Neural Network.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# EXPERIMENT 7\n",
    "\n",
    "### SHIVAM SINGHAL 2K18/CO/340\n",
    "\n",
    "**AIM:** Build an Artificial Neural Network by implementing the Backpropagation algorithm and test the same using appropriate data sets."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd \n",
    "import numpy as np"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "url = \"https://archive.ics.uci.edu/ml/machine-learning-databases/iris/iris.data\"\n",
    "header = ['sepal length in cm' , 'sepal width in cm' , 'petal length in cm' , 'petal width in cm' , 'class']\n",
    "df = pd.read_csv(url , names = header)\n",
    "df = df.replace(df['class'].unique() , [0,1,2])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "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>sepal length in cm</th>\n",
       "      <th>sepal width in cm</th>\n",
       "      <th>petal length in cm</th>\n",
       "      <th>petal width in cm</th>\n",
       "      <th>class</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>5.1</td>\n",
       "      <td>3.5</td>\n",
       "      <td>1.4</td>\n",
       "      <td>0.2</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>4.9</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.4</td>\n",
       "      <td>0.2</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>4.7</td>\n",
       "      <td>3.2</td>\n",
       "      <td>1.3</td>\n",
       "      <td>0.2</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4.6</td>\n",
       "      <td>3.1</td>\n",
       "      <td>1.5</td>\n",
       "      <td>0.2</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5.0</td>\n",
       "      <td>3.6</td>\n",
       "      <td>1.4</td>\n",
       "      <td>0.2</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   sepal length in cm  sepal width in cm  petal length in cm  \\\n",
       "0                 5.1                3.5                 1.4   \n",
       "1                 4.9                3.0                 1.4   \n",
       "2                 4.7                3.2                 1.3   \n",
       "3                 4.6                3.1                 1.5   \n",
       "4                 5.0                3.6                 1.4   \n",
       "\n",
       "   petal width in cm  class  \n",
       "0                0.2      0  \n",
       "1                0.2      0  \n",
       "2                0.2      0  \n",
       "3                0.2      0  \n",
       "4                0.2      0  "
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(105, 4) (105, 3)\n"
     ]
    }
   ],
   "source": [
    "X = np.array(df)[: , 0:4]\n",
    "Y = np.array(df)[: , 4]\n",
    "\n",
    "from sklearn.model_selection import train_test_split\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, Y,\n",
    "                                                    shuffle=True, stratify=Y,\n",
    "                                                    test_size=0.3, random_state=2)\n",
    "\n",
    "from sklearn import preprocessing\n",
    "X_train = preprocessing.normalize(X_train)\n",
    "X_test = preprocessing.normalize(X_test)\n",
    "\n",
    "from sklearn.preprocessing import OneHotEncoder\n",
    "y_train_without_one_hot = y_train\n",
    "one_hot = OneHotEncoder(sparse = False)\n",
    "y_train = y_train.reshape(-1,1)\n",
    "y_train = np.array(one_hot.fit(y_train).transform(y_train))\n",
    "\n",
    "print(X_train.shape , y_train.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/jpeg": "/9j/4AAQSkZJRgABAQEAYABgAAD/4RCIRXhpZgAATU0AKgAAAAgABAE7AAIAAAAHAAAISodpAAQAAAABAAAIUpydAAEAAAAOAAAQcuocAAcAAAgMAAAAPgAAAAAc6gAAAAgAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAFNoaXZhbQAAAAHqHAAHAAAIDAAACGQAAAAAHOoAAAAIAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAFMAaABpAHYAYQBtAAAA/+EKX2h0dHA6Ly9ucy5hZG9iZS5jb20veGFwLzEuMC8APD94cGFja2V0IGJlZ2luPSfvu78nIGlkPSdXNU0wTXBDZWhpSHpyZVN6TlRjemtjOWQnPz4NCjx4OnhtcG1ldGEgeG1sbnM6eD0iYWRvYmU6bnM6bWV0YS8iPjxyZGY6UkRGIHhtbG5zOnJkZj0iaHR0cDovL3d3dy53My5vcmcvMTk5OS8wMi8yMi1yZGYtc3ludGF4LW5zIyI+PHJkZjpEZXNjcmlwdGlvbiByZGY6YWJvdXQ9InV1aWQ6ZmFmNWJkZDUtYmEzZC0xMWRhLWFkMzEtZDMzZDc1MTgyZjFiIiB4bWxuczpkYz0iaHR0cDovL3B1cmwub3JnL2RjL2VsZW1lbnRzLzEuMS8iLz48cmRmOkRlc2NyaXB0aW9uIHJkZjphYm91dD0idXVpZDpmYWY1YmRkNS1iYTNkLTExZGEtYWQzMS1kMzNkNzUxODJmMWIiIHhtbG5zOmRjPSJodHRwOi8vcHVybC5vcmcvZGMvZWxlbWVudHMvMS4xLyI+PGRjOmNyZWF0b3I+PHJkZjpTZXEgeG1sbnM6cmRmPSJodHRwOi8vd3d3LnczLm9yZy8xOTk5LzAyLzIyLXJkZi1zeW50YXgtbnMjIj48cmRmOmxpPlNoaXZhbTwvcmRmOmxpPjwvcmRmOlNlcT4NCgkJCTwvZGM6Y3JlYXRvcj48L3JkZjpEZXNjcmlwdGlvbj48L3JkZjpSREY+PC94OnhtcG1ldGE+DQogICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgCiAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAKICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgIAogICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgCiAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAKICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgIAogICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgCiAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAKICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgIAogICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgCiAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAKICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgIAogICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgCiAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAKICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgIAogICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgCiAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAKICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgIAogICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgCiAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAKICAgICAgICAgICAgICAgICAgICAgICAgICAgIDw/eHBhY2tldCBlbmQ9J3cnPz7/2wBDAAcFBQYFBAcGBQYIBwcIChELCgkJChUPEAwRGBUaGRgVGBcbHichGx0lHRcYIi4iJSgpKywrGiAvMy8qMicqKyr/2wBDAQcICAoJChQLCxQqHBgcKioqKioqKioqKioqKioqKioqKioqKioqKioqKioqKioqKioqKioqKioqKioqKioqKir/wAARCAG6AuADASIAAhEBAxEB/8QAHwAAAQUBAQEBAQEAAAAAAAAAAAECAwQFBgcICQoL/8QAtRAAAgEDAwIEAwUFBAQAAAF9AQIDAAQRBRIhMUEGE1FhByJxFDKBkaEII0KxwRVS0fAkM2JyggkKFhcYGRolJicoKSo0NTY3ODk6Q0RFRkdISUpTVFVWV1hZWmNkZWZnaGlqc3R1dnd4eXqDhIWGh4iJipKTlJWWl5iZmqKjpKWmp6ipqrKztLW2t7i5usLDxMXGx8jJytLT1NXW19jZ2uHi4+Tl5ufo6erx8vP09fb3+Pn6/8QAHwEAAwEBAQEBAQEBAQAAAAAAAAECAwQFBgcICQoL/8QAtREAAgECBAQDBAcFBAQAAQJ3AAECAxEEBSExBhJBUQdhcRMiMoEIFEKRobHBCSMzUvAVYnLRChYkNOEl8RcYGRomJygpKjU2Nzg5OkNERUZHSElKU1RVVldYWVpjZGVmZ2hpanN0dXZ3eHl6goOEhYaHiImKkpOUlZaXmJmaoqOkpaanqKmqsrO0tba3uLm6wsPExcbHyMnK0tPU1dbX2Nna4uPk5ebn6Onq8vP09fb3+Pn6/9oADAMBAAIRAxEAPwD6RooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACszU9fs9MmS2YS3V7Iu6OztU3yuPXHRR/tMQPeotd1O6gaDTdHCNqd5nyy4ysEYxvmYdwuQAO7ED1In0fRbXRbd0t98s8zb7i5lO6Wd/7zN39h0A4AAoApCbxVefNFa6bpaHoLl3uZPxVNqj8GP1pfJ8WxfN9v0W6/wBj7DLBn/gXmv8AyrdooAwP+Ell09wnibTn01ScC8jk861z7yAAp9XVR71vKwdQyEMrDIIOQRQyq6FXUMrDBBGQRXNun/CH3EbW4I0GeQRvD2sHY4DL6REnBXopIIwM4AOlooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAwPDg+33up63Jy1xcNbQE/wwQsUAH1cSN77h6Vv1heCf+RH0gn7zWys/+8eW/XNbtABRRRQAVFdW0N7ZzWt1GssE6NHJG3RlIwQfwqWigDF8KXE0mhi1u5GludPlezldur+WcK592Ta3/Aq2qwtA413xMF+7/aMZ+jG0gz/Q/jW7QAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQBg+Fj9kTUNHfh9Pu32D1hkYyRn6AMV+qGt6sPXbO7tryHXNIhM93bp5VxbA4N1BnJUf7an5l/EfxZrS03U7TV7BLzT5hNC+RkcFSOCpB5DA8EHkGgC1RRRQAU13WONnkYIiglmY4AA7mnVwvi/xPp9zeW+iTyyR6TPOYNS1BVPkjjP2feOAW6MewyOp4AI9N8HJr9vJ4h/tXVtMvNTle5U2dyYx5ZOIgykdfLCZq3/wj/jXT/8AkG+LYr1B0i1GyX9XXk1CPDmv+E183wXe/wBoacBn+yL6TOB6RSdvoePrWroXjjTdYujp9ysul6qvD2F6Nj5/2SeGH059qAKH9s+PNP8A+P8A8NWOpKPvPp155f4hZOTR/wALJtLTjXdE1rScdXnsy0f4Muc/lXZ0UAYFh478L6ngWmu2ZY9Fkk8tj+DYNbsciSxh4nV0boynINZ1/wCGtE1TP9o6TZXDH+KSBS354zWDJ8L/AA8khk0s32kSHkvYXjoc/iSKAOxorjP+EV8VWH/II8aTyKOkWo2qTZ+r9aPt3xC07/j50jSNXUd7O5aBiP8AgfGaAOzorjP+FhSWfGu+FtbsPWRLfzoh/wACU/0q7Y/EXwnqBCxa3bxPnBW4zCQfT5wKAOmoqK3uoLuPzLWeOdD/ABRuGH5ipaACiiigAooooAKKKKACiiigAooqvfX9ppto91qFxHbwJ96SRsD6fX2oAsUVgDXdUv8AnRNCkeE/duNRl+yo3uF2tJ+arTt3i0fN5Oin/Y82Uf8Aj23+lAG7RWA3iK90/nxBos1tEPvXVk/2qFfc4AcD3KYHcitq1ure+tY7mynjuIJV3RyxMGVh6gjrQBLRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABWNe+Hg96+oaPdPpd/JzJJGoaOf082M8N9QQ3vWzRQBgjUPElp8t5okF+B/wAtdPugpb/gEu3b9N5+tL/bmsScQeFL5G9bq6tkX80kc/pW7RQBz7aVrWsca3fR2VofvWWms25x6POQGx/uKh9+1cvqOmWvgW/kingWXwbq7eXc2zLlbGU9GHoh/Q/QZ9IqG8s7fULKa0vYVmt5kKSRuOGBoA4S2urz4cXEdveySX/hSZgLe7++9hnor46p6Ht+ldVq+gaL4r09BqFvDdxMu6GdD8y56MjjkevHFc1oM0vhrVP+EL8Q4uNPuFb+yrmYAiWPvA+eNwzx6j8BUc0F98NbprmxWW98KyNumthlpNPJ6snqnqO36kAn+zeLfBx/0N38T6Qv/LCU4vIV/wBlukn06/St/QPFukeJEYadc4uI/wDW2sw2TRH0ZT/MZFadle22o2UV5YzJPbzLujkQ5DCsfxB4M0nxDItzPG9rqEf+qv7VvLmQ9vmHX6GgDforhjq3inwf8uv27a/pS9NQs0xPEP8AppH3+o/E11Oja7pniCxF3o95HdQnqUPKn0YdVPsaANCiiigAqlfaPpmpqRqWn2t2D/z3hV/5irtFAHI3Hww8LSSeba2UunzdpbK4eIj8AcfpUX/CF67Y/wDIE8a6lGB0S/jS6H0ycV2dFAHGb/iJp33odE1iMf3GeCU/n8tH/Cc6pZf8h3wbq9tj7z2gW6Qe+VxxXZ0UAcpafEzwpcyeVJqgtJu8d3E0RX6lhj9a6Kz1Kx1FN+n3lvdL13QSq4/Q067sLS/j2X1rBcp/dmjDj8jXOXnw18J3j+Z/ZEdtKDlXtXaEqfYKQP0oA6qiuM/4QK+sudC8YazaeiXLrcov0VgKPI+Imnf6u80XWEHXzongkP02/LQB2dFcZ/wmPiKx/wCQ14Jvwo6yadMlzn3CjBqSH4oeGWlEV9cXGmzH/lnfWzxn88EfrQB0mp6lb6Rps19eFhFEBwgyzEnCqo7sSQAO5IrN07Rprq6j1fxCqyXwO63t87o7EHsvYvjq/U9BgcVTj1Cx8VeK7RNPu4byw0uH7XI0MgdWncskQOP7oWQ4PcoewrqaACiiigArnNR0u50W6fWPDkZOTuvdNThLpe7oP4ZR1yOG6HsR0dFAENneW+oWMN3ZyrLBOgeN16MD0qasDRh/ZniLU9HXi3cLf2o7KJCRIg9g67v+2tb9ABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAZHibw7a+J9GexuyY3B8yCdOHgkH3XU+1ZnhDxBcX/wBo0HxCgj1vTxtnUj5biPoJV9Qe/ufeuqrmPGHh251FbfWNCkEGuablrZ+0y94m9Qf5n3NAGVf6Rf8AgS7l1bwtC91pEjb77SF/5Z+skPofVf8AI67R9Ysde0uLUNLnWe3lHDDqD3BHYj0qr4Y8R23ibR1vLcGKZD5dzbvw0Eo+8pFYWreHL/w7qc3iDwXGGMp3X2k5xHdDuyf3X/n+hAO1rltY8B2N7eHUtFnl0TVuou7PgOf9tOjD8q1PD3iOw8TaYLzTZDwdssLjEkL91YdjWrQBxCeLtY8MyC38d2H+j52pq9ipeFvTzF6of84rsLO9tdRtEurC4juIJBlZImDKfxFTOiyIySKGVhhlYZBHpXHXngNrC8fUfBF82i3bHc9tjdazH/aj/h+o6elAHZUVxtp48bTrtNP8b2B0W6Y4S5zutZj/ALL/AMP0PTua7FHWSNXjYOjDKspyCPWgBaKKKACiiigAooooAKKKKACo5oIbmMx3ESSoequoYH8DVbUta0zR4vM1W/t7RcZHnShSfoDyfwrmW+JNnesY/DGlalrr5xvt4CkIPu7Yx+VAF7wzYWlh4h8SJY2sNqn2qEbIYwgx5CHOB7s1dLXGeHLvWY/GF2fEenQ6a2rW6yW8Mc/m4aHhgzDjcVdenZD6V2dABRRRQAUUUUAcr4g1jT9D8b6Nc6pdxWkU1jdxb5WwCd9uQM/gfzNatv4p0C6x9m1zTpSey3SE/lmqAtbbXPGl7JdW8Vza6bbLaKJUDr5shEkgweOFEP5mp7jwR4Xus+b4f07J6lLZVP5gCgDaininXdBKki+qMCP0p9chL8LPB8jb00nyX7NDcSoR+TYpn/CtLCL/AI8db1+x9Ps+oMMfmDQB2VFcb/whWtwf8ePjjVk9PtKJP/PFH9iePLf/AFPi2zu/T7Tpqp/6CaAOyorjd3xIt+sfhy8X/ZaaNj+fFH9v+OLf/j48GQXI7tbami/owoA7KiuN/wCE51SD/j+8E62nr9mRZ/5EUf8ACztJi/4/tN1qx9ftOnsMflmgDsqK5KD4o+Dpzga0kbdxLDImPzWtO38ZeGrrHka/prE/wm6QH8ic0AbVFQQXtrdf8e1zDN/1zkDfyqegAooooAKKKKACiiigArL1fXI9MlhtYIJL3UbnPkWkRG5gOrMTwiDux+gycCrGralFpGk3F/OrMsKZCL9526Ko9ySAPc1U0DSJLGKW91AiXVb4iS7kByFPaJfRE6D15PUmgCuug6hqf7zxDqk2D/y56dI0EK+xcYkf8SAf7op3/CEeFz/rfD+nTN/fntllb/vpgTW7RQBgnwdpsI3aRJdaRKPutYzlFH/bM5jP4qaYurajocyQ+JPLms2bamqwLsRSTgCZMnZ6bwdpPXbwK6GmyxpNE8UyLJG6lXRxkMD1BHcUAOorn9EaTR9Vl8PXDs8IjNxp0jnJMIIDRE9zGSvP91l7g10FABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAcR4msbrwvrTeL9ChMkLKF1ezQf66Mf8ALVR/eX9R+NdfY31tqdhDe2MqzW86B43U8EGpyAykMAQRgg964FD/AMK38Q+UQw8L6pL8h/h0+c9vZG/T8OQDQ8QeFry31U+I/CDpb6sB/pFs3EV8v91h2b0b/wDXWp4a8U2fiS2k8pHtr23Oy6spuJIH9CO49D/+qtuuX8UeE5NQu49b8Pziw162GI58fJOv/POQdwfXtQB1FFc74Y8WR655tjfQHT9ZteLqxkPI/wBpf7yn1/8ArE9FQBBeWVtqFq9tfW8dxBIMPHKoZT+Brj5PCGr+GpGuPAl/i3zufSL5i8LeuxuqH/Oa7eigDldH8eWV5eDTdbgk0PVuhtbzhXP+w/Rh+VdVVDWND03X7E2msWcV1CegccqfVT1U+4rjru3134fQm407U4tU0VOtlqc6xyxD0jlPB+h/AUAegUVwVt8XtE1CKGPS7LUr6/lHFlBb7nU+5zjHuM1Z+2+PtZ/49NO0/wAPwN/Hdy/aJseoVflB9jQB2hOBk9K57VfHnhrR2Md1q0LzZwILc+a5Pphc4P1xWYPh3/aPzeK9f1LWSfvQ+Z5EB/4An+NdFpXhzR9DXGk6bbWpxjdHGNx+rdT+JoA53/hL/EOr8eGvCdysZ6XWquLdR77PvEfSj/hGfFmsc+IPFJs4m+9baNF5WPpK3zfpXaUUAczpvw88M6bL5401bu5Jy096xndj6/NkZ+grpVVUQKihVUYAAwBSkhVJY4A5JPauIv8AxRqPiW+m0bwKV2xnZday4zFB6iP++/6fzAA7x1rcIltdK0iN73xIkq3FlDAATAw4LSHoqFSynPUMenUdJo2sR6tbvuja2vLc7Lq0c5eB/Q+oPUMOCORWLDY6H8OPDdzfylpJMbp7qU7p7uQ9Bk9ST0HT9TVDQvCupapJL4l128udP1u7ANutu+BZwj7sRU/K45ywYdfQ80Ad1RWAL7xFp3yX2lR6qg6T6dKsbt7mKVgB+DtTv+Enk6Dw/rRf+79nQfqXx+tAG7XPeLvF9n4Xs4lklhN9dsI7WGRsLkkDe5/hjXOSfw6kU5rnxLqfyWljDosR6zXrrPMB7Rxkrn3LnHoamt/C2mx2Nzb3kbag96u27uLsh5Jx2BPGAOwUADsBQBz/APwg2o6SDqHhTX5o9Rl/e3S3Z8y3vZD1Zl/gJPcdBwKsWPj0Wl4mneM7F9CvWOEmc7raY+qydB9D09aoWt5d/Di+j07V5ZLrwzO+yzvn5ayJ6RyH+76Ht/LuLuzs9VsWgvYIbu2lXlJFDKw9f/r0ATq6uiujBlYZDA5BHrS1xD+ENX8NOZ/AuoYt87m0i+YvCfXY3VD/AJzVzSfHtlc3i6br9vLoWq9Ps94cJIf9iTow/KgDq6KKKACiiigAooooAintbe5GLmCKYekiBv51mXHhDw5dZ+0aDprk/wARtUz+eM1sUUAcpP8ADDwdccvokSn1ikdMf98sKg/4Vho0X/Hje6vY+n2a/cY/PNdlRQBxv/CCajB/x4+NdeT0+0SrN/MCj/hHfG1v/wAe3jWOcdludMj/AJg5rsqKAON8r4kW/wBy48O3i/8ATRJo2P5cUf2x4+t/9d4WsLz/AK9tQEef++xXZUUAcb/wmevwf8f3gbU09fs00c/8sUf8LJtIv+P/AEDxDZepn084/ME12VFAHDjxbpHjDxBpGl6XNK/k3BvLmKSB4yFjU7PvAD/WFD/wGu4rmtRv7QfEDQ4RdQtO0N1CYRIC4JCPnb16RGuloAKKKKACiiigDB8Tj7O+kakvD2uowxk+qTN5JH0zIp/4CK3q5zx2bk+Gkg09Y3vZ761W3WUkKXE6PzjnGFJP0qh/bPj22/1/hSxvMdfs2oiPP/fYoA7KiuN/4TbWoP8Aj/8AA+rp6/ZmSf8Alij/AIWbpkX/AB/6Trlh6/adPYY/LNAHZUVycHxQ8HTnaNaSNu6ywyJj81Falv4v8OXePs+u6c5P8P2pAfyzmgDYoqOG5guV3W80co9UcN/KpKACiiigAooooAKKKKACiiigAooooAKKKKACiisnWtaOnNDaWMH2zVLrP2a23bQQOruf4UXIyfcAZJAoA05p4raFpriVIokGWeRgqqPcnpWL/wAJnob/APHrcz3w/vWFnNdKf+BRIwotvDEU8qXniKUateg7h5q/uID/ANM4jkL/ALxy3vW6BgYHAoAwx4z0IMFubx7EscA39tLagn0zKq1tJIksavGyujDKspyCPXNKyq6lXAZWGCCMgisCbwz/AGfI114UlTTZydz2uD9ln7kNGOEJ/vpg+u7pQB0FFZ+kasmqQyBomtru3by7m1kOWhfGfxUjkMOCPyrQoAKKKKACiiigAooooAKKKwH1661SeS28LwRXAicpLf3BIt42HVVxzKw6ELgDoWB4oA36KwR4bubn5tW1/Urlj1S2k+yxj6CPDfmxPvS/8IlapzbalrUD9m/tWeX9JGYfpQBu0Vz7L4i0b51lXXrRfvI6LDdKPUEYjkPthPqelaumapa6vZC6sZC6ZKsrKVeNh1VlPKsO4NAFuiiigAqrqWnWur6bPYX8Qlt7hCkiHuP8e+atUUAcT4W1K68O6x/wh/iCcysq7tKvH/5eYR/AT/fXp9Pwz21YPjDQrTXdCdLq4Sznt2E1reltpt5B0bPp61zOifFexfTha6rHNda1A5ikg02Izicrx5iFflwfrQB0finwpFr6RXVpM1hq9p81pfRfeQ/3W/vKe4/+vmv4Y8WS3t4+h+I4VsNetx80X8Fyv/PSM9x7dvzxU/tvxtrHGj+HrfSYW6XGqzZbH/XNOQfrVO/+Gd54jAm8VeJbm6uY1PkfZYEhSFj6DBJH4jNAHT6v4u0DQdw1XVraB16xb90n/fC5P6Vif8J5f6px4V8L6hfg9Lm6AtoT7gtyf0rI8FW2l+GNXXw9r2j2dnrA5tb8R5W+X1V2yQ3quf8ACvS6AOL/ALH8cazzquvWujQnrBpcO98ehkfofcVYs/hr4dgnFxqEM+sXXefUpjMT+B+X9K6yigDB1jwXoOtWscNzYRwtD/qZ7YCKSH/dYdPp0rEM/i3wb/x9I/ifSF/5bRjF5CPdekn8/pXc0UAZWheJdJ8SWpm0i8Sbb/rIz8skZ9GU8itWub13wPpms3Qv7dpdM1VeUv7Jtkmf9rHDD6/nWUPEfiHwiPL8Y2f9o6evA1ewjztHrLH2+o4+tAHc1V1LU7LR7CS91O5jtraMZaSQ4H09z7CsPUvH+h2emwXNjcjVJrs7bS1szvkmb0wOnvnp+lUNM8KX+vX0WteOmWWWNt9rpSHMFr6Fv77+/T9MAFXy9X+I+TOJtH8ME8R/duL8e/8AcQ/r79u0t7aw0PSxFbRw2dlbRk4UbVRQMkn+ZNW64LV5n8feIH8P6fKy6HYODqlxGcee4PECn0/vH/62QA0eB/H3iBPEOoROmiWLkaVbSDHnuDzOw/8AQR/k97TIYY7eBIYEWOKNQiIowFAGABT6ACiiigAooooAhvLO31CzltL2FJ7eZSkkbjIYVwlvc3fw1vUstRkkuvC077ba7b5msGPRH9U9D2r0Gorq1gvrSW2vIkmglUpJG4yGB7EUASI6yRq8bB0YAqynII9Qap6ro2na5ZG01eziu4T/AAyL09weoPuK4uGW7+Gl6ttdvLdeFJ3xDOcs+nMT91u5j9D2/n38ciTRLJE6vG6hlZTkMD0IPcUAcQdB8TeEfn8K3n9r6avXS7+T50HpHL/Q8fWtbQvG+l61cmxl8zTdUXh7C9Xy5Af9nPDD6flXR1la74a0nxJbCHV7NJiv+rlHyyRn1VhyKANWiuG+y+LvB/8Ax4yN4n0lf+WEzbbuJf8AZbo/481vaB4v0fxGGSwuDHdJ/rbO4Xy5oz3BU/zGRQBt0UUUAFFFFABRVW+1Sw0uLzdSvbe0T+9PKEB/OuZm+JuhNKYdGjvtbnHHl6fas+D9TgfjQB2FFcX/AGv471b/AJBugWWjxHpLqdx5jY9didD7Gj/hC9b1PnxH4vv5VPWDTlW2T6ZGSR9aAOl1LXtJ0dN2q6ja2nGcTShSfoOprnH+Jem3TmPw7p2p64+cbrS1YRg+7tjA96vab8PvC+luJINIglmzky3OZmJ9cvnB+ldGqKiBUUKoGAAMAUAcZ9s+IGrf8e2m6ZoMJ/iupjcSgewX5c+xpf8AhArzUefE3irVNQB+9BbsLaE+xVev512ZIVSWOAOST2rjtQ8cSX17JpXgi0Gr3y8SXJOLW2Pqz/xH2H/1qAKeveHPDPgzw+dU022tNPu7GRbiCWRiXlZTkxBmyx3ruXA9c9q7TTdQttW0y3v7GTzLe4jEkbex9fQ9iOxrjj4d07Qon8TePtS/tS9gGRJOP3MHosUfTOencn0qDw5ceJLeS919dK2aNqE/nLpSk/aIlI5nUHjLdTHxnqOfvAHoNFVNN1Wx1e28/TblLiMHa204KH+6ynlT7EA1boAKKCcDJ4Fc9Pr0urzNY+FiszZKTaljdb23Y4PSSQf3RwD94joQDL8R3Otaj4jtj4YtbW9TQ3MtzHcSFBLM6FQinpuVGJOeP3i96sWHxE0x7tbHXoJ9A1A/8sb9dqN/uyfdI9+KoTeG9X8G3Daj4Qln1K1kYyX2l3Uu5pmP3pUY9HPUjv8AkK3NN1PQfHeitmGK6jB2z2l1GN8L+jKeh9/yNAG+rK6hkIZWGQQcgilrim8C32iOZvA+tS6euc/2fdkzWzewB5T6jJoTx5d6LIsHjjRptM5wL63BmtW/Ecr9DmgDr57S2uhi5t4ph6SIG/nWXceDvDV3nz9B05ierC1QH8wM1pWOoWep2q3OnXUN1A3SSFww/MVYoA5Gb4W+D5m3DR1ifs0M8iY/JsVH/wAK00+L/jw1rXrD0+z6gwx+YNdlRQBxv/CF67b/APHh451RPT7TEk/88Uf2R4/tv9T4n069x/z86eI8/wDfBrsqKAON+0fEa2/1tj4fvQP+eE0sZP8A31xR/wAJP4wt/wDj88DM4/v22oxvn/gOM12VFAHG/wDCwZoP+Qh4R8RQerJZiRR+INA+KnhiM4vpbyxPpc2Ui/yBrsqCMjB5FAHN2/xC8JXWPL1+zXP/AD1fy/8A0LFatvrukXmPsmqWU+enlXCN/I0txoek3mftel2U+evm26N/MVlXHw98JXWfN0CyXP8Azyj8v/0HFAHRggjIOR60Vxp+FfhiM5sYbuxPrbXki/zJo/4V7LB/yD/F3iKD0V7wSKPwIoA7KiuN/wCEY8YW/wDx5+OXcf3LnTo3z/wLOaPs3xGtv9Vf+H70D/nvDLGT/wB80AdfLKkELyzMEjjUs7HoAOSaxPDED3MEmvXqEXepgSKGHMMHWKL2wDuP+0zVzPiDUPGjaJNZaxo+npa3jJaTXNrdnKLK6xkhTyfvY/GvQwAqgKAABgAdqAFooooAKKKKAOe8Rp/ZV1b+JIPlNtiG+A/5aWzNyT7oTvB7AOP4q6GoL60jv9PuLOcZiuImiceoYEH+dUPCl1JfeDtHupzmWaxheQ/7RQZ/XNAGtRRRQAUUUUAFFFFAHP608us6mvh+0leGHyxNqM8ZwyxEkLEpHRnwcnqFU9yDW5b20NnbR29rEkMMShI40XCqB0AFYvhIfaNOutUfmTUruWfP/TMNsiH/AH7RfxJreoAKKKKACuf123l0q4PiHTEJkiUfb4EH/H1AOpx3kQcqepAK9xjoKKAGQzR3EEc0DiSKRQ6OpyGBGQRVHVPEGkaJHu1bUra04yFlkAY/Rep/CuL0rwrqWoyajps3ia/tNM028e1hsrFViZI9qyIpk6kbJFGPQV0Wl+APDOkyebBpUU0+cme6zM5Prls4P0xQBnn4jR6gxTwroepa2eglSLyYPxkfp+VJ9m8f6z/x8XuneHoG/ht4/tMwHoS3y/iK7QAKAAMAcADtRQBx0Pw00iWZZ9fub/XbhTkNf3DMoPsgwMexzTfEfhAWcdtrPg20hs9V00ZSGBAiXUf8UTAdc9vf8x2dFAGX4c8QWniXRYtQstyhiVlif70Lj7yN7j/69alcL4gtZ/BmvP4r0tHfTblgNYtEGeO06j1Hf/65I7W1uoL60iurSVZoJkDxyIchgehoAo6/4e0/xLpjWWpxblzujkU4eJuzKexrm9L8Q6h4Y1SLQPGcvmRynbYauRhLj0ST+6/8/wBT29UtW0ix1zTJdP1SBZ7eUYZW7ehB7EetAF2iuDtNVv8AwBcxaZ4lme70ORvLs9WbloPSOb+jf5HdqyuoZCGVhkEHIIoAWiio554rW3ee5lSKKNSzyOwCqPUk9KAJK5HXPF88upSaB4Qt01HV9v72Rj+4sx6yN3P+z/8AqNCbV9W8fSy2fhmSTTdDVtk+rFSJLj1WEdh/tf8A6j1miaDp3h3TVstJt1hiHLHq0jf3mPUmgDjLP4Vf2YiajpWtT23iAFnkvFjXypGbkqY8YC/59qvQeN73QrhLPx7p/wBgLHamp2wL2sp9z1Qn0P6V21c94z16DRdGERtF1C8vm8i0sWXd57nsR/dHU/8A16AM7xR4iuNQmt/DvhG4jl1DUI98l3GwZLSDvJkdz0H/AOqui0LRLPw9o0Gm6cm2GFeWP3nbuxPck1wuj/DjWPC9nHqHhzVo4dXkTN5bTRg205yTsGBlQM4BH6Vuad49hS9TTPFdm+g6k3Cic5hm7ZSTofx/WgDrqKAQQCOQehrnTe33iWVk0W4NlpSMUk1BVBkuCOCIc8Be3mEHP8I/ioA1tQ1jTdJjD6pqFrZK33TcTKmfpk81nf8ACa+Hep1WEL/fIYL/AN9YxVvTvD2l6VIZbS0T7Q337qUmSaT/AHpGyx/E1pUAVbDU7DVIPO0y9t7yL+/byrIv5g1arK1Hw1pepTfaZLbyLwfdvLZjFMv/AANcEj2OQe4NUotUvtBvYLHxDJ9ptZ2EdtqgULlz0jmUcKx6BhhWPGFOAQDoqKKKAI7i3hu7eS3uYllhlUq8bjIYHqCK4RWuvhpdBHMt34Tmf5W5Z9NYnoe5jJ/L69e/pksUc8LxToskcilXRxkMD1BFABDNFcQJNbyLLFIoZHQ5DA9CDT68/kW5+Gd2ZYRJdeE53+ePln05ieo7mMnt2+vXWvfiT4WtGVI9TW9mb7kNkhmZv++Rj8zQB1VYXiDwfpHiPbJewNFeR/6q9tm8uaM9iGHX6HIrI/4S3xLqnHh/whcRoelxqsogA99n3j+FH9geNNV51jxRDp0Z+9BpNvj8pH+YUAVZ9W8TeBE364y6/oynH2uMiO6iH+0pOH/Dn1NW/wDhavhiWNP7Plu9Rndci2tLR2k+mCAP1qW1+GfhyOYT6hDcavcj/ltqNw0zH8OF/SptV+H2i3zJcabEdFv4R+6u9OAhZfYgcMPr+dAFP/hIvGWq8aL4VSwjPSfV7jb+ca/MKP8AhFfFOqc6/wCLpoIz1t9JhEOPpIfm/Sox4g8SeEvk8W2f9q6cvTVdPj+ZB6yxdvqOPrXV6XrGn63ZLd6TeRXcDfxRtnHsR1B9jQBh2Pw38MWUvnS6d9vuOrTXzmdm+obj9K6aGCK2iEVvEkUa9ERQoH4Cn0UAFFFQXl7badaSXV/PHbwRjLySMFUfjQBPWH4i8XaX4bRVvJGmvJeILK3XfNKfQKP5nisJ/EeueMGa38Fw/YdOyVfWbuM/MP8ApjGeW+p4+lbXh7wdpnh53uYxJd6jNzPf3Tb5pD35PQew/WgDEGheIPGjeZ4qlbSdIbldJtZP3kq9vOkH/oI/Q10dzcaL4L8OtKyw6fp9svCRrjJ9AO7H8zVjW9bsfD+lS6hqkwigj/FnbsqjuT6Vy2j6Hf8AirVYvEfi6ExQxHdpuktysA7SSDu5/T9AAN0jRr7xhqkPiHxXAYbOE79N0l+iDtLIO7HsO1d1RRQBl6h4c0vU7n7VPbmO7AwLq2kaGbHYb0IYj2JxVb/hHLxOIPFOtRJ/d/0aT9XhY/rW7RQBgnwhY3H/ACF7q+1Ze8d7cExH6xLtQ/iprciijgiWKFFjjQYVEGAo9AKdRQAVy3iHwb9tvl1nw/c/2VrkfS4QfJOP7sq/xD36/XFdTRQBy3h7xl9tvm0bxDbDStcj6wMfkuB/eib+Ie3X64rqHRZEZJFDKwwVYZBFZXiHw1p3ibTxbalEdyHdDPGdskLf3lbsa5u28Rap4Nul07xqzXOnMwS11tE4Posw/hP+1/8AXNAFy++HenG7a+8O3Nx4fvzyZbE4jf8A3o/uke3FV/7e8WeGvl8SaSNYsl63+lD5wPV4j/McCu0R1ljV42V0YAqynIIPcGnUAZOh+KNG8Rxb9Iv4p2A+aLO2RPqh5Fa1c/rngjQ9emFxc2pt71Tlby0byplPruHX8c1k+R438Mf8e00XijT1/wCWcxEN0o9m6P8AjyaAO2ormdI8faLqlz9inkk0zURw1lfp5MgPoM8H8DmumoAKKKKACiiigAooooAKKKpalrOm6NB52q31vaJ2M0gXP0Hf8KALtFcW3xDGpMY/COh3+ttnAnCeRbg+8j/4Uf2R431znVtat9Dt2/5d9Mj3y49DI3Q+4oAv+Pr+0sfCN291dQwyRhZ4kkkCmRo3WQKoPUkriujilSeFJYWDxyKGVh0IPINczpvw78OafN9omszqV0fvXOouZ3b3+bj8hTfCmr2VtqF54SF5FLcaWf8AR1VwS1v/AAqf9pM7CPZT3oA6qiiigAooooAqarfppej3l/N9y1geZvfapP8ASq/huxfTPCulWM3+strOKJ8/3lQA/rVDWH/tzWYdBg+a3gdLnUXHQKDujh+rMASP7qnP3hXRUAFFFFABRRRQAUUUUAYXgnjwTpSH70UAif8A3k+Vv1BrdrA0Nv7N1vUtFk4VpGv7T/ajkbMgHuspbPs61v0AFFFFABRRVbUdQg0rTbi+vGKw28Zd8DJOOwHcnoB3JoA4xZ/FaeKPEUvhu00y5smvkVvtUro5dbaENjHGO31Bq1/bXjyL/WeErK4/646kq5/76FZujeKb7wpZtH4z0G6sknnkuX1C3/fxbpHL4fbyuM7e/C13Gmavp+s2oudKvIbuE/xROGx7H0PsaAOZ/wCEs8VRf8fPgO6X18m/ik/lR/wnl9F/x8+C/EC+vk26yfyNdlRQBxv/AAsqyT/j50HxDbevnacwx+RNH/C1fCyf8fNxdW3r51lKMfkprsqKAOSX4l+C7qModagZHGGWWJ1BB7HctczoHizRvCfiQ6Ta6xaXPh2+cyWrrMM2Eh5KNnohPQnp+Zr0yWxtJ/8AX2sMn+/GD/OszUfB/h/VLOa2udIswJVKmSOBVdfdWAyDQBZi8RaLP/qNYsJP9y6Q/wAjVyK5gn/1M0cn+44NeZeGvDehWurSeE/FGiWUl/CpeyvPJC/bYfXI/jHf6exNdLL8LfBk33tEjH+5NIv8moA6e8s7fULOW0vYUnt5lKyRuMhhXCxSX3w0uBDdNNfeFHbEc3LS6cSfut3aP37fzvf8Ks8Np/x7C/tvTyb6QY/MmuY8T6Hp+lyjRtH1nxLfaxdIRHp9vqOVCkdZMghUx69R+dAHoWreKNI0bRk1O8vIzbSgGAxne05PQIB94muag0TVvHU0d94tR7DR0ffbaMpw0vo0x/8AZf5d+Fuvh3rfg+HTdXn1CaeC2B+0GzQO9hk5LIrZDKO5GD/Ou8sNF8SX9jFe6N8RGuLaVd0btpsUgYfXOR/OgDuIoo4IUigjWONAFVEXAUegA6U6uN/sjx/F/q/E2nXH/Xaw25/75NNf/hY9rGztceGJY0BLNKJkwB3yOBQB0+savZ6FpM+o6lKIreBdzHufQAdyTwBXNeEtIvNT1JvF/iJGS9uU22Nq3SzgPQY/vMOSff3IrkYbjxj46urPVpNJ0+90zT52MNus5ihuZF48z5uWA7ZwP1rrv+Ej8bRf67wOko7tDqsX8iKAOyqrqWl2OsWTWmqWsV1bv1jlXI+o9D7iuX/4TPXov+PnwNqi+vkypJ/Kj/hYbR/8fXhHxNF6kWG5R+IagDMl8LXml65aeHNG1u6/sa/jeS6spTva3gTbkRydVDFgmPQseor0OKKOCFIoUWOONQqIowFA4AA7CuW8KapH4h1/WdWjguIFRYLJI7qLy3QKpkOV7ZMv44FdXQAUUUUAFQ3lnb6hZTWl7Es1vMhSSNujA1NRQBieHbuZILvTNSmMl1pcnltNIeZoiN0ch9yvBP8AeVqr6n4/8L6SxS51i3eXOPKtyZnz6YTOD9azfEfhfStd+IGnDV4Hnin06bdGJWRWMUkZXdtIz/rnrpNM8P6RoygaVplraHH3oogGP1bqaAOd/wCE21jU+PDfhHUJ1PSe/K2sf1Gckij+y/Hmrf8AIQ1yw0aJusWnW5lfHpvfofcV2dFAHHRfDLRZZVm1ufUNbmByH1C6ZwD7KMDHtzWW9hL8MNTl1DTrdrjwzdPm6iRd0tix/jU9SnqO38/RaR0WRGSRQysMMrDII9KAIrS7gvrSK6s5kmgmUPHIhyGB7ipq8/ubW7+G19Jf6ZHJdeF533XVmvzNYsesif7HqO1d1Z3lvqFlFd2MyT28yho5EOQwoAmooooAK5PVfANnNetqfh25l0HVOvn2gxHIfR4/usP8811lFAHEp4w1Xw262/jvTvLhztXVrFS8Dem9eqH/ADiuvs7211C1S5sLiO4gkGVkiYMp/EVR8QeINJ0DTzLrVxHHHICqxEbmm/2VXqf85rzaHwVrevT3moeHo38HafdR4W1Mrhrknu6KcRgjsOnoc0AdprXjiC1vzpPh+2fWtYPH2e3PyQ+8j9FH+eKq2fgm51e7j1Lx5drqU6HfDp8YItbc/wC7/Gfc/rWf4c1yy8EQRaN4i0NfD7MQq3sOXtrlvUydQfZs49RXoMUsc8SywSLJG4yro2Qw9QRQA5VVEVEUKqjAUDAA9Kzte16w8OaU9/qcuyNeEReWkbsqjuTTPEPiKw8M6U17qLnGdsUSDLzOeiqO5NYWgeHr7V9UTxN4wQfaxzY6dnKWKnufWT37fyAG6LoN/wCINVi8SeL4tjx/Np+lnlbQdmYd5D+n5AdpRRQAUUUUAFFFFABRRRQAUUUUAFRXVrBfWslteQpPBKu145FyrD0IqWigDgpNO1f4fSPc6Ck2q+HvvS6aW3TWnq0RP3l/2f8A9dddout6f4g02O/0m5W4gfuOqn0YdQfY1frkNZ8H3Ntqb694MmSw1MjM9uw/0e8Ho69m/wBof/XoA6+iud8NeMLbXZHsbuF9O1mAf6Rp8/Drj+JT/EvuK6KgDO1jQNK1+28jWLCG7Tt5i/Mv0Ycj8DXNf8Ir4i8OfP4P1s3FqvTTNVJkTHokg+ZfYdPWu2ooA423+IcFncJZ+L9OuNAumO0PMN9vIf8AZlHH5/nXXwzxXMKTW8qSxOMq8bBlYexHWm3NtBeW7QXcEc8LjDRyoGVvqDxXCaz4X0zwtu1DQfEf/CLux3GGSUPbyn/rkx6/Tp2FAHoFFeUW3xd1WHTLiWfw+2ppbnH9oWIkS2f3O5MrWppT+JfHFqLkeKLDTrJusGjjzJR7M7cofpQB22pazpujQedqt9b2idjNIFz9B3/CuYb4hjUmMfhHQ7/W2zgThPItwfeR/wDCrum/Dvw5p0/2mWzbUbvqbnUHM7k+vzcZ+grplUKoVQAoGAAOlAHGf2R431znVtat9Dt2/wCXfTI98uPQyN0PuKu6b8O/DmnT/aZbNtRu+pudQczuT6/Nxn6CunooARVCqFUAKBgADpTZ54raB57mVIoo13PI7BVUepJ6Vn694h03w3p/2vVZ/LUnbHGoy8rf3VXua5aDQ9W8czJe+Lkew0hH32+jKcNJ6NMf/Zf5dwAm1jVvHssln4Xd9O0RW2T6uykPP6rCP/Zv8naj8D6Pa6LFYabE1lJA/mw3kR/fJLjHmbj94noQeCDjGOK34oo4IUigjWONAFVEXAUegA6U+gDnk1290cCHxPasEXgalaRs8D+7qMtEfrlf9qtax1bTtTjEmm39tdoejQTK4/Q1brNvfDeh6nIZNR0bT7tz1ae1RyfxIoAlv9a0vS03alqNraj/AKbTKufYAnk+1ZTavqeufuvDtvJaW5+9qd7CUGP+mUTYZz6FgF7/ADdK0rHQNH0t92maTY2betvbJGf0ArQoAqaZplvpNkLa0DEFi8kkjbnlc9XZu7H1/pVuiigAooooAKKKKACiiigDL1zSH1OCGWzn+y6haP5trcYyFbGCrDujDgj8RyAaZpGvpfytZX0JsNViH72zlbkj+/Gf40/2h9Dg8Vr1T1LSLDWIFi1G2SYIdyMch4z6qwwVPuCDQBcorBGh6vacaZ4jn8vtHqFutyF/4ECjn8WJ96X+z/E8vyza/Yxr62ullG/N5nH6UAa15e22nWcl1fTx28EYy8kjBVUfWuS1fT9X8aW0dzYXL6RbWsqXFik8OWuZFOVeVDyqf3V65+YjgCtq18L2cV5He6hNcarexHdHPfOH8s+qIAEQ+6qD71tUAcp4f8YfbL06H4mthpmtrx5LH93cj+9Ex+8D6dfrzS6n8PNHu7o3ul+douodRdac/lEn/aUfKR68c1q+IfDeneJtP+y6nESUO6GaM7ZIW/vK3Y1zVt4h1TwbdLp3jR2udOdglrrarx7LMP4T/tf/AFzQBJ9t8beGuL+0i8TWK/8ALe0HlXKj1MfRvota2ieN9C16X7PaXnk3gOGs7pfKmU+m09fwzW+jrJGrxsHRgCrKcgg9waytb8LaL4ij26vp8M7AYWXG2RPo45H50Aa1FcT/AMI74q8OfN4Y1oapaL00/VjuIHoso5+gPFS2vxEtLe4S08V2Nz4eu24H2oboXP8Asyjgj3OBQB2NFMhmiuIVmt5ElicZV0YMrD2Ip9AGD4s8NL4j01BBMbXUbR/OsrtesMg/oehH+FR+EfE51yxmt9RQWur2DeVf2x42sP4x/snqDWxqOpWekWEl7qVzHbW8Yy0khwB/ifYV5R4h03WvHL3HiTR9MaysoofLEbO0VxqsIYFgQOi4HGeT78YAOr1DxTqHiS+m0bwLtPlnZdaw4zDb+oT++/6fzG54b8K6f4at3FqGmu5ubm8mO6WdvVm/pS+ErvSL3wxZzeHoo4LIphYUGDGw6qw/vA9fXrWzQAEAjBGQa4TUtGv/AAVqEut+EoWuNNkbff6OnT3khHZv9nv/AC7uigChoutWHiDSotQ0qdZreQdR1U91I7Eelcpr9zN4015/Culu6abasDrF2hxn0gU+p7//AFiDjeNYJ9C8SBfAMs0et6nE7XdhbqDGyBT++IPCPnoe5/XqPh1Lox8JQw6Ez5iYi7WcYmEx+95g9c/oOOlAHTWtrBY2kVraRLDBCgSONBgKo6AVLRRQAUVQ1bXdL0K28/WL+C0j7eY+C30HU/hXMf8ACYa54h+TwZoT+Q3A1LU8xQ49VT7zD6flQBs6Wfs/jLXbZuDOtveL7goYj+Rh/UetbtcIdE1rw9eR+KdV1mfV7iEGO8gjiCRrbMct5aDupCtnqQrDGTXcxSxzwpLC6yRyKGR1OQwPIIPcUAOooooAKKKhvLy30+ymu72VYbeFC8kjdFAoA5PxHe6vF42sW8P6XHqk1jYSmeF7gQ4WaRApDHjP7h/1pP8AhPryz/5DfhDWrT1eCIXCL9WU1B4S8UWc2uahDrUc+l6zfz747a9Tyy0IGIlQ9DgckddzNXdUAcpa/E3wndP5baqttKOsd1G8RX6lhj9a6Gz1XT9RXdp99bXQ65gmV/5Gn3VhZ3ybL61guV/uzRhx+tc9efDbwletufRYIX6hrYtDg+2wgUAdRRXGf8K9e050TxTrlj6Rtc+dGP8AgLD+tH9mfECw/wCPTX9L1QDoL6zMJP4x0Adkyh1KsAykYII4Irgbqxu/h1fS6lo0Ulz4bmffeWCctZk9ZYh/d9V/yLn/AAkXjSx/5CXg+O7QdZdPvVP5I3NB+Jml24K65pWsaSOjG7sW2fmucigDq7G+tdTsYbywnSe3mXdHIhyGH+e1WK8mTxNovhbUn1PwjqlteaNcvuvNJEmySInrJCrYP1Uf/q66/wDiLokNrbnSZG1i8u1DW9nZDdI2f739weueR6UAdTLLHDE0szrHGgyzucBR6k1xdz4yv/EF09h4CtVutp2y6rcAi2h9dvdz9OPqKbF4U1fxVKt147uBHagho9FtHIiXHTzGH3z9OP5V2dtbQWdtHb2kMcEMY2pHGoVVHoAKAOd0LwRZ6Zef2pqs8msaywG6+uudh9I16IPpzXT0UUARXNrBe2z295DHPDIMPHIoZWHoQa868R6WPh3bHVfC2rNZJJIFXRpwZoblyfuxr95Sfb9K7LxL4lsvDGmfar3dJLI2y3toxmSdz0VR/nFY3hzw1e3mqDxN4v2yaow/0W0BzHYIf4R6v6mgDm9A1SCbxob/AOI4lsNY6adb3Uey2hTjmNskFs9Sf/1eqgggEHIPQiquo6XY6vZNaapaxXUDdY5VyPr7H3rkf+EX1/wofM8F3/2uxHJ0jUHJUD0jk6r9Dx65oA7iiuY0Xx3p2pXn9m6jHLo+rDg2V6NhY/7DdGHpjr6V09ABRRRQAUVhzeKbV53t9HtrnWJ4ztcWSgxofRpWIQH1G7I9KT7f4of5otB05F9J9VZW/JYGH60AbtFYJ1/UrIbtX8PXUcQ+9NYyC6Vf+AjEh/BDWpp2p2WrWgudNuY7mEkqWjbOGHVT6EdweRQBaooooAKKKKACiiqWpazpujQedqt9b2idjNIFz9B3/CgDP8SeE7HxJHFJI0lpf2x3Wt9bnbLCfr3Ht/KsfTPFl9od/FovjtVhnkbZa6ogxBdfU/wP7Hj6cZVviGNSYx+EdDv9bbOBOE8i3B95H/wqtfeGPFfi+1a38T6hY6bYSEFrOygErnHTLv0PuKAOw1LWdN0aDztVvre0TsZpAufoO/4VzDfEMakxj8I6Hf622cCcJ5FuD7yP/hXMv4Rt/h9q76nf6WviDRWIBuZl824sQO5U/Kyj1ABHt39R06+s9S0+K60yeOe1kX928RyuPT2+nagDlP7I8b65zq2tW+h27f8ALvpke+XHoZG6H3FXdN+HfhzTp/tMtm2o3fU3OoOZ3J9fm4z9BXT0UAIqhVCqAFAwABwBXMat8P8ARtQuvt1iJdH1Eci709/KbPuBwffjn1rqKKAOJ+2+NfDHGoWkfiawX/l4s18u6Ue8fRvoPzrZ0PxnofiBvKsbwJdDh7S4HlzKfTaev4ZrdrF1/wAKaH4hjLaxZRu6DK3KnZJHjuHHPH5UAbVcz4i8Yx6XeJpOkWzaprk4/d2cR4jH96Rv4V/z71wsd94sluL/AEv4fatca3p0MZU3l6q/uW/uRzEjecdyMD9a2/BeteGvDw/sy+gudF1iY5uH1UYe5b+95p4YZ6dPpQBs6B4Okj1Bdc8VXI1PWz91j/qbUf3Yl7Y/vdf1z1lIrBlDKQQRkEHrS0AFFFFABRWXqPiKw065FozSXN6w3LZ2sZllI9So+6PdsD3qsNU8Rz/NaeHYYVPQX+oiNvxEaSD9aAN2isE6t4gtvmvfDiyoOv8AZ9+spH4SLHn8OauaXr+n6tLJBbSsl1EMy2s8bRTRjpkowBx79D2JoA0qKKKACiiigAooooAKKKKACiiigAooooAKKKKACorq1gvrWS2vIUnglXa8ci5Vh6EVLRQBwT6fq/w+la40RJtV8OctLpxbdNaerRE/eX/Z/wDrmuv0bWrDX9Mjv9KuFuLeQdV6qfQjsfY1erjtY8IXVlqja74KljstRPNxaPxb3o9GH8Lf7Q/+vQB2NQ3Vpb31u1vewRXELjDRyoGVvqDWJ4Z8X2viDzLSaJ9P1a34udPn4dD6j+8vuK6GgDjJvh6unztdeDdVudCmJ3GBD5tu5942/p09Kzr/AOIWr+EJls/GOkxTyyITbz6bMCJiPWNjuX3PT2rX13xhO2pPoPhG3XUdYx+9cn9xZj+9I3r/ALP/AOo2fDfg+HRp31LULh9T1ucfv76br/uoP4V9h/8AWoA5zwtawePriLX/ABFqNtftCd9tpED5itPQuvVn+vH9PR65vWfAWh6xP9rED6ffg5W9sH8mUH1yOCfqDWbt8ceGfuNB4psV7NiC6UfX7rfzNAEGrwP4B8QP4h0+Jm0O+cDVLaMZ8hyeJ1H/AKEP8juoZo7iCOaB1kikUOjqchgRkEVzOneOfD+uM+m3zNYXbqY5bDUo/Kc5GCvPDZ54BrL0iZ/APiBPD+oSs2h37k6XcSHPkOTzAx/9B/8A14AO9rF8VeJIfDWjm4KfaLuZhFaWq/enlPRR7eprQ1LUbXSNMnv9QlEVtboXkc9h/j2xXHaBbnVNQfxx4s2WSBdum29w4VbSE/xtngO3XPp+GADW8H+HJ9Jhn1HWZFuNc1FvMu5h0T0jX0Ven+RVfxB4Vul1Q+IfCUiWmsqMSxNxFer/AHHHr6N/+sQzfEJL+drXwdpdzr04O0zIPLt0PvI39OvrTP8AhFPEXiH5vF2um3tm66bpOY0x6NIfmb3H5UANt/iroz2IE9veLqwYxSaVDA0kwkHUDHBHvmnb/HHib/VpD4WsW/ifE90w+n3V/mKdqHw6srW3t7nwaF0bVbLJgmTJWUd0lzksD6nJH6Vo+GfFi6xNLpmqW507W7Uf6RZOev8Atof4lP6fqQBmk+ANE0y5+2XEUmqageWvNQfzpCfUZ4H4CumpruscbPIyoiglmY4AA7k1wc+oX/xEvJbDQ5ZLLw5ExS61FeHuz3ji9F9W/wD1EAk1PXdQ8X6lLoPg+cwWkTbNQ1heRH6xxHu3v2/WtK00m98HW0UOhRS6lpKKA9m8mZ4eOWiY8MD1KHHOdp/hre0zTLPR9OisdNt0t7aEYREH6+596tUAZuneIdL1WQxWl2n2hfv2soMc0f8AvRthh+IrSqnqGj6bq0YTVNPtbxV6C4hV8fTI4rO/4Qrw70OlQlf7hLFf++c4oAsah4l0vTZvs0lz594fu2dsplmb/gC5IHucAdyKpRaXfa9ewX3iGP7NawMJLbSwwbDjpJMw4Zh1CjKqecscEbFhplhpcHk6ZZW9nF/ct4ljX8gKtUAZut+H9M8RWJtNYtEuI+qkjDRn1VuoP0rlivifwNkx+d4l0Jf4TzeWyj0/56D9foK7uigDM0PxDpfiOy+1aRdJOg4dOjxn0ZTyDWnXL674HtdRvTquj3EmjayMkXlsMeYfSRejj9apW3jS90K6TT/HtotkzHbFqkAJtZ/qeqH2P6CgDtaKbHIksayROrowyrKcgj1Bp2cdaACqeqatYaLYveardR2tunV5DjJ9AOpPsK5vUvHLXN8+leDLP+2tQXiSUHFtb+7v3+g/PNLpfgUSXyat4vu/7b1NeUEi4t7fnOI4+n4mgDBvNPufiO3+gaJa6TpDHnUry0RrmcZ/5ZqR8oP94/8A1qzfEPwyh8LvDq/h/wDtCayhQLewW9wUuAo6yxsOp7lSMf09fooA4DSdE1S+0yHUPCfj6+mtZVyn26FLnP8AsknBB9e9XN3xFsOseh6tGP7rPBIfz+Wq+raPfeDNTm8QeFYGnsJjv1LSU6N6yxDsw7jvXW6PrFjr2lw6hpc6zW8oyGHUHuCOxHpQBzX/AAmut2X/ACGvBOqRAdWsXS6H1+XFQXvxc8OWVjNJILxLtFylnPbPG7n0yRtH1z+db/ibxPbeG7OMsjXV9cN5dpZRcyTv6AenqazNA8JSyXb654v8q+1e4XAiYBorRP8AnmgPH1P/ANckAzvB0NlrGpjxJrmrWGoazKMQW0FwrpYof4FAP3vU/wD1ye/rnb7wB4U1HP2nQrME9TCnlE/imKzv+Fa2drzomt63pWOiQXpKfirZz+dAHZ0Vxn9g+OLD/kH+LLa+UdI9QsQPzZOTR/bHj2w/4/fDWn6kB1bT73y/xAk5NAHQ614f0vxFZm11myjuY/4Sw+ZD6qw5B+lcz/ZfivwhzodyfEOlr/y43j4uIl9Ek/i+h/AVL/wsVLTjW/DmuadjrI1p5kY/4Ep/pV2y+IvhK/IEOuW0bdNtwTCQf+BgUATaB400jxBK1rDI9pqEfEthdr5cyHv8p6/hUGJvGN1IPMkh0CFyn7ttragwOG+YciIHjj75B/h+9S8a6boviexsYY0t7m8vLuO3t7uBwXiHLuwZfREcgdM4rsLe3itLWK2to1ihhQRxoo4VQMAD8KACCCG1t0gtokhhjG1I41Cqo9AB0qSiigArG1PQfMuzqmjutlqqjHm4/d3AH8Eqj7w9/vL2PY7NFAGfo2rJq9k0giaC4hkMNzbuctDIOqn16gg9wQe9aFcj4nub7w7rdrqujaf9vk1LFjNa+aIw8gBaJyx4GAHUnvlR2FQ/2R431znVtat9Dt2/5d9Mj3y49DI3Q+4oA6fUtZ03RoPO1W+t7ROxmkC5+g7/AIVzDfEMakxj8I6Hf622cCcJ5FuD7yP/AIVd034d+HNOn+0y2bajd9Tc6g5ncn1+bjP0FdMqhVCqAFAwAB0oA4z+yPG+uc6trVvodu3/AC76ZHvlx6GRuh9xV3Tfh34c06f7TLZtqN31NzqDmdyfX5uM/QV09FACKoVQqgBQMAAdKWiigAIBBBGQeoNcTqHhbUPDl9NrPgTYpkO+60dziG49Sn9x/wBP5HtqKAMTw34q0/xLA/2bfBeQcXNlONssDehHp71t1zfiTwfDrNwmp6dcPpmtwLiC+h6n/ZcfxL7H/wCtVXQ/GE66kmg+LrddO1jH7pwf3F4P70bev+z/APqAB11FFYHiXxdZ+HRFbiN77U7k7bawt+ZJT6n+6vuf1oA09W1ex0PTZL/VblLe3jHLsep9AO59hXGi11j4iOX1AT6P4ZOClt924vh6uf4UPp3/ACNW9K8I3mqaimt+OZEurxDuttPQ5t7P6D+Jvc/rwa7OgCCysrbTrOK0sYEt7eJdqRxrgKKZqOl2Or2httUtIbuE/wAEyBgPcZ6H3q1RQBxLeB9R0NjL4H1uSyQHP9nXpM1sfYZ+ZPqMmlTx7caPIsHjfR59JJO0XsIM1q5/3hyv0Oa7WmyRpLGySqrowwysMgj6UARWV/aalarc6fcxXMD/AHZInDKfxFZGrX15f6n/AGHokhglCB72+AB+yoeiqDwZG5xngDk9gcXxB4K0vSbS913Qru48PXFvC88j2R/duFBOGiPyn6DFdB4W02fTdBhF+3mahc/6ReyEYLzOAW6dhwo9lFAFvS9IstHtfIsIdgY7pJGJZ5W7s7Hlm9zV2iigArP1bRLPWYkF0rJNCd0FzC2yWBvVG7fToehBFaFFAGNpGp3K30mjayVOoQx+ZHMi7Vu4s48wDswOAy9iQRwRWzWF4thePSV1a1Um70lvtce3q6AfvE99ybhj12ntW3FIk0KSxMGR1DKw7g9DQA6iiigAooooAKKKKACiiigAooooAKKKKACiiigAoorN13xBpvhzTzeatcCJM4RRy8rf3VXuaAM/xX4Y07Wrdb2e4OmX1mPMg1ONgjwY9T3X1B/SvP4PHGteIzbaDeajBpVtcSvC2uxI6C8CnG2LIAVj65+mOh6aDRdW8eTR3vitH0/Rlffb6ODhpvRpj/7L/k9ZqOhaZqujtpV9ZxSWRUKIQu0Jjptx90jtigBND0DTvDunLZaTbrDGOWbq0jf3mPUmtGuBW81f4eSeXqjT6v4aGBHeY3T2Q7CQD7ye/wD9YV3FneW+oWcV1ZTJPbyrujkjbKsPrQBNRQSACScAdSa5XU/iFo1ndGy03zta1DoLXTk80g+7Dge/PFAG1q+g6Vr1v5GsWEF2nbzF5X6N1H4GvM/G3h218O6HNZ2/ilI7OVcppWpsZsEdDERl1IPTAxzya6b7H438S8313D4YsW/5YWpEtyw9C/RfqOa5268LaZd+IG8M+GYi0wAfWdYnfzZo0P8AyzVj0du+McfjQBxEnjTV/EiaNputXltbWlr832i6R2juGBwruFB3Yxj0JznvXq+leCNL1lYtU13V5PFMh5R5JB9nU/7ManaPoc/SrniXwPaal4etrfR4o7S90xc6e4HCkfwN6q3Q5+vPfnvD3h/TtftZb3w/LdeFNdtX8q+trU/u0lH96I8FT1GMD8qAPSoYIraFYbeJIokGFSNQqqPYCn1xP9veLvDny+ItHXWLNet9pQ/eAerRH+nFb2h+LNE8RL/xKdQimkA+aBvlkT6oef6UAbFc14z8PWOqaeNRluxpd9p4MtvqQO0wY/vHuvqK2tT1Sz0bTpb7U7hLe2hGWdz+g9T7VxkGnX/xDu4tQ12KSy8OxMHtNNbh7s9pJf8AZ9F//WQDA0nXb74j3lroviG5TTbLyhK8UIeNtWwSMqSBhOMkDn/2X1i1tYLK1jtrOFIYIlCxxoMKoHYCsrxD4W0/xFpaWlwhgkgw1rcQfK9sw6FCOnQcf/WrH0DxNfadqieG/GW2PUOlpfAYivl9j2f1H/1sgHZUUUUAFFFFABRRRQAUUUUAFRXVrb31rJbXkMc8Eq7XjkUMrD0INJd3dvYWslzezxwQRjc8kjBVUe5NcZJ4n1rxbI9t4It/s1jkrJrV2mF/7ZIeWPuePp1oAytdU/DS4jl8Naorw3D8eH7gtJ5hJ6w4yy/jx9elR2GoXfxG1efTtevzodvCSH0OItHcTjrl3YDK+y/p1rsvD3g3TdAme8zJfanLzNqF0d8rk9cH+Eew/Wp/EPhTSvE0KjUYSs8X+puoTsmhPXKt/TpQBf03S7HR7FLPS7WO1t0+7HGuB9fc+5q1XCf2t4j8E/J4ijk1zRl6albp+/gH/TVO4/2h+PpXYaZqtjrNil5pd1HdW79JI2yM+h9D7GgC3RRRQAV5n4rf/hX2vRat4aZXk1OQrcaIASLg4/1qBQdpHftXWeKvFUPh2CKGCFr3Vbs7LOxj+9K3qfRR3NVvC3hWawupNb8QzLe69dD95L/Bbr/zzjHYD170AUfh/p8GqQDxbf3seqarergyL9y0X/niin7uO/c/jk9vXE6xoV94Z1WXxH4Rh81ZTu1HSl4W5Hd0HZx+v5g9NoeuWHiLSo9Q0ubzIX4IPDRt3Vh2IoA0KKKbJIkMZeV1RB1ZjgCgB1FYN9458L6bn7XrtkGHVY5RIw/Bcmsr/hZul3PGiaZrGr5+61pYttP4tjAoA7OqV7o2makCNR060us/894Ff+Yrmf8AhI/Gd9/yDfB62qHpLqF6o/NF5o/s34g3/wDx9a9pWlA9rG0MxH4yUANj8KaHovxD0efStOitJJLW6ZjHkAkeUBxnA4duldrXCSeH7zw7rOla5qPiC/1Ty7n7NOLgqscaTAoGCgcfvPL5z0zXd0AFFFFABRRRQBznjuZ7XwuLuGFp5ra+s5Y4k+87C5j+Ue5BI/GqH/Ca63/0I+rf99pU3jzVPsUWlW6Wd1fM97HcywWke+TyoGEhYL3w/l/nVrSPHnhzWZPJttSjhuc4NtdAwyA+mGxk/TNAGf8A8Jrrf/Qj6t/32lH/AAmut/8AQj6t/wB9pXZUUAcb/wAJrrf/AEI+rf8AfaUf8Jrrf/Qj6t/32ldlRQBxv/Ca63/0I+rf99pR/wAJrrf/AEI+rf8AfaV2VFAHG/8ACa63/wBCPq3/AH2lH/Ca63/0I+rf99pXZUUAcb/wmut/9CPq3/faVna5q914i05rLVvh/qk0R5U70DRt/eU9Qa728vLfT7OW7vZkgt4l3SSSNhVFcO15rHxDk8vS2n0jw0ciS8I2z3o7iMH7q+//ANcUAcPbfE3xNo2k3unRWsl4tqwjS/uBve1BOAshXKsR0BJHPXPSug8J311pEB1GLwjq2q394N82qSSRs02f7pycL7D9a9G03QtN0jR10uws4o7MKVMRXIfPUtn7xPfNcpPoWreB53vvB6PfaS7l7nRWOSmerQnsf9n+fGACz/wmut/9CPq3/faUf8Jrrf8A0I+rf99pW9oHiLTfEun/AGvSp96qdskbDbJE391l7GtSgDjf+E11v/oR9W/77Sj/AITXW/8AoR9W/wC+0rsqKAON/wCE11v/AKEfVv8AvtKP+E11v/oR9W/77SuyooA8817xRqmo6V9iu/Cmo2EFzcQRS3EzKURWmQHOPUHH416HWL4wjWTwfqStNFAwgLxSSuFUSL8yZJ/2gK0NL1CHVtJtdQtTmG6hWVPowz+dAFqiiigAooooAbJGssbRyDcjgqwPcGsbwW7SeBNCZzub+z4AWP8AF8g5qx4j1BtL8OXt3EMzLEVgUfxyt8sa/ixUfjVjSbBdK0Wx0+M5S0t44FPqFUL/AEoAt0UUUAFFFFABRRRQAUUUUAFFVL7VtO0tQ2pX9rZqeQbiZY8/map23i3w5eSCO01/S53PAWK9jY/kDQBr0UA5GR0ooAKKCcDJ4FcTqPinUPEV9NovgUKzRnZdau4zDbeoT++/6fzABpeJPGEOj3CaZptu+p63OuYbGHqP9pz/AAr35/8Ar1W0HwfN/aKa94tuBqOs4/dqB+5sx/djX1/2uv8AM6XhvwrYeGoH+zb57yf5rm9nO6WdvUn09q26ACisPXPGWh+Hj5eo3yfaTwtrD+8lY9hsHI/HFYv9q+M/Eoxo2nR+HbJv+XvURvnI9ViHAP8AvUAdXqWoWGnWTzarcwW9vghmncKp9uev0ryq31G8s9bmu/hRp95eaawZrm2mhK2ZYfxRFiCG46Drx9K7Ow+HelR3a32uS3GvX4/5b6g+9V/3U+6B7c11iqqIFRQqqMAAYAFAHmfh3Tj8Q7NrvxL4hnulRts+j2oNskDd0kH3m78n869B0zSNP0a1FtpVnDaRd1iQDPufU+5rD8ReDhf3yazoNz/ZWuRfduUHyTD+5Kv8Q9+v1rL/AOFlppljdW3ibT5LPXbYAJZIpZbticKYm7gn8vfFAGr4w8RXOni30fQkE+uallbdO0K95W9AP5j2NX/DHhy28M6MlnbkyysfMubh+XnlP3nJ/wA8VneDvDtzZGfXNfbztc1IBpz2gT+GJfQDjPv9K6mgArjPFuk3ml6kvi/w6jPe26bb60XpeQDqMf3lHIPt7AV2dFAFLR9Xs9d0mDUdNlEtvOu5T3HqCOxB4IrnfHGkeEl099U8RQLbyp/q7m3Oy4Z+wUjlj6A5FYuvXw+F2uvqdsqy6LqzMZbBHCvFOBnfGD/CeAfT8hWr4d8O3erahF4n8Xsk14w3WVkjbobJDyCOzP8A7X/1sAHI2nhnxpeR2OsanB/bNpaO0lto+p3OJgv8LM2Apf8A3s/0rt9O+Iekz3QsdZSfQtQ6fZ9RTyw3+6/3SPyzXV1U1HS7DV7U22qWcN3Cf4JkDAe4z0PvQBaVgyhlIKkZBB61n65oOn+ItMex1SHzIm5Vhw0bdmU9iK5tvAt9orGXwRrk+nLnP2C6Jntj7AHlfqMmhfHOoaIwj8b6HNYIDj+0LMGe2PucfMn0OTQBHpmvah4S1SLQfGM3nWsrbNP1huFl9I5T2f37/rXS6rrsOnTx2kMMt7qEw3RWluAWI/vMTwi5/iYgemTxWT4g1bRte8KmGz+ya1/aDi2toBJlXlOSMkcrtALE8EBT3rT8N+Hbbw3pSWsDvcTFVE91KcyTEDAyT2A4A6AcUAVxY+ItR+e+1WPSkPSDTolkdfYyyqQfwRad/wAIzJ1HiDWg/wDe+0If0KY/St2igDAa28S6Z89pfQ61EOsN6iwTEe0kYC59igz6ir2k63bat5karJbXkOPPs7hdssOemRkgg9mBIPY1o1ka7of9peVeWMgtdWtMm1ucfnG/96NsYI/EcgGgDXrltc8cW1hfHStEtpNZ1k5AtLY8Rn1kfoo/X6VkwS+J/HokR2PhvSYpWgnSKQPdTMpw67hwgyDz1PXkGut0Pw/pnhyx+yaRaJbxnl2HLSH1ZupNAHOWvgq81y6TUfHt2t9Ip3Q6ZDkWsH1H8Z9z+tdoiLHGqRqERQAqqMAAdgKWigAoqK4u7ezj8y7nigT+9K4Ufma5+++InhLTyRPrlq7DtATMf/HAaAOlrkNT8C+RfPqvg68/sXUm5kRRm3uOejp/UfzqP/hY0d1xonh3XNSz0kS0KRn/AIEx4/Kj+2fHt/8A8ePhmw01T0fUL3zPzWPkUASaZ468m+TSvGNn/YupNxG7nNvcc4yj9PwP86veKvFUegRQ2tpAb7V7w7bOxj+85/vN6KO5/wDr44PxtF4oureHR9V1ixvL6+I8jSrCxV93+2XflFHPzf8A1yKU3wq1XRVt9QmL+IUWJUu7OKd4ZAvpG2fmA9DjpwKAOy0Sz0vwzPNrHi7W7KXX7ofv5pp1UQj/AJ5xgnhR7dasz/FDwskhitLybUJh/wAs7O2eQn8cY/WqXhDSvh/qsJbRdItPtEP+ttryPfNCenzK5J69+nvXcQ28NtGI7eKOJB0WNQo/IUAch/wm2tXv/IF8E6rKD0a+ZLUfX5s1zdzonj6y1K98R6JY6dpkssebiwgmMpuD/e242l/oRn3zz6vRQB5zoGnan4y0tL+bx3qDRsdskFlAlq8Ld0bGSCK1o/hd4baQSajHeapIOj3127n9CB+lN8Q+Hb/S9VfxP4PUC+xm9sM4jvkH8n9D3/nueHfEdj4m0tb3T2IIOyaFxh4XHVWHY0AOsfDGhaZj7Bo9jAw/iS3UN+eM1qUVia/4u0fw2qrqFzuuX/1VpAN80h9Ao/mcCgDbrndd8baVolwLJTJqGpvxHYWa+ZKT7gfd/GsgweL/ABh/x9O3hfSW/wCWUR3Xko926R/z+tdHoPhjSPDduYtJs0iZv9ZM3zSSH1Zjyf5UAc1J4f8AEvjSM/8ACU3X9jaW/wDzDLFwZXHUeZJ/QfpW9oOqTLM+iay2NTtV+WRuBeRDgTL79mHZvYgndqhq2jWmswRpdqyyQv5kE8TbJIH/ALyMOh/QjggjigC/RWAtz4i0n5Lu0XXIB0uLRlinx/tRsQh+qsM/3RTv+Eus14n0/Won/u/2Rcv+qIw/WgDdqC9vbbTrKW7vplhghXc7t0A/qe2O9ZB8Q312Nuj+H76Vj0lvQLWIfXdl/wAkNOtNBuLi9i1DxFdLe3MLboLeJSlvbnsVUklnH99vwC0AGhWlzc3txruqRNDcXSiO3t3+9b24OVUjs7E7m/4CP4as6v4c0fXo9mr6bb3XGAzp8w+jDkfga06KAOL/AOEDvtI+bwh4kvdOUdLS6P2mD6ANyv15o/4SPxdonHiHw2NRgXrd6M+849fKb5q7SigDndI8eeHNZk8m21KOG5zg210DDID6YbGT9M10VZmr+HNH16PZq+m291xgM6fMPow5H4Gue/4QO+0j5vCHiS905R0tLo/aYPoA3K/XmgDtKK4v/hI/F2iceIfDY1GBet3oz7zj18pvmrS0jx54c1mTybbUo4bnODbXQMMgPphsZP0zQB0VY/iPxRp3hmyE1+7PNIdsFrEN0s7dlVe/16VmeIPGLwX50TwxbjVNbbhowf3VqP70rdsenX9Kk8O+Dk02+fWNauTquuTD57uRfliH9yNf4V/X6dKAM608M6l4svY9U8cKI7VDutdERsxx+jSn+Nvbp/Ku3VVRAqKFVRgADAApaKACiiigDlPEHg55r8654XuBpetryzj/AFV0P7sq98+vX9MS+HfGKanevpGs2x0vXIB+8tJG4lH9+Nv4l/zz1rpqx/EfhfTvE1kIr5GSaM7oLqI7ZYG7FW/p0oA2KK80HxFk8GX0ui+Mpk1F4UzBe2RVnlH92RM/K/8AP9a01n8a+KlDWyw+GNOkGRI+Jrp19QPurn35FAHT6x4g0nw/befrN/DaJjgO3zN9FHJ/AVzX/CU+I/Efy+ENFNrat/zEtVBRSPVIx8zex6etaOj+AtE0m5+2SxSalqJOWvb9/OkJ9RngfgK6WgDjrb4eW93cJeeLtQuPEF2pyFnOyCM/7MQ4/PP0q3aunhLVv7PlURaPfSlrOUDCW0zHLQHsqscsvbJK/wB3PTVDd2lvf2ctrewpPbzKUkikXKsD2IoAmornksdb0IBNKkXVbBeFtbuUrPEPRJTnePZ+f9qpB4qji4v9H1m0fuv9nvcY/GHeD+dAG7RWCfFPncabousXbnoGs2th+Jm2cVG2latr3HiGWOzsT106ykJMo9JZcAkeqqAOxLCgBIm/4SfXIbiP5tG02QvE/a7uBkbh6onOD0Lcj7uT0dNjjSGJIoUWONFCqijAUDoAOwp1ABRRRQAUUUUAFFFFAEN5eW9hZy3d7MkFvCpeSRzgKB3rBiXVvEq+dNJcaNpbf6uGP5Lq4X+87dYgf7q/N6kdKdNGPEPidreUb9N0dkZ0PSa6IDKD6hFKtj+8yn+GuioAzLDw5o+mMXstNt45Ty0xQNI59Wc5Zj7k1dubO1vIjHeW0M6HgrLGGH5GpqKAOffwqlgfO8L3LaTKOfs65a1k9mhzhR7ptPv2p0Pim2tba4HiMx6Vd2aB7iOSTKMp4Dxt/GpPHTIPBANb1YHi7w5F4g02JxbwzX1hILmz85QVZ152Nn+FsbT+B6gUAc//AMTf4j/89tH8MFv924v1/wDZEP6+/btLOysNE01beyhhs7SBeFX5VUepP9TXIQ+Ndb8RwKPBvh6RUYYN7qn7qGM9CAo+ZsHjjuOlTR+AJNWkWfxtrFxrLg7haIfJtkP+4vX6n8aAJb34i6cbprLw5bXHiC9HBjsVzGv+9J90D3Gag/sTxh4k51/Vl0SybrZaWcykejSnof8Ad4rr7KxtNOtVttPtoraBfuxwoFUfgKnoAxdC8I6J4cXOlWEccx+9cP8APK3rlzz/AEraoooAKKKKAIL29t9OsZry9lWG3hQvJIx4UCvP7Tw6/wAR5Zdf8Refa2TIU0i3jco8KZz55P8AeOAR7evFWJyfiN4iNqgJ8MaXN+/cHi/nH8A9UXv6/ljvlUIoVAFUDAAGABQBw9r4i1TwdeLpvjVzcWDsEtdbVflPosw/hb37/ma7hHWSNXjYMjDKspyCPWoru0t760ktbyFJ4JV2vHIuVYehFcO9jq/w9kafR1m1bw31lsC26azHdoyfvL/s/wD1zQB31c/4p8VweHYYoIYWvdVuzts7GL78rep9FHc1j6v8TNNFhap4Zlh1LUb5f3EZcKkI7tKTjYB6HB/nTfDy+HNBnl1PW/Eum32uXQ/0i8lu4xtH9yMZ+VR+v6AAtaL4K89ptU8ZmPU9Wu4zG6MMw20ZH+rjHbg8nr+pNXw/dT+C9eTwrqju+m3LE6Pduc4HeBj6jt/9cAa8vxB8Jw/f1+yP+5Ju/lmuf8T+NfAXiHRpdNvdZyWIaKSG3lZ4pB911IXqP/rUAeh0V5j4X+K9qNOWw1tL271GAlBJbWrN9pQdJMcEE98itz/hYySf8evhbxNP6FdOwPzLUAdlSEBlKsAQRgg9647/AITfWJf+PbwRrDennFI/5mj/AISbxlL/AKjwIyj+9NqsS/pigB+leHtJg+JN/fafYQWzWlnHGxiXaGllZmY4HAIVU5/2zXX1yng2fUJ9U19tas0sr17mJ2gSUSBU8lFX5h1+6f1rq6ACiiigAooooA4zVNdj8IeLrkPZX15Fq0C3EcFjB5r+dH8khxkdVMP4j3o/4SzxRff8gjwTdKp6SahcpBj6rya1rz5viBpAXqmnXjN7AyW4H5n+VbtAHGfZ/iJf/wCtvdE0lD08iJ55B9d3y0f8ILql5/yG/GmsXGfvLaFbVT+C54rs6KAOSt/hh4Uhk82fT2vZu8l3O8pP1BOP0roLHRdL0wD+ztNtLXH/ADxgVP5CrtFABXNeKfFTaTLDpWjQfb9dvBi3tQeEH/PSQ9lH6/rTPFXiuTTbiLRtBhF9r12P3MH8MK/89JD2UfrU/hXwpH4fimubqY32r3h33l9J96Q/3R6KOw/yAA8L+FV0Pzr7UJzf61efNd3rjr/sJ6IPT2+gHQ0UUAc94h8Gadr0y3qNJp+qRYMWoWp2SrjoD/eHsf0rHj8U6x4TkS18c2/nWeQsetWiExn/AK6oOUPuOP513NNkjSWNo5UV0YYZWGQR6EUAMtbqC9tY7mzmjnglXckkbBlYeoIqWuKufBd9oN0+oeArtbMsd0ulzkm1m+g6ofcfoKmsfiNparLb+JlbQNRt13S2130YesbDhx9OaAOvrz3x1Fb+E78eK9Hvrey1FsLcWUjYXUV9No53js351aPiPxF4s+TwhZf2bp7f8xbUEwWHrFF3+p4+laeh+BdM0i8/tC7aXVdVblr69be4P+yDwo+nPvQBz+n+I/EPxCWVfD7RaDp0L+XcXEjCW63d1VOifU/UeldP4f8ABukeHGaa0hae9k/1t9ct5k0h75Y9PoMVm+JPDN7bap/wkvhHbFqyDFxbHiO/Qfwt/teh/wD11seGvEtn4m003Fpuimiby7m1l4kt5O6sP696ANiiiigAooooAKKKKACiiigAooooAKKKKACiis7W9e07w7pzXurXKwxDhR1aQ/3VHUmgC/JKkMTSzOscaAszscBQOpJrzbWgPifcmy0Owt10yKTbPrlxACzY6rBkZP1/+tm9FpOr+PpY7vxKkmmaEp3w6SGIkuPRpiOg/wBn/wDWe5t7eG0t47e1iSGGNQqRxqFVR6ACgDh7X4bT+Hdz+C/EF3pzvgyQ3KLPFKR6jAx9asf8JH4u0TjxD4bGowL1u9GfecevlN81dpRQBzukePPDmsyeTbalHDc5wba6BhkB9MNjJ+ma6KszV/Dmj69Hs1fTbe64wGdPmH0Ycj8DXJ6h4W/4RK1a70PxfPolunS3v5BNb/7oDcj8MmgDv6a7pFGzyMqIoyzMcACvKrb4leLp9MuHsfDaasIjhNRtI5hC/vsZdzfgRRow0XxlcKfF/ip7+6Dc6S+bOGNvTYcFyPXNAHUX3xEsDdNY+G7W48QXw4MdkP3af70v3QPcZqD/AIR/xX4l+bxLq40mzbrp+lHDkejynn6gcGuusrC0021W20+2itoE+7HEgVR+AqxQBh2Hgzw9pumy2NrpNv5Ey7Jt6b2lH+0x5Pr1rnTaax8PH8zTFn1fw0MmS0zunsh6of4kHp/9c131FAFLSNYsNd02O/0q4S4t5Bwy9j6EdQfY1drjdW8IXem6k+ueB5I7O9Y7rmxc4t7z6j+Fvcfpya0vDXi608Q+bbPG9hqlsdtzp8/EkZ9R/eX3FAHQUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAYPgweZ4YhvDy9/LLeOfXzJGYfkpA+gFb1YXgo7fCNnbH71nvtHHo0TtGf8A0Gt2gAooooAKKKKAOCtvGej+FdV1jRdQa48yC/eWKOG3eT5JVWbqBgfNI4x7Vb/4WXp7/wDHrouv3Pp5OnMc/mRVvw1qlhc+IPEaR3tubh9S2iASjeBHDFGfl6/eRq6egDjf+E+vJf8Aj28F+Im9POtRH/M0f8Jd4ol/49vAl43p519FH/OuyooA43+3PHkv+r8H2lv/ANdtTRsf98ij7T8R5f8AV2Hh63/67TStj/vmuyooA437L8Rpf9ZqXh+3/wCuMErY/wC+q5nUj4z1fxEfCsPieGYvETqEltZKq2sZ7Fs53H+6COvWu08X+JJdHt4LDSEW41vUG8qygPY95G9FXr/k1Z8K+G4/DekmFpDc3s7ma8um+9PKepJ9PQf/AF6AOf074f61punw2Nv40u4baFdqR29nHGAP1P41Z/4QC6l/4+vGfiNvXyrsR/yWuyooA43/AIVnpr/8fOr67c+vnagxz+QFcn4g8J+G5NTPh/w3pkmqa0wzLLPeStFZL/fkIbr6L/8AqPVa34kv9a1KXw74LZTcqdt7qZGYrIdwP7z+3b88bvhzw3YeGNN+yWCszud89xIcyTv3Zj3NAHA3Pwag07T7S50Vor3UbYZnhvlzDedyMfwegx+PrWz4Ug8G615lsfDWn2Gq2xxc2FzaoZIz6jI+ZfcV3lYHiXwjaeIhFcJI9jqlsd1tfwcSRn0P95fY0AaEWg6RB/qNKso/9y3QfyFXI4Y4RiKNEHoqgVyGk+LrvTdSXQ/G8cdnfMdttfIMW94PY/wt7H9OBXZUAc14w8OT6tDBqOjSLb65pzeZaTHo/rG3qrdP8mrfhXxJD4l0cXATyLuFjFd2rfeglHVT7ehrarifFOm3Xh7WP+Ew8PwNKyrt1WzT/l5hH8YH99ev0/HIB21FYdz4y0G00CDWbnUYo7O4TfCSfmk9gvUnsR2rCN/4r8Y8aRC3hzSW/wCXy5TN1Mv+wn8AI7n6g0AaF7qllpPxGs45LqFZNXtfszw7xuWSMl4iR2BDyDPqFHNdTXMWXw90Cy02e28iSee5H76+mfdcMwOQ4c/dIIBGMcgVa07WZrW6j0jxCyx3xO23uMbY74DuvYPjqnUdRkc0AbtFFFABRRXOajqlzrV0+j+HJCMHbe6kn3LVe6If4pT0wOF6nsCAZc/h9PGWvXurjUtQsEtSLKymsZ/LLbCTI/Q5Bc7f+2dP/sjx3pPOm6/ZaxEOkWpW/ltj03p1Pua66zs7fT7GG0s4liggQJGi9FA6VNQBxn/Caa3pnHiPwhfxKOs+nst0n1IGCBV/TfiD4X1RxHBq8EU2cGK5zCwPph8ZP0rpKz9S0HSdYUrqum2t3xjM0QYj6HqPwoAvqyuoZGDKRkEHINcv4p8VTWN1Hofh2Fb3Xrofu4v4Ldf+ekh7AeneuQ8ReH7DSNYh0jwBPqNrrk5Dm3tLxhBAnd5Qc4HoP/rZ0dJ8KeL/AAjPdXOlXGl61Jdv5ly92rxzyH0D5Ix9eKAOo8K+FYfDtvLNNM17ql2d95fSfelb0Hoo7Ct+uL/4T6807jxN4V1TTwPvTW6i5hHuWXp+Va+l+N/Des4Gn6zau7dI3fy3P/AWwaAN2iimySJDE0kzrGijLMxwAPUmgB1Q3d5bWFq9zfTx28EYy8krBVUfU1yVz48k1O5ex8Dae2s3Cna92x2WsJ93/i+g69jRa+Am1G6S/wDG9+2tXSncltjbawn/AGY/4vqevcUAMfxlqfiORrfwJp3nxZKtq16pS3Tsdo6uf84oj+GdhfrJP4svLnW9QlXBnkkMaxf9c0U4Wu0jjSKNY4kVEUYVVGAB6AU6gDhA3ifwNgSed4l0Jf4wP9MtlHqP+Wg/X6Cuq0TX9M8RWIu9Hu0uI+jAHDIfRl6g/WtGuW1vwPb3l8dW0K5fRdZHP2q3Hyy+0idGH6/WgDqa5PxH4Zu11IeIvCjLb6zEuJYm4jvk/uOPX0P/ANYivaeNrrRbpNO8e2i6fKx2w6jFk2tx/wAC/gPsf0rs0dZEV42DKwyrKcgj1oAxvDPie08S2LyQq1vdwN5d1Zy8SQP3BHp6HvW1XK+JvC9xPfLr/hmRbTXbdcc8R3af885B/I9vyIu+F/FNt4ktJB5bWmoWrbLyyl4eB/6j0NAG7RRRQAUUUUAFFFFABRRRQAUUEgDJ4AridR8U6h4ivptG8CBXMZ2XWrvzDb+oT++/6fzABpeJPGEGjTppun276nrc6/uLGHr/ALzn+Ffc/wD16q6H4PnfUk17xfcLqOsAfuowP3FmPSNfX/aP/wBc6fhvwrYeGoZDb77i8n+a5vZzulnb1J9PatugAoorl9W+IGjafdfYbJpdX1E8LZ6enmtn3I4Hvz+FAHUVja74s0Tw4g/ta/jilb7kC/PK/phBz/SsL7H428Tc311H4YsG/wCWFowlumHvJ0X8Oa2ND8F6H4ffzrKzEl2eWu7g+ZMx9dx6fhigDH/tnxh4l40DTF0Kybpe6muZmHqsXb/gXFWtP+HmlxXa32uyz69qA/5b37b1X/dj+6B+ddZRQAiqEUKoCqBgADgCs3V/Dmj69Hs1fTbe64wGdPmH0Ycj8DWnRQBxf/CB32kfN4Q8SXunKOlpdH7TB9AG5X680f8ACR+LtE48Q+GxqMC9bvRn3nHr5TfNXaUUAc7pHjzw5rMnk22pRw3OcG2ugYZAfTDYyfpmuirM1fw5o+vR7NX023uuMBnT5h9GHI/A1z3/AAgd9pHzeEPEl7pyjpaXR+0wfQBuV+vNAHaVgeJfCNn4iEVwsj2OqWx3W2oW/EkZ9D/eX2P6Vlf8JH4u0TjxD4bGowL1u9GfecevlN81aWkePPDmsyeTbalHDc5wba6BhkB9MNjJ+maAM/SvF13pmpJonjiOO0vWO22v0GLe8+h/hb2P6cCuxZgilnIVVGSScACqeraRY67pslhqtslxbyDlG7H1B6g+4ri9C8P319e3ukXmrS6h4Y0+fyo45l/eTuBzC7/xRoeD6njoCKAOh/t+81ZivheyS4hzj+0bpiluf9zALS/UYX/ap40fXZ/mu/E80LHqthZwxqPp5qyH9a3FVUUKihVUYAAwAKWgDBOk6/b/ADWfiRrgj+HULKJwfbMQjI+v6UkfiKfT5kt/FFmunl2CR3kUnmW0jE4A3YBQn0YAc4BNb9MmhiuIXhuI0likUq6OoZWB6gg9RQA+iudsHk8PaxDo87s+m3QP9nSyNkxOBk25J6jaCyE84DDsM9FQAUUUUAFFFFABRRRQBz0Df2H4rntpflstYfzrdz0W5C4eP/gSqHHqQ9dDVTVNMtdY0+SzvkLRPggqcMjA5VlPUMCAQR0IrGi1i98Pr9n8UbpbZeI9WijyhXt5yj/Vt6t9w9cr0oA6Sio7e5gu7dZ7SaOeFxlZInDKw9iKkoAKzde1b+yNNMkKCa8mYQ2kGf8AXTN91fp3J7AE9qh1DxNY2d0bK13ajqXaytMPIPTeekY/2nIFJpml3Ul6NW10xvf7SkMMRJis0PVVJ+8x/ifAzjAAHFAGaPhz4duNJt7fVNPiubqNP3t4oMcskh5Zyy4OSxJ/Gq3/AAhet6Tz4Y8WXkaL0tdRUXMf0BPKj6V2lFAHF/8ACQ+MdH41zwwmoxDrc6PNuP8A36b5jVvT/iN4avpvs8t8dPueht79DAyn/gXH611NVNQ0nT9Wh8rU7G3u4+yzxB8fTPSgCzHIk0ayROrowyrKcg/jWd4h1608N6LNqN8SVj4SNfvSufuovua5u9+H2j6TDNf6Nql94cESmSR7a5PlAAZJZGyCPbiuU0q28Z6/dWXiWaKLX7KxkcWEF0RaPNzxNtA2g+mT2/GgDtvCGgXa3E/iTxGg/tq/H+r6i0h/hiX09/f8c9ZXGL8SbOyYR+J9K1LQnzjfcQF4SfZ1zn8q6Ox17SdTtWuNP1K1uIUUs7xzKQgHUtzx+NAGgTgZPArhNS1q/wDGmoS6J4Sna306JtmoawnQescJ7t/tdv5xT3l98SLt7PSZZbLwxExS5vV+V74jqkfonqf/ANVdxYafaaXYxWWnQJb20K7UjQYAH+e9AEGi6LYeH9Ki0/SoBDBGOndj3Zj3J9av0UUAFFFFAFLV9HsNd02Sw1W3W4t5Byrdj2IPUH3Fcct7q/w8kEWrNPq3hocR3oXdPZDsJAPvL7//AFhXeSyxwxNJM6xxoMs7nAUepNcZeeN5dall07wRp39sS8pLeTDbaRfVv4/oOvvQB0763pkWkLqsl/brYMu9bgyDYR7H+lcq3irW/FTGHwRY+RZk4bWL9CseP+maHlvqePWubb4Y6joLW2rRpba+YJGmn0lk8uHnk+SucZHoRz6dq9E8N+JNN8SaaJ9LfaYvkltnG2SBh/Cy9v5UAefW/hez+Gvii01PVFTUtNusRPfSwgGxnJJDADhUOevb19fWAQygqQQRkEd6hvrG21Kwmsr6JZredCkiMOCDXHeGb668La0vhDXZjJCyltIvHP8Arox/yyY/3l/l+FAHcVXvrC01K0e11C3juIH+9HIuR9fr71YooAwBoWqWHGia7IkI+7b6jF9qRfYNuWT82NO2+LT8vnaKP9vypT/47u/rW7RQBgN4dvdQ48Qa1NcxH71rZJ9lhb2OCXI9i+D3BratbW3sbWO2soI7eCJdscUSBVUegA6VLRQAUUUUAFcp4k8T3Q1AeHvCyLc61MuXc8x2Sf8APRz6+g/+sDH4j8TXk+qf8I14RCzau4zPcHmOxQ/xN/tei/8A1gdfw14as/DOnGC1LTTyt5lzdy8yXD92Y/yHb8zQAzwz4YtfDdk6xu1ze3DeZd3kvMk79yT6egrboooAKydV8LaDrQY6rpVpcMesjRgP/wB9Dn9az9b8d6bpd5/Z1ikur6qeFsbIb2B/2z0UeufyrN/4RrxB4s+fxlffYbBv+YRp74DD0lk6t9Bx9KAOX1CCw0/UDp3wz1PWp9RU4NrZXHm2kP8Avs+VH5mp9R8JePdQe2uPEklp4gt4wGbTI7o2yluvOFVWP1/lXp2m6VYaNZLaaXaRWsC9EiXH4n1Puat0AcNa+O7PQrVLXWvDWo+H4IhgFbbzLdB7Mn+FdHpfirQtbwNL1a1uHPSNZQH/AO+Tz+la1YWqeCPDes5OoaNau7dZETy3P/AlwaAN2iuM/wCEBvNO58M+KtU08D7sNwwuYR7BW6fnSfa/iBpP/Hzp2ma9EP4rWY28pHuG+XPsKAO0orjV+Jem2rCPxFp2p6G+cbru1Yxk+zrnI966LTde0nWEDaVqVrd8ZxDKGI+o6j8aALV5Z22oWklrfQR3EEgw8cihlYfQ1xknhrW/CDvc+CZ/tdhks+i3ch2j/rk55U+x4+tdzRQBgeHvGWm+IJHtV8yy1KHibT7pdkyEdeO49x+lVfFPhWe9u49d8OSrZa9arhJP4LlP+ecg7g+vb+V7xD4S0vxLGpvYmiuov9ReQNsmhPqrD+R4rA/tvxD4KPl+KI21jR1+7qtsn72EdvNQf+hD9TQBt+FvFUHiO3liliay1O0Oy8sZPvxN6j1U9jW/XG6xo9v4nit/E3g3UIY9WgX9xdRNlJ17xSD07c8ijTPiDHfW/wBjbTrgeIo38qXSlGGRh1cseBH33H1A5JAIB2VFYC6DqGp/vPEOqTYP/Lnp0jQQr7FxiR/xIB/uinf8IR4XP+t8P6dM39+e2WVv++mBNAG7RWCfB2mwjdpEl1pEo+61jOUUf9szmM/ippi6tqOhzJD4k8uazZtqarAuxFJOAJkydnpvB2k9dvAoA6GqupanZaPYSXup3MdtbRjLSSHA+nufYVm+JPFdj4bjijlWS6v7k7bWxtxulmb2HYe9Y2meFL/Xb6LWvHbLNNG2+10pDmC19C399/c8fpgAq7dY+I/+s87R/DBbhfu3F+vv/cQ/r79u20/TrTSrGOz063jtreIYSOMYA/8Ar+9Qavr2laBbefrF/DaR9vMb5m+i9T+Armf+Er8Q+Ivk8H6IYbZump6qDHHj1SMfM31/MUAdlcXMFpbtPdzRwQoMtJK4VVHuT0rkJ/iEt/O1p4N0u412dTtaZB5dtGfeRv5Dr60tv8PIb24S88Yajca/cqdyxzHZbxn/AGYhx+f5V10EEVtAkNtEkMSDCxxqFVR7AdKAON/4RDXPEXz+M9bYW7cnTNMJiix6M/3m+n611GlaJpmh2v2fSLGG0i7iNMFvcnqT7mr1FABRRRQAUUUUAFFFFABRRRQAUUVHNcQ267p5UiX1dgB+tAElZmr+HNH16PZq+m291xgM6fMPow5H4Go7jxb4dtM/add06Mj+E3SZ/LOayrj4n+Drc4bW4nPYRRPJn/vlTQBk6z4Vu/COi3eo+E/EN/ZJbRl0sZyLiFz2RQ33cnAzzXaaNpqaPotpYI2/yIwrOesj9WY+5Ykn3NcTr/j7StY0tbSwttRkWS6ti05s2WJVE6FizHoMA16JQAUUUUAFFFFAGR4ospL7w5dC24u4FFxat/dmjO9P1AB9iRV+wvI9R022vYP9VcwpKn0YAj+dTkhVJY4AGSTWJ4IBHgHQQRj/AIl0GAew8sY/SgDcooooAKKKKACiiigAooooAxp/COhTztONPS2mc5eWzdrd2PqWjKkn61H/AMIbo7cTi+uV/uXWp3Myn/gLyEVu0UAV7HT7PTLYW+nWkFpAvSOCMIo/AVYoooAKKKKACiiuO8Xaxe3+oJ4S8ONt1C6Tdd3Q6WUB6t/vEcAe/uDQBT1GST4g+IX0ezkx4d06Qf2hMh/4+5RyIVP90dz/APWru440ijWOJFREAVVUYCgdABVPRdHs9A0iDTdNiEdvAuAO7HuxPck807VtXsdD0yXUNUuFgtohlmbv6ADuT6UAS31za2dhNcahJHFaxoWleU/KF968tg8HWHxE1g6qmlpo2gpkQvBGIp74/wB8jGFX8Mn+W7a6Xf8Aj+6i1PxJC9noUbCSz0puGn9JJvb0X/J7tVVECooVVGAAMACgDiYfCXifw/AkXhbxKs9rENsdlqsIdQB0HmLhgPoKf/wl/iLSOPEvhO5aMdbrSnFwp99n3gPrXaUUAc7pXj3w1rDeXa6rCk2cGC4PkuD6YbGT9M10QORkcis3VfDuj64uNW022uzjG6SMFh9G6j8DXnmv2GleEbgWvhTxBq1pqb8Q6VZt9rUntmNug+p6UAeq1yerePbWG+bTPDttJruq9DBan93F7ySfdUf54rkL/T/iRqthbHX7c3NgVzcWOnXKW80g/wBs4IP+6p/I1vaH4z8K6DAmmTabceGG/wCeN5atGGPrvGd31JoAnj8Gal4ilW68eah9ojB3JpVmSluncbj1cj3/AFFdja2tvZWsdtZwRwQRjakcahVUewFMs7+z1GATafdQ3UR6PDIHX8xVigArlvEPg77dfLrPh+5Gla5F0uFHyTj+5Kv8Q9+v1xXU0UAct4e8Y/br5tG8QW39la5F1t2PyTj+/E38Q9uv1xWj4n8OW3ibRns7gmKZD5ltcJw0Eo+64NL4i8M6d4msBbajGQ6HdDcRnbJA3Zlbt/KudtfEmp+ELxNM8bP59i5CWutquFb0WYfwt79/zNAGl4P8RXOorcaPrsYg1zTcLcp2mXtKvqD/ADPuK6euV8XaBPqK22v+HZFTWtPXfbup+W5j6mJvUEdPr71peHPEtl4i0MajAfJKZW5ikOGt3X7ytnpj+VAGxRWD/wAJZbXLEaJZXus4OPMs4wIvwlcqjf8AASaX+2NdHzHwxMV9Bew7vyzj9aAN2isJPFtlDIsWsW91o0jEKv2+MLGSe3mqWjz7bsmt0HIyORQAVx3iPxLe3uqHwz4QKvqjD/SrsjMdgh7n1f0FJ4h8R32paq3hjwewOoY/02+xmOxQ/wA39B/kbnhzw5Y+GdLFnYKWLHfNPIcyTuerMe5oATw34asvDGmfZbINJJI2+4uZDmSdz1ZjWvWRr3inSPDUCvq12sbv/q4FG6WU9PlUcnnv0rnhL4v8Yf6lX8LaS38bjdeSj2HSP+Y96ANnxB4z0jw6629xK9zfyf6qxtV8yZz2+UdPxxWN/Z3izxf82s3DeHNKb/lytHzcyr/tydFyOw+hFb2geEtH8No39m2379/9bdSnfNIfUsefwHFbVAGbonh3SvDtn9m0eyjtkP3mUZZz6sx5P41pUUUAFFFFABRRRQAUUUUAIyq6lXUMpGCCMg1zmpfD7wvqjGSfSIIps5EttmFgfXKYyfrXSUUAcZ/whet6Zz4c8X38SjpBqCrdJ9MnBApP7X8d6TxqWgWWsRDrLptx5bY9dj9T7Cu0ooA4+H4m6Esoh1mO+0Sc8eXqFqyZP1GR+NdLY6np+qw+Zp17b3kZHJhlVx+lWJoIriIxXESSxt1R1DA/ga5m++G/hi9l86PThY3HVZrFzAy/QLx+lAGb4l8MJ4dhuvEfhO7bSLuNd0lsi7oLticBDH2YkgAjufxrovDujyafDLe6kUl1e/xJeyr0BxxGvoiDgD6k8k1zX/CN3+neLNEsZfEV/qWmvK9y1rehXYGJcofM68OyHHtXfUAFFFFABTZY0mieKZFkjdSro4yGB6gjuKdRQB5vCND+HHiK8Oo29zc3F7h9MmVGnmePABt1z02ED0yrLzxWl9o8b+Jv+PWCLwvYN/y1nAlumHsvRPx5FaXjS5h0m00/Xp8qmm3sZkZVLHy5f3LDA5P+sBx/siqP/CzNOl/48NH12/8AT7Np7HP5kUAXNI8A6Lpl19tuI5NU1E8te6g/nSE+2eB+AzXTVxv/AAmuuT/8eHgfVX9PtMiQfzzR/bHj65/1HhbT7LP/AD86gJMf98CgDsqK43yPiPc/6y88PWQP/PGKWQj/AL64o/4Rnxlcf8fnjgxr/cttNjXH/Aic0AdlRnHWuN/4V/PP/wAhDxh4im9VjuxEp/ACj/hVnhqTm+jvb8/9PN7I2fyIoA6S41vSrPP2vU7ODHXzbhV/mayrj4g+E7XPma/ZNj/nlJ5n/oOaS3+HfhK1x5Wg2bY/56oZP/Qs1q2+g6PZ4+yaVYwY6eVbIv8AIUAc4fip4XckWVxdXreltZyN/NRR/wALCef/AJB/hLxFcejNZeWp/EmuyACgBRgDoBRQBxv/AAlHi+4/48vA0ij+/c6jHHj/AIDjNH2r4jXP+q07QLIH/n4nkkI/75rsqKAON/sr4gXP+u8SaZZZ/wCfax8zH/fZo/4Q3xBcf8f/AI51J/X7NBHB/LNdlRQBxv8AwrWyl/5CGu6/feon1BiP0AqSH4W+D4m3tpPnP3aa4kcn82xXXUUAYlv4M8M2uPI0DTgR0Y2yMfzIzWrb2VrajFrbQwj/AKZxhf5VNRQBS1nTl1fQ73TnbYLqB4g4/hJGA34HmotA1NtW0O3uZl2XGDHcx/8APOZTtdfwYGtKuc1GO48PapLrFjC8+n3GDqNrEpZ1YDAnRR1OAAyjkgAjkYIB0dFQ2d5b6hZxXdlPHPbyrujljbKsPY1NQAUUVW1DUbTSrN7vUJ0ggTALMepPQAdST0AHJoAzvFl1JDoEtrati81AiytvXfJxu/4Cu5z7Ka1ba3jtLSG2gXbFCixoPRQMCsfS7O51HVRruqRNAVQx2Nm/3oEP3nf/AKaNgcfwjjqWrdoAKKKKACiiigAooooAKKKKACiiigAooooAKKKoa5rVn4f0efUtSk8uCFc+7nso9STxQBm+LvE3/CP2McNnF9q1a+bybG1Xku/94/7I6k07wj4ZHh3TpGuZjdaneN519dt1lkPYf7I6Af41neEdGvbq+l8VeJUxqd4m22tzyLKDqEH+0epP/wBetfxJ4nsvDVkslyGnuZjstrOEZkuH7BR/M9v0oAm1/wAQaf4b0xr3U5diZ2xxqMvK3ZVHc1zel+HtQ8UapFr/AIyi8qKI7rDSCcrB6PJ/ef27foJvD/hi9vdVXxJ4xKy6nj/RbNTmKxX0Hq/qf/112NABRRVLVdY0/RLJrvVruK1gX+KRsZ9gOpPsKALtZGv+KNI8NW6yatdrG7/6qBBullPTCqOTz36Vzp17xL4v+TwranSNMbrqt9H+8cesUX9T+lbGg+CtL0O4N6RJf6m/Ml/eN5krH2J+6Pp+tAGPnxb4yXjf4X0h+/W8mX+Uf8x710WgeFtI8NQMmlWqpI/+tnc7pZT1yzHk89ula9FABUc9vDdQtFcwxzRt1SRQyn8DUlFAHJ3nw18Ozzm40+GfR7rtPpsxhI/AfL+lV/7H8caNzpWvWuswjpBqkOx8egkTkn3NdpRQBxf/AAnd/pfHirwvqFgB1ubUC5hHuSvI/WtvSPF+ga9tGlatbTu3SLftk/74bB/StmsTV/Bvh7XcnU9JtpZG6yqmyT/vpcH9aANuobu0t7+zltb2FJ4JV2vHIuVYe4rkf+EG1PSvm8LeKr+0UdLa9xcxfQBuVH50f23420bjWPD1vq0K9bjSpsNj/rm/JP0oApzwan8NQ11p5l1LwuvMto75msR6xk/eX2/+uau6Z4QttW1i68QavZtbpfFHXSyxCNt+7JOmcNIc529F9zzTbXXYPHGuW1hDbXltbafi7v7e8gMbFwcQowPUbgz/AFjHvXa0AIqhVCqAABgADpS0UUANkjSaJo5kWSNxhkYZDD0Irh/E2laz4d0eU+Ebv7Npr4W4gKGRrKMn5pYOcjAySnTuuCMHuqKAOZsv+Ea8A+F42+2QwWjjzDcSPue6YjO7jlyfbt7Vmf2z4o8Xnb4dtDoWmN11K+TM0g9Y4u31P6UmgeENGsPGmqR3Nr9puIPLuLFp2LrBA+QERTwNro49gVruqAOe0HwVpWhTm82yX2pPzJf3jeZKx9ifu/hXQ0UdOtABRVO41nTLTP2rUbSDHXzJ1X+ZrLuPHnhW2z5niCwOP+ec4f8A9BzQB0FFce/xU8IhtkGoyXT/AN2C1lY/+g4pv/CyLeX/AI8fDniK89DFp52/mSKAOyorjf8AhMfEM/8Ax4+BdRf0+03EcH880f2t4/uP9T4Z02z/AOvm/wDMx/3wKAOyorjfI+I9x/rLzw9Zg/8APGKWQj/vrij/AIRvxpcf8fXjdYR3S20yMf8AjxOaAOyorjf+ECvp/wDj+8aeIJPUQXCwg/kDR/wq/Qpf+P651W+9ftN+5z+WKAOquNQs7T/j6u4IcdfMkC/zrMuPGnhm1z52v6cCOoW5Rj+QNULf4ZeDrb/V6HA3/XR3k/8AQmNadv4T8PWuPs+hadGR3FqmfzxQBkTfFLwdC23+2FlbssUEj5/Jaj/4WZpkv/HjpOuX3p9m09jn88V10NvBbrtt4Y4h6IoX+VSUAcb/AMJvq8//AB4+CNZf0+07IP5k0f2746uP9R4PtrUdjc6mjfoorsqKAOHs5PEw8b6RP4og0yCKSK4t7dbJ3Zt7KrkNu46RHGPQ13FZHiaynutJE9gm+9sZVu7ZP77J1T/gSlk/4FV/T7+31PToL2zffBOgdD0OD2I7EdCOxoAsUUUUAFFFFAGF4v8An0i1tx9+fUrNFH0uEZvyVWP4Vu1z8jf2z4yhjj+a00UGSVh0a5ddqr9VjZif+ui+ldBQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAYlz4ajW7kvNFu5tJu5W3SmABopj6vEflJ9xhvemh/Ftv8AL5Ojahj+PzpbXP8AwHZLj863aKAME/8ACWXPykaPpwPV1eW7I+gIiGalsfDcEF4t/qNzPql+n3J7ogiL/rmgAVPTIG4jqTWzRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUANllSCF5ZnVI41LO7HAUDkkmuE0uKT4geIE12+iZdA09z/ZsDj/j5kBwZ2HoO3/68rrVxJ478QP4b06V00aycHVrqM481h0gU/wDoR/w50PEHiUaK1v4d8L2iXesyRhYLVOI7ZAMB5P7qgdu9AFvxT4ri8PxxWtrC1/q938tpYxfec/3m/uqO5/8Ar4r+GPCk1nePrniSZb/XrgfNJ/BbL/zzjHYe/f8APM/hjwnHoZlvr+c6hrV1zdX0g5P+yg/hUen/ANYDoWYKpZiAAMkk9KAFqK5uoLK2kuLyaOCCMbnkkYKqj1JNcpqHj1bi8fTfB1k2u36nDvGdtvB7vJ0/AdfWo7bwNcavcJfeO7/+1ZlO5LGLKWkJ9l/jPufxzQA2TxpqPiKZ7XwHp/2lAdr6rdgpbR9jtHVyPb9RVrSvAVrHfLqniS6k13VOomuh+7i9o4/uqP8APFdVFFHBEsUMaxxoMKiDAUegFOoAKKKKACiiigAooooAKKKKACiq15qNlp6b9QvLe1XrunlVB+prnbv4meFLWTyk1RbuXtHaRtKW+hUY/WgDq6K4z/hOtTvf+QF4N1e5z917sLao30LZ4o8z4iaj9yDRNHjP/PR3nkH5fLQBraN+88VeIpW+8k0EA/3RCrj9ZGrdrkPDNrqejeK9Qs9d1NdRn1K3ju4plgEIJjPlyKFHoDDz/tV19ABRRRQAUUUUAcf4ottXm8ZaQvh6/hsLmayulllmh8wFFeEgY9csfzNN/wCEV8WT/wDH746nwf4bbT4o8fjmtSM/bfiBK68pplh5JP8A00ncOV+oWJD/AMDFb1AHG/8ACvXm/wCP/wAW+I5/VVvfLU/gBR/wqvwzJ/x+xXl8f+ni9kOfyIrsqKAOat/h14RtseXoNo2P+eimT/0ImtO38OaJaY+y6Pp8OP8Annaov8hWlRQA1ESNdsaqijsowKdRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAVzlxZ3vh3UZr/AEi3e80+6kMl5YRn543PWWId89WTueRzkN0dFAFTTdVsdXtvP025S4jB2ttOCh/usp5U+xANW6y9Q8OaXqdz9qntzHdgYF1bSNDNjsN6EMR7E4qt/wAI5eJxB4p1qJP7v+jSfq8LH9aAN0nAyeBXPT69Lq8zWPhYrM2Sk2pY3W9t2OD0kkH90cA/eI6F58IWNx/yF7q+1Ze8d7cExH6xLtQ/iprciijgiWKFFjjQYVEGAo9AKAK2l6Zb6Rp6WloG2KSzO53PI5OWdj3Ykkk1boooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACorm5gs7d7i7mjghjGXklcKqj1JPAqprOrxaNYiZ43nmkcRW9vF9+eQ9EH5Ek9AASeBVCx8PPdTpqHido76+U74oQM29p7Ip6sP755PbA4oAUeLba450nT9S1Re0ltbbYz9HkKK31BIpf+Ekuo+bnwxrUCd22wS/pHKx/St2igDN03xBpmrSvDZXQNxGMvbSo0UyD1aNwGH4isTxfrt413B4Z8NSD+2L5cvL2s4f4pD7+n/wCrO9qmi2GsxIt9AGeM7oZkJWWFv7yOOVPuDXn82j634c1aTStOnU3PiC4bPiK6fMioFz5W3GPMAB24wD2AxwAXze/2HFD4K8AxLcalGubq7kGUtc/ekkPdz/d/+sK6Xw14WtPDdtIY3e6vrk77u9m5knb1J7D0FZC3vhn4a6Wmnxu8t5Kd3kRDzbq7kP8AEQOcn1OB6VALHxZ4x51WZvDekt/y6Wz5upl/23/gBHYfQigDS1vx1p2mXv8AZunxy6vqx4WxshuZT/tt0QeufyrOHhXXPFTCbxtffZ7InI0ewcqh9pJBy30HHpXTaJ4e0rw7Z/ZtHs47ZD95gMs59WY8k/WtKgCvYafZ6XZpaadbRW0CfdjiUKB/9f3qxRRQAUVlX/inQdLz/aGsWUDDqjTru/75zmsN/ihoMrFNIi1HWHHG2wsnfn6nFAHY0Vxn/CT+Lb//AJBPg14EPSXUbtY8fVBzR/Z3xB1D/j71zStJU9rG1MzAfWTvQB2dUb/XNK0sH+0tStLTHaadUP5E1zX/AArv7ZzrviXW9Rz1j+0+VEf+AqP61fsPh54T04g2+h2rsP4p1Mxz6/OTQBUn+KHhhZTFZXNxqUw/5ZWVs8hP44A/Wo/+Ex8Q33/IF8E6gVPSTUJUtse+05NdfBbw20QjtokiQdFjUKB+AqSgDjPI+Imo/wCsu9F0eM9PJjeeQfXd8tH/AAgd/e/8h3xhrN3/AHktnW2Rvqqg8V2dFAHK2fw08J2b+YdIjuZTy0l07TFvqGJH6V0VpYWdhHssbSC2T+7DGEH5CrFFABRRRQBkeIdOuLu3gvNMC/2lp8nnWwZsLJxh42Poykj2O09qt6Xqlvq9gt1aFsElXjcYeJxwyMOzA8EVcrG1DQ5WvjqWiXQsNQYAS5TdDdAdBIvGSOgYEMPUjigDZorAHiO6svk13Rby3YdZ7KNruFvcbBvH/AkFO/4TTQu11MW/uC0mLf8AfO3NAG7Wdresw6LZrI6NPcTOIrW1jPz3Eh6Kv8yegAJPAqi2vajf/JoOi3LZ/wCXnUVNrEvvtYeYfptH1FWNL0EWt4dS1O4OoaqybDcsm1YlPVIkydi8DPJJxyTxQBJoOlyaXppW6kWa9uJGuLuVRgPK3XHsAAo9lFadFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAc/pi/2v4mvdUl+aCwZrGzU9Awx50n1LfJ9EPqa6CsLwT83gvTZj964jNw/u0jF2/VjW7QAUUUUAFUNa0qLW9InsZnaIyDMcyfehkHKSKezKwBH0q/RQByHgLRNK0/Qo9QW3A1Nt0d9dXD+ZL5yMUkG89BuU9MZGK07/wAa+GtMyL3XLJGHVFmDsP8AgK5Nc7B4H0PXfFPiGXV7eW5EV+vlxGd1jG63icnapHO5mP411Fh4V0DS8fYNGsoGHR1gXd/30RmgDC/4WZpt1xoel6xrGfutaWTbPxLYwKP7d8cah/yDvCttYIekmo3oP5onIrs6KAOM/sPxzqH/ACEPFNpp6n70enWQb8mfkUf8K00+6513VtY1fP3kur1gn4KuMD8a7OigDCsPBHhnTMGz0OyVh0d4hIw/4E2TW4iLGgVFCqOAAMAUtFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFAGD4OPlaE2ntxJptzLaMPQK5KfnGUP41vVzmqSf8I5rh1psjTbxUhvyB/qXHCTn/ZwdrHsNp6Ka6MEMAQcg8gjvQAUUUUAFFFYniG/nPl6NpLkalfKQJF/5dYujTH0x0X1bA6ZwAM8Kn7Smq6kOUv9RkkjPqkarCp+hEWR9a3qgsbODTtPt7K0QRwW8axRoOyqMAVPQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAjoskbJIodGBDKwyCPQ1zseman4byugqL/Tc5GnTSbZIB6QueNv+w2AOzAcV0dFAGCPGWkRcam8+lSDquoQNCB/wMjYfqGIpf8AhN/DB/1Ou2Ny39y1mEzH/gKZNbtFAHPtrGrat+70HTZLWJuDf6lGY1UeqQnDsfZgg9z0rQ0nR4dJikIkkubqchri7mIMkzDoTjgAdAowAOgrQooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigBksUdxC8M8ayxSKVdHUFWB4IIPUV5faanqPhXwnZweGoNHtrKHRrnUXhe1YNuicFgAjqDvD4zgbTz833a9QlijuIXhnjWWKRSro6gqwPBBB6isUeB/CYjWMeF9GCJG0SqNPiwqEklQNvQkkkdMmklaakNW6mFL4l8SLq11FHNo5gTUmsIkMEnmjdbiZHP7zBK5+ZRjcoLApjFZ914n8Sanb28Vlf6bZyvBpF2SLSSQqbiVldD+9GVygI6fKSO+4dPoPhC30fXNQ1OWHTJJ7hwLWS308QyWsARVEO/cxK4RTgbRnPHpZHg3wwsU0a+HNIEc67ZlFjFiQbt+GG3kbgG5789auNla/l+G6+f9aC/r8d/u2X5GfrGm2U/xJ8OXdxZ28t1FbXeyd4lLpjy8YY8j7zfmfWode8VahZ+JLC20g2lzZNLDHdhYWkZPMlMW4yb1VAGGMASMxz8qhS1beq+FvD+vTRy65oWmalJEuyN7yzjmKL6AsDge1RT+DPC90bU3PhvSJjZxrFbGSwibyEU5CplflAPIAqIqz17g9n6GFP4n1zTtF1C4vWsLm70kzLeQw2rRCTJHkFS82EBVtxLEg8jK4zUejeIPGF7rmn6dqkGn6e7pcSXCyWwaRlieHBCx3LrGWSbGCzYIDcg7a6yx0+VLe4GqyW13PcsfOaK28tGTGAu0sxPy8HJOTnoOBDZeFvD+mzW82naFptpLaqy27wWcaNCGJLBSB8oJJzjrmnHR3fb8QOd8T6nrsnio6XpKa4sMFpHOW0YWG5y7uPn+1noNgxsHc5PQVd8VaLDrGg6eL3Q7PWrqGRCkGrRxn5ivIZlR1UnHJVSPTitnV/DuieIBENe0ew1MQ5MQvbVJvLzjONwOM4HT0ovvDuianDaw6lo9hdxWZBtkuLVJFgxgDYCPl6Dp6UraWG3rdHJaJoUPiPwppYstRk0Yadfea9hpjh4YnRwfIZZYwwAxkDauCQQOBXfVVt9L0+0v7m+tbG2gu7vb9puI4VWSbaMLvYDLYHAz0q1VE2sFFFFIYUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAV5Hq+kzrP8AEaPTPDOg3VncOGupbq4MMhYWkcmdggcPhiXBLD5iTx1r1yslvCnh11vlfQdMYaiwa9Bs4z9qIO4GTj5znnnPNLv6f5FwlynMwanLo11daXpltpVmAtha2UotSiQiRXwJAGG8KVO1QUyWC8ZzXL3d1J/wrDXNFgTSY4YoNXup0kjJjdVuplCxAEBcNhtxDY+UY5yPTY/CPhuK3uIIvD+lJDdRpDPGtlGFlROFVhjDKOwPApH8I+G5bdLeTw9pTwxu0iRtZRlVZhhmA24BI4J71fNrd9/6X36+exELwSXa34X/AD28tzjtW8ceJrWbWpdO061TTdNhnRLi6WMr50UXmDOLgO24AjZ5a4BDbiBg9Xol3e6muq2OvpZTNBKIysETCNo3iV9hDE7sbypbgNjO1elWJ/Cfh26laS50DS5neAWzNJZRsWhGMRkkcqMD5enAqW08O6Jp63IsNHsLYXiBLkQ2qJ56gYAfA+YAEjB7Gp+zby/r+vkJaJf12/I5vwlLZeHPhPYyWiWVkEs3ljjYeXGzhWckhQSehJwCcZODUGm+KvEDW8d1qxs4LW11I2l+zWDxO0bRq0Uir57eX8zqCGLHByQhBUdBZ+CfCmnPK2n+GdHtWmiaGUwWESF42+8hwvKnuDwabD4T0/TQlv4e0/SdJsZJA97Bb6eqm428qAVKhSD1JVuMgY60Lz/r+kN7aef5GDN4m8WR+I4dKhsrSdIFhF/dCFEjRps7WAa5DqFOPlCvvIIDKel/T/EGsn4Xv4k1AWc96+nfboobWFkRf3QYKdzktznnK8EDtk717oOkaneRXepaVZXdzCjJFNPbpI8asMMFYjIBBIIHXNT2On2el2MdlplpBZ2sQxHBbxCNEGc8KAAOTS+y11Gnrc5vwe+tSXEr6ufETRtECh1X+zfKJz/B9lJbP+8SMe9YVvpWm6F49tXi8PaLpMVzeSpFqkIMM8j4z5OEiC4b/ak+bnAJFdrpXhjQNCuJJ9E0PTdOmmXbJJaWkcTOM5wSoBIzUa+EPDSWNzZp4d0pbW7cSXEAsYwkzA5DOu3DEHnJquqZPdDfDnh5vD/9ob9WvdS+23Rud15sLRkqAVyqrkccZ6DA7VtUgAVQqgAAYAHalpDCiiigAooooAKKKKACiiigAooooAKKKKACuE8V2U3/AAs7w5faVo2l3+pJYXoD30xgKqGhwRIsUjZG9gBj+Nueee7qjNomlXOrQarcaZZy6jbrshvJLdWmiXnhXIyByeAe5o63/rYpO1zz7w6y+GorPytD0izE2o6k9zb2UG5kkjEhUxSYT+FSnKZIOOMcvsdXvrfxTqF2JNI+3atZaYsVzDE3kp5slzsMg35kIACghl3/AC42ZwO0s/B/hrT7iKfT/Duk2s0EjSxSQWMaNG7DDMpC8EgAEjqBQnhDw1HDPFH4e0pI7lCk6LZRgSqW3EMNvzDdzg9+aF/l/wAH7/wJf2rdb/mn+H4nn2k+I9b0XTYNM8OaTa3JS81C5uRF5SwiJb6RWCl5o/LH+0BJt4yprqPDXiLXNQ1eOHVW057aZ76KP7LC6Nm3n8sOSzsMMD93BwRndztG2fCnh0wwxHQdMMdvP9ohT7HHiOX/AJ6KMcNwPmHPFFr4U8O2F5Fd2Og6ZbXMLO0U0NnGjxs/DkMBkE98dad+/n/X9eoS1bfd/wCf9fgZGgWem6b4v8XXyW1raO9xC09wI1QkeQrEs31JPPqTVBvE3ie81nVbHSRp7BbeeSwlkspDGXhlVGjZjKpkJB5IVFUnAMmDXQy+C/C0+qHUp/DWjyX5kEpunsIjKXByG3lc5z3zmkl8K6Xb3dxqmh6Vo9hrcu4jUX05Xfc33ixUozZ5z8wzUJWVvUbf6GBrHi3Xls7OXw3Db6i+qt/oIW2XKqkQZ9wkuIwxLZGAwKgHhsGtPwprGt6zqGotqRsIrW1kSNIIIy0gLwQyjMokKtt8xlyBhuCMdDr3GgaXfaPHpmq2FpqFooBMN1bpIjMP4tpGM55qWy0jTdNmuJtO061tJbpg07wQKjTEdCxA+Y/WrWjZPRHHaTqfiDVvEsk7p4hjso7+WApbjTfsQSNynzbibjkLk4wcngAYqLx/o9guoPrDeF9G1GSO3Uz3t6hSSBQ3EiukTuSvHQjA5zXXS+F9An1pdYn0PTZNTVgy3z2kZmBAwD5hG7IAwOalbQdHbWDqzaVYnUinlm9NsnnFcY278bsY4xmptol2KfxNmbp/h63PiweJ47i3uDLpqWkcsaMXkjzvyz7yrDPIwoPJyTxjoaq6dplhpFillpNlb2NrHkpBbRLGi5OThVAAyTmrVUyQooopDCiiigAooooAKKKKACiiigAooooAK5jx9M8Xh+3Vpngs5tQtob6VGKlYGlUPlh0U5Ck9gT06jp6bJGk0TRyorxupVkYZDA9QR3FJ/wBf15h0aPONQVfB3jA23gm1sLG0Nik2oWccO2BGNzHHG4jQqFd0Mw3d/LXOQoq62r+JL3xFDFaappsUUM99Cy/YZHWTywuzOJhyu7B9SD0yMdTaeGtCsNKn0yx0XTrbT7nPn2kNoiRS5GDuQDByAAcjoKgPhTRYLMppWj6VZzRkyW8gsUKwy7NiybV2nIXA4IOBjIpS5mtH0/W/5aIa3/rt/TOWsPGHiSWLTZLltHkW9i0+6YW8Un7uO4k8tk5k65IZXxjhl28bqs6DrOv6v4qsGn1CwNkqajFNDBaOPNMN0IlYMZSAcbeoOPn/ALw27Gg+DNL0nwouiXdhpl2kiAXpjsFjjvGAA3vGS2WIAySTyKv23hvQrKaCWz0XT7eS3d5IXitUVoncbXZSBwWAAJHUDmtbpP7/AOvkLp939fN/cYugWem6b4v8XXyW1raO9xC09wI1QkeQrEs31JPPqTVBvE3ie81nVbHSRp7BbeeSwlkspDGXhlVGjZjKpkJB5IVFUnAMmDXQy+C/C0+qHUp/DWjyX5kEpunsIjKXByG3lc5z3zmkl8K6Xb3dxqmh6Vo9hrcu4jUX05Xfc33ixUozZ5z8wzWSVlb1G3+hgax4t15bOzl8Nw2+ovqrf6CFtlyqpEGfcJLiMMS2RgMCoB4bBrT8Kaxres6hqLakbCK1tZEjSCCMtIC8EMozKJCrbfMZcgYbgjHQ69xoGl32jx6ZqthaahaKATDdW6SIzD+LaRjOealstI03TZribTtOtbSW6YNO8ECo0xHQsQPmP1q1o2T0Rx2k6n4g1bxLJO6eIY7KO/lgKW4037EEjcp824m45C5OMHJ4AGKi8f6PYLqD6w3hfRtRkjt1M97eoUkgUNxIrpE7krx0IwOc110vhfQJ9aXWJ9D02TU1YMt89pGZgQMA+YRuyAMDmpW0HR21g6s2lWJ1Ip5ZvTbJ5xXGNu/G7GOMZqbaJdin8TZm6f4etz4sHieO4t7gy6alpHLGjF5I878s+8qwzyMKDyck8Y6GqunaZYaRYpZaTZW9jax5KQW0SxouTk4VQAMk5q1VMkKKKKQwooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigD/9k=\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from IPython.display import Image\n",
    "Image('NN_architechture.JPG') "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "input_size = 4 \n",
    "layers = [4,3]\n",
    "output_size = 3\n",
    "EPOCHS=50"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "def softmax(a):\n",
    "    ea = np.exp(a)\n",
    "    return ea/np.sum(ea,axis=1,keepdims=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "class NeuralNetwork:\n",
    "    def __init__(self,input_size,layers,output_size):\n",
    "        np.random.seed(0)\n",
    "        model = {}\n",
    "        model['W1'] = np.random.randn(input_size,layers[0]) \n",
    "        model['b1'] = np.zeros((1,layers[0]))\n",
    "        \n",
    "        model['W2'] = np.random.randn(layers[0],layers[1])\n",
    "        model['b2'] = np.zeros((1,layers[1]))\n",
    "        \n",
    "        model['W3'] = np.random.randn(layers[1],output_size)\n",
    "        model['b3'] = np.zeros((1,output_size))\n",
    "        \n",
    "        self.model = model\n",
    "        \n",
    "    def forward(self,x):\n",
    "\n",
    "        W1,W2,W3 = self.model['W1'],self.model['W2'],self.model['W3']\n",
    "        b1,b2,b3 = self.model['b1'],self.model['b2'],self.model['b3']\n",
    "        \n",
    "        z1 = np.dot(x,W1) + b1\n",
    "        a1 = np.tanh(z1)\n",
    "        \n",
    "        z2 = np.dot(a1,W2) + b2 \n",
    "        a2 = np.tanh(z2)\n",
    "        \n",
    "        z3 = np.dot(a2,W3) + b3\n",
    "        y_ = softmax(z3)\n",
    "        \n",
    "        #Store the outputs of each layer for back-propagation\n",
    "        self.activation_outputs = (a1,a2,y_)\n",
    "        \n",
    "        return y_\n",
    "        \n",
    "    def backward(self,x,y,learning_rate=0.001):\n",
    "        \n",
    "        W1,W2,W3 = self.model['W1'],self.model['W2'],self.model['W3']\n",
    "        a1,a2,y_ = self.activation_outputs\n",
    "        \n",
    "        delta3 = y_ - y\n",
    "        \n",
    "        dw3 = np.dot(a2.T,delta3) #)h2+1,C) => (h2+1,m) X (m,C) \n",
    "        db3 = np.sum(delta3,axis=0)\n",
    "        \n",
    "        delta2 = (1-np.square(a2))*np.dot(delta3,W3.T)  #(m,h2+1) => (m,h2+1) *(m,C X C,h2+1)\n",
    "        dw2 = np.dot(a1.T,delta2)  \n",
    "        db2 = np.sum(delta2,axis=0)\n",
    "        \n",
    "        delta1 = (1-np.square(a1))*np.dot(delta2,W2.T) \n",
    "        dw1 = np.dot(x.T,delta1)\n",
    "        db1 = np.sum(delta1,axis=0)\n",
    "        \n",
    "        self.model['W1'] -= learning_rate*dw1\n",
    "        self.model['W2'] -= learning_rate*dw2\n",
    "        self.model['W3'] -= learning_rate*dw3\n",
    "        \n",
    "        self.model['b1'] -= learning_rate*db1\n",
    "        self.model['b2'] -= learning_rate*db2\n",
    "        self.model['b3'] -= learning_rate*db3\n",
    "        \n",
    "        \n",
    "        \n",
    "    def predict(self,x):\n",
    "        y_out = self.forward(x)\n",
    "        return np.argmax(y_out,axis=1)\n",
    "    \n",
    "    def summary(self):\n",
    "        W1,W2,W3 = self.model['W1'],self.model['W2'],self.model['W3']\n",
    "        a1,a2,y_ = self.activation_outputs\n",
    "        \n",
    "        print(\"W1 \",W1.shape)\n",
    "        print(\"A1 \",a1.shape)\n",
    "        \n",
    "        print(\"W2 \",W2.shape)\n",
    "        print(\"A2 \",a2.shape)\n",
    "        \n",
    "        print(\"W3 \",W3.shape)\n",
    "        print(\"Y_ \",y_.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "def loss(p,y):\n",
    "    l = -1*np.mean(np.multiply(y ,np.log(p)) + np.multiply((1-y),np.log(1-p)))\n",
    "    return l"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "model = NeuralNetwork(input_size=input_size,layers = layers,output_size= output_size)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.preprocessing import OneHotEncoder\n",
    "def train(X,Y,model,epochs,logs=True):\n",
    "    \"\"\"Creates and Model and Trains it!\"\"\"\n",
    "    \n",
    "    training_loss = []\n",
    "    val_loss = []\n",
    "    val_acc = []\n",
    "    \n",
    "    for ix in range(epochs):\n",
    "        Y_ = model.forward(X)\n",
    "        l = loss(Y_,Y)\n",
    "        model.backward(X,Y)\n",
    "        if(logs and ix%500==0):\n",
    "            print(\"Epoch %d Loss %.4f\"%(ix,l),end='\\n')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 0 Loss 0.7769\n",
      "Epoch 500 Loss 0.1880\n",
      "Epoch 1000 Loss 0.1130\n",
      "Epoch 1500 Loss 0.0839\n",
      "Epoch 2000 Loss 0.0729\n",
      "Epoch 2500 Loss 0.0677\n",
      "Epoch 3000 Loss 0.0646\n",
      "Epoch 3500 Loss 0.0627\n",
      "Epoch 4000 Loss 0.0614\n",
      "Epoch 4500 Loss 0.0605\n"
     ]
    }
   ],
   "source": [
    "train(X_train,y_train,model,5000)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "W1  (4, 4)\n",
      "A1  (105, 4)\n",
      "W2  (4, 3)\n",
      "A2  (105, 3)\n",
      "W3  (3, 3)\n",
      "Y_  (105, 3)\n"
     ]
    }
   ],
   "source": [
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training Accuracy 96.1905\n"
     ]
    }
   ],
   "source": [
    "outputs = model.predict(X_train)\n",
    "training_accuracy = np.sum(outputs==y_train_without_one_hot)/y_train_without_one_hot.shape[0]\n",
    "print(\"Training Accuracy %.4f\"%(training_accuracy*100))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Testing Accuracy 97.7778\n"
     ]
    }
   ],
   "source": [
    "outputs = model.predict(X_test)\n",
    "testing_accuracy = np.sum(outputs==y_test)/y_test.shape[0]\n",
    "print(\"Testing Accuracy %.4f\"%(testing_accuracy*100))"
   ]
  }
 ],
 "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": "3.8.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}