{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Import Dependencies\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import tensorflow as tf\n",
    "from sklearn.datasets import load_boston\n",
    "boston = load_boston()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 독립 변수와 종속 변수를 분리한다."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_data = pd.DataFrame(boston.data, columns=boston.feature_names)\n",
    "y_data = pd.DataFrame(boston.target, columns=[\"Target\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>CRIM</th>\n",
       "      <th>ZN</th>\n",
       "      <th>INDUS</th>\n",
       "      <th>CHAS</th>\n",
       "      <th>NOX</th>\n",
       "      <th>RM</th>\n",
       "      <th>AGE</th>\n",
       "      <th>DIS</th>\n",
       "      <th>RAD</th>\n",
       "      <th>TAX</th>\n",
       "      <th>PTRATIO</th>\n",
       "      <th>B</th>\n",
       "      <th>LSTAT</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.00632</td>\n",
       "      <td>18.0</td>\n",
       "      <td>2.31</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.538</td>\n",
       "      <td>6.575</td>\n",
       "      <td>65.2</td>\n",
       "      <td>4.0900</td>\n",
       "      <td>1.0</td>\n",
       "      <td>296.0</td>\n",
       "      <td>15.3</td>\n",
       "      <td>396.90</td>\n",
       "      <td>4.98</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.02731</td>\n",
       "      <td>0.0</td>\n",
       "      <td>7.07</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.469</td>\n",
       "      <td>6.421</td>\n",
       "      <td>78.9</td>\n",
       "      <td>4.9671</td>\n",
       "      <td>2.0</td>\n",
       "      <td>242.0</td>\n",
       "      <td>17.8</td>\n",
       "      <td>396.90</td>\n",
       "      <td>9.14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.02729</td>\n",
       "      <td>0.0</td>\n",
       "      <td>7.07</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.469</td>\n",
       "      <td>7.185</td>\n",
       "      <td>61.1</td>\n",
       "      <td>4.9671</td>\n",
       "      <td>2.0</td>\n",
       "      <td>242.0</td>\n",
       "      <td>17.8</td>\n",
       "      <td>392.83</td>\n",
       "      <td>4.03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.03237</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2.18</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.458</td>\n",
       "      <td>6.998</td>\n",
       "      <td>45.8</td>\n",
       "      <td>6.0622</td>\n",
       "      <td>3.0</td>\n",
       "      <td>222.0</td>\n",
       "      <td>18.7</td>\n",
       "      <td>394.63</td>\n",
       "      <td>2.94</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.06905</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2.18</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.458</td>\n",
       "      <td>7.147</td>\n",
       "      <td>54.2</td>\n",
       "      <td>6.0622</td>\n",
       "      <td>3.0</td>\n",
       "      <td>222.0</td>\n",
       "      <td>18.7</td>\n",
       "      <td>396.90</td>\n",
       "      <td>5.33</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      CRIM    ZN  INDUS  CHAS    NOX     RM   AGE     DIS  RAD    TAX  \\\n",
       "0  0.00632  18.0   2.31   0.0  0.538  6.575  65.2  4.0900  1.0  296.0   \n",
       "1  0.02731   0.0   7.07   0.0  0.469  6.421  78.9  4.9671  2.0  242.0   \n",
       "2  0.02729   0.0   7.07   0.0  0.469  7.185  61.1  4.9671  2.0  242.0   \n",
       "3  0.03237   0.0   2.18   0.0  0.458  6.998  45.8  6.0622  3.0  222.0   \n",
       "4  0.06905   0.0   2.18   0.0  0.458  7.147  54.2  6.0622  3.0  222.0   \n",
       "\n",
       "   PTRATIO       B  LSTAT  \n",
       "0     15.3  396.90   4.98  \n",
       "1     17.8  396.90   9.14  \n",
       "2     17.8  392.83   4.03  \n",
       "3     18.7  394.63   2.94  \n",
       "4     18.7  396.90   5.33  "
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "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>Target</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>24.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>21.6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>34.7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>33.4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>36.2</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Target\n",
       "0    24.0\n",
       "1    21.6\n",
       "2    34.7\n",
       "3    33.4\n",
       "4    36.2"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_data.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Train Test 데이터를 분리한다"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train, X_test, y_train, y_test = train_test_split(X_data, y_data, test_size=0.2, random_state=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((404, 13), (102, 13), (404, 1), (102, 1))"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train.shape, X_test.shape, y_train.shape, y_test.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### StandardScaler를 사용하여 스케일링한다"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "StandardScaler(copy=True, with_mean=True, with_std=True)"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.preprocessing import StandardScaler\n",
    "# 객체로 사용해야 나중에 Test데이터에 같은 Mean, Variance를 사용할 수 있다.\n",
    "scaler = StandardScaler()\n",
    "scaler.fit(X_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "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>CRIM</th>\n",
       "      <th>ZN</th>\n",
       "      <th>INDUS</th>\n",
       "      <th>CHAS</th>\n",
       "      <th>NOX</th>\n",
       "      <th>RM</th>\n",
       "      <th>AGE</th>\n",
       "      <th>DIS</th>\n",
       "      <th>RAD</th>\n",
       "      <th>TAX</th>\n",
       "      <th>PTRATIO</th>\n",
       "      <th>B</th>\n",
       "      <th>LSTAT</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>42</th>\n",
       "      <td>-0.386768</td>\n",
       "      <td>-0.495593</td>\n",
       "      <td>-0.609290</td>\n",
       "      <td>-0.293294</td>\n",
       "      <td>-0.899583</td>\n",
       "      <td>-0.144968</td>\n",
       "      <td>-2.150030</td>\n",
       "      <td>0.894455</td>\n",
       "      <td>-0.746330</td>\n",
       "      <td>-1.008508</td>\n",
       "      <td>-0.248578</td>\n",
       "      <td>0.286742</td>\n",
       "      <td>-0.966850</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>58</th>\n",
       "      <td>-0.385349</td>\n",
       "      <td>0.579239</td>\n",
       "      <td>-0.869526</td>\n",
       "      <td>-0.293294</td>\n",
       "      <td>-0.856756</td>\n",
       "      <td>-0.179832</td>\n",
       "      <td>-1.357820</td>\n",
       "      <td>1.882903</td>\n",
       "      <td>-0.169594</td>\n",
       "      <td>-0.706413</td>\n",
       "      <td>0.582147</td>\n",
       "      <td>0.366695</td>\n",
       "      <td>-0.821168</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>385</th>\n",
       "      <td>1.439108</td>\n",
       "      <td>-0.495593</td>\n",
       "      <td>1.026692</td>\n",
       "      <td>-0.293294</td>\n",
       "      <td>1.258877</td>\n",
       "      <td>-1.440773</td>\n",
       "      <td>1.057367</td>\n",
       "      <td>-1.132950</td>\n",
       "      <td>1.675959</td>\n",
       "      <td>1.556337</td>\n",
       "      <td>0.812904</td>\n",
       "      <td>0.434727</td>\n",
       "      <td>2.501775</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>78</th>\n",
       "      <td>-0.396082</td>\n",
       "      <td>-0.495593</td>\n",
       "      <td>0.256216</td>\n",
       "      <td>-0.293294</td>\n",
       "      <td>-0.993801</td>\n",
       "      <td>-0.053448</td>\n",
       "      <td>-0.499009</td>\n",
       "      <td>0.560803</td>\n",
       "      <td>-0.515636</td>\n",
       "      <td>-0.031142</td>\n",
       "      <td>0.120633</td>\n",
       "      <td>0.319883</td>\n",
       "      <td>-0.060845</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>424</th>\n",
       "      <td>0.560723</td>\n",
       "      <td>-0.495593</td>\n",
       "      <td>1.026692</td>\n",
       "      <td>-0.293294</td>\n",
       "      <td>0.265300</td>\n",
       "      <td>-1.022396</td>\n",
       "      <td>0.093395</td>\n",
       "      <td>-0.832059</td>\n",
       "      <td>1.675959</td>\n",
       "      <td>1.556337</td>\n",
       "      <td>0.812904</td>\n",
       "      <td>-3.866459</td>\n",
       "      <td>0.607906</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         CRIM        ZN     INDUS      CHAS       NOX        RM       AGE  \\\n",
       "42  -0.386768 -0.495593 -0.609290 -0.293294 -0.899583 -0.144968 -2.150030   \n",
       "58  -0.385349  0.579239 -0.869526 -0.293294 -0.856756 -0.179832 -1.357820   \n",
       "385  1.439108 -0.495593  1.026692 -0.293294  1.258877 -1.440773  1.057367   \n",
       "78  -0.396082 -0.495593  0.256216 -0.293294 -0.993801 -0.053448 -0.499009   \n",
       "424  0.560723 -0.495593  1.026692 -0.293294  0.265300 -1.022396  0.093395   \n",
       "\n",
       "          DIS       RAD       TAX   PTRATIO         B     LSTAT  \n",
       "42   0.894455 -0.746330 -1.008508 -0.248578  0.286742 -0.966850  \n",
       "58   1.882903 -0.169594 -0.706413  0.582147  0.366695 -0.821168  \n",
       "385 -1.132950  1.675959  1.556337  0.812904  0.434727  2.501775  \n",
       "78   0.560803 -0.515636 -0.031142  0.120633  0.319883 -0.060845  \n",
       "424 -0.832059  1.675959  1.556337  0.812904 -3.866459  0.607906  "
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train = pd.DataFrame(data=scaler.transform(X_train), columns=X_train.columns, index=X_train.index)\n",
    "X_train.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "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>CRIM</th>\n",
       "      <th>ZN</th>\n",
       "      <th>INDUS</th>\n",
       "      <th>CHAS</th>\n",
       "      <th>NOX</th>\n",
       "      <th>RM</th>\n",
       "      <th>AGE</th>\n",
       "      <th>DIS</th>\n",
       "      <th>RAD</th>\n",
       "      <th>TAX</th>\n",
       "      <th>PTRATIO</th>\n",
       "      <th>B</th>\n",
       "      <th>LSTAT</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>307</th>\n",
       "      <td>-0.396864</td>\n",
       "      <td>0.923185</td>\n",
       "      <td>-1.300817</td>\n",
       "      <td>-0.293294</td>\n",
       "      <td>-0.694015</td>\n",
       "      <td>0.842866</td>\n",
       "      <td>0.082879</td>\n",
       "      <td>-0.303729</td>\n",
       "      <td>-0.284942</td>\n",
       "      <td>-1.073665</td>\n",
       "      <td>-0.017821</td>\n",
       "      <td>0.434727</td>\n",
       "      <td>-0.728209</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>343</th>\n",
       "      <td>-0.399481</td>\n",
       "      <td>1.869037</td>\n",
       "      <td>-1.066897</td>\n",
       "      <td>-0.293294</td>\n",
       "      <td>-0.591232</td>\n",
       "      <td>0.620603</td>\n",
       "      <td>-0.404365</td>\n",
       "      <td>0.899742</td>\n",
       "      <td>-0.515636</td>\n",
       "      <td>-0.196998</td>\n",
       "      <td>-0.387032</td>\n",
       "      <td>0.434727</td>\n",
       "      <td>-0.776769</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>47</th>\n",
       "      <td>-0.377155</td>\n",
       "      <td>-0.495593</td>\n",
       "      <td>-0.609290</td>\n",
       "      <td>-0.293294</td>\n",
       "      <td>-0.899583</td>\n",
       "      <td>-0.346892</td>\n",
       "      <td>0.615692</td>\n",
       "      <td>0.879585</td>\n",
       "      <td>-0.746330</td>\n",
       "      <td>-1.008508</td>\n",
       "      <td>-0.248578</td>\n",
       "      <td>0.389227</td>\n",
       "      <td>0.835448</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>67</th>\n",
       "      <td>-0.395926</td>\n",
       "      <td>0.041823</td>\n",
       "      <td>-0.732098</td>\n",
       "      <td>-0.293294</td>\n",
       "      <td>-1.233630</td>\n",
       "      <td>-0.567702</td>\n",
       "      <td>-1.631238</td>\n",
       "      <td>1.261294</td>\n",
       "      <td>-0.630983</td>\n",
       "      <td>-0.345084</td>\n",
       "      <td>0.212936</td>\n",
       "      <td>0.427180</td>\n",
       "      <td>-0.649124</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>362</th>\n",
       "      <td>0.000604</td>\n",
       "      <td>-0.495593</td>\n",
       "      <td>1.026692</td>\n",
       "      <td>-0.293294</td>\n",
       "      <td>1.858449</td>\n",
       "      <td>-1.317293</td>\n",
       "      <td>0.990765</td>\n",
       "      <td>-0.813129</td>\n",
       "      <td>1.675959</td>\n",
       "      <td>1.556337</td>\n",
       "      <td>0.812904</td>\n",
       "      <td>0.258523</td>\n",
       "      <td>-0.359147</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         CRIM        ZN     INDUS      CHAS       NOX        RM       AGE  \\\n",
       "307 -0.396864  0.923185 -1.300817 -0.293294 -0.694015  0.842866  0.082879   \n",
       "343 -0.399481  1.869037 -1.066897 -0.293294 -0.591232  0.620603 -0.404365   \n",
       "47  -0.377155 -0.495593 -0.609290 -0.293294 -0.899583 -0.346892  0.615692   \n",
       "67  -0.395926  0.041823 -0.732098 -0.293294 -1.233630 -0.567702 -1.631238   \n",
       "362  0.000604 -0.495593  1.026692 -0.293294  1.858449 -1.317293  0.990765   \n",
       "\n",
       "          DIS       RAD       TAX   PTRATIO         B     LSTAT  \n",
       "307 -0.303729 -0.284942 -1.073665 -0.017821  0.434727 -0.728209  \n",
       "343  0.899742 -0.515636 -0.196998 -0.387032  0.434727 -0.776769  \n",
       "47   0.879585 -0.746330 -1.008508 -0.248578  0.389227  0.835448  \n",
       "67   1.261294 -0.630983 -0.345084  0.212936  0.427180 -0.649124  \n",
       "362 -0.813129  1.675959  1.556337  0.812904  0.258523 -0.359147  "
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 주의 : Test는 Fit을하면 안된다!\n",
    "X_test = pd.DataFrame(data=scaler.transform(X_test), columns=X_test.columns, index=X_test.index)\n",
    "X_test.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Tensorflow에서 사용할 땐 Numpy 데이터 타입으로 사용할 예정이니 변환하자"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(numpy.ndarray, numpy.ndarray, numpy.ndarray, numpy.ndarray)"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train = np.array(X_train)\n",
    "y_train = np.array(y_train)\n",
    "X_test = np.array(X_test)\n",
    "y_test = np.array(y_test)\n",
    "type(X_train), type(y_train), type(X_test), type(y_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Tensorflow Model 정의"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Learning Rate\n",
    "lr = 0.01\n",
    "\n",
    "# 가중치를 몇번 업데이트 할 것인가?\n",
    "epochs = 2000\n",
    "\n",
    "# Features 독립 변수\n",
    "X = tf.placeholder(dtype=tf.float32, shape=[None, X_train.shape[1]])\n",
    "# Labels 종속 변수\n",
    "y = tf.placeholder(dtype=tf.float32, shape=[None, 1])\n",
    "\n",
    "# Weight 가중치, 초기값은 정규분포에서 랜덤하게 뽑는다\n",
    "W = tf.Variable(tf.random_normal([X_train.shape[1], 1]))\n",
    "# Bias 초기값은 정규분포에서 랜덤하게 뽑는다\n",
    "b = tf.Variable(tf.random_normal([1]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "# tf.Variable을 사용했거나, 메서드 내부적으로 변수가 존재하는 경우에는 Variables\n",
    "# 초기화해줘야 한다.\n",
    "init = tf.global_variables_initializer()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 우리가 예측하는 값 W*X + b\n",
    "hypothesis = tf.add(tf.matmul(X, W), b)\n",
    "\n",
    "# cost function으로는 MSE를 사용\n",
    "cost = tf.reduce_mean(tf.square(y - hypothesis))\n",
    "\n",
    "# Gradient Descent 방법으로 최적화\n",
    "optimizer = tf.train.GradientDescentOptimizer(learning_rate=lr).minimize(cost)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "# cost_history를 기록하면 마지막에 epoch 변화에 따른 cost 변화를 확인할 때 편리하다\n",
    "cost_history = np.empty(shape=[1], dtype=float)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 0, Error: 604.2737426757812\n",
      "Epoch: 100, Error: 33.710201263427734\n",
      "Epoch: 200, Error: 23.05824851989746\n",
      "Epoch: 300, Error: 22.47170639038086\n",
      "Epoch: 400, Error: 22.255775451660156\n",
      "Epoch: 500, Error: 22.13414764404297\n",
      "Epoch: 600, Error: 22.059106826782227\n",
      "Epoch: 700, Error: 22.009729385375977\n",
      "Epoch: 800, Error: 21.975486755371094\n",
      "Epoch: 900, Error: 21.950754165649414\n",
      "Epoch: 1000, Error: 21.932348251342773\n",
      "Epoch: 1100, Error: 21.918357849121094\n",
      "Epoch: 1200, Error: 21.907577514648438\n",
      "Epoch: 1300, Error: 21.899185180664062\n",
      "Epoch: 1400, Error: 21.892616271972656\n",
      "Epoch: 1500, Error: 21.8874568939209\n",
      "Epoch: 1600, Error: 21.88338851928711\n",
      "Epoch: 1700, Error: 21.88017463684082\n",
      "Epoch: 1800, Error: 21.877634048461914\n",
      "Epoch: 1900, Error: 21.87563133239746\n",
      "Epoch: 2000, Error: 21.874052047729492\n"
     ]
    }
   ],
   "source": [
    "with tf.Session() as sess:\n",
    "    sess.run(init)\n",
    "    \n",
    "    for epoch in range(0, epochs):\n",
    "        # optimizer에서 반환하는 값은 의미가 없으니 _로 받아주자\n",
    "        _, err = sess.run([optimizer, cost], feed_dict={X: X_train, y: y_train})\n",
    "        \n",
    "        cost_history = np.append(cost_history, err)\n",
    "        \n",
    "        # 100 번에 한번씩 Error 변화를 확인하자\n",
    "        if epoch%100 == 0:\n",
    "            print('Epoch: {0}, Error: {1}'.format(epoch, err))\n",
    "    \n",
    "    print('Epoch: {0}, Error: {1}'.format(epoch + 1, err))\n",
    "    \n",
    "    # 우리가 설정한 Epochs만큼의 학습이 끝난 후에 나온 값을 확인하기 위해 받아두자\n",
    "    updated_W = sess.run(W)\n",
    "    updated_b = sess.run(b)\n",
    "    \n",
    "    # Test 데이터를 예측한 값\n",
    "    y_pred = sess.run(hypothesis, feed_dict={X: X_test})\n",
    "    \n",
    "    # Mean Squared Error\n",
    "    mse = sess.run(tf.reduce_mean(tf.square(y_pred - y_test)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Trained Bias: \n",
      " [22.52223]\n",
      "Trained Weights: \n",
      " [[-1.0024459 ]\n",
      " [ 1.3155702 ]\n",
      " [ 0.05793529]\n",
      " [ 0.5868825 ]\n",
      " [-2.2774897 ]\n",
      " [ 2.1402504 ]\n",
      " [ 0.11897256]\n",
      " [-3.1695282 ]\n",
      " [ 2.4372222 ]\n",
      " [-1.6443572 ]\n",
      " [-2.1415503 ]\n",
      " [ 0.67585033]\n",
      " [-3.9236772 ]]\n"
     ]
    }
   ],
   "source": [
    "# 최종 Bias\n",
    "print('Trained Bias: \\n', updated_b)\n",
    "print('Trained Weights: \\n', updated_W)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Mean Squared Error:  23.407308966316197\n"
     ]
    }
   ],
   "source": [
    "print('Mean Squared Error: ',mse)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 280,
       "width": 404
      }
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(range(len(cost_history)), cost_history)\n",
    "plt.axis([0,epochs,0,np.max(cost_history)])\n",
    "plt.xlabel('Epochs')\n",
    "plt.ylabel('Cost')\n",
    "plt.title('Cost 변화', fontsize=15)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAwAAAAIxCAYAAAD60N8yAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAAWJQAAFiUBSVIk8AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvhp/UCwAAIABJREFUeJzs3Xl8XHW9//FXmqZJkUBSKgrSFtH0A1q8ahRFKwjuywW0qGhd6hVccKduoFcUtP5cKOqVuqDCFXq9suhV3JEKWLCoUREQPg2LXaCIbdNSaJOmaX5/fM8h0+ksZ87Mmcwk7+fjkUfamXNmvpk5c+b7Od/P9/NtGR0dRUREREREJocp490AERERERGpHwUAIiIiIiKTiAIAEREREZFJRAGAiIiIiMgkogBARERERGQSUQAgIiIiIjKJKAAQEREREZlEFACIiIiIiEwiCgBERERERCYRBQAiIiIiIpOIAgARERERkUlEAYCIiIiIyCSiAEBEREREZBJRACAiDcfMWjJ63A4zm5b/+GY2NavnFBHJmpm1mFnreLdDmkfL6OjoeLdBRCJmdg9wELDY3S+ow/PNAv4BbAR63P3BrJ+zHDObD/wOeACY5e47zWwecEu0ycHuviHlY68Enhv9dycwCLQA+wCPdfeNVTW+BszsZOB7wN3A09x9uA7P+UngbOBv7v60rJ8vDTO7CDgBuMrdF9X5uc8G3gf8zN3fXM/nLiUKWocJx/Ip7v6TOjzn84DfAne6++FZP1/ec08h/K1bgQ+5+0X1fP5SzOzbwCnARe7+3nF4/ucAK4E1wOHuPlTvNkhzmTreDRBpVtGX0b7ASPSzO/7t7iUj62jfDmBH3rZbgUMJHdNS+08Hdrr7SIltWgijfFOAFnffWWCzh6L7pzdC5z+yI/qd2+ZHXo+0nf/IfUA/sIvQcWoF2gjnwqKvZZ0NAtMJAUnmnf9IfBxsq8eTmdk7gNcRAo4PJNztYWAG4XOTGTPrA3qAL7j7Z3LumgE8rsrHnkp4bx85V5D8nNEKdLj7w/Ft7j5qZg8B+xNen2L7tgDtwHAF5wyKHH8PET43hc4nmXL33WY2THgv7q/385fxEPAowoWLzJjZh4FPApe5+9ty7hohXMxoVedfklAAIJLeXOD2QneY2ShjX+75X+wtwLTo348nXIGPDUS/S31JTwW2R//eTejIxo87Jecn103Asws8XPw8u4s9X7XM7OXAIuCJhA7m74Fl7r62yC5xW3bk3FYyIErK3V9bi8dJwsxuIXTMhhj7cp5G+Nu/WGLX+P18qMLnOxR4O/AiQgf2UcAW4FbgKuA77r61yO7xcVCvILAXOI6xvzWJuFNTcpTGzNoIHdQRYFexjnX0OWolBJq5x1cL0EnhDm6x1y+pFwK/KNKeUcaCgvw2TyEEqnH7cg0QjrNSAewTCIFv/jljCnueN3L9gHBFO1/V5wwz640euxeYSXhv1wDXAZe6+0CJ3bcRXTxJ+/w57ZgCjCYIvloI/aXW6PfOAhdU4velaCCW97wzGfuOiC9I7HT3cq/rIOHCU/7zxO9LpgGITBwKAETSGwbuYewEnvt7CvBUwhfG3cC/cvaLrzi3sXcHKO70lfoy3w38iLGRh1I/xwBPAW4o8li7Sv+J6UVfmhcCb8u76xjgnWb2andfUWDXuCOS+0X4yOtkZi2lvrCjgONcwt8/TPgb4/el2Jdr3DmfRrhSOp2QFrQP8HV3/1Sx5ythZvSzM3r+KYTjYVqpndgzoEvEzBYDnyW0HUKncA3QDTw/+vmwmb3J3X9T4CHi1zOz4yHPM6Pfv6pgn7jDVa6NHyekMwF7dKzjz2Ure3Z2+4Bn5Pw/PkZyg6H49al2lGgH4ZyRf74YIXRqnxRtd3vO88edz/i8kS/JOeMhkp8zXkG4MFHzc4aZ7Q98Gzi5wN3PABYAnzWzD7n7t4o8zPa0z1/Aa4Hvm9kuwvGVGxhNZazTnx8cvQcolqKZJDCaTTgO9hIFaLtyfl7p7r8rsGn+BYL4/dfVf0lEAYBISu5+F3BYofvMrAdYHf33I+5+ZcKHjb+Ain6JRFeIXp3kwczs99E/f1Rkkyw7fOcQOv//At4F/BI4GPgSIZf7CjOb5+735e0Xd3xzX4NdefeXumI3A3h6zmPEV9dyTSN0qEYJHa2dOT8thC/9eEJd2k7fweWuLBYRd3QTBQBm9iHgi4S/5evAV9zdc+4/EvgU4Zj5qZm9qECHom4BgJk9kRAcAzymgl3j96HcqMFWwqjaMGOd5/0JwdB29r5C+s+8/8fHXaERgKpGytz9OoqfM14E/Dr675vcvS/hwyY5Z9xPgnNGFLS/mnA8/F+RzVIdI9HIzC8JI5H3AF8ArgbWEzrYswnnhTOBb5rZo9z9/AIPVfYKewWGCO//9ugnDgLic0Y8khuPyjyJkAZWKlUuyWd+hJCOmHveiUcK44DjCGA/9g544mAk/3niz0e9gnhpcgoARLLxhJx/H1jBfrW60oiZzQWeBTh1HgGInvujhOHqF7n7zdFd/Wb2auBaYD7wMcLkylzxF1xu23I7feU6xj8ArqDIcHo0/P5HQpAwChzp7uvK/U2VStn5hwoCADN7NOHKP8B7C00cd/dbgAVmdgFwOvAN4Ml5m8WvUz06D6/P+fdbzOxT7l5JOkfJNkadxj06jmZ2KmE06hJ3f2eZx49fi9xjrmafyxLG/ZxBSFE6GLi6xGci7THyLkLn/4/AC9w9vxPtwBfN7OfAn4DPmdkVBdpRsxEAd/8RxS+O7MXMvgq8l9JBSNkgMfqbSs4nMbN/EIKi/ry7ip0X4vdflV0kEZUBFcnGC3L+/dIK9otTOGrx2fwc4cviM8U6o1EHOYsvjPcQrrB/PafzHz/nCPCR6L8LC5Sui7/gctuVPwJQlLsPu/tgiVzacxkbIZgCXNhg5fPivzXJCMDLCaMZ/cCyMtt+mHDl8klm9pS8++LXOtNSqGa2HxB3wIcJIwAfrfBh0nRA44nDSSauxq9FoQAgs7kyjPM5IwqMl0T/PafEpmkDgDdGv99eoPP/CHe/Dfgfwt/1mgKbVJ37X4X4PFFofk6h81Y1HgPcU6A4Q7HnUQAgFdEIgEiNmdk+wJsJV6oeAl5pZke4e8EJw3niL/OqPptm9gailA93v7TM5llUmonTDQrm8br7783sLsJVz14z+xhhyHuEkDefL/dLLVUn1czagc8D7ydcwTsZ+C/gJcCvzOwN7p5qAp2ZLWes/blXYuOc86k5P3H6UTxZ9e/ufkzOPpVcyT04+t1XbsTB3beb2V8IczDmAn/LuTvu2GYdCC0htHkz8Abg58BZZnZNkTznQtIcr3GqUZK/b3fe70L31ZSZPQ74d2ADIYXtrWZ2bsKytDU5ZxBSb54BXODuK0tsl/Z88SRgk7v/NcG2fcB/MDYnohbPn4iZtZaolLRv9LtQEFIodTFtG+YSgtZCr5UCAKkJjQCI1N4XCEP43wO+Rvic/Vd0ha2cqXm/K2ZmRxPSHdYTqu8k0Wlmt5nZzdHve8zsxJTP/1jC8PZ6d7+jxKbXRb+fDDyaMPFwNtBVYNvUAUC0yNdCwpfp+wk5vy9w918SJsf+jXD1dbWZfSSaqFgpA54GzAMOJ+R59xCCgqdEP0bo/O7HWAAQTzzOFXcgkvyd8dXBfUtuNSbeLr8TFT9nZt8JZvYqQhoIwDvd/VfApwmvw4/N7OlFdw7i1yPNFei50e8kr1OhACDrFKBvEDryXwGWE6oQlaoUlasW54xXE676/xVYnHC3w3POGX83s3+UeQ/bqbxzWqjka9alca82s4fN7IQC98WVmAq1oZYjAC+Mfv/JzKbnfXcUe556pKnJBKIRAJEaihZxejehCsvZhFGAdxA6mJ8BzirzEPGJvq3kVsWf//nAjwkjDy9z900JdovLAB7Bnp3O6WnaQCj3CfD3MtvFua2zgWPjlB0zezahVGgxRTvG0UTDTmAOIbA4BjiJEGCMElIL3h9fWXX3e6MFdL5ASE35PPCpKA/5WuDPhCpOm4usoxA7Htju7nt1Ts3ss4T3/fvu/qYSjxGrJACI53YcY2Zd7r6l2IZmZoxNvs1/b+LOQyYjAGb2MuB/CcfZl9z98uiucwnv02uB35rZKe5esExmTtvSXGGdF/2ek9OmeHRmJC9drK4jAFEFp1cCdxECgJmENRIWmdnvS1TDiVV7zjgF+G/gTkLFmXJVZOLjsp29zxmlKlytBw41sydEBRRKiRejK1QpJ8s0LAgB5j6EUap8cUBSKgipRfviEdRrGCv5HFezihULNBQASCIaARCpkeiK0fLov6e7+wPu/hChNvtu4Ewz+0yZkYD4voqCczNrNbNPEKqIPEzoUN+acPc24J/uPsXd46op7cBllbQhR3wFv9AXaK64s7pvgtrXuUp1jL8CbCJ03C8BTiNUf7kceIa7L8xPq3D3h9393YSr9N8iTFxeQEgPuoHQccnPmd+Duz9YqPMfiVOaCo1sFJI4Hz9Kp7iGMKpwaZRjv5doVOYywvH1W3fPn1iY2QiAmX2AEJROI7wn8fyPeKL0QuCHhL/hp2b2qahGf760HdxHMTYCMC+qdAPwYsKcgBEzGzWzXWY2BDwvuj+3g5XJ1VUzexfhSv8w8NZo7sp64EPRJheY2bvLPEzac8Y+0aTW/yEEH8e6+70Jdo3fh5vcPV4/oI1wzripxH5XRb+/WOocaGaHE46JUcJk/nxZd3Djz3GhgH+f6HehdUniALWqEQALq56/APgLcAehitr9wL3RT7wWhVJ9pCoaARCpUtSh+BjhamYr8HF3/9/4fnf/efRF/3VCjfJeM3tHkYWw4i+RxJ9NM3sy8F3gKOC3hBKCSb7I4yvme4jyX6v5kk1aPjP+oq30PFSqY/we4CDC3ILbCVfxfxSVQSwpmnz4DjN7L+EL+FmEnOjfu/ufKmxjrjgt4ogK90ua6vQmwvv+CuAeM/sWYfGv+whBx3MJ+dTdhPKYhUYhaj4CYGYHE2qlnxTddAHwvvy5Cu6+y8xeS0iDOZUwcnaSmb3L3XNHgtJ+X52Us+8BhGDuZsLx9yBjJR/juusHEka/CnWwanL12cymETr+74se8z9y50C4+zeieQGfAL5mZs8CPlhkRC/NOeO5hHPGXEJg+PYSi8Tl2+Mqf/R+JknL+iyhAtSrCIHef+aWOo1S715LKF4wHTgvv4BAJOuOb/y3FHqvD4p+F5rEHH9ep+XOIygzp6CQD0e/vx5NAN6jIlRe2d9Czy+SiAIAkSpEk7WWETqMI8C73X2vaizu/i0ze4AwQvBSwM1sGXBuXtpG4uF8MzuKUEHlJMIw8ceAL1Z4NT1tmk8p8ZdjZ5ntHpW3fb5iX2hFv+iiv/1VZZ63pCjV5xcUWbG1ElFwFi8y9QQzOyS6wltKRRV53H2DhZVVzyHM+fhYgc02EkpjfqpAVREY6+y80Mw2MzZheYu7H1xg+5LM7NOEjsx0Qjra+9z9ohJ/wwhwWjRJ+Tzg3whXgXMDgGIr4ZazMPo9SEjheD1wc7Qg2l7zPczsakIOdu7z1KwKkJk9kxDsPJ0wmfQN7r5XzX13/08zu48wEvUm4AQz+xKhY5w7CbWSc8aLCOeMFxCuLL/N3b9b4Z+Q6pzh7v80s+MJZTdfBrzMzLYSjs3phM51C6EDfg5h7YpC6hUAFHqeQ6Lb/1Xgvvh9+C/CnK9hopr+ZvZ4d/9HuSeOAr03AesIaVmlFAsAsk6RkglCAYBISmb2NUJ+/1TCCfut7n5Nse3d/f+ijtp3gaMJV/+uZWxoHMoM50ejDacDpxDq6A8TApBzU1awGSKU2iuaO55CPPpwaJnt4jrY+QuBxXI7YK1Fbs9M9FrHKwN3RD+7E3Tgc8VzPnZGj3U65eeBVFyS090fBhab2ZmE4+JNhGCgn3Cs/DVhYNhGGCmIpa25fimhk7mDMOG3XM43AO6+LOqAH+Pu38m7O/5sPIqEoiD5pYTX/4OEUbi3mdlnovS8QuLPXk0DAAulZi8jBKgtwG2E0bq/FNvH3b8eBUXfJYwgfZSwevIfczYrd86YTjjXvJ4QWD1MGK38YqlynCWsJ1TQ2lDpju5+i4WF6RYR8tz/jTAvYxC4BVgBfMN9bCG7AsZlBCA6H7wdQkWtAvvF56j1hEIDuwifp31IMHE5Oj6+xljp5iQla3MVWkFdpCgFACLpfYlQpm418NEkQ+jufoeZzSekOtwdXYXM9RnCVck/F9l/1MzWElIZlhCGiQt2SKP0nvwJjrn3Hw0cCdzh7teXa3sF7iR8oR9uZp0lOhlHRr9vMbNfEzqeuwllEGHPDlihKhiPMLMLCR3duHMQr+AZ/x7J+Z3773gCdCt7VuUpVJ0HwpyA+UX+nvw2LSCUutxBuAp9JfB+M/temepI8d9acaATdRpWmNlsQidrs7sXPJaK+JW7V1KDvlg7+s3seeVKkxbbl70XP4JQNWoLkOhYjTpscYfqEkJlrI8Qqk19iOJXmAuNNEzJ+10xdx8xs3OAWYQVcM9JMOEWd19lZk8jdOKvLlBG8wOE47fgxHl33xGN6rQRSn1+y90Lzs+JUpOGi71vZvZiQoe9r8Ljao/2EAKxr6fZP0dVFwLMbHruSEp0vHQy1pE/KBqhaGXsQsA/gelm9u+ECwLTgB3u/kPGjo3Pu/vXUjTpHMJo4e8Ix2o8YnMBIYDdReizPTraPv89qnblcplkFACIpBQN6R4f/9/MZhC+BDaXutoa3VesssdqYLDUcLG7X0U0amBml0Sl997t7tfmbfp94FVm9iJ3X1HgoU4idIi+TsJOVRJRR+fXwAmEybQX529jZp2EEpybCRMHDyN0zIp19MuNAExnrMTjLsLIxgjhi3M472ck2maUsQmMuXX642Agv37/lOhxy4qG8uPUig+7+4/M7HvAW4DLzewYdx8osnvqAKAGavacuZ1IM3sqIQCK8+zjIKyQeGJpS5F/352wCYuBZxICsHOi4/KzwLcJE/KvKDJRvtAIQPyeVDVHIsppPyr+v4WVnEeKdchz9huieFnQWwn19YvO+3H3CxnrVP7WzGYCJxe40n4j8BQze3KBieIAbyPk6X+UIhcp6qjaY/XuaN7BbsL72p73mD9L+DhOmMSe+tgws5cQgrPtwKk5n50OQtrRrqht0yk+2bgmk5Bl8lAAIFI7HyPkPf+OUH4yjf8GjrKwCNAnE2y/D2EUolAKz8OEjkux9J54clmSxYYq9U1CAPApM/tRgdGRswhfbl+PJoHOc/dBgGiE5HfsGQyUHAEgXO1+q7tnXSO8LAtlRX9OqGpzpbtfEN11BqHyzDxCrfGXF0nbasn7nWVb2xnrWEMVV7jL6CAEebHcIGw3Y6M1o4wFXq3RT/7rUDYFyMyOI0wmBfjPnAn3FxPSsJ4OXGZmzyowQlUoAIg7V7V+fS4AXmNml7j7m1M+xtXAgWaWNJ9/P8Ix+M8C9z1MCH6LlQ/O8pyRVPy+VPte3ERIQ9zC2ETwoZyfwej3DsYWdYx/tkU/WxmrdpYqAIjOF3G1oze5++r4vuhizz55238U+H/s3dFXVUepiAIAkdqJS8OVrThTQtyBLZUDW+g5C03sjCfrFesUPyXvd81ElY+uIqxuusLM3gf8iTB8/W7CFcT7CGlMxJ3/SJyDmzgFyIuX4KybKIXgA4S1BNoI8zviCai4+2YzeykhlaUXuNnM3lhg3kiiEYColGLulfF4DYQZhCpIEDqG7yUcC/F9jyWsvTCb8H68IelzVuGP0fMPVRqkRX/nVOA3hBKdJfc3s2OBn0T7/IIw+Rl4ZHTqDVF7jgD+18xOzDt+4hSgQgForddJaJpzRnR8x+sp1PyckfM8HYTSuY8hHKNzCakx/+PuPyJl2dN87n5S+a0qUvGxkTMpel/gI1EqUTnxyFmxAECBgCSiAECkduJJW9V0RuN9k+Zxxl8Cd5lZsW32mkxmZt2MLQr1fDNrT5KPXKE3Aj8ldNpW5t13P3CC59Xkz5M4AMhnYSGvwxi7upyb+5/7mLlXm+NUn9yUoDd4WLG23PM9kXA198XRTVcSrubt8Zq6+9+iL/2fEaqeXB3NXzgzJw2kbGfczK4j2SjT44Gvlrh/B+GKZrx+QCYBQFTlp9ik23L77gZ2mln83hX9bJjZ64CLCB3Z3wOvy0/Hc3e3sPjV/wEvJ8wF+ETOJqVGAGodAIznOWOkknMGYc5OvKbFS5I8kYUVuD9LeC3jH/L+nzsH51EU7pcMEUY6yLm/0foviY8NC+tcfBz4JOHv/4S7J135udhKwPHtmSzmJxNPo32ARCaCF5nZqpT7VlorPvZ1wlB27uTXEwhVNgp1DN5I+OK5CHgrIa/3kpTPXZC7PxilYryRUDVkVtTG3wLL3L1QKb1cxeYAJDEbiHs3cee/UM5sbvpLIUlKK55OuMo8jZAq8BHC31cwF9fd/2JmzyCUhH0+8Gz2fI+SXI2/mz0DgCFCOsJWQmrCIOEq7s6oTQPsuaDQemANcG80sfwdec9dU9FV/H0Zey92JR21ia48t1OmEx49xzMJ6UYrgJOKTUCPRqgWECbdfyXv7lJzALK6uvrmKDBM47Ep94vTSHIny7+FMMm30DnjLdHvi4C3RpO8f1dgu1xryVl9OcdOQvAZp9fsJASIA4RjdH20792ECeH/yAnk4s9kTfsvZnYgYX2Iu4HnJayalStR8Gxmi4D/JFygGCTM36qkFKsCAKkJBQAitRN/YUwjdEJyK9LEvwvdlvs7bX7rF/InDkeVYP4tf8Mo7/v9hBSc9xBqcp9tZpfnpeJULbry+9+Ur2ldSOoRAOA5RJ3gpJVoojJ8cUCwjdDRSPJc3yVM1Hsa8J4kJS/d/b6ow/cG4Nq8ORJJAoBzCJM67yHU6t9RYtsk4k5DVilAcwkLsz0iuvocdzyLia8Q5yoYlEUdtg+Z2eXAX8qVUXT3q8zs5773Ik2FqgBlNQIQ/+1xqdm6nTPc/cz826L0qb067GZ2AGExuZsIc51eDyyJJrOX+nytJAS5dzPW4d+RonOdqyYpQAUME4Kp3Snbl/T130I4xu4gjDAWLQNbRLHPaFavi0xQOlBEaif+Ivy1u78mzQOY2bXAsST/bCbpsOV3Wj5OyBE/y923m9nnCVewlxImSDaK1AGAF17sqtw+j6yAHC3ikygAiIKmcrX9C+03ShgFyFd2ErC730Po/NdK1gFArk2EK70PM3b1d5ixYGBK3k9L3v9Ljhy5+01JG1Kg8w8ZVgEqID5nXOzuHy65ZRFm9g9Cp73W54zcEZrzCatKf8XdHzCzbxIuInwY+EKxB4mO8esStiupxAufVShO10tbSGAzYXStZKqbh/VgriZcnKi01j8Uf/8qXhFaJjcdKCK1E5+Aq6lEE3dIkn42k2z3yBdlVGHno4QKIP8V3bwMeCfwLjPbTihbOZ6l5AqlW1Q6AlCt+D0cj1Kc41EGtJ4TCJ9RqsxtA6jnHIBGPmcMwSPzKt4E/A34QXT/uYQJ7v/PzIbcPT+NKkvx+5LFCEBq7n5qBds+XMVTFbtAEB9LtQ6MZIJSACBSO/Hnqb0Gj5H0sxmf9D8Wdd5zJ7E+N/exohU4ryKkKL3Xo5VQ3X1nNFnvRkLt9MeY2X+MY0nNeAGualKAqhWnAEyWAKBupUebQKkqQLUOkMbtnGFmX2DP80UbY/Nm4nPGC4HvEYKMd8apMe6+ycz+A/gx8GUzO4gwkb0eFw7i96fQQn21eNxGF39Gi/39tX5dZIJSACBSO3Fn/Dlm9iPC8H5cXzp3ESrYs/pFXAGjDXhydH/Sz2bccYgnccZpLPHCVQBtZnYSYfJeF7DU3S/PfRB37zOz1wOXEa7sfYtQi3887Cbki/8957ZyC4HVWvy6KgCovd+Y2SBjFZpyV2rOzWvPrQ7TQajs00FYKXuWu1dTOrOUWwhVkXIX58p6BOAkMzuMPc8ZcWpUbjCaW7kqPmfEtfkrPWfEKUdxoB/PgYFwzng7YYL0NOD97r7HSsPRHIoPAl8mpAN9m7AKeNbiNlYTNBUSv36HmNm9jK1VMZLzk7uAYKH0tPi23hIL/VXrZkK6ZrHzswIASUQBgEjtxJ+nxxJW2YXwZZH/RZJffrLQVcWkn819CB2EVwC/ylt99QOE/NwWQk3tnYRqQR8q9EDu/kMzew3wxATVPTLj7tcTFjfLVe8RgMkWAMTHWz1SgJ5A+FzkrsqcO+ky/kzk/64Ldy9UjSerOQDx634YYwulxUFA7mvTmvdT7TljhFDL//a8c8aXgXcROpGPJpwzznH3gqVk3f0r0XyZB929Hp3/XB01frxWwt87hXC+jDvzSeQHspkdr+7+U0J55WKSloOVSU4BgEjtxJ2DywilL3dVWIGmlVCa8bEk72icCGwvUrXiq+7+5ejffzCz75WrFhMttNOIHiKMStxJqNCTmajs5KXRf9eW2jYjdU/HiWqQJ61DnuopGKtysytNlZWozOcUoDWDNSvKeYDwN9xb48eNP+dfJEzOr/ScMZ2xz0PSc8ZTgYeKPM8Z7v6B6N+fNbOlCc4ZyxI+b618m1B5a0UtHzS6Yl9wVCF6reOr/fHncjcwUmVFo1r6PTCtEVZDl+bQMjo6nnP9RCYOM3sMYaXVLe6+IeVj7EP4kql4xVQRaS5mdghhfYSNZRbFK/UY+0b/3FGkqpGIyF4UAIiIiIiITCJ1y6sUEREREZHxpwBARERERGQSUQAgIiIiIjKJKAAQEREREZlEFACIiIiIiEwiWgegxvr6+lRWSURERETqore3t+J1YzQCICIiIiIyiWgEICO9vb3j3YSm1tfXB+h1lPJ0rEhSOlYkKR0rUonxOl7i501DIwAiIiIiIpOIAgARERERkUlEAYCIiIiIyCSiAEBEREREZBJRACAiIiIiMokoABARERERmUQUAIiIiIiITCIKAEREREREJhEFACIiIiIik4gCABERERGRSUQBgIiIiIjIJDJ1vBsgIiIiItJsBrYNcsU1/fz57/9kZBT2u+F6bE43Jx/fQ3dnx3g3ryQFACIiIiIiCQ0Nj3DepX30rxtg49bBR27fsHkAXzPAjTffR8/sbhYv7KW9rXVcNTKjAAAgAElEQVQcW1qcUoBERERERBIYGh7hrGUrWXXbhj06/7k2bh1k1a0bOGvZSoaGR+rcwmQUAIiIiIiIJLB0eR/967YwOlp6u9FR6F+3haXL++rTsAopABARERERKWNg2yCr1w6U7fzHRkdh9boBBrYVHikYTwoARERERETKuOKa/qJpP8Vs3DLIlSv6M2pRegoARERERETK8LUDqfa7Y026/bKkAEBEREREpIyRkd113S9LCgBERERERMpobU3XbU67X5Yar0UiIiIiIg3G5nSn2u+IQ2fUuCXVUwAgIiIiIlLGycf3MHP/ylb4ndnVwYLjejJqUXoKAEREREREyuju7KBndjctLcm2b2mBubO66epsz7ZhKSgAEBERERFJYPHCXnpmdZUNAlpaoGdWF2cs7K1PwyqkAEBEREREJIH2tlbOPu1oujvbmVIkCJjSAt2d7Zx92tG0t7XWt4EJKQAQEREREUlgaHiET1/4ewa2DbG7yIrAu0dhYNsQn77w9wwNj9S3gQkpABARERERSWDp8j76121htEjnPzY6Cv3rtrB0eV99GlYhBQAiIiIiImUMbBtk9dqBsp3/2OgorF43wMC2wWwbloICABERERGRMq64pp+NWyvrzG/cMsiVK/ozalF6CgBERERERMrwtQOp9rtjTbr9sqQAQERERESkjJGR3XXdL0sKAEREREREymhtTddtTrtflhqvRSIiIiIiDcbmdKfa74hDZ9S4JdVTACAiIiIiUsbJx/cwc/+OivaZ2dXBguN6MmpRegoARERERETK6O7soGd2Ny1FVgDO19ICc2d109XZnm3DUlAAICIiIiKSwOKFvfTM6iobBLS0QM+sLs5Y2FufhlVIAYCIiIiISALtba0sOX0+R887iJldhdOBZnZ1cPS8g1hy+nza21rr3MJkpo53A0REREREmkV7WytnLjqKgW2DXLGinz/ftp6R0VH269yXw+d0s+D4Hro7K5srUG8KAEREREREKtTd2cFpJx5J3yE7Aejtbcx0n0KUAiQiIiIiMokoABARERERmUQUAIiIiIiITCKTbg6AmU0FPggsAg4DNgFXAZ9093+NY9NERERERDI3oQIAM5sF/AnY4e6HFri/DfgpcDSwDLgZmAO8G3i5mT3b3TfUr8UiIiIiIvU1YQIAM5sO/B9wALC+yGZnA8cAz3H3v+Ts+13gJuBi4CXZtlREREREZPxMpDkA3wFGgSsK3WlmM4EPAefldv4B3P0B4APAi83smKwbKiIiIiIyXiZEAGBmHwVeALwKGCyy2WuAduCiIvdfBWwGFta8gSIiIiIiDaLpAwAzexnwaeA17r6uxKbPBe5397sK3enuu4FVwHNq30oRERERkcbQ1HMAzGwu8H1gsbtfX2bzuUDBzn+Ou4EX1qJtfX19tXiYSU+voySlY0WS0rEiSelYkUo00/HStAGAme0H/Bi4wt0vSLDL/sCdZbZ5EJhmZtPdfUe1bRQRERGRiemhHSOs/PuDrN+4k5FRaG2BQ2a2M/9Jnew7vXW8m1dSUwYAZjYFWA5sJZTwTKKFMEm43DYk2K6s3t7eah9iUoujaL2OUo6OFUlKx4okpWNFShkaHuG8S/voXzfAxq17Tj1dv2mYO+/fRc/sbhYv7KW9LbtAoJoRh6YMAIBzCbn6zwOmRyVAY9OAKWbWFf1/0N0HCVf39yvzuPsBQ9H2IiIiIiKPGBoe4axlK+lft4XRIpeLN24dZNOtGzhr2UqWnD4/0yAgrWadBLwQmAHcBgzk/bwemJXz/49F+9xJWPm3lMOA1Rm0V0RERESa3NLlfSU7/7HRUehft4WlyxtzXkCzjgCcAnQUue9jwFOjbQD+Ef1eBbzOzOa4+5r8ncysFXgWcHltmyoiIiIizW5g2yCr1w6U7fzHRkdh9boBBrYN0t1ZrNs6PpoyAHD3VcXuM7NFhLSfa/PuuhL4EvAW4JwCu55AGFVYXptWioiIiMhEccU1/Xvl/JezccsgV67o59QTj8yoVek0awpQxaI1Ai4EPmJmT869z8wOAs4Hfu7uvxuP9omIiIhI4/K1A6n2u2NNuv2y1JQjAFVYDDwNWGlmXwVuB3qAdwEPA4vGr2kiIiIi0qhGRnbXdb8sTZoRAAB33w4cCywlzBG4GHgHcAXwTHf/1/i1TkREREQaVWtrum5z2v2yNOFGANx9UZn7hwhlRM+tS4NEREREpOnZnG48RTrPEYfOyKA11Wm8kEREREREpMGcfHwPM/evrJrPzK4OFhzXk1GL0lMAICIiIiJSRndnBz2zu2lpSbZ9SwvMndVNV2d7tg1LQQGAiIiIiEgCixf28oRDuhJt+8RDujhjYW/GLUpHAYCIiIiISFIJVwIbTbpi2DhQACAiIiIiksDS5X3cde/WRNvede9Wli7vy7hF6SgAEBEREREpY2DbIKvXDiQdAGB0FFavG2BgW2WrB9eDAgARERERkTKuuKafjVsr68xv3DLIlSv6M2pRegoARERERETK8LWVrwEAcEeKtQOypgBARERERKSMkZHddd0vSwoARERERETKaG1N121Ou1+WGq9FIiIiIiINxuZ0p9rviENn1Lgl1VMAICIiIiJSxsnH9zBz/46K9pnZ1cGC43oyalF6CgBERERERMro7uygZ3Y3LS3Jtm9pgbmzuunqbM+2YSlMHe8GiIiIiEhjGdg2yBXX9ONrBxgZ2U1r6xRsTjcnH99Dd2dlV8EnksULezlr2Ur6120puR5ASwv0zOrijIW99WtcBRQAiIiIiAgAQ8MjnHdpH/3rBvaqee9rBrjx5vvomd3N4oW9tLe1jlMrx097WytLTp/P0uV9rF43wMYte68LMLOrg7mzujmjgV8jBQAiIiIiwtDwSNmr2xu3DrLp1g2ctWwlS06f37Ad3Cy1t7Vy5qKjwijJin7+fNt6RkZH2a9zXw6f082CJhglUQAgIiIiIixd3lc2tQVgdBT6121h6fI+zlx0VH0a14C6Ozs47cQj6TtkJwC9vY2Z7lOIAgARERGROmjkvPqBbYOsXjtQtvMfGx2F1esGGNg2OO5tl8opABARERHJUDPk1V9xTf9ebStn45ZBrlzRz6knHplRqyQrCgBEREREMpJlXn0tRxR87UBF28fuWJNuPxlfCgBEREREMpJFXn0WIwojI7sTbVer/WR8aSEwERERkQxUk1dfTDyisOq2DUVTdjZuHWRVNKIwNDyS6LlbW9N1CdPuJ+NL75qIiIhIBqrJqy8mzYhCEjanu5JmPuKIQ2ek2k/GlwIAERERkQzUOq8+ixGF2MnH9zBz/8rmDczs6mDBcT0V7SONQQGAiIiISAZqnVefxYhCrLuzg57Z3bS0JHvclhaYO6ubrs72itojjUEBgIiIiEgGap1Xn3WlnsULe+mZ1VU2CGhpgZ5ZXZyxsHkWvpI9qQqQiIiISAZsTjeeokxmsbz6rCv1tLe1suT0+Sxd3sfqdQNs3LL3aMPMrg7mzurmjHFcsyCpRl54bbwpABARERHJwMnH93DjzfdVlLZTKq++HpV62ttaOXPRUaHzvKIfXzPWeT58TjcLmqDz3AwLr403BQAiIiIiGYjz6jfduiHRxN1yefW1HlEopbuzg9OacIXfLBdem0g0B0BEREQkI7XMq1elnvKyKpM60SgAEBEREclInFd/9LyDmNlVuPM+s6uDo+cdVPZqtCr1lJZlmdSJRilAIiIiIhmqZV794oW9ZVNcYHJW6qmmTOqpTZjuVA0FACIiIiJ1UIu8+olWqaeWsi6TOpEoABARERFpIhOhUk8Wsi6TOpE0bQBgZo8B3gK8EpgH7AvcD/wCONfd1+dtPw3Yp8zDDrn7jgyaKyIiIlJTzVqpJyv1KJM6UTTlX2xmzwX+AXwUuA1YDLwNuBx4I/BXM3ty3m5vAAbK/HyxDs0XERERkRqzOd2p9ktTJrXZNesIwAHAN4Gz3X1rzu2XmNm3gZuApcBLCuz7esJIQSH31rSVIiIiIlIXtV54bSJr1gDgp+7+k0J3uPvtZnY5cEqRfVe5+z8ya5mIiEgDG9g2yBXX9ONrx/LGbU43J0/SvHGZOGq98NpE1pQBgLuXm63xILCzHm0RERFpBkPDI5x3aR/96wb2ukLqawa48eb76JndzeJJVjlGJhaVSU2mKecAlGJmM4CTgasSbDvdzCbcayAiIpJraHiEs5atZNVtG4qmR2zcOsiqWzdw1rKVDA2P1LmFIrVRy4XXJrKmHAGImdl+hCCmHTgYOA54H7AO+ECR3Y41sy8BxwPdwIiZ9QFfcPcrs2+1iIhIfS1d3lf2iiiElVH7121h6fI+zlx0VKZtUiqSZEVlUstrGU26XnIDMrN/AHPybv4b8B53/13etouAi4AdwCXACmAzcCjwTuDpwH+6+2eqaVNfX1/zvqAiIjLhPLRjhG/96gEe3J78qv5++7Ty9pccyL7Ta391dHjXKFfeuIn7Ng8XbNN++7TyuBltvPo5B9A2taXmzy8y0fT29lb8QWn2AODZQBzCTQdmA68lXN1fDrw5ni9gZocBzyFMAr4z73HagKuB5wFHuvvf07ZJAYCIiDSSX/Rt4SZ/qOL9nn34vrz06V01bcvwrlEuvuYB7t00XHbbxx3QxqIXHKggQKSMNAFAU6cAufuqAjd/08w+Anwe+BXhaj/ufjdwd5HHGTazM4A+YAGQOgCI9fZOzkkltdLX1wfodZTydKxIUpP1WPn+Dden2m9gR1vNX6vPXfwH7ttcvvMPcN/mYX57++7MU5EKmazHiqQzXsdL/LxpTNQJsOcDDwMvrGCfvwHDwCGZtEhERGQcjIyUK5xX2/2KGdg2yOq1A4nKM0KYj7B63QAD25LXdBeRZCZkAODuw8B2oJLkxanR9gOZNEpERGQctLam+6pPu18xV1zTX9ECTQAbtwxy5Yr+mrZDRCZoAGBmTwceTUjpSepkwutxTSaNEhERGQc2pzvVfkccOqOm7fC16a6v3bFG1+VEaq0p5wCY2XzgZcASd384776ZhGo//wS+m3P7CcBBwHfcfVfePr3AV4Br3f3qjJsvIiJSNycf38ONN99X0dX3mV0dLDiup6btaJRUJJFaiUvZ/vnv/2RkFPa74fqmKWXblAEAoZTnh4HTzOwy4E+ElJ95wGmEVJ5XuPvWnH2mA98APm5m/wvcAowCzwfeBPyVMAogIiIyYXR3dtAzu5tNt25IlH/f0gJzZ3XT1dle03Y0SiqSSLWKraq9YfNA06yq3ZSfKnfvA54IXAD0Al8mVPs5hVD+80h3/2PePj+Itv0RYfTgm9HPUwjBxPPcfVO9/gYREZF6Wbywl55ZXbSUKRbY0gI9s7o4Y2Htq5k0SiqSSDUmyqrazToCgLuvBT4d/STd58/AnzNrlIiISANqb2tlyenzWbq8j9XrBti4Ze+Oy8yuDubO6uaMjK5aNkoqkkg1GnFV7TSaNgAQERGR5NrbWjlz0VEhb3lFP75mgJGR3bS2TuHwOd0syDhvuVFSkUTSqqaUbaPNCVAAICIiMol0d3Zw2olHjstzL17Yy1nLVpa9gpplKpJIWtWUsj11nD5zxTTlHAARERFpPnEq0tHzDmJmV+ErojO7Ojh63kEsOX1+w06glMlpIpWy1QiAiIiI1M14pyKJpDWRStkqABAREZG6G89UJJE0JlIp28ZrkYiIiIhIg5lIpWwVAIiIiIiIlHHy8T3M3L+y9LRGLWWrAEBEREREpIy4lG25BfVijVzKVgGAiIiIiEgCjbCqdi0oABARERERSWCilLJVFSARERk3A9sGueKafnztWClIm9PNySoFKSINKr+U7Z9vW8/I6Cj7de7bNKVsFQCIiEjdDQ2PcN6lffSvG9hrZU1fM8CNN99Hz+xuFi/sbdgraCIyucWlbPsO2QlAb29jpvsUogBARETqZmDbID+4ejXX/HEtgztHim63cesgm27dwFnLVjb0MLpUTqM+IuNPAYCIiGSu1BX/YkZHoX/dFpYu7+PMRUdl3ELJmkZ9RBqHJgGLiEimhoZHOGvZSlbdtiFx5z82Ogqr1w0wsK2y/aSxJDkGNm4dZFU06jM0XHx0SESqpwBAREQytXR5H/3rtjA6mm7/jVsGuXJFf20bJXWV9BjIHfURkewoABARkcwMbBtk9dqB1J3/2B1rBmrTIKm7So8BjfqIZE8BgIiIZOaKa/orTvspZGRkdw1aI+MhzTGgUR+RbCkAEBGRzPja2ly5b23V11WzSnsMaNRHJDs6o4qISGZqdeX+iENn1ORxpP7SHgMa9RHJjgIAERHJTC2u3M/s6mDBcT01aI2Mh7THgEZ9RLKjdQBERMbJZFgQyeZ041WkcrS0wNxZ3XR1ttewVVJPaY8BjfqIZEcBgIhInU2mBZFOPr6HG2++L9VE4JYW6JnVxRkLezNomdRLmmNAoz4i2dL4mohIHU22BZG6Ozvomd1NS0tl+83s6uDoeQex5PT5TR8ETXaVHgMa9RHJngIAEZE6mowLIi1e2EvPrK5EHcCOaa288rmHsvQDx3LmoqPU+Z8gkh4DGvURqQ8FACIidTJZF0Rqb2tlyenzOXreQczsKjy3YWZXB8858iAuPedlvOPV/zZh5kBIkPQY0KiPSH1oDoCISJ1UsyDSqScemVGr6qO9rZUzFx0VJj6v6MfXjE18PnxONwsm0MRnKUzHgEjjUAAgIlInWhAp5IOf1uTBjFRHx4DI+FMKkIhInWhBJBERaQQKAERE6kQLIomISCNQCpCISJ2kXRDpgc3b+eD5107IhcJERKT+FACIiNRJ2kWxBrYNMbBtCJh4C4WJiEj9aVxZRKRO0i6KlW8iLRQmIiL1pwBARKSOKlkUq5SJtFCYiIjUl1KARESqNLBtkCuu6cfXjtU1L5arHy+ItHR5H6vXDbBxS/pFvnIXCtOcABERSappAwAzewzwFuCVwDxgX+B+4BfAue6+vsh+U4EPAouAw4BNwFXAJ939X9m3XEQmiqHhEc67tI/+dQN75fWXytUvtiDSpq2Dj+T6JzVRFgoTEZH6acoAwMyeC/wG2A5cBlwE7AKeCrwTWGBmx7r7bXn7tQE/BY4GlgE3A3OAdwMvN7Nnu/uGuv0hItK0hoZHOGvZSvrXbWF0tPA2G7cOsinK1V9y+vy9JuzmL4j0oa9eX3EAABNroTAREcleUwYAwAHAN4Gz3X1rzu2XmNm3gZuApcBL8vY7GzgGeI67/yW+0cy+G+1zcYF9RET2snR5X8nOfyw3V//MRUeV3FYLhYmISD006yTgn7r7B/I6/wC4++3A5cD83NvNbCbwIeC83M5/tM8DwAeAF5vZMdk1W0QmgoFtg6xeO1C28x/LzdUvRQuFiYhIPTTlt4a7l7vc9SCwM++21wDthHShQq4CNgMLq2udiEx0V1zTX3Et/zhXvxSb052qPUccOiPVfiIiMjk1ZQBQipnNAE4mdOhzPRe4393vKrRfFFSsAp6TbQtFpNn52nQ59+Vy9U8+voeZ+1dWzWdmVwcLjutJ1R4REZmcmnUOAABmth8hiGkHDgaOA94HrCOk9OSaCxTs/Oe4G3hhjZspIhNMVrn68UJhm27dkCi9qKUF5s7qpquzPVV7JpuHdoyw8u8P8v0bri9brlVEZCJr6gAA+Buhik/+bR919815t+8P3Fnm8R4EppnZdHffUU3D+vq0OE8t6HWUpOp5rAzu2J5yvx1l23n8EVNYe18b924aLvt4B89o47gjpuhzUsbwrlGuvHET920e5sHtI8DDj9znawa49k9reNyMNl79nANom1rlCm0yoeizJZVopuOl2VOATiFc9T8OeDmhBOhG4Hozu9TMcv++FqDcNbX4zJ9wap+ITEaHzEx3xf2QR08ru03b1BYWveBAjpjVwX77tBbcZr99WjliVgeLXnCgOqxlDO8a5eJrHuCO9YNR539vD24f4fb1g1x8zQMM79LpX0QmvqYeAXD3VQVu/qaZfQT4PPAr4JLo9geB/co85H7AkLunX5oz0tvbW+1DTGpxFK3XUcoZj2PlsLmD3Hn+dRVNBJ7Z1cG7T5mfOF3n2c9ir4XCWluncPicbhYoZSWxz138B+7bXH40BeC+zcP89vbdZcu1ysSn7yCpxHgdL9WMODR1AFDC+cAnCfn8cQBwJ3mlQQs4DFidYbtEZAKoV65+/kJhUplqyrUqwBKRiazZU4AKcvdhwirBuePnq4DHmVn+nAEAzKwVeBZwY/YtFJFmt3hhLz2zumhJkIEzbWor/9q6gwt/fEvZtQCkdrIq1yoi0uwmZABgZk8HHg3kjo1cCYwAbymy2wnADGB5tq0TkYmgva2VJafP5+h5BzGzq/TV4qHhEfrXbuEn19/NGedfx5KL/8DQcOF8dKmdrMq1iog0u6ZMATKz+cDLgCXu/nDefTMJi339E/hufLu7rzOzC4GPmNmV7n5bzj4HEdKGfu7uv6vH3yAiza+9rZUzFx31SK7+7fdsZu3920p27jduHWTTrRs4a9lKlpw+n/a2whN9pXq1LNc6sG2QK67px9cOqISoiDS9pgwAgB3Ah4HTzOwy4E+ElJ95wGmE1J9XuPvWvP0WA08DVprZV4HbgR7gXYS6cIvq0noRmVDiXP3PXfwH7ly/pez2o6PQv24LS5f3acJphlpb0w1y5+43NDzCeZf20b9uYK90Il8zwI0330fP7G4WL+xVMCciTaMpU4DcvQ94InAB0At8mTDZ9xRCCs+R7v7HAvttB44FlkbbXgy8A7gCeKa7/6se7ReRiaeaCaeSDZvTnWq/Iw6dAYTO/1nLVrLqtg1F5xJs3DrIqmhER2ldItIsUo0AmNknCLXyRwm183N/KHBbqfvudff/qrQN7r4W+HT0U8l+Q8C50Y+ISE1UM+H0VFX6ycTJx/dw4833VVyudcFxPQAsXd5H/7otZYM6jeiISLNJmwJ0FlCrpMc/ARUHACIijUQTThtPNeVaVUJURCaytClARwKzgMcQKud0AtMJQcG0nJ+26LbpwKOi7fYDuqL9DiDU6hcRaWq1nHAqtZO0XGtLC/TM6uKMhWEhH5UQFZGJLNUIgLvfVcHmSooUkQmvFhNOpfbicq1Ll/dx610P8OD2vb+SZnZ1MHdWN2fkTOTViI6ITGQ1qwJkZi3A1GgRrkL3TwMOcvc1tXpOEZFGYXO68RSdv3jCaUzlJmsvLtd63co/sPLv2xgYbHvktT18TjcLCry2GtERkYmslmVATyXU0t+3yP0fBd5tZk929001fF4RkcyV65hXO+FU5Sazt+/0Vl7a20Vvb2/ZbTWiIyITWS0DgN0USfcxs17gE8BKdf5FpJlU0jFPO+E0LjdZquKMFhCrr1qN6IiINKJaXqoYJgQBezCzQ4EfAZuBN9fw+UREMlVpHfj3vPapqSacpik3Kdk6+fgeZu5fWcpV7oiOiEgjSx0AmFmbmd1tZi+PbhomrMCbu82rgZsIVYBe5u73pm6piEidVdoxv+Cyv7Lk9PkcPe8gZnYV7jzO7Org6HkHPXIVXwuINaa4hGi5YC6WO6IjItLoUqcAufuwmc1hbD2AXUCHmX0KOBh4PvAE4AZgkbvfXV1TRUTqJ23HfPvgMGcuOirMGVjRj68ZKDnhVAuINa7FC3vLpmbB3iM6IiKNrto5ALl5/yPR430y5/6/ACe5++Yqn0dEpK6q7Zh3d3ZwWoIOerOUm5yM1YlyS4iuXjfAxi17Hw+FSoiKiDS61AFAVPazlbDgF4QRgB3A44HDgNcBpwF/NbPXuvuqKtsqIlI39eqYN3q5yclenSguIZp0REdEpBlUMwLQBtwGbI/+PwLscvcHgAeAVWa2DLgSuNrMXubuK6tqrYhIndSrY97I5SZVnWhM0hEdEZFmUM0cgJ1A/tlwNG+bfjN7PnA98EMz63X3dWmfU0SkXurVMW/kcpNpqhOdueiozNsljWMypoaJTAS1XAdgCrBX+QN332xmJwJ/Br4LvKiGzykikol6dcyrXUAsK9VUJ1LHb+Kb7KlhIs2ulmPIU4F2M9vrMd39LuAM4Hgze2UNn1NEJBNp6sB3TGvllrs28sHzr+VDX72eC398S9lynY1abrKaSdAysVW6PsbQcME1QkVkHNUyALgZeC9Q8GvM3b8D/IZQFlREpKFV2jEHGNw5wl3rt3Ln+q34mgF+cv3dnHH+dSy5+A8lO0FvO2Ee+z9qWtH7Y/UsN9ks1Ymk/rRwnUjzq1kKUFTn/4Iym73C3XfV6jlFZOJqhNzipHXgSyk1STY3jWLLQztLPs6UFujqbOfs046uS0pFo1cnkvGh1DCRiSH7MhI51PkXkXKGhkdYctEfOOP86/jJ7+7G1wxUfEW9VuI68Ec9+bFMm5r+dFnoSmiSNIpcu0dhYNsQn77w93X52xu5OpGMH6WGiUwMOlOLSMNo1NzigQcH2bmruivbuVdCIXkaRf5j1CulwuZ0p9qvHtWJZPwoNUxkYsg0ADCzE83su2bWluXziMjE0Ii5xXGbaiG+ElppGkWu/EAiK2kmQdejOpGML6WGiUwMWY8APBl4C7UtNyoiE1Da3OKHdmQ3ClBNR72YO9YMpEqjyFWPlIpGrU4k40upYSITQ1WfSDObYWbHmFlLzm2vM7P3RP+Nv+FUA0xESkqbW7zy9m0ZtShdm8oZGdmdOo0iVz1SKhYv7KVnVlfZIKCe1YlkfCk1TGRiqPbK/DOAXwD7Ajui23qA9wBfA4YgrBpsZo8DnksoEzol+t0C7HL3H1TZDhFpcmk7xev/Vbp6TjVq0VHP19o6pSbpEPVIqYgnQS9d3sfqdQNs3LJ3MDSzq4O5s7o5Qws+TQqNunCdiFSm2gBge/Q79wp/L/CX6N87gXjw/JnA/0b/z72etAVQACAyyaXOLa5lfk7+Y2fQyT7i0Bnc/o/NVT9OvVIq2j/qfegAACAASURBVNtaOXPRUaEs64p+fM1YWdbD53SzoI5lWWX8xalhm27dkCg1TqlhIo2p2gAgvuq/G8DMOoDjgHOi23cCcenP+PfbgI3R/8uvfCMik0Lq3OJKVuqq9LFr3MmOr4RevmI1XmUKT71TKro7OzjtxCPr+pzSmJKuj6HUMJHGVW0AMBT9jk8BbwH2Ab4f/T83ABiOtvupu2+s8nlFZIKxOd2pOsWHPDq76whp21RI7pXQNGkUuVpbW5RSIeNGqWEiza/iAMDMpgL/SVj1N06+nWZmncCngW+6+4bo9iH2HgFQLTAR2Uva3OL5T+psqDYVkn8lNE6j2HjLhjJ7Fja9fWrRlIpGWEFZJj6lhok0t4oDAHffZWYfJeTzxwHAYcBFhJSgj+dsvotw5T/+NygAEJEC0uYW79uR3dXFSttUSLEroYsX9nLqZ3/Nlm2VT2I+7umH7HXb0PAI513aR/+6gb0CFl8zwI0330fP7G4W64qs1JBSw0SaU9oE11HGOvYtwF+Bg4EXuPuDOduNoBEAEUmoEctOJm4T0NXZTs+sLp54yP7YnG5OPOYwln7gWM5cdNRene72tla+9L5jmTa1stPwAfu389oX2h63NeoKyiIi0pjSzgGYRvi+i6+JjUQ/+Y+X+5WpAEBESmrE3OIs2/SYGfvQe8RjWHXLBpIMMLS0gM2esVf6T5oVlM9cdFTidoqIyMSSdg5AS7RvXObzuYQUoJ+Z2VPd/eFo89boJ5cCABEpqhFzi7NsU7UVVdKuoDywbVA52iIik1TaEYBPE0p57hf9/x7gxcDfgLOBj0S3twJt1TRQRCanRswtzqJN1Y4wpF1B+coV/ZzaYK+viIjUR6pJwIQAADPryrn9fjM7B/i8mX3e3TcRAoD4OaotOSoiMiFVM8KQdrXiO2pU3lRERJpPtZ3y1rzfFwFfAhYB5xEmGcdFuhUIiMikV6pMJ0D+ZIBymT2pV1DOYJVjERFpDtV2xvfo1Lv7w2a2EngVYwFAvE0cJGj1XxGZsIp18E943mF858e3FS3T+csb/wHAzl2797qvVAnP1Cso13iV40ZT6H3o7hjOdN0IEZFmUYsRgBb27NSvAj6Ycz9m1ka4kPUgNZwTYGYHElYfPgF4CtBBmI9wJfD5vJKkmFlHtE0pO9x9qMw2IiJ7KFeH/+c33MOukeLX8/M7/rk2bh1kU1TCc8np8/cIAtKuVnzEoTMq3qfe0ixqVup9APj7uh3ccOcftB6CiExq1QYAa4DXEyYEx/4J3GpmLYyVAZ3m7r8Fuqt8vkeY2bOAawmLjy0HLiRUGHo2sBh4rZnNd/d/5uz2McIk5VI+TEhjEhFJJK7DX6qST6nOfxLFSnimXUF5wXE9VbUnS2kXNUvyPjy4feSR9RDygykRkcmiqgDA3bcAP8i7+WrgW+4+avbIYjVZ5P13Al8FPu3u23Nuv9TMfgBcA3wGOK3AvseVeNy7atdEEZkMktbhr1ahEp5pV1DOX0ugUSTpxBcbEdF6CCIiydS8Y+7ud+T8935gBWGRsFo/z2+A3xS573dmdiPwoiL3X1vr9ojI5FRpHf5qFSrhuXhhLx+7YCV3rttSdv8nHlKfFZTTStuJ13oIIiLJZVqRx92vJowIjIcHCZOQRUQyk6YOf7UKlvBM2PMdrVekkkI1nXithyAiktyELMlpZgcBLwR+nGDbfQgTfxv3W1FE6i7pBNS0dfirkV/Cc+nyPu66d2uife+6d2vDpr5U04nXeggiIslNuADAzPYFLidUIPpckW1OBN4BPA/YF9hpZjcA50aTlUVkkqp0Aupwieo9Wckt4TmRUl+q6cRrPQQRkeRa0gwHm9nbgYfixyjyU+y+/Ns3u/tlKduf365DgR8C/wa81d2/l3f/pwhVgLYQFi1bGf0dBrwPOKzQfpXo6+vTSIJIkxp4aBff/vUDPDxYvlP4qI4p7D99CvcN7KpDy/b07MP35aVPDwux/6JvCzf5Q2X2KP0YjeKbv/wnGzYPV7zfQTPaaG2B9Zsq3/eQA6Zx6ksOrHg/EZFG0dvb21J+qz2lHQH4BqGuf8VPWMBGoOoAwMxeCXwP2AW80t1/UWCzHwJ3Ar9x9/tzbv+1mV0M/BFYZmY/c/dN1bZJRJrD8K5RrrxxE3duGGRXwpIFDw/uThQo1Np++7Qy/0mdPLRjhJV/f5A/3/lwqsdZ/6+dNW5Z9VpTfqO0trRwyMxp6QKAR2ttShGZfNIGAMcAQ8DO6GcXISDIv/rdGv1MyfvJva2qIMLMpgKfJdTvXwG8yd03FNrW3f8G/K3IfdvM7CzCImIvBy6ppl29vY1bZaMZ9PX1AXodpbxqj5W47OTq9fWdyJtGSwsc8fhHc8Nqii50lVTH9OkN9/n68/pbWH/93RXv94x5s3j1cU/kzvOvq3g9hHefMr9hS6LK+NF3kFRivI6X+HnTSBUAuPvK1M9YQ1G+/8+A5wBnEVb/rSYF50/R70OqbZuIVC7Nyq/VistONrqWFnjC4/Zn04OD3LW++jUHcucRNIpqFjXr6myfUOshiIhkKVUAYGZfI6zAO0xYfbfQOHixuQFTcv59o7v/ME0bIpcCzwT+3d1/WcXjxOIehspCiNRR2pVfq1XvGv6lTJsaOuQ7C0wqntnVwdxZ3YyMjvKH2+6vSXuPOHRG9Q9SY9UuarZ4YW/ZRcTi/XpmNfZ6CCIiWUqbAnQ6Y53+3Mm9heYFlJorcAghL79iZvbvwInAG2rU+Qd4LaG9K2r0eCJSRjUrv1ZrPGr4z37svjx17oH4mrFRjsPndLPg+J7QphX9Re874/zratL5j6+aN6JqOvHtba0sOX0+S5f3sXrdABu37P3e7rdPK/OecCBn1DiYFBFpJmkDgDZ332OqnJk9BbjJ3afn3NYJbAUOcPeBvO2nUt1CXe8AbnD37yfdwcxOjdpzRX6qkJm9FPg48D13X11Fu0SkAmlXfq2F8ajhP729jdNKLDxV7L4L/++WmgQrjZ76kqQTH4+IFOrEt7e1PrIycH4w1T19mPlHdHLs/MZbA0FEpJ7SzgEoVCdjR4Hb4vp4e5VmcPdqa+cdBfzGzE4qs91Kd98Y/fsA4ELAzewHgAPTgJcCrwF+CbyzynaJSELjXcN+PGrAp029qUWw0iypL6U68fGISLn3v7uzY69gqpoJcyIiE0ktFwLbQcjJz3/8KwlVf2ptJvD66KeU44BrAdz982a2CnhjtN9sQhWjvwJvBS7RisAi9VPNyq+nlriKnlS9J8JWk3pTbbBS6qp5oyrUiRcRkerVJAAws15Ch9rM7B7gUcAgsB5YBTye0MmuGf//7d15nFxlmff/T6XT6U7I0g0NGiAkgp0LNOpoj8FoxB9RH5dReAaiglHMODIqjhtxA3dRVJQwovKoOCM6BLeAiuO4kYghxog0DrJeaQbILtBJJWlIutOprt8fp4oU1VVdp07tVd/369WvkDrnVN1VOVTf171cl3uk39zu/nvg9+Vsi4hEU0rl13Kwud14mZ6rkFKX3kQNVjqntPGKF8wNNWouIiKtoaThLzNrTxXQupVgBH03wabebwDfJyi69UrgNjO70szKUThMRJpE1FHtci3dWbqkl55Z1ekUd7S38bfd+/nAleu4+md3Eh8qbubD5nZHet1XLprH2858ljr/IiLyhFJnAD5BsHb+XcC/u3vOMoxmdhpBYLANuKzE1xSRJhF1VPtvu/YzMpooeSlLsWkno5gUg7EkDB9M8MD2vUC01Kal5MgXERHJVOoC2POA97n7N/J1/gHcfR1BpV5tsBWRJ0Qd1X7swCgXX7WekdFc+QiKs2JZH71zuohVaH5ybILUphtTqU3DvI90sBK2nfWe7UdERGqn1ADgaA5Xzy3kL8BTS3w9EWkipSzBSacELVU67eSiBbPp6crdls4pldk0m5naNIywwUqjZPsREZHaKDUAuBN4S8hzXwEMlPh6ItJEih3VzpROCfrYgdJnAdJpJ1e+7yWccdqJ2Nxunn78LGxuN2eediJfes+LK7ZXIDO1aZh2FgpWero6WbRgdlkLpomISHMpdQ/Ax4H/NrMe4Ap3HzeMZWazgPcAHwOWl/h6ItJk0pVfN23ZU/S1g3uGWX/vEK98XldZ2jJR2slK7hUoJrVpOXLki4hIayspAHD335jZGcA3gXPNbBuwGXgYaCfIs/9MgpSg7y6maq+ItIb0qPbyz/yax/bn3UqU17ZHD1agVeOVEqiEUWxqU+XIFxGRqEquguPu/w30EtQB+Dmwn2Ct/3Tgr8D7gBPd/VulvpaINKeO9jaeeuS0SNcmKpW+J0s6UDlyZmU21daiKrGIiLSmshQCc/dh4EepHxGRokVNCdpWqfQ9OewfHmVShV6v2lWJRUSkdek3jojUhagpQY8/ekqZW5Lf6jUDReXhL8Yp846syPOKiIhkUwAgInUhSkrQnq5OFj9jRoVaNJ5vKW6dflgq2CUiItWkAEBE6kLUQlfTO6uX6rJS6/RVsEtERKqpLHsARKSxxIeGWb1mAN9yOIWkze1maY1TSKYz7Qxs3TNhus3MQld3/fV/qta+SqzTP/G4mSrYJSIiVaUAQKSFjIwmuPzafga2xsetZffNcTbcsYPeE7pZsayvJkWk0pl2Vq7qZ9PWOIN7xq+37+nqZP6cbi6sQRttbjdeZLrOiRw5s4PL3n2aCnaJiEhVKQAQaREjo4mCo+uDe4fZdddOLr5qfc0qyU5U6Gre7BlAjId27uMjX7uFtrZJdHeOVm0fwNIlvWy4Y0dZNgKfdNwsvvjuF6vzLyIiVacAQKRFrFzVX3BpDUAyCQNb97ByVT8XLV9YncblkFnoKj1z0X/vIzk73/dsPcAf7r+14jMX6X0KpVQEruUMhoiICCgAEGkJ8aFhNm2Jh+60JpOwaWuc+NBwRfcEhNmLEGbmYt/+BBurNHMRep8CMGtGB0d3TSWZTNLWNomT53Zzdo33WYiIiCgAEGkBUfLXD+4Z5vq1A7wtNQpfTsXsRai3mYt636cgIiJSSFkDADN7OnCduxf87Wtmbe6eKOfri0huUfPX31fGDa9pxexF+PBXb2HP0HDdzVxMtE9Bo/wiIlLvKjEDEDaf3ffMbNjd/7kCbRCRDFHz11ci730xI/r/u31v0c9fqZmLek2dKiIiUqxyBwChRvTN7BLgXOBDZX59Eckhav76cue9L3YvQlTlmrmIDw3zw99u4ve3b+PAyCESY09ueD2kThURESlWpADAzL5C7s5+F5A0s0uBB9396qzr5gFfAs4CPubuX47y+iJSnKj560+Zd2RZ2xFlL0IUpc5cpPcobNoaZ1eB9tZD6lQREZFiRJ0BeA1wEDhEEAgcSv0A3A68GrgXuNrMPg90Ai8AFgL9wGnu/ocS2i0iRYiSv76nq5OzT+8tazui7kUoVikzF2H2KGSr1AZkLTsSEZFKiBQAuPtJRZz+AmARMAXYA1wD3BrldUUkmmLz18diMH9ON10zOsrajkrsKcillJmLsHsUspVzA3K9V2wWEZHGVt4Fvjm4++nANOBU4LvAZcBdZva8Sr+2iBy2YlkfvXO6iMUmPi8Wg945XVy4LOx+/vDKvacgl+yZi/jQMFf/9E4+cOU63n/FzXzgynVc/bM7iQ+Nnw0pdY9CegNyKdIzEBvv3pl3xmZw7/ATdQ9GRpVMTUREilOVOgDuPgb8GfizmX0NuBb4nZm91t3XVaMNIq2uHvLXR92LEFbmzEWUUfRy7FEodQNyvdU9EBGR5lP1QmDu/r9mtgT4ObDazGarHoBIddQ6f32UvQhhpWcu3nrGAr5xw19Z8+ctDB/M/9WSa/NuOfYopJc5RVm/X68Vm0VEpLlEDgDMbCrwZeC3wK/cPfRvdHc/YGZnAxep8y9Sefk6ox/9p4VV7TgWuxchrJnT2jjlaUeTBD7ytVtCBxjZo+jl2KMQi8W49Du3Rlq/X28Vm0VEpDmVMgPQBrwTeAewy8w+7+5XZJ9kZr3AUuAAMAokU6/bDjxawuuLSAH1uJl0xbK+orPs5NIxpY15s2fSPXWUU+dP55f/M1zy5t1y7FF4dM8B7t9WuMpxrrShUWcgfr1xM0lQdiAREQmllAAgPVR2OkEH/zIzezXwRnfP7NgvIAgUpgBHA1tSj7cDTwUuL6ENIpJHmHSWtchhH2YvQhhzjpnOl99zGv39/fzwlkEGtg2XvHm31D0KUyZPYu/QCIWakW/9ftQZiOGDCW5c90BDZAdSalMRkdorZbgr/ZvqLnd/D7AYOAlYa2ZHp09y95+4+wnA3wNj7v40d38a8CqgQD4SEYkqymbSaknvRVj5vpfQHTHVaHq0/rEDCbbvGi15SdF9m+MsXdLLUTNLS30athmZMw9ppc5A1HN2oJHRBJd+51YuvOL33HjLA/jmOPdv24tvjnPjuge48Irfc+k1t9Zdu0VEmlE5AoDJAO7+J+ClBNWAf2lmU7LOT/LkDv9oxnOISBmVspm0km3KTse5eu0Az3/mUyI9XzrX/y33DLFvf+mdxpGDh5jW2c7IaPFfS7EYdM2YwsFDxV2bnTbU5nYX/drZahHQFaLUpiIi9aUcAcAT88zu/iDwD8CzCPL9Z8sMAA6V+Poikkcpm0nLrdDI7233PMyUycV9FWTm+t8+OFKWdm575DH+5dLf8vjwaFHXHTWrg0ULZnN097RIr5uZNnTpkl56ZpW+DKYaAV0x6nk2SkSkFZXSAU9/lT9poam7/xX4PPAuM1tgZtPN7AzgeQBm9nwzW0gQJGBm9blQVaSBRd1MWmoO+2xhRn537xspauQ8u0pxokzZhA4lkuzeN1LUUqLOKW186vxFXLR8IcmxaA3JXPefzpJUqFhbGJUK6IpVj7NRIiKtrpQAIH1trg7854BHgC8ATwd+CvyMYAbgT8BG4PrUuVWvRSDS7KJuJi1HGsxMYUd+w8pVpbithjuJhg8muOnWIK9B1PX72deFrdgcRrkDuijqaTZKREQCpXS+02v8xwUA7j5qZl8HLgE+A8wF9qcOxzicBnSqu0eevzezY4C3AGcAzwY6gQcJgosvuvu+PNdNAz4KnAMcB+wAfgRc4u6PR22PSL0oV2e0FMWO/EKQRWf6tHZ27xv/tZCvSvHxPR1s21Xcsp1ySneyo2YQSu9lSCtXliQof0AXRb3MRomIyGGl1gG4g2Azby7fAD4FPN/dv17C6+RkZqcCNxPUF1gFXE2wL+EFwArg9Wa22N0fzrpuVuq6Y4GvAQPAKcC/Aq9IXaMgQBpauTqjpYgy8nvw0BgLn/FUpkxpC12lePEzZnDP1gNl2QgcRbqTHaXKceZehkzZFZt/s3HzhFWN8ylnQBdVvcxGiYjIYZEDgNTo+nMnOL7bzM52959HfY0CZgBXAp929/0Zj19rZj8E1gCfBc7Puu7rBJ3/57t7uiYBZraKYHnSSuDtFWqzSFWUszMaVdSR3wd37uPL7zkt9PnTp7Zx3JHtDB1IlLW6cFjpTnaUKsdjY0mmdub/Gu6e0cn5qQq/N657oOi2lTOgi6oeZqNEROTJKvoNW8HOP+5+k7t/OKvznz52C7ABeHnm42b2bGAZ8NHMzn/qmk0EMxb/bGZPq1S7Raqh2M2k2RtryyHqCO5DO/bygSvXcfXP7gy9EfSsFx5VtnXzxcrsZBe7fn/3vpFQaS+jZAcqd0AXVdTUpvUQvIiINKtmHmLZx/j3tww4CFyX55r/JNijcE4F2yVSFWE7o7k21pZD1BHckdGxootDtU+OcekFi1m0YDY9XdWrJpvdyU6v3y+muFmYtJf1ENBF1cjBi4hIs2rKAMDMZgMvA/6QdehFwO25Zg0gWLYE3Ae8sLItFKm8dGd0ok5xT1cnfXYMvXO6+ej/+8MTBbqKGX3PpxxFrYopDpVZXfiM007E5nbz9ONnRa40XEi+Tvb+4VEmFTEVETbtZa0DuqgaOXgREWlWsWQtFs1WkJlNB34FPJ9gnf9fM449AvzG3d80wfU/B+a7u0V5/f7+/ub6QKUpPHYgwfp7htg2eJBEMklbLMaxR7UTHzrEw3sP5dxAO3NasLb+rBceRfvk4tfWPHYgwbd+/UjZNueeMqeTN7y4p+btSDvuqHaWv/SYcZ/NL/v38Cd/rOjne8HJ03nl87omPGf0UJIb/riL7btG8/+bHdXOWYui/ZtVyuihJNeseYTtIbI15ftcRUQkt76+vqK/MCNtAjazrwIJgqw7CYKiYOnUnu0Z//1fwM+BbxOk2oQge9CkjHMmAe8uJR1oRrvmATcAzwH+KbPznzIL2FvgafYBE/8WFmkw06e28cq+w7d1mA7Zvv0J9u1PcM2aRyJ3yMqZo3/7rlEeO5Bg+tTiagemNwkXEwDMmDqJWCwWqZMdtTLxtkcPFjynfXKMN7y4J2dAd/zRU1h8yoyiP59qaJ8cY/lLj2nI4EVEpBlFzQL0ToKO/whBEBAjyME/OfXfaQ8DdwFvTJ0fS/0cyjg3CXwg9VyRmdlrgO+lnvs17v7LHKelX28iYc4pqK+vPqbfG1V/f7AmutU+x/jQMKvXDOBbDqfAtLndLM2TAjOqz19zKzt2h8udv2P3KL+7d4yLli8Mdf7IaILLr+1nYGuc+OPlG3Xftz/BwK6pvC2VFSctzL2y4NlBReJCRcnSy2cuvWAx+4dHWb12IHQ60rRr191M4XGG8TqnTi3qfn/J4qJfouZecCpPpDYt9nMth1b9XpHi6V6RYtTqfkm/bhRRA4BZhXLlm9kkIOnuSVJ7Dczsj8Dn3f1GM/s7oB94Wr6CXWGY2WSCysMfBNYCb3b3nXlO3wfMLPCUM4ny21ukBJmd5uzUnb45zoY7dtB7QjcrsopgRVFsga7MNeqFOmgjo+E62lFFLQ4VprhWdqGxjva2J1JwFkNpLyeWmdpURERqI1IAEKZQlrvnywGYniFI/wbeFqUN8MR6/18QbNq9mKD670TdjvuBEws87YmAR22TSLHCdJoH9w6zK7UZ9tILFpcUBEQp0DW4Z5jr1w6MG33PtnJVf8U6/1Bacajs4lqVGoGuhyJsIiIiEymlEvCEzOwEYLu7Z68BSAcAB2HCQCGMawk2+77W3X8V4vyNwNvNrNPdx/WAzOwo4GSCpUQiVRG205xMHk4ZGXY5Ti5RC3QVGn0vdmYhinKMkld6BLoeirCJiIhMJPJvUzN7s5kdmuCU6wiq7mZLBwBjlLDW3sxeC5wJ/HPIzj/Ajwj2Krw+z/HzUu36QdR2iRSjlOU4UUUdRS90XZSZhWI1wii50l6KiEi9K2U4bT/BhttxzOwsgmU5W3McTv9aTBJsDI7q7cAf3P37YS9w9w0EKUI/n6oV8AQzOwX4JPBNd3+ohHaJhFbKcpyoKrVGPerMQliljJLHh4a5+qd38oEr15W11kE+jZqzX0REWkMpS4AOAePSiJjZScDVwJ3AFws8RynLfxYCN5nZ/y1w3np3H8z4+3KCAmF/TqUz3UyQNvSdwD0EGYlEqqJSy3EmUqk16iMHy5tnP1PUUfJqbq7OFGXTsYiISLWUugfgSQsXzOy5wE8I1vef6e65ZgjSY2LpegBR9QDnpn4mcjpwc/ov7v6wmfUBnwDeARwLbAeuBD5XjnoEImGNHorWaS5lM2y516inO9nbHhmK3KaJRB0lr/bm6mzV2nQsIiJSrFIDgJiZdQDzgbcCFxB0ppdMsIwmHQC0EwQBkbh75ODB3fcCK1I/IjUxMppgx6MFE2rlVMpm2PQa9V137Qy192Ci0fdypP2c3XMEo4cSZR8lr/bm6nyU9lJEROpNqQHAEQR7AQD2AJ8BVrr7gQmuSQcAU3hy0TCRlrJyVT/DEZfNlLoZdsWyvqIKY+UbfS817WdPVyeX/euLSZIs6yh5JWsdiIiINLpSAoA2gjX8fwKeAXQBzwOOIlxu/3YICoaVmApUpOGkO6hRlCNlZDnWqJea9jN7ZqGco+SVrHUgIiLS6EoNAEbc/UVm1ga8AfgCcKeZLXX3NQBmFiPI2BMHjgROS1UJTvdgzjWzn4YpLibSLEpJmVmulJGlrlEv5T1UOvtNLTZXi4iINIpSA4DJAKliX9eZ2a+BG4Gfm9kSd98IPAW4imDDcIyg4/+ujOf5HmAEVXpF6k58aJjVawbwLYc7yDa3m6UlbOKM2kHtnNJW9k5z1DXqUd/D5LYYC5/x1Ipmv6lUrQMREZFmUGoA8KTf3u6+y8xeCWwAfmJmzyLYG/Ay4ADBkqEEQfafyRwOIsIsGRKpqkqmkIza0Tzu6Oll7TSXEtxEfQ/HHzO9IpttM1Wq1oGIiEgzKDUAiJlZzN2fWAXs7kNm9nrgNuBr7n4OsLbEdopU1RPZbbbsyVuuupQUklE7mpMnl6eDWo7gJup76JhSau6BwipV60BERKQZlNKbSAD7yJHK093vBVYCz0jtARBpKF+69jY2TdD5T8tMIVkMm9sdqV3l6KCmg5uNd+/Mu4Z/cO8wG1PBzcho7kxFtXwPhSxd0kvPrOKWZ5Vjc7WIiEgjiDwU5+6rgFUTnPIV4HuZswMijeDh3fu57d6HQ58fJYXkyxeewH/d8gBjRfzfMSkGL3v+nCf+HnX5Trny45e7oFg5lbPWgYiISLOp2Fy8uw8Cg5V6fpFK+fg3/kAiUVzcWmwKyd/+aUtRnX+AsST89tYtvPnVz4i8fKec+fHrvZNdrloHIiIizUY73kQyxIeGeXTPRHXs8ismhWTUDDq/+uNDvO1zv2XjXdGW75SSHz+XFcv66J3TRazAQr9adLLTtQ4WLZhNT1fuGZGerk4WLZhd9B4OERGRRlb53XgiDWT1mgEOFTn6n1ZMVpyoGXRGRscYGR0peF6+5Tvlzo9fjoJilVRqrQMREZFmpABA6tZjBxJc/dM7y5p/v5CoHWSYOCtO9nr9bY88Fvl1wsq1fKcS+fEboZMdfl02AAAAIABJREFUtdaBiIhIM1IAIHVnZDTBD9YNsmP3KPv2P3kJS6n59wsppRBUruw2E6XbrIbsvQmVzI+vTraIiEhj0B4AqSvpFJX3bRse1/lPC5OiMqrI+fnbYuOy24RJt1kNmct36jl1p4iIiFSHZgCkrqRTVBZSKEVlVFELSMViMX68dtOTlieFTbdZaZmzGtVI3VlKdWERERGpPAUAUjfKmaIyqigdZIDRQ2PcuO6BJ5Yn/fMZzyzqvVRS5qxGJVN3lqO6sIiIiFSeAgCpG6WkqAybf7+QYjvI49qzd5hdd+3k3od2s2eocLaeashevlNKfvyHdu7liutuZ8fg4ySTSWKxGMcefQTvWvocvvmTOyd8zvRnc/FV65V2U0REpIa0B0DqRrlTVEYVNrd9PskkkTv/UV8zn1zLd6Lkx9+3/yBv+dSveO/lN/PAjn0MH0wwMjrG8MEED2zfx4qv3MKmLcVVFxYREZHa0AyA1I1KpKiMIkxu+0qZecQU9j52sCzPNdHynWJSd+7bf5C3XvIbRg6WZ8N1JZZuiYiISHgKACSycm/2rGSKymJld5DX/WV7VZb09HRNpb1tUslZg8JW3g2TuvPdX1pbts5/WrmXbomIiEh4CgCkaJXa7Bk1A08lU1SmO8i+OV6VAOBZJ/Xw8O79kfcgQHkr7z60c2/F3ne5l25VirIaiYhIs1EAIEVJ57avxGbPaqSojKrcy4xySb+XqZ2TQ23SBeiaMYVjuqcxNpasSOXdldfdzliFMhlV4zMthbIaiYhIs1IAIEUJm9s+Sp7+SqaoLFUllhllyn4vhfYgZI/yp0ep79sc5zPf3li2Ueqdg49HvraQSn+mpahkoCsiIlJrCgAktGrk6U+nqNy0ZeJiYGHXuJdL9AJhFPy8cr2XsJt0R0YTXPqdWys2Sp2sYCGDeq4uXMlAV0REpNYUAEho1cjTn87A8/Gv38T2XaPs2z9+82k517iHFbVAWLoDObktxqHE+N5kofcy0SbdaoxSx8qdlzSlWku3oqiHgnQiIiKVpABAQqtWnv6O9jbe8OIeHjuQYNOuqROmqKyWUguEJRJJZvdMY+YRHWV7L9UYpT726CN4YPu+SO3Lp5pLt6Koh4J0IiIilaQAQEKrdp7+6VPbCqaorKawFXRzSQKjh8b46D8tLEvwUq1R6vef+zzee/nNZdsIXO2lW1HUS0E6ERGRSqnfXXhSd+opT38tZFbQ7ZxS/NKj9ChxOZQySl2MebNnFT1Sn2/VUHZ14XpVLwXpREREKqU5emZSFTa3O9J19bzZs1jpzbnHHTM90vXlGiWu5ij1Vz+4hI6QAU/HlDa+/qElnHHaidjcbp5+/CxsbjdnnnYiK9/3Ei5avrCuO/+gQFdERJqflgBJaPWcp7/aom6NLdcocTVHqWdOm8J/fPz/8O4vrWXP0EjO5UCTYtA1o4OvfnAJM6dNqaulW8Wqx4J0IiIi5aQAQEKr5zz91VbrUeJqv/7MaVP47idfyUM793LF929nx6OPk0wmicViHHv0Ebz/3Ocxb/asSM9dbxToiohIs1MAIEUJuxG2ETZ7lqLWo8S1ev15s2fxlQtPL+k56p0CXRERaXZatCpFydwI29OVO5tMo2z2LMXSJb30zCoum085R4lr/frNbsWyPnrndOXd0JzW7IGuiIg0J80ASNHCVqltZrUeJa716ze7dKC7clU/m7bGGdwzfjlQLQrSiYiIlIMCAIlsoiq1raDWy6Fq/frNToGuiIg0q5YJAMxsCjCtwGkj7n6gGu2RxlfrUeJav36raPVAV0REmk9TBQBmdhOwzd2X5zj8RuA7BZ7i68C/lrtd0rxqPUpc69cXERGRxtM0AYCZdQDPBbYVOPVc4G95jm0va6OkZdR6lLjWry8iIiKNo6EDgNSynhnAScCngDA5Dje6+0MVbJaIiIiISN1q6AAAOA+4utaNkNqJDw2zes0AvuXw0heb283SjKUvYc4RERERaRWNHgD8AsisSvSDYi42s6kEG3/HytqqJlcPHeqR0QSXX9vPwNb4uIqtvjnOhjt2cNLxXSST8MD2PXnP6T2hmxUV2CBbD5+RiIiISC6xZJgk4g3CzB4Cbs61CdjMlhNsAl4OvBZYAnQDCaAfuMzdry+1Df39/c3zgWYZPZTk+g272LF7lH37E+OOz5zWxnFHtnPWC4+ifXKBCkoltuOaNY+wfddoWZ7vuKPaWf7SY8rS5nr5jERERKQ19PX1Fd2haMVKwP8P2AW8E/g/qT8nA6vN7GO1bFg9S3e679s2nLNjC7Bvf4J7tw1zzZpHGD1UuTjohj/uKlvnH2D7rlFu+OOukp+nnj4jERERkXwafQlQMdYBbybYBHx/5gEzuwb4LfBpM7vB3e8p9cX6+pqr6NLnr7mVHbvDdbp37B7ld/eOcdHyhZFfr7+/Hxj/OcaHhnn0F7+P/Lz5PDoEJ85/ZknLc6r9GUkg370ikk33ioSle0WKUav7Jf26UbTMDIC7P+Du12Z3/lPHRoELCT6Ps6veuDoXHxpm05b4hNVmMyWTsGlrnPjQ+MJUpVq9ZmDcev5yGNwzzPVrByJfX0+fkYiIiMhEWiYACOGvwChwfK0bUm+idLpL7VDn41viZX/OtPs2R3/uevqMRERERCbSSkuACpkMtAGV62E2qKid7lI61PkkEpVL2FTKc9fTZ1SPlBVJRESkfigAOGwpwYzImlo3pN5E7RhXorPe1la5SatSnruePqN6EiZda6VSsYqIiEhuLbMEyMzOMLO3m9m4oMfM+oCvEKQQ/W31W1ffonaMK9FZt7ndZX/OtFPmhSkknVtsUrSUnpUMaGptZDTBxVetZ+PdO/MujxrcO8zGu3Zy8VXrGRnNnTlJREREyqt5ex/jTQW+ATxgZpeZ2ZvN7E1m9m1gA3A/wSyAZIna6S6lQ53P0iW99Mwq/5KRnq5Ozj69N9K1I6MJHo3vj3RtJT6jerFyVT8DW/cU3BidTMLA1j2sXBU9m4GIiIiE1zIBgLv/EOgDfgK8Cvhm6ufZwAeBF7t76cngm1CUTncpHeqJdM/opPeEbmJlrKEVi8H8Od10zeiIdP3KVf3sGTpY9HWV+ozqgbIiiYiI1K+m2gPg7vMKHL8duL06rWke6U73rrt2hurQldqhLmTFsj4uvmp9qNHlQmIx6J3TxYXLouXuTXd0i35dKvsZ1VopWZHeduazKtQqERERgRaaAZDSrFjWR++croIj76V2qMPoaG/j0gsWs2jBbHq6cs9M9HR1cuqCp3LqM5864TmLFszm0gsWR96AGrUuwawZHRX9jGpNWZFERETqV1PNAEjlpDvdK1f1s2lrnME94zu9PV2dzJ/TzYVVyOjS0d7GRcsXBukl1w7gmw+nlzx5bjdnZ6SXDHNOVFE7ukd3TW3qrDfKiiQiIlK/FABIaMV0uqule0Yn5xdYMhLmnKiidliTpa5dqnP1lDlKREREnkwBgBStkh3qRhOLuBu52Tu6Nrcbj7Ccp5mzIomIiNSL5u6FiFTYYweKz/4Dzd/RrafMUSIiIvJkCgBEIooPDTNysPjiVUfO7Gj6jm6x6VornTlKREREDlMAIBLR6jUD7N43UvR1HVMmt0RHt54yR4mIiMhhCgBEIoqaAWjGtPYyt6Q+hU3XWmoqVhERESmONgGLRBQ1A9DYWHNnAMpUj5mjREREWp0CAJGIlOoyPGWOEhERqR+t1xMRKROb2x3pumbPACQiIiL1TQGASERKdSkiIiKNSAGASERKdSkiIiKNSAGASAmU6lJEREQajQIAkRIo1aWIiIg0GmUBEimRUl2KiIhII1EAIFImSnUpIiIijUBLgEREREREWogCABERERGRFqIAQERERESkhSgAEBERERFpIQoARERERERaiAIAEREREZEWogBARERERKSFqA6AiEiNxYeGWb1mAN9yuIicze1mqYrIiYhIBSgAEBGpkZHRBJdf28/A1jiDe4efdMw3x9lwxw56T+hmxbI+OtrbatRKERFpNgoAGpxGDkUa08hogouvWs/A1j0kk7nPGdw7zK67dnLxVeu59ILFCgJERKQsFAA0KI0cijS2lav6J+z8pyWTMLB1DytX9XPR8oXVaZyIiDQ1BQANqJYjh5pxECldfGiYTVviBTv/ackkbNoaJz40rP/PRESkZAoAGlAtRg414yBSPqvXDIz7/6iQwT3DXL92gLed+awKtUpERFqF0oA2mFJGDqNKzzhsvHtn3k7L4N5hNqZmHEZGE5FfS6QV+JZ4pOvu2xztOhERkUwKABpMKSOHUUWZcRCR/BKJsapeJyIikkkBQIOp9shhLWYcRJpdW1u0r96o14mIiGTSb5MGU+2Rw1rMOIg0O5vbHem6U+YdWeaWiIhIK1IA0GCqPXKotcoi5bd0SS89s4rL5tPT1cnZp/dWqEUiItJKWi4LkJlNBt4PLAdOBHYBPwc+4e6P1rBpodjcbjxC5zrqyKHWKouUX/eMTnpP6GbXXTtDLa+LxWD+nG66ZnRUvnEiItL0mmoGwMxuMrNrJjjeDvwC+DhBp/+twNeA1wC3mdnsarSzFNUeOdRaZZHKWLGsj945XcRiE58Xi0HvnC4uXNZXnYaJiEjTa5pempl1AM8tcNongdOAl7j7R9z9++7+BaAPGAOuqWwrS5ceOSzUaUgrdeRQa5VFKqOjvY1LL1jMogWz6enKHdT3dHWyaMHsshbzExERaeglQGY2BZgBnAR8Csjb6zSzHuADwJfd/S+Zx9z9ETN7H/BTMzvN3ddVrtWlW7Gsr2AlYCjPyOHSJb1suGNHURuBtVZZJJyO9jYuWr4wqLC9dgDffLjC9slzuzlbFbZFRKQCGjoAAM4Drg557uuADuA7eY7/HNgNLAPqOgBIjxyuXNXPpq1xBveM75z3dHUyf043F5ZYmVdrlUUqr3tGJ+erwq+IiFRJowcAvwBOz/j7DyY490XA39z9f3MddPcxM9sIvLCM7auYao4cVnPGQUREREQqK5YMW+GpAZjZQ8DN7r48x7FbgYPuvniC678K/Iu7Rx6+7u/vb54PNMPooSQ3/HEX23eNsm9/YtzxmdPaOO6ods5adBTtk0NuUKihxw4kWH/PPrYNHiSRhLYYHN/TweJnzGD6VK21FhERkcbQ19dXdMer0WcAijELuL/AOfuAKWY21d0PVKFNDaN9cow3vLgn1XEeSnWck7TFYhx/9BQWn9IYHefRQ0mu37CLHbvHBzLbdo1yz9YDHHdkO2e9sDECGREREZFitVIAEAMKjc6ne3wlj+L39TXvMpiX5J1DKZ/+/n6gvJ/jyGgiWMq0fTjvUqZ9+xMMHUjw442PK/NKg6jEvSLNSfeKhKV7RYpRq/sl/bpRNE0a0BD2ATMLnDMTGHH38ClvpGGsXNVfcB8DQDIJA1v3sHJV9P+xREREROpVKwUA9xNU/p3IicCmKrRFqiw+NMymLfFQmYwgCAI2bY0TH1IsKCIiIs2llQKAjcBxZjY310EzawNOBTZUtVVSFavXDBRVywBgcM8w168dqFCLRERERGqjlQKA64EE8JY8x88gKCS2qmotkqrxLfFI1923Odp1IiIiIvWqZTYBu/tWM7sa+JCZXe/ud6ePmdls4Argv939lpo1UiomkRir6nVRxYeGWb1mAN9yuK6Dze1mqSrCioiISJm0TACQsgJ4LrDezK4E7gV6gXcCjwPLa9c0qaS2tmiTXVGvK9bIaILLr+1nYGt83FIl3xxnwx076D2hmxUlVnYWERERaaUlQLj7fuAlwErgHOAa4O3AauD57v5o7VonlWRzuyNdd8q8I8vckvHS6Uk33r0z7z6Fwb3DbLxrJxdftZ6R0fGF2ERERETCaqoZAHefF+KcEeCS1I+0iKVLetlwx46iNgL3dHVy9um9FWxVIEp60ouWL6x4u0RERKQ5tdQMgLSu7hmd9J7QTSxkcd9YDObP6aZrRkdF26X0pCIiIlJtCgCkZaxY1kfvnK6CQUAsBr1zurhwWeUr+ik9qYiIiFSbAgBpGR3tbVx6wWIWLZhNT1fujDo9XZ0sWjCbSy9YXJXNtkpPKiIiItXWVHsARArpaG/jouULg3SbawfwzYfTbZ48t5uzq5xus1HSk4qIiEjzUAAgLal7Rifnn/msWjej7tOTioiISPNRL0Kkhuo5PamIiIg0JwUAIjW0dEkvPbOKW3JUrfSkIiIi0pwUAIjUUL2mJxUREZHmpQBApMbqMT2piIiINC8FACI1Vo/pSUVERKR5KQuQSB2ot/SkIiIi0rwUAIjUkXpJTyoiIiLNS0uARERERERaiAIAEREREZEWogBARERERKSFaA+ANIz40DCr1wzgWw5vkLW53SzVBlkRERGR0BQASN0bGU1w+bX9DGyNM7h3+EnHfHOcDXfsoPeEblYs61OKTBEREZECtARI6trIaIKLr1rPxrt3juv8pw3uHWbjXTu5+Kr1jIwmqtxCERERkcaiAEDq2spV/Qxs3UMyOfF5ySQMbN3DylX91WmYiIiISINSACB167EDCTZtiRfs/Kclk7Bpa5z4UO6ZAhERERFRACB17JZ7hvIu+8lncM8w168dqFCLRERERBqfAgCpW9sHRyJdd9/meJlbIiIiItI8FABI3UqEXPoz7rrEWHkbIiIiItJEFABI3WqLRbyuTbe1iIiISD7qKUndOr6nI9J1p8w7sswtEREREWkeCgCkbi1+xgx6ZhVX4benq5OzT++tUItEREREGp8CAKlb06e20XtCN7GQS4FiMZg/p5uuGdFmDkRERERagQIAqWsrlvXRO6erYBAQi0HvnC4uXNZXnYaJiIiINCgFAFLXOtrbuPSCxSxaMJuertzLgXq6Olm0YDaXXrCYjva2KrdQREREpLFMrnUDRArpaG/jouULiQ8Ns3rtAL45TiIxRlvbJE6e283ZS3rpnlHcXgERERGRVqUAQBpG94xOzj/zWbVuhoiIiEhD0xIgEREREZEWogBARERERKSFKAAQEREREWkhCgBERERERFpIS20CNrOuAqck3X1vVRojIiIiIlIDLRUAAPECxx8HplejIY0kPjTM6jUD+JbD6TdtbjdLlX5TREREpOG0WgAA8F3gmjzHElVsR90bGU1w+bX9DGyNM7h3+EnHfHOcDXfsoPeEblYs61MBLhEREZEG0YoBwEPufnOtG1HvRkYTXHzVega27iGZzH3O4N5hdt21k4uvWq8qvCIiIiINQpuAJaeVq/on7PynJZMwsHUPK1f1V6dhIiIiIlKSlg4AzGyymXXUuh31Jj40zKYt8YKd/7RkEjZtjRMfGi58soiIiIjUVCwZtpfXBMwsCXwDGALOBHqBGLANuA64xN0fK+U1+vv7G/4D/WX/Hv4U4WN4wcnTeeXzCiVaEhEREZFy6evrixV7TSvOALwDeAHwb8ArgbOB3wAfAG4xs5bPArR9cCTSddsePVjmloiIiIhIubXaJuA3A4+4+2+yHr/BzDYC3wI+Bnyk1Bfq6+sr9Slq5tp1NwPFl0PonDq1bO+7vz/YU9DIn6NUh+4VCUv3ioSle0WKUav7Jf26UbTUDIC7X5uj858+djVwJ/C66raq/rS1Rbstol4nIiIiItWjHtuT3QYcX+tG1JrN7Y503SnzjixzS0RERESk3BQAPFknhasFN72lS3rpmVVchd+erk7OPr23Qi0SERERkXJRAJBiZrOAVwBrat2WWuue0UnvCd3EQu4pj8Vg/pxuumYoo6qIiIhIvWupAMDMrjGzY3M8fgRBGtCpwGer3rA6tGJZH71zugoGAbEY9M7p4sJl2iglIiIi0ghaLQvQG4BzzOxGYC2wC5gPvA3oAV7n7vfWsH11o6O9jUsvWMzKVf1s2hpncM/4Il89XZ3Mn9PNhcv66Ghvq0ErRURERKRYrRYAHA+cB7wW+BwwA9gB/BK4zN0frGHb6k5HexsXLV9IfGiY1WsH8M1xEokx2tomcfLcbs5e0kv3jOL2CoiIiIhIbbVUAODuu4ArUj8SUveMTs4/81m1boaIiIiIlEFL7QEQEREREWl1CgBERERERFqIAgARERERkRaiAEBEREREpIUoABARERERaSEKAEREREREWogCABERERGRFqIAQERERESkhSgAEBERERFpIQoARERERERaSCyZTNa6DU2lv79fH6iIiIiIVEVfX1+s2Gs0AyAiIiIi0kI0AyAiIiIi0kI0AyAiIiIi0kIUAIiIiIiItBAFACIiIiIiLUQBgIiIiIhIC1EAICIiIiLSQhQAiIiIiIi0EAUAIiIiIiItRAGAiIiIiEgLUQAgIiIiItJCFACIiIiIiLQQBQAiIiIiIi1EAYCIiIiISAuZXOsGiIRhZpOB9wPLgROBXcDPgU+4+6M1bJpUiJkdA7wFOAN4NtAJPAhcD3zR3ffluW4a8FHgHOA4YAfwI+ASd3+8Ck2XKjOzpxDcK68BFgDTgb8BvyT4d9+W5zp9r0gouldaj5l1FTgl6e5781xb9/dLLJlM1roN0uLM7CZgm7svz3O8HfgvYBFwFXAHMBd4FzAGvMDdd1antVINZnYqcDNwAFgF/JnUvzXwNmArsNjdH866blbqumOBrwEDwCnAvwJbUtcoCGgiZvYi4CZgP0GgdytwCPg74B0E99BL3P3urOv0vSIAmNkc4DbggLvPy3Fc90oLMrNCHeTH3X16jusa4n7RDIDUlJl1AM8Fco7QpXwSOA14obv/JePa/wD+BFwDvKKCzZTqmwFcCXza3fdnPH6tmf0QWAN8Fjg/67qvE3T+n+/uW9IPmtkqgntlJfD2SjZcqu4o4JvAJ7NG4/7TzL7N4X/37O8Ifa8IZjYV+CnBfZTv95Duldb1XYJ/31wSeR5viPtFMwBSdWY2haCDdxLwKeBVwHdzzQCYWQ/Bl/KX3f1jOY6fSfDl/RJ3X1fBZksdMbObgXmZo3Vm9myCkZbz3f3bOa55L3A50OvuD1apqVJhZjbJ3ccmOP7vwDnufkTGY/peEQDM7DpgPnA/wcjsvKzjuldaVGoG4NPu/qkirmmY+0WbgKUWzgMGCSLhVxU493VAB/CdPMd/DuwGlpWtddII9jH++2sZcBC4Ls81/wnECPYGSJOYqPOfso/gvsik7xXBzD4MvBT4R2A4z2m6V6QYDXO/KACQWvgFcHrGz8MTnPsi4G/u/r+5DqZ++W8EXljuRkp9MrPZwMuAP2QdehFwe9aSoSe4+27gPnSvtAwzOxJYSvBLN5O+V1qcmb0K+DTwOnffOsGpulcECDb2ppYtT6Rh7hftAZCqS21+eWIDjJnlG3mBYGo25/9IGR4g6BBKkzOz6cCPgTbg81mH5wO/KfAUD6TOkyZjZjMJBrU6CPaBnA68h2DD+PuyTtf3Sgszs/nA94EVIZZh6F5pbU8xs8uAM4FeIGZm2whmmi9x98eyzm+Y+0UzAFLvZgE502xl2AdMSW3mkiZlZvOAdQSZFc53979mnRL2XimU2k0a01+BOEH6z9sJ9nvsBT6cmv3JpO+VFpUKFH8GrHb3r4e4RPdKa3sHQfa5fwNeCZxNMND0AeCW1KBUpoa5XzQDIPUuBhTaqR5L/akd7U3KzF4DfI8gveNr3P2XOU4Le6/oPmlO5xDUigCYCpwAvB5Yl8oCdV7GfgF9r7QgM5tEkFZ4L0FKxjB0r7SuNwOPuHv2zPINZrYR+BbwMeAjGcca5n5RACD1bh8ws8A5M4ERd59oKZE0oFQxlc8BHwTWAm+eIH9y2Hul0OiMNCB335jj4W+a2YeALwK/JtgIDvpeaVWXEKy9fjEwNWsEdgowKaP403Dq3173Soty92snOHa1mb2bYNNvZgDQMPeLlgBJvbufoIreRE4ENlWhLVJFqanVNcCFwMXAywsUTwl7r3h5WigN4grgcZ685lbfK61pGXAkcDfBcrHMn3OBORl/T3fqdK9IPrcBx2c91jD3iwIAqXcbgePMbG6ug2bWBpwKbKhqq6QargWeD7zW3b/g7oWmSzcCfWbWmeugmR0FnIzulZbi7qMEVYLbMh7W90prOocnZ6DL/Pk1QUa69N+vSV2je0Xy6SQIFjM1zP2iJUBS764Hvgy8BfhMjuNnEIzorKpmo6SyzOy1BFkX3ujuvwp52Y+A9xKs+/5ejuPnEZRh/0FZGikNwcyeBxwN9Gc8rO+VFpRnmRgAZracYNnPzVmHdK/IOGY2i6Cab/bvp4a5XzQDIHUtlZ/5auBDZvbMzGOpfPBXAP/t7rfUon1SMW8H/uDu3w97gbtvIPgy/nzq3niCmZ1CUJ79m+7+UDkbKrVlZovN7HNmdkSOYz0EBXkeBv4j/bi+VyQs3Suty8yuMbNjczx+BEEa0KnAZzOPNdL9ohkAaQQrgOcC683sSuBegny87yRY27u8dk2TClkI3GRm/7fAeevdfTDj78sJCoT92cy+CmwGnkNwr9xDkLpNmssBgk3i55vZjwjW5e4HFgDnEyz9+Qd3z978re8VCUv3Smt6A3COmd1IkIRiF0Ge/7cBPQRF5O7NcV1D3C+xZFJZq6S2zOwh4GZ3Xz7BOR3Ah4A3AXOBQeAG4BPuvqfyrZRqMrMxDqdKm8jp2VP2qanZTwBnERSE2k6wn+Bz7j5S5qZKHTCzE4B/IsjTfQrByNxm4EbgS+6es9q4vlckzcyuAf4/d5+X57julRaT2jd2HvBagoGkGcAO4JfAZe7+4ATX1v39ogBARERERKSFaA+AiIiIiEgLUQAgIiIiItJCFACIiIiIiLQQBQAiIiIiIi1EAYCIiIiISAtRACAiIiIi0kIUAIiIiIiItBAFACIiIiIiLUQBgIiIiIhIC1EAICLSAsxspplNL3BOp5kNmdk1FXj9U81sl5mdW+7nzvN6083skJl9vczP+4CZfbWczykiUm2Ta90AERG7hk2qAAAIHUlEQVSpig8BHzWz57v7bblOcPdhM5sKjJT6YmZ2HHA/8E13fx8QA44EphX5PG8C2oBEjp9Mk4Fp7v49AHd/zMzGcpyXft5ZqWMHgbFU+yZl/STcfX/WpQfzPaeISKNQACAi0hr+DngMuL3AeQfJEwCY2cuB+QQd4EMc7ghPTv20Ab9z97uBJNCZ8VzJjOcvxteAWanr0687haCDvj/1eAxoT7XhexnXjhB07nP5PfCcAq/9DeCdWY8dKqLtIiJ1SQGAiEiTS412v4ygg3408PAEp48Bo3mOnQssI+iId5B7Gen5wN0c7ngPZx0vdvT8GIKR+CeuM7MNwCLg6Bwj9JnyzgAAZxIEDdmzCoeAZwI3AzfluE6j/yLS8BQAiIg0vzcTdNgBPsX4Ue1sOUe53f2twFvTfzez+4Gb3P0dOU5PBwDpEf9k1uOhuPuTZgzMbDLwjNRfe4E7Jrg8lvG62c+7Od9FZnY2sAv4ZY7DmgEQkYanTcAiIk3MzI4BPgNsAr4C/IuZvaLAZflmALJNn+DcdEc/u8Nc6gj664GZwB4ygpE8OilyoMvMTgUuAL6QZ3ZBAYCINDwFACIiTcrMpgHXE3TU30ywEfg24AYzOy3PZTFCdHLN7AiC5UQ9eU5JBwDZI/BFzQBkveZRwBeBnwErgXeaWd8El0wmWOYT9vmfBvwAWJd6/ny6zOxEM+s1s96wzy8iUi+0BEhEpAmlRv5/DLwQeJO735p6/B8I1rf/1swudPfsNJlhfy88i2AQ6djU8x4BnMfhtfSdqfOyA4FIMwBmNhO4kWD0/33AowQzAD8xsyXufn+Oy9Kbg8M8/7OBXwBDwLnuni9QmUTwPt+S+ns/8Pdh34eISD3QDICISJMxs38E/gIsBM5x9++nj7n7IHAasBH4mpndZGYnZVwedsR8SerP56ZSh3YDVwHfBK4GrshzXdEzAGZ2CvAHoA94nbtvTi3P+QeCvQ23ptbtZ14TS/3nhAGNmU0xs08RzIwMAqe7+0SbpKcAl7l7Om3owmLfj4hIrSkAEBFpImb2eoKR/4eBv3f3H2ef4+67gZcClxIs94mnrm0jGDUv1GmOESwpigMzgFe7+zZSmYHcfRJBQJCp6E3AZna0mV1CMMr+FOCl7v6bjPdxD0E2oO3AajPbYGbzUofTv99yBjRmdpSZXQw8BHycIHBZVKDzD3Af8GDq9ZMTzBSIiNQtLQESEWki7v4jM7sdeDAzdWaO8w4BH81xaDVwb4GXeQtwMkFK0CuAdwHXZ2XsSf9+mZTnzzC6gX8kWJrzLnd/JPsEd38gtQ/gw8AcgmAAgoDjcuDP2deY2RkEeyMmA78CPuLuE2UTyny9NxTRfhGRuqQAQESkyaTXw6eW5pxAMMo/xvgNuellLJlVcD9JsL4+JzN7CsFG3DuA7wNzgUvN7A3u/sOMU9O/X9JLcdqy/gzzPjaZ2XPcPWFmU81sNvBw9qh7KvC4JOuxMTP7D1Kj9VnHbjSzCwmCgxHggaz3eBZwl7tvynr8d8Ct7v7hsO9BRKQeaQmQiEjzeg7BkpX7CTq5D6b+HEg97gTpQQdS/30vQRGvt+d6slRAcT1wJPAv7p4E/i11/TfMbH7G6dkBQPr3TegAACBjFuOFwA4gV82BXG2NEbyXf8/zvF9NPd/tBMuhMn0FeF3W87Wl2tCJiEiD0wyAiEjzuhWYRjADkJhovXqqwNaJBIHAwRzHu4GfAC8C3p7OKuTuB8zsdQSbdH9lZovdfQeH195HngHIciD1ZzzMye6eNLMEQSc/n3Se/6SZHZ/x+JSM10vrSz3+gjCvLyJSzxQAiIg0r8nA8QSpNw+Z2SEOLwdKSy//aQdmZzyW7fUEI+AXuPu3Mg+4+x1m9mqCzbR7M14787lKDQDSbf6Mmb035DVhZ7l/mvozmfqZxPgCZ28kKD620MxOdPcHEBFpUAoARESa13zgzqzHDhF0bscYv/4/3WGekv1E7v5NM/tJro24qePrgJdnPJRvE3DUACC9FOg3BIW64HCHPZnj70nghgKvlw5OFrn7xvSDZvZQxnOSSpN6PsFG4+XAZcDSiO9DRKTmFACIiDS/y4HPAkMTZQYqJF/nP49yzwCkA4j1WZuN80rNeEw0CzDR78DJqeeYBlwL7CSob/A/wDoz+5y758qiJCJS9xQAiIg0v7kEo/NjqXXxYxxeUpOeAZhMkMe/A5iWo0JwscqyCTjH8/Wm0n6OEuxVGOXw7ED6vUwheB+Tmfj3XHqfwrFm1ptxfjvQbmZdBGlRFwBL3H0EWG9mHwc+a2YnAue5e/ZyIRGRuqYAQESk+S1N/RwitR8g9Xh62U9b6ifdWR8BSg0AHgY+SLA5GEqfAUhf9+nUDwTLdNIBTfp9ZO9fmKiy8dTUn9en/kw/T3r5z0XA3wFnuPsT9QTc/XNmNgqcos6/iDQiBQAiIs3rboK0lYlU4a8JpVJntlGGFNHu/ijw5Yy//4Lcm4vDSv++egtwHcF7yq5rkPkejidIezpRwDHA4SxJh/I838dydfLd/bKi34GISJ2IJZPjvu9ERETqSmot/vHATncfCnF+DJgFjLh7dkpPEZGWpgBARERERKSFqBKwiIiIiEgLUQAgIiIiItJCFACIiIiIiLQQBQAiIiIiIi1EAYCIiIiISAtRACAiIiIi0kIUAIiIiIiItBAFACIiIiIiLUQBgIiIiIhIC1EAICIiIiLSQhQAiIiIiIi0EAUAIiIiIiItRAGAiIiIiEgLUQAgIiIiItJC/n9psawkt+SnHQAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 280,
       "width": 384
      }
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.scatter(y_test, y_pred)\n",
    "plt.xlabel(u\"실제 집값\")\n",
    "plt.ylabel(u\"집값 예측치\")\n",
    "plt.title(\"집값 예측치와 실제 집값의 관계\", fontsize=15)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Tensorflow Estimator API를 사용하여 Linear Regression하는 방법"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['CRIM', 'ZN', 'INDUS', 'CHAS', 'NOX', 'RM', 'AGE', 'DIS', 'RAD', 'TAX',\n",
       "       'PTRATIO', 'B', 'LSTAT'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_data.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(13,)"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.array(X_train).shape[1:]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "feat_cols = [tf.feature_column.numeric_column('x', shape=np.array(X_train).shape[1:])]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Make Feature Columns\n",
    "feat_cols = [tf.feature_column.numeric_column('x', shape=np.array(X_train).shape[1:])]\n",
    "# Make Input Function\n",
    "input_func = tf.estimator.inputs.numpy_input_fn({'x': X_train}, y_train, batch_size=1, num_epochs=2000, shuffle=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Using default config.\n",
      "WARNING:tensorflow:Using temporary folder as model directory: /var/folders/gc/y94kqvf109v1_tthvbls56wc0000gn/T/tmp28oygs1t\n",
      "INFO:tensorflow:Using config: {'_model_dir': '/var/folders/gc/y94kqvf109v1_tthvbls56wc0000gn/T/tmp28oygs1t', '_tf_random_seed': None, '_save_summary_steps': 100, '_save_checkpoints_steps': None, '_save_checkpoints_secs': 600, '_session_config': None, '_keep_checkpoint_max': 5, '_keep_checkpoint_every_n_hours': 10000, '_log_step_count_steps': 100, '_train_distribute': None, '_device_fn': None, '_service': None, '_cluster_spec': <tensorflow.python.training.server_lib.ClusterSpec object at 0x119148320>, '_task_type': 'worker', '_task_id': 0, '_global_id_in_cluster': 0, '_master': '', '_evaluation_master': '', '_is_chief': True, '_num_ps_replicas': 0, '_num_worker_replicas': 1}\n"
     ]
    }
   ],
   "source": [
    "# Define Linear Regressor Model\n",
    "# Supported Optimizers: ('Adagrad', 'Adam', 'Ftrl', 'RMSProp', 'SGD')\n",
    "linear_model = tf.estimator.LinearRegressor(feature_columns=feat_cols, optimizer='Adam')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Set up Estimator Training Inputs\n",
    "train_input_func = tf.estimator.inputs.numpy_input_fn(X_train, y_train, batch_size=1, num_epochs=1000, shuffle=False)\n",
    "# Set up Estimator Test Inputs\n",
    "eval_input_func = tf.estimator.inputs.numpy_input_fn({'x': X_test}, y_test, batch_size=1, num_epochs=1, shuffle=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Calling model_fn.\n",
      "INFO:tensorflow:Done calling model_fn.\n",
      "INFO:tensorflow:Create CheckpointSaverHook.\n",
      "INFO:tensorflow:Graph was finalized.\n",
      "INFO:tensorflow:Running local_init_op.\n",
      "INFO:tensorflow:Done running local_init_op.\n",
      "INFO:tensorflow:Saving checkpoints for 0 into /var/folders/gc/y94kqvf109v1_tthvbls56wc0000gn/T/tmp28oygs1t/model.ckpt.\n",
      "INFO:tensorflow:loss = 225.0, step = 1\n",
      "INFO:tensorflow:global_step/sec: 890.828\n",
      "INFO:tensorflow:loss = 253.62346, step = 101 (0.114 sec)\n",
      "INFO:tensorflow:global_step/sec: 1038.08\n",
      "INFO:tensorflow:loss = 0.07589463, step = 201 (0.097 sec)\n",
      "INFO:tensorflow:global_step/sec: 1002.72\n",
      "INFO:tensorflow:loss = 2.1461585, step = 301 (0.100 sec)\n",
      "INFO:tensorflow:global_step/sec: 1143.59\n",
      "INFO:tensorflow:loss = 0.66625136, step = 401 (0.087 sec)\n",
      "INFO:tensorflow:global_step/sec: 1009.44\n",
      "INFO:tensorflow:loss = 21.957817, step = 501 (0.099 sec)\n",
      "INFO:tensorflow:global_step/sec: 988.006\n",
      "INFO:tensorflow:loss = 8.972441, step = 601 (0.101 sec)\n",
      "INFO:tensorflow:global_step/sec: 1050.7\n",
      "INFO:tensorflow:loss = 0.17537297, step = 701 (0.095 sec)\n",
      "INFO:tensorflow:global_step/sec: 1006.66\n",
      "INFO:tensorflow:loss = 6.7108874, step = 801 (0.100 sec)\n",
      "INFO:tensorflow:global_step/sec: 1124.59\n",
      "INFO:tensorflow:loss = 38.028217, step = 901 (0.089 sec)\n",
      "INFO:tensorflow:global_step/sec: 1075.72\n",
      "INFO:tensorflow:loss = 3.514638, step = 1001 (0.093 sec)\n",
      "INFO:tensorflow:global_step/sec: 1086.39\n",
      "INFO:tensorflow:loss = 0.009802317, step = 1101 (0.092 sec)\n",
      "INFO:tensorflow:global_step/sec: 1124.27\n",
      "INFO:tensorflow:loss = 6.4648666, step = 1201 (0.090 sec)\n",
      "INFO:tensorflow:global_step/sec: 1099.32\n",
      "INFO:tensorflow:loss = 1.2076899, step = 1301 (0.090 sec)\n",
      "INFO:tensorflow:global_step/sec: 1145.83\n",
      "INFO:tensorflow:loss = 21.732845, step = 1401 (0.087 sec)\n",
      "INFO:tensorflow:global_step/sec: 1134.79\n",
      "INFO:tensorflow:loss = 0.6822102, step = 1501 (0.089 sec)\n",
      "INFO:tensorflow:global_step/sec: 1114.55\n",
      "INFO:tensorflow:loss = 31.725891, step = 1601 (0.090 sec)\n",
      "INFO:tensorflow:global_step/sec: 1090.22\n",
      "INFO:tensorflow:loss = 1.2142678, step = 1701 (0.092 sec)\n",
      "INFO:tensorflow:global_step/sec: 1036.86\n",
      "INFO:tensorflow:loss = 19.478325, step = 1801 (0.095 sec)\n",
      "INFO:tensorflow:global_step/sec: 1112.72\n",
      "INFO:tensorflow:loss = 62.885754, step = 1901 (0.090 sec)\n",
      "INFO:tensorflow:Saving checkpoints for 2000 into /var/folders/gc/y94kqvf109v1_tthvbls56wc0000gn/T/tmp28oygs1t/model.ckpt.\n",
      "INFO:tensorflow:Loss for final step: 55.036022.\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<tensorflow.python.estimator.canned.linear.LinearRegressor at 0x119148eb8>"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Train the Linear Regressor Estimator\n",
    "linear_model.train(input_fn=input_func, steps=2000)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Calling model_fn.\n",
      "INFO:tensorflow:Done calling model_fn.\n",
      "INFO:tensorflow:Starting evaluation at 2018-11-01-10:59:40\n",
      "INFO:tensorflow:Graph was finalized.\n",
      "INFO:tensorflow:Restoring parameters from /var/folders/gc/y94kqvf109v1_tthvbls56wc0000gn/T/tmp28oygs1t/model.ckpt-2000\n",
      "INFO:tensorflow:Running local_init_op.\n",
      "INFO:tensorflow:Done running local_init_op.\n",
      "INFO:tensorflow:Evaluation [10/100]\n",
      "INFO:tensorflow:Evaluation [20/100]\n",
      "INFO:tensorflow:Evaluation [30/100]\n",
      "INFO:tensorflow:Evaluation [40/100]\n",
      "INFO:tensorflow:Evaluation [50/100]\n",
      "INFO:tensorflow:Evaluation [60/100]\n",
      "INFO:tensorflow:Evaluation [70/100]\n",
      "INFO:tensorflow:Evaluation [80/100]\n",
      "INFO:tensorflow:Evaluation [90/100]\n",
      "INFO:tensorflow:Evaluation [100/100]\n",
      "INFO:tensorflow:Finished evaluation at 2018-11-01-10:59:40\n",
      "INFO:tensorflow:Saving dict for global step 2000: average_loss = 26.29743, global_step = 2000, loss = 26.29743\n",
      "INFO:tensorflow:Saving 'checkpoint_path' summary for global step 2000: /var/folders/gc/y94kqvf109v1_tthvbls56wc0000gn/T/tmp28oygs1t/model.ckpt-2000\n"
     ]
    }
   ],
   "source": [
    "# Test the Model\n",
    "test_metrics = linear_model.evaluate(input_fn=eval_input_func, steps=100)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Calling model_fn.\n",
      "INFO:tensorflow:Done calling model_fn.\n",
      "INFO:tensorflow:Graph was finalized.\n",
      "INFO:tensorflow:Restoring parameters from /var/folders/gc/y94kqvf109v1_tthvbls56wc0000gn/T/tmp28oygs1t/model.ckpt-2000\n",
      "INFO:tensorflow:Running local_init_op.\n",
      "INFO:tensorflow:Done running local_init_op.\n"
     ]
    }
   ],
   "source": [
    "# Get Predicted Values as an Array\n",
    "predicted_vals = []\n",
    "\n",
    "for pred in linear_model.predict(input_fn=eval_input_func):\n",
    "    predicted_vals.append(pred['predictions'])"
   ]
  },
  {
   "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": 280,
       "width": 384
      }
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.scatter(y_test, predicted_vals)\n",
    "plt.xlabel(u\"실제 집값\")\n",
    "plt.ylabel(u\"집값 예측치\")\n",
    "plt.title(\"집값 예측치와 실제 집값의 관계\", fontsize=15)\n",
    "plt.show()"
   ]
  },
  {
   "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": "3.6.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}