Chapter_9/Lab_9.6.4 and 9.6.5.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Lab 9.6.4 and 9.6.5"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "\n",
    "from sklearn.svm import SVC\n",
    "\n",
    "from mlxtend.plotting import plot_decision_regions\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Lab 9.6.4 SVM with multiple classses"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "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>X1</th>\n",
       "      <th>X2</th>\n",
       "      <th>y</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1.624345</td>\n",
       "      <td>-0.611756</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>-0.528172</td>\n",
       "      <td>-1.072969</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.865408</td>\n",
       "      <td>-2.301539</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1.744812</td>\n",
       "      <td>-0.761207</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.319039</td>\n",
       "      <td>-0.249370</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         X1        X2  y\n",
       "0  1.624345 -0.611756  0\n",
       "1 -0.528172 -1.072969  0\n",
       "2  0.865408 -2.301539  0\n",
       "3  1.744812 -0.761207  0\n",
       "4  0.319039 -0.249370  0"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.random.seed(1)\n",
    "X = np.random.normal(size = (250,2))\n",
    "X[150:200] += 2\n",
    "X[200:250] += 5\n",
    "y = [0]*150 + [1]*50 + [2]*50\n",
    "\n",
    "data = pd.DataFrame({'X1':X[:,0],'X2':X[:,1],'y':y})\n",
    "data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x295c95ce7b8>"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAmAAAAFzCAYAAACZwbV4AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nOzdd3hUZfbA8e/NZJKZdNIhAUIRpEqvSlFARQUFxQWsqNgb+ltdXV11RXd1UayruIgNRcUCKiICUhSlK9JbKIEkpJeZzGTK/f1xITDMJJBkWpLzeZ48mPfOvXMShBze97znVVRVRQghhBBC+E9IoAMQQgghhGhqJAETQgghhPAzScCEEEIIIfxMEjAhhBBCCD+TBEwIIYQQws8kARNCCCGE8LPQQAdQG4mJiWpGRkagwxBCCCGEOKONGzfmq6qa5Olag0rAMjIy2LBhQ6DDEEIIIYQ4I0VRDlZ3TZYghRBCCCH8TBIwIYQQQgg/kwRMCCGEEMLPGlQNmBBCCCGaFpvNRlZWFhaLJdChVMtgMJCeno5erz/reyQBE0IIIUTQysrKIjo6moyMDBRFCXQ4blRVpaCggKysLNq0aXPW98kSpBBCCCGClsViISEhISiTLwBFUUhISKj1DJ0kYEIIIYQIasGafJ1Ql/gkARNCCCGE8DNJwIQQQohTOJwOrHZroMMQjZwkYEIIIQRgtVvJLMrkkaWPcN1X17Fw10LyTHmBDksE2BNPPMErr7xS9fnjjz/Oq6++Wu/nyi5IIYQQAthTuIc+s/pgdWizX/O3z+faLtfyxug3SIhICHB0IlBuueUWxo0bx/3334/T6WTevHmsW7eu3s+VGTAhhBBNXlFFEdN+mFaVfJ3w6bZPyTfnBygqEQwyMjJISEhg8+bNLFmyhJ49e5KQUP+EXGbAhBBCNHkWu4UtuVs8XttfvJ+OiR39HJEIJrfeeivvvfceOTk5TJkyxSvPlBkwIYQQTV6EPoK+aX09XusQ38HP0Yhgc9VVV7F48WLWr1/PxRdf7JVnygyYEEKIJi/WEMuMUTNYeWAlZZVlVeP39buPBGMCxZZiIvWR6HVnf9SMaDzCwsIYPnw4cXFx6HQ6rzxTEjAhhBACaBvXlu13b+eTPz8hsziT23reRmJkIi+ueZE1WWsY1HIQd/W5ixbRLYK+MajwLqfTyW+//cbnn3/utWdKAiaEEEIAobpQ0mPS+b/B/wdAdlk2vWf1Jrs8G4AVB1YwZ/McNk7dSPPo5oEMVfjR9u3bufzyy7nqqqs455xzvPZcScCEEEIIDz7d9mlV8nVCdnk2n237jPsH3B+gqOqnwFxAkaWIw6WHadesHbHhscQaYgMdVlDr3Lkz+/fv9/pzJQETQgghTmN32ll/dL3HaxuyN+BwOtCFeKcW6ExKLCUcLTvKR1s+wqg3MrnbZFKiUojQR9TqOXmmPO767i7m75gPgILCk0Of5L7+9xFvjPdF6KIGsgtSCCGEOE1oSChXdLjC47UrOlzht+TLareyYNcCOr/Zmed+fo4nfnqCDq93YG3WWlRVrdWzlmUuq0q+AFRUnl75NEdKj9QrRofTQYWtotbxNHWSgAkhhBAeXNjmQvqn9XcZG5A2gKGth/othoKKAu5ZdI/LmN1p57ZvbiPXlHvWzymzlvH+H+97vPbFji/qFJuqqhwtO8qLa15k4hcTeWP9G+SU59TpWU2RLEEKIYQQHiRHJrNw4kK25G5hzWFtF2T3lO4kRyb7LQZTpcmlLcYJ+4r2YXPYzvo5+hA9qZGpHq+lRafVKbbs8mz6vtOXo2VHAViwawGvrXuNlTetJDXK83uJk2QGTAghhPBAVVUUFHqn9ubJoU8you0IvyZfAJFhkRhCDW7jLaJb1KonmUFv4K+D/4pOcV06jQ2P5bIOl9U6LofTwfu/v1+VfJ2wu2A3P2X+VOvnBbvFixfTsWNH2rdvz7/+9S+vPFMSMCGEEOI0+eZ83t38LpfMvYRLPr6EeVvnUWAu8HscseGxPDHkCbfxmRfPJCkiqVbPah3Xmp+n/Ez/tP7EGeIYfc5o1t22jpTIlFrHZXVY+e3Ibx6vrTm8ptbPC2YOh4O7776b77//nu3bt/PJJ5+wffv2ej9XEjAhhBDiFGXWMv656p/c+s2tbMrexLoj65j4xUTe3vg2FrvFr7FEhkVyR587WHnTSq7pfA3Xd7+eTVM3MardqFpvBIjQRzAgfQDfTvqW7XdtZ+64uXRI6FCnDQWGUAMXtbnI47UR7UbU+nneNHcuZGRASIj269y59XveunXraN++PW3btiUsLIy//OUvLFiwoN5xSg2YEEIIcYpSayn/Xf9ft/F//fwvbu5xs9+bsMYb4xnSegj90vqhoBAeGl6v5yVGJNY7phAlhAldJvDG+jfYXbC7arxfi34MSBtQ7+fX1dy5MHUqmM3a5wcPap8DTJ5ct2ceOXKEli1bVn2enp7O2rVr6xmpJGBCCCGEC4vdgs3pXuBeVlnmcdxfPNWCBVJqVCqrblrFssxlrDm8hhFtRzAofRDJUf6tkzvV44+fTL5OMJu18bomYJ7aa3jjKCpJwIQQQjRJpkoTpdZSFEUh3hBPWGgYoC37tYhu4VZg3imxEwZdcCVBgZYSlcKkbpOY1G1SoEMB4NCh2o2fjfT0dA4fPlz1eVZWFi1atKj7A4+TGjAhhBBNTm55LtN+mEb719rT+Y3OPLv6WfLN+QAkRSQxb/w8lxmnqLAoPhr3UUBnd8SZtWpVu/Gz0bdvX/bs2UNmZiaVlZXMmzePMWPG1P2Bx8kMmBBCiCalwFzAjV/fyA/7fgDAbDPzz1X/pMJWwTPDn8GoN9IvrR/77t3Hmqw16EP09E3rW+tdh8L/pk93rQEDiIjQxusqNDSU119/nYsvvhiHw8GUKVPo0qVLvWNVGtLRAX369FE3bNgQ6DCEEEI0YIdKDtF6Zmu38Qh9BHvv3ev3IntRsx07dtCpU6ezfv3cuVrN16FD2szX9Ol1r/+qDU9xKoqyUVXVPp5eLzNgQgghvK7UWkpWaRbvbn6XECWEKT2nkBadRnR4dKBDo9JR6XHcbDPjVJ1+jkZ42+TJ/km46ksSMCGEEF5lc9hYtGcRE7+YWDX2nzX/Yf6E+YzpMIZQXWB/9ESFRdE6tjUHSw66jI9oMwKj3higqERTI0X4QgghvCrfnM/di+52GVNRueu7u8gz5wUoqpOSI5P5ZuI3Lh3gOyZ05H9j/ke8MT6AkYmmRGbAhBBCeFWFvYLCikK38VxTLlaHNQARuQpRQuic1JnNt28m35yPXqenmaEZKVG1P5LHV0osJZhtZnQhOr+fPyn8QxIwIYQQXmUINRAbHkuJtcRlPMGYQLiufl3cvUUXoqN5dPOgK7hXVZXM4kweXPwgyw8sJyMugxdHvsigloOICY8JdHjCi2QJUgghhFclGBP4z6j/uI2/fPHLXjkGpzHLKc9h4OyBLNy9kPLKcrYe28qlcy9lW962QIcmvEwSMCGEEF4VHhrO1Z2uZsNtG7jpvJu4ucfNbL59M2M6jkGv0wc6vKC2/uh6jpmOuY0/veJpSiwlHu4Q/jBlyhSSk5Pp2rWr154pS5BCCCG8Ls4YR29jb2ZdMQug3omX2Wam2FJMTnkO8cZ4YsJiiI+ouWA+z5THMdMxjpYdpWNiR+IMcUG/jOcp+QoNCaVX816UWkupsFWQEJEQNImsqqrkmnKrav7ijfGkRKZ45azEYHLTTTdxzz33cMMNN3jtmZKACSGE8BlvJApl1jK+3f0ttyy8hQp7BQBXdrySWVfMIinSc3f6Y6ZjXPP5Naw6uArQCu9njJrBzT1uJtYQW++YfOXCjAtdPjeEGvhywpesO7qOwe8Oxqk6ub337dzR545qv3Z/2le0j5EfjuRA8QEAMuIy+PH6H2kf3z5wQWXOhT8eB/MhiGgF502HNvVrDDZkyBAOHDjgnfiOkyVIIYQQQa2wopDJX06uSr4Avt71NR//+TEOp8Pt9Q6ng/d/f78q+QJwqk4e/OHBqvMeg1ViZCJvXfYW+hAtcb2v/318vfNrnlrxFK3jWnPDeTeQa8rl+Z+fp8xaFtBY80x5XP3Z1VXJF8CB4gNM+HwCeaYAtRvJnAvrpoL5IKBqv66bqo0HGUnAhBBCBK0KWwVZpVmkRqW6XXv/j/c9trsorCjk022fenze8szlXo/Rm2LCY5jcfTKZ92ey+ubV3NnnTubvmM93k77j2i7XsjlnM6XWUq4898qAJ2Bmm5k/cv9wG9+csxmzzezhDj/443FwnPbeDrM2HmQCugSpKEoc8D+gK6ACU1RV/TWQMQkhhAi8AnMBG7M38taGt4gOi+bDqz5k0d5FvPTrS1WviTPEERri/mMsPDTcpcnqqdJi0nwWs7dEhUURFRZFWkwaB4sPMv3C6by14S2+2f1N1Ws+/vNjVt+8mhYxLQIWpy5Eh07R4VBdZyF1ig5diC4wQZkP1W48gAI9A/YKsFhV1XOB84AdAY5HCCFEgJVaS3lhzQtc/NHFfLXzKz7Y8gEjPxxJr9ReDMsYVvW6J4Y8QTNjM7f7Y8JjeGrYUyi4FoK3iG5Br+a9fB2+Vxn1RrqndHdJvgAcqoOHljwU0CXVmPAYru1yrdv4xK4TA7fZIaJV7cYDKGAJmKIoMcAQYDaAqqqVqqoWByoeIYQQwaHUWsqMNTNcxlRUnlzxJHf2vpO2zdry2dWf0bN5z2qfcW7iuSy/cTn90vqRFJHExK4TWTNljcelzGCWHJlc7VLjroJd2Bw2P0d0Ukx4DC9f8jIPDHigatbugQEPMOPiGYFLwM6bDroI1zFdhDZeDxMnTmTgwIHs2rWL9PR0Zs+eXa/nQWCXINsCecAcRVHOAzYC96uqajr1RYqiTAWmArRqFXwZrBBCCO8qs5a5LWsB7C/az4CWA1gzZQ3Jkck1tjqIDo9mWMYwvpv4HTanjaiwKKLDo30Zts+cm3guCgoqqsv4oJaDiNBHVHOXfyRHJvPchc/xyOBHAIgLj8OgNwQuoBO7Hb28C/KTTz7xQnCuArkEGQr0Av6rqmpPwAQ8evqLVFWdpapqH1VV+yQlBX7LrRBCCN+KCY/BEOr+Q7x7SncMoQZSos6+z1RiZCLNo5s32OQLtFq3vw/5u9vYS6NeCoqWGka9kdSoVFKjUgObfJ3QZjJceQAmObVf65l8+UogE7AsIEtV1bXHP5+PlpAJIYRowuIMccwY5boEGa4LZ9bls5rkwdSxhlgeGPAAf9zxB/8Y+g9mXT6LbXdto22ztoEOTdRDwJYgVVXNURTlsKIoHVVV3QVcBGwPVDxCCCGCQ2RYJJO6TWJo66F8/OfHxBpiubbLtaREed7Z2BTEG+OJN8bTPaV7oEMRXhLoTvj3AnMVRQkD9gM3BzgeIYQQQSDOEEecIY7pF9WveFqIYBXQBExV1d+BPoGMQQghRHA7ZjqGw+kgPDSceGPN5z8K0VAEug+YEEII4VGptZQf9v7AsPeG0WpmK8Z9Oo7MokyPxw8J0dBIAiaEECIo7SncwyVzL2FH/g56pvZk+oXT+WHfD7y69lV25u+kqKIo0CGKJuDw4cMMHz6cTp060aVLF1555RWvPDfQNWBCCCGEG1OliedXPw9AalQqMy+ZyZXzriTPrB3yPG3JNJ678Dnu6nuX31sxVNgqKLIUUemoxBhqPGNPMtGwhYaGMmPGDHr16kVZWRm9e/dm5MiRdO7cuV7PlRkwIYQQQcfmtFUdtH1rr1v5z5r/VCVfJ/z9p79TYi3xa1wF5gJm/DqDDq91oM0rbTh/zvmsO7KOSkelX+PwlwJzAduObeODPz5g/ZH15JnyznzTWah0VJJTlkN2WbbXv3dz/5xLxswMQp4OIWNmBnP/nFuv5zVv3pxevbQuWdHR0XTq1IkjR47UO05JwIQQQgSdOEMcN/W4CYAeKT1YfWi122ucqpNd+bv8GteyzGU88dMTmGzaoS17C/cy/P3hXktMgkm+OZ+HlzxM1/925cavb6Tf//ox/rPxHDMdq9dz80x5PL/6eXrO6kmvWb14bvVzXvv+zf1zLlO/mcrBkoOoqBwsOcjUb6bWOwk74cCBA2zevJn+/fvX+1mSgAkhhAhKl7a/lOu6XcfR8qO0j2/v8TXpMel+i6fAXMDM32a6jVfYK/jl8C9+i8Nf9hXu470/3nMZW31oNd/u/hZVVT3fdAZl1jIeW/4YT618ipzyHHLKc3h65dM8tvyxas+8rI3Hlz2O2WZ2GTPbzDy+7PF6P7u8vJzx48czc+ZMYmLqf9alJGBCCCGCUlJkEq9e+irjO43n3yP+TYji+iNrZNuRJEX674g6FRWn6vR4rTHuzFywa4HH8c+3f055ZXmdnllqLeW9399zG3/v9/cotZbW6ZmnOlRyqFbjZ8tmszF+/HgmT57MuHHj6vWsEyQBE0IIEbSaGZvRIroFvZr3YtPUTVzd6WrOb3U+b45+k4/GfURiRKLfYkmMSOTefve6jYfrwrmg9QV+i8NfOiZ09Djerlk7wnXhdXqmzWnD7rS7jduddmxOW52eeapWsa1qNX42VFXllltuoVOnTkybNq3OzzmdJGBCCCGCXlRYFOelnsecK+ew4NoF3N7n9jqfC5lTnsPn2z7nieVPsPrg6lrVH13S/hIeHfxoVQLSMqYlP17/o18TQX8Z1W4UqVGpLmPGUCMPDXyIsNCwOj0zQh9Bl6QubuNdkroQoY+o0zNPNf2i6W7PidBH1OtEhV9++YUPP/yQ5cuX06NHD3r06MGiRYvqGypKXddxA6FPnz7qhg0bAh2GEEKIBiq7LJvz55zP/qL9VWNXdrySWVfMOuvlTFOliWJLMVaHlQh9BEkRSehCdL4KOWBUVSWrNItnVz3L6kOr6ZLchWeHP0ubZm0I09UtAQPYkbeD4e8PJ9eUC0BKZArLb1xO5yTPbR127NhBp06dzvr5c/+cy+PLHudQySFaxbZi+kXTmdxtcp3jPVue4lQUZaOqqh5P/JEETAghRJNgd9qZvno6T614yu3aqptWsSNvB6nRqfRL6+c289OUmW1myqxlROgjiA6PrvfznKqTY6ZjZJVmAZAWnUZKVIpbjd8JtU3AAqW2CZg0YhVCCNGoqapKTnkORZYilu5f6vE1i/ctZv2R9fy4/0cy4jL4+eafSYtJ83OkwSlCH+GV5cETQpQQUqNSm3ySKzVgQgghGrW9hXvpNasXr659lXMTz/X4mo4JHatmZA4UH+ClX1/CYrf4M0zRxEgCJoQQotHKN+dz/VfXk1Oew7yt87i1563EhLv2cOqa3JXUqFR25O+oGluyf4lX2iII7wj2cqm6xCdLkEIIIRoti93C2iNrASixlvC3ZX9jyXVL+GDLB+wp2MOodqMYmD6QiV9MdLmvQ3wHjKHGQIQsTmMwGCgoKCAhISEoz9xUVZWCggIMBkOt7pMETAghRKOlU3REhUVVNQ796cBPjPhwBFd3vpp3x7xLhD6CER+O4HDp4ap7wnRhPHvhs14pOBf1l56eTlZWFnl5wXvck8FgID29dqcySAImhBCi0WpmbMZDAx/i6ZVPV42VV5ZjsVmICo8izhDHosmL+OCPD1iwawHtmrXj8Qsep3Vc6wBGLU6l1+tp06ZNoMPwOmlDIYQQolErMBfw6bZPmfnbTKwOK1N6TuHOPne6NHK1O+yUWksxhBqICPPejj/RtEkfMCGEEE2aw+kg35yPqqokRCSg1+kDHZJoAqQPmBBCCJ8wVZow2UxEh0Vj1Adv0bouREdKVAqgzYiZbWZAO+KombFZIEMTTZS0oRBCCFFrNoeNfYX7uOO7OxjxwQimLZnG4ZLDQd8uIKs0i/GfjafVzFa0ntmayV9OJrssO9BhCT9xqk5yynM4VHKI7LJsbI76HwBeV7IEKYQQotYOFh+k63+7Vu0uBO1Mv823b6Z5dPMARla9fHM+Y+eNZc3hNS7jYzqO4f2x7xNnjAtQZMIfzDYza7PWcvOCmzlYcpAEYwL/HvlvxncaT5zBN7/3NS1BygyYEEKIWrE77byx/g2X5Asg15TLwl0LPd5jtpnJKcuhsKLQHyF6ZKo0uSVfAN/s+gaTzRSAiIQ/5ZnyGPXRKA6WHASgoKKAWxfeyo68HWe40zckARNCCFErlY5KdhXs8njt1G7yJxwpPcK0H6bRf3Z/rpx3JWsOr6HcWu7h7sBRaTirQcGg0l5JTnkOeabg7c11uvnb52N32t3GZ/w6o6om0J8kARNCCFErEfoIxnca7/HaVede5fJ5TnkOF8y5gLc3vs2hkkOsPrSa898932Oi5muR+kgGpg90Gx99zmgi9ZF+j6ehyi3P5amVTzFo9iAumXsJX+/8mqKKIr++/7e7v+WZlc+w+uDqs04CKx2VHsdtTltAahclARNCCFFrl7S/hCGth7iMXdP5Grokd3EZ2563ncziTJcxFZUnfnqCYkuxz+M8VWJkIp9d8xkXtLqgamxU21G8c8U7ddoJabFbyC7L5kjpEb8mIPVRZi3DareSb8rnQNEB9hXuI7c896zvzzfnM+HzCTz/8/NkFmeyKXsTV316Fd/v/R6n6vRh5Jrc8lwu/OBCrvjkCv6x4h8MeW8IN3x9A8dMx85474QuE1BwP8rovn73ERnm/wRc2lAIIYSoteTIZOZfM5/M4kz+zP2TPi36kBaTRmJEosvrskqzPN5/pOxItTMSvpQek85X136FyWZCQSEyLJJ4Y3ytn5NvzueVta/wym+vUF5Zzoi2I3jnineCtoN+nimPZZnLeP+P90mNTOW23rcxe/Ns3t38Lh0TOvLtpG9pH9/+jM8pqihi1aFVbuNP/vQkF7W5qKrVhy84nA4+2foJ2/O2u4wv3ruYbce2kdwmuZo7NSmRKcyfMJ/bvrmNwopCjKFGnhjyBD1Te/os5ppIAiaEEKJOkiKTSIpMol9av2pfM7jlYI/j4zqNo5khMP23EiISSCChzvfbnXZmb5rNs6uerRr7cf+PjPxwJKtuXkVqVKo3wvSaQnMh9yy6h8+2f1Y19uGWD/n6L1+z4egGtuRuYcwnY1hx0wqX0wE8KbJ4nuk7WnbU5zNgZpuZxXsXe7y2ZN8ShrcZXuP9UeFRXNHhCrbeuRWzzYxRbyTOEEeEPjAnH8gSpBBCCJ9JiEjghZEvuCz99EztyV197mqw3ejzTfnMXDvTbXxP4R7yzfkBiKhmhRWFLskXgEN1MH31dG7rdRugbZ44fVerJ+kx6RhD3Rvujj5ntM+X8QyhBnqk9vB4rWfzs5vF0uv0NI9uTrv4drSIbhGw5AskARNCCOFDcYY4pvaaSub9mXx97ddsnLqRxdct9ulSla+pqFTYKjxes9qtfo7mzA6XHfY4vjN/Jy1jWlZ9rlN0Z3xWM0MzPhn/CaEhJxfQWse2ZsaoGcSEx9Q/2BrodXru7nu3W8+uds3audUjNgSyBCmEEMKnYg2xxBpig7Y+qrbiDHFM7j6ZN9e/6TIeb4ynRXSLAEVVvbZxbVFQ3Fpt9E/rz7a8bQBc0OqCs5rBMuqNjGw3ksz7M9mcs5l4Qzzt4tv5bdm1RXQLttyxhZm/zWRzzmZGtB3BlJ5Tgm7Z92xIJ3whhBCilnLLc7nh6xtYsm8JoCUG30z8hu7J3QnVBdfcRomlhJd+e4lnVj5TNRYbHsvCiQu58esb6Znak9dHvx6UyWN1Ku2VmO1movRRQff9PlVNnfAlARNCCCHqoMBcQHllOZWOSqLDo0mKSEIXcuZlvEAorCgkqzSL+dvnkxadxmUdLkMfosehOojURxJriA10iI1STQlY8KaNQgghRBBLiEggIaLuuyn9Kd4YT7wxnu4p3QMdijhOivCFEEIIIfxMZsCEEEIIL8k351PpqCRECSE5MpkQReY5hGfyf4YQQghRTzaHjS25Wxg9dzRpL6XR751+fL7tc0osJYEOTQQpScCEEEI0Gaqqkl2Wzbe7v+W939/jQPEByqxl9X5unimPwe8OZv3R9QAcLj3MX774C1uPba33s0XjJEuQQgghmoz9RfsZOHsgeeY8ABQU5oydwzVdrqlXV/SlmUs9dpJ/7ufnmDd+HtHh0XV+tq/kmfLYmb+TJfuX0COlB4NbDW6Q/bQaKknAhBBCNAnFFcXct/i+quQLtK72t397OyPbjqxXAlZqLfU4bqo0YXfa6/xcX8kz5XHrwltZuHth1Vi7Zu1YffNqmkc3D2BkTYcsQQohhGgUSq2l5JTnVFt3VWGvYNn+ZW7jVoeVrLKser336HNGu5x3ecJtvW6jmTEwh47XZG/hXpfkC2Bf0T7+t/l/QZkwNkaSgAkhhGjQHE4Hewv3csvCW+j3Tj+u++o6duTtoNJR6fK6ECWk2uOQEoz16+eVFJHEh1d9SKQ+suq9pvaaysXtL67Xc33lpwM/eRxfun+pV2rixJnJEqQQQogGLbs8mz6z+lBi1Wa+DpceZtn+ZWy/ezsZcRlVr0uOTOaFES9w5adXutx/2TmXuR3wXFvR4dGM6zSO4W2GU2IpISosipjwmKDtMN8jtYfH8e7J3eu1FCvOnsyACSGEaNA+3fZpVfJ1QoW9gjfXv4nD6agaUxSFYRnDWHHjCi5qcxG9mvfi5YtfZs7YOV7paG/UG2kR3YJOSZ1oGdvS68mXqdKE2Wb2yrN6N+9N56TOLmMx4TE8POhhwkPDvfIeomYyAyaEEKJB21e4z+P4/qL92J12l/MZYw2xDM0Yynkp52Fz2og3xgft+Y0nFFUU8UfuH7yy9hUUFB4c8CBdk7vWWFtWai3FbDMTqY/0uAMzJSqFZTcs4+M/P+b7vd/TLbkbDwx4gLToNF9+KeIUchi3EEKIBm3VwVUMfW+o2/iXE77kqk5X1XjvsfJjOFQHoSGhJEUm+SrEOrPYLPx3w3+ZtmSay/ibl73JlB5T3GarbA4bB4oP8OjSR9mcs5k+Lfrw3EXPkRGbQajOfc7F4XRgqjRhCDUQFhrm06+lKarpMG5ZghRCCNGgdU7qzJQeU1zGrjr3Ks5vdX6195RXlrNs/zKGvj+UFi+14NK5l7IxeyNWu9XX4dZKoaWQJ1c86Tb+2LLHKN01fLQAACAASURBVKgocBvPKc+h16xefLnzSzKLM/l8++f0ntWbHFOOx+frQnTEGGIk+QoAScCEEEI0aIkRibw48kV23bOLr6/9mm13bWPWFbNqnNE6WHyQUR+NYmf+TgA2Zm/k/HfPJ9eU66+wz4rFbvHY4LXYUuy2y1NVVf636X9ury+1lvLRlo98GqeoPakBE0KIIJJbnsuW3C1sy9vGkNZDaBXbisSIRJ+/569Zv7I5ezMj2o7g3MRzg3I5ribxEfHER8TTIaHDGV9b6ajkpV9fwqk6XcYtdgvzts7jr4P/6qswa80YaiQ5MpljpmMu42nRaRhCDS5jDtXBnsI9Hp+zu2A3TtUph4MHEUnAhBCNm9MOIQ3jr7qc8hxGfDCCbXnbqsYmdZ3EK5e+4rMkLKc8h6HvDWV3wW4Anln1DJedcxlzxs5pcEnY2XI4HRRZijxeK6wo9HM0NUuKSGL2mNmMnTe2KmHUKTpmj51NUoTr709oSCiTuk3ik62fuD1nUrdJknwFGfndEEI0PqoKFUdhxwz49QY4MA8sx858XwA5nA4++OMDl+QL4OOtH5NZlOmz93z/9/erkq8TvtvzXdXSXGNk1BuZ2nuqx2uTuk3yczQ1C9WFMixjGLvv2c2/LvoX/x7xb/bcu4fzW57vcfdm/7T+jDt3nMvYX7r8pdq+X2fDYrdQbCmm1FLK4ZLDbD22laNlR6mwVdT5mUJmwIQQjVHFUVjcGyzH63kOfgIJ/WHoQjAkBza2apRaS/luz3cery3dv5S+aX29/p7lleUs2b/E47Vlmcu4oPUFXn/PYNGnRR8e6P8Ar657FafqJEwXxvMXPU/LmJaBDs1NVFgUUfFRPHL+I2d8bVJkErOumMUzw59hR/4OOid1JiUypU59zqx2K0fKjvDvX/7N3sK9jGo7ikEtBzFh/gRKLCW8c8U7jOk4JigPGm8IJAETQjQuzkrY9drJ5OuEgrVQvAVSRwQmrjMwhho5N/FcVh1c5XatS3IXn71n95TuLM9c7natZ2pPn7xnsEiMSOSpYU8xbeA0Ci2FJEUkERMeQ1RYVKBDq7eEiAQSIhLq/f/NwZKDnPfWeVjsFgCWZy6nd/PevH3524ydN5brv7qevfftlQSsjmQJUgjRuNhNUPCb52v51YwHAYPewKODH3U7BqZds3b0T+vvk/cMCw3jwQEPEhvu2rG9Y0JHBqQP8Ml7BpNYQywtY1uSFp2GzWEjz5RHnikv0GEFhfLKcv6x4h9VydcJG7M3YnfaaRnTEhWVxXsXByjChi/gM2CKouiADcARVVUvD3Q8QogGLjQSEgfBsZXu1xIH+j+eWkiPSWfLHVuYvno62/O2M6rdKO7scycpUSk+e88W0S3YcucWXl37Kr/n/M7F7S/m+u7X+/Q9g8mR0iNcO/9afjn8CwC9mvfiiwlfuJwh2RSZbWa25233eG1P4R5aRLfgcOnhep+h2ZQFvBO+oijTgD5AzJkSMOmEL4Q4K+Yj8ENfqMg+OZY0GC74MmhrwE5ltpmpsFUQEx6DXqf3y3vaHDbMNjNRYVFBfzSPtxRWFPKX+X/hx/0/uoz3adGHRZMWNdpdoGfDYrfwt6V/Y+bamW7Xlly3hIlfTMTmtLHjrh20iGkRgAgbhqDthK8oSjpwGfC/QMYhhGhkjC3gko3Q+1Vocz2cP7/BJF8AEfoIEiIS/JZ8Aeh1emINsU0m+QIwV5rdki+ADUc3YLKZAhBR8DCEGnh40MNumxImdZ3E/uL9dEjowJopa5rMTKkvBHoJcibwV6DaCj5FUaYCUwFatWrlp7CEEA2aooCxOXS8F1QnSP8j4YGiKIQoIW4NWYGg7ZnlVJ3klOXw+fbP2Zm/kwldJtA1uatPZuvSYtJYd9s6Vh1cxc78nVza/lJaRLcgRAlhfKfxPm8Q3NgFLAFTFOVy4JiqqhsVRRlW3etUVZ0FzAJtCdJP4QkhGosg/UEqAi8qLIqrzr2KL3Z84TI+LGMYkfrIAEVVs0Mlh+g9q3dVw9i3Nr7FtV2u5fXRr/skIUqNSmVClwlef64I7BLkYGCMoigHgHnAhYqiyGFVQggh/CLWEMvro19nQpcJ6BQdCgqXd7icuePm1qlvlq+ZKk08teIpt279n277lOyy7GruEsEqYDNgqqr+DfgbwPEZsIdVVb0uUPEI0SDZysBers3yhCdrS29CiLOWGpXKO1e8w0ujXgIgMiwyaHf2lVeW81uW51Yqf+T+QbeUbn6OSNSHzM0L0VCZDsG6qfDtubB0OGR9BZXFgY5KiAYnJjyGtJg00mLSgjb5Am1zRnVHCnVJ8k2zXuE7QZGAqaq6QnqACVELFTmwdCgcnAe2UijdAavHB3WjUSFE/USHR/Pshc+6desf1W4U6THpAYpK1FWgd0EKIeqibC+YDriP//kUJPSBcNmdJERj1Dq2Ndvu2sZbG95id8FuJnSZwLCMYU26Z1lDJQmYEA2RNb/6cafdv7EIIfxGr9PTKrYVzw5/FpvTRnhoeKBDEnUkCZgQDVF8b1B0oDpcx1tfC2HxgYnJl+xmqCyEwo2gbwYxHcCYGuiogoKqqiiy+aLJCQkJITxEkq+GTBIwIRqisGYw6CP49UZwVmpjiYOgw32gCwtsbN5mr4DsJfDzNaAen92LSIcRqyEqI6ChBYrVbiWnPIc5v8/hSOkRbuxxI50SOwVl6wQhhGcBPwuyNuQsSCFOYTdrux7L90N4glb3ZWiEdSDmo/BNe3BUuI6njYWB70NYbGDiCqA/cv5gwOwBWOyWqrEHBzzIk0OfDOpdfEI0NUF7FqQQoh5CIyCiBSSfD7GdGmfyBVBx1D35Asj+3vN4I1doLuTe7+91Sb4AZv42kxJLSYCiEkLUliRgQojgFlbNjI4xDWh6tU9Wh5XNOZvdxlVUDpceDkBEQoi6kARMCBHcwuIg5UL38e7/bLyzfjUI04XRPaW7x2vSC0qIhkMSMCFEcAtPhMEfQ7d/QHR7SOgLQxZAi9FN8qDthIgEXrv0NcJO22xxZ587iQ1vevVwQjRUUoQvhGgYnJVgLdLabxiadqNZi91Cdlk2b254k6NlR5nSYwrnpZ5HYkTT/r4IEWxqKsKXBEwIIRoop+rE4XSg1+kDHYoQwoOaEjDpAyaEEA1UiBJCiK7pLcM2Vflm7QQMmelsHCQBE0IEF6cDUCFE/noSAqDAXMDivYt5bd1rANzf/35GtRsljXcbOPmnkxAiOFSWQNEWWH+X9lH0uzYmRBNmrjTz0m8vcd1X17H2yFrWHlnLpC8n8eraVzHbzIEOT9SDJGBCiMBz2ODIt/D9ebBvFux7B77vCVlfnzxqqQEqqigiuyybUktpoEMRDVSRpYiXfn3JbfzFNS9SbCkOQETCWyQBE0J4l80EFdlQkQNO+9ndU5kPm+53H9/4AFjyvRufH1TYKvg953fGfzaeXrN6cdOCm9hXuA+H03Hmm4U4RaWj0u3UA4AKewWVjob7jxMhCZgQwpsqsmHDPfDNObC4F+x+DawFZ77PYfH8OlsxON1/+GAt0FpSBKms0iz6vdOPnw78RE55Dl/t/Iq+7/Ql15Qb6NBEA2PUG2kd29ptvG2zthhCDQGISHiLJGBCCO+w5MGqsZD5HtiPz4Jtmgb73zvzTJjOAOEeutqHxWvXTrAWwIG5sHIMrL4SshaCtdCbX0XtqE7t/W3lVUOVjkr+s+Y/2Jw2l5cWWYr4dve3/o5QNHDJkcl8PP5jl2TLGGpk7ri5pESmBDAyUV+yzUgI4R22YihY7z6+/d+QMRmMqdXfG5YIfV+Hn691He/zGoQdT8xs5bDtedg54+T1Y6ug5wzocDfowuv/NdSG5Rgc+gIOzgVDCnR9AqLaU6lCVlmWx1sOFB/wb4yiwQtRQujdvDd7793LL4d/QUFhUMtBJEUmoShN7yzUxkQSMCGEd9grPI/bzqJQWKeH5pfAZdtg7yxtZqn97RCRrl0DsJXA7lfd7936DGRMBGPzusdeW9Z8+PUGyP7h5Njhr2DEKqKSz2dS10ks2rPI7bbxncb7L0bRaISHhpMWk8aELhMCHYrwIlmCFEJ4hyEJwpq5j6ePg9DIM9+vj4HYztB7JvR+BeK6QNgpZxs6LHDash6gJWaexn3Jku+afAGgwuaHwJLPqHajGNNhjMvVaQOmkRGX4bcQhRDBTWbAhBDeEZYIQ7+DFaNPznrFdYNeM0AfXbtneVpaCY3UlvospxWyx5zrWifmDxVHPI+X7QXVRlJkc2aPnU2eKY8DxQfomNiRZoZmNDN6SFCFEE2SJGBCCO/Q6SG+D1y+XSvAD43QiugNyd55fngiDPwIVo4+OeOli4CBH3jvPc5W9DmghGhLpadKGQ46bbYvMSKRxIhEOiV18m9sQogGQRIwIYT36PRaLVZ19Viqqs1gqXZQ9FridLaFxCGhkDQYrtgHucshJAySh0C4n5MvAH0cnPc8/P7IybHwJOj5AoTFnBxz2sGSA4fma7Nmra6FqLYQHu//mIUQQUVRVTXQMZy1Pn36qBs2bAh0GEKcvcoirT3DsRUQmQFx3WveDdiY2coh/1dYfweU74eodtDvLUgYCPqzqBELNpXFWjJ5dLGWcCZfAIZU14SyZCf80A/sZSfHOj+qfZxa31YbFTlQsg3MWZB0vjYzWNdniaBSVFGExW5BURSSIpLQhegCHZKoJ0VRNqqq2sfTNZkBE8JXKoth16vw51MnxyIzYORqbXdfU1NxBFZccnLZrnwf/HQJXLYd9B0CG1tdhMVpHzEdPV+vLIZND7gmX6C15TjnjrolTabDsPQCMB08OdbrJWh3i7aJQTRIqqqyv2g/d3x7B8syl5ESlcI/h/+T8Z3GS91gIya7IIXwlcpC1+QLwHQAtj1XfcuGxkp1wp7/utdMqQ7Y+7a2NNnYOCogf62HC6o2A1hbdhP8+Q/X5Atg00PaTKtosHLLc7lgzgUszVyKikpOeQ63fXMbyzKXBTo04UOSgAnhKwUbPY9n/6C1TmhKVMBe7vmarez4C+rAknf8zMkgPBMvJExrq+FJRKvaP89WCtlLPFxQoXhr7Z8nfKMiF3J/gj2zoGTHWZ3UsKtgF9nl2W7jz//8PHmmPF9EKYKAJGBC+EpUhufxyDYQ4ueu7WfLkgfHVsPWf8KR77QfJt4QEqI1VvXknNu1HYW1YS2CrAVay4ulF8Cf/9Q60weT8AStp5lyWqVHxmTP/dLOJCQMIt3PBAQgIq32zxPeV5ENy4bDsgth/e3wXWf48+kzJmElVs//ICuzluE8fdZYNBpSAyaEr0S2hmY9oWjzyTElBHr8C8KDsK7Dkne8u/vik2OxXeDCpd7ZOBDVDnq8AFueAKdV693VfTpEtq3dc1Qn5K2BLU9CyVbt823PQv4aOP9TrSg9WMR21Wrcdr0MFUehzU3HC+frsAsyPEH7f2fpUFxmDOP7glESsIBzVsKeN6F0h+v47le1f2TU8Hveq3kv9CF6t/NDr+9+PfFG9/tMlSbyzHl8sf0LAMZ3Hk9yZDIR+oj6fx3Cb2QXpBC+VJEL+/4Hh7+EyJbQ7WmIaq/t+rMWat3dFUX74RoSFthYc3/S/uV+ut6vamct1naWyhObSWvSai+H0GitiD20Fj80HDatrUP2D9oPvPhesGMGHJ6vXb9iD0S3r3+c3ua0aR+1+Vo9sZVByXb480lttqXl1drMolEOZQ44Sz78dDEUbXK/1ve/2saLalTYKvjpwE9M/GIipdZSAC7vcDmzx8wmOdK1zYrdYWfJ/iWM+WQMDtUBgE7RsWjyIi5qc5HsnAwysgtSiEAxpkDnR6D9VG3GRx+tFZyX7YO1t8CxldpyVOdHtZ1s4QmBi/Xo4mrGF0Hbm2rfzd4TfWT9Wk6UbIGlw07Wk4WEw+B52oaH3OXBW4weotc+6ksfDYn9YfCnWgIa1sw7zxX1pzNos7yeErDoc2q81ag3clGbi9h5904KKgqICosiOiyahAj3vw/yzHncsvCWquQLwKE6mLJgChumbiA1qom2uWmAJAETwtdCQrVzEk+w5MLSIdqSFGhJw++PaMlX25u9M9NUF/E9PY/HnQc6o39j8cRyDNZMdi3md1phwz3Qfxbk/9Z0luLC4gIdgTidPgq6Pw1HvnY9mzS6g7YUfQbhoeE0j25O8+iaD5W32C3klOe4jR8pO4LFbql12CJwpAhfCH8rzzyZfJ1qx3+0OqxASR6mdWk/VVgzbfkxJAj+reaogNJd7uMVR7TjfwbNrVtxuxDeEtkGLv1DWxpu1hO6PA4jVnh1iTg8NJyosCi38ZjwGMJ0AS5jELUSBH+rCtHEOEyex+1m6tyOwRuMqTDyZ9g3G3KWaT9Azn0QjC0CF9OpFD2ERrm3swjRaztOw/tCaBDM1ImmK9QAsZ1gwLtafWdYrNdrO+ON8Tw97GkeWvKQy/izw58lwRjAEgZRa5KACeFvMZ21Q6QdZtfxtjcG/oxAY3OtHq3DPdqyoy6I2mWExWuxbfm76/g5d0NYQv0L3IV32U3aaQBle7Ul+PAk16X4xkwf7Z2aSQ8MoQZu6nETPVJ7MPO3mSiKwoMDHuS8lPMIDw2iP6/ijGQXpBD+5rBCwTpYPR6sx5cc08ZqdUyGABws3ZBYCyBnKex8WTvQ+5y7If2K4Go9IbRzP7O+hrVTTtZDJQ6CC76UHZteVF5ZjoJCZFgDPEu1iahpF6QkYEIEgsMGlQVaR3ydUVtaC/TsVyCpztptPrAUAKq2ceHUw69FcDAdgoVt3I+e6vIYdPtH4FuuCOEn0oZC+I3Fon1ER4NO2tFUT6fXaq680eC0IbPkadv2D3ys7RZreyMYUs9c9G+QWpegVrjBPfkCOPQ5dLhPZsGEQBIw4SU2Gxw+DC+8ADt3wujRcMMNkNrE8wtRA2sBbLgXDn16cmzHC3DJxuBspirOXng1tV7hicGxo1aIICB/EoRXHD4M3bqB+Xhd+cqV8OGHsHQppMg/doUnllzX5Au0A6d/fwwGzPZZEbPwg+hzIDIDTAdcx7s9Fdhmw0IEEekDJurNbIbp008mXyds3Qp//hmYmALOXqEVIovqlWz3PF64QdtBJxouYyqMWAWtJ4E+TmtEOmQhJPQLdGRCBA2ZARP1VlEB27Z5vrZ1K4wY4d94AqqyFMr3wo4XtXP7zrkDEgfKv/o9ienkeTy+N4Q2kl1dlSUn+0HpDIGOxr8iW0K/t7S+bYpOdvgKcZoaZ8AURYlRFKWdh/HuvgtJNDTR0TBypOdrQ4f6N5aAUp1wbAUs7g0H58HR72DlFbDtOS0xE64MKVrH8FOFRkGXR7XktQHt0HZjN0PR7/DLRFg2DH7/m+fTD7zJcgzK90PpHqjICY7vnz5a6y0nyZcQbqptQ6EoygRgJnAM0AM3qaq6/vi1Taqq9vJblMdJG4rgdfQoXHAB7N9/cmziRHj1VUhsKi2aKnJhyQD3uhdFB2MPQER6IKIKbpY8KFirJazGNGg1Hv74OxT/CZdu0n54N0RFW2BxLzjlwGSi2sHIX3yzA9CcBSvHnjwIOrI1DPseYs6VNh1CBFBd21A8BvRWVTVbUZR+wIeKojymquqXgPyJFi5atIBffoG1a7W6r1GjoG3bJpR8gfbD9vTk68S4rczv4TQIYXFQulNbcizfB0uHakt2ANk/QNubAhpendjK4M8nXZMv0L6+kj+9n4BVFsP6O08mXwCmg7DiMhi1RlqdCBGkakrAQlVVzQZQVXWdoijDgW8VRUknoAfWiWCVmgpjx2ofTVKIHpr10JaeTqUzgj42MDEFO4cVcldoy7Wnq65IP9g5LFoC5El5pvffz26GIx6+f6ZM93MzhRBBo6YasNJT67+OJ2PDgLFAFx/HJUTDY0iCfu+4F1v3fkWb6Wlg8vNh3Tp44gmtpUh2tg/KikIjoeU4z9fSG2gmHxYHaWM8X0se4v33U6j+zM4QvfffTwjhFTXNgD3CaUuNqqqWKYpyCfA3n0YlREMV1w0u36Wdg1dZAq2vAUPzBndQdGEhPPoozJ59ciw1VVtibtXKi2+kKJB2OSQNhrxfTo63ngjRHb34Rn4Uoodz7oSsr7RathO6PO6bYnR9HLS7DXa/5jqedD7oGsluUiEaoZqK8PcDbwMzVFW1Hx9LAWYAHVVV7eu3KI+TInzR0KmqNrMEWn1csNZH79gBnTu7j0+dCi+/DBHezictx7TluZKtEN8XItIafusOyzEo2wem/ZDQX/t6wpr56L3ytN22e2eBsxJajofeLzfcTQxCNBJ1LcLvDTwPbFYU5X6gGzANeAG4wetRCtHIFRTAokUwcyY4nXDXXXDVVcG5UWHjRs/jK1dCWZkPEjBDsvaR2N/LDw6gE19T0kA/vFcS9HgOOv9V+zw0Sk4SECLIVZuAqapaBNxxPPlaChwFBqiqmuWv4IRoLCoq4LXX4OmnT45NnarNND3zDERFBS42TzpWs/rXubMPki/hrrIInHZt1kw5ywNLdEYwGn0blxDCa6r9k60oSpyiKG8DNwOXAPOB7xVFudBfwQnRWBQVwYsvuo+/8QYUF/s/njPJyIDzz3cdMxjguee0xrtBy5KvNSG15AU6krqpLIacZbBqHPx0Cex+o+F+LUKIGtW0BLkJeBO4+3gN2BJFUXoAbyqKclBV1Yn1eWNFUVoCHwCpgBOYparqK/V5phDBymZzPysToLJS+wg2SUnwxRewYAHMnw/t2sFDD0H6WfSStVggNFT78BuHFUp3wPq7oGA9xHWFPq9Bs54N51gj1QnZi7Xu+Sds3ATHVkG/tyE8PnCxCSG8rqYi/PTqlhsVRblNVdV36vXGitIcaK6q6iZFUaKBjcCVqqpW2/zH10X4FRXaD4+YGNDpfPY2TUpxMZSUwIED0LIlxMVBvI9/juTlae9bWqo1iI2Ph/Bqdun7S26udizTrl2u4y1baq0eUoO0V6aqQnk5hIWd+XuYn6/tkpwzB5o3h/vu0xI2v6yKmQ7Btx1PNnEF7QSCy7ZBTJDsprQWaoeMKyFajVbYab3hKnLhh35gPuR+79gDWnd7IUSDUlMRfrVLkDXVetU3+Tr+jGxVVTcd/+8yYAeQVt/n1oXFArt3a0XRY8dqRdK5uYGIpHEpKNCWrNq0gWHDtFmUhx8+uQvQF3JytML2Dh2gTx/t15Urtd/jQEpJgY8/dq31Mhph3jxIDuJj8hRFW3I8U/JVVAR//Stcfrk2c/b669Cli3YYu18c+Ng1+QKtE/3OmeC0+SmIGpgOw89Xw4JWsKA1rLtNS7hcOMF82PP9lUG4Ti2EqJezrO70LUVRMoCewFoP16YqirJBUZQNeXm+qYXIzIRu3eC992D1ai1JGD0ajh3zyds1Gfv2aXVPp06yzpmjzZL4gsUC//63diTSCeXlWlJdWOib96yNbt20GbCFC+Grr2DvXujdG0KC4k/hKewWMB+F7KWQv06rqTqDoiLt9/ZUNhvcf79vE+4qlmr+sFrz3I8E8jdLHqweB7k/aZ+rDjj0OWx+yPWIqpBwSPZwer0uQtvlKIRoVAL+V7+iKFHAF8ADqqqWnn5dVdVZqqr2UVW1T1KS9/8SKiuDv//dvQ5n0yYtgRB1N3eu5/E5c3wzI1VUBJ9+6j5usbgv/QWCXq8tiV5xBVx5pfbfgV4adeOwQt5q+KYd/DQSlvSHJQOrP1rnuKxq5su3btUSMZ9rc73n8Xa3up9M4G/2cij0UDpx8FPXBCw8Hvq95dr/TAmBAXNA76P+YUKIgPFnmawbRVH0aMnX3OOHfPtdRUX1P5x37YKBfmjh01ilVbOgnJbmmwLtkBCtxiw72/1aTIz3369RqizUlspOXc4zHYAN98PA993rlo5r3Vpbrjy9pLRfPx8nmbZysJWAEgpj9mvnIlYWaPGX7IAEj6UX7uwmLfnUx0KIlwtAQ8Lhgi+1mrSj30Hmh+CoQDtS97RvWPQ5MHorFKzTWlGkDIeweAiV9hJCNDYBmwFTFEUBZgM7VFV9KVBxxMTAhdU01ujfiHpCBsKkSRB52gY0vV4rzvZFApacDI895j7evn31yWBjU1QEhw/Dnj1aIuqo7eqb5RjY3Cai4ei34PCwjfO4uDitBuxUkZHwyis+3HRRWQR739aK7ysLYOMDsKgrLB0Ka66D+F4QeoaeGXazduj3b7fCqjGw5/XqlzPrwpIH+/4H6++A327WEsVh32uJXvpYrRj/VEoIGFMhfQy0vREiW4E+yJrECSG8IpBLkIOB64ELFUX5/fjHaH8HYTDAI4+4n293xx1a4bSou5QUrd7roou0RHfQIFizRlt68wVF0Wr33nlH+/0MD4err4bly4N3l6E3HTum/X/burW2+aBnT+37X6s2F/pqpgoNKZx2NKyL2Fj4v//TdnTedx+88ILWZLZDh1p9CbVjOgSbH4b0q7SZpSMLT16z5ml9tCoLan5G6U5Y1B0OzdPOotz4AKwe753eW3aTdjzQn//QkrrKQtjzJux6Bfq8AX1er3ZGUQjR+FXbhiIY+bINRU4O/PqrNnMwapTWHiChgR9FFyyKirQ6rLAw/3xPHQ6tFYWqarMwTWH50WbTdpw+9ZTreGSktsP3rJNeaz78fC3kLncd7/uWVk/l7eW5+tj8f7DjPzDoI/jjcc91ahcuhdSLPN9vLdKWW0//WgEu3QzNetQvvoqjsLCd++5MFBibKW0lhGgC6noWZJOSmqq1LxDe18yL9cO5uWC1akuY8fHaDObpdLqmMeN1qoICbRfv6UwmbSbqrBOw8EQY9DHsfAkOfqwtlXV+FFqMDq7kC7TaKABHpXYMjyfVjQM4rVBezU6b8gP1T8BUh4fkC0DVjhkSQjRpAd8FKcTZsNlg82YYPlxbYuvYUWtxUXCGFaamQlGqL3b3lKTWyJgC3Z+Bi9fDRcshY3JwdmHPmKQVuB/+XJudO52xUUz6zgAAIABJREFUOUS1rf7+0GhIHub5WrPu9Y9PZ4TEAe7jUe3ca7+EEE2OJGCiQcjP1zrJ79ihfV5eDk8+CYsXBzauYJGUBNOmuY+npkLbGnKQaunCtWJwQ7KW3QUjQwqMXK01NDW2gK7/AH2cdi15CIxYdbx2rRr6SC3RNDZ3He/8t5Oza/URnggDP3JdajQkw5CvtF+FEE2a1ICJBuGbb2DMGPfxPn3g++8hMdH/MQWbggJ4+22tAL6kRNv08N572i7QYM2hvMKSB6odQvRa13vVqc0+ne2sXUUO5K/Rlh1bXAKGVO/O+FXkgCVXW5I0Nofw5DMv51YWa8uUYc2Cb+lXCHHWpAZMNHj2akpm7Hb33lONlt0MKNX2hEpI0A7MvvlmcDq1pccmsZGkvl3ijanQcpx3Yqnu+cazLEqsLIHiLbD1Ga3NRpsboPVE6YQvRCMkCZhoEPr31xKK0zvo33VXE5j9shZC0SbY/TooYdDpIe2A6bA4t5eGh2sHYdckNxe2b9e61w8erCVpsbXohuBwgNkMEREN6NB6p03b4el0gC4sOJcAVRWOrYRVY0+OFW7Udmn2n+3aIV8I0eBJDZhoEBIS4McfTzZU1evh3nu1nauNennNbtZ6Ry0fCVkLtILzJQPgyDd12kl35Ih2usOFF8INN2gHpL/7LpR66L3qydGjWruLCRO0TRCeTh0IOtZC2PsOLOoGC1rC8hFQuBkcQXBI96msx+D3v7qPZy3Quv0LIRoVqQETDcaJ/l4mkzbTExsL0WdodN7gmY9ovaScVtdxQzJc+rt7AXlNjzLDAw9ojWpPpSjagfStq2tLZTeBrQS7XeXjTyN55Ik4co6fz926tdY/z2XWzenQGqGCVsOkC/CBl0e/hxWn9XgOjYTLd0OEj7oC10VFNizIAKeHzrkXr4OEvn4PSQhRPzXVgMkMmGgwTvT3atcO0tObQPIF2pmBpydfoHVWd9ZuBqekxPOuUVWFLVuquclyDDY/AgvbEfpNSya3vY51K7KqkrWDB7UNElWs+bDnDVgyCL7vCVue9O7RPrVVWQQ7XnQft5sgb5X/46mJzqid/Xi6kHAwNpGztIRoQiQBE+I0drt2MkJ2tjZrFFChESdbK5wqqq32g7kWwsLcj9w6weNZmfYK7SidPW8cbyiqosv5jpb7x/Lfl08e1fPrr8c3QjissPtN2Hg/mDLBkgM7XoC1t2nLgIGgOrW4PLF7apLqoxgqcrQPT+dsnhAWpx1PFH5qwb0C/d6SI4uEaIQkARPiFPn5MGMG9OoFnTrBww9rResBE5ao/QA+9RzGEL1WlF3LQvKEBPjXv9xr5nr10mYU3diKtdqp0xVtokfXsqpPL730+DMrC2HnDPfXH1kI9vJaxeo14Qlwzu3u44oOUi/0/ftbiyDzQ61ub0EG/HqzdoZldaLawugtMGSB9ns8Zj+0HK8tmfqaw6Ydn5T9A2T/qC2JSsd+IXxGdkEKcZzVCq+9Bs888//tnXd4VGXaxu+TSSWkkEog9N5lQURBBJGiImLXxcLa1gJiXXXtXeQTKxZE14bKAnakSUdAQUR6hwikd9LLvN8fN7PTziQzybRknt91zQXzzpkzb84B5uYp92Nee/dd4PBh4IsvfGTpYAjlGKBL9gNp8xn1an8VDUYb0H0wYAAHoj/xBCN8V1wB3HknkORIy+nVIwH0tAJw5pk0yOWa0XGEx1EUyvYYVcuonztpcxHQfRpw8F36hYXFA2d/SqNUT5O5Atg8xfz8xNdAwR/A2I361hRaENdTdUzvPEVlHoVWaRqwarRZLIfEABesBVoN8N5eBCGAkCJ8QThNRgZHHJ06Zf9aWprj9J0vMBqBoEbErwsLgaoqIDaWqUldqoqB324D/vqv9XpkR6T13IRdh1rjzDMtxFtlPrD+clopWBLemsOtHXlhVRcDJUcZPasuAbrdAcQNcq/tQkUuUFvGR3A003yGEPedX4/ybGD1GPp62TJ2M5Bwlmc/vz6UAkoOAZumAL0eBLbdD5Qesz4mpjcwerV/2nYIQhNAjFgFwQmU4ogjPSqdCOB4mrIy2kDMmUMBddttdLlvyLDzWJ2yMjtCo4FBrzMVlbOeay07A+ctRofoZHToYXN8WBww9CPg51FA2ek0W0g002lhDoxElRHIWguss4j4nPgG6PNvDgEPcUOnRXkG8MdDFJKGCKDbVKDnvYDB0+amtXS016Pawbo3qcgEVozgr6Fx9uILAIr2sBFEEAS3IwKsGXHqFC0aWrbkw12UljLb1cLNmSFXyclhPdbJk6zPio0FoqPdd/6ICGDMGGD5cuv1lBT/6LjcupX+XbXM/uGDD4CXXgKmTnXv/bYiIgUY8TUjU6qGA6zrmg/ZsjMw7lfaZxgr+Dw03vE4nYoc4Pdp9ut7XgG63dV4AVaRC2y4Csj5hc+N1cCeF5nq7P80YHB1UrkLhMYBHa6x78IMCgVi+nruc53l1CGKL4A1caZRTpYYIgDNw5FCQQhQpAi/GVBTAxw6BNx6KzBiBHDHHcCRI47H9zhLfj6wciXw978DN9wA/PILIy++IDsbuPxyoF8/YPx4oFMn4MMPaa3gDLW1jB4tXMgI0pEj9uajrVpR1HTrZl6Li6PNQmPd9ouKeD0bSlYW8M9/msWXiaeecv4aNJiwBKBlRyCqKxDhRO1ZRGsgfhCQOIwCzuAoxwmKutI0/fUanVxwXVQXU3AZLS5STYlZfFlycDYtKjyJIQzo+QCQeK55LTgSOHeRe4Z9NxbLyFb6YqDTTfbH9LhHd+KCV6gupZA/dYQdpMba+t8jCE0IiYA1AzIy2Mlmql06eBBYvBjYtcuBvYATlJdTqDz6qHnt668pUG68sY66IQ9QU0O39g0bzGtGI3D//RzQ7cwYncOH6QBvKYI++gi45hrryF779sD69Yy2VVYCbdoAiYlAcAP/phQWAn/8we7D8nKKqHHjXBd0tbXAvn3261VVFJINvc8+JygEiOkDFO22XjdEsFbLGaqLgaK9wM6naQDb4e9Ap+sZqXPklVZb5twQ0Zpypgs1Q8PqoCKSKbiqCznnsUVbIKQVEOzByJuzxPTlda4tB/a9xn1GdgCOzePP230a0P5K9zdFOENlHrBnJrD/NTaCRKQAZ38OJJztcBaqIDQ1JALWxDEagblz7QvHCwvZuddQ8vOtuwFNPPwwkJfn+vlycigUc3LqP9aWggJgwQL911avrv/9+fmcGWkbgbrzTv2IXnIy0LcvMGgQ048NFV+1tcB33zFtuHw5hd3117MD0dWolcHABgFbQkLcm4b1OuFJwFkfMC1nyd9ecz7ykreVNg8ZSzk78Y8HgF/+zmhYSBTQQsdjo82E+q0dyjOA7Q8BSwcDq8cBJ35wXNNVF+GJQFQ3IH4whYQ/iC+ATQ6jlrE+z1h5egZlMNdGrwS63ua7+ZMZy4G9M8xduOUZwJrxNPoVhGaCCLAmTk0NcOCA/mv79jn3n3w9KioYsbElP59RF2cpLwc2b2ZtVdu2wNixfK53bkeEhlIU6eFM5KeyEli7Vn89Pd35fbhKdjbwyCP263Pm6Hda1kVyMvDOO/bDr5980rVB2m6jpozpoeKD9I6qqWAXZN4WYOs9wK4XmFp0xn4idiBwyQHgjBlAn8eAi3cDHa51LvJSkQP88aD9etZKGCtyKfBG/mQ9sil2AHDm7LrNTSuyKLoOzObPV7CdjQLZfuae3xgMYYwoXfQnMGE/MPEoPdMi25+2OfHR10NlPgfP22KsBrLWeH07guApJAXZxAkNZRrtq6/sX7v22oYPqm7Rgr5XttGuDh04h9FZ0tOBc88116Nt3846tX37gM6dnTtHTAzwzDMco2MpKNu2BQYOrP/9QUGsGTt40P61xtZ21UV1Nf43M9ESo5GNDa5y1lnAnj30JisqAm6/Heje3YMF+I6ozAcOzQF2PUOH/OAoipzMn7lmYvfzwNhNQKsz6j5fcDgQ3AHorTOIuj5UjX73HoCirAyUlvREapvewPhtrPkKCmFnZn3pxPJMoHCn/fqOJ4CEoc3HliEomOLUhZmiXkFz0LThqJlDEJogEgFrBgwbBlxyifXaVVcBZ9TzvVcXCQms97L0mgoOZt2UQ9NOG5QC3nvPvhmgutp+IHR99OwJrFlDEZKczFTexo2cDVkfSUnATJ1xgFde6dn0XU4Oa/NsiY5uWFdlZCQF16uv8voNHcomAVcwGhmZy85ueHQURbuBPx89PZ4IQG0phc3u562Pq60AttzJeh5PYYgEknUc7TUDSrWuGD4cyMw2sDEgphcbCZwRT5UOcuUVWf8zoRU8RFgc0OsB+3VDBJB0nv26IDRRJALWDEhMpDBKTwd272anYOvWjYvuhIQAF1zA9ObChRRfV1xBMeOsAajR6LjmKyfHNTPRqChGzn78kQIuKsr5yI+mASNHsoj/+edZ9zVlCrsqXRUwzlJRAcybB8yYAVx2mdlfLCgIePvtxrnqN9SANTcXmD+fothgAKZPZxODS3uprWBazpLgKMfCJP93x2767iA0Ghj4f+x0rDCHG4u7vY73/xOLtDT+nXBGqFsR3UvfliF1Eh3iBc+SeC5T0rtfYJNFVDcW4TvykxOEJogIsGZCQgIf/fu775xRUXw8/HDD3m8w0Brjk0/sX7vlloYJiYaKypgYRgrnz6eAa9WqcU7y9RESws949VVgxQrg119ppDpyJH91JY3rDk6dYr3Yu++a126+GXj6ad7fcKfrwoNYO2RJzSnWDEEDYBNWi+kLaB7+ZyayA2rG/oGiI1uBigxUtTofny1IwPMvM8zYoFmeobHA2Z8BG69nmhOgK3zfJ33TFRhohMUBPaazm9VYDQSFs6NUEJoRMopI8Ch5eRQhr7zCrkCDgYXp993no9mKXuTkSc5eLChgR2VYGDtB169nd6W399Kxo306ODKSUc42bVw4WcF2YIlN8V3/F4CKDOviaS2YswQTz2notp2mpobdwJ9+CuzYYa6x0zTg6FHWLrp+0lLWjRXtoRdaRFsRAYIguISMIhJ8Rnw8Iyx33cW6o6Qk1kA1aesEJ4mLA7ZtY+fo5s1Mu958s/fFF8CUqJ4xb2mpa12tAIDITsDwBcDv09kh2LIzu+liegNtLwEOz2WdVffp9L3yAsHBTPXOn28WXyEhwPvvNyLNHBzJh56NhSAIQiORCJjQpCkvp8DZsYPp0q5dG1Dv4wFycoDXXmO9V3k56+defVXfNqOmhvVZQUGs52to52pdnDjBSQZr1wKffWauSevbl5Gj7t1dnClprAUqswFjDT28LMcT1VaeHm3j/f/f5eTwz0N2NrtsY2J80CUqCIJwGomACR4nK4tdiuvXs87prLNokREb67l6p8pKYN064NJLzcOyU1O51qmTZz7TGSoqKLZmzDCvzZ8P/PUXmwCKi4FzzmE0MDubUZpPPqH1x4MPAhdf7L70bHEx8OefnGhw8iQbK1as4Hip/HyKxDvuAD7/3EUBFmRwbF1gWyPmRRIT+dAzrRUEQfAnJAIWYJSW8ktZ0yiOnC++dkxWFt3e9+wxrw0cCMyaRR+siy7yTMoxPZ1DuW1nOo4fD3z5JX8+R+TlsRjfYOAXdk0NoyfV1RSOrnR72pKRwYiSKcpkycaNFF9DhwKLFgH33MNfLZk1iwO2Q9wwA3nDBvqwWdKzJz+zoICzJFeupM/ak082/vMCktqK03MlNSC0lU8FqCAI/kVdETDxAQsgsrJoPdCpEwdOP/54w0YD2fL999biC+D8wz17gDfeaGAXmhPk5dmLLwBYtsyx035lJc1gL7mEcx/HjWNqbt06dpB26MCZkWvXuubWb0tZmf569WlXg82bgd9/159i8MILZgPc2lraZlRUuL6H/Hzgscfs1/ft4/D2q66i+AL8I23bJKnIBrY/AvzYC1jcB9j5LN35BUEQ6kEEWICQn0/rhw8/pAgpKWGa7KmnGubKbqKqiiktPbZuZSTI0RzHxuLIzDQ52XEdVWYmBdamTRRDf/zBlGlICMUOABw7xpFJDRWOLVrQX8uW1FTaQURGAvfey+cffcT0oGX6zyS+srKAt95icfm0aRRNlU5M9jFRXe141NKJE/SLu/NOntvWyFdwgqpCYMtUYP8bQHURUJUP7HkR2PkUUK0T/hQEQbBABFiAUFoKLF5sv/7hh64PhrYkJIQpNT369AHS0oCIiIafvy6iophutOXpp5lW1GPePPtoUk0NC9MnTbJeW7iwYfuKiWHxvaULfps2tEh49VWayZrqwM4+GzhyhGumPY8YQZPaSy6hXceaNSyU79uXlgrOEhurLwQ1jSnjadN478vKWIumF030JUVFFMOLFgFbtnCPfkXNKeC4zh+Swx9SkAmCINSBCLAAobpaf72qil/2DUXTOIuyXTvr9S5dWIi/eTMd5z1BfDyL1598kmnVAQOAL75gUb7t0GoTerMZAab5IiOt11y2Z7CgbVtgyRJGrXbvBj7+mOnAs8+m6P30U/7+wguBn38Gnn0W+Pe/mQr8+GMK1y1brM9ZWQk88YTzg7zDwljU36ePeU3TmBY+fpwC74svuJ8zzgCWLzdHAX1NURFNYzt14sioIUMYlXR0/3yCsQZ2xrMAnf9VI/5SCYIQEIgACxBMcwRtOe88pswaQ0oKnd5nzKD4efZZCoxnn6W4SHbRu7KqioXsW7eyXqmuyEdSEuvavvySBe0ffEBBeOiQ/qzDG2/UP8/EiYw0mdA04OqrXdu33t66dGGjQJ8+/IzLL2c6eMMGYMwYFsTPn8806IQJ9A3r2JE/tx5797pWm5aSwjqv334DvvmGEaXLL+d9suWee/wnypSby9SsJX/+Ccye3Thh7FYMLYBonXbL+CGcWygIglAXSqkm8xg0aJASGobRqNTevUq1a6cUpYlSPXsqdeyY+z6jtlapU6eUys1V6q+/lMrMdP0clZVKrVmjVHS0eZ8DByp18qT+8UajUp98Yj7W9EhJUSo93f74vDylnnpKKYOBx2maUnffrdRvvykVEcG1uDilFixQqrjY9f3XRW0t97RqlVItW1rv98knlUpLMx+7a5f9zwQoNW2aUuXljdvHwYP65was9+BL9O4poFTfvkplZfl6dxYU7lHq6xSl5oGPbzsqVXzI17sSBMFPALBVOdA0YkMRQBiNjHAUFDBFFxPjenTK02Rk0MPJNs02ZQrrqmzThNnZwOjRwK5d9udat87eggFgrVNxMQvUW7emRUZoKNOQZWX8jPh4rjWE2lqmyn76id5bEybwvBERjEhNnsy9WRIcDBw+zM5MgFGyRx5hRM9Eu3a0sUhtpDF7ejojb7bXuEMHpoz9oSNy9WrWqdkyfjzw1Vf8s+sXKCMHkVfkAFrQ6ZFFfnABBUHwC+qyoRABJviU0lKKqG+/pSi86CKmLj/7zPq4qChg/377MT6ZmRyyfeSI/bmXLqXNREPJyqJFxdatFHn9+jGtWBe5uXzfsGHWzQ0PP0xfsSuvpAeYXk3er7+y1glgJ2R6utngtlcvpin1nPRdpbKStV8332xeCwqiYBwzxrNDyp3FdA0PHzavaRpTqYN1/ymrn6Iinve77yiwx43jf0CCxY5aEAQPIU74gscpKTHPPIyPZ81TfZGUmhqKi4kTzY0ADz7Ieq7jx61rslq10reWiI9ndMzWRDQigoKpIZSWMhp2wQXmWqyXXqII++ILxyIsKwv44Qd2T9p2ls6cSTH355+MPu3caf9+UzSyrIz2E888Q8E5YAAL+rOzOVOzsZMFwsJYBzZkCG0wwsN5Ddu08Q/xBfBarF3L675kCSODL7/Ma9cQSkt57+66y7wWFUU7EssmBcHHGGvYXWqIAAxucIkWBH/GUW7SHx9SA+aflJYq9dVXSgUFmWt12rRR6siRut+Xnq5UYqJ9nU+HDkp9/bX12jvvKFVTo3+erCylJk0yHxsfzzorV2ulKiuVOnBAqZkzlXr+ef0apBUr9N9bU6PUjBlKff+9df2a5eOrr5S68EKlFi+2vlaAUlOmKJWfz3OdOKFUeLj9+6OjHdfCNYbSUqUyMpTKyXH/uRtKXh7/fGRn82G6Ng3l5EmlQkPtr+m557JmUfADyjKU2vWSUitGKPXrHUqdOqJUTZWvdyUIjQJ11ID5yf93haZMYSEjKJZ2Funp9LCqy2OsrEzfiT8tjfVOLVqw3mn2bHYkOrKWSEqilUJaGiNWO3cCw4e7PmbpxAlGm3JzGa3SY/ly/fXCQqZRjx93HKVp2xZYtYr7/fNP4IYbaEOxcCEjZCYz1ooKfef74mLHdiINwWhk6nbyZKB3b6Z/161z3ubCE1RUcFLBpElMu15zDSOLtrV/rpKZqd89+csvftRVGciUZwErRwF/PgpkrwMOvQf81A8oP+HrnQmCx5AUpNBojh/XFwzLllFkOSqYDg9nqtDWViE2lunLQ4eYdkxMdCy+TMTF8dFQKipoo1FebhZRS5bYH3fmmfrvDw+nfcQHH7CG7bLLrD21LruMoqtDB4rL5GRgzhwKKltH/xYt+LqtE3/nzu4dbJ6VRa+23Fw+37KFdWbbttEXzBdkZHBPJlG0ejVTpXv38to1FEfDzbt185+0a0BTsA0otvFeqSkF9rwMDHpD0pFCs0T+6REajSPh065d3V9ucXGcR2nL888zSpSSQiFWn/hyBxUV5rmM33wDXHut/Zd2t250qdcjMpJ1aLt3sw7s55+B669nofecOfQqAygoamspNDRNf5xSUhKbECyLw0NDaTqr17WalcWo0apV7Lp0NIfSlvXrzeLLhFK8/nqDxAHWUmVkMKLUkPmUdWE0Au+8Yx+RKi/nz94YoqOBm26yX3/tNf/rBA5ICnWKIgGgaDdQ4+QfaEFoYogAExpNq1YsWLflxRfr7hqMiADuuIPdihddRGf21auBv//deQuI3FymOzMzG+fiHhVFuwiAX/j33MOuwOnT6cD+yitMz9X1Zd2hAwdsZ2ZSWF58MYvcb7yREaWLL6YIaNuW6bVXXrEXQAAFp6kD8L33OIbo0CF2/9k2IqSn89oPHMgmgc6dzZHH+nA0JzIrSz8tl5XFGZadO1OgrljhXmf62loKSD1OnNA31nWWVq2A//s/jqIaPZrdqL//zuss+AHJo/TXW48DQqK9uxdB8BaOisP88SFF+P5LVhYL13v1UmrYMKWWLlWqoMD59xcX08TVWaqrldq5U6khQ1hMnZqq1BdfuPaZtmRmKjVggLlAOyKCxf+ZmY4bAPQoKuJ7ysrMaxkZSnXqZF8E/sYbrp3bkpISpW67zf6coaEs5K+Pffv0mwXefZemsZbk5Sk1YQINbD//XKk5c5QaNUqpMWN4rxtbJG9i2TL9Pa1f757zK6VUYSGvneBHlOcotf5qs6HtPCj1fTcW5gtCEwZixCq4g9paWiHU1DBClZTEiExuLiMmQUEczg04rrlxF47MRNeudZwmdIbsbNYb7d7N6Ehqqnt+lj/+sB7ObaIx5qcZGWwa0GtkWLUKGOUgqGCisJDp0UcfNTdQXHgh0322w8yPH6cVxH338Z7PmmX9+ltvAbff3nDzWhO5uZyJaWlAO3Uq12w94IRmRkUuUHIYyFwJxPYF4s8CIiQ/LDRt6vIBkxSk4BTFxezyGzyYX8Rnn80Uzo4dTB+2bcsC9Z9/9k7N1urV+t16L71Ud+dlfSQlcT7mXXdR3LhLSDqqlyora3hqzWBwnOJ1piEhNpYp4GPHOC9y3z7O8LQVX4C5+/Lyy9mVassTT+inU10lKoq1c7/8Qj+4jRspvGbPdlyXJjQTwhOAhLOAvv8GUieK+BKaPdIFKTjF0aOsmzGRlsZI2KhR5i/GEydYG7R+PW0gPElxsf56SUnjasE8RceO7Aa1FYevvUZxs3AhX+/bl9EwPdNZWxITgaeesh8a3r+/8xG16Gg+2rWr+7jISEYca2vppG9LYSH/PDSWvDzWAlZUUETm5fH6hIUBd94JtGzZ+M8QBEHwByQCJtRLdTXw5pvWa4MHM3WmF5V48UWzQDJ5aj3+OPDf/7qvaHv8eH2RctttjbOjcBfFxUxnmsRKQgKL4y0jVm+8QcuJzp2Bq65isf/AgfpjlWw5dYq2FiEhwH/+Q+GWlAT885+0z2hMZ19NDe+TZWNDcjIjoEajfoSsWzf3WGRUV/PaVVXx802Rt8pK93ddCoIg+BKJgAn1YjTae3WFh9OSQI+SEn6J5+ZSEHz9NbvQBg9mWmn2bI69aQyJiRwtc+ut3IemcbbhRRc17ryNJS+Pw87feYddghddxM7BlBT+/H/+SfEUHs5r1K+fdcQuK4vpz/nzmSJ0xJYt7H5UiinT6dOBLl0Y/WpM2jQ3lzVg777L53fdRcPYxESge3fex48+olGqad9hYVxzx4Ds8HCga1d2fVrSujXFqiAIQnNBivAFp/jlF+u0YkQEBca559rXMH36Kb+0t2yhqeaLL/LXVav4RX7BBRQLjY2YVFRw/mRhIWuHoqLqFi2epKCABqYzZ1JYXXcdf74pUygo1q41R6Xy8+kZdskljOTZEhTEdK6jovOcHF7DHTvsXztyBOjUqWE/Q0kJ8Mgj9jVe06bxHprSf6YJBj/9xAjVkCHA22/z3j76aP0Dy+tCKQ4lHzXKHPEKDuZnjRolg7MFQWhayDBuodH06kWn+Kee4hejwUChMW8e036lpRQOt9zCTjqAvlkPP8yIz/nnm88VH88hyN26NW5P4eGMpDU2mtZYqqoYjZs61by2ciVw//0UL6+/zgJ3kwBLT2eEafJk/fOlptZtYFteDhw8qP9aZmbDBVhREbsibXn/fd5HkwBr0YJi9+hRCvMHHjDXf5WVsUPSdnRQRQVFan4+39uypX6quKaG3aIHD1Kw19QAY8YwhSviSxCE5oREwASnKSvjl2hZGb9g4+IYsSgo0I9CrVwqcCkLAAAgAElEQVTJ42bM4Jdqdjbw1VdMc116KVNd7khb+ZqMDKBPH14HS8LC2BV67rmcVXnttXTbX7OGdg5vv83Izk8/Wb/v009pRmvqJjVZRAQFMUW5eDEF38qV1u8LDmZHY9u2Dfs5jh5lPZqj1zp2ND//6y/90UChoTzWUhSXlponA5hqBq+8kmlaUz1ZZiZrzNauZaRr4sSGWXMIgiD4ExIBE9xCixb6dTgREfpRqDPOoCjp35/pyg4d+CU7axZtJOqaE9lkMNYgrkUuNq82YtNv4Xj25bj/FdFXVprrpIYPpzi65BIKLwB47DFej4EDOf4oJoZ+V8OGUXwVFVHomFKCDz1E8fbCC8DHHwNbt1p3Vb7yCjsa69yukdGo8HD7KFuLFqzzMo1kMtGzJ++xJY46HquqrIeyA4x6XXGFda3bwoX8OadNYzpz+HA6/wMU6bNmWadtBUEQmhsiwASPUVtLi4Q//uDz9euBBQsoxiorzaatviQri6Jm/XqmSc85h/suLqYgiomxT6f9j8p8IG0+wnY9je4V2eje6TyMXToXl07ugi1bNCQmUpA8+iiL1++/n7VdRiM907ZsYXpt0iQW6ptGCQGsrfrxR0aNTISF0RD16FGmgpcupU9Wfj6FXY8e+rMlTWRkMGW8YQPNaq+7zrrOLDmZ92fkSHM0Ly6ODQG2QigyUl+sjRxpL9I3btS3Bvn4Y9bIffONWXyZ2L+f3ZxTpjj+eQRBEJoyPhVgmqaNB/AGAAOAuUqpl325H8G9HDhgFl8mKitpvzBzJut6Gkt5OcWSqS7JFbKyKLr27OHzvn1ZrzVrFgVQaCijTbfcQgFlf4KVwNa7zM+z1yKleDg+fX87+p7ZGu+9R5EyeDAjWtnZPOyOOyh01q9nnVNMDNNulqnD3FzzAG8TJ05QZAEUjcOGsbkhKoqfc9ZZjn/WzEya56al8fl339G9ftMm61Rfnz6cAmCqMevWTb+oPjkZ+P57Dhs3nbNPH6aVbWu7HN0X0/ratfqvr17N2ZnOeKIJgiA0NXzmA6ZpmgHAbAAXAugN4DpN03r7aj+C+9m/X3/9wAH3pB5PnGBabtQoflHv3euaV9R335nFV9euFBozZpi9p6qqeP6//tJ5c0UOsGeGznoW2sXuwbFjFCcdOlC8RUdzKDdAITZqFCNDw4czFdm1q3VXaEUFLS0s+fFHRstMwtVopIDau5ceYnWxaJFZKJk4dozXwBKDgVGxESP4SElxPNmge3d6we3bR8G2ciWnJNgycKC+2H70UdYLOhqZNHq0iC9BEJovvjRiHQLgkFLqiFKqCsBXAC714X4ENzN0qP76mDGNN0s1RXRmz6YA+eYb1pwdP+7c+6uqWBhuYvRo+pVZEh7ODsDSUvpcbd5sjmJBGYEa/dk4kSElSE21T12OHQtccw1/bzTSmDUxkdG7jAyLc5/+bFvRUlVFQ9utWykMR46knUV9syRra9mtqMeGDfY1W86iafzcHj0oIB3VayUn8/NNMzrbtKF57Nln8/mll1LMWdKrV/2iUhAEoSnjSwHWFoDl1+WJ02tCMyElhQOaLenQgbVQ4eGNO/eGDYyAWVJVBbz8sr1pLIw1gLHaaikkhPVeJior7QXT++8zDTZsGNOQZ5/NCFR2NoCweKDTTfYbCwoF4gbp7jkxkZ1/hw4x/Xj8OH3Abr2VImbsWNbHFRdTfL31FtOSs2bR0mPdOuDpp/G/1Oann1KQ1WfDYTA4FjPjxtVteeEOgoK432++oQXH77/TJ86U1m3dmmnIDz/k+iefMDUrXZCCIDRnfGZDoWnaVQDGKaVuPf38BgBDlFLTbI67HcDtANC+fftBabZ5FMGvyctjCm/5coqMoUPd88X60kvsGLTlvPPYWRgbC6C6BCg7Dhx4G6guBrr+E4jpDYQx/JaRwbqp48fpTfb110yHGY0Uiu+9Z/Y0s+TNN4HLLgNSWmXD8PsdwIlv+EJILDD8KyBxBBAcYf9GGzIy2CFqO8R6yxbWjZ06xdduvZWCxGDgyKK77gImTGBac8sW565nVhYjjzt3mtf69+d98adOQ6PR84JQEATBW/irDcUJAJYjgFMBpNsepJSaA2AOQB8w72xNcBfx8XwMHOje844fry/ALr/8dHG3UkDuJmDNeKYLAeDY50C/Z4Ge9wEhLZGSQtf1zz9niqysjJ13U6cynblmjf5n//QTIznduiXhhmvmImTQ60BNKRAaC4TGA4ZQp36GjRvtxRcAPPssuxUrKpiy3LKF65rGSF1WFqNEd94J/PYbPbPqIzmZ0bV16xhtGjmS/mT+JL4AEV+CIAQOvhRgWwB00zStE4CTAK4F8Hcf7kdoQnTuTBuG9HRGcRYtosi75hpGimpLs2HYcqdZfJnY/RzQ5WYghLbuKSl0cp86lUXwQUFMD4aH035Bj/btmYacNQsYNy4Obds2rKAtK0t/PSeH6dTSUrP4at2anZPLl/NzO3Xi/ix9wOojOZkRtKuuatB2BUEQBDfiMwGmlKrRNG0qgGWgDcVHSqndvtqP0HTIz2cB+4wZFCmTJ7MLr0UL1lnl5ABaeQ0SSg7bv9lYDdSc+t9TpXi80Ugn+cREChWTOWhKClOFJlq0AP7xD9ZOVVVRAFraR5jmU1ZU0Lw0IcGx35mjuqzrr2d9lKW4eu892lJs28bnmzZRkJkEmiAIgtC08GnAXyn1k1Kqu1Kqi1LqBV/uRWga1NRwDM/f/w78+ScL2p95hkamBgPTiM89B2zfEQxE97Q/QVAYEEwDquJiphPPOYeF7OPGsUA8O5tdkWVlrAu74w56hF15JYXf88/zvYB5lA5A89I5c1jr1qUL0K8fvbJOnbLfhum9b71lPePwootY35Wby3mJffqYi+xN4stEdTW7IE17EXxPdjbF8dy5HJZuayUiCIJgQmZBCk2KzEwWj+fk2L+WlkYx06UL05E/fLgO8dtHA8pibs7fZgFdbgdCIrF9u31tWng4C9V796bVQ2oqI13nnMPxOZ98wi7LKVMoyDp2ZAQsKIh1ZMOHW59P0+iTZWuzALCeKz+f5z9wgJ+9dy/wr3/RAHXJEoqrxx4Dhgxh96gtPXuyVs3farmqqykiCwsZNWzZkrWAzZmsLDZnbNpkXrvuOhoPWwp1QRACh7qK8KXkVWhSGI364gtgxEopCptNm4CX5gxGxZj9qO35b9R2mYrK87ehPGUKEBKJigq68dtSUUFz0vPOA158kUX6AwcynThpEnDxxRQ85eV8PnIkPcJychjNskUpFtTbkpPDyF2/fhxLtHYt/boeeICpx61bgc8+YzfmW2+x6UCP0aPtLT3y8ihGDxxg+tTR3EZPYbr+fftSyHbsyLSqpc9Zc0Mp4IcfrMUXAHz5pf24JkEQBEAEmOBB8vKYIty8GTh5kvVajSU8nOLIllat6K7fooW5tmrk6Ba4/MbOuPalF3DTW2+h098GYtS4VsjOpihxlLorLmbEBmDE67zzKJQ2bmT68Y47gA8+YPTqyBHgttvYBHDmmfrns3WSP3UKePhhWmnk5VGMvfACzUm3bOGcx9BQpj9NnmDJyfajidq3p29WukXvcHY2o3MdOzIVOmAARWRVVX1X1n3k5TGVmp9vXlu6FHj1VYqz5khpqb2Rr4nvv/fuXgRBaBqIABM8Qk4OhUC3bjQw7dSJ0YDG1ivFxdGw09L7KiwM+O9/meJq1Yri6OqrKUaWLGHqcN48RoN+/ZWdhC1b2pvEmrj6ahbaW0aWhg9nFOfUKVo/2PLCC3R0tyUoiPVqlhQVMbply7x5FC/Hj7PDsXNn8x7i4ii2Vqxg3dfcuTRi/cc/KArLyihuZs3iyCITOTmMnhUU0Lj2gw/YvLB/v2sdlK5w4IC+2P78c2tR1pwIC9NPMwOMAgqCINgiAkxwO7W1FFuWQqC6mpEidxQld+pEIfX99xQhv/7Kui/T3MB27Sg0bNNBJr77jmLl7LMZVTJ5T4WHU5wcO8bXDh/mOfbtY3QjOdlxQX1uLtOUH35oHrPUti3TUikp1sfW1uqnBWtrmWL96COa1z7/PM9pYsUK1hRt2MC6opEjWTOWkcEIV3k5zVnPOMP6vKmpjED26kXR+cgjrB374gv3RCVtsZ0oYKJly+br8xUSwkYQ2589JUVGKgmCoE8z/edQ8CUFBYx26GE5f9FZiospgp55hmmstDSKk7vvpmnpwIGsNzp50vye6GjH5q9nnMEvzIQEfmlu3Mh9rVzJlOKECazBatOGzv09epiLqFNSThu92jBpEj/z+uuBXbuAo0d5jnHj7L+UW7RgI4EtffuaB3//97/2tV2m7shVq6wd7Z95hsJw0ybudf58epmZ0qh33w1MmwaU2IyuvO8+z0TB2rVj7ZotDz3UvIvR27YFtm+nLUr//rzmv/5qL8AFQRAAAEqpJvMYNGiQEvyfwkKlxo1TiqXJ1o9vvnHtXGVlSv3nP9bnCA1V6ocflBoyxHr9vvuUqqoyvzc9Xak2bayPiY9X6sQJ8zHPPsv14GDr4y65RKlTp+z3U1Gh1MqVSkVGmo/t1Uup48dd+7n27VMqNdV8jrZtlVq7Vqm+ffl89GilCgqs35Ofz/1qmvl9Dz6o1M6dSiUmmtdatlRq+XKlfvqJz9ev178XgFLbtrm2b2c5elSpsWO51/h4pV59VancXM98lr9RUqJUVpZS5eW+3okgCL4GwFblQNP40glf8HOKipgyXL2a6bfBg/mrKdXniJgY1iktW2a9npjI2YuuUFBAl3pLqqpYxP7oo6yLMnHwIF8zGZ+mpLBe6623mLYbMoRRH8uIhClKZJsSjIy0L54HWOszbBg/68gRRr2Skly3gejenQX3GRmsVVMKePBBRs8ApgljY63f06oVcM89wI03ss6qXTvuf+JE687QkhLWhq1aRVuE1FTW4h08aH0+TfOcNUTHjsBXXzEtqml1G9I2NyIjHadhBUEQ/ocjZeaPD4mAeY9Tp5R65x3raElyslJHjjj3/sJCpRYvVqp3b6UiIpS6+GKlDh1Symh0bR/79jmO3qxZY/38k0/0z1FZqVReHqNXtqSlKRUWZn/u3393bZ8NpapKqR07lBo40HyN585ltMsZjh51fH127FCqqIjX3BQNs3zccot9lE0QBEFwH5AImOAqRUWMtliSlUWfqo8/ZuSnLmJiaEUwaBALyyMi7CM6zhAZSXNV2whVaiqjYyZGjnTslRUaai6MtyU5mR5ct9wC7N7NqNJrrwFdu7q+14YQEkKLi2XLGL0zGBgtCnbyb2ZICK+RbTF9UBCjY6b7NHw4Xf5ffpnmqLfcQg8xV+9JcTE/KyzM8TUVBEEQ6kcEmKDL4cP6nXorV/ILuD4BZqKxDu0xMSzefukl6/XXXqPH1SefcFxP+/YNK/AOC2NadNUqdmoaDEwpmrr1KiuZhj1wgEKnY0e+VlXlXhHiaO/5+UwpFhebbTYsi/Pj4phWff556/ddd511B2VUFPC3v3GmZG2t69eqtpbdoQ89RMf/Hj1oedG7N5sK3E1uLtOXQUH8s6bX+CAIgtCUEQEm6GLps2VJp07OR2fcQVQUo27nnQe88w5F0IMPMkIVHc39NITaWtZe5edT0ERHWw/VBijINm2i+31ZGXD55cBdd7HrcPt2dlO++SbtHcLC+J6KCkbmjh6leExMpKBrCDk57GBcsIDPIyMpOMeNM9euRUTQSqNNG+6lupr+a7fcYn0PCwq451df5c9+zz2siXO2Biwri8ebfLyys2nVsWsXxZg7OXKEAvK33/hnbfJk4JVXGn4dBUEQ/BJHuUl/fEgNmPfIyVFqzBj7uqHly+t/b2amUhs2KPXFF6wZKyx0z55KStzTWVZdrdSWLdadg4MHK3XypPVx6elKxcby9chIpTZtsq8XCwtjHZZS3NuyZUq1aGF+fcAA+/M6Q02NUm++af1ZPXoo9cYbrFvLzrY+vraW1z0jgz+fJZWVSr39tv29fO45pUpLeV0zMuquB/voI/06sylT3Nvtl5mpVNeu9p8zfTo7YgVBEJoSqKMGTHzABF0SEujl9eqrrOMaN47RoPq6GNPTWY81fDgd4Dt3Bt5/3z1+U5GR9t5YDSEnB7jwQuvOwa1bmV6zNFrNz2e9FMAo2IIF9qN0KiuBOXPMx192GaNlJv78E3j8cdcNT/PzOZrIxIQJwOzZ3EOPHqx327jR/FlBQUz3tm5tH6HMy+MebHnxRV6DO++k39m119JfrKLC/liTP5ktJ0+6d8xRYSHHV9kyd67nnPsFQRB8gQgwwSFJSUxvLV1Kc8+hQ+uu/TINuN63z3r94Yfd44DvLvLyWGNky4IF1kLJsrYpKkr/PQBFp9HIGilL8WXCNNNRj4ICvm/jRo4gMokMg8Fc9xQaSgE1cSLtNCoqgG3bgBEj+J76qKoyC0lLystpavvZZ/x12TKKbb1zTpqkf+7rr3e+HtAZlNJfNxrd9xmCIAj+gAgwoU5MXXkxMfUfW1QE/PST/mtbtrh3X43BVK9li22hd1QUcMEF/P369YyC6TFlCsVZmzYc1G17/oQE/RE8ubksoO/Uid5iHTsCr7/O6FdcHPDvf/O4Cy4AFi+2F3e1tTy+urqun5b76djRfj0piSLMkupq1lvZRsHataPvmiUTJjCS6E5iY/X3OmUK70d6OvDtt/QYO3HCfv8+x1gLVOYB1SX1HysIQkAjAkxwG6GhtIfQo1077+6lLmJi9McUPfCAdVejKQ370EPsRDQYmK6zZOpUCqHUVNpJhIZyRqVlqvSxx/S7DtevZ1G9CaMRePppc7pvyBDOumzf3tpyw5KsLP1uVUuSkmgdYmmEajAA775Lk1pbjh2zTyvGxQH/+hcL5JcsAfbsYYrU3aOFkpM5LL1vXz7XNDY/PPccP7tXL6Z5r7uO6e1ffqlfgHqNimxg/xvA2gnA5ilA4Q6g2gPDNgVBaBZoylHM3w8ZPHiw2rp1q6+3IdTBr7+yO87yj1Xv3rR5aKwlhS01NYwWhYa67meVnk7BtWiRudPy9tspumyprKQA0jR+VlER03Rt2zIleOut1iLg7rtZr7ZwIVOHl15qb1dRWkoR8cMP9p/3yCNm243aWqZM09IoyGz58UfHkTlLystZ7/XDD7xuEyfy3EOH2qeHP/iAXZT1TTzwJNnZjPiZUrHV1cCYMaypsyQxkWs+n7dYkQOsuwzI/cW8pgUBYzcB8To3ThCEgEDTtN+VUoN1XxMBJriTU6eAvXuZPjtxgl/0tuN/6qOgwFyLFRlJ7ytbsrOBDz9kKiohgdYQ/fo5lyq13GtJCYVGXBzFlSPKyviIiTFHktLS9NNl4eHA/v1M/SUm6qcfq6poafHhh/avvfGGvQlufj6jVc8+a66Huv124IUX9EVjXZSXU3Tt3Mk9lpRwKPnRo8D55wNfful/lg/HjzMSqMeBAxy15FMK/gSWnGG/nnQecO7XQJi41gpCIFKXABMfsAAkM5NfWoWFLLpu1cp9ZppRUYzULFzIyFFsrOOaKz3S0xl9Mc2RHDeOIqVNG/MxubnA1VfTwd7EqlV8z9ixru21PoPP9HT6Ue3ezTotAOjZk92GjlJ/lZWM3NQV8QsNpZ/Zxx8zEmWiZUum3GwxGa7efDM7D1u3phjUE6d1UVrKKNgNN5j336sX034VFdyzu9OK7iA4mD+rbSrWYLA2nPUZpWn66yVHAKMb20QFQWg2SA1YgHHsGHDmmTQ2vfRSRnBWrNC3HmgMsbH8MndFfOXnc4j00qVmB6ilS7lmMgAFKMAsxZeJhx+2tpZoLCdPUnRddhlTiaNGUVgWFjIi06KFvhHsuHHODWNu1451YIMG8VwjRwKbNzsWbtHRfM/QobxvroovgHu/8UZr8bh3LwvvO3f2T/EF0DD2ySft12+91U9c8mP7668njwaC/WGDgiD4GyLAAoiiImDaNKYGTdTU0K/LUZG3NyktZSTGluXLre0hMjP135+WVn9BurOYLDWOHbNenz2bhe8jRrAY/YcfrBsP+venL5gzNWmRkayXW7qU3leLFnGskmWxfGPJzqZ/2xdf8GfZs0e/aP277/zbZys0lMJx0SLgnHM4Vun99zmCyZW0s8cIbQX0e8Z6LaIN0P8ZIMQJNS4IQsAhKcgAoqzMnNqzXU9P94NCZifp2lV/QPfYseYRPQAjaFlZFBzBwaxrMhic+4ziYkYG9di6lRGZW25hevTXX3l8SAijMa7WT7law+UsmZmMxu3YweexsfpF/4C+gau/ERfH9Ox557EOLj5ev77OJ4TGAN2nAR2uBjJWAJHtgfizgAgHM70EQQh4/OWfL8ELaJp1LZUlrnYRegJTGs6WkSOta9RiYxmJsqRtW2DGDHM6qqwMWL2aEab27RkxWbTI+ShPeLjjwu7OnSnsAFpUBAWxLqxLF/8pXq+uZiTOJL4Aph+zsxmls+W55/w3/WhLfLzj5gafEtYKiO4J9JgGpF4q4ksQhDrxt3/CBA+SlMQvWlsuuMA/0jjx8XRlHzHCvDZiBNcsh0a3bMmxOUeO0Edr2TJGpTp0MB+TmcmImCmFmJkJXHMN3+MM0dHsMLRNB55xBiNFpjSu5egid1BaChw+DDz1FH23du9uWGqwsJBpRVvuvZeO/9OmUbQOHEjfMstrLgiCIHgeP086CO4kKIju5d9+S3FRWAhMnkxzUU+lwVwlNZWjeyxtKCzFl4noaD70iuABijbL7kITr79On6u6LCdMdO7MCNJzz9GiYfRozmC8+mrzMVdc4VzBvTMoxSL8cePMe585E/joI3qGuTIHMzycgnTbNuv148f5mDGDzvbBwa5HvkpLKTzDw/0jcioIgtAUEQEWYLRqxe7HYcP4JR8X596ib3cQH68vupylqkp/JiNAiwhn5wpGRDC1OGcOz1dQwPojUxPA3/7GYeWmtGd+PsVJTQ1FmavpyKwsdvXZCsfp0ynKHKWP9YiKojfa999bn69zZ7rMR0S4bt+gFF36n3iC3Zu9elHIdevmniHpgiAIgYQYsQrNjowM+pzp1ZOtWUMR1RBqamhzkZdHwREdbRZZGRnATTeZC/f79we++YaCx1mcMRutrOQeVq+mIDr/fEYv9QRQWRlw8CBNcY8cYfTuoYdcE3KWZGQwBZudbV4LDQV27fIDI1RBEAQ/RIxYhYCivJzWFW++Sf+u4mJGex54wDVBZEtwMDtFbbtFi4tZW2XZNbljB0cErVnj/Aim4GCKo/R06/WICDYhGI2sdRs9mkIMoABasQIYPty+KL1FC2DAAFpQVFSwzq8xkaoVK6zFF8Bo48yZvNYSBRMEQXAeKcIXmh0REUwbbt7Merc1a1ioX1vrnnRrebm1cW1JCTssbdm3j685S1IShYwtTz/N1HF2Nj3bTOILoAC6/np7YWRJTAxFYGMFkqMGBr3h3YIgCELdSARMaHYkJFCAXXEFoz8A/b9++KFxVgsFBexKfPNNWnrcey9rxDSN59cr+nfFKsFg4MDp7dv5GeXlHOzdqxejWVlZrMGy5fhxHutpJk5kXZkt11zjJ270giAITQgRYEKzIySEQubgQQowg4EioXVr541YbamoYGfl9Onmtf/+l0Lv2ms5W9F2sPbAgdbGsM4QHc204XvvMeVoOcopNJRCzLbBICLCtZFPDaVDBw4Jt4zSnX8+cMklFKGCIAiC80gRviA4QXo60KOHfUoxNpZRseBg1ph99RWL9c8/n4O227Vz7XMKCmjxUF5OMZaQYE6blpez6/KJJ6zfY/IMc9dA9fr2l5vLGreuXVmz5isD15wcRv9276bYTUlpXPesIAiCu6mrCF8EmBDwFBSwkD4/n7VSMTH23l5HjtDpXo+jRzkc+9QpnkcpiqG4ONf2kZ0N3HEHuycBiq9Fi4CzzjJHuPLyOIj8jTf4OffcwyHhgSY8srPpYffzz+a1667jdWkqjv6CIDR/6hJgUoQvBDS5uUwrduxIX6/OnYEvv6SQsiQiQv+LvU0bc3F7VBTd5VNTXRdfVVXAu++axZdpb+PHU3SZiI/nPMRvv+XjyisDT3wBwMqV1uIL4H3bu9c3+xEEQXAVEWBCQLN4MWu7TFRWArfdZt9VmJAAvP++da2TwQDMnUvhdeQIU4H338/0XGGha/vIz+dYJVvKy+mzZUurVq6LvOZCeTnr7/RYuLBh58zNBX75BbjrLuCVV5jatB32LgiC4E6kCF8IWAoKgP/8R/+1xYutC+5Nhf0HDrCwPyiIKa82bYAtW+jNZfrCfu014K23gJtvdr4uS9McO9N7o7arKRES4tj4tWdP189XUECrD8sB7889x/vakPMJgiA4g0TAhIAlONjxEHK96FLLliw8f/JJGrx26cJU5e2320dL/vUv16JgiYnAgw/ar6ekOK49C1SCg2nPYVunl5AATJrk+vkKCqzFF8BmiwcfdD2SKQiC4CwiwISAJSoKeOwx+/XoaEa0nKGmBti/3369vJxF+c4SFESfrZkzmV4EgBEjgHXraJ8hWNOmDfDHHxyM3rMn8I9/cEpAQ67VwYP667/95h1/NUEQAhNJQQoBTY8eHFj9wAPAoUMc6fPee86PDzIYWMB/7Jj1emwsIzTr1wM//gj06cMUpu0YI0vi49nVOHkyOxzDwgKzwN4ZTGnIuXPpi9aypX1EzFkcjafq31/GKwmC4DnEhkIIeJRi0b3RSLNTV0SP0QgsWUIzUsu/Sr/8Arz+OrBggXmtbVuOR0pNdd/ehcaTnw9Mm2aemgBQ/P72G0WYIAhCQxEfMEHwIPn5wMmTdIgvKWEUKzQUGKzzV+7++4EXXpDIir+Rk0Nx/OmnQPv2wNSpTHN6Y8KAIAjNl7oEmKQgBaGBlJTQfuLFF4HqanZN9u3LAv5339V/z5o1rA3zlADLzOSeiosZvWnVynF3pWAmMZFRzLFjWeTf0JFVgiAIzm66fewAAAsBSURBVCICTBAayLZtdKE3Gvn8669pkjp3LtCvn/57+vf3nK3EsWMs3D9+nM/DwoDvvgNGjpRIjrPIdRIEwVtIF6QgNICcHKYTTeLLxNdf07m+e3dgyBDr11q25BzHhhaL10VhIU1ETeILoKns1VdbO+kLgiAI/oEIMEFoADU1ju0L0tOBpCTghx/onn/hhRRru3a5PpzbWcrLgWXL7NeLi+nyLgiCIPgXkoIUXKa8nOaVJSWM5kRH01MrkIiIoGXFTz9ZrxsMQKdO/H1SEscaXXcda75CQjy3H02jdUZGhv1rgXZvBEEQmgISARNcoqgImD+fKbYePSg2Xn458NJcsbEcOWQyTTXx4ot8zYSmUQB5UnwBLCJ/6in79bFjKZAFQRAE/0JsKASX2L9ffz7ekiXA+PHe348vqa0FsrKAVauYdpw4kVEoW1HmLfLzgeXLgZdeYupx8mRaYiQl+WY/giAIgY7YUAhuY/58/fV33gHOPdczBeb+isFAr6jrr/f1TkhcHHDttcD557M5IC6OfmSCIAiC/yECTHCJli311yMjxTvJX5CIV8OoqDAP346JEf80QRA8i9SACS5xxRX6UZUHHxR3d6HpkpMDPP000+tduwIPPcT0siAIgqcQASa4RHIy3dx79+bz1FRg4UIORhaEpkhJCfD448CMGWwyKS0FZs8G7ryT3b6CIAieQIrwhQaRnc3xOwYDO/Ak/Sg0VU6eBDp3BqqqrNc1DUhL85x3myAIzR+/K8LXNG0mgEsAVAE4DOAfSqlCX+xFaBhSZ+RejEaK2poaWlYkJ/t6R4GD0WgvvgBAKf4nQxAEwRP4KgW5AkBfpVR/AAcAPOqjfQiCzykupmv+kCGMtowYAaxbR8Pb5sqpUxwcXlzs652w2H7oUPv1Hj0Cq6tXEATv4hMBppRarpSqOf10M4BUX+xDEPyBY8eASZPMcxwPHAAuuIACpblhGuE0ZQpFz0038ef1ZaQpIQGYN4/F9yZSU4Fvv5VIpCAInsMfbChuBuDAXQrQNO12ALcDQPv27b21J0HwClVVwOuv269XV1MUPP649/fkSTIygEGDGAEDWGP188/A7t2AL/96d+4MbNhAM1ujEYiPlzS7IAiexWMRME3TftY0bZfO41KLYx4DUANgnqPzKKXmKKUGK6UGJyYmemq7guATjEZ24enhaL0p8/HHZvFloqQE+Ogj1lz5kuRkoFcvoE8foHVrIEh6xAVB8CAei4AppS6o63VN024CMAHAaNWUWjEFwY2EhwN33w0sWGD/2o03en8/nsRoBA4d0n/t0CGOdgr2h5i8IAiCF/DJ//E0TRsP4GEAE5VSZb7YgyD4C/36Ac88Yza4bdECeP99oG1b3+7L3QQFcT6lHpMni/gSBCGw8FWQ/W0AUQBWaJq2XdO093y0D0HwOXFxwP33A0eOAPv2sUj9hhs4Dqe5MXAgcN111mtXXQUM1nXJEQRBaL6IEasgCF4lL4+eZ3v2sOYqOZlF74IgCM0NvzNiFQTBt5SXc8zOsWNAbCytGLzV9Rcfz0evXt75PEEQBH9E+nwEIcAoLwdWrAC6dAGGDWPX38UXN0/fMUEQBH9FBJggBBj5+ay7qqgwr23dCrz0kvWaIAiC4DlEgAlCgLF3r/7sw4ULmZYUBEEQPI/UgAkCOJOwuJjCJDKS9VCa5utdeYZWrfTX4+MBg8G7exEEQQhUJAImBDw5OcD06UDHjua6qJ07aQzaHElN5aBpW554QsbvCIIgeAsRYEJAU1UFvPMOR+SYBNfhw8CoUbRKaI4kJ3P+4uTJQHQ00K0b505eUOfsCkEQBMGdSApSCGjy84G5c/XXjx4FUlK8vydvkJoKvPceMHMmHeqbc8pVEATBHxEBJgQ8joYuN/d6qJYt+RAEQRC8j6QghYAmIYHDsG1JTgbat/f+fgRBEITAQASYENAEBwM338wi/LAwrg0YAKxZQxEmCIIgCJ5AZkEKAoDSUqCwkIX4ERFAYqKvdyQIgiA0dWQWpCDUQ2QkH4IgCILgDSQFKQiCIAiC4GVEgAmCIAiCIHgZEWCCIAiCIAheRgSYIAiCIAiClxEBJgiCIAiC4GVEgAmCIAiCIHgZEWCCIAiCIAheRgSYIAiCIAiClxEBJgiCIAiC4GVEgAmCIAiCIHiZJjULUtO0HABpvt6HH5AAINfXmxB0kXvjv8i98V/k3vgvcm8aRwellO504SYlwASiadpWR8M9Bd8i98Z/kXvjv8i98V/k3ngOSUEKgiAIgiB4GRFggiAIgiAIXkYEWNNkjq83IDhE7o3/IvfGf5F747/IvfEQUgMmCIIgCILgZSQCJgiCIAiC4GVEgDVRNE2bqWnaPk3Tdmia9o2mabG+3lOgo2naeE3T9muadkjTtEd8vR+BaJrWTtO01Zqm7dU0bbemadN9vSfBGk3TDJqm/aFp2o++3otgRtO0WE3TFp7+rtmradrZvt5Tc0IEWNNlBYC+Sqn+AA4AeNTH+wloNE0zAJgN4EIAvQFcp2lab9/uSjhNDYAHlFK9AAwFcLfcG79jOoC9vt6EYMcbAJYqpXoCGAC5R25FBFgTRSm1XClVc/rpZgCpvtyPgCEADimljiilqgB8BeBSH+9JAKCUylBKbTv9+1Pgl0hb3+5KMKFpWiqAiwHM9fVeBDOapkUDGAHgQwBQSlUppQp9u6vmhQiw5sHNAJb4ehMBTlsAxy2en4B8yfsdmqZ1BDAQwK++3YlgwesA/gXA6OuNCFZ0BpAD4D+n08NzNU2L9PWmmhMiwPwYTdN+1jRtl87jUotjHgNTLPN8t1MBgKazJi3GfoSmaS0BLAJwr1Kq2Nf7EQBN0yYAyFZK/e7rvQh2BAP4G4B3lVIDAZQCkNpWNxLs6w0IjlFKXVDX65qm3QRgAoDRSvxEfM0JAO0snqcCSPfRXgQbNE0LAcXXPKXU177ej/A/hgGYqGnaRQDCAURrmva5Uup6H+9L4L9pJ5RSpmjxQogAcysSAWuiaJo2HsDDACYqpcp8vR8BWwB00zStk6ZpoQCuBfC9j/ckANA0TQPrWPYqpWb5ej+CGaXUo0qpVKVUR/DvzCoRX/6BUioTwHFN03qcXhoNYI8Pt9TskAhY0+VtAGEAVvD7BZuVUnf4dkuBi1KqRtO0qQCWATAA+EgptdvH2xLIMAA3ANipadr202v/Vkr95MM9CUJTYBqAeaf/U3kEwD98vJ9mhTjhC4IgCIIgeBlJQQqCIAiCIHgZEWCCIAiCIAheRgSYIAiCIAiClxEBJgiCIAiC4GVEgAmCIAiCIHgZEWCCIAQMmqa10zTtqKZpcaeftzr9vIOmaUs1TSvUNO1HX+9TEITmjwgwQRACBqXUcQDvAnj59NLLAOYopdIAzAT9wgRBEDyOCDBBEAKN1wAM1TTtXgDDAbwKAEqplQBO+XJjgiAEDuKELwhCQKGUqtY07SEASwGMVUpV+XpPgiAEHhIBEwQhELkQQAaAvr7eiCAIgYkIMEEQAgpN084AMAbAUAD3aZqW4uMtCYIQgIgAEwQhYNA4uf5dAPcqpf4CC+//z7e7EgQhEBEBJghCIHEbgL+UUitOP38HQE9N087TNG09gAUARmuadkLTtHE+26UgCM0eTSnl6z0IgiAIgiAEFBIBEwRBEARB8DIiwARBEARBELyMCDBBEARBEAQvIwJMEARBEATBy4gAEwRBEARB8DIiwARBEARBELyMCDBBEARBEAQvIwJMEARBEATBy/w/8HFpECtq5v0AAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#plotting the data \n",
    "plt.figure(figsize = (10,6))\n",
    "sns.scatterplot(x = 'X1',y = 'X2',hue = 'y',data = data,s = 50,palette=['blue','orange','green'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\Lenovo\\Anaconda3\\lib\\site-packages\\sklearn\\svm\\base.py:193: FutureWarning: The default value of gamma will change from 'auto' to 'scale' in version 0.22 to account better for unscaled features. Set gamma explicitly to 'auto' or 'scale' to avoid this warning.\n",
      "  \"avoid this warning.\", FutureWarning)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "SVC(C=1.0, cache_size=200, class_weight=None, coef0=0.0,\n",
       "    decision_function_shape='ovr', degree=3, gamma='auto_deprecated',\n",
       "    kernel='rbf', max_iter=-1, probability=False, random_state=None,\n",
       "    shrinking=True, tol=0.001, verbose=False)"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "svm = SVC()\n",
    "svm.fit(data.drop('y',axis=1),data['y'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0, 0.5, 'X2')"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize = (10,6))\n",
    "plot_decision_regions(np.array(data.drop('y',axis=1)), np.array(data['y']), clf=svm, legend=2)\n",
    "\n",
    "plt.xlabel('X1')\n",
    "plt.ylabel('X2')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# LAB-9.6.5 Application to Gene Expression Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(64, 2308)\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "      <th>10</th>\n",
       "      <th>...</th>\n",
       "      <th>2299</th>\n",
       "      <th>2300</th>\n",
       "      <th>2301</th>\n",
       "      <th>2302</th>\n",
       "      <th>2303</th>\n",
       "      <th>2304</th>\n",
       "      <th>2305</th>\n",
       "      <th>2306</th>\n",
       "      <th>2307</th>\n",
       "      <th>2308</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>EWS.T1</th>\n",
       "      <td>3.2025</td>\n",
       "      <td>0.0681</td>\n",
       "      <td>1.0460</td>\n",
       "      <td>0.1243</td>\n",
       "      <td>0.4941</td>\n",
       "      <td>3.1207</td>\n",
       "      <td>3.7106</td>\n",
       "      <td>1.8416</td>\n",
       "      <td>1.2607</td>\n",
       "      <td>2.9001</td>\n",
       "      <td>...</td>\n",
       "      <td>0.7653</td>\n",
       "      <td>1.6679</td>\n",
       "      <td>0.1493</td>\n",
       "      <td>0.6918</td>\n",
       "      <td>1.4151</td>\n",
       "      <td>0.2756</td>\n",
       "      <td>0.1521</td>\n",
       "      <td>0.3175</td>\n",
       "      <td>0.7240</td>\n",
       "      <td>0.2044</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>EWS.T2</th>\n",
       "      <td>1.6547</td>\n",
       "      <td>0.0710</td>\n",
       "      <td>1.0409</td>\n",
       "      <td>0.0520</td>\n",
       "      <td>0.2045</td>\n",
       "      <td>2.1609</td>\n",
       "      <td>2.4452</td>\n",
       "      <td>1.1473</td>\n",
       "      <td>0.7371</td>\n",
       "      <td>1.9989</td>\n",
       "      <td>...</td>\n",
       "      <td>1.0665</td>\n",
       "      <td>3.6014</td>\n",
       "      <td>0.3048</td>\n",
       "      <td>1.7957</td>\n",
       "      <td>1.0701</td>\n",
       "      <td>0.2688</td>\n",
       "      <td>0.1932</td>\n",
       "      <td>0.4140</td>\n",
       "      <td>1.2708</td>\n",
       "      <td>0.2990</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>EWS.T3</th>\n",
       "      <td>3.2779</td>\n",
       "      <td>0.1160</td>\n",
       "      <td>0.8926</td>\n",
       "      <td>0.1014</td>\n",
       "      <td>0.2818</td>\n",
       "      <td>1.9773</td>\n",
       "      <td>3.2590</td>\n",
       "      <td>1.4106</td>\n",
       "      <td>0.9548</td>\n",
       "      <td>2.0775</td>\n",
       "      <td>...</td>\n",
       "      <td>1.2674</td>\n",
       "      <td>1.5152</td>\n",
       "      <td>0.2382</td>\n",
       "      <td>0.8720</td>\n",
       "      <td>0.6819</td>\n",
       "      <td>0.3221</td>\n",
       "      <td>0.2156</td>\n",
       "      <td>0.3227</td>\n",
       "      <td>1.2142</td>\n",
       "      <td>0.2230</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>EWS.T4</th>\n",
       "      <td>1.0060</td>\n",
       "      <td>0.1906</td>\n",
       "      <td>0.4302</td>\n",
       "      <td>0.1035</td>\n",
       "      <td>0.2984</td>\n",
       "      <td>1.6804</td>\n",
       "      <td>5.8901</td>\n",
       "      <td>0.2958</td>\n",
       "      <td>0.7381</td>\n",
       "      <td>1.6610</td>\n",
       "      <td>...</td>\n",
       "      <td>0.4743</td>\n",
       "      <td>1.0282</td>\n",
       "      <td>0.1049</td>\n",
       "      <td>0.5632</td>\n",
       "      <td>1.2264</td>\n",
       "      <td>0.8123</td>\n",
       "      <td>0.2758</td>\n",
       "      <td>0.3016</td>\n",
       "      <td>0.7235</td>\n",
       "      <td>0.0871</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>EWS.T6</th>\n",
       "      <td>2.7098</td>\n",
       "      <td>0.2367</td>\n",
       "      <td>0.3693</td>\n",
       "      <td>0.2190</td>\n",
       "      <td>0.3711</td>\n",
       "      <td>1.7800</td>\n",
       "      <td>3.2376</td>\n",
       "      <td>0.6769</td>\n",
       "      <td>0.8546</td>\n",
       "      <td>0.6808</td>\n",
       "      <td>...</td>\n",
       "      <td>0.7039</td>\n",
       "      <td>0.5961</td>\n",
       "      <td>0.0707</td>\n",
       "      <td>0.4001</td>\n",
       "      <td>1.5271</td>\n",
       "      <td>0.4084</td>\n",
       "      <td>0.6412</td>\n",
       "      <td>0.3552</td>\n",
       "      <td>1.3928</td>\n",
       "      <td>0.2157</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 2308 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "          1       2       3       4       5       6       7       8     \\\n",
       "EWS.T1  3.2025  0.0681  1.0460  0.1243  0.4941  3.1207  3.7106  1.8416   \n",
       "EWS.T2  1.6547  0.0710  1.0409  0.0520  0.2045  2.1609  2.4452  1.1473   \n",
       "EWS.T3  3.2779  0.1160  0.8926  0.1014  0.2818  1.9773  3.2590  1.4106   \n",
       "EWS.T4  1.0060  0.1906  0.4302  0.1035  0.2984  1.6804  5.8901  0.2958   \n",
       "EWS.T6  2.7098  0.2367  0.3693  0.2190  0.3711  1.7800  3.2376  0.6769   \n",
       "\n",
       "          9       10    ...    2299    2300    2301    2302    2303    2304  \\\n",
       "EWS.T1  1.2607  2.9001  ...  0.7653  1.6679  0.1493  0.6918  1.4151  0.2756   \n",
       "EWS.T2  0.7371  1.9989  ...  1.0665  3.6014  0.3048  1.7957  1.0701  0.2688   \n",
       "EWS.T3  0.9548  2.0775  ...  1.2674  1.5152  0.2382  0.8720  0.6819  0.3221   \n",
       "EWS.T4  0.7381  1.6610  ...  0.4743  1.0282  0.1049  0.5632  1.2264  0.8123   \n",
       "EWS.T6  0.8546  0.6808  ...  0.7039  0.5961  0.0707  0.4001  1.5271  0.4084   \n",
       "\n",
       "          2305    2306    2307    2308  \n",
       "EWS.T1  0.1521  0.3175  0.7240  0.2044  \n",
       "EWS.T2  0.1932  0.4140  1.2708  0.2990  \n",
       "EWS.T3  0.2156  0.3227  1.2142  0.2230  \n",
       "EWS.T4  0.2758  0.3016  0.7235  0.0871  \n",
       "EWS.T6  0.6412  0.3552  1.3928  0.2157  \n",
       "\n",
       "[5 rows x 2308 columns]"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train = pd.read_csv(r'E:\\programming\\dataset\\Into_to_statstical_learning\\khan_train.csv',index_col = 0)\n",
    "train = train.T\n",
    "print(train.shape)\n",
    "train.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(25, 2308)\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "      <th>10</th>\n",
       "      <th>...</th>\n",
       "      <th>2299</th>\n",
       "      <th>2300</th>\n",
       "      <th>2301</th>\n",
       "      <th>2302</th>\n",
       "      <th>2303</th>\n",
       "      <th>2304</th>\n",
       "      <th>2305</th>\n",
       "      <th>2306</th>\n",
       "      <th>2307</th>\n",
       "      <th>2308</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>TEST.9</th>\n",
       "      <td>1.7733</td>\n",
       "      <td>0.4875</td>\n",
       "      <td>0.5832</td>\n",
       "      <td>0.3514</td>\n",
       "      <td>0.7182</td>\n",
       "      <td>4.7686</td>\n",
       "      <td>8.8864</td>\n",
       "      <td>2.2408</td>\n",
       "      <td>0.8908</td>\n",
       "      <td>3.5985</td>\n",
       "      <td>...</td>\n",
       "      <td>1.7174</td>\n",
       "      <td>0.7078</td>\n",
       "      <td>0.1307</td>\n",
       "      <td>0.2834</td>\n",
       "      <td>0.1558</td>\n",
       "      <td>0.2899</td>\n",
       "      <td>0.2239</td>\n",
       "      <td>0.3658</td>\n",
       "      <td>0.6412</td>\n",
       "      <td>0.2182</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TEST.11</th>\n",
       "      <td>0.1397</td>\n",
       "      <td>0.0846</td>\n",
       "      <td>2.3266</td>\n",
       "      <td>0.1188</td>\n",
       "      <td>0.3497</td>\n",
       "      <td>2.6046</td>\n",
       "      <td>3.9808</td>\n",
       "      <td>1.8516</td>\n",
       "      <td>1.3238</td>\n",
       "      <td>1.9454</td>\n",
       "      <td>...</td>\n",
       "      <td>0.7283</td>\n",
       "      <td>1.3666</td>\n",
       "      <td>0.2613</td>\n",
       "      <td>1.1244</td>\n",
       "      <td>0.2988</td>\n",
       "      <td>0.2298</td>\n",
       "      <td>0.1414</td>\n",
       "      <td>0.1283</td>\n",
       "      <td>0.7414</td>\n",
       "      <td>0.7103</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TEST.5</th>\n",
       "      <td>1.9420</td>\n",
       "      <td>0.2103</td>\n",
       "      <td>2.4897</td>\n",
       "      <td>0.1770</td>\n",
       "      <td>0.5963</td>\n",
       "      <td>2.9903</td>\n",
       "      <td>4.2457</td>\n",
       "      <td>2.4517</td>\n",
       "      <td>1.6298</td>\n",
       "      <td>2.2302</td>\n",
       "      <td>...</td>\n",
       "      <td>0.2924</td>\n",
       "      <td>0.4544</td>\n",
       "      <td>0.5507</td>\n",
       "      <td>0.2250</td>\n",
       "      <td>0.2284</td>\n",
       "      <td>0.6212</td>\n",
       "      <td>0.1504</td>\n",
       "      <td>1.9152</td>\n",
       "      <td>0.9166</td>\n",
       "      <td>0.1850</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TEST.8</th>\n",
       "      <td>0.7721</td>\n",
       "      <td>0.1855</td>\n",
       "      <td>1.1922</td>\n",
       "      <td>0.0979</td>\n",
       "      <td>0.1841</td>\n",
       "      <td>0.9914</td>\n",
       "      <td>9.9955</td>\n",
       "      <td>1.5774</td>\n",
       "      <td>0.7100</td>\n",
       "      <td>2.0473</td>\n",
       "      <td>...</td>\n",
       "      <td>0.1288</td>\n",
       "      <td>0.2941</td>\n",
       "      <td>0.3197</td>\n",
       "      <td>0.3858</td>\n",
       "      <td>1.3418</td>\n",
       "      <td>0.2996</td>\n",
       "      <td>0.2328</td>\n",
       "      <td>0.5194</td>\n",
       "      <td>0.9413</td>\n",
       "      <td>0.3751</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TEST.10</th>\n",
       "      <td>0.3296</td>\n",
       "      <td>0.3510</td>\n",
       "      <td>0.4258</td>\n",
       "      <td>0.0737</td>\n",
       "      <td>0.1702</td>\n",
       "      <td>0.2839</td>\n",
       "      <td>4.1636</td>\n",
       "      <td>0.4754</td>\n",
       "      <td>1.8462</td>\n",
       "      <td>0.5868</td>\n",
       "      <td>...</td>\n",
       "      <td>0.3817</td>\n",
       "      <td>0.3617</td>\n",
       "      <td>0.3710</td>\n",
       "      <td>0.5255</td>\n",
       "      <td>0.0979</td>\n",
       "      <td>0.2380</td>\n",
       "      <td>0.6157</td>\n",
       "      <td>1.6150</td>\n",
       "      <td>0.3655</td>\n",
       "      <td>0.4591</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 2308 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "           1       2       3       4       5       6       7       8     \\\n",
       "TEST.9   1.7733  0.4875  0.5832  0.3514  0.7182  4.7686  8.8864  2.2408   \n",
       "TEST.11  0.1397  0.0846  2.3266  0.1188  0.3497  2.6046  3.9808  1.8516   \n",
       "TEST.5   1.9420  0.2103  2.4897  0.1770  0.5963  2.9903  4.2457  2.4517   \n",
       "TEST.8   0.7721  0.1855  1.1922  0.0979  0.1841  0.9914  9.9955  1.5774   \n",
       "TEST.10  0.3296  0.3510  0.4258  0.0737  0.1702  0.2839  4.1636  0.4754   \n",
       "\n",
       "           9       10    ...    2299    2300    2301    2302    2303    2304  \\\n",
       "TEST.9   0.8908  3.5985  ...  1.7174  0.7078  0.1307  0.2834  0.1558  0.2899   \n",
       "TEST.11  1.3238  1.9454  ...  0.7283  1.3666  0.2613  1.1244  0.2988  0.2298   \n",
       "TEST.5   1.6298  2.2302  ...  0.2924  0.4544  0.5507  0.2250  0.2284  0.6212   \n",
       "TEST.8   0.7100  2.0473  ...  0.1288  0.2941  0.3197  0.3858  1.3418  0.2996   \n",
       "TEST.10  1.8462  0.5868  ...  0.3817  0.3617  0.3710  0.5255  0.0979  0.2380   \n",
       "\n",
       "           2305    2306    2307    2308  \n",
       "TEST.9   0.2239  0.3658  0.6412  0.2182  \n",
       "TEST.11  0.1414  0.1283  0.7414  0.7103  \n",
       "TEST.5   0.1504  1.9152  0.9166  0.1850  \n",
       "TEST.8   0.2328  0.5194  0.9413  0.3751  \n",
       "TEST.10  0.6157  1.6150  0.3655  0.4591  \n",
       "\n",
       "[5 rows x 2308 columns]"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test = pd.read_csv(r'E:\\programming\\dataset\\Into_to_statstical_learning\\khan_test.csv',index_col = 0)\n",
    "test = test.T\n",
    "print(test.shape)\n",
    "test.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Although both the train and test data are now having same number of columns, but i am still not able to get the target variable..."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Till Next time, Happy Learning :)"
   ]
  }
 ],
 "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
}