7. Getting_prices_of_used_macbooks.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 중고맥북 가격 구하기를 통해 \n",
    "- 모델을 만드는 것\n",
    "- 손실함수에 대한 개념\n",
    "- Gradient Descent를 통한 손실값 줄이기 등을 파악한다"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "슝=3\n"
     ]
    }
   ],
   "source": [
    "from IPython.display import display, Image\n",
    "print(\"슝=3\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "슝=3\n"
     ]
    }
   ],
   "source": [
    "import pandas as pd\n",
    "print(\"슝=3\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(80, 2)\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>used_years</th>\n",
       "      <th>price</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>2.65</td>\n",
       "      <td>175</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>3.80</td>\n",
       "      <td>63</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>0.95</td>\n",
       "      <td>273</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>4.50</td>\n",
       "      <td>133</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>4.45</td>\n",
       "      <td>106</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   used_years  price\n",
       "0        2.65    175\n",
       "1        3.80     63\n",
       "2        0.95    273\n",
       "3        4.50    133\n",
       "4        4.45    106"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "macbook = pd.read_csv('macbook.csv')\n",
    "print(macbook.shape)\n",
    "macbook.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "슝=3\n"
     ]
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "# 실행한 브라우저에서 바로 그림을 볼 수 있게 해줌\n",
    "%matplotlib inline \n",
    "%config InlineBackend.figure_format = 'retina' # 더 높은 해상도로 출력한다.\n",
    "print(\"슝=3\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 248,
       "width": 375
      },
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.scatter(macbook['used_years'], macbook['price'])\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- 상관계수 구해보기"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[ 1.        , -0.78972238],\n",
       "       [-0.78972238,  1.        ]])"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import numpy as np\n",
    "\n",
    "# np.corrcoef(x, y)를 사용합니다.\n",
    "np.corrcoef(macbook['used_years'], macbook['price'])\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 7-3. \"모델\"을 만든다는 것, 그 의미 이해하기"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Job done\n"
     ]
    }
   ],
   "source": [
    "x = macbook['used_years'].values\n",
    "y = macbook['price'].values\n",
    "\n",
    "print(\"Job done\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Job done\n"
     ]
    }
   ],
   "source": [
    "def model(x,w,b):\n",
    "    y = w*x + b\n",
    "    return y\n",
    "print(\"Job done\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "12"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model(x=5, w=2, b=2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 248,
       "width": 377
      },
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# x축, y축 그리기\n",
    "plt.axvline(x=0, c='black')\n",
    "plt.axhline(y=0, c='black')\n",
    "\n",
    "# y = wx + b 일차함수 그리기\n",
    "x = np.linspace(0, 8, 9)\n",
    "y = model(x, w=-20, b=140) # y = -20x + 140\n",
    "plt.plot(y)\n",
    "\n",
    "# 나의 (x, y) 점 찍기\n",
    "x_data = [2, 5, 6]\n",
    "y_data = [100, 40, 20]\n",
    "plt.scatter(x_data, y_data, c='r', s=50)\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAvIAAAHwCAYAAADEu4vaAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAAWJQAAFiUBSVIk8AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy8QZhcZAAAgAElEQVR4nOzdd3wUdf7H8dc3nRRC71261IReEgtFQQU79opYEEi8885TPO+nV/S8BEQUxYIdFWz0opJQxSRKEekgndBJIXXn98cuI8eFFpLMbvJ+Ph55TPLZmd13Hih558vsjLEsCxERERER8S1+TgcQEREREZELpyIvIiIiIuKDVORFRERERHyQiryIiIiIiA9SkRcRERER8UEq8iIiIiIiPkhFXkRERETEB6nIi4iIiIj4IBV5EREREREfpCIvIiIiIuKDVORFRERERHyQiryIiIiIiA8KcDqAtzLGbAMqA9sdjiIiIiIi5VcT4LhlWU0v9EAV+TOrXKlSpWpt2rSpVpYvmpaWZn8eFRVVli8tIiIiImXs119/5cSJE8U6VkX+zLa3adOmWmpqapm+qDHG/rysX1tEREREylZ0dDRpaWnbi3OszpEXEREREfFBKvIiIiIiIj5IRV5ERERExAepyIuIiIiI+CAVeRERERERH6QiLyIiIiLig1TkRURERER8kIq8iIiIiIgPUpEXEREREfFBKvIiIiIiIj5IRV5ERERExAepyIuIiIiI+KBSKfLGmJuMMROMMYuNMceNMZYx5sMz7NvCGPMnY8x3xpidxpg8Y8x+Y8zXxpjLz/E69xhjVhpjMo0xx4wxi4wx15TG9yQiIiIi4k1Ka0X+GWAk0AnYfY59nwf+BdQGZgP/AZYCg4HvjDGjijrIGPMyMAWoC0wGPgTaAzOMMSMv/lsQEREREfFeAaX0vHHALmAzEAt8f5Z95wIvWpb106lDY0wssAD4tzHmc8uy9p7yWC/gCWAL0NWyrCOe+b+BVOBlY8xMy7K2l9y3JCIiIiLiPUplRd6yrO8ty9pkWZZ1HvtOOb3Ee+ZJwCIgCOh12sMPe7Z/P1niPcdsByYCwcB9xUvvrOD6bQioWs/pGCIiIiLi5UprRb6k5Hu2BafNr/Bs5xZxzBxgrGefv57rBYwxqWd4qPX5BCxJGTn51LjuSfwqVebYko8pKHQR4K/3I4uIiIjI//LalmiMaQxcCWQDyafMw4D6QOapp9ucYpNn27LUQ5awl+ZuIKByTfwCg6l6+X0MfW0p6/YcdzqWiIiIiHghryzyxphg4CPcp8g8d+rpM0CkZ3vsDIefnFc5n9eyLCu6qA9gfXGyX4xh3RqSu2+z/fXa3ce57tUlvDxvAzn5hWUdR0RERES8mNcVeWOMP/AB0Bv4FHi5mE91zvPzvc2l9SLZ9348Rxa9i1WQB0CBy+LV7zcz+JXFpP522OGEIiIiIuItvKrIe0r8h8DNwGfAnUW8YfbkinskRTvXir13s1wc/2E6e94ZSbcm1ezxlgNZ3DRpOc998wtZuae/ZUBEREREKhqvKfLGmADgE2AY8DFwu2VZ/9NYLcvKwn1t+nBjTN0inqqFZ7uxtLKWhYIje5j6UA+eH3IpYUH+AFgWTFm2nQGJySRvPOBwQhERERFxklcUeWNMEDAN90r8+8BdlmWd7aTw7zzbq4p47OrT9vFZfn6Gu3o2YX58LLEta9rz3UdPcPc7K/nD56s4lp1/lmcQERERkfLK8SLveWPrl8AQ4G3gPsuyXOc4bJJn+7Qxpuopz9UEeAzIBd4t8bAOqV+lElPu60rCLR2pEhpoz6el7qJfYhJz1xZ18R4RERERKc9K5TryxpihwFDPl3U8257GmCmezw9alvUHz+eTgEHAQdynzDxrjDn9KRdZlrXo5BeWZS0zxiQA8cBqY8w03DeOuhWoBjxe3u7qaozhhqgG9G1Rk+dm/MKs1e7yfiAjl4c/TOPqdnX425BLqRUR4nBSERERESkLpXVDqE7APafNmnk+AH4DThb5pp5tDeDZszznolO/sCzrCWPMamAk8BDgAtKAf1uWNbPYyb1czYhgJt4exXUd9zH2q7WkZ+QCMGftPpZtOcTYa9pyY1R9ivhlSERERETKEfO/F4URcN/xNSoqKio19Uw3fi2117U/P9efzbET+fxj1q98mrLzv+Z9W9TgH9e3p2G10FLJKCIiIiIlIzo6mrS0tDTPfYwuiOPnyEvxRVYK5MWbOvDRg91pWK2SPV+86SADxyUzZek2XC79oiYiIiJSHqnIlwO9m9dg3pgY7u/dlJML+tl5hTw3Yx23vLGczemZzgYUERERkRKnIl9OhAYF8Oy1bZn+SC9a1Aq35ym/HWHQ+MVM/H4z+YXnuhiQiIiIiPgKFflyJqpRVWaO6sOoK5oT4Odens8rdPHveRsY8upS1u72zRveioiIiMh/U5Evh4ID/Ikf0IoZj/ehQ4NIe75u73GGTFzKi3PXk5N/tvttiYiIiIi3U5Evx9rUrcwXj/TiL4NaExzg/qMudFm8vmgLg8Yv5sfthx1OKCIiIiLFpSJfzgX4+/FQzCXMGxND96bV7PnWg1ncPGk5z369lszcAgcTioiIiEhxqMhXEE1qhPHJ8B78/fp2hAf/fh+w95f/xsDEZBZtSHcwnYiIiIhcKBX5CsTPz3BH98YsiI/hita17Pnuoye4990fif/sZ45k5TmYUERERETOl4p8BVQ3shJv39OF8cM6UTU00J5/kbab/olJzFq995x3lRURERERZ6nIV1DGGIZ0qs/C+Fiu61jPnh/MzOOxj9MY8UEq6cdzHEwoIiIiImejIl/BVQ8P5pXbOvPW3V2oUznEns9ft58rE5L47MedWp0XERER8UIq8gJAv7a1mR8fw23dGtmzjJwCnpy+mrveXsnOw9kOphMRERGR06nIi61ySCD/vKE9Hw/vTuPqofZ8yeaDDEhM5p0l2yh0aXVeRERExBuoyMv/6HVJDeaOjmF436b4GffsRH4h/zdzHTdPWsam/RnOBhQRERERFXkpWqUgf54e3JYvHu1Nq9oR9jxtx1EGv7KECd9uIq/A5WBCERERkYpNRV7OqlPDKsx4vA9x/VoS6O9ens8rdPGfBRu57tUlrN511OGEIiIiIhWTirycU1CAH6P7tWDm433p2LCKPV+/L4OhE5fyz9m/kpNf6GBCERERkYpHRV7OW6s6EXzxSC+eGdyGkED3fzouC95I3spV45JZsfWQwwlFREREKg4Vebkg/n6GB/s2Y96YGHo2q27Ptx/KZtibK3j6yzVk5OQ7mFBERESkYlCRl2JpXD2Mj4d35183tCciOMCef/TDDgYkJvPd+v0OphMREREp/1TkpdiMMQzr1ogF8bH0a1Pbnu89lsP9U1IYM/UnDmflOZhQREREpPxSkZeLVicyhMl3RzPhts5UDwuy51/9vId+CUl8s2oPlqUbSYmIiIiUJBV5KRHGGK7tWI8F8bFc37m+PT+clceoT35i+Pup7DuW42BCERERkfJFRV5KVLWwIBJv7cQ793ahbmSIPV/46376JyTxycodWp0XERERKQEq8lIqrmhdm/lxMdzZo5E9y8gt4Kkv1nD75B/47VCWg+lEREREfJ+KvJSaiJBAXhjanqkP9aBJ9VB7vnzrIQaOS+atxVspdGl1XkRERKQ4VOSl1PVoVp25Y2IYEdsMP+Oe5eS7eGHWr9zw+jI27MtwNqCIiIiID1KRlzIREujPU1e34avHetO6ToQ9X7XzKNdMWMy4hRvJK3A5mFBERETEt6jIS5nq0KAKMx7vwxP9WxLk7/7PL7/QYtzCTVw7YQk/7zzqcEIRERER36AiL2Uu0N+Px69swaxRfejcqIo937A/gxteW8oLM9dxIq/QwYQiIiIi3k9FXhzTonYE0x7uxbPXtKVSoD8ALgveWrKNgeOSWbbloMMJRURERLyXirw4yt/PcH+fpsyPi6FP8xr2fMfhbG6f/ANPfbGa4zn5DiYUERER8U4q8uIVGlYL5YMHuvHSTR2oHBJgzz9ZuZP+CUksXLffwXQiIiIi3kdFXryGMYZbujRkYXwsAy+tbc/3H8/lwfdTePyTnziUmetgQhERERHvoSIvXqdW5RAm3RnNa3dEUSM8yJ7PWLWHfglJfPXTbixLN5ISERGRik1FXrySMYZB7euyIC6WG6Lq2/Mj2fmM+fRnHngvhT1HTziYUERERMRZKvLi1aqGBZFwSyem3NeV+lUq2fPv1qczIDGZD1f8hsul1XkRERGpeFTkxSdc1qoW8+JiuLtnY3uWmVvAM1+tZdjkFWw7mOVgOhEREZGyF3DuXUS8Q3hwAP83pB3XdKjHn6evZqunvK/cdpirxiUT378lD/RpSoC/fj/9HwUFMGsWpKRAZiaEh0OXLjB4MATorwERERFfpJ/g4nO6Na3G7NF9eeXbTbyRvJVCl0VugYt/zlnPrDV7efHGDrSpW9npmN4hKwsSE+GNN2DXrv99vEEDGDEC4uIgLKzs84mIiEixlfjSpTHmJmPMBGPMYmPMcWOMZYz58BzH9DLGzDbGHDbGZBtjVhtjxhhj/M9yzDXGmEXGmGPGmExjzA/GmHtK+vsR7xQS6M+TV7Xm68d60/aU0r561zGunbCEhPkbyC0odDChF0hPh8sug7Fjiy7x4J6PHeveLz29LNOJiIjIRSqNcxCeAUYCnYDd59rZGDMESAZigC+BiUAQkAhMPcMxI4EZQDvgQ2AyUA+YYox5+eK/BfEV7epH8vXI3vxxYCuCAtz/ORe4LF75bjODX1lC6m9HHE7okKws92kzKSnnt39Kinv/LL3XQERExFeURpGPA1oClYFHzrajMaYy7hJeCFxmWdYDlmX9EfcvAcuBm4wxw047pgnwMnAY6GJZ1mOWZcUBHYAtwBPGmJ4l+h2JVwv09+Oxy5sze1RfohtXteeb0zO5adIy/jbjF7LzChxM6IDExPMv8SelpMC4caWTR0REREpciRd5y7K+tyxrk3V+d+y5CagJTLUsy24dlmXl4F7Zh//9ZeB+IBh41bKs7acccwT4h+fLh4sZX3xY81rhfD6iJ3+77lJCg9xnZVkWvLt0OwMSk1my6aDDCctIQYH7nPjieOMN9/EiIiLi9Zy+vMcVnu3cIh5LBrKBXsaY4PM8Zs5p+0gF4+dnuKdXE+bHxRDTsqY933XkBHe+/QNPTlvFsRP5DiYsA7Nmnfmc+HPZuRNmzy7ZPCIiIlIqnC7yrTzbjac/YFlWAbAN95V1mp3nMXuBLKCBMSb0fAIYY1KL+gBaX8D3IV6mQdVQ3ruvKy/f3JHISoH2/LOUXfRPSGLeL/scTFfKLvSUmpI+XkRERMqE00U+0rM9dobHT86rFOOYyDM8LhWEMYabohuwID6GQe3r2PP0jFxGfJDKYx+lcSAj18GEpSQz8+KOz8gomRwiIiJSqpwu8udiPNvzOd++WMdYlhVd1Aew/kKCiveqFRHCa3dEM+nOKGqE/36W1qw1e+mXkMT01F2c31s6fER4+MUdHxFRMjlERESkVDld5M+1el75tP0u5JjjF5FLyqGr2tXl2/hYbo5uYM+Oncjnic9Xce+7P7L76AkH05WgLl2cPV5ERETKhNNFfoNn2/L0B4wxAUBToADYep7H1AXCgF2WZWWXbFQpDyJDA/n3zR354IFuNKhayZ4nbTzAgIQk3l++HZfLx1fnBw9237G1OBo2hEGDSjaPiIiIlAqni/x3nu1VRTwWA4QCyyzLOvVE5rMdc/Vp+4gUqW+LmswbE8O9vZpgPCdjZeUV8uzXv3Drm8vZcuAizzN3UkAAjBhRvGNHjHAfLyIiIl7P6SI/DTgIDDPG2P+eb4wJAV7wfPn6ace8C+QCIz03hzp5TFXgL54vJ5VSXilHwoIDeO66S5n2cE8uqRlmz3/cfoSrxy/mtUWbKSh0OZjwIsTFXfgpMl27wpgxpZNHRERESlyJF3ljzFBjzBRjzBTgz55xz5MzY8zLJ/e1LOs4MBzwBxYZY94yxrwE/Az0xF30Pz31+S3L2gb8EagGpBhjJhpjEoHVwCXAfyzLWl7S35eUX9GNqzFrVF9GXt6cAD/38nxegYuX5m5g6GtL+WXPmS6Q5MXCwtzXkz/fMt+1q3v/sLBz7ysiIiJeoTRW5DsB93g+BnpmzU6Z3XTqzpZlfQXE4r4B1I3A40A+EA8MK+oOsZZlTQCuA34B7gYeAvYB91qW9YeS/5akvAsJ9OcPA1vx9cjetKtf2Z6v3X2c615dyr/nrScnv9DBhMVQqxYsWgQvvHDmc+YbNnQ//v33ULNm0fuIiIiIVzLl6rJ7JcgYkxoVFRWVmppa1q9rf64/G2cUFLp4a8k2EhdsJLfg91NrmtUM46UbO9ClSTUH0xVTQYH7jq0pKe7rxEdEuFfrBw3SOfEiIiIOio6OJi0tLc1z+fMLoiJ/BirysvVAJn+evoaV2w/bM2Pgnp5N+OPAVoQFqwCLiIjIxbmYIu/0m11FvFazmuFMfagHzw9tR1iQPwCWBVOWbWdAYjLJGw84nFBEREQqMhV5kbPw8zPc1aMx8+NjuazV7+eQ7z56grvfWckfPl/F0ew8BxOKiIhIRaUiL3Ie6lepxLv3diXx1o5UCQ2059NSd9EvIZk5a/Y6mE5EREQqIhV5kfNkjOH6zg1YGB/LNR3q2vODmbk88lEaD3+QSvrxHAcTioiISEWiIi9ygWqEB/Pq7VG8eVc0tSKC7fncX/bRLyGJz1N26o3KIiIiUupU5EWKacCldVgQH8uwrg3t2fGcAv44bTV3v7OSnYezHUwnIiIi5Z2KvMhFiKwUyL9u7MBHD3anYbVK9nzxpoMMHJfMlKXbcLm0Oi8iIiIlT0VepAT0bl6DeWNieKBPU07eCiA7r5DnZqzj5jeWszk9w9mAIiIiUu6oyIuUkNCgAMZe05bpj/SiRa1we5762xEGjV/CxO83k1/oOssziIiIiJw/FXmREhbVqCozR/Vh9JUtCPBzL8/nFbr497wNXPfqUtbuPuZwQhERESkPVORFSkFwgD9x/Vsyc1QfOjSItOe/7j3OkIlL+dec9eTkFzqYUERERHydirxIKWpdpzJfPNKLpwe1ITjA/b9boctiUtIWBo1fzMpthx1OKCIiIr5KRV6klAX4+zE8phnzxsTQvWk1e771YBa3vLGcsV+tJTO3wMGEIiIi4otU5EXKSJMaYXwyvAf/uL494cEB9vyDFb8xICGJ7zekO5hOREREfI2KvEgZ8vMz3N69EQviY7iydS17vudYDve9+yPxn/7Mkaw8BxOKiIiIr1CRF3FA3chKvHVPF8YP60S1sCB7/sVPu+mXkMTM1XuwLN1ISkRERM5MRV7EIcYYhnSqz4K4GIZ0qmfPD2XlMfLjnxjxQSr7j+c4mFBERES8mYq8iMOqhwczflhn3rq7C3Uqh9jz+ev20y8hiU9/3KHVeREREfkfKvIiXqJf29rMj4/h9u6N7FlGTgF/mr6GO9/+gR2Hsh1MJyIiIt5GRV7Ei1QOCeQf17fnk+E9aFw91J4v3XyIgeOSeXvJNgpdWp0XERERFXkRr9TzkurMHR3DQzHN8DPu2Yn8Qp6fuY6bJi1j0/4MZwOKiIiI41TkRbxUpSB//jKoDV8+2pvWdSLs+U87jjLolcW88u0m8gpcDiYUERERJ6nIi3i5jg2r8M3IPsT1a0mgv3t5Pr/QImHBRq57dQmrdh51OKGIiIg4QUVexAcEBfgxul8LZo3qS6eGVez5+n0ZXP/aUv45+1dO5BU6mFBERETKmoq8iA9pWTuC6Y/0Yuw1bakU6A+Ay4I3krdy9fhkVmw95HBCERERKSsq8iI+xt/P8ECfpswbE0Pv5tXt+fZD2Qx7cwVPf7mGjJx8BxOKiIhIWVCRF/FRjaqH8uED3XnxxvZEhATY849+2MGAxGS+W7/fwXQiIiJS2lTkRXyYMYZbuzZiYXws/dvWtud7j+Vw/5QURk/9iUOZuQ4mFBERkdKiIi9SDtSuHMKbd0Xz6u2dqR4WZM+//nkP/ROT+WbVHixLN5ISEREpT1TkRcoJYwzXdKjHwvhYbuhc354fzspj1Cc/Mfz9FPYdy3EwoYiIiJQkFXmRcqZqWBAJt3bi3fu6Ui8yxJ4v/DWd/glJfLJyh1bnRUREygEVeZFy6vJWtZgXF8NdPRrbs4zcAp76Yg23TV7Bb4eyHEwnIiIiF0tFXqQciwgJ5Pmh7fj0oR40rRFmz1dsPczAcclMTt5KoUur8yIiIr5IRV6kAujerDpzRvdlRGwz/P0MADn5Lv4++1dueG0pG/ZlOJxQRERELpSKvEgFERLoz1NXt+GrR3vTpm5le75q1zGumbCYxAUbyStwOZhQRERELoSKvEgF075BJN+M7M0fBrQkyN/9V0B+ocX4bzdxzYTF/LzzqMMJRURE5HyoyItUQIH+foy8ogWzR/chqlEVe75xfyY3vLaUF2au40ReoYMJRURE5FxU5EUqsOa1Ivj84V48d21bQoP8AXBZ8NaSbQwcl8yyzQcdTigiIiJnoiIvUsH5+xnu7d2UeWNi6Nuihj3fcTib29/6gT9PX82xE/kOJhQREZGiqMiLCAANq4Xy/v3d+PdNHagcEmDPp/64kwGJSSxYt9/BdCIiInI6FXkRsRljuLlLQxbGxzLw0tr2fP/xXIa/n8LIj9M4mJnrYEIRERE5yauKvDFmsDFmvjFmlzHmhDFmqzHmc2NMzzPs38sYM9sYc9gYk22MWW2MGWOM8S/r7CLlSa3KIbxxVxdevyOKGuHB9nzm6r30T0jiq592Y1m6kZSIiIiTvKbIG2NeBGYCUcBcYDyQBgwBlhpj7jxt/yFAMhADfAlMBIKARGBq2SUXKb+ubl+XhfEx3BjVwJ4dyc5nzKc/88B7Kew5esLBdCIiIhWb8YZVNWNMHWA3cADoYFlW+imPXQ58B2yzLKuZZ1YZ2AxEAr0ty0rxzEM8+/YEbrMsq9iF3hiTGhUVFZWamlrcpyju69qfe8OfjchJSRsP8Jcv1rD7lPIeHhzAn69uze3dGuHnZ85ytIiIiBQlOjqatLS0NMuyoi/0WG9ZkW+MO8sPp5Z4AMuyvgcygJqnjG/yfD31ZIn37JsDPOP58pFSTSxSwcS2rMm8uBju6dmYk79vZuYW8MxXaxk2eQXbDmY5G1BERKSC8ZYivwnIA7oZY2qc+oAxJgaIABaeMr7Cs51bxHMlA9lAL2NMcBGP/xdjTGpRH0Dr4nwjIuVZeHAAfxvSjs9G9KRZzTB7vnLbYa4al8ykpC0UFLocTCgiIlJxeEWRtyzrMPAnoDawzhjzpjHmn8aYz4D5wAJgxCmHtPJsNxbxXAXANiAAaFaqwUUqqK5NqjF7VF8evewS/D2n1OQWuPjXnPVc/9oy1u057nBCERGR8i/g3LuUDcuyxhljtgPvAMNPeWgzMOW0U24iPdtjZ3i6k/MqZ3j81Nct8nwkz6p81LmOF6moQgL9efKq1gxqX5c/TV/NL57yvmb3Ma57dQmPXHYJI69oTnCALiIlIiJSGrxiRR7AGPMkMA2YAlwChAHRwFbgI2PMSxfydJ6t3i0qUsra1Y/kq8d68+RVrQgKcP+VUuCymPDdZga/soTU3444nFBERKR88ooib4y5DHgR+MayrHjLsrZalpVtWVYacD3uK9o8YYw5earMyRX3yP99NgAqn7afiJSiQH8/Hr2sOXNG96Vrk6r2fHN6JjdNWsbfZvxCVm6BgwlFRETKH68o8sA1nu33pz9gWVY2sBJ31s6e8QbPtuXp+xtjAoCmQAHu1XwRKSOX1Azn04d68n9DLiUsyH1KjWXBu0u3M3BcMos3HXA4oYiISPnhLUX+5NVlap7h8ZPzPM/2O8/2qiL2jQFCgWWWZele8iJlzM/PcHfPJsyLiyGm5e//S+86coK73l7Jk9NWcSw738GEIiIi5YO3FPnFnu1Dxpj6pz5gjLka6A3kAMs842nAQWCYMabLKfuGAC94vny9VBOLyFk1qBrKe/d15T83dySyUqA9/yxlF/0Sk5i7dp+D6URERHyftxT5abivE18b+NUY854x5kVjzDfALNxvXv2zZVmHACzLOo77yjb+wCJjzFueN8P+jPuurtOATx34PkTkFMYYboxuwML4WAa3r2vPD2Tk8vCHqTz2URoHMvQPZyIiIsXhFUXesiwXMAiIA9bhfoPrE0APYDYw0LKs8acd8xUQi/sGUDcCjwP5QDwwzLIsXbFGxEvUjAhm4h1RTLozmpoRv9+nbdaavfRLSGJ66i70v6yIiMiFMfrhWTRjTGpUVFRUampqWb+u/bn+bKQ8Opadz99nr+OzlF3/NY9tWZO/X9+OBlVDHUomIiJS9qKjo0lLS0s7072NzsYrVuRFpOKIDA3kpZs68uED3WlQtZI9T9p4gIGJyby/fDsul36JFRERORcVeRFxRJ8WNZg3Job7ejfh5D9EZeUV8uzXv3Drm8vZciDT2YAiIiJeTkVeRBwTFhzAX6+9lGkP96J5rXB7/uP2I1w9fjGvLdpMfqHLwYQiIiLeS0VeRBwX3bgqs0b1YdQVzQnwcy/P5xW4eGnuBoZOXMra3bpJs4iIyOlU5EXEKwQH+BM/oBXfjOxD+/qR9vyXPccZMnEp/563npz8QgcTioiIeBcVeRHxKm3rVebLR3vx1NWtCQ5w/xVV6LKY+P0WBr2ymJTthx1OKCIi4h1U5EXE6wT4+zEi9hLmjO5Lt6bV7PnWA1nc/MZy/vr1WjJzCxxMKCIi4jwVeRHxWs1qhjN1eA9eGNqO8OAAACwL3lv+GwMTk0naeMDhhCIiIs5RkRcRr+bnZ7izR2Pmx8VwWaua9nz30RPc885KnvhsFUez8xxMKCIi4gwVeRHxCfWqVOLde7sy7tZOVA0NtOfT03bRLyGZOWv2OphORESk7KnIi4jPMMYwtHN9FsTHcm3Hevb8YGYuj3yUxsMfpJJ+PMfBhCIiImVHRV5EfE6N8GAm3NaZyXd3oXblYHs+95d99EtI4vOUnViW5WBCERGR0qciLyI+q3/b2syPi+W2bg3t2fGcAmr7rG4AACAASURBVP44bTV3v7OSnYezHUwnIiJSulTkRcSnRVYK5J83dODjB7vTqFqoPV+86SADxyXz7tJtFLq0Oi8iIuWPiryIlAu9mtdg7pi+PNinKX7GPcvOK+RvM9ZxyxvL2Zye4WxAERGREqYiLyLlRmhQAM9c05bpj/SiZe1we5762xEGjV/Cq99tIr/Q5WBCERGRkqMiLyLlTudGVZn5eF9GX9mCQH/38nxeoYuX52/kuleXsnb3MYcTioiIXDwVeREpl4IC/Ijr35IZj/ehY4NIe/7r3uMMmbiUf81ZT05+oYMJRURELo6KvIiUa63rVOaLR3vz9KA2hAS6/8ordFlMStrC1eMX88PWQw4nFBERKR4VeREp9/z9DMNjmjF3dAw9mlWz59sOZnHrmysY+9VaMnLyHUwoIiJy4VTkRaTCaFIjjI8f7ME/rm9PRHCAPf9gxW8MTEzm+w3pDqYTERG5MCryIlKh+PkZbu/eiPnxMVzZupY933Msh/ve/ZH4T3/mSFaegwlFRETOj4q8iFRIdSMr8dY9XXjlts5UCwuy51/8tJt+CUnMXL0Hy9KNpERExHupyItIhWWM4bqO9VgYH8uQTvXs+aGsPEZ+/BMPfZDK/uM5DiYUERE5MxV5EanwqoUFMX5YZ96+pwt1KofY8wXr9tMvIYlPf9yh1XkREfE6KvIiIh5XtqnN/PgYbu/eyJ5l5BTwp+lruOOtH9hxKNvBdCIiIv9NRV5E5BSVQwL5x/Xt+WR4D5pUD7Xny7YcYuC4ZN5eso1Cl1bnRUTEeSryIiJF6HlJdeaMjmFETDP8jHt2Ir+Q52eu48bXl7Fxf4azAUVEpMJTkRcROYNKQf48NagNXz3Wm9Z1Iuz5zzuPMviVxbzy7SbyClwOJhQRkYpMRV5E5Bw6NKjCNyP7EN+/JYH+7uX5/EKLhAUbue7VJazaedThhCIiUhGpyIuInIegAD9GXdmCWaP60qlhFXu+fl8G17+2lH/M/pUTeYUOJhQRkYpGRV5E5AK0rB3B9Ed6MfaatlQK9AfAZcGbyVu5enwyy7cccjihiIhUFCryIiIXyN/P8ECfpswbE0Pv5tXt+fZD2dw2eQV/+XINx3PyHUwoIiIVgYq8iEgxNaoeyocPdOelGzsQERJgzz/+YQcDEpL5bv1+B9OJiEh5pyIvInIRjDHc0rUhC+Nj6d+2tj3fdzyH+6ekMHrqTxzKzHUwoYiIlFcq8iIiJaB25RDevCuaibdHUSM8yJ5//fMe+icm8/XPu7Es3UhKRERKjoq8iEgJMcYwuENdFsTFckPn+vb8cFYeo6f+zIPvpbD32AkHE4qISHmiIi8iUsKqhgWRcGsn3r2vK/UiQ+z5t+vTGZCQzMc/7MDl0uq8iIhcHBV5EZFScnmrWsyLi+GuHo3tWUZuAX/5cg23v7WC7QezHEwnIiK+TkVeRKQURYQE8vzQdnw2oidNa4TZ8xVbD3PV+GQmJ2+lUKvzIiJSDCryIiJloFvTaswZ3ZdHLrsEfz8DQE6+i7/P/pUbXlvKhn0ZDicUERFf43VF3hjT1xgz3Riz1xiT69nON8YMKmLfXsaY2caYw8aYbGPMamPMGGOMvxPZRUTOJiTQnz9d1ZqvHu1Nm7qV7fmqXce4ZsJiEhdsJLeg0MGEIiLiS7yqyBtjngGSgRhgLvAfYAZQFbjstH2HnLLvl8BEIAhIBKaWWWgRkQvUvkEk34zszR8HtiLI3/3XcH6hxfhvN3HthCX8tOOIwwlFRMQXGG+5rrEx5mbgM2AhcINlWRmnPR5oWVa+5/PKwGYgEuhtWVaKZx4CfAf0BG6zLKvYhd4YkxoVFRWVmppa3Kco7uvan3vLn42IlJ7N6Rk8OW01aTuO2jNj4P7eTXliQEtCgwLOcrSIiPi66Oho0tLS0izLir7QY71iRd4Y4we8CGQDt59e4gFOlniPm4CawNSTJd6zTw7wjOfLR0ovsYhIyWheK4LPH+7Fc9e2JTTIfVagZcHbS7Zx1bjFLNt80OGEIiLirbyiyAO9gKbAbOCIMWawMeZPxpjRxpieRex/hWc7t4jHknH/QtDLGBNcOnFFREqOv5/h3t5NmTcmhr4tatjzHYezuf2tH/jz9NUcO5F/lmcQEZGKyFv+zbarZ7sfSAPan/qgMSYZuMmyrAOeUSvPduPpT2RZVoExZhtwKdAM+PVsL2yMOdO5M63PL7qISMloWC2U9+/vxrTUXTw/cx3HcwoAmPrjTr7fkM4LQ9vTv21th1OKiIi38JYV+Vqe7cNAJaAfEAG0A+bhfkPr56fsH+nZHjvD852cVynZmCIipcsYw81dGrLwiViuurSOPd9/PJfh76cw8uM0DmbmOphQRES8hbcU+ZOXizS4V96/tSwr07KsX4DrgV1A7BlOsynKyXeMnvPdopZlRRf1Aay/0G9CRKSk1IoIYdJd0bx+RxQ1wn8/S3Dm6r30S0jiy5926Q3xIiIVnLcU+ZPXWttqWdaqUx+wLOsE7lV5gG6e7ckV90iKVvm0/UREfNLV7euyMD6GG6Ma2LOj2fnEfbqK+6f8yJ6jJxxMJyIiTvKWIr/Bsz16hsdPFv1Kp+3f8vQdjTEBuN84WwBsLamAIiJOqRIaxH9u6ch793ejfpVK9vz7DQcYkJjMByt+w+XS6ryISEXjLUU+GXfxbmGMCSri8Xae7XbP9jvP9qoi9o0BQoFllmXpRFIRKTdiW9ZkflwM9/ZqwslbTmTmFjD2q7UMm7yCbQeznA0oIiJlyiuKvGVZB4FPcZ8q8+ypjxlj+gMDcZ8mc/Jyk9OAg8AwY0yXU/YNAV7wfPl6KccWESlzYcEBPHfdpXw+oifNaobZ85XbDnPVuGQmJW2hoNDlYEIRESkrXlHkPeJx3631aWNMsjHmZWPM58AcoBAYblnWUQDLso4Dw3G/SXaRMeYtY8xLwM+47+o6DfcvBiIi5VKXJtWYPaovj11+Cf5+7uX53AIX/5qznutfW8a6PccdTigiIqXNa4q8ZVnpQHcgEWgIjMJ946dZQF/Lsj4/bf+vgFjcp+XcCDwO5OP+hWCYpcs5iEg5FxLozx8Htuabkb25tF5le75m9zGue3UJ/5m/gdyCQgcTiohIaTLqu0UzxqRGRUVFpaae6X5Rpfa69uf6sxGR85Vf6GLy4q2MW7iJvILfT61pXiucF2/sQHTjqg6mExGRM4mOjiYtLS3Nc/nzC+I1K/IiIlJ8gf5+PHpZc+aM7kvXJr+X9s3pmdw0aRl/m/ELWbkFDiYUEZGSpiIvIlKOXFIznE8f6snzQy4lLMh9rz3LgneXbmfguGQWbzrgcEIRESkpKvIiIuWMn5/hrp5NmBcXQ2zLmvZ815ET3PX2Sp6ctopj2fkOJhQRkZKgIi8iUk41qBrKlPu6knBLR6qEBtrzz1J20S8xiblr9zmYTkRELpaKvIhIOWaM4YaoBiyIi2Vw+7r2/EBGLg9/mMqjH6WSnpHjYEIRESkuFXkRkQqgZkQwE++IYtKd0dSMCLbns9fso39CMtNTd+lKWSIiPkZFXkSkArmqXR0WxsVya5eG9uzYiXye+HwV97z7I7uOZDuYTkRELoSKvIhIBRMZGsiLN3Xgwwe606BqJXuevPEAAxOTeX/5dlwurc6LiHg7FXkRkQqqT4sazI+L4f7eTTl5L7qsvEKe/foXbn1zOVsOZDobUEREzkpFXkSkAgsNCuDZa9sy7eFeNK8Vbs9/3H6Eq8cv5rVFm8kvdJ3lGURExCkq8iIiQnTjqswa1YdRVzQnwM+9PJ9X4OKluRsYOnEpa3cfczihiIicTkVeREQACA7wJ35AK74Z2Yf29SPt+S97jjNk4lJemruenPxCBxOKiMipVORFROS/tK1XmS8f7cVTV7cmOMD9Y6LQZfHaoi0MemUxKdsPO5xQRERARV5ERIoQ4O/HiNhLmDsmhm5Nq9nzrQeyuPmN5fz167Vk5hY4mFBERFTkRUTkjJrWCGPq8B68MLQd4cEBAFgWvLf8NwYmJpO08YDDCUVEKi4VeREROSs/P8OdPRozPy6Gy1vVtOe7j57gnndW8sRnqzianedgQhGRiklFXkREzku9KpV4596ujLu1E1VDA+359LRd9EtIYvaavQ6mExGpeFTkRUTkvBljGNq5PgviY7m2Yz17fjAzj0c/SuPhD1JJP57jYEIRkYpDRV5ERC5YjfBgJtzWmcl3d6F25WB7PveXffRLSOKzlJ1YluVgQhGR8k9FXkREiq1/29osiI/ltm4N7dnxnAKenLaau99Zyc7D2Q6mExEp31TkRUTkolQOCeSfN3Tg4we706haqD1fvOkgA8cl8+7SbRS6tDovIlLSVORFRKRE9Gpeg3ljYniwT1P8jHuWnVfI32as4+ZJy9icnuFsQBGRckZFXkRESkylIH+euaYt0x/pRcva4fY8bcdRBo1fwqvfbSK/0OVgQhGR8kNFXkRESlznRlWZ+XhfRl/ZgkB/9/J8XqGLl+dv5NoJS1iz65jDCUVEfJ+KvIiIlIqgAD/i+rdkxuN96Ngg0p6v35fB0NeW8q8568nJL3QwoYiIb1ORFxGRUtW6TmW+eLQ3zwxuQ0ig+8dOoctiUtIWrh6/mB+2HnI4oYiIb1KRFxGRUufvZ3iwbzPmjYmhZ7Pq9nzbwSxufXMFz3y1hoycfAcTioj4HhV5EREpM42rh/Hx8O7884b2RAQH2PMPV+xgYGIy369PdzCdiIhvUZEXEZEyZYzhtm6NWBAfS782tez5nmM53DflR+I+/ZnDWXkOJhQR8Q0q8iIi4og6kSFMvrsLr9zWmWphQfb8y5920z8hiZmr92BZupGUiMiZqMiLiIhjjDFc17EeC+NjGdqpnj0/lJXHyI9/4qEPUtl/PMfBhCIi3ktFXkREHFctLIhxwzrzzr1dqBsZYs8XrNtPv4Qkpq7codV5EZHTqMiLiIjXuKJ1bebHxXBH90b2LCOngD9/sYY73vqBHYeyHUwnIuJdVORFRMSrRIQE8vfr2zP1oR40qR5qz5dtOcSAcUm8tXgrhS6tzouIqMiLiIhX6tGsOnNGxzAiphl+xj3LyXfxwqxfufH1ZWzcn+FsQBERh6nIi4iI16oU5M9Tg9rw1WO9aV0nwp7/vPMog19ZzPiFm8grcDmYUETEOSryIiLi9To0qMI3I/sQ378lgf7u5fn8QovEhRu5dsISVu086nBCEZGypyIvIiI+ISjAj1FXtmD2qL50blTFnm/Yn8H1ry3l77PWcSKv0MGEIiJlS0VeRER8SovaEUx7uBfPXtOWSoH+ALgsmLx4G1eNT2b5lkMOJxQRKRsBTgcQERG5UP5+hvv7NKVfm9o89eVqlm52l/ffDmVz2+QV3NatEU8Nak3lkECHk55BQQHMmgUpKZCZCeHh0KULDB4MAfrRLCLnR39biIiIz2pUPZQPH+jO5ym7eH7WOjJyCgD4ZOUOvl+fzt+vb8eVbWo7nPIUWVmQmAhvvAG7dv3v4w0awIgREBcHYWFln09EfIpOrREREZ9mjOGWrg1ZGB9L/7a/l/Z9x3N44L0URn3yE4cycx1M6JGeDpddBmPHFl3iwT0fO9a9X3p6WaYTER/ktUXeGHOXMcbyfDx4hn2uMcYsMsYcM8ZkGmN+MMbcU9ZZRUTEebUrh/DmXdFMvD2KGuFB9vybVXvol5DE1z/vxrIcupFUVpb7tJmUlPPbPyXFvX9WVunmEhGf5pVF3hjTEJgAZJ5ln5HADKAd8CEwGagHTDHGvFwWOUVExLsYYxjcoS4L4mK5Iaq+PT+Snc/oqT/z4Hsp7D12ouyDJSaef4k/KSUFxo0rnTwiUi54XZE3xhjgXeAQMOkM+zQBXgYOA10sy3rMsqw4oAOwBXjCGNOzTAKLiIjXqRoWRMItnXj3vq7Uiwyx59+uT6d/QjIf/fAbLlcZrc4XFLjPiS+ON95wHy8iUgSvK/LAKOAK4D7gTP+meD8QDLxqWdb2k0PLso4A//B8+XApZhQRER9weatazI+P5e6eje1ZZm4BT3+5ltvfWsH2g2Vw6sqsWWc+J/5cdu6E2bNLNo+IlBteVeSNMW2AfwHjLctKPsuuV3i2c4t4bM5p+5zrNVOL+gBan3dwERHxWuHBAfzfkHZ8NqInTWv8fiWYFVsPM3BcMm8mb6Gg0FV6AS70lJqSPl5Eyi2vKfLGmADgA2AH8Jdz7N7Ks914+gOWZe3FvZLfwBgTWqIhRUTEZ3VrWo05o/vyyGWX4O9nAMgtcPGP2eu58fVlrN93vHReOPOMb/c6PxkZJZNDRModrynywLNAZ+Bey7LO9U6kSM/22BkeP3bafmdkWVZ0UR/A+vNKLSIiPiMk0J8/XdWarx/rTdu6le35ql3HuOaVJSQs2EhuQWHJvmh4+MUdHxFRMjlEpNzxiiJvjOmGexX+P5ZlLS+Jp/RsHbrOmIiIeLN29SP5emRv/jiwFUH+7h+FBS6LV77dxDWvLCFtx5GSe7EuXZw9XkTKLceL/Cmn1GwExp7nYedacT+5zFJK/04qIiK+LtDfj8cub87s0X2JblzVnm9Kz+TG15fx/Mx1ZOeVwBVjBg9237G1OBo2hEGDLj6DiJRLjhd5IBxoCbQBck65CZQF/NWzz2TP7OQFdTd4ti1PfzJjTF0gDNhlWVZ2KWcXEREf17xWOJ+P6Mlz17YlNMgfAMuCt5dsY+C4ZJZuPnhxLxAQACNGFO/YESPcx4uIFMEbinwu8PYZPn7y7LPE8/XJ026+82yvKuL5rj5tHxERkbPy8zPc27sp88bE0LdFDXu+8/AJ7njrB/48fTXHTuQX/wXi4i78FJmuXWHMmOK/poiUe8ax21WfB2PMc7hX5YdblvXWKfOmwK+4r04TffJa8saYqsCPwCVAr4s5394YkxoVFRWVmppa/G+geK9rf+7NfzYiIuWVZVlMS93F8zPXcTzn91NrakUE88LQdgy4tE7xnjg93X2azflcTrJrV/f152vWLN5riYjPiI6OJi0tLc1zsZUL4g0r8hfMsqxtwB+BakCKMWaiMSYRWI27xJfUm2ZFRKSCMcZwc5eGLHwilqvb/V7a0zNyeeiDVB77OI0DGbkX/sS1asGiRfDCC2c+Z75hQ/fj33+vEi8i5+SzJ95ZljXBGLMd+ANwN+5fStYBz1iW9Z6T2URExPfVigjh9TujmbNmL2O//oWDme7yPmv1XpZuPshfr23L0E71/+tfUs8pLAyefhr+9Cf3HVtTUtzXiY+IcJ96M2iQzokXkfPm1afWOEmn1oiIyElHs/N4YdavTEvd9V/zy1rV5O/Xt6d+lUoOJRMRX1fhTq0REREpS1VCg3j55o68f3+3/yrtizYcYEBCEh8s347LpcUXESlbKvIiIiLnKaZlTebHxXBvryac/AfUrLxCxn79C8PeXMHWA5nOBhSRCkVFXkRE5AKEBQfw3HWX8vmInjSrGWbPV24/zFXjF/P6oi0UFLocTCgiFYWKvIiISDF0aVKN2aP6MvLy5vj7uZfn8wpcvDh3PUNfW8q6Pbq5uIiULhV5ERGRYgoJ9OcPA1vxzcjeXFqvsj1fu/s41726hJfnbSAnv9DBhCJSnqnIi4iIXKRL60Xy9WO9+dNVrQkKcP9oLXBZvPr9Zga/spjU3w47nFBEyiMVeRERkRIQ4O/HI5ddwpzRfenapKo933Igi5smLee5b34hK7fgLM8gInJhVORFRERK0CU1w/n0oZ48P+RSwoL8AbAsmLJsOwMSk0neeMDhhCJSXqjIi4iIlDA/P8NdPZswLy6G2JY17fnuoye4+52V/OHzVRzLzncwoYiUByryIiIipaRB1VCm3NeVhFs6UiU00J5PS91Fv8Qk5q7d62A6EfF1KvIiIiKlyBjDDVENWBAXy+AOde35gYxcHv4wjUc+TCU9I8fBhCLiq1TkRUREykDNiGAm3h7FG3dFUzMi2J7PWbuP/gnJTEvdhWVZDiYUEV+jIi8iIlKGBl5ah4VxsdzSpYE9O3Yinz98voq731nJzsPZDqYTEV+iIi8iIlLGIkMDeemmjnz4QHcaVK1kzxdvOsjAcclMWboNl0ur8yJydiryIiIiDunTogbz42K4v3dTjHHPsvMKeW7GOm55Yzmb0zOdDSgiXk1FXkRExEGhQQE8e21bpj/Sixa1wu15ym9HGDR+MRO/30x+ocvBhCLirVTkRUREvEBUo6rMHNWHUVc0J8DPvTyfV+ji3/M2MOTVpazdfczhhCLibVTkRUREvERwgD/xA1ox4/E+tK8fac/X7T3OkIlLeXHuenLyCx1MKCLeREVeRETEy7SpW5kvH+3FU1e3JjjA/aO60GXx+qItDBq/mB+3H3Y4oYh4AxV5ERERLxTg78eI2EuYOyaGbk2r2fOtB7O4edJynv16LZm5BQ4mFBGnqciLiIh4saY1wpg6vAd/v74d4cEB9vz95b8xMDGZRRvSHUwnIk5SkRcREfFyfn6GO7o3ZkF8DFe0rmXPdx89wb3v/kj8Zz9zJCvPwYQi4gQVeRERER9RN7ISb9/ThfHDOlE1NNCef5G2m/6JScxesxfL0o2kRCoKFXkREREfYoxhSKf6LIyP5dqO9ez5wcw8Hv0ojYc/TCX9eI6DCUWkrKjIi4iI+KDq4cFMuK0zk+/uQu3KwfZ83i/76ZeQxGc/7tTqvEg5pyIvIiLiw/q3rc2C+Fhu69bInh3PKeDJ6au56+2V7Dyc7WA6ESlNKvIiIiI+rnJIIP+8oT0fD+9O4+qh9nzJ5oMMSEzmnSXbKHRpdV6kvFGRFxERKSd6XVKDuaNjGN63KX7GPTuRX8j/zVzHzZOWsWl/hrMBRaREqciLiIiUI5WC/Hl6cFu+eLQ3rWpH2PO0HUcZ/MoSJny7ifxCl4MJRaSkqMiLiIiUQ50aVmHG430Y068Fgf7u5fm8Qhf/WbCRaycsYc2uYw4nFJGLpSIvIiJSTgUF+DGmX0tmPt6Xjg2r2PP1+zIYMnEJ/5zzKzn5hQ4mFJGLoSIvIiJSzrWqE8EXj/TimcFtCAl0/+h3WfBG0lauHr+YFVsPOZxQRIpDRV5ERKQC8PczPNi3GfPGxNCzWXV7vu1gFsPeXMHTX64hIyffwYQicqFU5EVERCqQxtXD+Hh4d/51Q3siggPs+Uc/7GBAYjLfr093MJ2IXAgVeRERkQrGGMOwbo1YEB9Lvza17PneYzncN+VHxkz9icNZeQ4mFJHzoSIvIiJSQdWJDGHy3V2YcFtnqocF2fOvft5D/4QkZqzag2XpRlIi3kpFXkREpAIzxnBtx3osiI9laKd69vxQVh6Pf/ITw99PZf/xHAcTisiZqMiLiIgI1cKCGDesM+/c24W6kSH2fOGv++mXkMTUlTu0Oi/iZVTkRURExHZF69rMj4vhzh6N7FlGTgF//mINd7z1A78dynIwnYicKuDcu4iIiEhFEhESyAtD23NNh3r8efpqth/KBmDZlkMMHJfMHwa04r7eTfH3M2UTqKAAZs2ClBTIzITwcOjSBQYPhgBVGam49F+/iIiIFKlHs+rMHRND4sKNTE7eisuCnHwXL8z6lRmr9/LSjR1oVSei9AJkZUFiIrzxBuza9b+PN2gAI0ZAXByEhZVeDhEv5RWn1hhjqhtjHjTGfGmM2WyMOWGMOWaMWWKMecAYU2ROY0wvY8xsY8xhY0y2MWa1MWaMMca/rL8HERGR8igk0J+nrm7DV4/1pvUppX3VzqNcM2Ex4xZuJK/AVfIvnJ4Ol10GY8cWXeLBPR871r1fuq5/LxWPVxR54GZgMtAd+AEYB0wH2gFvAZ8ZY/7r3++MMUOAZCAG+BKYCAQBicDUMksuIiJSAXRoUIUZj/fhif7/3959h0dVpn0c/94phN6rdATEhkhEEDBYABVdO+qrorKiYFmE6LrF1dV3393VdTcCawFdFcWyKnZRQRQTmiBBReldmiC9hrTn/eNMxhgTIKScc5Lf57pyHXKfOTM3wzD55ZnnPKcjVWK9+JCV4xg1dTm/+vcMvlm3s/QebN8+b9rMvHlHdvt587zb79P8falcghLklwEXAy2cc9c55/7gnPs10AlYB1wBXJ53YzOrjRf8c4CznHM3O+d+C3QBZgNXmtk15f2XEBERqcjiY2P4zbkdmDS8N6e2qhutL928h8uenMlfJy3iQGZOyR/osceOPMTnmTcPRo0q+WOLhEgggrxz7jPn3PvOudwC9R+AsZFvz8q360qgEfBf59y8fLfPAP4U+fa2sutYRESk8urQpBYTh/XkgYtOoFq8N5s118Ez01dz/ug0Zq3cevR3np3tzYk/GuPGeceLVBKBCPKHkRXZ5v+feU5k+3Eht08D9gM9zSyhLBsTERGprGJjjF/3bsuUkUn0bt8wWl+7bT/XPjOHP7z1Lbszsg5xD0WYNKnoOfGHs24dfPjh0R0rEkKBDvJmFgfcEPk2f2g/LrJdVvAY51w2sBpvRZ52R/AY6YV94U3rERERkUNoWb86E24+nX9c0ZlaVX9aDO/Vud/TLyWVqYs2F+8OizulprSPFwmRQAd54GG8E14/dM5NzlevE9nuKuK4vHrdIvaLiIhIKTEzrurWkqnJfeh/QpNoffPugwx5cR7DX/2KbXsPHtmd7d1bsmb27CnZ8SIhEtggb2bDgbuBJcCg4h4e2R72WtLOucTCviKPKyIiIkeoSe2qjBuUyJPXdaVhzSrR+nvfbKRvSirvfr0B5w7zo7lmzZI1UasM17UXCZhABnkzuwMYDSwCznbObS9wk7wR9zoUrnaB24mIiEg5MDMGnNyMT0b24fKuzaP1HfuzuOu/XzPkhXls2nWg6Ds47bSSNVDS40VCJHBB3sxGAI8D3+GF+B8KudnSyLZjIcfHAW3xTo5dVVZ9ioiISNHq1ahCylVdGD+4G83rVovWP12yhX4pHyj5TgAAIABJREFUabw8Zy25uYWMzl94oXfF1qPRsiUMGHCUHYuET6CCvJn9Du+CTl/jhfiiLtP2WWR7fiH7koDqwCzn3BFOyBMREZGycNZxjZk8Mokbzmgdre09mM19b3/H/zzzBau3FriIU1wcDB16dA82dKh3vEglEZggb2b3453cmg6c65w71CK0E4GtwDVmFv0MzcyqAv8X+fapsupVREREjlzNhDj+95KTeH3oGbRrWCNan7N6O+ePSuPptJVk5+S7lMzIkcWfItOtG4wYUUodi4SDHfakk/JowuxGYDzelVr/TeFz29c458bnO+ZSvECfAfwX2I53ddjjIvWrXAn+cmaW3rVr167p6elHexdH+7jRPwfh30ZERKQ0ZWTlMObT5YxLW0VOvqk1nVvU4ZErOnN8s8hpblu2eNNsjmQ5yW7dvPXnGzUqo65Fyk5iYiLz58+fH1lspViC8vlT28g2Fijq1+lUvLAPgHPuHTPrA9wHXAFUBVYAycCYkoR4ERERKRtV42O59/xODDi5GfdOXMCiTbsBWLB+F7/69wxuP+tY7jinPQmNG8Pnn8OoUTB2bOEXiWrZ0ptOM2IE1Kjxy/0iFVwgRuSDSCPyIiIiZSsrJ5en01Yx+tPlZGb/NLWmQ+OaPHJlZ7q2qucVsrO9K7bOm+etE1+rljf1ZsAAzYmX0CvJiLyCfBEU5EVERMrHii17+d2bC0hfuyNaM4PBPdtyz3kdqV5FYV0qrpIE+cCc7CoiIiKVU/vGNXlj6Bk8dPGJVK8SC4Bz8NzM1Zw3Ko2ZKw61/oVI5aUgLyIiIr6LiTFu7NmGKSOTSOr400mr67Yf4Lr/zOF3Exew60CWjx2KBI+CvIiIiARGi3rVeWFwN/458BTqVIuP1l+bt45+KalMXljYdSJFKicFeREREQkUM+PKxBZ8kpzEgJObRutb9hxk6IR07nh5Pj/u0TUfRRTkRUREJJAa16rKk9clMvb6rjSsmRCtT/p2E/0eS+Wt+eu1MIRUagryIiIiEmjnn9SMT5P7MDCxRbS2c38Wya9/w+DxX7Jh5wEfuxPxj4K8iIiIBF6d6vE8OvAUXvz16bSoVy1a/3zpj/RPSWXC7DXk5mp0XioXBXkREREJjaSOjZg8IomberYh79Ir+zJzuP/dhVzz9Bes/HGvvw2KlCMFeREREQmVGglxPHjxiUwcdgbHNqoRrc9ds50LRk/nyc9XkJ2Te4h7EKkYFORFREQklBJb12fS8DO58+z2xMV4w/OZ2bn84+OlXPrkTBZu3OVzhyJlS0FeREREQqtqfCz3nHcc797Zi5Oa147Wv9uwm4sfn8mjk5eQkZXjY4ciZUdBXkRERELvxGPq8M7tvfj9BZ1IiPPiTU6u44lpK7lwzHTS1273uUOR0qcgLyIiIhVCXGwMw/ocy0d3ncnpbepH6yt/3MeVY2fz4HsL2Xcw28cORUqXgryIiIhUKO0a1eS/t/bgL5eeRI0qsQA4B+NnraH/Y2mkLfvR5w5FSoeCvIiIiFQ4MTHGoB6tmZLch7OOaxStb9h5gBuem8s9b3zDzv2ZPnYoUnIK8iIiIlJhNa9bjedv6sZjV59C3erx0frE9PX0TUnjo283+didSMkoyIuIiEiFZmZcdmoLpib34aLOzaL1rXsPctvL87ntpXS27MnwsUORo6MgLyIiIpVCw5oJPH5tV54elEjjWgnR+kff/UC/lDTemLcO55yPHYoUj4K8iIiIVCr9T2zKJ8l9uKZby2ht14EsfjtxATc8N5d12/f72J3IkVOQFxERkUqnTrV4Hr6iMy8P6U7L+tWi9enLt3LeqDTGz1xNbq5G5yXYFORFRESk0urVviGTRyRxc++2mHm1/Zk5PPj+IgaOm82KLXv8bVDkEBTkRUREpFKrXiWO+y86gTdv60mHxjWj9fS1OxgwegZPTFtBVk6ujx2KFE5BXkRERATo2qoeHwzvzfBzOxAX4w3PZ+bk8ujkpVzy+Ey+27DL5w5Ffk5BXkRERCQiIS6W5H4d+WB4bzq3qBOtL9q0m0uemMkjHy8hIyvHxw5FfqIgLyIiIlJAp6a1eeu2ntw34HgS4ry4lJPreOrzlQwYPZ0v12z3uUMRBXkRERGRQsXFxnBLUjsmj0iie9v60fqqrfsYOHY2D7z7HXsPZvvYoVR2CvIiIiIih9CmYQ1evaUHf7vsZGomxEXrL85eS/+UVKYt3eJjd1KZKciLiIiIHEZMjHFt91Z8kpzEuZ0aR+sbd2Uw+PkvSX7ta3bsy/SxQ6mMFORFREREjlCzOtX4z42nMfqaLtSvUSVaf+urDfR7LJVJCzbhnC4kJeVDQV5ERESkGMyMS7o055ORSVzS5ZhofeveTO54ZT5DJ6SzZXeGjx1KZaEgLyIiInIUGtRMYPQ1p/KfG06jae2q0fqURZs5NyWV179cp9F5KVMK8iIiIiIl0PeEJkxJTuLa7q2itT0Z2dz75gIGPTuX77ft97E7qcgU5EVERERKqHbVeP522cm8ckt3WjeoHq3PWLGV80al8eyM1eTkanReSpeCvIiIiEgp6XlsQz6+K4lbk9oRY17tQFYOf/lgEVeOncXyzXv8bVAqFAV5ERERkVJUrUosfxxwPG/f3otOTWtF6199v5MLx8xgzKfLyczO9bFDqSgU5EVERETKwCkt6/Lenb0Z2bcj8bHe8HxmTi4pnyzj4sdnsGD9Tp87lLBTkBcREREpI1XiYrirbwcmDT+TLi3rRutLftjDpU/M5O8fLiYjK8fHDiXMFORFREREyljHJrV487ae/OnC46ka78WvXAfj0lZx/qg0vli1zecOJYwU5EVERETKQWyMMeTMdkwZ0YeexzaI1tds2881T3/BfW9/y56MLB87lLBRkBcREREpR60aVOflId155IqTqVU1Llp/ec739H8sjc+WbPaxOwkTBXkRERGRcmZmXN2tFVOT+9DvhCbR+qZdGfx6/DxG/Pcrtu/L9LFDCYNQB3kza2Fmz5nZRjM7aGZrzGyUmdXzuzcRERGRw2lSuypPD0rk8WtPpUGNKtH6O19vpG9KKu99sxHndCEpKVxog7yZHQukA4OBucBjwCrgLmC2mTU4xOEiIiIigWBmXNT5GKYm9+GyU5tH69v3ZTL81a+45cV0ftiV4WOHElShDfLAk0BjYLhz7lLn3O+dc+fgBfrjgL/62p2IiIhIMdSrUYXHru7C8zd1o1mdqtH61MWb6ZeSyqtzv9fovPxMKIO8mbUD+gNrgCcK7P4zsA8YZGY1yrk1ERERkRI5u1NjpoxM4voeraK1PQez+cNb33LtM3NYu22fj91JkFgYf7MzsyHAM8DTzrmhheyfjBf0+zrnPj3MfaUXsasTUL2kvYqIiIgcrYQWJ9LgguHE1/9pyk1uVgY7p7/Ennnvgcv1sTspRfOdc4nFPSiUI/J4U2cAlhWxf3lk27EcehEREREpEwfXL2TT879h1xdv4HK9K8DGxFel/jlDaHr9o8Q3bO1zh+KnsAb5OpHtriL259XrFrE/yjmXWNgXsKQ0GhUREREpCZedyc7UF/jhxWQyN6+K1hOOOY5mN42iTq//gZi4Q9yDVFQV9V/dItsSzRvq2rUr6elFzbwpG2YW/XMYpz2JiIhI2cnKyWVc6krGfLqCzJxcLDaeur2vo/sVw3jkys50aXnYMUwJmMTERObPn39Ux4Z1RD5vxL1OEftrF7idiIiISOjFx8Zw5zkd+PCu3nRt9VNoX7p5D5c/OZP/+2ARBzJzfOxQylNYg/zSyLaoOfAdItui5tCLiIiIhFb7xrV4Y1hP/vyrE6gWHwtAroP/zFjNeaPSmLVyq88dSnkIa5CfFtn2N7Of/R3MrBbQCzgAfFHejYmIiIiUh9gYY3CvtkwZmUTv9g2j9e+37+faZ+bwh7cWsDsjy8cOpayFMsg751YCU4A2wB0Fdj8E1ABedM5poVURERGp0FrWr86Em0/nH1d2pnbVn05/fHXuOvqlpDJ10WYfu5OyFMogH3E7sAUYY2bvmNnfzewzYCTelJr7fO1OREREpJyYGVed1pKpyX0478Qm0frm3QcZ8uI8fvPqV2zbe9DHDqUshDbIR0blTwPGA92Bu4FjgTHAGc65bf51JyIiIlL+GteuyrhBp/HUdV1pWLNKtP7+Nxvpm5LKO19t0Kp4FUhogzyAc26dc26wc66Zc66Kc661c+4u59x2v3sTERER8csFJzdjanIfrujaIlrbsT+LEa99zc0vzGPjzgM+dielJdRBXkREREQKV7d6Ff511Sm88OvTaV63WrT+2ZIt9H8sjZe+WEturkbnw0xBXkRERKQC69OxEZNHJnHjGa2jtb0Hs/nTO9/xP898weqtWhskrBTkRURERCq4mglxPHTJSbwx7AzaNaoRrc9ZvZ3zR6UxLnUl2Tm5PnYoR0NBXkRERKSS6NamPh8OP5PbzzqW2BgD4GB2Ln//aAmXPzWLxZt2+9yhFIeCvIiIiEglUjU+lnvP78S7d/TihGa1o/UF63fxq3/PIGXKUg5m5/jYoRwpBXkRERGRSuik5nV4985e/Pa846gS50XC7FzHmM9WcOGYGaSv3eFzh3I4CvIiIiIilVR8bAx3nN2eD4efyWmt60XrK7bs5cqxs3jo/YXsz8z2sUM5FAV5ERERkUqufeOavD70DB66+ESqV4kFwDl4fuYa+j+WxozlW33uUAqjIC8iIiIixMQYN/Zsw5SRSSR1bBStr99xgOufncO9E79h14EsHzuUghTkRURERCSqRb3qvDC4G/8aeAp1qsVH66/PW0+/lFQmL/zBx+4kPwV5EREREfkZM+OKxBZMTe7DgJObRutb9hxk6IR07nh5Pj/uOehjhwIK8iIiIiJShEa1EnjyukTGXt+VRrUSovVJ326ib0oqb6avxznnY4eVm4K8iIiIiBzS+Sc1Y+rIPgxMbBGt7TqQxd1vfMNNz3/Jhp0HfOyu8lKQFxEREZHDqlM9nkcHnsKEm0+nRb1q0Xrqsh/pn5LKi7PXkJur0fnypCAvIiIiIkfszA6NmDwiicG92mDm1fZl5vDAuwu5+unZrPxxr78NViIK8iIiIiJSLDUS4vjzr05k4rCetG9cM1r/cs0OLhg9nSc/X0F2Tq6PHVYOCvIiIiIiclQSW9dj0vDe/Oac9sTFeMPzmdm5/OPjpVz65EwWbtzlc4cVm4K8iIiIiBy1hLhY7u5/HO/d2ZuTm9eJ1r/bsJuLH5/Jo5OXkJGV42OHFZeCvIiIiIiU2AnH1Obt23vy+ws6kRDnRcycXMcT01YyYMx05q3Z7nOHFY+CvIiIiIiUirjYGIb1OZaP7jqT09vUj9ZX/biPgeNm8+B7C9l3MNvHDisWBXkRERERKVXtGtXkv7f24C+XnkTNhDgAnIPxs9bQ/7E00pb96HOHFYOCvIiIiIiUupgYY1CP1kwZmcRZxzWK1jfsPMANz83lnje+Yef+TB87DD8FeREREREpM8fUrcbzN3Vj1NVdqFc9PlqfmL6evilpfPTtJh+7CzcFeREREREpU2bGpac255PkPlzUuVm0vnXvQW57eT7DJqSzZXeGjx2Gk4K8iIiIiJSLhjUTePzarjw9KJHGtRKi9Y8X/kDflFTemLcO55yPHYaLgryIiIiIlKv+Jzblk+Q+XNOtZbS2OyOb305cwA3PzWXd9v0+dhceCvIiIiIiUu7qVIvn4Ss68/KQ7rSqXz1an758K+eNSmP8zNXk5mp0/lAU5EVERETEN73aN+TjEWcypHdbYsyr7c/M4cH3FzFw3GxWbNnjb4MBpiAvIiIiIr6qXiWOP110Am/e1pMOjWtG6+lrdzBg9AyemLaCrJxcHzsMJgV5EREREQmEU1vV44Phvbnr3A7ERYbnM3NyeXTyUi5+fCbfbdjlc4fBoiAvIiIiIoGREBfLyH4d+WB4bzq3qBOtL960m0uemMnDHy0hIyvHxw6DQ0FeRERERAKnU9PavHVbT+4bcDxV473ImpPrGJu6kgGjpzN39XafO/SfgryIiIiIBFJcbAy3JLXj47uS6NGufrS+aus+rho3m/vf+Y69B7N97NBfCvIiIiIiEmhtGtbglSE9+NtlJ1MrIS5an/DFWvqnpDJt6RYfu/OPgryIiIiIBF5MjHFt91ZMSU7i3E6No/WNuzIY/PyXJL/2NTv2ZfrYYflTkBcRERGR0GhWpxr/ufE0Rl/Thfo1qkTrb321gb4pqXywYCPOVY4LSSnIi4iIiEiomBmXdGnOJyOTuKTLMdH6tn2Z3PnKVwydkM7m3Rk+dlg+FORFREREJJQa1Exg9DWn8uyNp9G0dtVofcqizfRNSeW1L7+v0KPzCvIiIiIiEmrnHt+EKclJXNu9VbS2JyOb3735Ldc/O4fvt+33sbuyoyAvIiIiIqFXu2o8f7vsZF69pQetG1SP1meu2MZ5o9J4dsZqcnIr1ui8gryIiIiIVBhnHNuAj+9K4takdsSYVzuQlcNfPljEFU/NYtnmPf42WIp8D/Jm1sHMfmdmn5nZOjPLNLPNZvaumZ19mGNvNLO5ZrbXzHaZ2edmdlF59S4iIiIiwVOtSix/HHA8b9/ei05Na0XrX6/byYVjpjPm0+VkZuf62GHp8D3IA38BHgaaAB8C/wJmAhcCn5nZ8MIOMrN/AuOBZsAzwEvAycD7ZnZn2bctIiIiIkF2Ssu6vHdnb0b27Uh8rDc8n5XjSPlkGRc/PoNv1u30ucOSMb/P5DWzm4BvnHNfFaj3AT4BHNDGObcp376eeGF/JdDNObcjUm8DpAM1gE7OuTUl6Cu9a9euXdPT04/2Lo72caN/9vvfRkRERKSiWLZ5D/dOXMDX+cJ7jMEtZ7ZjRN+OVKsS60tfiYmJzJ8/f75zLrG4x/o+Iu+cG18wxEfqqcDnQBWgZ4HdwyLbv+aF+Mgxa4AngARgcFn0KyIiIiLh07FJLd68rSf3X3QC1eK90J7rYFzaKi4YncYXq7b53GHx+R7kDyMrss0uUD8nsv24kGM+KnAbERERERFiY4ybe7dl8ogkerVvEK2v2bafa57+gj++/S27M7IOcQ/BEud3A0Uxs9bAucB+IC1fvQbQHNibf7pNPssj245H+DhFzZ3pdOTdioiIiEhYtGpQnZdu7s7r89bxf5MWsyfDGzN+Zc73LNm0mzdv6/mz6c5BFcgReTNLAF7GmyLzYP7pM0CdyHZXEYfn1euWUXsiIiIiEnJmxtXdWjE1uQ/9TmgSrd95TvtQhHgopRF5M1sDtC7GIS87564v4r5igQlAL+A14J9H2dYRnSla1IkFkZH6rkf52CIiIiISAk1qV+XpQYlM+nYT6Wt3cE6nJoc/KCBKa2rNSiCjGLffWFgxEuJfAgYCrwPXu18u3ZI34l6Hwh1uxF5EREREJMrMuKjzMVzU+Ri/WymWUgnyzrlzS3ofZhYHvIIX4l8BbnDO5RTyWPvMbAPQ3MyaFTJPvkNku6ykPYmIiIiIBFUg5sibWRVgIl6IfxEYVFiIz+ezyPb8QvZdUOA2IiIiIiIVju9BPnJi69vAJcCzwGDn3OGumTs2sr3PzOrlu682wB3AQeD5Um9WRERERCQggrD85FhgALAV2AA8UMiZwp875z7P+8Y5N8vMUoBkYIGZTcS7cNTVQH3gNyW5qquIiIiISNAFIci3jWwbAg8c4naf5//GOXe3mS0A7gRuBXKB+cCjzrkPyqBPEREREZHA8D3IO+fOKsGxLwAvlF43IiIiIiLh4PsceRERERERKT4FeRERERGREFKQFxEREREJIQV5EREREZEQUpAXEREREQkhBXkRERERkRBSkBcRERERCSEFeRERERGREFKQFxEREREJIQV5EREREZEQivO7gQBrs3jxYhITE31rwM/HFhEREZGyt3jxYoA2R3OsOedKtZmKwsxWA7WBNeX80J0i2yXl/LhhpeerePR8FY+er+LTc1Y8er6KR89X8ej5Kh6/nq82wG7nXNviHqggHzBmlg7gnNNw/BHQ81U8er6KR89X8ek5Kx49X8Wj56t49HwVTxifL82RFxEREREJIQV5EREREZEQUpAXEREREQkhBXkRERERkRBSkBcRERERCSGtWiMiIiIiEkIakRcRERERCSEFeRERERGREFKQFxEREREJIQV5EREREZEQUpAXEREREQkhBXkRERERkRBSkBcRERERCSEF+YAwsxZm9pyZbTSzg2a2xsxGmVk9v3sLGjO70sz+bWbTzWy3mTkze8nvvoLKzBqY2RAze9vMVpjZATPbZWYzzOxmM9P7QAFm9oiZfWpm6yLP13Yz+8rM/mxmDfzuL+jMbFDk/6UzsyF+9xM0kfd3V8TXD373F1RmdqaZvWlmmyI/JzeZ2RQzG+B3b0FhZjcd4rWV95Xjd59BY2YXRl5L6yPv+avM7A0zO8Pv3g5HF4QKADM7FpgFNAbeBZYApwNnA0uBXs65bf51GCxm9jVwCrAXWA90Al52zl3va2MBZWbDgKeATcA04HugCXA5UAd4Exjo9GYQZWaZwHxgEbAFqAH0AE4DNgI9nHPr/OswuMysJfAtEAvUBG5xzv3H366CxczWAHWBUYXs3uuc+2f5dhR8ZvYn4C/AVuADvPezhsCpwDTn3L0+thcYZtYFuLSI3WcC5wCTnHMXlV9XwWZmjwD3AtuAd/BeY+2Bi4E44AbnXGAHCxXkA8DMJgP9geHOuX/nq6cAI4FxzrlhfvUXNGZ2Nl6AXwH0wQunCvJFMLNz8ILoJOdcbr56U2Au0BK40jn3pk8tBo6ZVXXOZRRS/yvwR+Ap59zt5d9ZsJmZAZ8AbYG3gHtQkP+FSJDHOdfG307CwcwGAq8DU4HLnXN7CuyPd85l+dJciJjZbLwBiUucc+/53U8QRH4ObgB+BDo757bk23c28Bmw2jnXzqcWD0sfqfvMzNrhhfg1wBMFdv8Z2AcMMrMa5dxaYDnnpjnnlmsE+cg45z5zzr2fP8RH6j8AYyPfnlXujQVYYSE+4vXItkN59RIyw/FG/AbjvXeJlEhk6t8jwH7g2oIhHkAh/vDM7CS8EL8BmORzO0HSGi8Lz8kf4sHLGsAeoJEfjR2pOL8bEM6JbKcUErT2mNlMvKDfA/i0vJuTCi/vB2C2r12Ex68i2wW+dhFAZnY88DAw2jmXFvkkSIqWYGbXA63wfulZAKQ55zR/+ed64n3CMxHYYWYXAicBGcBc59xsP5sLkaGR7bN6jf3MciATON3MGjrntubtMLMkoBbedJvAUpD333GR7bIi9i/HC/IdUZCXUmRmccANkW8/9rOXoDKze/DmedfBmx/fGy9wPexnX0ETeS1NwDv/4o8+txMWTfGes/xWm9lg51yqHw0FVLfIdjPeeSsn599pZml4UwN/LO/GwsLMqgHXA7mAprnl45zbbma/A1KARWb2Dt5c+WPx5sh/wk+/BAWSgrz/6kS2u4rYn1evWw69SOXyMN7I1ofOucl+NxNQ9+CdGJznY+AmhYZfeADvpMPezrkDfjcTAs8D04GFeB/dtwPuBG4FPjKzM5xz3/jYX5A0jmyHAauBvsAcvCkR/wLOA95A0wMP5Sq8DDFJJ+n/knNuVOS8leeAW/LtWgGMLzjlJmg0Rz74LLLVfHApNWY2HLgbb4WkQT63E1jOuabOOcMbPb0cL3B9ZWZd/e0sOMzsdLxR+H9pmsORcc49FDl3ZbNzbr9z7rvIggYpQDXgQX87DJTYyNbwRt4/dc7tdc4tBC7DW/igTxiWCfTRrZHtOF+7CCgzuxdv6tZ4vJH4GkAisAp42cz+4V93h6cg77+8Efc6ReyvXeB2IiViZncAo/GWVjzbObfd55YCLxK43sab5tYAeNHnlgIh35SaZcD9PrdTEeSdfJ7kaxfBsiOyXVXwU4rIpz95nyaeXq5dhYSZnYB3nsF64EOf2wkcMzsL72Tq95xzyc65VZFfrufj/aK4Abg7sjBJICnI+29pZNuxiP15q2MUNYde5IiZ2QjgceA7vBCvi88Ug3NuLd4vQCeaWUO/+wmAmnjvXccDGfkvOoO36hbAM5FaYWumy8/lfYSvVcp+kvczcmcR+/OCfrVy6CWMdJLroeWtpz+t4A7n3H68JZpj8KYOBpLmyPsv78XT38xiCqzzXQvoBRwAvvCjOak4Iif0PAx8DfTLf3a+FMsxka1+KMJB4Nki9nXF++E3Ay+MadrN4eVND1nlaxfBkoa3qlYHM6vinMsssP+kyHZNuXYVAmZWFW/qZC5F/z+t7BIi26KWmMyrF3zdBYZG5H3mnFsJTAHaAHcU2P0Q3sjMi845rcksR83M7scL8enAuQrxRTOzTpGLhBSsx0QuCNUYmOWc2/HLoysX59wB59yQwr6AvAvOvBCpveZnr0FhZieaWf1C6q3xPi0DCOxVJMtb5L3qNbzppw/k32dm/fBOdt2FVt4qzECgHt6CBjrJtXDTI9tbzax5/h1mdgHeYGoGMKu8GztSGpEPhtvxXiRjzOxcYDHQHTgbb0rNfT72Fjhmdik/XYI6L3CdYWbjI3/e6py7p9wbCygzuxH4X7wR5OnAcO8CnD+zxjk3vpxbC6rzgUcjy9qtxFuKrAneVYTbAT/w85UNRIpjIPB7M5uGtwrLHrwT7C4EquLNY/6nf+0FUjLez8T7Imt7z8VbteYyvPe1W5xzRU29qczyTnJ92tcugm0i3hWD+wKLzextvPf44/Gm3Rjwe+fcNv9aPDQF+QBwzq00s9Pwwtb5wABgEzAGeEgnI/5CF+DGArV2kS+AtXjLBoqnbWQbC4wo4japeGfsi/em/jTeSMwpeMu27cP7pXoCMEb/J6UEpuFdP+RUvKk0NfDmf8/Ae31N0FWrf845t8XMugN/wgvvPfB+AZoE/N05p6mnBUQu0NYbneR6SM65XDMbgDcj4hotQZ5+AAAAsUlEQVS811d1YDve8zbGOTfFxxYPy/R+ISIiIiISPpojLyIiIiISQgryIiIiIiIhpCAvIiIiIhJCCvIiIiIiIiGkIC8iIiIiEkIK8iIiIiIiIaQgLyIiIiISQgryIiIiIiIhpCAvIiIiIhJCCvIiIiIiIiGkIC8iIiIiEkIK8iIiIiIiIaQgLyIiIiISQgryIiIiIiIhpCAvIiIiIhJCCvIiIiIiIiGkIC8iIiIiEkL/D7Gk++NSx0BCAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 248,
       "width": 377
      },
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "## test\n",
    "\n",
    "# x축, y축 그리기\n",
    "plt.axvline(x=0, c='black')\n",
    "plt.axhline(y=0, c='black')\n",
    "\n",
    "# y = wx + b 일차함수 그리기\n",
    "x = np.linspace(1, 8, 9)\n",
    "y = model(x, w=-20, b=140) # y = -20x + 140\n",
    "plt.plot(y)\n",
    "\n",
    "# 나의 (x, y) 점 찍기\n",
    "x_data = [2, 5, 6]\n",
    "y_data = [100, 40, 20]\n",
    "plt.scatter(x_data, y_data, c='r', s=50)\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([120. , 102.5,  85. ,  67.5,  50. ,  32.5,  15. ,  -2.5, -20. ])"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x7f9c7e924150>]"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 248,
       "width": 362
      },
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# test\n",
    "plt.plot(x)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x7f9c80ba8c10>]"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAugAAAHwCAYAAAD0N5r7AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAAWJQAAFiUBSVIk8AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy8QZhcZAAAgAElEQVR4nOzdeXiU5aH38d89WckOhGTCvsg+iKzBpe5LYpdz9NRaEVQQrN1s7XJOX9vTavva077drLY9rYArorV2bw2uqHUhICAQ9n3NJISQkH2Zud8/ZhgRJxLIZJ6Z5Pu5Lq7nmvu55+bXhcyPh2ee21hrBQAAACA2uJwOAAAAAOB9FHQAAAAghlDQAQAAgBhCQQcAAABiCAUdAAAAiCEUdAAAACCGUNABAACAGEJBBwAAAGIIBR0AAACIIRR0AAAAIIZQ0AEAAIAYQkEHAAAAYkii0wGiyRizR1KWpL0ORwEAAEDPNlzScWvtiDN9Y68q6JKy+vTp02/8+PH9nA4CAACAnmvLli1qamo6q/f2toK+d/z48f3WrFnjdA4AAAD0YNOmTdPatWv3ns17uQcdAAAAiCEUdAAAACCGUNABAACAGEJBBwAAAGIIBR0AAACIIRR0AAAAIIZQ0AEAAIAYQkEHAAAAYggFHQAAAIghFHQAAAAghlDQAQAAgBhCQQcAAABiSEQLujHmY8aYPxpjyo0xLcHji8aYa89gjcHGmEeMMYeDa+w1xjxgjOkbyawAAABALEqM1ELGmO9I+oGkKkn/kFQuKVfSFEmXSnq+E2uMkvS2pDxJf5W0VdJMSV+RVGSMudBaezRSmQEAAIBYE5GCboy5QYFy/rKk6621daecT+rkUr9RoJzfZa196KT3/1zS3ZLul3RnJDIDAAAAsajLt7gYY1ySfiypUdLsU8u5JFlr2zqxzkhJV0vaK+nXp5z+nqQGSXONMeldzQwAAIDe4VBNk3ZUfKiexrRIXEG/QNIISc9JOmaM+bgkj6RmSauste90cp3Lg8cXrbX+k09Ya+uMMW8pUOBnSXrloxYyxqzp4NS4TmYBAABAnNpb1aCSMq+Wl5Vr/cFaXTk+T4tvneF0rE6LREE/8Z+2QtJaSZNOPmmMeUPSp621R06zztjgcXsH53coUNDH6DQFHQAAAL3Ljoo6lZR59fzGcm31fvCK+Rs7qlTf0q6MlIh9/bJbRSJlXvB4p6Q9kq6UVCppmKSfSbpG0h8U+KLoR8kOHms7OH9iPOd0gay108KNB6+sTz3d+wEAABDbrLXadPi4lpd5VVJWrl1HGsLOS3QZzRrZX9X1rb2qoCcEj0aBK+Xrg683GWOuU+CK+CXGmPPP4HaXcEzwaLuwBgAAAOKU32+1/mBN8PYVr/ZXN4adl5zo0sWjB6jY49aV4/OVndbZ55XEhkgU9GPB4+6TyrkkyVrbZIx5QdLtCjwu8aMK+okr5NkdnM86ZR4AAAB6OJ/f6t291Sop8+qFTV6V1zaHndcnKUGXj8tTkcety8blxc3V8nAikXxb8FjTwfkTBb5PJ9cZ08H50cFjR/eoAwAAoAdo8/m1cvdRlZR59eKmClXVt4Sdl5mSqCvG56nIU6BLxgxQn+SEsPPiTSQK+huS2iWNNsYkW2tbTznvCR73nmadFcHj1cYY18lPcjHGZEq6UFKTpJVdjwwAAIBY0tLu01s7q1Sy0auXtlSopjH8U7pz0pJ09YR8FXsKdME5/ZWS2DNK+cm6XNCttVXGmN9LulnSdyV958Q5Y8xVCnxJtFbS8uBYkqRRktqstbtOWmeXMeZFBZ7U8kVJoY2KJN0nKV3S76y14b8BAAAAgLjS1OrT69srVVLm1atbKlXX0h52Xm5Giq6ZmK9rJxWocEQ/JSZ0eSufmBapm3O+JqlQ0reNMRdLWqXAU1yuk+STtNBae+IWmEGStkjaJ2n4Ket8QdLbkh40xlwRnFco6TIFbm35doTyAgAAwAF1zW16dWullpd59dq2I2pq84WdNzA7Vdd43Cr2FGjasL5KcJmw83qiiBR0a22lMaZQgavn1ymwmVCdpH9K+h9rbaduSwleRZ8u6fuSiiRdK6lc0oOS7rPWVkciLwAAAKKntrFNL22p0PKycr2xo0qt7f6w84b2S1PxpEApnzw4W8b0nlJ+soh9vTVYnr8W/PVR8/bq/Ucmhjt/QNK8SOUCAABA9FXVt+jFTRUqKSvXO7uOqt0f/knZ5+RlqNjjVpHHrQkFWb22lJ8sfp8/AwAAgJjirW3WC5sCGwet2lOtDjq5JhRkqdjjVvEkt87Jy4xuyDhAQQcAAMBZO1DdGNrNc+3+jp66LU0ekhMo5R63hvVPj2LC+ENBBwAAwBnZfaReJcFSXnboeNg5xkgzhvVTUfD2lYE5p9sSBydQ0AEAAPCRrLXaVlGnko1eLS/zaltFXdh5CS6jWSP7qdhToKsn5isvMzXKSXsGCjoAAAA+xFqrjYdqVVIWKOV7qsJvRZOUYHTRObkq9hToqgn56pueHOWkPQ8FHQAAAJIkv99q3YFjKtnoVUmZV4dqmsLOS0l06dKxA1TsKdDl4/OUlZoU5aQ9GwUdAACgF2v3+bVqb7WWl3n1wiavKo63hJ2Xnpygy8bl6dpJBbp07AClJVMjuwv/zQIAAPQyre1+vbP7qEo2luvFzRWqbmgNOy8rNVFXTshXsadAHxudq9SkhCgn7Z0o6AAAAL1Ac5tP/9pRpZKycr28uULHm9vDzuufnqyrJ+aryFOg80f2V3KiK8pJQUEHAADooRpb27Vi6xGVlJVrxdZKNbT6ws7Ly0wJPQ5x5vB+SkyglDuJgg4AANCDHG9u06tbKlVSVq7Xth1RS7s/7LxBOX1Cu3lOGdJXLpeJclJ0hIIOAAAQ5441tOqlzRUqKSvXWzuPqtUXvpSPyE1XUXA3z0mDsmUMpTwWUdABAADiUGVds17cFCjlK3dXy+e3YeeNzc8MlPJJbo3Nz6SUxwEKOgAAQJw4XNOk5cGNg1bvq5YN38nlGZSlYk+Bij1ujRyQEd2Q6DIKOgAAQAzbd7RBJWWBjYPWH6jpcN7UoTkq9hSoyOPWkH5pUUyISKOgAwAAxJidlXWh3Tw3lx8PO8dlpJkj+qnYU6BrJrrlzk6Nckp0Fwo6AACAw6y12lx+XMuDV8p3VtaHnZfoMjp/VH8Vewp09cR85WakRDkpooGCDgAA4ABrrdYfrFVJWbmWl3m172hj2HnJiS5dPDpXRZ4CXTk+TzlpyVFOimijoAMAAESJz2+1Zt8xlZSV64Uyrw7XNoed1ycpQZeNG6AiT4EuH5enjBQqW2/C/9oAAADdqN3n18rd1YFSvqlCVfUtYedlpCTqivF5Kva4dcmYPPVJTohyUsQKCjoAAECEtbT79PbOoyopK9dLmyt0rLEt7LyctCRdNT5fxZPcuvCcXKUkUspBQQcAAIiI5jafXtt2RMvLyvXKlkrVtbSHnZebkayrJ7p1radAhSP7KSnBFeWkiHUUdAAAgLNU39KuFVsrtbzMq1e3VqqpzRd2XkF2qq6Z6Faxx63pw/spwcVunugYBR0AAOAM1Da26eUtFSop8+qNHUfU2u4PO29Ivz6h3TwnD86Ri1KOTqKgAwAAnMbR+ha9uDlQyt/eWaV2vw07b9SA9NBunhMHZskYSjnOHAUdAAAgjIrjzXphk1clG70q3XNUHXRyjS/IUrEncPvK6PzM6IZEj0RBBwAACDp4rDG0m+fa/cdkOyjlkwdnqyh4+8rw3PTohkSPR0EHAAC92p6qhtBunhsO1oadY4w0fVhfFQVvXxmU0yfKKdGbUNABAECvYq3V9or6UCnf6q0LOy/BZVQ4op+KPW5dM9GtvKzUKCdFb0VBBwAAPZ61VpsOH1dJWblKNnq1u6oh7LykBKMLz8lVscetqya41S89OcpJAQo6AADoofx+q3UHarS8rFwlZV4dPNYUdl5KokuXjBmg4kluXT4uX9l9kqKcFPggCjoAAOgxfH6rVXuqtbysXC9sqpD3eHPYeWnJCbpsXJ6KPW5dNjZP6SlUIsQO/t8IAADiWpvPr3d2HVVJmVcvbfaqqr417LzM1ERdNT5fRR63Lh4zQKlJCVFOCnQOBR0AAMSd5jaf3txRpZIyr17eUqHapraw8/qmJenqCW4VT3LrglG5Sk50RTkpcOYiUtCNMXslDevgdIW11h3NdQAAQM/T2Nqu17cd0fNlXq3YWqn6lvaw8/IyU3TNxMDGQTNH9FNiAqUc8SWSV9BrJT0QZrzeoXUAAECcq2tu06tbK1Wy0avXtlequc0fdt6gnD4qCu7mOXVoX7lcJspJgciJZEGvsdbeG0PrAACAOFTT2KoXN1doeZlXb+6oUqsvfCkf3j8ttJvnuYOzZQylHD0D96ADAADHHalr0YubvVpe5tXbu47K57dh543JzwiV8nHuTEo5eqRIFvQUY8wcSUMlNUjaIOkNa63PoXUAAEAMK69t0vIyr0rKvFq9t1o2fCfXxIFZunZSgYo8bo0akBHdkIADIlnQ3ZKePGVsjzFmnrX29WiuY4xZ08GpcWeQAwAARNiB6kaVlJXr+Y1evXegpsN5U4bmqNjjVtHEAg3tnxbFhIDzIlXQH5X0L0mbJNVJGinpS5LukFRijDnfWrs+iusAAIAYsbOyPrSb56bDx8POcRlpxvB+Kva4dY3HrYLsPlFOCcSOiBR0a+19pwyVSbrTGFMv6euS7pV0XRTXmRZuPHhlferp3g8AAM6etVZbvXUq2Rgo5Tsqwz+ILcFldMGo/iryuHX1BLcGZKZEOSkQm7r7S6K/VaBYXxwj6wAAgG5grdWGg7UqKfNqeVm59h5tDDsvOcGlj43OVZHHrasm5CsnLTnKSYHY190FvTJ4TI+RdQAAQIT4/VZr9h9TyUavXtjk1aGaprDzUpNcunRMnoonuXX5uDxlpiZFOSkQX7q7oJ8fPO6OkXUAAEAXtPv8WrWnOnClfJNXR+paws7LSEnU5ePyVOxx65KxA5SWzJOdgc7q8p8WY8xESeXW2upTxodJ+lXw5dKTxpMkjZLUZq3ddbbrAACA6Ght9+utXVVavtGrl7ZUqLqhNey87D5JumpCvoo9bl14Tq5SkxKinBToGSLx19kbJH3LGLNC0h4Fnr4yStLHJaVKel7ST0+aP0jSFkn7JA3vwjoAAKCbNLf59Pr2I1pe5tXLWypU19wedl5uRrKumuBWscet80f1V1KCK8pJgZ4nEgV9haSxkqYocCtKuqQaSW8q8DzzJ63taOuBblkHAACchYaWdq3YVqmSMq9WbK1UY2v4PQLdWakq8rhV5HFrxvB+SnCxmycQSV0u6MHNgzq9EZG1dq+kD/1JPtN1AABA19U2temVLRUqKfPqje1H1NLuDztvcN8+Kva4VTypQOcNzpGLUg50G76xAQBAL1Pd0KqXNnv1/Eav3t5VpTZf+H+gHjkgPVDKPQWaODBLxlDKgWigoAMA0AtUHm/WC5u8KinzqnRPtXz+8KV8nDtTxZ4CFU9ya3ReBqUccAAFHQCAHupQTZNKNpZreZlXa/YfU0ff5Dp3cLaKglfKR+Sy5QjgNAo6AAA9yN6qhtBunusP1oadY4w0bWjf0Bc9B/dNi3JKAB+Fgg4AQJzbUVGn5zd6VVJWrq3eurBzXEYqHNFf105y65qJbuVlpUY5JYDOoqADABBnrLXadPi4lpcFSvmuIw1h5yUlGF0wKlfFHreumpCv/hkpUU4K4GxQ0AEAiAN+v9V7B2tCpfxAdVPYecmJLl0yZoCKPW5dMT5f2X2SopwUQFdR0AEAiFE+v9W7e6uD95R75T3eHHZeWnKCLhubpyKPW5eNy1NGCh/vQDzjTzAAADGkzefXyt1HVVLm1YubvKqqbw07LzMlUVdOyFeRx61LxgxQalJClJMC6C4UdAAAHNbS7tObO6pUUubVy1sqVNPYFnZe37QkXTUhX8WeAl1wTn+lJFLKgZ6Igg4AgAOaWn16fXulSsq8enVLpepa2sPOG5CZomsmBkp54Yh+SkxwRTkpgGijoAMAECV1zW16dWullpd59dq2I2pq84WdNzA7Vdd43Lp2UoGmDu2rBBe7eQK9CQUdAIBuVNvYppe2VKhkY7n+taNKrT5/2HnD+qeFdvOcPDhbxlDKgd6Kgg4AQIRV1bfoxU0VKikr1zu7jqrdb8POG52XoWKPW0WeAo0vyKSUA5BEQQcAICK8tc1aXlaukjKvVu+tVgedXBMKslTscat4klvn5GVGNySAuEBBBwDgLB2obgxtHLR2f02H884bkhMo5Z4CDe2fFsWEAOIRBR0AgDOw60h9qJSXHToedo4x0oxh/VQ8ya1rJro1MKdPlFMCiGcUdAAAPoK1Vtsq6lSyMVDKt1fUh52X4DI6f2R/FXncunpivvIyU6OcFEBPQUEHAOAU1lptPFSrkjKvlpd5taeqIey85ASXLhqdqyKPW1eNz1ff9OQoJwXQE1HQAQCQ5PdbrTtwTM9vDJTyQzVNYeelJrl0yZgBKvYU6PLxecpKTYpyUgA9HQUdANBrtfv8WrW3WsvLvHphk1cVx1vCzktPTtDl4/NV7HHr0rEDlJbMxyeA7sNPGABAr9La7tfbu6q0vMyrFzdXqLqhNey8rNREXTXBrWKPWxeNzlVqUkKUkwLorSjoAIAer7nNp3/tqFJJWble3lyh483tYef1T0/W1RPzVeQp0Pkj+ys50RXlpABAQQcA9FANLe16bdsRlZSVa8XWSjW0+sLOy89KUdHEwG6eM4b3VWICpRyAsyjoAIAe43hzm17ZUqGSjV69vv2IWtr9YecNyukT3M2zQFOG5MjlMlFOCgAdo6ADAOLasYZWvbS5QiVl5XpzZ5XafDbsvJG56SoK7ubpGZQlYyjlAGITBR0AEHcq65r1wqYKLS8r18rd1fL5w5fysfmZKp4UKOVj8jMo5QDiAgUdABAXDtc0aXlZYDfPd/cdkw3fyTVpUHbwSrlbIwdkRDckAEQABR0AELP2HW1QSZlXJWVerT9Q0+G8acP6qtjj1jUT3RrSLy2KCQEg8ijoAICYsqOiLlTKt5QfDzvHZaSZI/qp2FOgaya65c5OjXJKAOg+FHQAgKOstdpcfjx4+4pXOyvrw85LdBldcE6uij1uXTUhX7kZKVFOCgDRQUEHAESdtVbvHagJlfL91Y1h5yUnunTx6FwVewp05fh8ZaclRTkpAEQfBR0AEBU+v9WafcdUUlauF8q8OlzbHHZen6QEXTZugIo8Bbp8XJ4yUvioAtC78FMPANBt2n1+rdxdHSjlmypUVd8Sdl5mSqKuGJ+nIk+BLhkzQH2SE6KcFABiBwUdABBRLe0+vb3zqJ7fWK6XtlSoprEt7LyctCRdNT5fxZPcuvCcXKUkUsoBQIpQQTfG7JU0rIPTFdZa9xmsNVjS9yUVSeovqVzSXyTdZ6091sWoAIBu0NTq0+vbj2h5Wble2VKpupb2sPNyM1J0zcR8FXsKVDiyn5ISXFFOCgCxL5JX0GslPRBmPPzX8cMwxoyS9LakPEl/lbRV0kxJX5FUZIy50Fp7NAJZAQBdVN/Srle3Vmp5WblWbD2ipjZf2HkF2anBjYMKNG1YXyW42M0TAD5KJAt6jbX23i6u8RsFyvld1tqHTgwaY34u6W5J90u6s4u/BwDgLNU2tunlLRUqKfPqjR1H1NruDztvaL80FXvcKvK4NXlwjlyUcgDotJi5B90YM1LS1ZL2Svr1Kae/J+kOSXONMV+31jZEOR4A9FrHGlq1fFPgcYhv76xSu9+GnXdOXkaolE8oyJIxlHIAOBuRLOgpxpg5koZKapC0QdIb1trw/+b5YZcHjy9aaz9wScZaW2eMeUuBAj9L0isftZAxZk0Hp8Z1MgsAQNIb24/oi0+t7fCe8vEFWSr2uFXscWt0fmaU0wFAzxTJgu6W9OQpY3uMMfOsta934v1jg8ftHZzfoUBBH6PTFHQAQNe9urVCdz65Vq2+D97GMnlITuBK+US3huemO5QOAHquSBX0RyX9S9ImSXWSRkr6kgK3pZQYY8631q4/zRrZwWNtB+dPjOecLoy1dlq48eCV9amnez8A9HYvbPLqS8vWqs0XuJ2lIDtVCz42UkUetwbl9HE4HQD0bBEp6Nba+04ZKpN0pzGmXtLXJd0r6bou/jYnbmYMf/MjACAi/rmhXF95Zl3oXvMh/fro6YWzNLhvmsPJAKB36O4H0P42eLy4E3NPXCHP7uB81inzAAAR9pd1h/Tlp9eGyvmI3HQ9+7nzKecAEEXd/RSXyuCxMzcpbgsex3RwfnTw2NE96gCALvjDuwf0n3/cIBv8d8pRA9L19MJZystKdTYYAPQy3X0F/fzgcXcn5q4IHq82xnwglzEmU9KFkpokrYxcPACAJC0r3a9vPvd+OR+bn6ln7jifcg4ADuhyQTfGTDTG9AszPkzSr4Ivl540nmSMGRfcNTTEWrtL0ouShkv64inL3afAVfgneAY6AETW42/v1T1/3hh6PaEgS0/fMUsDMlMcTAUAvVckbnG5QdK3jDErJO1R4CkuoyR9XFKqpOcl/fSk+YMkbZG0T4EyfrIvSHpb0oPGmCuC8wolXabArS3fjkBeAEDQojd26/7nt4ReTx6crSfmFyo7LcnBVADQu0WioK9Q4BnmUxS4pSVdUo2kNxV4LvqT1tpOPXnFWrvLGDNd0vclFUm6VlK5pAcl3WetrY5AXgCApF+v2KmfvLAt9Hrq0Bw9Nn+mslIp5wDgpC4X9OAmRJ3ZiOjE/L16/5GJ4c4fkDSvq7kAAOFZa/XAyzv0y1d2hMZmjuinR26boYyU7n52AADgdPhJDAC9iLVWP3lhm37z2q7Q2AWj+mvxrdOVlsxHAgDEAn4aA0AvYa3V/f/cosVv7gmNXTJmgH43d5pSkxIcTAYAOBkFHQB6Ab/f6r6/b9Lj7+wLjV05Pk+/vnmqUhIp5wAQSyjoANDD+f1W3/7LRj296kBorGiiWw/eNEXJid29HQYA4ExR0AGgB/P5rf7rjxv03JqDobFPTh6on39mspISKOcAEIso6ADQQ7X7/Pr6H9brr+8dDo1dP2WQfnLDZCW4OnyYFgDAYRR0AOiB2nx+ffWZ9/TPjeWhsRunD9EPr59EOQeAGEdBB4AepqXdpy8tW6eXNleExubMGqrvf8ojF+UcAGIeBR0AepDmNp8+v3SNVmw7Ehqbd+FwffcTE2QM5RwA4gEFHQB6iKZWn+548l39a0dVaOxzl4zUt4rGUc4BII5Q0AGgB2hoadftj6/Wyt3VobG7Lj9Hd181hnIOAHGGgg4Aca6uuU3zHl2td/cdC419/aox+vIVox1MBQA4WxR0AIhjtU1tuvWRVXrvQE1o7FvF43TnJaMcTAUA6AoKOgDEqZrGVs1ZUqqyQ8dDY//9iQm6/aIRDqYCAHQVBR0A4tDR+hbdvLhUW711obEf/LtHc2cNczAVACASKOgAEGcq65p186JS7aislyQZI/3o+km6ccZQh5MBACKBgg4AccRb26zZi1Zqd1WDJMllpJ/eMFnXTx3scDIAQKRQ0AEgThyqadLsRSu172ijJCnBZfSLG8/TpyYPdDgZACCSKOgAEAcOVDfqpkUrdfBYkyQp0WX00E1TVDypwOFkAIBIo6ADQIzbU9Wg2YtWqry2WZKUnODSb26eqisn5DucDADQHSjoABDDdlbWafaiUlXWtUiSUhJd+t3cabp0bJ7DyQAA3YWCDgAxapu3TjcvXqmq+lZJUmqSS0tunaELz8l1OBkAoDtR0AEgBm06XKs5i0t1rLFNkpSWnKBHbpuhWSP7O5wMANDdKOgAEGM2HKzR3CWrVNsUKOeZKYl6bP4MTRvWz+FkAIBooKADQAxZs++Ybntklepa2iVJWamJevL2Qk0ekuNwMgBAtFDQASBGlO4+qvmPrVZDq0+SlJOWpKW3F8ozKNvhZACAaKKgA0AMeGtnlRY8/q6a2gLlvH96sp5aWKhx7iyHkwEAoo2CDgAOe337Ed3xxLtqafdLkgZkpmjZgkKNzs90OBkAwAkUdABw0CtbKvT5pWvV6guUc3dWqpYtLNTIARkOJwMAOIWCDgAOWV7m1ZefXqs2n5UkDcrpo6cXztLQ/mkOJwMAOImCDgAO+Pv6w/rq79+Tzx8o50P7pWnZwkIN7ks5B4DejoIOAFH253UH9fVn1yvYzTUyN11PLSxUQXYfZ4MBAGICBR0AoujZ1Qf0X3/aIBss56PzMvTUgkLlZaU6GwwAEDMo6AAQJUtX7tN3/lIWej3OnamlCwqVm5HiYCoAQKxxddfCxpi5xhgb/LXgDN5nP+LXyu7KCwDd6dG39nygnHsGZenphbMo5wCAD+mWK+jGmCGSHpJUL+lsnhW2T9JjYcYPdiEWADjid6/v0v+UbA29njwkR0/Mn6nsPkkOpgIAxKqIF3RjjJH0qKSjkv4k6Rtnscxea+29kcwFAE741as79NMXt4deTxvWV4/Nm6HMVMo5ACC87rjF5S5Jl0uaJ6mhG9YHgJhnrdXPX9r+gXJeOKKfnpg/k3IOAPhIEb2CbowZL+lHkn5prX3DGHP5WS6VY4yZL8ktqVbSGmst958DiAvWWv14+Tb99vVdobGLzsnVolumq09ygoPJAADxIGIF3RiTKOlJSfsl3dPF5SZLWnLK+uslzbXWbuxEljUdnBrXxVwA8JGstfrBP7bokbf2hMYuHTtAv50zTalJlHMAwOlF8haX70qaIuk2a21TF9b5uaQLJQ2QlClphqTnFCjtrxpjBnU1KAB0B7/f6rt/3fSBcn7l+Hz9bi7lHADQeRG5gm6MmanAVfOfWWvf6cpa1tqvnzL0rqQbjDHPSfoPBb50evdp1pjWQc41kqZ2JR8AhOP3W93z5416ZvWB0Fixx61ffnaKkhO77Ym2AIAeqMufGifd2rJd0n93OVHHfhs8XtyNvwcAnDGf3+qbz234QDn/1OSBeugmyjkA4MxF4pMjQ9IYSeMlNZ+8sZCk7wXnLAqOPdCF3+dI8JjehTUAIKLafX7d/fv39Me172/T8B9TB+sXN56nxATKOQDgzEXiFiwr3lIAACAASURBVJcWnfKFzpNMVeC+9DclbZPUldtfZgWPu7uwBgBETGu7X195Zp1Kyryhsc/OGKIfXjdJLpdxMBkAIJ51uaAHvxC6INw5Y8y9ChT0x621i08aT5M0VFKjtXb/SeNTJW2z1jacss65ku4Pvlza1cwA0FUt7T598al1enlLRWhs7qxhuu9TEynnAIAuifhOop00U9IKSa9LuvSk8bskXW+MeVXSAQWuzo+TVCQpQdIiSU9HNSkAnKK5zac7l67Ra9uOhMZuv2iEvvPx8QpspgwAwNlzqqB35C+SsiSdq8BupKmSjkoqkbTIWvs3B7MBgJpafVr4xLt6c2dVaOzzl47Sf14zlnIOAIiIbi3o1tp7Jd0bZvw1SR/6JLPW/kWBkg4AMaehpV3zH1ut0j3VobG7rhitu68cTTkHAERMrF1BB4CYdLy5TfMeXa01+46Fxr5x9Rh96fLRDqYCAPREFHQAOI3axjbd8ugqrT9QExq759pxuuPiUQ6mAgD0VBR0APgIxxpaNWdJqTYdPh4a+94nJ2jehSMcTAUA6Mko6ADQgar6Fs1ZXKqt3rrQ2P3XeXRz4TAHUwEAejoKOgCEUXm8WbMXl2pnZb0kyRjpx9efq8/MGOJwMgBAT0dBB4BTlNc2afaiUu2pCuyZ5jLSzz4zWddNGexwMgBAb0BBB4CTHDzWqNmLSrW/ulGSlOAyeuDG8/TJyQMdTgYA6C0o6AAQtP9oo25atFKHapokSUkJRg/dNFVFHrfDyQAAvQkFHQAk7T5Sr9mLSuU93ixJSk5w6X/nTNUV4/MdTgYA6G0o6AB6vR0VdZq9uFRH6lokSSmJLj18y3RdMmaAw8kAAL0RBR1Ar7bVe1w3LyrV0YZWSVKfpAQtuXW6Ljgn1+FkAIDeioIOoNcqO1SruUtKdayxTZKUnpygR+fN1MwR/RxOBgDozSjoAHql9w7U6JYlpTre3C5JykxJ1GPzZ2rasL4OJwMA9HYUdAC9zpp91br1kdWqbwmU86zURC1dUKhzB+c4nAwAAAo6gF6mdPdRzXtstRpbfZKkvmlJWrqgUBMHZjucDACAAAo6gF7jrZ1Vuv3x1Wpu80uScjOS9dSCWRrrznQ4GQAA76OgA+gVXttWqc89uUYt7YFynpeZomULZ+mcvAyHkwEA8EEUdAA93subK/SFp9aq1Rco5wXZqVq2cJZG5KY7nAwAgA+joAPo0Uo2luvLT69Tu99Kkgbl9NEzd8zSkH5pDicDACA8CjqAHutv6w/r7t+/J1+wnA/rn6ZlC2dpUE4fh5MBANAxCjqAHumPaw7qm8+tV7Cba+SAdC1bMEvu7FRngwEAcBoUdAA9zu9X79e3/rRRNljOR+dl6KmFhcrLpJwDAGIfBR1Aj/LkO3v133/dFHo9zp2ppxYUqn9GinOhAAA4AxR0AD3Gkjf36Af/2Bx67RmUpSfnF6pverKDqQAAODMUdAA9wm9f36UflWwNvT5vSI4enz9T2X2SHEwFAMCZo6ADiHsPvrJDP39pe+j1jOF99chtM5SZSjkHAMQfCjqAuGWt1c9f2q6HXt0ZGjt/ZH8tvnW60lP48QYAiE98ggGIS9Za/ahkq373xu7Q2MdG5+rhudPVJznBwWQAAHQNBR1A3LHW6vv/2KxH39obGrts7AD975xpSk2inAMA4hsFHUBc8futvvu3Mi1duT80dtWEfP1q9hSlJFLOAQDxj4IOIG74/Fb3/Gmjfv/ugdDYxycV6IHPnqekBJeDyQAAiBwKOoC40O7z6z+f26A/rTsUGvu38wbqZzdMViLlHADQg1DQAcS8Np9fd//+Pf1jQ3lo7NPTBuvH/3GuElzGwWQAAEQeBR1ATGtt9+uup9dp+SZvaOymmUN1/7975KKcAwB6oG75d2FjzFxjjA3+WnCG751gjHnWGFNpjGk2xmwzxtxnjOnTHVkBxK6Wdp++8NSaD5TzW88fph9eRzkHAPRcEb+CbowZIukhSfWSMs7wvYWSXpWUJOk5SQckXS7pu5KuMMZcYa1tiWxiALGouc2nO55coze2HwmNLfzYCN1z7XgZQzkHAPRcEb2CbgKfmo9KOirpt2f43oTge9MkfdpaO9ta+1+SCiX9UdKFku6OZF4AsamxtV3zH1v9gXL+hUtHUc4BAL1CpG9xuUuBK97zJDWc4XsvkTRe0hvW2r+dGLTW+iX9Z/DlnYZPZ6BHq29p122PrNbbu46Gxr565Wh985qxlHMAQK8QsYJujBkv6UeSfmmtfeMslrg8eFx+6glr7W5J2yUNkzTyrEMCiGnHm9t0y5JSrdpbHRr75jVj9dUrx1DOAQC9RkTuQTfGJEp6UtJ+Sfec5TJjg8ftHZzfIWlM8Neu0+RZ08GpcWcXDUB3q21s0y2PlGr9wdrQ2Hc+Pl4LPsbfyQEAvUukviT6XUlTJF1krW06yzWyg8faDs6fGM85y/UBxKjqhlbNWVyqzeXHQ2P3fWqibr1guHOhAABwSJcLujFmpgJXzX9mrX2n65E6/q2CR3u6idbaaWEXCFxZnxrJUAC65khdi+YsLtW2irrQ2A+vm6TZhUMdTAUAgHO6VNBPurVlu6T/7mKWE1fIszs4n3XKPABxrvJ4s25atFK7jgS+U26M9OP/OFefmT7E4WQAADinq18SzVDgnvDxkppP2pzISvpecM6i4NgDp1lrW/A4poPzo4PHju5RBxBHymubdOPD75dzl5F+8ZnzKOcAgF6vq7e4tEha0sG5qQrcl/6mAuX7dLe/vCrp25KKJP3PySeMMSMVKO77JO3uQl4AMeBAdaNmL16pA9WBr6wkuox++dkp+vi5BQ4nAwDAeV0q6MEvhC4Id84Yc68CBf1xa+3ik8bTJA2V1Git3X/SW16XtEXSxcaYT514FroxxiXpx8E5v7XWnvYedACxa9/RBs1eVKpDNYFynpRg9KvZU3XNRLfDyQAAiA2ReorLmZgpaYUChfzSE4PWWp8xZp4CV9KfM8Y8p8BjG6+QNF3SW5J+EfW0ACJm15F6zV60UhXHWyRJyYku/XbOVF0+Lt/hZAAAxI5I7yTaJdbaUkkzJP1V0tWS7lbgS6Pfl3SVtbbFwXgAumBHRZ1u/N375Twl0aXFt0ynnAMAcIpuu4Jurb1X0r1hxl/T+49MDPe+zZJu6K5cAKJvS/lxzVlcqqMNrZKkPkkJWnLbdF0wKtfhZAAAxB4nbnEB0IuUHarVnCWlqmlskySlJyfosfkzNWN4P4eTAQAQmyjoALrNuv3HdMsjq1TX3C5JykxN1OPzZ2rq0L4OJwMAIHZR0AF0i3f3Vuu2R1erviVQzrP7JGnp7YWaNLijvcgAAIBEQQfQDd7ZdVS3P75aja0+SVK/9GQtvb1QEwZmneadAACAgg4got7cUaUFT6xWc5tfkpSbkaJlCws1Jj/T4WQAAMQHCjqAiFmxtVKfW7pGre2Bcp6flaJlC2dp1IAMh5MBABA/KOgAIuLFTV59cdlatfkCm/0OzE7VsoWzNDw33eFkAADEFwo6gC57fmO57np6ndr9gXI+uG8fPb1wlob0S3M4GQAA8YeCDqBL/vreIX3t2fXyBcv58P5pWrZwlgbm9HE4GQAA8YmCDuCsPbfmoL753HrZQDfXqAHpWrZwlvKzUp0NBgBAHKOgAzgrT6/ar3v+vDFUzsfkZ+ipBbM0IDPF2WAAAMQ5CjqAM/bEO3v13b9uCr0eX5ClpbfPVP8MyjkAAF1FQQdwRhb/a7f+7z+3hF6fOzhbT8yfqZy0ZAdTAQDQc1DQAXTab17bqf+3fFvo9ZShOXp8/kxlpSY5mAoAgJ6Fgg7gtKy1evCVnfrFy9tDYzOG99Wj82YqI4UfIwAARBKfrAA+krVWP31xm369Yldo7PyR/bXktulKS+ZHCAAAkcanK4AOWWv1PyVb9fAbu0NjHxudq4fnTlef5AQHkwEA0HNR0AGEZa3VfX/frMfe3hsau3xcnn5z81SlJlHOAQDoLhR0AB/i91t9569lWla6PzR2zcR8PXTTVCUnuhxMBgBAz0dBB/ABPr/Vt/64QX9YczA09vFzC/TAjecpKYFyDgBAd6OgAwhp9/n1jT+s11/eOxwau27KIP3k0+cqkXIOAEBUUNABSJLafH599ffv6Z8bykNjN0wbrB/9x7lKcBkHkwEA0LtQ0AGotd2vLz+9Vi9sqgiN3Vw4VD/4N49clHMAAKKKgg70cs1tPn3hqbV6dWtlaOy2C4bre5+cIGMo5wAARBsFHejFmlp9uuPJd/WvHVWhsTsuHqn/UzyOcg4AgEMo6EAv1djartsfe1fv7D4aGvvSZefo61ePoZwDAOAgCjrQC9W3tGv+o6u1am91aOxrV43RXVeMdjAVAACQKOhAr1Pb1KbbHl2ldftrQmP/VTROn790lIOpAADACRR0oBepaWzV3CWrtPFQbWjsOx8frwUfG+lgKgAAcDIKOtBLVDe06ubFpdpSfjw09v1/m6hbzh/uXCgAAPAhFHSgFzhS16KbF6/U9op6SZIx0g+vm6SbZg51OBkAADgVBR3o4SqON2v2opXadaRBkuQy0v/79GR9etpgh5MBAIBwKOhAD3a4pkmzF63U3qONkqQEl9HPPzNZ/3beIIeTAQCAjlDQgR7qQHWjblq0UgePNUmSEl1GD940RddOKnA4GQAA+CgUdKAH2lvVoNmLVupwbbMkKSnB6Nezp+rqiW6HkwEAgNNxRWIRY8yPjTGvGGMOGGOajDHVxph1xpjvGWP6n8E6e40xtoNf3khkBXq6nZX1uvHhd0LlPDnRpYdvmU45BwAgTkTqCvrdktZKeklSpaR0SbMk3SvpDmPMLGvtgU6uVSvpgTDj9RHICfRo27x1unlxqarqWyRJqUkuLb5lhi4anetwMgAA0FmRKuhZ1trmUweNMfdLukfS/5H0hU6uVWOtvTdCuYBeY/Ph45qzpFTVDa2SpLTkBD1y2wzNGtnpf8QCAAAxICK3uIQr50HPBo+jI/H7AAhvw8Ea3bRoZaicZ6Qk6on5MynnAADEoe7+kugng8cNZ/CeFGPMHElDJTUE3/uGtdYX6XBAT7B2/zHdumSV6lraJUmZqYFyPmVoX4eTAQCAsxHRgm6M+YakDEnZkqZLukiBgv2jM1jGLenJU8b2GGPmWWtf72SONR2cGncGOYCYt3pvtW57ZJUaWgN/f81JS9LS2wvlGZTtcDIAAHC2In0F/RuS8k96vVzSbdbaI518/6OS/iVpk6Q6SSMlfUnSHZJKjDHnW2vXRzAvELfe3lWl2x97V01tgXLeLz1ZTy0o1PiCLIeTAQCArohoQbfWuiXJGJMv6QIFrpyvM8Z8wlq7thPvv++UoTJJdxpj6iV9XYGnwlzXiXWmhRsPXlmferr3A7Huje1HtPCJd9XS7pck5WakaNnCQo3Jz3Q4GQAA6KqIfEn0VNbaCmvtnyVdLam/pCe6uORvg8eLu7gOEPde3VqhBY+/X87zs1L0+8/NopwDANBDdEtBP8Fau0/SZkkTjTFdeRBzZfCY3vVUQPx6YZNXn3tyjVp9gXI+KKePnv3c+Ro1IMPhZAAAIFK6+ykukjQweOzKU1jODx53dzELELf+uaFcX3lmndr9VpI0pF8fPb1wlgb3TXM4GQAAiKQuX0E3xowzxnxoD3FjjCu4UVGepLettceC40nB94w6Zf5EY0y/MOsMk/Sr4MulXc0LxKO/rDukLz+9NlTOR+Sm69nPnU85BwCgB4rEFfQiST8xxrwhaZekowo8yeUSBZ7C4pW08KT5gyRtkbRP0vCTxm+Q9C1jzApJexR4issoSR+XlCrpeUk/jUBeIK48++4B/dcfN8gGurlGDUjX0wtnKS8r1dlgAACgW0SioL8s6WFJF0qaLClHgQ2GtivwPPMHrbXVnVhnhaSxkqYocEtLuqQaSW8G13nS2hMVBegdlpXu1z1/3hh6PTY/U0sXFGpAZoqDqQAAQHfqckG31pZJ+uIZzN8ryYQZf11SpzYiAnqDx9/eq+/9bVPo9YSCLC1dUKh+6ckOpgIAAN0tGl8SBXCGFr2xW/c/vyX0evLgbD0xv1DZaUkOpgIAANFAQQdizK9X7NRPXtgWej11aI4emz9TWamUcwAAegMKOhAjrLV64OUd+uUrO0JjM0f00yO3zVBGCn9UAQDoLfjUB2KAtVY/eWGbfvPartDYBaP6a/Gt05WWzB9TAAB6Ez75AYdZa3X/P7do8Zt7QmOXjBmg382dptSkBAeTAQAAJ1DQAQf5/Vb3/X2THn9nX2jsyvF5+vXNU5WSSDkHAKA3oqADDvH7rb79l416etWB0FjRRLcevGmKkhO7vMkvAACIUxR0wAE+v9V//XGDnltzMDT2yckD9fPPTFZSAuUcAIDejIIORFm7z6+v/2G9/vre4dDY9VMG6Sc3TFaC60N7eAEAgF6Ggg5EUZvPr68+857+ubE8NHbj9CH64fWTKOcAAEASBR2ImpZ2n760bJ1e2lwRGpsza6i+/ymPXJRzAAAQREEHoqC5zafPL12jFduOhMbmXThc3/3EBBlDOQcAAO+joAPdrKnVpzuefFf/2lEVGvvcJSP1raJxlHMAAPAhFHSgGzW0tOv2x1dr5e7q0Nhdl5+ju68aQzkHAABhUdCBblLX3KZ5j67Wu/uOhca+dtUY3XXFaAdTAQCAWEdBB7pBbVObbn1kld47UBMa+1bxON15ySgHUwEAgHhAQQcirKaxVXOWlKrs0PHQ2H9/YoJuv2iEg6kAAEC8oKADEXS0vkU3Ly7VVm9daOwH/+7R3FnDHEwFAADiCQUdiJDKumbdvKhUOyrrJUnGSD+6fpJunDHU4WQAACCeUNCBCPDWNmv2opXaXdUgSXIZ6ac3TNb1Uwc7nAwAAMQbCjrQRYdqmjR70UrtO9ooSUpwGf3ixvP0qckDHU4GAADiEQUd6IID1Y367MMrdaimSZKU6DJ66KYpKp5U4HAyAAAQryjowFnaU9Wg2YtWqry2WZKUnODSb26eqisn5DucDAAAxDMKOnAWdlbWafaiUlXWtUiSkhNdenjuNF06Ns/hZAAAIN5R0IEztM1bp5sXr1RVfaskKTXJpSW3ztCF5+Q6nAwAAPQEFHTgDGw6XKs5i0t1rLFNkpSWnKBHbpuhWSP7O5wMAAD0FBR0oJM2HKzR3CWrVNsUKOeZKYl6bP4MTRvWz+FkAACgJ6GgA52wZt8x3fbIKtW1tEuSslIT9eTthZo8JMfhZAAAoKehoAOnUbr7qOY/tloNrT5JUk5akpbeXijPoGyHkwEAgJ6Igg58hLd2VmnB4++qqS1QzvunJ+uphYUa585yOBkAAOipKOhAB17ffkR3PPGuWtr9kqQBmSlatqBQo/MzHU4GAAB6Mgo6EMYrWyr0+aVr1eoLlHN3VqqWLSzUyAEZDicDAAA9HQUdOMXyMq++/PRatfmsJGlQTh89vXCWhvZPczgZAADoDSjowEn+vv6wvvr79+TzB8r50H5pWrawUIP7Us4BAEB0uCKxiDHmx8aYV4wxB4wxTcaYamPMOmPM94wxZ7SDizFmsDHmEWPMYWNMizFmrzHmAWNM30hkBTryp7UH9ZVn1oXK+cjcdP3+c7Mo5wAAIKoiUtAl3S0pXdJLkn4p6SlJ7ZLulbTBGDOkM4sYY0ZJWiNpnqRVkn4habekr0h650zLPtBZz64+oK//Yb2C3Vyj8zL0zB2zVJDdx9lgAACg14nULS5Z1trmUweNMfdLukfS/5H0hU6s8xtJeZLustY+dNI6P1fgLwH3S7ozIomBoKUr9+k7fykLvR7nztTSBYXKzUhxMBUAAOitInIFPVw5D3o2eBx9ujWMMSMlXS1pr6Rfn3L6e5IaJM01xqSfZUzgQx59a88HyvnEgVl6euEsyjkAAHBMpG5x6cgng8cNnZh7efD4orXWf/IJa22dpLckpUmaFbl46M1+9/ou3ff3zaHXk4fkaNmCWeqbnuxgKgAA0NtF9CkuxphvSMqQlC1puqSLFCjnP+rE28cGj9s7OL9DgSvsYyS9cpocazo4Na4TOdAL/OrVHfrpi+//X23asL56bN4MZaYmOZgKAAAg8o9Z/Iak/JNeL5d0m7X2SCfemx081nZw/sR4zllmA2St1S9e3qEHX9kRGisc0U+P3DZD6Sk8dRQAADgvoo3EWuuWJGNMvqQLFLhyvs4Y8wlr7douLm9O/DadyDEt7AKBK+tTu5gDccpaqx8v36bfvr4rNHbRObladMt09UlOcDAZAADA+7rlHnRrbYW19s8K3JLSX9ITnXjbiSvk2R2czzplHtBp1lr94B9bPlDOLx07QItvpZwDAIDY0q1fErXW7pO0WdJEY0zuaaZvCx7HdHD+xJNgOrpHHQjL77f67l836ZG39oTGrhyfr9/NnabUJMo5AACILdG46XZg8Og7zbwVwePVxhjXyU9yMcZkSrpQUpOklZGPiJ7K77e6588b9czqA6GxYo9bv/zsFCUndvdDjAAAAM5clxuKMWacMcYdZtwV3KgoT9Lb1tpjwfGk4HtGnTzfWrtL0ouShkv64inL3afATqVPWGsbupoZvYPPb/XN5zZ8oJx/avJAPXQT5RwAAMSuSFxBL5L0E2PMG5J2STqqwJNcLpE0UpJX0sKT5g+StEXSPgXK+Mm+IOltSQ8aY64IziuUdJkCt7Z8OwJ50Qu0+/z62rPr9bf1h0Nj108dpJ98erISXOYj3gkAAOCsSBT0lyU9rMAtKJMVeAxigwKF+klJD1prqzuzkLV2lzFmuqTvK1D8r5VULulBSfd1dh30bq3tfn3lmXUqKfOGxj47Y4h+eN0kuSjnAAAgxnW5oFtry/ThW1I+av5evf/IxHDnD0ia19Vc6J1a2n364lPr9PKWitDY3FnDdN+nJlLOAQBAXGBnFvQYzW0+3bl0jV7b9v6+WLdfNELf+fh4GUM5BwAA8YGCjh6hqdWnhU+8qzd3VoXGPn/pKP3nNWMp5wAAIK5Q0BH3GlraNf+x1Srd8/5XFO66YrTuvnI05RwAAMQdCjri2vHmNs17dLXW7DsWGvvG1WP0pctHf8S7AAAAYhcFHXGrtrFNtzy6SusP1ITG7rl2nO64eNRHvAsAACC2UdARl441tGrOklJtOnw8NPa9T07QvAtHOJgKAACg6yjoiDtV9S2as7hUW711obH/++8ezZk1zMFUAAAAkUFBR1ypPN6s2YtLtbOyXpJkjPTj68/VZ2YMcTgZAABAZFDQETfKa5s0e1Gp9lQ1SJJcRvrZZybruimDHU4GAAAQORR0xIWDxxo1e1Gp9lc3SpISXEYP3HiePjl5oMPJAAAAIouCjpi3/2ijblq0UodqmiRJSQlGD900VUUet8PJAAAAIo+Cjpi2+0i9Zi8qlfd4syQpOcGl/50zVVeMz3c4GQAAQPegoCNm7aio0+zFpTpS1yJJSkl06eFbpuuSMQMcTgYAANB9KOiISVu9x3XzolIdbWiVJPVJStCSW6frgnNyHU4GAADQvSjoiDllh2o1d0mpjjW2SZLSkxP06LyZmjmin8PJAAAAuh8FHTHlvQM1umVJqY43t0uSMlMS9dj8mZo2rK/DyQAAAKKDgo6YsWZftW59ZLXqWwLlPCv1/7d351FyVfWix7+/zCGEBMJMwBCmMM9JgIsSeCLIEgdAJIxmEJdoRNF7eTz1gl69vPsUEByQMcyiqJeLAqIyCEIIhEGmMDdhCIQQE8g87ffHOWmasqu7K32qq6r7+1mr1kntc87ev9q9z8mvTp2hD9dMHMPuWw6tcWSSJEldxwRddWHaS+8wYepDLF6+CoD11+nLtZPGsPPmQ2ocmSRJUtcyQVfN/e2FuUy86iGWrlgNwIbr9uO6SWPZYdPBNY5MkiSp65mgq6bufnYOp14zg2Urs+R848H9uX7yWLbdeN0aRyZJklQbJuiqmT8//RZfuu4Rlq/KkvPNhgzg+slj2XrDQTWOTJIkqXZM0FUTtz0xm6/c8CgrVycAthg6kF9+YSxbbrBOjSOTJEmqLRN0dbn/efwNvnbjY6zKk/MPDVuH6yePZYuhA2scmSRJUu2ZoKtL/WbGa3zzpsfJc3NGbjSI6yeNZdMhA2obmCRJUp0wQVeXufGhWZz52ydIeXK+3cbrct3kMWw82ORckiRpDRN0dYlrHmji2zc/1fx+1KaDuW7SGIat2792QUmSJNUhE3RV3eX3vcz3fv908/tdtliPayaMYf1B/WoYlSRJUn0yQVdVXXzPi5x728zm93tsOZSrJoxmyMC+NYxKkiSpfpmgq2ou/MvznPen55rf7ztifa44ZV8GDzA5lyRJKscEXYVLKXHen57jojtfaC4bO3IDLj95Xwb1d8hJkiS1xWxJhUopce5tM/nFX19qLjtwuw255MR9GNivdw0jkyRJagwm6CpMSonv/v5prvxbU3PZuB024ucn7M2AvibnkiRJHWGCrkKsXp34zv88ybXTZjWXfXSnTfjJ+D3p38fkXJIkqaNM0NVpq1YnzvrtE9z48KvNZUfsuhkXfG4P+vbuVcPIJEmSGo8Jujpl5arV/OtNf+e3j77eXPbJPTbnR8fsTh+Tc0mSpIp1OoOKiGERMSkifhcRL0TEkohYEBH3RcTEiOhwGxHRFBGpzOvNzsaqYq1YtZrTb3zsA8n50XsP57zP7mFyLkmStJaKOIJ+DPBzYDZwFzAL2AT4DHAZcHhEHJNSSh2sbwFwQSvlCwuIVQVZvnI1U254lNufev9703Gjt+L7n9qFXr2ihpFJkiQ1tiIS9OeAI4E/pJRWrymMiLOA6cBRZMn6bzpY3/yU0tkFxKUqWbZyFadd9wh/fmZOc9nJ+32Is4/cmQiTc0mSpM7o9HkIKaU7U0q3tEzO8/I3gYvztwd1th3Vh6UrVjH56hkfSM4nGitKiQAAF19JREFU/cvWJueSJEkFqfZFoivy6coK1ukfEScAWwGLgL8Df00prepoBRExo8ysURXEoRKLl69k0lUPc/+L7zSXfemgbfjmx3YwOZckSSpI1RL0iOgDnJS/vb2CVTcFrikpezkiPp9SuqeQ4FSxhctWMuHKh5jeNK+57PT/tR1fPWQ7k3NJkqQCVfMI+rnALsCtKaU/dnCdK4F7gaeA94CRwJeBLwC3RcR+KaXH26skpbR3a+X5kfW9OhiLcu8uXcEpV0znkVnzm8u++bEdOG3ctjWMSpIkqXuqSoIeEVOAM4CZwIkdXS+ldE5J0ZPAFyNiYV7f2cCnCwpTHbBg8QpOuuJBHn9tQXPZt47YkUkHjqxhVJIkSd1X4TerjojTgB8DTwPjUkrz2lmlI9ZcbPrhAupSB81btJzjLp32geT8nCN3NjmXJEmqokKPoEfE6cD5ZEe+D0kpzWlnlY5aU8+ggupTO95+bxknXPYgz771XnPZDz69K+PHbFXDqCRJkrq/whL0iPg3svPOHwM+mlKaW1TdwH759KUC61QZc95dynGXTuPFtxcBEAH/96jd+Ow+W9Y4MkmSpO6vkFNcIuLbZMn5DLIj52WT84joGxGjImKbkvKdI2KDVpb/EPCT/O21RcSr8mYvWMKxl7yfnPcKOP+ze5icS5IkdZFOH0GPiJOB7wKryO7AMqWV2+41pZSm5v/eAngGeAUY0WKZY4AzI+Iu4GWyu7hsAxwBDABuBX7Y2XhV3qvzFjP+smm8Om8JAL17BRd+bk+O2G2zGkcmSZLUcxRxisvW+bQ3cHqZZe4BprZTz13ADsCeZKe0DALmA/eR3Rf9mpRS6mywat0r7yxi/KUP8vr8LDnv2zv4yfi9+NjOm9Y4MkmSpJ6l0wl6SulsstsfdnT5JuCfDrHnDyHyQUQ18OLbCxl/6TTeencZAP369OLiE/bi4FGb1DgySZKknqeaDypSA3j+rfc47tIHmbswS8779+nFpSftw4e336jGkUmSJPVMJug92DOz3+WEyx7knUXLARjYtzeXn7IP+2+zYY0jkyRJ6rlM0HuoJ19fwAmXP8j8xSsAGNSvN1MnjGbfEf90Ix1JkiR1IRP0HujRWf/gpCum897SlQAMHtCHqyaMZq+t1q9xZJIkSTJB72EebprHKVc+xMJlWXI+ZGBfrp04hl2HD6lxZJIkSQIT9B7lgRffYeJVD7F4+SoANhjUj2snjmGnzdercWSSJElawwS9h7jv+blMuvohlq5YDcCG6/bn+slj2H6TwTWOTJIkSS2ZoPcAd82cw6nXzmD5yiw533hwf66fPJZtN163xpFJkiSplAl6N3fHU29y2vWPsGJV9hDWzYcM4PrJYxmx4aAaRyZJkqTWmKB3Y7c+MZspNzzKytVZcj58/YHcMHksW26wTo0jkyRJUjkm6N3UzY+9ztd/9Tir8uR8xLB1uH7yWDYfOrDGkUmSJKktJujd0E0zXuObNz1OynJzRm40iBsmj2WT9QbUNjBJkiS1ywS9m7lh+izO+t0Tzcn59pusy3WTxrLR4P61DUySJEkdYoLejVz9QBPfufmp5vc7brYe104czbB1Tc4lSZIahQl6N3HZvS/xH394pvn9bsOHcPWE0Qxdp18No5IkSVKlTNC7gZ/d/QL/dfuzze/33GooV00YzXoD+tYwKkmSJK0NE/QGllLiwr+8wPl/fq65bN8R63Pl50ezbn//tJIkSY3ILK5BpZT44R3P8tO7Xmwu22/kMC4/ZR/W6eefVZIkqVGZyTWglBL/edtMLvnrS81lB263IZecuA8D+/WuYWSSJEnqLBP0BpNS4pxbnmbq/U3NZQeP2pifHb8XA/qanEuSJDU6E/QGsnp14ls3P8n1D85qLvvYzptw0XF70a9PrxpGJkmSpKKYoDeIVasTZ/7m7/x6xmvNZUfsthkXHLsHfXubnEuSJHUXJugNYOWq1Xzj14/z34+90Vz26T234P8dvRt9TM4lSZK6FRP0Ordi1WpOv/Ex/vD32c1lx+w9nHOP2o3evaKGkUmSJKkaTNDr2PKVq/nKDY/wx6feai47fsxWfO+Tu9DL5FySJKlbMkGvU0tXrOJL1z3CnTPnNJedsv8I/v0TOxFhci5JktRdmaDXoSXLV/GFax7m3ufnNpd94cMj+d+HjzI5lyRJ6uZM0OvM4uUrmTj1YR546Z3msi+P25YzDt3e5FySJKkHMEGvIwuXrWTClQ8xvWlec9nXP7o9Uw7ZroZRSZIkqSuZoNeJBUtWcMqV03l01vzmsn89bAe+dNC2NYxKkiRJXc0EvQ7MX7ycEy+fzhOvL2gu+9YROzLpwJE1jEqSJEm1YIJeY+8sXMYJl0/nmdnvNpd995M7c9J+I2oXlCRJkmrGBL2G3n5vGcdfNo3n3loIQAT84NO7ctzorWocmSRJkmrFBL1G3np3KeMvncaLby8CoFfAfx29O0fvPbzGkUmSJKmWenW2gogYFhGTIuJ3EfFCRCyJiAURcV9ETIyIitqIiOERcUVEvBERyyKiKSIuiIj1OxtrvXhj/hKO/cUDzcl5717B+cfuYXIuSZKkQo6gHwP8HJgN3AXMAjYBPgNcBhweEceklFJ7FUXENsD9wMbAzcBMYDTwVeCwiDggpfROG1XUvVfnLea4S6fx2j+WANCnV3DhcXvy8V03q3FkkiRJqgdFJOjPAUcCf0gprV5TGBFnAdOBo8iS9d90oK6fkSXnU1JKF7Wo6zzga8D3gS8WEHNNNM1dxPhLp/HGgqUA9O0d/HT8Xhy686Y1jkySJEn1otOnuKSU7kwp3dIyOc/L3wQuzt8e1F49ETESOBRoAn5aMvvfgUXAiRExqLMx18ILcxZy7CUPNCfn/fr04pKT9jE5lyRJ0gd0OkFvx4p8urIDyx6cT+9oJdl/D/gbsA4wtrjwusazb77H5y6ZxlvvLgNgQN9eXHHyvozbYeMaRyZJkqR6U7W7uEREH+Ck/O3tHVhlh3z6XJn5z5MdYd8e+Es7bc8oM2tUB+Io1MtzF3HcpdOYt2g5AOv0683lJ+/LftsM6+pQJEmS1ACqeQT9XGAX4NaU0h87sPyQfLqgzPw15UM7G1hX2nzoAHYbnn20dfv34eoJo03OJUmSVFZVjqBHxBTgDLK7sJxYVLX5tN27waSU9i4T1wxgr4Li6ZD+fXpz8Ql7c8avHmfSgVuz51bd5m6RkiRJqoLCE/SIOA34MfA0cEhKaV4HV11zhHxImfnrlSzXMAb07c1Pj+/S7wWSJElqUIWe4hIRpwM/AZ4ExuV3cumoZ/Pp9mXmb5dPy52jLkmSJDW8whL0iPg34HzgMbLkfE6FVdyVTw8tffpoRAwGDgCWANM6G6skSZJUrwpJ0CPi22QXhc4gO61lbhvL9o2IUflTQ5ullF4E7gBGAKeVrHYOMAi4OqW0qIiYJUmSpHrU6XPQI+Jk4LvAKuBeYEpElC7WlFKamv97C+AZ4BWyZLylLwH3AxdGxCH5cmOAcWSntvyfzsYrSZIk1bMiLhLdOp/2Bk4vs8w9wNT2KkopvRgR+5Al/IcBHwdmAxcC51RwwakkSZLUkDqdoKeUzgbOrmD5Jt6/ZWJr818FPt/ZuCRJkqRGVM0HFUmSJEmqkAm6JEmSVEdM0CVJkqQ6YoIuSZIk1RETdEmSJKmOmKBLkiRJdcQEXZIkSaojJuiSJElSHTFBlyRJkuqICbokSZJURyKlVOsYukxEvDNw4MANdtxxx1qHIkmSpG7smWeeYcmSJfNSSsMqXbenJegvA+sBTTVoflQ+nVmDthuR/VUZ+6sy9ldl7K/K2F+Vsb8qY39Vppb9NQJ4N6W0daUr9qgEvZYiYgZASmnvWsfSCOyvythflbG/KmN/Vcb+qoz9VRn7qzKN2l+egy5JkiTVERN0SZIkqY6YoEuSJEl1xARdkiRJqiMm6JIkSVId8S4ukiRJUh3xCLokSZJUR0zQJUmSpDpigi5JkiTVERN0SZIkqY6YoEuSJEl1xARdkiRJqiMm6JIkSVIdMUFvR0QMj4grIuKNiFgWEU0RcUFErF9hPRvk6zXl9byR1zu82m13pc7GHBGDIuL4iLg+ImZGxKKIeC8iHo6IMyKiX5n1UhuvacV+yuIU8TeOiLvb+fwDyqy3U0T8KiLmRMTSiHg2Is6JiIHFfcJiFTC+Dmqnr9a8tixZr+HGV0QcHREXRcS9EfFuHuu1a1lXxf3eaOOriP6KiGERMSkifhcRL0TEkohYEBH3RcTEiPin/3MjYkQ74+uXxX3K4hQ1vvKxVO6zv9nGevtHxK0RMS8iFkfE3yPi9Ijo3blPVh0Fja9TOrDvWlWyTqOOr4q3pXbqa7h9WJ+uaKRRRcQ2wP3AxsDNwExgNPBV4LCIOCCl9E4H6hmW17M9cCfwS2AU8HngiIjYL6X0UjXa7koFxXwgcC0wD7gL+G9gA+ATwA+Bz0TEISmlpa2s+wowtZXy1yr/NNVXhb/xOWXKV7bS9hiysdgXuAl4FTgY+A5wSN7Hyypou+oK6q8myvfTrsBngKdSSq+2Mr+hxhfwLWB3YCFZjKPWppK16fdGHF8U01/HAD8HZpPtv2YBm5CNq8uAwyPimNT6EwIfJ9vflXpyLeLoCoWMr9wC4IJWyhe2tnBEfBL4DbAUuJHs/4tPAOcDB5D9HepNEf31GOX3XweSbWO3lZnfaOOrM9vSBzTsPiyl5KvMC/gjkICvlJSfl5df3MF6fpEvf15J+ZS8/PZqtd1o/QXsARwP9CspHwzMyOs5o5X1EnB3rfugRuPr7mxT7nC7vYGn8zaObFHei2xHlIAza90/1eqvNuq/Ia9nSjcZX+OA7YAADso/w7XV7vcGHl+d7i+y/8A/AfQqKd+ULMFIwFEl80bk5VNr3Qc1Gl9NQFMFy68HzAGWAfu0KB9AloQl4HO17p9q9Vcb9T9Qus01+PiqeFtqo66G3IfV/I9Qry9gZP5HeLmVATKY7FvwImBQO/UMAhbnyw8umdcrrz8BI4tuuxH7q502xudt3NLKvIZKoIrsLypP0A/O276njbiagKh1P3XV+AKGkR2NWwys3+jjq5X4D2LtEs6K+70Rx1dR/dVOnWfldV5UUj6CBkygiuovKk/QJ+RtXdXKvLJjr55eRY8vYJe8vteA3iXzGn58tfJ5W92WyizbsPswz0Ev7+B8ekdKaXXLGSml94C/AesAY9upZz9gIPC3fL2W9awG7sjfjqtC212pK2JekU//6ZSN3NCImBARZ0XEaRFRT/1TqvD+iohjI+LMiPh6RBweEf3bafv20hkpO9XqOeBDZDuielHt8XUK0B/4dUrpH2WWaaTxVZS16fdGHF9dob391+YRcWo+vk6NiN26KrA60D8iTsg/+1cjYlwb55KXHV/AX8m+ZO/fxv6vOzo1n16eUlpVZpnuNL7a25Zaath9mAl6eTvk0+fKzH8+n25fhXqKarsrdUXME/JpaztmyM7vuxz4PvAT4IGIeCwidu1Em9VSjf76JfCfwI+AW4FZEXF0F7VdbdWOeVI+/UUbyzTS+CpKT9l/VVVE9AFOyt+W2399FLiYbHxdDDweEXdFxFZdEGKtbQpcQ/bZLyA79/f5iPhIK8uWHV8ppZVkR0r70EO+AOYXLJ4ArCY7N7ucbjG+OrgttdSw+zAT9PKG5NMFZeavKR9ahXqKarsrVTXmiPgycBjZRTJXtLLIeWQXB21E9rPVvmTniu0O3BkRW6xNu1VUZH/dTHau3nCyX2tGkSXqQ4EbI+LwKrbdVaoWc54EjCK7OPT+Mos12vgqSk/Zf1XbuWSnIdyaUvpjybzFwPeAvYH189dHyC6MOwj4S0QM6rpQu9yVwCFkSfogsou1f0F2asZtEbF7yfKOrw/6LNlnvS21fnF7dxtfbW1LrWnYfZgJ+tqLfJpqUE9RbXeltY45Ij5DdlTlTbKLQlaULpNSOiOldH9KaW5KaWFK6eGU0jFkV/pvCHyjE7HXQof7K6V0fkrp9yml11NKS1NKz6aUzgLOINvGf1CttutIZ2L+Qj4te/S8G46vovSU/ddai4gpZNviTODE0vkppTkppe+klB5JKc3PX38FDgUeBLbl/V94up2U0jkppTtTSm+llBanlJ5MKX2R7EvxQODsCqvsUeOLdvZf3Wl8tbctrW21+bTu9mEm6OWt+YY0pMz89UqWK7KeotruSlWJOSI+RXbqxhzgoFRyO8oOuDiffrjC9aqtK/7Gl5Gdo7dHRAzu4raLVq3xtQFwFLCE7Cf2StXr+CpKT9l/VUVEnAb8mOyOEONSSvM6um5+usaaUxa66/hqS7lty/GVi4idgP3JLg69tZJ1G218dWJbath9mAl6ec/m03LnGG2XT8udo9SZeopquysVHnNEHAP8GngL+EhK6dl2VmnN2/m03n7Cq/rfOGX3il9zYXLLz+/4et/JZBeH/iqlNH8t4qrX8VWUnrL/KlxEnE52rcKTZAlF2YfutKG7j6+2zMmnpZ+97PjKz0/emuzARKUHcxpRRy4ObUtDjK9ObksNuw8zQS/vrnx6aOkTq/KjkQeQHXVr7ymC0/LlDig5ikle76El7RXZdlcqNOaIGE92X+o3yJLz59tZpZw1V2bX28666n/jiNiB7HzD94C5LWbdmU8Pa2WdkWQ7pVeorz6rVn9NzqeXrGVc9Tq+irI2/d6I46tQEfFvZA/NeYwsoZjTzirldPfx1Zb98mnpZy87vsiOBK8D3J/q70FYhYrsCdEnkl0cevlaVlP346uAbalx92HVvIdjo7+o/Ob2o4BRrdSz5kFFPyop79EPKmqjv04GVpEN/g91oN29aOX+18BuZIlpAsbXun+q0V9kdyrYopW6N+T9h3ZcUjKvrYcw/Jr6fZBMIeOrxfwD8/We6I7jqyTWg2jjvstkT8sbBWxTQL835PgqsL++na/7MLBBB9oaQ8mD2fLyg8nuzZ+A/WvdJ9XoL2Dn1vqI7BZ2z+d1nlUybz2yI78N9aCiosZXi2VOpMxzQbrL+KpkW+qO+7DIG1UrWnk87DNkg30c2U8b+6cWj4eNiASQUoqSeobl9WxP9s1sOrAj8Emyn/H2Tym92Jm260ER/RUR44A/k20IV5A9XrfU/JTSBS3WmUr2+N878+WXkW2oh5FtaJcCp6Y6G+wF9dcpZOcR3gO8SPbI662Aj5OdP/cw8NFUcvpGK48xnkV2J4V9yO4LW3ePYi9qe2wx/xqy25NNSSld1Ea7U2nM8fUp4FP5202Bj5F96b03L5ubUvpGvuwIstvTvZJSGlFST8X7ogYdX53ur4g4GZhKdoDhIlo/R7UppTS1xTp3kyWqd5OdSwzZl78192L+dkrpP9b+k1VHQf11NnAm2VHOl8l+7dsGOIIs4b4V+HRKaXkrbd9ElmD+kmy/dyTZ7fFuAj7bXbfHFvXdC/wLWQJ5Sxvt3k1jjq+KtqVuuQ+r9Teken8BW5LdBmo2sJzsZ40f0/q3/kSZJzoCG+TrvZLXM5ssAR1eRNv18upsf5E9MCa182oqWedTwG+BF4B3W/TvLZQ89rjeXgX0165kO7EngHfIHuAwj2yn/xVaOXLSYt2dyI4GzCVLOp8DzgEG1rpfqtVfLeatT/az5mJgaDttNuT4Irv7RYe2I95/2mBTmboq3hc12vgqor86UEei5Im0wETg92RPJlyY99Us4EbgwFr3S5X76yNkpzLOBOaT7b/eBv5Edq/rsk9qJDs14VbgH/m2/ATwNUqepFkvr4K3xx3z+a+293m78fj6wLbUgT5ruH2YR9AlSZKkOuJFopIkSVIdMUGXJEmS6ogJuiRJklRHTNAlSZKkOmKCLkmSJNURE3RJkiSpjpigS5IkSXXEBF2SJEmqIybokiRJUh0xQZckSZLqiAm6JEmSVEdM0CVJkqQ6YoIuSZIk1RETdEmSJKmOmKBLkiRJdcQEXZIkSaojJuiSJElSHfn/gz3UCR8ANCUAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 248,
       "width": 372
      },
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# test\n",
    "plt.plot(x_data)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "----------------------------------------------------------------------"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 248,
       "width": 375
      },
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.scatter(macbook['used_years'], macbook['price'])\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 7-4. 정확한 방정식이 아닌, \"최적의 방정식\"을 구하는 여정의 시작: 손실함수 정의하기"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "슝=3\n"
     ]
    }
   ],
   "source": [
    "w = 3.1\n",
    "b = 2.3\n",
    "print(\"슝=3\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 248,
       "width": 375
      },
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# 위의 w, b를 랜덤으로 지정했을 때 function의 그래프를 그려보기\n",
    "\n",
    "x = np.linspace(0, 5, 6)\n",
    "y = model(x, w, b) # y = 3.1x + 2.3\n",
    "plt.plot(y, c='r')\n",
    "\n",
    "plt.scatter(macbook['used_years'], macbook['price'])\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([2.65, 3.8 , 0.95, 4.5 , 4.45, 2.7 , 3.05, 2.25, 2.3 , 1.85, 1.5 ,\n",
       "       0.95, 2.6 , 0.55, 2.05, 0.75, 3.95, 2.8 , 2.05, 1.15, 1.35, 3.35,\n",
       "       1.35, 2.45, 2.4 , 0.1 , 3.45, 2.35, 0.7 , 1.05, 1.75, 4.1 , 4.5 ,\n",
       "       4.  , 2.  , 2.45, 2.5 , 2.1 , 3.  , 4.05, 2.7 , 2.15, 3.3 , 2.4 ,\n",
       "       2.85, 1.65, 1.55, 2.15, 3.4 , 1.55, 0.8 , 2.95, 2.4 , 1.4 , 1.65,\n",
       "       2.25, 3.85, 3.9 , 4.6 , 2.35, 4.05, 1.2 , 1.95, 1.6 , 2.  , 1.8 ,\n",
       "       3.75, 3.65, 2.35, 2.2 , 1.  , 4.15, 1.1 , 1.55, 2.95, 1.5 , 1.65,\n",
       "       1.1 , 3.25, 0.75])"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 실제 데이터와 얼마나 차이나는지 확인하기\n",
    "\n",
    "x = macbook[\"used_years\"].values\n",
    "x"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([10.515, 14.08 ,  5.245, 16.25 , 16.095, 10.67 , 11.755,  9.275,\n",
       "        9.43 ,  8.035,  6.95 ,  5.245, 10.36 ,  4.005,  8.655,  4.625,\n",
       "       14.545, 10.98 ,  8.655,  5.865,  6.485, 12.685,  6.485,  9.895,\n",
       "        9.74 ,  2.61 , 12.995,  9.585,  4.47 ,  5.555,  7.725, 15.01 ,\n",
       "       16.25 , 14.7  ,  8.5  ,  9.895, 10.05 ,  8.81 , 11.6  , 14.855,\n",
       "       10.67 ,  8.965, 12.53 ,  9.74 , 11.135,  7.415,  7.105,  8.965,\n",
       "       12.84 ,  7.105,  4.78 , 11.445,  9.74 ,  6.64 ,  7.415,  9.275,\n",
       "       14.235, 14.39 , 16.56 ,  9.585, 14.855,  6.02 ,  8.345,  7.26 ,\n",
       "        8.5  ,  7.88 , 13.925, 13.615,  9.585,  9.12 ,  5.4  , 15.165,\n",
       "        5.71 ,  7.105, 11.445,  6.95 ,  7.415,  5.71 , 12.375,  4.625])"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 현재 모델에 넣어보기\n",
    "\n",
    "prediction = model(x, w, b) # 현재 w = 3.1, b = 2.3\n",
    "prediction"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "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>used_years</th>\n",
       "      <th>price</th>\n",
       "      <th>prediction</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>2.65</td>\n",
       "      <td>175</td>\n",
       "      <td>10.515</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>3.80</td>\n",
       "      <td>63</td>\n",
       "      <td>14.080</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>0.95</td>\n",
       "      <td>273</td>\n",
       "      <td>5.245</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>4.50</td>\n",
       "      <td>133</td>\n",
       "      <td>16.250</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>4.45</td>\n",
       "      <td>106</td>\n",
       "      <td>16.095</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   used_years  price  prediction\n",
       "0        2.65    175      10.515\n",
       "1        3.80     63      14.080\n",
       "2        0.95    273       5.245\n",
       "3        4.50    133      16.250\n",
       "4        4.45    106      16.095"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# macbook 데이터프레임에 prediction 컬럼을 넣어서 비교해보기\n",
    "macbook['prediction'] = prediction\n",
    "macbook.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "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>used_years</th>\n",
       "      <th>price</th>\n",
       "      <th>prediction</th>\n",
       "      <th>error</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>2.65</td>\n",
       "      <td>175</td>\n",
       "      <td>10.515</td>\n",
       "      <td>164.485</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>3.80</td>\n",
       "      <td>63</td>\n",
       "      <td>14.080</td>\n",
       "      <td>48.920</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>0.95</td>\n",
       "      <td>273</td>\n",
       "      <td>5.245</td>\n",
       "      <td>267.755</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>4.50</td>\n",
       "      <td>133</td>\n",
       "      <td>16.250</td>\n",
       "      <td>116.750</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>4.45</td>\n",
       "      <td>106</td>\n",
       "      <td>16.095</td>\n",
       "      <td>89.905</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   used_years  price  prediction    error\n",
       "0        2.65    175      10.515  164.485\n",
       "1        3.80     63      14.080   48.920\n",
       "2        0.95    273       5.245  267.755\n",
       "3        4.50    133      16.250  116.750\n",
       "4        4.45    106      16.095   89.905"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 정답과 예측 간 차이 컬럼 만들기\n",
    "\n",
    "macbook['error'] = macbook['price'] - macbook['prediction']\n",
    "macbook.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### RMSE를 구하는 function 만들기"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "슝=3\n"
     ]
    }
   ],
   "source": [
    "def RMSE(a, b):\n",
    "    mse = ((a - b) ** 2).mean()  # 두 값의 차이의 제곱의 평균\n",
    "    rmse = mse ** 0.5        # MSE의 제곱근\n",
    "    return rmse\n",
    "print(\"슝=3\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[10.515 14.08   5.245 16.25  16.095 10.67  11.755  9.275  9.43   8.035\n",
      "  6.95   5.245 10.36   4.005  8.655  4.625 14.545 10.98   8.655  5.865\n",
      "  6.485 12.685  6.485  9.895  9.74   2.61  12.995  9.585  4.47   5.555\n",
      "  7.725 15.01  16.25  14.7    8.5    9.895 10.05   8.81  11.6   14.855\n",
      " 10.67   8.965 12.53   9.74  11.135  7.415  7.105  8.965 12.84   7.105\n",
      "  4.78  11.445  9.74   6.64   7.415  9.275 14.235 14.39  16.56   9.585\n",
      " 14.855  6.02   8.345  7.26   8.5    7.88  13.925 13.615  9.585  9.12\n",
      "  5.4   15.165  5.71   7.105 11.445  6.95   7.415  5.71  12.375  4.625]\n"
     ]
    }
   ],
   "source": [
    "x = macbook[\"used_years\"].values\n",
    "y = macbook[\"price\"].values\n",
    "\n",
    "predictions = model(x, w, b)\n",
    "print(predictions)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "188.81322969819274"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rmse = RMSE(predictions, y)\n",
    "rmse"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 손실함수 정의하기"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "슝=3\n"
     ]
    }
   ],
   "source": [
    "def loss(x, w, b, y):\n",
    "    predictions = model(x, w, b)\n",
    "    L = RMSE(predictions, y)\n",
    "    return L\n",
    "print(\"슝=3\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 미분으로 손실함수값을 줄이는 function 만들기\n",
    "- dw, db는 loss 값이 최소가 되는 w, b의 기울기값"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "슝=3\n"
     ]
    }
   ],
   "source": [
    "def gradient(x, w, b, y):\n",
    "    dw = (loss(x, w + 0.0001, b, y) - loss(x, w, b, y)) / 0.0001\n",
    "    db = (loss(x, w, b + 0.0001, y) - loss(x, w, b, y)) / 0.0001\n",
    "    return dw, db\n",
    "print(\"슝=3\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 7-6. 한 번에 못해도 괜찮아, 한 단계 한 단계 학습해나가면 되니까!"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- 하이퍼파라미터"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "슝=3\n"
     ]
    }
   ],
   "source": [
    "# 매개변수의 기울기 앞에 놓일 상수항을 그냥 정해보기\n",
    "\n",
    "LEARNING_RATE = 1\n",
    "print(\"슝=3\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 학습시키기\n",
    "\n",
    "1. 입력 데이터 x와 정답 데이터 y 준비\n",
    "2. 초기 w, b를 랜덤하게 선택\n",
    "3. 현재의 w와 b로 모델이 prediction을 예측하고, 그 값과 y 값을 비교해서 손실함수값 계산\n",
    "4. 손실함수값을 계산함과 동시에 그 점에서의 기울기, gradient 계산\n",
    "5. 계산된 gradient를 활용해서 w'=w-ηg, b′=b−ηg 로 업데이트\n",
    "6. 3~5 번의 과정을 손실함수가 충분히 줄어들 때까지 반복"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "슝=3\n"
     ]
    }
   ],
   "source": [
    "# 실험값을 저장할 컨테이너\n",
    "\n",
    "losses = []\n",
    "print(\"슝=3\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Iteration 100 : Loss 38.3037\n",
      "Iteration 200 : Loss 38.3037\n",
      "Iteration 300 : Loss 38.3036\n",
      "Iteration 400 : Loss 38.3036\n",
      "Iteration 500 : Loss 38.3036\n",
      "Iteration 600 : Loss 38.3036\n",
      "Iteration 700 : Loss 38.3036\n",
      "Iteration 800 : Loss 38.3036\n",
      "Iteration 900 : Loss 38.3036\n",
      "Iteration 1000 : Loss 38.3036\n",
      "Iteration 1100 : Loss 38.3036\n",
      "Iteration 1200 : Loss 38.3036\n",
      "Iteration 1300 : Loss 38.3036\n",
      "Iteration 1400 : Loss 38.3036\n",
      "Iteration 1500 : Loss 38.3036\n",
      "Iteration 1600 : Loss 38.3036\n",
      "Iteration 1700 : Loss 38.3036\n",
      "Iteration 1800 : Loss 38.3036\n",
      "Iteration 1900 : Loss 38.3036\n",
      "Iteration 2000 : Loss 38.3036\n",
      "Iteration 2100 : Loss 38.3036\n",
      "Iteration 2200 : Loss 38.3036\n",
      "Iteration 2300 : Loss 38.3036\n",
      "Iteration 2400 : Loss 38.3036\n",
      "Iteration 2500 : Loss 38.3036\n",
      "Iteration 2600 : Loss 38.3036\n",
      "Iteration 2700 : Loss 38.3036\n",
      "Iteration 2800 : Loss 38.3036\n",
      "Iteration 2900 : Loss 38.3036\n",
      "Iteration 3000 : Loss 38.3036\n"
     ]
    }
   ],
   "source": [
    "for i in range(1, 3001):\n",
    "    dw, db = gradient(x, w, b, y)   # 3, 4번: 모델이 prediction을 예측하고, 손실함수값을 계산함과 동시에 기울기 계산\n",
    "    w -= LEARNING_RATE * dw         # 5번: w = w - η * dw 로 업데이트\n",
    "    b -= LEARNING_RATE * db         # 5번: b = b - η * db 로 업데이트 \n",
    "    L = loss(x, w, b, y)            # 현재의 loss 값 계산\n",
    "    losses.append(L)                # loss 값 기록\n",
    "    if i % 100 == 0:\n",
    "        print('Iteration %d : Loss %0.4f' % (i, L))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 248,
       "width": 375
      },
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(losses)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(-45.04142321365011, 293.4010665659575)"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "w, b"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 248,
       "width": 375
      },
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# 모델에 넣을 x 값들 준비\n",
    "x = np.linspace(0, 5, 6)\n",
    "\n",
    "# x, w, b를 모델에 넣어 y값 출력\n",
    "y = model(x, w, b)\n",
    "\n",
    "# 일차함수 y 그리기\n",
    "plt.plot(y, c=\"r\")\n",
    "\n",
    "\n",
    "# 원본 데이터 점찍기\n",
    "plt.scatter(macbook['used_years'], macbook['price'])\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 새로운 데이터를 모델에 넣어보기"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(20, 2)\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>used_years</th>\n",
       "      <th>price</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>1.20</td>\n",
       "      <td>203</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>1.85</td>\n",
       "      <td>206</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>2.40</td>\n",
       "      <td>191</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>2.85</td>\n",
       "      <td>164</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>3.05</td>\n",
       "      <td>176</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   used_years  price\n",
       "0        1.20    203\n",
       "1        1.85    206\n",
       "2        2.40    191\n",
       "3        2.85    164\n",
       "4        3.05    176"
      ]
     },
     "execution_count": 54,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test = pd.read_csv(\"macbook_test.csv\")\n",
    "print(test.shape)\n",
    "test.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [],
   "source": [
    "test_x = test['used_years'].values\n",
    "test_y = test['price'].values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "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>used_years</th>\n",
       "      <th>price</th>\n",
       "      <th>prediction</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>1.20</td>\n",
       "      <td>203</td>\n",
       "      <td>239.351359</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>1.85</td>\n",
       "      <td>206</td>\n",
       "      <td>210.074434</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>2.40</td>\n",
       "      <td>191</td>\n",
       "      <td>185.301651</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>2.85</td>\n",
       "      <td>164</td>\n",
       "      <td>165.033010</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>3.05</td>\n",
       "      <td>176</td>\n",
       "      <td>156.024726</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5</td>\n",
       "      <td>3.35</td>\n",
       "      <td>136</td>\n",
       "      <td>142.512299</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6</td>\n",
       "      <td>2.55</td>\n",
       "      <td>133</td>\n",
       "      <td>178.545437</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>7</td>\n",
       "      <td>2.60</td>\n",
       "      <td>181</td>\n",
       "      <td>176.293366</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>8</td>\n",
       "      <td>2.50</td>\n",
       "      <td>181</td>\n",
       "      <td>180.797509</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>9</td>\n",
       "      <td>3.10</td>\n",
       "      <td>86</td>\n",
       "      <td>153.772655</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>10</td>\n",
       "      <td>2.70</td>\n",
       "      <td>171</td>\n",
       "      <td>171.789224</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>11</td>\n",
       "      <td>3.40</td>\n",
       "      <td>253</td>\n",
       "      <td>140.260228</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>12</td>\n",
       "      <td>1.30</td>\n",
       "      <td>263</td>\n",
       "      <td>234.847216</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>13</td>\n",
       "      <td>1.80</td>\n",
       "      <td>129</td>\n",
       "      <td>212.326505</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>14</td>\n",
       "      <td>3.10</td>\n",
       "      <td>135</td>\n",
       "      <td>153.772655</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>15</td>\n",
       "      <td>1.55</td>\n",
       "      <td>236</td>\n",
       "      <td>223.586861</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>16</td>\n",
       "      <td>1.80</td>\n",
       "      <td>206</td>\n",
       "      <td>212.326505</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>17</td>\n",
       "      <td>3.55</td>\n",
       "      <td>203</td>\n",
       "      <td>133.504014</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>18</td>\n",
       "      <td>3.40</td>\n",
       "      <td>96</td>\n",
       "      <td>140.260228</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>19</td>\n",
       "      <td>2.50</td>\n",
       "      <td>115</td>\n",
       "      <td>180.797509</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    used_years  price  prediction\n",
       "0         1.20    203  239.351359\n",
       "1         1.85    206  210.074434\n",
       "2         2.40    191  185.301651\n",
       "3         2.85    164  165.033010\n",
       "4         3.05    176  156.024726\n",
       "5         3.35    136  142.512299\n",
       "6         2.55    133  178.545437\n",
       "7         2.60    181  176.293366\n",
       "8         2.50    181  180.797509\n",
       "9         3.10     86  153.772655\n",
       "10        2.70    171  171.789224\n",
       "11        3.40    253  140.260228\n",
       "12        1.30    263  234.847216\n",
       "13        1.80    129  212.326505\n",
       "14        3.10    135  153.772655\n",
       "15        1.55    236  223.586861\n",
       "16        1.80    206  212.326505\n",
       "17        3.55    203  133.504014\n",
       "18        3.40     96  140.260228\n",
       "19        2.50    115  180.797509"
      ]
     },
     "execution_count": 56,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "prediction = model(test_x, w, b)\n",
    "test['prediction'] = prediction\n",
    "test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "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>used_years</th>\n",
       "      <th>price</th>\n",
       "      <th>prediction</th>\n",
       "      <th>error</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>1.20</td>\n",
       "      <td>203</td>\n",
       "      <td>239.351359</td>\n",
       "      <td>-36.351359</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>1.85</td>\n",
       "      <td>206</td>\n",
       "      <td>210.074434</td>\n",
       "      <td>-4.074434</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>2.40</td>\n",
       "      <td>191</td>\n",
       "      <td>185.301651</td>\n",
       "      <td>5.698349</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>2.85</td>\n",
       "      <td>164</td>\n",
       "      <td>165.033010</td>\n",
       "      <td>-1.033010</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>3.05</td>\n",
       "      <td>176</td>\n",
       "      <td>156.024726</td>\n",
       "      <td>19.975274</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5</td>\n",
       "      <td>3.35</td>\n",
       "      <td>136</td>\n",
       "      <td>142.512299</td>\n",
       "      <td>-6.512299</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6</td>\n",
       "      <td>2.55</td>\n",
       "      <td>133</td>\n",
       "      <td>178.545437</td>\n",
       "      <td>-45.545437</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>7</td>\n",
       "      <td>2.60</td>\n",
       "      <td>181</td>\n",
       "      <td>176.293366</td>\n",
       "      <td>4.706634</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>8</td>\n",
       "      <td>2.50</td>\n",
       "      <td>181</td>\n",
       "      <td>180.797509</td>\n",
       "      <td>0.202491</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>9</td>\n",
       "      <td>3.10</td>\n",
       "      <td>86</td>\n",
       "      <td>153.772655</td>\n",
       "      <td>-67.772655</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>10</td>\n",
       "      <td>2.70</td>\n",
       "      <td>171</td>\n",
       "      <td>171.789224</td>\n",
       "      <td>-0.789224</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>11</td>\n",
       "      <td>3.40</td>\n",
       "      <td>253</td>\n",
       "      <td>140.260228</td>\n",
       "      <td>112.739772</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>12</td>\n",
       "      <td>1.30</td>\n",
       "      <td>263</td>\n",
       "      <td>234.847216</td>\n",
       "      <td>28.152784</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>13</td>\n",
       "      <td>1.80</td>\n",
       "      <td>129</td>\n",
       "      <td>212.326505</td>\n",
       "      <td>-83.326505</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>14</td>\n",
       "      <td>3.10</td>\n",
       "      <td>135</td>\n",
       "      <td>153.772655</td>\n",
       "      <td>-18.772655</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>15</td>\n",
       "      <td>1.55</td>\n",
       "      <td>236</td>\n",
       "      <td>223.586861</td>\n",
       "      <td>12.413139</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>16</td>\n",
       "      <td>1.80</td>\n",
       "      <td>206</td>\n",
       "      <td>212.326505</td>\n",
       "      <td>-6.326505</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>17</td>\n",
       "      <td>3.55</td>\n",
       "      <td>203</td>\n",
       "      <td>133.504014</td>\n",
       "      <td>69.495986</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>18</td>\n",
       "      <td>3.40</td>\n",
       "      <td>96</td>\n",
       "      <td>140.260228</td>\n",
       "      <td>-44.260228</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>19</td>\n",
       "      <td>2.50</td>\n",
       "      <td>115</td>\n",
       "      <td>180.797509</td>\n",
       "      <td>-65.797509</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    used_years  price  prediction       error\n",
       "0         1.20    203  239.351359  -36.351359\n",
       "1         1.85    206  210.074434   -4.074434\n",
       "2         2.40    191  185.301651    5.698349\n",
       "3         2.85    164  165.033010   -1.033010\n",
       "4         3.05    176  156.024726   19.975274\n",
       "5         3.35    136  142.512299   -6.512299\n",
       "6         2.55    133  178.545437  -45.545437\n",
       "7         2.60    181  176.293366    4.706634\n",
       "8         2.50    181  180.797509    0.202491\n",
       "9         3.10     86  153.772655  -67.772655\n",
       "10        2.70    171  171.789224   -0.789224\n",
       "11        3.40    253  140.260228  112.739772\n",
       "12        1.30    263  234.847216   28.152784\n",
       "13        1.80    129  212.326505  -83.326505\n",
       "14        3.10    135  153.772655  -18.772655\n",
       "15        1.55    236  223.586861   12.413139\n",
       "16        1.80    206  212.326505   -6.326505\n",
       "17        3.55    203  133.504014   69.495986\n",
       "18        3.40     96  140.260228  -44.260228\n",
       "19        2.50    115  180.797509  -65.797509"
      ]
     },
     "execution_count": 57,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test['error'] = test['price'] - test['prediction']\n",
    "test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "45.06326393703716"
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rmse = ((test['error'] ** 2).sum() / len(test)) ** 0.5\n",
    "rmse"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 248,
       "width": 375
      },
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# 모델 일차함수 그리기\n",
    "x = np.linspace(0, 5, 6)\n",
    "y = model(x, w, b)\n",
    "plt.plot(y, c=\"r\")\n",
    "\n",
    "# 실제 데이터 값\n",
    "plt.scatter(test['used_years'], test['price'])\n",
    "\n",
    "# 모델이 예측한 값\n",
    "plt.scatter(test['used_years'], test['prediction'])\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "20"
      ]
     },
     "execution_count": 65,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.7.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}