Consumer_complaints.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Date received</th>\n",
       "      <th>Product</th>\n",
       "      <th>Sub-product</th>\n",
       "      <th>Issue</th>\n",
       "      <th>Sub-issue</th>\n",
       "      <th>Consumer complaint narrative</th>\n",
       "      <th>Company public response</th>\n",
       "      <th>Company</th>\n",
       "      <th>State</th>\n",
       "      <th>ZIP code</th>\n",
       "      <th>Tags</th>\n",
       "      <th>Consumer consent provided?</th>\n",
       "      <th>Submitted via</th>\n",
       "      <th>Date sent to company</th>\n",
       "      <th>Company response to consumer</th>\n",
       "      <th>Timely response?</th>\n",
       "      <th>Consumer disputed?</th>\n",
       "      <th>Complaint ID</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>03/12/2014</td>\n",
       "      <td>Mortgage</td>\n",
       "      <td>Other mortgage</td>\n",
       "      <td>Loan modification,collection,foreclosure</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>M&amp;T BANK CORPORATION</td>\n",
       "      <td>MI</td>\n",
       "      <td>48382</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Referral</td>\n",
       "      <td>03/17/2014</td>\n",
       "      <td>Closed with explanation</td>\n",
       "      <td>Yes</td>\n",
       "      <td>No</td>\n",
       "      <td>759217.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>10/01/2016</td>\n",
       "      <td>Credit reporting</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Incorrect information on credit report</td>\n",
       "      <td>Account status</td>\n",
       "      <td>I have outdated information on my credit repor...</td>\n",
       "      <td>Company has responded to the consumer and the ...</td>\n",
       "      <td>TRANSUNION INTERMEDIATE HOLDINGS, INC.</td>\n",
       "      <td>AL</td>\n",
       "      <td>352XX</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Consent provided</td>\n",
       "      <td>Web</td>\n",
       "      <td>10/05/2016</td>\n",
       "      <td>Closed with explanation</td>\n",
       "      <td>Yes</td>\n",
       "      <td>No</td>\n",
       "      <td>2141773.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>10/17/2016</td>\n",
       "      <td>Consumer Loan</td>\n",
       "      <td>Vehicle loan</td>\n",
       "      <td>Managing the loan or lease</td>\n",
       "      <td>NaN</td>\n",
       "      <td>I purchased a new car on XXXX XXXX. The car de...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>CITIZENS FINANCIAL GROUP, INC.</td>\n",
       "      <td>PA</td>\n",
       "      <td>177XX</td>\n",
       "      <td>Older American</td>\n",
       "      <td>Consent provided</td>\n",
       "      <td>Web</td>\n",
       "      <td>10/20/2016</td>\n",
       "      <td>Closed with explanation</td>\n",
       "      <td>Yes</td>\n",
       "      <td>No</td>\n",
       "      <td>2163100.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>06/08/2014</td>\n",
       "      <td>Credit card</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Bankruptcy</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>AMERICAN EXPRESS COMPANY</td>\n",
       "      <td>ID</td>\n",
       "      <td>83854</td>\n",
       "      <td>Older American</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Web</td>\n",
       "      <td>06/10/2014</td>\n",
       "      <td>Closed with explanation</td>\n",
       "      <td>Yes</td>\n",
       "      <td>Yes</td>\n",
       "      <td>885638.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>09/13/2014</td>\n",
       "      <td>Debt collection</td>\n",
       "      <td>Credit card</td>\n",
       "      <td>Communication tactics</td>\n",
       "      <td>Frequent or repeated calls</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>CITIBANK, N.A.</td>\n",
       "      <td>VA</td>\n",
       "      <td>23233</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Web</td>\n",
       "      <td>09/13/2014</td>\n",
       "      <td>Closed with explanation</td>\n",
       "      <td>Yes</td>\n",
       "      <td>Yes</td>\n",
       "      <td>1027760.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  Date received           Product     Sub-product  \\\n",
       "0    03/12/2014          Mortgage  Other mortgage   \n",
       "1    10/01/2016  Credit reporting             NaN   \n",
       "2    10/17/2016     Consumer Loan    Vehicle loan   \n",
       "3    06/08/2014       Credit card             NaN   \n",
       "4    09/13/2014   Debt collection     Credit card   \n",
       "\n",
       "                                      Issue                   Sub-issue  \\\n",
       "0  Loan modification,collection,foreclosure                         NaN   \n",
       "1    Incorrect information on credit report              Account status   \n",
       "2                Managing the loan or lease                         NaN   \n",
       "3                                Bankruptcy                         NaN   \n",
       "4                     Communication tactics  Frequent or repeated calls   \n",
       "\n",
       "                        Consumer complaint narrative  \\\n",
       "0                                                NaN   \n",
       "1  I have outdated information on my credit repor...   \n",
       "2  I purchased a new car on XXXX XXXX. The car de...   \n",
       "3                                                NaN   \n",
       "4                                                NaN   \n",
       "\n",
       "                             Company public response  \\\n",
       "0                                                NaN   \n",
       "1  Company has responded to the consumer and the ...   \n",
       "2                                                NaN   \n",
       "3                                                NaN   \n",
       "4                                                NaN   \n",
       "\n",
       "                                  Company State ZIP code            Tags  \\\n",
       "0                    M&T BANK CORPORATION    MI    48382             NaN   \n",
       "1  TRANSUNION INTERMEDIATE HOLDINGS, INC.    AL    352XX             NaN   \n",
       "2          CITIZENS FINANCIAL GROUP, INC.    PA    177XX  Older American   \n",
       "3                AMERICAN EXPRESS COMPANY    ID    83854  Older American   \n",
       "4                          CITIBANK, N.A.    VA    23233             NaN   \n",
       "\n",
       "  Consumer consent provided? Submitted via Date sent to company  \\\n",
       "0                        NaN      Referral           03/17/2014   \n",
       "1           Consent provided           Web           10/05/2016   \n",
       "2           Consent provided           Web           10/20/2016   \n",
       "3                        NaN           Web           06/10/2014   \n",
       "4                        NaN           Web           09/13/2014   \n",
       "\n",
       "  Company response to consumer Timely response? Consumer disputed?  \\\n",
       "0      Closed with explanation              Yes                 No   \n",
       "1      Closed with explanation              Yes                 No   \n",
       "2      Closed with explanation              Yes                 No   \n",
       "3      Closed with explanation              Yes                Yes   \n",
       "4      Closed with explanation              Yes                Yes   \n",
       "\n",
       "   Complaint ID  \n",
       "0      759217.0  \n",
       "1     2141773.0  \n",
       "2     2163100.0  \n",
       "3      885638.0  \n",
       "4     1027760.0  "
      ]
     },
     "execution_count": 1,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "df = pd.read_csv('Consumer_Complaints.csv')\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "df = df[pd.notnull(df['Consumer complaint narrative'])]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 4569 entries, 1 to 21662\n",
      "Data columns (total 18 columns):\n",
      "Date received                   4569 non-null object\n",
      "Product                         4569 non-null object\n",
      "Sub-product                     3106 non-null object\n",
      "Issue                           4569 non-null object\n",
      "Sub-issue                       2294 non-null object\n",
      "Consumer complaint narrative    4569 non-null object\n",
      "Company public response         2220 non-null object\n",
      "Company                         4569 non-null object\n",
      "State                           4556 non-null object\n",
      "ZIP code                        4556 non-null object\n",
      "Tags                            770 non-null object\n",
      "Consumer consent provided?      4569 non-null object\n",
      "Submitted via                   4569 non-null object\n",
      "Date sent to company            4569 non-null object\n",
      "Company response to consumer    4569 non-null object\n",
      "Timely response?                4569 non-null object\n",
      "Consumer disputed?              4568 non-null object\n",
      "Complaint ID                    4569 non-null float64\n",
      "dtypes: float64(1), object(17)\n",
      "memory usage: 678.2+ KB\n"
     ]
    }
   ],
   "source": [
    "df.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "col = ['Product', 'Consumer complaint narrative']\n",
    "df = df[col]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['Product', 'Consumer complaint narrative'], dtype='object')"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "df.columns = ['Product', 'Consumer_complaint_narrative']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "df['category_id'] = df['Product'].factorize()[0]\n",
    "from io import StringIO\n",
    "category_id_df = df[['Product', 'category_id']].drop_duplicates().sort_values('category_id')\n",
    "category_to_id = dict(category_id_df.values)\n",
    "id_to_category = dict(category_id_df[['category_id', 'Product']].values)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Product</th>\n",
       "      <th>Consumer_complaint_narrative</th>\n",
       "      <th>category_id</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Credit reporting</td>\n",
       "      <td>I have outdated information on my credit repor...</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Consumer Loan</td>\n",
       "      <td>I purchased a new car on XXXX XXXX. The car de...</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>Credit reporting</td>\n",
       "      <td>An account on my credit report has a mistaken ...</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>Debt collection</td>\n",
       "      <td>This company refuses to provide me verificatio...</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>Debt collection</td>\n",
       "      <td>This complaint is in regards to Square Two Fin...</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             Product                       Consumer_complaint_narrative  \\\n",
       "1   Credit reporting  I have outdated information on my credit repor...   \n",
       "2      Consumer Loan  I purchased a new car on XXXX XXXX. The car de...   \n",
       "7   Credit reporting  An account on my credit report has a mistaken ...   \n",
       "12   Debt collection  This company refuses to provide me verificatio...   \n",
       "16   Debt collection  This complaint is in regards to Square Two Fin...   \n",
       "\n",
       "    category_id  \n",
       "1             0  \n",
       "2             1  \n",
       "7             0  \n",
       "12            2  \n",
       "16            2  "
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAfAAAAHlCAYAAAAKt67OAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nSUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f72dd798160>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "fig = plt.figure(figsize=(8,6))\n",
    "df.groupby('Product').Consumer_complaint_narrative.count().plot.bar(ylim=0)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(4569, 12633)"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.feature_extraction.text import TfidfVectorizer\n",
    "\n",
    "tfidf = TfidfVectorizer(sublinear_tf=True, min_df=5, norm='l2', encoding='latin-1', ngram_range=(1, 2), stop_words='english')\n",
    "\n",
    "features = tfidf.fit_transform(df.Consumer_complaint_narrative).toarray()\n",
    "labels = df.category_id\n",
    "features.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "# 'Bank account or service':\n",
      "  . Most correlated unigrams:\n",
      "       . bank\n",
      "       . overdraft\n",
      "  . Most correlated bigrams:\n",
      "       . overdraft fees\n",
      "       . checking account\n",
      "# 'Consumer Loan':\n",
      "  . Most correlated unigrams:\n",
      "       . car\n",
      "       . vehicle\n",
      "  . Most correlated bigrams:\n",
      "       . vehicle xxxx\n",
      "       . toyota financial\n",
      "# 'Credit card':\n",
      "  . Most correlated unigrams:\n",
      "       . citi\n",
      "       . card\n",
      "  . Most correlated bigrams:\n",
      "       . annual fee\n",
      "       . credit card\n",
      "# 'Credit reporting':\n",
      "  . Most correlated unigrams:\n",
      "       . experian\n",
      "       . equifax\n",
      "  . Most correlated bigrams:\n",
      "       . trans union\n",
      "       . credit report\n",
      "# 'Debt collection':\n",
      "  . Most correlated unigrams:\n",
      "       . collection\n",
      "       . debt\n",
      "  . Most correlated bigrams:\n",
      "       . collect debt\n",
      "       . collection agency\n",
      "# 'Money transfers':\n",
      "  . Most correlated unigrams:\n",
      "       . wu\n",
      "       . paypal\n",
      "  . Most correlated bigrams:\n",
      "       . western union\n",
      "       . money transfer\n",
      "# 'Mortgage':\n",
      "  . Most correlated unigrams:\n",
      "       . modification\n",
      "       . mortgage\n",
      "  . Most correlated bigrams:\n",
      "       . mortgage company\n",
      "       . loan modification\n",
      "# 'Other financial service':\n",
      "  . Most correlated unigrams:\n",
      "       . dental\n",
      "       . passport\n",
      "  . Most correlated bigrams:\n",
      "       . help pay\n",
      "       . stated pay\n",
      "# 'Payday loan':\n",
      "  . Most correlated unigrams:\n",
      "       . borrowed\n",
      "       . payday\n",
      "  . Most correlated bigrams:\n",
      "       . big picture\n",
      "       . payday loan\n",
      "# 'Prepaid card':\n",
      "  . Most correlated unigrams:\n",
      "       . serve\n",
      "       . prepaid\n",
      "  . Most correlated bigrams:\n",
      "       . access money\n",
      "       . prepaid card\n",
      "# 'Student loan':\n",
      "  . Most correlated unigrams:\n",
      "       . student\n",
      "       . navient\n",
      "  . Most correlated bigrams:\n",
      "       . student loans\n",
      "       . student loan\n",
      "# 'Virtual currency':\n",
      "  . Most correlated unigrams:\n",
      "       . handles\n",
      "       . https\n",
      "  . Most correlated bigrams:\n",
      "       . xxxx provider\n",
      "       . money want\n"
     ]
    }
   ],
   "source": [
    "from sklearn.feature_selection import chi2\n",
    "import numpy as np\n",
    "\n",
    "N = 2\n",
    "for Product, category_id in sorted(category_to_id.items()):\n",
    "  features_chi2 = chi2(features, labels == category_id)\n",
    "  indices = np.argsort(features_chi2[0])\n",
    "  feature_names = np.array(tfidf.get_feature_names())[indices]\n",
    "  unigrams = [v for v in feature_names if len(v.split(' ')) == 1]\n",
    "  bigrams = [v for v in feature_names if len(v.split(' ')) == 2]\n",
    "  print(\"# '{}':\".format(Product))\n",
    "  print(\"  . Most correlated unigrams:\\n       . {}\".format('\\n       . '.join(unigrams[-N:])))\n",
    "  print(\"  . Most correlated bigrams:\\n       . {}\".format('\\n       . '.join(bigrams[-N:])))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.feature_extraction.text import CountVectorizer\n",
    "from sklearn.feature_extraction.text import TfidfTransformer\n",
    "from sklearn.naive_bayes import MultinomialNB\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(df['Consumer_complaint_narrative'], df['Product'], random_state = 0)\n",
    "count_vect = CountVectorizer()\n",
    "X_train_counts = count_vect.fit_transform(X_train)\n",
    "tfidf_transformer = TfidfTransformer()\n",
    "X_train_tfidf = tfidf_transformer.fit_transform(X_train_counts)\n",
    "\n",
    "clf = MultinomialNB().fit(X_train_tfidf, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['Debt collection']\n"
     ]
    }
   ],
   "source": [
    "print(clf.predict(count_vect.transform([\"This company refuses to provide me verification and validation of debt per my right under the FDCPA. I do not believe this debt is mine.\"])))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['Credit reporting']\n"
     ]
    }
   ],
   "source": [
    "print(clf.predict(count_vect.transform([\"I am disputing the inaccurate information the Chex-Systems has on my credit report. I initially submitted a police report on XXXX/XXXX/16 and Chex Systems only deleted the items that I mentioned in the letter and not all the items that were actually listed on the police report. In other words they wanted me to say word for word to them what items were fraudulent. The total disregard of the police report and what accounts that it states that are fraudulent. If they just had paid a little closer attention to the police report I would not been in this position now and they would n't have to research once again. I would like the reported information to be removed : XXXX XXXX XXXX\"])))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Product</th>\n",
       "      <th>Consumer_complaint_narrative</th>\n",
       "      <th>category_id</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>Debt collection</td>\n",
       "      <td>This company refuses to provide me verificatio...</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            Product                       Consumer_complaint_narrative  \\\n",
       "12  Debt collection  This company refuses to provide me verificatio...   \n",
       "\n",
       "    category_id  \n",
       "12            2  "
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[df['Consumer_complaint_narrative'] == \"This company refuses to provide me verification and validation of debt per my right under the FDCPA. I do not believe this debt is mine.\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Product</th>\n",
       "      <th>Consumer_complaint_narrative</th>\n",
       "      <th>category_id</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>61</th>\n",
       "      <td>Credit reporting</td>\n",
       "      <td>I am disputing the inaccurate information the ...</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             Product                       Consumer_complaint_narrative  \\\n",
       "61  Credit reporting  I am disputing the inaccurate information the ...   \n",
       "\n",
       "    category_id  \n",
       "61            0  "
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[df['Consumer_complaint_narrative'] == \"I am disputing the inaccurate information the Chex-Systems has on my credit report. I initially submitted a police report on XXXX/XXXX/16 and Chex Systems only deleted the items that I mentioned in the letter and not all the items that were actually listed on the police report. In other words they wanted me to say word for word to them what items were fraudulent. The total disregard of the police report and what accounts that it states that are fraudulent. If they just had paid a little closer attention to the police report I would not been in this position now and they would n't have to research once again. I would like the reported information to be removed : XXXX XXXX XXXX\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ec2-user/anaconda3/envs/python3/lib/python3.6/site-packages/sklearn/model_selection/_split.py:605: Warning: The least populated class in y has only 1 members, which is too few. The minimum number of members in any class cannot be less than n_splits=5.\n",
      "  % (min_groups, self.n_splits)), Warning)\n",
      "/home/ec2-user/anaconda3/envs/python3/lib/python3.6/site-packages/sklearn/model_selection/_split.py:605: Warning: The least populated class in y has only 1 members, which is too few. The minimum number of members in any class cannot be less than n_splits=5.\n",
      "  % (min_groups, self.n_splits)), Warning)\n",
      "/home/ec2-user/anaconda3/envs/python3/lib/python3.6/site-packages/sklearn/model_selection/_split.py:605: Warning: The least populated class in y has only 1 members, which is too few. The minimum number of members in any class cannot be less than n_splits=5.\n",
      "  % (min_groups, self.n_splits)), Warning)\n",
      "/home/ec2-user/anaconda3/envs/python3/lib/python3.6/site-packages/sklearn/model_selection/_split.py:605: Warning: The least populated class in y has only 1 members, which is too few. The minimum number of members in any class cannot be less than n_splits=5.\n",
      "  % (min_groups, self.n_splits)), Warning)\n"
     ]
    }
   ],
   "source": [
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.naive_bayes import MultinomialNB\n",
    "from sklearn.svm import LinearSVC\n",
    "\n",
    "from sklearn.model_selection import cross_val_score\n",
    "\n",
    "\n",
    "models = [\n",
    "    RandomForestClassifier(n_estimators=200, max_depth=3, random_state=0),\n",
    "    LinearSVC(),\n",
    "    MultinomialNB(),\n",
    "    LogisticRegression(random_state=0),\n",
    "]\n",
    "CV = 5\n",
    "cv_df = pd.DataFrame(index=range(CV * len(models)))\n",
    "entries = []\n",
    "for model in models:\n",
    "  model_name = model.__class__.__name__\n",
    "  accuracies = cross_val_score(model, features, labels, scoring='accuracy', cv=CV)\n",
    "  for fold_idx, accuracy in enumerate(accuracies):\n",
    "    entries.append((model_name, fold_idx, accuracy))\n",
    "cv_df = pd.DataFrame(entries, columns=['model_name', 'fold_idx', 'accuracy'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYcAAAELCAYAAAAybErdAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nREQkjZKDiIikUXIQEZE0Sg4iIpLm/wPqteqRdmULsAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f72cbc4ac50>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import seaborn as sns\n",
    "\n",
    "sns.boxplot(x='model_name', y='accuracy', data=cv_df)\n",
    "sns.stripplot(x='model_name', y='accuracy', data=cv_df, \n",
    "              size=8, jitter=True, edgecolor=\"gray\", linewidth=2)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "model_name\n",
       "LinearSVC                 0.822890\n",
       "LogisticRegression        0.792927\n",
       "MultinomialNB             0.688519\n",
       "RandomForestClassifier    0.443826\n",
       "Name: accuracy, dtype: float64"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cv_df.groupby('model_name').accuracy.mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "model = LinearSVC()\n",
    "\n",
    "X_train, X_test, y_train, y_test, indices_train, indices_test = train_test_split(features, labels, df.index, test_size=0.33, random_state=0)\n",
    "model.fit(X_train, y_train)\n",
    "y_pred = model.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAkEAAAHlCAYAAADyYSjIAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMS4xLCBo\ndHRwOi8vbWF0cGxvdGxpYi5vcmcvAOZPmwAAIABJREFUeJzs3Xd4VVX28PHvShFCF0EggFJVkBKU\nXjSggiAIow7YB4cZxoJtBN4ZK/qzAyoOCIIOCKMCohQRKQKhCiRK6CBVDISi9J6y3j/uCV5ieu49\nJyTr8zz34d59ytrn3EOysvc+Z4uqYowxxhhT1IR4XQFjjDHGGC9YEmSMMcaYIsmSIGOMMcYUSZYE\nGWOMMaZIsiTIGGOMMUWSJUHGGGOMKZIsCTLGGGNMkWRJkDHGGGOKJEuCjDHGGFMkhXldAVO4nHzu\nz549grzc4OVehaZa6QqexU48ediz2BFhl3gW+/i5057F9pJ4GNvmF/BO8rk9+dk8KJdN0q878nVJ\nhFeo5eXlDFhLkDHGGGOKKGsJMsYYY0zupaZ4XYN8syTIGGOMMbmnqV7XIN+sO8wYY4wxRZK1BBlj\njDEm91Iv/pYgS4KMMcYYk2taCLrDLAkyxhhjTO4VgpYgGxNkjDHGmCLJWoKMMcYYk3uFoDvMWoKM\na6TsZRTv8xIRT75LxBPvENaqCwDhN/ci4vEhFO83mOK9n0dKX3p+m5Ca9SnebzART7xD8b+9HJR6\nVasWyby5X7B2bQzx8Qt4vF+foMQBqBJZic+mfcS876cyZ9lX9O57LwD/+ehtvomZxDcxk1iyehbf\nxEwKWh38lS1bhs8+G8WaNQuIj59PixbXBS3Wfz54g592rmT5qll/WNbviT4cPrGN8pddmsGWgdep\nYzQb1i9m88alDBzwmCsxAcaMHsrehDXEr57vWsw0bl7nGfHqnBfl2EGXmpK/VwEgqkX3QegiUhl4\nD2gGnAV2AU+p6k953F9voKmq9hORh4FTqjreKZ+rqnsDUvG81a0ccK+qfuB8jgTeV9W7Ahknq2kz\npHQ5pPSlpO7dCZcUJ+Kxtzjzv8Hosd/grG8KhLBWnQm5vBrnpo+B4iWI+MdrnBn3Gnr0VyhZBk4e\nyzR2XqfNqFz5cqpUvpzV8espVaokK1fO5q67/sqmTVtzvI+cTptRsVIFLq9UgQ1rN1OyVAm+nj+R\nvg8+xbYtO86v89wrz3Ds2An+M+TDHO0zP9NmfPTROyxbtoqxYycSHh5OiRIRHD2a+TlOLzfTZrRu\n04wTJ04xasxgWjfvcr68atUqDBvxOlddVYvodj049FvOjiev02aEhISwacMSbu1yDwkJiaz4fhb3\nP/Borr7vvGrXtgUnTpxk7NhhRDW5KU/7yOs8A4G4zvP628LLc15YYhfEaTPO7YrLVwJxSY2mNm2G\nV0REgKlAjKrWVtX6wLNApXTrheZl/6o6SlXHOx97A5E5qFNQuiedYygHPOpXv72BToCyo8eP+BIg\ngHNnSD24BylT/nwCBCDhxc7/pA1r3JbkDSt9CRBkmQDlx759B1gdvx6AEydOsnnzViIjKwcl1sH9\nv7Jh7WYATp44xbatO6hc5fIL1unSoyNff/VtUOL7K126FG3bNmfs2IkAJCUl5SoByq3ly2I5fPjI\nH8pfe+s5Bj3/Fm79Qda8WRO2b9/Fzp27SUpKYvLk6dzerZMrsZcsXcmhDM6BG9y8ztPz8pwX1diu\nSE3N36sAKLJJENAeSFLVUWkFqhqvqktEJFpEForIZ8A6ABG5X0RWiUi8iHyYlhyJyEMi8pOILALa\npO1LRAaJSH8RuQtoCnzqbBvhXwkRiRGR153tnxSRiiLypYjEOq82fvubICILRGSriPzdKRcRGSwi\n60VknYj0csrTH8ObQG2nDoNFpIaIrHfW7S0iX4nIbGffb/vVr49zfDEiMkZEhgfi5Eu5ioRUqUlq\ngu8vovBb7iFiwEjCotpx7jtfV1DIZZFIREmK9xlE8UffIizqhkCEztKVV1YjqnEDVq1aHfRYVatH\nUr/hNcT/sO58WfNW1/Hrwd/YtWN30OPXrHkFBw8eYsyYoaxYMYuRI9+iRImI7DcMoM5dbiJx7z7W\nr9/sWszIqpX5JeH3RtmEPYmuJQMFhZvXOXh7zotqbJMzRTkJagD8kMXy5sBzqlpfROoBvYA2qhoF\npAD3iUgV4GV8yc8tQP30O1HVKUAccJ+qRqlqRm345VT1RlUdCgwD3lXVZsCdwEd+6zUCbgNaAS86\nXVp3AFFAY+BmYLBTrwuOAfgXsN2pw4AM6hDlHGNDoJeIVHf2/wLQ0jm+a7I4Xzl3SXGK3dufc9+M\nPd8KlDTvc04PfoTk+CWEt7rVt15oKCGRtTgz/g3OjHuV8PZ3IZdVyWLH+VOyZAkmTxrDM/1f4vjx\nE0GLA1CiZAQjxw3l/54bzInjJ8+Xd7uzM19/OTuosdOEhYXRpEkDRo+eQMuWXTh58jQDBjya/YYB\nEhFRnH8OeIQ3Xn3PtZgAvkbgCxWlYQFuXudpvDznRTW2G1RT8/UqCIpyEpSdVarq9N1wE3A9ECsi\n8c7nWkALfN1pB1X1HJDX0az+290MDHfizADKiEhpZ9l0VT2tqr8CC/ElOW2Bz1U1RVX3A4vwjXFK\nfwzZma+qR1X1DLARuNLZ/yJVPaSqScAXGW0oIn1FJE5E4v67ekdGq/wuJJRi9z5D8polpGxc9YfF\nyWuXEnZtCwD06G+kbI2HpLNw6jgpuzYRUuXKHB5O7oSFhTF50hg+/3wq06YFtysqLCyMkePeYfqU\nWcyZ+fsA2dDQUG697SZmTnMnCdqzJ5E9exKJjY0HYOrUWURFNXAlNkDNWldwZY3qLPl+Jms2xBBZ\ntTKLlk7n8stzNr4qr/YkJFK92u+909WqViExcX9QYxYUbl7n/rw850U1tiusO+yitgFfYpOZk37v\nBfjEaUWJUtWrVXWQsywQab1/rBCglV+sqqp6PJNYStYD3k5msSy9s37vU/A9PiFHg9ZUdbSqNlXV\npn9tUivLdS+54xH0wB6Sl808XyaX/d48HHpNU1IP+pqPkzfFElqjHoSEQPglhFavQ+qBfA0OzNSY\n0UPZvHkb7w0bHZT9+3vr/UFs+2kHH4+ccEF5mxtbsH3rTvbtPRD0OgDs33+QhIRE6tb1fWft27dx\nZbBomo0bfuKqmi1ofG00ja+NZu+efdzYtjsHDvwa1LixcfHUqVOTGjWqEx4eTs+e3fl65tygxiwo\n3LzO/Xl5zotqbFdoav5eBUBRToIWAMXSxtYAiEgzEbkxg3XnA3eJyOXOeuVF5EpgJRAtIpeJSDjw\n50xiHQdKZ7IsvblAP786Rfkt6y4ixUXkMiAaiAUW4+u+ChWRisANwB+bWHJXhzSrgBtF5FJn0Pad\nudz+AiFXXkN4kxsJrd3Adzt8v8GEXtWESzreR8QTQ4l4fAihdRv7uskAPbiHlJ/iiXh8KBGPvEFS\n3Hz0wC/5qUKG2rRuxv3330X79q2Ji51LXOxcbr21Q8DjADRt0YQ7enWjdbvm52+Jj765LQDd7riV\nGV+50wqU5umnX2TcuPeJjZ1Do0b1efvtEUGL9dHYd5m74Avq1K3J+i1Luf/BzP67BFdKSgpPPvU8\ns775jPVrY5gy5Ws2bszTDaG59r8JI1i6eAZXX1WbXTvieKj33a7EBXev8/S8POdFNbYr7Bb5i5sz\n5uU9fC1CZ3BukQeqAv1Vtavfur2Af+NLHJOAx1R1hYg85JQnAvFAqHOL/CDghKoOEZE7gdeB0/ha\neU777TfGiRXnfK4AjADq4WuNWayqDzv7iwRqA1cAb6vqGOcut7eBzvhahl5V1UkiEp3BMXyGb1zR\nt06MmarawP/Wfme9mcAQVY0Rkb5Af2AvsAk4pKrPZXZOs7pFPtjyeot8IOT0FvlgyM8t8vmVm1vk\nAy2vt8hf7Ly8p7jo/rbwXkG8Rf7s5kX5uiSKXXOj57fIF+kk6GLin1S5HLeUqp5wWoKmAv9V1amZ\nrW9JkPssCSpaLAkqmgpkErRpYf6SoHrtPU+CbNoMk51BInIzUBxfV900j+tjjDGmICggg5vzw5Kg\ni4TfQGy34/b3Iq4xxpgCroAMbs6Pojww2hhjjDFFmLUEGWOMMSb3CkF3mLUEGWOMMSbXVFPy9cqO\n80iYVSKyRkQ2iMjLTnlNEVnpTPM0SUQuccqLOZ+3OctrZBfDkiBjjDHG5F7wH5Z4Fuigqo3xTe10\nq4i0BN7CN71UXeAw0MdZvw9wWFXrAO8662XJkiBjjDHG5F6Qp81Qn7QJ7sKdlwIdgClO+SdAD+d9\nd+czzvKbJKMJ3PxYEmSMMcYY1/nPO+m8+mawTqgzl+YBYB6wHTiiqsnOKgn4HnCM8+8vAM7yo8Bl\nWdXBBkabgKr0bpxnsdtXauhZ7OW/bfEsdmiId3/LnEo+m/1KhZCX5zy1EAxGNYVEPm+RV9XRQJYT\n2alv8FCUiJTD98Deehmt5vybUatPlg90tCTIGGOMMbnn4vxfqnrEmWaqJVBORMKc1p5q+KZ1Al+r\nUHUgwZnloCxwKKv9WneYMcYYY3IvyAOjRaSi0wKEiEQAN+Obw3IhcJez2l+A6c77Gc5nnOULNJu5\nwawlyBhjjDG5F/yu2SrAJyISiq/RZrKqzhSRjcBEEXkVWA187Kz/MTBBRLbhawG6O7sAlgQZY4wx\npsBR1bVAkwzKdwDNMyg/A/w5NzEsCTLGGGNM7hWCucMsCTLGGGNM7hWCOxUtCTLGGGNM7hWCJMju\nDjPGGGNMkWQtQcYYY4zJtZxMglrQWUuQ8cwHo95i565YVsXOPl926aVlmfH1BOLXLmDG1xMoV65M\nUGKXLFOSF0Y9x8cLx/DRgtHUu64epcuV4s1PX2fs4o9589PXKVW2VFBie3nc6fXr14fYuLnExs5h\n3Lj3KVasmCtxAbZsWc4PcfNYtXI2y5d941pcgE4do9mwfjGbNy5l4IDHXI0NEBISwsoV3zL1q7Gu\nxaxWLZJ5c79g7doY4uMX8Hi/PtlvFEBenfMxo4eyN2EN8avnuxbTn9fXWlAFee4wN1gSBIhIZRGZ\nKCLbRWSjiMwSkau8rld+iEgNEVnvdT2y8umEL+nRo/cFZf985hFiYpYR1agDMTHL+OczjwQl9qOD\nHiY25gf6tP87D3d6lN3bdtPr0V6sXhbPQzf0YfWyeHo92jMosb08bn9VIivxyKO9ade2G82adSIk\nNIQ//7lb0OP669ipJ81b3ErrNre5FjMkJIT3h71G127307Bxe3r16kG9enVdiw/weL8+bN6yzdWY\nycnJDBz4Mo0aRdO2bTcefqS3a8ft5TkfP34yt3W9z5VY6RWEay2ogj+LfNAV+STImWF2KhCjqrVV\ntT7wLFDJ25rljvOI8IvKsmWrOHzoyAVlt3W9hU8//RKATz/9kq7dOgY8bolSJWjYoiGzJ/paYpKT\nkjl57CStOrZi3pTvAJg35Ttad2od8Njg3XFnJCwslIiI4oSGhlKiRASJiftdieul5s2asH37Lnbu\n3E1SUhKTJ0/n9m6dXItftWplOnfuwNixn7sWE2DfvgOsjvf9XXTixEk2b95KZGRlV2J7ec6XLF3J\nocNHsl8xCLy+1oLOWoIKhfZAkqqOSitQ1XhVXSI+g0VkvYisE5FeACISLSIxIjJFRDaLyKdOMoWI\nvOm0Jq0VkSFO2TgRSXvENyJywm8/i0Rksoj85Gx7n4iscuLVdtarKCJfikis82rjlA8SkdEiMhcY\nn5ODFZEoEVnh1G+qiFzqlP/d2fcaJ1YJv7q/LyLLRWSH/3EEw+WXV2D/voMA7N93kIoVs5wAOE8q\nX1GZI4eO0v+dZ/jg2+E8/fZTFI8oxqUVynHogG+amUMHDlHusrIBj50ZN447vcS9+xn23hg2b1nO\n9h2rOHb0OPPnLwl63PNU+Wbmp3y//Bv69LnXtbCRVSvzS8Le858T9iS6lgwADBk8iH8/+7qnE6Fe\neWU1oho3YNWq1a7E8/qce6WoHvfFxJIgaAD8kMmyO4AooDG+OUsGi0gVZ1kT4CmgPlALaCMi5YE/\nAdeqaiPg1RzEbww8CTQEHgCuUtXmwEfA4846w4B3VbUZcKezLM31QHdVzelvkfHA/3Pqtw54ySn/\nSlWbqWpjfHOz+A8YqAK0BboCb+YwToEVGhZK3QZ1mDl+Jo927seZU2fo9Vgvr6vlunLlytC16y1c\nW78ddWq3oETJEtx9dw/X4ke3v4OWrbpwe/cHefgff6Ft2xauxHX+XrlANtMLBUyXzjdx8OBvrF69\nzpV4GSlZsgSTJ43hmf4vcfz4CVdiennOvVToj9u6wwq9tsDnqpqiqvuBRUAzZ9kqVU1Q1VQgHqgB\nHAPOAB+JyB3AqRzEiFXVRFU9C2wH5jrl65x9gi8BGy4i8fgmiCsjIqWdZTNU9XRODkZEygLlVHWR\nU/QJcIPzvoGILBGRdcB9wLV+m05T1VRV3UgG3YQi0ldE4kQkLin5eE6qkqkDB36lUuWKAFSqXJGD\nB3/L1/4y8mvirxxM/JXN8VsAWDJrCXUa1OHwr0cof3l5AMpfXp4jvx0NeOzMuHHc6bVv35ZdP//C\nr78eIjk5mRnTZ9Oi5fVBj5smrevt4MHfmD5jNs2aRrkSd09CItWrRZ7/XK1qFde6AVu1bsptt93C\nli3LmTB+BNHRbRg7dpgrsQHCwsKYPGkMn38+lWnTvnUtrpfn3EuF/ritO6xQ2ICvNSUjf0zjf3fW\n730KEKaqyfjmM/kS6AGk3f6TjHOunW6zSzLZT6rf51R+f4RBCNBKVaOcV1VVTcs2TmZRx9wYB/RT\n1YbAy0DxTOr4h3OiqqNVtamqNg0PK51+ca7M+uY77rvvTgDuu+9Ovpk5L1/7y8jhg4c5mHiQarWq\nAdCkTRN2b93NinkruOWumwG45a6b+X7u9wGPnRk3jju9XxL20qxZEyIifF91dHQbtmx2Z7BuiRIR\nlCpV8vz7m2+6gQ0btrgSOzYunjp1alKjRnXCw8Pp2bM7X8+cm/2GAfDCC29Ru05zrr66NQ88+Bgx\nMct46KEnXYkNvjulNm/exnvDRrsWE7w9514q9MdtLUGFwgKgmIj8Pa1ARJqJyI3AYqCXiISKSEV8\nrSarMtuRiJQCyqrqLHxdZWl/2u7i90SrOxCeyzrOBfr5xcnTn8yqehQ4LCLtnKIH8LVuAZQGEkUk\nHF9LUNCNHTeMBTFfUfeqWmzZupwH/9KTd4aOpEOHtsSvXUCHDm15Z+jIoMQe8cIH/Os/Axk1dyS1\nr63F58MnMnHEJK5r14Sxiz/munZNmPTBpKDE9vK4/cXFxjNt2rcsW/4NsbFzCAkR/vtfdwbrVqpU\nkYULviJ21RyWLf2ab2cvYO68GFdip6Sk8ORTzzPrm89YvzaGKVO+ZuPGn1yJ7aU2rZtx//130b59\na+Ji5xIXO5dbb+3gSmwvz/n/Joxg6eIZXH1VbXbtiOOh3tlOLB4whf5aKwQtQVKo+ifzSEQigffw\nJSpn8CUtTwHbgLeBzoACr6rqJBGJBvqraldn++FAHDAHmI6vFUWAIar6iYhUcspDgPnA46paKoP9\nxDif4/yXiUgFYARQD1/r0GJVfVhEBgEnVHVIBsdUA9gK+Le9Pu2UjQJKADuAh1T1sIg8AgwEfsbX\nFVdaVXuLyDhgpqpOcfZ7QlUzfYBOqRI1PbugWl92tVehWf6bO60YGVG8+z+cnOrdw9JSPPwhGhri\n3d+PXg6ott8W3kk+tyc/m2fVq5Fnp799P1+XRETnJ4JSr9ywJMgElCVB7rMkyH2WBBm3Fcgk6Jv3\n8pcE3faU50nQRfdsGWOMMcYUAAVkXE9+WBJkjDHGmNwrION68sMGRhtjjDGmSLKWIGOMMcbknnWH\nGWOMMaZIKgTdYZYEGWOMMSb3rCXIGGOMMUVSIWgJsoHRxhhjjCmSrCXIGGOMMblXCFqCLAkyAVUy\nvJhnsWMOrPcs9sk9iz2LXSKyXfYrBYlvPuCix8unNnt5zm2GAXOBQnA9WBJkjDHGmNyzliBjjDHG\nFEmFIAmygdHGGGOMKZKsJcgYY4wxuWfPCTLGGGNMkVQIusMsCTLGGGNM7hWCu8NsTJAxxhhjiiRr\nCTLGGGNM7ll3mDHGGGOKpEKQBFl3mPHMe8NfY8O2ZSz6fsb5smsbXsOs7yYyf8lU5sRMocl1DYNe\nj2LFirFs6UziYucSv3o+L77wTED3f/bsOe7+25Pc8ZdH6X7fPxj+0QQA/t+gt+h699/ocf/DPP/6\nOyQlJwPw30+ncOdfHuPOvzxGj/sfplG72zh67HhA6wRQrVok8+Z+wdq1McTHL+Dxfn0CHiMzwT7n\n2enUMZoN6xezeeNSBg54zLW4ds7dP+djRg9lb8Ia4lfPdy2mP6+O2xWamr9XASD2GPTfiUgKsA4I\nB5KBT4D3VDP/tkQkGuivql0zWPasqr4egHr1Bpqqaj8RGQScUNUhudxHOeBeVf3A+RwJvK+qd+W3\nfv4qlb0mxxdUy9ZNOXnyFMNHvcmNrW4HYNLUj/lwxDgWfLeEm265gcee/Bt3dH0wR/s7fOZE3ioN\nlCxZgpMnTxEWFkbMwqn885mXWLXqxxxvn9W0GarK6dNnKFEigqTkZB58pD//evIfHD12nHatmgEw\ncNBbXB/VgLv/dOFlFLN0BeMnTeO//3kz0/3nddqMypUvp0rly1kdv55SpUqycuVs7rrrr2zatDXH\n+8jPFA75PeepefzZFRISwqYNS7i1yz0kJCSy4vtZ3P/Ao7k77jxFtnOen3OeV+3atuDEiZOMHTuM\nqCY3BT2ev0Aed/K5PfmpSlDmWjk1+ul8JRAl+r7r+bw71hJ0odOqGqWq1wK3AF2Al/Kxv2cDU62A\nKAc8mvZBVfcGOgHKrRXL4zhy+OgFZapK6TKlAChTpjT79x1wpS4nT54CIDw8jPDwsIDOkSQilCgR\nAUBycjLJycmICDe0bo6IICI0rHc1+w/8+odtZ323iC633Biwuvjbt+8Aq+N9862dOHGSzZu3EhlZ\nOSixMhLMc56V5s2asH37Lnbu3E1SUhKTJ0/n9m6dXIlt59z9c75k6UoOHT7iSqz0vDxukzOWBGVC\nVQ8AfYF+4hMqIoNFJFZE1orIP/xWLyMiU0Vko4iMEpEQEXkTiBCReBH5NP3+ReRWEflRRNaIyHyn\nrLyITHP2v0JEGmVVRxGpLSKzReQHEVkiItc45ZWc+qxxXq2BN4HaTn0Gi0gNEVnvrF9cRMaKyDoR\nWS0i7Z3y3iLylRNjq4i8HYhzm5UX/vU6L74ygB83LOSlVwfy2svvBDsk4PuLLXbVHPYkrGH+/CXE\nxq4O6P5TUlK48y+PcUPXe2jVrAmNrr3m/LKk5GS+njOfti2aXrDN6TNnWLoijlui2wa0Lhm58spq\nRDVuwKpVgT3urAT7nGcmsmplfknYe/5zwp5EVxORNHbO3T/nbiv0x52amr9XAWBJUBZUdQe+c3Q5\n0Ac4qqrNgGbA30WkprNqc+AZoCFQG7hDVf/F7y1L9/nvV0QqAmOAO1W1MfBnZ9HLwGpVbYSvFWl8\nNlUcDTyuqtcD/YEPnPL3gUXOvq8DNgD/ArY79RmQbj+POcfbELgH+EREijvLooBezrH1EpHq6Ssh\nIn1FJE5E4k6fy99fXL373MOLz77Jdde258Vn3+Dd4a/ma385lZqaSrPmnahZqxlNm0Zxbf2rA7r/\n0NBQvvxkBPOnTmDdxp/YumPX+WWvDhnB9Y0bcH1Ugwu2iVm6kiaN6lO2TOmA1iW9kiVLMHnSGJ7p\n/xLHj+e9SzG3gn3OM5NRd5LbwwLsnBeNGekL/XEXgjFBlgRlL+0q7gg8KCLxwErgMqCus2yVqu5Q\n1RTgcyC7P91bAotVdSeAqh5yytsCE5yyBcBlIlI2w0qJlAJaA184dfoQqOIs7gCMdPaToqpHM9qH\nH/+4m4GfgaucZfNV9aiqngE2Alem31hVR6tqU1VtGnFJuWxCZa3nPT34ZsZcAGZMnU2T67JsDAu4\no0ePsXjx93TsFB2U/ZcpXYpm1zVi6Yo4AD7476ccPnKUgU/0/cO6385fRJebg1OPNGFhYUyeNIbP\nP5/KtGnfBjVWZoJ9ztPbk5BI9WqR5z9Xq1qFxMT9rsQGO+fg/jn3SqE/7lTN36sAsCQoCyJSC0gB\nDuBLhh53WlKiVLWmqs51Vk3/bWb37Uom62Q0SCyzfYUAR/zqE6Wq9bKJm1V9MnPW730KQX6swr59\nB2jdtjkA7W5syY4dPwczHAAVKpSnbNkyABQvXpwOHdqyZcu2gO3/0OEjHHP+2j9z9iwrYldT88rq\nTJkxm2Urf+Dtl/8fISEX/lc8fuIkcavX0b5dq4DVIyNjRg9l8+ZtvDdsdFDjpBfsc56V2Lh46tSp\nSY0a1QkPD6dnz+58PXNu9hsGiJ1z98+5V4rqcV9M7DlBmXC6rEYBw1VVRWQO8IiILFDVJBG5Ckgb\nrt/c6Rr7GV/XUdpPtyQRCVfVpHS7/x4YISI1VXWniJR3WoMWA/cB/+fcdfarqh7LpEn1mIjsFJE/\nq+oX4lupkaquAeYDjwDviUgoUBI4DmTWr5IWd4FzXFcAW/B1pQXNqI+H0rptM8pfdimrN8Yw+I3/\n8MwTL/DqW88RFhrK2bNn6f/ki8GsAgBVKlfi44/fJTQ0lJAQYcqUmcyaFbjbaQ/+dpjnXh1CSmoq\nmqp06tCO6DYtaHzDbVSpdDn39f0nADff2JpH/urrOZ2/aDmtm19HiYjiWe06X9q0bsb999/FunUb\niYv1/WB+/oU3mT17QdBipgn2Oc9KSkoKTz71PLO++YzQkBDGfTKJjRt/ciW2nXP3z/n/Jozgxhta\nUaFCeXbtiOPlV4YwdtxEV2J7edyuKCDjevLDbpH3k8Et8hOAd1Q1VURCgFeBbvhaTg4CPYAmwIvO\n54b4EopHnW3eAm4HfsxgXFBn4HV8LToHVPUWESkPjAVqAqeAvqq6NrNb5J3EayS+brBwYKKqviIi\nlfAlYmktWY+o6vci8hnQCPgWGAHMVNUGzvifUcD1znH/U1UX+sd16jwTGKKqMZmdw9zcIh9o+blF\nPr+yukU+2PJ6i3wg5Od27fw4H+JEAAAgAElEQVTK6+3ageDlfb1F9ZwXdQXyFvlhD+fvFvknR3l+\ni7wlQSagLAlynyVB7rMkyLitQCZB7/0jf0nQUx96ngRZd5gxxhhjcq8QdIfZwGhjjDHGFEnWEmSM\nMcaY3Csgt7nnhyVBxhhjjMm9AvLAw/ywJMgYY4wxuWctQcYYY4wpitQGRhtjjDHGBJ6IVBeRhSKy\nSUQ2iMiT6Zb3FxEVkQrOZxGR90VkmzMRebYP/LWWIGOMMcbkXvC7w5KBZ1T1RxEpDfwgIvNUdaMz\nmfctwG6/9Tvjm9OzLtAC38OEW2QVwJIgE1BHz57yLLaXD/4sXS3as9jHFw3xLHaZ6AGexfZSaEio\nZ7GTU1M8i23MBYI8MFpVE4FE5/1xEdkEVMU3mfe7wEBgut8m3YHx6vtlsEJEyolIFWc/GbIkyBhj\njDG55+LAaBGpgW+aqpUicjuwR1XXpHuCelXgF7/PCU6ZJUHGGGOMKThEpC/Q169otKqOzmC9UsCX\nwFP4usieAzpmtMsMyrLM1CwJMsYYY0zu5fPuMCfh+UPS409EwvElQJ+q6lci0hDfJONprUDVgB9F\npDm+lp/qfptXA/ZmtX9LgowxxhiTe0HuDhNflvMxsElV3wFQ1XXA5X7r7AKaquqvIjID6CciE/EN\niD6a1XggsCTIGGOMMXkR/CdGtwEeANaJSLxT9qyqzspk/VlAF2AbcAp4KLsAlgQZY4wxJveC3BKk\nqkvJeJyP/zo1/N4r8FhuYtjDEo0xxhhTJFlLkDHGGGNyrTBMm2FJkDHGGGNyrxBMoGrdYabAKFu2\nDJ99Noo1axYQHz+fFi2ynfYlIKpVi2Te3C9YuzaG+PgFPN6vjytx0wTzuPf9dpQ+b46jx7+H86dn\nR/Dp3BUAbNm9jwf+7yPufP4DHn/3M06cPnPBdom/HaHlP17jk2+XBawu/ooVK8aypTOJi51L/Or5\nvPjCM0GJk5lOHaPZsH4xmzcuZeCAXA0hyDevrnPw9rgttvuxgy5V8/cqAKwlyEMiosD/VPUB53MY\nvidbrlTVrrnYTxQQmcWI+YvC0KGDmDcvhnvvfZjw8HBKlIhwJW5ycjIDB77M6vj1lCpVkpUrZ/Pd\n/MVs2rTVlfjBPO7Q0BD6392RejUiOXn6LHcP+pCW19bi5bEz+GevjjS9pgZTF//IuFnL6Xdnh/Pb\nDf5sDm0b1g1YPdI7e/YsHTv15OTJU4SFhRGzcCqz5yxk1aofgxYzTUhICO8Pe41bu9xDQkIiK76f\nxdcz5xaK7zsrXh63xfbmWgu64N8dFnTWEuStk0ADEUn7KXgLsCc3O3ASpyh8twVetEqXLkXbts0Z\nO3YiAElJSRw9esyV2Pv2HWB1/HoATpw4yebNW4mMrOxK7GAfd8VypalXIxKAkhHFqBVZkQOHj7Mr\n8Veuv/pKAFpdW5v5P2w8v82CHzZRreKl1K5aMWD1yMjJk7555sLDwwgPD3Nt7rfmzZqwffsudu7c\nTVJSEpMnT+f2bp1cie3lde7lcVts92ObnLEkyHvfArc57+8BPk9bICLlRWSaiKwVkRUi0sgpHyQi\no0VkLjAeeAXoJSLxItJLRCqKyDwR+VFEPhSRn0WkgrPtNBH5QUQ2OI8sT4vVR0R+EpEYERkjIsOd\n8ooi8qWIxDqvNsE4CTVrXsHBg4cYM2YoK1bMYuTIt1z7C9nflVdWI6pxA1atWu1KPDePe8/Bw2z+\nOZGGtatSp9rlxKzeAsDc2A3sO+T7RXzq7DnGzlrGwz1uDEod/IWEhBC7ag57EtYwf/4SYmPdOeeR\nVSvzS8LvD5FN2JPoWtLr5XXu5XFbbPdju6IQdIdZEuS9icDdIlIcaASs9Fv2MrBaVRsBz+JLeNJc\nD3RX1XuBF4FJqhqlqpOAl4AFqnodMBW4wm+7v6rq9UBT4AkRuUxEIoEXgJb4WqOu8Vt/GPCuqjYD\n7gQ+CtSB+wsLC6NJkwaMHj2Bli27cPLkaQYMeDQYoTJVsmQJJk8awzP9X+L48ROuxHTruE+dOcsz\nwycz4N5bKRVRnJf/2p2J81dx90sfcurMOcJDfbOij5y6kPs7taRE8WIBr0N6qampNGveiZq1mtG0\naRTX1r866DEB0k24COBaK5SX17mXx22x3Y/tBk3VfL0KAhsT5DFVXevMjnsPvqdd+muLL/FAVRc4\nCUtZZ9kMVT2dyW7bAn9ytpstIof9lj0hIn9y3lcH6gKVgUWqeghARL4ArnLWuRmo7/efuYyIlFbV\n42kF/pPghYVdSmhoqZwe/nl79iSyZ08isbG+h4JOnTqL/v0fyfV+8iosLIzJk8bw+edTmTbtW9fi\nunHcSckp/HP4ZLq0asjNTesDUDOyIh8OeBCAXft+ZfGanwBYt2MP38Vu5L1J8zh+6gwSIlwSHsY9\nN7cIaJ38HT16jMWLv6djp2g2bNwStDhp9iQkUr1a5PnP1apWITFxf9DjgrfXuafHbbFdj+2KApLI\n5Ie1BBUMM4Ah+HWFObKaEfdkFvvL8AmbIhKNL6lppaqNgdVA8czWd4Q460c5r6r+CRD4JsFT1aaq\n2jQvCRDA/v0HSUhIpG7dWgC0b9/G1cGDY0YPZfPmbbw3LMu5/AIu2Metqgz673RqVanAg7e2Pl/+\n2zFfS1dqaipjZizmz+2bAjDu2b/y7dCn+Xbo09zXsSV/69ouKAlQhQrlKVu2DADFixenQ4e2bNmy\nLeBxMhIbF0+dOjWpUaM64eHh9OzZna9nznUltpfXuZfHbbHdj21yxlqCCob/4pvobZ2TqKRZDNwH\n/J9T/quqHsugifU4UNrv81KgJ/CWiHQELnXKywKHVfWUiFyDr/sLYBXwrohc6uzrTmCds2wu0A8Y\nDL470VQ1bQ6XgHr66RcZN+59LrkknJ07d9O3b/9ghPmDNq2bcf/9d7Fu3UbiYn0/oJ5/4U1mz17g\nSvxgHvfqrbuZuXwtdatdTs8XRgLw+F03sXv/ISbOXwXATdfXo0e7JgGLmRNVKlfi44/fJTQ0lJAQ\nYcqUmcyaNd+V2CkpKTz51PPM+uYzQkNCGPfJJDZu/MmV2ODdde7lcVtsb661oCsED0uUwtQ/ebER\nkROqWipdWTTQX1W7ikh5YCxQE99kcH2d7rNBwAlVHeJsUx6YA4QDbwAL8bUqXQosAno5+wCYBlQF\ntgAVgUGqGuN0afUH9gKbgEOq+pwzoHoEUA9f0rxYVR/O7JiKF7/CswsqJTXFq9CEhoR6FvvIwrc8\ni10meoBnsVM9/NkV5uH3nezhdW68k3wuVzcOp5fl/Ft5dfzRzvn6T1j6g2+DUq/csJYgD6VPgJyy\nGCDGeX8I6J7BOoPSfT4ENEv7LCLFgE6qmiwirYD2qnrWWdw5k+p8pqqjnVvup+JrAUJVf8WXRBlj\njDG/KwRjgiwJKpyuACaLSAhwDvh7DrYZJCI34xsjNBdfi5ExxhiTocLQk2RJUCGkqluBXA3yUFV3\nBiYYY4wxBYQlQcYYY4zJPesOM8YYY0yRZEmQMcYYY4qigvLU5/ywJMgYY4wxuVcIkiB7YrQxxhhj\niiRrCTLGGGNM7l38D4y2JMgEVlF9km7ZYiW8i91+oGexR1aM9iz2Pw4s9Cy2McbGBBljjDGmqCoE\nSZCNCTLGGGNMkWQtQcYYY4zJPRsTZIwxxpiiyMYEGWOMMaZospYgY4wxxhRFhaElyAZGG2OMMaZI\nspYgY4wxxuSedYcZY4wxpijSQpAEWXeYKRAefbQ3q2JnExs3h0cfe8j1+J06RrNh/WI2b1zKwAGP\nBTXWe8NfY8O2ZSz6fsb5smsbXsOs7yYyf8lU5sRMocl1DYNahzQhISGsXPEtU78aG5T9txvyd+6N\nH8Ed371xvqzZ8/dwZ8zb/Gne69z00VNcUsb3tO3af2pNjzmvnX/9dfd4yte/Iij1cvP7Tq9s2TJ8\n9tko1qxZQHz8fFq0uM612F4d95jRQ9mbsIb41fNdi+nPy+/by9hBl5rPVwFgSVAQiUhlEZkoIttF\nZKOIzBKRq/Kxv94iMtx5/7CIPOhXHhmoeuckfiDVr38VvR+6mxtv6EHLFl3o3LkDtWvXCHSYTIWE\nhPD+sNfo2u1+GjZuT69ePahXr27Q4k38bCp33/n3C8pefGUAQ94cwU3t/sTbr73PC68MCFp8f4/3\n68PmLduCtv+tXyxmzv2DLyjbu3gdX930L6be8izHdiTSuF83ALZPXc60Ts8xrdNzLHpyJMd/+ZVD\nG3cHvE5uf9/pDR06iHnzYmjcuAPNmt3K5s3BO//+vDzu8eMnc1vX+1yJlZ6Xx+31tRZsmpq/V0Fg\nSVCQiIgAU4EYVa2tqvWBZ4FK6dbL02RbqjpKVcc7H3sDAU2CxMeV6+Pqq+uwKjae06fPkJKSwtKl\nq+h2eyc3QgPQvFkTtm/fxc6du0lKSmLy5Onc3i148Vcsj+PI4aMXlKkqpcuUAqBMmdLs33cgaPHT\nVK1amc6dOzB27OdBi7Fv5RbOHjlxQdmexevRFN9PwAM/bqdElfJ/2K5W99bsmP59UOrk9vftr3Tp\nUrRt25yxYycCkJSUxNGjx1yJ7eVxL1m6kkOHj7gSKz0vj9vL2CZnLAkKnvZAkqqOSitQ1XhVXSIi\n0SKyUEQ+A9YBiMj9IrJKROJF5MO05EhEHhKRn0RkEdAmbV8iMkhE+ovIXUBT4FNn2wj/SohIHRH5\nTkTWiMiPIlJbREqJyHzn8zoR6e6sW0NENonIB8CPQPXM4gfSxo1baNOmOeXLlyMiojgdO0VTrVqV\nYITKUGTVyvySsPf854Q9iURGVnYtPsAL/3qdF18ZwI8bFvLSqwN57eV3gh5zyOBB/PvZ10lN9e5P\nsqt63UDCwrV/KK/VrUXQkiAvv++aNa/g4MFDjBkzlBUrZjFy5FuUKBGR/YYBUBCucy94edyF/pxb\nd5jJQgPghyyWNweeU9X6IlIP6AW0UdUoIAW4T0SqAC/jSz5uAeqn34mqTgHigPtUNUpVT6db5VNg\nhKo2BloDicAZ4E+qeh2+ZG2o03IFcDUwXlWbAOeyix8IW7Zs5913RjFj5gSmTf+E9es2kZycHIxQ\nGfr90H+n6u7zL3r3uYcXn32T665tz4vPvsG7w18NarwunW/i4MHfWL16XVDjZKXx47eTmpLK9q+W\nXVBesUltks+c4/CWhKDE9fL7DgsLo0mTBowePYGWLbtw8uRpBgx41JXYBeE694KXx13Yz7l1h5n8\nWKWqO533NwHXA7EiEu98rgW0wNeddlBVzwGTchNAREoDVVV1KoCqnlHVU4AAr4vIWuA7oCq/d9P9\nrKornPc5ii8ifUUkTkTikpKP56aK543/ZDJtW3ejU8deHDp8hO3bd+VpP3mxJyGR6tV+702sVrUK\niYn7XYsP0POeHnwzYy4AM6bOpsl1jYIar1Xrptx22y1s2bKcCeNHEB3dhrFjhwU1pr86d7Xjipub\nENPvgz8sq3V7S3ZMC04rEHj7fe/Zk8iePYnExsYDMHXqLKKiGrgTuwBc517w9Psu5OfckiCTlQ34\nEpvMnPR7L8AnTktOlKperaqDnGX5+bPhj3+G+NwHVASud1qe9gPFM6hXjuKr6mhVbaqqTcPDSuep\nohUrXgZAtWqRdL/9Vr6YPCObLQInNi6eOnVqUqNGdcLDw+nZsztfz5zrWnyAffsO0LptcwDa3diS\nHTt+Dmq8F154i9p1mnP11a154MHHiIlZxkMPPRnUmGmqRjei0aNdmffQO6ScOXfhQhFqdm3BjhnB\nS4K8/L737z9IQkIidevWAqB9+zZs2rTVldgF4Tr3gpfHXVTP+cXEnhMUPAvwtbb8XVXHAIhIM6BE\nBuvOB6aLyLuqekBEygOlgZXAMBG5DDgG/BlYk8H2x531L6Cqx0QkQUR6qOo0ESkGhAJlgQOqmiQi\n7YErMzmGnMbPt08/G0n58uVISkrmn0+/yJEj7gwWBUhJSeHJp55n1jefERoSwrhPJrFx409Bizfq\n46G0btuM8pddyuqNMQx+4z8888QLvPrWc4SFhnL27Fn6P/li0OK7KXr4Y1RpVY/i5Utxd+z7/Dj0\nSxr3u52QS8K49fN/AXDgx20s/7fvFv3KLa/hZOIhju8+GLQ6uf19p/f00y8ybtz7XHJJODt37qZv\n3/6uxPXyuP83YQQ33tCKChXKs2tHHC+/MoSx4ya6EtvL4/b6Wgu2gtKakx9SmPonCxrntvX38LUI\nnQF2AU/h637qr6pd/dbtBfwbX+tcEvCYqq4QkYec8kQgHghV1X4iMgg4oapDRORO4HXgNNDKf1yQ\niNQFPgQqOPv9M76E5msg3NlnG6Czs8lMVW3gt32G8TM75lIlanp2QZ1JPpf9SkFyWUTeWsAC4cjZ\n9I137vmgwo2exf7HgYWexQ4LydNNnQGRnJriWWzjneRze/KzeWa9AvmyPzo6Xz/vK8XEBKVeuWFJ\nkAkoS4LcZ0mQ+ywJMm4riEnQvhvylwRVXux9EmTdYcYYY4zJNU31PIfJNxsYbYwxxpgiyVqCjDHG\nGJNrhWFgtCVBxhhjjMk11Yu/O8ySIGOMMcbkmrUEGWOMMaZIsoHRxhhjjDEXKWsJMsYYY0yuFYbH\nDFoSZIwxxphcKwzdYZYEmYDy8qnNXvrt9HGvq+AJL5/a7CV7arMxhSMJsjFBxhhjjClwROS/InJA\nRNb7lUWJyAoRiReROBFp7pSLiLwvIttEZK2IXJeTGJYEGWOMMSbXVPP3yoFxwK3pyt4GXlbVKOBF\n5zP4JgGv67z6AiNzEsC6w4wxxhiTa8HuDlPVxSJSI30xUMZ5XxbY67zvDoxX36zwK0SknIhUUdXE\nrGJYEmSMMcaYXPPoidFPAXNEZAi+3qzWTnlV4Be/9RKcsiyTIOsOM8YYY0yuaWr+XiLS1xnXk/bq\nm4OwjwBPq2p14GngY6c8o4ws2043awkyxhhjjOtUdTQwOpeb/QV40nn/BfCR8z4BqO63XjV+7yrL\nlLUEGWOMMSbXUlXy9cqjvcCNzvsOwFbn/QzgQecusZbA0ezGA4G1BBljjDEmD4I9JkhEPgeigQoi\nkgC8BPwdGCYiYcAZfHeCAcwCugDbgFPAQzmJkWkSJCJfk0V/mqrenpMAxhhjjCl8XLg77J5MFl2f\nwboKPJbbGFl1hw0BhmbxMiZgxoweyt6ENcSvnu9J/E4do9mwfjGbNy5l4IBc/z+y2BbbYlvsAhfb\nZE/UgxnQRKQaMAKojy8RmwkMUNVzIhIFRKrqLGfdQcAJVR0SoNjXABPxtXLdBUxQ1dZZbxUYIvIR\n8I6qbsxinRigv6rGuVCf24H6qvpmoPYZdknVPF1Q7dq24MSJk4wdO4yoJjcFqjo5EhISwqYNS7i1\nyz0kJCSy4vtZ3P/Ao2zatDX7jS22xbbYFtuF2Mnn9uSnKkFpstlUt0u+Eoh6W2d5Pu9GtgOjRaSu\niEwRkY0isiPtldeAIiLAV8A0Va0LXAWUAl5zVonC168XECISmq6oBzBdVZuo6na3EiAAVf1bVglQ\nMDj9ppnVZ0YgE6D8WLJ0JYcOH/EkdvNmTdi+fRc7d+4mKSmJyZOnc3u3ThbbYltsi33RxnaDpkq+\nXgVBTu4OG4vv8dPJQHtgPDAhHzE7AGdUdSyAqqbgu9f/ryJSBngF6OXMC9LL2aa+iMQ4CdgTaTsS\nkftFZJWz7odpCY+InBCRV0RkJdDKb/0u+B609DcRWZi2rvNvtBNjiohsFpFPnYQNEXlRRGJFZL2I\njPYrjxGRt5w6/CQi7ZzyUBEZIiLrnDlMHvdbv6nzfqTzXIQNIvJydidNRN50EtG1zkOiEJGKIvKl\nU7dYEWnjlA9y6jkXGC8iK0XkWr99xYjI9SLSW0SGO2WVRGSqiKxxXq2zOseFSWTVyvyS8PudlAl7\nEomMrGyxLbbFttgXbWw3eHR3WEDlJAmKUNX5+LrOflbVQfgSmby6FvjBv0BVjwG7gRr45gKZpKpR\nqjrJWeUaoBPQHHhJRMJFpB7QC2jjzCGSAtznrF8SWK+qLVR1qV+cWcAo4F1VbZ9B3ZrgS5LqA7WA\nNk75cFVtpqoNgAigq982Yara3NnuJaesL1ATaKKqjYBPM4j1nKo2BRoBN4pIo4xPF4hIeeBPwLXO\n/l51Fg1zjqUZcCe/Py8BfAPHuqvqvfi6/3o6+6qCr7vxgu8AeB9YpKqNgeuADdmcY//6nX/gVWrq\nycwOo8ByctoLuNVNbLEttsW22BcrVcnXqyDIyS3yZ0QkBNgqIv2APcDl+YgpZHzXWWblAN+o6lng\nrIgcACoBN+H7RR/rXGgRwAFn/RTgyzzUbZWqJgCISDy+pGwp0F5EBgIlgPLABuBrZ5uvnH9/cNYH\nuBkYparJAKp6KINYPZ2nY4YBVfAlXmszqdcxfLcCfiQi3+AbQ5UWp77ff7QyIlLaeT9DVU877ycD\n8/AlaT3xPWAqvQ7Ag059U4CjIvIAmZ/j8/wfeJXXMUFe2pOQSPVqkec/V6tahcTE/RbbYltsi33R\nxjY5k5OWoKfw/fJ/At8vxAfwPbExrzYATf0LnG6w6sD2TLY56/c+BV/iIMAnTotRlKpe7bRSga+7\nLSUPdftDHBEpDnwA3KWqDYExQPEMtkmrF2Sd0CEiNYH+wE1Oy8436fZ5ASeZao4vsesBzHYWhQCt\n/M5BVVU97iw76bf9HuA3p7WpF76WoZzI6hwXGrFx8dSpU5MaNaoTHh5Oz57d+XrmXIttsS22xb5o\nY7vBhVnkgy7bliBVjXXeniCHDx/KxnzgTRF5UFXHO2NMhgLjVPWUiBwHSme9i/P7mS4i76rqAafL\nqLSq/hyAOvpLS05+FZFS+O4om5LNNnOBh0UkRlWTRaR8utagMviSlKMiUgnoDMRktjMnbglVnSUi\nK/A9DCotTj9gsLNelKrGZ7KbicBAoKyqrstg+Xx8c7K853wnJXHvHPO/CSO48YZWVKhQnl074nj5\nlSGMHZfTXC1/UlJSePKp55n1zWeEhoQw7pNJbNz4k8W22BbbYl+0sd1QUMb15Ee2t8g7A4j/sJKq\n5nlckIhUx9e6cg2+1oxZ+G4LP+v8op0DhANvAPXwu0VeRNYDXVV1lzNw+t/OPpKAx1R1hYicUNVS\nmcQelG5/J1S1lIhEO3Xo6pQPB+JUdZyIvArcDezCN0vtz6o6SPxuZxeRCs76NcR3R9bbwK1Ovcao\n6vB0648DWgA78LUmzXBinV/Hr85VgOn4EjIBhqjqJ07MEc45CgMWq+rD6Y/R2UclfF2Z/6eqLztl\nvYGmqtrPWT4a31ioFOARVf0+s3Oc2Xd7MXaHGWNMQVcQb5FffUX3fP28b7J7uudZVE6SIP8nMxbH\nNwA3WVUHBrNi5uJkSZAxxgReQUyCfqyevyToul+8T4Jy0h2W/i6iZSKyKEj1McYYY4xxRbZJkNM9\nlSYE3+DowvOgA2OMMcbkWmEYE5STW+R/wDcmSPA9MHEn0CeYlTLGGGNMwVZQnvWTHzlJguqp6hn/\nAhEpFqT6GGOMMeYiUBhagnLynKDlGZR9H+iKGGOMMca4KdOWIBGpDFQFIkSkCb+PLi+D7+GJxhhj\njCmiCsOtwFl1h3UCegPV8D3MMC0JOgY8G9xqGWOMMaYgKwzdYZkmQar6CfCJiNypqnmZh8sYY4wx\nhVRRGRh9vYjMV9UjACJyKfCMqj4f3KqZi1FIBrMmu8XL2ZlDQnIyvC5IscW72MVDwz2Lffzc6exX\nCpKHI9t6FnvU3qWexTbGX6rXFQiAnPz07JyWAAGo6mGgS/CqZIwxxhgTfDlpCQoVkWKqehZARCIA\nu0XeGGOMKcI0OLNxuConSdD/gPkiMtb5/BDwSfCqZIwxxpiCLrUQ3B6Wk7nD3haRtcDN+O4Qmw1c\nGeyKGWOMMabgSi0iLUEA+/CNgeqJb9oMu1vMGGOMKcIKdXeYiFwF3A3cA/wGTAJEVdu7VDdjjDHG\nmKDJqiVoM7AE6Kaq2wBE5GlXamWMMcaYAq2w3yJ/J75usIUiMkZEboJC0PZljDHGmHxTJF+vgiDT\nJEhVp6pqL+AaIAZ4GqgkIiNFpKNL9TPGGGNMAZSaz1dBkO3DElX1pKp+qqpd8c0jFg/8K+g1M0VK\nsWLFWLZ0JnGxc4lfPZ8XX3jGtdjVqkUyb+4XrF0bQ3z8Ah7v18e12OB72vTKFd8y9aux2a8cQHXr\n1mLFilnnX/v3r6dfv78GLd5/PniDn3auZPmqWefLnn3hKZaumMni5TP4cvo4Kle+PGjx/XXqGM2G\n9YvZvHEpAwc8FvR40Q915tk5Q3hu7hCi/+p71myJsiXpN+E5Xlz4Hv0mPEdEmZJBr4fbx51mzOih\n7E1YQ/zq+a7F9FdUj9tkL1fP21fVQ6r6oap2yG5dEUkRkXgRWSMiP4pI67xWUkRiRKRpXrf3gogU\n+ElmRaSpiLzvdT0Azp49S8dOPWnarCNNm3WiY8domje/zpXYycnJDBz4Mo0aRdO2bTcefqQ39erV\ndSU2wOP9+rB5yzbX4qXZunUHLVt2oWXLLrRu3ZVTp04zY8acoMX7/NOvuKvHhUnWf977iLYtu3JD\n69uZM3sBA//dL2jx04SEhPD+sNfo2u1+GjZuT69ePYL6fVe5qjqt776Jwd2f5Y3OA2nQ4Toq1qjM\nLY/0YMvy9bzS/im2LF9Px0e7B60O4P5x+xs/fjK3db3PlVjpFdXjdkORaAnKh9OqGqWqjYF/A28E\nMVZBlK8kSERCA1EJEclqktw4VX0iEHEC4eTJUwCEh4cRHh7m2lxg+/YdYHX8egBOnDjJ5s1biYys\n7ErsqlUr07lzB8aO/dyVeJlp374NO3fuZvfuPUGLsXxZLIcPH7mg7PjxE+fflyxRwpXvvHmzJmzf\nvoudO3eTlJTE5MnTub1bp6DFq1ynKrtWbyXpzDlSU1LZtnIjjTs1p9EtTVk5ZREAK6csotEtzYJW\nB3D/uP0tWbqSQ+m+e1g7d5gAACAASURBVLcU1eN2Q6EeExRgZYDDACJSSkTmO61D60Sku1NeQ0Q2\nOYOw/z97Zx4nRXX17+c7MLIKBjWyKiqKK6ICKqACGhAX9H3FJREjaGJco6/7rpjl5xpFoyagQXBB\nEDckqBiUIC4wowyLCG5gBHGLyuYCzJzfH3Ubm3EWhpmqGqbPM5/6TNftqvu9t7q6+/S55577tqTJ\nYYmO9UjKkzRK0h9LC0i6TlKBpHmShkvRSp6SOkj6V5ZHaudQflnQny3pplDWWdIbkuZIeiosFruB\nJ0rSNpIWh8eDJT0p6XlJ70m6JZTfBDQKnrBHymjrL4P2PEk3Z5WvknSjpBnAQaXO+b2k+aFtj4Wy\nJpL+Efo9K+taDpb0uKRngcmSxko6MquuByUdL6mXpIlZr8vI0K45ko4P5X0lvR6u3eOSmm7ka15l\n8vLyKJj5AkuXzGbKlFcoKJgVl1S57LBDWzrvsxczZyajfdutN3DlVX+mpCTd30UnnDCAceMmpKJ9\nzfUXMW/BK5xw0gD+/Mdhseu1btOSj5d8sn5/ydJlsRq9nyz8mA7ddqPJVk3Jb7gFe/bel5+12pot\nt23Oii+iL8gVX3zDlts0i60NkHy/awu52u8kKFH1ttpAnEZQxghYANwP/CGUfw/8j5ntB/QGbs8Y\nLMAuwD1mtifwDdEMtQz1gUeAd8tZwf6vZtbVzPYCGgFHh/JHQp37AN2BZZL6A8cBB4TyW8Kxo4HL\nzawTMBe4fiP62Rk4CdgbOElSOzO7gh89YRv4QiW1Bm4G+oRzu0o6LjzdBJhnZgeYWemloq8A9g1t\nOyuUXQ28ZGZdia7lrZIygQUHAaeFocvHQhuRtAVwGDCJDbkWWG5meweNlyRtA1wDHB5er0Lgoo24\nJptESUkJXbv1Y8edutKlS2f23KNjXFJl0qRJY8aNHcHFl1y/gYciLo7sfxhffPFfZs2aG7tWReTn\n53PUUYfz5JP/TEX/j0P/wl67HczjYyfw29+dGrvejx83PxKnB+qzD5by4t8mcN7D13DuqKtY+s5H\nFBcXx6ZXHkn3u7aQq/1OghJUra02kMRw2G7AEcDoYOwI+HNYiuNfQBtgu3DOIjMrCo/fBNpn1fd3\nIgPhT+Xo9ZY0Q9JcIgNjT0lbAm3M7CkAM/vezL4lWgJkZHiMmX0lqTmwlZn9O9Q3CjhkI/o5xcyW\nm9n3wHwqX1KkKzDVzL4ws3VERlpGp5jys3HPAR6RNAhYF8r6AldIKiKawdcQ2D4896KZfRUePwf0\nkdQA6A9MM7PvStV/OHBPZsfMvgYOBPYAXg0ap5XVP0lnSiqUVFhSvLqS7lfO8uUrmDbtdfr261Xt\nujaW+vXrM27sCMaMeYqnn34uEc2DunfhqKN+wcKFr/HQ6Hvo1asHI0fG7wkpTb9+vSgqmsfnn3+Z\nuHY248dNYMCx8Q9TLF2yjHZtW6/fb9umFcuWfRar5uvjXubmo6/gzpNuYPU3q/hi0aes/GI5zbbd\nCoBm227Fyi9XxNqGNPpdG8jVfjsbRyLDYWb2OrANsC1wSvi/v5l1Bj4j+vIG+CHrtGI2TOb4GpGh\n05BShLJ7gYFmtjcwItRZnqkpoCo/Bdbx47UqrV9Rm8vTLo/vzay8n4hHERkp+wNvhlgfAccHY7Oz\nmW1vZu+E49dbI8FAmwr0I/IIPVZOu0pfExEZU5n69zCzn0ydMrPhZtbFzLrk1du0GS7bbNOC5s2j\n4YCGDRvSp09PFiYYLDxi+O0sWPA+dw4bnpjmtdfezM4dutGxY3dO/fW5TJ36KkOGXJCYfoYTT0xv\nKGynnX+0qY846jDefffD2DULCovo0GFH2rdvR35+PieeeCzPTpwcq2bTraN7+2ett2afI7pROOFV\n5v6rkAMGHgrAAQMPZc6LhbG2IY1+1wZytd9JYNXcagOJGEGSdgPqES2/0Rz43MzWSurNxi/G+gDR\nEM7jZQT7ZgyTL0PMykAAM1sBLMkMN0lqIKkxMBk4PTxGUgszWw58LengUNepQMYrtJjI+CBT90aw\nVlJ+GeUzgENDbFE9omVJ/l3GceuRlAe0M7OXgcuArYCmwAvA+VnxT/tWUM1jwBDg4HBeaSYD66fm\nhHioN4AekjqEssaKllOpcVq13I4XJ4/jzcIXef21iUyZ8gqTJiUzrbRH964MGjSQ3r27U1gwmcKC\nyRxxRKUTIOsEjRo1pE+fg3nmmedj17p/5B1MfulxOuyyI/MWTmfQr0/g+hsv5bWZk5j+xkT69DmY\nKy/9Q+UVVZPi4mIuuPAaJv3zUebNmcr48c8yf/67sWr+5r6LuPrF2/ndA5cz7tp/8N2K1bx43zPs\n1nNvrnv5TnbruTcv3vd0rG1Io98ZHn7oHqZPm0DHXXdm8YeFDBl8ciK6kLv9ToK6MDtMcY2NSiom\niquByKNwlZn9M8SZPAvkE+Uc6kE0RAMwMcT0IOkSoKmZ3SBpKnCJmRVKGgrsCpxiZiVZen8kWuts\nMfAx8FE4dxeiobRtgLXACWb2oaQrgF8Da4BJZnaVpM7A34DGwIfAEDP7Ohhx44BVwEvAIDNrL2kw\n0MXMzgttmAjcZmZTQ8DzAOCtMuKCfkU0Y05B+7JQvsrMfhJ4HIypl4kMSAEPm9lNigLH7ySKdRKw\n2MyOLt2urDo+BSaY2ZBQ1itc16OD8ZjxNBUDQ83sSUl9iGKYGoSqrjGzct0GWzRom5qBn+Y4f15e\nUnMMytBWetoN65Vl5yfDyjWlR3ST46zWPVPT/tsnpcMFnVxg3ZpqzdyMJQBnfKtTqvWhO3DZI6kH\nBsVmBDm5iRtBKWi7EZQ4bgQ5SVMbjaDHq2kEnVALjKD0Pj0dx3Ecx3FSpLIgXsdxHMdxnJ9QW+J6\nqoMbQY7jOI7jVJnakvCwOrgR5DiO4zhOlaktCQ+rg8cEOY7jOI6Tk7gnyHEcx3GcKlMX5pa7EeQ4\njuM4TpXxmCDHcRzHcXISnx3mOKVIM2Fhmq7ZxvUbVH5QTKSZNHBt8brKD6qDDF/2amra57Y+uPKD\nYuKeT15JTdupfdSF4TAPjHYcx3EcJydxT5DjOI7jOFXGY4Icx3Ecx8lJPCbIcRzHcZycpC4YQR4T\n5DiO4zhOlTFVb6sMSf+Q9LmkeVllt0paIGmOpKckbZX13JWS3pe0UFK/jemDG0GO4ziO49RGHgSO\nKFX2IrCXmXUC3gWuBJC0B3AysGc4515J9SoTcCPIcRzHcZwqU1LNrTLMbBrwVamyyWaWyc3xBtA2\nPD4WeMzMfjCzRcD7QLfKNNwIchzHcRynysRtBG0EpwPPhcdtgI+znlsSyirEjSDHcRzHcaqMVXOT\ndKakwqztzI3VlnQ1sA54JFNUThMrxI0gp1bQtm1rXpz8OHPmTKWo6CXOP++MxLRHDL+dT5bMpmjW\nlET07r73//Huohm8NnPS+rKrrr2Q6W9MZNprE3jimQdp2fLnibSlX99evD1vGgvmT+eyS89NRDOX\ntRs0aMCr0ydSWDCZollTuO7ai2PXPHRIf6544VaumHwrh57eH4ABV57CVVNu5/LnbuaMv19Eo2aN\nY29HLr7eaWvXdsxsuJl1ydqGb8x5kk4DjgZOsR+XKVgCtMs6rC3wSWV1uRFUDSRdLentEKVeJOmA\nUH6hpCp/qkhaVY22DJbUupznHpQ0cFPrToJ169Zx2WVD6dSpFz17HsNZZw9m9913SUR79OhxHHX0\nKYloAYx55EkGHnf6BmV333k/PQ88mkO6D+CF51/isivPi70deXl53DXsTxx9zCD23qc3J510XGLX\nPFe1f/jhB/r2O5EuXfvSpWs/+vbtRbdu+8Wm12rXthx0ch9uP/Zqbul/OXv22Y9t27dk4fS53NT3\nUm7ufzmfL/qUw885LrY2QO6+3mlqJ0GJqrdtCpKOAC4HBpjZt1lPTQBOltRA0o7ALsDMyupzI2gT\nkXQQkSW6X4hSP5wfxyMvBOL/abUhg4EyjaDNgU8//ZxZRdEsyFWrVrNgwXu0bt0yEe1Xps/gq6+/\nSUQL4LVXC/i6lN7KlT/av00aN05kDbZuXfflgw8Ws2jRf1i7di3jxj3DgGM2alapa1eD1aujz+38\n/Prk59eP9bXerkMbFs96j7Xfr6GkuIT3Z7zD3v26svCVOZQUR1EZH816j61atoitDZC7r3fa91rc\nxB0TJGkM8DrQUdISSWcAfwW2BF4Mzoe/AZjZ28A4YD7wPHCumRVXpuFG0KbTCvjSzH4AMLMvzewT\nSb8nMkZelvQybOjhkTRQ0oPh8Y6SXpdUIOkP2ZVLujSUz5E0NJS1l/SOpBHBAzVZUqPg5ekCPBJu\nikblNVrSYZJmSZobcjA0COXXBb15koZLUiifKulmSTMlvSsp9tUbd9ihLZ332YuZM2fFLVWruOb6\ni5i34BVOOGkAf/7jsNj1WrdpycdLfvQWL1m6LDHDM1e1IfIOFMx8gaVLZjNlyisUFMR3ny9b+DE7\nd9udxls1Jb/hFuzRuzM/a7X1BscccEIv3plaFFsbIHdf77TvtbhJYHbYL82slZnlm1lbM3vAzDqY\nWTsz6xy2s7KO/5OZ7WxmHc3suYrqzuBG0KYzGWgXDIN7JR0KYGZ3EY1D9jaz3pXUMQy4z8y6Ap9m\nCiX1JXLldQM6A/tLOiQ8vQtwj5ntCXwDHG9m44FCovHRzmZW5rLikhoS5V04ycz2JsoYfnZ4+q9m\n1tXM9gIaEXm5MtQ3s25EHq7rK70y1aBJk8aMGzuCiy+5fgPvSC7wx6F/Ya/dDubxsRP47e9OjV0v\n2LkbkIQHKpe1AUpKSujarR877tSVLl06s+ceHWPT+uyDT5jytwmc8/DVnDXqSj5556P1HiCAX5x7\nHCXFxRQ+PT22NkDuvt5p32txU93A6NqAG0GbiJmtAvYHzgS+AMZKGlzFanoAY8Ljh7LK+4ZtFvAW\nsBuR8QOwyMwyP9veBNpXQa9jOP/dsD8KyBhXvSXNkDQX6EOUcCrDkxXpZUf4l5SsrkJzNqR+/fqM\nGzuCMWOe4umnN8qIr5OMHzeBAcfG7zJfumQZ7dr+OILatk0rli37LHbdXNbOZvnyFUyb9jp9+/WK\nVeeNcS9z29FXcvdJQ/n2m9V8sWgZAF2PP4Q9D9uP0Rf8NVZ9yN3Xu7bca075uBFUDcys2Mymmtn1\nwHnA8eUdmvW4YQXPZRDw/7LcfR3M7IHw3A9ZxxVTtfXfygxFCx6ie4GBwUM0olQ7M5pl6mVH+Ofl\nNalCczZkxPDbWbDgfe4ctlETBOoUO+28w/rHRxx1GO+++2HsmgWFRXTosCPt27cjPz+fE088lmcn\nTo5dN5e1t9mmBc2bNwOgYcOG9OnTk4UL349Vs+nWkd7PWm9NpyO68uaE19jt0H04/KwBjPjNraz9\nfk2s+pC7r3ea2kmQRmB0TeMLqG4ikjoCJWb2XijqDHwUHq8kCtz6Mux/Jml3YCHwP+F5gFeJ0nw/\nDGRPT3oB+IOkR8xslaQ2wNpKmpTRrIgFQHtJHczsfeBU4N/8aPB8KakpMBAYX0ldNUqP7l0ZNGgg\nc+fOp7Ag+pC45tqbeP75l2LXfvihezj0kIPYZpsWLP6wkKE33sbIBx+LTe/+kXfQ4+AD2HrrnzFv\n4XRu+tMwftHvUHbZZSdKSkr4+D+fcNEF18amn6G4uJgLLryGSf98lHp5eTw4aizz579b+Ymuvcm0\narkdDzxwB/Xq1SMvT4wfP5FJk+JNzXD6fRfR5GdNKV5XzPhrR/LditUMHDqE+lvkc87DVwNRcPS4\nqx+opKZNJ1df7zS1k6AuLKCqujQ+mSSS9gfuBrYiStj0PnCmmX0p6XzgXGCZmfUOgcs3E80emwc0\nNbPBYRrfo0TG6BPANWbWNNR/AfCbILcKGETkiZkY4naQdEmo6wZJxwN/Br4DDsqOCwqB2BPNbLyk\nw4DbgmYBcLaZ/SDpj0QG2eLQzo9CvVOBS8ysUNI2QKGZtS/vuuRv0Sa1GyrNO3nLLcqNRY+dlWvK\nDAFzYiSvjFiPpDi7Vc/UtO/55JXUtHOddWuWVuf0WG7Y/7fDoGp97F750cOp+4PcCHJqFDeCkseN\noORxI8hJmtpoBP1ph1Oq9bF79UePpG4EeUyQ4ziO4zg5iccEOY7jOI5TZepCTJAbQY7jOI7jVJm6\nEEzjRpDjOI7jOFWmLniCPCbIcRzHcZycxD1BjuM4juNUmdqS8LA6uBHkOI7jOE6VKakDUUFuBDmO\n4ziOU2U2fxPIjSDHcRzHcTaBuhAY7UaQ49QAq9d+n3YTnARRihmj08zanGam7BJf3cCJATeCHMdx\nHMepMh4T5DiO4zhOTrL5m0BuBDmO4ziOswl4TJDjOI7jODlJXRgO84zRjuM4juPkJO4JchzHcRyn\nymz+fiA3ghzHcRzH2QQ8JshxHMdxnJzE6oAvyGOCHMdxHMfJSdwIcmoFbdu25sXJjzNnzlSKil7i\n/PPOSFS/X99evD1vGgvmT+eyS89NTLdBgwa8On0ihQWTKZo1heuuvTgx7RHDb+eTJbMpmjUlMc1s\n0rrmaWsD5OXlMeON53jqyZGJ6ubifQ65fa/FSUk1t9pAThpBkkzSQ1n79SV9IWliSu25KiGdWyW9\nLenWJPSqwrp167jssqF06tSLnj2P4ayzB7P77rskop2Xl8ddw/7E0ccMYu99enPSScclpv3DDz/Q\nt9+JdOnaly5d+9G3by+6ddsvEe3Ro8dx1NGnJKJVmjSveZraGc4/7wwWLHw/Uc1cvc9z/V6LkxKs\nWlttICeNIGA1sJekRmH/F8DSFNtTphGkiJp8jX4H7Gdml27MwZISixn79NPPmVU0D4BVq1azYMF7\ntG7dMhHtbl335YMPFrNo0X9Yu3Yt48Y9w4Bj+iWiDbB69bcA5OfXJz+/PpbQGkmvTJ/BV19/k4hW\nadK85mm/3m3atKR//z6MHDkmMU1Iv99p3ee5fK/FjVVzqw3kqhEE8BxwVHj8S2D9J5KkFpKeljRH\n0huSOoXyGyT9Q9JUSR9K+n3WOYMkzZRUJOnvkupJOkPSHVnH/FbSX7IbIekmoFE47xFJ7SW9I+le\n4C2gnaT7JBUGL87QrHMXSxoq6S1JcyXtFsoPDfUVSZolaUtJE4AmwAxJJ0naVtITkgrC1iOrj8Ml\nTQZGS9ozq19zJMX+M2aHHdrSeZ+9mDlzVtxSALRu05KPl3yyfn/J0mWJGWAQ/VosmPkCS5fMZsqU\nVygoSKbfaZLmNU/79b7t1hu48qo/U1KS7IBA2v1O6z7P5XstbtwTtHnzGHCypIZAJ2BG1nNDgVlm\n1onISzM667ndgH5AN+B6SfmSdgdOAnqYWWegGDglaAyQlB/OHQJsEARgZlcA35lZZzPLjE10BEab\n2b5m9hFwtZl1Ce08NGOUBb40s/2A+4BLQtklwLmhLQeH+gdk6YwFhgF3mFlX4Hjg/qw69weONbNf\nAWcBw0JdXYAlpS+kpDODkVZYUrK6jEu98TRp0phxY0dw8SXXs3LlqmrVtbGUtSJ4Ur9SAUpKSuja\nrR877tSVLl06s+ceHRPTTos0r3ma2kf2P4wvvvgvs2bNTUQvm1y9z3P1XnM2jpw1gsxsDtCeyAs0\nqdTTPYGHwnEvAVtLah6e+6eZ/WBmXwKfA9sBhxEZDgWSisL+Tma2GngJODp4afLNbGM+/T4yszey\n9k+U9BYwC9gT2CPruSfD/zdDfwBeBf4SPFVbmdm6MjQOB/4a2jsBaCZpy/DcBDP7Ljx+HbhK0uXA\nDlnl6zGz4WbWxcy65OU12YjulU39+vUZN3YEY8Y8xdNPP7fJ9VSVpUuW0a5t6/X7bdu0YtmyzxLT\nz7B8+QqmTXudvv16Ja6dNGle8zS1D+rehaOO+gULF77GQ6PvoVevHowcOSwR7Vy9z3P1XksCD4ze\n/JkA3EbWUFjgp+b7j0OYP2SVFRPlWhIwKnhZOptZRzO7IRxzPzCYMrxAFbDenSJpRyLPzmHBM/VP\noGHWsZn2ZNqCmd0E/AZoBLyRGSYrRR5wUFab25jZytL6ZvYoMAD4DnhBUp+N7EOVGTH8dhYseJ87\nhw2PS6JMCgqL6NBhR9q3b0d+fj4nnngsz06cnIj2Ntu0oHnzZgA0bNiQPn16sjDhgNk0SPOap6l9\n7bU3s3OHbnTs2J1Tf30uU6e+ypAhFySinav3ea7ea0lg1fyrDeR6ssR/AMvNbK6kXlnl04iGs/4Q\nyr80sxVluTYDU4BnJN1hZp9LagFsaWYfmdkMSe2A/YiGs8piraR8M1tbxnPNiIyS5ZK2A/oDUyvq\nlKSdg8dprqSDiIbwFpQ6bDJwHnBrOKezmRWVUddOwIdmdld43InIu1Wj9OjelUGDBjJ37nwKC6IP\niWuuvYnnn69xqZ9QXFzMBRdew6R/Pkq9vDweHDWW+fPfjV0XoFXL7XjggTuoV68eeXli/PiJTJqU\nzJT1hx+6h0MPOYhttmnB4g8LGXrjbYx88LFEtNO85mlqp0mu3ud+r8VHbfHmVAfl4vikpFVm1rRU\nWS/gEjM7OhgxI4EdgW+BM81sjqQbgFVmdls4Zx5wtJktlnQScCWRh2UtUUzOG+G4K4DOZnZyOe25\nmcjb8hZwNTDRzPbKev5B4ADgQyLPzwQze1DSYqCLmX0pqQtwm5n1knQ30JvIOzQfGGxmP2T3W9I2\nwD3A7kTG8DQzO6uMPl4JDAp9+hT4lZl9Vd61zd+iTWo3VJp3cl75BnLslOTgezht6uWl50QvTjig\nOhu/z9Nj3ZpqTWCO5YU7vf3Aar0o/1g8Pr0bKpCTRlDShPxDd5hZOlnpEsSNoOTJ9S+HNHAjKHly\n/T6vjUbQkPbHV+tFGbn4idSNoFyPCYoVSVtJepdoVladN4Acx3Gc3KEuBEbnekxQrJjZN8CuabfD\ncRzHcWqauuCdcyPIcRzHcZwqs/mbQD4c5jiO4zhOjuKeIMdxHMdxqkxtWfqiOrgR5DiO4zhOlakt\nCQ+rgxtBjuM4juNUmdoyw6s6uBHkOI7jOE6V8eEwxylFmm+JNLNupZl01BPYJU+aCQvTJM3Xu35e\nvdS015UUp6btxIsbQY7jOI7jVBmPCXIcx3EcJyepC/5QN4Icx3Ecx6kydWHtUU+W6DiO4zhOTuKe\nIMdxHMdxqozPDnMcx3EcJyfxmCDHcRzHcXKSujA7zGOCHMdxHMepMiVYtbaNQdJWksZLWiDpHUkH\nSWoh6UVJ74X/P9vUPrgR5DiO4zhObWUY8LyZ7QbsA7wDXAFMMbNdgClhf5NwI8ipFYwYfjufLJlN\n0awpiWu3bduaFyc/zpw5UykqeonzzzsjJ7QbNGjAq9MnUlgwmaJZU7ju2osT0wbo17cXb8+bxoL5\n07ns0nNzQjvN+zxNbUj39W7evBmPPvo3Zs9+iaKiKRxwwH6JaafZ77gxs2ptlSGpGXAI8EDQW2Nm\n3wDHAqPCYaOA4za1D6oL8/xrA5KKgblEcVbvAKeZ2bfVrPMGYJWZ3VYb6tkY6m/RZpNuqIN7HsCq\nVasZOXIYnfc9bJO0N3XxiJYtf06rlj9nVtE8mjZtwowZzzNw4Om88857m1hjstqqxrIZTZo0ZvXq\nb6lfvz5TX36Kiy6+npkz39ro8zd1GYW8vDzeefsVjjjylyxZsow3Xp/EoFPPSeSap6ldE/f55qhd\nE9e8Ostm3H//X3j11ZmMHPkY+fn5NG7ciOXLV2z0+Zu6bEZN3mvr1izdpDYEYllbp1+7/tUyIF74\n+LkK2yWpMzAcmE/kBXoTuABYamZbZR33tZlt0pCYe4Jqju/MrLOZ7QWsAc5Ku0GbE69Mn8FXX3+T\nivann37OrKJ5AKxatZoFC96jdeuWdV4bYPXqyE7Pz69Pfn79xJKfdeu6Lx98sJhFi/7D2rVrGTfu\nGQYc06/Oa6d5n6epneY133LLpvTs2Y2RIx8DYO3atVUygKpDmv1OAqvmn6QzJRVmbWeWkqgP7Afc\nZ2b7AqupxtBXWbgRFA+vAB0AJD0t6U1Jb2deYElnSLojc7Ck30r6S3h8taSFkv4FdCx1TIGk2ZKe\nkNRY0paSFknKD8c0k7Q4s18WkjpLekPSHElPZQLKyqo/lD8o6S5Jr0n6UNLAmr9ctYcddmhL5332\nYubMWTmhnZeXR8HMF1i6ZDZTprxCQUEy2q3btOTjJZ+s31+ydFlixl+a2rlKmtd8xx2354svvmLE\niNt5441J3HffzTRu3CgR7bp+r1U3MNrMhptZl6xteCmJJcASM5sR9scTGUWfSWoFEP5/vql9cCOo\nhpFUH+hPNDQGcLqZ7Q90AX4vaWvgMWBAlrEyBBgpaX/gZGBf4H+BrllVP2lmXc0sExh2hpmtBKYC\nR4VjTgaeMLO1FTRxNHC5mXUKbby+vPqzzmkF9ASOBm7a+KuxedGkSWPGjR3BxZdcz8qVq3JCu6Sk\nhK7d+rHjTl3p0qUze+7RsfKTaoCyhvCS8kKlqZ2rpHnN69evz7777sXw4Q9x4IFHsnr1d1x66TmJ\naPu9Vj3M7FPgY0mZD6bDiIbGJgCnhbLTgGc2VcONoJqjkaQioBD4DyGQi8jwmQ28AbQDdjGz1cBL\nwNGSdgPyzWwucDDwlJl9a2YriF7oDHtJekXSXOAUYM9Qfj+REUX4P7K8BkpqDmxlZv8ORaOIgs4q\nqh/gaTMrMbP5wHZl1LvepVlSsrriq1RLqV+/PuPGjmDMmKd4+unnckY7w/LlK5g27XX69uuViN7S\nJcto17b1+v22bVqxbNlndV47V0n19V66jKVLl1FQUATAU09NonPnvZLRruP3WtyB0YHzgUckzQE6\nA38m+jH+C0nvAb+gGj/O3QiqOTIxQZ3N7HwzWyOpF3A4cFDwsMwCGobj7wcG81PDpbw740HgPDPb\nGxiaqcfMXgXaSzoUqGdm8zax/WXWH/gh6/FPftpkuzTz8ppsony6jBh+OwsWvM+dw0p7Y+uu9jbb\ntKB582YANGzY88O6oQAAIABJREFUkD59erJw4fuJaBcUFtGhw460b9+O/Px8TjzxWJ6dOLnOa+cq\naV7zzz77giVLlrHLLjsB0Lt3j0SC4KHu32tJ5Akys6Lw/dLJzI4zs6/N7L9mdpiZ7RL+f7WpfXAj\nKF6aA1+b2bfB43Ng5okwxtkO+BUwJhRPA/5HUiNJWwLHZNW1JbAsDKGdUkpndKijXC9Q0FwOfC3p\n4FB0KpDxClVUf+w8/NA9TJ82gY677sziDwsZMvjkxLR7dO/KoEED6d27O4UFkyksmMwRR/Sp89qt\nWm7Hi5PH8Wbhi7z+2kSmTHmFSZOSmT5dXFzMBRdew6R/Psq8OVMZP/5Z5s9/t85rp3mfp6md5jUH\n+L//u44HH7yLgoIX6NRpD2655Z5EdNPud9xUNzC6NuBT5GsISavMrGmpsgbA00AbYCGwLXCDmU0N\nz18BdDazk7POuRr4NfARUVDYfDO7TdLZwGWhfC6wpZkNDue0BBYBrUIOhdJtu4EwRT5MOfwb0Bj4\nEBhiZl+XV7+kB4GJZja+vH5ms6lT5GuCWOaAbgZUZ4p8ddnUKfKOU1WqM0W+umzqFPkabUMtnCLf\nq+3h1foAmLrkX6l/bLsRlCKSJgJ3mFm1fn6HGVvHmtmpNdOyTceNoORxI8jJBdwIqn1G0CFtDqvW\nB8C0pVNS/9j2BVRTQNJWwExgdg0YQHcTzUY7siba5jiO4zgbQ134CeRGUAqEIatda6iu82uiHsdx\nHMepChsb3FybcSPIcRzHcZwqUxeMIJ8d5jiO4zhOTuKeIMdxHMdxqkxdmFjlRpDjOI7jOFWmLgyH\nuRHkOI7jOE6VqS0JD6uDG0GO4ziO41SZujAc5oHRjuM4juPkJO4JcuoMm/9vkk0jzV9jDetvkZr2\n9+vWpKZdLy+934/FJSWpaadJmlmbU09rXEvxmCDHcRzHcXKSujAc5kaQ4ziO4zhVpi54gjwmyHEc\nx3GcnMQ9QY7jOI7jVBmfIu84juM4Tk5S4jFBjuM4juPkIu4JchzHcRwnJ6kLniAPjHYcx3EcJydx\nI8ipFYwYfjufLJlN0awpqej369uLt+dNY8H86Vx26bmunQDnnDOYmQXPU1D4AuecOyRR7TT7DZCX\nl8eMN57jqSdHJqqbVr9z9f3dtm1rXpz8OHPmTKWo6CXOP++MxLSTwKr5VxtwI6gaSCqWVCRpnqTH\nJTWOWW+ApCvKeW5VnNpBY6qkLnHUPXr0OI46+pQ4qq6UvLw87hr2J44+ZhB779Obk046jt1338W1\nY2SPPXZl8JCTOfSQ4zjwgCPp378PO+/cPhHtNPud4fzzzmDBwvcT1Uyz37n6/l63bh2XXTaUTp16\n0bPnMZx19uDE77U4KTGr1lYbcCOoenxnZp3NbC9gDXBW9pOKqLFrbGYTzOymmqqvIiQlGi/2yvQZ\nfPX1N0lKrqdb13354IPFLFr0H9auXcu4cc8w4Jh+rh0jHTt2YGZBEd999z3FxcVMnz6TYwbU/X4D\ntGnTkv79+zBy5JjENCHdfufq+/vTTz9nVtE8AFatWs2CBe/RunXLRLSTwD1BTjavAB0ktZf0jqR7\ngbeAdpL6Snpd0lvBY9QUQNJiSTdLmhm2DqH8GEkzJM2S9C9J24XywZL+Gh7vGOoskPSH8hol6deS\n5kiaLemhSuq/QdJwSZOB0ZIaSXosnD8WaBTj9UuN1m1a8vGST9bvL1m6LLEPqlzVnj9/IT16dKNF\ni61o1Kghffv1om3bVolop9lvgNtuvYErr/ozJQmvAZZ2v9OitvR7hx3a0nmfvZg5c1bi2nHhniAH\nWO816Q/MDUUdgdFmti+wGrgGONzM9gMKgYuyTl9hZt2AvwJ3hrLpwIHh/MeAy8qQHQbcZ2ZdgU/L\nadeewNVAHzPbB7hgI+rfHzjWzH4FnA18a2adgD+F5+oc0k+XR0xqTZxc1V648APu+MvfmDDxIZ5+\nZhTz5r7DunXrEtFOs99H9j+ML774L7Nmza384BomzX6nSW3od5MmjRk3dgQXX3I9K1fGHrngVAGf\nIl89GkkqCo9fAR4AWgMfmdkbofxAYA/g1fBm3AJ4PauOMVn/7wiP2wJjJbUKxy8qQ7sHcHx4/BBw\ncxnH9AHGm9mXAGb21UbUP8HMvguPDwHuCufOkTSnrIsg6UzgTADVa05eXpOyDqu1LF2yjHZtW6/f\nb9umFcuWfebaMTN61DhGjxoHwPVDL+GTpWXa8jVOmv0+qHsXjjrqF/Q7ojcNGzSgWbMtGTlyGEOG\nXFD5ydUk7dc7LdLud/369Rk3dgRjxjzF008/l5huEtSWIa3q4J6g6pGJCepsZueb2ZpQvjrrGAEv\nZh23h5llTxGwMh7fDfzVzPYGfgc0LEe/sjtQ5RxTUf2rSx1b6V1uZsPNrIuZddncDCCAgsIiOnTY\nkfbt25Gfn8+JJx7LsxMnu3bMbLvt1kA0g+bYAUfw+LgJieim2e9rr72ZnTt0o2PH7pz663OZOvXV\nRAwgSP/1Tou0+z1i+O0sWPA+dw4bnphmUpiVVGurDbgRFD9vAD2y4n0aS9o16/mTsv5nPETNgaXh\n8Wnl1PsqcHJ4XN60iynAiZK2DtotqlA/wLRM3ZL2AjpVcGy1ePihe5g+bQIdd92ZxR8WMmTwyZWf\nVEMUFxdzwYXXMOmfjzJvzlTGj3+W+fPfde2YeeTR+yh8czKPj7+fi/7vOr75ZkUiumn3Oy3S7Heu\nvr97dO/KoEED6d27O4UFkyksmMwRR/RJRDsJSrBqbbUB5cKYcFxIWmVmTUuVtQcmhhljmbI+RMNV\nDULRNWY2QdJiYCRwJJFB+ksze1/SsURDY0uJjKiuZtZL0mCgi5mdJ2lH4FGiIc0nQp0btCVonwZc\nChQDs8xscAX13wCsMrPbwrmNQvv2AIqADsDvzaywvGtSf4s2fkPlEA3rb5Ga9vfr1lR+UEzUy0vv\n92NxwgHVTuRST5u1a5ZWflD5xNKFHbbuVK3P+4/+Oyf1S+tGUIoEI6hLJmanLuBGUG7hRlDyuBGU\nPKl/U1M7jaDtW+xdrc/7/3w1N/VL64HRjuM4juNUmdoypFUd3AhKETNrn3YbHMdxHGdTqAsjSW4E\nOY7jOI5TZWpLwsPq4LPDHMdxHMfJSdwT5DiO4zhOlakLyRLdCHIcx3Ecp8p4TJDjOI7jODmJzw5z\nHMdxHCcnqQueIA+MdhzHcRwnJ3FPkOM4juM4VaYuTJF3I8ipUdZVL7W74ziOs5lQF4bD3AhyHMdx\nHKfK1IXAaI8JchzHcRwnJ3FPkOM4juM4VcaHwxzHcRzHyUk8MNpxHMdxnJzEl81wHMdxHCcnqQue\nIA+MdhzHcRwnJ3FPkOM4juM4VcYDox3HcRzHyUnqQkxQTg6HSZoqqV+psgsl3SuptaTx5ZzXXtKv\nakC7S3XqcBzHcZy0MbNqbRuDpCMkLZT0vqQraroPOWkEAWOAk0uVnQyMMbNPzGxg6RMk1QfaA9Uy\nguJAUr2K9h3HcRynponbCArfZfcA/YE9gF9K2qMm+5CrRtB44GhJDSDy8ACtgenB2zMvlA+W9Lik\nZ4HJwE3AwZKKJP1feP6vmUolTZTUKzy+T1KhpLclDa2sQZK6SnpN0mxJMyVtWUn9qyTdKGkGcJCk\nxZKukzQdOEHSzpKel/SmpFck7RbOe1DSXUHrQ0kDs+q/TNLc0IabQh1vZT2/i6Q3N+2SO47jOE6V\n6Aa8b2Yfmtka4DHg2JoUyMmYIDP7r6SZwBHAM0ReoLFmZpJKH34Q0MnMvgoGyCVmdjRERlIFMleH\nc+oBUyR1MrM5ZR0oaQtgLHCSmRVIagZ8V0k3mgDzzOy6UAfA92bWM+xPAc4ys/ckHQDcC/QJ57YC\negK7AROA8ZL6A8cBB5jZt5JahPYvl9TZzIqAIcCDlbTLcRzHyQESiAhqA3yctb8EOKAmBXLVEwQb\nDomdHPbL4kUz+2oT6j8xeFFmAXsSufLKoyOwzMwKAMxshZmtq6T+YuCJUmVjASQ1BboDj0sqAv5O\nZPhkeNrMSsxsPrBdKDscGGlm34Y2ZPp8PzAkGHMnAY+WboikM4PXq1DS7wBt6lbd813btV3btWub\ndtr6ks4kBtatWarqbKW+OwrLaKfKkK1R2yuXjaCngcMk7Qc0MrO3yjludQV1rGPDa9gQQNKOwCXA\nYWbWCfhn5rlyEGW/sGXWH/jezIrLaWse8I2Zdc7ads867odS2hW14Qmi8dijgTfN7L+lDzCz4WbW\nJWzDy6ijKsTyZnVt13Zt105RO239tPteJqW+O8r6/lgCtMvabwt8UpNtyFkjyMxWAVOBf1C+F6g0\nK4Ets/YXA50l5UlqRzR+CdCMyCBZLmk7IiOiIhYArSV1BQjxQPUrqL9CzGwFsEjSCaE+SdqnktMm\nA6dLahzOaRHq+h54AbgPGLkx+o7jOI5TAxQAu0jaMYSNnEwUwlFj5GRMUBZjgCf56Uyx8pgDrJM0\nmyg25k5gETAXmAe8BWBmsyXNAt4GPgRerahSM1sj6STgbkmNiOKBDg/n/aT+jeQU4D5J1wD5RAFl\nsytow/OSOgOFktYAk4CrwtOPAP9LZCg5juM4TuyY2TpJ5xH9EK8H/MPM3q5JDdWFjI9OvEi6BGhu\nZtcmoHVmDQypubZru7Zr1xrttPXT7nttxo0gp0IkPQXsDPQxsy/Tbo/jOI7j1BRuBDmO4ziOk5Pk\nbGC04ziO4zi5Ta4HRjspI+muMoqXA4Vm9kzS7UkSSW2AHch6H5rZtITb0MTMKkoDUVM6d1NBfg8z\n+30CbRDRhIGdzOxGSdsDLc1sZgLajYGLge3N7LeSdgE6mtnEuLXTRNLOwBIz+yEkm+0EjDazb9Jt\nWXyEtCvlUkE6lppsQyHRbN5HzezruPU2Z9wT5KRNQ6Az8F7YOgEtgDMk3RmnsKT/lfReyIq9QtJK\nSSvi1MzSvplo9t81wKVhuyQJ7aDfXdJ84J2wv4+ke2OULATeJHq99+PH17szUeLPJLiXKAP8L8P+\nSqJ1iZJgJFF+roPC/hLgj0kIS+oh6UVJ7ypaKmeRpA+T0CbKM1YsqQPwALAjZSRcjYMU39+3h+0e\nYAYwHBgRHpf1oy8OTiZaCqpA0mOS+oUfAU5pqrsAmm++VWcDXgLqZ+3XD2X1gPkxa78P7J5SvxcC\nDVK87jOIkpDNyiqbl4Duy0B+1n4+8HJCfX4r/M/u8+yEtAtT1F5AlKvs58DWmS3ha34pcH7paxCz\ndmrv76D/GLB31v5ewIMJtyEPGAAsJVp+YijQIq1rUhs3Hw5z0qYN0Tpoy8N+E6C1mRVL+qH802qE\nz8zsnZg1yuNDIgMg7j6Wi5l9XOrHYRIemdZECUczy7I0DWVJsDYs/2IAkrYFShLSXhNygGW0dya5\n1365mT2XkFZp1kr6JXAacEwoy09IO833N8BuZjY3s2Nm80IutkSQ1IlovccjiTxyjxCtGfkSkQfW\nwWOCnPS5BSiSNJVo6Y5DgD9LagL8K2btQkljiZZQWf+FZGZPxqwL8C1Rv6eU0o49NibwsaTugIVM\nrL8nDI3FzE3ALEkvh/1DgRsS0IVoKOIp4OeS/gQMJBqOTILrgeeBdpIeAXoAgxPSflnSrUSJYbPv\ntdhjU4i+hM8C/mRmi8KSQg8noAvpvr8BFki6n6i/BgwimfcYkt4EviEagrzCzDL9nyGpRxJt2Fzw\nKfJO6khqRbQkiICZZlaja8NUoFvWMiBmZqcnoH1aWeVmNipu7aC/DTCMKDO5iLKBX2BlrA1Xg5oi\nWvtnLT+uBD3DzD6NS7OMNuwGHEbU5ylJegokbQ0cGLTfsITybmUZnNmYmfWJWbceMMrMBsWpU4F+\nau/voN8QOJvohx3ANOA+i5Yiilt7JzNLKu5rs8aNICd1asMsqTQIHphdw+5CM1ubZnuSQNKbZrZ/\nStoHAm+b2cqwvyWwh5nNSED7f4CXzGx52N8K6GVmT8etnSaSXgCOMbM1abclSWqBAfhn4BYLs/Ak\n/Qy42MyS8nxuNrgR5KRKmCV1EtE6a5n4DDOzAQloNwTOAPYkmrWUEU/CE9QLGEW0SK6IgpRPS8r4\nSys1gaR7iIJDC+LSqEB7FrCfhQ89SXlE/a1wSnMNaReZWedSZbPMbN+4tYPWUfz0Pr8xAd2/E80G\nnEC0qHRG+y8JaKf2/g76qRmAZd1bkt5K4l7f3PCYICdtjiPKl5JGgPBDRDNn+gE3EuWQSWp45Hag\nr5ktBJC0K9GCvkl5SRoCuwGPh/3jiQzRMyT1NrMLY9LtDfxO0kdEX4oiMno7xaSXjSzrV5+ZlUhK\n6jOwrHQkiWhL+hvQmOja308UCxV7bqTAJ2HLIwqIT5I0398Q/cB5VVLiBiBQT1KDzOdqCMpvkIDu\nZocbQU7apDlLqoOZnSDpWDMbJelRotWKkyA/YwABmNm7kpKaNQPQgWg9uHUAku4jigv6BTC3ohOr\nSf8Y666MDyX9Hrgv7J9DdP8lQaGkvxDljjHgfKK8SUnQ3cw6SZpjZkMl3U4UJB07ZjYUkkvKWYo0\n39+QrgH4MDAlxEUZcDqR59kphRtBTtqkOUsqE4PzjaS9gE+B9gnoQvSl+ADRr1WIfqUm9aUIKaUm\nMLOPACT9nKwhioQ4i2iG2DVEXwxTgDMT0j4fuBYYy4+B6OcmpP1d+P+tpNbAf4mSFsaOpIOIZig1\nBbaXtA/wOzM7JwH5NN/f6w3ANDCzWyTN5cdJAH8wsyQNwM0GN4KctJkQtjQYHgIGrw1taBoeJ8HZ\nRF+Cvyf6kJpGlNE4KVJJTSBpANFQYGvgc6KA+HeI4jZixcw+J8qkmzjBC3JFGtrAxBCIfSvwFpEB\nOCIh7TuJhqMmAJjZbEmHVHxKjZHm+zuTh+oyfhqTFOusvCyd54C08kNtNnhgtOPkKGmkJpA0G+gD\n/MvM9pXUG/ilmcXmkZF0WfhlXOb6ZXF6HSXdaWYXSnq2HO3YJwCUak8DoGFmlloCejPM7IDsQF1J\ns81snyT000TSZCLP3yVEXsjTgC/M7PIEtP8XuJkoS7j4MfauWdzamxvuCXJSQdI4MzsxuGzL+nKI\nPVBWUnOiRH0Hh6KpRG7j2L4gakO/s/geWEb0K7WDpA4JzE5ba2b/lZQnKc/MXg4zBOMkEwxbGLNO\nWWSGO29LQRuAEGuWna9mqqS/J5SSIa2knKm8v0uxtZk9IOkCM/s38G9J/05I+xaimWlpZszeLHAj\nyEmLC8L/o1Nswz+AecCJYf9UooUu/zdGzdrQbyT9JrSlLVBElMTvdSIvTZx8I6kp0fDfI5I+B9bF\nKWhmz4aHc8xsVpxaZWhn4rxaAJNSmgV5H9Hkg8xw66mh7DcJaJ9FlJSzDdGisUnGQqXx/s4mY2Qu\nCykKPiF6vyVB2kuGbDb4cJiTKpJuLu0eLqssJu2ycrf8pCwm7dT6HbTmAl2JMhd3DpmUh5rZSTHr\nNiEK1M0jCgZvDjwSZ6bqLO2XgVZEaQEeM7O349bM0h5JZGBOI1pY84XMzLwEtH8y/BT3kFTmXpZ0\ngpk9XvkZsbQhtfd30DoaeIUoB9jdQDOi91jsMZCShgEtSW/JkM2GsnJXOE6S/KKMsqSmUX8nqWdm\nR9GaOt9VcHxNkma/Ab7PpO8P+UQWAB0T0P05sIWZrbNoiZARJDR92Mx6A72AL4iCZudKSiSDrpkN\nIUpL8DjwK+ADRetKJUGxogVbgWhJBeJfLPfIMAx3Zcw6FZHm+xszm2hmy81snpn1NrP9kzCAAs2I\nZt72JVq49hhS9j7XVnw4zEkFSWcT5WnZWdKcrKe2BF5NqBlnAaND7ADA10TBi7FRSb9fi1O7FEvC\njKGngRclfU3kro+bx4HuWfvFoaxrAtpYtE7ZXcErdBlwHfDHhLTXSnqOKBasEXAsyQxJXUq0iOqH\nRAGyOxAtbBonzwNfAk0krQi6lvmfUIDu2cCo8P4W8BXJLVqLpFFE6/FlL11xexIZq4PR7WwEPhzm\npEL4YPoZ8P/YcOrwSjP7KuG2NAMwsxWSjjezJ2LUqjX9zmrToUTDUs/HneK/nCGKRGYLSdqdaImW\ngUS5ch4DnghT5+PWPoJoen5vogDdscDkBIfEGhB5+gQsSCo2SdIzZnZsEloVtGH9+zth3bKWrkhk\nqRRFGejvA7Yzs70kdQIGmFkiBv/mhBtBTmooWrtpjpntlXZbMkj6j5ltn4BOaot5Br0WZRSvjHvG\nkKQXgbszwwKSjgV+b2aHxakbtN4gWprk8STSAZTSHkNk+DyXoAFSYQBwEvEhmRiwsETJrkRLtTwX\n530m6aKKnk9o2YpMOoheZvZ12G8B/NvM9k5A+99EHsC/Z6UmmFebPmtrCz4c5qRG+GCcLWl7M/tP\n2u0JKCGd+4gWlsywuoyyOHmLKGDza6I+b0U0i+Vz4LdZs5pqmrOIZoX9NewvIZq1EyuKVvX+wMyG\nxa1Vjva2lvyK8cdU8JyRzNIZ04CDw1DQFKI0BScRBcXHRdJLVJTH7cBrksYTXe8TgT8lpN3YzGZK\nG3ycJeJ13NxwI8hJm1bA25JmsuEig4kmkcsiKddomot5QhSz8VQmlb6kvsARwDiiqdQHxCFqZh8A\nB4Zp8sp4wuLGouVAtpa0RdxDfuVofyupeYI5ampLXIjM7FtJZxB5AG+RFGuaAktxuYpszGy0pEKi\nWYEC/tfM5ick/2UIhjcASQOJcoI5pXAjyEmbxD+wyktUSPRBtV1CzUhzMU+ALmZ2VmbHzCZL+rOZ\nXRTiR2LFzFbFrVEGH5Heqt7fA3PDcGC2dhJr5KWJFK0fdgpwRijLme+dYPQkZfhkcy4wHNhN0lJg\nEfF63zZbcuZmdGonZvZvSdvx4+ygmQkEqtaGqaJpLuYJ8JWky4mCgyEaovgmDN2UJNiOJElzVe9/\nhi3XuIBomvxTZvZ2mJ7/csptqtOEWMsuZnZ4iMnKS8rjujnigdFOqkg6kWhhx6lEnpiDgUvNbHya\n7arrSNoGuB7I5FGZTuSVWwFsb2bvx6TboHRgcFllcSKpiUULmiaKpEZE13ZhwrqpX/M0kFTPzOLO\nh1QrkTTNzJJaqHazxpMlOmlzNdDVzE4zs18TLeiZ2ErPaSFpV0lTJM0L+52SStwX6G1m55vZvmE7\nP5SticsACry+kWU1jqSDJM0nrF0laR9J91ZyWk1pH0O0PMnzYb9zGJZLgjSv+a6ShkuaLOmlzJaE\nNvC+pFsl7ZGQXm3iRUmXSGonqUVmS7tRtREfDnPSJq/U8Nd/yQ3jfARhCiuAmc2R9CgJJe4jGqIo\nvZxBWWU1gqSWROtHNZK0Lz/OwmsGNI5DswzuBPoBEwDMbLakpH4t30Bk4E8N2kWSdoxTsJZc88eB\nvwH3E3+W6tJ0IsrNdH8YIvoH0XIpseYLkrSSCiZYJJQoMpOQMXudNgN2SkB7s8KNICdtnpf0AlH+\nFohiUybFLRpiX0aZ2aC4tcohlSmskvoDRwJtJN2V9VSzmPX7EWXrbQtkByKvBK6KUXcDzOzjUtc8\nqS/mdWa2vJR23LEIteGarzOz+yo/rOYJcTAjgBHB2B0D3BGmrP8hLo+nmW0JIOlG4FPgISID9BQS\niEULBt8gM0sq8/5mjRtBTqqY2aUhqVtPog+K4Wb2VAK6xZK2TWPKdCCtKayfEOVqGQBk5wJaCfxf\nXKJhnbBRijkjdyV8LKk7YJK2AH5PGBpLgHmSfgXUk7RL0I51mZRacs2flXQO8BQbLuQZe3b08EPn\nKKIlQtoT5e15hCjucBKwa8xN6Gdm2akm7pM0A7glTtGQbuM24KA4deoKHhjtpE5w2x9ANCupIKzv\nlITu34mSEyY+ZTrMkhlOtI7W10RTWAeZ2eIEtOsBo80ssSmzkgaZ2cOSLqYMD0hC13wbYBhwOJHB\nPZlobackVrBvTBT/1jdov0Dkjfg+Ae0GwPFEhsD6H75mdmMC2ovKKDYzi31YRtFaaS8DD5jZa6We\nuyvu9ASSXgPuIZqBacAvgXPNrHuFJ9aM9lBgDvCk+Zd8hbgnyEkVSb8hWsTyJaIvh7sl3Whm/0hA\nPrUp02b2IZDKFNaUEgc2Cf+bJqT3E8zsS1LKlWJm3xIZQVcHI7RJEgZQ4BlgOZHnL9EZYWYWa9xT\nJXQqLx9VQvmZfkVkdA8jMoJeDWVJcBHRe26dpO9JduHazQr3BDmpImkh0D3za1zS1sBrZtYxwTYk\nNmVatWddo9S8YGkh6RaiwPPviGZp7QNcaGYPJ6D9KFFuqGIiY6Q58BczuzUB7VTXjJK0F7AH0DBT\nZmajE9BtSJSgcc9S2rGv4u5sPrgnyEmbJUTxKBlWAh8nIRwy2T5A5J3YXtI+wO/M7JwYZWvLukaJ\nesFKBWH/hIR+mfc1s8sk/Q/RfXcC0XBJ7EYQ0eK4KySdQhSPcjmRMRS7EUS0ftXeZjY3Aa0NkHQ9\n0IvICJoE9CfKSRW7EUQUkLyAKED8RiIvYOwxYJIuC8uD3E3ZQ7+x3+vlzXo0s2lxa29uuBHkpM1S\nYIakZ4g+MI4FZmY8JjF7JhKfMl2L1jUaCutXr7cElrHIBGH3IPpCHBv2T2DDAO04yQ//jwTGmNlX\npWZrxaotKR84Dvirma2VlJQbvicwOMTn/MCPQyOdEtAeSORxm2VmQ0J2+PsT0AXoYGYnSDrWzEYF\nb9wLCehmDK3CBLTK49Ksxw2J0jO8SbSOmZOFG0FO2nwQtgzPhP+JeEySnjJdSzwimSGKh4AWYf9L\n4Ndm9nYcemGmEpIGEyVlXBv2/0YUoJwEz0paQDQcdo6kbYnW9EqCvwOLgdnANEk7EGXnToL+CemU\nxXdhttI6Sc2Az0kuV83a8P+bcL9/ShQcHitm9mz4PypurQracEz2vqR2xDwrbXPFjSAnVbI8Emks\nZZDGlOmkvB6VMRy4yMxeBpDUiyinStwzV1oTGbiZKdJNQ1nsmNkVkm4GVoTg8G+JPI9JaN9FtFYc\nAJL+A/SPQKu5AAAaQElEQVROSPsjST2BXcxsZDD+kgpQL5S0FdG99SawCpiZkPZwST8jykA/gajP\n1yWkTbjOl/PTeKg0vDFLgNTiwmozHhjtpEp2XI6ZJRWXk9FObcp0VhuSGo4qrTvbzPaprCwG3SFE\n2ZMzi2geCtyQ5q/muk6Iy+kCdDSzXSW1Bh43sx4x6wpoa2Yfh/32QDMzmxOnbm1B0mSiYd9LiILi\nTwO+MLPLE9DOjkfKAzoDi1NMDltrcSPISZWQPGwgMMHM9g1lqc5mSYJSw1ECviDG4agy9J8C3gpt\nABhEtPL0cQloZ/JCAcxIKi9UriKpCNgXeCvrPTYniZggSW+a2f5x65TSrC0zMN80s/2zr7Wkf5vZ\noQlon5a1u47IAPIM0mXgw2FO6qS1lIGitZvO56dJ5AYkIJ/WcFSG04lWjX+SyAibRpRZN1aCd+Bw\nYCczu1HS9pK6mVlSQySJo2gZgwNLJ+xLkDVmZplA7JCbKinekNTVzAoS1MzEE3YEuhImPgDHEN3n\nSZGJSVom6Sii2ZhtE9IeD3xvZsUQJUiV1Djkq3KycCPISZs0lzJ4mmgo7lmibNVJ0iRjAAGY2dQk\nv5zM7Guia5009xJd6z5E05ZXAk8QfVnFgqT9KnrezN6KSzvUXyLpdtJbxmBcyAu1laTfEhnAIxLS\n7g38TtJHRPmoYp+ZlhVnOBnYL5OIVNINxLRAcDn8UVJz4GL4/+3de5zcVZnn8c833BIggK7oAALh\nJihMCAEEwVUBYUR3kIsKUWBZnAVHl1HZdV8gXiC7KiviCOx4R4ZhMKKDYRAWDWII14iEBBIgDqLg\nbbyggDGJEOC7f5xTdHWlukHIOSfd9bxfr3p116+6+zmB7qqnzuV5uIDUn69Ya5oe15HebHSW2SeR\nlvtrvckaMyIJCq29i7QvZyvS5r05DO98XNKf8obVFn4s6cMMX47q12JgjZJ05WiPV5gF28f2dEkL\nc7yHc/Jb0rmjPGbqHBueI+koGrQxsP0pSQeTTqPtDHzE9rWVwrc8mbYN0F0R/XEqnA7rsH1V/vRR\nKm2C7zKxe5+h7T8qtW4JPSIJCs3k9gHHuWIPqx7n5U2jcxje3LHozEDWvRwFlZajSLMRPyN11P4+\n6Z15Tavy//fO0szmFJ6Fs137BaifThuDJyWtpHIbg5z01Ep8uv1v28d1X5B0CXDcCF+/Jl1Cqjk2\nm/T7dgRQbQO+Un/A80h/c08BtwLvd2qZU9pySdM7z2WS9iSVhgg9YmN0aErS9bZf1yj2J0hPxvcz\n9ELsRkdYq8gJyMGkZo5TgatJhQNrbch+B3A0qWXHxaRN8R+yXWWZolULhxYkLaNPxeKOGgmYpDts\nT++6vw6w2PYrSsfO8aaTusYD3GB7YY24OfZ8UgPVWfnSMcApHt5ZvlTsvUmNW3+ZL20BHG17bSnR\nsdaIJCg0JeljpD5KlzG8h1Xx2ZhcOG+q6zUR7Y59LfBW24/k+y8Avmb7ryqOYQNSMnQOMNP2BZXi\n7gIcRJoNuc52lT1gGqGFg+23VIp/GNCpSH5913JJ6bgzSYUCLyH9N38HMNl2seJ5kk4HPkjai9LZ\njCvSktQXbZ9eKnbXGD4FXFQrwe8T//u9CY+k+bb3rRR/PdLyp4ClnQKlYbhIgkJTkub2uVxlNkbS\nZaR3Zr8pHatP7IWd48qjXSsUewPgTaQEaArp9MxXbP+icNwJwF2tyh9IWsxQC4fdlVs49FbXLRT7\nbNLm70vzpRnAAtunVYjd78V4tWuFYn+iRsIzQuy/IS0xrwtcRJrxfLRi/LOBR0gzMibNgG5Amh3C\n9u9H/u5QS+wJCk013q/xEmCppB8wfE9QjSPyT0naxvZPAXIbheLvSCRdTKocew1wlu0lpWN25FNS\nd3b/uytr2cLhjcA020/B0/8fFgLFkyDSPqR3MPRiPINKZShaJUA59peBL0vamZQM3SXpZuBL3Scz\nCzo6fzy55/qJpP8PtX73wigiCQqD7KMNY58B3CRpXr7/GuCkCnGPIy07vgz4u676TLU26m4B3C3p\nNoYvf9ZIPFu2cADYjKF2IZtWjPt20gbd80gvvjfna+Ne3oO0S749ROrddqqkk20fUzK27e1K/vyw\nZsRyWAiN5LYd+5ISkFttP9R4SMVJ6lst1/a8ftcLjmMKFVs4SJoBnE1qFyJS0nu67a/ViD+IJH0a\nOIxUM+fC7oKckn5oe+fC8TcknQrcxvZJknYitS4pthesdU2ssSiSoNCUpA1sP/ZM1wrF7j49sz6w\nHrC81rHlQSJpR+AlvaX7Jb0G+IXt+wvG3sX20pFeIGq9MEjagrQvSFRsFyLpIvostdo+sULsZpuT\nJZ1IOmywWpVkSZuW3h+U9xwuILXD2U3SJNKbnWkFY462zDeuT74+V7EcFlq7lXRc+pmurXG2J3ff\nl3Q48MrScQfUZ0inhXqtyI+V3Jx8KmmpsV/RxFrFErH97wy1cKipe+ZhIqlezi9H+No1bSmpm3v1\nzcm2vyLpBbksQndJhBsqjWEH20fnWUBsr5RUtC7XWlITa0yJJCg0odREcytgkqQ9GCratwnQpLKp\n7Ssk1dioOoim9Ft6sn17XpoqxvZJ+eNAvkDYvrz7vqRZwHcrxW62OTmfDnsvqV/XItLS861USnqB\nx/PsT6cw6A50HcAobZBqYj0fkQSFVv4KOIH0BNXd1XkZ/WcM1jhJR3bdnQDsRYUTWjn2Jf0q6fZe\nG0cmjvLYpBoDkPQe4NKe2kwzbH+2Rvy1yE6klhJVNNyc/F7S8uN82wfk+lRnFYzX66PAt4GtJV0K\n7E96zitupJpYQCRBPWJPUGhK0lG971Qrxr6o6+4TwAOkd6jF6wa1rqRbW559+J7tL/VcfydwiO2j\n+3/nGh3Dot79GBVrMzVLevtUjv4VaVN28b+7lpuTJf3A9t6SFpF61j3W73egUGyR3uCtYOjww/xa\nhx9a1sQaa2ImKDQh6Vjb/wxMkXRq7+O2P93n29Yo2zV6dQ3TXUlX0h8YWgZ8HPhi7fFU9D5gdq5X\n0yndvxdpQ/oRlcYwQZKc3/nlxLN089aOXbvv5Nh71gjcu/etsiWktiirbU6m/P67n+eSCFcA10p6\nmEp7oWxb0hW29yS1pqmtZU2sMWVC6wGEgbVR/rgxMLnPrThJn5S0iaT1JF0n6SFJx5aMafsT+UXp\nHNub2J6cb/+hZWG50mz/2vZ+pOWIB/LtLNuvqnVKCvgO8HVJB0k6kNTT6dslA0o6Pc/ETJX0h3xb\nRnpR+teSsbvGcN2zuVbIPwJHSvpIjruNpFcClN6cbPsI24/YPhP4MHAhcHjJmD3mK/XwaqG3JtYd\n1K2JNWbEclgYWJ2pcUlHkJ4c3w/Mtb17hdgizYC8mrRUcaPtK0rHHWRKbTtOZqhv2RzSEkHx6slq\n0D5C0kTSIYO5pP0h3YcPrrH98gpj+BypOfGBtl+e92HNsV0sOcj/7ncBOwKLSctwT5SKN8o47iH1\n7nqAVBi0U5B0auVxTKFiTayxJpbDQhOSzh/tcdt/V2EY6+WPbyQd3f194ROs3f6B9CTd6TD9LkkH\n235PrQEMmtyy4nP5Vjv26ZK2Aral63nX9g0Fw55MWobckjQb0Pnl/gO5f1UF+9ieLmkhgO2HJZVe\ngrwYWAXcSNoQ/ArSJunaDm0QE3i6/tZq1wr/vo1JkQSFVjr7QvYnPUldlu+/teux0r6l1El+JfBu\nSZsDf6oU+7XAbl37Uy4mvWsNhUjaHziToUSk8868+F4JpWaaxwD3MNS3y0CxFyXb5wHnSTrF9gWl\n4jyDVXn/U+f3fHPSzFBJr7D9lznehVReBlpLZqI+0PX5RNL+qwXUKw8wZkQSFJqwfTGApBOAA2yv\nyvc/T1qmqDGG0yT9H+APtp+UtBx4c43YwA9Jx5QfzPe3BmK6uqwLSUueC6jUQLTLEaSWCTXrxOwN\n/KyTAEk6HjiK9Dt3put0MT8fmA28WNLHgLcAHyocc1XnE9tPVJzd7Wg+E9V7CkzS1sAna45hrIgk\nKLS2JWkjdOcJeeN8rZaXk06odf8tFKulIelbpHfFmwL3KjUSNbAPcEupuAGAR21f0yj2j0nLr9WS\nIOALwOvh6eWRs4FTgGmkk4hvKT0A25dKWsDQPqzDbd9bOOzu+eQlOWb3SUy7fFucpjNRI/g5sFvr\nQayNIgkKrZ0NLNRQz5vXkpYsipN0CbADqZps9xJFyYJinyr4s8Po5ko6B/gmXclIpd5hK4BF+VRW\nd+ySe9/W6ZrtORr4Yq4NdHmunVPLfaR9SOtCOiFm+6elgtlep9TPfpZaz0Qh6QKGakNNICW+d1Yf\nyBgQp8NCc0otNPbJd2s2lryX9K6tyR+BpG2BnWx/N5fXX9f2shZjGQTq31zSrtBUUtJ/7ne9syxc\nKOYSYFp+IV4KnNTZGCtpie3iMwOSTiFVTv416Y1GkxNSNUl6knQaDPJMFCkJrjUT1fv79gTwgHua\nF4ckZoJCU/mo+OuB7W3P7NQR6a4sW9AS4C+Af68QaxhJ/5XU1POFpNmolwKfJy0bhALcsHdYyWRn\nFLOAeZIeIm3+vxFA0o5AlSampL0wO9v+XaV4za0FM1EAm+WN8U+T9N7eayFmgkJjLeqIdMWeS5om\nvo3hSxSHVYi9iHRi4/udtg2SFnf2EoQyJL2JVL25u6nkzApxf0KfvnSlT6ZJ2hfYgvQ3tTxfexmw\ncY1lwPw3dnCLOj2DTD1tefK1Ki1ixpqYCQqttagj0nFmpTj9PGb78c5+gbwxO96RFJRPHm4IHAB8\nmbQxuNam1b26Pp9IKgXxwtJBbc/vc+3fSsft8mPgeklXM/yNRvG2OINI0gzg7cD2kq7semgyMDCz\ncX+OSIJCay3qiABge55SY8HOrNNtrtA8NZsnqdND7GDg3cC3KsUeVPvZnirpLttnSTqXtEm6uD7L\nQZ+RdBPwkRrxG/ppvq1PvT5tg+wW0vL+i4Bzu64vI0pw9BVJUGitRR0RACS9DTgHuJ60afECSR+w\n/S8Vwp8GvJNUTO1k4P+RZidCOSvzxxWStiS9M96uRmBJ3UsTE0gzQy0bm1Zh+ywASZPTXf+x8ZDG\nNdsPSvo5sNz2vNbjGQsiCQpNNaoj0nEGsHdn9ifPQn0XKJ4E5Q7PVwBX2P5t6XgBgKtyU8lzSA0l\nTb3Es/td+ROkflJvqxS7GUm7AZeQl/7yJu3jbd/ddGDjWC78ukLSpqWb1I4HsTE6NJMbWt5V46ju\nCPGHbUTO47mz5ObkfBruo8B/IyV9Ih0dvqDGBt2QSNoAmBgvEmVJugU4w/bcfP91wMdt79d0YOOc\npK8D+wLXMnRcv1ZPxjElZoJCM3k25M7SxdNG8W1J32GoienRQOmKwu8j9Uvb2/ZPACRtD3xO0vtt\n/33h+ANN0n7AFIYK92G7ZHHMTtxNSclvp7HlPGDmACRhG3USIADb10vaqOWABsTV+RaeQcwEhaYk\nfY+0Mfk2hr9jKX5MPcc/Eng1aUbmBtuzC8dbSDoy/FDP9c1Jx5jjCGshI1UIr/HuWNLlpLpUnXpB\nxwG72z6ydOyWJM0mLT1eki8dC+xl+/B2owphSCRBoSlJr+13veSmvlws7iW9FVRzf6Vf2L6/YOwR\nK/XWquI7qFpWCJe0yPa0Z7o23uS6X2fR9UaD1Lz14aYDG6ckfd322yQtpn9dqnFbqfu5iuWw0ERX\nIjKv5/prgF8UDv8Z4IN9rq/Ij/11n8fWlMef42Ph+WtWIRxYKenVtm8CkLQ/Q6fVxq2c7MQ+lHo6\n3er/U9NRjCGRBIVWWiYiU2yvVjPD9u2SphSMC8M7XHcTXVWMQxEvAu6RVL1COPC3wMV5bxDAw8AJ\nFeI20VOobzW1lrsH0BmSvmr7ltYDGSsiCQqttExERks2JpUMvJb0FRpUZ7YKbHsRKQHeJN/vlwiP\nJ68CfkY6dPB9UpIfyrsPOFfSFsBlwKz8uxdGEHuCQhOSfmR7xz/3sTUUexbwPdtf6rn+TuAQ20eX\nih0Gk6SPA5+0/Ui+/wLgv9uuUhi0tlwF/mBgBjCVdFJpVtQHqkPStsAx+TaRlIx+rXLLlDEhkqDQ\nRMtEJLfKmE3ag7MgX96LVNb/CNu/KhU71CfpJtuvlrSM4ZtFRTodtkmFMazWvLJfk8vxKNdkmkEq\nUjnT9gWNhzRQJO0BfAWYGjPRq4skKDSxNiQikg4AOqex7rb9vdIxQ32Strf948ZjuItUG+qxfH8S\ncLvtXVuOq6Sc/LyJlABNAa4EvmK79MGHgSdpPeANpJmgg0h1qWbZvqLpwNZCkQSFpiIRCaVJWmB7\nT0nX2T6o0Rj+J3AYcBFpNupE4Erbn2wxntIkXUz6u76GtAyzpPGQBkJuxjyDlHzeBnyN1Jpn+ajf\nOMAiCQohjGu5QOUVwN8Aq1Xktv3pSuN4A/B60jLcHNvfqRG3BUlPMVT8tMkS5CCSNBf4KnC57d+3\nHs9YEKfDQgjj3THA4aTnuyad2yVtB1xv+9v5/iRJU2w/0GI8pdme0HoMg8j2Aa3HMNbETFAIYSBI\nOtR26d5wI8W+HdjP9uP5/vrAzbb3bjGeEEIS2XoIYSC0SoCydTsJUB7L46RDACGEhiIJCiGE8n4r\n6ekqyZLeDDw0yteHECqI5bAQwrgnaQKwb6t2ApJ2AC4FtiRtDv4ZcLztH7UYTwghiSQohDAQJN1q\n+1WNx7Ax6Xl3WctxhBCSOB0WQhgUcyQdBXzTDd79SXoTsCswUUqttGzPrD2OEMKQSIJCCIPiVGAj\n4ElJK6nbNuPzwIbAAcCXgbeQitmFEBqK5bAQQihM0l22p3Z93Jg0I3VI67GFMMjidFgIYSAoOVbS\nh/P9rSW9slL4lfnjCklbAquA7SrFDiGMIJKgEMKg+CzwKuDt+f4fgX+oFPsqSZuROqnfATwAzKoU\nO4QwglgOCyEMBEl32J4uaaHtPfK1O23vXnkcGwATbT9aM24IYXWxMTqEMChWSVqH3NBT0ubAU7UH\nYfsx4LHacUMIq4vlsBDCoDgfmA28WNLHgJuAj7cdUgihpVgOCyEMDEm7AAeRjsdfZ/vexkMKITQU\nM0EhhEFyH2k26EpguaRtagSVNLPn/jqSLq0RO4QwskiCQggDQdIpwK+Ba4GrgKvzxxq2kXR6HscG\npETsvkqxQwgjiOWwEMJAkPQjYB/bv2sQW6QGqotJVaOvsf33tccRQhgukqAQwkCQNBc42PYTFWNO\n77q7HvAF4GbgQgDbd9QaSwhhdZEEhRDGNUmn5k93BXYmLYM9fUTd9qcLxp47ysO2fWCp2CGEZxZ1\ngkII493k/PGn+bZ+vkGuGVSK7QNK/vwQwvMTM0EhhIEg6a22v/FM1wrF3gA4CphC15tP2zNH+p4Q\nQnlxOiyEMChOf5bXSvhX4M3AE8DyrlsIoaFYDgshjGuSDgXeCGwl6fyuhzYhJSU1vNT2GyrFCiE8\nS5EEhRDGu18CC4DD8seOZcD7K43hFkl/aXtxpXghhGch9gSFEAaCpI1Je3IM3G/7TxVj3wPsCPyE\ndDJNpNNhU2uNIYSwukiCQgjjmqR1SY1S/wvpdNgE4KXARcAZtldVGMO2/a7bfrB07BDCyGJjdAhh\nvDsHeCGwve09be8B7ABsBnyqxgBsP5gTnpWkmajOLYTQUMwEhRDGNUn3AS9zz5OdpHWApbZ3qjCG\nw4BzgS2B3wDbAvfa3rV07BDCyGImKIQw3rk3AcoXn6TebMz/AvYF/s32dsBBpPYZIYSGIgkKIYx3\n90g6vveipGOBpZXGsCo3bp0gaYLtucC0SrFDCCOII/IhhPHuPcA3JZ1IOiJvYG9gEnBEpTE8kk+n\n3QBcKuk31KtRFEIYQewJCiEMBEkHkpqoCrjb9nUVY29E2hQ9AXgHsClwaZ4dCiE0EklQCCFUJOlF\nwO/67VMKIdQVe4JCCKEQSftKul7SNyXtIWkJsAT4taRooxFCYzETFEIIhUi6Hfggafnri8ChtudL\n2gWYlWsWhRAaiZmgEEIoZ13bc2x/A/iV7fkAtmudSgshjCKSoBBCKOeprs9X9jwW0/AhNBbLYSGE\nUIikJ4HlpBNpk4AVnYeAibbXazW2EEIkQSGEEEIYULEcFkIIIYSBFElQCCGEEAZSJEEhhBBCGEiR\nBIUQxixJT0paJGmJpG9I2vB5/KzXSboqf36YpNNG+drNJL37OcQ4U9L/eK5jDCGsWZEEhRDGspW2\np9neDXgceFf3g0r+7Oc521faPnuUL9kM+LOToBDC2iWSoBDCeHEjsKOkKZLulfRZ4A5ga0mHSLpV\n0h15xmhjAElvkLRU0k3AkZ0fJOkESf83f/4SSbMl3Zlv+wFnAzvkWahz8td9QNIPJN0l6ayun3WG\npB9K+i6wc7X/GiGEZxRJUAhhzJO0LnAosDhf2hn4p9yWYjnwIeD1tqcDtwOnSpoIfAn4a+A/An8x\nwo8/H5hne3dgOnA3cBpwf56F+oCkQ4CdgFcC04A9Jb1G0p7AMcAepCRr7zX8Tw8hPA/rth5ACCE8\nD5MkLcqf3whcCGwJPNhpUQHsC7wCuFkSwPrArcAuwE9s3wcg6Z+Bk/rEOBA4HsD2k8Cjkl7Q8zWH\n5NvCfH9jUlI0GZhte0WOceXz+teGENaoSIJCCGPZStvTui/kRGd59yXgWtszer5uGmuudYWAT9j+\nQk+M963BGCGENSyWw0II4918YH9JOwJI2lDSy4ClwHaSdshfN2OE778O+Nv8vetI2gRYRprl6fgO\ncGLXXqOtJL0YuAE4QtIkSZNJS28hhLVEJEEhhHHN9m+BE4BZku4iJUW72P4Tafnr6rwx+sERfsR7\ngQMkLQYWALva/h1peW2JpHNszwG+Ctyav+5fgMm27wAuAxYBl5OW7EIIa4noHRZCCCGEgRQzQSGE\nEEIYSJEEhRBCCGEgRRIUQgghhIEUSVAIIYQQBlIkQSGEEEIYSJEEhRBCCGEgRRIUQgghhIEUSVAI\nIYQQBtL/BxhSpB+n0/QiAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f72cbc37940>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.metrics import confusion_matrix\n",
    "\n",
    "conf_mat = confusion_matrix(y_test, y_pred)\n",
    "fig, ax = plt.subplots(figsize=(8,6))\n",
    "sns.heatmap(conf_mat, annot=True, fmt='d',\n",
    "            xticklabels=category_id_df.Product.values, yticklabels=category_id_df.Product.values)\n",
    "plt.ylabel('Actual')\n",
    "plt.xlabel('Predicted')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "'Consumer Loan' predicted as 'Credit reporting' : 10 examples.\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Product</th>\n",
       "      <th>Consumer_complaint_narrative</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2720</th>\n",
       "      <td>Consumer Loan</td>\n",
       "      <td>Quoting them, your first loan application, the...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7091</th>\n",
       "      <td>Consumer Loan</td>\n",
       "      <td>While reviewing my XXXX credit report, I notic...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5439</th>\n",
       "      <td>Consumer Loan</td>\n",
       "      <td>I have been recently checking my credit report...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12763</th>\n",
       "      <td>Consumer Loan</td>\n",
       "      <td>We went to buy XXXX cars, and the dealership s...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13158</th>\n",
       "      <td>Consumer Loan</td>\n",
       "      <td>I got a 30 day late XX/XX/2017 and it 's repor...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4134</th>\n",
       "      <td>Consumer Loan</td>\n",
       "      <td>I took out an instalment loan in the amount XX...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13848</th>\n",
       "      <td>Consumer Loan</td>\n",
       "      <td>I was turned down for a loan by Honda Finacial...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19227</th>\n",
       "      <td>Consumer Loan</td>\n",
       "      <td>ONEMAIN # XXXX XXXX , IN XXXX ( XXXX ) XXXX Da...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11258</th>\n",
       "      <td>Consumer Loan</td>\n",
       "      <td>I have not been given credit for the payments ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11242</th>\n",
       "      <td>Consumer Loan</td>\n",
       "      <td>Reliable Credit falsely submitted an applicati...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             Product                       Consumer_complaint_narrative\n",
       "2720   Consumer Loan  Quoting them, your first loan application, the...\n",
       "7091   Consumer Loan  While reviewing my XXXX credit report, I notic...\n",
       "5439   Consumer Loan  I have been recently checking my credit report...\n",
       "12763  Consumer Loan  We went to buy XXXX cars, and the dealership s...\n",
       "13158  Consumer Loan  I got a 30 day late XX/XX/2017 and it 's repor...\n",
       "4134   Consumer Loan  I took out an instalment loan in the amount XX...\n",
       "13848  Consumer Loan  I was turned down for a loan by Honda Finacial...\n",
       "19227  Consumer Loan  ONEMAIN # XXXX XXXX , IN XXXX ( XXXX ) XXXX Da...\n",
       "11258  Consumer Loan  I have not been given credit for the payments ...\n",
       "11242  Consumer Loan  Reliable Credit falsely submitted an applicati..."
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "'Debt collection' predicted as 'Credit reporting' : 18 examples.\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Product</th>\n",
       "      <th>Consumer_complaint_narrative</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>18410</th>\n",
       "      <td>Debt collection</td>\n",
       "      <td>Dear CFPB, I am asking you for assistance to i...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5262</th>\n",
       "      <td>Debt collection</td>\n",
       "      <td>XXXX XXXX, XXXX ( This letter describes in det...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11834</th>\n",
       "      <td>Debt collection</td>\n",
       "      <td>XXXX XXXX XXXX is reporting negatively on my c...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19652</th>\n",
       "      <td>Debt collection</td>\n",
       "      <td>I recently paid of both debts on my credit acc...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15557</th>\n",
       "      <td>Debt collection</td>\n",
       "      <td>Never have been a XXXX XXXX customer. I was at...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4431</th>\n",
       "      <td>Debt collection</td>\n",
       "      <td>someone tried getting credit information and i...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15949</th>\n",
       "      <td>Debt collection</td>\n",
       "      <td>This debt is from account from $ XX/XX/2008 an...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12475</th>\n",
       "      <td>Debt collection</td>\n",
       "      <td>In XXXX XXXX, there was an account opened thro...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13548</th>\n",
       "      <td>Debt collection</td>\n",
       "      <td>DIVERSIFIELD CONSULTANTS INC HAVE VIOLATED FCR...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6988</th>\n",
       "      <td>Debt collection</td>\n",
       "      <td>Also collections refuses to stop reporting to ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16498</th>\n",
       "      <td>Debt collection</td>\n",
       "      <td>They called my son and told him that they are ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12028</th>\n",
       "      <td>Debt collection</td>\n",
       "      <td>Rubin &amp; Rothman LLC ( R &amp; R ) received default...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7131</th>\n",
       "      <td>Debt collection</td>\n",
       "      <td>THIS IS FRAUD. I HAVE REQUESTED VERIFICATION A...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15630</th>\n",
       "      <td>Debt collection</td>\n",
       "      <td>Barclays Bank Delaware obtained a judgment aga...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11112</th>\n",
       "      <td>Debt collection</td>\n",
       "      <td>This account was a joint account with XXXX and...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>Debt collection</td>\n",
       "      <td>This complaint is in regards to Square Two Fin...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>311</th>\n",
       "      <td>Debt collection</td>\n",
       "      <td>Hunter Warfield has be unable to provide prope...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15988</th>\n",
       "      <td>Debt collection</td>\n",
       "      <td>Unknown account, never have been notified and ...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               Product                       Consumer_complaint_narrative\n",
       "18410  Debt collection  Dear CFPB, I am asking you for assistance to i...\n",
       "5262   Debt collection  XXXX XXXX, XXXX ( This letter describes in det...\n",
       "11834  Debt collection  XXXX XXXX XXXX is reporting negatively on my c...\n",
       "19652  Debt collection  I recently paid of both debts on my credit acc...\n",
       "15557  Debt collection  Never have been a XXXX XXXX customer. I was at...\n",
       "4431   Debt collection  someone tried getting credit information and i...\n",
       "15949  Debt collection  This debt is from account from $ XX/XX/2008 an...\n",
       "12475  Debt collection  In XXXX XXXX, there was an account opened thro...\n",
       "13548  Debt collection  DIVERSIFIELD CONSULTANTS INC HAVE VIOLATED FCR...\n",
       "6988   Debt collection  Also collections refuses to stop reporting to ...\n",
       "16498  Debt collection  They called my son and told him that they are ...\n",
       "12028  Debt collection  Rubin & Rothman LLC ( R & R ) received default...\n",
       "7131   Debt collection  THIS IS FRAUD. I HAVE REQUESTED VERIFICATION A...\n",
       "15630  Debt collection  Barclays Bank Delaware obtained a judgment aga...\n",
       "11112  Debt collection  This account was a joint account with XXXX and...\n",
       "16     Debt collection  This complaint is in regards to Square Two Fin...\n",
       "311    Debt collection  Hunter Warfield has be unable to provide prope...\n",
       "15988  Debt collection  Unknown account, never have been notified and ..."
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "'Mortgage' predicted as 'Credit reporting' : 6 examples.\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Product</th>\n",
       "      <th>Consumer_complaint_narrative</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>4637</th>\n",
       "      <td>Mortgage</td>\n",
       "      <td>This complaint is in follow-up to Complaint # ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5269</th>\n",
       "      <td>Mortgage</td>\n",
       "      <td>The attached complaint was initially written t...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7343</th>\n",
       "      <td>Mortgage</td>\n",
       "      <td>In 2014, I went to XXXX in order to buy a mobi...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15048</th>\n",
       "      <td>Mortgage</td>\n",
       "      <td>Company repeatedly corrects my credit report a...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>861</th>\n",
       "      <td>Mortgage</td>\n",
       "      <td>Mortgage broker did Credit inquiry on my credi...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19781</th>\n",
       "      <td>Mortgage</td>\n",
       "      <td>I am a card carrying XXXX and wanted to see if...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        Product                       Consumer_complaint_narrative\n",
       "4637   Mortgage  This complaint is in follow-up to Complaint # ...\n",
       "5269   Mortgage  The attached complaint was initially written t...\n",
       "7343   Mortgage  In 2014, I went to XXXX in order to buy a mobi...\n",
       "15048  Mortgage  Company repeatedly corrects my credit report a...\n",
       "861    Mortgage  Mortgage broker did Credit inquiry on my credi...\n",
       "19781  Mortgage  I am a card carrying XXXX and wanted to see if..."
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "'Credit card' predicted as 'Credit reporting' : 9 examples.\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Product</th>\n",
       "      <th>Consumer_complaint_narrative</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>18643</th>\n",
       "      <td>Credit card</td>\n",
       "      <td>I was told this account wiuld be deleted from ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18574</th>\n",
       "      <td>Credit card</td>\n",
       "      <td>This inquiry was n't me</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19868</th>\n",
       "      <td>Credit card</td>\n",
       "      <td>Capital One/Kohls has been reporting a past du...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19963</th>\n",
       "      <td>Credit card</td>\n",
       "      <td>on XX/XX/XXXX my wallet was stolen with all my...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4706</th>\n",
       "      <td>Credit card</td>\n",
       "      <td>American Express is reporting an account on my...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21566</th>\n",
       "      <td>Credit card</td>\n",
       "      <td>Have disputed the reporting of the status of a...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13906</th>\n",
       "      <td>Credit card</td>\n",
       "      <td>I have been the victim of identity theft fraud...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16853</th>\n",
       "      <td>Credit card</td>\n",
       "      <td>I have requested XXXX XXXX to run a credit rep...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10505</th>\n",
       "      <td>Credit card</td>\n",
       "      <td>I have been working since XXXX 2016 to get a i...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           Product                       Consumer_complaint_narrative\n",
       "18643  Credit card  I was told this account wiuld be deleted from ...\n",
       "18574  Credit card                            This inquiry was n't me\n",
       "19868  Credit card  Capital One/Kohls has been reporting a past du...\n",
       "19963  Credit card  on XX/XX/XXXX my wallet was stolen with all my...\n",
       "4706   Credit card  American Express is reporting an account on my...\n",
       "21566  Credit card  Have disputed the reporting of the status of a...\n",
       "13906  Credit card  I have been the victim of identity theft fraud...\n",
       "16853  Credit card  I have requested XXXX XXXX to run a credit rep...\n",
       "10505  Credit card  I have been working since XXXX 2016 to get a i..."
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    },
    {
     "ename": "IndexError",
     "evalue": "index 11 is out of bounds for axis 0 with size 11",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mIndexError\u001b[0m                                Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-22-9932ab8bdc5b>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[1;32m      3\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0mpredicted\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mcategory_id_df\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcategory_id\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      4\u001b[0m   \u001b[0;32mfor\u001b[0m \u001b[0mactual\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mcategory_id_df\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcategory_id\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 5\u001b[0;31m     \u001b[0;32mif\u001b[0m \u001b[0mpredicted\u001b[0m \u001b[0;34m!=\u001b[0m \u001b[0mactual\u001b[0m \u001b[0;32mand\u001b[0m \u001b[0mconf_mat\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mactual\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mpredicted\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m>=\u001b[0m \u001b[0;36m6\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m      6\u001b[0m       \u001b[0mprint\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"'{}' predicted as '{}' : {} examples.\"\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mformat\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mid_to_category\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mactual\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mid_to_category\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mpredicted\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mconf_mat\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mactual\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mpredicted\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      7\u001b[0m       \u001b[0mdisplay\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdf\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mloc\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mindices_test\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0my_test\u001b[0m \u001b[0;34m==\u001b[0m \u001b[0mactual\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m&\u001b[0m \u001b[0;34m(\u001b[0m\u001b[0my_pred\u001b[0m \u001b[0;34m==\u001b[0m \u001b[0mpredicted\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'Product'\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m'Consumer_complaint_narrative'\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;31mIndexError\u001b[0m: index 11 is out of bounds for axis 0 with size 11"
     ]
    }
   ],
   "source": [
    "from IPython.display import display\n",
    "\n",
    "for predicted in category_id_df.category_id:\n",
    "  for actual in category_id_df.category_id:\n",
    "    if predicted != actual and conf_mat[actual, predicted] >= 6:\n",
    "      print(\"'{}' predicted as '{}' : {} examples.\".format(id_to_category[actual], id_to_category[predicted], conf_mat[actual, predicted]))\n",
    "      display(df.loc[indices_test[(y_test == actual) & (y_pred == predicted)]][['Product', 'Consumer_complaint_narrative']])\n",
    "      print('')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "LinearSVC(C=1.0, class_weight=None, dual=True, fit_intercept=True,\n",
       "     intercept_scaling=1, loss='squared_hinge', max_iter=1000,\n",
       "     multi_class='ovr', penalty='l2', random_state=None, tol=0.0001,\n",
       "     verbose=0)"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.fit(features, labels)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "# 'Bank account or service':\n",
      "  . Top unigrams:\n",
      "       . bank\n",
      "       . account\n",
      "  . Top bigrams:\n",
      "       . debit card\n",
      "       . overdraft fees\n",
      "# 'Consumer Loan':\n",
      "  . Top unigrams:\n",
      "       . vehicle\n",
      "       . car\n",
      "  . Top bigrams:\n",
      "       . personal loan\n",
      "       . history xxxx\n",
      "# 'Credit card':\n",
      "  . Top unigrams:\n",
      "       . card\n",
      "       . discover\n",
      "  . Top bigrams:\n",
      "       . credit card\n",
      "       . discover card\n",
      "# 'Credit reporting':\n",
      "  . Top unigrams:\n",
      "       . equifax\n",
      "       . transunion\n",
      "  . Top bigrams:\n",
      "       . xxxx account\n",
      "       . trans union\n",
      "# 'Debt collection':\n",
      "  . Top unigrams:\n",
      "       . debt\n",
      "       . collection\n",
      "  . Top bigrams:\n",
      "       . account credit\n",
      "       . time provided\n",
      "# 'Money transfers':\n",
      "  . Top unigrams:\n",
      "       . paypal\n",
      "       . transfer\n",
      "  . Top bigrams:\n",
      "       . money transfer\n",
      "       . send money\n",
      "# 'Mortgage':\n",
      "  . Top unigrams:\n",
      "       . mortgage\n",
      "       . escrow\n",
      "  . Top bigrams:\n",
      "       . loan modification\n",
      "       . mortgage company\n",
      "# 'Other financial service':\n",
      "  . Top unigrams:\n",
      "       . passport\n",
      "       . dental\n",
      "  . Top bigrams:\n",
      "       . stated pay\n",
      "       . help pay\n",
      "# 'Payday loan':\n",
      "  . Top unigrams:\n",
      "       . payday\n",
      "       . loan\n",
      "  . Top bigrams:\n",
      "       . payday loan\n",
      "       . pay day\n",
      "# 'Prepaid card':\n",
      "  . Top unigrams:\n",
      "       . prepaid\n",
      "       . serve\n",
      "  . Top bigrams:\n",
      "       . prepaid card\n",
      "       . use card\n",
      "# 'Student loan':\n",
      "  . Top unigrams:\n",
      "       . navient\n",
      "       . loans\n",
      "  . Top bigrams:\n",
      "       . student loan\n",
      "       . sallie mae\n",
      "# 'Virtual currency':\n",
      "  . Top unigrams:\n",
      "       . https\n",
      "       . tx\n",
      "  . Top bigrams:\n",
      "       . money want\n",
      "       . xxxx provider\n"
     ]
    }
   ],
   "source": [
    "from sklearn.feature_selection import chi2\n",
    "\n",
    "N = 2\n",
    "for Product, category_id in sorted(category_to_id.items()):\n",
    "  indices = np.argsort(model.coef_[category_id])\n",
    "  feature_names = np.array(tfidf.get_feature_names())[indices]\n",
    "  unigrams = [v for v in reversed(feature_names) if len(v.split(' ')) == 1][:N]\n",
    "  bigrams = [v for v in reversed(feature_names) if len(v.split(' ')) == 2][:N]\n",
    "  print(\"# '{}':\".format(Product))\n",
    "  print(\"  . Top unigrams:\\n       . {}\".format('\\n       . '.join(unigrams)))\n",
    "  print(\"  . Top bigrams:\\n       . {}\".format('\\n       . '.join(bigrams)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\"I requested a home loan modification through Bank of America. Bank of America never got back to me.\"\n",
      "  - Predicted as: 'Mortgage'\n",
      "\n",
      "\"It has been difficult for me to find my past due balance. I missed a regular monthly payment\"\n",
      "  - Predicted as: 'Credit reporting'\n",
      "\n",
      "\"I can't get the money out of the country.\"\n",
      "  - Predicted as: 'Bank account or service'\n",
      "\n",
      "\"I have no money to pay my tuition\"\n",
      "  - Predicted as: 'Debt collection'\n",
      "\n",
      "\"Coinbase closed my account for no reason and furthermore refused to give me a reason despite dozens of request\"\n",
      "  - Predicted as: 'Bank account or service'\n",
      "\n"
     ]
    }
   ],
   "source": [
    "texts = [\"I requested a home loan modification through Bank of America. Bank of America never got back to me.\",\n",
    "         \"It has been difficult for me to find my past due balance. I missed a regular monthly payment\",\n",
    "         \"I can't get the money out of the country.\",\n",
    "         \"I have no money to pay my tuition\",\n",
    "         \"Coinbase closed my account for no reason and furthermore refused to give me a reason despite dozens of request\"]\n",
    "text_features = tfidf.transform(texts)\n",
    "predictions = model.predict(text_features)\n",
    "for text, predicted in zip(texts, predictions):\n",
    "  print('\"{}\"'.format(text))\n",
    "  print(\"  - Predicted as: '{}'\".format(id_to_category[predicted]))\n",
    "  print(\"\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                         precision    recall  f1-score   support\n",
      "\n",
      "       Credit reporting       0.82      0.82      0.82       288\n",
      "          Consumer Loan       0.83      0.60      0.70       100\n",
      "        Debt collection       0.80      0.91      0.85       359\n",
      "               Mortgage       0.90      0.93      0.92       317\n",
      "            Credit card       0.73      0.77      0.75       165\n",
      "Other financial service       0.00      0.00      0.00         1\n",
      "Bank account or service       0.74      0.74      0.74       121\n",
      "           Student loan       0.92      0.83      0.87       111\n",
      "        Money transfers       0.50      0.23      0.32        13\n",
      "            Payday loan       0.75      0.38      0.50        16\n",
      "           Prepaid card       0.67      0.12      0.20        17\n",
      "\n",
      "            avg / total       0.82      0.82      0.81      1508\n",
      "\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ec2-user/anaconda3/envs/python3/lib/python3.6/site-packages/sklearn/metrics/classification.py:1428: UserWarning: labels size, 11, does not match size of target_names, 12\n",
      "  .format(len(labels), len(target_names))\n",
      "/home/ec2-user/anaconda3/envs/python3/lib/python3.6/site-packages/sklearn/metrics/classification.py:1135: UndefinedMetricWarning: Precision and F-score are ill-defined and being set to 0.0 in labels with no predicted samples.\n",
      "  'precision', 'predicted', average, warn_for)\n"
     ]
    }
   ],
   "source": [
    "from sklearn import metrics\n",
    "print(metrics.classification_report(y_test, y_pred, \n",
    "                                    target_names=df['Product'].unique()))"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "conda_python3",
   "language": "python",
   "name": "conda_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": "3.6.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}