analyse_dataset.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Analyzing stories:   4%|▍         | 32665/781382 [00:08<03:21, 3718.69it/s]\n"
     ]
    },
    {
     "ename": "KeyboardInterrupt",
     "evalue": "",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mKeyboardInterrupt\u001b[0m                         Traceback (most recent call last)",
      "Cell \u001b[1;32mIn[9], line 107\u001b[0m\n\u001b[0;32m    105\u001b[0m file_path \u001b[38;5;241m=\u001b[39m \u001b[38;5;124mr\u001b[39m\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mdata\u001b[39m\u001b[38;5;124m\\\u001b[39m\u001b[38;5;124mbatches_2024-09-03-11-24-05\u001b[39m\u001b[38;5;124m\\\u001b[39m\u001b[38;5;124mprocessed.jsonl\u001b[39m\u001b[38;5;124m'\u001b[39m\n\u001b[0;32m    106\u001b[0m stories \u001b[38;5;241m=\u001b[39m read_jsonl_file(file_path)\n\u001b[1;32m--> 107\u001b[0m metrics \u001b[38;5;241m=\u001b[39m \u001b[43mcompute_metrics\u001b[49m\u001b[43m(\u001b[49m\u001b[43mstories\u001b[49m\u001b[43m)\u001b[49m\n\u001b[0;32m    108\u001b[0m plot_word_distribution(metrics)\n\u001b[0;32m    109\u001b[0m plot_story_length_histogram(metrics)\n",
      "Cell \u001b[1;32mIn[9], line 48\u001b[0m, in \u001b[0;36mcompute_metrics\u001b[1;34m(stories)\u001b[0m\n\u001b[0;32m     46\u001b[0m all_metrics \u001b[38;5;241m=\u001b[39m []\n\u001b[0;32m     47\u001b[0m \u001b[38;5;28;01mfor\u001b[39;00m story_data \u001b[38;5;129;01min\u001b[39;00m tqdm(stories, \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mAnalyzing stories\u001b[39m\u001b[38;5;124m\"\u001b[39m):\n\u001b[1;32m---> 48\u001b[0m     story_metrics \u001b[38;5;241m=\u001b[39m \u001b[43manalyze_story\u001b[49m\u001b[43m(\u001b[49m\u001b[43mstory_data\u001b[49m\u001b[43m[\u001b[49m\u001b[38;5;124;43m'\u001b[39;49m\u001b[38;5;124;43mstory\u001b[39;49m\u001b[38;5;124;43m'\u001b[39;49m\u001b[43m]\u001b[49m\u001b[43m)\u001b[49m\n\u001b[0;32m     49\u001b[0m     story_metrics[\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mid\u001b[39m\u001b[38;5;124m'\u001b[39m] \u001b[38;5;241m=\u001b[39m story_data[\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mgeneration_id\u001b[39m\u001b[38;5;124m'\u001b[39m]\n\u001b[0;32m     50\u001b[0m     story_metrics[\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mexpected_num_paragraphs\u001b[39m\u001b[38;5;124m'\u001b[39m] \u001b[38;5;241m=\u001b[39m story_data[\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mnum_paragraphs\u001b[39m\u001b[38;5;124m'\u001b[39m]\n",
      "Cell \u001b[1;32mIn[9], line 32\u001b[0m, in \u001b[0;36manalyze_story\u001b[1;34m(story)\u001b[0m\n\u001b[0;32m     29\u001b[0m \u001b[38;5;66;03m# Calculate word frequency distribution\u001b[39;00m\n\u001b[0;32m     30\u001b[0m word_freq \u001b[38;5;241m=\u001b[39m Counter(words)\n\u001b[1;32m---> 32\u001b[0m reading_level \u001b[38;5;241m=\u001b[39m \u001b[43mflesch_kincaid_grade\u001b[49m\u001b[43m(\u001b[49m\u001b[43mstory\u001b[49m\u001b[43m)\u001b[49m\n\u001b[0;32m     34\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m {\n\u001b[0;32m     35\u001b[0m     \u001b[38;5;124m'\u001b[39m\u001b[38;5;124mnum_characters\u001b[39m\u001b[38;5;124m'\u001b[39m: num_characters,\n\u001b[0;32m     36\u001b[0m     \u001b[38;5;124m'\u001b[39m\u001b[38;5;124mnum_words\u001b[39m\u001b[38;5;124m'\u001b[39m: num_words,\n\u001b[1;32m   (...)\u001b[0m\n\u001b[0;32m     41\u001b[0m     \u001b[38;5;124m'\u001b[39m\u001b[38;5;124mreading_level\u001b[39m\u001b[38;5;124m'\u001b[39m: reading_level\n\u001b[0;32m     42\u001b[0m }\n",
      "File \u001b[1;32mc:\\Users\\PC\\AppData\\Local\\Programs\\Python\\Python312\\Lib\\site-packages\\textstat\\textstat.py:718\u001b[0m, in \u001b[0;36mtextstatistics.flesch_kincaid_grade\u001b[1;34m(self, text)\u001b[0m\n\u001b[0;32m    695\u001b[0m \u001b[38;5;129m@lru_cache\u001b[39m(maxsize\u001b[38;5;241m=\u001b[39m\u001b[38;5;241m128\u001b[39m)\n\u001b[0;32m    696\u001b[0m \u001b[38;5;28;01mdef\u001b[39;00m \u001b[38;5;21mflesch_kincaid_grade\u001b[39m(\u001b[38;5;28mself\u001b[39m, text: \u001b[38;5;28mstr\u001b[39m) \u001b[38;5;241m-\u001b[39m\u001b[38;5;241m>\u001b[39m \u001b[38;5;28mfloat\u001b[39m:\n\u001b[0;32m    697\u001b[0m \u001b[38;5;250m    \u001b[39m\u001b[38;5;124mr\u001b[39m\u001b[38;5;124;03m\"\"\"Calculate the Flesh-Kincaid Grade for `text`.\u001b[39;00m\n\u001b[0;32m    698\u001b[0m \n\u001b[0;32m    699\u001b[0m \u001b[38;5;124;03m    Parameters\u001b[39;00m\n\u001b[1;32m   (...)\u001b[0m\n\u001b[0;32m    716\u001b[0m \n\u001b[0;32m    717\u001b[0m \u001b[38;5;124;03m    \"\"\"\u001b[39;00m\n\u001b[1;32m--> 718\u001b[0m     sentence_length \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mavg_sentence_length\u001b[49m\u001b[43m(\u001b[49m\u001b[43mtext\u001b[49m\u001b[43m)\u001b[49m\n\u001b[0;32m    719\u001b[0m     syllables_per_word \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mavg_syllables_per_word(text)\n\u001b[0;32m    720\u001b[0m     flesch \u001b[38;5;241m=\u001b[39m (\n\u001b[0;32m    721\u001b[0m             \u001b[38;5;28mfloat\u001b[39m(\u001b[38;5;241m0.39\u001b[39m \u001b[38;5;241m*\u001b[39m sentence_length)\n\u001b[0;32m    722\u001b[0m             \u001b[38;5;241m+\u001b[39m \u001b[38;5;28mfloat\u001b[39m(\u001b[38;5;241m11.8\u001b[39m \u001b[38;5;241m*\u001b[39m syllables_per_word)\n\u001b[0;32m    723\u001b[0m             \u001b[38;5;241m-\u001b[39m \u001b[38;5;241m15.59\u001b[39m)\n",
      "File \u001b[1;32mc:\\Users\\PC\\AppData\\Local\\Programs\\Python\\Python312\\Lib\\site-packages\\textstat\\textstat.py:406\u001b[0m, in \u001b[0;36mtextstatistics.avg_sentence_length\u001b[1;34m(self, text)\u001b[0m\n\u001b[0;32m    389\u001b[0m \u001b[38;5;250m\u001b[39m\u001b[38;5;124;03m\"\"\"Calculate the average sentence length.\u001b[39;00m\n\u001b[0;32m    390\u001b[0m \n\u001b[0;32m    391\u001b[0m \u001b[38;5;124;03mThis function is a combination of the functions `lexicon_count` and\u001b[39;00m\n\u001b[1;32m   (...)\u001b[0m\n\u001b[0;32m    403\u001b[0m \n\u001b[0;32m    404\u001b[0m \u001b[38;5;124;03m\"\"\"\u001b[39;00m\n\u001b[0;32m    405\u001b[0m \u001b[38;5;28;01mtry\u001b[39;00m:\n\u001b[1;32m--> 406\u001b[0m     asl \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mfloat\u001b[39m(\u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mlexicon_count\u001b[49m\u001b[43m(\u001b[49m\u001b[43mtext\u001b[49m\u001b[43m)\u001b[49m \u001b[38;5;241m/\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39msentence_count(text))\n\u001b[0;32m    407\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_legacy_round(asl, \u001b[38;5;241m1\u001b[39m)\n\u001b[0;32m    408\u001b[0m \u001b[38;5;28;01mexcept\u001b[39;00m \u001b[38;5;167;01mZeroDivisionError\u001b[39;00m:\n",
      "File \u001b[1;32mc:\\Users\\PC\\AppData\\Local\\Programs\\Python\\Python312\\Lib\\site-packages\\textstat\\textstat.py:301\u001b[0m, in \u001b[0;36mtextstatistics.lexicon_count\u001b[1;34m(self, text, removepunct)\u001b[0m\n\u001b[0;32m    279\u001b[0m \u001b[38;5;250m\u001b[39m\u001b[38;5;124;03m\"\"\"Count types (words) in a text.\u001b[39;00m\n\u001b[0;32m    280\u001b[0m \n\u001b[0;32m    281\u001b[0m \u001b[38;5;124;03mIf `removepunct` is set to True and\u001b[39;00m\n\u001b[1;32m   (...)\u001b[0m\n\u001b[0;32m    298\u001b[0m \n\u001b[0;32m    299\u001b[0m \u001b[38;5;124;03m\"\"\"\u001b[39;00m\n\u001b[0;32m    300\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m removepunct:\n\u001b[1;32m--> 301\u001b[0m     text \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mremove_punctuation\u001b[49m\u001b[43m(\u001b[49m\u001b[43mtext\u001b[49m\u001b[43m)\u001b[49m\n\u001b[0;32m    302\u001b[0m count \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mlen\u001b[39m(text\u001b[38;5;241m.\u001b[39msplit())\n\u001b[0;32m    303\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m count\n",
      "\u001b[1;31mKeyboardInterrupt\u001b[0m: "
     ]
    }
   ],
   "source": [
    "import json\n",
    "import re\n",
    "from collections import Counter\n",
    "from statistics import mean\n",
    "from textstat import flesch_kincaid_grade\n",
    "from tqdm import tqdm\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "def read_jsonl_file(file_path):\n",
    "    with open(file_path, 'r') as file:\n",
    "        stories = [json.loads(line) for line in file]\n",
    "    return stories\n",
    "\n",
    "def analyze_story(story):\n",
    "    num_characters = len(story)\n",
    "    \n",
    "    words = re.findall(r'\\b\\w+\\b', story)\n",
    "    num_words = len(words)\n",
    "    \n",
    "    sentences = re.split(r'[.!?]', story)\n",
    "    num_sentences = len([s for s in sentences if s.strip()])\n",
    "    \n",
    "    paragraphs = story.split('\\n\\n')\n",
    "    num_paragraphs = len(paragraphs)\n",
    "    \n",
    "    # Calculate average paragraph length (in words)\n",
    "    avg_paragraph_length = num_words / num_paragraphs if num_paragraphs > 0 else 0\n",
    "    \n",
    "    # Calculate word frequency distribution\n",
    "    word_freq = Counter(words)\n",
    "    \n",
    "    reading_level = flesch_kincaid_grade(story)\n",
    "    \n",
    "    return {\n",
    "        'num_characters': num_characters,\n",
    "        'num_words': num_words,\n",
    "        'num_sentences': num_sentences,\n",
    "        'num_paragraphs': num_paragraphs,\n",
    "        'avg_paragraph_length': avg_paragraph_length,\n",
    "        'word_freq': word_freq,\n",
    "        'reading_level': reading_level\n",
    "    }\n",
    "\n",
    "# Compute Metrics for All Stories and Summarize\n",
    "def compute_metrics(stories):\n",
    "    all_metrics = []\n",
    "    for story_data in tqdm(stories, \"Analyzing stories\"):\n",
    "        story_metrics = analyze_story(story_data['story'])\n",
    "        story_metrics['id'] = story_data['generation_id']\n",
    "        story_metrics['expected_num_paragraphs'] = story_data['num_paragraphs']\n",
    "        all_metrics.append(story_metrics)\n",
    "    \n",
    "    # Calculate overall averages\n",
    "    avg_num_words = mean([m['num_words'] for m in all_metrics])\n",
    "    avg_reading_level = mean([m['reading_level'] for m in all_metrics])\n",
    "    avg_num_sentences = mean([m['num_sentences'] for m in all_metrics])\n",
    "    avg_num_characters = mean([m['num_characters'] for m in all_metrics])\n",
    "\n",
    "    print(f\"Average number of words: {avg_num_words}\")\n",
    "    print(f\"Average reading level: {avg_reading_level}\")\n",
    "    print(f\"Average number of sentences: {avg_num_sentences}\")\n",
    "    print(f\"Average number of characters: {avg_num_characters}\")\n",
    "\n",
    "    \n",
    "    return all_metrics\n",
    "\n",
    "# Visualize Word Distribution\n",
    "def plot_word_distribution(metrics):\n",
    "    all_words = Counter()\n",
    "    for m in metrics:\n",
    "        all_words.update(m['word_freq'])\n",
    "    \n",
    "    topk = 42\n",
    "    most_common_words = all_words.most_common(topk)\n",
    "    words, frequencies = zip(*most_common_words)\n",
    "    \n",
    "    plt.figure(figsize=(16, 4))\n",
    "    plt.bar(words, frequencies)\n",
    "    plt.xticks(rotation=45)\n",
    "    plt.title(f\"Top {topk} Most Common Words\")\n",
    "    plt.show()\n",
    "\n",
    "# Plot Histogram of Story Length and Paragraphs vs Story Length\n",
    "def plot_story_length_histogram(metrics):\n",
    "    story_lengths = [m['num_words'] for m in metrics]\n",
    "    \n",
    "    plt.figure(figsize=(10, 6))\n",
    "    plt.hist(story_lengths, bins=10, edgecolor='black')\n",
    "    plt.title(\"Histogram of Story Lengths (in Words)\")\n",
    "    plt.xlabel(\"Story Length (words)\")\n",
    "    plt.ylabel(\"Frequency\")\n",
    "    plt.show()\n",
    "\n",
    "def plot_paragraphs_vs_story_length(metrics):\n",
    "    num_paragraphs = [m['expected_num_paragraphs'] for m in metrics]\n",
    "    story_lengths = [m['num_words'] for m in metrics]\n",
    "    \n",
    "    plt.figure(figsize=(10, 6))\n",
    "    plt.scatter(story_lengths, num_paragraphs)\n",
    "    plt.title(\"Number of Paragraphs vs. Story Length\")\n",
    "    plt.xlabel(\"Story Length (words)\")\n",
    "    plt.ylabel(\"Expected Number of Paragraphs\")\n",
    "    plt.show()\n",
    "\n",
    "file_path = r'data\\batches_2024-09-03-11-24-05\\processed.jsonl'\n",
    "stories = read_jsonl_file(file_path)\n",
    "metrics = compute_metrics(stories)\n",
    "plot_word_distribution(metrics)\n",
    "plot_story_length_histogram(metrics)\n",
    "plot_paragraphs_vs_story_length(metrics)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "600000it [00:14, 42222.17it/s]\n",
      "65000it [00:01, 37804.50it/s]\n",
      "c:\\Users\\PC\\AppData\\Local\\Programs\\Python\\Python312\\Lib\\site-packages\\numpy\\core\\fromnumeric.py:3504: RuntimeWarning: Mean of empty slice.\n",
      "  return _methods._mean(a, axis=axis, dtype=dtype,\n",
      "c:\\Users\\PC\\AppData\\Local\\Programs\\Python\\Python312\\Lib\\site-packages\\numpy\\core\\_methods.py:129: RuntimeWarning: invalid value encountered in scalar divide\n",
      "  ret = ret.dtype.type(ret / rcount)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fraction of stories with [UNK] tokens: 0.0\n",
      "Average number of [UNK] tokens: nan\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 600x400 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAjMAAAGJCAYAAAB/67JMAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjguNCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8fJSN1AAAACXBIWXMAAA9hAAAPYQGoP6dpAABwM0lEQVR4nO3deVxU1fsH8M8MMOwDCLLJKuCG+44bmiiWuZSVqRVuaYZfJUvLMrcyzSyXNM0W9dfXpSy1csd9Q9xAXBBRQU3ZFNkFBub8/iDu15FBQQeHgc/79eJVc+6Zc5/7MMLDuefeKxNCCBAREREZKLm+AyAiIiJ6GixmiIiIyKCxmCEiIiKDxmKGiIiIDBqLGSIiIjJoLGaIiIjIoLGYISIiIoPGYoaIiIgMGosZIiIiMmgsZoj+NXz4cHh5eek7DIO0evVqyGQyJCYmPvVY3bt3R/fu3Z96nIqQyWQYP378M9nXo/Czp1teXl4YPny4vsOgZ4jFDBmUmTNnQiaT4c6dO1q3N23a9Jn9InzW5s+fD5lMhqioKI12IQTs7Owgk8mQkJCgsS0/Px+mpqYYOnToswy1Wjl27BhmzpyJjIwMvcXg5eUFmUym9Ss/P1/n+6vsMQ8fPhwymQzNmzeHtifcVLToKywsxOLFi9GqVSsolUrY2trC398fY8aMwaVLl544PqLHMdZ3AETVxQ8//AC1Wq3vMMrVpUsXAMCRI0fQqlUrqf3ChQvIyMiAsbExjh49Cm9vb2nbyZMnUVhYKL3XEOzevVun4x07dgyzZs3C8OHDYWtrq9OxK6Nly5Z4//33y7QrFAqdf/ae9JjPnTuHTZs2YdCgQU+030GDBmHHjh0YMmQI3n77bahUKly6dAlbt25Fp06d0KhRo6eKr6Li4uIgl/Nv9dqExQzRv0xMTPQdwiO1bdsWZmZmOHLkCP7zn/9I7UePHoW9vT3atm2LI0eO4I033pC2HTlyBACeuphRq9UoLCyEmZnZU41TEQqFosr3oQ/16tXT+N48qCK/eIuKiqBWq6ssP+bm5nB3d8fs2bPx8ssvQyaTVer9J0+exNatWzFnzhx8/PHHGtuWLl1a5bMwQgjk5+fD3NwcpqamVbovqn5YulKNduDAAchkMvz222+YM2cO3NzcYGZmhp49e+LKlSsafbWtW8jIyMDw4cNhY2MDW1tbhISEIDo6GjKZDKtXr5b6lbfOQ9uYarUaixYtgr+/P8zMzODk5ISxY8fi3r17jzwWhUKBdu3a4ejRoxrtR48eRUBAADp37qx1m62tLZo2bQoAyM3Nxfvvvw93d3eYmpqiYcOGWLBgQZlTC6WnFdauXQt/f3+Ymppi586dAEpmgp577jmYm5vDzc0Nn3/+udZZhVOnTiE4OBgODg4wNzeHt7c3Ro4c+chjBMrmsjLfw4fNnDkTkydPBgB4e3tLp3YeXtuzZcsWNG3aFKampvD395eO9UG3bt3CyJEj4eTkJPX7+eefH3s8FfHw5yQxMREymQwLFizAokWL4OPjA1NTU1y8eBEA8O2338Lf3x8WFhaws7ND27ZtsW7dukod88PkcjmmTZuGmJgYbN68udLHcPXqVQBA586dy2wzMjKCvb19heIrKirCZ599Jh2zl5cXPv74YxQUFGiM6eXlhRdffBG7du1C27ZtYW5uju+//17a9vCamYyMDISFhUmffV9fX3z55ZdlPrsbNmxAmzZtYG1tDaVSiWbNmmHx4sWVzgc9W5yZoVph3rx5kMvl+OCDD5CZmYn58+dj2LBhiIyMLPc9QggMGDAAR44cwTvvvIPGjRtj8+bNCAkJeapYxo4di9WrV2PEiBGYMGECEhISsHTpUkRFReHo0aOPnCHq0qULDh8+jMTEROmX39GjRzF69Gi0b98eM2bMQEZGBmxtbSGEwLFjxxAQEAC5XA4hBPr374/9+/dj1KhRaNmyJXbt2oXJkyfj1q1bWLhwoca+9u3bh99++w3jx4+Hg4MDvLy8kJycjB49eqCoqAgfffQRLC0tsXLlSpibm2u8NzU1Fb1790bdunXx0UcfwdbWFomJidi0adMT5+1Jvocvv/wyLl++jPXr12PhwoVwcHAAANStW1fqc+TIEWzatAnvvvsurK2tsWTJEgwaNAg3btyQfgGnpKSgY8eOUpFXt25d7NixA6NGjUJWVhbCwsIeG79KpSqz1svCwgIWFhblvmfVqlXIz8/HmDFjYGpqijp16uCHH37AhAkT8Morr2DixInIz89HTEwMIiMjMXTo0Aodc3mGDh2Kzz77DLNnz8ZLL71UqdkZT09PAMDatWvRuXNnGBtr//XyuPhGjx6NNWvW4JVXXsH777+PyMhIzJ07F7GxsWWKrLi4OAwZMgRjx47F22+/jYYNG2rdZ15eHgIDA3Hr1i2MHTsWHh4eOHbsGKZOnYqkpCQsWrQIABAeHo4hQ4agZ8+e+PLLLwEAsbGxOHr0KCZOnFjhXJAeCCIDMmPGDAFApKWlad3u7+8vAgMDpdf79+8XAETjxo1FQUGB1L548WIBQJw7d05qCwkJEZ6entLrLVu2CABi/vz5UltRUZHo2rWrACBWrVoltQcGBmrst7wxDx8+LACItWvXavTbuXOn1vaHbdu2TQAQv/zyixBCiKSkJAFAHDx4UGRnZwsjIyOxbds2IYQQ58+fFwDEnDlzNI7n888/1xjzlVdeETKZTFy5ckVqAyDkcrm4cOGCRt+wsDABQERGRkptqampwsbGRgAQCQkJQgghNm/eLACIkydPPvJ4tHk4l5X5Hmrz1VdfacT2IABCoVBoHPvZs2cFAPHtt99KbaNGjRIuLi7izp07Gu9//fXXhY2NjcjLy3tkDJ6engJAma8ZM2YIIcp+ThISEgQAoVQqRWpqqsZYAwYMEP7+/k98zNqEhIQIS0tLIYQQa9asEQDEpk2bpO0ARGho6CPHUKvVIjAwUAAQTk5OYsiQIWLZsmXi+vXrFY4vOjpaABCjR4/WaP/ggw8EALFv3z6prTSnO3fuLDO+p6enCAkJkV5/9tlnwtLSUly+fFmj30cffSSMjIzEjRs3hBBCTJw4USiVSlFUVPTIY6Xqh6eZqFYYMWKExlqDrl27AgCuXbtW7nu2b98OY2NjjBs3TmozMjLSWK9SWRs3boSNjQ169eqFO3fuSF9t2rSBlZUV9u/f/8j3d+rUCXK5XFoLUzqT065dO1hZWaF58+bSqabS/5aul9m+fTuMjIwwYcIEjTHff/99CCGwY8cOjfbAwEA0adJEo2379u3o2LEj2rdvL7XVrVsXw4YN0+hXuqhz69atUKlUFUnNYz3J97AigoKC4OPjI71u3rw5lEqlNK4QAn/88Qf69esHIYTG9y04OBiZmZk4c+bMY/fToUMHhIeHa3y99dZbj3zPoEGDysyo2Nra4p9//sHJkyef4Ggfb9iwYfDz88Ps2bO1XtlUHplMhl27duHzzz+HnZ0d1q9fj9DQUHh6emLw4MEVWjOzfft2AMCkSZM02ksXTm/btk2j3dvbG8HBwY8dd+PGjejatSvs7Ow0vn9BQUEoLi7GoUOHAJTkNjc3F+Hh4RU5ZKpGWMxQjaNtatzDw0PjtZ2dHQA8cp3K9evX4eLiAisrK4328qayKyI+Ph6ZmZlwdHRE3bp1Nb5ycnKQmpr6yPeXXur6YMHSqlUr6TRPp06dNLYpFAqp8Lh+/TpcXV1hbW2tMWbjxo2l7Q968KqoUtevX4efn1+Z9odzEhgYiEGDBmHWrFlwcHDAgAEDsGrVqjLrHirjSb6HTzJu6dil46alpSEjIwMrV64s8z0bMWIEADz2+wYADg4OCAoK0viqX7/+I9+j7Xvw4YcfwsrKCu3bt4efnx9CQ0PLrJV6GkZGRpg2bRqio6OxZcuWSr3X1NQUn3zyCWJjY3H79m2sX78eHTt2lE5XPs7169chl8vh6+ur0e7s7AxbW9sKfUa1iY+Px86dO8t8/4KCggD87/v37rvvokGDBnj++efh5uaGkSNHal0/RdUP18yQQSm9mub+/ftat+fl5Wm94sbIyEhr/8r85fkoMplM61jFxcUar9VqNRwdHbF27Vqt41RkXUOXLl2wYsUKZGRk4OjRo+jUqZO0rVOnTvj555+hUqlw5MgRtGnT5omvQHp4HUxlyGQy/P777zh+/Dj+/vtv7Nq1CyNHjsTXX3+N48ePlykQK6KqvoePG7d0gegbb7xR7nqp5s2bP1UM5dH2PWjcuDHi4uKwdetW7Ny5E3/88Qe+++47TJ8+HbNmzdLJfocNGyatnRk4cOATjeHi4oLXX38dgwYNgr+/P3777TesXr263LU0D6roWp2KfkbVajV69eqFKVOmaN3eoEEDAICjoyOio6Oxa9cu7NixAzt27MCqVavw1ltvYc2aNRXaF+kHixkyKKWLDOPi4uDu7q6xLS8vDzdv3kTv3r11tq+9e/ciJydH45dvXFxcmb52dnZaT3c8/Jekj48P9uzZg86dOz9xsdClSxcsX74ce/bsQVRUlHRlCFBSzNy/fx/btm3DtWvXNO4X4unpiT179iA7O1tjdqb0ZmaluX0UT09PxMfHl2nXlhMA6NixIzp27Ig5c+Zg3bp1GDZsGDZs2IDRo0dX+HifVmUvMX5Y3bp1YW1tjeLiYukveX2ztLTE4MGDMXjwYBQWFuLll1/GnDlzMHXqVJiZmT31MZfOzgwfPhx//vnnU41lYmKC5s2bIz4+Hnfu3IGzs3O58Xl6ekKtViM+Pl6aMQRKFmBnZGRU6DOqjY+PD3Jycir0/VMoFOjXrx/69esHtVqNd999F99//z0+/fTTMjNGVH3wNBMZlJ49e0KhUGD58uVlLqlcuXIlioqK8Pzzz+tkXy+88AKKioqwfPlyqa24uBjffvttmb4+Pj64dOkS0tLSpLazZ8+Wmf5/7bXXUFxcjM8++6zMGEVFRRVaV1C6Buabb76BSqXSmJnx8vKCi4sL5s+fr9G39HiKi4uxdOlSjfEWLlwImUxWoby98MILOH78OE6cOCG1paWllZlpunfvXpkZk5YtWwLAU51qehKWlpYA8MT3OTEyMsKgQYPwxx9/4Pz582W2P/g9fxbu3r2r8VqhUKBJkyYQQkjrk572mIGSmShfX98Kz/bEx8fjxo0bZdozMjIQEREBOzs7aeaxvPheeOEFAJCuLir1zTffAAD69u1bmUOQvPbaa4iIiMCuXbu0xldUVASgbG7lcrk06/asP7dUOZyZIYPi6OiI6dOnY9q0aejWrRv69+8PCwsLHDt2DOvXr0fv3r3Rr18/neyrX79+6Ny5Mz766CMkJiaiSZMm2LRpEzIzM8v0HTlyJL755hsEBwdj1KhRSE1NxYoVK+Dv74+srCypX2BgIMaOHYu5c+ciOjoavXv3homJCeLj47Fx40YsXrwYr7zyyiPj8vDwgLu7OyIiIuDl5QVXV1eN7Z06dcIff/wBmUymcc+Pfv36oUePHvjkk0+QmJiIFi1aYPfu3fjzzz8RFhamsQi2PFOmTMEvv/yCPn36YOLEidKl2Z6enoiJiZH6rVmzBt999x1eeukl+Pj4IDs7Gz/88AOUSqX0C+tZadOmDQDgk08+weuvvw4TExP069dP+oVaEfPmzcP+/fvRoUMHvP3222jSpAnS09Nx5swZ7NmzB+np6VUVfhm9e/eGs7MzOnfuDCcnJ8TGxmLp0qXo27evNOOmi2M2MjLCJ598Iq0LepyzZ89i6NCheP7559G1a1fUqVMHt27dwpo1a3D79m0sWrRIOqVXXnwtWrRASEgIVq5ciYyMDAQGBuLEiRNYs2YNBg4ciB49elQmVZLJkyfjr7/+wosvvojhw4ejTZs2yM3Nxblz5/D7778jMTERDg4OGD16NNLT0/Hcc8/Bzc0N169fx7fffouWLVtqzBRRNaSvy6iInsZ///tf0bFjR2FpaSlMTU1Fo0aNxKxZs0R+fr5Gv9LLejdu3KjRXnrp64OXVz98eawQQty9e1e8+eabQqlUChsbG/Hmm2+KqKioMu8tjal+/fpCoVCIli1bil27dmkdUwghVq5cKdq0aSPMzc2FtbW1aNasmZgyZYq4fft2hY5/yJAhAoAYOnRomW3ffPONdCnzw7Kzs8V7770nXF1dhYmJifDz8xNfffWVUKvVGv3wiEtxY2JiRGBgoDAzMxP16tUTn332mfjpp580LrU9c+aMGDJkiPDw8BCmpqbC0dFRvPjii+LUqVOPPbbyLs2uyPewPJ999pmoV6+ekMvlGnGWd5wPX9orhBApKSkiNDRUuLu7CxMTE+Hs7Cx69uwpVq5c+dj9e3p6ir59+5a7vbxLs7/66qsyfb///nvRrVs3YW9vL0xNTYWPj4+YPHmyyMzMrNAxl7f/0kuzH6RSqYSPj0+FLs1OSUkR8+bNE4GBgcLFxUUYGxsLOzs78dxzz4nff/+9TP/y4lOpVGLWrFnC29tbmJiYCHd3dzF16tQy/7YflVNt37/s7GwxdepU4evrKxQKhXBwcBCdOnUSCxYsEIWFhUIIIX7//XfRu3dv4ejoKBQKhfDw8BBjx44VSUlJjzx20j+ZEDpaAUlUSyQmJsLb2xurVq3ik3mJiKoBrpkhIiIig8ZihoiIiAwaixkiIiIyaFwzQ0RERAaNMzNERERk0FjMEBERkUHjTfN0RK1W4/bt27C2tn7qW4kTERHVJkIIZGdnw9XVFXJ55edZWMzoyO3bt8s8K4iIiIgq7ubNm3Bzc6v0+1jM6EjpbcRv3rwJpVKpkzFVKhV2794t3fKe/oe50Y55KR9zox3zUj7mRruqyEtWVhbc3d01HoJbGSxmdKT01JJSqdRpMWNhYQGlUsl/SA9hbrRjXsrH3GjHvJSPudGuKvPypMs0uACYiIiIDBqLGSIiIjJoLGaIiIjIoHHNzDMkhEBRURGKi4sr1F+lUsHY2Bj5+fkVfk9tURtzY2JiAiMjI32HQURU7bCYeUYKCwuRlJSEvLy8Cr9HCAFnZ2fcvHmT9655SG3MjUwmg5ubG6ysrPQdChFRtcJi5hlQq9VISEiAkZERXF1doVAoKvQLWK1WIycnB1ZWVk90E6GarLblRgiBtLQ0/PPPP/Dz8+MMDRHRA1jMPAOFhYVQq9Vwd3eHhYVFhd+nVqtRWFgIMzOzWvELuzJqY27q1q2LxMREqFQqFjNERA+oHb8Fqona8kuXqkZtOZ1GRFRZ/O1KREREBo3FDBERET2WWi2QkpWPyIR0HE+tXjPFXDNDz4yXlxfCwsIQFham71D0aubMmdiyZQuio6P1HQoRkVb5qmJcTMrC5eRsxCZlIepmBmKTsqAqFgAAuUyOyfkq1Kkmj3lgMUOP1L17d7Rs2RKLFi3SaF+9ejXCwsKQkZFR4bFOnjwJS0tL3QZYSZcuXULjxo0RERGBjh07Su0dO3ZEdHQ0MjIyYGZmBgDIz8+Hra0tli1bhlGjRukrZCKiKlVQVIyYfzJx9mYGLtzOQsw/GUi4kwu1KNvXSC6Di40ZHOS5yMovQp0ney6kzrGYoWembt26+g4BjRo1grOzMw4cOCAVM9nZ2Thz5gycnJxw/PhxdO/eHQAQERGBgoICPPfcc0+0L5VKxYfTEVG1czenALFJ2Th/OxPHr93FqcR7yCkoKtPP3lIBPycrNHW1QRNXJdp61oGLrRmgLsb27dtRz9ZcD9FrxzUzeiKEQF5h0WO/7hcWV6hfRb+E0FJq68Dw4cMxcOBALFiwAC4uLrC3t0doaChUKpXUx8vLS2OGJz4+Ht26dYOZmRmaNGmC8PBwyGQybNmyBQBw4MAByGQyjdmf6OhoyGQyJCYmSm1HjhxB165dYW5uDnd3d0yYMAG5ubnlxtqjRw8cOHBA4/0NGjRAv379NNoPHDgAT09PeHt7AwCWL18OHx8fKBQKNGzYEL/88ovGuDKZDMuXL0f//v1haWmJOXPmAADmzZsHJycnWFtbY9SoUcjPz9d434EDB9C+fXtYWlrC1tYWnTt3xvXr1x+VbiKiCsktKMLh+DR8uzceIT+fQJvPwtHm8z1446dIzNtxCQfi0pBTUAR7SwV6NXHCe0ENsGp4O0R+3BOnpgVhw5gATHuxCV5u7QYPewuYGFXPsoEzM3pyX1WMJtN3PfP9XpwdDAtF1Xzb9+/fDxcXF+zfvx9XrlzB4MGD0bJlS7z99ttl+qrVarz88stwcnJCZGQkMjMzn2gtzdWrV9GnTx98/vnn+Pnnn5GWlobx48dj/PjxWLVqldb39OjRA++99x6KiopgbGyM/fv3o3v37ujatSu+++47zJw5UzqeHj16AAA2b96MiRMnYtGiRQgKCsLWrVsxYsQIuLm5SX2AkvUw8+bNw6JFi2BsbIzffvsNM2fOxLJly9ClSxf88ssvWLJkCerXrw8AKCoqwsCBA/H2229j/fr1KCwsxIkTJ3gZNhFVmhACCXdycer6PVy4lYmjV+/iWlqO1tNFXvYWaOhsjZbudujsa4+mrjaQyw335w6LGdIZOzs7LF26FEZGRmjUqBH69u2LvXv3ai1m9uzZg0uXLmHXrl1wdXUFAHzxxRd4/vnnK7XPefPmYdiwYVIh5OfnhyVLliAwMBDLly+X1r88qEePHsjNzcXJkycREBCAAwcOYPLkyejSpQtCQkKQn58PIQROnDiB0aNHAwAWLFiA4cOH49133wUATJo0CcePH8eCBQs0ipmhQ4dixIgR0uvXX38do0aNktbcfP7559izZ480O5OVlYXMzEy8+OKL8PHxAQA0bty4UjkgotqpqFiNS8nZiLqZgSPxaThzIwNp2QVl+tWzNUcbTzs0q2eD9t514OdkVWV/1OpLzToaA2JuYoSLs4Mf2UetViM7KxvWSmud3XDP3KTq7hzr7++vcWdaFxcXnDt3Tmvf2NhYuLu7S4UMAAQEBFR6nzExMYiJicHatWulNiGE9AgJbYWBr68v3NzccODAAfj7+yMqKgqBgYFwdHSEh4cHIiIiIIRAQUGBVKjExsZizJgxGuN07twZixcv1mhr27ZtmeN85513NNoCAgKwf/9+AECdOnUwfPhwBAcHo1evXggKCsJrr70GFxeXSueCiGq2jLxCnLlxD7FJ2Thz/R5OJqYjK19zrYvCWI7m9WzQ0t0Wzd1t0dG7Dupam9b42V4WM3oik8keWxmr1WoUKYxgoTDW292DlUolMjMzy7RnZGTAxsZGo+3hxa4ymQxqtfqJ9116zA+u83lwDQ4A5OTkYOzYsZgwYUKZ93t4eJQ7dvfu3bF//340b94cfn5+cHR0BAAEBgZi//79EELA19cX7u7ulYr5Sa7WWrVqFSZMmICdO3fi119/xbRp0xAeHq5xtRUR1R5FxWok3s1F1I0MXE3LRVxyFuJTc/DPvftl+irNjNHEVYnOPg5o510HLd1tYVaFf7RWVyxm6JEaNmyI3bt3l2k/c+YMGjRo8MTjNm7cGDdv3kRSUpI0C3H8+HGNPqVXPyUlJcHOzg4AytybpVWrVrh48SJ8fX0rtf8ePXpgwoQJaNKkiXT1EgB069YNP/zwA4QQGqePGjdujKNHjyIkJERqO3r0KJo0afLY44yMjMRbb70ltT18nKXH0apVK0ydOhUBAQFYt24dixmiWiKvsAjn/slEZEI6IhPuIupGBvIKi7X29alriUYuSrR0s0VrTzu0cLOBcTVdlPsssZihRxo3bhyWLl2KCRMmYPTo0TA1NcW2bduwfv16/P333088blBQEBo0aICQkBB89dVXyMrKwieffKLRp3RmZObMmZgzZw4uX76Mr7/+WqPPlClT0KlTJ4wfPx6jR4+GpaUlLl68iPDwcCxdurTc/Zeum/n555/xww8/SO2BgYHSOpnS9TEAMHnyZLz22mto1aoVgoKC8Pfff2PTpk3Ys2fPI49z4sSJGD58ONq2bYvOnTtj7dq1uHDhgrQAOCEhAStXrkT//v3h6uqKuLg4xMfHaxQ/RFRzqNUCV9NycDwhHWeu38OZG/dw/W5emX7mJkZoVs8GDZ2t0cDZGn6OVvBztIK9lakeoq7+WMzQI9WvXx+HDh3CJ598gqCgIBQWFqJRo0bYuHEj+vTp88TjyuVybN68GaNGjUL79u3h5eWFJUuWaIxpYmKC9evXY9y4cWjevDnatWuHzz//HK+++qrUp3nz5jh48CA++eQTdO3aFUII+Pj4YPDgwY/cv7e3Nzw9PXH9+nUEBgZK7R4eHnB1dUViYqLGjM3AgQOxePFiLFiwABMnToS3tzdWrVql0UebwYMH4+rVq5gyZQry8/MxaNAgjBs3Drt2lVzJZmFhgUuXLmHNmjW4e/cuXFxcEBoairFjx1Yim0RUXWXmqXDmxj2c/ScDZ29mIPpmBu7lqcr0c7BSoI2nHQLq2yPAxwG+jlYwMuCri541maiqG4/UMllZWbCxsUFmZiaUSqXGtvz8fCQkJMDb21vr1TXlUavVyMrKglKprDVP3JbJZNi8eTMGDhz4yH61MTcV+RypVCps374dL7zwAm/Y9xDmRjvmpXyVzU16biHO38rEuVuZuHg7C7HJWbiWVvaeV2YmcrR0t0UHb3u08bSDv6vSoGZcquIz86jfoRXBmRkiIqJKEkLgn3v3cSk5G9E372HfpTTEJmVp7evtYIlW7rZo5maD5m62aO5mU21vPmeoWMwQERFVQGaeCofi07DrQjJOJqYjJavsPV28HSzRtJ4NGrtYo4mLEv6uNqhrbTizLoaKxQxVKzzrSUTVRWp2ASJTZTj25wWcup6Ba3dy8eCPKBMjGfwcrdHYRYkAH3v0aFjXoE4X1SQsZoiIqNYTQuB2Zj4OxKXiVOI9xCZlIS4lG0IYAbgl9atf1xLd/Ori+abOaOZmU+PupGuo+F14hjjrQE+Dnx8i3RFC4GpaLvbEpuD09XuIupGBOzllTxt5Wgn0bumNtt72aOtpx5mXaorFzDNQuto7Ly8P5ubV55HpZFgKCwsBQOOREURUMbkFRbhwOwvRN+/hcPwdRN/IQHaB5qMAjOQy+Lsq0aOhY8m6FycLnD6yDy8EN+CVXtUci5lnwMjICLa2tkhNTQVQcm+RijwnQ61Wo7CwEPn5+bXm8uOKqm25UavVSEtLg4WFBYyN+c+W6HGEELiSmoM9sakIv5iM87eyUFis+XgVhZEcrT1t0b2hI9p52aGJiw3MFf/7Y+Hhx6dQ9cWfis+Is7MzAEgFTUUIIXD//n2Ym5vX+IeEVVZtzI1cLoeHh0etOV6iihJC4PrdPFxKzsKF21mITcrGxduZuJ2Zr9HPWWmGZm426OBdBwE+9vBztIbCuOb/MVQbsJh5RmQyGVxcXODo6Fjhal+lUuHQoUPo1q0bpzgfUhtzo1AoasUsFNHjCCFwK+M+jl65g0OX7+Do1TvI0HJXXVNjOdp710Gfps7o7OMAT/uKzYqT4WEx84wZGRlVeM2DkZERioqKYGZmVmt+YVcUc0NUu6Rm5WPvpVScSEjHkSt3kJatuVhXYSxHQydrNHK2RhNXJRo4WaO1h53GaSOquVjMEBFRtSKEwM30+4i5lYETCemIuHoXV9JyNO7xYiyXoaGzNYIaO6FbAwc0q2fLU0a1GIsZIiLSu7zCIhy/dhcRV+9i76VUrc80auFui66+DujkY4/WnnYwM+GsC5XQaxlbXFyMTz/9FN7e3jA3N4ePjw8+++wzjftpCCEwffp0uLi4wNzcHEFBQYiPj9cYJz09HcOGDYNSqYStrS1GjRqFnJwcjT4xMTHo2rUrzMzM4O7ujvnz55eJZ+PGjWjUqBHMzMzQrFkzbN++vWoOnIiollOrBS4lZ2HV0QS8uuIYWs4Kx8jVp/DD4QRcS8uFXFZSvAxp74Hv32yDk58E4c/QzvgguCE6+TqwkCENep2Z+fLLL7F8+XKsWbMG/v7+OHXqFEaMGAEbGxtMmDABADB//nwsWbIEa9asgbe3Nz799FMEBwfj4sWL0pODhw0bhqSkJISHh0OlUmHEiBEYM2YM1q1bB6DkaZy9e/dGUFAQVqxYgXPnzmHkyJGwtbXFmDFjAADHjh3DkCFDMHfuXLz44otYt24dBg4ciDNnzqBp06b6SRARUQ1RUrxk48LtTByKv4NDl9OQeV9z0a6rjRk6+zqgk689ejR0hK2FQk/RkqHRazFz7NgxDBgwAH379gUAeHl5Yf369Thx4gSAklmZRYsWYdq0aRgwYAAA4P/+7//g5OSELVu24PXXX0dsbCx27tyJkydPom3btgCAb7/9Fi+88AIWLFgAV1dXrF27FoWFhfj555+hUCjg7++P6OhofPPNN1Ixs3jxYvTp0weTJ08GAHz22WcIDw/H0qVLsWLFimedGiIig1asFohPzUbE1bs4cyMDR+LTcO+hK44sFUZoWs8Gvf2d8VwjR3jxaiN6QnotZjp16oSVK1fi8uXLaNCgAc6ePYsjR47gm2++AQAkJCQgOTkZQUFB0ntsbGzQoUMHRERE4PXXX0dERARsbW2lQgYAgoKCIJfLERkZiZdeegkRERHo1q0bFIr/VfnBwcH48ssvce/ePdjZ2SEiIgKTJk3SiC84OBhbtmzRGntBQQEKCv63mj4rq+TR7yqVSmc3WiodhzduKou50Y55KR9zo50u85KaXYDw2FTsiU3F6ev3cF+leZM6cxM5mtazQUs3GzzXqC5auNnAxOh/qx2KiooeHlKv+JnRriry8rRj6bWY+eijj5CVlYVGjRrByMgIxcXFmDNnDoYNGwYASE5OBgA4OTlpvM/JyUnalpycDEdHR43txsbGqFOnjkYfb2/vMmOUbrOzs0NycvIj9/OwuXPnYtasWWXad+/eDQsLiwodf0WFh4frdLyahLnRjnkpH3OjXWXzUqwGrucAlzLkuJkL/JMrQ5ZKc1ZFIReoby1QXyngoxTwtgKM5GlAcRpSL1xB+AVdHkHV4WdGO13mJS8v76ner9di5rfffsPatWuxbt066dRPWFgYXF1dERISos/QHmvq1KkaMzlZWVlwd3dH7969oVQqdbIPlUqF8PBw9OrVi/dSeQhzox3zUj7mRruK5qWoWI24lBxE38zA0avpOHr1LvIKizX6yGRAM1cl+jR1QqCfA3zrWkEuN9zTRvzMaFcVeSk9u/Gk9FrMTJ48GR999BFef/11AECzZs1w/fp1zJ07FyEhIdIjAFJSUuDi4iK9LyUlBS1btgRQ8piAhx8RUFRUhPT0dOn9zs7OSElJ0ehT+vpxfUq3P8zU1BSmpmWfnmpiYqLzD31VjFlTMDfaMS/lY260ezgv+apinL+VieibGTgcfwcnEtJxX6VZvNhZmKCzrwPae9dBExclGjpbw9qs5uWWnxntdJmXpx1Hr8VMXl5emduzGxkZQa0uOc/q7e0NZ2dn7N27VypesrKyEBkZiXHjxgEAAgICkJGRgdOnT6NNmzYAgH379kGtVqNDhw5Sn08++QQqlUpKWHh4OBo2bAg7Ozupz969exEWFibFEh4ejoCAgCo7fiKi6kKtFohNysLh+DQcvXIXkQl3kf/QmhdrM2O0cLNFO6866N6wLprWs4GRAc+8UM2h12KmX79+mDNnDjw8PODv74+oqCh88803GDlyJICS5xmFhYXh888/h5+fn3RptqurKwYOHAgAaNy4Mfr06YO3334bK1asgEqlwvjx4/H666/D1dUVADB06FDMmjULo0aNwocffojz589j8eLFWLhwoRTLxIkTERgYiK+//hp9+/bFhg0bcOrUKaxcufKZ54WI6FlIyryPvReTsemyHJ+fO4i0nEKN7XWtTdGsXsmDGbs1qIsGTtYsXqha0msx8+233+LTTz/Fu+++i9TUVLi6umLs2LGYPn261GfKlCnIzc3FmDFjkJGRgS5dumDnzp3SPWYAYO3atRg/fjx69uwJuVyOQYMGYcmSJdJ2Gxsb7N69G6GhoWjTpg0cHBwwffp06bJsoOTKqnXr1mHatGn4+OOP4efnhy1btvAeM0RUY2TkFeLMjXs4lXgPh+Pv4NytzH+3yAEUwtzECO2966CrnwM6+TigsYs1L5Umg6DXYsba2hqLFi3CokWLyu0jk8kwe/ZszJ49u9w+derUkW6QV57mzZvj8OHDj+zz6quv4tVXX31kHyIiQ5FTUIQTCSWPCIhMSMe5W5kazzeSy4Cmrkq4yDIwNKgdAnwd+XwjMkh8NhMRUQ2RU1CEqBv3sO9SKs5cv4cLt7NQpBYafbwdLNHG0w4dvOuge0NH2JrJsX37dnTysYcJCxkyUCxmiIgMWHpuIbbF3MbWmCScuXEPqmLN4sW9jjm6+DqgjWfJ6SMnpZnGdt4QjmoCFjNERAYmPbcQO84nYVtMEo5fu4sHJ1/c7MzRsb49uvo5oLWHHdzr6PYmnkTVEYsZIqJqLregCCcS0nHs6h0cuXIXsUmaNxhr4qJE3+YueL6pM+rXtdJTlET6w2KGiKiaUasFYm5l4uiVOzh29Q7OXM8oc8O6Bk5WGNTaDb39neHtYKmnSImqBxYzRETVQGkB8+vJGwi/mIo7OQUa2+vZmqND/Tro5lcXnXzt4WhtVs5IRLUPixkiIj3Jylfh8OU72HUhGUeu3EF67v9uWmdlaowAn5K1Lx287eHnaNjPOSKqSixmiIieESEEEu/mYdeFZByMS8PpG/dQWPS/RwZYKozQvaEjXmpVD90a1OU9X4gqiMUMEVEVEkIgLiUb288lY+vZ27h2J1dje30HSzzXyBHBTZ3Rws2WBQzRE2AxQ0SkY/mq4n8fGZCG8IspGgWMsVyGAB97dG/oiMAGDvCpa8VHBhA9JRYzREQ6UKwW2Bubgj/P3sbe2BSNJ06bGssR4GOPfs1d0cvfCUozEz1GSlTzsJghInpCarVA9D8Z+PvsbWyJuoV7ef+7m66jtSm6+tVFVz8H9GzsCGsWMERVhsUMEVElqIrVOHrlDo5euYNdF1JwIz1P2mZnYYJBrd3Qr4UrmrvZ8PQR0TPCYoaI6DGKitU4ff0eNp25hR3nk5CVXyRts1QYoXsjR7zS2g1d/BxgYsQFvETPGosZIiIt7uQU4HB8Go7E38X+uFSNe8DYWyrQs7EjuvrVRc/GjrBQ8EcpkT7xXyAREUouob6alot9l1Kw+0IKTl2/p7FdaWaMXk2c8UobN7TzsoMxZ2CIqg0WM0RUa+UUFGHfpVTsi03B8WvpSM7K19je2EWJ7g3rooN3HXT25SkkouqKxQwR1RrFauBk4j2cuZmJY1fv4mRiOlTFQtpuYiRDx/ol94Dp09QZ9WzN9RgtEVUUixkiqvEuJWfhx0PXsPOcEXIiT2ps87K3QHBTZwQ2qIuW7rZc/0JkgPivlohqpIy8QmyNScLvp/9B9M2Mf1tlsLMwQXvvOuji64AAHwf4OlrpM0wi0gEWM0RUY9xMz8Of0bew91IqYv7JRLG65BSSkVyGXo0d4aW+jQmDg2BhZqrnSIlIl1jMEJFBK52B+fvsbUQmpGtsa+yixEutXPFSKzfYmsmxffstLuIlqoFYzBCRwUnNzsehy3fwZ/QtHL92V2MRb2dfezzf1AWBDerCvY6F1K5SqbQNRUQ1AIsZIjII2fkq/Bl9G5ujbuH0Q/eAaeRsjf4tXTGgZT1egURUC7GYIaJqq6hYjSNX7mBL1C3siU1FTkHJYwRkMqChkzX6NHXGC81c0MDJWs+REpE+sZghompFrRbYfTEZm6Nu4Uj8HeQWFkvbvB0sMayDB/o2d4GLDWdgiKgEixkiqhbikrOx+lgi9l1KQUpWgdRuZ2GCF5u7on9LV7TxsINczidRE5EmFjNEpDc5BUXYczEF/z1+XeNZSNamxhj67wxMU1cbFjBE9EgsZojomcpXFePolTtYf+IGDsXfQWGRGkDJvWCCGjtiSHsPBPjYw9TYSM+REpGhYDFDRM9Ewp1cbI66hf8ev4703EKp3dPeAgNa1sPr7dzhyiuRiOgJsJghoiojhMDe2FRsjrqF3ReTpfvB1LU2Rb/mrhjczh0NnKwgk/E0EhE9ORYzRKRzBUXF2BJ1C2sjbyDmn0ypvZOPPQa3c0ffZi4w5p14iUhHWMwQkU5k56tw8HIa9lxMQfjFFOmSanMTIwzt4IGBLeuhaT0lZ2GISOdYzBDRU4m8dhdL919B5LV0FBarpXYnpSne7OiJwe08UNeaD3YkoqrDYoaIKi0rX4X9l1KxLvKGxsMd6ztYomdjR/Rp6ozWHnachSGiZ4LFDBFV2LW0HKyLvIH/i7guzcIYyWUY1Loexgb6oL6DJQsYInrmWMwQ0SPlq4qx4cQNbDh5E5eSs6V2n7qW6O3vjLcCPPloASLSKxYzRKTVvdxC/HHmH6w+loh/7t0HAJgYydDeuw5GdfHGc42c9BwhEVEJFjNEpCEuORtrIhKx+cwt3FeVXJHkaG2Kcd198FKrerC1UOg5QiIiTSxmiAhCCByOv4MfDl/D4fg7UnsjZ2sM7eCBQa3dYGnKHxdEVD3xpxNRLZaSlY/dF5Lx66mbOH8rCwAgkwG9GjshpJMXOvnYc0EvEVV7LGaIaqH03EJsOHkDS/bGI19VclWSiZEMwzp4YkRnL3jaW+o5QiKiimMxQ1RLFKsF9sSmYOOpf7DvUgrUJY9JQhMXJfq1cMWrbd3gYMWb2xGR4WExQ1TDpWbl46ejCdh6Ngm3Mu5L7Y1dlBjVxRsvtaoHIzlPJRGR4WIxQ1RDZeapsPLwVfxfxHVk5xcBAGzMTfBaWze80sYdDZ2t9RwhEZFusJghqmFu5QLT/ryArTHJ0sMem7goEdrDFz0bO8LMxEjPERIR6RaLGaIaQAiBY1fvYum+eERcMwZwCwDQ0MkaE4P8EOzvzFNJRFRjsZghMmCp2fn47/Eb2BZzG1fTcgEAMggE+zvjzQBeWk1EtQOLGSIDI4RA1M0MrD1+A5uj/pGuSjI3McLLrVzhW5SAN15qARMTE/0GSkT0jLCYITIQOQVF2H0hGT8eTsDFpCypvZWHLd7s6ImejZ1gYQxs356gxyiJiJ49FjNE1VxsUhaWH7iKnReSUVhUcoM7hbEcffydMbKLN1q620p9VSqVnqIkItIfFjNE1VRGXiEW7YnH6mOJUptHHQu81Koehnfygp0lH/hIRAQAcn0HcOvWLbzxxhuwt7eHubk5mjVrhlOnTknbhRCYPn06XFxcYG5ujqCgIMTHx2uMkZ6ejmHDhkGpVMLW1hajRo1CTk6ORp+YmBh07doVZmZmcHd3x/z588vEsnHjRjRq1AhmZmZo1qwZtm/fXjUHTfQIqdn5+HzrRQTM3ScVMr2aOGFLaGccnNwd7/VqwEKGiOgBei1m7t27h86dO8PExAQ7duzAxYsX8fXXX8POzk7qM3/+fCxZsgQrVqxAZGQkLC0tERwcjPz8fKnPsGHDcOHCBYSHh2Pr1q04dOgQxowZI23PyspC79694enpidOnT+Orr77CzJkzsXLlSqnPsWPHMGTIEIwaNQpRUVEYOHAgBg4ciPPnzz+bZFCtl5mnwrwdl9Dz64P48UgC7quK0cjZGt8Na40f3mqLlu62vDKJiEgLvZ5m+vLLL+Hu7o5Vq1ZJbd7e3tL/CyGwaNEiTJs2DQMGDAAA/N///R+cnJywZcsWvP7664iNjcXOnTtx8uRJtG3bFgDw7bff4oUXXsCCBQvg6uqKtWvXorCwED///DMUCgX8/f0RHR2Nb775Rip6Fi9ejD59+mDy5MkAgM8++wzh4eFYunQpVqxY8axSQrVQsVrgt1M38dWuOKTnFgIAmtZT4j/P+aF3EycWMEREj6HXYuavv/5CcHAwXn31VRw8eBD16tXDu+++i7fffhsAkJCQgOTkZAQFBUnvsbGxQYcOHRAREYHXX38dERERsLW1lQoZAAgKCoJcLkdkZCReeuklREREoFu3blAo/jc1HxwcjC+//BL37t2DnZ0dIiIiMGnSJI34goODsWXLFq2xFxQUoKCgQHqdlVVydYlKpdLZIszScbios6yakBtVsRpbY5Lxw5EExKeW3COmvoMF3u/lh56NHGEkl6GoqKhyY9aAvFQV5kY75qV8zI12VZGXpx1Lr8XMtWvXsHz5ckyaNAkff/wxTp48iQkTJkChUCAkJATJyckAACcnJ433OTk5SduSk5Ph6Oiosd3Y2Bh16tTR6PPgjM+DYyYnJ8POzg7JycmP3M/D5s6di1mzZpVp3717NywsLCqaggoJDw/X6Xg1iaHmJjZDhj+vy5GUVzLrYmYk0LueGoEuWShKPI1diU83vqHm5VlgbrRjXsrH3Giny7zk5eU91fv1Wsyo1Wq0bdsWX3zxBQCgVatWOH/+PFasWIGQkBB9hvZYU6dO1ZjJycrKgru7O3r37g2lUqmTfahUKoSHh6NXr168AdpDDDE3arXAgfg7+PFIIk4m3gMAKM2MMbqLF4a2d4eN+dMfhyHm5VlhbrRjXsrH3GhXFXkpPbvxpPRazLi4uKBJkyYabY0bN8Yff/wBAHB2dgYApKSkwMXFReqTkpKCli1bSn1SU1M1xigqKkJ6err0fmdnZ6SkpGj0KX39uD6l2x9mamoKU1PTMu0mJiY6/9BXxZg1haHkZs/FFHyxIxbX/n3kgFwGvNHRE2FBDVCnCq5MMpS86ANzox3zUj7mRjtd5uVpx9Hr1UydO3dGXFycRtvly5fh6ekJoGQxsLOzM/bu3Sttz8rKQmRkJAICAgAAAQEByMjIwOnTp6U++/btg1qtRocOHaQ+hw4d0jgnFx4ejoYNG0pXTgUEBGjsp7RP6X6InkT0zQy8+VMkRv/fKVxLy4WVqTHe7uqNwx8+h9kDmlZJIUNEVNtUembmxo0bcHd3L3OFhRACN2/ehIeHR4XHeu+999CpUyd88cUXeO2113DixAmsXLlSumRaJpMhLCwMn3/+Ofz8/ODt7Y1PP/0Urq6uGDhwIICSmZw+ffrg7bffxooVK6BSqTB+/Hi8/vrrcHV1BQAMHToUs2bNwqhRo/Dhhx/i/PnzWLx4MRYuXCjFMnHiRAQGBuLrr79G3759sWHDBpw6dUrj8m2iiopLzsbC8MvYeaFkzZWRXIaQAC+E9fKD0ox/4RER6VKlixlvb28kJSWVWXSbnp4Ob29vFBcXV3isdu3aYfPmzZg6dSpmz54Nb29vLFq0CMOGDZP6TJkyBbm5uRgzZgwyMjLQpUsX7Ny5E2ZmZlKftWvXYvz48ejZsyfkcjkGDRqEJUuWSNttbGywe/duhIaGok2bNnBwcMD06dM17kXTqVMnrFu3DtOmTcPHH38MPz8/bNmyBU2bNq1siqgWS8nKx/ydcfjjzD8ASk4nvdTKDRN6+sLT3lLP0RER1UyVLmaEEFrve5GTk6NRYFTUiy++iBdffLHc7TKZDLNnz8bs2bPL7VOnTh2sW7fukftp3rw5Dh8+/Mg+r776Kl599dVHB0ykxd2cAqw6mojVxxKRU1ByOXUff2eE9fJDI2fdLAgnIiLtKlzMlF65I5PJ8Omnn2pcflxcXIzIyEhpUS5RbfHPvTysOpqItZHXka8qeQikv6sSswf4o41nHT1HR0RUO1S4mImKigJQMjNz7tw5jRvQKRQKtGjRAh988IHuIySqhlKz8rFwz2WsP3FTamtaT4nQ7r4I9neGXM679hIRPSsVLmb2798PABgxYgQWL16ss3upEBmS1Ox8/HQ4Ab8cv468wpL1Ya09bPGf5/zQvWFdPnqAiEgPKr1m5sHnKBHVFqnZ+ViyNx6/nfoHhUUlp5NauNvioz6NEOBjr+foiIhqt0oXM7m5uZg3bx727t2L1NRUqNVqje3Xrl3TWXBE+pavKsZ3+6/gxyMJ0kxMs3o2GBtYH32buXAmhoioGqh0MTN69GgcPHgQb775Jlxc+MOcaq79camY9dcFJN4teWZIczcbaSaGn3siouqj0sXMjh07sG3bNnTu3Lkq4iHSu7TsAszeehF/n70NAHCwMsUnfRthQIt6XNhLRFQNVbqYsbOzQ506vOSUap6ComL8eDgB3+2/gtx/TymN6OyF93s3hJWpXh9jRkREj1Dpn9CfffYZpk+fjjVr1mjca4bIkJ1MTMdHf8Tg6r8PgmxaT4lZ/XmvGCIiQ1DpYubrr7/G1atX4eTkBC8vrzJPujxz5ozOgiOqasmZ+fh820VsjUkCADhYKTCtbxP0b+HKU0pERAai0sVM6QMeiQxZvqoYi/fG46cjCdKl1oPbuuPD5xvxSdZERAam0sXMjBkzqiIOomdCCIEDcWmYt+MS4lKyAZRcpTRnYDM0c7PRc3RERPQkuKqRao0zN+7hi22xOHX9HgBAaWaMLwc1R5+mzrzUmojIgFW6mJHL5Y/8wV9cXPxUARHpWlLmfYRtiEZkQjoAQGEsx1sdPRHawxd2PKVERGTwKl3MbN68WeO1SqVCVFQU1qxZg1mzZuksMKKnJYTAX2dvY+ZfF3AvTwUTIxn6t6iHSb0boJ6tub7DIyIiHal0MTNgwIAyba+88gr8/f3x66+/YtSoUToJjOhppGTlY/LvMTh0OQ0A0NhFieXDWsPLwVLPkRERka7pbM1Mx44dMWbMGF0NR/REitUCv526ia93x+FOTiEAYMJzvvhPTz+YGMn1HB0REVUFnRQz9+/fx5IlS1CvXj1dDEdUaUII7LqQjPk743DtTsmN7xo4WWHJkFZo5KzUc3RERFSVnuhxBg8uABZCIDs7GxYWFvjvf/+r0+CIKuJOTgEm/XZWOqVkY26C/zznizc6esLMxEjP0RERUVWrdDGzaNEijddyuRx169ZFhw4dYGdnp6u4iCokLjkbI/7vDNKyC6AwlmN0F2+828OXz1IiIqpFKv0TPyQkpCriIKoUIQQOJ8sw9YcTyCssRj1bc/w8vB0aOlvrOzQiInrGnujP14yMDPz000+IjY0FAPj7+2PkyJGwseEdVKnqZd5X4eM/zmFbghGAYrT3qoPlb7SGvZWpvkMjIiI9qPTlHadOnYKPjw8WLlyI9PR0pKen45tvvoGPjw8fMklVLuafDLz47WFsO58MGQQ+fr4hNozpyEKGiKgWq/TMzHvvvYf+/fvjhx9+gLFxyduLioowevRohIWF4dChQzoPkkgIgZ+PJmLu9lgUqQXc7MzxSr1sjOjkyadbExHVcpUuZk6dOqVRyACAsbExpkyZgrZt2+o0OCKg5Gqlj/6IwZ7YVABAsL8TPuvfGBEH9ug5MiIiqg4qXcwolUrcuHEDjRo10mi/efMmrK25+JJ060RCOt5dewZ3cgqgMJJjQk9fhPbwRVFRkb5DIyKiaqLSxczgwYMxatQoLFiwAJ06dQIAHD16FJMnT8aQIUN0HiDVTsVqgaX7ruDbffEoUgv4OVph4eCWaFqPi8yJiEhTpYuZBQsWQCaT4a233pL+OjYxMcG4ceMwb948nQdItU96biEmbojC4fg7AIAXm7vgy0HNYcl7xxARkRaV/u2gUCiwePFizJ07F1evXgUA+Pj4wMLCQufBUe2z+0IyZv19Ebcy7sPUWI65LzfDS63qadx1moiI6EEVLmaKi4tx4cIF+Pn5wdzcHBYWFmjWrBmAkmczxcTEoGnTppDL+TA/ejIbTtzA1M3nIARQz9YcP7zVFk1c+VwlIiJ6tApXHr/88gtGjhwJhUJRZpuJiQlGjhyJdevW6TQ4qh2EEFi05zI+2lRSyAxp74697weykCEiogqpcDHz008/4YMPPoCRUdkH95Vemr1y5UqdBkc1n1otMPOvC1i0Jx4AMDawPuYMbMYHRBIRUYVV+DRTXFwcOnbsWO72du3aSY83IKqIK6k5+OiPGJy6fg8AMLNfEwzv7K3nqIiIyNBUuJjJzc1FVlZWuduzs7ORl5enk6Co5gu/mIIJ66NwX1UMS4UR5rzUDANb1dN3WEREZIAqfJrJz88Px44dK3f7kSNH4Ofnp5OgqOYSQuC7A1cw9pdTuK8qRkB9e+wM68ZChoiInliFi5mhQ4di2rRpiImJKbPt7NmzmD59OoYOHarT4KhmURWr8d6v0Zi/Mw5qAQxq7Yb/G9Ue7nV4WT8RET25Cp9meu+997Bjxw60adMGQUFB0uMMLl26hD179qBz58547733qixQMmwZeYUY998ziLh2F3IZMLO/P94K8NJ3WEREVANUuJgxMTHB7t27sXDhQqxbtw6HDh2CEAINGjTAnDlzEBYWBhMTk6qMlQzUrYz7ePPHSFy7kwtzEyMsHdoKPRs76TssIiKqISp1B2ATExNMmTIFU6ZMqap4qIZJuJOL11dGICWrAK42Zlj5Vls+X4mIiHSKD7uhKhOblIU3fozE3dxC+NS1xJqR7eFmx/UxRESkWyxmqEok3snFsB8jkZ5biCYuSqwZ2R51rU31HRYREdVALGZI55Iz8zH0h+NIzy2Ev6sS60Z3hI0F11MREVHV4FMhSadSs/Lx1s+RuJ2Zj/oOllg1vB0LGSIiqlJPXMwUFhYiLi4ORUVFuoyHDFhWvgpDfjiOyyk5qGttilUj2sFRaabvsIiIqIardDGTl5eHUaNGwcLCAv7+/rhx4wYA4D//+Q/mzZun8wDJMNzJKcAbP0bialouHKwU+G1sADztLfUdFhER1QKVLmamTp2Ks2fP4sCBAzAz+99f3UFBQfj11191GhwZhpSsfLz83THE/JMJWwsTrBreHt4OLGSIiOjZqPQC4C1btuDXX39Fx44dIZPJpHZ/f39cvXpVp8FR9Xc3pwDDV53EjfQ8uNcxx+oR7eFT10rfYRERUS1S6WImLS0Njo6OZdpzc3M1ihuq+e7lFuL1lccRn5oDOwsT/N/IDpyRISKiZ67Sp5natm2Lbdu2Sa9LC5gff/wRAQEBuouMqrXsfBWG/hiJ+NSSxb4b3wlgIUNERHpR6ZmZL774As8//zwuXryIoqIiLF68GBcvXsSxY8dw8ODBqoiRqpmcgiKM+b/TiE3KgoOVKdaN7gBfR2t9h0VERLVUpWdmunTpgujoaBQVFaFZs2bYvXs3HB0dERERgTZt2lRFjFSNCCHw3q/RiLh2F5YKI/wU0hZ+TixkiIhIf57oDsA+Pj744YcfdB0LGYBl+68g/GIKFEZy/N+oDmjhbqvvkIiIqJardDFTel+Z8nh4eDxxMFS9/Rl9C1+HXwYAzBrgjzaednqOiIiI6AlOM3l5ecHb27vcryc1b948yGQyhIWFSW35+fkIDQ2Fvb09rKysMGjQIKSkpGi878aNG+jbty8sLCzg6OiIyZMnl7kr8YEDB9C6dWuYmprC19cXq1evLrP/ZcuWwcvLC2ZmZujQoQNOnDjxxMdSE8UlZ2PyxhgIAbwV4Ikh7Vm0EhFR9VDpYiYqKgpnzpyRviIjI7FixQo0aNAAGzdufKIgTp48ie+//x7NmzfXaH/vvffw999/Y+PGjTh48CBu376Nl19+WdpeXFyMvn37orCwEMeOHcOaNWuwevVqTJ8+XeqTkJCAvn37okePHoiOjkZYWBhGjx6NXbt2SX1+/fVXTJo0CTNmzMCZM2fQokULBAcHIzU19YmOp6ZJzc7HO/89jcJiNZ5r5IiZ/fz1HRIREdH/CB3ZunWrCAwMrPT7srOzhZ+fnwgPDxeBgYFi4sSJQgghMjIyhImJidi4caPUNzY2VgAQERERQgghtm/fLuRyuUhOTpb6LF++XCiVSlFQUCCEEGLKlCnC399fY5+DBw8WwcHB0uv27duL0NBQ6XVxcbFwdXUVc+fOrfBxZGZmCgAiMzOz4gf/GIWFhWLLli2isLBQZ2NWVk6+SgR9fUB4frhVBHyxR6Rk3ddbLA+qDrmpjpiX8jE32jEv5WNutKuKvDzt79AnWgCsTcOGDXHy5MlKvy80NBR9+/ZFUFAQPv/8c6n99OnTUKlUCAoKktoaNWoEDw8PREREoGPHjoiIiECzZs3g5OQk9QkODsa4ceNw4cIFtGrVChERERpjlPYpPZ1VWFiI06dPY+rUqdJ2uVyOoKAgRERElBt3QUEBCgoKpNdZWVkAAJVKBZVKVek8aFM6jq7GqywhBD7dcqHkXjJWCvz0VmvYmRnpLZ4H6Ts31RXzUj7mRjvmpXzMjXZVkZenHavSxUzpL+1SQggkJSVh5syZ8PPzq9RYGzZswJkzZ7QWQcnJyVAoFLC1tdVod3JyQnJystTnwUKmdHvptkf1ycrKwv3793Hv3j0UFxdr7XPp0qVyY587dy5mzZpVpn337t2wsLAo931PIjw8XKfjVdTRFBk2XTMCALzqfh+XTx3CZb1EUj595aa6Y17Kx9xox7yUj7nRTpd5ycvLe6r3V7qYsbW1LfPYAiEE3N3dsWHDhgqPc/PmTUycOBHh4eEaD6w0FFOnTsWkSZOk11lZWXB3d0fv3r2hVCp1sg+VSoXw8HD06tULJiYmOhmzos7+k4k/fzoJQI33evri3e71n+n+H0efuanOmJfyMTfaMS/lY260q4q8PDxRUlmVLmb279+v8Voul6Nu3brw9fWFsXHFhzt9+jRSU1PRunVrqa24uBiHDh3C0qVLsWvXLhQWFiIjI0NjdiYlJQXOzs4AAGdn5zJXHZVe7fRgn4evgEpJSYFSqYS5uTmMjIxgZGSktU/pGNqYmprC1NS0TLuJiYnOP/RVMeajJGfmY/z6sygoKlnw+5+eDSCXV8/nbj3r3BgK5qV8zI12zEv5mBvtdJmXpx2n0sVMYGDgU+2wVM+ePXHu3DmNthEjRqBRo0b48MMP4e7uDhMTE+zduxeDBg0CAMTFxeHGjRvSM6ACAgIwZ84cpKamSg+/DA8Ph1KpRJMmTaQ+27dv19hPeHi4NIZCoUCbNm2wd+9eDBw4EACgVquxd+9ejB8/XifHakiEEJi25TySs/JRv64lFr/estoWMkRERMATFDN//fVXhfv279+/3G3W1tZo2rSpRpulpSXs7e2l9lGjRmHSpEmoU6cOlEol/vOf/yAgIAAdO3YEAPTu3RtNmjTBm2++ifnz5yM5ORnTpk1DaGioNGvyzjvvYOnSpZgyZQpGjhyJffv24bffftN4WOakSZMQEhKCtm3bon379li0aBFyc3MxYsSICh9rTfHf49exJzYFxnIZlg1tDWsz/jVCRETVW6WLmYEDB0Imk0EIodH+cJtMJkNxcfFTBbdw4ULI5XIMGjQIBQUFCA4OxnfffSdtNzIywtatWzFu3DgEBATA0tISISEhmD17ttTH29sb27Ztw3vvvYfFixfDzc0NP/74I4KDg6U+gwcPRlpaGqZPn47k5GS0bNkSO3fuLLMouKaLT8nGlzvjAABTX2iMxi66WftDRERUlSpdzOzevRsffvghvvjiC+lUTUREBKZNm4YvvvgCvXr1euJgDhw4oPHazMwMy5Ytw7Jly8p9j6enZ5nTSA/r3r07oqKiHtln/PjxtfK0UqmMvEIM+zESOQVFaO9VB8M7eek7JCIiogqpdDETFhaGFStWoEuXLlJbcHAwLCwsMGbMGMTGxuo0QHo2PtlyHqnZBXCzM8eKN9vAiOtkiIjIQFT6cQZXr14tc+8XALCxsUFiYqIOQqJnbd+lFGyLSYJcBnw7pBXqWCr0HRIREVGFVbqYadeuHSZNmqRxKXNKSgomT56M9u3b6zQ4qnqZeSp89EfJVWXDO3mjlQefhE1ERIal0sXMzz//jKSkJHh4eMDX1xe+vr7w8PDArVu38NNPP1VFjFSFPt5yDqnZBfB2sMSUPg31HQ4REVGlVXrNjK+vL2JiYhAeHi7d7r9x48YICgoqc2dgqt4Ox6dJp5e+fq0FzEyM9B0SERFRpT3RgyZlMhl69+6N3r176zoeekaK1QJzt5cUo28FeKE1Ty8REZGBqlAxs2TJEowZMwZmZmZYsmTJI/tOmDBBJ4FR1fr15E1cTMqCtZkxxj/nq+9wiIiInliFipmFCxdi2LBhMDMzw8KFC8vtJ5PJWMwYgJyCIizcU/L867CgBnCwKvuMKSIiIkNRoWImISFB6/+TYfp2XzzSsgvgUccCb3b01Hc4RERET6XSVzORYYtPycbPR0oK0hn9mkBhzI8AEREZtkovAC4uLsbq1auxd+9epKamQq1Wa2zft2+fzoIj3RJCYNbfF6EqFujRsC6ea+So75CIiIieWqWLmYkTJ2L16tXo27cvmjZtysuxDchfZ2/jyJU7UBjL8emLTfi9IyKiGqHSxcyGDRvw22+/4YUXXqiKeKiK5KuKMf/fJ2KPC/RB/bpWeo6IiIhINyq9YEKhUMDXl5fyGpr/Hr+OWxn34WJjhncCffQdDhERkc5Uuph5//33sXjxYgghqiIeqgJ5hUX44fA1AMCEnn4wV/BOv0REVHNU+jTTkSNHsH//fuzYsQP+/v4wMTHR2L5p0yadBUe6sS7yBlKyClDP1hwvt66n73CIiIh0qtLFjK2tLV566aWqiIWqQHpuIRbvjQcAjH/OF6bGnJUhIqKapdLFzKpVq6oiDqoiq48lIju/CI1dlHitrbu+wyEiItI53jGtBkvPLZRukDeuuw+M5LwUm4iIap4Kz8zY2dlpvS+JjY0NGjRogA8++AC9evXSaXD0dNafuIGcgiI0dLJG32Yu+g6HiIioSlS4mFm0aJHW9oyMDJw+fRovvvgifv/9d/Tr109XsdFTyCkowvcHrwIAxgbW56wMERHVWBUuZkJCQh65vWXLlpg7dy6LmWpi69nbyMovQv26lhjQklcwERFRzaWzNTMvvvgiLl26pKvh6CmtP3EDADCotRtnZYiIqEbTWTFTUFAAhUKhq+HoKZy+no6z/2RCYSTH4Ha8gomIiGo2nRUzP/30E1q2bKmr4egp/Hw0EQAwsJUrHKxM9RsMERFRFavwmplJkyZpbc/MzMSZM2dw+fJlHDp0SGeB0ZP5514etp9LAgC82dFLv8EQERE9AxUuZqKiorS2K5VK9OrVC5s2bYK3t7fOAqMns/nMLQgBBNS3RzM3G32HQ0REVOUqXMzs37+/KuMgHRBCYHP0LQDAoDZueo6GiIjo2eAdgGuQqJsZuJaWC4WxHMH+TvoOh4iI6JlgMVODbD5TMivTt5kLrM1MHtObiIioZmAxU0MUqwXCL6YAAPq14KMLiIio9mAxU0McuXIHyVn5UJoZo5OPg77DISIiemZYzNQQv526CQB4qVU9mJkY6TkaIiKiZ4fFTA2Qmp2PXeeTAQCvtOEdf4mIqHZhMVMDbI9JQpFaoKW7Le8tQ0REtQ6LmRpg+7mSWZkXm3PhLxER1T4sZgzczfQ8nEhMh0wGPN+MxQwREdU+LGYM3IG4VABAe686qGdrrudoiIiInj0WMwbuyJU7AIBuDerqORIiIiL9YDFjwFTFahy7ehcA0MnHXs/REBER6QeLGQMWeS0d2flFqGOpQHM3W32HQ0REpBcsZgzY9vNJAIBgfycYyWV6joaIiEg/WMwYKCEE9l8qWfzb299Zz9EQERHpD4sZA3UpORtJmfkwNZYjoD7XyxARUe3FYsZAHYkvuYqpk489n8VERES1GosZA3UiMR0A0IGzMkREVMuxmDFAQgicvn4PANDOy07P0RAREekXixkDdO1OLtJzC6EwlqNpPT5YkoiIajcWMwbo1L+nmFq62cLUmOtliIiodmMxY4BOJZacYmrDU0xEREQsZgzRKa6XISIikrCYMTB3cgqQcCcXANDag8UMERGRXouZuXPnol27drC2toajoyMGDhyIuLg4jT75+fkIDQ2Fvb09rKysMGjQIKSkpGj0uXHjBvr27QsLCws4Ojpi8uTJKCoq0uhz4MABtG7dGqampvD19cXq1avLxLNs2TJ4eXnBzMwMHTp0wIkTJ3R+zE+r9BRTAycr2Foo9BwNERGR/um1mDl48CBCQ0Nx/PhxhIeHQ6VSoXfv3sjNzZX6vPfee/j777+xceNGHDx4ELdv38bLL78sbS8uLkbfvn1RWFiIY8eOYc2aNVi9ejWmT58u9UlISEDfvn3Ro0cPREdHIywsDKNHj8auXbukPr/++ismTZqEGTNm4MyZM2jRogWCg4ORmpr6bJJRQaevlyz+betVR8+REBERVROiGklNTRUAxMGDB4UQQmRkZAgTExOxceNGqU9sbKwAICIiIoQQQmzfvl3I5XKRnJws9Vm+fLlQKpWioKBACCHElClThL+/v8a+Bg8eLIKDg6XX7du3F6GhodLr4uJi4erqKubOnVuh2DMzMwUAkZmZWcmjLl9hYaHYsmWLKCwslNoGLD0iPD/cKv44fVNn+zFE2nJDzMujMDfaMS/lY260q4q8PO3vUGP9llKaMjMzAQB16pTMOpw+fRoqlQpBQUFSn0aNGsHDwwMRERHo2LEjIiIi0KxZMzg5OUl9goODMW7cOFy4cAGtWrVCRESExhilfcLCwgAAhYWFOH36NKZOnSptl8vlCAoKQkREhNZYCwoKUFBQIL3OysoCAKhUKqhUqqfIwv+UjlP634IiNc7fKslRi3rWOtuPIXo4N1SCeSkfc6Md81I+5ka7qsjL045VbYoZtVqNsLAwdO7cGU2bNgUAJCcnQ6FQwNbWVqOvk5MTkpOTpT4PFjKl20u3PapPVlYW7t+/j3v37qG4uFhrn0uXLmmNd+7cuZg1a1aZ9t27d8PCwqKCR10x4eHhAIBbuUCR2hjmRgLnIg7gvEynuzFIpbkhTcxL+Zgb7ZiX8jE32ukyL3l5eU/1/mpTzISGhuL8+fM4cuSIvkOpkKlTp2LSpEnS66ysLLi7u6N3795QKpU62YdKpUJ4eDh69eoFExMTbIq6BcRcgL+bHfr2ba+TfRiqh3NDJZiX8jE32jEv5WNutKuKvJSe3XhS1aKYGT9+PLZu3YpDhw7Bzc1Nand2dkZhYSEyMjI0ZmdSUlLg7Ows9Xn4qqPSq50e7PPwFVApKSlQKpUwNzeHkZERjIyMtPYpHeNhpqamMDU1LdNuYmKi8w996ZiXUkoWRjdzs+U/rH9VRb5rAualfMyNdsxL+Zgb7XSZl6cdR69XMwkhMH78eGzevBn79u2Dt7e3xvY2bdrAxMQEe/fuldri4uJw48YNBAQEAAACAgJw7tw5jauOwsPDoVQq0aRJE6nPg2OU9ikdQ6FQoE2bNhp91Go19u7dK/WpDi7cLqlc/V35PCYiIqJSep2ZCQ0Nxbp16/Dnn3/C2tpaWuNiY2MDc3Nz2NjYYNSoUZg0aRLq1KkDpVKJ//znPwgICEDHjh0BAL1790aTJk3w5ptvYv78+UhOTsa0adMQGhoqzZy88847WLp0KaZMmYKRI0di3759+O2337Bt2zYplkmTJiEkJARt27ZF+/btsWjRIuTm5mLEiBHPPjFaCCEQm1RSzDRx0c1pLCIioppAr8XM8uXLAQDdu3fXaF+1ahWGDx8OAFi4cCHkcjkGDRqEgoICBAcH47vvvpP6GhkZYevWrRg3bhwCAgJgaWmJkJAQzJ49W+rj7e2Nbdu24b333sPixYvh5uaGH3/8EcHBwVKfwYMHIy0tDdOnT0dycjJatmyJnTt3llkUrC+p2QXIzi+CXAb4OFrqOxwiIqJqQ6/FjBDisX3MzMywbNkyLFu2rNw+np6e2L59+yPH6d69O6Kioh7ZZ/z48Rg/fvxjY9KHK6k5AABPe0s+KZuIiOgBfDaTgSh9HlN9B87KEBERPYjFjIG4frekmPGw1+09bIiIiAwdixkDEf/vaSZfRys9R0JERFS9sJgxENfSSmZmfOqymCEiInoQixkDUFCkxj/3Sm717M01M0RERBpYzBiAf+7dh1oAFgojOFqXveswERFRbcZixgAk/rv418veEjIZny5JRET0IBYzBuBG+n0APMVERESkDYsZA5CaXQAAcLYx03MkRERE1Q+LGQNwL68QAFDHUqHnSIiIiKofFjMGID2XxQwREVF5WMwYgHt5KgCAnQWLGSIiooexmDEApTMz9lYsZoiIiB7GYsYAcGaGiIiofCxmqrkiNZCdXwQAsOeaGSIiojJYzFRzuSV1DOQywMbcRL/BEBERVUMsZqq5nJIzTLCzUEAu591/iYiIHsZipprLLSopYHhZNhERkXYsZqo5aWaGxQwREZFWLGaqudI1M1z8S0REpB2LmWouR1VymokzM0RERNqxmKnmcv89zcSZGSIiIu1YzFRzOf+eZuIN84iIiLRjMVPNlS4A5qMMiIiItGMxU83l/HtpNmdmiIiItGMxU82VrpnhfWaIiIi0YzFTjQkhpEuzeTUTERGRdixmqrH7qmIUi5LTTLZ8LhMREZFWLGaqsdKnZRvJZbBQGOk5GiIiouqJxUw1VlrMWCqMIJPxIZNERETasJipxvIKiwEAlqbGeo6EiIio+mIxU42VFjM8xURERFQ+FjPVWJ7q35kZFjNERETlYjFTjeUVlKyZMWcxQ0REVC4WM9VYfpEaAGBmzGKGiIioPCxmqrGCf4sZUxN+m4iIiMrD35LVWMG/a2ZMjfltIiIiKg9/S1Zj+ap/TzOZ8DQTERFReVjMVGOlp5kURvw2ERERlYe/JasxVfG/xQxPMxEREZWLvyWrsSK1AACYGPFRBkREROVhMVONlc7MmPA0ExERUbn4W7IaKy1mjOWcmSEiIioPi5lqTFVcepqJ3yYiIqLy8LdkNVb875oZI87MEBERlYvFTDWmFixmiIiIHofFTDWmLlkyAxlrGSIionKxmKnGiktnZljNEBERlYvFTDVWumZGztNMRERE5WIxU42pOTNDRET0WCxmqrH/LQDWcyBERETVGH9NVmP/WwDMmRkiIqLysJh5yLJly+Dl5QUzMzN06NABJ06c0FssXABMRET0eCxmHvDrr79i0qRJmDFjBs6cOYMWLVogODgYqampeolHzQXAREREj8Vi5gHffPMN3n77bYwYMQJNmjTBihUrYGFhgZ9//lkv8fxvZkYvuyciIjIIxvoOoLooLCzE6dOnMXXqVKlNLpcjKCgIERERZfoXFBSgoKBAep2VlQUAUKlUUKlUOomp+N8HTQqh1tmYNUVpPpgXTcxL+Zgb7ZiX8jE32lVFXp52LBYz/7pz5w6Ki4vh5OSk0e7k5IRLly6V6T937lzMmjWrTPvu3bthYWGhk5jMC+Twtpbh2sUYbL91Vidj1jTh4eH6DqFaYl7Kx9xox7yUj7nRTpd5ycvLe6r3s5h5QlOnTsWkSZOk11lZWXB3d0fv3r2hVCp1so9eKhXCw8PRq1cvmJiY6GTMmkLF3GjFvJSPudGOeSkfc6NdVeSl9OzGk2Ix8y8HBwcYGRkhJSVFoz0lJQXOzs5l+puamsLU1LRMu4mJic4/9FUxZk3B3GjHvJSPudGOeSkfc6OdLvPytONwAfC/FAoF2rRpg71790ptarUae/fuRUBAgB4jIyIiokfhzMwDJk2ahJCQELRt2xbt27fHokWLkJubixEjRug7NCIiIioHi5kHDB48GGlpaZg+fTqSk5PRsmVL7Ny5s8yiYCIiIqo+WMw8ZPz48Rg/fry+wyAiIqIK4poZIiIiMmgsZoiIiMigsZghIiIig8ZihoiIiAwaixkiIiIyaCxmiIiIyKDx0mwdEUIAePrnSzxIpVIhLy8PWVlZvJX2Q5gb7ZiX8jE32jEv5WNutKuKvJT+7iz9XVpZLGZ0JDs7GwDg7u6u50iIiIgMU3Z2NmxsbCr9Ppl40jKINKjVaty+fRvW1taQyWQ6GbP0Sdw3b97U2ZO4awrmRjvmpXzMjXbMS/mYG+2qIi9CCGRnZ8PV1RVyeeVXwHBmRkfkcjnc3NyqZGylUsl/SOVgbrRjXsrH3GjHvJSPudFO13l5khmZUlwATERERAaNxQwREREZNBYz1ZipqSlmzJgBU1NTfYdS7TA32jEv5WNutGNeysfcaFcd88IFwERERGTQODNDREREBo3FDBERERk0FjNERERk0FjMEBERkUFjMVONLVu2DF5eXjAzM0OHDh1w4sQJfYdUYYcOHUK/fv3g6uoKmUyGLVu2aGwXQmD69OlwcXGBubk5goKCEB8fr9EnPT0dw4YNg1KphK2tLUaNGoWcnByNPjExMejatSvMzMzg7u6O+fPnl4ll48aNaNSoEczMzNCsWTNs37690rHoyty5c9GuXTtYW1vD0dERAwcORFxcnEaf/Px8hIaGwt7eHlZWVhg0aBBSUlI0+ty4cQN9+/aFhYUFHB0dMXnyZBQVFWn0OXDgAFq3bg1TU1P4+vpi9erVZeJ53GesIrHowvLly9G8eXPpJlwBAQHYsWNHpeKoaTkpz7x58yCTyRAWFlapmGpafmbOnAmZTKbx1ahRo0rFUdNy8qBbt27hjTfegL29PczNzdGsWTOcOnVK2l7jfgYLqpY2bNggFAqF+Pnnn8WFCxfE22+/LWxtbUVKSoq+Q6uQ7du3i08++URs2rRJABCbN2/W2D5v3jxhY2MjtmzZIs6ePSv69+8vvL29xf3796U+ffr0ES1atBDHjx8Xhw8fFr6+vmLIkCHS9szMTOHk5CSGDRsmzp8/L9avXy/Mzc3F999/L/U5evSoMDIyEvPnzxcXL14U06ZNEyYmJuLcuXOVikVXgoODxapVq8T58+dFdHS0eOGFF4SHh4fIycmR+rzzzjvC3d1d7N27V5w6dUp07NhRdOrUSdpeVFQkmjZtKoKCgkRUVJTYvn27cHBwEFOnTpX6XLt2TVhYWIhJkyaJixcvim+//VYYGRmJnTt3Sn0q8hl7XCy68tdff4lt27aJy5cvi7i4OPHxxx8LExMTcf78+VqbE21OnDghvLy8RPPmzcXEiRMrHFNNzM+MGTOEv7+/SEpKkr7S0tJqdU5KpaenC09PTzF8+HARGRkprl27Jnbt2iWuXLki9alpP4NZzFRT7du3F6GhodLr4uJi4erqKubOnavHqJ7Mw8WMWq0Wzs7O4quvvpLaMjIyhKmpqVi/fr0QQoiLFy8KAOLkyZNSnx07dgiZTCZu3bolhBDiu+++E3Z2dqKgoEDq8+GHH4qGDRtKr1977TXRt29fjXg6dOggxo4dW+FYqlJqaqoAIA4ePCjt28TERGzcuFHqExsbKwCIiIgIIURJoSiXy0VycrLUZ/ny5UKpVEq5mDJlivD399fY1+DBg0VwcLD0+nGfsYrEUpXs7OzEjz/+yJz8Kzs7W/j5+Ynw8HARGBgoFTO1NT8zZswQLVq00Lqttuak1Icffii6dOlS7vaa+DOYp5mqocLCQpw+fRpBQUFSm1wuR1BQECIiIvQYmW4kJCQgOTlZ4/hsbGzQoUMH6fgiIiJga2uLtm3bSn2CgoIgl8sRGRkp9enWrRsUCoXUJzg4GHFxcbh3757U58H9lPYp3U9FYqlKmZmZAIA6deoAAE6fPg2VSqURT6NGjeDh4aGRm2bNmsHJyUnjmLKysnDhwgWpz6OOuyKfsYrEUhWKi4uxYcMG5ObmIiAggDn5V2hoKPr27VvmGGpzfuLj4+Hq6or69etj2LBhuHHjRoXjqKk5AYC//voLbdu2xauvvgpHR0e0atUKP/zwg7S9Jv4MZjFTDd25cwfFxcUa/8gAwMnJCcnJyXqKSndKj+FRx5ecnAxHR0eN7cbGxqhTp45GH21jPLiP8vo8uP1xsVQVtVqNsLAwdO7cGU2bNpXiUSgUsLW1fWTMT3rcWVlZuH//foU+YxWJRZfOnTsHKysrmJqa4p133sHmzZvRpEmTWp2TUhs2bMCZM2cwd+7cMttqa346dOiA1atXY+fOnVi+fDkSEhLQtWtXZGdn19qclLp27RqWL18OPz8/7Nq1C+PGjcOECROwZs0ajeOrST+D+dRsIj0JDQ3F+fPnceTIEX2HUi00bNgQ0dHRyMzMxO+//46QkBAcPHhQ32Hp3c2bNzFx4kSEh4fDzMxM3+FUG88//7z0/82bN0eHDh3g6emJ3377Debm5nqMTP/UajXatm2LL774AgDQqlUrnD9/HitWrEBISIieo6sanJmphhwcHGBkZFRmtXtKSgqcnZ31FJXulB7Do47P2dkZqampGtuLioqQnp6u0UfbGA/uo7w+D25/XCxVYfz48di6dSv2798PNzc3qd3Z2RmFhYXIyMh4ZMxPetxKpRLm5uYV+oxVJBZdUigU8PX1RZs2bTB37ly0aNECixcvrtU5AUpOU6SmpqJ169YwNjaGsbExDh48iCVLlsDY2BhOTk61Oj+lbG1t0aBBA1y5cqXWf2ZcXFzQpEkTjbbGjRtLp+Fq4s9gFjPVkEKhQJs2bbB3716pTa1WY+/evQgICNBjZLrh7e0NZ2dnjePLyspCZGSkdHwBAQHIyMjA6dOnpT779u2DWq1Ghw4dpD6HDh2CSqWS+oSHh6Nhw4aws7OT+jy4n9I+pfupSCy6JITA+PHjsXnzZuzbtw/e3t4a29u0aQMTExONeOLi4nDjxg2N3Jw7d07jB014eDiUSqX0A+xxx12Rz1hFYqlKarUaBQUFtT4nPXv2xLlz5xAdHS19tW3bFsOGDZP+vzbnp1ROTg6uXr0KFxeXWv+Z6dy5c5lbPly+fBmenp4AaujP4AovFaZnasOGDcLU1FSsXr1aXLx4UYwZM0bY2tpqrLyvzrKzs0VUVJSIiooSAMQ333wjoqKixPXr14UQJZfi2draij///FPExMSIAQMGaL0ssFWrViIyMlIcOXJE+Pn5aVwWmJGRIZycnMSbb74pzp8/LzZs2CAsLCzKXBZobGwsFixYIGJjY8WMGTO0Xhb4uFh0Zdy4ccLGxkYcOHBA45LSvLw8qc8777wjPDw8xL59+8SpU6dEQECACAgIkLaXXlLau3dvER0dLXbu3Cnq1q2r9ZLSyZMni9jYWLFs2TKtl5Q+7jP2uFh05aOPPhIHDx4UCQkJIiYmRnz00UdCJpOJ3bt319qcPMqDVzNVJKaamJ/3339fHDhwQCQkJIijR4+KoKAg4eDgIFJTU2ttTkqdOHFCGBsbizlz5oj4+Hixdu1aYWFhIf773/9KfWraz2AWM9XYt99+Kzw8PIRCoRDt27cXx48f13dIFbZ//34BoMxXSEiIEKLkcrxPP/1UODk5CVNTU9GzZ08RFxenMcbdu3fFkCFDhJWVlVAqlWLEiBEiOztbo8/Zs2dFly5dhKmpqahXr56YN29emVh+++030aBBA6FQKIS/v7/Ytm2bxvaKxKIr2nICQKxatUrqc//+ffHuu+8KOzs7YWFhIV566SWRlJSkMU5iYqJ4/vnnhbm5uXBwcBDvv/++UKlUGn32798vWrZsKRQKhahfv77GPko97jNWkVh0YeTIkcLT01MoFApRt25d0bNnT6mQqWgcNS0nj/JwMVMb8zN48GDh4uIiFAqFqFevnhg8eLDGfVRqY04e9Pfff4umTZsKU1NT0ahRI7Fy5UqN7TXtZ7BMCCEqPo9DREREVL1wzQwREREZNBYzREREZNBYzBAREZFBYzFDREREBo3FDBERERk0FjNERERk0FjMEBERkUFjMUNEREQGjcUMET1ziYmJkMlkiI6O1ncokkuXLqFjx44wMzNDy5Yt9R0OAODAgQOQyWRlHlJIRJpYzBDVQsOHD4dMJsO8efM02rds2QKZTKanqPRrxowZsLS0RFxcXJkH45VKS0vDuHHj4OHhAVNTUzg7OyM4OBhHjx6V+shkMmzZskUnMXXq1AlJSUmwsbHRyXhENRWLGaJayszMDF9++SXu3bun71B0prCw8Infe/XqVXTp0gWenp6wt7fX2mfQoEGIiorCmjVrcPnyZfz111/o3r077t69+8T7LY9KpYJCoYCzs3OtLTCJKorFDFEtFRQUBGdnZ8ydO7fcPjNnzixzymXRokXw8vKSXg8fPhwDBw7EF198AScnJ9ja2mL27NkoKirC5MmTUadOHbi5uWHVqlVlxr906RI6deoEMzMzNG3aFAcPHtTYfv78eTz//POwsrKCk5MT3nzzTdy5c0fa3r17d4wfPx5hYWFwcHBAcHCw1uNQq9WYPXs23NzcYGpqipYtW2Lnzp3SdplMhtOnT2P27NmQyWSYOXNmmTEyMjJw+PBhfPnll+jRowc8PT3Rvn17TJ06Ff379wcAKS8vvfQSZDKZRp6WL18OHx8fKBQKNGzYEL/88ovG+DKZDMuXL0f//v1haWmJOXPmaD3NdOTIEXTt2hXm5uZwd3fHhAkTkJubK23/7rvv4OfnBzMzMzg5OeGVV17RmhOimoTFDFEtZWRkhC+++ALffvst/vnnn6caa9++fbh9+zYOHTqEb775BjNmzMCLL74IOzs7REZG4p133sHYsWPL7Gfy5Ml4//33ERUVhYCAAPTr10+a5cjIyMBzzz2HVq1a4dSpU9i5cydSUlLw2muvaYyxZs0aKBQKHD16FCtWrNAa3+LFi/H1119jwYIFiImJQXBwMPr374/4+HgAQFJSEvz9/fH+++8jKSkJH3zwQZkxrKysYGVlhS1btqCgoEDrfk6ePAkAWLVqFZKSkqTXmzdvxsSJE/H+++/j/PnzGDt2LEaMGIH9+/drvH/mzJl46aWXcO7cOYwcObLM+FevXkWfPn0waNAgxMTE4Ndff8WRI0cwfvx4AMCpU6cwYcIEzJ49G3Fxcdi5cye6deumNVaiGqVSz9gmohohJCREDBgwQAghRMeOHcXIkSOFEEJs3rxZPPhjYcaMGaJFixYa7124cKHw9PTUGMvT01MUFxdLbQ0bNhRdu3aVXhcVFQlLS0uxfv16IYQQCQkJAoCYN2+e1EelUgk3Nzfx5ZdfCiGE+Oyzz0Tv3r019n3z5k0BQMTFxQkhhAgMDBStWrV67PG6urqKOXPmaLS1a9dOvPvuu9LrFi1aiBkzZjxynN9//13Y2dkJMzMz0alTJzF16lRx9uxZjT4AxObNmzXaOnXqJN5++22NtldffVW88MILGu8LCwvT6LN//34BQNy7d08IIcSoUaPEmDFjNPocPnxYyOVycf/+ffHHH38IpVIpsrKyHnkcRDUNZ2aIarkvv/wSa9asQWxs7BOP4e/vD7n8fz9OnJyc0KxZM+m1kZER7O3tkZqaqvG+gIAA6f+NjY3Rtm1bKY6zZ89i//790oyIlZUVGjVqBKBkhqJUmzZtHhlbVlYWbt++jc6dO2u0d+7cudLHPGjQINy+fRt//fUX+vTpgwMHDqB169ZYvXr1I98XGxtbof23bdv2keOcPXsWq1ev1shJcHAw1Go1EhIS0KtXL3h6eqJ+/fp48803sXbtWuTl5VXqGIkMEYsZolquW7duCA4OxtSpU8tsk8vlEEJotKlUqjL9TExMNF7LZDKtbWq1usJx5eTkoF+/foiOjtb4io+P1zh1YmlpWeExdcHMzAy9evXCp59+imPHjmH48OGYMWOGTsZ+3LHk5ORg7NixGvk4e/Ys4uPj4ePjA2tra5w5cwbr16+Hi4sLpk+fjhYtWvDSbqrxWMwQEebNm4e///4bERERGu1169ZFcnKyRkGjy3vDHD9+XPr/oqIinD59Go0bNwYAtG7dGhcuXICXlxd8fX01vipTwCiVSri6umpcPg0AR48eRZMmTZ76GJo0aaKxANfExATFxcUafRo3bqyT/bdu3RoXL14skw9fX18oFAoAJTNcQUFBmD9/PmJiYpCYmIh9+/Y94dERGQZjfQdARPrXrFkzDBs2DEuWLNFo7969O9LS0jB//ny88sor2LlzJ3bs2AGlUqmT/S5btgx+fn5o3LgxFi5ciHv37kkLX0NDQ/HDDz9gyJAhmDJlCurUqYMrV65gw4YN+PHHH2FkZFTh/UyePBkzZsyAj48PWrZsiVWrViE6Ohpr166t8Bh3797Fq6++ipEjR6J58+awtrbGqVOnMH/+fAwYMEDq5+Xlhb1796Jz584wNTWFnZ0dJk+ejNdeew2tWrVCUFAQ/v77b2zatAl79uypeLIAfPjhh+jYsSPGjx+P0aNHw9LSEhcvXkR4eDiWLl2KrVu34tq1a+jWrRvs7Oywfft2qNVqNGzYsFL7ITI0nJkhIgDA7Nmzy5wGaty4Mb777jssW7YMLVq0wIkTJ7Re6fOk5s2bh3nz5qFFixY4cuQI/vrrLzg4OACANJtSXFyM3r17o1mzZggLC4Otra3G+pyKmDBhAiZNmoT3338fzZo1w86dO/HXX3/Bz8+vwmNYWVmhQ4cOWLhwIbp164amTZvi008/xdtvv42lS5dK/b7++muEh4fD3d0drVq1AgAMHDgQixcvxoIFC+Dv74/vv/8eq1atQvfu3St1HM2bN8fBgwdx+fJldO3aFa1atcL06dPh6uoKALC1tcWmTZvw3HPPoXHjxlixYgXWr18Pf3//Su2HyNDIxMMnxImIiIgMCGdmiIiIyKCxmCEiIiKDxmKGiIiIDBqLGSIiIjJoLGaIiIjIoLGYISIiIoPGYoaIiIgMGosZIiIiMmgsZoiIiMigsZghIiIig8ZihoiIiAza/wNpSyAN5qR/9QAAAABJRU5ErkJggg==",
      "text/plain": [
       "<Figure size 600x400 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import json\n",
    "import matplotlib.pyplot as plt\n",
    "from collections import Counter\n",
    "from tqdm import tqdm\n",
    "from make_hf_tokenizer import make_tokenizer\n",
    "import numpy as np\n",
    "\n",
    "def plot_unique_words_and_tokens(jsonl_filepath, n):\n",
    "    unique_words = []\n",
    "    unique_tokens = []\n",
    "    word_counts = Counter()\n",
    "    token_counts = Counter()\n",
    "    \n",
    "    tokenizer = make_tokenizer([\"vocab/ch-hf.txt\", \"vocab/wordbank-en.txt\"])\n",
    "\n",
    "    unk_counter = []\n",
    "    with open(jsonl_filepath, 'r') as file:\n",
    "        for i, line in tqdm(enumerate(file)):\n",
    "            if i < 65000:\n",
    "                continue\n",
    "            if i >= n:\n",
    "                break\n",
    "            data = json.loads(line)\n",
    "            story = data['story']\n",
    "            \n",
    "            words = story.split()\n",
    "            word_counts.update(words)\n",
    "            \"\"\"tokens = tokenizer.encode(story).tokens\n",
    "            # counting number of unk\n",
    "            unk_counter.append(tokens.count('[UNK]'))\n",
    "            token_counts.update(tokens)\"\"\"\n",
    "            \n",
    "            unique_words.append(len(word_counts))\n",
    "            unique_tokens.append(len(token_counts))\n",
    "    with open(jsonl_filepath, 'r') as file:\n",
    "        for i, line in tqdm(enumerate(file)):\n",
    "            if i >= 65000:\n",
    "                break\n",
    "            data = json.loads(line)\n",
    "            story = data['story']\n",
    "            \n",
    "            words = story.split()\n",
    "            word_counts.update(words)\n",
    "            \"\"\"tokens = tokenizer.encode(story).tokens\n",
    "            # counting number of unk\n",
    "            unk_counter.append(tokens.count('[UNK]'))\n",
    "            token_counts.update(tokens)\"\"\"\n",
    "            \n",
    "            unique_words.append(len(word_counts))\n",
    "            unique_tokens.append(len(token_counts))\n",
    "\n",
    "    print(f\"Fraction of stories with [UNK] tokens: {sum(np.array(unk_counter) > 0)/n}\")\n",
    "    print(f\"Average number of [UNK] tokens: {np.mean(unk_counter)}\")\n",
    "    # Plot the results\n",
    "    for log in [True, False]:\n",
    "        plt.figure(figsize=(6, 4))\n",
    "        plt.plot(range(1, n + 1), unique_words, label='Unique Words')\n",
    "        #plt.plot(range(1, n + 1), unique_tokens, label='Unique Custom Tokenizer Tokens')\n",
    "        plt.xlabel('Number of Stories')\n",
    "        plt.ylabel('Unique Count')\n",
    "        if log:\n",
    "            plt.xscale('log')\n",
    "            plt.yscale('log')\n",
    "            plt.xlim(1, 10**7)\n",
    "            plt.ylim(100,10**5)\n",
    "        plt.title('Unique Words in the First N Stories')\n",
    "        plt.legend()\n",
    "        plt.grid(True)\n",
    "        plt.show()\n",
    "\n",
    "n = 600000\n",
    "plot_unique_words_and_tokens(file_path, n)\n"
   ]
  }
 ],
 "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": "3.12.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}