07361181734324618, 0.05608946665104373, 0.37286858537421846, 0.1839323641931933, 0.46585772488628074, 0.28270467074947364, 0.870452636194415, 0.4561083833703655, 0.20226970161119645, 0.5759984609176978, 0.292738145446666, 0.1913031399164804, 0.24029057385173974, 0.28004109021814083, 0.6052533878133238, 0.3136687476036361, 0.7780820287838425, 0.5275860830302185, 0.7826580094093181, 0.7580313912480452, 0.48148611322569124, 0.48673412742854505, 0.4421134968077764, 0.30991643402981484, 0.20172218145056553, 0.15616884398349545, 1.055079513097927, 0.11487013953579013, 1.061921866109665, 0.6039733085535243, 0.3818237045048189, 0.4351344119292669, 0.8242091445918815, 0.8968015814281762, 0.6792373338305973, 0.3871681474930775, 0.4887824938015798, 0.5290311898468185, 1.0591739596012448, 0.6942751643968715, 0.5222005857663672, 0.2230687933431063, 0, 0.19465817314642828, 0.19174322436702682, 0.8067397829771699, 0.3521533191638738, 0.20676910137672783, 0.4477805120676363, 0.15076421598762682, 0.34755265907737803, 0.38242575590294997, 0.7227565004258097, 2.3828239708598398, 0.4407419727350909, 0.6328763171538556, 1.0214838094133307, 0.3217046395320755, 0.6651538745169246, 0.17970048820456055, 0.23509997948424058, 0.1334074728619809, 0.6037722701813345, 0.11788826004749452, 1.2193239887260188, 0.8100896761880639, 0.12492161188342318, 0.2737871757024896, 0.34627113390075126, 0.6799088249389066, 0.41063277930113723, 0.13358954087125752, 0.6989367826508306, 0.11583181552512165, 0.6457561426764978, 0.026668247082161273, 0.3812152917168682, 0.33360451217672316, 0.9464628024481084, 0.18149869888632403, 0.35972964615097824, 0.3311200123676761, 0.1202861658745019, 0.5729773770402814, 0.32711856212745505, 0.3285222077960759, 1.7877484421276348, 0.5364394549970123, 0.4441818710672654, 0.4846238379628982, 0.35248020150536435, 0.6068066817741015, 0.5847196138851167, 0.16095282749789402, 0.8928017201051699, 0.3520110258319682, 0.11285396419673491, 0.8172129200544928, 0.17553891668498334, 0.7583591658293781, 1.0795822017291907, 0.233572746804749, 0.5233312383011355, 0.03207271988799413, 0.20297938579776648, 0.43827624427047013, 0.5279824432970771, 0.9195338287147744, 0.527444820271743, 1.084474672270432, 0.41501693114235944, 0.49516177055284977, 0.35138854277313125, 0.4420927548593041, 0.09186444663870333, 0.7808111581505996, 0.47429086487642097, 0.32848262397943023, 0.2634646564931092, 0.5972099989909873, 0.5154954681362457, 0.593135239829769, 0.31823132960372946, 0.5664820022206905, 1.0029728369363204, 0.231401422181992, 0.4654777222812233, 0.15154989812720085, 0.7466079695289396, 0.22064737640483167, 0.3102768568883153, 0.42734736991444194, 0.40651777194832084, 0.513818873358123, 0.38994943880116245, 0.2997515201127897, 0.6265230271417278, 0.17955842773532094, 0.6715935427243367, 0.7283109427626551, 0.8518022150926116, 0.909208925966726, 1.0734196927938753, 0.15830411437897318, 0.3229791067174209, 0.02020270731751947, 0.3916467901251791, 0.6208931434723229, 0.4215142318984128, 0.033901551675681416, 0.46335680664441725, 0.24260741802541386, 0.6103187499699335, 0.28540745443483667, 0.6227439388088143, 0.27601143940990447, 0.6381180851330562, 1.522132688877931, 0.24448076440925953, 0.15923330592067625, 0.6739616429907745, 0.766251097971461, 0.3920049789512326, 0.24628886224589935, 0.838223576039351, 0.557602248344496, 0.46601091044705273, 1.2369228728932777, 0.7664271554063734, 0.4453784872074972, 0.19423980513186084, 0.572439693767298, 0.5432656858691047, 0.2889530470326795, 0.581186483448571, 0.4947312463623718, 0.8658547670629021, 0.6875676521858916, 0.7715401287056814, 0.3656499201492584, 0.5032480638065697, 0.36260421886987787, 0.17485420229981163, 0.6107987591012783, 0, 0.4786493965611849, 0.502108632273064, 0.8917977237145515, 0.11778303565638347, 0.9613540028839942, 0.2259966202966162, 0.15164679181838775, 0.256312182716271, 0.7025196440488415, 0.40731867675748734, 0.45471489812168975, 0.13886284296411258, 0.3034997825903715, 0, 0, 0.41399095543163034, 0.6995592437872082, 0.4639839654562936, 0, 0.2016051979076433, 1.0115151433274265, 0.7575297775819427, 0.22577865804698727, 0.18942891535135672, 0.4396373328855287, 0.7491640982395191, 0.32178470999180364, 0.056747145070014424, 0.5145329108649801, 0.00985229644301164, 0.23247506323079592, 0.4453592556402723, 0.16164251578935768, 0.7489356523536458, 0.4999116412345651, 0.6559716054054445, 0.38096088921660354, 0.4688371681661183, 0.5687122228851249, 0.5409700169824934, 0.5709012143845899, 1.0663111971052839, 0.280621508705645, 0.11975957563729364, 0.06022102531776139, 0.9972477294386018, 0.19215003903381478, 0.355571471216959, 0.20517171737658252, 0.5485026513997056, 0.3021227566065753, 0.5660520524370415, 0.43815131422612164, 0.5351493156000208, 0.3623366027000468, 0.2631154483288812, 0.7424534813109702, 0.4028590760583026, 0.24076097089627777, 0.5086103820829355, 0.20835811988601027, 1.970601123927713, 0.23811467563169572, 0.6794882088356652, 0.12876197248457757, 0.14844253895401363, 0.7073479327071177, 0.049190244190771656, 0.523955167432047, 0.2276055739348507, 0.13350120348959, 0.03174869831458027, 0.1378506830822805, 0.2971414355401851, 0.9139029547360911, 0.12615128532690004, 0.3427476644363627, 0.28661747414119954, 0.12226270585541217, 0.36387593104100274, 0.6301276289459827, 0.3626807514676143, 0.06669137449867238, 0.4767991079526364, 0.4430662949748586, 0.47015049219817606, 1.6300893391620892, 0.09077785978510655, 0.220857981079584, 0.29339766556797797, 0.1628815869438988, 0.29771566108452674, 0.21540242725820666, 0.07875983386064704, 0.4569556515969763, 0.4599137068244252, 0.46852158094775925, 0.36003688641345055, 0.35225285123876277, 0.31884368088377524, 0.20520585062754235, 0.29471374381546156, 0.7109112477906379, 0.6885742000575461, 0.43463824564314896, 0.1537827191614463, 0.3178598088705124, 0.668675435788984, 0.5402782388733406, 0.14826108576950808, 0.4887290539556347, 0.337889771564667, 0.30888856805978965, 0.11143295014210414, 0.3071101980255718, 0.661749356506907, 0.08788298934064327, 0.7563587079235535, 0.8103061383531727, 0.19755674534578269, 0.454910438661273, 0.4065543337735275, 0.033901551675681416, 0.7714611972966283, 1.1276588986081066, 0.044977995283077415, 0.2530345894252123, 0, 0.4229655886869693, 0.27641145136297535, 0.05129329438755048, 0.36063974040208685, 0.15875658203654305, 0.5077585818997231, 0.320386651539633, 0.5442288353314084, 0.45367950612445834, 0, 0.25031712049487576, 0.24314039710742164, 0.20920647144265536, 0.3498008928933422, 2.562826669720248, 0.6949283641772808, 0.48882575761020686, 0.08579041946372903, 0.5744787599883551, 0.5510209520921728, 0.6980576728733597, 0.4255076536605995, 0, 0.6631290036565012, 3.8033323865356445, 0.5819105379405212, 0.6505472338587057, 0.6366646099015953, 0.8726632523813522, 0.19007498343180862, 0.6808979219609087, 0.6347653174231336, 0.22954052031853284, 0.33144661591453123, 0.3495975572095822, 0.3856075138908285, 0, 0.38299225225610584, 0.12309847690280823, 0.6124516492800403, 0.905674782430169, 0.3566745975132929, 0.05378345282342274, 0.4323746847801565, 0.2155593079559018, 0.9630589937835312, 0, 0.33297399019858853, 0.7420490183455986, 0.19701746672581547, 0.4205590258746953, 0.1790560992515075, 0.26429334887866923, 0.7406212072088386, 0.2795100923046342, 0.3497569137831805, 0.1602621350365369, 0.693768645302515, 0.4920124630456776, 0.3342708899030689, 0.20551916501149062, 1.160568990706443, 0.5981294991175569, 0.19074278859998794, 0.30272848832285093, 0.5867975138237269, 0.09496180843692761, 0.8625805924022356, 0.09328793442355712, 0.4492835584349041, 0.1228004848412149, 0.2512017198130674, 0.2835263655553314, 0.6196675750596746, 0.1593431497420958, 0.04652001563489291, 0.5603211010329422, 0.36098921831710606, 0.512129697555848, 0.18669732914732934, 0.5557016533090415, 0.38435982841788735, 0.021506205220963682, 0.5552956474058258, 0.41764843855733713, 0.39716144951635624, 0.442928980982186, 0.28333830616673045, 0.12388681859440126, 0.20275186166597564, 0.9790395267117666, 0.03774032798284711, 0.1409862578932487, 0.6254973002062552, 0.5384773289864307, 0.3808720231018319, 0.14260468883019753, 0.14013994275765998, 0.39222863195646557, 0.16078228619940593, 0.5306082024362224, 0.5041118107637059, 0.4494278626395605, 0.1327935318196974, 0.6035869383730494, 0.21148214411702582, 0.2960512697735675, 0.16245501708554572, 0.2480857292835595, 0.5768133149813541, 0, 0.22272683764972032, 0.43013382736615363, 0.30811852564221087, 0.2915254955229784, 0.5535027524511779, 0.6432081974114452, 0.03278982282299097, 0.2046214904517358, 0.13126638599367046, 0.07677325326384309, 0.5049550097098672, 0.045778113699440466, 0.1352690125230633, 0.6264389294530631, 0.261723302792917, 0, 0.4239375995978798, 0.33976963794156434, 0.5009954837075994, 0.8245277703174835, 0.3981745512441591, 0.4407054265756905, 0, 0.8057162332823757, 0.6222782370102622, 0.5169485121853146, 0.142943226806869, 0.46599894774325046, 0.6369407871742359, 0.07262263953753002, 0.12768241672055752, 0.296251892592703, 0.8331658836828897, 0.7910868710651479, 0.5766235191409551, 0.5476213627238044, 0.35973277910362506, 0.40265829934941155, 0.6765972064442928, 0.2880069036348472, 0.6579743126514199, 0.10487963082047552, 0.6598252008605916, 0.07430831273337617, 0.1949599556656773, 0.3444078628563428, 0.5909680371429817, 0.4836974639984018, 0.3044467142801116, 0.17381546735801867, 0.4627642243753931, 0.18583690602699257, 0.39556524126939374, 0.09343787070092029, 0.45992491311980876, 0.34578706418622085, 0.7242692253187905, 1.045685649953346, 0.38855852187345624, 0.3907658279829633, 0.021053409197832263, 0.3678615855605197, 0.4186102648877689, 0, 0.35742975901395035, 0.5769905419355976, 0.2180284506474393, 0.2574752309881486, 0.028987536873252187, 0.3011862822398599, 0.16320175274069987, 0.21233242537280592, 0.6445105007427006, 0.8524467753388709, 0.16634841084450003, 0.8308699091363928, 0.7266446461183514, 0.21617134373444852, 0.3787839478862376, 0.4225188001222442, 0.17422857345525217, 0.26236426446749117, 0.9085492847550738, 0.12718508948023854, 0.4269900349798123, 0.17425774399326507, 0.3498458019348002, 0.5304804240282687, 0.5139587396441168, 0.27993296969620596, 0.14820475555884915, 0.15737853271082383, 0.08068891125014266, 0.4051315941770116, 0.36274562668933313, 0.6960246723874556, 0.9648278646786931, 0.38350760626139324, 0.12503264746243986, 0.6475546057560598, 0.4676673031342521, 0.07856232250310231, 0.22170516152651434, 1.4544317749420483, 0.2232659396577843, 0.3649690076637758, 0.028170876966696224, 0.1342207673542524, 0.36516595510364785, 0.40906569304535273, 0.3053824047101447, 0.24925850379001635, 0.6976869195394224, 0.4094555002931151, 0.9757561708177778, 0.40598676889502067, 0.44318467903633385, 0.685869000032673, 0.059719234701622284, 0.4659711644214954, 0.030771658666753687, 0.29656681796436485, 0.7459843670479877, 0.36685878208000855, 0.32143828962598936, 0.2915387499055101, 0.2446617642526312, 0.44671869851085, 0.3321626931359309, 0.27127494546456415, 0.4598263476477358, 0.24883136463642366, 1.0548624039301093, 0.20589399546652065, 0.09252336889714546, 0.23097943691544034, 0.5146411292479178, 0.16715907339727545, 0.10520852842446113, 0.41805236597204265, 0.724157173274471, 0.01600034134644112, 0.3528856079180929, 0.13019249135900804, 0.35991865923458705, 0.1978257433299198, 0.3783672846275654, 0.7978542934812608, 0.13332312331631616, 0.3650217599090412, 0.714772383460694, 0.7822494652391718, 0.3999799575357659, 0.26214236628614573, 0.28635193895317884, 0.6150081011765411, 0.9629763937263841, 0.4272519355212164, 0.3114690538210125, 0.30417835877150956, 0.23361485118150507, 0.033901551675681416, 0.26715243676298395, 0.05426962993405762, 0.6000084397452108, 0.315258628020401, 0.4227852107451575, 0.4747323145521442, 0.29662632807401323, 0.3671923773136219, 0.4937892127184216, 0.05406722127027579, 0.2841712684353155, 0.28446413529916326, 0.8635751623704268, 0.775401410694024, 0.16467746466091643, 0.48994133059906986, 0.7574034209198374, 0.43676390499784834, 0.48370903059506903, 0.4345177766482811, 1.0839763418864217, 0.6345956985592122, 0.20740558336280215, 0.3923976286145351, 0.15406055819765996, 0.32234249346395216, 0.015504186535965254, 0.3448235040412834, 0.43426989669894306, 1.0597832594686454, 0.4878077903260607, 0.4855861482752433, 0.4716180799297881, 0.3718636166996163, 0.7458227248738752, 0.34052173007279096, 0, 0.08961215868968711, 0.36782502572712544, 0.12656272330842933, 0.30236533895166173, 0.21051679993552827, 0.26316197171835093, 0.08834469926882577, 0.17965698615540918, 0.06980953139350739, 1.0577161908365478, 0.3587171147113617, 0.19999471550847875, 0.0828876598057677, 0.6829541946470431, 0.38797204951212994, 0.08320381535693097, 0.5720917113304688, 0.492989057698434, 0.1503305935368207, 0.610554018546504, 0.13695488107604348, 0.20346444835900646, 0.30398824796328455, 0.4504566260203859, 0.34672865618975296, 0.2676004336927032, 0.054768729541766134, 0.3674794890409262, 0.25154457138316555, 0.12034385651805721, 0.7658402878763837, 0.5350756876059778, 0.3403750522153775, 0.14912783932070364, 0.5587978849401948, 0.018692133012152546, 0.3641303102144652, 0.14468645320221857, 0, 0.6274734221334732, 0.3720261274028258, 0.23044251941750005, 0.2859137475498309, 0.23600684761336574, 0.2480234505828639, 0.6225227329109998, 0.6184547241235676, 0.32139904573474903, 0.06684001812585047, 0.34122868684668195, 0.10921664064986981, 0.23937674056495703, 0.6638532876787049, 0.2458337358608225, 0.41999298836893173, 0.2750873807672076, 0.08685704409326676, 0.39764872212005736, 0.6957678893586587, 0.13218095352665082, 0.4957111004214667, 0.6213146100630537, 0.06391332574365274, 0.3444336554783614, 1.069510922403402, 0.28913682991756634, 0.5035473669310652, 0.1827802675722323, 0.2800049207438355, 0.1929863531089072, 0.5077287407392759, 0.2900804244268961, 0.3937052904500974, 0.22029654410046423, 0.14161551102448144, 0.0851228023017615, 0.49277154043253585, 0.5637163400407113, 0.1620482241808051, 0.440056069318672, 0.17621209769778762, 0.03278982282299097, 0.47128593749971975, 0.41610720332938766, 0.043012410441927364, 0.3156053085053468, 0.08685704409326658, 0.1720694796264818, 0.5809189867556642, 0.5021767116177126, 0.2828906302095531, 0.3650365820171444, 0.8039989792286998, 0.6255550409149077, 0.11344463405778521, 0.09438356189500638, 0.3262932017717485, 0.084450556127956, 0.3800439593727507, 0.6690373376575701, 0.6699566778353513, 0.6953119741289271, 0.23813347755787792, 0, 0.2999297519918104, 0.06062462181643484, 0.6613253317832575, 0, 0.32329766313891606, 0.29209137340876956, 0.2302643050384077, 0.2881060652400518, 0.2807498934931147, 0.5933670681819858, 0.7395423949957345, 0.41371812794373336, 0.3321103646537815, 0.2531553012818537, 0.14773829536387478, 0.321498156096841, 0.3166178053516516, 0.21195529724381312, 0.40494309583882826, 0.3427592181155703, 0.11194715158507884, 0, 0.6641089203665548, 0.19407133823760403, 0.3485550011322539, 0.02061928720273561, 0.387441961880159, 0.9088649866039437, 0.398273437819482, 0, 0.10103517835516089, 0.4970728096441842, 0.2582452650958856, 0.20448948156451652, 0.14294883330367777, 0.41576430791714525, 0.5345015466352482, 0.22588067154583422, 0.5460284606917333, 0.22892265813875062, 0.42771283012705325, 0.05989814158106883, 0.19569620145928662, 0.42025569843321764, 0.3342950716861672, 0.37813211689716175, 0.20184556475429324, 0.24257435585314724, 0, 0.5142855649775453, 0.24929220715030143, 0.026668247082161273, 0.13036181786324363, 0, 0.027398974188114347, 0.26800784305410646, 0, 0.17443419196078883, 0.20054404765321743, 0.10201267188814193, 0.37626884215136297, 0.018692133012152546, 0.27601445654578904, 0.02061928720273561, 0.1870080127278976, 0.6104379668767412, 1.2417304744538835, 0.27567292007062955, 0.20657906607964702, 0.287924211819608, 0.24681240691129125, 0.057975073746504374, 0.29096190173807157, 0.4667518602548592, 0.5276140467109411, 0.09631981090767865, 0.021506205220963682, 0.2040472467290668, 0.44224648088979984, 0.6141727796807703, 0.12671050675150788, 0.47744901628599323, 0.35147552960356304, 0.4476782536029476, 0.18294422182029918, 0.33990209767683893, 0.40513946764874365, 0.38663564548903084, 0.6479746557630476, 0.7179347968500924, 0.14268058727811056, 0.17299597330593078, 0.17678165094567524, 0.42104533244313225, 0.5022129543464304, 0.292934724781136, 0.17195670418683873, 0.3688754288091447, 0.17846201273748877, 0.2663575543382126, 0.17395330712343796, 0.7767622529852628, 0.054386149673553, 0.5474949334700724, 0.4622293345667256, 0, 0.24345797206816405, 0.3846483776107673, 0.11805985968290451, 0.3707518601350463, 0.38344637752368566, 0.24857920183743562, 0.4244695672036423, 0.17086812410869617, 0.48773750846158775, 0.31452548180244877, 0.33326526747522206, 0.306582872749963, 0.7251537652716835, 0.545203789726388, 0.28167481817772555, 0.13109733857546713, 0.41679984886300875, 0.14310084364067333, 0.2524948176363108, 0.3085080214879474, 0.2963517994387061, 0.4430601778831735, 0.11884940997306127, 0.1317148616981248, 0.3246269609123741, 0.8658091882840899, 0.3427047756929912, 0.4272431630388205, 0.0848200275568214, 0.4961514462846433, 0.26653193837864225, 1.0427608762219631, 0.30417743297338645, 0.800438927043438, 0.39861813447498057, 0.42922332344675873, 0.07343929944408595, 0.7245753118987582, 0.28848915459745594, 0.3162067838425274, 0.37205694751744395, 0.3534864926350054, 0.47751544916924177, 1.0229738565762654, 0.23819481551739352, 0.5863566084255041, 0.3450422859083038, 0.11371765415550272, 0.3991920863217586, 0.3433382386927378, 0.2937297354501602, 0.17328988476360668, 0.4285891081060929, 0.15727488356700106, 0, 0.47626586498001033, 0.26994529451507077, 0.46361138127285945, 0.22884157242884767, 0.445246919287838, 0.08766093855842527, 0.5377364960543003, 0.28637996569594076, 0.10303222589823513, 0.6509621856787008, 0.9739883900924184, 0.1742786067159684, 0.20091694434736515, 0.1260940272201729, 0.010471299867295437, 0.3967411312715293, 0.2883912396724339, 0.3106507360741287, 0.357494254101395, 0.2932633870248846, 2.0659147394688366, 0.04879016416943204, 0.3798368737815448, 0.5056994397888369, 0.20143571721395087, 0.19658952859165937, 0.5289549898068114, 0.10536051565782631, 0.11155248590574732, 0.08004270767353636, 0.47688073428428507, 0.2781700835433037, 0.6682991772413454, 0.5555031186721652, 0.13602423390723536, 0.2592802797641232, 0.12284044567959856, 0.27055812287789105, 0.31199506761436285, 0.5408405920832953, 0.23544898460477498, 0.12479771265576775, 0.3277764140854934, 0.5610580645998042, 0.08376418880726413, 0.4523090747288844, 0.47304430898850947, 0.09952959534703315, 0.13898843208758066, 0.17071701950139387, 0.6181830242055754, 0.01980262729617973, 0.3667346688084656, 0.3705639401310374, 0.7650868805230802, 0, 0.4512379431548721, 0.24106355890988884, 0.27028447331028393, 0.052277863887717366, 0.5526784162546255, 0.20469615940549177, 0.16982728414948767, 0, 0.32220579475674965, 0.31292671164819236, 0.9959410280488599, 0.45808539218985933, 0.8192631260017124, 0.27605492983665775, 0.12525838964527608, 0.0840831172105416, 0.7557917347616878, 0.08260352853521001, 0.6174128620942863, 0.04678735957446143, 0.42202907021098696, 0.4980703272341151, 0.3219567463096642, 0.5112803037458638, 0.26685520292146203, 0.5572681998878372, 0.2788296880734077, 0.2188257956593136, 0.25693311931182145, 0.30444687015547145, 0.8712782554705427, 0.5700900627335636, 0.4898444186879858, 0.24642958489040148, 0.17969579333697827, 0.40902987684464776, 0.33237161080831473, 0.11023583002100075, 0.30974197776831, 0.18335154883733756, 0.13439999923377333, 0.08338532917022642, 0.29841421676650115, 0.8600721909649213, 0.24709980902778805, 0.1342202348167044, 0.4115798611443783, 0.248267874844725, 0.1558176041212499, 0, 0.6433610409314708, 0.18180284761285195, 0.06713930283762848, 0.03922071315328133, 0.024692612590371414, 0.8780266034862261, 0.5375711917212553, 0.43611440066106266, 0.26091077583950784, 0.5679473630318853, 0.36671621671675286, 0.22450001672210224, 0.29359029628143507, 0.19861374967857354, 0.23271744166627167, 1.1898648849833973, 0.8240640049381542, 0.07218876238397053, 0, 0, 0.9611564816485537, 0.1289478870826144, 0.04585330449612225, 0.556313569865562, 0.057975073746504374, 0.9646959137995038, 0.2847038316578157, 0.3106383650315264, 0.25277208935563056, 0.18508328541020408, 0.3552187846246595, 0.24046861528878172, 0.3699049010253207, 0.5160365415748928, 0.4816040763839694, 0.11419628644443117, 0.09424149356271326, 0, 0.09489874182279573, 0.2096009739262378, 0.20203031497819063, 0, 0.15941815181952645, 0.11481852646832087, 0.1834154754316702, 0.6108969553511907, 0.6781211534221212, 0.5042256349513052, 0.3441573107214041, 0.2349012779041694, 0.35282135116735475, 0.24828150245868988, 0.07361181734324618, 0.11712535723741269, 0.11026823915389695, 0.30109211732053365, 0.2625649515604463, 0.35571729823218645, 0.29122851623480206, 0.18768396693153083, 0.31399938076915435, 0.48114468115694586, 0.3308914774993941, 0.06062462181643484, 0.32347572428081434, 0.1040841912982216, 0.07185073675894814, 0.14773410086950034, 0.17147685747661415, 0.6573219160390269, 0.22963025625581987, 0.2983895753879312, 0.21001973226283918, 0.24062120701725825, 0.10046609363252924, 0.36285560855588384, 0.40163999022545094, 0.5353734117669239, 0.056747145070014424, 0.1987737304065009, 0.566900162467619, 0.6413441674820062, 0.4334299749541759, 0.21720800478050212, 0.38208152589084016, 0.3141413112773551, 0.1183165179671528, 0, 0.8032470469313706, 0.19246386232735446, 0.18977236294982003, 0.26723604843289256, 0.08768500449910396, 0.2923801604077567, 0.5036132277910352, 0.21662887029301592, 0.2292692634869036, 0.3148267654966791, 0.24427258820002345, 0.4388896049771632, 0.15431720749375769, 0.044451762570833796, 0.8732733074993012, 0.17762233489172904, 0.19159739925647626, 0.1933713929805398, 0.05430534485754197, 0.3475988760547942, 0.3664067947048362, 0.21572695365916042, 0, 0.386019749113228, 0.2982972036587706, 0.1492737027477541, 0.36891929139320706, 0.4949805257075372, 0.15256211714212037, 0.1208260017545368, 0.11447462266050941, 0.2145541819020228, 0.07558835732415611, 0.3073946592957956, 0.2432594252116264, 0.22361232274365778, 0.09740034597664957, 0.5481690823612395, 0.4087281229250655, 0.2581197415171289, 0.6422024983767826, 0.48144817812204027, 0.41758563141258925, 0.14581090225458435, 0.6557530206899971, 0.23751063088913016, 0.08185102643032004, 0.4917516195328283, 0.45416254118837596, 0.263574498532925, 0.02985296314968113, 0.17297081277362678, 0.39981618209782976, 0.3821475856967127, 1.41033172827485, 0.5310644980061582, 0.09726477030364758, 0.3285042565783866, 0.20972073863945595, 0.3441694910227261, 0.48130154836644756, 0.10015832847054269, 0.12000279239469636, 0.03174869831458027, 0.5173492646435788, 0.3465797136992662, 0.09951768767909028, 0.2499496171758905, 0.666972520257531, 0.25264113194631477, 0.3834674530722533, 0.07709791311755869, 0.12693433736278184, 0.10511855090284494, 0.16175196026677388, 0.024097551579060524, 0.1604975294509147, 0.17052418400366495, 0.148176633124615, 0.14731480098250682, 0.5261770178523455, 0.6620752767742306, 0.29559985030473157, 0.1277097444482362, 0.6427594907116996, 0.34141219429132225, 0.10112616928220144, 0.30425267580907733, 0.4034041622881417, 0.6408094319508019, 0.20679153543894357, 0.5144775375658522, 0.18759861389479845, 0.24910362332544222, 0.16592774701827837, 0.11205236094739865, 0.050055670467200594, 0.6475656630564455, 0.36434904056251166, 0.3127272217463699, 0.1779312084926617, 0.3288482852826799, 0.36798487593179957, 0.16414973231450816, 0.41967174990022316, 0.6624424827798308, 0.10105859271518897, 0.5034939537838021, 0.37198303665425364, 0.21503625660297515, 0.05129329438755048, 0.600291128057256, 0.20210634655153148, 0.09739748791233616, 0.2444521266613842, 0.18332220482590644, 1.1828464939297398, 0.9272980634163126, 1.40059960755281, 1.6740328081338152, 0.43036353694623314, 0.37192396521385396, 0.0956531741482809, 0.5072209680144757, 0.306221762770402, 0, 0.06062462181643484, 0.6532506884957634, 0.09571594162770503, 0.3037509905972452, 0.4195647138188264, 0.3536417248130552, 0.07017952181161274, 0.09439148428725229, 0.35953668059807165, 0.3583746596304532, 0.21463688375988133, 0.3619264170978554, 0.2531993903369408, 0.4730855402445944, 0.5563861894253619, 0.321184798031363, 0.11535911402731006, 0.25683794943601235, 0.26871184715092444, 0.25739180697418135, 0.250808892552114, 0.04927104900678264, 0.25669698137271213, 0.39176230110852844, 0.15128703767829907, 0.6989468355706547, 0.6486973971773955, 0.48927433536651127, 0.1727379965298971, 0.08701137698962969, 0.664765319901297, 0.07164187965852853, 0.25908674928886766, 0.23918690173666993, 0.1840169513791994, 0.5225031302601503, 0.3886357535747823, 0.36777516117335624, 0.08069226924233214, 0.4274124330740959, 0.2455516053863181, 0.21883799941760015, 0.43248365092393504, 0.05139094586948929, 0.7259637858107973, 0.4449147895032057, 0.05970592629936226, 0.26314888592790897, 0.570091527005823, 0.7325141105767637, 0.2105233940250299, 0.8720849767352576, 0, 0.1132753044238602, 0.27192977046860706, 0.44047937468486864, 0.11159106540846228, 0.2110582513981043, 0.7331896872282805, 0.24142004347239343, 0.11512369566152666, 0, 0.30493107071358777, 0.025975486403260736, 0.07157952580036303, 0.242027545232137, 0.3139178321862952, 0.41313357360961156, 0.3156319229820907, 0.7112394386533433, 0.7092014885686494, 0.3883803871719555, 0.3703522458753869, 0.350082274209087, 0.2288782591859112, 0.37731165331951744, 0, 0.12839105026857695, 0.42717283997050415, 0.06914437516120521, 0.08405934532554851, 0.6433134443465722, 0.5454367683083535, 0, 0.26085724304617036, 0, 0.0565212102318424, 0.1572367304932442, 0.5150600028806283, 0.2723391998743536, 0.06713930283762852, 0.2497896303607993, 0.03774032798284711, 0.34304143998118375, 0.40270823490970925, 0.3849595650592866, 0.32743890651932706, 0.29500251801318744, 0.5965153023292754, 0.3307864102767115, 0, 0.5897165773959788, 0.19575824805661277, 0.21272047791843401, 0.3162765740465426, 0.6661144587196962, 0.16490431396095864, 0.25713081360234563, 0, 0.4506283393140109, 0.2919407561274533, 0.5085657955396565, 0.8332669159683065, 0.15747425792675593, 0.2723039074437468, 0.09352605801082342, 0.33509301042986456, 0.19312162856333018, 0.36456543110708556, 0.18230703946445057, 0.05325490353554395, 0.213664819058528, 0.15224260359876784, 0.5459336880767875, 0.7658198608411168, 0.7212719413578826, 0.21719994521176056, 0.6084329661270229, 0.4318512359978518, 0.05862181722146423, 0.024097551579060524, 0.31774424766107423, 0.27995764040919957, 0.41893678112907906, 0.19380551497563597, 0.16321798580027294, 0.2375805152378288, 0.06725222700916732, 0.23952590538735236, 0.1440665607606321, 0.09531017980432485, 0, 0.24252336625146506, 0.14797782072017737, 0.08061876181090293, 0.13220414761679808, 0.3546190388697508, 0.5124602290616782, 0.1913875326445572, 0.11938431702335751, 0.552187122097052, 0.07229429409600523, 0.0741079721537219, 0.38889392663247563, 0.8364500866355038, 0.865209038316043, 0.25789855615792084, 0.12351371036536844, 0.23978937300491582, 0.6032026116856409, 0.4570338508789285, 0, 0.23142715193849123, 0.15057285847937435, 0.014184634991956381, 0.3920020038826246, 0.14459287845957955, 0.26727551373680786, 0.24483616983442807, 0, 0.24422791642587682, 0.12948280409852247, 0.0408219945202552, 0.32687944289672455, 0.33691728860427167, 0.10521570879372796, 0.5148693207197269, 0.11436980981541398, 0.40518288232722705, 0.09187716532053922, 0.24893374243338423, 0.05406722127027579, 0.3018561632328759, 0.5499021452033822, 0.11163921293248547, 0.13202658787898833, 0.06391332574365274, 0.07193731845192178, 0.5416182854008599, 0.2083927415560616, 0.15478842333371792, 0.4212447285614843, 0.06820825002653352, 0.5638812196107479, 0.20820393751991503, 0.3455036417660372, 0.2969404768125894, 0.05756417566332877, 0.1677865386871034, 0.16118896521076798, 0.0890336223703976, 0.17552510849647493, 0.07145896398214499, 0.026668247082161273, 0.47990263385116766, 0.2537231506384775, 0.2411620568168881, 0.04515096077689279, 0.04760168150563382, 0.11856765711680152, 0.07265764557607525, 0.3240060388464262, 0.21526708545226966, 0.47984232477172006, 0.4263569073662129, 0.20610743615860477, 0.08050879950206276, 0.33226412032332486, 0.459203890698834, 0.133311745546894, 0.2668569989624524, 0, 0, 0.20834784717009666, 0.3861356465213433, 0.3306650859325058, 0.5687041641900294, 0.40752325295745795, 0.3714782997051834, 0.1758032424132523, 0.27743876297082537, 0.11367962407179101, 0.04879016416943204, 0.039730489954911914, 0.99434440539951, 0.17923036422428185, 0.3101665303703562, 0.127261779610927, 0.31440527971190707, 0.41558112777524636, 0.04652001563489291, 0.138646493291293, 0.1317164598479097, 0, 0.4454052803357319, 0.2669832103237856, 0.012578782206860185, 0.17846922409854396, 0.10965543994070696, 0.09846119951218427, 0.31883166233725047, 0.5056610155903158, 0.3612799865290373, 0.2113852152985919, 0.25309554130599354, 0.37735942468324984, 0.47386786981448514, 0.6778572007775947, 0, 0.6354145920555244, 0.3541037508154169, 0.27034485366379596, 0.05333649416432255, 0.20306599377966292, 0.25660965536468106, 0, 0.13653793614796023, 0.1219446436733615, 0.05129329438755048, 0.21177268123621418, 0.3079205573045897, 0.34480433806968136, 0.33447884997437505, 0.09647265049107681, 0.2285343999490862, 0.33378135514470414, 0.12068690861167147, 0.13387195008514152, 0.42313896959307185, 0.23567459886530595, 0.1250037168875064, 0.127352486672534, 0.448494410801635, 0.4718235025169377, 0.2249846710273027, 0.06757342483272517, 0.20637884734705886, 0.07954308238210399, 0.18829502297265127, 0.31218000883078023, 0.02985296314968113, 0.6662118998956539, 0.20548140271499746, 0.46556370513984024, 0.25982956275349617, 0.42313049365415883, 0, 0.47999238455441623, 0.49848153304497717, 0.3512353168710086, 0.32114355461671934, 0.20951291547386677, 0.21343629756245655, 0.15785744643914576, 0.05406722127027579, 0.23899874665634413, 0.307551311457609, 0.6393856833194476, 0.3365508578717014, 0.4819843574974786, 0.13385046082880292, 0.15304622646056104, 0.2988609216850807, 0.05004094607108445, 0.30540412615135615, 0.20927833698211706, 0.09211528890780568, 0.35677987835208114, 0.0609606997896872, 0.12022432104316379, 0.3492945841380749, 3.151326063834452, 0.10411761458145329, 0.10262239542570184, 0.3541904359546307, 0.14800342523305693, 0.07327220110632565, 0.014184634991956381, 0.3498762296293346, 0.11028663399388206, 0.23100722089345524, 0.5164982782714187, 0.3193654010271895, 0.2709119704364573, 0.06838772774685052, 0.11496821243802047, 0.028170876966696224, 0.2566011806256094, 0.21317424288853698, 0.40720131882461535, 0.2599424427583529, 0.2644243913031279, 0.20672877824829317, 0.01092907053219023, 0, 0.0789272585896707, 0.4810026907504978, 0.07257069283483547, 0.21836990751868487, 0.4509814227008362, 0.3626116410294123, 0.18571648177443537, 0.3752003030886543, 0.4963795410963471, 0.2168641899519787, 0.1740656038244734, 0.11569891632896617, 0.21438757142402043, 0.502191915917402, 0.4392349357954336, 0, 0.021978906718775167, 0.16293179373889866, 0.13480771698412736, 0.5356136745188117, 0.053950514728741655, 0.15907112200516496, 0.19927343766709524, 0.3061899588942989, 0.1745410238608955, 0.41812797545036257, 0.3428617182266791, 0.25737317352799877, 0.7539488255884021, 0, 0.16886895287323428, 0.15552141051613755, 0.07930983845788388, 0.15139556523083778, 0.02985296314968113, 0.014598799421152631, 0.3770986522236808, 0.09391878278944071, 0.3924218893209052, 0.29074207803579477, 0.7825856914585876, 0.0408219945202552, 0.2629763680287488, 0.2833627819940228, 0.04430025889657999, 0.06431350529669068, 0.15129761084485183, 0.4817591444098786, 0.1176646714495814, 0.26053248502495036, 0.5739876386375801, 0.024692612590371414, 0.3539347237032878, 0.1195206555549243, 0.06007529666016399, 0.6457635150783896, 0.07029636939039573, 0.250421411617205, 0.17556912707217337, 0.10886516964650432, 0.19104222854802144, 0.4132685939782096, 3.152293116126424, 0.1912922903080946, 0.22650385902524814, 0.06453852113757116, 0.1386546101414683, 0.37045260019849474, 0.4267660045312875, 0.38392386936071615, 0.18477188582744564, 0.3018425515645714, 0.1971093742668939, 0.060569798757842686, 0.20545215320380394, 0.3594249345237188, 0.14894565980037103, 0.15485385218682052, 0.5289659667201774, 0.22127311775814335, 0.2129439795150605, 0.1365110339616178, 0.02985296314968113, 0.4990209959756081, 0.05978393240164161, 0.021978906718775167, 0.2654410056158192, 0.5835799611097604, 0.03774032798284711, 0.15550657911212262, 0.4842899307162445, 0.10210980170221497, 0.133429398642079, 0.2896148453396049, 0.05556985115481057, 0.024097551579060524, 0.023530497410194036, 0.42340280879483155, 0.11531896870122424, 0.2096183024314622, 0.09806121317621457, 0.4463426378618149, 0.13156480883476587, 0.12665903561745795, 0.4475279906736862, 0, 0.23684824400498486, 0.356248523849087, 0.1006902864352803, 0.3878507820921721, 0.2721024859004566, 0.6393266807449506, 0.12201215375100978, 0.3808191128806593, 0.1951587035546352, 0.17117285990625347, 0.021506205220963682, 0.1707597587331498, 0.417047926563514, 0.024097551579060524, 0.05735129647146571, 0.17147578141819916, 0.111126489646379, 0.4214177587826337, 0.06504950735206223, 0.26952792406946796, 0, 0.08004270767353636, 0.0425596144187959, 0.09496835789021216, 0.2659404329959837, 0.16144998230886995, 0.11748000055400629, 0.1639201148188695, 0.46610043380860605, 0.13410992894381232, 0.7926040380114152, 0.042222630422346585, 0, 0.12124924363286968, 0.0878280368961855, 0.6484656955601592, 0.23101271835870932, 0, 0.20301046006410706, 0.20125937296191487, 0.2085909993609549, 0.25228866072350886, 0.11966052590166895, 0.1613664274134622, 0.46849273780334805, 0, 0.4430443794025267, 0.8593694913756, 0.5241775692687194, 0.33945541343619634, 0.12207795993926412, 0.04090990274412594, 0.32327238179225176, 0.14578279087788806, 0.05129329438755048, 0.3485260546831229, 0.2077950953464155, 0.16535994208034277, 0.02985296314968113, 0.37847937883463206, 0.302808258605245, 0.17175260544472026, 0.15901914560244013, 0.08467497037986335, 0.1897586620111687, 0.12827417457231194, 0.23927590982307378, 0.4708599032391672, 0.2756467553885901, 0.23132818036049882, 0.05756417566332877, 0.07880739051070029, 0.1838959416184774, 0.129923182175635, 0.17134647857847318, 0.09432359261100672, 0, 0.1763238802807803, 0.09884583463663255, 0.13613217432457994, 0.20855870533474305, 0.15361174152264778, 0.3596545852758276, 0.07999142430728504, 0.05406722127027579, 0.2840719097537117, 0.1928680474579318, 0.12042061061791504, 0.2298106189066486, 0.14815266117870465, 0.24484263916669038, 0.4579390196355409, 0.3536969934840889, 0.25146267423845814, 0.39970969653754385, 0.12845094139210367, 0.2676078434020742, 0.10490085120477757, 0.29008819566297456, 1.324612243478005, 0.07386290414612345, 0.17421026485015617, 0.096086878872497, 0.2722116127803393, 0.2006233951286751, 0.36952215136111183, 0.1485359538690554, 0.20060313180668507, 0.09134977858822799, 0.9910314375178024, 0.0978133100224434, 0.28621396810011396, 0.6291519902455972, 0.19993065020983483, 0.4270805564116228, 0.18790635354948115, 0.06177735969624316, 0.2199879028112946, 0.21613655224562026, 0.01904819497069441, 0.27166564229760487, 0, 0.10746356585460501, 0.4633453464802893, 0.4008926673835905, 0.07869226857566483, 0.2286964324410747, 0.298641666948853, 1.557940452839114, 0.027398974188114347, 0.25577100206339726, 0.3814772644108815, 0.028987536873252187, 0.20393839120668109, 0.047830345588699905, 0.19764894974829975, 0, 0.06749024160241647, 1.0065654861877025, 0.060569798757842686, 0.10578147967162037, 0.30069285860679873, 0.44366526399897155, 0.03929141201152404, 0.24449285279856786, 0.18682892990444466, 0.13591369442881293, 0.25720136519819403, 0.11246387417878373, 0.4766854086046896, 0.08247622182423842, 0.3144869908436561, 0.021506205220963682, 0.46499242170650207, 0.26517952108501647, 0.138646493291293, 0.23890587806740182, 0.19311336175733032, 0.3458868390600615, 0.29693939215029747, 0.5398458668733267, 0.43958285413643605, 0, 0.4458994138463679, 0.14173627667887626, 0, 0.18626875769192366, 0.08004270767353652, 0.10375923429085249, 0.37695286455450483, 0.32562842040919143, 0.3389517584552062, 0.14940603476898795, 0.15113905373507236, 0, 1.0993425294153512, 0.7156360238902082, 0.3622376882985783, 0.25738876825480395, 0.09224973365121872, 0.027398974188114347, 0.5658137867874355, 0.028987536873252187, 0.07244636078160174, 0.09067696688283672, 0.17940865474578377, 0.6309465377756444, 0.2280338629996646, 0.6098981123420175, 0.5068928614935293, 0.11868488450752798, 0.48273150308905155, 0.2997658424346445, 0, 0.07053015080583808, 0.03174869831458027, 0, 0.23992277018020844, 0.28125987163255844, 0.3961671597488908, 0.40756970798099734, 0.71728566147046, 0.10536051565782628, 0.2186684979747617, 0.509721214700891, 0.4231727328485257, 0.10278219464302599, 0.1166968346305769, 0.43340477785829784, 0.5327227161508735, 0, 0.21871870849549402, 0.15225069918051487, 0.08028381677657304, 0.06062462181643484, 0.20286429601687825, 0.31897099421581304, 0.33275953790272383, 0.375097494890241, 0.024097551579060524, 0.0425596144187959, 0.4014490324305481, 0.14153139428303696, 0.25599051536175266, 0.1267298216461894, 0.06062462181643484, 0.26529435679511776, 0.32072015942189847, 0.6099202867516305, 0.056887374402051494, 0.4563373875134627, 0.19311910569819052, 0.503530103654374, 0.3207604519944046, 0.3976561509053336, 0, 0.4621077531082475, 0.16967043924039577, 0.18663987151581535, 0.06684001812585047, 0.1209129286653111, 0.04652001563489291, 0.3823716101883181, 0, 0.08916423035073096, 0.4316898000721437, 0.021978906718775167, 0.23293774497029007, 0.2619920534523745, 0.13447730643042674, 0.19174669725353746, 0.34076384121716496, 0.17920916393815375, 0.2219480939094362, 0.10970123981579197, 0.2502035711731745, 0.07646116609447731, 0.11869420132111565, 0.20773352230385422, 0.21669154437075677, 0.2999150320012309, 0.22280106749232648, 0.16764706538211782, 0.347865847942173, 0.10303222589823531, 0.06062462181643484, 0.34577314034700146, 0.14135817362285508, 0.3293623286469711, 0.13073558174884317, 0.4369268453386353, 0.16609036186537202, 0.16103046910948463, 0.2564455196704993, 1.4405416638252113, 0.27725244554835565, 0.03509131981127019, 0.1609011234387797, 0.05978393240164161, 0.15244074714145947, 0.4898893554858825, 0.19658481573307698, 0.1345996947740014, 0.21030872711289655, 0.16053078339427515, 0.08701137698962969, 0.3724235235151756, 0.24075347743191455, 0.028987536873252187, 0, 0.06947970948383964, 0.43537046582556194, 0.10008345855698243, 0.5236365402498301, 0.14932526732337967, 0.23764725441368956, 0.10671095475569763, 0.06519619955654207, 0.18622049838612817, 0, 0.1589617829236682, 0.3407766849167318, 0.38257690226164115, 0, 0.2521187007010897, 0, 0.21272702190297904, 0.25666823919285764, 0.474705108106931, 0.917605374521725, 0.19556327817920027, 0.044451762570833796, 0.5978233593941593, 0.9308673866172865, 0, 0.33061072080656545, 0.2297818188151998, 0.22813647714449586, 0.7230029715567107, 0.2676714229454751, 0.16775654058592634, 0.10536051565782621, 0.48441709359385854, 0.3663270192369873, 0.23792178500522612, 0.14487704304192905, 0.2592962081819442, 0.2826351584531965, 0, 0.15600424847658118, 0.3434387623989338, 0.05876530922625171, 0.05430215607694772, 0.40788340836698606, 0.2909283069593754, 0.05170137437689026, 1.3774750250428247, 0.22667008929584226, 0, 0.2927733873503848, 0.12272928567260069, 0.5382226036319399, 0.3698548407515515, 0, 0.23288285935470254, 0.21974612997393483, 0, 0.25813249941186556, 0.02985296314968113, 0.39516639016036803, 0.14386374645075525, 0.21800064262243837, 0.4259808914419831, 0.24776355428486085, 0.033901551675681416, 0.4642618960451849, 0.184719462620962, 0.06453852113757116, 0.11079762020837311, 0.030771658666753687, 0.24100772392052494, 0.46497665015966233, 0.1818113442572516, 0.27351085456059054, 0, 0.30903195615071805, 0.1244741055120129, 0.26769242439575164, 0.01980262729617973, 0.35997951577902026, 0.27304936888096365, 0.43061955047629125, 0.3468121102140166, 0.5079802408773288, 0.39406501527207605, 0.20879369317025356, 0.06298693044615217, 0.2465971670452702, 0.23038178473746845, 0.5962642600858641, 0.021978906718775167, 0.4698497916398792, 0.13550697893066146, 0.2958858442574403, 0.1219251476802306, 0.08260352853521001, 0, 0.16809754862804246, 0, 0.3358661236488965, 0.19046918274158106, 0.3200076443693211, 0.12556890477738616, 0.35335000050192505, 0.3751304063606283, 0.025975486403260736, 0.04879016416943204, 0.09021560479976432, 0.22261988481307368, 0.25989377535329766, 0.19754953747947762, 0.502162811867074, 0.10966446835275714, 0.07042724897409453, 0.30411642324986754, 0.11544004064576027, 0.19062066249705414, 0.033901551675681416, 0.159977713988611, 0.49862534611300874, 0.2875669333071375, 0.200491538832486, 0.07875983386064735, 0.24293825621814377, 0.25988554860990054, 0.26045649577041613, 0.08923113372794247, 0.106937260280628, 0.13830467137718017, 0.22356451532800387, 0.10359943507352838, 0.4303258984874775, 0.31289904719522027, 0.31154408343517087, 0.09118078646231392, 0.0777004127040069, 0.10870973945070854, 0.11957034623047937, 0.25746635678105373, 0.10119890764084821, 0.5063155827067741, 0.07191258159028613, 0.22963386616052614, 0.5967973815052803, 0.05945806990515225, 0.33920053115228793, 0.2223670620832257, 0.04652001563489291, 0.37955457118322206, 0, 0.2743493202996275, 0.23156919144070898, 0.4073062276097637, 0.09945669576617341, 0.4712789176590303, 0.06062462181643484, 0.20694044826041913, 0.09531017980432493, 0.10513191810282449, 0.2467260139606406, 0.06749024160241647, 0.6926667173324932, 0.29421635882548425, 0.07817031358342813, 0, 2.3130316135559283, 0.06252035698133396, 0.03636764417087479, 0.3593531724626111, 0.09452617349211626, 0.09806121317621465, 0.389350527037502, 0.10300943840227891, 0.47977315860089975, 0.29190430472266227, 0.21252447075011172, 0.07548065596569423, 0.18938273568829977, 0, 0, 0.2755632890762909, 0.1280676996161202, 0.21905185841095323, 0.17374086329890712, 0.03287676800410893, 0.3829758130708856, 0.30660682432380854, 0.2300439440188521, 0.2025876128619965, 0.08408311721054146, 0.021978906718775167, 0.16280042140674142, 0.08988881214691348, 0, 0.2401910640139691, 0, 0.17380857227243568, 0.061607809389490314, 0.08988926818749889, 0.4566628697503246, 0.2535064867587869, 0.11040973036263127, 0.1074808518199994, 0.5807729176564572, 0.10923799240424671, 0.33233390243414684, 0, 0.5724583864353234, 0.06834204165597307, 0.15215067916059116, 0.06622060732055592, 0.044451762570833796, 0, 0.07251380192314302, 0.17580119397072058, 0.4425735964643335, 0.45406415754263446, 0.15917415761618176, 0.2830774044898684, 0.033901551675681416, 0.09670245291471179, 0.19491850018129178, 0.08842239055409844, 0.028987536873252187, 0.0536801494636236, 0.20794546119959056, 0.3630641286636522, 0.1940592605344431, 0.03278982282299097, 0.22978094785726444, 0, 0.11536774713407155, 0.19081057211886596, 0.6982238247985043, 0.3234754529233168, 0.22924331516651716, 0.0667278648560993, 0.7076299473433979, 0.556703924958835, 0.17182546962766998, 0.17737105163409844, 0.047481691624224415, 0.030771658666753687, 0.5816849716057049, 0.06665716599785643, 0.16627354442727882, 0.20470622533610996, 0.12749375527296455, 0.20456839707760455, 0.3371899858888779, 0.03706670047337263, 0.1935112491617522, 0.22061758972407575, 0.06669137449867214, 0.4716663179511701, 0.11256411978059164, 0.10287963015380841, 0, 0.04677200001319481, 0.077142875848876, 0.6096017286901598, 0.31338651392533345, 0.09923893991699674, 0.1303644693598146, 0.21109321479404114, 0.08651742785634642, 0.13430809169269464, 0.10991511160901489, 0.397675402224082, 0, 0.24553749613841142, 0.5505778227683671, 0.45281992354760364, 0.07398550325968023, 0.22301433509142643, 0.04607645829783569, 0.26668392561200743, 0.5095451191474245, 0.3445101465280399, 0.12981857916772765, 0.09834508349947886, 0.05479794837622869, 0, 0.327070684271533, 0.028170876966696224, 0.18497056496553177, 0.3979508183234122, 0.28654712737478455, 0.123211596431481, 0.15280564367008623, 0.23402965821510713, 0.3751813495443137, 0.35459557970183325, 0.21560753442742137, 0.15266147627579066, 0.22628235907616967, 0.2229812946459378, 0.3276814978136376, 0.4817956662223084, 0.33138625128512184, 0.2394775355857684, 0.2793701116557834, 0.09237332013101511, 0.24324682259767788, 0.20815099772032225, 0.20775163705167993, 0.09162573639352242, 0.1019830348895967, 0.23220574479481704, 0.22503743342105348, 0.2642831703212456, 0.09263427509029704, 0.06757342483272508, 0.12468223521415776, 0, 0.2226048627598178, 0.5576492041456658, 0, 0.25028562389789205, 0, 0.06762138054431302, 0.0408219945202552, 0.1592696216617438, 0.0871049004874766, 0.2957618554723551, 0.10396093530340296, 0.08483887993702931, 0.12468237499883042, 0.07185073675894814, 0.22733031326257072, 0.024692612590371414, 0.07159365318700889, 0.34375657496116296, 0, 0.2555253562866808, 0.18703557453145395, 0.16403920689388454, 0.22567289785791056, 0.20709406871121683, 0.11467912440487465, 0, 0.1486968291447943, 0.6158737377763694, 0.2987546980417222, 0.10625930317983379, 0.30658669512704023, 0.2106522117433391, 0.23507212217948348, 0.28650339957411514, 0.31777651240454396, 0.11433929681292673, 0.34990087097268296, 0.27844764376514497, 0.21066858182986017, 0.2727047965751966, 0.07731197990091607, 0.21504191817543575, 0, 0.3414636656534346, 0.16407052018919377, 0.16257087620046953, 0.06905101271200328, 0, 0, 0.4278836251154787, 0.17501413746216768, 0.07953936215092863, 0.22664859149757902, 0.27555691056003484, 0.4316898201668038, 0.3444811759259563, 0.388534323429092, 0.1925300351649843, 0.17561574792676327, 0.4015454648461677, 0.20232020330064443, 0.0425596144187959, 0.21608616044939385, 0.06453852113757125, 0.4190835702581468, 0.2377765136456588, 0.353286157539733, 0.2577869427865235, 0.03774032798284711, 0.19979509435672269, 0.23461480209608523, 0.2096371905839121, 0.16092117390404698, 0.24958655883355896, 0.23598069807489044, 0.4740974289599242, 0.12745234654814022, 0.05109166115771131, 0.163474234385995, 0.547942798539316, 0.0569869313736104, 0, 0.440085815566911, 0.06177735969624316, 0, 0.2031738256802696, 0.1661834675426726, 0.24746085896491588, 0.2520843773651092, 0.415787771886169, 0.17681890492716232, 0.24402873124469282, 0.028987536873252187, 0.2066158232354574, 0.1648550759204055, 0.11627308994154521, 0.21009378911327165, 0.1344448064375929, 0.39530357484049944, 0.2959251723393248, 0.41874707937046574, 0, 0.06757342483272517, 0.16762852321799132, 0.21340972123958088, 0.3374296369709479, 0.3363709802233152, 0.18057259115253904, 0.2504356936022171, 0.2726029062620106, 0.12584690366440662, 0.15981718348435286, 0.18407793816487408, 0.13374003805740492, 0.300464135206576, 0.2899528361345292, 0.0789272585896707, 0.15286013185291344, 0.2687833371486663, 0.3627260019079481, 0.3357302531360347, 0.09579106464167564, 0.09078402460743347, 0.1432628000747467, 0.027398974188114347, 0.29005984908202037, 0, 0.12470347850095743, 0.20733088149834183, 0.5210800715276844, 0.49733991790266924, 0.4788248196910145, 0.2513239543131709, 0.3139541125117947, 0, 0.06527953636229035, 0.3000631347470772, 0.25157902093275714, 0.1423130333996611, 0.11211729812070613, 0.09628721945215143, 0.2203696244314844, 0, 0.05577934104768975, 0.2889369137248172, 0.10285738543970784, 0.03922071315328133, 0.07241635993847241, 0.4332192046497044, 0.08701137698962969, 0.25355216768254873, 0.20400364928140816, 0.26846922368858367, 0.3171903469976426, 0.1704540634895672, 0.4536335591204484, 0.24708493995917716, 0.5194534393741757, 0.7315161831361969, 0.12371245977216577, 0.12841033113637165, 0.10015355420953857, 0.7542148505910502, 0.01904819497069441, 0.2610767619285739, 0.581697693973191, 0.397110236324442, 0.10270505371143165, 0, 0.14458122881110752, 0.0408219945202552, 0.017699577099400857, 0.21586019792310143, 0.10285738543970784, 0.23175065892827565, 0.2878786231128721, 0.08125596541900915, 0.21953660728795193, 0.48958393178079224, 0.1562496709111764, 0.4236080146001832, 0.22307018046851843, 0.04652001563489291, 0.07078216002419026, 0.08975110473074996, 0.22141338744266592, 0.20791143964896336, 0.22063347667575928, 0.39476334310140454, 0.40064564140073683, 0.24470209218738725, 0.20299603209814956, 0.1711434297167633, 0.5062453063358553, 0.32551174176459885, 0.38457145235017187, 0.1890972026487061, 0.09814500386711796, 0.308875430192432, 0.08256646198220989, 0.25646656667060136, 0.0832306541497237, 0.40916426096416614, 0.12315735959923854, 0.18157397305646186, 0, 0.03774032798284711, 0.3199284401308198, 0.0897144552474972, 0.7374200464659105, 0.1574933013239483, 0.3245098329505486, 0.01980262729617973, 0.6023501259305707, 0.023530497410194036, 0.1992703271151589, 0, 0.122807080328797, 0.08073546835243707, 0.1156688827345372, 0.22904346789932764, 0.656355788331214, 0.1133669700236827, 0.453394555090991, 0.09362410203266773, 0.2842613925734462, 0, 0.26573894326829517, 0.0821920905536809, 0.030771658666753687, 0.11568187924775865, 0.2968758942395612, 0.3572785972019682, 0.10643071484915245, 0.06303589125303606, 0.12366434229466367, 0.03869635100021315, 0, 0.07988058812507898, 0.2882718193728121, 0, 0.16916421336781381, 0.42553739279585084, 0.09868286428296408, 0.15651591955198224, 0.07548065596569423, 0.2187752567682635, 0.20972053098206897, 0.20488394727777673, 0.10099761896436962, 0.0425596144187959, 0.028987536873252187, 0.09043486544115059, 0.2133583621708569, 0.04762804898925456, 0.2932111316868448, 0.1515016716556815, 0.030771658666753687, 0.1924600047075036, 0, 0.23992215503199935, 0.50251870410112, 0.24638109858379503, 0.2842103575287407, 0.09887207137068536, 0.3443872693829294, 0.15503479612041532, 0, 0.2221737930799541, 0, 0.30759063391600494, 0.18196842924907225, 0.1363053195072478, 0.27582076682613255, 0.21555324130157064, 0.28981822308026056, 0.22939598534655042, 0.2551972161426085, 0, 0.09516716943582934, 0.22219220220278987, 0.07410797215372183, 0.11155248590574732, 0.22851825989699254, 0.17141443212132973, 0.14156065436534698, 0.3798189391249129, 0.47175348645717613, 0.08366860478883899, 0, 0.23884221760679383, 0.05406722127027579, 0.2518048862843096, 0.03636764417087479, 0.0538940760876079, 0.10570351000178224, 0.04879016416943204, 0.21578605125894523, 0.16186686509813325, 0.033901551675681416, 0.18823931052485038, 0.17275210577780392, 0.09571594162770503, 0, 0.44610658039161094, 0.2682831663482278, 0.18152243497181408, 0, 0.016529301951210506, 0.20309462144176846, 0.13544841361417936, 0.34340856714813633, 0.043333124706373766, 0.21514786997437274, 0.206330921063112, 0.19832442850796958, 0.3223159263044003, 0.3328809646423415, 0.37302617591865805, 0.15113862935726663, 0, 0.20074582412001307, 0.10857901493847075, 0.24967592784411366, 0.1955064535388289, 0.11086259281180968, 0.03509131981127019, 0.25951455615777475, 0.11332868530700316, 0, 0.27462802068154596, 0.17383502871877926, 0.1281980046924064, 0.11228095850240005, 0.09924276914218927, 0.0646732103424351, 0, 0.14172815982870124, 0.4143193845197183, 0.07410797215372183, 0.2562860474316991, 0.4300703994347073, 0.20729830065488358, 0.05862181722146423, 0.06119961987205649, 0.1007762192358831, 0.27596456957710236, 0.14630515162206637, 0.19952932125470388, 0.10893833700571026, 0.12127187319926133, 0.21906165276360431, 0.3948346865255206, 0.6353718462551258, 0.2356543539421092, 0.018692133012152546, 0.1960777545782824, 0.03278982282299097, 0.4643475327949507, 0.3381793138570286, 0.14944969458628946, 0.39239533349945827, 0.2342416139723143, 0, 0.22367302403817127, 0.2067796203038561, 0.17379388105693847, 0, 0.444582261304897, 0.1250366327381374, 0, 0.04574811285631096, 0.34787202285411584, 0.29268338158451856, 0.2330597982294237, 0.21456420424181827, 0.06899287148695142, 0.20572068494123472, 0, 0.45294836193478694, 0.16859544520658332, 0.11028285162388682, 0.14502628079416638, 0.053244514518812264, 0.13995945373270371, 0.27517480808754924, 0.1095228123444032, 0.09419671286333373, 0.19010049703631735, 0.05582844955294186, 0.04320875433123673, 0.14387754270884548, 0, 0.06406581963975959, 0.372589017908963, 0.44629766149669015, 0.6781745687958939, 0.05325490353554395, 1.3793445643069773, 0.19411076454841797, 0.12109429696003947, 0.162518929497775, 0.2909649912268323, 0.12260232209233227, 0.13305050778717198, 0, 0.12738890924724994, 0.2557633385950229, 0.10492157208284747, 0.05743990574891496, 0.06062462181643484, 0.05226842854575675, 0.4133694441516156, 0.2963889552558158, 0.24142274429878652, 0.20230119718142728, 0.29718197904096816, 0.2951184904770694, 0, 0.18649096626663852, 0.15778044787783696, 0.2139848237536475, 0.12270915399243928, 0.11469184308671063, 0.12914061149765146, 0, 0.06453852113757116, 0.02985296314968113, 0.11492996667397704, 0.19980310950490698, 0.2748668936301272, 0.2079978538048761, 0, 0.40845662316198417, 0.38394534633883404, 0.06183787956190764, 0.13147044680450015, 0.11553450617255262, 0, 0, 0.09025836226629333, 0.40770014126502796, 0, 0.4001075071386027, 0.2554384898786971, 0.390811392137748, 0.19222194387967076, 0.13133600206108687, 0.2126163336549734, 0.5101196369272565, 0.18366424219052227, 0.7140679788005743, 0.024692612590371414, 0.1181136688219018, 0.0685119866496008, 0.13582562870889064, 0, 0.04941535956335031, 0.11208501454174583, 0.2219101591573968, 0.20842804535152293, 0.2458551795502694, 0.08175026897928944, 0.2623284755400861, 0.1133550014427433, 0.0425596144187959, 0.2229540390507535, 0.18217854642545914, 0.12233763160757191, 0.23835230123759812, 0.06367863883909021, 0.1372011215134849, 0.18778300977311108, 0.033901551675681416, 0.10477263926166765, 0.05715841383994862, 0.19508109893831654, 0.16339039265349947, 0.37132110410297964, 0.11262268509707372, 0.5596946607896534, 0.1843796817885755, 0.15898388981367653, 0.13703402317572463, 0.2953313136976805, 0.027398974188114347, 0.07780055426784668, 0.05715841383994863, 0.28274683726505, 0.20989943786151544, 0.07260534226918666, 0.06453852113757116, 0.16308279488604693, 0.3755470718210355, 0.024097551579060524, 0.11153477704973111, 0.2862397725723922, 0.29435203374839886, 0.04879016416943204, 0.3205612134580993, 0.021506205220963682, 0.03509131981127019, 0.15035736445849956, 0.16124260513817024, 0, 0, 0.1370340231757248, 0.0696424330557471, 0.17275866580037041, 0.3385641125134701, 0.13740927411812714, 0.08963689555132273, 0.17050800678339478, 0.3790792112903017, 0.05129329438755048, 0.2395349230783605, 0.22246917763392288, 0.06453852113757116, 0.07053015080583808, 0.14821594430744367, 0.418052196076908, 0.10166598776840821, 0.14157189762039812, 0.22647453018747848, 0.26442529310001056, 0, 0.19948735470204002, 0.4772765230446541, 0, 0.030771658666753687, 0.17297058567357115, 0.15385707127362938, 0.291141548188373, 0.2669335190855895, 0.3338992324884449, 0.4850612489724496, 0.24348611742840015, 0.06915746699386577, 0.28703685640090953, 0.21514065480000907, 0.0408219945202552, 0.8082932069604444, 0.20026922443575568, 0.03509131981127019, 0.13629927266234027, 0.24697681197009835, 0.4616421496524675, 0.23516035662317478, 0, 0.18800064703923056, 0.10058119006026107, 0.1396262082183209, 0.11944525553956542, 0.046189382469374436, 0.5138943978151397, 0.06854931414989432, 0.05667059709020238, 0.23777605148091552, 0.07112000965299507, 0.2505610515030889, 0.22017208704496932, 0, 0.29850470219935893, 0.688785170577222, 0.028987536873252187, 0.19159739925647618, 0.33754354875568376, 0.03922071315328133, 0.46548875397962924, 0.09352605801082342, 0.11040897520366823, 0.030771658666753687, 0.07096941146786159, 0.05406722127027579, 0.14339654243639954, 0.32860636352984596, 0.07773774020430049, 0.1177282125977913, 0.12567040048954942, 0.5151651977944322, 0.20382208585676925, 0.07954308238210399, 0.31092028097839725, 0.08388148398070223, 0.17269089590409883, 0.10870313715451638, 0.20255377219003817, 0.08701137698962974, 0.11768104167393989, 0.105736496427871, 0.1630847003756671, 0.36293814728371565, 0.13681733304670834, 0.08507700352429566, 0.28043844457073536, 0.2768672262792015, 0.033901551675681416, 0.04212549242369929, 0.19637577117927033, 0.10827172586528466, 0.30862993910565417, 0.024097551579060524, 0.14497877846487342, 0]\n"
     ]
    }
   ],
   "source": [
    "demo1=sum_log_probabilities(dm1_human)\n",
    "print(demo1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The word with the highest log probability is 0 with a log probability of 1489.4840635928042\n"
     ]
    }
   ],
   "source": [
    "max_probability = max(demo1)\n",
    "max_index = demo1.index(max_probability)\n",
    "print(f\"The word with the highest log probability is {max_index} with a log probability of {max_probability}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sum_log_probabilities(dm1_machine)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA18AAAIjCAYAAAD80aFnAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/bCgiHAAAACXBIWXMAAA9hAAAPYQGoP6dpAABNpklEQVR4nO3dd3wVVf7/8fdNQkJJI7QkELpUCSCuIYAUQaoUZVUQV9rKiqGzKHxXpfxUEBuioLjSXFGwAbuirEhdqhBgURYR2NAJ1eQSSgjJ+f3hI7PcNHLhZm4SX8/H4z4emTNnZj5z70nk7cyc6zDGGAEAAAAACpSPtwsAAAAAgN8CwhcAAAAA2IDwBQAAAAA2IHwBAAAAgA0IXwAAAABgA8IXAAAAANiA8AUAAAAANiB8AQAAAIANCF8AAAAAYAPCFwDAxeHDh+VwOPTaa695bJ+TJk2Sw+HQuXPnbtq3evXqGjBggLW8bt06ORwOrVu3zmobMGCAqlev7rH6cHOZ42LBggVeOb7D4dCkSZO8cmwA8BTCFwDkYcGCBXI4HNqxY4e3S7ECTOardOnSatCggZ577jk5nU5vl+dVly9f1qRJk1wCWlH0/fffy+Fw6M0338y2rmfPnnI4HJo/f362da1bt1blypXtKPGmMsPy559/7u1SAKDQ8fN2AQAA97z77rsKDAxUSkqKvv32W7300ktas2aNNm3aJIfD4e3ybtv+/fvl45P3/xv861//qoyMDGv58uXLmjx5siSpbdu2BVlegbrrrrtUunRpbdy4UaNHj3ZZt3nzZvn5+WnTpk0aOHCg1X7t2jVt375d3bt3t7tcAICbCF8AUMT8/ve/V/ny5SVJTz31lHr37q0vv/xSW7duVWxsbI7bXL58WaVLl7azzFsWEBBw0z4lSpSwoRL7+fn5KSYmRps2bXJp379/v86dO6fHHntMGzdudFkXHx+vq1evqlWrVrd9/KI0TgCgKOK2QwDwgF27dqlLly4KDg5WYGCg2rdvr61bt2brt2fPHrVp00alSpVSlSpV9OKLL2r+/PlyOBw6fPjwLR37vvvukyQlJCRI+vXKz5133qn4+Hi1bt1apUuX1v/93/9Jks6cOaPBgwerUqVKKlmypBo3bqyFCxfmuu8333xT1apVU6lSpdSmTRv9+OOP2c5nwIABqlmzpkqWLKnw8HANGjRI58+fz3F/586d0yOPPKLg4GCVK1dOI0eO1NWrV136ZH3mKyc3PvN1+PBhVahQQZI0efJk67bMSZMmWe/trl27su3j5Zdflq+vr06cOJHjMT7//HM5HA6tX78+27o5c+bI4XBY70diYqIGDhyoKlWqKCAgQBEREerZs+ctfaatWrXS6dOndfDgQatt06ZNCg4O1pAhQ6wgduO6zO0yzZ49Ww0bNlRAQIAiIyMVFxenpKQkl+PkNU6SkpI0YMAAhYSEKDQ0VP3798+2vTsyb5k9ePCgBgwYoNDQUIWEhGjgwIG6fPmyS9/U1FSNHj1aFSpUUFBQkHr06KHjx4/nuN8TJ05o0KBBqlSpkgICAtSwYUPNmzfPWn/lyhXVq1dP9erV05UrV6z2CxcuKCIiQi1atFB6evotnxcAuIsrXwBwm/bu3at7771XwcHBeuaZZ1SiRAnNmTNHbdu21fr16xUTEyPp138otmvXTg6HQxMmTFCZMmX0wQcf5OtKT14OHTokSSpXrpzVdv78eXXp0kV9+vTR448/rkqVKunKlStq27atDh48qGHDhqlGjRr67LPPNGDAACUlJWnkyJEu+/3www918eJFxcXF6erVq3rrrbd033336YcfflClSpUkSatWrdJ///tfDRw4UOHh4dq7d6/ef/997d27V1u3bs12G+Qjjzyi6tWra+rUqdq6datmzpypX375RR9++OEtn3+FChX07rvvaujQoXrwwQf10EMPSZKio6NVo0YNxcXFadGiRWratKnLdosWLVLbtm1zfVaqW7duCgwM1Keffqo2bdq4rFuyZIkaNmyoO++8U5LUu3dv7d27V8OHD1f16tV15swZrVq1SkePHnV7YpDMELVx40bVrl1b0q8Bq3nz5oqJiVGJEiW0efNm9ejRw1oXFBSkxo0bS/o16EyePFkdOnTQ0KFDtX//fr377rvavn27Nm3a5HLVMKdxYoxRz549tXHjRj311FOqX7++li5dqv79+7t1Hjl55JFHVKNGDU2dOlU7d+7UBx98oIoVK+qVV16x+vzxj3/URx99pMcee0wtWrTQmjVr1K1bt2z7On36tJo3by6Hw6Fhw4apQoUK+uabbzR48GA5nU6NGjVKpUqV0sKFC9WyZUv95S9/0RtvvCFJiouLU3JyshYsWCBfX9/bPi8AyDcDAMjV/PnzjSSzffv2XPv06tXL+Pv7m0OHDlltJ0+eNEFBQaZ169ZW2/Dhw43D4TC7du2y2s6fP2/CwsKMJJOQkJBnLRMnTjSSzP79+83Zs2dNQkKCmTNnjgkICDCVKlUyly5dMsYY06ZNGyPJvPfeey7bz5gxw0gyH330kdV27do1ExsbawIDA43T6TTGGJOQkGAkmVKlSpnjx49bfbdt22YkmdGjR1ttly9fzlbnJ598YiSZDRs2ZKu9R48eLn2ffvppI8n8+9//ttqqVatm+vfvby2vXbvWSDJr16612vr372+qVatmLZ89e9ZIMhMnTsxWT9++fU1kZKRJT0+32nbu3Gkkmfnz52frn3XbihUrmuvXr1ttp06dMj4+PmbKlCnGGGN++eUXI8m8+uqree4rv5xOp/H19TWDBw+22urWrWsmT55sjDHmnnvuMePGjbPWVahQwdx///3GGGPOnDlj/P39TceOHV3O95133jGSzLx586y23MbJsmXLjCQzffp0q+369evm3nvvzdd7lvl5ffbZZ1Zb5uc/aNAgl74PPvigKVeunLW8e/duI8k8/fTTLv0ee+yxbJ/v4MGDTUREhDl37pxL3z59+piQkBCXsTlhwgTj4+NjNmzYYD777DMjycyYMSPP8wCAgsBthwBwG9LT0/Xtt9+qV69eqlmzptUeERFhPZ+TORPhypUrFRsbqyZNmlj9wsLC1K9fP7eOWbduXVWoUEE1atTQn/70J9WuXVsrVqxweVYnICDAZVIGSfr6668VHh6uvn37Wm0lSpTQiBEjlJKSku32ul69erlcFbrnnnsUExOjr7/+2morVaqU9fPVq1d17tw5NW/eXJK0c+fObLXHxcW5LA8fPtyqraA88cQTOnnypNauXWu1LVq0SKVKlVLv3r3z3PbRRx/VmTNnXGZR/Pzzz5WRkaFHH31U0q/vgb+/v9atW6dffvnltusNCgpSdHS09WzXuXPntH//frVo0UKS1LJlS+tWw59//llnz561rpZ99913unbtmkaNGuUyacmTTz6p4OBgrVixwuVYuY0TPz8/DR061Grz9fW1Pqvb8dRTT7ks33vvvTp//rz1O5I5DkaMGOHSb9SoUS7Lxhh98cUX6t69u4wxOnfunPXq1KmTkpOTXcbfpEmT1LBhQ/Xv319PP/202rRpk+0YAGAHwhcA3IazZ8/q8uXLqlu3brZ19evXV0ZGho4dOyZJOnLkiHUb2Y1yasvLF198oVWrVmndunU6ePCgfvzxRzVr1sylT+XKleXv7+/SduTIEd1xxx3ZZhKsX7++tf5Gd9xxR7Zj16lTx+U5pgsXLmjkyJGqVKmSSpUqZYVCSUpOTs62fdZ91qpVSz4+Prf8vFt+3H///YqIiNCiRYskSRkZGfrkk0/Us2dPBQUF5blt586dFRISoiVLllhtS5YsUZMmTVSnTh1JvwaYV155Rd98840qVaqk1q1ba/r06UpMTLzlmlu1amU927V582b5+vpaobZFixaKj49Xampqtue9Mj/DrOPR399fNWvWzPYZ5zZOIiIiFBgY6NKe0xh3V9WqVV2Wy5YtK0lWaD1y5Ih8fHxUq1atPI999uxZJSUl6f3331eFChVcXplh8syZM1Z/f39/zZs3TwkJCbp48aL1LCAA2I1nvgCgiGndurU122FubrwiVZAeeeQRbd68WePGjVOTJk0UGBiojIwMde7c2WUq+NzY8Q9gX19fPfbYY/rrX/+q2bNna9OmTTp58qQef/zxm24bEBCgXr16aenSpZo9e7ZOnz6tTZs26eWXX3bpN2rUKHXv3l3Lli3TP//5Tz3//POaOnWq1qxZk+1Zs/xo1aqV3n77bW3atEmbN29Wo0aNrDDUokULpaamavv27dq4caP8/PysYOYuu8ZJptyerzLGuLWfzLH1+OOP5/osWnR0tMvyP//5T0m/XqE9cOCA9T8JAMBOXPkCgNtQoUIFlS5dWvv378+27qeffpKPj4+ioqIkSdWqVXOZwS5TTm0FoVq1ajpw4EC2UPTTTz9Z62904MCBbPv4+eefrQkkfvnlF61evVrjx4/X5MmT9eCDD+r+++93uf0yq6z7PHjwoDIyMtyelCKrm4W4J554Qk6nU//4xz+0aNEiVahQQZ06dcrXvh999FGdO3dOq1ev1meffSZjjHXL4Y1q1aqlsWPH6ttvv9WPP/6oa9eu6fXXX7+l87lx0o1NmzapZcuW1rrIyEhVq1ZNmzZt0qZNm9S0aVPrltPMzzDreLx27ZoSEhKyfcY5qVatmk6dOqWUlBSX9pzGuKdVq1ZNGRkZ1iQyuR07cybE9PR0dejQIcdXxYoVrf579uzRlClTNHDgQDVt2lR//OMfc7wyCwAFjfAFALfB19dXHTt21PLly11unTt9+rQ+/vhjtWrVSsHBwZKkTp06acuWLdq9e7fV78KFC9btcAWta9euSkxMdLmF7vr163r77bcVGBiYbUa/ZcuWuUzD/v3332vbtm3q0qWLpP9dxch61WLGjBm51jBr1iyX5bfffluSrH3eqszwkdt06NHR0YqOjtYHH3ygL774Qn369JGfX/5u/ujQoYPCwsK0ZMkSLVmyRPfcc4/LVZPLly9nmy6/Vq1aCgoKUmpqqtV26tQp/fTTT0pLS7vpMSMjI1WjRg2tXr1aO3bssJ73ytSiRQstW7ZM+/fvd5livkOHDvL399fMmTNdPpe5c+cqOTk5x1kDs+ratauuX7+ud99912pLT0+3PquClDkOZs6c6dKedUz5+vqqd+/e+uKLL7J9/YH0622JmdLS0jRgwABFRkbqrbfe0oIFC3T69OlsX2INAHbgtkMAyId58+Zp5cqV2dpHjhypF198UatWrVKrVq309NNPy8/PT3PmzFFqaqqmT59u9X3mmWf00Ucf6f7779fw4cOtqearVq2qCxcuFPgteEOGDNGcOXM0YMAAxcfHq3r16vr888+1adMmzZgxI9vzT7Vr11arVq00dOhQpaamasaMGSpXrpyeeeYZSVJwcLD1fFNaWpoqV66sb7/91vq+sZwkJCSoR48e6ty5s7Zs2WJNKZ45TfqtKlWqlBo0aKAlS5aoTp06CgsL05133mlNBS/9evXrz3/+syTl65bDTCVKlNBDDz2kxYsX69KlS3rttddc1v/8889q3769HnnkETVo0EB+fn5aunSpTp8+rT59+lj9JkyYoIULFyohISFfV/patWqlv/3tb5LkcuVL+jV8ffLJJ1a/TBUqVNCECRM0efJkde7cWT169ND+/fs1e/Zs/e53v8vXeXfv3l0tW7bU+PHjdfjwYTVo0EBffvmlLVeKmjRpor59+2r27NlKTk5WixYttHr16hyvDk+bNk1r165VTEyMnnzySTVo0EAXLlzQzp079d133+nChQuSpBdffFG7d+/W6tWrrclMXnjhBT333HP6/e9/r65duxb4eQGAxZtTLQJAYZc51Xxur2PHjhljfp26vFOnTiYwMNCULl3atGvXzmzevDnb/nbt2mXuvfdeExAQYKpUqWKmTp1qZs6caSSZxMTEPGvJnK777NmzefZr06aNadiwYY7rTp8+bQYOHGjKly9v/P39TaNGjbJNHZ451fyrr75qXn/9dRMVFWUCAgLMvffe6zIlvDHGHD9+3Dz44IMmNDTUhISEmIcffticPHky27TgmbX/5z//Mb///e9NUFCQKVu2rBk2bJi5cuWKyz5vZap5Y4zZvHmzadasmfH3989x2vlTp04ZX19fU6dOnTzfv5ysWrXKSDIOh8P6zDOdO3fOxMXFmXr16pkyZcqYkJAQExMTYz799FOXfv3798/XVwpkmjNnjpFkKleunG1d5lT5kszp06ezrX/nnXdMvXr1TIkSJUylSpXM0KFDzS+//OLSJ69xcv78efOHP/zBBAcHm5CQEPOHP/zB7Nq167anms86djN/v258T65cuWJGjBhhypUrZ8qUKWO6d+9ujh07luNnevr0aRMXF2eioqJMiRIlTHh4uGnfvr15//33jTHGxMfHGz8/PzN8+HCX7a5fv25+97vfmcjIyGzvCwAUJIcxbj7lCgDwqFGjRmnOnDlKSUnhC18L0Llz5xQREaEXXnhBzz//vLfLAQD8BvHMFwDY6MqVKy7L58+f19/+9je1atWK4FXAFixYoPT0dP3hD3/wdikAgN8onvkCABvFxsaqbdu2ql+/vk6fPq25c+fK6XRyJaYArVmzRv/5z3/00ksvqVevXrc9syIAALeK2w4BwEb/93//p88//1zHjx+Xw+HQXXfdpYkTJ6pDhw7eLq3Yatu2rTZv3qyWLVvqo48+UuXKlb1dEgDgN4rwBQAAAAA24JkvAAAAALAB4QsAAAAAbMCEG5IyMjJ08uRJBQUFFfiXnAIAAAAovIwxunjxoiIjI+Xj49lrVYQvSSdPnlRUVJS3ywAAAABQSBw7dkxVqlTx6D4JX5KCgoIk/foGBwcHe7kaAAAAAN7idDoVFRVlZQRPInxJ1q2GwcHBhC8AAAAABfI4EhNuAAAAAIANCF8AAAAAYAPCFwAAAADYgPAFAAAAADYgfAEAAACADQhfAAAAAGADwhcAAAAA2IDwBQAAAAA2IHwBAAAAgA0IXwAAAABgA8IXAAAAANiA8AUAAAAANiB8AQAAAIANCF8AAAAAYAPCFwAAAADYgPAFAAAAADYgfAEAAACADQhfAAAAAGADwhcAAAAA2IDwBQAAAAA2IHwBAAAAgA0IXwAAAABgA8IXAAAAANiA8AUAAAAANiB8AQAAAIANCF8AAAAAYAPCFwAAAADYgPAFAAAAADYgfAEAAACADQhfAAAAAGADwhcAAAAA2IDwBQAAAAA2IHwBAAAAgA0IXwAAAABgA8IXAAAAANiA8AUAAAAANvBq+NqwYYO6d++uyMhIORwOLVu2zGW9w+HI8fXqq69afapXr55t/bRp02w+EwAAAADIm1fD16VLl9S4cWPNmjUrx/WnTp1yec2bN08Oh0O9e/d26TdlyhSXfsOHD7ejfAAAAADINz9vHrxLly7q0qVLruvDw8NdlpcvX6527dqpZs2aLu1BQUHZ+gIAAABAYVJknvk6ffq0VqxYocGDB2dbN23aNJUrV05NmzbVq6++quvXr+e5r9TUVDmdTpcXAAAAABQkr175csfChQsVFBSkhx56yKV9xIgRuuuuuxQWFqbNmzdrwoQJOnXqlN54441c9zV16lRNnjy5oEsGAAAAAIvDGGO8XYT06+QaS5cuVa9evXJcX69ePd1///16++2389zPvHnz9Kc//UkpKSkKCAjIsU9qaqpSU1OtZafTqaioKCUnJys4OPiWzwEAAABA0eZ0OhUSElIg2aBIXPn617/+pf3792vJkiU37RsTE6Pr16/r8OHDqlu3bo59AgICcg1mAAAAAFAQisQzX3PnzlWzZs3UuHHjm/bdvXu3fHx8VLFiRRsqAwAAAID88eqVr5SUFB08eNBaTkhI0O7duxUWFqaqVatK+vWy32effabXX3892/ZbtmzRtm3b1K5dOwUFBWnLli0aPXq0Hn/8cZUtW9a28wAAAACAm/Fq+NqxY4fatWtnLY8ZM0aS1L9/fy1YsECStHjxYhlj1Ldv32zbBwQEaPHixZo0aZJSU1NVo0YNjR492toPAAAAABQWhWbCDW8qyIfqAAAAABQdBZkNisQzXwAAAABQ1BG+AAAAAMAGhC8AAAAAsAHhCwAAAABsQPgCAAAAABsQvgAAAADABoQvAAAAALAB4QsAAAAAbED4AgAAAAAbEL4AAAAAwAaELwAAAACwAeELAAAAAGxA+AIAAAAAGxC+AAAAAMAGhC8AAAAAsAHhCwAAAABsQPi6wZ0T/+ntEgAAAAAUU4QvAAAAALAB4QsAAAAAbED4AgAAAAAbEL4AAAAAwAaELwAAAACwAeELAAAAAGxA+AIAAAAAGxC+AAAAAMAGhC8AAAAAsAHhCwAAAABsQPgCAAAAABsQvgAAAADABoQvAAAAALAB4QsAAAAAbED4AgAAAAAbEL4AAAAAwAaELwAAAACwAeELAAAAAGxA+AIAAAAAGxC+AAAAAMAGhC8AAAAAsAHhCwAAAABsQPgCAAAAABsQvgAAAADABoQvAAAAALAB4QsAAAAAbED4AgAAAAAbEL4AAAAAwAaELwAAAACwAeELAAAAAGxA+AIAAAAAGxC+AAAAAMAGXg1fGzZsUPfu3RUZGSmHw6Fly5a5rB8wYIAcDofLq3Pnzi59Lly4oH79+ik4OFihoaEaPHiwUlJSbDwLAAAAALg5r4avS5cuqXHjxpo1a1aufTp37qxTp05Zr08++cRlfb9+/bR3716tWrVKX331lTZs2KAhQ4YUdOkAAAAA4BY/bx68S5cu6tKlS559AgICFB4enuO6ffv2aeXKldq+fbvuvvtuSdLbb7+trl276rXXXlNkZKTHawYAAACAW1Hon/lat26dKlasqLp162ro0KE6f/68tW7Lli0KDQ21gpckdejQQT4+Ptq2bVuu+0xNTZXT6XR5AQAAAEBBKtThq3Pnzvrwww+1evVqvfLKK1q/fr26dOmi9PR0SVJiYqIqVqzoso2fn5/CwsKUmJiY636nTp2qkJAQ6xUVFVWg5wEAAAAAXr3t8Gb69Olj/dyoUSNFR0erVq1aWrdundq3b3/L+50wYYLGjBljLTudTgIYAAAAgAJVqK98ZVWzZk2VL19eBw8elCSFh4frzJkzLn2uX7+uCxcu5PqcmPTrc2TBwcEuLwAAAAAoSEUqfB0/flznz59XRESEJCk2NlZJSUmKj4+3+qxZs0YZGRmKiYnxVpkAAAAAkI1XbztMSUmxrmJJUkJCgnbv3q2wsDCFhYVp8uTJ6t27t8LDw3Xo0CE988wzql27tjp16iRJql+/vjp37qwnn3xS7733ntLS0jRs2DD16dOHmQ4BAAAAFCpevfK1Y8cONW3aVE2bNpUkjRkzRk2bNtULL7wgX19f7dmzRz169FCdOnU0ePBgNWvWTP/6178UEBBg7WPRokWqV6+e2rdvr65du6pVq1Z6//33vXVKAAAAAJAjhzHGeLsIb3M6nb/OejjqUx1982FvlwMAAADASzKzQXJyssfnhihSz3wBAAAAQFFF+AIAAAAAGxC+AAAAAMAGhC8AAAAAsAHhCwAAAABsQPgCAAAAABsQvgAAAADABoQvAAAAALAB4QsAAAAAbED4AgAAAAAbEL4AAAAAwAaELwAAAACwAeELAAAAAGxA+AIAAAAAGxC+AAAAAMAGhC8AAAAAsAHhCwAAAABsQPgCAAAAABsQvgAAAADABoQvAAAAALAB4QsAAAAAbED4AgAAAAAbEL4AAAAAwAaELwAAAACwAeELAAAAAGxA+AIAAAAAGxC+AAAAAMAGhC8AAAAAsAHhCwAAAABsQPgCAAAAABsQvgAAAADABoQvAAAAALAB4QsAAAAAbED4AgAAAAAbEL4AAAAAwAaELwAAAACwAeELAAAAAGxA+AIAAAAAGxC+AAAAAMAGhC8AAAAAsAHhCwAAAABsQPgCAAAAABsQvgAAAADABoQvAAAAALAB4QsAAAAAbED4AgAAAAAbEL4AAAAAwAaELwAAAACwAeELAAAAAGzg1fC1YcMGde/eXZGRkXI4HFq2bJm1Li0tTc8++6waNWqkMmXKKDIyUk888YROnjzpso/q1avL4XC4vKZNm2bzmQAAAABA3rwavi5duqTGjRtr1qxZ2dZdvnxZO3fu1PPPP6+dO3fqyy+/1P79+9WjR49sfadMmaJTp05Zr+HDh9tRPgAAAADkm583D96lSxd16dIlx3UhISFatWqVS9s777yje+65R0ePHlXVqlWt9qCgIIWHhxdorQAAAABwO4rUM1/JyclyOBwKDQ11aZ82bZrKlSunpk2b6tVXX9X169fz3E9qaqqcTqfLCwAAAAAKklevfLnj6tWrevbZZ9W3b18FBwdb7SNGjNBdd92lsLAwbd68WRMmTNCpU6f0xhtv5LqvqVOnavLkyXaUDQAAAACSJIcxxni7CElyOBxaunSpevXqlW1dWlqaevfurePHj2vdunUu4SurefPm6U9/+pNSUlIUEBCQY5/U1FSlpqZay06nU1FRUYoa9amOvvnwbZ8LAAAAgKLJ6XQqJCREycnJeeaOW1Hor3ylpaXpkUce0ZEjR7RmzZqbvgExMTG6fv26Dh8+rLp16+bYJyAgINdgBgAAAAAFoVCHr8zgdeDAAa1du1blypW76Ta7d++Wj4+PKlasaEOFAAAAAJA/Xg1fKSkpOnjwoLWckJCg3bt3KywsTBEREfr973+vnTt36quvvlJ6eroSExMlSWFhYfL399eWLVu0bds2tWvXTkFBQdqyZYtGjx6txx9/XGXLlvXWaQEAAABANl4NXzt27FC7du2s5TFjxkiS+vfvr0mTJunvf/+7JKlJkyYu261du1Zt27ZVQECAFi9erEmTJik1NVU1atTQ6NGjrf0AAAAAQGHh1fDVtm1b5TXfx83mArnrrru0detWT5cFAAAAAB5XpL7nCwAAAACKKsIXAAAAANiA8AUAAAAANiB8AQAAAIANCF8AAAAAYAPCFwAAAADYgPAFAAAAADYgfAEAAACADQhfAAAAAGADwhcAAAAA2IDwBQAAAAA2cDt8TZw4UUeOHCmIWgAAAACg2HI7fC1fvly1atVS+/bt9fHHHys1NbUg6gIAAACAYsXt8LV7925t375dDRs21MiRIxUeHq6hQ4dq+/btBVEfAAAAABQLt/TMV9OmTTVz5kydPHlSc+fO1fHjx9WyZUtFR0frrbfeUnJysqfrBAAAAIAi7bYm3DDGKC0tTdeuXZMxRmXLltU777yjqKgoLVmyxFM1AgAAAECRd0vhKz4+XsOGDVNERIRGjx6tpk2bat++fVq/fr0OHDigl156SSNGjPB0rQAAAABQZLkdvho1aqTmzZsrISFBc+fO1bFjxzRt2jTVrl3b6tO3b1+dPXvWo4UCAAAAQFHm5+4GjzzyiAYNGqTKlSvn2qd8+fLKyMi4rcIAAAAAoDhx+8pX5rNdWV25ckVTpkzxSFEAAAAAUNy4Hb4mT56slJSUbO2XL1/W5MmTPVIUAAAAABQ3t3Tly+FwZGv/97//rbCwMI8UBQAAAADFTb6f+SpbtqwcDoccDofq1KnjEsDS09OVkpKip556qkCKBAAAAICiLt/ha8aMGTLGaNCgQZo8ebJCQkKsdf7+/qpevbpiY2MLpEgAAAAAKOryHb769+8vSapRo4ZatGihEiVKFFhRAAAAAFDc5Ct8OZ1OBQcHS5KaNm2qK1eu6MqVKzn2zewHAAAAAPiffIWvsmXL6tSpU6pYsaJCQ0NznHAjcyKO9PR0jxcJAAAAAEVdvsLXmjVrrJkM165dW6AFAQAAAEBxlK/w1aZNmxx/BgAAAADkT77C1549e/K9w+jo6FsuBgAAAACKq3yFryZNmsjhcMgYk2c/nvkCAAAAgJzlK3wlJCQUdB0AAAAAUKzlK3xVq1atoOsAAAAAgGItX+Hr73//u7p06aISJUro73//e559e/To4ZHCAAAAAKA4yVf46tWrlxITE1WxYkX16tUr13488wUAAAAAOctX+MrIyMjxZwAAAABA/vh4uwAAAAAA+C24pfC1evVqPfDAA6pVq5Zq1aqlBx54QN99952nawMAAACAYsPt8DV79mx17txZQUFBGjlypEaOHKng4GB17dpVs2bNKogaAQAAAKDIy9czXzd6+eWX9eabb2rYsGFW24gRI9SyZUu9/PLLiouL82iBAAAAAFAcuH3lKykpSZ07d87W3rFjRyUnJ3ukKAAAAAAobtwOXz169NDSpUuztS9fvlwPPPCAR4oCAAAAgOImX7cdzpw50/q5QYMGeumll7Ru3TrFxsZKkrZu3apNmzZp7NixBVMlAAAAABRxDmOMuVmnGjVq5G9nDof++9//3nZRdnM6nQoJCVHUqE919M2HvV0OAAAAAC/JzAbJyckKDg726L7zdeUrISHBowcFAAAAgN8avmQZAAAAAGzg9lTzknT8+HH9/e9/19GjR3Xt2jWXdW+88YZHCgMAAACA4sTt8LV69Wr16NFDNWvW1E8//aQ777xThw8fljFGd911V0HUCAAAAABFntu3HU6YMEF//vOf9cMPP6hkyZL64osvdOzYMbVp00YPP8xkFQAAAACQE7fD1759+/TEE09Ikvz8/HTlyhUFBgZqypQpeuWVV9za14YNG9S9e3dFRkbK4XBo2bJlLuuNMXrhhRcUERGhUqVKqUOHDjpw4IBLnwsXLqhfv34KDg5WaGioBg8erJSUFHdPCwAAAAAKlNvhq0yZMtZzXhERETp06JC17ty5c27t69KlS2rcuLFmzZqV4/rp06dr5syZeu+997Rt2zaVKVNGnTp10tWrV60+/fr10969e7Vq1Sp99dVX2rBhg4YMGeLuaQEAAABAgXL7ma/mzZtr48aNql+/vrp27aqxY8fqhx9+0JdffqnmzZu7ta8uXbqoS5cuOa4zxmjGjBl67rnn1LNnT0nShx9+qEqVKmnZsmXq06eP9u3bp5UrV2r79u26++67JUlvv/22unbtqtdee02RkZHunh4AAAAAFAi3r3y98cYbiomJkSRNnjxZ7du315IlS1S9enXNnTvXY4UlJCQoMTFRHTp0sNpCQkIUExOjLVu2SJK2bNmi0NBQK3hJUocOHeTj46Nt27bluu/U1FQ5nU6XFwAAAAAUJLevfNWsWdP6uUyZMnrvvfc8WlCmxMRESVKlSpVc2itVqmStS0xMVMWKFV3W+/n5KSwszOqTk6lTp2ry5MkerhgAAAAAcndL3/MlSTt27NC+ffskSQ0aNFCzZs08VlRBmzBhgsaMGWMtO51ORUVFebEiAAAAAMWd2+Hr+PHj6tu3rzZt2qTQ0FBJUlJSklq0aKHFixerSpUqHiksPDxcknT69GlFRERY7adPn1aTJk2sPmfOnHHZ7vr167pw4YK1fU4CAgIUEBDgkToBAAAAID/cfubrj3/8o9LS0rRv3z5duHBBFy5c0L59+5SRkaE//vGPHiusRo0aCg8P1+rVq602p9Opbdu2KTY2VpIUGxurpKQkxcfHW33WrFmjjIwM67k0AAAAACgM3L7ytX79em3evFl169a12urWrau3335b9957r1v7SklJ0cGDB63lhIQE7d69W2FhYapatapGjRqlF198UXfccYdq1Kih559/XpGRkerVq5ckqX79+urcubOefPJJvffee0pLS9OwYcPUp08fZjoEAAAAUKi4Hb6ioqKUlpaWrT09Pd3twLNjxw61a9fOWs58Dqt///5asGCBnnnmGV26dElDhgxRUlKSWrVqpZUrV6pkyZLWNosWLdKwYcPUvn17+fj4qHfv3po5c6a7pwUAAAAABcphjDHubLB8+XK9/PLLmjVrljXF+44dOzR8+HA9++yz1lWposTpdCokJERRoz7V0Tcf9nY5AAAAALwkMxskJycrODjYo/vO15WvsmXLyuFwWMuXLl1STEyM/Px+3fz69evy8/PToEGDimT4AgAAAICClq/wNWPGjAIuAwAAAACKt3yFr/79+xd0HQAAAABQrN3Slyynp6dr2bJl1pcsN2zYUD169JCvr69HiwMAAACA4sLt8HXw4EF17dpVJ06csKabnzp1qqKiorRixQrVqlXL40UCAAAAQFHn9pcsjxgxQrVq1dKxY8e0c+dO7dy5U0ePHlWNGjU0YsSIgqgRAAAAAIo8t6eaL1OmjLZu3apGjRq5tP/73/9Wy5YtlZKS4tEC7XDjVPM+AaV1eFo3b5cEAAAAwAsKcqp5t698BQQE6OLFi9naU1JS5O/v75GiAAAAAKC4cTt8PfDAAxoyZIi2bdsmY4yMMdq6daueeuop9ejRoyBqBAAAAIAiz+3wNXPmTNWqVUuxsbEqWbKkSpYsqZYtW6p27dp66623CqJGAAAAACjy3Jrt0Bgjp9OpxYsX68SJE9ZU8/Xr11ft2rULpEAAAAAAKA7cDl+1a9fW3r17dccddxC4AAAAACCf3Lrt0MfHR3fccYfOnz9fUPUAAAAAQLHk9jNf06ZN07hx4/Tjjz8WRD0AAAAAUCy5dduhJD3xxBO6fPmyGjduLH9/f5UqVcpl/YULFzxWHAAAAAAUF26HrzfffFMOh6MgagEAAACAYsvt8NW3b19dv35dZcqUKYh6AAAAAKBYyvczX2fPnlWXLl0UGBio4OBgNW/eXAcPHizI2gAAAACg2Mh3+Hr22We1e/duTZkyRa+99pqSkpL05JNPFmRtAAAAAFBs5Pu2w1WrVmnBggXq1KmTJOmBBx5Q/fr1lZqaqoCAgAIrEAAAAACKg3xf+Tp58qQaN25sLd9xxx0KCAjQqVOnCqQwAAAAAChO3PqeL19f32zLxhiPFgQAAAAAxVG+bzs0xqhOnTou08ynpKSoadOm8vH5X4bje74AAAAAILt8h6/58+cXZB0AAAAAUKzlO3z179+/IOsAAAAAgGLNrWe+AAAAAAC3hvAFAAAAADYgfAEAAACADQhfAAAAAGADwhcAAAAA2CDfsx1mGjNmTI7tDodDJUuWVO3atdWzZ0+FhYXddnEAAAAAUFy4Hb527dqlnTt3Kj09XXXr1pUk/fzzz/L19VW9evU0e/ZsjR07Vhs3blSDBg08XjAAAAAAFEVu33bYs2dPdejQQSdPnlR8fLzi4+N1/Phx3X///erbt69OnDih1q1ba/To0QVRLwAAAAAUSQ5jjHFng8qVK2vVqlXZrmrt3btXHTt21IkTJ7Rz50517NhR586d82ixBcXpdCokJERRoz6VT0BpHZ7WzdslAQAAAPCCzGyQnJys4OBgj+7b7StfycnJOnPmTLb2s2fPyul0SpJCQ0N17dq1268OAAAAAIqJW7rtcNCgQVq6dKmOHz+u48ePa+nSpRo8eLB69eolSfr+++9Vp04dT9cKAAAAAEWW2xNuzJkzR6NHj1afPn10/fr1X3fi56f+/fvrzTfflCTVq1dPH3zwgWcrBQAAAIAizO1nvjKlpKTov//9rySpZs2aCgwM9GhhduKZLwAAAABSwT7z5faVr0yBgYHWd3kV5eAFAAAAAHZw+5mvjIwMTZkyRSEhIapWrZqqVaum0NBQ/b//9/+UkZFREDUCAAAAQJHn9pWvv/zlL5o7d66mTZumli1bSpI2btyoSZMm6erVq3rppZc8XiQAAAAAFHVuh6+FCxfqgw8+UI8ePay26OhoVa5cWU8//TThCwAAAABy4PZthxcuXFC9evWytderV08XLlzwSFEAAAAAUNy4Hb4aN26sd955J1v7O++8o8aNG3ukKAAAAAAobty+7XD69Onq1q2bvvvuO8XGxkqStmzZomPHjunrr7/2eIEAAAAAUBy4feWrTZs2+vnnn/Xggw8qKSlJSUlJeuihh7R//37de++9BVEjAAAAABR5t/Q9X5GRkdkm1jh+/LiGDBmi999/3yOFAQAAAEBx4vaVr9ycP39ec+fO9dTuAAAAAKBY8Vj4KijVq1eXw+HI9oqLi5MktW3bNtu6p556ystVAwAAAICrW7rt0E7bt29Xenq6tfzjjz/q/vvv18MPP2y1Pfnkk5oyZYq1XLp0aVtrBAAAAICbKfThq0KFCi7L06ZNU61atdSmTRurrXTp0goPD7e7NAAAAADIt3yHr4ceeijP9UlJSbdby01du3ZNH330kcaMGSOHw2G1L1q0SB999JHCw8PVvXt3Pf/883le/UpNTVVqaqq17HQ6C7RuAAAAAMh3+AoJCbnp+ieeeOK2C8rLsmXLlJSUpAEDBlhtjz32mKpVq6bIyEjt2bNHzz77rPbv368vv/wy1/1MnTpVkydPLtBaAQAAAOBGDmOM8XYR+dWpUyf5+/vrH//4R6591qxZo/bt2+vgwYOqVatWjn1yuvIVFRWlqFGfyiegtA5P6+bx2gEAAAAUfk6nUyEhIUpOTlZwcLBH913on/nKdOTIEX333Xd5XtGSpJiYGEnKM3wFBAQoICDA4zUCAAAAQG4K/VTzmebPn6+KFSuqW7e8r0rt3r1bkhQREWFDVQAAAACQP0XiyldGRobmz5+v/v37y8/vfyUfOnRIH3/8sbp27apy5cppz549Gj16tFq3bq3o6GgvVgwAAAAAropE+Pruu+909OhRDRo0yKXd399f3333nWbMmKFLly4pKipKvXv31nPPPeelSgEAAAAgZ0UifHXs2FE5zQsSFRWl9evXe6EiAAAAAHBPkXnmCwAAAACKMsIXAAAAANiA8AUAAAAANiB8AQAAAIANCF8AAAAAYAPCFwAAAADYgPAFAAAAADYgfAEAAACADQhfAAAAAGADwhcAAAAA2IDwBQAAAAA2IHwBAAAAgA0IXwAAAABgA8JXLqqPX+HtEgAAAAAUI4QvAAAAALAB4QsAAAAAbED4AgAAAAAbEL4AAAAAwAaELwAAAACwAeELAAAAAGxA+AIAAAAAGxC+AAAAAMAGhC8AAAAAsAHhCwAAAABsQPgCAAAAABsQvgAAAADABoQvAAAAALAB4QsAAAAAbED4AgAAAAAbEL4AAAAAwAaELwAAAACwAeELAAAAAGxA+AIAAAAAGxC+AAAAAMAGhC8AAAAAsAHhCwAAAABsQPgCAAAAABsQvgAAAADABoQvAAAAALAB4QsAAAAAbED4AgAAAAAbEL7yUH38Cm+XAAAAAKCYIHwBAAAAgA0IXwAAAABgA8IXAAAAANiA8AUAAAAANiB8AQAAAIANCF8AAAAAYAPCFwAAAADYoFCHr0mTJsnhcLi86tWrZ62/evWq4uLiVK5cOQUGBqp37946ffq0FysGAAAAgJwV6vAlSQ0bNtSpU6es18aNG611o0eP1j/+8Q999tlnWr9+vU6ePKmHHnrIi9UCAAAAQM78vF3Azfj5+Sk8PDxbe3JysubOnauPP/5Y9913nyRp/vz5ql+/vrZu3armzZvbXSoAAAAA5KrQX/k6cOCAIiMjVbNmTfXr109Hjx6VJMXHxystLU0dOnSw+tarV09Vq1bVli1b8txnamqqnE6nywsAAAAAClKhDl8xMTFasGCBVq5cqXfffVcJCQm69957dfHiRSUmJsrf31+hoaEu21SqVEmJiYl57nfq1KkKCQmxXlFRUQV4FgAAAABQyMNXly5d9PDDDys6OlqdOnXS119/raSkJH366ae3td8JEyYoOTnZeh07dizP/tXHr7it4wEAAABAoQ5fWYWGhqpOnTo6ePCgwsPDde3aNSUlJbn0OX36dI7PiN0oICBAwcHBLi8AAAAAKEhFKnylpKTo0KFDioiIULNmzVSiRAmtXr3aWr9//34dPXpUsbGxXqwSAAAAALIr1LMd/vnPf1b37t1VrVo1nTx5UhMnTpSvr6/69u2rkJAQDR48WGPGjFFYWJiCg4M1fPhwxcbGMtMhAAAAgEKnUIev48ePq2/fvjp//rwqVKigVq1aaevWrapQoYIk6c0335SPj4969+6t1NRUderUSbNnz/Zy1QAAAACQXaEOX4sXL85zfcmSJTVr1izNmjXLpooAAAAA4NYUqWe+AAAAAKCoInwBAAAAgA0IXwAAAABgA8IXAAAAANiA8AUAAAAANiB8AQAAAIANCF8AAAAAYAPC1y2oPn6Fqo9f4e0yAAAAABQhhC8AAAAAsAHhCwAAAABsQPgCAAAAABsQvgAAAADABoQvAAAAALAB4QsAAAAAbED4AgAAAAAbEL4AAAAAwAaELwAAAACwAeELAAAAAGxA+AIAAAAAGxC+AAAAAMAGhC8AAAAAsAHhy03Vx6/wdgkAAAAAiiDCFwAAAADYgPAFAAAAADYgfAEAAACADQhfAAAAAGADwhcAAAAA2IDwBQAAAAA2IHwBAAAAgA0IXwAAAABgA8IXAAAAANiA8AUAAAAANiB8AQAAAIANCF8AAAAAYAPCFwAAAADYwM/bBRQV1cev8HYJAAAAAIowrnwBAAAAgA0IXwAAAABgA8IXAAAAANiA8OUBPA8GAAAA4GYIXwAAAABgA8IXAAAAANiA8AUAAAAANiB8AQAAAIANCF8FiIk4AAAAAGQifAEAAACADQhfAAAAAGADwtdt4LZCAAAAAPlF+PIQghgAAACAvBTq8DV16lT97ne/U1BQkCpWrKhevXpp//79Ln3atm0rh8Ph8nrqqae8VDEAAAAA5KxQh6/169crLi5OW7du1apVq5SWlqaOHTvq0qVLLv2efPJJnTp1ynpNnz7dSxUDAAAAQM78vF1AXlauXOmyvGDBAlWsWFHx8fFq3bq11V66dGmFh4fbXR4AAAAA5FuhvvKVVXJysiQpLCzMpX3RokUqX7687rzzTk2YMEGXL1/Ocz+pqalyOp0uLwAAAAAoSEUmfGVkZGjUqFFq2bKl7rzzTqv9scce00cffaS1a9dqwoQJ+tvf/qbHH388z31NnTpVISEh1isqKsqjtTL5BgAAAICsCvVthzeKi4vTjz/+qI0bN7q0DxkyxPq5UaNGioiIUPv27XXo0CHVqlUrx31NmDBBY8aMsZadTqfHAxgAAAAA3KhIhK9hw4bpq6++0oYNG1SlSpU8+8bExEiSDh48mGv4CggIUEBAgMfrBAAAAIDcFOrwZYzR8OHDtXTpUq1bt041atS46Ta7d++WJEVERBRwdQAAAACQf4U6fMXFxenjjz/W8uXLFRQUpMTERElSSEiISpUqpUOHDunjjz9W165dVa5cOe3Zs0ejR49W69atFR0d7eXqAQAAAOB/CnX4evfddyX9+kXKN5o/f74GDBggf39/fffdd5oxY4YuXbqkqKgo9e7dW88995wXqgUAAACA3BXq8GWMyXN9VFSU1q9fb1M1AAAAAHDrisxU8wAAAABQlBG+AAAAAMAGhC8AAAAAsAHhCwAAAABsQPgCAAAAABsQvgAAAADABoQvAAAAALAB4QsAAAAAbED4KiSqj1/h7RIAAAAAFCDCFwAAAADYgPAFAAAAADYgfBUh3JoIAAAAFF2ELwAAAACwAeELAAAAAGxA+LIJtwwCAAAAv22ELwAAAACwAeELAAAAAGzg5+0CipustxdyuyEAAAAAiStfAAAAAGALwpcXcDUMAAAA+O0hfP0GEPYAAAAA7yN8AQAAAIANCF8AAAAAYAPCFwAAAADYgKnmbXSrz17xzBYAAABQ9HHly8sIVgAAAMBvA+HLS6qPX5Fn8CKUAQAAAMUL4QsAAAAAbED4AgAAAAAbEL6KKW5bBAAAAAoXwtdvxI3PmN3seTMAAAAAnkf4AgAAAAAbEL4KEXeuRnHlCgAAAChaCF8AAAAAYAPCF/KU3+fDbuzDVTkAAAAgO8JXIWdXkCEwAQAAAAWL8IVcEcgAAAAAzyF8FXF5BaRbmVK+sASuwlIHAAAA4CmEr0LoVgOT3YGFgAQAAADkH+GriPHkFyR7OjwRxgAAAIDcEb5+Y7wRkDwZGAEAAICiivCFApU1dHkriBH+AAAA4G2Er2KGkOF5t/JdZ0VFUawZAACgqCJ8IcerU3Ycx1v7KIh9AQAAADdD+MJtyS3A3OxqUX5maLxxH1n3l3WdO8e+mcztCWd54/0BAABwD+ELbivM/+i+ne89y+95EcwAAABwKwhfyBdvhg1PXMXyBjuvoGW9KphTHbe6v6KoqNcPAACKJ8IXChU7nukqDP8wdzcQefNZt5v1LwzvZ34VpVozFcWaAQBAzghfQD7l9I/ggg5ROe0/r6tc+d0uv+23OxlLXu9PbufhyWN4SmEJQIWljqKA9woAUBgVm/A1a9YsVa9eXSVLllRMTIy+//57b5cEG93qP7RuNvGHp/4B505AyG9gud1juLPv/Owv6+Qonpie/2ahLD+3W+bnXAvq6wTymjAmP/u28znE271N9nYC9M32l7W9MAWronpLdmFSXM4DAPKjWISvJUuWaMyYMZo4caJ27typxo0bq1OnTjpz5oy3S0M+5Gfmw6LCEzMtepLdtyt6OsC5ewx3tstv8MttuxvDiqeu3N1uCMxPwLuVkFcQwSq3l7u15fd4OR37xmVPyu9neCvHzm07T1zZzs92t9u3II/h7v+gKc6KyrnaVWdhfz8K2/9UcldRr99uft4uwBPeeOMNPfnkkxo4cKAk6b333tOKFSs0b948jR8/3svV4beoIP/Pf2FwK/9gPDytm7Xd4Wndbut4mfvzBE/94+xW9uOJf3zm9x+lWd+vm72Hea2/8Zg39sktEGR+9rf7meX3Pc7pOPk936xj9Mb2rG03qzW/73lOx87p59yOk59j59Y367qbvUd5yW/9N/t8sv58s+O7M7azjt28xqg7x8ttH7n1y2ldbuMrt9+3rNvmNHbz2ubGmnOqP6/f85v9TuS0bX5+h/L7nudnH+78juY2RvNzHjkdK6f95ed9zrrPWzmXnNpz+7uWU/+sx8vp885p33nVltM5uPM33N2/v+7w9P7yo8iHr2vXrik+Pl4TJkyw2nx8fNShQwdt2bIlx21SU1OVmppqLScnJ0uSMlIvS5KcTmeOP+ckt743284T+7D72FVHf+a1Y3tiu8J67Dsn/lOS9OPkTm4dO3M7T573jZ+xO9t54tie3q6wHjs/n1vV0Z+5jAd3Pqusv6c5HS/zZ6fTKUnZ6shpfW593Rkzue0j63ZZa876ftxMbttlfc+z1n6zz+rG/rn9nPV4WX+/C/p37Maac/o5q5zeo/xsdzv15/XfEul/482dY+c1Ztx5zzO3y+vYeX3eN+5DUo77ufEYmf2y/n7k5/f7xjrzes+z7utW/rbk9ruS9X3O6b272d+kW3nPcxqjWfeRdQxlLt/4nuf2tyW39yDruWYdu5ly+5ua23uX23uVuS7re5T1Z3fGeX7/LuS0j8zzzO2/Yzcba/n9905Ov2M5vf831pJVXu9XbmP0xnWZ9Rljcq3zVjlMQezVRidPnlTlypW1efNmxcbGWu3PPPOM1q9fr23btmXbZtKkSZo8ebKdZQIAAAAoQg4dOqSaNWt6dJ9F/srXrZgwYYLGjBljLSclJalatWo6evSoQkJCvFgZijun06moqCgdO3ZMwcHB3i4HxRhjDXZhrMEujDXYJTk5WVWrVlVYWJjH913kw1f58uXl6+ur06dPu7SfPn1a4eHhOW4TEBCggICAbO0hISH8MsMWwcHBjDXYgrEGuzDWYBfGGuzi4+P5uQmL/GyH/v7+atasmVavXm21ZWRkaPXq1S63IQIAAACANxX5K1+SNGbMGPXv319333237rnnHs2YMUOXLl2yZj8EAAAAAG8rFuHr0Ucf1dmzZ/XCCy8oMTFRTZo00cqVK1WpUqV8bR8QEKCJEyfmeCsi4EmMNdiFsQa7MNZgF8Ya7FKQY63Iz3YIAAAAAEVBkX/mCwAAAACKAsIXAAAAANiA8AUAAAAANiB8AQAAAIANfvPha9asWapevbpKliypmJgYff/9994uCUXMhg0b1L17d0VGRsrhcGjZsmUu640xeuGFFxQREaFSpUqpQ4cOOnDggEufCxcuqF+/fgoODlZoaKgGDx6slJQUG88CRcHUqVP1u9/9TkFBQapYsaJ69eql/fv3u/S5evWq4uLiVK5cOQUGBqp3797ZvoT+6NGj6tatm0qXLq2KFStq3Lhxun79up2ngkLu3XffVXR0tPVltrGxsfrmm2+s9YwzFIRp06bJ4XBo1KhRVhtjDZ4yadIkORwOl1e9evWs9XaNtd90+FqyZInGjBmjiRMnaufOnWrcuLE6deqkM2fOeLs0FCGXLl1S48aNNWvWrBzXT58+XTNnztR7772nbdu2qUyZMurUqZOuXr1q9enXr5/27t2rVatW6auvvtKGDRs0ZMgQu04BRcT69esVFxenrVu3atWqVUpLS1PHjh116dIlq8/o0aP1j3/8Q5999pnWr1+vkydP6qGHHrLWp6enq1u3brp27Zo2b96shQsXasGCBXrhhRe8cUoopKpUqaJp06YpPj5eO3bs0H333aeePXtq7969khhn8Lzt27drzpw5io6OdmlnrMGTGjZsqFOnTlmvjRs3WutsG2vmN+yee+4xcXFx1nJ6erqJjIw0U6dO9WJVKMokmaVLl1rLGRkZJjw83Lz66qtWW1JSkgkICDCffPKJMcaY//znP0aS2b59u9Xnm2++MQ6Hw5w4ccK22lH0nDlzxkgy69evN8b8OrZKlChhPvvsM6vPvn37jCSzZcsWY4wxX3/9tfHx8TGJiYlWn3fffdcEBweb1NRUe08ARUrZsmXNBx98wDiDx128eNHccccdZtWqVaZNmzZm5MiRxhj+psGzJk6caBo3bpzjOjvH2m/2yte1a9cUHx+vDh06WG0+Pj7q0KGDtmzZ4sXKUJwkJCQoMTHRZZyFhIQoJibGGmdbtmxRaGio7r77bqtPhw4d5OPjo23bttleM4qO5ORkSVJYWJgkKT4+XmlpaS7jrV69eqpatarLeGvUqJHLl9B36tRJTqfTuqoB3Cg9PV2LFy/WpUuXFBsbyziDx8XFxalbt24uY0ribxo878CBA4qMjFTNmjXVr18/HT16VJK9Y83PQ+dS5Jw7d07p6ekub6AkVapUST/99JOXqkJxk5iYKEk5jrPMdYmJiapYsaLLej8/P4WFhVl9gKwyMjI0atQotWzZUnfeeaekX8eSv7+/QkNDXfpmHW85jcfMdUCmH374QbGxsbp69aoCAwO1dOlSNWjQQLt372acwWMWL16snTt3avv27dnW8TcNnhQTE6MFCxaobt26OnXqlCZPnqx7771XP/74o61j7TcbvgCgKIuLi9OPP/7ocr864El169bV7t27lZycrM8//1z9+/fX+vXrvV0WipFjx45p5MiRWrVqlUqWLOntclDMdenSxfo5OjpaMTExqlatmj799FOVKlXKtjp+s7cdli9fXr6+vtlmMTl9+rTCw8O9VBWKm8yxlNc4Cw8PzzbJy/Xr13XhwgXGInI0bNgwffXVV1q7dq2qVKlitYeHh+vatWtKSkpy6Z91vOU0HjPXAZn8/f1Vu3ZtNWvWTFOnTlXjxo311ltvMc7gMfHx8Tpz5ozuuusu+fn5yc/PT+vXr9fMmTPl5+enSpUqMdZQYEJDQ1WnTh0dPHjQ1r9rv9nw5e/vr2bNmmn16tVWW0ZGhlavXq3Y2FgvVobipEaNGgoPD3cZZ06nU9u2bbPGWWxsrJKSkhQfH2/1WbNmjTIyMhQTE2N7zSi8jDEaNmyYli5dqjVr1qhGjRou65s1a6YSJUq4jLf9+/fr6NGjLuPthx9+cAn8q1atUnBwsBo0aGDPiaBIysjIUGpqKuMMHtO+fXv98MMP2r17t/W6++671a9fP+tnxhoKSkpKig4dOqSIiAh7/67d0nQhxcTixYtNQECAWbBggfnPf/5jhgwZYkJDQ11mMQFu5uLFi2bXrl1m165dRpJ54403zK5du8yRI0eMMcZMmzbNhIaGmuXLl5s9e/aYnj17mho1apgrV65Y++jcubNp2rSp2bZtm9m4caO54447TN++fb11Siikhg4dakJCQsy6devMqVOnrNfly5etPk899ZSpWrWqWbNmjdmxY4eJjY01sbGx1vrr16+bO++803Ts2NHs3r3brFy50lSoUMFMmDDBG6eEQmr8+PFm/fr1JiEhwezZs8eMHz/eOBwO8+233xpjGGcoODfOdmgMYw2eM3bsWLNu3TqTkJBgNm3aZDp06GDKly9vzpw5Y4yxb6z9psOXMca8/fbbpmrVqsbf39/cc889ZuvWrd4uCUXM2rVrjaRsr/79+xtjfp1u/vnnnzeVKlUyAQEBpn379mb//v0u+zh//rzp27evCQwMNMHBwWbgwIHm4sWLXjgbFGY5jTNJZv78+VafK1eumKefftqULVvWlC5d2jz44IPm1KlTLvs5fPiw6dKliylVqpQpX768GTt2rElLS7P5bFCYDRo0yFSrVs34+/ubChUqmPbt21vByxjGGQpO1vDFWIOnPProoyYiIsL4+/ubypUrm0cffdQcPHjQWm/XWHMYY8xtXbMDAAAAANzUb/aZLwAAAACwE+ELAAAAAGxA+AIAAAAAGxC+AAAAAMAGhC8AAAAAsAHhCwAAAABsQPgCAAAAABsQvgAAAADABoQvAMBvTtu2bTVq1KhicxwAQNFA+AIA2Oq9995TUFCQrl+/brWlpKSoRIkSatu2rUvfdevWyeFw6NChQ7bWuGDBAoWGhtp6TABA8Uf4AgDYql27dkpJSdGOHTustn/9618KDw/Xtm3bdPXqVat97dq1qlq1qmrVquX2cYwxLgEPAABvI3wBAGxVt25dRUREaN26dVbbunXr1LNnT9WoUUNbt251aW/Xrp0kKTU1VSNGjFDFihVVsmRJtWrVStu3b3fp63A49M0336hZs2YKCAjQxo0bdenSJT3xxBMKDAxURESEXn/9dbdrnjRpkpo0aaK//e1vql69ukJCQtSnTx9dvHjR6pOf46SmpurPf/6zKleurDJlyigmJsZ6H65evaqGDRtqyJAhVv9Dhw4pKChI8+bNc7tmAEDhQ/gCANiuXbt2Wrt2rbW8du1atW3bVm3atLHar1y5om3btlnh65lnntEXX3yhhQsXaufOnapdu7Y6deqkCxcuuOx7/PjxmjZtmvbt26fo6GiNGzdO69ev1/Lly/Xtt99q3bp12rlzp9s1Hzp0SMuWLdNXX32lr776SuvXr9e0adOs9fk5zrBhw7RlyxYtXrxYe/bs0cMPP6zOnTvrwIEDKlmypBYtWqSFCxdq+fLlSk9P1+OPP677779fgwYNcrteAEAhZAAAsNlf//pXU6ZMGZOWlmacTqfx8/MzZ86cMR9//LFp3bq1McaY1atXG0nmyJEjJiUlxZQoUcIsWrTI2se1a9dMZGSkmT59ujHGmLVr1xpJZtmyZVafixcvGn9/f/Ppp59abefPnzelSpUyI0eOzLW++fPnm5CQEGt54sSJpnTp0sbpdFpt48aNMzExMfk+zpEjR4yvr685ceKEy7Hat29vJkyYYC1Pnz7dlC9f3gwbNsxERESYc+fO3eztBAAUEX7eDn8AgN+etm3b6tKlS9q+fbt++eUX1alTRxUqVFCbNm00cOBAXb16VevWrVPNmjVVtWpV7dmzR2lpaWrZsqW1jxIlSuiee+7Rvn37XPZ99913Wz8fOnRI165dU0xMjNUWFhamunXrul1z9erVFRQUZC1HRETozJkz+T7ODz/8oPT0dNWpU8dlv6mpqSpXrpy1PHbsWC1btkzvvPOOvvnmG5d1AICijfAFALBd7dq1VaVKFa1du1a//PKL2rRpI0mKjIxUVFSUNm/erLVr1+q+++5ze99lypTxdLmSfg17N3I4HMrIyMj39ikpKfL19VV8fLx8fX1d1gUGBlo/nzlzRj///LN8fX114MABde7c+fYKBwAUGjzzBQDwinbt2mndunVat26dyxTzrVu31jfffKPvv//eet6rVq1a8vf316ZNm6x+aWlp2r59uxo0aJDrMWrVqqUSJUpo27ZtVtsvv/yin3/+2aPnkp/jNG3aVOnp6Tpz5oxq167t8goPD7f6DRo0SI0aNdLChQv17LPPZruyBwAourjyBQDwinbt2ikuLk5paWnWlS9JatOmjYYNG6Zr165Z4atMmTIaOnSoxo0bp7CwMFWtWlXTp0/X5cuXNXjw4FyPERgYqMGDB2vcuHEqV66cKlasqL/85S/y8fHs/3vMz3Hq1Kmjfv366YknntDrr7+upk2b6uzZs1q9erWio6PVrVs3zZo1S1u2bNGePXsUFRWlFStWqF+/ftq6dav8/f09WjMAwH6ELwCAV7Rr105XrlxRvXr1VKlSJau9TZs2unjxojUlfaZp06YpIyNDf/jDH3Tx4kXdfffd+uc//6myZcvmeZxXX31VKSkp6t69u4KCgjR27FglJyd7/Hzyc5z58+frxRdf1NixY3XixAmVL19ezZs31wMPPKCffvpJ48aN09y5cxUVFSVJmj17tqKjo/X888/rlVde8XjNAAB7OYwxxttFAAAAAEBxxzNfAAAAAGADwhcAAAAA2IDwBQAAAAA2IHwBAAAAgA0IXwAAAABgA8IXAAAAANiA8AUAAAAANiB8AQAAAIANCF8AAAAAYAPCFwAAAADYgPAFAAAAADb4/5zu8/AQKd4gAAAAAElFTkSuQmCC",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sum_log_probabilities(dm2_human)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sum_log_probabilities(dm2_machine)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "----------------------------------------------------------------------------------------"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# read test_set.json into the system\n",
    "test = pd.read_json('data/test_set.json', lines=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "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>id</th>\n",
       "      <th>text</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>[59, 2, 3434, 1013, 823, 2, 887, 6, 2375, 0, 3...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>[2, 785, 6, 5, 0, 317, 17, 5, 2970, 3654, 858,...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>[3306, 4, 2, 2288, 6, 963, 94]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>[0, 2, 1021, 38, 126, 1, 907, 1, 761, 1, 3357,...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>[30, 488, 197, 18, 4, 31, 420, 18, 2, 3577, 6,...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>5</td>\n",
       "      <td>[0, 1041, 0, 2424, 862, 54, 63, 3048, 62, 3661...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>6</td>\n",
       "      <td>[695, 4297, 16, 0, 24, 2, 2357, 549, 2785, 3, ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>7</td>\n",
       "      <td>[5, 349, 1310, 98, 17, 770, 6, 0, 4, 32, 21, 3...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>8</td>\n",
       "      <td>[20, 38, 0, 2458, 3, 5, 2247, 1109, 49, 0, 706...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>9</td>\n",
       "      <td>[63, 281, 62, 2349, 35, 2, 48, 6, 213, 16, 2, ...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   id                                               text\n",
       "0   0  [59, 2, 3434, 1013, 823, 2, 887, 6, 2375, 0, 3...\n",
       "1   1  [2, 785, 6, 5, 0, 317, 17, 5, 2970, 3654, 858,...\n",
       "2   2                     [3306, 4, 2, 2288, 6, 963, 94]\n",
       "3   3  [0, 2, 1021, 38, 126, 1, 907, 1, 761, 1, 3357,...\n",
       "4   4  [30, 488, 197, 18, 4, 31, 420, 18, 2, 3577, 6,...\n",
       "5   5  [0, 1041, 0, 2424, 862, 54, 63, 3048, 62, 3661...\n",
       "6   6  [695, 4297, 16, 0, 24, 2, 2357, 549, 2785, 3, ...\n",
       "7   7  [5, 349, 1310, 98, 17, 770, 6, 0, 4, 32, 21, 3...\n",
       "8   8  [20, 38, 0, 2458, 3, 5, 2247, 1109, 49, 0, 706...\n",
       "9   9  [63, 281, 62, 2349, 35, 2, 48, 6, 213, 16, 2, ..."
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test.head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "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>id</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>1000.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>499.500000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>288.819436</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>249.750000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>499.500000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>749.250000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>999.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                id\n",
       "count  1000.000000\n",
       "mean    499.500000\n",
       "std     288.819436\n",
       "min       0.000000\n",
       "25%     249.750000\n",
       "50%     499.500000\n",
       "75%     749.250000\n",
       "max     999.000000"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test.describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "------------------------------------------------------"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Belance the data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### bag of words process"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# bag-of-words\n",
    "\n",
    "# count the appearance of each word in the text\n",
    "def count_word(text_list):\n",
    "    count = [0] * 5000\n",
    "    for token in text_list:\n",
    "        count[token] += 1\n",
    "    return count\n",
    "\n",
    "# apply the count_word function to each text and store the result in a new column\n",
    "def bag_of_word(df):\n",
    "    df['bag_of_word'] = df['text'].apply(count_word)\n",
    "    return df"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### <font color='red'>我这里认为标记了domain的model为“7”，不考虑model进去的话可以忽略</font>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "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>text</th>\n",
       "      <th>label</th>\n",
       "      <th>model</th>\n",
       "      <th>bag_of_word</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>[70, 746, 825, 109, 2083, 0, 2, 0, 0, 0, 9, 0,...</td>\n",
       "      <td>1</td>\n",
       "      <td>7</td>\n",
       "      <td>[7, 0, 1, 0, 0, 0, 0, 1, 0, 1, 0, 0, 0, 0, 0, ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>[1209, 179, 1952, 4, 4959, 7, 0, 2, 978, 1522,...</td>\n",
       "      <td>1</td>\n",
       "      <td>7</td>\n",
       "      <td>[1, 1, 1, 0, 1, 0, 1, 1, 0, 0, 0, 0, 0, 0, 0, ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>[287, 3, 3330, 0, 23, 12, 13, 465, 74, 8, 0, 8...</td>\n",
       "      <td>1</td>\n",
       "      <td>7</td>\n",
       "      <td>[6, 1, 0, 1, 2, 0, 0, 2, 4, 1, 0, 0, 1, 1, 0, ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>[0, 0, 3, 592, 19, 2, 706, 1439, 2575, 7, 2, 0...</td>\n",
       "      <td>1</td>\n",
       "      <td>7</td>\n",
       "      <td>[3, 1, 2, 2, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>[9, 2, 110, 12, 42, 32, 44, 361, 9, 3860, 2358...</td>\n",
       "      <td>1</td>\n",
       "      <td>7</td>\n",
       "      <td>[4, 1, 1, 1, 1, 0, 1, 0, 0, 2, 0, 0, 1, 0, 0, ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>[9, 892, 3, 0, 0, 3, 2, 0, 811, 3, 26, 939, 13...</td>\n",
       "      <td>1</td>\n",
       "      <td>7</td>\n",
       "      <td>[4, 1, 3, 4, 2, 0, 0, 0, 0, 1, 0, 0, 0, 2, 0, ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>[0, 1115, 0, 72, 192, 64, 31, 3295, 33, 0, 1]</td>\n",
       "      <td>1</td>\n",
       "      <td>7</td>\n",
       "      <td>[3, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>[2, 1510, 0, 9, 30, 138, 118, 2, 0, 0, 0, 5, 0...</td>\n",
       "      <td>1</td>\n",
       "      <td>7</td>\n",
       "      <td>[13, 1, 3, 3, 1, 1, 0, 1, 0, 1, 0, 0, 1, 0, 0,...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>[0, 1398, 324, 41, 1629, 0, 4162, 0, 4, 3020, ...</td>\n",
       "      <td>1</td>\n",
       "      <td>7</td>\n",
       "      <td>[5, 1, 1, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>[9, 5, 940, 4, 394, 3, 2, 537, 8, 15, 773, 3, ...</td>\n",
       "      <td>1</td>\n",
       "      <td>7</td>\n",
       "      <td>[1, 1, 2, 3, 1, 1, 1, 0, 1, 2, 0, 0, 1, 0, 0, ...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                text  label  model  \\\n",
       "0  [70, 746, 825, 109, 2083, 0, 2, 0, 0, 0, 9, 0,...      1      7   \n",
       "1  [1209, 179, 1952, 4, 4959, 7, 0, 2, 978, 1522,...      1      7   \n",
       "2  [287, 3, 3330, 0, 23, 12, 13, 465, 74, 8, 0, 8...      1      7   \n",
       "3  [0, 0, 3, 592, 19, 2, 706, 1439, 2575, 7, 2, 0...      1      7   \n",
       "4  [9, 2, 110, 12, 42, 32, 44, 361, 9, 3860, 2358...      1      7   \n",
       "5  [9, 892, 3, 0, 0, 3, 2, 0, 811, 3, 26, 939, 13...      1      7   \n",
       "6      [0, 1115, 0, 72, 192, 64, 31, 3295, 33, 0, 1]      1      7   \n",
       "7  [2, 1510, 0, 9, 30, 138, 118, 2, 0, 0, 0, 5, 0...      1      7   \n",
       "8  [0, 1398, 324, 41, 1629, 0, 4162, 0, 4, 3020, ...      1      7   \n",
       "9  [9, 5, 940, 4, 394, 3, 2, 537, 8, 15, 773, 3, ...      1      7   \n",
       "\n",
       "                                         bag_of_word  \n",
       "0  [7, 0, 1, 0, 0, 0, 0, 1, 0, 1, 0, 0, 0, 0, 0, ...  \n",
       "1  [1, 1, 1, 0, 1, 0, 1, 1, 0, 0, 0, 0, 0, 0, 0, ...  \n",
       "2  [6, 1, 0, 1, 2, 0, 0, 2, 4, 1, 0, 0, 1, 1, 0, ...  \n",
       "3  [3, 1, 2, 2, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, ...  \n",
       "4  [4, 1, 1, 1, 1, 0, 1, 0, 0, 2, 0, 0, 1, 0, 0, ...  \n",
       "5  [4, 1, 3, 4, 2, 0, 0, 0, 0, 1, 0, 0, 0, 2, 0, ...  \n",
       "6  [3, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ...  \n",
       "7  [13, 1, 3, 3, 1, 1, 0, 1, 0, 1, 0, 0, 1, 0, 0,...  \n",
       "8  [5, 1, 1, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ...  \n",
       "9  [1, 1, 2, 3, 1, 1, 1, 0, 1, 2, 0, 0, 1, 0, 0, ...  "
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# manully label train_domain1's model type as \"7\"\n",
    "train_domain1['model']=7\n",
    "\n",
    "# apply the bag_of_word function to train_domain1\n",
    "bag_train_domain1 = bag_of_word(train_domain1)\n",
    "bag_train_domain1.head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "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>text</th>\n",
       "      <th>label</th>\n",
       "      <th>model</th>\n",
       "      <th>bag_of_word</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>[3147, 471, 4, 343, 2, 0, 1, 14, 124, 133, 436...</td>\n",
       "      <td>1</td>\n",
       "      <td>NaN</td>\n",
       "      <td>[3, 5, 2, 0, 2, 2, 1, 1, 0, 0, 1, 0, 0, 1, 3, ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>[10, 389, 232, 24, 2, 0, 4730, 1, 13, 10, 129,...</td>\n",
       "      <td>1</td>\n",
       "      <td>NaN</td>\n",
       "      <td>[4, 4, 4, 2, 1, 0, 0, 2, 0, 0, 2, 0, 0, 3, 0, ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>[861, 0, 2505, 2, 0, 1015, 2, 0, 3, 1772, 8, 3...</td>\n",
       "      <td>1</td>\n",
       "      <td>NaN</td>\n",
       "      <td>[84, 25, 28, 26, 12, 12, 11, 13, 15, 12, 5, 0,...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>[325, 21, 3234, 1, 14, 187, 16, 13, 1965, 1, 1...</td>\n",
       "      <td>1</td>\n",
       "      <td>NaN</td>\n",
       "      <td>[0, 5, 0, 0, 2, 0, 0, 0, 0, 0, 0, 0, 1, 1, 4, ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>[5, 1091, 272, 6, 3232, 32, 2, 1012, 4, 3240, ...</td>\n",
       "      <td>1</td>\n",
       "      <td>NaN</td>\n",
       "      <td>[12, 11, 7, 1, 3, 2, 2, 1, 0, 3, 0, 4, 1, 0, 0...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>[10, 410, 36, 3484, 8, 15, 0, 59, 36, 3484, 12...</td>\n",
       "      <td>1</td>\n",
       "      <td>NaN</td>\n",
       "      <td>[5, 4, 3, 1, 1, 2, 1, 1, 2, 0, 3, 0, 0, 4, 0, ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>[503, 7, 11, 133, 4, 1, 860, 291, 10, 5, 0, 21...</td>\n",
       "      <td>1</td>\n",
       "      <td>NaN</td>\n",
       "      <td>[2, 5, 1, 2, 3, 4, 0, 1, 1, 1, 2, 1, 1, 1, 0, ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>[56, 239, 4, 31, 0, 3830, 53, 90, 2, 1745, 1, ...</td>\n",
       "      <td>1</td>\n",
       "      <td>NaN</td>\n",
       "      <td>[2, 5, 4, 0, 4, 1, 0, 1, 1, 1, 0, 1, 0, 0, 0, ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>[117, 191, 10, 2, 117, 69, 2, 339, 505, 465, 4...</td>\n",
       "      <td>1</td>\n",
       "      <td>NaN</td>\n",
       "      <td>[11, 11, 7, 1, 4, 1, 4, 8, 2, 5, 9, 13, 0, 2, ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>[10, 1345, 5, 853, 4, 21, 4815, 1, 53, 6, 5, 2...</td>\n",
       "      <td>1</td>\n",
       "      <td>NaN</td>\n",
       "      <td>[3, 5, 1, 0, 1, 4, 1, 1, 0, 0, 2, 0, 2, 1, 2, ...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                text  label  model  \\\n",
       "0  [3147, 471, 4, 343, 2, 0, 1, 14, 124, 133, 436...      1    NaN   \n",
       "1  [10, 389, 232, 24, 2, 0, 4730, 1, 13, 10, 129,...      1    NaN   \n",
       "2  [861, 0, 2505, 2, 0, 1015, 2, 0, 3, 1772, 8, 3...      1    NaN   \n",
       "3  [325, 21, 3234, 1, 14, 187, 16, 13, 1965, 1, 1...      1    NaN   \n",
       "4  [5, 1091, 272, 6, 3232, 32, 2, 1012, 4, 3240, ...      1    NaN   \n",
       "5  [10, 410, 36, 3484, 8, 15, 0, 59, 36, 3484, 12...      1    NaN   \n",
       "6  [503, 7, 11, 133, 4, 1, 860, 291, 10, 5, 0, 21...      1    NaN   \n",
       "7  [56, 239, 4, 31, 0, 3830, 53, 90, 2, 1745, 1, ...      1    NaN   \n",
       "8  [117, 191, 10, 2, 117, 69, 2, 339, 505, 465, 4...      1    NaN   \n",
       "9  [10, 1345, 5, 853, 4, 21, 4815, 1, 53, 6, 5, 2...      1    NaN   \n",
       "\n",
       "                                         bag_of_word  \n",
       "0  [3, 5, 2, 0, 2, 2, 1, 1, 0, 0, 1, 0, 0, 1, 3, ...  \n",
       "1  [4, 4, 4, 2, 1, 0, 0, 2, 0, 0, 2, 0, 0, 3, 0, ...  \n",
       "2  [84, 25, 28, 26, 12, 12, 11, 13, 15, 12, 5, 0,...  \n",
       "3  [0, 5, 0, 0, 2, 0, 0, 0, 0, 0, 0, 0, 1, 1, 4, ...  \n",
       "4  [12, 11, 7, 1, 3, 2, 2, 1, 0, 3, 0, 4, 1, 0, 0...  \n",
       "5  [5, 4, 3, 1, 1, 2, 1, 1, 2, 0, 3, 0, 0, 4, 0, ...  \n",
       "6  [2, 5, 1, 2, 3, 4, 0, 1, 1, 1, 2, 1, 1, 1, 0, ...  \n",
       "7  [2, 5, 4, 0, 4, 1, 0, 1, 1, 1, 0, 1, 0, 0, 0, ...  \n",
       "8  [11, 11, 7, 1, 4, 1, 4, 8, 2, 5, 9, 13, 0, 2, ...  \n",
       "9  [3, 5, 1, 0, 1, 4, 1, 1, 0, 0, 2, 0, 2, 1, 2, ...  "
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "bag_train_domain2 = bag_of_word(train_domain2)\n",
    "bag_train_domain2.head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count    14900.0\n",
       "mean      5000.0\n",
       "std          0.0\n",
       "min       5000.0\n",
       "25%       5000.0\n",
       "50%       5000.0\n",
       "75%       5000.0\n",
       "max       5000.0\n",
       "Name: bag_of_word, dtype: float64"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "bag_train_domain2['bag_of_word'].apply(len).describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Data replication"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# repeat the human label for multiple times\n",
    "\n",
    "def replicate(df, multiplication):\n",
    "    \n",
    "    # rows with label 1\n",
    "    label_1_rows = df[df['label'] == 1]\n",
    "    # Replicate the rows\n",
    "    replicated_df = pd.concat([label_1_rows] * (multiplication-1), ignore_index=True)\n",
    "    # Concatenate the original dataframe and the replicated dataframe\n",
    "    result_df = pd.concat([df, replicated_df], ignore_index=True)\n",
    "    \n",
    "    return result_df\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<font color='red'>一半一半</font>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1    12900\n",
       "0    12750\n",
       "Name: label, dtype: int64"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "new_df = replicate(train_domain2, 6)\n",
    "new_df['label'].value_counts(dropna=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Bootstrap"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<font color='red'>从label=1的行中有放回抽取，最终达到0和1的一定比例</font>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# random sample with replacement from label 1\n",
    "# label_1_rate: the rate of label 1 in the sampled dataframe, should be between 0 and 1\n",
    "\n",
    "def rated_sample(df, label_1_rate):\n",
    "\n",
    "    sampled_label_0 = df[df['label'] == 0]\n",
    "    label_0_count = len(df[df['label'] == 0])\n",
    "\n",
    "    #calculate the number of label 1 samples\n",
    "    sampled_label_1_count = int(label_1_rate * label_0_count)\n",
    "    \n",
    "    # random sample with replacement from label 1\n",
    "    sampled_label_1 = df[df['label'] == 1].sample(n=sampled_label_1_count, replace=True)\n",
    "\n",
    "    # concatenate the sampled label 0 and sampled label 1\n",
    "    sampled_df = pd.concat([sampled_label_0, sampled_label_1])\n",
    "    \n",
    "    return sampled_df\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "label\n",
       "0    12750\n",
       "1    12750\n",
       "Name: count, dtype: int64"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "new_df = rated_sample(train_domain2, 0.5)\n",
    "new_df['label'].value_counts(dropna=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "total_balanced_train = pd.concat([train_domain1,new_df], ignore_index=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "total_balanced_train.to_csv('data/total_balanced_train.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}