PreProcessing.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# <center> RDS Project Draft Notebook</center>\n",
    "    \n",
    "### <center>By Ankush Jain (aj2885) and Vishnu Thakral (vvt223)</center>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Importing all relevant libraries\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import json\n",
    "import csv\n",
    "from bs4 import BeautifulSoup\n",
    "import matplotlib.pyplot as plt\n",
    "import matplotlib as mpl"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Loading data\n",
    "# identity_individual_annotations = pd.read_csv('data/identity_individual_annotations.csv')\n",
    "# sample_submission = pd.read_csv('data/sample_submission.csv')\n",
    "#test_file = pd.read_csv('data/test.csv')\n",
    "train_data_copy = pd.read_csv('data/train.csv')\n",
    "train_data = train_data_copy.copy()\n",
    "# test_private_expanded = pd.read_csv('data/test_private_expanded.csv')\n",
    "# test_public_expanded = pd.read_csv('data/test_public_expanded.csv')\n",
    "# toxicity_individual_annotations = pd.read_csv('data/toxicity_individual_annotations.csv')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Data Pre-Processing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1804874, 45)"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Getting the shape of the train dataset\n",
    "train_data.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1804874, 39)"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Removing irrelevant columns\n",
    "train_data = train_data.drop(['severe_toxicity','obscene','identity_attack', 'insult', 'threat','sexual_explicit'],axis=1)\n",
    "train_data.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>id</th>\n",
       "      <th>target</th>\n",
       "      <th>comment_text</th>\n",
       "      <th>asian</th>\n",
       "      <th>atheist</th>\n",
       "      <th>bisexual</th>\n",
       "      <th>black</th>\n",
       "      <th>buddhist</th>\n",
       "      <th>christian</th>\n",
       "      <th>female</th>\n",
       "      <th>...</th>\n",
       "      <th>parent_id</th>\n",
       "      <th>article_id</th>\n",
       "      <th>rating</th>\n",
       "      <th>funny</th>\n",
       "      <th>wow</th>\n",
       "      <th>sad</th>\n",
       "      <th>likes</th>\n",
       "      <th>disagree</th>\n",
       "      <th>identity_annotator_count</th>\n",
       "      <th>toxicity_annotator_count</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>59848</td>\n",
       "      <td>0.0</td>\n",
       "      <td>This is so cool. It's like, 'would you want yo...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2006</td>\n",
       "      <td>rejected</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>59849</td>\n",
       "      <td>0.0</td>\n",
       "      <td>Thank you!! This would make my life a lot less...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2006</td>\n",
       "      <td>rejected</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>2 rows × 39 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      id  target                                       comment_text  asian  \\\n",
       "0  59848     0.0  This is so cool. It's like, 'would you want yo...    NaN   \n",
       "1  59849     0.0  Thank you!! This would make my life a lot less...    NaN   \n",
       "\n",
       "   atheist  bisexual  black  buddhist  christian  female  ...  parent_id  \\\n",
       "0      NaN       NaN    NaN       NaN        NaN     NaN  ...        NaN   \n",
       "1      NaN       NaN    NaN       NaN        NaN     NaN  ...        NaN   \n",
       "\n",
       "   article_id    rating  funny  wow  sad  likes  disagree  \\\n",
       "0        2006  rejected      0    0    0      0         0   \n",
       "1        2006  rejected      0    0    0      0         0   \n",
       "\n",
       "   identity_annotator_count  toxicity_annotator_count  \n",
       "0                         0                         4  \n",
       "1                         0                         4  \n",
       "\n",
       "[2 rows x 39 columns]"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_data.head(2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Data Profiling"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "dtype('int64')"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#replacing the categorical value into numberic\n",
    "train_data.replace({\"rating\":{\"rejected\": 0, \"approved\": 1}}, inplace=True)\n",
    "train_data.rating.dtype"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'🤣gotta love it!'"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#laoding the comment with max size\n",
    "train_data.comment_text.max()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "#store the length of each comment\n",
    "lens = []\n",
    "for i in list(train_data['comment_text']):\n",
    "    lens.append(len(i))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'k'"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#find the smallest sized comment (as size xero is not considered as a comment)\n",
    "train_data.loc[lens.index(1),'comment_text']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The possible key for the dataset are: ['id']\n"
     ]
    }
   ],
   "source": [
    "#extracting the unique columns that can be used to map to the database\n",
    "a =list(train_data.columns)\n",
    "unique = []\n",
    "for i in a:\n",
    "    if train_data[i].is_unique:\n",
    "        unique.append(i)\n",
    "print('The possible key for the dataset are:',unique)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Store information like mean, median, max frequency, count, maximum occuring words/values in the attribute \n",
    "with open('numeric.csv', 'w') as f, open('non_numeric.csv', 'w') as c:\n",
    "    #f is for numeric and c for categorical/textual\n",
    "    f.write(\"attribute,count,unique,max,min,mean,median,std_dev,max_f1,f1_count,max_f2,f2_count,max_f3,f3_count,max_f4,f4_count,max_f5,f5_count\\n\")\n",
    "    c.write(\"Attribute,Count,Unique,Max,Min,Max_frequency\\n\")\n",
    "    for i in a:\n",
    "        if train_data[i].dtype !='O':\n",
    "            #dict_col[i]\n",
    "            f.write(\"%s,\"%(i))\n",
    "            f.write(\"%f,\"%(train_data[i].isnull().sum())) # all null values \n",
    "            f.write(\"%f,\"%(train_data[i].nunique())) #unique values\n",
    "            f.write(\"%f,\"%(train_data[i].max())) # max in the attribute\n",
    "            f.write(\"%f,\"%(train_data[i].min())) #min in the attribute\n",
    "            f.write(\"%f,\"%(train_data[i].mean())) #mean of the values that belong to the attribute\n",
    "            f.write(\"%f,\"%(train_data[i].median())) # median of the attribute\n",
    "            f.write(\"%f,\"%(train_data[i].std())) #std devition of values\n",
    "            max_values = train_data[i].value_counts(ascending=False) # taking the highest 5 occuring values\n",
    "            max_values = max_values.reset_index().head(5) \n",
    "            for j in range(max_values.shape[0]):\n",
    "                f.write(\"%f,\"%(max_values.loc[j,'index']))\n",
    "                f.write(\"%f,\"%(max_values.loc[j,i]))\n",
    "            f.write(\"\\n\")\n",
    "        else:\n",
    "            c.write(\"%s,\"%(i))\n",
    "            c.write(\"%f,\"%(train_data[i].isnull().sum()))\n",
    "            c.write(\"%f,\"%(train_data[i].nunique()))\n",
    "            c.write(\"\\n\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "male 44032.25796984777 (405130, 12)\n",
      "female 51723.05737755975 (405130, 12)\n",
      "homosexual_gay_or_lesbian 10375.613491363378 (405130, 12)\n",
      "christian 38595.950841561535 (405130, 12)\n",
      "jewish 7236.672289150351 (405130, 12)\n",
      "muslim 20037.556176699083 (405130, 12)\n",
      "black 13933.484259646462 (405130, 12)\n",
      "white 23072.31180173668 (405130, 12)\n",
      "psychiatric_or_mental_illness 4895.233197204576 (405130, 12)\n"
     ]
    }
   ],
   "source": [
    "#removing values that are NaN from the social group attributes\n",
    "df_test = train_data.copy()\n",
    "#dropping attributes that are neither label/social_group or label\n",
    "#df_test = df_test.drop(['created_date','publication_id','parent_id','article_id', 'rating', 'funny','wow','sad', 'likes', 'disagree', 'identity_annotator_count','toxicity_annotator_count'],axis=1)\n",
    "columns = ['id', 'comment_text', 'target', 'male', 'female', 'homosexual_gay_or_lesbian', 'christian', 'jewish', 'muslim', 'black', 'white',  'psychiatric_or_mental_illness']\n",
    "df_test = df_test[columns]\n",
    "col_list = list(df_test.columns)\n",
    "col_list =col_list[3:]\n",
    "for col in col_list:\n",
    "    df_test = df_test[df_test[col].notnull()]\n",
    "    #df_test = df_test[df_test['atheist'].notnull()]\n",
    "    print(col,df_test[col].sum(),df_test.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Input Attribute Profiling\n",
    "The input attribute **comment_text** has zero empty values and it ranges from 1 to 8813 in length. The high range of input text and high variability of length must be handled well by the feature extraction technique and the machine learning model.\n",
    "\n",
    "#### Output Attribute Profiling\n",
    "The output attribute/label **toxicity** has zero empty value, two unique values (0:False or 1:True) and a True:False ratio of 0.08:0.92. Hence, 92\\% of the training data is Non-Toxic (toxicity is false) and 8\\% data is Toxic (toxicity is true).\n",
    "\n",
    "#### Social Group Attribute Profiling (Auxiliary Attribs)\n",
    "Social Group Attributes or the annotations will be used in validating the fairness of the model predictions. All the above metrics have been calculated for 405130 annotated data entries. The value for each social group attribute in the data set denotes the probability of the comment being toxic and offensive to the particular social group. The probability will range from 0.0 (not toxic) to 1.0 (extremely toxic)."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Distributions\n",
    "To study the distribution of data across the various ranges of probability, we have divided the data into 5 buckets ranging from 0.0 to 1.0 at an interval of 0.2. We plot the frequency of data samples that fall in each probability range or bucket to observe the distribution of probabilities across the various social groups."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "#creating bins for the social group attributes so that we can make analysis amongst them and with the label \n",
    "def create_binned(df,i):\n",
    "    bins = [0.0,0.2,0.4,0.6,0.8,1.0]\n",
    "    binned_data= pd.cut(df[i],bins) \n",
    "    df['_'+i] = binned_data\n",
    "    return df\n",
    "#acco.loc[account_raw['IfrsBalanceEUR'].isna(), 'LoanGBVBuckets'] = 'missing values'\n",
    "df_test1 = df_test.copy()\n",
    "relevant = columns[3:]\n",
    "for i in relevant:\n",
    "    df_test1 = create_binned(df_test1,i)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "#creating bin data for the plots\n",
    "bin_list = list(df_test1.columns)\n",
    "buckets = [[],[],[],[],[]]\n",
    "for i in bin_list[12:]:\n",
    "    temp = df_test1.groupby(i).count()\n",
    "    temp =temp.reset_index()\n",
    "    present = temp['id'].to_list()\n",
    "    for j in range(len(buckets)):\n",
    "        buckets[j].append((i,present[j]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABDAAAAFgCAYAAABNIolGAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjMsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+AADFEAAAgAElEQVR4nOzdd7hkRZ3/8fcHEIwISJABFEREYVWQETHHJegq4qroImJYZw2IWTFizmEX0y4uCLgmTD9YRZFV1NU1MJhIKogIOICjIKIoEr6/P6oamsu9M8OkPvfO+/U897nd1eecru7Tffqcb32rKlWFJEmSJEnSkK016QpIkiRJkiQtjQEMSZIkSZI0eAYwJEmSJEnS4BnAkCRJkiRJg2cAQ5IkSZIkDZ4BDEmSJEmSNHgGMCRJE5HkDkn+lGTtfv8bSf550vWaSZIHJvn5Stzel5Mc0G8/Lcm3V+K290vy1ZW1vbHtrtT3YEUkqSR3Xs51z03yiBkeu8FrHF82yauS/Ofy1XjuSLJPkvP793fnSddHkrTmMIAhSVomPcBwaZL1ppQfmeTNU8pmvEAcqarzqurWVXXNSqjb65P81wquf1WSy/vfL5J8IMnmY/X936rafmXVpar2qqqjlrfOY8+3db+YX2ds2x+vqt1XdNtTLet7MJ0epLmmX/T+McmPk/zDyq7jilrSa6yqt1bVP8P07/sa5N3Agf37+6OpD/b35qQkVyT52dKOBX2djZIsXlIgb2UE+pK8KMlFSS5LcsTU49mUZW+Z5ENJfteX/9aKPLckacUZwJAkLVWSrYEHAgU8ZiVsb4gXfZ+uqtsAGwH7ALcHThkPYqwMadbU39/vVtWtgQ2Aw4Fjkmw0daGBfj5WqVn2mu8InL6Exz8J/Ai4HfBq4LNJNlnKNt8BnLlyqje9JHsABwMPB7YG7gS8YQmrHEY7Htyt/3/RqqyfJGnp1tQTKEnSTfNU4HvAkcABo8IkC4D9gJf3lvX/TvIx4A7Af/eyl4+1Vj8zyXnA12dowd42yQ96a+exo4vbJA9JcsF4hUZZHkn2BF4F7Nuf7yf98dsmOTzJhUl+k+TNo+4qS1JVV1XV6cC+wGLgJdPVIckr+nYvT/LzJA9fQl2+keQtSb4DXAHcKTfuMpMk7++v/WdJHj71tY7dH8/yGLUK/6E/532ntlQnuV+Sk/u2T05yv7HHvpHkTUm+01/LV5NsPN17M817cG6Slyb5ad/2p5PcfBne42uBI4Bb9PfiIUku6O/pRcBH+/afleTsJJckOS7JvCmbemSSc3oL+btGgaEk2yb5epLf98c+nmSDKeveO8kZaVlFHx3Ve7rP2tjrXdL7/uBez7uPLb9pkr9Md/He99F3krwvySXA65dW76W93/27dmGSRUn+OWPdbJKsl+TdSc5LcnGSf09yixle51pJXpPk10l+m+To/n1aL8mfgLWBnyT55TTr3gW4F3BIVf2lqj4HnAr843TP1de5L/B39P0+wzJ3A/4duG9/v//Qy2/b67e41/c1mTlAeABweFWdXlWXAm8CnjbD821PC9YuqKrFVXVNVZ0yU/0kSauHAQxJ0rJ4KvDx/rdHks0AquqwXvbOnk7+6KraHzgPeHQve+fYdh5Ma83cYwnP8wxgHnA1cOjSKlZVXwHeSsuguHVV3bM/dFTfxp2BnYHdgWUeY6N3bTmWlnlyA/3i5kDg3j1rYw/g3CXUBWB/YAFwG+DX0zzlfYBzgI2BQ4DPZ5rshGk8qP/foD/nd6fUdSPgS7T38nbAe4EvJbnd2GL/BDwd2BRYF3jpMjzvyBOBPYFtgHswwwXhlDqtQ9sXfwLO6sW3p7Vy3xFYkORhwNv69jenvWefmrKpfYD5tAvmvWmfHYD0defRPm9bAa+fsu5+tP22LXAX4DVLf6k3MPV9/2av31PGlnky8D9VtXiGbYz2+abAW5ax3tO+32nBsxcDj6B95h88Zb130F7nTv3xLYDXzVCvp/W/h9KyFG4NfKCqruwZNAD3rKptp1l3R+Ccqrp8rOwnvfxG0oKKH6R9n2qG+lBVZwLPpmfxVNUosPN+4La9ng+mHUOePsNmdux1Ga/XZlO+CyP3oX3m3tCDSacmmTEII0laPQxgSJKWKMkDaBeVx/QWyF/SLniXx+ur6s9V9ZcZHv9YVZ1WVX8GXgs8McuQNTFNnTcD9gJe2J/vt8D7gCfdxE0tol1UT3UNsB6wQ5KbVdW5VXWj1ugpjuwtv1dX1VXTPP5b4F97BsingZ8Dj7qJ9Z3Oo4Czqupj/bk/CfwMePTYMh+tql/0/XIM7SJ3WR1aVYuq6hLgv5ey7m695fwi2sX9PlV1WX/sWlqr/ZW9HvsBR1TVD6vqSuCVtNb3rce2946quqSqzgP+tW+Tqjq7qk7s21pMC9pMvaD/QFWd3+v9ltG6K+go4J/GMgD2Bz62hOUXVdX7+375yzLWe6b3+4m0/Xh6VV3BWNeIJAGeBbyov1+X0wJtM30f9gPeW1XnVNWfaO/9k7Js3VxuDVw2pewyWuBuOgcB31+e7IZ+bNgXeGVVXV5V5wLvob3vy1K30e3p6rYlLSvkMlpA6UDgqJ4JIkmakNnU31KSNBkHAF+tqt/1+5/oZe9bjm2dfxMe/zVwM1pGwk11x77uhe3aDWhB+6U9/1RbAJdMLayqs5O8kNY6vmOSE4AXV9WiJWxrac/9m6oab4H+Ne3CaUXN48YZH7+mvbaRi8ZuX0G70FtWU9ddUp2/V1UPmOGxxVX117H784Afju5U1Z+S/J5W73N78dTPyzxoXTdoGScPpF2crgVcOuX5pl13RVTV95P8GXhwkgtpmQ7HLWGVG3wmlrHeM73f84CFM2x7E+CWtDFdrns6WleQ6Uz9zPyads64GfCbGV9N8ydg/Sll6wOXT12wdwk6CNhlKducyca0jKGpdd1i+sVvVLfR7RvVDfgLcBXw5qq6GvhmkpNomVyrdKwOSdLMzMCQJM2o95F/Iu2C7KI+PsGLgHsmGXWPmC7te6ZU8BlTxLutxm7fgXYB8Tvgz7QLsFG91qZdlM203fOBK4GNq2qD/rd+VU2bxj6d3or+aOB/p3u8qj7RL8bv2J//HTPUZaY6TrVFxq4uaa9/FBC5weundbdY1u0u6nUcdweWfiG6uk19HTeod5Jb0brAjNd76udl9H69rW/vHlW1Pq1bx/h7u6R1l7e+I0f159sf+OyUoMzStrEs9Z7JhbSsgZHx1/c72gX5jmPfh9uOdQeZaupn5g607lgXL0M9TqeNazKe1XBPph/0c1da96Az+rHl34Bd+7FmuuDK1Pfrd7RjxNS6zvTZPr3XZbxeF1fV76dZ9qczbEOSNEEGMCRJS/JYWneJHWip6jvR+ub/L62vObSLmjtNWW+6smXxlCQ7JLkl8EbaBeA1wC+Amyd5VJKb0cYrGJ/+8GJg61HqflVdCHwVeE+S9fughNsmmZqOfyNJbtbTxD9JCxS8d5pltk/ysLQpGP9KuzgcTQd7g7rcBJsCB/XnfwLtfT6+P/ZjWgr/zZLMBx4/tt5iWveLmd7v44G7JPmnJOsk2Ze2P794E+u3un0CeHqSnfr7/FZaV4Nzx5Z5WZINk2wFvAD4dC+/Da21/Q9JtgBeNs32n5dkyz5GyKvG1l1WM73vH6ONzfEU4OibuM1lqfdMjqG9X3fr35/rxrfog6Z+BHhfz/IgyRZps3JM55PAi5Jsk+TWXD+uy9VLq0RV/YL2eT0kyc2T7EMbq+Nz0yz+ZdpsIKNjy+tos5fsVNNPr3wxsGWSdftzXdNf91uS3CbJHWnjgMw0jfHRwDP7MWZD2nHkyBmW/RZtLJ9X9u/N/YGHACcs4eVLklYxAxiSpCU5gNav/ryqumj0B3wA2K/3iT+cNhbEH5L8v77e24DX9LKbMiDkx2gXFBcBN6ell9PHSXgu8J+01tU/A+MzRXym//99klG3g6fS0svPoKXhf5bW2juTffsMC3+gpf3/Hthlhm4h6wFvp7UAX0QLPrxqCXVZFt8HtuvbfAvw+LGW4dfSBpu8lDa2wSdGK/XxDt4CfKe/37uNb7Rv4x9os6n8Hng58A9jXYIGqaq+Rnvdn6NlF2zLjcdsOBY4hXbB/CXaZxHae3Qv2vgFXwI+P81TfIIW5Dqn/735JtZv2ve9qi6gdX0pZsjeWYJlqfdM9fkyrfvJScDZwGgw1yv7/1f08u8l+SPwP8D2M2zuCNp38VvAr2hBuuffhNfxJNrgqpfSviePHw1kmmS/JKf3Ol855bhyGXBVvz2dr9OyKC5KMvr8Pp92PDgH+DZtvx4x3cp9kN130t6jX/e/Q0aPJzk9yX592atoA8M+stfrI8BTq+pnN+F9kCStZLlhd1tJkiStiCRH0AbovKkzm6zMOtwNOA1Yb1kyJyRJmg0MYEiSJK0kfZaUHwM7V9WvVvNz70PL3LgVbSyOa6vqsauzDpIkrUp2IZEkSVoJkryJlvXwrtUdvOj+hTY2xy9pY7I8ZwJ1kCRplTEDQ5IkSZIkDZ4ZGJIkSZIkafDWmXQFltfGG29cW2+99aSrIUmSJEmSVqJTTjnld1W1ydTyWRvA2HrrrVm4cOGkqyFJkiRJklaiJL+ertwuJJIkSZIkafAMYEiSJEmSpMEzgCFJkiRJkgbPAIYkSZIkSRo8AxiSJEmSJGnwDGBIkiRJkqTBM4AhSZIkSZIGzwCGJEmSJEkaPAMYkiRJkiRp8AxgSJIkSZKkwTOAIUmSJEmSBs8AhiRJkiRJGjwDGJIkSZIkafAMYEiSJEmSpMEzgCFJkiRJkgbPAIYkSZIkSRq8pQYwkmyV5KQkZyY5PckLevnrk/wmyY/73yPH1nllkrOT/DzJHmPle/ays5McPFa+TZLvJzkryaeTrLuyX6gkSZIkSZq9liUD42rgJVV1N2A34HlJduiPva+qdup/xwP0x54E7AjsCXwoydpJ1gY+COwF7AA8eWw77+jb2g64FHjmSnp9kiRJkiRpDlhqAKOqLqyqH/bblwNnAlssYZW9gU9V1ZVV9SvgbGDX/nd2VZ1TVX8DPgXsnSTAw4DP9vWPAh67vC9IkiRJkiTNPTdpDIwkWwM7A9/vRQcm+WmSI5Js2Mu2AM4fW+2CXjZT+e2AP1TV1VPKp3v+BUkWJlm4ePHim1J1SZIkSZI0iy1zACPJrYHPAS+sqj8CHwa2BXYCLgTeM1p0mtVrOcpvXFh1WFXNr6r5m2yyybJWXZIkSZIkzXLrLMtCSW5GC158vKo+D1BVF489/hHgi/3uBcBWY6tvCSzqt6cr/x2wQZJ1ehbG+PKSJEmSJEnLNAtJgMOBM6vqvWPlm48ttg9wWr99HPCkJOsl2QbYDvgBcDKwXZ9xZF3aQJ/HVVUBJwGP7+sfABy7Yi9LkiRJkiTNJcuSgXF/YH/g1CQ/7mWvos0ishOtu8e5wL8AVNXpSY4BzqDNYPK8qroGIMmBwAnA2sARVXV6394rgE8leTPwI1rARJIkSZIkCYC0BIjZZ/78+bVw4cJJV0OSJEmSJK1ESU6pqvlTy2/SLCSSJEmSJEmTYABDkiRJkiQNngEMSZIkSZI0eAYwJEmSJEnS4BnAkCRJkiRJg2cAQ5IkSZIkDZ4BDEmSJEmSNHgGMCRJkiRJ0uAZwJAkSZIkSYNnAEOSJEmSJA2eAQxJkiRJkjR4BjAkSZIkSdLgGcCQJEmSJEmDZwBDkiRJkiQNngEMSZIkSZI0eAYwJEmSJEnS4BnAkCRJkiRJg2cAQ5IkSZIkDZ4BDEmSJEmSNHgGMCRJkiRJ0uAZwJAkSZIkSYNnAEOSJEmSJA2eAQxJkiRJkjR4BjAkSZIkSdLgGcCQJEmSJEmDZwBDkiRJkiQNngEMSZIkSZI0eAYwJEmSJEnS4BnAkCRJkiRJg2cAQ5IkSZIkDZ4BDEmSJEmSNHgGMCRJkiRJ0uAZwJAkSZIkSYO31ABGkq2SnJTkzCSnJ3lBL98oyYlJzur/N+zlSXJokrOT/DTJvca2dUBf/qwkB4yV75Lk1L7OoUmyKl6sJEmSJEmanZYlA+Nq4CVVdTdgN+B5SXYADga+VlXbAV/r9wH2ArbrfwuAD0MLeACHAPcBdgUOGQU9+jILxtbbc8VfmiRJkiRJmiuWGsCoqgur6of99uXAmcAWwN7AUX2xo4DH9tt7A0dX8z1ggySbA3sAJ1bVJVV1KXAisGd/bP2q+m5VFXD02LYkSZIkSZJu2hgYSbYGdga+D2xWVRdCC3IAm/bFtgDOH1vtgl62pPILpimf7vkXJFmYZOHixYtvStUlSZIkSdIstswBjCS3Bj4HvLCq/rikRacpq+Uov3Fh1WFVNb+q5m+yySZLq7IkSZIkSZojlimAkeRmtODFx6vq87344t79g/7/t738AmCrsdW3BBYtpXzLacolSZIkSZKAZZuFJMDhwJlV9d6xh44DRjOJHAAcO1b+1D4byW7AZb2LyQnA7kk27IN37g6c0B+7PMlu/bmeOrYtSZIkSZIk1lmGZe4P7A+cmuTHvexVwNuBY5I8EzgPeEJ/7HjgkcDZwBXA0wGq6pIkbwJO7su9saou6befAxwJ3AL4cv+TJEmSJEkCIG3ij9ln/vz5tXDhwklXQ5IkSZIkrURJTqmq+VPLb9IsJJIkSZIkSZNgAEOSJEmSJA2eAQxJkiRJkjR4BjAkSZIkSdLgGcCQJEmSJEmDZwBDkiRJkiQNngEMSZIkSZI0eAYwJEmSJEnS4BnAkCRJkiRJg2cAQ5IkSZIkDZ4BDEmSJEmSNHgGMCRJkiRJ0uAZwJAkSZIkSYNnAEOSJEmSJA2eAQxJkiRJkjR4BjAkSZIkSdLgGcCQJEmSJEmDZwBDkiRJkiQNngEMSZIkSZI0eAYwJEmSJEnS4BnAkCRJkiRJg2cAQ5IkSZIkDZ4BDEmSJEmSNHgGMCRJkiRJ0uAZwJAkSZIkSYNnAEOSJEmSJA2eAQxJkiRJkjR4BjAkSZIkSdLgGcCQJEmSJEmDZwBDkiRJkiQNngEMSZIkSZI0eAYwJEmSJEnS4BnAkCRJkiRJg7fUAEaSI5L8NslpY2WvT/KbJD/uf48ce+yVSc5O8vMke4yV79nLzk5y8Fj5Nkm+n+SsJJ9Osu7KfIGSJEmSJGn2W5YMjCOBPacpf19V7dT/jgdIsgPwJGDHvs6HkqydZG3gg8BewA7Ak/uyAO/o29oOuBR45oq8IEmSJEmSNPcsNYBRVd8CLlnG7e0NfKqqrqyqXwFnA7v2v7Or6pyq+hvwKWDvJAEeBny2r38U8Nib+BokSZIkSdIctyJjYByY5Ke9i8mGvWwL4PyxZS7oZTOV3w74Q1VdPaV8WkkWJFmYZOHixYtXoOqSJEmSJGk2Wd4AxoeBbYGdgAuB9/TyTLNsLUf5tKrqsKqaX1XzN9lkk5tWY0mSJEmSNGutszwrVdXFo9tJPgJ8sd+9ANhqbNEtgUX99nTlvwM2SLJOz8IYX16SJEmSJAlYzgyMJJuP3d0HGM1QchzwpCTrJdkG2A74AXAysF2fcWRd2kCfx1VVAScBj+/rHwAcuzx1kiRJkiRJc9dSMzCSfBJ4CLBxkguAQ4CHJNmJ1t3jXOBfAKrq9CTHAGcAVwPPq6pr+nYOBE4A1gaOqKrT+1O8AvhUkjcDPwIOX2mvTpIkSZIkzQlpSRCzz/z582vhwoWTroYkSZIkSVqJkpxSVfOnlq/ILCSSJEmSJEmrhQEMSZIkSZI0eAYwJEmSJEnS4BnAkCRJkiRJg2cAQ5IkSZIkDZ4BDEmSJEmSNHgGMCRJkiRJ0uAZwJAkSZIkSYNnAEOSJEmSJA2eAQxJkiRJkjR4BjAkSZIkSdLgGcCQJEmSJEmDZwBDkiRJkiQNngEMSZIkSZI0eAYwJEmSJEnS4BnAkCRJkiRJg2cAQ5IkSZIkDZ4BDEmSJEmSNHgGMCRJkiRJ0uAZwJAkSZIkSYNnAEOSJEmSJA2eAQxJkiRJkjR4BjAkSZIkSdLgGcCQJEmSJEmDZwBDkiRJkiQNngEMSZIkSZI0eAYwJEmSJEnS4BnAkCRJkiRJg2cAQ5IkSZIkDZ4BDEmSJEmSNHgGMCRJkiRJ0uAZwJAkSZIkSYNnAEOSJEmSJA3eUgMYSY5I8tskp42VbZTkxCRn9f8b9vIkOTTJ2Ul+muReY+sc0Jc/K8kBY+W7JDm1r3NokqzsFylJkiRJkma3ZcnAOBLYc0rZwcDXqmo74Gv9PsBewHb9bwHwYWgBD+AQ4D7ArsAho6BHX2bB2HpTn0uSJEmSJK3hlhrAqKpvAZdMKd4bOKrfPgp47Fj50dV8D9ggyebAHsCJVXVJVV0KnAjs2R9bv6q+W1UFHD22LUmSJEmSJGD5x8DYrKouBOj/N+3lWwDnjy13QS9bUvkF05RPK8mCJAuTLFy8ePFyVl2SJEmSJM02K3sQz+nGr6jlKJ9WVR1WVfOrav4mm2yynFWUJEmSJEmzzfIGMC7u3T/o/3/byy8Athpbbktg0VLKt5ymXJIkSZIk6TrLG8A4DhjNJHIAcOxY+VP7bCS7AZf1LiYnALsn2bAP3rk7cEJ/7PIku/XZR546ti1JkiRJkiQA1lnaAkk+CTwE2DjJBbTZRN4OHJPkmcB5wBP64scDjwTOBq4Ang5QVZckeRNwcl/ujVU1Ghj0ObSZTm4BfLn/SZIkSZIkXSdt8o/ZZ/78+bVw4cJJV0OSJEmSJK1ESU6pqvlTy1f2IJ6SJEmSJEkr3VK7kEiStDzmzZs36SrMKYsWOca1JElas5mBIUmSJEmSBs8AhiRJkiRJGjwDGJIkSZIkafAMYEiSJEmSpMEzgCFJkiRJkgbPAIYkSZIkSRo8AxiSJEmSJGnw1pl0BdY08+bNm3QV5pRFixZNugqSJEmSpNXADAxJkiRJkjR4BjAkSZIkSdLgGcCQJEmSJEmDZwBDkiRJkiQNngEMSZIkSZI0eAYwJEmSJEnS4BnAkCRJkiRJg2cAQ5IkSZIkDZ4BDEmSJEmSNHgGMCRJkiRJ0uAZwJAkSZIkSYNnAEOSJEmSJA2eAQxJkiRJkjR4BjAkSZIkSdLgGcCQJEmSJEmDZwBDkiRJkiQNngEMSZIkSZI0eAYwJEmSJEnS4BnAkCRJkiRJg2cAQ5IkSZIkDd46k66ANDTz5s2bdBXmnEWLFk26CpIkSZJmOTMwJEmSJEnS4BnAkCRJkiRJg7dCAYwk5yY5NcmPkyzsZRslOTHJWf3/hr08SQ5NcnaSnya519h2DujLn5XkgBV7SZIkSZIkaa5ZGRkYD62qnapqfr9/MPC1qtoO+Fq/D7AXsF3/WwB8GFrAAzgEuA+wK3DIKOghSZIkSZIEq6YLyd7AUf32UcBjx8qPruZ7wAZJNgf2AE6sqkuq6lLgRGDPVVAvSZIkSZI0S61oAKOAryY5JcmCXrZZVV0I0P9v2su3AM4fW/eCXjZT+Y0kWZBkYZKFixcvXsGqS5IkSZKk2WJFp1G9f1UtSrIpcGKSny1h2UxTVksov3Fh1WHAYQDz58+fdhlJkiRJkjT3rFAGRlUt6v9/C3yBNobFxb1rCP3/b/viFwBbja2+JbBoCeWSJEmSJEnACgQwktwqyW1Gt4HdgdOA44DRTCIHAMf228cBT+2zkewGXNa7mJwA7J5kwz545+69TJIkSZIkCVixLiSbAV9IMtrOJ6rqK0lOBo5J8kzgPOAJffnjgUcCZwNXAE8HqKpLkrwJOLkv98aqumQF6iVJkiRJkuaY5Q5gVNU5wD2nKf898PBpygt43gzbOgI4YnnrIkmSJEmS5rZVMY2qJEmSJEnSSmUAQ5IkSZIkDZ4BDEmSJEmSNHgGMCRJkiRJ0uAZwJAkSZIkSYNnAEOSJEmSJA2eAQxJkiRJkjR4BjAkSZIkSdLgGcCQJEmSJEmDZwBDkiRJkiQNngEMSZIkSZI0eAYwJEmSJEnS4BnAkCRJkiRJg7fOpCsgSZIkSavavHnzJl2FOWXRokWTroLWQGZgSJIkSZKkwTMDQ5IkacBsNV65bDWWpNnLDAxJkiRJkjR4BjAkSZIkSdLgGcCQJEmSJEmD5xgYkiRJ0gpwnJKVz7FKJE3HDAxJkiRJkjR4ZmBIkiRJkibObKaVay5mMhnAkDQr+QO3cs3FHzhJkiTNLXYhkSRJkiRJg2cAQ5IkSZIkDZ4BDEmSJEmSNHgGMCRJkiRJ0uAZwJAkSZIkSYNnAEOSJEmSJA2e06hKkrQGc0rilcspiSVJWnXMwJAkSZIkSYNnAEOSJEmSJA3eYAIYSfZM8vMkZyc5eNL1kSRJkiRJwzGIAEaStYEPAnsBOwBPTrLDZGslSZIkSZKGYhABDGBX4OyqOqeq/gZ8Cth7wnWSJEmSJEkDMZQAxhbA+WP3L+hlkiRJkiRJg5lGNdOU1Y0WShYAC/rdPyX5+Sqt1ZptY+B3k67E0iTTfXTWKO6n4XMfDZ/7aHZwPw2f+2j4ZsU+AvcTs2A/uY/cR6vYHacrHEoA4wJgq7H7WwI3mki9qg4DDltdlVqTJVlYVfMnXQ8tmftp+NxHw+c+mh3cT8PnPho+99Hs4H4aPvfR5AylC8nJwHZJtkmyLvAk4LgJ10mSJEmSJA3EIDIwqurqJAcCJwBrA0dU1ekTrpYkSZIkSRqIQQQwAKrqeOD4SddD17Grzuzgfho+99HwuY9mB/fT8LmPhs99NDu4n4bPfTQhqbrRWJmSJEmSJEmDMpQxMCRJkiRJkmZkAEOSJEmSJA2eAQxJkiRJkjR4BjC0XJL42ZmFkmTSddDck2TtSddhkkav3+/X5CVZy98nacWs6cf02cDfHU3CUH5fB1EJzS5J1qqqa6eUeQAdsNEBpxy1VyvR6HtfVdf0+zsn2azfnvO/L2Pfq2t60Zx/zUOVZMsk962qa6vq2iSbTLpOWjrPHYYlyY5wg2P6HSdbI00nScZ+d9afaGW0Rhg737m23193kvXxZEvLZPxipJ8cbpHk0CRPTbKxF8bDNnbA2SvJQUl2mnSdNPuNvvdJdkjyDeADwAlJ1p8a5JyLxr5XeyQ5GXh/kntNuFprqkcARydZN8kbgFOSvCPJIyZdMd3Y1KB6khcneVaSu022ZnaknfoAACAASURBVGuu/l05JMmu/f4LgZcYZBqG8f1QVZXkIUm+Crw1yUvXhEYDTc7Y+c5jknwMuN8k6+OHXUuV5OHA28fu7w0cA5wM3BP4fBIjwAOSZP0eXNqh379lkqOAg4BFwJFJHj3RSmpWGktbXav/PRl4E3BkVd0fOA146yTruCqNn0Qm2TDJR4BHAq8E/gTsl+Sek6rfmmS8u0hVHQmcBfwXcCXwcODXwIuS3HxildQNpBs7GV4/yX6079AdgH9PsuFEK7mGSXKzfvOXwE+BR/fj/K7A4TZQTVaSmyXZZnw/JLkT8DLg1cBngecAj5tQFTVHjXcl67+3bwBeAhxdVd+YWMUwgKFlcxVw9yT37/fXB14HnEH7gTuddsKo4bgD8EDggf2C6+bAj6tqL2Bz4Fb4/ddNMPohG0tbXadfhFwB3B5Yp5e/GHhQkges/lquOmOvf+rJ/KOAv1bV/wBHAn8Gdrc1bNUadWXsGYE7JdmAdkJ/f+DcqjoL+AxwIfCiSdZV118kV9ezOD8JHA3cF9i9ql4LnEkLiGoVG+sCeFUvugj4FrAB8AzgvKr6ieNhTE6SebT9cVCSuyd5Q5K/A3YAFgJ3At4LfKSqPjvBqmoOGe8e3ANo2/fzvU2BDwM/SLJLkgdOqo6eYOlGegPJ+A/Wj4CvAs/q93ehHTDfBLy+qp6Dn6WJ69HR0UHnNODbwPa0LJmtaamg3wXuDty/qo5Ncru+rimimtY0faKfkuT7wOuT7FtVx9KOD5smmVdVv6VdlLx50n0kV5Z+sTx6/S9NcnCSh1XVpbQWsD0BquoM4FRgC2CPiVV4jkqyyeiEqQcuNk3yX8B/ADtW1enAl4EH91V+DxwF/EOS7SdSaZFkd+AdY/cPAt5J21ffof1G7dIffg9wX7tirXpj3XeelOQM4Am0BqmzgOcCz+2ZmnbrWc2SrJfkucD2VbUY2Ab4Bu387efA+bTGgr2Bh1bV2/s6u8y0TWlZjR0bnkFrrN6zZ9p/DfhX2vH8FcC/JnnBJOroRaduoJ+oV4+63SbJ2lV1OXAccJt+IvIZIMCTq+prSW4DHDXJSNyabqw1skZBCeAE4G/A7sCPaan936uqBVX12yR7AguSrGOKqKaTZDfgAUlu3gNkz6alqe5Ha617cZJ70FJY59GyfqiqdwPnMIsHF+utDick2bRfLN8lybeBu9Bavj6XZLuq+ijwyySv7qt+m5aVsomBwZXursDGY/dfC/ymqu5TVd/pZS8HHpHkPr3F6Me0E/87r9aaiiR/n+T2VfVV4GVJNu0P3R3YraqOpp0MnwrcJ8mGPXPmK4wFPLRyjLrvTCl7MK37wbOq6uiq+j3tGPat/rcVcHiS/0ty39Ve6TXMWBerK4EjgB8leQjweeA84E09Y+aXtODsxVX1hyT3Br4EzDdjRjdV/9itNaXsocDTaNlx/1ZVf+xZPjtV1bOr6onAR2jnO6tdvG7RdJK8HHgq8AVaOu7hSQ6gRXv/kXbScXtauvSutADHa8fSy7WaJdkCeAtwW+ATtB+z3WitKp+g9c//PG2gxXvRTiJfVlUnTKTCGqweuLymp31fTcswOIGWWfB7Wkr+P9Iu5DeqqsenDfh2V+B9VfXzCVV9hY1lMVX/Tl1WVX9Km9ViR1of8dcCTwdOqKp9e6DnaODhVXV+2sDGv5vUa5jLkmwHPL6q3pbkxcCDaBdcG9G6xr0X+HtgQVWNBiNc29+m1S/JKMPibcBDgY/SujduBnwTeG5vBHkULTD6iX7/VsC8HszQSjD+HeiNFlf32+8GLqqqdydZr6quTLIe8ETad+t1VXVhkntX1cmTewVz2/g+GStbm9Zd+7a0cQeeQcsue1nfJxsDx9ICG9sDH66qj6zemmsuSZtF7nZVdUaSR9IySf9A64q5PfCNnr19V1pjwS7A/lX109VdVzMw1nAZGwSt379ZklfQ+tU9BPgjcHDaVFqfAS6lnRi+AHgD8APg0VX1Kk8QV59Mmf+7pw1+FDgReD/wL8ABVfV14ALaCf1pwD7AZcAPq+oeBi80Ln2ww9F3ubf03J924rRPVZ1PC4rdtf//MLBDkifRgmTHA78Y296s+o1JsjXtx7iSrFtVvwF+keTJPY33TNogkecBG9LG+ti3qr4HfJEW4GAUvDADY8VM+W2al2Rf4C/AS5NsQ2uBPJv2u/RrWkvQgVX1n7SsmI3hun687ovVYErr72to5xH36GPE/Bx4SVVdBBzaH6eqvgRcA9wjyS2q6s9VdZb7bMXl+rFHRsGL1wCHJXlCX+THtOASvdV/9P97tPO/u/YygxerSNqscAvH7j8vrYviNcCngbVpDQYfpQVq79+DTBtVGzj7YGDXUfDC742WR5JX0QLLL07yOlo3pVNpAbS/0sY5e2TaGCx7A+dX1T0nEbwAMzDUpQ0UtLiqruqtjpcD76L1u1sM3KKqHpdkL+AQWveRX42tvxZ9fK4JVH+NlWSjqrokye2BdWkD7LyPdkJ/BS0j42raDAkn9XT38fVtmRT9ZOgI2sjSJyR5HK2rxNeq6uTedeTutPEengncuaqek2R/YC/gl9UG4JuV0rtg9dvnA5+j9cs/gOtbIe+SZHPgQ8AzqurSJP9H66a1B/A3j38rX5JbVtUVPdX96VX1tCTvALauqn3HlrsV8EHgB1X1oUnVV02S+1XV/yV5O3Drqjqwp7l/EngA8DtaV5ETqupdSTavqgsnWee5ZOyY/mlacHVL2vfjl7Rg80dps0WdTDun+0lVvbMf+59ASx1fq6r+svprv+ZJcg7wMWA92vSU5wO/omX7PQO4B21/7UT7XdoN+D/afrqmB909n9NSTXe9ljYEwBOq6qAkT6Fl2T+mqv5vbJkDaIOWP4kWPxgFRSfyuZtVrWNaOaa0aiXJ6EfsfUn26K2Oe9JmGdid9kF+UJK9qurLwIumBC9SffyF1fxS1ii54XRG2yY5CfhUkn2AS4CLgZfSUg4fT2shXlBVZ9LSDL8+ZXvxx049dfVKWobO/r0ryHOBAo5JsgftQuMa2lSHnwS2TxsPYgHwhtkavOjHv+uCF91ptAGLF1TVeVX1X8BlSV5KS6VcDLw7yRdp78uzq+rKfgJpy9dK1FsmT0ryGNoJ/d/Sxlx6HbBzkocnWSfJM2ktyRcBh42t7znOKjb1PU7yhLQBIR+TNojvvwI7JnlUb8X/BnBIT5d/P61FGdq+c5+tBP2YdiXwE9q5wPq049Zb+t/TgFvQsmNuSxs49XH9mPYCWiD7yqr6i8e01WZvWkPThVX1IFpX341pXas+R5sN8J9oA2a/inax+ZSqunp07u35nJYmNxwvb6skt+wP3RZYN8kXaMeHfXoAet0k90nrDrg/rYvwtWPBi4ldR5iBsQbrJ4VX0vo1fQx4NPBsWsr4a2nR99enjYT8COA7VfWeSdV3TTWldXgDWh/i3Wkn9FfT9s3XaQNu/aiq5vWuAP9Fu+B6fVVdMJHKa7CmRs37iernaSe7z6mqXyR5MvBCWovpY4F/AN4I/AbYpa4fOPEGn9Ohy5T+xj3r7GDgU7SW4U8D76+qw/vju9C6x2xDy3R6MXBmVX2yPx4DuMtv6vuX1r/2r7RGlq/Tuiq+kRaceFdVfSHJ82iBpvnAzrQMwnOn255Wj7TxST5KC/6dMVa+gNaN8am0cUp+RRvE8/SJVHQOmu7424NBX6QNsvxR4Ga0LIxTq+rQJF8Hvk/LwLgKuNv4ftOql+ROtGznJ9MyY25WVXsn2ZD2e/sw2m/wI2jn5u+uqkVj65t1oSVKcgvgNtVmiBvdfw+tofokWlfg0DK2/r2qPtiX242Wjf9XYLuq+soEqj8jI91roCT3SvKftEGB/g24U1VdWm1E8ItoQYzPANsmOZOWQv0vBi8mYyx4sR9wCq0F5em0AVa/TDu5//uquhj4Vo+Unkk78V9g8ELTGYugvzDJR2gR+PcDtwY2SBvQ7ZO034kn0n7oLgLuUlV/HQUvRplBsyh48WDg8LH7T6cNeHsxcFq1AUhfALwjyToAVXUK8F3go1X1h6p63VjwYi0vlpff+PuXNgbTurRZbl5D+7yN0qq3oZ1M7QPQT7J+BmxaVSdX1bnpU0m7P1atUZZEf7/XSvKGtOmWr6VNwXlgkhcl+XCSg6rqMNo+fG618WHuNx68MOtixSS5G+0YNrr/hCS79WPye4F9ad+ftWgDeI+yMS/r5Xep5oy+vrNYrGRTM1mSPC7JrrRBsRfRBrZ9PvDwJHeuNkX3Qlpg6Z9pjQuvGA9egFkXWrIkt6ZdM+zX7/8DrZv5RcDfAT+kZVWcDPwImJfkAWlTXR9FG7/ol6PgxZCODWZgrGGS7EAbsf2Qqnp/kpfRZhP5SFX9rB9Q/4s2r/RvkuzST95HB+DMlguV2Wrq+5zktrSU/kfRsmT+TJui8ok9xevutD6Rv+779P7AolE3n9nUMq5VZ3QCNdbVYSPaeA7XAG+tqtP6ckfRLkKOqKpFSd4HfKXa2Bi3qqo/T+glrJApmUw3B9avNp3wq2gjbB8FbALctmeffJaWdn08sC0tpfd2PVBoK/8KyPWz3KxVbYraW9FOstahBSzOp421cjdai/GDgf9HCzzdkpY+/Yvpt67VJW2g278lOZLWSvci2r7agxbwG02t/HLaYJC3H2U1aeXpLapfoQWZN6O915fTGjJeSfsOXUibDebltP0DrbvcW6p1G9ZqkDao6ua0/XUO7cLy34FPVtVxSd5Du2j8+x5An08bLPE3fX3P57RUPbNnP+DttDErdqZlWmxGy2R8d1Ud0Zf9BnAkLVPrGbQxWAp4dVWds7rrvqyMeq9heoT9f2mpaNCCFRsB9+0XJz+gTYm6U19+FLxYu0foPXCuYqP3ufdP27CqLqNFSzcCtqyqv9EyZ97UVzmNNrXjpkluXVXfqapfjbVGus/WcKNW7h68uFW/8L6G1g/6LbQxHu7XU8BfTRvr4u1JPkRL/f4FwCh4MbU1aeimBC+2Au5IO6GHlnnxbFo/4w8Bh6eNqfB02nfuubSBIa+uqotHrcUGL5ZPkjsD/w0taydthqtP0AYXPJF2cnXvqnoTLVjxUmCz/tv1Rtp++uXY9jyPWcVGmRZj9zdLG6X+Mb3oIOA+wAOr6itV9aKqOoZ23LgNbSriLxm8WDWqDbT5Olrf9fOq6qG0DKZbAgfSBn98EHCvqnoL7YL5bVX13N5QNauO57NRkrXTBr1+AS1A+3XawKr3o01xvwCgql5Cy8J4eP/N+d74PvJ8TkuS5A5JNgI2oM0muTcte+ca4LFV9W3gBGDrXD/+xWto54HrVtWhtIz7J1fVOaPriNX/SpbOH/4108G0qQ/vWW3U7/+hReTv0R9/WbVpza5jmtqqk+TmSQ5NmyKQJOv1KPxxwL/1i6njaV1GHg1QVe8Cbpfk+f1C6gtV9dqq+tNou+XAqmu00Y/OKIiVNtjh24GjkvwjsAVtFPPjgZcBr6D1l16b9gN3a9rn6u9qbNBemH0X76NW/iQvp8168HPge0kO7hdVz6T1z9+PlomxQVVdTuuCtVeNjcTtCeSKqaqzgX9Kskkv2hA4gxa82J/WKvnD/thLaWOSPL8H3r5cVW8c/z1yf6x6/bfk2rSpbEPrerAubXDOO1TVH4H/BF7Tf8+2T/IZ2mCD76k+rTDMvuDnbFFV36Q1ZNy3F/2SNnj3LtWmvz6Ddoyjqj5RbUB2u8CtJv2YdR5tDLODaWMQnE7rWrUu7YLynn3xe1bV16as7z7SEvWAxKOB29G+79+kXdvdnJbxc6e0cS3eRwto7tzPD79NO1aMGq4v69u7bsDP1f5iloEBjDVQtVkpPk2L2ENr/TqZduI4nmKuVSytv/fLaIOpfqYXPxH4c1XtTGsdPph2Ufk94A5JHtaXewNtwEX6xZatkQIgyabAd5Ns0r/P6wP/QbvweCetxfSxwDtorXIHVdXetIvI29AGFTugqk7s2xtMv8dlNX4MS7IxLY36QbR+n9BavF6Z5HbVus+sRetrfBB9RoSxH/JZ9/oH7lbAxWkD1V1NG5T4k8DHq2qfqjo/yRY92HEwrUvjnzM29sLEar6GmJJxkVw/W9kHaeMoHE6btvsBfbEPA7vQgoG/Az5UVQ+oqhPGtzvUk+E54hW0GaLu3rPlxoN7r6J9l27AAOCqN5Y98U3agKlPoc0O8zNaJsZ3aF251+urnDa+nrQsquqKamND3Q7YgZbpeDmwXw+ILaJ1Rf8NrbHqBfRZoHo21tenbG/QxwZPAtZc/wHcM21qs6qq91Tv2w2eZKwuvTvI5rQTwQ8keQRt6qy/pk1ptj1tOqNFtNbyi4B902ZR+EJPBx3f3qAPOFo9qo02vYh20gotm6JoP1rPB64Ajunf80uT7Nk/b3cGftN/CC8f6y4xKzKwkuycZHe4LhB7tyS36S3Ap9KCF6OBOU+nBQ0/0Ff/Z9rF2D5V9fHx7c6W1z8b9Fad39DGunhPDx79DPhgVX2hL/PvtHF9qKrf9xP/645vHudWvbq+y9VjaAGm0aBv3wfe0bOyfgQ8JG1g3EfRWvyq77OT+voG/1aT3jj1OeDzSR5OC1j8sY+l8JfxY7pWn9H5dO+KfSltAP0H9P9rVRvk9j7VunBft7zn4bqp0sbDeQRtwNff08bF2TltZq/P0zIeH0c77zkH+NvYurMqYOYgnmuwJDtX1Y/G7jso3WrUgxBXJ/kwre/qS6vqg0le2+8/d9R61U8ivw5sDaxdVT8Z7a84qJPG5PoBEjentez8I+0C8Uu09NVXVtVn+rJ3oQ2+93Lg+1X1sQlVe6VI8mpav+KP07of3JE2EOezgN/Sxlc4uVo/z9FgnlcAOwK/qqq/9vK16MPRrPYXMcfl+oE7b0kLsu1GC7C9nDbN47a0ASBfWVWXTK6ma7Yk96KN/7IdbeC3r1TVC/tjnwe+XVXvTfJs4F+Ac4EXVtWvJ1RlAb3/+4+AbwCfrqrjJ1sjTSfJvWljmT0P+Ek/JnoOrqVKMo822PiZmWYa3bSxzF5PG/j6WNpAvmtV1SFpEzesRwtCX7Waq75SGcBYA3lyPixpM8M8i9bv/ulpgwx+njal5Xdp/fIfT+uP/38zb0lroulOesaCGG8E7k4bhfrVwHpV9Yq+zLtofW9fM+qCNL7u6nsFK2a8vkluR+sicxXw86p6X9osKjtW1e5pU4g9C3h+VZ3X15kayDUguIqNBW8PpmW83CdtJpJdgEur6tS+nCf0E5Bln61s92rT195xFLjorXgOHj1BSXasG05TO6uO6XOZDU9aUUleBPxTVd17hsfXpWVZPLn/3YPWVf39tAHJrxhbdtZ+Dk0lW8P0H7Jr+wH0Xkk+2k8ctZqN9Ys8g9b6+IAke1QbcOv1XD914xbAgw1eaFyun2Vmugu8Uar964C7AA+lZR+sn+SEJD+iTRn6zrp+/JTR53FWneiOBS/2p42t8Cla/88/9sdfBNwjye5V9UXauDIvHlv/R1O2Nyt/zGeTqrq6/387bTDiZ1Trs//tqjq1j7ng4IITUss2W9mx9PFkxoIXzlY2QWO/CePBi8y2Y/ocl75Prh0vmGSFNHxTun59hNb194D+2A266VXrmv4t2mw3L62q79FmHroueDHWPXjWHqvNwFgDJdmANh3ddrQP9+lLWUWr0Fhr+XOAJ1XVg8ce27j6CO62omg6Se5GS+F+1ZTI+uhztR8tTfWBtMDGdsAtquonfblZG4EHSPJ3wBG0gale2VuID6X17zymqhYleRrw9qq6/ajbzCgDQ5MxloXxMODQqvq7Xm7WxQD048qngf17l8X9aCPc/1tVfdf9NCxTMtHuRRvr6MAeGNQATLOPDqRlA7qPNK2ZMtqSPBZ4LbBbVV019XjcAxS708bDeFUPaswpZmCsYZI8jjaS+FeqTQ9o8GLCRj9oVfVhYMvekjx67HdjrZEGL9ZwueHMAGsneR3wQuDU8eAF3OBz9XFaZsJBvXX0F/2CZPS5mrXBi+6xwBHVZq/4WS/7LLAzbWwLqupI4JQk2/bXf56tXhN3bd8HfwJ+McoE9KJ4GMrZymaVHqzeoAdv3wK82wvjYZlmH73HfaQlGWW0pU1h/Z4kj0yyYVX9P9pUyYf0RTNlvWuBk6rqpXMxeAEGMNZEP6D1B3dgpwFJGyUcYM+aMpCiKbka695x3eegByh2pc0Zf/gM641SCx/F9bNtjNafK5+ruwK3hOtG4KaqvkUbVHDvJHfuZY+qql+OVvJCeXLGuzLS5qnfibHR0DUYzlY2S9g4NXzuIy2LJJtNuf884MvAr2gZFR/qD72Ddo6zTQ9yrN2XXxugqq4c28acCzbbhUQagLnQH02rXpLH037AzqT1adyW9sP291X1i6kZFVPTD+da2nf/oX4ucAvgo1W1uH+X7kwbA2NvWnbGVX35uZBxMidM6cr4/Ko6e8JV0jTibGWzQpItgd/O1dbWucB9pKVJckdgf+CDtPO7PwObAz8BtgfeBdwXeFRVfTnJO4G7VtVj+vnQqGFgdL64YVV9ZAIvZZUzA0OasLHWyGvTBlY9Ig6sqin6OA4vAY4HHkPr/3gtLbPirXDDANj4gHpJtkmyxVy78OhZKD8F5gEfSPIA2vvzAmBxVf1HjU0VZvBiGKZpiTR4MTBjA0JOHeR2Th1D5oqqusAL42FzH2k6vTvvKEPiIlpW6Y9p05/+BTgJeDjwPmBf4KB+m/7/8iQbVNU1vWvfPZMcS2vs+vRqfCmrlQEMacLsF6ll9ADgsKo6jjb415XA/YCjgK2SPAFukM1zTZKbJ3k7bVreOXnxXlXfpAVwzqBlY3yqqp5X1w+WNudSJ+cAuzIOWJytTJJWuZ4VWmOB4WuBjYFLgbdV1bn9sf/f3p1H3z3feRx/vpLYIrEFjYpoa1d6CNVWlRZDOailRpURorW00qqilhzb0JQRS1GHWot2QqsmLTIxllqmtbYjbbW1tAhqaUi0QYa85o/P52a+fhJrcu/95fd6/JP7vd/lfC/n5PvJ+/tehgJX2Z4CTAFWl7SP7ads72H7BUkLSToNOBM43PYBtqd34ne1Q0pIIjqsvo08Gfh6FvR911ulZks6HFiEMvr0FUn/Cgy2fbCk/erncY3jDwK+SInQ/3hBf3M6h/KZlItEvEvKtLKIiLaQdCSwMPBT2/fXNd0mwD71ZdQYSikJlKbs1wNXNzNMJa0AvM/2b9p8+x2RDIyIzsvbyJidmi1pSP2zZ+bA48BgyihDgKco9ZHYPr9H8GJNYEVgc9tXLcjBi0aqe4IXEfNAmg1GRMx7Pdd1ktaWNA4YQenbdaOkQcBEygurXeqhEymZtIsAJ9oe7zI+dfa/42s2Rp8IXkAyMCIiOqb1D+3GQ21r4NPA0c0SiJrKvTSwLaUPxiPA2sBetu9pXo86YKSdv6NTaqp767/TCEppzeiUYEW8e2k2GBExbzXXK3V7BeACSjLBdjXT4kfATNsjJX0RGA1MBX4NHN9oSN7nmykPeOtDIiJifmhkCagGMtYpX/u11sOu9ZCy/TxwhaR7geG2J73J9fqEVv8YXp/qnuBFxHtQ66wjImIeqeuVxYDDgFsp2dfnAPsB6wP3APsDD9fpTz+UtCiwkO3zWtdJhmmREpKIiDZplTs0todJGgscWr+aCGwtaWAzUt9k+w+t4EVr3ndflVT3iIiI6DZzKBfZGrgNWIjSgH1CLR1/DthA0pDadPMi4DIA2xe1gheNBu19PngBycCIiGiLZtRc0hrA321PkfRj4AeSHgMeAn4GDAP+1DhX1CyN5ndzC3L0Ia3+MUl1j4iIiI6bS5bESpTG6v0ozdVn1O8nADtQ1nw32/6WpAvrddTIwk3goiEZGBER81ErCl9LRJaTdBHwA+BsSXvYvhc4EBgOnApsBbxYz+1XS0lcz19e0gWSVu/r9Y9QUt0TvIiIiIhO67HeW1nSYZKWr7u3Ba4CzgausL1DDXRMoIxP/bCkRWvQ4k+tEaud+SXdL008IyLmgzlF4CV9m9Ic74wayPg4sKft++r+fwN2BU62fW6Pcw+nROlPtD2xLT8iIiIiIuZqDmPcNwLGAq8ATwCX18+TbC/ROG4scCllXTi1vXfduyWAERExjzXT/iT9E7Cu7dMkLQ4sB1wIPEqphXwBGGN7Wm1I+U3gT7YvqzWPH6ZE7c8CzrP9agd+UkRERETMhaQvAetQWjT8yPYdkg4A1gOOpDTthFIusj2lcechrebjmS7y9qUHRkTEPFbHnq4DfB1YAthV0iTbv5X0z8DDtveTtCFwO3CDpOttvyBpMLBivc4sSX8BPmZ7Wod+TkRERETMgaQVKeu9FYBrgfMpDTsB/htYFdgJ2Av4DLAFcKDtu5rXSfDi7UsAIyJiHqvTQU4GbgbOAGYCp1H6W7wEfKA28twGmAQ8aft/JQ2j1Eke3LqW7RfbfPsRERER8Rbqem9r4PPAKNu3SBoOfAkYD0ymNBzfDbjX9g3ADfXcNzRoj7cnTTwjIt6D1mirHoZSRmNdZftV2/8CrCVpe+BGysPsGmAxYBfb90BpSgmsXUdrRURERESHzWWtR50GNwm4DtiyfncKsISkUTWr4k7gLNv/07xeq0H7/L/7BU8CGBER71CdBnKupKG1zKPn36XPAKsAH2p8dz1wku1nbY8BPm37qJp50b91UHpcRERERHSeqlagQdJyre9bx9SXT/8JLCdp8/r1KcAJkhax/ajtW5rXTeDivUkAIyLinRtA+fvzAHj9g6iOPf1fyqjUIyVtImlV4K+UiPw+9Zyn65hU1Qh+RERERHSJmiVhSStJuhwY3foeXhfIuAt4ENipBi1+AnzW9isdufEFXAIYERFvQzNLAngKuAzYUNL6dX8/mJ1OiO3zKRH5LwM/A35Bicgv2bqI7Vlp2hQRERHRHVrrvVZwopb/fhu46m4qIQAADO1JREFU3/YxzWNbazjbTwO/oYxLHVa/+20bb7tPSQAjIuJtaAUmJC1fH1h3U+oaD6z7m1kYrWDGuLp/XcrI1AOAe9t75xERERHxdjSyYpeuf04FhgMD4Y39MBpZGHcAh9t+uB332ZclgBERMQfN+sa6vbmkW4Gxkk6qaYFXAstK+lw9phW4mFXLJvtRJpCMBo4C9rZ9GxERERHRdSRtKeku4FxJF9i+g9J4fSFJH2qt8Xqc08/2S41eGf3ncOmYRxLAiIiYg2Zph6SNgW8BhwLnALtJGm37j5TmnLtLWqjV0LP2wWh1lx4EXGB7U9v3deK3RERERMSbq006xwAnAPsAq0s6jBLAGARsBmWNWF9UzV7vSVpN0omSBqS32fyVAEZERDWHtMBRklYBfgXsCqwKXAT8FDhE0lDgx5RxqPvD7L4WrXKT44BLKOUjEREREdG9VgYeBW62PQMYCXyVUkZyN7BJq/dZDVy8JmkRSWOBC4GrM01u/ksAIyKi6tHHYiCwHnBk/X5hYCdgL9uHUUaljrP9PHA8cHnj3D0l3Q48Aexie2obf0ZERERE9DCHsfc9vQCsTZkaJ9t/Bn4NfILSjP064HeN621f9/8G2CyZtu0xoNM3EBHRSfUB1RqHtSxwrO3RtmdI+h5wRp3r/VdgEeApSR+jROKHS1q89cCStDAwClgH2Mb2i534TRERERHx/2qfilaPiqG2/9rjO9l+SNIvgbHAWZIWAZYCJtt+Ani8cb2FgWeBDWy/1PYf1IcpE/wioi/qEbgYCGxn+0pJfwbG2L6iPpz2Bja3/QVJ5wAfpJSSjLJ9+xyuO7CmHUZEREREh0ha1PbLje2VgBOBZSgZsjMb+/rVXhaLA1+jZOGuCJxle3ybbz3eRAIYEdHnSdoL2ML2SEk7A8cBG9l+WdJmwHeBk23/UNK6wO8bfS76p1lTRERERPeQNALYHTjG9kuSBlP6lt0MnDmnrImeWbm2n2vrTcfbkh4YEdEn1G7Ranz+uKRTJW1CKQ+ZAWD7auAx4Oh66nLAZGCN+mCbXJs29a/HJ3gRERER0QVaa71a3nsUsFTtfbEGMMj2d2pA4w2jTut0kX51vfdcvV5aLnSZ/A+JiAVejxrHZW0/V0tF/gCcQpkicn3jlMOBY+sc8BnA/nVk6mwJXERERER0j2ZWrKTVKGW/YynTRB4FpksaYfu+nuu4VoPPnn0yMlWk+6SEJCL6BEmDKHWP6wE/B+60fZukYcD5wHDgAeAc4D7b0yV91PbdjWvMDoRERERERGc1yz7q9jLAFpQ+Fp8DTqJMjrsU2AVY0/a+kpYGxgGX2L61cf6ylPXia8A3mn0yojukhCQiFjhzGZP1HeARYA9gY8qDDdtTKCUkJwFXAntSHni0gheNcpEELyIiIiK6gKQhwF6N7X7AFcAYYM86xv5UYFPg/XXfUEnjgV8BTwK3N87/FjABuNr2VxO86E4pIYmIBUqNxLfS/9YDngemAkOAe4BzgZmUMpGWhShZF3+UNMH2K81rplwkIiIiorvY/puk3SV9hjLC/hjgWOA24MV6zMOSbqBMlfsasDNlusjM+hILSUsCR1DWi5umbKS7JYAREQuU2oBpZeB0ypisQ2w/Kml54Cpgt9b4U0mfpAQ1FgZeqSUirzQaQKXGLiIiIqJLtPpcNEpH/gDsD+xu+7p6zCRK5u1+9bTTKZkWm9m+npKRO/tatqdJOjYZF71DemBERK82pzGmki4BHrR9UuO77SnR9a2AVylNnYYCo23/rX13HBERERHvVu1rtiFwB7AUcBrwhO0j6v7lKH3NPmX7gfrdarYf7NAtxzyUAEZELBAkrQ88a3tKrWH8OCWFcCgwC/ghsAOwPvAhSu3j0bU+Mg06IyIiIrqMpG2Bu20/W7cPAL4B/JKyvjsYWJSyrtvF9q/rcWOBrWxv0ON6SoZt75YSkojo1SStBVwGPAcsIek4SgOmpSm1jC8DqwA72j5B0mLACrZb6YP9bM9K8CIiIiKi63wAuBFA0gbAWsAIYE3gv4Apto+RdCFwrKTLga1tf1nS93teLMGL3i8ZGBHRa8ylXOQk4BnbZ0o6iPKgu9n2tXX/EOB7lI7S4xvnifJ3YAIXEREREV1C0oBWI806Ce6bwLW2fydp0bq9HeWF1YbAObZvqi+x1gOusH1VPT8ZtguYjFGNiK7XaKr5Wt3etNY3AqwALFY/X0rpOr2ypCGSRlNSDH/fDF7UazkPtIiIiIjOk9RP0kYAtl+tI1Fba78Ngd0lLQwMAj4FfNL2WOCDwH6ShgMn2N6xFbyo52ett4BJACMiupaqVrqfpLUk/RY4HrhS0mDg7rpvuO0XgReAdWpjznsoHaePb12vIz8kIiIiIuaors+2A06XNFDSF4GbJB1T9x0KbELpbzad0rjzwNqg/SngOuCpVrCiFfyIBVNKSCKiKzVT/mpEfkvgMeDJmiZ4LSV18EHgs8BKlJFZpwDjbV/UuFZ/YFbqHiMiIiK6Qw1O9KtjUd9Paci5BiUo8RNKqcj1tUz4MGB14HDgI8BRlCaeh9ue3JEfEB2RAEZEdK2aYbENMAqYQhmBOq4+yDYGjqM8yJ4GDqI89G6xfXZn7jgiIiIi3kqPF1VLA6b0r7gYONj2f0jaDtgeuIIyZWQicJntiyUNqdm2zVLj/MO2D0gAIyK6wlwadJ5JqXM82PatkvampBiOsj1d0unAksABtmf2KDdJ06aIiIiILtJzjKmkoykvqkYC9wPHAsvY3qfuP57S9uDbwLbAwrZ/1Dj/DevHWLClPigiukKjQeeWknaSNIBSEvJ3SpACYBLwDPCVuj0OeBLoX6/hRtOnBC8iIiIiukiP4MVXKSNRN7Z9u+3pwA+AwbW/BcC1wCrAurZ/0gxe1OsleNHHJAMjIrqCpEGUh9YQ4C/AYErTpq2AjYBv2J4m6bOU6Pweth/p0O1GRERExFvokR0rYHlgjO3Rkq4ALrU9SdJA2zMkLQ58AdgR2LH2x1jB9lNzumb0PcnAiIi2epPO0MOB521vBnwZuAk4CzgPeB+lUSeUsagHN4MX6TYdERER0X2agYb6eTqwYX1x9RywW903o3HarcCjlKadNIMXPa8ZfU8W/RHRNpLWoqQCtrb3lPSJurky8LHG4ZcACwGrAlcCIyUtaXua7Tub1025SERERER3qNPfmtv7Stq6bg4B/kgp/x0LbCFpO0mLSDoSOAF43PZBth9o641HrzCg0zcQEX3KX4CBko6lpBCOAKZJuga4AJgqaaTtS+vD71ngxbp9k+1pHbvziIiIiHhTtbyj1ddsc0rvsunAQZJWtn2+pGHAZ2xfI+lEShbG0ZR14hjbLzeulWyLeJ30wIiItpK0GWVE1hm2vytpZ2Az4BpgJvDvwBGUTtNLAHtTSktm5UEWERER0d1qgOJiQMCJtm+RtBFwAKX5+jRglu1xjXNWtf1Q/Zz1XsxVSkgioq1s/wKYTBmPCnAL8DilWeevgD0o5SSTbW9v+2+tEpE8zCIiIiK6x1z6kI0Cbra9pe1bAGzfBXwdGAYcDAyq5w+o+1vBi35Z78WbSQZGRLRd7YUxHtjT9v2SPgnsA9xp+/s9js1874iIiIguJunzwGO275I0EjiG0sNsMeAjwCm2J9YpIyOBvW1v1Lk7jt4qPTAiou1sPyBpPOXh9nnbd0haDrindUwdtZX53hERERFdStII4CuUpuvTJT1AGXc/GHgCeJpSNvI1YKLtf0iaAKwpaTHbL3Xo1qOXSglJRHTKecB6krYFsH2N7Smtna46dncRERERMVeS1gZuBB6x/WlgDCXj4hDbZ9v+KaVMeA1goqQB9QXVicAHgJc7cuPRqyWAEREdYfs5YFfb13X6XiIiIiLinbH9e+BW4KP1q4eBCcBqklaUtC9wEyXA8V3br9aXU5fY3iEvquLdSA+MiGi72vApGRYRERERvVjta3YNpSR4ch2duoftfSWtBPzD9tR67ADbr3byfqP3SwZGRLRVbco5y7YljZB0cW3oFBERERG9iO0HgKuBCZK2BI4AZkjqD0yxPVVS/zoaNcGLeM+SgRERbSdpKeAEYDXgUNu/6/AtRURERMS7IGkZYBLwEHCW7Ts6fEuxAEsGRkS0laSdgbspnai3SfAiIiIioveqJSKnAsNbwQtJmXYZ80UyMCKirSQNA56xPbPT9xIRERER717ta4btWZKuB86x/fMO31YswBIZi4i2ao5KjYiIiIjeqfY1e61+/ggwI8GLmN8SwIiIiIiIiIh3xPZrPfuadfiWog9ID4yIiIiIiIh4R9LXLDohPTAiIiIiIiLiHUlfs+iEBDAiIiIiIiIiouulhCQiIiIiIiIiul4CGBERERERERHR9RLAiIiIiIiIiIiulwBGRERERERERHS9/wNvD8nKPXVwJgAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 1080x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#plotting the distribution of values of various social group attributes in the 5 bins\n",
    "def plotDistGraph(bucket, b_range):\n",
    "    attrib = list(map(lambda x : x[0], bucket))\n",
    "    y_pos = np.arange(len(attrib))\n",
    "    counts = list(map(lambda x : x[1], bucket))\n",
    "    plt.figure(figsize=(15, 5))\n",
    "    plt.bar(y_pos, counts, align='center', alpha=0.9, color='black')\n",
    "    plt.xticks(y_pos, attrib, rotation=30)\n",
    "    plt.ylabel ='Frequency'\n",
    "    plt.ylim(bottom=0, top=27500)\n",
    "    plt.title('Attribute Distribution in Probability range of '+str(b_range[0])+' to '+str(b_range[1]))\n",
    "    plt.tight_layout()\n",
    "    \n",
    "    plt.show()\n",
    "    \n",
    "plotDistGraph(buckets[0], [0.0, 0.2])\n",
    "plotDistGraph(buckets[1], [0.2, 0.4])\n",
    "plotDistGraph(buckets[2], [0.4, 0.6])\n",
    "plotDistGraph(buckets[3], [0.6, 0.8])\n",
    "plotDistGraph(buckets[4], [0.8, 1.0])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "From the above distribution plots, it is evident that the social groups most impacted by toxicity are female, male, christian and muslim.  From the social group attribute profiling, it was evident that female and male groups had high means for probability of toxicity. These plots have validatedthe results."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Correlations\n",
    "We will calculate the pearson correlation which can range between -1 to 1 with -1 being negative correlation and 1 being a total positive correlation.\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['id', 'comment_text', 'target', 'male', 'female',\n",
       "       'homosexual_gay_or_lesbian', 'christian', 'jewish', 'muslim', 'black',\n",
       "       'white', 'psychiatric_or_mental_illness'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_test.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [],
   "source": [
    "#figuring out the correlation between various social group attributes and the target/label\n",
    "def plot_corr(a,b):\n",
    "    temp = df_test.columns[a:b].to_list()\n",
    "    print(temp)\n",
    "    return (df_test[temp].corr())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['target', 'male', 'female', 'homosexual_gay_or_lesbian', 'christian', 'jewish', 'muslim', 'black', 'white', 'psychiatric_or_mental_illness']\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x1152 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# plot correlation matrix\n",
    "correlations = plot_corr(2,12)\n",
    "names = df_test.columns[2:].to_list()\n",
    "fig = plt.figure(figsize=(16,16))\n",
    "ax = fig.add_subplot(111)\n",
    "cax = ax.matshow(correlations, vmin=-1, vmax=1)\n",
    "fig.colorbar(cax)\n",
    "ticks = np.arange(0,10,1)\n",
    "ax.set_xticks(ticks)\n",
    "ax.set_yticks(ticks)\n",
    "ax.set_xticklabels(names,rotation=75)\n",
    "ax.set_yticklabels(names)\n",
    "plt.savefig('correlation.png')\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Target has a high correlation with Male category\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "There is a noticeable amount of linear correlation between \n",
    "\n",
    "1) muslim and target \n",
    "\n",
    "2) homosexual_gay_or_lesbian and target\n",
    "\n",
    "3) psychiatric_or_mental_illness has less correlation than muslim and target and homosexual_gay_or_lesbian and target\n",
    "\n",
    "4) Christian has negligible correlation with target"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}