00_Predicting_star_rating_initial_direction_Raju.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "ca_restaurant_reviews=pd.read_csv(\"california_restaurants.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "ca_restaurant_reviews=ca_restaurant_reviews.reset_index(drop = True)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Unnamed: 0</th>\n",
       "      <th>business_id</th>\n",
       "      <th>name</th>\n",
       "      <th>address</th>\n",
       "      <th>city</th>\n",
       "      <th>state</th>\n",
       "      <th>postal_code</th>\n",
       "      <th>latitude</th>\n",
       "      <th>longitude</th>\n",
       "      <th>stars</th>\n",
       "      <th>review_count</th>\n",
       "      <th>is_open</th>\n",
       "      <th>attributes</th>\n",
       "      <th>categories</th>\n",
       "      <th>hours</th>\n",
       "      <th>user_id</th>\n",
       "      <th>review_stars</th>\n",
       "      <th>text</th>\n",
       "      <th>date</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2456</td>\n",
       "      <td>IDtLPgUrqorrpqSLdfMhZQ</td>\n",
       "      <td>Helena Avenue Bakery</td>\n",
       "      <td>131 Anacapa St, Ste C</td>\n",
       "      <td>Santa Barbara</td>\n",
       "      <td>CA</td>\n",
       "      <td>93101</td>\n",
       "      <td>34.414445</td>\n",
       "      <td>-119.690672</td>\n",
       "      <td>4.0</td>\n",
       "      <td>389</td>\n",
       "      <td>1</td>\n",
       "      <td>{'RestaurantsTakeOut': 'True', 'NoiseLevel': \"...</td>\n",
       "      <td>Restaurants</td>\n",
       "      <td>{'Monday': '0:0-0:0', 'Tuesday': '8:0-14:0', '...</td>\n",
       "      <td>bFPdtzu11Oi0f92EAcjqmg</td>\n",
       "      <td>5</td>\n",
       "      <td>What a great addition to the Funk Zone!  Grab ...</td>\n",
       "      <td>2016-10-13 22:50:47</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Unnamed: 0             business_id                  name  \\\n",
       "0        2456  IDtLPgUrqorrpqSLdfMhZQ  Helena Avenue Bakery   \n",
       "\n",
       "                 address           city state  postal_code   latitude  \\\n",
       "0  131 Anacapa St, Ste C  Santa Barbara    CA        93101  34.414445   \n",
       "\n",
       "    longitude  stars  review_count  is_open  \\\n",
       "0 -119.690672    4.0           389        1   \n",
       "\n",
       "                                          attributes   categories  \\\n",
       "0  {'RestaurantsTakeOut': 'True', 'NoiseLevel': \"...  Restaurants   \n",
       "\n",
       "                                               hours                 user_id  \\\n",
       "0  {'Monday': '0:0-0:0', 'Tuesday': '8:0-14:0', '...  bFPdtzu11Oi0f92EAcjqmg   \n",
       "\n",
       "   review_stars                                               text  \\\n",
       "0             5  What a great addition to the Funk Zone!  Grab ...   \n",
       "\n",
       "                  date  \n",
       "0  2016-10-13 22:50:47  "
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ca_restaurant_reviews.head(1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "import nltk\n",
    "from nltk.corpus import stopwords\n",
    "from tqdm import tqdm\n",
    "from tqdm import tqdm_notebook\n",
    "\n",
    "from collections import Counter"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "import string\n",
    "def clean_text(sample_review):\n",
    "    stopwords = nltk.corpus.stopwords.words('english')\n",
    "    newStopWords = ['ive','hadnt','couldnt','didnt', 'id'] \n",
    "    stopwords.extend(newStopWords)\n",
    "    text = sample_review\n",
    "\n",
    "    text = text[2: len(sample_review)-1].lower()\n",
    "    text = text.replace('\\\\n', ' ').replace('\\\\t', ' ')\n",
    "    nopunc = [char for char in text if char not in string.punctuation]\n",
    "    nopunc = ''.join(nopunc)\n",
    "\n",
    "    \n",
    "    l = [word for word in nopunc.split() if word.lower() not in stopwords]\n",
    "    clean_text = \"\"\n",
    "    for word in l:\n",
    "        clean_text += str(word)+\" \"\n",
    "    \n",
    "    return clean_text.strip()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'place short walk hotel wonderful find awesome ambiance pleasant service lemon blueberry scone best ever anywhere full berries ham cheese croissant delicious took fresh made bread us take wine tasting later afternoon also noticed located near figueroa mountain brewing co perfect want sandwich beer'"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sample_review = ca_restaurant_reviews.text[20]\n",
    "display(clean_text(sample_review))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "import string\n",
    "def get_words(text):\n",
    "\n",
    "    stopwords = nltk.corpus.stopwords.words('english')\n",
    "    newStopWords = ['ive','hadnt','couldnt','didnt', 'id'] \n",
    "    stopwords.extend(newStopWords)\n",
    "    text = text[2: len(sample_review)-1].lower()  \n",
    "    text = text.replace('\\\\n', ' ').replace('\\\\t', ' ')\n",
    "    display(text)\n",
    "    nopunc = [char for char in text if char not in string.punctuation]\n",
    "    nopunc = ''.join(nopunc)\n",
    "    \n",
    "    display(nopunc)\n",
    "    \n",
    "    l = [word for word in nopunc.split() if word.lower() not in stopwords]\n",
    "    \n",
    "    return l, len(l)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'at a great addition to the funk zone!  grab a bite, grab some tastings, life is good. right next door to the santa barbara wine collective, in fact it actually shares the same tables.  we had a fabulous savory croissant'"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "'at a great addition to the funk zone  grab a bite grab some tastings life is good right next door to the santa barbara wine collective in fact it actually shares the same tables  we had a fabulous savory croissant'"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "['great',\n",
       " 'addition',\n",
       " 'funk',\n",
       " 'zone',\n",
       " 'grab',\n",
       " 'bite',\n",
       " 'grab',\n",
       " 'tastings',\n",
       " 'life',\n",
       " 'good',\n",
       " 'right',\n",
       " 'next',\n",
       " 'door',\n",
       " 'santa',\n",
       " 'barbara',\n",
       " 'wine',\n",
       " 'collective',\n",
       " 'fact',\n",
       " 'actually',\n",
       " 'shares',\n",
       " 'tables',\n",
       " 'fabulous',\n",
       " 'savory',\n",
       " 'croissant']"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "for i in range(1):\n",
    "    sample_review = str(ca_restaurant_reviews.text[i])\n",
    "    check = get_words(sample_review)\n",
    "    display(check[0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Unnamed: 0</th>\n",
       "      <th>postal_code</th>\n",
       "      <th>latitude</th>\n",
       "      <th>longitude</th>\n",
       "      <th>stars</th>\n",
       "      <th>review_count</th>\n",
       "      <th>is_open</th>\n",
       "      <th>review_stars</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>2.05e+05</td>\n",
       "      <td>205464.00</td>\n",
       "      <td>205464.00</td>\n",
       "      <td>205464.00</td>\n",
       "      <td>205464.00</td>\n",
       "      <td>205464.00</td>\n",
       "      <td>205464.00</td>\n",
       "      <td>205464.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>2.36e+06</td>\n",
       "      <td>93101.13</td>\n",
       "      <td>34.42</td>\n",
       "      <td>-119.72</td>\n",
       "      <td>3.95</td>\n",
       "      <td>667.85</td>\n",
       "      <td>0.81</td>\n",
       "      <td>3.95</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>1.34e+06</td>\n",
       "      <td>41.12</td>\n",
       "      <td>0.06</td>\n",
       "      <td>0.08</td>\n",
       "      <td>0.51</td>\n",
       "      <td>736.33</td>\n",
       "      <td>0.39</td>\n",
       "      <td>1.34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>2.46e+03</td>\n",
       "      <td>93013.00</td>\n",
       "      <td>34.39</td>\n",
       "      <td>-120.08</td>\n",
       "      <td>1.50</td>\n",
       "      <td>5.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>1.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>1.23e+06</td>\n",
       "      <td>93101.00</td>\n",
       "      <td>34.41</td>\n",
       "      <td>-119.74</td>\n",
       "      <td>3.50</td>\n",
       "      <td>200.00</td>\n",
       "      <td>1.00</td>\n",
       "      <td>3.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>2.39e+06</td>\n",
       "      <td>93101.00</td>\n",
       "      <td>34.42</td>\n",
       "      <td>-119.70</td>\n",
       "      <td>4.00</td>\n",
       "      <td>405.00</td>\n",
       "      <td>1.00</td>\n",
       "      <td>4.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>3.63e+06</td>\n",
       "      <td>93109.00</td>\n",
       "      <td>34.43</td>\n",
       "      <td>-119.69</td>\n",
       "      <td>4.50</td>\n",
       "      <td>885.00</td>\n",
       "      <td>1.00</td>\n",
       "      <td>5.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>4.56e+06</td>\n",
       "      <td>96161.00</td>\n",
       "      <td>39.38</td>\n",
       "      <td>-119.51</td>\n",
       "      <td>5.00</td>\n",
       "      <td>3834.00</td>\n",
       "      <td>1.00</td>\n",
       "      <td>5.00</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       Unnamed: 0  postal_code   latitude  longitude      stars  review_count  \\\n",
       "count    2.05e+05    205464.00  205464.00  205464.00  205464.00     205464.00   \n",
       "mean     2.36e+06     93101.13      34.42    -119.72       3.95        667.85   \n",
       "std      1.34e+06        41.12       0.06       0.08       0.51        736.33   \n",
       "min      2.46e+03     93013.00      34.39    -120.08       1.50          5.00   \n",
       "25%      1.23e+06     93101.00      34.41    -119.74       3.50        200.00   \n",
       "50%      2.39e+06     93101.00      34.42    -119.70       4.00        405.00   \n",
       "75%      3.63e+06     93109.00      34.43    -119.69       4.50        885.00   \n",
       "max      4.56e+06     96161.00      39.38    -119.51       5.00       3834.00   \n",
       "\n",
       "         is_open  review_stars  \n",
       "count  205464.00     205464.00  \n",
       "mean        0.81          3.95  \n",
       "std         0.39          1.34  \n",
       "min         0.00          1.00  \n",
       "25%         1.00          3.00  \n",
       "50%         1.00          4.00  \n",
       "75%         1.00          5.00  \n",
       "max         1.00          5.00  "
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.set_option('display.precision', 2)\n",
    "ca_restaurant_reviews.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "5    102470\n",
      "4     45823\n",
      "3     21149\n",
      "1     20272\n",
      "2     15750\n",
      "Name: review_stars, dtype: int64\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "pandas.core.series.Series"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print(ca_restaurant_reviews[\"review_stars\"].value_counts())\n",
    "type(ca_restaurant_reviews[\"review_stars\"].value_counts())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "labels = \"5 stars\",\"4 Stars\", \"3 Stars\",\"1 Star\",\"2 Stars\"\n",
    "sizes = ca_restaurant_reviews[\"review_stars\"].value_counts()\n",
    "colors = ['lightpink', 'yellowgreen', 'orange', 'lightskyblue','green']\n",
    " \n",
    "# Plot\n",
    "plt.pie(sizes, labels=labels,colors =colors, autopct='%1.1f%%') \n",
    "# plt.axis('equal')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "from tqdm import tqdm"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|█████████▉| 205464/205465 [01:09<00:00, 2963.59it/s]\n"
     ]
    }
   ],
   "source": [
    "texts = []\n",
    "stars = [ca_restaurant_reviews['review_stars'] for review in ca_restaurant_reviews]\n",
    "pbar = tqdm(total=ca_restaurant_reviews.shape[0]+1)\n",
    "for index, row in ca_restaurant_reviews.iterrows():\n",
    "    texts.append(clean_text(row['text']))\n",
    "    pbar.update(1)\n",
    "pbar.close()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Wall time: 1min 7s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "from sklearn.feature_extraction.text import TfidfVectorizer\n",
    "vectorizer = TfidfVectorizer(ngram_range=(1,3))\n",
    "vectors = vectorizer.fit_transform(texts)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "X_train, X_test, y_train, y_test = train_test_split(vectors, stars[1], test_size=0.15, random_state=42, shuffle =False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Wall time: 51.1 s\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "LinearSVC()"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "%%time\n",
    "from sklearn.svm import LinearSVC\n",
    "classifier = LinearSVC()\n",
    "\n",
    "classifier.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Actual Ratings(Stars): "
     ]
    },
    {
     "data": {
      "text/plain": [
       "174644    5\n",
       "174645    3\n",
       "174646    5\n",
       "174647    4\n",
       "174648    4\n",
       "Name: review_stars, dtype: int64"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Predicted Ratings: [5 5 5 5 4]\n"
     ]
    }
   ],
   "source": [
    "preds = classifier.predict(X_test)\n",
    "print(\"Actual Ratings(Stars): \",end = \"\")\n",
    "display(y_test[:5])\n",
    "print(\"Predicted Ratings: \",end = \"\")\n",
    "print(preds[:5])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_null, X_full_test, y_null, y_full_test = train_test_split(vectors, stars[1], test_size=0.999995, random_state=42, shuffle = False)\n",
    "predict_all = classifier.predict(X_full_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Actual Ratings(Stars): \n",
      "154731    1\n",
      "154732    3\n",
      "154733    3\n",
      "154734    4\n",
      "154735    2\n",
      "154736    3\n",
      "Name: review_stars, dtype: int64\n",
      "\n",
      "Predicted Ratings: [1, 3, 3, 4, 2, 3]\n"
     ]
    }
   ],
   "source": [
    "predicted_stars = list(predict_all)\n",
    "\n",
    "print(\"Actual Ratings(Stars): \")\n",
    "print(y_full_test[154730:154736])\n",
    "print(\"\\nPredicted Ratings: \",end = \"\")\n",
    "print(predicted_stars[154730:154736])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(205464, 19)"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ca_restaurant_reviews.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "predicted_stars.append(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Original Reviews (with user bias)\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Unnamed: 0</th>\n",
       "      <th>business_id</th>\n",
       "      <th>name</th>\n",
       "      <th>address</th>\n",
       "      <th>city</th>\n",
       "      <th>state</th>\n",
       "      <th>postal_code</th>\n",
       "      <th>latitude</th>\n",
       "      <th>longitude</th>\n",
       "      <th>stars</th>\n",
       "      <th>review_count</th>\n",
       "      <th>is_open</th>\n",
       "      <th>attributes</th>\n",
       "      <th>categories</th>\n",
       "      <th>hours</th>\n",
       "      <th>user_id</th>\n",
       "      <th>review_stars</th>\n",
       "      <th>text</th>\n",
       "      <th>date</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>205454</th>\n",
       "      <td>4559987</td>\n",
       "      <td>uriD7RFuHhLJeDdKaf0nFA</td>\n",
       "      <td>Pizza Guru</td>\n",
       "      <td>3534 State St</td>\n",
       "      <td>Santa Barbara</td>\n",
       "      <td>CA</td>\n",
       "      <td>93105</td>\n",
       "      <td>34.44</td>\n",
       "      <td>-119.74</td>\n",
       "      <td>4.0</td>\n",
       "      <td>299</td>\n",
       "      <td>0</td>\n",
       "      <td>{'NoiseLevel': \"u'average'\", 'RestaurantsGoodF...</td>\n",
       "      <td>Restaurants</td>\n",
       "      <td>{'Monday': '15:0-21:0', 'Tuesday': '15:0-21:0'...</td>\n",
       "      <td>qW-Q3krGy2FgY1ouN0znjQ</td>\n",
       "      <td>5</td>\n",
       "      <td>Santa Barbara has lots of good pizza. \\nThis o...</td>\n",
       "      <td>2018-10-18 20:35:16</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>205455</th>\n",
       "      <td>4559988</td>\n",
       "      <td>uriD7RFuHhLJeDdKaf0nFA</td>\n",
       "      <td>Pizza Guru</td>\n",
       "      <td>3534 State St</td>\n",
       "      <td>Santa Barbara</td>\n",
       "      <td>CA</td>\n",
       "      <td>93105</td>\n",
       "      <td>34.44</td>\n",
       "      <td>-119.74</td>\n",
       "      <td>4.0</td>\n",
       "      <td>299</td>\n",
       "      <td>0</td>\n",
       "      <td>{'NoiseLevel': \"u'average'\", 'RestaurantsGoodF...</td>\n",
       "      <td>Restaurants</td>\n",
       "      <td>{'Monday': '15:0-21:0', 'Tuesday': '15:0-21:0'...</td>\n",
       "      <td>X0NlTMV_-DNpGM0sPFa_cA</td>\n",
       "      <td>5</td>\n",
       "      <td>Great pizza with good beer. Stopped in and we ...</td>\n",
       "      <td>2015-05-02 22:25:45</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>205456</th>\n",
       "      <td>4559989</td>\n",
       "      <td>uriD7RFuHhLJeDdKaf0nFA</td>\n",
       "      <td>Pizza Guru</td>\n",
       "      <td>3534 State St</td>\n",
       "      <td>Santa Barbara</td>\n",
       "      <td>CA</td>\n",
       "      <td>93105</td>\n",
       "      <td>34.44</td>\n",
       "      <td>-119.74</td>\n",
       "      <td>4.0</td>\n",
       "      <td>299</td>\n",
       "      <td>0</td>\n",
       "      <td>{'NoiseLevel': \"u'average'\", 'RestaurantsGoodF...</td>\n",
       "      <td>Restaurants</td>\n",
       "      <td>{'Monday': '15:0-21:0', 'Tuesday': '15:0-21:0'...</td>\n",
       "      <td>_ymxPyUR2_tuoPJP8t67gQ</td>\n",
       "      <td>1</td>\n",
       "      <td>forgot to write a review until now- didn't hav...</td>\n",
       "      <td>2013-09-16 12:11:56</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>205457</th>\n",
       "      <td>4559990</td>\n",
       "      <td>uriD7RFuHhLJeDdKaf0nFA</td>\n",
       "      <td>Pizza Guru</td>\n",
       "      <td>3534 State St</td>\n",
       "      <td>Santa Barbara</td>\n",
       "      <td>CA</td>\n",
       "      <td>93105</td>\n",
       "      <td>34.44</td>\n",
       "      <td>-119.74</td>\n",
       "      <td>4.0</td>\n",
       "      <td>299</td>\n",
       "      <td>0</td>\n",
       "      <td>{'NoiseLevel': \"u'average'\", 'RestaurantsGoodF...</td>\n",
       "      <td>Restaurants</td>\n",
       "      <td>{'Monday': '15:0-21:0', 'Tuesday': '15:0-21:0'...</td>\n",
       "      <td>avcQbkVRQ0skWZS1rkoAQg</td>\n",
       "      <td>4</td>\n",
       "      <td>Solid pizza and beer at Pizza Guru.  I choose ...</td>\n",
       "      <td>2018-07-22 13:31:09</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>205458</th>\n",
       "      <td>4559991</td>\n",
       "      <td>uriD7RFuHhLJeDdKaf0nFA</td>\n",
       "      <td>Pizza Guru</td>\n",
       "      <td>3534 State St</td>\n",
       "      <td>Santa Barbara</td>\n",
       "      <td>CA</td>\n",
       "      <td>93105</td>\n",
       "      <td>34.44</td>\n",
       "      <td>-119.74</td>\n",
       "      <td>4.0</td>\n",
       "      <td>299</td>\n",
       "      <td>0</td>\n",
       "      <td>{'NoiseLevel': \"u'average'\", 'RestaurantsGoodF...</td>\n",
       "      <td>Restaurants</td>\n",
       "      <td>{'Monday': '15:0-21:0', 'Tuesday': '15:0-21:0'...</td>\n",
       "      <td>agYoN0HOl6KBkAkIZaJmFg</td>\n",
       "      <td>3</td>\n",
       "      <td>3.4 star review. Nice concept. I like the loca...</td>\n",
       "      <td>2014-12-14 16:41:07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>205459</th>\n",
       "      <td>4559992</td>\n",
       "      <td>uriD7RFuHhLJeDdKaf0nFA</td>\n",
       "      <td>Pizza Guru</td>\n",
       "      <td>3534 State St</td>\n",
       "      <td>Santa Barbara</td>\n",
       "      <td>CA</td>\n",
       "      <td>93105</td>\n",
       "      <td>34.44</td>\n",
       "      <td>-119.74</td>\n",
       "      <td>4.0</td>\n",
       "      <td>299</td>\n",
       "      <td>0</td>\n",
       "      <td>{'NoiseLevel': \"u'average'\", 'RestaurantsGoodF...</td>\n",
       "      <td>Restaurants</td>\n",
       "      <td>{'Monday': '15:0-21:0', 'Tuesday': '15:0-21:0'...</td>\n",
       "      <td>o72LSEPHKu1oQcpXeIVlag</td>\n",
       "      <td>5</td>\n",
       "      <td>This is one of my new favorite pizza joints he...</td>\n",
       "      <td>2015-04-11 04:43:19</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>205460</th>\n",
       "      <td>4559993</td>\n",
       "      <td>uriD7RFuHhLJeDdKaf0nFA</td>\n",
       "      <td>Pizza Guru</td>\n",
       "      <td>3534 State St</td>\n",
       "      <td>Santa Barbara</td>\n",
       "      <td>CA</td>\n",
       "      <td>93105</td>\n",
       "      <td>34.44</td>\n",
       "      <td>-119.74</td>\n",
       "      <td>4.0</td>\n",
       "      <td>299</td>\n",
       "      <td>0</td>\n",
       "      <td>{'NoiseLevel': \"u'average'\", 'RestaurantsGoodF...</td>\n",
       "      <td>Restaurants</td>\n",
       "      <td>{'Monday': '15:0-21:0', 'Tuesday': '15:0-21:0'...</td>\n",
       "      <td>u30sv3s4V2tz5VOgNQYtgg</td>\n",
       "      <td>4</td>\n",
       "      <td>Ordered vegan chicken and Daiya cheese and onl...</td>\n",
       "      <td>2015-05-26 15:27:36</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>205461</th>\n",
       "      <td>4559994</td>\n",
       "      <td>uriD7RFuHhLJeDdKaf0nFA</td>\n",
       "      <td>Pizza Guru</td>\n",
       "      <td>3534 State St</td>\n",
       "      <td>Santa Barbara</td>\n",
       "      <td>CA</td>\n",
       "      <td>93105</td>\n",
       "      <td>34.44</td>\n",
       "      <td>-119.74</td>\n",
       "      <td>4.0</td>\n",
       "      <td>299</td>\n",
       "      <td>0</td>\n",
       "      <td>{'NoiseLevel': \"u'average'\", 'RestaurantsGoodF...</td>\n",
       "      <td>Restaurants</td>\n",
       "      <td>{'Monday': '15:0-21:0', 'Tuesday': '15:0-21:0'...</td>\n",
       "      <td>Mp97BPH2Y7O1DsFt6wFp4Q</td>\n",
       "      <td>1</td>\n",
       "      <td>This was the worst pizza I've ever had ever ea...</td>\n",
       "      <td>2019-07-08 19:27:04</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>205462</th>\n",
       "      <td>4559995</td>\n",
       "      <td>uriD7RFuHhLJeDdKaf0nFA</td>\n",
       "      <td>Pizza Guru</td>\n",
       "      <td>3534 State St</td>\n",
       "      <td>Santa Barbara</td>\n",
       "      <td>CA</td>\n",
       "      <td>93105</td>\n",
       "      <td>34.44</td>\n",
       "      <td>-119.74</td>\n",
       "      <td>4.0</td>\n",
       "      <td>299</td>\n",
       "      <td>0</td>\n",
       "      <td>{'NoiseLevel': \"u'average'\", 'RestaurantsGoodF...</td>\n",
       "      <td>Restaurants</td>\n",
       "      <td>{'Monday': '15:0-21:0', 'Tuesday': '15:0-21:0'...</td>\n",
       "      <td>06egjh3nrEb7EXX1VYZOCg</td>\n",
       "      <td>5</td>\n",
       "      <td>One of the best pizzas in SB....seriously!!!\\n...</td>\n",
       "      <td>2014-07-07 18:40:39</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>205463</th>\n",
       "      <td>4559996</td>\n",
       "      <td>uriD7RFuHhLJeDdKaf0nFA</td>\n",
       "      <td>Pizza Guru</td>\n",
       "      <td>3534 State St</td>\n",
       "      <td>Santa Barbara</td>\n",
       "      <td>CA</td>\n",
       "      <td>93105</td>\n",
       "      <td>34.44</td>\n",
       "      <td>-119.74</td>\n",
       "      <td>4.0</td>\n",
       "      <td>299</td>\n",
       "      <td>0</td>\n",
       "      <td>{'NoiseLevel': \"u'average'\", 'RestaurantsGoodF...</td>\n",
       "      <td>Restaurants</td>\n",
       "      <td>{'Monday': '15:0-21:0', 'Tuesday': '15:0-21:0'...</td>\n",
       "      <td>vTjt5vVDuW-Xu1Z_tfraTQ</td>\n",
       "      <td>5</td>\n",
       "      <td>We wanted a calzone and it was getting around ...</td>\n",
       "      <td>2016-11-06 05:15:44</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        Unnamed: 0             business_id        name        address  \\\n",
       "205454     4559987  uriD7RFuHhLJeDdKaf0nFA  Pizza Guru  3534 State St   \n",
       "205455     4559988  uriD7RFuHhLJeDdKaf0nFA  Pizza Guru  3534 State St   \n",
       "205456     4559989  uriD7RFuHhLJeDdKaf0nFA  Pizza Guru  3534 State St   \n",
       "205457     4559990  uriD7RFuHhLJeDdKaf0nFA  Pizza Guru  3534 State St   \n",
       "205458     4559991  uriD7RFuHhLJeDdKaf0nFA  Pizza Guru  3534 State St   \n",
       "205459     4559992  uriD7RFuHhLJeDdKaf0nFA  Pizza Guru  3534 State St   \n",
       "205460     4559993  uriD7RFuHhLJeDdKaf0nFA  Pizza Guru  3534 State St   \n",
       "205461     4559994  uriD7RFuHhLJeDdKaf0nFA  Pizza Guru  3534 State St   \n",
       "205462     4559995  uriD7RFuHhLJeDdKaf0nFA  Pizza Guru  3534 State St   \n",
       "205463     4559996  uriD7RFuHhLJeDdKaf0nFA  Pizza Guru  3534 State St   \n",
       "\n",
       "                 city state  postal_code  latitude  longitude  stars  \\\n",
       "205454  Santa Barbara    CA        93105     34.44    -119.74    4.0   \n",
       "205455  Santa Barbara    CA        93105     34.44    -119.74    4.0   \n",
       "205456  Santa Barbara    CA        93105     34.44    -119.74    4.0   \n",
       "205457  Santa Barbara    CA        93105     34.44    -119.74    4.0   \n",
       "205458  Santa Barbara    CA        93105     34.44    -119.74    4.0   \n",
       "205459  Santa Barbara    CA        93105     34.44    -119.74    4.0   \n",
       "205460  Santa Barbara    CA        93105     34.44    -119.74    4.0   \n",
       "205461  Santa Barbara    CA        93105     34.44    -119.74    4.0   \n",
       "205462  Santa Barbara    CA        93105     34.44    -119.74    4.0   \n",
       "205463  Santa Barbara    CA        93105     34.44    -119.74    4.0   \n",
       "\n",
       "        review_count  is_open  \\\n",
       "205454           299        0   \n",
       "205455           299        0   \n",
       "205456           299        0   \n",
       "205457           299        0   \n",
       "205458           299        0   \n",
       "205459           299        0   \n",
       "205460           299        0   \n",
       "205461           299        0   \n",
       "205462           299        0   \n",
       "205463           299        0   \n",
       "\n",
       "                                               attributes   categories  \\\n",
       "205454  {'NoiseLevel': \"u'average'\", 'RestaurantsGoodF...  Restaurants   \n",
       "205455  {'NoiseLevel': \"u'average'\", 'RestaurantsGoodF...  Restaurants   \n",
       "205456  {'NoiseLevel': \"u'average'\", 'RestaurantsGoodF...  Restaurants   \n",
       "205457  {'NoiseLevel': \"u'average'\", 'RestaurantsGoodF...  Restaurants   \n",
       "205458  {'NoiseLevel': \"u'average'\", 'RestaurantsGoodF...  Restaurants   \n",
       "205459  {'NoiseLevel': \"u'average'\", 'RestaurantsGoodF...  Restaurants   \n",
       "205460  {'NoiseLevel': \"u'average'\", 'RestaurantsGoodF...  Restaurants   \n",
       "205461  {'NoiseLevel': \"u'average'\", 'RestaurantsGoodF...  Restaurants   \n",
       "205462  {'NoiseLevel': \"u'average'\", 'RestaurantsGoodF...  Restaurants   \n",
       "205463  {'NoiseLevel': \"u'average'\", 'RestaurantsGoodF...  Restaurants   \n",
       "\n",
       "                                                    hours  \\\n",
       "205454  {'Monday': '15:0-21:0', 'Tuesday': '15:0-21:0'...   \n",
       "205455  {'Monday': '15:0-21:0', 'Tuesday': '15:0-21:0'...   \n",
       "205456  {'Monday': '15:0-21:0', 'Tuesday': '15:0-21:0'...   \n",
       "205457  {'Monday': '15:0-21:0', 'Tuesday': '15:0-21:0'...   \n",
       "205458  {'Monday': '15:0-21:0', 'Tuesday': '15:0-21:0'...   \n",
       "205459  {'Monday': '15:0-21:0', 'Tuesday': '15:0-21:0'...   \n",
       "205460  {'Monday': '15:0-21:0', 'Tuesday': '15:0-21:0'...   \n",
       "205461  {'Monday': '15:0-21:0', 'Tuesday': '15:0-21:0'...   \n",
       "205462  {'Monday': '15:0-21:0', 'Tuesday': '15:0-21:0'...   \n",
       "205463  {'Monday': '15:0-21:0', 'Tuesday': '15:0-21:0'...   \n",
       "\n",
       "                       user_id  review_stars  \\\n",
       "205454  qW-Q3krGy2FgY1ouN0znjQ             5   \n",
       "205455  X0NlTMV_-DNpGM0sPFa_cA             5   \n",
       "205456  _ymxPyUR2_tuoPJP8t67gQ             1   \n",
       "205457  avcQbkVRQ0skWZS1rkoAQg             4   \n",
       "205458  agYoN0HOl6KBkAkIZaJmFg             3   \n",
       "205459  o72LSEPHKu1oQcpXeIVlag             5   \n",
       "205460  u30sv3s4V2tz5VOgNQYtgg             4   \n",
       "205461  Mp97BPH2Y7O1DsFt6wFp4Q             1   \n",
       "205462  06egjh3nrEb7EXX1VYZOCg             5   \n",
       "205463  vTjt5vVDuW-Xu1Z_tfraTQ             5   \n",
       "\n",
       "                                                     text                 date  \n",
       "205454  Santa Barbara has lots of good pizza. \\nThis o...  2018-10-18 20:35:16  \n",
       "205455  Great pizza with good beer. Stopped in and we ...  2015-05-02 22:25:45  \n",
       "205456  forgot to write a review until now- didn't hav...  2013-09-16 12:11:56  \n",
       "205457  Solid pizza and beer at Pizza Guru.  I choose ...  2018-07-22 13:31:09  \n",
       "205458  3.4 star review. Nice concept. I like the loca...  2014-12-14 16:41:07  \n",
       "205459  This is one of my new favorite pizza joints he...  2015-04-11 04:43:19  \n",
       "205460  Ordered vegan chicken and Daiya cheese and onl...  2015-05-26 15:27:36  \n",
       "205461  This was the worst pizza I've ever had ever ea...  2019-07-08 19:27:04  \n",
       "205462  One of the best pizzas in SB....seriously!!!\\n...  2014-07-07 18:40:39  \n",
       "205463  We wanted a calzone and it was getting around ...  2016-11-06 05:15:44  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Unbiased Reviews (with predicted rating using user's review text)\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Unnamed: 0</th>\n",
       "      <th>business_id</th>\n",
       "      <th>name</th>\n",
       "      <th>address</th>\n",
       "      <th>city</th>\n",
       "      <th>state</th>\n",
       "      <th>postal_code</th>\n",
       "      <th>latitude</th>\n",
       "      <th>longitude</th>\n",
       "      <th>stars</th>\n",
       "      <th>review_count</th>\n",
       "      <th>is_open</th>\n",
       "      <th>attributes</th>\n",
       "      <th>categories</th>\n",
       "      <th>hours</th>\n",
       "      <th>user_id</th>\n",
       "      <th>review_stars</th>\n",
       "      <th>text</th>\n",
       "      <th>date</th>\n",
       "      <th>predict</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>205454</th>\n",
       "      <td>4559987</td>\n",
       "      <td>uriD7RFuHhLJeDdKaf0nFA</td>\n",
       "      <td>Pizza Guru</td>\n",
       "      <td>3534 State St</td>\n",
       "      <td>Santa Barbara</td>\n",
       "      <td>CA</td>\n",
       "      <td>93105</td>\n",
       "      <td>34.44</td>\n",
       "      <td>-119.74</td>\n",
       "      <td>4.0</td>\n",
       "      <td>299</td>\n",
       "      <td>0</td>\n",
       "      <td>{'NoiseLevel': \"u'average'\", 'RestaurantsGoodF...</td>\n",
       "      <td>Restaurants</td>\n",
       "      <td>{'Monday': '15:0-21:0', 'Tuesday': '15:0-21:0'...</td>\n",
       "      <td>qW-Q3krGy2FgY1ouN0znjQ</td>\n",
       "      <td>5</td>\n",
       "      <td>Santa Barbara has lots of good pizza. \\nThis o...</td>\n",
       "      <td>2018-10-18 20:35:16</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>205455</th>\n",
       "      <td>4559988</td>\n",
       "      <td>uriD7RFuHhLJeDdKaf0nFA</td>\n",
       "      <td>Pizza Guru</td>\n",
       "      <td>3534 State St</td>\n",
       "      <td>Santa Barbara</td>\n",
       "      <td>CA</td>\n",
       "      <td>93105</td>\n",
       "      <td>34.44</td>\n",
       "      <td>-119.74</td>\n",
       "      <td>4.0</td>\n",
       "      <td>299</td>\n",
       "      <td>0</td>\n",
       "      <td>{'NoiseLevel': \"u'average'\", 'RestaurantsGoodF...</td>\n",
       "      <td>Restaurants</td>\n",
       "      <td>{'Monday': '15:0-21:0', 'Tuesday': '15:0-21:0'...</td>\n",
       "      <td>X0NlTMV_-DNpGM0sPFa_cA</td>\n",
       "      <td>5</td>\n",
       "      <td>Great pizza with good beer. Stopped in and we ...</td>\n",
       "      <td>2015-05-02 22:25:45</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>205456</th>\n",
       "      <td>4559989</td>\n",
       "      <td>uriD7RFuHhLJeDdKaf0nFA</td>\n",
       "      <td>Pizza Guru</td>\n",
       "      <td>3534 State St</td>\n",
       "      <td>Santa Barbara</td>\n",
       "      <td>CA</td>\n",
       "      <td>93105</td>\n",
       "      <td>34.44</td>\n",
       "      <td>-119.74</td>\n",
       "      <td>4.0</td>\n",
       "      <td>299</td>\n",
       "      <td>0</td>\n",
       "      <td>{'NoiseLevel': \"u'average'\", 'RestaurantsGoodF...</td>\n",
       "      <td>Restaurants</td>\n",
       "      <td>{'Monday': '15:0-21:0', 'Tuesday': '15:0-21:0'...</td>\n",
       "      <td>_ymxPyUR2_tuoPJP8t67gQ</td>\n",
       "      <td>1</td>\n",
       "      <td>forgot to write a review until now- didn't hav...</td>\n",
       "      <td>2013-09-16 12:11:56</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>205457</th>\n",
       "      <td>4559990</td>\n",
       "      <td>uriD7RFuHhLJeDdKaf0nFA</td>\n",
       "      <td>Pizza Guru</td>\n",
       "      <td>3534 State St</td>\n",
       "      <td>Santa Barbara</td>\n",
       "      <td>CA</td>\n",
       "      <td>93105</td>\n",
       "      <td>34.44</td>\n",
       "      <td>-119.74</td>\n",
       "      <td>4.0</td>\n",
       "      <td>299</td>\n",
       "      <td>0</td>\n",
       "      <td>{'NoiseLevel': \"u'average'\", 'RestaurantsGoodF...</td>\n",
       "      <td>Restaurants</td>\n",
       "      <td>{'Monday': '15:0-21:0', 'Tuesday': '15:0-21:0'...</td>\n",
       "      <td>avcQbkVRQ0skWZS1rkoAQg</td>\n",
       "      <td>4</td>\n",
       "      <td>Solid pizza and beer at Pizza Guru.  I choose ...</td>\n",
       "      <td>2018-07-22 13:31:09</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>205458</th>\n",
       "      <td>4559991</td>\n",
       "      <td>uriD7RFuHhLJeDdKaf0nFA</td>\n",
       "      <td>Pizza Guru</td>\n",
       "      <td>3534 State St</td>\n",
       "      <td>Santa Barbara</td>\n",
       "      <td>CA</td>\n",
       "      <td>93105</td>\n",
       "      <td>34.44</td>\n",
       "      <td>-119.74</td>\n",
       "      <td>4.0</td>\n",
       "      <td>299</td>\n",
       "      <td>0</td>\n",
       "      <td>{'NoiseLevel': \"u'average'\", 'RestaurantsGoodF...</td>\n",
       "      <td>Restaurants</td>\n",
       "      <td>{'Monday': '15:0-21:0', 'Tuesday': '15:0-21:0'...</td>\n",
       "      <td>agYoN0HOl6KBkAkIZaJmFg</td>\n",
       "      <td>3</td>\n",
       "      <td>3.4 star review. Nice concept. I like the loca...</td>\n",
       "      <td>2014-12-14 16:41:07</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>205459</th>\n",
       "      <td>4559992</td>\n",
       "      <td>uriD7RFuHhLJeDdKaf0nFA</td>\n",
       "      <td>Pizza Guru</td>\n",
       "      <td>3534 State St</td>\n",
       "      <td>Santa Barbara</td>\n",
       "      <td>CA</td>\n",
       "      <td>93105</td>\n",
       "      <td>34.44</td>\n",
       "      <td>-119.74</td>\n",
       "      <td>4.0</td>\n",
       "      <td>299</td>\n",
       "      <td>0</td>\n",
       "      <td>{'NoiseLevel': \"u'average'\", 'RestaurantsGoodF...</td>\n",
       "      <td>Restaurants</td>\n",
       "      <td>{'Monday': '15:0-21:0', 'Tuesday': '15:0-21:0'...</td>\n",
       "      <td>o72LSEPHKu1oQcpXeIVlag</td>\n",
       "      <td>5</td>\n",
       "      <td>This is one of my new favorite pizza joints he...</td>\n",
       "      <td>2015-04-11 04:43:19</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>205460</th>\n",
       "      <td>4559993</td>\n",
       "      <td>uriD7RFuHhLJeDdKaf0nFA</td>\n",
       "      <td>Pizza Guru</td>\n",
       "      <td>3534 State St</td>\n",
       "      <td>Santa Barbara</td>\n",
       "      <td>CA</td>\n",
       "      <td>93105</td>\n",
       "      <td>34.44</td>\n",
       "      <td>-119.74</td>\n",
       "      <td>4.0</td>\n",
       "      <td>299</td>\n",
       "      <td>0</td>\n",
       "      <td>{'NoiseLevel': \"u'average'\", 'RestaurantsGoodF...</td>\n",
       "      <td>Restaurants</td>\n",
       "      <td>{'Monday': '15:0-21:0', 'Tuesday': '15:0-21:0'...</td>\n",
       "      <td>u30sv3s4V2tz5VOgNQYtgg</td>\n",
       "      <td>4</td>\n",
       "      <td>Ordered vegan chicken and Daiya cheese and onl...</td>\n",
       "      <td>2015-05-26 15:27:36</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>205461</th>\n",
       "      <td>4559994</td>\n",
       "      <td>uriD7RFuHhLJeDdKaf0nFA</td>\n",
       "      <td>Pizza Guru</td>\n",
       "      <td>3534 State St</td>\n",
       "      <td>Santa Barbara</td>\n",
       "      <td>CA</td>\n",
       "      <td>93105</td>\n",
       "      <td>34.44</td>\n",
       "      <td>-119.74</td>\n",
       "      <td>4.0</td>\n",
       "      <td>299</td>\n",
       "      <td>0</td>\n",
       "      <td>{'NoiseLevel': \"u'average'\", 'RestaurantsGoodF...</td>\n",
       "      <td>Restaurants</td>\n",
       "      <td>{'Monday': '15:0-21:0', 'Tuesday': '15:0-21:0'...</td>\n",
       "      <td>Mp97BPH2Y7O1DsFt6wFp4Q</td>\n",
       "      <td>1</td>\n",
       "      <td>This was the worst pizza I've ever had ever ea...</td>\n",
       "      <td>2019-07-08 19:27:04</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>205462</th>\n",
       "      <td>4559995</td>\n",
       "      <td>uriD7RFuHhLJeDdKaf0nFA</td>\n",
       "      <td>Pizza Guru</td>\n",
       "      <td>3534 State St</td>\n",
       "      <td>Santa Barbara</td>\n",
       "      <td>CA</td>\n",
       "      <td>93105</td>\n",
       "      <td>34.44</td>\n",
       "      <td>-119.74</td>\n",
       "      <td>4.0</td>\n",
       "      <td>299</td>\n",
       "      <td>0</td>\n",
       "      <td>{'NoiseLevel': \"u'average'\", 'RestaurantsGoodF...</td>\n",
       "      <td>Restaurants</td>\n",
       "      <td>{'Monday': '15:0-21:0', 'Tuesday': '15:0-21:0'...</td>\n",
       "      <td>06egjh3nrEb7EXX1VYZOCg</td>\n",
       "      <td>5</td>\n",
       "      <td>One of the best pizzas in SB....seriously!!!\\n...</td>\n",
       "      <td>2014-07-07 18:40:39</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>205463</th>\n",
       "      <td>4559996</td>\n",
       "      <td>uriD7RFuHhLJeDdKaf0nFA</td>\n",
       "      <td>Pizza Guru</td>\n",
       "      <td>3534 State St</td>\n",
       "      <td>Santa Barbara</td>\n",
       "      <td>CA</td>\n",
       "      <td>93105</td>\n",
       "      <td>34.44</td>\n",
       "      <td>-119.74</td>\n",
       "      <td>4.0</td>\n",
       "      <td>299</td>\n",
       "      <td>0</td>\n",
       "      <td>{'NoiseLevel': \"u'average'\", 'RestaurantsGoodF...</td>\n",
       "      <td>Restaurants</td>\n",
       "      <td>{'Monday': '15:0-21:0', 'Tuesday': '15:0-21:0'...</td>\n",
       "      <td>vTjt5vVDuW-Xu1Z_tfraTQ</td>\n",
       "      <td>5</td>\n",
       "      <td>We wanted a calzone and it was getting around ...</td>\n",
       "      <td>2016-11-06 05:15:44</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        Unnamed: 0             business_id        name        address  \\\n",
       "205454     4559987  uriD7RFuHhLJeDdKaf0nFA  Pizza Guru  3534 State St   \n",
       "205455     4559988  uriD7RFuHhLJeDdKaf0nFA  Pizza Guru  3534 State St   \n",
       "205456     4559989  uriD7RFuHhLJeDdKaf0nFA  Pizza Guru  3534 State St   \n",
       "205457     4559990  uriD7RFuHhLJeDdKaf0nFA  Pizza Guru  3534 State St   \n",
       "205458     4559991  uriD7RFuHhLJeDdKaf0nFA  Pizza Guru  3534 State St   \n",
       "205459     4559992  uriD7RFuHhLJeDdKaf0nFA  Pizza Guru  3534 State St   \n",
       "205460     4559993  uriD7RFuHhLJeDdKaf0nFA  Pizza Guru  3534 State St   \n",
       "205461     4559994  uriD7RFuHhLJeDdKaf0nFA  Pizza Guru  3534 State St   \n",
       "205462     4559995  uriD7RFuHhLJeDdKaf0nFA  Pizza Guru  3534 State St   \n",
       "205463     4559996  uriD7RFuHhLJeDdKaf0nFA  Pizza Guru  3534 State St   \n",
       "\n",
       "                 city state  postal_code  latitude  longitude  stars  \\\n",
       "205454  Santa Barbara    CA        93105     34.44    -119.74    4.0   \n",
       "205455  Santa Barbara    CA        93105     34.44    -119.74    4.0   \n",
       "205456  Santa Barbara    CA        93105     34.44    -119.74    4.0   \n",
       "205457  Santa Barbara    CA        93105     34.44    -119.74    4.0   \n",
       "205458  Santa Barbara    CA        93105     34.44    -119.74    4.0   \n",
       "205459  Santa Barbara    CA        93105     34.44    -119.74    4.0   \n",
       "205460  Santa Barbara    CA        93105     34.44    -119.74    4.0   \n",
       "205461  Santa Barbara    CA        93105     34.44    -119.74    4.0   \n",
       "205462  Santa Barbara    CA        93105     34.44    -119.74    4.0   \n",
       "205463  Santa Barbara    CA        93105     34.44    -119.74    4.0   \n",
       "\n",
       "        review_count  is_open  \\\n",
       "205454           299        0   \n",
       "205455           299        0   \n",
       "205456           299        0   \n",
       "205457           299        0   \n",
       "205458           299        0   \n",
       "205459           299        0   \n",
       "205460           299        0   \n",
       "205461           299        0   \n",
       "205462           299        0   \n",
       "205463           299        0   \n",
       "\n",
       "                                               attributes   categories  \\\n",
       "205454  {'NoiseLevel': \"u'average'\", 'RestaurantsGoodF...  Restaurants   \n",
       "205455  {'NoiseLevel': \"u'average'\", 'RestaurantsGoodF...  Restaurants   \n",
       "205456  {'NoiseLevel': \"u'average'\", 'RestaurantsGoodF...  Restaurants   \n",
       "205457  {'NoiseLevel': \"u'average'\", 'RestaurantsGoodF...  Restaurants   \n",
       "205458  {'NoiseLevel': \"u'average'\", 'RestaurantsGoodF...  Restaurants   \n",
       "205459  {'NoiseLevel': \"u'average'\", 'RestaurantsGoodF...  Restaurants   \n",
       "205460  {'NoiseLevel': \"u'average'\", 'RestaurantsGoodF...  Restaurants   \n",
       "205461  {'NoiseLevel': \"u'average'\", 'RestaurantsGoodF...  Restaurants   \n",
       "205462  {'NoiseLevel': \"u'average'\", 'RestaurantsGoodF...  Restaurants   \n",
       "205463  {'NoiseLevel': \"u'average'\", 'RestaurantsGoodF...  Restaurants   \n",
       "\n",
       "                                                    hours  \\\n",
       "205454  {'Monday': '15:0-21:0', 'Tuesday': '15:0-21:0'...   \n",
       "205455  {'Monday': '15:0-21:0', 'Tuesday': '15:0-21:0'...   \n",
       "205456  {'Monday': '15:0-21:0', 'Tuesday': '15:0-21:0'...   \n",
       "205457  {'Monday': '15:0-21:0', 'Tuesday': '15:0-21:0'...   \n",
       "205458  {'Monday': '15:0-21:0', 'Tuesday': '15:0-21:0'...   \n",
       "205459  {'Monday': '15:0-21:0', 'Tuesday': '15:0-21:0'...   \n",
       "205460  {'Monday': '15:0-21:0', 'Tuesday': '15:0-21:0'...   \n",
       "205461  {'Monday': '15:0-21:0', 'Tuesday': '15:0-21:0'...   \n",
       "205462  {'Monday': '15:0-21:0', 'Tuesday': '15:0-21:0'...   \n",
       "205463  {'Monday': '15:0-21:0', 'Tuesday': '15:0-21:0'...   \n",
       "\n",
       "                       user_id  review_stars  \\\n",
       "205454  qW-Q3krGy2FgY1ouN0znjQ             5   \n",
       "205455  X0NlTMV_-DNpGM0sPFa_cA             5   \n",
       "205456  _ymxPyUR2_tuoPJP8t67gQ             1   \n",
       "205457  avcQbkVRQ0skWZS1rkoAQg             4   \n",
       "205458  agYoN0HOl6KBkAkIZaJmFg             3   \n",
       "205459  o72LSEPHKu1oQcpXeIVlag             5   \n",
       "205460  u30sv3s4V2tz5VOgNQYtgg             4   \n",
       "205461  Mp97BPH2Y7O1DsFt6wFp4Q             1   \n",
       "205462  06egjh3nrEb7EXX1VYZOCg             5   \n",
       "205463  vTjt5vVDuW-Xu1Z_tfraTQ             5   \n",
       "\n",
       "                                                     text  \\\n",
       "205454  Santa Barbara has lots of good pizza. \\nThis o...   \n",
       "205455  Great pizza with good beer. Stopped in and we ...   \n",
       "205456  forgot to write a review until now- didn't hav...   \n",
       "205457  Solid pizza and beer at Pizza Guru.  I choose ...   \n",
       "205458  3.4 star review. Nice concept. I like the loca...   \n",
       "205459  This is one of my new favorite pizza joints he...   \n",
       "205460  Ordered vegan chicken and Daiya cheese and onl...   \n",
       "205461  This was the worst pizza I've ever had ever ea...   \n",
       "205462  One of the best pizzas in SB....seriously!!!\\n...   \n",
       "205463  We wanted a calzone and it was getting around ...   \n",
       "\n",
       "                       date  predict  \n",
       "205454  2018-10-18 20:35:16        5  \n",
       "205455  2015-05-02 22:25:45        1  \n",
       "205456  2013-09-16 12:11:56        5  \n",
       "205457  2018-07-22 13:31:09        3  \n",
       "205458  2014-12-14 16:41:07        5  \n",
       "205459  2015-04-11 04:43:19        5  \n",
       "205460  2015-05-26 15:27:36        1  \n",
       "205461  2019-07-08 19:27:04        5  \n",
       "205462  2014-07-07 18:40:39        5  \n",
       "205463  2016-11-06 05:15:44        3  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "print(\"\\nOriginal Reviews (with user bias)\")\n",
    "display(ca_restaurant_reviews.tail(10))\n",
    "\n",
    "print(\"\\nUnbiased Reviews (with predicted rating using user's review text)\")\n",
    "unbiased_reviews_dataset = ca_restaurant_reviews\n",
    "\n",
    "# dropping actual ratings(stars) by user\n",
    "#unbiased_reviews_dataset = unbiased_reviews_dataset.drop('review_stars', 1)\n",
    "\n",
    "# adding the unbiased predicted rating\n",
    "unbiased_reviews_dataset['predict'] = predicted_stars\n",
    "\n",
    "display(unbiased_reviews_dataset.tail(10))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.7119402985074627\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import accuracy_score\n",
    "print(accuracy_score(y_test, preds))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Precision: 0.6822727835840492\n",
      "Recall: 0.7119402985074627\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import precision_score\n",
    "from sklearn.metrics import recall_score\n",
    "print ('Precision: ' + str(precision_score(y_test, preds, average='weighted')))\n",
    "print ('Recall: ' + str(recall_score(y_test, preds, average='weighted')))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           1       0.65      0.80      0.72      2796\n",
      "           2       0.50      0.26      0.34      2149\n",
      "           3       0.53      0.33      0.41      2866\n",
      "           4       0.52      0.39      0.44      6087\n",
      "           5       0.79      0.93      0.86     16922\n",
      "\n",
      "    accuracy                           0.71     30820\n",
      "   macro avg       0.60      0.54      0.56     30820\n",
      "weighted avg       0.68      0.71      0.69     30820\n",
      "\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import classification_report\n",
    "print(classification_report(y_test, preds))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "import itertools  \n",
    "def plot_confusion_matrix(cm, classes,\n",
    "                          normalize=False,\n",
    "                          title='Confusion matrix',\n",
    "                          cmap=plt.cm.Blues):\n",
    "    \"\"\"\n",
    "    This function prints and plots the confusion matrix.\n",
    "    Normalization can be applied by setting `normalize=True`.\n",
    "    \"\"\"\n",
    "    if normalize:\n",
    "        cm = cm.astype('float') / cm.sum(axis=1)[:, np.newaxis]\n",
    "        print(\"Normalized confusion matrix\")\n",
    "    else:\n",
    "        print('Confusion matrix, without normalization')\n",
    "\n",
    "    print(cm)\n",
    "\n",
    "    plt.imshow(cm, interpolation='nearest', cmap=cmap)\n",
    "    plt.title(title)\n",
    "    plt.colorbar()\n",
    "    tick_marks = np.arange(len(classes))\n",
    "    plt.xticks(tick_marks, classes, rotation=45)\n",
    "    plt.yticks(tick_marks, classes)\n",
    "\n",
    "    fmt = '.2f' if normalize else 'd'\n",
    "    thresh = cm.max() / 2.\n",
    "    for i, j in itertools.product(range(cm.shape[0]), range(cm.shape[1])):\n",
    "        plt.text(j, i, format(cm[i, j], fmt),\n",
    "                 horizontalalignment=\"center\",\n",
    "                 color=\"white\" if cm[i, j] > thresh else \"black\")\n",
    "\n",
    "    plt.tight_layout()\n",
    "    plt.ylabel('True label')\n",
    "    plt.xlabel('Predicted label')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Confusion matrix, without normalization\n",
      "[[ 2249   246    99    50   152]\n",
      " [  795   566   412   194   182]\n",
      " [  296   275   949   922   424]\n",
      " [   58    42   284  2365  3338]\n",
      " [   45     7    39  1018 15813]]\n",
      "Normalized confusion matrix\n",
      "[[8.04e-01 8.80e-02 3.54e-02 1.79e-02 5.44e-02]\n",
      " [3.70e-01 2.63e-01 1.92e-01 9.03e-02 8.47e-02]\n",
      " [1.03e-01 9.60e-02 3.31e-01 3.22e-01 1.48e-01]\n",
      " [9.53e-03 6.90e-03 4.67e-02 3.89e-01 5.48e-01]\n",
      " [2.66e-03 4.14e-04 2.30e-03 6.02e-02 9.34e-01]]\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAjQAAAHpCAYAAACVw6ZvAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/NK7nSAAAACXBIWXMAAA9hAAAPYQGoP6dpAACQIUlEQVR4nOzdd1gUVxfA4d8CUkIVpIgCYu+9YS8o9p6IMbZYEmPvGo1Ro2KLNfYktmhsicYWFSvG3rD3rgioCCuogDDfH4T9XMEIujKU8+aZJ+7M3TtnhoE9e++dOxpFURSEEEIIITIwI7UDEEIIIYT4UJLQCCGEECLDk4RGCCGEEBmeJDRCCCGEyPAkoRFCCCFEhicJjRBCCCEyPElohBBCCJHhSUIjhBBCiAxPEhohhBBCZHiS0GQy165do379+tja2qLRaNi4caNB6799+zYajYalS5catN7MIE+ePHTu3FntMJJIzc8ssey0adM+fmAiWWPGjEGj0eitU+vaSq/XtBDJkYTmI7hx4wZfffUVefPmxdzcHBsbG6pWrcqsWbN48eLFR913p06dOHfuHBMmTGDFihWUL1/+o+4vM7p48SJjxozh9u3baofy0Wzbto0xY8aoHUYSEydONHgSLv7boUOHGDNmDOHh4WqHIsQHMVE7gMxm69atfPrpp5iZmdGxY0eKFy9OTEwM//zzD0OGDOHChQssWrToo+z7xYsXHD58mJEjR9K7d++Psg8PDw9evHhBtmzZPkr96cHFixcZO3YstWrVIk+ePCl+35UrVzAySn/fEZL7mW3bto25c+emu6Rm4sSJtGnThhYtWqgdSrryMa+tQ4cOMXbsWDp37oydnV2a7VcIQ5OExoBu3bqFr68vHh4e7Nmzh5w5c+q29erVi+vXr7N169aPtv9Hjx4BJPmjZEgajQZzc/OPVn9GoygKL1++xMLCAjMzM7XDSZb8zD5MVFQUlpaWqsag1rWVXq9pIZKlCIP5+uuvFUA5ePBgisrHxsYq48aNU/LmzauYmpoqHh4eyogRI5SXL1/qlfPw8FAaN26sHDhwQKlQoYJiZmameHp6KsuWLdOV+f777xVAb/Hw8FAURVE6deqk+/frEt/zup07dypVq1ZVbG1tFUtLS6VgwYLKiBEjdNtv3bqlAMqSJUv03rd7926lWrVqyieffKLY2toqzZo1Uy5evJjs/q5du6Z06tRJsbW1VWxsbJTOnTsrUVFR7zxfNWvWVIoVK6acOXNGqVGjhmJhYaHky5dPWbdunaIoirJv3z6lYsWKirm5uVKwYEHF399f7/23b99WevbsqRQsWFAxNzdX7O3tlTZt2ii3bt3SlVmyZEmS8wgoe/fu1ftZbN++XSlXrpxiZmamzJgxQ7etU6dOiqIoSnx8vFKrVi0lR44cSkhIiK7+6OhopXjx4krevHmVyMjIdx7z6wYMGKDY29sr8fHxunW9e/dWAGXWrFm6dcHBwQqgzJs3T1GUpD+zTp06JXuMr5edOnWqsnDhQt21Wb58eeXYsWNJYkrJzz2l119yMSWez+Ts3btXAZQ1a9Yo48ePV3LlyqWYmZkpderUUa5du5ak/Nq1a5WyZcsq5ubmioODg9K+fXvl/v37SWK1tLRUrl+/rjRs2FCxsrJSmjdvrouvV69eytq1a5UiRYoo5ubmSuXKlZWzZ88qiqIoCxYsUPLly6eYmZkpNWvW1LuuFEVRAgIClDZt2ihubm6Kqampkjt3bqV///7K8+fP//O8KIr+tfW2c5W4JO73zJkzSqdOnRRPT0/FzMxMcXZ2Vrp06aI8fvw4yb7eVseb+1UURblx44bSpk0bJXv27IqFhYVSqVIlZcuWLR/0sxHCEKSFxoA2b95M3rx5qVKlSorKd+vWjWXLltGmTRsGDRrE0aNH8fPz49KlS2zYsEGv7PXr12nTpg1du3alU6dO/Prrr3Tu3Jly5cpRrFgxWrVqhZ2dHQMGDKBdu3Y0atQIKyurVMV/4cIFmjRpQsmSJRk3bhxmZmZcv36dgwcP/uf7du3aRcOGDcmbNy9jxozhxYsXzJkzh6pVq3Lq1Kkk3TafffYZnp6e+Pn5cerUKX7++WecnJyYPHnyO2N8+vQpTZo0wdfXl08//ZT58+fj6+vLypUr6d+/P19//TWff/45U6dOpU2bNty7dw9ra2sAjh8/zqFDh/D19SV37tzcvn2b+fPnU6tWLS5evMgnn3xCjRo16Nu3L7Nnz+bbb7+lSJEiALr/Q0IzfLt27fjqq6/o3r07hQoVShKnRqPh119/pWTJknz99df8+eefAHz//fdcuHCBffv2pfpbf/Xq1ZkxYwYXLlygePHiABw4cAAjIyMOHDhA3759desAatSokWw9X331FUFBQfj7+7NixYpky6xatYpnz57x1VdfodFomDJlCq1ateLmzZu6rqvU/tzfZcWKFXTr1o2KFSvSo0cPAPLly/fO902aNAkjIyMGDx5MREQEU6ZMoX379hw9elRXZunSpXTp0oUKFSrg5+dHSEgIs2bN4uDBg5w+fVqvVfPVq1f4+PhQrVo1pk2bxieffKLbduDAATZt2kSvXr0A8PPzo0mTJgwdOpR58+bxzTff8PTpU6ZMmcKXX37Jnj17dO9dt24dz58/p2fPnjg4OHDs2DHmzJnD/fv3WbduXarP1ZtGjRpFaGio7vfe39+fmzdv0qVLF1xcXHTd3RcuXODIkSNoNBpatWrF1atX+f3335kxYwY5cuQAwNHRMdn9hoSEUKVKFZ4/f07fvn1xcHBg2bJlNGvWjPXr19OyZctU/2yEMBi1M6rMIiIiQgF03+beJTAwUAGUbt266a0fPHiwAih79uzRrfPw8FAAJSAgQLcuNDRUMTMzUwYNGqRb9/q369el9BvyjBkzFEB59OjRW+NOroWmdOnSipOTk/LkyRPdujNnzihGRkZKx44dk+zvyy+/1KuzZcuWioODw1v3mahmzZoKoKxatUq37vLlywqgGBkZKUeOHNGt37FjR5I43/wmrCiKcvjwYQVQli9frlu3bt06vVaZ1yX+LLZv357stje/zS5cuFABlN9++005cuSIYmxsrPTv3/+dx5qc0NBQvZaX8PBwxcjISPn0008VZ2dnXbm+ffvqteQk9zPr1atXklaA18s6ODgoYWFhuvV//fWXAiibN2/WrUvpzz01LYSWlpb/2SrzusRWgCJFiijR0dG69bNmzVIA5dy5c4qiKEpMTIzi5OSkFC9eXHnx4oWu3JYtWxRAGT16tF6sgDJ8+PAk+wMUMzMzvZaXxJ+vi4uLotVqdetHjBih19KhKMlff35+fopGo1Hu3LmjW5eSFpo3TZkyJcl1nNz+fv/99yR/S6ZOnZok1rftt3///gqgHDhwQLfu2bNniqenp5InTx4lLi5OUZSU/2yEMCQZ7WUgWq0WQNca8C7btm0DYODAgXrrBw0aBJBkrE3RokWpXr267rWjoyOFChXi5s2b7x3zmxK/pf7111/Ex8en6D0PHz4kMDCQzp07Y29vr1tfsmRJ6tWrpzvO13399dd6r6tXr86TJ0905/C/WFlZ4evrq3tdqFAh7OzsKFKkCJUqVdKtT/z36+fHwsJC9+/Y2FiePHlC/vz5sbOz49SpUyk42gSenp74+PikqGyPHj3w8fGhT58+dOjQgXz58jFx4sQU7+t1jo6OFC5cmICAAAAOHjyIsbExQ4YMISQkhGvXrgEJrQjVqlVLcutvarRt25bs2bPrXidee4nn831+7h9Lly5dMDU1fWusJ06cIDQ0lG+++UZvLFHjxo0pXLhwsuPaevbsmey+6tatq9fylHidtW7dWu93/13XX1RUFI8fP6ZKlSooisLp06dTfLxv2rt3LyNGjNBdY8nt7+XLlzx+/JjKlSsDpOp6f922bduoWLEi1apV062zsrKiR48e3L59m4sXL+qVf9fPRghDkoTGQGxsbAB49uxZisrfuXMHIyMj8ufPr7fexcUFOzs77ty5o7fe3d09SR3Zs2fn6dOn7xlxUm3btqVq1ap069YNZ2dnfH19Wbt27X8mN4lxJtftUqRIER4/fkxUVJTe+jePJfGDMyXHkjt37iQf1La2tri5uSVZ92adL168YPTo0bi5uWFmZkaOHDlwdHQkPDyciIiId+47kaenZ4rLAvzyyy88f/6ca9eusXTpUr0PmtSqXr26rkvpwIEDlC9fnvLly2Nvb8+BAwfQarWcOXNGL/l9H+/6Gb3Pz/1j+ZBYCxcunOR3zcTEhNy5c6doX4nXWUquv7t37+oSQCsrKxwdHalZsyZAqq6/192/f1/3ezt9+nS9bWFhYfTr1w9nZ2csLCxwdHTUXbvvu787d+689WeeuP11H/K7LkRqyRgaA7GxscHV1ZXz58+n6n0p/RZtbGyc7HpFUd57H3FxcXqvLSwsCAgIYO/evWzdupXt27ezZs0a6tSpw86dO98aQ2p9yLG87b0pqbNPnz4sWbKE/v374+XlpZt80NfXN8UtUkCqE5J9+/YRHR0NwLlz5/Dy8krV+19XrVo1Fi9ezM2bNzlw4ADVq1dHo9FQrVo1Dhw4gKurK/Hx8R+c0HzIz+hNKb3+3pchY4WEO3vedqvy+15/cXFx1KtXj7CwMIYNG0bhwoWxtLTkwYMHdO7cOVXXX6KYmBjatGmDmZkZa9euxcRE/8/5Z599xqFDhxgyZAilS5fGysqK+Ph4GjRo8F77ex+G/tkI8V8koTGgJk2asGjRIg4fPvzODy0PDw/i4+O5du2a3oDTkJAQwsPD8fDwMFhc2bNnT3bSrDe/TQEYGRlRt25d6taty/Tp05k4cSIjR45k7969eHt7J3sckDBQ9k2XL18mR44cqt/ymmj9+vV06tSJH3/8Ubfu5cuXSc7Nh3TVvOnhw4f06dOH+vXrY2pqyuDBg/Hx8Xnvn29iouLv78/x48cZPnw4kDAAeP78+bi6umJpaUm5cuX+s54PPcbU/NxTc/0Z8twnej3WOnXq6G27cuWKQX/X3ubcuXNcvXqVZcuW0bFjR916f3//966zb9++BAYGEhAQgLOzs962p0+fsnv3bsaOHcvo0aN16xO7JV+XmnPu4eHx1p954nYh1CJdTgY0dOhQLC0t6datGyEhIUm237hxg1mzZgHQqFEjAGbOnKlXJrHZuHHjxgaLK1++fERERHD27FnduocPHya5kyosLCzJe0uXLg2ga2F4U86cOSldujTLli3T+9A6f/48O3fu1B1nemBsbJzkm+GcOXOStBQkfhAbYubU7t27Ex8fzy+//MKiRYswMTGha9eu7/0N1dPTk1y5cjFjxgxiY2OpWrUqkJDo3Lhxg/Xr11O5cuUk39bf9KHHmJqfe0qvv8S4DD1jbfny5XFycmLBggV61/Hff//NpUuXDPq79jaJLRWv/9wVRdH9PUitJUuWsHDhQubOnUvFihVTtD9I+vcGUnctNGrUiGPHjnH48GHduqioKBYtWkSePHkoWrRoKo5CCMOSFhoDypcvH6tWraJt27YUKVJEb6bgQ4cOsW7dOt1zUUqVKkWnTp1YtGgR4eHh1KxZk2PHjrFs2TJatGhB7dq1DRaXr68vw4YNo2XLlvTt25fnz58zf/58ChYsqDc4cNy4cQQEBNC4cWM8PDwIDQ1l3rx55M6dW28Q4JumTp1Kw4YN8fLyomvXrrrbd21tbdPVTLRNmjRhxYoV2NraUrRoUQ4fPsyuXbtwcHDQK1e6dGmMjY2ZPHkyERERmJmZUadOHZycnFK1vyVLlrB161aWLl2qG5MxZ84cvvjiC+bPn88333yjK6vRaKhZsyb79u17Z73Vq1dn9erVlChRQjcmoWzZslhaWnL16lU+//zzd9aR2ILTt29ffHx8MDY21htsnRIp/bmn9PpLjGvXrl1Mnz4dV1dXPD099QZ7v49s2bIxefJkunTpQs2aNWnXrp3utu08efIwYMCAD6o/JQoXLky+fPkYPHgwDx48wMbGhj/++OO9xpI8fvyYb775hqJFi2JmZsZvv/2mt71ly5bY2NhQo0YNpkyZQmxsLLly5WLnzp3cunUrSX2J18LIkSPx9fUlW7ZsNG3aNNmW1eHDh/P777/TsGFD+vbti729PcuWLePWrVv88ccfMquwUJcq91ZlclevXlW6d++u5MmTRzE1NVWsra2VqlWrKnPmzNGbNC82NlYZO3as4unpqWTLlk1xc3P7z4n13lSzZk2lZs2autdvu21bURImzCtevLhiamqqFCpUSPntt9+S3B66e/dupXnz5oqrq6tiamqquLq6Ku3atVOuXr2aZB9vTqy3a9cupWrVqoqFhYViY2OjNG3a9K0T6715W3jiZHbJ3Tb65vEWK1Ysyfq3nR/+nQgt0dOnT5UuXbooOXLkUKysrBQfHx/l8uXLyd4Su3jxYiVv3ryKsbFxshPrJef1eu7du6fY2toqTZs2TVKuZcuWiqWlpXLz5k1FURJuewUUX1/f/zz+RHPnzlUApWfPnnrrvb29FUDZvXu33vrkfmavXr1S+vTpozg6OioajSbZifXeBCjff/+93rqU/NwVJWXXn6Ik3IafOGkiKZxYL3Fixf86XkVRlDVr1ihlypRRzMzMFHt7+/+cWC85b15Pr+/rzfOVXGwXL15UvL29FSsrKyVHjhxK9+7dlTNnziSJ9V23bSfu821L4u/R/fv3lZYtWyp2dnaKra2t8umnnypBQUHJ/hx/+OEHJVeuXIqRkVGKJ9azs7NTzM3NlYoVK751Yr2U/myEMASNosjoLCHUtG3bNpo0acKZM2coUaKE2uEIIUSGJO2DQqhs7969+Pr6SjIjhBAfQFpohBBCCJHhSQuNEEIIITI8SWiEEEIIkeFJQiOEEEKIDE8SGiGEEEJkeBl6Yr34+HiCgoKwtrb+KFOmCyGEyNwUReHZs2e4urqm+cSAL1++JCYmxqB1mpqa6j1VPivJ0AlNUFBQkqfcCiGEEKl17969tz5l/WN4+fIlFtYO8Oq5Qet1cXHh1q1bWTKpydAJjbW1NQABp65i9e+/Req42KXuydEiKSNpHPwg0rr6YeLjZeaND/HsmZaCed11nydpJSYmBl49x6xYFzA2NUylcTEEX1hCTEyMJDQZTeIfQitra6ysbVSOJmOysZGE5kNJQvNhJKH5MJLQGIZq16GxKRoDJTRZ/UrI0AmNEEIIkaFpAEMlU1n8u4EkNEIIIYRaNEYJi6HqysKy9tELIYQQIlOQFhohhBBCLRqNAbucsnafk7TQCCGEECLDkxYaIYQQQi0yhsZgJKERQggh1CJdTgaTtdM5IYQQQmQK0kIjhBBCqMaAXU5ZvI1CEhohhBBCLdLlZDBZO50TQgghRKYgLTRCCCGEWuQuJ4ORhEYIIYRQi3Q5GUzWTueEEEIIkSlIC40QQgihFulyMpisffRCCCGEyBSkhUYIIYRQi4yhMRhJaIQQQgi1SJeTwWTtoxdCCCFEpiAtNEIIIYRaNBoDttBIl5MQQggh1GCkSVgMVVcWJl1OQgghhMjwpIVGCCGEUIsMCjYYSWiEEEIItcht2waTtdM5IYQQQmQK0kIjhBBCqEW6nAwmax+9EEIIITIFaaERQggh1CJjaAxGWmjeYcHsqbTyqU6ZfM5ULuZBz85tuXn9qm57+NMwxn07CJ+qpSmRx4Ga5Qrxw8jBPNNGJFvf07AnVC9TgIIulmgjwvW2/fbrQhpUL0uJPA74VC3NhrUrP+ahqWLaFD9qVKmIi4MNeXI749umJVevXEm2rKIotGzaCCszIzb/tTHJ9t+WL6VSuVI42FiQJ7czA/r2+sjRp1/Pnj1jyKD+FMqfB3ubT6hdoyonThzXbQ8JCaFH1y7k9ciFg60lzZo05Pq1aypGnL6MHzcGi2wavaVU8cK67S9fvqR/n17kcnYgh50Vvp+1JiQkRL2AVfbPgQDatGxGvjy5sEzm97NHty5YmhnpLc2bNNRtv3P7Nj2/6krRgnlxsP2E4oXzM37c98TExKTxkaQDiV1OhlqyMGmheYfjh//hiy49KFG6HK/iXjF94hi+bNuMbQEn+cTSktDgh4SGPGTY9xPJX7AwD+7f5fuh/QgNfsicX5ImJN8O/IZCRYoT8jBIb/2qpYv5ceL3jJ/2EyXKlOPs6RN8N6g3tnbZqVO/UVod7kf3T0AAPb7+hrLlKxD36hVjvhtJ8yY+nAi8gKWlpV7ZubNnonnLN445M6cze9Z0JvhNoXyFSjx/HsWd27fT4AjSp2++6s7FC+f5ZclycuZ05fdVv9GkQT1OnrmAq6srbdu0JFu2bKz9YyM21jbMnjWdxg3rcepM0vOeVRUtVoyt23fpXpuY/P/P49BBA/j7762sXL0OGxtbBvTrje+nrdgbcFCNUFUXFRVFiZIl6di5C+0+a51smXr1G7Bg8a+612ZmZrp/X7lymfh4hdlzF5AvX34uXjhPr296EBUVhd/kaR89fpE5aRRFUdQO4n1ptVpsbW05de0hVtY2abLPsMePqFw8Dys37KCCV7Vky/y96U8G9+7KmZuP9P4orlq6mG1//UGvQcPp1KYxJ648wMbWDoC2TepQtkJlhn0/UVfe7/vhnDl9gtWbdr25C4NxzW7x0epOiUePHuGZ25ntu/ZRrXoN3fqzZwJp07IpBw4dJ5+HK7+v/ZOmzVsA8PTpUwp65mbtn5uoXaeuSpH/n9qTc7548QInexvW/rGRho0a69ZXqVSe+j4NaP9FR0oVL8yJ0+coWqwYAPHx8Xi65WTMDxPo8mU3tUIHeGvSmpbGjxvD5r82cvRkYJJtERERuOV0ZOmKVbRq3QaAK5cvU7pEEfYdOEylypXTNtg3xMer+yfc0syI1a/9fkJCC01EeDhr1m9IcT0zfpzKz4sWcOHKjY8Q5dtptVpyOtoRERGBjU3afI4k7tfW1hazOj+gMTE3SJ3Kq5dE7/kuzY8lvcja7VPv4dkzLQC2dtn/s4yVlY1eMnP9yiXmTvdjypzFGCXTLBgTE42Zuf5FbW5hwbnTJ4iNjTVQ9OmPNiKhay67vb1u3fPnz+nSsT3TZ/6Es4tLkvfs2e1PfHw8D4MeULZkUQrmdaPD5225f+9emsWdnrx69Yq4uDjM37h+LCwsOHzoINHR0QB6242MjDA1M+PwwazZwpCc69ev4enuSpGCeencoT13794F4PSpk8TGxlKnrreubKHChXFzd+fokcNqhZvuHQjYh0duZ0oXL0y/3j158uTJf5bXaiP0/g5kGdLlZDBZ++hTKT4+ngnfDaVsRS8KFimWbJmwJ4+ZN30SbTt00a2LiY5mQM/ODB09Adfcbsm+r1otb9atXMr5M6dRFIVzgadYt3IpsbGxPA17/FGOR23x8fEMGzwArypVKVasuG79sMEDqOzlRZNmzZN93+1bN4mPj2fqZD+mTJvBb7+v42lYGE0b1c+SffDW1tZUquzFpInjCQoKIi4ujt9X/sbRI4cJfvhQ9+E7etS3PH36lJiYGH6cOpkH9+8THPxQ7fDThQoVK7Hol6Vs2rKd2T/N5/btW3jXrs6zZ88IDg7G1NQUOzs7vfc4OTkTEhKsTsDpXL36Piz+ZRlbt+/ihwmT+OdAAC2bNSIuLi7Z8jeuX2fBvJ/o2q1HGkcqMhNVx9AEBAQwdepUTp48ycOHD9mwYQMtWrRQM6T/NHb4AK5dvsjvb+kCinympccXrclXsDB9Bo/UrZ82cTT5ChSmeZt2b62714DhPA4N4bPGtVAUBQdHJ1p+1p7Fc2dgZJQ5884BfXtx8eJ5/Pcc0K3bunkTAfv2cvDYqbe+Lz4+ntjYWKZNn0XdevUBWLJiFfnccxKwby/e9X0+euzpzS9LlvN1j67kz5MbY2NjSpcpy2dt23H61EmyZcvG72v/oGePbuRydsDY2Jjadb2p36AhGbjH2aB8Gvx/wGqJkiWpULEShfJ58Me6tZhbqNstmxF9+pmv7t/Fi5egeImSFC+Sn4D9+5J0Ewc9eECLpg1p2boNXbp2T+tQ1Sd3ORmMqp+UUVFRlCpVirlz56oZRoqMHTGQvbv+Zvkff+PimivJ9sjIZ3Rt1wJLKyvmLVlNtmzZdNuO/LOf7Zv/pEguG4rksqHTpwnjHCoVdWfWlPFAQveS38wFnLn1mL3HL7H/5BVyuXlgaWWNvYNj2hxkGhrYrzfb/97Kth17yJU7t279/n17uHnzBrmcsmP7STZsP0k4j+1929CgXm0AXFxyAlC4SFHd+xwdHXHIkYN79+6m4VGkH3nz5WPn7n08evqMqzfvcuDQUWJjY8mTNy8AZcuW4+iJ0zx89JSbd4PYtOVvwp48wdPTU+XI0yc7OzvyFyjIjRvXcXFxISYmhvDwcL0yoaEhODsn7RIVSXnmzUuOHDm4eeO63vqHQUE0rF+HSl5V+GneIpWiU5shu5sy55fflFK1haZhw4Y0bNjw3QVVpCgK474dhP/fm/jtz+24eeRJUibymZYvfZtjamrGgmXrkoyF+emXVbx88UL3+lzgSUYM6Mmqv/xx99D/QMmWLZsuYdq6cT216zXIVC00iqIwqH8fNm/ayN8795LnjQ/UQUOG0+mNQaqVypZk0tTpNGrcFIDKXlUBuHr1ii4ZCgsL48njx7i7e6TBUaRflpaWWFpa8vTpU3b572C832S97ba2tgBcv3aNUydPMHrMODXCTPciIyO5dfMGLu07UKZsObJly8bePbtp2Srhjp6rV65w7+5dKlX2UjnSjOHB/fs8efJE92UEElpmGtavQ+my5Vi4+NdM9XdOqCND3bYdHR2tG+AICaPEP7axwwewecNa5i9dg6WVFY9CE/rMra1tMbewIPKZli5tm/HyxXOmzf2FyEgtkZEJcdk7OGJsbIx7nrx6dT4NSxgcl69AId1dTrduXOPs6ROUKluBiPBwliycw7UrF5k8O3N9axnQtxfr1vzO6vUbsba2JiQ44Xza2NpiYWGBs4tLsgOB3dzcdclPgYIFadK0OUMH9WfOvIVY29gwZtS3FCxUmBq1aqfp8aQX/jt3oCgKBQsW4saN63w7fCgFCxWmY6eEsVx/rl9HDkdH3NzcOX/+HEMG9adpsxZ4/9tll9UNHzqYxk2a4u7uQVBQEOPHfY+xsTGf+bbD1taWzl26MmzIQOzt7bG2tmFg/z5Uquyl+h1OaomMjOTGa60tt2/f4syZQOyz25Pd3p6J48fSomVrnJ1duHnzBqO+HUa+fPl13cFBDx7QoF5t3Nw98Js0lUePHunqcknm9z9Tky4ng8lQCY2fnx9jx45N032uWrYYgC9aNdBbP2nmAlr5duDC2UDOnEqYwMy7cgm9MnuOXSR3ClsM4uLi+HXBbG7duIaJSTYqVa3B6s27U/z+jOLnRQsAaFhPP/FYsPhXvujYOcX1LPp1GcOHDKBNiyYYGRlRtXpNNm7+W6+rLyvRRkQw+rtveXD/Ptnt7WnRshVjxk3QnY/g4IcMGzqI0JAQXHLm5PP2HRgx8juVo04/Hjy4T8cv2hH25Ak5HB2pUrUa+/85gqNjQnfvlB8TxrK1+6w10dHReNf3YdaceSpHrZ5TJ0/QsH4d3evhQwcB0L5DJ2bNmcf5c+dY+dtyIsLDyenqSt269fhuzA+6uWh27/bnxo3r3LhxnQJ59W+UiIqOT7sDEZlKupmHRqPRvHNQcHItNG5ubmk6D01mo/Y8NJmB2vPQZHTpYR6ajEzteWgyOtXnoak/BU02w/wdVmJfEL1zqMxDkxGYmZlhY2OjtwghhBAZlorz0AQEBNC0aVNcXV3RaDRs3LjxrWW//vprNBoNM2fO1FsfFhZG+/btsbGxwc7Ojq5duxIZGalX5uzZs1SvXh1zc3Pc3NyYMmVKkvrXrVtH4cKFMTc3p0SJEmzbti1VxwIZLKERQgghhGGk9E7jDRs2cOTIEVxdXZNsa9++PRcuXMDf358tW7YQEBBAjx7/n09Iq9VSv359PDw8OHnyJFOnTmXMmDEsWvT/8aGHDh2iXbt2dO3aldOnT9OiRQtatGjB+fPnU3U8qo6hiYyM5Pr1/w8su3XrFoGBgdjb2+Pu7q5iZEIIIUQaUHFQcEruNH7w4AF9+vRhx44dNG7cWG/bpUuX2L59O8ePH6d8+fIAzJkzh0aNGjFt2jRcXV1ZuXIlMTEx/Prrr5iamlKsWDECAwOZPn26LvGZNWsWDRo0YMiQIQD88MMP+Pv789NPP7FgwYIUH4+qLTQnTpygTJkylClTBoCBAwdSpkwZRo8erWZYQgghRNr4CF1OWq1Wb3l97GlqxMfH06FDB4YMGUKxYklnxz98+DB2dna6ZAbA29sbIyMjjh49qitTo0YNTE1NdWV8fHy4cuUKT58+1ZXx9vbWq9vHx4fDh1P3aBFVE5patRJmxX1zWbp0qZphCSGEEBmWm5sbtra2usXPz++96pk8eTImJib07ds32e3BwcE4OTnprTMxMcHe3p7gf6fkCA4OxtnZWa9M4ut3lUncnlIZ6rZtIYQQIlP5CF1O9+7d07tpJvF2+dQ4efIks2bN4tSpUxnmTkQZFCyEEEKo5SN0Ob15N/D7JDQHDhwgNDQUd3d3TExMMDEx4c6dOwwaNIg8efIACZMghoaG6r3v1atXhIWF6SZIdHFxISQkRK9M4ut3lUntJIuS0AghhBBCT4cOHTh79iyBgYG6xdXVlSFDhrBjxw4AvLy8CA8P5+TJk7r37dmzh/j4eCpVqqQrExAQQGxsrK6Mv78/hQoVInv27Loyu3fv1tu/v78/Xl6pe7SIdDkJIYQQalHxLqd33Wns4OCgVz5btmy4uLhQqFAhAIoUKUKDBg3o3r07CxYsIDY2lt69e+Pr66u7xfvzzz9n7NixdO3alWHDhnH+/HlmzZrFjBkzdPX269ePmjVr8uOPP9K4cWNWr17NiRMn9G7tTglpoRFCCCGyIEPcabxy5UoKFy5M3bp1adSoEdWqVdNLRGxtbdm5cye3bt2iXLlyDBo0iNGjR+vNVVOlShVWrVrFokWLKFWqFOvXr2fjxo0UL148VceTbh598D4Sp46WRx+8P3n0wYeTRx98mIwy4DC9kkcffBi1H31g3vQngz764OXm3ln20QfS5SSEEEKoRKPRGC6pz+JfDqTLSQghhBAZnrTQCCGEEGrR/LsYqq4sTBIaIYQQQiXS5WQ40uUkhBBCiAxPWmiEEEIIlUgLjeFIQiOEEEKoRBIaw5EuJyGEEEJkeNJCI4QQQqhEWmgMR1pohBBCCJHhSQuNEEIIoRaZh8ZgJKERQgghVCJdToYjXU5CCCGEyPCkhUYIIYRQiUZjwCfOZ+0GGklohBBCCLVoMGCXUxbPaKTLSQghhBAZnrTQCCGEECqRQcGGIwmNEEIIoRa5bdtgpMtJCCGEEBmetNAIIYQQajFgl5OSxbucpIVGCCGEEBmetNAIIYQQKjHkoGDD3f6dMUlCI4QQQqhEEhrDkS4nIYQQQmR40kIjhBBCqEVu2zYYSWiEEEIIlUiXk+FIl5MQQgghMrxM0UKT3dIUGytTtcPIkGLj4tUOIcMzzuLfij6UkXyt+iBy9WVs0kJjOJkioRFCCCEyIkloDEe+GwkhhBAiw5MWGiGEEEIl0kJjONJCI4QQQogMT1pohBBCCLXIPDQGIwmNEEIIoRLpcjIc6XISQgghRIYnLTRCCCGESqSFxnAkoRFCCCFUIgmN4UiXkxBCCCEyPGmhEUIIIdQidzkZjCQ0QgghhEqky8lwpMtJCCGEyIICAgJo2rQprq6uaDQaNm7cqNsWGxvLsGHDKFGiBJaWlri6utKxY0eCgoL06ggLC6N9+/bY2NhgZ2dH165diYyM1Ctz9uxZqlevjrm5OW5ubkyZMiVJLOvWraNw4cKYm5tTokQJtm3blurjkYRGCCGEUEliC42hltSIioqiVKlSzJ07N8m258+fc+rUKb777jtOnTrFn3/+yZUrV2jWrJleufbt23PhwgX8/f3ZsmULAQEB9OjRQ7ddq9VSv359PDw8OHnyJFOnTmXMmDEsWrRIV+bQoUO0a9eOrl27cvr0aVq0aEGLFi04f/586s6loihKqt6Rjmi1WmxtbbkV9AQbGxu1w8mQjIyydhOlIRhn8WbeDyXX4IeRs/dhtFotOR3tiIiISNPPkcTPr9xfrcbI9BOD1Bkf85z7C33f61g0Gg0bNmygRYsWby1z/PhxKlasyJ07d3B3d+fSpUsULVqU48ePU758eQC2b99Oo0aNuH//Pq6ursyfP5+RI0cSHByMqakpAMOHD2fjxo1cvnwZgLZt2xIVFcWWLVt0+6pcuTKlS5dmwYIFKT4GaaERQgghVKLBgC00/6a3Wq1Wb4mOjjZIrBEREWg0Guzs7AA4fPgwdnZ2umQGwNvbGyMjI44ePaorU6NGDV0yA+Dj48OVK1d4+vSproy3t7fevnx8fDh8+HCq4pOERgghhFDJx+hycnNzw9bWVrf4+fl9cJwvX75k2LBhtGvXTtf6ExwcjJOTk145ExMT7O3tCQ4O1pVxdnbWK5P4+l1lErenlNzlJIQQQmQi9+7d0+tyMjMz+6D6YmNj+eyzz1AUhfnz539oeB+NJDRCCCGEWj7CPDQ2NjYGGw+UmMzcuXOHPXv26NXr4uJCaGioXvlXr14RFhaGi4uLrkxISIhemcTX7yqTuD2lpMtJCCGEUImadzm9S2Iyc+3aNXbt2oWDg4Pedi8vL8LDwzl58qRu3Z49e4iPj6dSpUq6MgEBAcTGxurK+Pv7U6hQIbJnz64rs3v3br26/f398fLySlW8ktAIIYQQWVBkZCSBgYEEBgYCcOvWLQIDA7l79y6xsbG0adOGEydOsHLlSuLi4ggODiY4OJiYmBgAihQpQoMGDejevTvHjh3j4MGD9O7dG19fX1xdXQH4/PPPMTU1pWvXrly4cIE1a9Ywa9YsBg4cqIujX79+bN++nR9//JHLly8zZswYTpw4Qe/evVN1PHLbdhYnt8x+OLlt+8PINfhh5Ox9GLVv287Taz1GZga6bTv6Obfntknxsezbt4/atWsnWd+pUyfGjBmDp6dnsu/bu3cvtWrVAhIm1uvduzebN2/GyMiI1q1bM3v2bKysrHTlz549S69evTh+/Dg5cuSgT58+DBs2TK/OdevWMWrUKG7fvk2BAgWYMmUKjRo1SsXRS0KT5cmHyYeThObDyDX4YeTsfRi1ExrP3oZNaG79lPKEJrORLichhBBCZHhyl5MQQgihEo3GcA+VzOqNxZLQCCGEEGrRGDARyeIJjXQ5CSGEECLDkxYaIYQQQiWGnD/G0PPQZDTSQiOEEEKIDE9aaIQQQgiVaAw4hiaLN9BIC01qlS6aHwerbEmWIQP6AHDr5g06+LahoEdOPHLa82WHdoS+8YyK5OqY+eMUNQ4nzU0aP5bsn5joLRVLF9Mrc+zoYZo19CZXDhvcnbPTqF4tXrx4oVdmx99b8a7hRU57K/K45qD9Z63S8jDSlelTJ2NjYcywwQN065b8sohG9euQy8kOGwtjwsPD9d5z585ten3djRKF8+GU3ZKSRQsw4YcxuhlAM7t/DgTwactm5M+TCyszIzb/tVFve0hICF9160L+PLlwtLOkRZOGXL92Ldm6FEWhZdNGydaTWf1zIIA2LZuRL08uLJM57sjISAb2602BvG442H5CuVLF+HnRAt32sLAwBvXvQ+nihXGw/YRC+T0YPKAvERERaXwk6jMy0hh0ycqkhSaVdu0/TFx8nO71pYsXaN20Ac1btiEqKoo2zRtRrHhJNm7dCcDE8WP4/LMW7Nx7ECOj/+ePI0aNoUOXrrrXVlbWaXcQKitctBgbt+zQvTYx+f9leOzoYdo0b8yAwcOY/OMsTExMOH/urN6527TxT/r1+orvxoynRq3avHr1iksXL6TpMaQXJ08cZ8kviyheoqTe+ufPX+Bdzwfvej6MGf1tkvddvXKZ+Ph4Zv40n7z58nPpwnn69PqK51FRTJg0Na3CV83zqCiKlyxJh85d+Pyz1nrbFEWh3actMcmWjTXrN2JtbcOcWdNp2qgeJwIvYGlpqVd+7uyZWW7sQlRUFCVKlqRj5y60e+P8AQwfMpD9+/byy5IVeHjkYfeunfTv24ucOV1p3LQZDx8G8fDhQyZOmkrhIkW5e/cO/Xr35OHDh6xcvU6FIxKZgSQ0qZTD0VHv9awfp+CZNx9Vq9dg355d3L1zm70Hj+tmaZy38Ffy5nYkYP9eatWuq3uflbUVzs6pe5JoZmFibILzW56iOnLoIL7q2ZsBg/8/LXaBgoV0/3716hUjBg9g3ITJdOj8pW594SJFP17A6VRkZCTdunRg9ryFTJ00UW9brz79ADgQsC/Z99ar34B69RvoXnt65uXa1av8snhBlkho6jdoSP0GDZPddv3aNY4dPcKx0+coWjSh9XDWT/PJ656TdWt+p/OX3XRlz54JZPas6Rw4dJx8Hq5pEnt64NOgIT5vOX8AR44cpn2HjtSoWQuAL7v14JefF3HixDEaN21GsWLFWbVmva583nz5+H7ceLp27sCrV6/0vuRkdtLlZDjS5fQBYmJiWLd6FZ936IxGoyE6OhqNRoOZmZmujJm5OUZGRhw9dFDvvbN+nEp+d2dqVSnPnJk/8urVq7QOXzU3b1yjSF43ShctQPcuHbh37y4Aj0JDOXH8GI5OTtSvXY2CeVxpXL82hw/9o3vvmdOnCAp6gJGRETUql6ewZ27aNG/MxQvn1Toc1Qzq3xufBo2oXcfbIPVFaCPIbm9vkLoysuiYaADMzcx164yMjDAzM+Pwa7/Hz58/p0vH9kyf+dNbE/SsqnJlL7Zu2UzQgwcoisL+fXu5fu0qdb3rv/U92n+n689KyQyk76dtZzSS0HyAbZv/IiIinHZfdASgfIVKfGJpydjvRvD8+XOioqIY/e1Q4uLiCAl+qHtfj569WLx0JX9t86fTl92ZMW0SY0YNV+sw0lS5ChWZu+hX1v21lR9n/cSd27do5F2LZ8+ecfv2TQAmTRhHpy7dWL9xK6VKl6FFo/rcuJ4wfuH27Vu6MoOHf8vqP/7CLnt2mjaoy9OwMNWOK62tX7uaM4GnGfPDxHcXToEbN66zaP5PdOna3SD1ZWSFChXGzd2d77/7lqdPnxITE8P0aZN5cP8+wQ///3s8bPAAKnt50aRZcxWjTZ9+nDmHwkWKUiCvG3ZWZrRo2pDps36iWvUayZZ//Pgxk/zGy/UnPoiqCY2fnx8VKlTA2toaJycnWrRowZUrV9QMKVV+W74E7/oNyJkzoak5h6MjS1asZsffW3F3tsPT1YGIiHBKlS6D5rUxIN/0GUC1GjUpVrwkXbp9xQ8Tp7B4wVyio6PVOpQ0U8+nIS1ataF4iZLUrefDug1biIgIZ+Mf64iPjweg85fdad+xMyVLl2HilOnkL1iI35YvAdCVGTR0BM1atKJ02XLMXfgLGo2GjX+uf+t+M5P79+4xbMgAfl6yAnNz83e/4R2CHjygVbNGtGjVhs5fygdKtmzZWLXmD65fu4qbiwOOdpYE7NtHfZ+GurFcWzdvImDfXiZPm6lusOnU/LlzOH70COv++It/jpzAb/I0BvbrzZ7du5KU1Wq1tG7RhMKFizLyuzFpH6zKErucDLVkZaq27e3fv59evXpRoUIFXr16xbfffkv9+vW5ePFikoF36c29u3fYv3c3y1bpD2CrXbceJ89d4cnjx5iYmGBrZ0eRvLlp2SbvW+sqV74ir1694u6d23rjRbICWzs78ucvyM2b16lRK+Ex9oXeGA9TqFBh7t+7B4DLv037hYoU0W03MzMjTx5P7v/bdZXZBZ4+yaPQUKp7ldeti4uL4+A/ASxaMJfHES8wNjZOUV0Pg4Jo3KAulSp7MXvuwo8VcoZTpmw5Dh8/TUREBDExMTg6OlKrWmXKlC0HwP59e7h58wa5nLLrva+9bxuqVKvOdv+9aoSdLrx48YIxo0eyeu2fNGjUGIASJUpy9mwgs2b8SJ26/+8iffbsGS2aNsTKyprV6/4kW7ZsaoWtGplYz3BUTWi2b9+u93rp0qU4OTlx8uRJatRIvmkyvVi1YhmOjk7Ub9Ao2e0OOXIAELBvL48ehdKgUZO31nXu3BmMjIxwdHT6KLGmZ5GRkdy6dYO2Lu1x98hDzpyuXL+q30p3/do1vOv7AFCqTDnMzMy4fvUqXlWqARAbG8vdu3dwc/dI8/jVULN2XY6cOKO3rmePrhQsVIgBg4amOJkJevCAxg3qUrpMWeYv+lXvTjKRwNbWFki4Bk+dPMF3348DYNCQ4XR6bXAwQKWyJZk0dTqNGjdN8zjTk9jYWGJjY/VapQGMjYx1LayQ0DLTvEkDzMzMWPfnXwZpbRRZW7oafZU4B4H9WwYmRkdH63XLaLXaNInrTfHx8az6bRlt23dIMoBt5YqlFCxUmBw5HDl+7AjfDh1Iz979dC0vx48e5uSJY1SrUQsrK2uOHzvCqGGD+dT3c+yyZ09ud5nKdyOG0KBRE9zcPXj4MIhJ48dibGxM60990Wg09BkwCL/xYyleshQlSpbi99+Wc+3qZZatWgOAjY0NXbp9xaTxY8mVOzdu7h7MmfEjAC1atVHz0NKMtbU1RYsV11tnaWmJvb2Dbn1IcDAhIcHcvHEdgIvnz2FlbU1uN3fs7e0JevCARj51cHf3YILfVB4/eqSrKysMcI2MjNSdG4A7t29x9kwg2bPb4+buzp9/rCNHDkfc3Ny5cP4cQwf3p0mzFtStlzCo1dnFJdnz5ObmTh5PzzQ7DrVERkZy47Xzd/v2Lc6cCcT+3/NXvUZNRo4YioWFBe7uHhw4sJ9VK1cwaUrC76pWq6VZYx+eP3/OL0tWoNVqdX/PHR0dU5yUZwbSQmM46SahiY+Pp3///lStWpXixYsnW8bPz4+xY8emcWRJ7d+7m/v37tK+Q+ck265fu8r470fx9GkY7h55GDhkOD1799dtNzUz48/1a5k88QdioqNx9/Dk6979+KZP/yR1ZUYPHjygW6cvCAt7Qo4cjlSqUhX/fQd1t8P37N2Ply9f8u3QQYQ/DaNYiZL8uWU7nnnz6eoYN3EyJibGfN2tMy9fvKBchYr8tc0/SySEKfXLzwuZNGGc7nWDerUAmL/oF9p36MzePf7cvHGdmzeuUzi/u957tS/iyOxOnTxBo/p1dK+HDx0EQPsOnVj48xKCHz5kxNBBhIaE4JIzJ+3ad2D4t9+pFW66c+rkCRq+5fwt+nkJS1f8zvfffcuXnb/gaVgY7u4efD92PN16fA1A4OlTHD92FIASRQvo1X3xyk088uRJmwMRmYpGURRF7SAAevbsyd9//80///xD7ty5ky2TXAuNm5sbt4Ke6OZ9EamT1WeWNATjLP6t6EPJNfhh5Ox9GK1WS05HOyL+vW08Lfdra2tL8eF/YWxmmDGjcdFRnJ/UPM2PJb1IFy00vXv3ZsuWLQQEBLw1mYGEwZ+vz/EihBBCZGQaDNjllMXTW1UTGkVR6NOnDxs2bGDfvn14ZoG+ZyGEEEIYnqoJTa9evVi1ahV//fUX1tbWBAcHAwl3FlhYWKgZmhBCCPHRyaMPDEfVhGb+/PkA1KpVS2/9kiVL6Ny5c9oHJIQQQqQhucvJcFTvchJCCCGE+FDpYlCwEEIIkRVJl5PhSEIjhBBCqES6nAxH5joXQgghRIYnLTRCCCGESqTLyXCkhUYIIYQQGZ600AghhBAqkTE0hiMJjRBCCKEWA3Y5ZfEnH0iXkxBCCCEyPmmhEUIIIVQiXU6GIwmNEEIIoRK5y8lwpMtJCCGEEBmetNAIIYQQKpEuJ8ORhEYIIYRQiXQ5GY50OQkhhBAiw5MWGiGEEEIl0uVkONJCI4QQQogMT1pohBBCCJVIC43hSAuNEEIIoZLEQcGGWlIjICCApk2b4urqikajYePGjXrbFUVh9OjR5MyZEwsLC7y9vbl27ZpembCwMNq3b4+NjQ12dnZ07dqVyMhIvTJnz56levXqmJub4+bmxpQpU5LEsm7dOgoXLoy5uTklSpRg27ZtqTsYJKERQgghsqSoqChKlSrF3Llzk90+ZcoUZs+ezYIFCzh69CiWlpb4+Pjw8uVLXZn27dtz4cIF/P392bJlCwEBAfTo0UO3XavVUr9+fTw8PDh58iRTp05lzJgxLFq0SFfm0KFDtGvXjq5du3L69GlatGhBixYtOH/+fKqOR6MoipLKc5BuaLVabG1tuRX0BBsbG7XDyZCMjLJ2E6UhGGfxZt4PJdfgh5Gz92G0Wi05He2IiIhI08+RxM+vqn47MTG3NEidr15GcXBE/fc6Fo1Gw4YNG2jRogWQ0Drj6urKoEGDGDx4MAARERE4OzuzdOlSfH19uXTpEkWLFuX48eOUL18egO3bt9OoUSPu37+Pq6sr8+fPZ+TIkQQHB2NqagrA8OHD2bhxI5cvXwagbdu2REVFsWXLFl08lStXpnTp0ixYsCDFxyAtNEIIIYRKPkaXk1ar1Vuio6NTHdetW7cIDg7G29tbt87W1pZKlSpx+PBhAA4fPoydnZ0umQHw9vbGyMiIo0eP6srUqFFDl8wA+Pj4cOXKFZ4+faor8/p+Essk7ielJKERQgghMhE3NzdsbW11i5+fX6rrCA4OBsDZ2VlvvbOzs25bcHAwTk5OettNTEywt7fXK5NcHa/v421lErenlNzlJIQQQqjkY9zldO/ePb0uJzMzM4PUn95JC40QQgihEg0G7HL6t04bGxu95X0SGhcXFwBCQkL01oeEhOi2ubi4EBoaqrf91atXhIWF6ZVJro7X9/G2MonbU0oSGiGEEELo8fT0xMXFhd27d+vWabVajh49ipeXFwBeXl6Eh4dz8uRJXZk9e/YQHx9PpUqVdGUCAgKIjY3VlfH396dQoUJkz55dV+b1/SSWSdxPSklCI4QQQqjESKMx6JIakZGRBAYGEhgYCCQMBA4MDOTu3btoNBr69+/P+PHj2bRpE+fOnaNjx464urrq7oQqUqQIDRo0oHv37hw7doyDBw/Su3dvfH19cXV1BeDzzz/H1NSUrl27cuHCBdasWcOsWbMYOHCgLo5+/fqxfft2fvzxRy5fvsyYMWM4ceIEvXv3TtXxyBgaIYQQIgs6ceIEtWvX1r1OTDI6derE0qVLGTp0KFFRUfTo0YPw8HCqVavG9u3bMTc3171n5cqV9O7dm7p162JkZETr1q2ZPXu2brutrS07d+6kV69elCtXjhw5cjB69Gi9uWqqVKnCqlWrGDVqFN9++y0FChRg48aNFC9ePFXHI/PQZHEyB8iHk3loPoxcgx9Gzt6HUXsemtpTd2FiYaB5aF5EsXeId5ofS3ohLTRCCCGESuRZToYjY2iEEEIIkeFJC40QQgihEiNNwmKourIySWiEEEIItWgM2FWUxRMa6XISQgghRIYnLTRCCCGESl5/qKQh6srKMkVCY2pihKmJNDYJdbyMjVc7hAwti/8N/mDyt+/DxMWrO3OJ5t//DFVXVia/CUIIIYTI8DJFC40QQgiREcldToYjLTRCCCGEyPCkhUYIIYRQicwUbDiS0AghhBAqkbucDEe6nIQQQgiR4UkLjRBCCKESI40GIwM1rRiqnowqRQnNpk2bUlxhs2bN3jsYIYQQIiuRLifDSVFC06JFixRVptFoiIuL+5B4hBBCCCFSLUUJTXy8zIQqhBBCGJrc5WQ4HzSG5uXLl5ibmxsqFiGEECJLkS4nw0n1XU5xcXH88MMP5MqVCysrK27evAnAd999xy+//GLwAIUQQggh3iXVCc2ECRNYunQpU6ZMwdTUVLe+ePHi/PzzzwYNTgghhMjMEu9yMtSSlaU6oVm+fDmLFi2iffv2GBsb69aXKlWKy5cvGzQ4IYQQQoiUSPUYmgcPHpA/f/4k6+Pj44mNjTVIUEIIIURWoPl3MVRdWVmqW2iKFi3KgQMHkqxfv349ZcqUMUhQQgghRFaQeJeToZasLNUtNKNHj6ZTp048ePCA+Ph4/vzzT65cucLy5cvZsmXLx4hRCCGEEOI/pbqFpnnz5mzevJldu3ZhaWnJ6NGjuXTpEps3b6ZevXofI0YhhBAiUzLSGHbJyt5rHprq1avj7+9v6FiEEEKILEUm1jOc955Y78SJE1y6dAlIGFdTrlw5gwUlhBBCCJEaqU5o7t+/T7t27Th48CB2dnYAhIeHU6VKFVavXk3u3LkNHaMQQgiRaWXxhhWDSfUYmm7duhEbG8ulS5cICwsjLCyMS5cuER8fT7du3T5GjEIIIUSmJHc5GU6qW2j279/PoUOHKFSokG5doUKFmDNnDtWrVzdocEIIIYQQKZHqhMbNzS3ZCfTi4uJwdXU1SFBCCCFEVmDIu5Oy+l1Oqe5ymjp1Kn369OHEiRO6dSdOnKBfv35MmzbNoMEJIYQQQqREilposmfPrtc3FxUVRaVKlTAxSXj7q1evMDEx4csvv6RFixYfJVAhhBAis5Hbtg0nRQnNzJkzP3IYQgghRNYjz3IynBQlNJ06dfrYcQghhBBCvLf3nlgP4OXLl8TExOits7Gx+aCAhBBCiKzCSKPByEBdRYaqJ6NK9aDgqKgoevfujZOTE5aWlmTPnl1vyeymTZlEzaqVyJnDFk83F3w/bcnVq1f0yty8cYN2n7UiT25nXB3t6Ni+LaEhIUnq2v73VmpX98LRzhI3Fwd8P22ZVoehqnedwzu3b2NtbpzssuGPdbpyyW1fv3a1GoeU5iKfPWPksIGUKZoPN0drGtWtzumTx5MtO7jfNzhaZ2PB3Fl6688EnqJNswbky52Dgu7ODOzzNZGRkWkRvuqe/Xv+ShfNR25HaxrWrc6pf89fbGwsY78bQfVKpXF3tqVYAXe+6dGZhw+DdO+/e+c2/b7pTtniBcjtaE35koWYNGFski94WcH0qZOxsTBm2OABAISFhTF4QF/KliyCU3ZLihbIw5CB/YiIiEj2/U+ePKFwPndsLIwJDw9Pw8jTB43GsEtWluqEZujQoezZs4f58+djZmbGzz//zNixY3F1dWX58uUfI8Z05eCB/XT/qid7Ag6xaesOYmNjadG4AVFRUUBCwteiSQM0Gg1bt+/Cf+8BYmJi+Kx1c+Lj43X1/LXhD3p82YkvOnbi0PHT+O89wGdt26l1WGnqXecwt5sb128/0FtGfjcGKysr6vk01Ktr/qJf9Mo1adZChSNKe/17f8X+PbuZu2gp+4+cplbderRu1oCHQQ/0ym3dtJETx4/iklN/SoXgh0G0adYAz7z52LHnIGs2bOHypYv0+bprWh6Gavr3/op9e3Yzb9FSAt44fy+eP+fsmdMMGjaS3QeOsWzlWq5fu8oXbf//hePa1SvEx8fz46x5/HPsDD/4TWPZL4sYP2aUikeV9k6eOM6SXxZRvERJ3brgh0EEPwxigt8Ujpw8y/zFv7LLfwe9v05+4tXeX3ejWIkSaRWyyMQ0iqIoqXmDu7s7y5cvp1atWtjY2HDq1Cny58/PihUr+P3339m2bdvHijUJrVaLra0tD0KfqtbV9ejRI/K6ufC3/16qVa/Bbv+dtGremHvBT3QxRURE4ObiwF9btlO7rjevXr2iWKG8fDvqezp1yRofIP/lzXOYnKqVylGqdBnmLfxZt87a3JhVa/+gqcpJzMvY+HcXMqAXL17gmTM7y1f/Sf0GjXTr61avSN16Dfh29DgAHgY9wKd2VdZu3MrnbZrT45s+fN2rHwDLf13MpPFjOH/9HkZGCd9rLl44R83KZTkaeIm8+fKn2fGk9ZfKFy9ekCdndla8cf7qVK+I92vn73WnTh6nfq0qBF68QW4392TrnTPzR5b+spCT565+tNiTY2qS6u+lBhEZGUl1r/JMn/UTUydNpETJUkyeNiPZshv+WEf3LzsS/OSZ7u5YgJ8XzefP9esY9u0omjasx92HT3SP1EkrWq2W3M7ZiYiISNPPkcTPr05Lj2D6iZVB6ox5HsmyzpXT/FjSi1T/JoSFhZE3b14gYbxMWFgYANWqVSMgIMCw0WUAWm1CM6q9vT0A0THRaDQazMzMdGXMzc0xMjLi8KGDAASePkXQgwcYGRlRtVI58ufJRatmjbh44XzaH0A68OY5fNPpUyc5eyaQjp2/TLJtUP8+eORyola1yixf+iupzM8zpLhXr4iLi8Pc3Fxvvbm5BUcPJ1xj8fHxfNO9M736DaRwkWJJ6oiOiSabqakumUl8P6CrI7N69ZbzZ2FuwZG3HPszrRaNRoOtrd1b632mjcAuC3S7JxrUvzc+DRpRu473O8tqtRFY29joJTOXL11kst94Fv68VO86zGrU6nKKi4vju+++w9PTEwsLC/Lly8cPP/yg9zdUURRGjx5Nzpw5sbCwwNvbm2vXrunVExYWRvv27bGxscHOzo6uXbsm6bo+e/Ys1atXx9zcHDc3N6ZMmfJB5+xtUn0V5c2bl1u3bgFQuHBh1q5dC8DmzZvTPLNWW3x8PMMGD6CyV1WKFisOQIWKlbG0tGT0yOE8f/6cqKgoRg4fQlxcHMHBDwG4fesmAH7jxzF0+Les+3MTdnbZaVi/ji5BzCqSO4dvWr70VwoVLkJlryp660eNHsuy31azaesOmrVoxcB+vVkw76e0CFtVVtbWVKhYmR8nTyD4YRBxcXGsW72SE8eOEBIcDMDs6VMxMTGhR88+ydZRvWZtQkOC+Wnmj8TExBD+9Ck/fD8SQFdHZmX97/mbNnkCD/89f2tXr+T4a+fvdS9fvmTs6BG0+rQt1m/51nvzxnUWL5xLpy7dP3b46cL6tas5E3iaMT9MfGfZJ48fM8VvAl2+/P+5iY6O5stO7flh4mTc3JNv8RIf1+TJk5k/fz4//fQTly5dYvLkyUyZMoU5c+boykyZMoXZs2ezYMECjh49iqWlJT4+Prx8+VJXpn379ly4cAF/f3+2bNlCQEAAPXr00G3XarXUr18fDw8PTp48ydSpUxkzZgyLFi0y+DGlOqHp0qULZ86cAWD48OHMnTsXc3NzBgwYwJAhQwweYHo2sF9vLl24wNIVq3TrHB0dWb5yDX9v3YKLgw25nLITER5O6TJldd9CEsfSDB42guYtW1OmbDnmL/4VjUbDxtcGvWYFyZ3D17148YJ1a35PtnVm2Lej8KpSlVKlyzBw8FD6DxzCrOlZY7bquYuXoigKJQp6kMvBksULfqLVp20xMjLizOmTLJo/hzkLfnnrRFuFixTjp4W/Mm/ODNydbCiWPzceHnlwdHLOEt+W5712/lzfOH+vi42NpWvHdiiKwrQZc5Ot62HQA9q2akKzFq3p2CXzP6D3/r17DBsygJ+XrEjSyvUmrVZLm5ZNKVSkCCNGfa9bP+a7bylYqDC+7b742OGme4l3ORlqSalDhw7RvHlzGjduTJ48eWjTpg3169fn2LFjQELrzMyZMxk1ahTNmzenZMmSLF++nKCgIDZu3AjApUuX2L59Oz///DOVKlWiWrVqzJkzh9WrVxMUlDCIfuXKlcTExPDrr79SrFgxfH196du3L9OnTzf8uUztGwYMGEDfvn0B8Pb25vLly6xatYrTp0/Tr1+/VNU1f/58SpYsiY2NDTY2Nnh5efH333+nNiRVDOrfh+3btrJ1x25y5c6tt61uvfqcvXSNm/eCuf0glMVLlhMU9IA8nglddS4uOQEoXKSo7j1mZmZ4eubl3r17aXcQKvuvc5ho45/ref78Oe3ad3hnfeUrVuTBg/tER0cbOtR0xzNvPjZt38Pt4HACL99i577DxMa+wiOPJ4cP/cPjR6GULpIXFztzXOzMuXf3Dt9/O5Syxf4/Nqb1Z+24eOM+Z6/e4cqdEIZ8O5onjx/h4emp4pGlDc+8+di8fQ93gsM5c/kW/q+dv0SJycz9e3f446/tybbOPHwYRPNG9ahQqTIz5ixIy0NQTeDpkzwKDaW6V3myW5mS3cqUfw7sZ8G8OWS3MiUuLg5IuJOsVbNGWFtbs2rNn2TLlk1XR8D+vWz8c73u/U0b1gPAM7cTE34Yo8ZhZSparVZvSe5vYpUqVdi9ezdXryaM+Tpz5gz//PMPDRsm3Hhx69YtgoOD8fb+f5eira0tlSpV4vDhwwAcPnwYOzs7ypcvryvj7e2NkZERR48e1ZWpUaMGpqamujI+Pj5cuXKFp0+fGvS4P2geGgAPDw88PDze6725c+dm0qRJFChQAEVRWLZsGc2bN+f06dMUK5a03z89UBSFwQP6snnTRrbt3EOe//jjnyNHDgD2793Do9BQGjVpCkDpsuUwMzPj2tUrVKlaDUj443nnzu0s0fyamnO4fOkSGjVpiqOj4zvrPXfmDNmzZ9cbv5TZWVpaYmlpSfjTp+zdvZPvx/nRpHkratauq1fusxaN+dS3PZ9/kXSSTCcnZwBWLl+Cubk5tWq/e0xEZpHc+YP/JzM3b1xn41Z/7B0ckrz3YdADmjeqR6kyZZkz/5cs0bIFULN2XY6cOKO3rmePrhQsVIgBg4ZibGyMVqulZdOGmJmZsXr9xiQtOSt+X8fLFy90r0+dPM43X3Vj+6795M2bL02OI70w5O3WifW4ubnprf/+++8ZM2aM3rrhw4ej1WopXLgwxsbGxMXFMWHCBNq3bw9A8L/dr87Oznrvc3Z21m0LDg7GyclJb7uJiQn29vZ6ZTzf+BufWGdwcLBBp3tJUUIze/bsFFeY2HqTEk2bNtV7PWHCBObPn8+RI0fSbUIzsF9v1q35ndXrNmBtZa3rc7extcXCImFQ5YplSyhUuAg5cjhy7Ohhhg4eQK++/SlYsFBCWRsbunb/ionjx5I7txtuHh66rpKWrT9V58DSUErOIcCNG9c5+E8Af/y1JUkd27ZuJjQkhIqVKmNmbs7e3f5Mm+JH3/6D0uw41LRn104URSF/gYLcunmDMaOGUaBAIdp16Ey2bNmSfABny5YNJ2dn8v97DQL8vHAuFSp5YWVpxb69uxg7ajijxk7ANguMhXvb+fu8Q2diY2Pp8kVbzp45zap1G4mLjyMkJOEazZ7dHlNTUx4GPaBZI2/c3NwZN2Eyjx8/0tXt7Oyi1mGlCWtr6yTj3SwtLbG3d6BoseJotVpaNGnAixfPWbxkOc+0Wp5ptQDkcHTE2Ng4SdLy5MljAAoVLpLlxmJ+jGc53bt3T+8up+S+5K1du5aVK1eyatUqihUrRmBgIP3798fV1TXDPh0gRQnNjBnJ34r3Jo1Gk6qE5nVxcXGsW7eOqKgovLy8ki0THR2t13Sm/feXJC39vCihWblh/Tp66+cv+oUvOnYG4Nq1q4wZPZKnYWG4e+RhyLBv6d23v1758X5TMDYxoXvXTrx88YLyFSqydfuuLDE5YUrOIcCKpUvIlSs3db3rJ6kjW7ZsLF44nxFDB6EoCnnz5cdv8jQ6d80agzK12ggmjBlF0IP72GW3p0nzlowc/YNes/67nD55nCkTxhEVFUmBgoWYNmsen2WRMQ1abQTjXzt/TV87f3fv3Gb7ts0A1KpSXu99G7ftolr1muzbs4tbN65z68Z1ShTKo1fm8bPYtDqMdOlM4ClOHE/obihdrKDetnOXb+DhkUeFqLKWxGEc/2XIkCEMHz4cX19fAEqUKMGdO3fw8/OjU6dOuLgkJOYhISHkzJlT976QkBBKly4NgIuLC6GhoXr1vnr1irCwMN37XVxcCHljYtnE14llDCXV89AY2rlz5/Dy8uLly5dYWVmxatUqGjVqlGzZMWPGMHbs2CTr1ZyHRoi0nocms8nik5t+MLXmocks1J6Hpsdvxww6D82iLyqm6FgcHBwYP348PXv21K3z8/NjyZIlXL16FUVRcHV1ZfDgwQwaNEgXs5OTE0uXLsXX15dLly5RtGhRTpw4Qbly5QDYuXMnDRo04P79+7i6ujJ//nxGjhxJSEiI7gvXt99+y59//snly5cNctyJVP9NKFSoEIGBgRw9epSePXvSqVMnLl68mGzZESNGEBERoVuy0gBaIYQQmU9il5OhlpRq2rQpEyZMYOvWrdy+fZsNGzYwffp0WrZsqYurf//+jB8/nk2bNnHu3Dk6duyIq6srLVq0AKBIkSI0aNCA7t27c+zYMQ4ePEjv3r3x9fXF1TVhdvLPP/8cU1NTunbtyoULF1izZg2zZs1i4MCBBj+XHzwo+EOZmpqSP3/CnRflypXj+PHjzJo1i4ULFyYpa2ZmlqUGfAohhBAfw5w5c/juu+/45ptvCA0NxdXVla+++orRo0frygwdOpSoqCh69OhBeHg41apVY/v27XqDvFeuXEnv3r2pW7cuRkZGtG7dWm/cra2tLTt37qRXr16UK1eOHDlyMHr0aL25agxF9S6nN9WpUwd3d3eWLl36zrLp4dEHQkiX04eRLqcPI11OH0btLqevVx3HzEBdTtHPI1nweYUs++gDVVtoRowYQcOGDXF3d+fZs2esWrWKffv2sWPHDjXDEkIIIdKEkSZhMVRdWZmqCU1oaCgdO3bk4cOH2NraUrJkSXbs2EG9evXUDEsIIYQQGcx7JTQHDhxg4cKF3Lhxg/Xr15MrVy5WrFiBp6cn1apVS3E9v/zyy/vsXgghhMgUPsY8NFlVqjtf//jjD3x8fLCwsOD06dO6eWEiIiKYOPHdDyoTQgghhDC0VCc048ePZ8GCBSxevFhvEq+qVaty6tQpgwYnhBBCZGaJY2gMtWRlqe5yunLlCjVq1Eiy3tbWlvDwcEPEJIQQQmQJH+NZTllVqltoXFxcuH79epL1//zzD3nz5jVIUEIIIYQQqZHqhKZ79+7069ePo0ePotFoCAoKYuXKlQwePFhvCmUhhBBC/DcjjcagS1aW6i6n4cOHEx8fT926dXn+/Dk1atTAzMyMwYMH06dPn48RoxBCCJEpGWG4ZxBl9SkWU53QaDQaRo4cyZAhQ7h+/TqRkZEULVoUKyvDzHQohBBCCJFa7z2xnqmpKUWLFjVkLEIIIUSWIoOCDSfVCU3t2rX/c/KePXv2fFBAQgghRFZhhOHGvhhl8SejpTqhKV26tN7r2NhYAgMDOX/+PJ06dTJUXEIIIYQQKZbqhGbGjBnJrh8zZgyRkZEfHJAQQgiRVUiXk+EYbFD0F198wa+//mqo6oQQQohMT2YKNhyDJTSHDx/G3NzcUNUJIYQQQqRYqrucWrVqpfdaURQePnzIiRMn+O677wwWmBBCCJHZaTQYbFBwVu9ySnVCY2trq/fayMiIQoUKMW7cOOrXr2+wwIQQQgghUipVCU1cXBxdunShRIkSZM+e/WPFJIQQQmQJMijYcFI1hsbY2Jj69evLU7WFEEIIA5BBwYaT6kHBxYsX5+bNmx8jFiGEEEKI95LqhGb8+PEMHjyYLVu28PDhQ7Rard4ihBBCiJTRGPi/rCzFY2jGjRvHoEGDaNSoEQDNmjXTewSCoihoNBri4uIMH6UQQgiRCRmyqyirdzmlOKEZO3YsX3/9NXv37v2Y8QghhBBCpFqKExpFUQCoWbPmRwtGCCGEyEqkhcZwUnXb9n89ZVsIIYQQqaPRaAz22ZrVP6NTldAULFjwnScsLCzsgwISQgghhEitVCU0Y8eOTTJTsBBCCCHej3Q5GU6qEhpfX1+cnJw+VixCCCGEEO8lxQlNVu+bE0IIIQxNHn1gOKm+y0kIIYQQhmGk0RjsaduGqiejSnFCEx8f/zHjEEIIIYR4b6kaQyOEEEIIw5FBwYYjCY0QQgihFgOOocnij3JK/cMphRBCCCHSG2mhEUIIIVRihAYjAzWtGKqejCpTJDQmxkaYGEtjk1CHudoBZHBBT1+qHUKGFhX9Su0QMrTIZ5Gq7l9u2zYcyQKEEEIIkeFlihYaIYQQIiOSu5wMR1pohBBCCJHhSUIjhBBCqCRxpmBDLanx4MEDvvjiCxwcHLCwsKBEiRKcOHFCt11RFEaPHk3OnDmxsLDA29uba9eu6dURFhZG+/btsbGxwc7Ojq5duxIZqT8u6ezZs1SvXh1zc3Pc3NyYMmXK+5+w/yAJjRBCCKGSxEHBhlpS6unTp1StWpVs2bLx999/c/HiRX788UeyZ8+uKzNlyhRmz57NggULOHr0KJaWlvj4+PDy5f8H8rdv354LFy7g7+/Pli1bCAgIoEePHrrtWq2W+vXr4+HhwcmTJ5k6dSpjxoxh0aJFBjl/r5MxNEIIIUQWM3nyZNzc3FiyZIlunaenp+7fiqIwc+ZMRo0aRfPmzQFYvnw5zs7ObNy4EV9fXy5dusT27ds5fvw45cuXB2DOnDk0atSIadOm4erqysqVK4mJieHXX3/F1NSUYsWKERgYyPTp0/USH0OQFhohhBBCJUYYsMvp33lotFqt3hIdHZ1kv5s2baJ8+fJ8+umnODk5UaZMGRYvXqzbfuvWLYKDg/H29tats7W1pVKlShw+fBiAw4cPY2dnp0tmALy9vTEyMuLo0aO6MjVq1MDU1FRXxsfHhytXrvD06VMDn0shhBBCqOJjdDm5ublha2urW/z8/JLs9+bNm8yfP58CBQqwY8cOevbsSd++fVm2bBkAwcHBADg7O+u9z9nZWbctODgYJycnve0mJibY29vrlUmujtf3YSjS5SSEEEJkIvfu3cPGxkb32szMLEmZ+Ph4ypcvz8SJEwEoU6YM58+fZ8GCBXTq1CnNYjUkaaERQgghVGJk4AXAxsZGb0kuocmZMydFixbVW1ekSBHu3r0LgIuLCwAhISF6ZUJCQnTbXFxcCA0N1dv+6tUrwsLC9MokV8fr+zAUSWiEEEIIlWg0GoMuKVW1alWuXLmit+7q1at4eHgACQOEXVxc2L17t267Vqvl6NGjeHl5AeDl5UV4eDgnT57UldmzZw/x8fFUqlRJVyYgIIDY2FhdGX9/fwoVKqR3R5UhSEIjhBBCZDEDBgzgyJEjTJw4kevXr7Nq1SoWLVpEr169gIREq3///owfP55NmzZx7tw5OnbsiKurKy1atAASWnQaNGhA9+7dOXbsGAcPHqR37974+vri6uoKwOeff46pqSldu3blwoULrFmzhlmzZjFw4ECDH5OMoRFCCCFUovl3MVRdKVWhQgU2bNjAiBEjGDduHJ6ensycOZP27dvrygwdOpSoqCh69OhBeHg41apVY/v27Zib//+RvCtXrqR3797UrVsXIyMjWrduzezZs3XbbW1t2blzJ7169aJcuXLkyJGD0aNHG/yWbQCNoiiKwWtNI1qtFltbW0KeROgNgBIiLb2Ki1c7hAxNnrb9YeRp2x8m8pmWGiXciIhI28+RxM+vBXsvYGFlbZA6X0Q+4+vaxdL8WNILaaERQgghVPI+jyz4r7qyMklohBBCCBVl7TTEcGRQsBBCCCEyPGmhEUIIIVSS2odKvquurEwSGiGEEEIlqZ0/5l11ZWXS5SSEEEKIDE9aaIQQQgiVvP7IAkPUlZVJQiOEEEKoRLqcDCerJ3RCCCGEyASkhUYIIYRQiVqPPsiMpIVGCCGEEBmetNAIIYQQKpExNIYjCY0QQgihErnLyXCy+vELIYQQIhOQFhohhBBCJdLlZDiS0AghhBAqkbucDEe6nAxg/LgxWGTT6C2lihfWbQ8ODubLTh3Ik9sFB1tLvCqUZcOff6gXcDo3dcokLLJpGDywPwBhYWEM6NeHksUKkd3aggJ53RnYvy8RERHqBqqSaVMmUbNqJXLmsMXTzQXfT1ty9eoVvTIhwcF079KRfB6uONtbU61yef7akPw1Fx0dTZWKZbE2N+bsmcA0OIK0tXD2VFo3qE6Z/M54Fffgm85tuXn9ql6Z0UP64F25OCU9HahczIOenT/jxrUrSer6c80KmtapSIk89ngV92DsiAG6bffv3aFQTsskS+DJYx/9GD+mdSt+5rMGVahePDfVi+emU0tvDu71120fP6IfzWqUwquQM3XK5mVAt3bceu38hj8No1fHVtSvWIhKBR1p6FWUSaMHE/lMq7efbRvX0rZBVaoUdqF+hYKMGdKL8KdhaXacIuOTFhoDKVqsGFu379K9NjH5/6nt1qUj4eHhrPtzEzly5GDN6lV80e4zDh45QekyZdQIN906cfw4vyxeSIkSJXXrHgYF8fBhEH6Tp1GkSFHu3r1Dn15f8/BhEL+vWa9itOo4eGA/3b/qSbnyFXj16hVjRo+kReMGHA88j6WlJQA9unYiIiKCNes34uCQg3Vrfqdje18CDh2jVGn9a+67b4eRM2dOzp09o8bhfHTHDv9D+y49KFG6HHGvXjHdbwxdfZuxNeAkn3yScL6KlSxD01ZtyZnbjYinYcz5cSJdfZux+9hFjI2NAViyYDa/LpzN0O8mUKpsBZ4/j+LBvbtJ9rd07RbyFyqie22X3SFtDvQjccqZi77DxuCeJx+KorD5j1UM6NGO37ceIF/BIhQpUZqGLT4jp2tuIiKesnDmJHp1bMnmA2cxNjbGyEhDrXqN6DV4FHb2Obh35yaTvxvMxPCnTJz9CwCBJ44weuBXDPrOjxreDQgNfsjEkQP4YXgffly4UuUz8HHJ07YNR6MoiqJ2EO9Lq9Via2tLyJMIbGxsVItj/LgxbP5rI0dPBia7PYedFbN/ms/nX3TQrcvl7MD4iZPp0rVb2gSZAURGRuJVsSyz5sxj0sTxlCxVmmnTZyZb9o/16/iy0xc8iYjSSx7V8CouXtX9P3r0iLxuLvztv5dq1WsA4OJgw4zZc2nX/v/XnLurI+PG+9H5y/9fczt3/M2IoYNZuXodFcqU4ODRk5QsVTpN4w96+jJN9xf2+BFeJfLw2587qOBVLdkyly+eo3ndyvgfPod7nrxEhD+lRpkCLFi+Dq/qtZN9z/17d6hbsSgb/Q9RpHipj3kIeqKiX6XZvhLVKuVB/29/oEXbjkm2Xb10Ht+GVflr/2ncPPIm+/7flyxg+aLZ/H34IgDLF81m/W+/sCng/0n16qULWbpgJtuPXPo4B/GvyGdaapRwIyIibT9HEj+/Vh+6xidW1gap83nkM3yrFEjzY0kvpMvJQK5fv4anuytFCualc4f23L37/29ulb2qsH7dGsLCwoiPj2ftmtW8fPmSGjVrqRdwOtS/Ty8aNGxMnbre7yyr/fcXVu1kJj3QahO63uzt7XXrKlX24o/1a3XX3Pq1q4l++ZLqr11zoSEh9PnmKxb/ugwLi0/SOmzVPPu3q8M2e/Zktz9/HsWfq1eQ2z0PLq65ATgYsId4JZ6Qh0E0rF6WGmUL0K9HBx4+uJ/k/T07fYZXcQ/aNfNm946tH+9AVBAXF8eOTet58eI5JctWTLL9xfMoNq1bSS43D1xy5k62jkchD9mzfTNlK1XVrStZtiLBDx/wz96dKIrCk0eh7Nr2F1Vr1/toxyIyH/k0MIAKFSux6JelFCxYiODgh0z4YSzetatzMvA81tbW/Pb7Wjp83pZczg6YmJjwySefsGb9BvLlz6926OnG2jWrCTx9in+OHH9n2cePH+M38Qe+7NYjDSJL3+Lj4xk2eACVvapStFhx3fplK9fQ+QtfPFwdddfcqjV/kC9fwjWnKApfd+9C125fUbZcee7cvq3SEaSt+Ph4Jo4eStkKXhQsXExv28qli5j2wyieP4/CM19BlqzZjKmpKQD379xCiY9nwexpjPxhCtY2tsycPJYubZuyac9RTE1N+eQTS4Z/70fZil5ojIzYuXUjvbq0Ze6SNdT1aazG4RrMtcsX6NyqHjHRL7H4xIofF64kb4H/jxNcu2Ixs/y+58XzKPLkLcC83zaS7d9zl2hEny/Z77+Nly9fUMO7IaMnzdFtK12+MhNmLmZ47y7ERL/k1atX1PBuyPBxP6bZMapFupwMJ9200EyaNAmNRkP//v3VDiXVfBo0pHWbTylRsiT16vuwcfM2IsLD+WPdWgDGfv8d4eHhbNuxi4NHTtC3/0C+aPcZ58+dUzny9OHevXsMGdiPJctXYm5u/p9ltVotLZs1pkiRoowaPSZtAkzHBvbrzaULF1i6YpXe+vFjRxMREcHmbTsJOHSM3n0H0OkLXy6cT7jmFsz7iWfPIhk0dLgaYatm7IgBXLt8kRkLliXZ1qxVWzb4H+K3P3eQJ19++vfoQPTLhO6w+HiF2NhYRo2fSvXa9ShdriLT5y3lzq3rHD24HwB7hxx0+bovpcpWoGTpcgwe+QPNWvvyy7yZaXmIH0WevAX4fdsBlm3czadffMnoQV9z89pl3faGzT/j960HWLxmG+558zOsV2fduUs06Ds/Vm4JYMbi37l/5xbTx3+r23bz2mWmjh1O975D+W3zfn5a9gdB9+8ycWT/tDpEkQmkixaa48ePs3DhQkqWLPnuwhmAnZ0d+QsU5MaN69y8cYMF837iZOB5ihZL+EZYslQpDv5zgIXz5zJn3gKVo1Xf6VMnCQ0NxatiWd26uLg4/jkQwIJ5PxERFY2xsTHPnj2jWeMGWFtbs2b9BrJly6Zi1Oob1L8P27dtZfuufeTK/f/m/Zs3brBw/lyOnTpLkaIJ11yJkqU4dPAfFi2Yx6yf5rN/3x6OHT2Mg42FXp01qlTkM9/PWfTL0rQ8lDQx7tuB7Nv1N79t2ImLa64k261tbLG2sSVP3vyUKleRioVz4f/3Jpq0/AxHZ2cA8hf8/2Bf+xyOZLd3SLbbKVGpshU4FLDH8AeTxrKZmuKeJx8ARUuU4cLZU6z6dT6j/GYB/z937p75KFmmAjVLebB3xxYaNG+jqyOHkzM5nJzxzF8QG7vsdP20Ad36DsXRyYVf502ndPlKdPqqHwAFixTH4hNLun7agG8Gf4ejk0vaH3Qa0fz7n6HqyspUT2giIyNp3749ixcvZvz48WqHYxCRkZHcunkDl/YdeP78OQBGRvqNYcbGxsTHqzuYNL2oXacuJ07rt1b16NaFQoUKM2jIMIyNjdFqtTRt5IOZmRnrN2x6Z0tOZqYoCoMH9GXzpo1s27mHPJ6eettfvEi45jRJrjkj3TU39cdZjB7zg27bw6AgWjRtyLLffqd8hUof+QjSlqIo/DByEP5/b2LFH9txc8+TkjehKAoxMdEAlK3gBcCtG1d1yVD40zCehj3BNbfbW6u5dP5spvwwjo+PJzYmJtltiqLAa+fube8HiI1OKPPyxXOMjfU/joyMjBMrNEDE6Zd0ORmO6glNr169aNy4Md7e3u9MaKKjo4mO/v8viVar/Y/SaWf40ME0btIUd3cPgoKCGD/ue4yNjfnMtx12dnbky5+f3t98hd/kaTg4OLBp00Z27/Lnz7+2qB16umBtbU2x4sX11llaWmLv4ECx4sXRarU0aVifF8+fs2TZb2i1Wt3P3tHRUXdbbVYxsF9v1q35ndXrNmBtZU1IcDAANra2WFhYULBQYfLly0+/Xj2ZMGkK9vYObNn8F3t272Ldhk0AuLm769VpaWkFgGfefHqtPZnB2BED2LJhLfOWrMHSyopHoQnny9raFnMLC+7ducW2v9ZTtaY39g45CH74gEU//Yi5hQU16/oA4JmvAHV9mjDhuyGMm/oTVtbWTJ/4PXnzF6JS1ZoAbFj7G9mymerucPLftok/Vi9n/I9z1TlwA5kzeQxVatUjp2tuoqIi2f7XOk4e+Ye5y//k/t1b7Nz8J5Vr1CG7fQ5Cg4NYMn8GZubmVKtdH4B/9u7kyaNQipUqyyefWHLj2mVmTvyO0uUr4+rmAUCNug0ZP6Iv61b8jFfNujwODWHauOEUL1UOR+ecah6+yEBUTWhWr17NqVOnOH783QNBAfz8/Bg7duxHjir1Hjy4T8cv2hH25Ak5HB2pUrUa+/85gqOjIwAbN21j1MjhtGnZlMjISPLly8/Pvy6jQcNGKkeeMQSePsXxY0cBKFZYfyD15Wu38MiTR4Wo1PPzooRuyob16+itn7/oF77o2Jls2bKx/q8tfD9qBJ+1bk5UZCR58+Vn4c9L8GmQ9a6535ctBqBD6wZ66/1mLqBV2w6Ymplz4ughli2eizYiHAdHJ8pXqsrvm3bjkMNJV37KnMVM/H4YX3VojZGRERUqV+PnVRv1uj7nzZhM0P27GJuYkDd/QWYsXE6DJi3T5kA/krAnjxg98GsePwrGytqGAoWLMXf5n1SuXodHIQ85ffwwq5bMTzh3OZwoW7EKS/7wxz5Hwt8/MzNzNqxexo8/fEtsTDTOrrmo49OULj3/Pylhs0/b8zwqkjXLFzNjwiisbGypUKUG/Yanv7/3hqZBg5F0ORmEavPQ3Lt3j/Lly+Pv768bO1OrVi1Kly7NzJkzk31Pci00bm5uqs9DI7I2teehyejSeh6azEaNeWgyE7Xnofnj6A0sDTQPTVTkM1pXypdl56FRrYXm5MmEgaBly+oPBA0ICOCnn34iOjo6SVeCmZkZZmZmaR2qEEIIIdI51RKaunXrcu6N25a7dOlC4cKFGTZsWJYbFyGEECLrkUHBhqNaQmNtbU3xZAaCOjg4JFkvhBBCZEZy27bhpJuJ9YQQQggh3pfqt22/bt++fWqHIIQQQqQZI03CYqi6sjJpoRFCCCFEhpeuWmiEEEKIrETG0BiOJDRCCCGESuQuJ8ORLichhBBCZHjSQiOEEEKoRIPhuoqyeAONJDRCCCGEWuQuJ8ORLichhBBCZHjSQiOEEEKoRO5yMhxpoRFCCCFUkniXk6GW9zVp0iQ0Gg39+/fXrXv58iW9evXCwcEBKysrWrduTUhIiN777t69S+PGjfnkk09wcnJiyJAhvHql/wT4ffv2UbZsWczMzMifPz9Lly59/0D/gyQ0QgghRBZ2/PhxFi5cSMmSJfXWDxgwgM2bN7Nu3Tr2799PUFAQrVq10m2Pi4ujcePGxMTEcOjQIZYtW8bSpUsZPXq0rsytW7do3LgxtWvXJjAwkP79+9OtWzd27Nhh8OOQhEYIIYRQicbAS2pFRkbSvn17Fi9eTPbs2XXrIyIi+OWXX5g+fTp16tShXLlyLFmyhEOHDnHkyBEAdu7cycWLF/ntt98oXbo0DRs25IcffmDu3LnExMQAsGDBAjw9Pfnxxx8pUqQIvXv3pk2bNsyYMeM9ov1vktAIIYQQmYhWq9VboqOj31q2V69eNG7cGG9vb731J0+eJDY2Vm994cKFcXd35/DhwwAcPnyYEiVK4OzsrCvj4+ODVqvlwoULujJv1u3j46Orw5AkoRFCCCFUYoQGI42Bln/baNzc3LC1tdUtfn5+ye579erVnDp1KtntwcHBmJqaYmdnp7fe2dmZ4OBgXZnXk5nE7Ynb/quMVqvlxYsXqT9h/0HuchJCCCFU8r5dRW+rC+DevXvY2Njo1puZmSUpe+/ePfr164e/vz/m5uYGikBd0kIjhBBCZCI2NjZ6S3IJzcmTJwkNDaVs2bKYmJhgYmLC/v37mT17NiYmJjg7OxMTE0N4eLje+0JCQnBxcQHAxcUlyV1Pia/fVcbGxgYLCwtDHTIgCY0QQgihHpVGBdetW5dz584RGBioW8qXL0/79u11/86WLRu7d+/WvefKlSvcvXsXLy8vALy8vDh37hyhoaG6Mv7+/tjY2FC0aFFdmdfrSCyTWIchSZeTEEIIoRK1JtaztramePHieussLS1xcHDQre/atSsDBw7E3t4eGxsb+vTpg5eXF5UrVwagfv36FC1alA4dOjBlyhSCg4MZNWoUvXr10rUKff311/z0008MHTqUL7/8kj179rB27Vq2bt1qkGN+nSQ0QgghhEhixowZGBkZ0bp1a6Kjo/Hx8WHevHm67cbGxmzZsoWePXvi5eWFpaUlnTp1Yty4cboynp6ebN26lQEDBjBr1ixy587Nzz//jI+Pj8Hj1SiKohi81jSi1WqxtbUl5EmE3gAoIdLSq7h4tUPI0IKevlQ7hAwtKvrVuwuJt4p8pqVGCTciItL2cyTx82t34F2srA2z38hnWuqWdk/zY0kvpIVGCCGEUMnHuMspq5JBwUIIIYTI8KSFRgghhFCLNNEYjLTQCCGEECLDkxYaIYQQQiVq3badGUlCI4QQQqhEo0lYDFVXViZdTkIIIYTI8KSFRgghhFCJjAk2HElohBBCCLVIRmMw0uUkhBBCiAxPWmiEEEIIlchdToYjCY0QQgihErnLyXCky0kIIYQQGZ600AghhBAqkTHBhiMJjRAfyMRYGjo/RHbLbGqHkKGVajhU7RAyNCUuRu0QhIFIQiOEEEKoRZpoDEYSGiGEEEIlcpeT4UhbuRBCCCEyPGmhEUIIIVQit20bjiQ0QgghhEpkCI3hSJeTEEIIITI8aaERQggh1CJNNAYjCY0QQgihErnLyXCky0kIIYQQGZ600AghhBAqkbucDEcSGiGEEEIlMoTGcKTLSQghhBAZnrTQCCGEEGqRJhqDkRYaIYQQQmR40kIjhBBCqERu2zYcSWiEEEIIlchdToYjXU5CCCGEyPCkhUYIIYRQiYwJNhxJaIQQQgi1SEZjMNLlJIQQQogMT1pohBBCCJXIXU6GIwmNEEIIoRYD3uWUxfMZ6XISQgghRMYnLTRCCCGESmRMsOFIC40QQgghMjxJaIQQQgi1aAy8pJCfnx8VKlTA2toaJycnWrRowZUrV/TKvHz5kl69euHg4ICVlRWtW7cmJCREr8zdu3dp3Lgxn3zyCU5OTgwZMoRXr17pldm3bx9ly5bFzMyM/Pnzs3Tp0pQHmgqS0AghhBAq0Rj4v5Tav38/vXr14siRI/j7+xMbG0v9+vWJiorSlRkwYACbN29m3bp17N+/n6CgIFq1aqXbHhcXR+PGjYmJieHQoUMsW7aMpUuXMnr0aF2ZW7du0bhxY2rXrk1gYCD9+/enW7du7NixwzAn8DUaRVEUg9eaRrRaLba2toQ8icDGxkbtcIQQ7+HZi1i1Q8jQ3GsMUDuEDE2JiyH63GIiItL2cyTx8yvwRgjW1obZ77NnWkrnc36vY3n06BFOTk7s37+fGjVqEBERgaOjI6tWraJNmzYAXL58mSJFinD48GEqV67M33//TZMmTQgKCsLZ2RmABQsWMGzYMB49eoSpqSnDhg1j69atnD9/XrcvX19fwsPD2b59u0GOO5G00AghhBAqSXw4paEWSEiWXl+io6PfGUdERAQA9vb2AJw8eZLY2Fi8vb11ZQoXLoy7uzuHDx8G4PDhw5QoUUKXzAD4+Pig1Wq5cOGCrszrdSSWSazDkCShEUIIIVTyMYbQuLm5YWtrq1v8/Pz+M4b4+Hj69+9P1apVKV68OADBwcGYmppiZ2enV9bZ2Zng4GBdmdeTmcTtidv+q4xWq+XFixfvPD+pIQmNgU2dMgmLbBoGD+yvW1e/bi0ssmn0lj7ffK1ekOlcofx5kpwvi2wa+vfppXZo6dKiBfOpUKYkTvY2ONnbULOaFzu2/63bfvPGDT5r0xK3nI442dvQvt1nSQb2ZWaHDx7gi89aUKKgB042pmzb8pfedkVRmDR+DMULuOPuZEPrZg24ef2aXpkZU/1o5F0DD2db8rs5Jruf0ydP0LqpD/ndHCng7sRnLRpz/tyZj3ZcH0vVsvlYP/Mrbu6cwIvTP9G0Vkm97YvGfsGL0z/pLX/99I1emfzuTqyd0YN7eyYRcmAqu38dQI3yBfTK/Di0DQdXDiX86AyOrB6eJI4CHk5sX9SX27sm8vTIDC5uHsP33zTBxEQ+tt7l3r17RERE6JYRI0b8Z/levXpx/vx5Vq9enUYRfhxyZRjQiePH+WXxQkqUKJlk25ddu3Pr3kPdMmHSFBUizBj+OXxc71xt3e4PQKs2n6ocWfqUK3dufpg4iUNHT3LwyAlq1a7Dp62ac/HCBaKiomjSqD4ajYa/d+5hz/6DxMTE0LpFU+Lj49UOPU08j4qiWPGSTPpxVrLb58ycxs8L5zJ15k/8vecfLD/5hM9aNeHly5e6MjExMTRr0ZpOXb9Kto7IyEh8WzUhV243tu/5h8079mJlbUXblk2Ijc1YY4QsLcw4d/UB/f3WvLXMjoMXyOM9Qrd0GrFEb/ufs7/GxNiIhl/Npkr7KZy9+oA/Z3+Ns4O1Xrnlfx1h/c5Tye4j9lUcK7cco+k3cynVchxDpv1Bl1ZV+O7rxh9+kOnJR2iisbGx0VvMzMzeuvvevXuzZcsW9u7dS+7cuXXrXVxciImJITw8XK98SEgILi4uujJvfjlKfP2uMjY2NlhYWLz7/KSCTKxnIJGRkXTp1J55CxYzaeL4JNstPvlE9wMW/83RUf8b8LQpk8ibLx/Va9RUKaL0rXGTpnqvx/4wgcUL53Ps6BGCgh5w5/Ztjhw/rRsk+POvy8jpmJ19e/dQp653clVmKnXrN6Bu/QbJblMUhUXz5jBgyAgaNm4GwE8Ll1Asf27+3vIXLdu0BWDYyO8BWL1yebL1XL96hadPwxg28nty5XYDYPDwUdTyKse9u3fImy+/oQ/ro9l58CI7D178zzIxMa8IefIs2W0OdpYU8HCi59iVnL8WBMB3s//i67Y1KJrflZAnCbcGD5qyHoAc2RtRvECuJPXcfvCE2w+e6F7fffiUGuULULVMvvc6rvRKrWc5KYpCnz592LBhA/v27cPT01Nve7ly5ciWLRu7d++mdevWAFy5coW7d+/i5eUFgJeXFxMmTCA0NBQnJycA/P39sbGxoWjRoroy27Zt06vb399fV4chSQuNgfTv04sGDRu/9QNize8rye2Sg3Kli/PdyBE8f/48jSPMmGJiYli96jc6df4SjcEeeJJ5xcXFsXbNaqKioqhU2Yvo6Gg0Go3eNzRzc3OMjIw4dPAfFSNNH+7cvkVoSDA1atXRrbOxtaVs+YqcOHY0xfXkL1AQe3sHVi5fQkxMDC9evGDV8qUULFQYd488HyFydVUvX4A7u/04s+E7Zn3bFntbS922J+FRXLkVzOdNKvKJuSnGxkZ0a12NkCdaTl+8+977zOuWg3pVinDg5HVDHEKW16tXL3777TdWrVqFtbU1wcHBBAcH68a12Nra0rVrVwYOHMjevXs5efIkXbp0wcvLi8qVKwNQv359ihYtSocOHThz5gw7duxg1KhR9OrVS/c35+uvv+bmzZsMHTqUy5cvM2/ePNauXcuAAYa/O09aaAxg7ZrVBJ4+xT9Hjie7va3v57h7eJAzpyvnzp1l1LfDuHr1CmvW/ZnGkWY8m/7aSHh4OF907Kx2KOna+XPnqFXdi5cvX2JlZcWa9RsoUrQoORwdsbS0ZOSIYYwbPxFFURj17XDi4uIIfvhQ7bBVFxqa0BTu5KQ/aNHRyYnQ0OAU12Nlbc2Gbf50avcp06dMBCBvvvys2bAVE5PM9WfW/9Al/tpzhtsPnpA3dw7G9mnKXz/1pGanH4mPT5gFpPHXP7FmRg8eHZxGfLzCo6eRNO81j/BnqR8EunfpQEoXdsPcLBs/r/+HcfO3GvqQVKXBcA+nTE018+fPB6BWrVp665csWULnzp0BmDFjBkZGRrRu3Zro6Gh8fHyYN2+erqyxsTFbtmyhZ8+eeHl5YWlpSadOnRg3bpyujKenJ1u3bmXAgAHMmjWL3Llz8/PPP+Pj4/O+h/lWqv6mjRkzhrFjx+qtK1SoEJcvX1YpotS7d+8eQwb2Y8vf/pibmydbpmv3Hrp/Fy9Rgpw5c9Kwfl1u3rhB3nyZq/nU0JYt+QWfBg1xdXVVO5R0rWChQhw9EUhERAQb/lxP9y87sXP3fooULcrK1evo27sn836ajZGREZ+1bUeZMmUxMpIGWkN58eIFA3p9RcXKXiz8dQVxcXHMmzOd9p82Z8e+wwYfK6CmdTtO6v594XoQ56494NKWsdQoX4B9x64CMGPEZzwKe4b3lzN5ER1D55ZV+GPWV1T7YirBj7Wp2l+HYb9iZWlOyYK5mNi/BQM61mX6sl0GPaasKCVT0JmbmzN37lzmzp371jIeHh5JupTeVKtWLU6fPp3qGFNL9a8OxYoVY9eu/1+cGe3bzOlTJwkNDcWrYlnduri4OP45EMCCeT8RERWNsbGx3nsqVKwEwI0b1yWh+Q937txhz+5drJaWrHcyNTUlX/6EcRply5Xj5InjzJ0zi5/mL8S7Xn0uXrnB48ePMTExwc7Ojjy5XciTN6/KUasvsWUmNDQEZ5ecuvWPQkMpXqJUiuv5c91q7t69w7bdB3SJ4oJfVlDQ3YntWzfpxuJkRrcfPOHR02fkc3Nk37Gr1KpYkEbVi5Oz5lCeRSUMrO7vt5a6lQvzRdNKTFvin6r674eEA3D5ZjBGRkbMHdWOmSt261qDMjp5OKXhqJ49mJiYZOjBsrXr1OXE6XN663p060KhQoUZNGRYkmQG4ExgIAAur/0BFUmtWLYEJycnGjbKZHc1pIH4+Pgkk2nlyJEDgH179xAaGkqTJs3UCC1d8cjjiZOzCwf276VEydIAPNNqOXXiGJ279vjvN7/mxfPnGBkZ6Y3zMjIyAo0m099NlsvJDgdbS13LyyfmpgBJjjs+XvngcXBGRhqymRhjZKTJPAmNxoBdTlk8o1E9obl27Rqurq6Ym5vj5eWFn58f7u7uyZaNjo7W+yOt1aau6fJjsLa2pti/ExElsrS0xN7BgWLFi3Pzxg3WrF6FT4NGODg4cO7cWYYOHkC16jUoUTLp7d0iQXx8PMuXLaF9h04ZrtUurX03cgQ+DRri5ubOs2fPWLN6FQH797F5W8KzUpYvXUKhwkVwdHTk6JHDDB7Yjz79BlCwUCGVI08bkZGR3Lr5/4Gkd2/f5tzZQLJntye3mzs9vunDjKl+5M2XH3ePPEwaPwbnnK40bNJc95779+7y9GkY9+/dJS4ujnNnAwHwzJsfKysratapy9jvhjNsYF+6ff0N8fHxzJ4+FRMTE6rVqJXGR/xhLC1MyffaXDt5cjlQsmAunmqfExYRxcivGrFxdyDBj7XkdcvBhH4tuHHvMf6HLgFw9Owtnmqf8/MPHZm46G9evIzly1ZVyJPLge3/XNDVm9ctB1YWZjjnsMHCLBslCybc6XTpZjCxr+LwbVie2FdxnL8eRHTMK8oVdeeHPs1Yv/Mkr15l7iRRvB9VPykqVarE0qVLKVSoEA8fPmTs2LFUr16d8+fPY21tnaS8n59fkjE36V02U1P27N7FT7NnEhUVRW43N1q0bM3wb0epHVq6tmf3Lu7dvUunzl+qHUq69yg0lK5dOhL88CG2trYUL1GSzdt2UNe7HgBXr15h9KgRhIWF4ZEnD0OHj6Rv/6zz/J8zp0/SsnE93evR3w4BoO3nHZiz4Bf69B/M86goBvX9Bm1EOBW9qrLmj816Y+ImTxjLmlUrdK/rVqsIwIat/lStXpMCBQuzYs0Gpk0aTyPvGhhpjChRqjSr/9ii15WVEZQt6sHOn/vpXk8ZnHDL7opNR+g7cQ3FC+SifdNK2Flb8PBRBLsOX2bcvC3ExCY8YflJeBTNe89jTK+m/L2wL9lMjLh0M5hPByzi3NUHunrnj26vN9ne0TUJk78VajSauw/DeBUXz8DO9Sjg4YRGo+HuwzDmrwlgzm970uI0pCHpdDKUdPVwyvDwcDw8PJg+fTpdu3ZNsj25Fho3Nzd5OKUQGZg8nPLDyMMpP4zaD6e8dOcR1gba7zOtliIejml+LOlFumrLt7Ozo2DBgly/nvw8A2ZmZv8546EQQgghsqZ0dd9mZGQkN27cIGfOjNVEK4QQQryPj/FwyqxK1YRm8ODB7N+/n9u3b3Po0CFatmyJsbEx7dq1UzMsIYQQIk0k3uVkqCUrU7XL6f79+7Rr144nT57g6OhItWrVOHLkSJJn+QghhBBC/BdVE5qM/qhyIYQQ4kOo9XDKzChdjaERQgghhHgf6eouJyGEECJLkWloDEYSGiGEEEIlks8YjnQ5CSGEECLDkxYaIYQQQiXycErDkYRGCCGEUInc5WQ40uUkhBBCiAxPWmiEEEIItcioYIORhEYIIYRQieQzhiNdTkIIIYTI8KSFRgghhFCJ3OVkONJCI4QQQogMT1pohBBCCNUY7rbtrD6KRhIaIYQQQiXS5WQ40uUkhBBCiAxPEhohhBBCZHjS5SSEEEKoRLqcDEdaaIQQQgiR4UkLjRBCCKESeTil4UhCI4QQQqhEupwMR7qchBBCCJHhSQuNEEIIoRJ5OKXhSAuNEEIIITI8aaERQggh1CJNNAYjCY0QQgihErnLyXCky0kIIYQQGZ600AghhBAqkdu2DUcSGiGEEEIlMoTGcKTLSQghhBAZnrTQCCGEEGqRJhqDkYRGCCGEUInc5WQ40uUkhBBCiAxPWmiEEEIIlchdToaToRMaRVEAeKbVqhyJEOJ9Rb6IVTuEDE2Ji1E7hAwt8fwlfp6kNa0BP78MWVdGlKETmmfPngGQ39NN5UiEEEJkZM+ePcPW1jbN9mdqaoqLiwsFDPz55eLigqmpqUHrzCg0ilppqQHEx8cTFBSEtbU1mnTY1qbVanFzc+PevXvY2NioHU6GJOfww8j5+zBy/j5cej+HiqLw7NkzXF1dMTJK22GlL1++JCbGsC1spqammJubG7TOjCJDt9AYGRmRO3dutcN4Jxsbm3T5i5yRyDn8MHL+Poycvw+Xns9hWrbMvM7c3DzLJh8fg9zlJIQQQogMTxIaIYQQQmR4ktB8RGZmZnz//feYmZmpHUqGJefww8j5+zBy/j6cnEORVjL0oGAhhBBCCJAWGiGEEEJkApLQCCGEECLDk4RGCCGEEBmeJDRCCCGEyPAkoRFCCCFEhicJjUi35AY8oZaHDx9y8eJFtcPI0OLi4gD5PRZpRxKajyDxF1mkXlRUFM+ePUOr1abL53NlBGFhYVy+fJlr164Z/DkxWcGDBw8oUaIEo0aN4sSJE2qHkyEFBgbSokULnj9/Lr/HIs1IQmNgV69eZebMmTx8+FDtUDKcixcv0qpVK2rWrEmRIkVYuXIlIN/wUuP8+fN4e3vz2WefUaJECaZMmSIJdipdu3aNiIgIIiIimDNnDqdOndJtk2vx3c6cOUOVKlUoVqwYn3zyiW69nDvxsUlCY0DXr1/Hy8uLIUOGMGfOHB4/fqx2SBnGxYsXqVGjBsWKFWPw4MH4+vrSpUsXAgMD5RteCl28eJFatWpRt25dVq9ezYQJExg9ejRBQUFqh5ahlCxZkkaNGtG2bVvOnz/P9OnTuXDhAiAfyu9y9uxZqlatSu/evZk0aZJufUxMjPwei49OZgo2kKioKPr27Ut8fDwVKlSgd+/eDB48mKFDh5IjRw61w0vXwsLCaNeuHYULF2bWrFm69bVr16ZEiRLMnj0bRVHkD+J/ePz4Ma1bt6ZMmTLMnDkTSPjwbdSoEaNHj8bCwgIHBwfc3NzUDTSdi4uLIywsjGrVqrFnzx6OHTuGn58fpUuX5sKFC+TMmZP169erHWa6FBwcTJkyZShVqhTbt28nLi6OwYMHc+3aNW7cuMFXX31FgwYNKFy4sNqhikzKRO0AMgsjIyPKlSuHg4MDbdu2JUeOHPj6+gJIUvMOsbGxhIeH06ZNGwDi4+MxMjLC09OTsLAwAElm3kGj0dCgQQPdOQQYP348O3bsIDg4mMePH1OsWDFGjRpFtWrVVIw0fTMyMsLR0ZEKFSpw/vx5WrZsiZmZGZ06dSI6Opru3burHWK65uXlxb179/jrr79YsGABsbGxlC5dmjx58jB79mzOnz/P6NGjcXd3VztUkRkpwmAiIyP1Xq9evVrRaDTK4MGDlcePHyuKoihxcXHKzZs31QgvXbt69aru3zExMYqiKMqoUaOUDh066JV79uxZmsaVkWi1Wt2/f//9d0Wj0Shr1qxRnjx5ouzfv1+pUKGCMmbMGBUjzDg6duyoDB8+XFEURenatauSPXt2pWjRosqXX36pHD16VOXo0q+goCClY8eOioWFhVKvXj3d3z1FUZSVK1cqdnZ2yrZt21SMUGRm0kJjQJaWlkBCs7WRkRFt27ZFURQ+//xzNBoN/fv3Z9q0ady5c4cVK1boDZjL6goUKAAktM5ky5YNSOgyCQ0N1ZXx8/PDzMyMvn37YmIil+6brK2tdf/28vLixIkTlC1bFoAaNWrg5OTEyZMn1QovQ1D+7dqsU6cOt27d4ptvvmHbtm2cPHmSwMBAhgwZgqmpKSVLlsTc3FztcNOdnDlz4ufnR65cufD29sbBwUF3Tj///HO+//579u7dS8OGDdUOVWRC8qnwERgbG6MoCvHx8fj6+qLRaOjQoQObNm3ixo0bHD9+XJKZtzAyMtIbL2NklDBuffTo0YwfP57Tp09LMpMCHh4eeHh4AAlJYkxMDFZWVpQsWVLlyNK3xOvO83/t3X9M1dUfx/HnRwS8yFWzDL2GRKEIDfEHW+OfiGXB2opija2ooNCtyGmUBq5QiCmtZk3aUocF5FTUTFZoNXJAGNUf6XWt6BY3iSz+cLPcsAEXON8/HJ9vN8QvGMb36uvx3+dzzj3n/bm7bC/O59zPjY7miSeeICIigvr6eqKjo4mOjsayLBITExVmLsHlclFUVGS/R5ZlYYzh7NmzzJo1i8WLF09sgXLV0qbgK2jorbUsi7vuugu3201TUxMJCQkTXNn/t6E9NCUlJXR1dTF//nxeeuklWltb7RUHGZsNGzZQU1PDp59+aq+Gych8Ph+7du0iKSmJRYsWaVP6ONi4cSN79+6loaHBDtsi40n/6l5BlmUxMDDAunXraGxsxO12K8yMwtCqTHBwMJWVlUybNo1jx44pzFyGAwcO0NzcTG1tLQ0NDQozoxQcHExubq79WVSYuXy1tbU0NjZy4MABjh49qjAjV4yeQ/MvuO222zh+/LiW+8coLS0NgNbWVpKSkia4msAUHx/PmTNnaGlpYcmSJRNdTkAZCjPyz8THx/Prr7/qMyhXnG45/Qu0XH35zp8/b2+2lsvj8/nsjdYiE6Gvr4+QkJCJLkOucgo0IiIiEvC0pioiIiIBT4FGREREAp4CjYiIiAQ8BRoREREJeAo0IiIiEvAUaERERCTgKdCITLDc3FweeOAB+/jOO+/k2Wef/dfraGpqwrIs/vjjjxH7WJZFXV3dqMcsKSn5x7/d09HRgWVZuN3ufzSOiFzdFGhELiI3NxfLsrAsi5CQEGJiYnj55Zfp7++/4nO///77lJWVjarvaEKIiMi1QL/lJDKC9PR0qqqq6O3t5ciRIzzzzDMEBwezfv36YX3H80moM2fOHJdxRESuJVqhERlBaGgos2fPJioqiqeffprly5fzwQcfAP+9TbRp0yZcLhexsbEA/PLLL2RlZTFjxgxmzpxJRkYGHR0d9pgDAwM899xzzJgxg+uvv54XXniBvz+s+++3nHp7eyksLCQyMpLQ0FBiYmJ4++236ejoIDU1FYDrrrsOy7LIzc0FLvxieXl5OdHR0TgcDhITE3nvvff85jly5AgLFizA4XCQmprqV+doFRYWsmDBAsLCwrjlllsoLi7G5/MN67djxw4iIyMJCwsjKyuLc+fO+bXv3LmTuLg4pkyZwsKFC3nrrbfGXIuIXNsUaERGyeFw0NfXZx8fPXoUj8dDQ0MD9fX1+Hw+0tLScDqdtLS08PnnnxMeHk56err9ui1btlBdXc0777zDsWPHOHv2LIcOHbrkvI8//jh79+6loqKCtrY2duzYQXh4OJGRkRw8eBAAj8dDV1cXW7duBaC8vJx3332X7du38+2331JQUMCjjz5Kc3MzcCF4ZWZmct999+F2u1mxYgVFRUVjfk+cTifV1dV89913bN26lcrKSt544w2/Pu3t7ezfv58PP/yQjz/+mBMnTpCfn2+37969mw0bNrBp0yba2trYvHkzxcXF1NTUjLkeEbmGGREZJicnx2RkZBhjjBkcHDQNDQ0mNDTUrF271m6PiIgwvb299mt27dplYmNjzeDgoH2ut7fXOBwO88knnxhjjJkzZ4559dVX7Xafz2duuukmey5jjElJSTFr1qwxxhjj8XgMYBoaGi5aZ2NjowHM77//bp/r6ekxYWFhprW11a9vXl6eefjhh40xxqxfv97Ex8f7tRcWFg4b6+8Ac+jQoRHbX3vtNbNs2TL7eOPGjSYoKMicPn3aPvfRRx+ZSZMmma6uLmOMMbfeeqvZs2eP3zhlZWUmOTnZGGPMqVOnDGBOnDgx4rwiItpDIzKC+vp6wsPD8fl8DA4O8sgjj1BSUmK3JyQk+O2bOXnyJO3t7TidTr9xenp68Hq9nDt3jq6uLm6//Xa7bfLkySQlJQ277TTE7XYTFBRESkrKqOtub2/nzz//5O677/Y739fXx5IlSwBoa2vzqwMgOTl51HMM2bdvHxUVFXi9Xrq7u+nv72fatGl+febNm8fcuXP95hkcHMTj8eB0OvF6veTl5bFy5Uq7T39/P9OnTx9zPSJy7VKgERlBamoq27ZtIyQkBJfLxeTJ/n8uU6dO9Tvu7u5m2bJl7N69e9hYs2bNuqwaHA7HmF/T3d0NwOHDh/2CBFzYFzRevvjiC7KzsyktLSUtLY3p06dTW1vLli1bxlxrZWXlsIAVFBQ0brWKyNVPgUZkBFOnTiUmJmbU/ZcuXcq+ffu48cYbh61SDJkzZw5fffUVd9xxB3BhJeLrr79m6dKlF+2fkJDA4OAgzc3NLF++fFj70ArRwMCAfS4+Pp7Q0FA6OztHXNmJi4uzNzgP+fLLL//3Rf5Fa2srUVFRvPjii/a5n3/+eVi/zs5OfvvtN1wulz3PpEmTiI2NJSIiApfLxU8//UR2dvaY5hcR+SttChYZJ9nZ2dxwww1kZGTQ0tLCqVOnaGpqYvXq1Zw+fRqANWvW8Morr1BXV8f3339Pfn7+JZ8hc/PNN5OTk8OTTz5JXV2dPeb+/fsBiIqKwrIs6uvrOXPmDN3d3TidTtauXUtBQQE1NTV4vV6OHz/Om2++aW+0feqpp/jxxx9Zt24dHo+HPXv2UF1dPabrnT9/Pp2dndTW1uL1eqmoqLjoBucpU6aQk5PDyZMnaWlpYfXq1WRlZTF79mwASktLKS8vp6Kigh9++IFvvvmGqqoqXn/99THVIyLXNgUakXESFhbGZ599xrx588jMzCQuLo68vDx6enrsFZvnn3+exx57jJycHJKTk3E6nTz44IOXHHfbtm089NBD5Ofns3DhQlauXMn58+cBmDt3LqWlpRQVFREREcGqVasAKCsro7i4mPLycuLi4khPT+fw4cNER0cDF/a1HDx4kLq6OhITE9m+fTubN28e0/Xef//9FBQUsGrVKhYvXkxrayvFxcXD+sXExJCZmcm9997LPffcw6JFi/y+lr1ixQp27txJVVUVCQkJpKSkUF1dbdcqIjIalhlpN6KIiIhIgNAKjYiIiAQ8BRoREREJeAo0IiIiEvAUaERERCTgKdCIiIhIwFOgERERkYCnQCMiIiIBT4FGREREAp4CjYiIiAQ8BRoREREJeAo0IiIiEvD+A85ECIz8DG2zAAAAAElFTkSuQmCC",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn import metrics\n",
    "names = ['1','2','3','4','5']\n",
    "\n",
    "# Compute confusion matrix\n",
    "cnf_matrix = metrics.confusion_matrix(y_test, preds)\n",
    "np.set_printoptions(precision=2)\n",
    "\n",
    "# Plot non-normalized confusion matrix\n",
    "plt.figure()\n",
    "plot_confusion_matrix(cnf_matrix, classes=names,\n",
    "                      title='Confusion matrix, without normalization')\n",
    "\n",
    "# Plot normalized confusion matrix\n",
    "plt.figure()\n",
    "plot_confusion_matrix(cnf_matrix, classes=names, normalize=True,\n",
    "                      title='Normalized confusion matrix')\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "205464\n"
     ]
    }
   ],
   "source": [
    "sentiments = []\n",
    "for star in stars[1]:\n",
    "    if star <= 3:\n",
    "        sentiments.append('n')\n",
    "    if star > 3:\n",
    "        sentiments.append('p')\n",
    "\n",
    "print(len(sentiments))    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "X2_train, X2_test, y2_train, y2_test = train_test_split(vectors, sentiments, test_size=0.20, random_state=42)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Wall time: 12.4 s\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "LinearSVC()"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "%%time\n",
    "\n",
    "classifier2 = LinearSVC()\n",
    "classifier2.fit(X2_train, y2_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Actual Class:    ['p', 'p', 'p', 'p', 'p', 'p', 'n', 'p', 'p', 'n']\n",
      "\n",
      "Predicted Class: ['p', 'p', 'p', 'p', 'p', 'p', 'p', 'p', 'p', 'n']\n"
     ]
    }
   ],
   "source": [
    "preds2 = classifier2.predict(X2_test)\n",
    "print(\"Actual Class:    \",end = \"\")\n",
    "print(y2_test[:10])\n",
    "print(\"\\nPredicted Class: \",end = \"\")\n",
    "print(list(preds2[:10]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.9273598909789988\n"
     ]
    }
   ],
   "source": [
    "print(accuracy_score(y2_test, preds2))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Precision: 0.9266755135168885\n",
      "Recall: 0.9273598909789988\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import precision_score\n",
    "from sklearn.metrics import recall_score\n",
    "print ('Precision: ' + str(precision_score(y2_test, preds2, average='weighted')))\n",
    "print ('Recall: ' + str(recall_score(y2_test, preds2, average='weighted')))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           n       0.91      0.82      0.86     11334\n",
      "           p       0.93      0.97      0.95     29759\n",
      "\n",
      "    accuracy                           0.93     41093\n",
      "   macro avg       0.92      0.89      0.91     41093\n",
      "weighted avg       0.93      0.93      0.93     41093\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(classification_report(y2_test, preds2))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[ 9286  2048]\n",
      " [  937 28822]]\n"
     ]
    }
   ],
   "source": [
    "print(metrics.confusion_matrix(y2_test, preds2))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Confusion matrix, without normalization\n",
      "[[ 9286  2048]\n",
      " [  937 28822]]\n",
      "Normalized confusion matrix\n",
      "[[0.82 0.18]\n",
      " [0.03 0.97]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "class_names = ['negative','positive']\n",
    "\n",
    "cnf_matrix = metrics.confusion_matrix(y2_test, preds2)\n",
    "np.set_printoptions(precision=2)\n",
    "plt.figure()\n",
    "plot_confusion_matrix(cnf_matrix, classes=class_names,\n",
    "                      title='Confusion matrix, without normalization')\n",
    "\n",
    "plt.figure()\n",
    "plot_confusion_matrix(cnf_matrix, classes=class_names, normalize=True,\n",
    "                      title='Normalized confusion matrix')\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3.7.12 ('testinggrounds')",
   "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": "3.7.12"
  },
  "orig_nbformat": 4,
  "vscode": {
   "interpreter": {
    "hash": "be1130cc366ce64c2af566ee12e797ac4446b6a928687bd6f2dd0eaf2725f7ce"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}