SVM-SVR-LR.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 234,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Install the dependencies\n",
    "import numpy as np \n",
    "from sklearn.linear_model import LinearRegression\n",
    "from sklearn.svm import SVR\n",
    "from sklearn.model_selection import train_test_split\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 235,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.read_csv('WTI_price.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 236,
   "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>Date</th>\n",
       "      <th>Price</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>2-Jan-86</td>\n",
       "      <td>25.56</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>3-Jan-86</td>\n",
       "      <td>26.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>6-Jan-86</td>\n",
       "      <td>26.53</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>7-Jan-86</td>\n",
       "      <td>25.85</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>8-Jan-86</td>\n",
       "      <td>25.87</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       Date  Price\n",
       "0  2-Jan-86  25.56\n",
       "1  3-Jan-86  26.00\n",
       "2  6-Jan-86  26.53\n",
       "3  7-Jan-86  25.85\n",
       "4  8-Jan-86  25.87"
      ]
     },
     "execution_count": 236,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 237,
   "metadata": {},
   "outputs": [],
   "source": [
    "# A variable for predicting 'n' days out into the future\n",
    "forecast_out = 472 #'n=30' days"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 238,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Create another column (the target ) shifted 'n' units up\n",
    "df['Prediction'] = df[['Price']].shift(-forecast_out)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 239,
   "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>Date</th>\n",
       "      <th>Price</th>\n",
       "      <th>Prediction</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>8472</td>\n",
       "      <td>13-Aug-19</td>\n",
       "      <td>57.05</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>8473</td>\n",
       "      <td>14-Aug-19</td>\n",
       "      <td>55.16</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>8474</td>\n",
       "      <td>15-Aug-19</td>\n",
       "      <td>54.51</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>8475</td>\n",
       "      <td>16-Aug-19</td>\n",
       "      <td>54.83</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>8476</td>\n",
       "      <td>19-Aug-19</td>\n",
       "      <td>56.24</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           Date  Price  Prediction\n",
       "8472  13-Aug-19  57.05         NaN\n",
       "8473  14-Aug-19  55.16         NaN\n",
       "8474  15-Aug-19  54.51         NaN\n",
       "8475  16-Aug-19  54.83         NaN\n",
       "8476  19-Aug-19  56.24         NaN"
      ]
     },
     "execution_count": 239,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.tail()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 240,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['Date'] = pd.to_datetime(df.Date) # ,format='%Y-%m-%d'\n",
    "df.index = df['Date']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 241,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x112738780>]"
      ]
     },
     "execution_count": 241,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA6UAAAHSCAYAAAAUmW0WAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy8QZhcZAAAgAElEQVR4nOzdebxcdX3/8fd3trtmz03ISkIIiQk7MQoSZVFBEUGtVqyVtlatUq3W1sLPWqpUsWpb69pa3GqRShVEZV/FgATCmhASCEnIvif33txttvP7Y+acOTNzZu7s6+v5ePhgzjrfO7l3nM98vt/Px1iWJQAAAAAA6sFX7wEAAAAAANoXQSkAAAAAoG4ISgEAAAAAdUNQCgAAAACoG4JSAAAAAEDdEJQCAAAAAOomUO8BSNL06dOtBQsW1HsYAAAAAIAqePLJJw9altXndawhgtIFCxZo7dq19R4GAAAAAKAKjDGv5DrG9F0AAAAAQN0QlAIAAAAA6oagFAAAAABQNwSlAAAAAIC6ISgFAAAAANQNQSkAAAAAoG4ISgEAAAAAdUNQCgAAAACoG4JSAAAAAEDdEJQCAAAAAOqGoBQAAAAAUDcEpQAAAACAuiEoBQAAAADUDUEpAAAAAKBuCEoBAAAAAHVDUAoAAAAAqBuCUgAAAABA3RCUAgAANJg3fPVBLbj6dm0/NFzvoQBA1RGUAgAANJhXksHol+54oc4jAYDqIygFAABoUCORWL2HAABVR1AKAADQoKx6DwAAaoCgFAAAAABQNwSlAAAADcrUewAAUAMEpQAAAA2K6bsA2gFBKQAAAACgbghKAQAAAAB1Q1AKAAAAAKgbglIAAIAGZVmsKgXQ+ghKAQAAAAB1Q1AKAAAAAKibcYNSY8wPjDH7jTHrPY79jTHGMsZMd+27xhiz2RizyRhzUaUHDAAAAABoHYVkSn8k6eLMncaYeZLeJGm7a98ySe+VtDx5zXeMMf6KjBQAAKDNxFlTCqANjBuUWpb1sKTDHof+TdJnlN7X+TJJ/2tZ1phlWVslbZa0shIDBQAAaDfxeL1HAADVV9KaUmPM2yXtsizr2YxDcyTtcG3vTO7zuseHjTFrjTFrDxw4UMowAAAAWs6xsajzmEwpgHZQdFBqjOmW9FlJ/+B12GOf57upZVnfsyxrhWVZK/r6+oodBgAAQEvadWTEeUxQCqAdBEq4ZpGkhZKeNcZI0lxJTxljViqRGZ3nOneupN3lDhIAAKBduAPRODEpgDZQdKbUsqx1lmXNsCxrgWVZC5QIRM+0LGuvpF9Jeq8xpsMYs1DSYkmPV3TEAAAALSoWtzQcjjnbZEoBtINxM6XGmJsknSdpujFmp6RrLcv6vte5lmU9b4y5WdIGSVFJV1mWFfM6FwAAAOk+duOTuvv5fc52nFQpgDYwblBqWdYV4xxfkLH9RUlfLG9YAAAA7ccdkEpM3wXQHkqqvgsAAIDqY/ougHZAUAoAANCgyJQCaAcEpQAAAA2KNaUA2gFBKQAAQINi+i6AdkBQCgAA0KAISgG0A4JSAACABkVMCqAdEJQCAAA0qBP6euo9BACoOoJSAACABjW1J1TvIQBA1RGUAgAANCiK7wJoBwSlAAAADYo1pQDaAUEpAABAg7KISgG0AYJSAACABkVICqAdEJQCAAA0oOm9HfQpBdAWCEoBAAAaUG+HnzWlANoCQSkAAEADuPa29WnbPmOYvgugLRCUAgAANIAf//6V9B1GTN8F0BYISgEAABqQzxgqHQFoCwSlAAAADciITCmA9hCo9wAAAACQbulxE7R3YFSDo9F6DwUAqo5MKQAAQIP5p8tP1tHhiPYOjNZ7KABQdQSlAAAADWbFgqn1HgIA1AxBKQAAAACgbghKAQAAAAB1Q1AKAAAAAKgbglIAAAAAQN0QlAIAAAAA6oagFAAAAABQNwSlAAAADWDe1K607UtOnVWnkQBAbRGUAgAANICuoD9te96UboX8fFQD0Pp4pwMAAGgAlpW+7TOSJcv7ZABoIQSlAAAADcgYKU5MCqANEJQCAAA0gHAsnrbtM0ZWZvoUAFoQQSkAAEAD2NM/mrZtlMiUrt/VX58BAUCNEJQCAAA0gKDPpG0bk9h+2zdX12M4AFAzBKUAAAB1NhqJaSgcS9tnTI6TAaDFEJQCAADU2VU3PpW1z0dUCqBNEJQCAADU2f0b99d7CABQNwSlAAAAAIC6ISgFAABoQEeHI/UeAgDUBEEpAABAA/rBI1vrPQQAqAmCUgAAgDrrDPKRDED74h0QAACgzt7/muPrPQQAqBuCUgAAgDoLBXwK+tNbwCw9bkKdRgMAtUVQCgAAUGexuCW/Lz0o7ZvQUafRAEBtEZQCAADUWTRuKehL/1gWi1tp2/sGRnXD77bIstL3A0CzIygFAACos2gsLn/G9N0rz1mQtv0X//Ok/un2F7T14FANRwYA1Reo9wAAAADaXTRuKeAz+sofnKqDx8YkSW84qS/tnP6RRN/SOJlSAC2GoBQAAKDO7DWl71kxz9nXGfTrklNn6YU9AxlnGwFAK2H6LgAAQJ0lMqXZH8tCfp+2HBjSbc/skkiQAmhR4walxpgfGGP2G2PWu/Z91Riz0RjznDHmVmPMZNexa4wxm40xm4wxF1Vr4AAAAK0iGosr4M/OgAaSFXn/6n+fcfYZEqUAWkwhmdIfSbo4Y9+9kk62LOtUSS9KukaSjDHLJL1X0vLkNd8xxvgrNloAAIAWFPVoCSNJAT+T2gC0vnHf6SzLeljS4Yx991iWFU1uPiZpbvLxZZL+17KsMcuytkraLGllBccLAADQcmLJQkeZQh7ZUwBoNZX4+u3PJN2ZfDxH0g7XsZ3JfQAAAMghkSnN/ljmzpSypBRAqyorKDXGfFZSVNKN9i6P0zzfQ40xHzbGrDXGrD1w4EA5wwAAAGha63f1694N+9Q/HM465rXOlNwpgFZTclBqjLlS0tsk/ZFlOQ2zdkqa5zptrqTdXtdblvU9y7JWWJa1oq+vz+sUAACAlvcnP3xCkrS7fzTrWIg1pQDaQEnvdMaYiyX9naS3W5Y17Dr0K0nvNcZ0GGMWSlos6fHyhwkAANCaDh4by3nM3SYmlQMAgNYSGO8EY8xNks6TNN0Ys1PStUpU2+2QdK9J1CV/zLKsv7As63ljzM2SNigxrfcqy7Ji1Ro8AABAK3NP37VDUkNPGAAtZtyg1LKsKzx2fz/P+V+U9MVyBgUAAID06btHhhJrTglJAbQaFioAAAA0KHemdGA0mudMAGheBKUAAAANKkChIwBtgHc6AACABhXyagnD/F0ALYagFAAAoEG5q+8CQKvinQ4AAKCOFvX15DwW8MiUAkCrISgFAACoo0gsd//REGtKAbQB3ukAAADqKG7lDkrDsXgNRwIA9UFQCgAAUEd5YlKNhGNFnQ8AzYigFAAAoI7yZUq7Qv6sfcSkAFoNQSkAAEAd5QtKWVMKoB3wTgcAAFBH+ZaN+nzZ1Xct5u8CaDEEpQAAAHWUL1Ma8ApKqzkYAKgDglIAAIA6iuRJlXpnSqs5GgCoPYJSAACAOorm6VPqM9lBaTt6ZPNB3bV+T72HAaBKAvUeAAAAQDuLxnNnSoMemdJ2nMD7RzeskSRt+/IldR4JgGogUwoAAFBH0XjuIHPlwqlZ+9p5+i5FnoDWRFAKAABQJ7G45QSZbzipL+t4wO/Ts9e+WVO6gzUeWeNYt7PfeUxMCrQmglIAAIA6sYscXXX+It1w5QrPcyZ1BTWtt8PZbqe4bPfREV36rdXOdoyoFGhJBKUAAAB1EktO3Z3UFVTQn/tjmbs1TDvFZcfGomnbsTxTnQE0L4JSAACAOhmLJjKlHQF/3vMC/lRQevu69qlCm1l9uJ0CcqCdEJQCAADUyeBoRJLUFcoflPp9qY9sa7cdruqYGkkgo/ow03eB1kRQCgAAUCfX/up5SdKBwbG853m3hml9mZnSVp6++/jWw3p+d//4JwItiD6lAAAAdbB+V79+++IBSdJQxtrJTP42XVMaz/hhW7UlTDxu6T3/+XtJ9GJFeyIoBQAAqLF1O/vTqspO7QnlPd9dBCkzUGtlmT9rq2ZKv/HAS/UeAlBXTN8FAACosX0Do2nb714xL+/5aZnSqoyoMWUFpS0akN/3wr56DwGoK4JSAACAGssMrnrGKXSUVvCnNeMyT5FY5vTdytz34zc9ret+s6EyN6sAv2nPNcOAjaAUAACgxuKuaajGpGdCvbhbwlhtFJVGYvG07UpN3/31s7v1/dVbK3KvSjAEpWhzBKUAAAA15s6UBv2+cYOSgKslTIvOYPVUraDUNpBsyVNv7i8lxit6BbQiglIAAIAac8dWIf/4H8fSM6XtYywaH/+kMuw5Ojr+STVwZDjsPH705UN1HAlQHwSlAAAANeaevhv0jz91M70lTPuEpd996OW07Ur/6Bd9/WE9+vLByt60QLG4pfd+7/d613cf1SuHhp39H/rvtWXf+6bHt+vBTfvLvg9QK7SEAQAAqLFY2prS8YPSoHv6blVG1ByqsZ52zZbDOmfR9IrfdzxHh8N6bMvhit83EovrmlvWSaLnKZoHmVIAAIAac6+VPDwUznNmgt89fbeNotILls5I267Gz/7tBzdX/qYFqFZ7m7uf31uV+wLVRFAKAABQY5kFfMYTbNM+pZlrSqvxs0crXDypUOEqrZcdi6Tu205TvdHcCEoBAABqrNgCPn7X9N2zT5hW6eE0rNFILG27lYKsY1WqsusuilWt5wAqjaAUAACgxsJFZkondQWdxyfO6K30cBrWaKT6mVIpEbz9+tndVbq7t4u//ruq3Ne9Xnl3g1QXBsZDUAoAAFBjxU7dPKGvx3ncStnC8WRnSqvzPNfcsk4fv+lpbdg9UJ0nqCH3dOTnd/fXcSRA4QhKAQAAauzZHUeLOj/ga79CR0NjUY1FYxl7q/PD7z46knjOcPNPd43GUq/RrU/v0oKrb9eu5M8HNCqCUgAAgBp7cNOBos73pRU6av2o9NkdR7X82rt10+M7JEn//K5TJFUmII96TJ0eDieC33idih4VazQS04Krb9d//jazj6ul/YOpKbu/eynRg/WVg0M1HR9QLIJSAACABud39TJtkripLGtfOZK23dMRkFSZPOmox9TpF/Ykpu0W0jO22i4/ffa45xwdjkiSvr96a9r+q376lL5+30tZ54cCfORHY+M3FAAAoMYuPW38wMPN32bTdzOzmUaJn78SP3vmOlW37pC//Ccow5zJXQr4x/94bvc4tad19w9HFI7Gdce6VI/SzmDqPsVWewZqjaAUAACgxuJFRlfuoLTYa5tRZu9QO4FZianLI8mpuledv6jse5XjTctmZu3rDvn18yd36uCxsbzXxpLrRv1+o5FwTKd94R596uZn0s5ZOD1VpTlfIA40AoJSAACAGiu2gm5aprTSg2lAma+PcfaXf2+7eNKS4yZmHatlwO/1VHYweuG//DbvtZF4IvMZ8Pn0/dVbJEm3P7cn7ZyF07udx7F2mPONpkZQCgAAUGPxImdT+ox7+m7rBxiZazvtzcHR8qvjjoQTL35XMHuqbi1f2sx/xzs+sUrdocTa2f6RSN5r79uwT5LkM9I3H9jseU7EVYWXmBSNjqAUAACgxsqZvtsGMWmaP3vdQtm50vffsKakexwdDuvirz+szfuPaTSZKXWvubTVMlN6/8b9advLZk8seO3n9XdulJTIlNrXnH3CtLRz/GmBfZv90qDpEJQCAADUWLGZq3ZbU+r25uWptZfhWFxbS2hvct8L+7Vx76C+89BmZ02pV6a0HhnF46d16+nPvUmSxl1LmmnTvkFdlqzWG3P9XiyfPTHjd6YCAwWqiKAUAACghvqHI7rvhX1FXdPOmdJEhdnUD33+1x4q6Lr/W7tDC66+XUNjrim/VqroT2fQr3NPnJ52Ta2nRn/w3IX67d+eryk9oaxj4QKzpvbU7se3Hnb23f6JVWm9bdvtiww0H4JSAACAGtp1dKToa9L7lLZ+gOEODgN+X1amL15A6u87D70sSdo7MOoUSrrl6V0acQWl7105L/15Sx9yUezCQxM7g2n7r7tsufP4z370REH3yqxUfNrcSZIkv2v2LplSNDqCUgAAgBoK+M34J2XwtdknNneb0p6QP6t67OduW1/yvcciiZt3Bn1ZWedCgt1KiNrVczN+Fy46+Tjn8eb9xzyvzczm/vrZ3WnbvZ2JYknus9qhOBaaW5u9xQEAANSX11rG8QRcUWk7ZErd6yN7OwNZP/ONa7YXdT+7uJEkJ1PaFfSrM+PfolYZxWiyMm7Alx6UzpjQ6TyO5fh3Hq8Y0vtWHi9Juu2ZVLDaBr8yaHIEpQAAAA0u4kodtkOAEXP1zOnpyA5KTRHJ5gv/5bf67K2pzKp7TemFS2eknVurjKI95Tbgz/4o/s4z5khK9WbNZGd6czlj/mRJ0p+cs8DZ1w5fZKC5jRuUGmN+YIzZb4xZ79o31RhzrzHmpeR/p7iOXWOM2WyM2WSMuahaAwcAAGhGn/rZM2nb03s7xr3GDqQkaduh4YqPqdGkT98NpG1L6X1bizUcTgWlPp/R5962TJcnK9jWKnSLJn+gzEypJH3kDYskSX0TvH8v7Eyvx6WSpFAg8fH+VbMmOPtYU4pGV0im9EeSLs7Yd7Wk+y3LWizp/uS2jDHLJL1X0vLkNd8xxhQ/RwUAAKBFrX3liPP47k++Xnd/ctW41yyekQowbnq8uKmrzcid2fP7THamtIB75Mp63vr0LoX8Pqei8QfPXaj3v/b4rOetppiTKc3+SeZO6UqOxfvaoyNhSVJvR8Dz+JTuRCVf91Rg1pSi0Y0blFqW9bCkwxm7L5P04+TjH0u63LX/fy3LGrMsa6ukzZJWVmisAAAALWXJcRM0rYBMafsVOkoPojILEGVWnC3GjiPD6gymv6B24rVWGcVI3HtNqZSYrtwZ9GnV4lS7mljc0o7DiQz5v937oiRpYDSada2Uah903pI+p5ovMSkaXalvcTMty9ojScn/2hPy50ja4TpvZ3JfFmPMh40xa40xaw8cOFDiMAAAABrf5v2D2j8wWvL15UxXbUZZQWmOoGr30RENjEaKurdlZQd0Jvn61iqjGHMKHXl/FO8OBTQcTo3xK3dv1KqvPKh9A6O6+/lEj9tLTp2V9zmMMXrjspmSWFOKxlfp79283jE9/wosy/qeZVkrLMta0dfXV+FhAAAANI43/uvDWvml+0u+vt2DUq9KtHv7R3XOlx/Qqf94j+c9ignDfE5QWsRFZYjkaAlj6wr6nbWvkvTAC/slSUeGw7rklFma2hPS7Emp6bmXnpZYE2sXSbKZ5Edz1pSi0XlPRh/fPmPMLMuy9hhjZknan9y/U5K7C/FcSbuzrgYAAGhxn/zfp7Wnf1Q/+8jZnscf+pvzCr5X203fzYgOvTKYWw569/G0vVJEQSifM303d/QWjsb10v5BLZ89qeD75hIdJ1PaFfKnFbd6KdmzNBK1FI7FNWNCh9Py5ewTpunf3nOavnnFGVn3sX8uq2YlnIDSlPoW9ytJVyYfXynpNtf+9xpjOowxCyUtlvR4eUMEAABoPr98ZrfWbM0sy5Eyf2p3wfdqt0xp5hrSzMypJP3Xw1sq9nyFZBSv+80GXfKN1c7aznJEk5lSf44SugcGx3THur1prYAk6dhYVJFYXB0Bn/YPjkmS9g6MeraWkVLTksmUotEV0hLmJkm/l7TEGLPTGPNBSV+W9CZjzEuS3pTclmVZz0u6WdIGSXdJusqyrJj3nQEAANqXL1dPDw/+NgtKs6bvekRVD25K1CSZ2hMq+/nsl9freWxPbEt8wVDsGtZMI+GYLvnGaklSMMf03f6RxHM8sfWwdh0dcfYPjUX13M7+tGA2V2ArpTKlB5MBLNCoCqm+e4VlWbMsywpaljXXsqzvW5Z1yLKsCy3LWpz872HX+V+0LGuRZVlLLMu6s7rDBwAAaGyZgc7iGb1aPntiUfdos5jUY/pu7nMLje2vWDnfeXzh0hlpx+zA9vBQOOf19tTefEFgOBrXTY9vz8r0uu08ksq05ruXlFgXe/m3H3G2H992WIeHwnpq+1H9/SWvkiT91wdW5LzezpT++/0v6Uienw2otzZboQAAAFBb+zKq7vp9RrMndxV1D9NmUWkhhY5SCnttTp4zUUtmJvq9Prn9SNqxGRMSbXky/628xpRvKD94ZKuuuWWdbl67I+c5G/YMOI9DOabd2oJ+nw64spzugPbPV52gbV++RAun9+S5PvXanHHdvbp3w768zwfUC0EpAABAFWVm36JxK+e0TSRkt4TJHQkWGq/7jdGsyYmKtUeH06fgBvw+dYf8Ghrz7v3pHpNdpMiLff3Vt6zT4GhE/SORrMzpWCS1TrS3M3/N0cyfzZ7WW6gJncG07TVbDhV1PVArBKUAAABVlBlQRWJxBcfJkLW7rKA0z3TYQqfv+nwm7/rT4XBMN6ze6nls68EhHUkGsuFY7nIpXSG/83j/4Jj+/MdP6Jpb1unGNa84+91Z396O/EFpZqGjRzaXF1TmKogE1Bu/mQAAABX2lGt66GgkPbCIxqycrUCQcOhYWGfMn6wXvnCxJGn5nFQblpULpqadawqcvus3RpO7SiuKdP7XHnKylOFo7gC5O5gKSp/YetgZW3coFXxGY4VnSvMVXipFoIjiWkAt8Y4IAABQYe/8zqPO4x+4sm+RWFzhWFyhAMFBPvsGRzVvSreTeTx/Saow0dJZE0q6Z8Bv1BFMfPT9u4uXZh3vDvk1d0r2Wt/MHqmZ2Uu3TldQevUt65znC/iNNu4d0Nfu3qSXDww550zoCGbdQ5IuO322pPxThQt19VtSP2uAaeNoUPm/ngEAAEBZ3C1Ejo1GFY3FyZSOIxJN9OJ0e+DTb9AT2w7rhT2DaftHo4V1H/QZo4+et0hHhyO68pzjs46fMX9y2npP2/6MdirhaO6g9Opb1qVt2/FsLG7p4q//Luv8zqD378GHVp2g257ZnTMA/sq7Ts05hkzu7CjTxtGo+M0EAACoovlTu53HMctSJGaVFRyct6SvEsNqaNG4ldUu5YS+Xv3hq+fLl1H9Z5rHOlF3dnPF8VMkJaoeT+wM6vp3npI2ndbWEfBrzCPg3Lz/WNp2rkAxc6rt606c5qwnjuaYhpurqrKd0YzFLU3oDOi9r56Xdvz0+ZM9r/Pifh0HR6N6ZPPBgq8FaoWgFAAAoIrcgUc0ZiULHZU+jTJvd5QWEbeyg1JbZjwf8ZjiGk4GjpefPtuZApwZzGbqCvk1FM6uvjuWkYkN5whKjw6nV1k+YXqvE6gOFFk1185ufv7XGzQ4Gk2bFhwK+HTSzMKnML/atQb3P377sv7ohjV6ZsfRosYDVBtBKQAAQBWNRlJBTcyyFI1bZa3ty9cepVV4ZUpz2T84mrXuc2gs8ZqfPm+yE8SPd7/uoF8j4eypwJnTdb2CYElpWdbpvSHFLMt57nz9T73Y07v3Jq9zT2VePntiUfc62VUkyrb76EhR9wCqjaAUAACgitxBaTQWVyxO9d3xxPIEpb939dr86HmLNBqJZ02dPTaayHj2dgadIH68GdPdIb+GPYLSzCm9mQGwzQ5eL1o+Uz5jFI9bTvuX25/bk//JM2R+aRFyBaXBCvzuHMvTjxWoB94RAQAAqujO9Xudx7uSGSqm7+YXi1vy55hu665Ia/f5jGW8KHbQ1dvhd16v8afvBjTsMX03M1Oa6/W3s5qXnDpbfp9RLG4560939xeXKZ3W05G2HfD5nCm9HTmKIxXD/UUJ0AgISgEAAGrkw//9pCQpUEaho9VtUKgmFrfkzxG4u7OidqCWmSm9cc0rkhKBpiUreW7+17wn5FckZmUVMrLXkP7yqtdJknO/TB/+77WSpCe3HZbPGMUsyzPzWgh7Hawt4DfyJX/WE6b3lHRPNxrDoNEQlAIAANSIncHLDKKQLl+m1L2m1p7im1nd9sY12yVJIb/Pec1ztV+x2cWEMrOIn711fdr1uf7pBpJThqf0hOT3Gb24bzCrcq/bGUVU0A36jaLJ4Pi4Sdm9VIs2TtYYqDWCUgAAgApav6t/3HO+v3pryfc/cUZvydc2i5hlpfXXdHMHhXZQGs8RKYYCPn36TUs0e1Knls/OLviTea6Uuw9pZ8CffP78Xyh0Bv06MhTW+l0DWcfsH+kjrz9BP/vw2Xnvk36dcX7u8YJrL9+84gzPcQCNgqAUAACggvJlx2y5el0iEWBalpzpqpn6elPrLQM5MqVu5y+doUevuTBrSmwmJyh1/du8uG/QeWxnUr2eak9/qprtFa+er8EchYTsayd0BtKKF3l5+2mzncfuvqru9jCFutR1L2n89bVArRGUAgAAVNBIAUVk/vi1x5d8/1xZwVZhFy3KlSn9h0uXOY/9yXWiuaZDFxN7BZPrfF/al/hS4bmdR/Xmf3vYOW5XxP3cL9erP6Pv6JGhxPbVb1mqSd3BcZ/LX0AF3X9/7+nO497OgE7oS6wl7SohKAUaHUEpAABABQ1kBCxe3nXW3JLvn1lpttXYAWbOTOmEVKbUrheVK1Oaa12qF7t3558nCxbtOpLKfl562uy07OJpn78n7dqRSCIzumxWoofolWfn/9KhkDpXxhi9/qQ+SYnM6pYDQ5Jyvy7FYE0zGg1BKQAAQAVdf+fGcc8pZQqmrVUDikc3H9QT2w47P1+uTKk7zrQzjrmyx4tnFr7+1l5Lak+tdgd/S4+bkHX+netSvUeHxhLZ8e7kFOEVC6Y6xz523qKsa4MFVl+2n3felO6scZZjvHWxQK0Fxj8FAAAAldQxznrCfFpx+u7Na3foMz9/TpL07LVvlpR73aNxNTSx48ZcmVL3Wszx2MGoHTC6s6yTu4Oa2hNKO3/fQKr3qN3f1H4+OzhdOL1HHYHUFxAfPW+RRsIxXbFyfkFj+ps3L9HbT5udVtyq1B6j7z5rrv7vyZ2SWveLDTQvMqUAAAAV9AcFTM0tZlqpzW4h0orTd+2AVJL2J4O9QjKlduBqVeA1efeKeZKky5JFgdzLPid2Zq8TneDaZ/cj7elIBKB2UaWQ3+esRU1cE9A/vn15wZnyUCikJpEAACAASURBVMCnk+ekVw0+c/6Ugq7N9E/vOFmLkutSo7HW+x1CcyMoBQAAqKBCpkaWUv301o+9TlesnK9WK9ybmfl9U7K4kD9XUOp+nNyoRIhlB2yHhsKS0v+Ngv7ssbi/HLhr/V5JqWDULkZkjPe1pXj2H96suz65SstmTyzp+o6AX7f95bmSpC/e8UJFxgRUCtN3AQAAKsieGtkR8Gks1/q/EuMUv6/11gPudU2DdctVoTbgsR6zEi+JSQahD2zcn3z+tPA363x3MH3Phn2SUtN33WtG3Y9LzXJK0qTuYEGVffOpVIAMVBqZUgAAgArq6UgEJqfPm5zznFILqO4+OqrDQ+GKTFdtFLkC91y1gCZ1BfXNK87QPZ96vRNISpZuf26PvvPQ5oqtuXVnSr0S2+5MqT1luzf5b28f8hnjBNEzJnTotSdMq8jYSuVe3wo0EoJSAACACpozuUuSdP07T8l5jilh+q6UyuJtOzRc0vWNaCzqXbgnXy/PS0+brZNmTnDyl5YlXfXTp/SVuzY5waJXxdxipAWlHsfdxYKefOVI2rGFfT0K+X366zedpGDyG4jeTiYoArkQlAIAAFTQL55KVDi1s2aS9K33nZG2XW6ryVaqnuruB+pWWC/PxH/dr4b92lx2+pyyxuWeJj0wmqiue/Vbljr7xiKJDK9lWdp6cCjt2t6OgF784lv0xmUznUxpCyW3gYojKAUAAKigLQcSAYp77eOyWRPT1iiaUheVJrXK9N071+3RB3+81vNYvkypzX4d7eq3Uqo9TK7qvYWwLEsj7nsmq0stcWVfRyIxPbPjqAZGonnvZa/jbJQvEi45ZZZT1AloFASlAAAAVeAOQn3GpAWSJc7edTRIfFO2320+mPNYIW1z7Jf4h49sdfaFk2tUc1XvzedTbzxJUiKrOeLqBxrxeMH3Dozq8m8/onO/8kDeewaSwXU5vWkrye8zLfP7g9bRGH8dAAAALcZd6TQRlCptuxzlBrWNoieUu/BOIUGl/Trc9sxuZ9/QWCJzGSih0qz9lLGMTGnEoxjTT9dslyQNjubPlNrjmNAga0oDPqNovMX6CqHpEZQCAABUgd0eREoET+7kVLlBZYvM3k17jTIVlunMPuff739JkrRh90DR4/ElnzNuWRoOp4JNO4gLFjClOJOd8e3tLK+dS6X4fUaxWIv8AqFlEJQCAABUWTgWTyucU2qm9COvP0FSa2RKf/n0Lm3YkztwLGRNqNfrsGbrIUnSnn7v/qf52P8u0Zil/3x4i7M/kgzizlk0TZ9640nq9sjwvmfFXM97jiarC/d2NEY7loDfOOtugUZBUAoAAFAhuYrZTO/t0CWnzHK2S1nvKEmnJXufxlsgVfrJnz2jezfsc7Yfu+bCtOO+QoJSj312VdxSXmL7mluf3pUW1EaShY58PqO/euNi9U3oyLp25ULvHqT270QhhZtqwWdMwxRdAmyNMbkdAACgBdjBi+26y5br2FhMk7qCuv6dp+j/ntxZ1v3tTF6zLwnMrB685Utv1bFw+trMwjKl2eeMJgsUlZKNtr8syAzapnSH0p/X49pgjjWs0Vj51YArKeAzTi9XoFEQlAIAAFRIOCMo/eOzFziPA36fbv/EuXpq+9GS72/HNc2eKX3etd7TmEQGsjuYPr21kKBy3a7+rH12T1GvgHU89jXuSrn//t7T9bZTZ6edt+3QcNa1wRyNVe0AsNziVpXi9/lYU4qG0xjzCAAAAFpAdJwP+8tnT9Ifv/b4ku9vBzaNMv3ygY37tODq27Wnf6So6w4cG3Me24WAAhlBXSHVc5fMnJDzWCm9XO2g333lZafPKWi6da7quifPniRJuvBVM4oeTzWwphSNiEwpAABAhdjTd+dM7qrK/RtkWaLDbovyB9/9vab3hnTbX55b0HXu4D3X2tFCMotzpuR+naf1hnIey8V+zlIS0Qum9XjuXzZ7ol74wsXqytP+ppb8PpM1zRyotwZ7awMAAGhe9of9v3rj4qo+T6PluXYdHdGzO7On0uYSdQVFYY8eoFJhazD9eQLXj19Q/L+BuyVMsdxTfjM1SkAqSQ+/eEDRuKX9g8VXJwaqhaAUAACgQuwMYK6iN+UysjN5jRaWFqeQ6aOFTJnNlzmeN7W7mCEl7mdP323y1zeft5x8nCTppX3H6jwSIIWgFAAAoELsTGmuojdla4xaOY5SY7doAeWDCwlKAxWez2xP373r+b15z1u5cGrWvmm92W1iGpHduqaF4240IYJSAACACok47T+q+xGrUeKJUscRKaD6a2dw/CmvlY797Wm7j2w+lPe86y472Xl821Wv0w0fWFFy79laSxVzapTfIoCgFAAAoGJSmdJqTd9tLKXGYYVUD57SHSzg+VMDeOeZc0objEskx/rWTEuOS1X9PW3eZL1x2cyyn7tWjNNWqL7jANwISgEAACogHI1rOByTVMXpu0mNMvVyYtf4gaOXsUjMc/9PPrjSeVzIa+jOTnYEyi8mNBIpvCrtbz5+rn77t+eV/Zy1ZkxrrEtGa6ElDAAAQAW89vr7dXgoLKmwHpulME5msEECihKH4Q7+prvWYq5a3Oc8LiQodWdK81W/LVT/SKTgc0+eM6ns56sH5zeoQX6FAIlMKQAAQEXYAalUvUxpo03fHY2mZzwLyb5t3j+oEVem9LLTZ3ueV8gUaHemNOQKSj9+wYnjXuvl6HA4bXv95y8q6T6NzA7kXz5A9V00DoJSAACACmu16btrtx3Wgqtv1xZXILPlwDGt2XI47bzx1inesW6P3vivD+sb97/k7JvQmT5xb/akTknurHBu6dN3fZ6Pi7Fgek/adm9H600qtF/Wf7r9hfoOBHAhKAUAAKiwQJUqsRqncmpt/ejRbZKk7z70srPvgn/5rQ4NpWcW4xnR8mgkpu2Hhp3tdbv6s+6dGcDf9pfn6qYPvbagcbmn74b87qC0tPWlH1p1QknXNRPTcPl2gKAUAACg4kIVWN/opV4BxZzJXZKkNVsP5z0vMyj97kMv6/VffdBZq7lxz0DWNZkJ0b4JHTp70bSCxpVr+m6p1Y+bpa1LOdw9YsMFVhsGqo2gFAAAoMKqlSm11Xr67uxkUHrW8VMkScPhqOd5meO6ee0OSVL/cCIo9Xv0b/UXME03F/e17zgj1RLGV8br/+jVF5R8bTMYdRWZ+vp9L9ZxJEAKQSkAAECFVa3QkT19t8ZR6VAyCLUzkP9+30ue52VmSu3ptXZBJK/iOttc03uL5Y5xJ3Sm2tMUsh41l1bPlrpfmr39o/UbCOBCUAoAAFBh1a6+O1rjaZfDY4mg0g7Ydh4Z8Twvs9DRgcExSdJ3HtwsSdp6cCjrmnKqwLoDSHewVU72tbVDUmnlgqn1HgKQpax3TGPMp4wxzxtj1htjbjLGdBpjphpj7jXGvJT875RKDRYAAKAZVKtPqZ1x/NTPnqnK/SXpyVeO6Nx/fkADo6menXamNBJLRJ0nzZzgHAv6jT79ppMkZWdKw7FE8HzX83slSZecOivr+fYPlJ6tcxc6Sn9c8i3LyrI2A/fU5uOn9eQ5E6idkoNSY8wcSZ+QtMKyrJMl+SW9V9LVku63LGuxpPuT2wAAAG0j6LF2shLGkusBD2dUva2kbz7wknYeGdETrqJGdqY0mgwy3f1Jp/d2qCfZOsXKkcCd2h2SJPWEsqvinnV86Zk7d6bUHYj6yggsW3z2bpqJXa3X8gbNqdx3zICkLmNMQFK3pN2SLpP04+TxH0u6vMznAAAAaCrBQHUim1gN1pJ2BROB49/9Yp2zdtXJlCbn5465iuWEAj4nkMvMlK5aPF2SdMXK+ZKkmEfQ+vnLlpc8Vn+uTGkZkWU5AW2zOGP+ZEnSGNV30SBKDkoty9ol6WuStkvaI6nfsqx7JM20LGtP8pw9kmZUYqAAAACNatPewbTtQJUypbHMRZtVYLdWOXhsTAeOJdaEDodTmdLN+4/pB49sTbvGDgIzg1K7X6g/OZ055mpHctHymbrjE6vU21F6ts6XY01pOdnOdghKf/4X50hK9JEFGkE503enKJEVXShptqQeY8z7i7j+w8aYtcaYtQcOHCh1GAAAAHX3wMb9adul9skcTy2K7kZjqSe59aldkqShsahzLDMgPev4Kc46zMyY2c60xpMHXLfWh1adoGWzJ1Zs3CZH1rTo+7RBGVC/zygU8KW1hwHqqZw/uzdK2mpZ1gHLsiKSbpF0jqR9xphZkpT8736viy3L+p5lWSssy1rR19dXxjAAAADqKzPjVK1iOdEaZEpvX7fHeXz9nRslpTKlkbiVVpjo//7ibH3pHac4mcnMVjX2dGN73HZwumTmBC2fPaliY87sC8v03fF1Bf1kStEwyglKt0t6rTGm2yTeeS+U9IKkX0m6MnnOlZJuK2+IAAAAjW3O5K6aPE+8BkGpF3tN6cMvHtCIK5B59YKp6gz6nUDOPbz1u/r10KbEbDgnUxq3dNLMXt39qdery6PoUSm+9u7TdNcnXy9JunBpYtVYedN3KzGqxtcR8GlwNFrvYQCSEoWKSmJZ1hpjzM8lPSUpKulpSd+T1CvpZmPMB5UIXN9diYECAAA0qr4JHZKkr7zrVFmqXuBYi0ypMenThIfGok71XUk6PBTJuiaz0NEzO47q50/ucI7bGdOYZclf4fW2f3DWXOexXY3XlNFttF0ypfsHx/SLp3bqX95zWr2HApQelEqSZVnXSro2Y/eYEllTAACAlmVZlv7z4S1615lznX6cJ8+ZVNF1kpnchYKq5YqV8/XTNdud7eXX3q0JnamPjKfPm6QX9gykXWNPV47FLQ2Ho7r824+kHY+6MqX+Kq7Z7ExWDo54lfkF0LDaYCk3AABA5T2/e0BfvnOjPvWzZ5y1eR3B6n60qkWmNOLRJsReUyql2ojMnZKasmxnFy1LOjKcnUmNxVxBaRUzkR3JysGslSxcmLYwaAAEpQAAACWwp6r2j0Q0kFybN7EzWNXnrEVLmLBHltH9vCPJAPUXHz3H2WdnP+OWpdd9+YHs6+0qvJZVVhGi8diZ0nL6b9qBbbv41bO76z2ElrSnf0RHhsL1HkbTKGv6LgAAQCM7OhyWkdGk7uoFi3HL0sBIIjvonuZaDSsXTpUknTKncpVrM+Wa+nrq3El6bme/7ly/V1O6g5o5sdM5lip05B00R2OWHtq0X6s3H9RZ86dUftBJdkA5Fi09U2qM0YYvXOSZ8W1F5fSJRW5nX/+AfEbacv0l9R5KU2ivr4IAAEBbGA5HdfMTO3T6F+7VaV+4pyrPcevTiR6ez+8e0MBoRB0Bn5Opq5ZT505WR8CnsxdNq9pz5JrO6c4CZyZs7WD0xX3HPK+Nxi39yQ+fkGWlihFVg52FLXdJaXcoULOKyvXSmZxqPrmKX9i0uzoVy25KBKUAAKDlfOmOF/SZXzxX1eewM0xnzp+sgZGoJlR56q7N7zNZ/UArKRyzNH9qd9b+blcLl8xloRv3DEqS/up/n/a8p7tA05qthyswSm/tUTe3Mn74JyslSTc9vn2cM1FJ0VhcT28/Uu9hNByCUgAA0HJqMfVyRnL66qK+Xg2Ho+rpqG6W1GZU3QxMOBrTca6pubaQa61lILOtSzIazBUr16JAk1s12/K0CjthfdszrCmtpa/f95Le8Z1HtX5Xf72H0lAISgEAQMvpqvI0Wkn63C/XS5IsSUNjMXWHarM2z2dMzuCvEiIxS6GAT9ddtjxtvzsoDfrTc5L51pSGAj49/OLBKozUwzjBMVLcxaCqmXlHug3Jdkr7B0frPJLGQlAKAABaTi0rqMYtS0NjUfWEapMplcldUKgSwtG4gn6jeRlTeN2vaSAjKI0mF3F6jSscjevgsTFn+49fe3wlh5vGMIG3YHv7U0HRaIS2MLUyNJao1M33AOkISgEAQMu5cU0N18lZ0sBoRJO6arOm1FfFPp9SovpuKOCTyXiejkAq6N5xeCTtmJ11K2SWbqiKXxi88VUzJEmrFk+v2nO0CnfBKa82QKgOe001XwSkIygFAABNbzSSagESrcMH7P6RiCbWKCg1NcmU+tSZETzmi4Xfs2KeJOnkOROzjp0xf3LadhWL72rFgqna9uVLdOrcyeOf3Obcv0P1+JtpV3bV43+5Z1OdR9JYCEoBAEBTe3HfoJZ+7i79+tlEwZbDw7VtWG9JGhipbaa0mlP/wslM6cqFU7VgWmoKb3+e4lEnz5mk6b0dWr9rIG3/F99xsi45ZVbavgc27q/sgFES969QJMZc0lpZPGOCJGnLwaE6j6SxEJQCAICm9tzORBXLBzclgp2Dg9lBaaTCmaCxaCozG47GNTgW1cTO2hQ6OjwU1p3r91Tt/uFoXCF/YvruX16w2NnfPU514ZA/OwU6Z3JX1nTd3hq1zkF+7uJGlf77QEpmld1qzhRoZgSlAACgqd23YZ8kKZhsU3J4KDsoHRyNVvQ5n3wl1WewfyQiy1LNpu9K0sFjlcsGW5al1S8dVDy5IHQsGncCSXdxo0tPnZ33PgF/9sfKgM+nYMb+Kd0EpY3gEte/J0Fp9bztm6vTtpcel5ji/o4z5tRjOA2LoBQAADS1u57fKylVEda9vtQ2MFLZvqXuIHf15kS7k1q1hKm0Xz+3R+///hr9bO0OSYnXz26pk6vi7rvOnJt1n8w2MfY1mUHpRDKlDaG3I6Bvv+9MSUzfrabejvT3hTuSsxxq1de4WRCUAgCAlvDivkFJ3lmfgdHKBqUxjzKz3bVqCVNhz+04KimR8Y3HLY1F4+q0g1JXv1d3cOmRFM0KPiUp4DPqzfjwXc3quyiOXbzKEkFptWT2gLW/0CI5nY53BQAA0BKe2HZE335ws3b3Zzelr/T03Y/d+FTWvq4mCUp/89xuDbqC9OFkZnnfwKhGk2tl7Z/FXYF3OJzKQPt92R8hvYJSv8+ob0JH2r7XnTitjNGjkuz1jfTMrJ7XnejdniheSP+kNkJQCgAAmlZmFuKrd2/Sdb/ZICmxZutL7zhFkvRHN6yp+liaIVP68oFj+sufPq1P3/yss8+edPvDR7bpz3+8VpKc6bvurObS4ybo7acl1iF6ZUoDXtN3fb606bqr/+58XX46a+kaR+LfrJothtrNrqOJHr5Te0KSpJkTOz3Pi/GapyEoBQAATSuaJ9tw9VuWevbNrJZmCErt7MzLB45JkgZHI3p862Hn+KMvH5KU6qXoDkond4d0ZrLnqN+jaWnAo6xoZqA6d0q3TL6Gp6gpMqWVdffze/W6Lz+gh1884Hxh5g4+3V+ibT88XPPxNTKCUgAA0LTyVQ3t6QjIX8P+C13Bxi90ZE+xHY0kXrd/u/clvbT/WNZ5nR6FjqTUlwBe03cDrn3zpyb6m07tCbFasYHZXxAQlFaG3f7lqe1HZH9ftufoiIbGEssHxqKp96vHtx7OmunRzghKAQBA0wpHvYPSgM+oJ+RPC5Qq+QFwxoQO/eGKeXrbqbOcfc1QTdPO2uw6OqLvPfyyfv3cbs/z7GA0c53o0eHEWtTJHm1d3FnRf3z7Mq3//EWaObHTKQo1c2JH1jWoLx+Fjirqmw9slpSYYWC/3zy46YDe/R+/l5RdGdyrYFq7IigFAABNK1dQOqkrKGNM2trHSrW9iMUtHTw2ppkTO/Sx80509te60FEpQbb7Q/CX7tiY80OxHYxmBqX29vHTurOucWelVy3uc1ph2GvrWEvaeOyZ1MRG5XMXLuoM+NOyzxv2DEiSdh9NL8KWb/lBu2n8eSYAAAA5hHNM3z0yHJaUHoiORGIVaUdy6NiY4pY0Y2KnFs3ocfbXuk9p3JI8agvllRmE+nKs7/QlA8zM1+sjbzhBx03q0KWnzs66xl5T+rcXLUkLZmdO7NQjV1+g43IUfEH9pKbvEhyV66irF7Ix8sw9/+SxbWnbFJhKIVMKAACaVq5MqR17uYOwnUcqU1hk/+CYpMQU3o5AKjvaFaxtpvTvfvFc0ddkBqXHTfKeUrtuZ2JtXGamtDPo1x++er4TtLoFPNaZ2uZM7qrp+l4Uxv4XIWFXvpvX7nAeh6PxrIDz8FBYT29P9AR+w0l9ksiUuhGUAgCAuorE4npmx9GSrj025t1/dOH0RAbTnen72t2bSnqOTHbREnt6qq3WQdfPn9xZ9DWZH4LjOepE9SR/tsxCR/nYPU4prts8UplygqNyDbvei8LReFbxqDOvu1cb9w5KkmZPTswaoFdpCkEpAACoq6/ds0mXf/sRbdg9UPS1mWu0bNFktHXSzAnOvkp9/LOnBAcrMBW41mIZUehw2Duo70muj83MlObz0KYDkiQjotJmwZrSyhkKp4oYhWNx5z3Iiz3DgkxpSvO9mwIAgJby34++Ikn6xVPFZ/729o947t9xOLX/9k+cK0l6y8nHlTC6bMfGEmvHignYKuntp2Wv5yxU5hLcXMWfXr1wqqTSsr+ZgS8al/0FAksby+f+gufIcDhvYbWOZB/gAdc61HZHUAoAAOpqJNkm4bEth4q+dnDUO9Nn98mUEoV2JGk4HPM8t1h/8T9PSZJCyaD0Y+ct0qrF0yty70KcMmdSyddGM6JSr9YukrSor9d5fM+nXq/ffeb8ce997omJ16BSVY5RffZ3Di8fyO5Vi+K431/29o/lPfeRzQclST957JWqjqmZUH0XAAA0hFyVYPPJtab0a+8+zXk8rSekiZ0Bbd5f2Q/edl/Oz1y8tKL3HY9XkaFCve+GNc5jv88UNG3TPQU6n9csnKrVmw/mnbaIxmL3rb3mlnW6YuX8Oo+mubm/IBsvA2qf+8qhyhRfawVkSgEAQF2ddfwUSdKbl83Me976Xf36oxse05GhsLNvMEdQ6m4/YozRzImdTpuYcjRCs/tK1lN6YU/x63hzCSQzx0NjlclIo/pY01gZ//PYK3pg435n+/Fth/Oef+2lyySl3vtAUAoAAOpsUV+iUm7fBO/2JLa3fXO1Htl8SLc8vcvZd8yVnfjxn63U2r9/o27/xLmaP6077dpQwJezfUwxRiP1D7hKySh7yRVgl3r7x7cmpl//n6s1BhpbjKnWFXFzkb/zZ8xLBKN2QTEQlAIAgCYzvTfkPHZP333DSX2a3tuh5bOz11yGAj6NVTgozWwJUyvV7jxT6u3/4Kx5kqQTZ/SOcyYaBZnSypiR4wu1D5670HO//cXS+hIqjrcqglIAAFBXDyZbifxu80H1D49fjdK4UnnuKXP5hPyVCUrtokx/uGKeZk/uKvt+pTBVagT6y6teV9b1MyYmPpgH6lSVGMVrhOnoreC+F7zfh16zcKo2Xndx2r6PvP4EmeSfSCm9hlsV7xoAAKCuDgwmKlXe/twenfaFe8Y9P7OCrCTd9clVea+p1PTdh19MVM0858RpZd+rVJWavpvJn7xvqUGvfZVFf5GmQVGqynrnmXPStn3GpLWO+sGfrNA1b31V1f6GmxlBKQAAqJv//O3LWfvGq5JrTzn8/cuJNYyzJ3Vq6XET817TUUZQGo9b2jcwKkn6f7eukyR1Buu3FqwS03eD/uyb2J+TS82e8Tm7+ZAprax/fc/pads+X3qv345A4n3Dzx9LFoJSAABQN9ffuTFrn505dTvqqpxrZ+Ju+N0WSYUFiIk1pcUXKTp4bEzf+90WveZL9+t7D6cC6K56BqUViEpfv7gvbfuWj51TsaCSMKd5sKa0eD/5/TatfulgzuNvP22289hkrNC236u8vhQaz/889oq2HRwq+rpmQVAKAADqZnpvdoEQrymFz7sKgtifo4+blGj7suPI+L3+ejsCOXua5nL7c3u04p/u05eTgfOX7kgF0B2B+n2EqsTUv45g+vjPnD/FKeJ0+rzJJd41MS5m7zYPMqXF+9xtz+v930/1+123sz/teI+rAFo4Y6nB7MmJ96yA36d3njlHk7uDBT3nWDSmv//ler3ru4+WOuyGR1AKAADqxqsmjlf2xr0uy56GO39qou3LGfPH7/U3pSekI8ORotY75utruvPISMH3qbRyEqUhv0/ve818hTxeeHtq4TmL6rdeFrV1+emJNZDHZ7RQwvgGRiNav6tfl35rtSTpA2cfL0l6dsdR55z/eeyVtGum9aS+hJvSHVK0wJY8du/fQ0Pl91puVASlAACgbg4ey/6Q5dU70T3d7dpfPS8ptVbrW1ecMe7zTOkOKRyNO9VzC5EvizErmfGoh3ITpRM7g04A6nbynEn6xUfP1qffvKS8J0DT6Ar5dd6SPk3qKixjh5TRcCytRVTAlwirNuxJzeq48uwFadeEXDMsgn5fViY1l6HkLI96ztCottb9yQAAQMPzmj4Y8figNhrJ3hdJBq8TC/hAba8B9bpPMWOznbNoesH3aRSWZSkciyvkN2kfjt3OOn5qWmGWYtjBMhNCm0vQ73P+lpCfe6bFcDiWVqk64LFOdMH0RAb6hL6erGMhv1EkFi9o9saxNghK69P1GQAAIAevbOaoR5EiO3gNFtAX0/4wV0yxo0Zdb+deU2pZVsEtXOyfJ+j3VeXD7dxk39a3nnxcxe+N6gn5fZ5fBCGbu9fxcDiWVtE7kPwy58KlM3T/xv2aO6VLJ86YIEm69WOv08BIeg/moN8ny0r8XXoFtG5/+sMnJNW36ne1EZQCAICG8tc3P6uVC6dq7pTUOrefrtnuPLYL8URicfmMCsrs2YV9xorIlOaqTPoPb1tW8D2qzbIKn85rZ8MCfl/OTGk5Zkzs1POfv0jdodb94NyKgsmMHcZ3o+t9aDgcTXvdAskvx264coWefOWIViyY6hyb1BXMmiIdTP4NRmKWPGbTOyzL0t5kS6pWDkpbNwcMAACa1rce2Jy2fe+GfZKkaT0hp0jPTx57RYUmM+01lJv2DRY8hlyZ0nq3GHRnRovJ5YadzLLxXFNaCT0dgYIzt2gMQb9PkRJ7+Lab636zwXn82xcPaP3uVOXdYPLLMWNMWkCaiz3DFbub4QAAIABJREFUY7x1pcPh1OyO7YfHrzTerAhKAQBAw8kV2Cyc3qPDw2Gded29Ojoc8TzHix1gfuQnT6btf3zrYa3b2e+5ritXprTQipm1UEw14Wjyw28oUJ1MKZpTMOBTuIF+p5vFNx/YrK/ctcnZLvYVDCWn7I6Xpc5sZRVv0GUF5eIdCQAANBx3sOV+PKEzoM37j+lwka0RAh5TfHceGdZ7/vP3uvRbq/WlO17IOh7L8WExXudGnO4fpZjPp870XV/6mtI3LZtZqaGhCbGmtDLWbD1U1Pl2pnS8L9cGR9OD0p+t3dGS/14EpQAAoO5WLU6vZnv7uj3OY/f0tlKnhp63ZIbz+OUDxySlFy35r99tzbomV6b0PSvmlTSGSul0Tb21isjPRNzTd5NrbOdO6dJ/vP+syg4QTYU1pYXr7QjopJm9nsdmT+oq6l5dybXXz7umAHsZDqcHpdfcsk4/fnRbUc/VDAhKAQBA3fSE/PrguQv1Xx9YoY9fcKKz350dGE42jp85sUMPbNxf0vN0uYrvfPH2RFZ0vGm4XmtKJ3YGNKUnVNIYKsVd7KSYpK27WnEw2VMx5PeV3AIGrSFIprRg4Whcy2ZN9DzW01Fc/Vi7YNt4Vb437U2sg1+5MLVOtRVbw7TeTwQAAJrGaDSuzqBPnUG/Pv3mJZ7n2GuqPv3mJZpagYDwrOOnSMrfHiYai3tmSm/68GvLfv5yBV3tIw4MjhV8nV0wpTPod4o1UZMIdp/SYtYnt6N4PNHntztH8GnPPiiUXWxsbJwiU/+cXLf6+NbDzr7eztZroEJQCgAA6mL/4KhicUsDI1HP43bQaAdTPaGA/vDV5U+dndydaM0wmqM9zCuHhnTiZ+/UV+/elHXsVcd5Z0lqaUJnqrXE1+7JHmMudp/ESV1Bp9cpWVKEXK1JkNvPn9wpSdp9dMTz+GsWjl9x183pnezRl9ntgqV9WfviLZjYJigFAAB18aNHtkmSbnp8u+fxz966XmPRmJ7afkSS1N3h17QKZErthve5MqXP7cy9xsvXAEHcstmpwLiriL6Fa19JvI6Tu4NKzt51glO0L7sIWKFTeIfGolqX52+kVf3iqURQurd/NOvY45+9UBcsLa5gmNM7eZxM6bxkv+aPnrfI2VfvYmvVQFAKAABqyrIsvepzd+k7D70sSeqb0OF53u9eOqDP3rpe19yyTlIiUzq91/vcYkRicT24cb/+7ufPZY1Lyv6Q+LcXLdGlp83WiTO8C5zUQ09yjWwx69j+9d4XJSWKtdjBKD1FYVeBLTQoXX7t3br0W6t189od1RxWw3nDkkTG8h8uXZZ1bMaEzqLvZ/dbHi8oHY3G5PcZXXV+as19C8akKmtCsjFmsqQbJJ2sRHueP5O0SdLPJC2QtE3SeyzLOlLWKAEAQMuIW9KIa8parkDz2GhUd7qq8HaH/BUp8LFvYExfumNj1v7Htx7Wa06YplHX2JbNmpj2YbBRDDlTmgvPlJ45f7Ke2n5Usyd36ekdRyVJftITbS+Y/JsKjxOUXvbtR/Rs8vdGkj7z8+f07rPmts0XGxOSXwDZmctyBfw+BXwm79p2KbHMoDPgU6/rCygypdn+XdJdlmUtlXSapBckXS3pfsuyFku6P7kNAAAgSYpmLIj6f299lfP43BNTrWGGwjEn+JISa98qsQby+6vT278smTlBknTtr56XlF75N+BvzA/cf/q6BZKkWZMLb0Nx4oxezZrUKb/PyP6pUo/QrkJ+e/pu/kDHHZDaxgtkW8kvntolKfHlWKV0BHway7G23TYaiTkVt+/8q1WSpFgyKB2NxJzlCM2u5KDUGDNR0uslfV+SLMsKW5Z1VNJlkn6cPO3Hki4vd5AAAKB1uFsgBHxGZy+a5mz/z5+/Rh9atdDzumCV2pdc+KpED9M5yQDvyHDYOdaoay6vPHuBJKmziIqfkZjlTNW87ZndkqR1u9pvbSDS2b8T0RICzNFwawREhXgmGZS7C41J0puXFbeW1K0j6NdonkzpsbGoblyz3Qn+7Vkl9lvo0s/dpVd/8b6Sn7+RlJMpPUHSAUk/NMY8bYy5wRjTI2mmZVl7JCn53xleFxtjPmyMWWuMWXvgwIEyhgEAAJrJbzel/n+/06NQz7Qc03mDfpPVpuWSU2eVPZ4/X3WCJOnEmb365dO79L2HtzjHGrU6rR0sF1OFMxyLO+1kPnD28ZJas98hihMock2p23DEu3J2K1rU16OLls90qhVL0ttPm61/fPvyku85Z3KXth0cznn858l1u/bsDfvtyN2+pz9ZVbvZlfNOFJB0pqTvWpZ1hqQhFTFV17Ks71mWtcKyrBV9fdmljgEAQGv66I1POY+9Mn0XLz/O87qQ35dWFOSuT67St993ZsHP+/kcHx67Q351Bf2yLOnvf7k+7Zi/QTOl9rCKWVsWicadrJidnV61eHq+S9AG7N/xfDHp/sHsirOStPqlg9UYUkM6OhzJ+sLsG1ecodlFTKHPNK03pIHR3EFl5vTo1JdRrCl12ylpp2VZa5LbP1ciSN1njJklScn/7i9viAAAoFWMZvTksxvIuy2Y3uN5bdDvS8sQLC2yZ+j7XjPfc39HwCefSUwrPjaWnvmJNGhDQDsoLeajaSSWCkqDfp9+f80F+sYVZ1R+cGgqdrGrfF9wDOTIxv1tRgXrVrS3f1ThaFxHRyKa2l1+Syo3o/yVdDOP2UHpUDiWVT282ZVcfdeyrL3GmB3GmCWWZW2SdKGkDcn/XSnpy8n/3laRkQIAgKaX2Xg+15rId54xR7c8vSttXzDgK6vqZK6spzFGPmM8731stDGnJ9ofTq0CX4+xaEwPbkpfLjVrUukZHrQO42RKc/8ueX151A5GwjG99vr7ddHymYrFLU3uTqwn/a8PrHAKpJXDZ4ysPF8t7R1Iz1Cb5NvlV+/elLbfsqymr4JcVksYSR+XdKMxJiRpi6Q/VSL7erMx5oOStkt6d5nPAQAAWsRARpCX68OuV9XboN8UtYYyky/P+lCfz3hmLAYbPCg9PFTYerLrPVrgAFLqy5p832/kC1hbmd266u7n90mSpiQzpW8qo7iRmzH514X/8JFtkqSv/MGpknJ/sbbwmjv04j+9JW29a7Mpa+SWZT2TXBd6qmVZl1uWdcSyrEOWZV1oWdbi5H8PV2qwAACguQ1lTI/dfti7yEfQo4Fm0FdeplTK3RPVZ6ThcHYAOphnvVc92fH1P99VWLC5Zisfx+DNLuYVy/O3ldnGyfaqWcVNoW82mT1Ep/QEc5xZGmNM3in4J83slSS9Z8U8Sfmrgeer4tsMmjecBgAATScz4zKpy/tD3ssHjmXt8/m8p9gWI9d0V58x2jcwlrXf3Se1kRQ7VW/WpM4qjQTNzv5VypcNzax6LSXaOZ2/pLWLlWb2EJ1SlTWluV/3OZO7dPKcwgL/Ri3KViiCUgAAUBWxuKX1GX0wM7MxuzLWmNoe25Ke2Zs5MZHhPHnOJEnSd/+o8Kq7buec6F1t1uczcs/uveMTq0q6f60U+/nzgqWeHfoAZ8pnvpYw0Vj63+1/fWCFAn7T0tN6o7G4vvvQy2n7Kh6UmtzTpiOxRHGlkGvWSNBjWYOt2f8lCEoBAEBV/OjRbXrbN1frsS2HnH2FtjLIzKB+7LwTJUnLZ0/Shi9cpLecUlp/0q8m12ZlcgekKxdM1Zwy2jzUQrEJ42b/wIrqsXsF7+n3/oJIys6UnjxnogI+nyKx1v3N+r8nd+pnyT6htqm9lQ1K1+8a0KZ9g3phz0DWscWfvVNPbz+atk404Pfp/a/1riL+w9VbKzq2WiMoBQAAFbf14JCu+80GSdKTrxxx9mdmVnIFf99xZUJ/8CcrdOU5C5zt7lDpdRrtD+CZfMY4H7zfeeYcTegstxZkdeWr2OmlFfsaojK6kn8Tn/rZsznP+ec7E2uX//vPVmrr9W/VrEldCvhNzrWmrSCzfZUkTeys7JpSe6bI4xlrvr3Wt9tyrSv9l3tfrNzA6oCgFAAAFOWOdXv0iyd35j3ni7dvcB6PRVMfXN1rQt991lz99wdXel7vDgrLCUK9fP/KFfqHty1L27enf1TP7UxMNfYZk7dSbzOyA+4W+7FQAbm+qLGNRWP6fXK2Q8BnnPXMQb8v75TfZjd3SnfNnmvmxPQ13+6q35lLGfIVO9qX0UKmmTT214AAAKDhfOzGpyRJ7zprbs5zFvX16r4X9ktKz9LZAer9n36DFvX15rx+cldimtxxEzv1moVTyx6z24WvmqmRcExf+M2GtP39I4lKu3ZAev07T3GqXzacIhOfsWRG67d/e34VBoNm5u4V7NXv0p0MdX9ZE/L70r5wajWZlXc/tGph1Z6rpyP9i4GBkdxVv/OtJ9+weyArwG0WBKUAAKDiZrg+GEU9gtKOcfrpzZ/Wrfv++g1a1NdTlabwdsGQK88+PuuYXVfkipXea7cawaTu4qYR2gmtvgneLXHQvtzFe8ai8azMqbs4mTuDFwq09prSzMq71cycZs6u33Yo1SrrxBnpX4zly5T2dDRvaNe8IwcAADVX6NpE93nuKbuf++V6SVJHIP+UQSn7w1glBfw+bfjCRer0GEdmpdFG1BHw67UnTNXm/UMaGouO+2HUzpTm+0CL9uQOQj2DUtffw9wpqTXgIb9P4SbvjZmPu+/n1//wdF162uyKP8fKBVP1+LbDaW1hBkYj+tIdLzjbP/3Qa9KucU/BnzO5K62CeW8TB6WsKQUAAAU7Mhwu6Dx3dtQO8g4eG3MypY2wtrE7FHCmI37ighOd/Rs8KmE2oqDfp4PHxvThn6wd91w7UxpohBceDStzyqokp5jRZy5eolfNSvXMDAZM22RKLz9jjvxV+Nu55q1LJaVX077+jo3aenBIknTV+Ys0Y0L6dFz3FOrMab8EpQAAoC385U+fLui8f75ro/PYzpQ+s/2os29qT2VbK/x/9u47vI0qawP4OyqWux07dnrs9N5NKqSQTgIbeifA7lJ3WTpJKAkQIAt8gaUsu/S2dEJNIL2S3ntvTnViO+5Flub7QzOjmdFIlmzZku339zx5MhqNpGtLtufMPfec6lKf6NW14i36QihGlJlSBqXkgz5lFXCn7+orz0aEeaGj3KJyzQxkoOQLaHPuGxysIXmQMxfU1bTLVFV/HxndyetjAOCRMdr7o22VZ6CEKwalRERE5LeNxyoPgPQ+Xn0UH6w6gsxc9zqpmlgnWh1m1XjK60jxFnkGurL1uQDwxpKDNT0cqgeKyz1nSuU2TvpZdmsYFzo6cr4IfZ5fiE/XHKvyc+SV2GExCejdMjGII9PKL3UVNFq4+6yyT1153OgiknpXXKQFjVW9UzlTSkRERA1CRRX7XT7/6258vrbqJ4g1zWwOryDZH/IMltXs/XRu5YFzHr1hibyZOXe3xz65z7A+fdVV6Ch8gtJHvtmGPs8twJ7T+bjh3TUAgOk/76ry8+07k4+05OgazS44fM6Vpvvl+kxlX2Wvp54ptZhMmgt8lbX3CWcMSomIiMhv/mTDnc4rqfygMKOeKf3HqI4hHIn/5BndEoPZLQBYui8Lt36wHv9Zfqg2h0V10JOXdQEArDxwHgCw/kgODmYVAHCn7HsEpWZTWGUVfL/5BHKL7Rj/r5U4m19W7efbc7oAvVrV3Cwp4HlBqaTcgY/+OOrzMerZaYtZCIv1+cHAoJSIiIiC6nSecQP3Tk3jankk/lOfcCeH2XpXb7ZmutbolnuZrVq619UndsvxXFzSoTESA2wjQw1H37RGANzpn9f9dw1GzV6hOSbcZ0q9sTucSJ8yF+8sC+ziTIXTsxJxsCWrUm+dThEf/nGk0seoU30tJgH/uqFPjYyttjEoJSIiooD5ujpvVCwFAMZ2a1pDo6m+nCJ3VeG6nAKnJq+nW30oG2aTgLSkmuuzSHVb39auGcHJg9OQMXOR4TEWkzZsyC+1Y//ZQizec9bw+Nr05A87vN534GwhAG3xNX84nGKNV6se07WJsv31xky/Xu+afi2VbbNJQMcm4XuxLxAMSomIiMhvUVLA1kLVr1Cv1EvvwnBe2/jvAGdR6pLicgfKK5w+155SwyYIAqKsZtgdIs4XGqe+6j8+fxzMBgDMXri/podXqf+tO+71PnWBtUBUOMUaaQOjpl4POnXODsRGVl6oqJMqCLWaTZqlB3VZ3S3RRERERLVOTtdz+sja256ZZ7hfjknTkzljV9uKyioQU4crc1LNs1lNKLUbX1ACALPJ+KLGseyqBX3BUtnFrrP5xssJvDmWXQSbxYyC0gpsUbWxqg1P/rCz0mOcqoX9JkGAl7elzqknXwYRERHVNKdTVKrvnrzgvZjRa4uMZ04e/XYbAOCLvw4M/uCqyVoHq+9WZkj7ZGV724k8zpSST5EWs9fUe8BzplQ2qktqDY3IPwVSWxW9yYPSAGhT8/0x7JVlGPjSYgDudduhsHrKpYb71TF4SbnDI626rqofXwURERHVOG8FdQIRZ7OgaXxkEEYTXP+8umeohxB0yTE2zW0GpeSLzWpCmS71fudJd9ZDapzxz233Fgk1Oi4jm4/n4s3FBwAAD3y11eP+LU+PxqB2jQEABaUVVX6dKeM7V/mx1dU80XiJhHqmtHOzOM6UEhERUcOiD0q9tYNonuA96EyNt9Vo37+q6tu6UaiHEHT6yqg2C0/7yLtIixmlupnSiW+uUrZb6taRf3WXK+MhGBer1HKLyvHSvD2o8PG8V/17Nf5PWsu6Yv85j/ttVpNSNEg9kyqKIjYczYHopbeVPhV4RKfQzAJvfGqU1/uGdkxRtuvTmlL+diIiIiK/2HVBqLd+pDarGZf3am58nyU8K9vqT7jrA/15d31MUabgMZopVdPPtGdIbWS89cmtqplz9+C/Kw5rWp8EKspqhln6vOeXuGdK52w+iWv/swY/bT1l+LhC3axqxyaxVR5DdTSK9t6WKtZmQfvUWOXnWS7GdNfQtrUytprCoJSIiIj8op8RmfHzLsPjyuwORHqZlbNZw/PUw1IPU1tFaKPSE7ne1wETGc2Uquln2uVg6M0lB4M6DrnYksPLbKa6OrA6IG7bOAYA8NSELhAEQZkp/X3XGeWYLZm5ALz3Uj6WU6S5LYRoFrKyqr8LHhyKvc+PB+Aa49FZEzDtsi61MbQawzJsRERE5JfsQm3BkCIvMyRlFU6vwWdUGPcA7dw0DpP6tAj1MIJGf06/8VhuaAZCdYLNakJRmff1l/oLNzUVsMlrJk1env+Rb7Yp25tUn+k/X9IGNw9IU24bpbWeL3D9DkuIsir7Fuw6g+JyByb1aYH/rfXeWiachOMSiOpiUEpERER+mTpH26C+qKwCWfmlSNUVLiqrcHpN040M46D09weHhnoIQRXGbWEpDNksJuQUBWd9qKMaPT7lteo/bjmJ9OQYdG0er7lfXRFXvV5UgPb1jF5fnmWNU/UDveuzTQCASX1aILfYfeGtX1rtrDNvHGvTzP7Wx6UE/qh/uSpERERUI3ZIlTjlQka7TuWj/4uLlZPI49nFOJhViLIKh9eiOuE8U1rXREjf4wwvJ89OL+mPREbMJqHSnp9612W0RGqctsrz/9YdQ7tp8zSBViAyc119TxfsPovL3ljpcX+aqs/xA19tQftU17rPsd2aaI6zGKyhPiP1LH1toWfbqrWHs7FAtY61dVLt9FNe9thwPDqmo3J7TNemtfK64YZBKREREfkl1uaaXXjl2l6a/fK6rqGvLMWo2cthd4jKTOlVfVpAnUUXrmtK6yKnFEB4S+XTV98l8sUkCAFfyLBZzMrn7I3FB7D5eK6SAnvGy7pNX7IKSrH/bKHPYy5KT1K27Q4RrZOi0aNFApJjtcGxWdUrRb5IJq+rPnxeu3YUAG54d63X16lJsTaLZuxGwXRDwPRdIiIi8qm4vALRERa0S4lBQnSEUkBE5qrYadXss1lNOPLSZQCA2wanY9LbfwAAzuYHfqJKxiqkoNTbrLQ+KG3hpe8hEeAq/lNZQKhnNZuQW2xHqd2B2Qv3Y/bC/egh9S2tyky9PxV39eteHU4RRtdl1GtKrWYTyry0sDLy9V0D0b9N7QSlADTjr2rac13Hy5VERETk1bbMC+j6zHws3H0WeSV2JEZZ0UW3xsspep4oRlpMEAQBgiCgd6tEZf/R88W1Mu6G4Pt7Bynb83acRsenflMqlwKuWSS1f9/ct9bGRnWPeq2mXi/Vz7DavrP5AIBnftqp7JODKv3nzx8tGxmnzO4+lY9vN2YCAArLKtBYNbPoFEXDbAF1cHdZD+OUWG/9Sge0Ta7Vyrvq9bD6i34NBYNSIiIi8ko+UV2x/xxK7U5EWk2Ij9TOijpEEW8v1baFWHXwvOb2y9f0BAC0kdo2UPX1S0tCv7RGcIoinvlpF8ornMhXFX6p0M2UxkYyQY6qptRLpW058DyiSoeVg6oye+D9Syd/uN5w/2VvrMRj320H4LoA1jTBhmYJkbi6b0usPHDesFKvOg02OsKCUV20a06dTlHJNgg1gTOlDEqJiIjIO/n8yCmKKK1wGFbPdTpF5JXYNfv2nS3Q3E6SmsFHeEk1paoxCwL+OJiNnCKpqIzqHLvcIWpOxCPqYS9Wqh3eeoZapcBPnRorB1XPz92D53/dHdRxlNodOHy+CDERFtgsJny/+QQAbWsYmbrtS4TFpIxVtvzAOcN1197S4WvS+iM5ynYTXTXzhoK/nYiIiMgrOS3OKbpOCA2DUoMTVn2PwHLp5I+BUXBtP+mayZYnfNQTPxUOJyIs7veBFwTIl2v6tVS25XWhsuszWhk+Js7mCvxyitytVOQZyj2n8/HBqiNBHWPnp3/Hsexi5JXYvbadkqmDO0HwLAhmr3DCXuH5uysU2RwrDpxTtpsyKCUiIiLSktPiHE6nK33XILBxOEXoT+0eHdtJc1tuG2NlYBRU+nV78oyWKIo4kFUIq+oigJUXBMiH5qpCWJf3aobJg9KU238d2tbwMcmxrgyIvGJ3psSOE3lVHkOszYLRXbVptq5Calpdm8UjMoBK3h+sPOKxfjQu0oqD5zwLO02/vJvfzxss6ot4oZipDQcN86smIiIiv8izoKV2V1Bpk2ZKB7dL1hyjP5Ga2LO55na6NPtwaeeUGhtrQ6QviiK3iXll/j4AwE9bTyn3caaUfFH/DJsEAd11s6VGHpMuPo3r7i4klF9a4e3wShWWVaCVrtjRAYOKwFazqdKZUrWeLRNQIV3AubRzKgDXz8PV76z2OHaQ6ndbbVHP4jbUn9OG+VUTERGRX578wVVVs1gqdCKn764+lK0cczCr0CNdV693q0SsfHwEruzT0udxFBh9OvQtH6zDjhN5mLP5pMex+jV1RGrqoFQQBCUNNzrCe/CXGB0Bm8UERxAKBp284Ooh+uEf2pTfv3660eNYs1lAfJTVY7/ekZcuw+8PXoLv7hmMbFWKMQDsOOGuNtwvrVFVhhw06uJG/nxd9RGDUiIiIqpUXonrhE4+QVW3ZLjn882Yv/tMpUFPqyTjdg9Udfp06GPZxXjyxx2GVUW5npd8UQdDAgCLyfV5iY7wXbU50mpGQVnVZ0dl32zINNzvrb9olI9gWSYIAjo3jYfJJCjFkNYddl1Qm/GLuwjThB7NAh1uUKkv6jWSisI1NPztRERERJWSq+umSMHo13cP1NyfmVOC1LiGWaAjlKIMCk9FWc2IinCd4o3o5E6Xrs2+i1T3tFUV+MnMLVYuMlW2xjGvxI6Fu89W67VL7Q78a/EBAMCMy7viziFtlPtyisrRqUmc5vgFuyq/COZNjM0zyJbXxobKM5d3VbYTozlTSkRERKQxtpur6EhOkSsolU/e2qXE4v3bMjTHyul3VHtsBsVeYmwWlJS7ZpeaJkRh7gMXY4bqpJfISEZ6krL93cYTMEszpaZaiBbUrVlS4yPx1IQumvuLyrUzsecLywMuCNRaytT48PaLPO4b260ptk0fg63PjA7oOYNFfUGvoRYkYxdlIiIi8kqu7loinRSqi3AYnawOapuM+CieXtQWk8HsZ1SEGeVSxVKTAHRrnoBuzSsvWkMkc4qisqZUQM3PsDtVGbpWswkmk4AoqxkldtfnuMggPTi70L1GtIWqcrA33987GAeyCtC9RQK6NY/HrlP5AIBuzeMRaTUbtruqLa2SKh9/fce/GkREROTVkr1ZAIAiqdCRxaSt0Kl297C2mDq+i0frBao5JoN4wWY2Kf1K+U5QIJJiIpBTVI6icges0s96bWR9O1S/M+RgeNplnfH0T7sAuH//qG0+nqtsL3p4WKWvkRJnQ0qca/mBHJACoelLqhcX2TBTdtUa5vwwERERVcppUCxHfYKqD0pjpYIoXLtYewxnsQQo1VB5fYACMV1K8x7SPhkxNtfMod1LoaFgqlBNlcrB8KQ+LZR95QZjKFTNnvpT9MibQFrLUM1hUEpERESGvjKohin3+gO0bQwAwNJA10KFklH8bxIEJe2Rs9YUCHkmMcJsUmbvSqsQlPZXrU9dffB8pcer03fLHa7PblykFf8Y2cHw+B/uG6z0Tq6Kvq0Tle2LO9R+X1IjX901EP/7y4BQDyNk+NeDiIiIDL274pDHPnn2BPCcKWUfzNpXbJDWqH4XGJNSIPq2boRJvZtjxhXdlLXhZXbPz5gvP90/BL1audcw7zyVV+lj1Om7cpEuAF7XeXZoEofP/tw/oHGpje3WFIBrRjhceicPbJuMIe0bh3oYIcOglIiIiAydyS/12JeW7F5/pV/PaNQbk2rW8Zxij33qawUiV5VSACKtZrx+Qx+kJccgXpop9dYnVPafW/oq2+/dloFerRI1rYq6NIuv9HUdqgwM9ec3wkuF3SirGRepZmMDJVe4DYf1pOTCQkdERERkyO7wHdDo03eLDSpkUu1TXxvgdQKqKpvFhDuGpGNiz+Y+j5NTfgH3hapuLdwzpf5XeVs9AAAgAElEQVRcrHp+7m4AQIfUWGUWEwBivKwVNZsEmAQT4iItuKZf4DOdcguahtp+JRwxKCUiIiJDjkpOJgt1QWgBg9Kw4FSlQjJ9l6pKEARMv7xbpcdFmN2BozzLOapLE0wZ3xmzfturWYfuzcLdZwEAd17cRnOxKyHKXZU21mbB9Re1QqoUBAuCgK3PjDGsQF2ZE7munsotG0UH/mCqEbw8QERERACA03kleH/lYY/98kmiXJlT1jjWprldWMqgNByoA1H1LBZRTSh3uNN75crbZpOAYR1TAAAVDv8LEunXSMergtIOTWLx9MSuuHtYO2Wf2SRUqdp3WrIrGE1LYlAaLjhTSkRERACAQS8tAQBc1qMZmidGoX+bJJRXODG0YwreWHwAV+kKgnRXpegBQCue4IWF1Yfc1U4fGm1cvZQoWJomRCrb6uJncuEzeyUZFwWldmW7VFdUKT5SO1MaLDcPSEOrpGiM7JIatOek6uFMKREREaFENUNhEgSIooj1R3KwNfMCHhrVAdtnjEFCtO8G7/cNb+fzfqodZ/PLlG32YKSa1iIxStlWp9JapH6jy/Zm+Xz8WVVBNY+gNModiDZPiEKwREWYMbZbU/ZUDiMMSomIiAhdnvldc/vTNceUbUEQNDMW3rBPKVHDJqgaElmkmdI5W0567Zf79tKDGDV7hXL7tkHpmvvVa0qf1i0foPql2n89BEEwC4KwRRCEX6XbSYIgLBQE4YD0f6PqD5MoOF6Yuxu3frAu1MMgIgprdocTsxfuB+CqwOmP1kzdJWrw1DOl6sq2babOMzz+jcUHNLf1a6DVKbvBTN+l8BOMS5r/ALBHdXsKgMWiKHYAsFi6TRQW3lt5BCsPnK/8QCKiBszhFJFX4lrn1ad1ol+PWfH4iJocEnkx577BoR4CkUKdDhuMvsXMvmg4qvVOC4LQEsAEAO+rdv8JwCfS9icAJlXnNYiIiKh2VTjd1TKfnug7Ze7jOy7Ch7dn1PSQyIu+rRthUNvkUA+DCIC7JQwANFcVQPKmrML/yrxUv1V3Hvx1AI8DiFPtayKK4mkAEEXxtCAILGtFYcHJDuJERIb0/Ujtqr6C3Zon6A/XGN6Jf+ZDzcTJJAqxlo2icCK3RFN9l0WEKBBV/jUmCMJEAFmiKG6q4uPvEgRhoyAIG8+dO1fVYRD57e9fbgn1EIiIwtKhc4Wa25y9qFtMXk7+72U1ZKolEVKarYlxKFVRda6tDQFwhSAIRwF8BeBSQRA+B3BWEIRmACD9b1gHWhTFd0VRzBBFMSMlJaUawyDyz9wdp5XtvGK7jyOJiBqWCod2pnTd4ewQjYSCqUNqbKiHQA2EXNSoOpOj79zc13C/uuUM1V9VDkpFUZwqimJLURTTAdwAYIkoircA+BnAZOmwyQB+qvYoiYJsxQHOzhMRyQrLKjS3S3S9Aqlu4vtItSVCqdKtjUq/umsgAP8q5/ZsZVxUbemjw7H3+XHVGh+Fv5pYhTALwGhBEA4AGC3dJgorXOZAROSWL1XalU8sX1/katPw4/1DQjYm8l+5l3Rrb/uJgk3+3aHvRzqwbTLuGtrWY9260bFmLydnERYTIq3mII2UwlVQglJRFJeJojhR2s4WRXGkKIodpP9zgvEaRMFUaucfaiJq2M7ml+LtpQcBAAVlrqD06QldNMdwfVjdYHcY/03ztp8o2Hy1f7FZTCircHgEob/vPKO5zYJdDRvffmqQWImXiMLZ1DnbMXr2chzLLqqx1xjw4mK8Mn8fPll9FBuO5gIAxnRrqjmmU9M4o4dSmPEWEHCmlGrLE+M6oV1KjGG1bpvFBKcIfLb2GE7kFiv71x/VzltZGJU2aHz3qUFqlxoT6iEQEXn15fpMHMgqxLBXltX4a03/eRe+WHccAJAQZdXcx5PEusFb8BkVUd3Of0T+GdyuMRY/MhxREZ5ptjaLa98zP+3CVf9erez/6I+jmuO8pe9Sw8C/NtQgMaOJqO5Zcygb3286Eeph1DijtVfB9uePNxju16/bYvpu3eDtM3PboLRaHgmRp0irO9zIKijzehyvgTVsvIRGDVJtnPQRUXDd+N5aAMDV/VqGeCQ161+L9tf4ayzea9itzYPAmYs6TW7TQRRK8boMDG+YmdGw8d2nBolBKRGFqzeWHAzJ63ZvEQ8AODprQkhen6qOf9EonMVF+jcHxpi0YePbTw2SQ+SfcKK66sj5IoiiiHMFZbjjo/U1Wgwo1G64qFXQnqu4vAJP/bgDeVL7F73P7hwQtNei2qWvakpUF0zq3Vxzm2tKGzYGpdQgsfouUd014tVl+GL9cVz0wiIs3XcOv24/HeohBVXXZvHKtt0RvN9VX6w7js/XHse/l3nOxD57RTc0iokI2msREcnySyoM91t06eVmLmJv0BiUUoN0x8cbMHXO9lAPg4iqaNOxXGW7WUJkCEcSfGUVDgxok4Qm8TZUOINXla2wzHViuFn1vZNx7WHdxsusFM7UhY4sqsCzvMKJ9ORo5TbXsDds/CtEDdaX6zNDPQQi8tP2Exc0t9XrwutT4oPTKeLQuSI0T4xCo+gIFJU5gvbcpXZXgFsoPWeyamZUH5MueGgo3r6pb9Bem2pYPfoZoPpnrKr/cYVTVNLNyyuciLAwFCEXfhKoQWBhI6K640xeKZbu01aHveKtPzS3f9p6StmuL+n4eSV2tJ02DwDww5aTaNkoStNovrr+s/wQAOCk9JxLHxuOq/q2AADolyR2bBKHCT2bBe21qWbJb98bN/bB9hljQjoWIj39DGi51Jev3MGglNz4SaAGwc7GpER1xsQ3V+GOj4z7aBr5bnP96F362w7t2ti4SCuKyo3XYlVHfmkF4iMtiI+0IkKaImXxt7pNnnmS31eicCZnbSzffw72Cv7uIRcGpdQgVNSTmRSihuB8oau5unyiXdlM6PojOTU+ptowZ/NJZXtS7+YwCQKCuKRUo6jclcLbpnEMAKBRNIsc1QcmrsmjMPXDfYPRITUWgGvdPAA0T4xEbnG53y1jqH5jUEoNgr3CdWZ368C0EI+EiABgxs+7cOO7a30eI19M+u+Kw7UxpJBTFzW6dVA6TALgDMIMZkm5A8t06dDykoa/XNIW79+WgVFdmlT7dSh05E+JunrpRKZfUxjp07oR7hraFgBwIrcEZRUOWM0m9G+ThMUPD8P8B4eGeIQUarw0QQ3C1xtdRY2+26RN8ysqq0CMjT8GRLXt49VHKz2mwiHCaga2ZV7wedyfL24TpFGFVvcWCdh83PW1RkeYYTYJOJ1XirwSOxKiqp6S+cT32/HztlOG95lNAkZ1ZUBa18nXLuSJ0l3PjoWNa/UozNisZgDAVf9ejVFdUl2FjswmpMZHIjW+flVRp8DxNxY1CLlF5QCAEru2kmW36fNZBIkoTMnFMGIrSe2qqAdrxt9Zdgifrjmm3I6ymrHndD4A4Invqta+6sj5Ijz89VbDgDSe6XL1kpy+G2OzePSAJAo19YWSRXuyYHc42Y6KFPwkUIPQrUUCAODVa3t53FdYGvxCIkRUNaIqXTWv2A5A29fOSHk9CEr/+ftezW2zSUCZtOzgnLTGNlAPfr0Vc7acNLyP1+LqFzlt11zJzwpRKOln7+0OkdV3ScFPAjUIZdIMaduUGI/75HVc9WG2haiu234iT9l+9NttAIBzBa6grFmCZ3pXi8QolNfx6o2iwbrRlo2i/AowyiucuOOj9Sgq87y4ZvS8r1/fW3l+qj/kCzeMSSmc2Sxmze2ScgfTzEnBTwI1CKVSUBpnsH7UKQLrDmej/ZO/YcPR+lHFk6iuKNQFU+oU+/VHc1Bqd2DxXleRnmv7tVTu+/D2DCx7dDgiLCZlpvSbDZlInzJXqexYV1yQZoRlMyd1hyAISiqmPs64+f21+GHLCZTaHej41G9Yuu8cuk2frwTv6w5n41xBmSbAl13RqznuGdYO70/OqJGvhUJDToFk9V0KZzarNuwosTuqtV6e6hcGpVTv7TiRhyPnXc3ijYoaiRCxfP85AK6TOSKqPT/q0ktLdeu+L/7nUmXbbHL/yerWPAHpjWNgNQtKde1XF+wDAGTlVy3dNVT06bm3SFXCTQbTXk6niD8OZuOhr7fhrs82ae676IVFAIDr312LgS8t9njsJR0aw2QSMGV8Z7RsFB2s4VMYsEqzTXU7Z4Dqu0jdTCnAln3kxqCU6r3L31qFD/84AsBLUCq6fymyMARRzctTzQwmxWj7Y+qD0vOqgG3hnjPKdqz0s6yeKc2SZgoveXkpDmYVBnfQNShHKsSmZzaY9Cosd88sr5AuphkxKuB277B2gQ+O6gSrdAGjwsETfApf+plSAEpBNyKegVODYjU4yxNFwC6d1FZWUIWIqk+dXrv/bIHmPn2FbLVp47so29ERrivuVrNJ+fltHGtT7t9dh050LhS7g9KBbZOUbTkVc+OxXGXfToOUXDWnKhiNVV2Ee3h0Rwxu37jaY6XwJKfvsjYChTOj9aNM3yUZg1JqUIwKhzhFUbm6zNLkRDVPrioLAK8vOqC5r6Tc+KT62Su6KT3uAECQArYIs0l5vhib+357RfidnJ/OK0H6lLl46scdmv25qpljdSEQ9frAR77ZhlK7A/N3nYEv6lQ49Xrdk7klVR43hT85fbc+VKKm+suo0m5cJINScuEZODUoZoMiECLcFXgtRvlyRBQUpXYHdp7Mw5zNxm1Kftp6EtN+cAVs+gtIY7s1hdOgmqwI4Gx+KQBoKtCekfaFg3MFZfjP8kPYevwCAODztcc196vTd2/s30rZVi2hxfebT2DdkRx8ouplqhdns2DHyQuG912T0dJwP9UPcvqunem7FMaMCnHFsWcySRiUUoNiOFPqVM2UmvgjQVRTHv12Gya+uQqvLdqv7LtGVVH3H19tVbZTVKm4gCv1Xm4J89jYTsr+9UdycCy7GHaHE+cL3cFdmcFMaWZOcUgqbD/x/XbM+m0vdp5yp94u2n1W2X5lvqtA09FZEzCuezNlv/73VaKPNLe7h7VFmcOJT70ErT1bJlRp7FQ3jOveFIBx2zOicJGsqyEAAFf2aRGCkVA44uUJqtf2nXGvVxveKUVJ+dNzFzriTClRTfl1+2nNbYtJQKRU+GLnSe1aycRoq2a202oxITnWhg1PjkLjWM8TmzmbT2huOw0K/Qx9ZSlE0RX8GTlyvghJMRFBX+Mkz4S+vfSQsu8vn270Og6ZflYhr8RueNxVfVogNsKC8gonCko9+5UCnv0BqX65NqMVJvRshugIntZR+DI6B0tvzAsp5MLfXlSvjX19hbI9rGOK4TGZOcVKoRSuKSWqPalxNpTZXT97x3OKNfcNaJOEvaqLShHSz2ZKnHYGVfbE99p1mmcN0nfl7F9RFA1Pjka8ugytkqKw8vFL/f8iVJxOEYJgfOLlS/82SR69SAF4pCvf8fEGj2PkwPaDVa4K42t1ba3+95cBaJEYFdB4qG5iQEpEdRnPwKnB0FfWlWdD/rvisCoo5UwpUW14fFwn2KxmJc1W38LkifGdkZbs7qUZEeAFo283ncCaQ8Z9h9tMnYf7v9hseF9mTtULArWdNg+PfLvN7+Pl2dzC0grDdVX69YFGbV5kclXL4nIHolQFoWJsFs5EEBFR2GNQSg2GWbde9Ou7BwIAhnZMUdaUWrimlKhWXNO3JWwWk9IeRr0G9Pt7ByM6woIRnVKVfSYv7Zq8zZwCwI3vrUVBqR19nluAiW+u1Nw3V5dKXF1ygOmtiJMRuedoYVmFpn2LzO6jkmpyTAQeHdNRua1O9dX3fiUiCkdsB0NqPAOneku/pky9XjQ6woxm8a6Utud/3Y3Fe7MAGBdCIqLqK9X1H422WaSg1BV4vbnE3RqmX1ojAO7Zv0m9m3t93sp+YhfvyUJusR07T2r7lsoprU6niJMXSvDivD1+fR3eeFvLCQDlqoB7+4wxmHVVDwCuGVJACkoNZkp9FTba9PRo/O3SDsrt6y9yV+09X1iGx8Z2QnykBb1Y4IiIwtRlPZpVfhA1GAxKqd6q0AelUsD5xV8HYOHDw2A2SNUVwXL6RMEyb8dpZa3j+cIyzX1RVjOcInD0fBEA4Fi2a03p+mkjlWPk9ZHzdnjvzfnM5V01t7+7ZxDWP+l+jmxVuxW1kxdKMODFRbjtw/UYMmsJ3l1x2N8vy1BusefrnM4rwYGzBbigui/SYsbWTFfblu82uYozFZZWINbmGYAadfdokRiFFY+N8NhvNgno3yYJgGvW+f4R7bF9xtiA17cSEdWWCvbVJRUGpVRv6ddfybOgg9s1RovEKI81poC7EAoRVd99/9uM53/dDcBzJtFsErDjZB6OZhejrMKBIe2TAQCp8ZHKMc0SXLOZ5T5OXCb21M6imk2CJhVWfn0jZ/PLsOrgeT+/GmMzft6Fv3yyUVMZVz7RGvTSEox+bQVyVEGp1SygWwvX7OW5gjIMmbUE5Q6n4ZpSowrCXZvHo7Vqra1ah9RYAK72MERE4c7XEgVqeFiqjeqtCqf2l51+xsAoVZdBKVHNMOobKusxYwHKK5zo0zpRs///ruuF36efwcSevlO8JvRohrk7XGtEbRazzyrawzqmYPn+cwGM3LePVx8FACza4+47WmJ3IFJ1TKndiVsGtsbF7V1tqS5u3xgA8Nlad09RozWlRoWNFqr6m+rJgT/XaRFRXaAv5kYNG2dKqd7Sn9Dp11aZDdLa+OuRKPiyC8s0LVq++MsAAEDnpnEA3Gsu9T+RMTYLlj82HK9e28vn8xeUuWdhuzaPh8UkYIJBINundSI+uv0in2vHm8ZHer3PXyXlDpy6oK3iGxdpxbjuTQEAMTbPnqFGacaXdGzsse/2weleX/fnbacAAFn5ZV6PISIKF76yYKjhYVDqh5yicuw9k1/5gRRW9GtKE6O0FSmNqnmKnCqlADmdIhbsOoNNx3IMe2MS0G/mItz92SYArr6Zg6WZwruGatNMNx+/4PHYtOQYRFo9gzi1FbqZT0EQ8ORlXTyOe2BkB5hMgs/WKo4g/A7IzC3GsFeWafbZVTPFqXGege8tA1p77LtnaDusmzYSSx4Zpuwb0TnV4zjZy9f0BABc069loEMmIqp1XFNKagxK/TB69nKMe31l5QdSWNGvx7JZK/+4MySlQH2zMRN3fbYJV7+zBhPfXBXq4YQNb4FfjCpNNdjXgOSKuvrXcb+g679kHy1TgrHGad2RHI9916mq4wKuAFk2qXdzzVpamckkoEl8JOIi3em4ZboqxprXyGiFo7MmoHsLVtwlovD0/b2DMVK6uKafPKCGjUGpH+S0qqIy7yX/Kfzof9lF+FhnJuNEKVXG6RSxYv85JXg5nlOs3HeuoIyz7RJ9CxiZ+ufwCl2rlzuHtKnWa6r7cxqtq5Sra783OUOzXw5Se7VKVHoW+8Pb34SXf9+nbL95Yx/sfm4sOjaJ0xzTLiVG2X79hj4+X0ddlK243HtQSkQU7vqlNcLzk7oDYEsY0mJQGoB9ZwtCPQQKgHqmZv2TIw3TdfW+3ZhZk0OieuDtpQdx24fr8fnaY2g/bR7+veyQ5v4cLy1IGpoSb0Gpxf1nR1+QKNZgraU/5BnS0V2baPa/e2s/zW1BWrWqv0DVNME1S2kSApspferHnZrb79zcV3P7rZv6YGLPZoiO8Jy1DSTVOyrC/X2J9CPjg4gonDVPjMKBF8bjBl0GCTVs/OsWgOPZxZUfRGFDnimdPCjNcA2XkcV7s2pySBSGdp7MwwE/LzjZHU7838L9AICTuSWGqUdG/SobIm8zpTaL9z873aqYdrr00eFY+fgI/G1Ee83+Md2a4tM7+yvtVixSb+IuzeI1x75zcz9cn9EKA9okBxSU7td9bga1S9bcHtoxxWufUMP0Yi/Ua2pHdmni40giorrBajaxjzJpMCj1w7JHhwMAHvx6a2gHQgFxSC1hMtKTvB7zzd2Dqp0ySHXbxDdXYfRrKyo97qetJzHrt73K7SIvaZQ/bzsdtLHVNXnFdhzMcgVqpy4YzwT6atdiNVftBCXCYkKrpGjDbIihHVPw8R0XoWl8JHq1crWcMZsEPDq2k3JM6+Ro/POanoiymuEUva+H1dt1SlsATx1otkiMQnyk99Ys12e0QtP4SKU4UWVuG5SG5/7Uzef3j4iIqK7iXzc/tE5yNSof3iklxCOpe7ILy3DPZ5uQX2qv/OBKHM8uxkldmwVf5Fksi4+03f5tkjD1ss7VHhvVPVn5pfhg1RHldmVrQf/x1VbN8V+uP665/4u/utqcvLH4gNdZwvrM7nCi13MLMGr2CmzNvIDTee6f1Y/uuEjZjtJV0m0ca1O243wEcdXRLy0Ja6eN9BkkAu6ZVH9mS40+L+qAsX8b7xfDXK9lwtppI3Fdhn/pa8/9qTtuG5Tu17FERER1DYNSP5hMAhKjrUpwSv57e+kh/L7rDL7ZUP21mkNfWYohs5b4fbxcsMRXT0LAdSK54KGh1Rqb2s6TeQ0yKKlr7vl8E57/dbdyW9/CI1AD27hTN/NKqn8Rpq5Rr5FccygbKw+cBwCsfHwERnRKxb9u6I3bB6cjIVobGMo9O+8Z1g4ZaY1qb8BwXUj4zy3udafyTK0/FSH1s6Syf93QGwBwIpfLPYiIiPzFoNRPNosJZXb2UwqUHA86q1GRVBRFLK3CWk85Bc/iR0pgsFY1ZOYUY+Kbq9D56d/ZszLMnS/Urv1UV9HV08+KxRqsB1Snjur7ZjYE6gAfcBfkaS4VIfpT7xaYcUU3j8dFWszS/c1rfX3R4HaNMa57U+W2PNOp7inqzfnCMgBAqyTX1ycPfXTXJkhPjsZDozoGebRERET1F4NSP0VYTCir4OxXoOQT9eq0ovp1+2nc8fGGgB9XIa0pNflxohusc+HXFu1Xtm//KPAxU+3RN+32lQlxVFfk7OaBrQ2PGyClbOaXNrz2Uc1VPUJXHjiHY9nFsFlMlWYq/OvG3riyTwt0SI2t6SFWyiIFpV9tyERJJa1XsgpcQenbN7kq7sZIFXajIyxY9tgIDG7fuAZHSkREVL8wKPVTpMWMMj+unpOWHOz5WzjESLY0IxEou5S+609/0mDNlRaqgpE9p43T+yrjdIoo52etxtl1n8nB7ZIN1wmKoogFu84ot1+9thdiDVp8AMAnd/YHgAb5/sVEWGASXMH9ail915/fmZ2bxuO163srAWEoyevP//n7Xrwwb7fPY7Olmfb0xjGwWUyYfnnXGh8fERFRfRX6s4A6IirC7LXvHnlnlqLSyorI+BJh0RZGuem9tX49n7ym1J+TXfVM6aZjOYENUGXB7rOa284qBOOPfrsNHZ/6rcpjIP+cK9Be7MjMLUabqfOweI/2PXx3xWG8JFXdvT6jFa7u20KpstovrRE2PDkKm54aBcB9AaSqWRXlFU7MXrAPRWV1b6a1sKwCMTYLcutwn1b1xbOz+b4vhuWX2mE1C4izWbBv5nhc62fBIiIiIvLEoNRPkRZzWBWvOZhViCPni0I9jErJqXsBtP7zIEIb2K1WFVHxRa6g6U+bCfURV7+zBoUBBAVfrDuO03klmiAnQurF+OEfR7w9zNCIV5dhzpaTADzTSw+dK0RZhQPnCsqqFeSTsfVHXBcjFkoXFrIKSrH/bAGWq9aHTpvQBYIgoG1KDACgV8tEpMTZkCxVkJXT1V9fdACr/PiM6v2y7RTeWHIQry7YV62vJRQKyyoQE2FBgepnJyXO5uMR4UcdlEb46KcKABeK7YiPtLLPHhERURD43727gYuMMIdVRc1Rs5cDAGZc3hW3h3GfTTl2qup5W3ZhGZ78YafH/u0nLmBoR98tetxBqT8zpdoBZheWGRazkRWU2vHivD2YPDgd037Yoblv8qA0HMspxrJ95zBz7h785ZK2lb4+4ApC1Rcavt10AheK7bh3eDsczCpU3nMAmNizGd6S1rJRYERRxLebTsBiEjRVVu26as1jX1uB3GLtz3x8pOszMaxjCmZf10tTJEdv2b4s9E1LRLSU6ptfasfe0wU+W4VYpUDo+00nMP1yz6JA4UoURXy36QS6NovHGVWBr6v7tgzhqAKn/jzM3X4a5wvW4K2b+hoG17tP5aFN45jaHB4REVG9xZlSP0VZTSitpPBFKMz4xfe6p1CTZzlNArB0XxaKywNLS/x1+2nD/Uv3VV7dVA4yqtJsXk7VPpNXiqd+3IGpc3bg3RWHlPuX7TuHL9dn4tp31ng8dkLP5vjX9X0Cfs23lx7S3J46Zwf++ftefLX+uCYgBbx/X6hy64/k4PHvtmsCkAGqQFFeV6gPSAH3xQtBEHBV35ZKwGkkM7cYXZ+Zj3WHswEAD3y5Bdf9d43Pnr1xkXIAW7fSd6946w8AwG7dOurE6JrpO1pTHE5tdsK6Izn41+L9hsdeKLGjRaMow/uIiIgoMAxK/RRpNaOU1XcDJmegHj5fhDs+2oCpc3b4foBObrF7fVqExaT0Mdx0LBc7T+b5fKxcfbcqLWHk9agPfLkFn689ji/XH8eL8/YiS5oF+vuXWwBAk6oIAA9c2h792yRpejHa/cxdlgPhLU+P1uyfYvA969kywa/nJE8XDDIeGkVHKNvHc4rR8cmqr+m9WKq6On+XKw1YTv9dJl1I+Wr9ca+PVa9BPpNXN1oKfbMxEzu8/Cw2jY+s5dFUj3whS22udAFo35kCfLL6KERRxLrD2TiWXYwoq9njeCIiIgocg1I/RVnNlbYIqC36qq7hlFaslldiR1aB68R6zmbXOkk5PdXf6qTqWc5ljw7Hd/cOVm7riwrpKTOlpsAKHQHutWVHs7Xrdvu/uBi7TnkPhv9h0Jtw3o7KZzXnbD6BX7efQqNoKxrFRBimJvdpnYhYmwUXt2+M7SfykGcwkyfLzClW+iiSlv6zN6l3cxw6V6jcXrrvHMqrsQj6U6kCr6xAN+v54ry9XqtRv7/SvQZ54EuLkZcB7MsAACAASURBVOmjd2o4KK9w4vHvtiu3P7w9Q9l+bGwnXN6reSiGVWVG6di5xa7fY2NfX4HpP+/C/rOFuP7dtQDgc6aciIiI/FfloFQQhFaCICwVBGGPIAi7BEH4h7Q/SRCEhYIgHJD+bxS84YaO3SEiq6AMpy6UhHgcTjz9o3aN5V2fbgzRaHy7eNYSJRiVZeYUo83Uuej41G/4ZPVRj8e8v/Iw7v18k3J75QHX7NKWp0crfRDlCqfy+j5vlDWllsAXtOaV2PH20oOG9329IVNze2TnVGx9ZjTWTxup6ck4vJMrsLzgI3h0OEVk5Zfi4W+24URuiVIpeMblXfHcn7RrCt+/LQM7nx2rpEBP+9H7rPMlLy9FxsxFyCooRVmFw6NoUk3KzCnGUeniQ6ndEXbtUdQz12O6NsHs63rjQFahj0cAz0/qrlTYrYxJ15dz/9kCj2P2nXHtK6twYNZv7hn4NVKqr+ySl5f6TPcNNX3mw9AO7osp949oX2mP0nDTLiUWR2dNwIrHRij7WjaKwolc9+/9kxfcFwqyi3jhh4iIKBiqM1NaAeARURS7ABgI4H5BELoCmAJgsSiKHQAslm7XeQt2u/oUvrnEOFCpLY9+uw0bj+Vq9p0OwzS/UrvDI7UVcM06yMWPpv+8y6OK7My5e/DbzjO4UFyOx7/bhrWHXRVRG8W40yv3PD9OOTZj5kKvVZHl1F+LHzOl+n6Kt324Hq/M34esgjJEWc1okeheO/bpmmMAgCt6NcfRWRPwwe0XITE6Aqm6VMX/3NIPgO+Z7HbT5qH/i4uV23IF37YpsbhtUDoOvjAev/79YhydNUGp8Pr+5IsAuNIK/09XpXXR7rP4z3L32tT+LyxGp6d+R/snf0P6lLmVfh+C4faP1mP4q8uwdG8WOj/9O6789x/Ver6NR3OQPmUubvtwfZWfo7zCiWd+2onj2cVQT1LG2CwwmQQ8NraTz8ffOjBN+f4Hat2RHKw+pK3EW1RegT8Onsd3m07gP8sP4fXFBwC4Zm31wrXFyq5Tefh+8wnNPovZhGev6Iar+rQI0aiCo3VytLLdIjFKM7O97oi7ZdTSvVm1Oi4iIqL6qspBqSiKp0VR3CxtFwDYA6AFgD8B+EQ67BMAk6o7yHBgkvI7/WkvUhNK7Q7kl9rx09ZTyr4JPZsBAHq0cK8vvO3D9UifMhfL9oX2ZOna/3gWADIiB2yiKOKHLe4T3N7PLcQ3G12326fGah6jnn05X1huOBMFAC//7grYfFXRlcmp2U3iPQOPoR0bY9UTI3DwhfGa/bOv6+XzOSOl9WbvrThc5RYuFrMJ3Vto148mqQL0N5ccxKZjucgrtmPvmXz85dONmCX11DTSY8Z8peVJTTl0zjVLesfHGwAAu07l+zq8UtdIn6UV+895nb2uzO+7zuDTNcfw5pIDmnWbTul9uX9Ee3x110DNYxKiql6kR15XKrvpvXUA3J/lJ3/YgZvfX6dUlv5i3XEUllUgu6gcXZrFI1VV7dXXTHsoPSCtqwZc2QvLHxsOAJg8OB2zr+8dolEF37ojOZrfZ/9dfljZTktm9V0iIqJgCMqaUkEQ0gH0AbAOQBNRFE8DrsAVQKqXx9wlCMJGQRA2njtXeSXVUJPjoFC1h+z89O/oOWOBcnvK+M54+6a+GNI+GXN3nMa411cgfcpcrJCKqtz+0YbQDFTirfCJ3oajrlnfNlPn4aGvtxke8/EdF/l8joLSCpy8UII1h7IN74+KqLwYiZwa/PdLO3jc17VZAgRBUFJrZfrbXsdXVoE/DrrG9sLc3UifMheiKKJIN5N8Ra/meOmqHn4956vXugPiq99ZjV7PLcD+s75TUAHX9+qvYZrubUS/LvaV+fs81tLml9qRPmWuUpBGL6/Yjp+lizlREWZNBWj1RZ6BbZM1j3v2im64dWAaXrzSv/dE7bXre6Njk1hNVV/AfUHB6L3qPn0+1hzKRvOESKx/chS+v3cQAGDbiQsBv35tkC8+AMBvD15S7wK0Szo09nl/QpQVn+jWDxMREVHVVDsoFQQhFsD3AB4URdHvKRFRFN8VRTFDFMWMlBTf/SbDgVwZ9bO1x2r9tfVtVF6+pifuGdYOgDs1de8Z7WxhpNWEE7m1WySl0CBdFwDGdfPeyzG7sKzS9Y7NEzzbLqjXW14otmPCGytx43trlRnJQNdQpsTZcOSly3DLwDSP+1JVs6eXdnZdY5FTc/316oJ9OJZdhPekQjZvLTmIg6p1jG/e2Adv3NgHN/Zv7dfzXdOvJbbPGKPZ9+3GTMNjbx+c7rGvoJbXKTq9FPapjFG15kPntQHdESk4Uqctq93ywTos2uOaHU6OsaFIVbBM30fzloGu7/+c+wZjUp8WeH5Sd9w0wL/3RC0lzoYFDw3DbYPSNft3VzJrXOF0rV0HgESpIvAzP+0K+PVr2uk87dr6dimxXo6su96p5Gf8179frMlaICIioqqrVlAqCIIVroD0f6IozpF2nxUEoZl0fzMA9WLRzed/HgAAuCi9Zuo29XluAV6d714fKBeIyS+1o+sz85X9kwel4bqMVsptoz6A47o1RandiYv/ubTW0njn7TiN7tPnKyfdcprioRcvw+s39EbzhEjNWBc8NBSAayasva79xvTLu2qCBX3hGADo2TJR2c4tLldSHOX+jkY9Jisj6EvwSpomuNeKfjA5A/tmjjOs0unL1swLGPbKMuX2/9Ydx5/edq21/OTO/lWqUhofqX3vVx5wr1uMsprx1V0D8fr1vTHjim44OmuC5tg1h7L9blXjL6dTRJmXtkk7dRWLS8odyvpZvQqHE+UVToiiaJhqfNW/V2vaAckpuEVlFR5rNwHtrP1ri/YrRYUA4Ilx2rWkMyf1wNFZE9C3dXB+ztXp/sM7peDLvw70OKaR7md4snQRIc6PtPPalFdiVy4uZBe617ne2L+Vt4fUabE2i8/3oFVStNf7iIiIKDBVPusRXGfwHwDYI4ribNVdPwOYDGCW9P9P1RphmMhIT0KszYIeLRIrPzhAmTnFyC22462lBzGmWxP0bJmIzk//7nHc9/cOQrfm2vWFD43qiE3HcnEitwTv35aBPq0T8e5K95qn2z/a4BGQ1AR5lu5YdhGyCkpxMKsQ47s3hdkkwGwyY/XUkQCArIJSmAUBidEREATg1QXuxvRjuzXBs1d0V4LAF67sjjK7ceDUu1Ui1j85Ev1fWIyNR92FR87mlyIhyqrse+WantX6ur65exD6q1IwBUGAzeJ/b8JFDw/DqNnLPfafUQVGg9sle9xfHcsfG46EKKsy0yZ7flJ37DuTj8/XHsddn23CQ6M64h+jOiC/1I5HvtmGZyZ2xcoD57H2cDZeuLI74iIDW1P5wFdb8KsuhdZiElDhFD3WRU7+cD3WH83RfDb3nSnA2NdXKLd7tXL/rH1/72DER1ow+jXX/RPfXKU8Vp6EPXy+CDe9tw47nx2rrCPedMz92ZB9IhWqen5Sd4/iVME2sF0y2jaOwayre6J/myTDGepFDw+D3SHCKYrIKSpX1hCnqNaVPvT1VrwWwnWaecV29HpuAR64tD0eHtMJx6VWNV/8dQAGt/Od5lqXGRVrA4ARncI/u4eIiKguqc6l+CEAbgWwQxCErdK+aXAFo98IgvBnAMcBXFu9IYaPCIsJ5Y7g9ypVF4K54q0/sPTR4R7H3D44Hf3Skjz2pzeOwaonLsX5wjI0lqqDDmyTrCnGsfLAOVzSwfgk6u9fbsEv207h5Wt6amZgA3XqgivI+mX7Kczb4apUvOGoZ0CQGucOAvTrc1+/vo9m/Wek1awUCzKSGheJKKsZP6rWBZ7OK0XHJnG493+bAQC2ajS33z5jjMdsZKDapfheZ/fSVT00vVgDdVXfFpiz+SSmX94Vz/6yG4D34iu3DkxDcXkFPl97HIBrdvv3XWfQLiUGC3efxZm8UmVWsXGsDc9c3jWgsegD0p/uH4JIqxljX1+h9OrcfDwXDqeI9dJnIzOnWJlx+tPbqzSP35bpWkv5/m0Z6JfWCAWldtgsJpRVOBGt+pxcKNZWpy0uq0CszYJX5+/DWz4KI91qkKodbPGRVixR/TzHRVrRpVm80mv4578N0VT1ba6q8iwIAqZd1hkvztuLH7acRN+0RrUyZiObjrverzeWHITNasYrUlZHdX8+6qoHDfoRExERUdVVp/ruKlEUBVEUe4qi2Fv6N08UxWxRFEeKothB+t8zMqmjIsymGum5+Kqurcef3tKenLdIjMLTE30HCI1VJ7YjOqciTtXDc9oP3vtZ/rLNFdA9/t12v8drRO6lKAekANCmsf+FT/51Q2+/ChLplejawSzeo033VFcxDVQw0ie9pQS77oPfa0i9mX1dbxx56TI09XPGLzrC/TXtO1uAPafzlWBSneb64R9HAhpHZo52/fIvf7sYvVolIsbmek/v/8J1keCqf6/WVDK95OWlysWLUi+z4iO7uNbxxkVasee5cZg8KA3F5Q5cKC5HUVmFJm0ZcH0myiocmoB093NjNcdUVsSmJv32j0sw577BOPTiZZo0dCN/ubitsv30jzvhdLpmU2vb/6QLGQCUgBTQtmqq73q2dGepJMc2nK+biIioNgSl+m5DkVdix5zNJ4P+vHLBm63PjAbgXhe5bfoYDO2YgucndQu4Cf0fUy7FF391rYPNzClB+pS5yC0qxxfrjntUfZW9v/Iw1h42rmDrTYXDiZ+3nUJhqedz/vtm/4oB3TW0Lf7UOzh9DT9dcwzZhWVKSqy+omogfAWUgVg95VJEWl0/akdnTcDU8Z3x3m0ZOPJScNKqBUHAuO5N8dr1vXBA17bGiL/p3N4+J3pL9p7FJS8v1eyTT9rVKaje+qTOnLsHU77XXhR548Y+yrb6fTBJ6cAAkDFzEbpNn4+PVx/VPPaleXsxQNX79e6hbREdYdG08LFZQvurr2/rRn79TJtMAt691f1zNHvhfvR9fmGtBqaiKGKxl36c6v699d0Dl3bAfcPbIdJqalBfNxERUW0Ir0oaYU6elbtQXO6xXq+qzuS50l4fGtVR85xN4m1IiLLi0yq2HIiPtGJwu8ZIjbMp1Twf+mYrlu07h6PZRZh2WRePCrUz5+4BAGx6apQmpdCXT9Ycw/O/7vbY70/gs/7Jkfh+00nD6rDV0W/mIgxpn4x+aTVTlCpQzROjsPyxETgrrSO9W6qcHEyCIODKPi0rP1By55A2lc6Gdps+36/38f9U64IBV2sbeebWn/W32zIvKKm6ALBv5jjYLGbkldgx3qCglPycFbqKvk3ibTibX4bfd53R7Ht8XGcAwFV9W+Lhb1xth7o0i690XOFiTLemuHVgGj5be0yZ/c0vsdda5df8EuOLEw0pMJN/DkZ1baJ8noiIiCh4OFNaBeqWErKpc7bjkW+M+2z6suW4q0/nMKlwxp96u6qwVqV6rJGEKPear2X7XD1MC0orUOFw4sb31ho+JpDZ4EPnKu+N6U1qXCTuHd6uSmm7Mrka8pV9tDOtfxzMVmYnAzV1fGfcNzy4gWOT+MhKUzVr09TLtCfWdwxJV7ZfVxXU+W2Hce9PNfW6353PjsUbN/YxrJisZxTUvHVTHyXovHVgmiYt3dvYAdd7pm/h8c7NfbFu2ijNjKRcPKmurQm8eaA2zTuvRK42bfdYUxtsm6XfUZN6uytEX9W3BRY+PLRGXzccfHpnf/z8tyGhHgYREVG9x6C0CozSGr9cn4nvN59Q+mT6S07d7djE1UJFrvQarLWrRgHfl+uP4+uNmdhw1HWyGRNhxsXtGystW16Yt8fv51e35pADww8mZ1RnyAH588VtALjbaKidvlDqsc8fdw9rV+9nQ6xmE366fwh++dvF2Pv8OCRGuWfd1O1u5IJRvuw44f4MxBqsw53Qs5nmduukaNw8oDX+mHIp1LHr3ufHYWLPylvjWM0mrHx8BADXZ3fb9DG4e1g79G3dSKmK+uiYjhjfo5nHY3+6fwiOzpoQcDp8qOn7gH4ipSyPf30lej+3sMZet6DUjjs+3gAAGKvqNxwdYdasT66vhnZMCauLSURERPVV/T+rqAGrD55HxyZxhveV2p0egeB/lx9CXKQVNw3wLGrzfwtdqY/yCV50NWYNjWw/kWe4/8kfdirba6eNVNp/yOv+HE4Rz/6yCyM6p2JEp9RKn99sEvDa9b1rvW3FuO7NvFbJlXulkjF1y5Ub+7fCx6uP4KM7+iPSasbtg9OVtZp5JXbNjLteuZQGPqS98frdV67piVsGpOGPg+fRNCESN1zUCiZpnej6J0chY+YiNI61+ay0rNcqKdowtfjtm/ti3ZEcn5/ZukhfoXnOlpN49dpeOHmhBIDr5/XI+SK/PvMOp+h3UC5nV+jHUFQW/CrkRERE1HBxprQKZvziuYZSZtSH8KXf9hpWwDU6Vp4pevnq6vXX1Lt7WFtsfWY0lunazbx8dU9NP8pG0mzp7R+tx6drjuGOjzb49fzzH7wkaGMNlLe2FDOv7F7LI6m7UuMjseWZMegtBarTL++KSzu7Ajt1UZ15O07j87XHlNtlFe7g5OM7jNc/R0dYMKhdMh4d2wm3DEyDxWxS0nsbx9qw69mxWPXEiKB8HdERlnoXkMpevbaX5vadn7h/NjNmLsSo2cux+tB5/cM03lpyAO2mzcO+MwV+vaZ6re/ILqnKDLU+XZ6IiIioOhiUBsF3m04o2/mqQFMURU3F0YJSO7ZKJ3kzf92ND1Z5Fpqxmk04OmsCrruo6j1D1SKk2Y2r+rREYnQE0pKjNffL7TZkcnEgdZuN9ClzlRkZtY2qPqTtU41njmvTlX1aKOnPLRtFaXqiUmAEQcBtg1w9MdVB6X3/24ynfnTPsv/1000AgKv7tqxyv9UYmyWgWdKG6pp+2kJW6llMeQ366oO+q2e/KhWl+m5Tpl+vKb/3ZpMAQRCUGeqhHY37HhMRERFVBYPSAMgB3cjO2kBuqapdQr6qNcqe09rZiB4zFmDS23/gq/XH8f6qI3h90QEArsq7NSVCan0h/69vc6KvsuutjcuQWUvw6LfbcL6wDO+vPAxRFHGNqt9kOHjt+t54cGQHAECf1uFRebcuk4sMzZy7G3tO5+OOj9Yr91U4nFh14DxW7HcFRkap6RR8A9sm4eL23nusrjuSjaV7s2B3eK5JV18we2+lf31o52xxFT3bOWNsJUcSERERVR2D0gAseMhVbVLfs69lI3cV0QJVUHrXZxsNn2fKHG0qb+/WNVdIQ07HVC8he3i0KwheN22kx/ERPvo3frfpBDJmLsLMuXuw8Viusl+fEhxKA9sm47GxnTDj8q6hHkqd16mpa/Z7y/ELePKHHViqmpk7ml2MWz5YB8BV3Chc2u/Ud1/dNQif/2WA1/s3HM3FHR9vwENfb4Xd4dQUIpPbT/kru7BM2a5OhWwiIiKiyjAoDYC656J61uGLdceVbfU60RO5nimvRmqyIM/bN/fFy9f0RFpyjLLvgZEdcHTWBDSJN05vvaKXuwLqK9f0xGQpjVPtWHaxsp3eOMbj/lAxmQTcP6K9331WyTt1Ou7m4xc0942avVzZTo3j97q2/fr3i5XtR0Z7Zlr8uv00Ojz5Gya+uQoHswrgdIoY89oKzTFGs6lqL0h9i5+eyAs8REREVLMYlFbRF+uO45dtp/D3L7egQNUi5rFvt2PS238AcPXyA4ADL4w3fI6jsybg0IuX1WgT+oQoK67LCGx96r9ucFfQvTajFVonewadlRVUofrhxv6Vf3bifFTmpZrRvUUC5Ez8I+eLfB47avYKnM73nCXt8ORvWL7/nMEjXOTU3f7pSVUfKBEREZEfGJRW0azf9uLvX27BL9tOafaX2B3YmnkBdocTZkFAs4RIzYzT3y9tDwBKi41w7JcoCALuHd4Ob97YBwAwvJNnUZM5m0/W9rAoBJ69QlvBeNHDwzyO6cf1uyHRVMp0aONHpsJ/lh1Stu8e2lbZnvzhejidvnsrd28RX8UREhEREfmHQWmQtErSznaeKyhDaYUTUbqqon+VTggTo8N7dumJcZ1xuZTG2y4lFt/cPUhpF6P27BXdantoVIsiLCbcO7wd4iMtWD3lUrRPjVU+0y9f3RPf3jMIUy/rHOJRNkypUlA62Et/WLXPpDY+792WgQk9m2nuO6daO6p2SYfG6NUq0aM4GhEREVGwMSgN0ItX9jDcn5mjXT+6+1Q+SsodsEkn8PMfHIq1U0ciPtKKl67qgc//7L1YSTjq3yYJg9t5Vv2UW7BQ/fXEuM7YPmMsmktp5ttnjMHBF8bjuota4aL0pCq3gqHqaZ3kqgYeYTYuQtTXoIBaqd2BtCTtzKq8Dv7dFYeQPmUuHNLM6f6zBWgbRuvFiYiIqP7i2WSAvK2xm3PfYM3tv3y6EYv2nEWU1fUt7tQ0Dk0TIqXnaI1WSdEezxHu4iItmtsjOqWgQw0WaaLwZDWbYGEgGnIvXNkd/7y6h9f02gFtk/Hpnf01+wa1S0ZCtBXPXtEN12W4+p7KPU5fnLcXAHAwqxCldgfO5pehXQqDUiIiIqp5lsoPITVvqWx9WzeCIACibnlWpLX+tFKQg2oA+O+t/TC2W9MQjoaoYYuPtOL6i1z9Yb+/dxASoyPw2ZpjOJFbjEV7smA1CRjaMQUtEqNw8kIJDrwwXpnVnjw4HTtO5OGbjSew+1S+pv3L2NdX4AGp329jVrEmIiKiWsDpjiBSt1KR6deU1mUTejRD84RILHp4KANSojDSLy0J7VJiMeOKbkoP4a82ZAIAfvrbEHx910CPNOuk2AgAwPSfd+Gezzdr7ntj8QEA7oJsRERERDWJM6VB9OKVPdAvrRHeWHwA5wvLAQCR9ajpfIcmcVg9dWSoh0FEPhSUulpUZRW4Zj8bx9oMZzybJxj3KVbLYDsYIiIiqgWcKa2CmZO6Y6KugiUAxNgsuG1QOuJVswtzt5+uzaERUQMnt+gZ2NZ3QCkIAsZ3953xkBLH9F0iIiKqeQxKq+CWgWm4eUCaclsuGCJrHMMTOSIKjfcmZ+DuoW39qvB9+FyR5va26WPwyOiOAIBxTNEnIiKiWsL03SqKj3J965rE2zDrqp6a+xrHRSjbg9tV3kOQiChYEqKsmHpZF7+O3Xe2wOOxfx/ZAW1SYjC8U2pNDI+IiIjIA2dKqyg+0pWi63ACJpO2Iu/TE7sq2/qAlYgoXDw6pqPh9sSezRFr4zVLIiIiqh0866iiGOmErcLp9LivWUIUjs6aAIdThNlk3EKGiCjU/nZpB9w/oj3KHU7YLPWnKBsRERHVLQxKq6hRtBW3DGyNa/u18noMA1IiCneCIDAgJSIiopBiUFpFgiBg5qQeoR4GERERERFRncY1pURERERERBQyDEqJiIiIiIgoZBiUEhERERERUcgwKCUiIiIiIqKQYVBKREREREREIcOglIiIiIiIiEKGQSkRERERERGFDINSIiIiIiIiChkGpURERERERBQyDEqJiIiIiIgoZBiUEhERERERUcgwKCUiIiIiIqKQYVBKREREREREIcOglIiIiIiIiEKGQSkRERERERGFDINSIiIiIiIiChkGpURERERERBQyDEqJiIiIiIgoZARRFEM9BgiCcA7AsVCPo55rDOB8qAfRwPE9CC1+/0OP70Fo8fsfenwPQovf/9DjexBaof7+p4mimGJ0R1gEpVTzBEHYKIpiRqjH0ZDxPQgtfv9Dj+9BaPH7H3p8D0KL3//Q43sQWuH8/Wf6LhEREREREYUMg1IiIiIiIiIKGQalDce7oR4A8T0IMX7/Q4/vQWjx+x96fA9Ci9//0ON7EFph+/3nmlIiIiIiIiIKGc6UEhERERERUcgwKK2jBEH4UBCELEEQdqr29RIEYY0gCDsEQfhFEIR4ab9VEIRPpP17BEGYqnrMjdL+7YIg/C4IQuNQfD11UYDvQYQgCB9J+7cJgjBc2h8tCMJcQRD2CoKwSxCEWSH6cuqkYLwHqvveFQRhv/ReXB2CL6fOEQShlSAIS6XfK7sEQfiHtD9JEISFgiAckP5vpHrMVEEQDgqCsE8QhLGq/f2k9+agIAhvCIIghOJrqkuC+f1X3f+z+ueJfAvyzwD/Hgco0O+/IAjJ0vGFgiC8pXoe/i2uomC9B9J9/FscoCp8/0cLgrBJ+l2zSRCES1XPFdq/w6Io8l8d/AdgKIC+AHaq9m0AMEzavhPA89L2TQC+krajARwFkA7AAiALQGPpvpcBzAj111ZX/gX4HtwP4CNpOxXAJrguCkUDGCHtjwCwEsD4UH9tdeVfMN4D6fazAGZK2yb5Z4L/Kv3+NwPQV9qOA7AfQFfpd8kUaf8UAP+UtrsC2AbABqANgEMAzNJ96wEMAiAA+I0/B7X7/ZfuvwrAF+qfJ/6rnfeAf49r7fsfA+BiAPcAeEv1PPxbHOL3QLqPf4tr/vvfB0Bzabs7gJOq5wrp32HOlNZRoiiuAJCj290JwAppeyEA+QqTCCBGEAQLgCgA5QDy4frQCdJ9AoB4AKdqeOj1RoDvQVcAi6XHZQG4ACBDFMX/b+9+QqwqwziOf5+YKFAqkkbUKceoFtPGEsR2UtiiqAkiiBYKGlS0UIsWUjuFCiwkgiIwMIggs6gIWiS1aJSiRpvU8U+l2MSQUItsJHXqafE+F47mnbnn3DPnzL39PnDw+J4/vvd5Zu5z33Peezzj7p9H+zlgGOib4a53jTJyENvWAs/Htn/cXf+xdwvcfdzdh2P9NDAKLAIGgR2x2w7ggVgfJF0gO+vux4EfgOVmtgC4yt33eqqMb2WOkSbKij+Amc0FngK2VPcKOl+JOVA9LiBv/N19wt2/BP666DyqxQWVlYOgWpxTgfjvc/fGe8tB4Eozu2I21GENSrvLAeD+WH8IuD7W3wMmgHHgJLDV3X939/PAE8D3pOI3AGyvtMfdp1kOvgMGzazHX4hREgAAA5dJREFUzJYAyzLbADCza4D7iIGTFJYrBxF3gM1mNmxmO81sfrVd7nxm1k+6AvsVMN/dxyEVTNKdaUiF8ufMYWPRtijWL26XFrUZf4DNwEvAmQq625XayYHqcftajH8r51EtLqidHKgWt69A/B8E9rn7WWZBHdagtLusBZ40s29Jt/DPRfty4G9gIWm60NNmdqOZXU4qgrfFthFg03/OKnk0y8GbpF/wb4BtwB5gsnFQ3MV+B3jF3X+qtMfdJ28OekhXxIfc/XZgL7C16k53srjLtgvY4O5/TLXrJdp8inZpQbvxN7OlwE3u/sGMdPB/oIQcqB63IUf8pzuPanFBJeRAtbgNeeNvZrcCLwKPNZousVuldbinyn9MZpa7HwbuBjCzW4B7Y9MjwKdxJfaUmQ2Rpi3Oi+N+jGPeJc07l4Ka5cDdJ4GNjf3MbA9wLHPoG8Axd99WXW+7U4Ec/Ea6O9T4QL4TWFdhlztafJjeBbzt7u9H869mtsDdx2NK0KloH+PCGQJ9pLtCY1w4Va7RLtMoKf53AMvM7ATpc0GvmX3h7iureA2drqQcLAXV4yJyxn86qsUFlJQD1eKC8sbfzPpIcV7deM9hFtRh3SntImbWG39eBjwHvB6bTgJ3WjIHWAEcBn4BBszsuthvFWkuuhTULAeWnuw3J9ZXAZPufij+vgW4GthQS6e7TN4cxHcnPgZWxinuAg5V3e9OFN992w6MuvvLmU0fAWtifQ3wYab94fj+yhLgZuDrmFp02sxWxDlXZ46RJkqM/2vuvtDd+0kPIDmqAWlrysoBqseFFIj/VOdSLS6grByoFheTN/4xTfoTYJO7DzV2nhV1eKqnIGmZvQtpesk4cJ50dWMdsJ701K2jwAuAxb5zSVecDpJ+wZ/JnOdxUuEbIb0ZzKv7tXXKkjMH/cCRiPVnwOJo7yNNjxgF9sfyaN2vrVOWMnIQ2xaTHo40Qvoe0Q11v7ZOWEgDGI+4NX5+7yHNwthNuhO9G7g2c8yzpCeOHiHzZD/S7I0Dse3VRt60VBP/zPZ+9PTdWnKgelxZ/E+QHpD3Z9SNAdXi+nMQ7arFMxx/0sX6icy++4He2FZrHW58WBMRERERERGpnKbvioiIiIiISG00KBUREREREZHaaFAqIiIiIiIitdGgVERERERERGqjQamIiIiIiIjURoNSERERERERqY0GpSIiIiIiIlIbDUpFRERERESkNv8C2nKrAReVwGoAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 1152x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(16,8))\n",
    "plt.plot(df['Price'], label='Price History')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 242,
   "metadata": {},
   "outputs": [],
   "source": [
    "### Create the independent data set (X)  #######\n",
    "# Convert the dataframe to a numpy array\n",
    "X = np.array(df.drop(['Prediction'],1).drop(['Date'],1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 243,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[25.56]\n",
      " [26.  ]\n",
      " [26.53]\n",
      " ...\n",
      " [51.85]\n",
      " [51.59]\n",
      " [52.14]]\n"
     ]
    }
   ],
   "source": [
    "#Remove the last '30' rows\n",
    "X = X[:-forecast_out]\n",
    "print(X)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 244,
   "metadata": {},
   "outputs": [],
   "source": [
    "### Create the dependent data set (y)  #####\n",
    "# Convert the dataframe to a numpy array \n",
    "y = np.array(df['Prediction'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 245,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[18.89 18.69 18.28 ... 54.51 54.83 56.24]\n"
     ]
    }
   ],
   "source": [
    "# Get all of the y values except the last '30' rows\n",
    "y = y[:-forecast_out]\n",
    "print(y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 246,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Split the data into 80% training and 20% testing\n",
    "x_train, x_test, y_train, y_test = train_test_split(X, y, test_size=0.2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 247,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "SVR(C=1000.0, cache_size=200, coef0=0.0, degree=3, epsilon=0.1, gamma=0.1,\n",
       "    kernel='rbf', max_iter=-1, shrinking=True, tol=0.001, verbose=False)"
      ]
     },
     "execution_count": 247,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Create and train the Support Vector Machine (Regressor) \n",
    "svr_rbf = SVR(kernel='rbf', C=1e3, gamma=0.1) \n",
    "svr_rbf.fit(x_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 248,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "svr/svm confidence:  0.7501112261448108\n"
     ]
    }
   ],
   "source": [
    "# Testing Model: Score returns the coefficient of determination R^2 of the prediction. \n",
    "# The best possible score is 1.0\n",
    "svm_confidence = svr_rbf.score(x_test, y_test)\n",
    "print(\"svr/svm confidence: \", svm_confidence)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 249,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "LinearRegression(copy_X=True, fit_intercept=True, n_jobs=None, normalize=False)"
      ]
     },
     "execution_count": 249,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Create and train the Linear Regression  Model\n",
    "lr = LinearRegression()\n",
    "# Train the model\n",
    "lr.fit(x_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 250,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "lr confidence:  0.5866480581754341\n"
     ]
    }
   ],
   "source": [
    "# Testing Model: Score returns the coefficient of determination R^2 of the prediction. \n",
    "# The best possible score is 1.0\n",
    "lr_confidence = lr.score(x_test, y_test)\n",
    "print(\"lr confidence: \", lr_confidence)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 254,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[51.62]\n",
      " [51.67]\n",
      " [50.59]\n",
      " [50.44]\n",
      " [50.  ]\n",
      " [50.79]\n",
      " [49.34]\n",
      " [49.58]\n",
      " [50.93]\n",
      " [51.3 ]\n",
      " [50.61]\n",
      " [51.43]\n",
      " [51.86]\n",
      " [51.87]\n",
      " [52.05]\n",
      " [51.29]\n",
      " [51.63]\n",
      " [51.91]\n",
      " [52.32]\n",
      " [51.97]\n",
      " [52.41]\n",
      " [53.92]\n",
      " [54.11]\n",
      " [54.36]\n",
      " [54.32]\n",
      " [54.55]\n",
      " [55.63]\n",
      " [57.34]\n",
      " [57.19]\n",
      " [56.82]\n",
      " [57.16]\n",
      " [56.75]\n",
      " [56.77]\n",
      " [55.67]\n",
      " [55.28]\n",
      " [55.14]\n",
      " [56.21]\n",
      " [56.21]\n",
      " [56.84]\n",
      " [57.88]\n",
      " [58.94]\n",
      " [58.1 ]\n",
      " [57.96]\n",
      " [57.25]\n",
      " [57.4 ]\n",
      " [58.35]\n",
      " [57.48]\n",
      " [57.66]\n",
      " [55.79]\n",
      " [56.5 ]\n",
      " [57.15]\n",
      " [57.84]\n",
      " [57.12]\n",
      " [56.59]\n",
      " [57.  ]\n",
      " [57.29]\n",
      " [57.17]\n",
      " [57.49]\n",
      " [58.09]\n",
      " [58.34]\n",
      " [58.25]\n",
      " [59.55]\n",
      " [59.67]\n",
      " [59.84]\n",
      " [60.46]\n",
      " [60.37]\n",
      " [61.61]\n",
      " [61.98]\n",
      " [61.49]\n",
      " [61.73]\n",
      " [62.92]\n",
      " [63.6 ]\n",
      " [63.81]\n",
      " [64.22]\n",
      " [63.82]\n",
      " [63.92]\n",
      " [63.96]\n",
      " [63.38]\n",
      " [63.66]\n",
      " [64.45]\n",
      " [65.69]\n",
      " [65.62]\n",
      " [66.27]\n",
      " [65.71]\n",
      " [64.64]\n",
      " [64.82]\n",
      " [65.92]\n",
      " [65.5 ]\n",
      " [64.18]\n",
      " [63.48]\n",
      " [61.91]\n",
      " [61.3 ]\n",
      " [59.2 ]\n",
      " [59.41]\n",
      " [59.33]\n",
      " [60.7 ]\n",
      " [61.48]\n",
      " [61.89]\n",
      " [61.91]\n",
      " [61.73]\n",
      " [62.72]\n",
      " [63.52]\n",
      " [63.81]\n",
      " [62.94]\n",
      " [61.43]\n",
      " [60.98]\n",
      " [61.19]\n",
      " [62.49]\n",
      " [62.54]\n",
      " [61.09]\n",
      " [60.13]\n",
      " [62.02]\n",
      " [61.35]\n",
      " [60.69]\n",
      " [60.89]\n",
      " [61.16]\n",
      " [62.29]\n",
      " [62.01]\n",
      " [63.37]\n",
      " [65.1 ]\n",
      " [64.25]\n",
      " [65.8 ]\n",
      " [65.49]\n",
      " [65.21]\n",
      " [64.3 ]\n",
      " [64.87]\n",
      " [63.05]\n",
      " [63.41]\n",
      " [63.35]\n",
      " [63.53]\n",
      " [62.03]\n",
      " [63.4 ]\n",
      " [65.48]\n",
      " [66.81]\n",
      " [67.07]\n",
      " [67.35]\n",
      " [66.23]\n",
      " [66.5 ]\n",
      " [68.44]\n",
      " [68.3 ]\n",
      " [68.26]\n",
      " [67.61]\n",
      " [67.66]\n",
      " [68.  ]\n",
      " [68.18]\n",
      " [68.11]\n",
      " [68.56]\n",
      " [67.28]\n",
      " [67.91]\n",
      " [68.45]\n",
      " [69.71]\n",
      " [70.74]\n",
      " [68.83]\n",
      " [71.16]\n",
      " [71.36]\n",
      " [70.69]\n",
      " [71.01]\n",
      " [71.34]\n",
      " [71.43]\n",
      " [71.47]\n",
      " [71.23]\n",
      " [72.26]\n",
      " [72.09]\n",
      " [71.85]\n",
      " [70.77]\n",
      " [67.92]\n",
      " [66.8 ]\n",
      " [68.24]\n",
      " [66.98]\n",
      " [65.81]\n",
      " [64.76]\n",
      " [65.51]\n",
      " [64.75]\n",
      " [65.96]\n",
      " [65.77]\n",
      " [66.1 ]\n",
      " [66.38]\n",
      " [66.63]\n",
      " [66.91]\n",
      " [65.01]\n",
      " [65.91]\n",
      " [65.09]\n",
      " [65.92]\n",
      " [65.68]\n",
      " [69.02]\n",
      " [69.91]\n",
      " [75.23]\n",
      " [77.41]\n",
      " [73.45]\n",
      " [74.13]\n",
      " [73.89]\n",
      " [74.19]\n",
      " [73.05]\n",
      " [73.78]\n",
      " [73.93]\n",
      " [74.11]\n",
      " [70.47]\n",
      " [70.28]\n",
      " [71.03]\n",
      " [68.22]\n",
      " [68.03]\n",
      " [68.78]\n",
      " [69.42]\n",
      " [70.31]\n",
      " [67.9 ]\n",
      " [70.77]\n",
      " [71.13]\n",
      " [71.59]\n",
      " [68.66]\n",
      " [71.19]\n",
      " [69.88]\n",
      " [68.8 ]\n",
      " [68.95]\n",
      " [68.49]\n",
      " [69.01]\n",
      " [69.17]\n",
      " [66.92]\n",
      " [66.81]\n",
      " [67.61]\n",
      " [67.25]\n",
      " [67.04]\n",
      " [65.07]\n",
      " [65.44]\n",
      " [65.93]\n",
      " [66.5 ]\n",
      " [67.32]\n",
      " [67.85]\n",
      " [69.13]\n",
      " [69.71]\n",
      " [69.97]\n",
      " [68.54]\n",
      " [69.68]\n",
      " [70.25]\n",
      " [69.84]\n",
      " [69.82]\n",
      " [68.69]\n",
      " [67.81]\n",
      " [67.73]\n",
      " [67.55]\n",
      " [69.29]\n",
      " [70.37]\n",
      " [68.6 ]\n",
      " [68.98]\n",
      " [68.86]\n",
      " [69.87]\n",
      " [71.08]\n",
      " [70.77]\n",
      " [70.8 ]\n",
      " [73.23]\n",
      " [73.4 ]\n",
      " [72.22]\n",
      " [72.18]\n",
      " [73.16]\n",
      " [75.37]\n",
      " [75.16]\n",
      " [76.4 ]\n",
      " [74.44]\n",
      " [74.26]\n",
      " [74.27]\n",
      " [74.95]\n",
      " [73.18]\n",
      " [70.97]\n",
      " [71.41]\n",
      " [71.84]\n",
      " [71.93]\n",
      " [69.63]\n",
      " [68.63]\n",
      " [69.16]\n",
      " [69.25]\n",
      " [66.49]\n",
      " [66.56]\n",
      " [67.25]\n",
      " [67.58]\n",
      " [67.  ]\n",
      " [66.18]\n",
      " [65.31]\n",
      " [63.67]\n",
      " [63.12]\n",
      " [63.12]\n",
      " [62.16]\n",
      " [61.69]\n",
      " [60.71]\n",
      " [60.19]\n",
      " [59.85]\n",
      " [55.63]\n",
      " [56.16]\n",
      " [56.45]\n",
      " [56.49]\n",
      " [57.16]\n",
      " [53.39]\n",
      " [54.41]\n",
      " [51.46]\n",
      " [51.31]\n",
      " [50.06]\n",
      " [51.46]\n",
      " [50.78]\n",
      " [52.98]\n",
      " [53.21]\n",
      " [52.64]\n",
      " [51.54]\n",
      " [52.76]\n",
      " [51.07]\n",
      " [51.65]\n",
      " [51.04]\n",
      " [52.69]\n",
      " [51.26]\n",
      " [49.8 ]\n",
      " [46.12]\n",
      " [47.96]\n",
      " [45.64]\n",
      " [45.38]\n",
      " [46.04]\n",
      " [44.48]\n",
      " [45.15]\n",
      " [46.31]\n",
      " [46.92]\n",
      " [47.76]\n",
      " [48.27]\n",
      " [49.58]\n",
      " [52.19]\n",
      " [52.42]\n",
      " [51.44]\n",
      " [50.31]\n",
      " [51.8 ]\n",
      " [52.08]\n",
      " [51.83]\n",
      " [53.6 ]\n",
      " [52.59]\n",
      " [52.44]\n",
      " [52.94]\n",
      " [53.53]\n",
      " [51.79]\n",
      " [53.07]\n",
      " [54.18]\n",
      " [53.84]\n",
      " [55.29]\n",
      " [54.57]\n",
      " [53.69]\n",
      " [53.94]\n",
      " [52.68]\n",
      " [52.75]\n",
      " [52.43]\n",
      " [53.14]\n",
      " [53.84]\n",
      " [54.4 ]\n",
      " [55.58]\n",
      " [56.12]\n",
      " [56.9 ]\n",
      " [56.95]\n",
      " [57.01]\n",
      " [55.32]\n",
      " [55.4 ]\n",
      " [56.92]\n",
      " [57.21]\n",
      " [55.76]\n",
      " [56.6 ]\n",
      " [56.55]\n",
      " [56.22]\n",
      " [56.6 ]\n",
      " [55.77]\n",
      " [56.79]\n",
      " [56.89]\n",
      " [58.27]\n",
      " [58.59]\n",
      " [58.51]\n",
      " [59.09]\n",
      " [59.12]\n",
      " [60.12]\n",
      " [59.98]\n",
      " [58.87]\n",
      " [58.71]\n",
      " [59.87]\n",
      " [59.39]\n",
      " [59.29]\n",
      " [60.19]\n",
      " [61.59]\n",
      " [62.53]\n",
      " [62.46]\n",
      " [62.12]\n",
      " [63.1 ]\n",
      " [64.37]\n",
      " [64.05]\n",
      " [64.62]\n",
      " [63.61]\n",
      " [63.86]\n",
      " [63.43]\n",
      " [64.01]\n",
      " [63.74]\n",
      " [64.02]\n",
      " [65.66]\n",
      " [66.24]\n",
      " [65.96]\n",
      " [65.23]\n",
      " [63.29]\n",
      " [63.39]\n",
      " [63.83]\n",
      " [63.55]\n",
      " [61.75]\n",
      " [61.98]\n",
      " [62.3 ]\n",
      " [61.41]\n",
      " [62.13]\n",
      " [61.58]\n",
      " [61.65]\n",
      " [60.97]\n",
      " [61.82]\n",
      " [62.03]\n",
      " [62.93]\n",
      " [62.77]\n",
      " [63.12]\n",
      " [63.02]\n",
      " [61.42]\n",
      " [57.65]\n",
      " [58.4 ]\n",
      " [58.91]\n",
      " [58.84]\n",
      " [56.47]\n",
      " [53.49]\n",
      " [53.25]\n",
      " [53.5 ]\n",
      " [51.57]\n",
      " [52.59]\n",
      " [53.95]\n",
      " [53.33]\n",
      " [53.3 ]\n",
      " [51.13]\n",
      " [52.38]\n",
      " [52.47]\n",
      " [51.94]\n",
      " [53.86]\n",
      " [53.74]\n",
      " [56.88]\n",
      " [57.35]\n",
      " [57.73]\n",
      " [57.63]\n",
      " [59.17]\n",
      " [59.18]\n",
      " [58.2 ]\n",
      " [58.91]\n",
      " [56.  ]\n",
      " [57.06]\n",
      " [57.35]\n",
      " [57.57]\n",
      " [60.28]\n",
      " [59.93]\n",
      " [59.99]\n",
      " [59.3 ]\n",
      " [57.44]\n",
      " [56.5 ]\n",
      " [55.08]\n",
      " [55.42]\n",
      " [55.87]\n",
      " [56.58]\n",
      " [55.9 ]\n",
      " [55.88]\n",
      " [56.04]\n",
      " [56.85]\n",
      " [58.04]\n",
      " [58.53]\n",
      " [53.64]\n",
      " [55.67]\n",
      " [54.63]\n",
      " [53.6 ]\n",
      " [51.14]\n",
      " [52.6 ]\n",
      " [54.41]\n",
      " [54.98]\n",
      " [57.05]\n",
      " [55.16]\n",
      " [54.51]\n",
      " [54.83]\n",
      " [56.24]]\n"
     ]
    }
   ],
   "source": [
    "# Set x_forecast equal to the last 30 rows of the original data set from Price column\n",
    "x_forecast = np.array(df.drop(['Prediction'],1).drop(['Date'],1))[-forecast_out:]\n",
    "print(x_forecast)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 255,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[ 57.43217455  57.11216193  62.45433123  62.77637237  63.16271221\n",
      "  61.84818068  62.83085651  63.01564711  61.30104934  59.41345362\n",
      "  62.40296714  58.63198644  55.91468868  55.85340839  54.8046114\n",
      "  59.47157143  57.36817695  55.6109356   53.5241038   55.25641254\n",
      "  53.20339208  56.54087378  57.36808416  58.17982426  58.07570211\n",
      "  58.52527081  57.22678256  67.64893622  66.02044462  62.13224423\n",
      "  65.69267566  61.47087022  61.65630803  57.18393514  57.86065006\n",
      "  58.14875389  57.85614018  57.85614018  62.32725201  72.57513988\n",
      "  73.5062656   73.80210359  73.08524295  66.67544563  68.28601044\n",
      "  74.50044728  69.11274085  70.83502724  57.11640629  59.4359612\n",
      "  65.58350583  72.29413025  65.25652005  60.10334357  63.96537895\n",
      "  67.11007248  65.80190133  69.21385783  73.75855544  74.48683369\n",
      "  74.31107422  70.57969058  70.04237068  69.38632533  68.41578882\n",
      "  68.4099638   71.21440351  72.43040571  70.81183187  71.61579248\n",
      "  74.9302301   75.69228646  75.62886422  75.04699095  75.62180081\n",
      "  75.53086921  75.4842515   75.59247466  75.6905513   74.49878428\n",
      "  72.68999351  72.57083665  75.42340698  72.73188363  73.97857421\n",
      "  73.47821401  73.3925857   72.46136737  75.128436    75.65787694\n",
      "  72.20640632  70.18611593  72.31361326  71.25885953  71.66059143\n",
      "  68.64802106  70.77839689  72.14174939  72.20640632  71.61579248\n",
      "  74.49841609  75.67483595  75.62886422  74.96937226  70.61179926\n",
      "  69.2410951   69.84003567  73.92459094  74.05548465  69.5425901\n",
      "  68.63630538  72.55668052  70.34819756  68.63263167  69.01971952\n",
      "  69.74882311  73.37092986  72.52523517  75.58418707  72.80973181\n",
      "  74.98287255  72.96494047  72.457372    72.62305152  74.87057629\n",
      "  73.3448602   75.16979425  75.6154631   75.56668891  75.67822856\n",
      "  72.58804183  75.60811501  72.45413258  80.59569821  83.6686805\n",
      "  87.11102992  75.13296069  77.36379416  97.93141229  96.94109396\n",
      "  96.63348855  90.2226756   90.79634523  94.3716981   95.98553719\n",
      "  95.38318214  98.67439661  86.25064619  93.48768617  97.99703011\n",
      " 101.77099087 100.60987019 100.00585414  99.36589596  98.66410933\n",
      " 100.73140754  99.85198301  98.73641401  98.4083378   98.2606566\n",
      "  99.12621448  95.50665745  96.02946566  96.85470796 100.53399085\n",
      "  93.58831633  80.48269284  96.47558287  82.58070391  72.99541655\n",
      "  73.6426763   72.46612604  73.67042758  73.56551273  72.87905417\n",
      "  74.29165733  76.29575959  78.64445281  81.75014359  72.99991958\n",
      "  73.35174371  72.82935577  73.3925857   72.67037142 100.68514639\n",
      " 101.77481327  88.9422605   87.40315157  93.52958615  92.51491062\n",
      "  92.97382282  92.37792655  93.94641219  93.14099779  92.90674944\n",
      "  92.55853518 101.19033079 101.48393299  99.78957805  96.31494991\n",
      "  94.65525778  99.79317803 101.54040251 101.44400529  93.38646992\n",
      " 100.53399085  99.46628689  97.81385903  99.22091467  99.26407057\n",
      " 101.78170845  99.88002186 100.45776165  98.25273458 100.65423056\n",
      " 101.08946546  81.86778611  80.59569821  90.2226756   85.8808598\n",
      "  83.30395444  72.86979368  72.4485842   73.43438232  77.36379416\n",
      "  86.74286758  92.87177725 100.99221306 101.77099087 101.75335117\n",
      "  98.55724947 101.76001764 101.5214358  101.78687004 101.78770442\n",
      "  99.37236262  92.45001847  91.58146456  89.52173536 101.33825366\n",
      " 101.35689584  98.90079219 100.5583701  100.12644855 101.7834328\n",
      "  99.63020503 100.53399085 100.45593308  93.74973588  93.57992776\n",
      "  95.62418685  95.74519979  93.8227898   88.45926511  89.19363946\n",
      "  86.60615011  91.70949665  92.20648756  92.18097787  89.96567735\n",
      "  93.80156494  99.97438391  98.48180076  96.890744    96.57048612\n",
      " 101.73541258  99.06374836 101.06584698 101.26226009  77.27041151\n",
      "  77.93988007  85.8808598   89.87381204  82.82076131  74.79061722\n",
      "  72.50714195  75.68900267  75.28322671  75.28322671  72.98765209\n",
      "  71.48242107  68.66386279  68.54586284  69.352279    57.22678256\n",
      "  57.67527966  59.10002181  59.36673542  65.69267566  53.99572243\n",
      "  58.29492103  58.44592782  59.35501962  63.15096573  58.44592782\n",
      "  61.88340578  52.73401483  53.30809849  52.67686651  57.94242585\n",
      "  52.5820934   60.65669293  57.24015296  60.80257582  52.62200967\n",
      "  59.64395493  63.13406703  60.21362838  61.70655135  60.61977352\n",
      "  61.1028957   60.23994351  64.00579647  61.6830066   60.20973033\n",
      "  60.61393287  61.51800764  61.96737345  63.01564711  54.08691729\n",
      "  53.17152933  58.57016482  62.97018708  56.28706011  54.64236638\n",
      "  56.09993744  54.96927537  52.75348576  53.1101463   52.67581458\n",
      "  54.63091813  56.34979796  52.91194314  57.63196913  56.15781459\n",
      "  57.83949234  58.54698175  55.41481873  56.63410534  52.63122936\n",
      "  52.58515575  53.14044535  53.09287825  56.15781459  58.27326043\n",
      "  57.29237328  57.54981083  62.92658167  63.44050965  64.07147531\n",
      "  57.77610913  57.61052711  63.1307038   66.23898423  57.12390167\n",
      "  60.1822741   59.79703107  57.89554384  60.1822741   57.12066371\n",
      "  61.84462356  62.82529497  74.35847373  74.49036337  74.56305798\n",
      "  72.84638619  72.70411269  68.65359002  68.96115731  73.77534873\n",
      "  74.26431214  69.28582898  71.35876695  71.8624893   68.54586284\n",
      "  71.14725508  74.02956307  73.84454295  72.86634972  75.2520036\n",
      "  74.70297517  75.35849363  74.03464471  75.69289061  75.58984866\n",
      "  75.62920363  75.41790332  75.66795082  75.40357275  72.63374817\n",
      "  75.20420234  73.56551273  72.5954432   75.50698594  75.60045143\n",
      "  75.61436734  75.68398416  71.68225223  72.43040571  73.3996856\n",
      "  70.54550119  72.89681756  71.11367478  71.34857103  69.21527734\n",
      "  71.91338857  72.58804183  74.94990044  74.61282435  75.28322671\n",
      "  75.11772263  70.57862192  70.74549288  74.55107802  73.62514089\n",
      "  73.88128697  59.23132816  54.44280368  53.44527426  54.48940769\n",
      "  57.75159715  52.75348576  56.68027188  53.74749033  53.63004925\n",
      "  60.35285997  53.30358891  53.02401762  55.4321781   56.2548941\n",
      "  55.66404267  62.7245463   67.75593227  71.43810087  70.56402737\n",
      "  72.46164689  72.41248571  74.17154054  73.62514089  57.27292895\n",
      "  64.60659094  67.75593227  70.00151534  68.45275776  69.09999354\n",
      "  68.93518517  71.81202517  68.70309659  59.4359612   58.26138958\n",
      "  57.5706038   57.13077177  60.02534479  57.14994542  57.13629405\n",
      "  57.34899685  62.42568895  73.5229856   74.55109401  55.16634402\n",
      "  57.18393514  58.59542049  54.96927537  60.30071125  52.73643413\n",
      "  58.29492103  58.42364068  64.49901017  58.10930252  58.47344235\n",
      "  58.58423471  57.97759819]\n"
     ]
    }
   ],
   "source": [
    "# Print support vector regressor model predictions for the next '30' days\n",
    "svm_prediction = svr_rbf.predict(x_forecast)\n",
    "print(svm_prediction)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 253,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[52.04371403 52.08115597 51.27240998 51.16008414 50.83059503 51.42217775\n",
      " 50.33636137 50.5160827  51.5270152  51.80408558 51.28738675 51.90143464\n",
      " 52.22343536 52.23092375 52.36571475 51.79659719 52.05120242 52.2608773\n",
      " 52.56790125 52.30580764 52.63529675 53.76604346 53.90832285 54.09553257\n",
      " 54.06557901 54.23781196 55.04655795 56.32707245 56.21474661 55.93767623\n",
      " 56.19228145 55.8852575  55.90023428 55.07651151 54.78446434 54.6796269\n",
      " 55.48088451 55.48088451 55.952653   56.73144544 57.52521466 56.89619\n",
      " 56.79135255 56.25967695 56.37200278 57.08339972 56.43190989 56.56670089\n",
      " 55.16637217 55.69804778 56.18479306 56.70149189 56.16232789 55.76544328\n",
      " 56.07246723 56.2896305  56.19976984 56.43939828 56.88870161 57.07591133\n",
      " 57.00851583 57.98200638 58.07186705 58.19916966 58.66344977 58.59605427\n",
      " 59.52461448 59.80168487 59.43475382 59.61447515 60.50559342 61.01480386\n",
      " 61.17206003 61.47908397 61.17954842 61.25443231 61.28438586 60.85005931\n",
      " 61.0597342  61.65131691 62.57987713 62.52745841 63.01420368 62.59485391\n",
      " 61.7935963  61.9283873  62.75211007 62.43759774 61.44913042 60.9249432\n",
      " 59.74926615 59.29247443 57.71991277 57.87716894 57.81726183 58.8431711\n",
      " 59.42726543 59.73428937 59.74926615 59.61447515 60.35582564 60.95489675\n",
      " 61.17206003 60.5205702  59.38982348 59.05284599 59.21010215 60.1835927\n",
      " 60.22103465 59.13521826 58.41633293 59.83163843 59.32991637 58.83568271\n",
      " 58.98545049 59.18763698 60.03382492 59.82415004 60.84257092 62.13806219\n",
      " 61.50154914 62.66224941 62.43010935 62.22043447 61.53899108 61.96582925\n",
      " 60.60294248 60.87252447 60.82759414 60.96238514 59.83912681 60.86503609\n",
      " 62.42262097 63.41857668 63.61327479 63.82294968 62.98425013 63.18643663\n",
      " 64.63918406 64.53434662 64.50439306 64.01764779 64.05508973 64.30969495\n",
      " 64.44448595 64.39206723 64.72904473 63.77053096 64.24229945 64.64667245\n",
      " 65.59020945 66.3615135  64.93123123 66.67602583 66.82579361 66.32407155\n",
      " 66.5637     66.81081683 66.87821233 66.90816588 66.72844455 67.4997486\n",
      " 67.37244599 67.19272466 66.38397866 64.24978784 63.41108829 64.48941629\n",
      " 63.54587929 62.6697378  61.88345697 62.44508613 61.87596858 62.78206363\n",
      " 62.63978424 62.88690107 63.09657596 63.28378568 63.49346057 62.07066669\n",
      " 62.74462169 62.1305738  62.75211007 62.57238874 65.07351062 65.73997722\n",
      " 69.72380009 71.35626886 68.39086687 68.90007732 68.72035598 68.94500765\n",
      " 68.09133132 68.63798371 68.75030954 68.88510054 66.159327   66.01704761\n",
      " 66.57867677 64.47443951 64.33216012 64.89378928 65.37304617 66.03951278\n",
      " 64.23481107 66.38397866 66.65356066 66.99802655 64.80392862 66.698491\n",
      " 65.71751206 64.90876606 65.02109189 64.67662601 65.06602223 65.18583645\n",
      " 63.50094896 63.41857668 64.01764779 63.74806579 63.59080963 62.11559702\n",
      " 62.39266741 62.75959846 63.18643663 63.80048451 64.19736912 65.15588289\n",
      " 65.59020945 65.78490756 64.71406795 65.56774428 65.99458244 65.6875585\n",
      " 65.67258172 64.82639378 64.16741557 64.10750846 63.97271746 65.27569712\n",
      " 66.08444311 64.75899828 65.04355706 64.95369639 65.71002367 66.61611872\n",
      " 66.38397866 66.40644383 68.22612232 68.35342493 67.46979505 67.43984149\n",
      " 68.1737036  69.82863753 69.67138137 70.59994158 69.13221737 68.99742637\n",
      " 69.00491476 69.5141252  68.18868038 66.53374644 66.86323555 67.18523627\n",
      " 67.25263177 65.53030234 64.78146345 65.17834806 65.24574356 63.17894824\n",
      " 63.23136696 63.74806579 63.99518262 63.56085607 62.94680818 62.29531836\n",
      " 61.06722258 60.6553612  60.6553612  59.93647587 59.58452159 58.85065949\n",
      " 58.46126327 58.20665805 55.04655795 55.44344256 55.66060584 55.69055939\n",
      " 56.19228145 53.36915885 54.13297451 51.92389981 51.81157397 50.87552537\n",
      " 51.92389981 51.41468936 53.06213491 53.23436785 52.80752969 51.98380692\n",
      " 52.89739035 51.63185264 52.06617919 51.60938747 52.84497163 51.77413203\n",
      " 50.68082726 47.92510016 49.30296371 47.5656575  47.37095939 47.86519305\n",
      " 46.69700439 47.19872644 48.06737955 48.52417127 49.15319593 49.53510376\n",
      " 50.5160827  52.47055219 52.64278513 51.90892303 51.06273509 52.17850503\n",
      " 52.38817991 52.20097019 53.52641502 52.77008774 52.65776191 53.03218135\n",
      " 53.4739963  52.17101664 53.12953041 53.96074157 53.70613635 54.79195273\n",
      " 54.25278874 53.59381052 53.78102024 52.83748324 52.88990197 52.65027352\n",
      " 53.18194913 53.70613635 54.12548613 55.00911601 55.41348901 55.99758334\n",
      " 56.03502528 56.07995561 54.8144179  54.87432501 56.01256011 56.22972339\n",
      " 55.14390701 55.77293167 55.73548973 55.4883729  55.77293167 55.1513954\n",
      " 55.91521106 55.99009495 57.02349261 57.26312105 57.20321394 57.63754049\n",
      " 57.66000566 58.40884454 58.3040071  57.47279594 57.35298172 58.22163482\n",
      " 57.86219216 57.78730827 58.46126327 59.5096377  60.21354626 60.16112753\n",
      " 59.90652231 60.64038442 61.5914098  61.35178136 61.77861952 61.02229225\n",
      " 61.20950197 60.88750125 61.32182781 61.11964131 61.32931619 62.55741196\n",
      " 62.99173852 62.78206363 62.23541124 60.78266381 60.8575477  61.18703681\n",
      " 60.97736192 59.62945193 59.80168487 60.04131331 59.37484671 59.9140107\n",
      " 59.50214932 59.55456804 59.0453576  59.68187065 59.83912681 60.51308181\n",
      " 60.39326759 60.6553612  60.58047731 59.38233509 56.5592125  57.12084166\n",
      " 57.50274949 57.45033077 55.67558262 53.44404274 53.26432141 53.45153113\n",
      " 52.00627208 52.77008774 53.78850863 53.32422852 53.30176335 51.67678297\n",
      " 52.61283158 52.68022708 52.28334247 53.72111313 53.63125246 55.98260656\n",
      " 56.33456083 56.61911961 56.54423572 57.6974476  57.70493599 56.97107389\n",
      " 57.50274949 55.32362834 56.11739756 56.33456083 56.49930539 58.52865877\n",
      " 58.26656516 58.31149549 57.79479666 56.40195633 55.69804778 54.63469657\n",
      " 54.88930179 55.22627929 55.75795489 55.24874445 55.23376767 55.3535819\n",
      " 55.96014139 56.85125966 57.21819072 53.55636857 55.07651151 54.29771907\n",
      " 53.52641502 51.68427136 52.77757613 54.13297451 54.55981268 56.10990917\n",
      " 54.69460368 54.2078584  54.44748685 55.50334967]\n"
     ]
    }
   ],
   "source": [
    "# Print linear regression model predictions for the next '30' days\n",
    "lr_prediction = lr.predict(x_forecast)\n",
    "print(lr_prediction)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Article: https://medium.com/@randerson112358/predict-stock-prices-using-python-machine-learning-53aa024da20a"
   ]
  }
 ],
 "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": "3.7.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}