Projeto .ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\matfs\\Anaconda3\\lib\\site-packages\\statsmodels\\tools\\_testing.py:19: FutureWarning: pandas.util.testing is deprecated. Use the functions in the public API at pandas.testing instead.\n",
      "  import pandas.util.testing as tm\n"
     ]
    }
   ],
   "source": [
    "%matplotlib inline\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "from sklearn.model_selection import GridSearchCV\n",
    "from yellowbrick.model_selection import FeatureImportances\n",
    "\n",
    "\n",
    "RANDOM_SEED = 42\n",
    "np.random.seed(RANDOM_SEED)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Introduction:\n",
    "\n",
    "DataSet: https://www.kaggle.com/uciml/pima-indians-diabetes-database\n",
    "\n",
    "The goal of this project is to develop a Machine Learning model capable of predicting if a woman has diabetes, with some parameters, such as glucose in the last blood exam or insulin. This model will be a classifier and it will be tested through some different alternatives to check which one has the best outcome. \n",
    "\n",
    "Dependent variable: Outcome\n",
    "\n",
    "Examples of Independent Variables: Life expectancy and Poverty rates"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "scrolled": false
   },
   "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>Pregnancies</th>\n",
       "      <th>Glucose</th>\n",
       "      <th>BloodPressure</th>\n",
       "      <th>SkinThickness</th>\n",
       "      <th>Insulin</th>\n",
       "      <th>BMI</th>\n",
       "      <th>DiabetesPedigreeFunction</th>\n",
       "      <th>Age</th>\n",
       "      <th>Outcome</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>6</td>\n",
       "      <td>148</td>\n",
       "      <td>72</td>\n",
       "      <td>35</td>\n",
       "      <td>0</td>\n",
       "      <td>33.6</td>\n",
       "      <td>0.627</td>\n",
       "      <td>50</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>85</td>\n",
       "      <td>66</td>\n",
       "      <td>29</td>\n",
       "      <td>0</td>\n",
       "      <td>26.6</td>\n",
       "      <td>0.351</td>\n",
       "      <td>31</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>8</td>\n",
       "      <td>183</td>\n",
       "      <td>64</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>23.3</td>\n",
       "      <td>0.672</td>\n",
       "      <td>32</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>89</td>\n",
       "      <td>66</td>\n",
       "      <td>23</td>\n",
       "      <td>94</td>\n",
       "      <td>28.1</td>\n",
       "      <td>0.167</td>\n",
       "      <td>21</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>137</td>\n",
       "      <td>40</td>\n",
       "      <td>35</td>\n",
       "      <td>168</td>\n",
       "      <td>43.1</td>\n",
       "      <td>2.288</td>\n",
       "      <td>33</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>763</th>\n",
       "      <td>10</td>\n",
       "      <td>101</td>\n",
       "      <td>76</td>\n",
       "      <td>48</td>\n",
       "      <td>180</td>\n",
       "      <td>32.9</td>\n",
       "      <td>0.171</td>\n",
       "      <td>63</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>764</th>\n",
       "      <td>2</td>\n",
       "      <td>122</td>\n",
       "      <td>70</td>\n",
       "      <td>27</td>\n",
       "      <td>0</td>\n",
       "      <td>36.8</td>\n",
       "      <td>0.340</td>\n",
       "      <td>27</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>765</th>\n",
       "      <td>5</td>\n",
       "      <td>121</td>\n",
       "      <td>72</td>\n",
       "      <td>23</td>\n",
       "      <td>112</td>\n",
       "      <td>26.2</td>\n",
       "      <td>0.245</td>\n",
       "      <td>30</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>766</th>\n",
       "      <td>1</td>\n",
       "      <td>126</td>\n",
       "      <td>60</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>30.1</td>\n",
       "      <td>0.349</td>\n",
       "      <td>47</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>767</th>\n",
       "      <td>1</td>\n",
       "      <td>93</td>\n",
       "      <td>70</td>\n",
       "      <td>31</td>\n",
       "      <td>0</td>\n",
       "      <td>30.4</td>\n",
       "      <td>0.315</td>\n",
       "      <td>23</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>768 rows × 9 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     Pregnancies  Glucose  BloodPressure  SkinThickness  Insulin   BMI  \\\n",
       "0              6      148             72             35        0  33.6   \n",
       "1              1       85             66             29        0  26.6   \n",
       "2              8      183             64              0        0  23.3   \n",
       "3              1       89             66             23       94  28.1   \n",
       "4              0      137             40             35      168  43.1   \n",
       "..           ...      ...            ...            ...      ...   ...   \n",
       "763           10      101             76             48      180  32.9   \n",
       "764            2      122             70             27        0  36.8   \n",
       "765            5      121             72             23      112  26.2   \n",
       "766            1      126             60              0        0  30.1   \n",
       "767            1       93             70             31        0  30.4   \n",
       "\n",
       "     DiabetesPedigreeFunction  Age  Outcome  \n",
       "0                       0.627   50        1  \n",
       "1                       0.351   31        0  \n",
       "2                       0.672   32        1  \n",
       "3                       0.167   21        0  \n",
       "4                       2.288   33        1  \n",
       "..                        ...  ...      ...  \n",
       "763                     0.171   63        0  \n",
       "764                     0.340   27        0  \n",
       "765                     0.245   30        0  \n",
       "766                     0.349   47        1  \n",
       "767                     0.315   23        0  \n",
       "\n",
       "[768 rows x 9 columns]"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "diabetes = pd.read_csv('diabetes.csv')\n",
    "\n",
    "diabetes"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# First look at the database"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 768 entries, 0 to 767\n",
      "Data columns (total 9 columns):\n",
      " #   Column                    Non-Null Count  Dtype  \n",
      "---  ------                    --------------  -----  \n",
      " 0   Pregnancies               768 non-null    int64  \n",
      " 1   Glucose                   768 non-null    int64  \n",
      " 2   BloodPressure             768 non-null    int64  \n",
      " 3   SkinThickness             768 non-null    int64  \n",
      " 4   Insulin                   768 non-null    int64  \n",
      " 5   BMI                       768 non-null    float64\n",
      " 6   DiabetesPedigreeFunction  768 non-null    float64\n",
      " 7   Age                       768 non-null    int64  \n",
      " 8   Outcome                   768 non-null    int64  \n",
      "dtypes: float64(2), int64(7)\n",
      "memory usage: 54.1 KB\n"
     ]
    }
   ],
   "source": [
    "diabetes.info()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can learn through the info function that there are no null values in the database, therefore we don't need to address any solution to the missing values"
   ]
  },
  {
   "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>Pregnancies</th>\n",
       "      <th>Glucose</th>\n",
       "      <th>BloodPressure</th>\n",
       "      <th>SkinThickness</th>\n",
       "      <th>Insulin</th>\n",
       "      <th>BMI</th>\n",
       "      <th>DiabetesPedigreeFunction</th>\n",
       "      <th>Age</th>\n",
       "      <th>Outcome</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>768.000000</td>\n",
       "      <td>768.000000</td>\n",
       "      <td>768.000000</td>\n",
       "      <td>768.000000</td>\n",
       "      <td>768.000000</td>\n",
       "      <td>768.000000</td>\n",
       "      <td>768.000000</td>\n",
       "      <td>768.000000</td>\n",
       "      <td>768.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>3.845052</td>\n",
       "      <td>120.894531</td>\n",
       "      <td>69.105469</td>\n",
       "      <td>20.536458</td>\n",
       "      <td>79.799479</td>\n",
       "      <td>31.992578</td>\n",
       "      <td>0.471876</td>\n",
       "      <td>33.240885</td>\n",
       "      <td>0.348958</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>3.369578</td>\n",
       "      <td>31.972618</td>\n",
       "      <td>19.355807</td>\n",
       "      <td>15.952218</td>\n",
       "      <td>115.244002</td>\n",
       "      <td>7.884160</td>\n",
       "      <td>0.331329</td>\n",
       "      <td>11.760232</td>\n",
       "      <td>0.476951</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.078000</td>\n",
       "      <td>21.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>99.000000</td>\n",
       "      <td>62.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>27.300000</td>\n",
       "      <td>0.243750</td>\n",
       "      <td>24.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>3.000000</td>\n",
       "      <td>117.000000</td>\n",
       "      <td>72.000000</td>\n",
       "      <td>23.000000</td>\n",
       "      <td>30.500000</td>\n",
       "      <td>32.000000</td>\n",
       "      <td>0.372500</td>\n",
       "      <td>29.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>6.000000</td>\n",
       "      <td>140.250000</td>\n",
       "      <td>80.000000</td>\n",
       "      <td>32.000000</td>\n",
       "      <td>127.250000</td>\n",
       "      <td>36.600000</td>\n",
       "      <td>0.626250</td>\n",
       "      <td>41.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>17.000000</td>\n",
       "      <td>199.000000</td>\n",
       "      <td>122.000000</td>\n",
       "      <td>99.000000</td>\n",
       "      <td>846.000000</td>\n",
       "      <td>67.100000</td>\n",
       "      <td>2.420000</td>\n",
       "      <td>81.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       Pregnancies     Glucose  BloodPressure  SkinThickness     Insulin  \\\n",
       "count   768.000000  768.000000     768.000000     768.000000  768.000000   \n",
       "mean      3.845052  120.894531      69.105469      20.536458   79.799479   \n",
       "std       3.369578   31.972618      19.355807      15.952218  115.244002   \n",
       "min       0.000000    0.000000       0.000000       0.000000    0.000000   \n",
       "25%       1.000000   99.000000      62.000000       0.000000    0.000000   \n",
       "50%       3.000000  117.000000      72.000000      23.000000   30.500000   \n",
       "75%       6.000000  140.250000      80.000000      32.000000  127.250000   \n",
       "max      17.000000  199.000000     122.000000      99.000000  846.000000   \n",
       "\n",
       "              BMI  DiabetesPedigreeFunction         Age     Outcome  \n",
       "count  768.000000                768.000000  768.000000  768.000000  \n",
       "mean    31.992578                  0.471876   33.240885    0.348958  \n",
       "std      7.884160                  0.331329   11.760232    0.476951  \n",
       "min      0.000000                  0.078000   21.000000    0.000000  \n",
       "25%     27.300000                  0.243750   24.000000    0.000000  \n",
       "50%     32.000000                  0.372500   29.000000    0.000000  \n",
       "75%     36.600000                  0.626250   41.000000    1.000000  \n",
       "max     67.100000                  2.420000   81.000000    1.000000  "
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "diabetes.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x1080 with 9 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "diabetes.hist(bins=50, figsize=(20, 15))\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Missing Values"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As we can see, the dabatabase don't have null values, but, there are multiple 0s. This would make sense in some cases, such as pregnancies, but in others, for example, inslulin and blood pressure. This situations needs to be addressed. The approach we'll take will be by replacing the mean values of the whole column to the missing value. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Dividing in Train Test"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We will be making the division of the train and test set stratifically. This is due to the fact that the parameter \"Diabetes Pedigree Function\" is believed to have a huge impact on the outcome, therefore, we believe that the train and the test sets should have the same proportion. Also, as we have a small amount of samples (m = 768) this type of separation can be useful  since, if we were to not to use it, the proportion would be unequal."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    0.200521\n",
       "2    0.200521\n",
       "3    0.200521\n",
       "1    0.199219\n",
       "4    0.199219\n",
       "Name: pedigree_cat, dtype: float64"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "q = diabetes['DiabetesPedigreeFunction'].quantile([0.2,0.4,0.6,0.8]).values\n",
    "\n",
    "def check_pedigree(x):\n",
    "    result = 0\n",
    "    for val in q:\n",
    "        if x > val:\n",
    "            result += 1\n",
    "        else:\n",
    "            break\n",
    "    return result\n",
    "diabetes['pedigree_cat'] = diabetes['DiabetesPedigreeFunction'].apply(check_pedigree)\n",
    "diabetes['pedigree_cat'].value_counts(True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import StratifiedShuffleSplit\n",
    "\n",
    "split = StratifiedShuffleSplit(\n",
    "    n_splits=1,\n",
    "    test_size=0.2,\n",
    "    random_state=RANDOM_SEED,\n",
    ")\n",
    "for train_index, test_index in split.split(diabetes, diabetes['pedigree_cat']):\n",
    "    strat_train_set = diabetes.loc[train_index]\n",
    "    strat_test_set = diabetes.loc[test_index]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0    0.200326\n",
      "2    0.200326\n",
      "3    0.200326\n",
      "4    0.200326\n",
      "1    0.198697\n",
      "Name: pedigree_cat, dtype: float64 \n",
      " 0    0.201299\n",
      "1    0.201299\n",
      "2    0.201299\n",
      "3    0.201299\n",
      "4    0.194805\n",
      "Name: pedigree_cat, dtype: float64\n"
     ]
    }
   ],
   "source": [
    "print(strat_train_set['pedigree_cat'].value_counts(True), '\\n', strat_test_set['pedigree_cat'].value_counts(True))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As we can see, both test and train has the same proportions of values."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Now we need to drop the created column\n",
    "strat_train_set.drop(['pedigree_cat'], axis=1, inplace=True)\n",
    "strat_test_set.drop(['pedigree_cat'], axis=1, inplace=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Exploratory analysis"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "diabetes_train = strat_train_set.copy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x135c243bf48>"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x576 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(16, 8))\n",
    "sns.set(style=\"whitegrid\")\n",
    "corr = diabetes_train.corr()\n",
    "sns.heatmap(corr,annot=True,cmap=\"coolwarm\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x135c2c8bac8>"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x576 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(16, 8))\n",
    "sns.set(style=\"whitegrid\")\n",
    "corr = diabetes_train.corr(method='kendall')\n",
    "sns.heatmap(corr,annot=True,cmap=\"coolwarm\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x135c2a41b08>"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x576 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(16, 8))\n",
    "sns.set(style=\"whitegrid\")\n",
    "corr = diabetes_train.corr(method='spearman')\n",
    "sns.heatmap(corr,annot=True,cmap=\"coolwarm\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As we can see, the kendall correlation seems to be the one with the higher values for the outcome relation with other variables. This indicates that the relationship between the outcome and other variables may not be linear."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can see that the correlation between age and pregnancies is the highest. Also, the outcome appears to have some kind of relation with glucose."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Separating Independent and Dependent Variables"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Replacing the 0s with Nan so that SimpleImputer can operate\n",
    "strat_train_set = strat_train_set.replace(to_replace={\n",
    "             'BloodPressure':{0:np.nan}, \n",
    "             'Insulin':{0:np.nan},\n",
    "             'SkinThickness':{0:np.nan},\n",
    "             'BMI':{0:np.nan},\n",
    "             'Glucose':{0:np.nan},\n",
    "                 })\n",
    "strat_train_set.dropna(subset=['BMI', 'BloodPressure', 'Glucose'], inplace = True)\n",
    "\n",
    "diabetes_train = strat_train_set.drop('Outcome', axis=1)\n",
    "\n",
    "diabetes_labels = strat_train_set['Outcome'].copy()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Replacing Invalid values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "IterativeImputer(random_state=0)"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.experimental import enable_iterative_imputer\n",
    "from sklearn.impute import IterativeImputer\n",
    "\n",
    "\n",
    "diabetes_in = diabetes_train.copy()\n",
    "\n",
    "imputer = IterativeImputer(max_iter=10, random_state=0)\n",
    "\n",
    "imputer.fit(diabetes_in)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABIkAAANiCAYAAADllpmXAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy8QZhcZAAAgAElEQVR4nOzdf2AU9Z3H/9eaX/y0iE2Es5GriMWaCl6qkopE7RECyRoNyAUjyFkleBAVqTmIkSgIRMRLpaDX9qu2AiqQAoHIDzlRWgiKrFZcD5EqwSJcCIr8Znezme8ffNkvCZuQbHZ3ZpPn46/MZ2ZnXzM7u5/dd2Y+YzMMwxAAAAAAAADatYvMDgAAAAAAAADzUSQCAAAAAAAARSIAAAAAAABQJAIAAAAAAIAoEgEAAAAAAEAUiQAAAAAAACAp2uwAwL59+zR48GBdffXVvjbDMDRmzBiNGDHCxGQtU11drUceeURvvvmm2VEAAM1QVlamJUuW6MSJE3K73UpMTNSjjz6qfv366fbbb9cLL7ygn/3sZ2bHBAAEWcPfH3V1derQoYOmTJmi2tpazZgxQxUVFUF5rpdfflm7d+9WSUmJpkyZoi1btqh79+6y2Wyqra1VYmKinnnmGV166aVBeT6gtSgSwRI6dOig8vJy33R1dbUyMzOVlJSkvn37mpis+S677DIKRAAQIf7rv/5LH374oX7zm9/o8ssvlyRt3bpVeXl5Wr58ucnpAACh1vD3x5o1azR16lTNmDEjpM87duxY/epXv/JNl5SU6Omnn9a8efNC+rxAc1EkgiVddtll6tWrl7Zs2aLp06fr1KlT6tKlixYuXKhly5bpjTfeUF1dnbp166Ynn3xSvXv31nfffaepU6fq66+/Vrdu3RQfH68+ffooPz9fP/vZzzRu3Dht2bJFBw8e1AMPPKB77rlHJ0+e1FNPPaW9e/fq+++/V+fOnTV37lxdeeWVGj16tPr376+PPvpIBw4cUEpKimbMmKGLLrpI7777rn7zm9+orq5OnTp10tNPP60uXbrIbrfr448/liS99NJLevvtt1VXV6fLL79cxcXFuuyyy/T222/rpZdeks1mU1RUlAoKCnTDDTeYvMcBoP04dOiQ/vSnP2nDhg1KSEjwtaekpGjKlCk6deqUr+2DDz6o9x/lc6dra2v13HPP6b333lNUVJSuv/56FRcXy2azqaSkRFu3blVUVJSuu+46TZ06VV26dNHrr7+uN998UzExMYqLi9P06dN11VVXqbq6WtOnT9eBAwfk8XiUkZGh8ePHh33fAEB79f333ys+Pr5e27Fjx/T000/r888/l81m0y233KLHHntM0dHR2r59u+bMmaNTp04pJiZGjz76qAYNGiSPx6NnnnlGlZWVuvTSS3XppZeqa9eujT5vSkqKnnvuOUnS7bffruuuu067du3SY489puuuu85v33D2bKePPvpIMTEx+tGPfqTZs2crLi7Ob/vhw4fr/U7Zt2+fb3r58uUqKytr1u8ttA8UiWBJH3/8sb7++mudPn1af//737Vx40Z16dJF27Zt08qVK7V48WJ17NhRmzdv1sSJE7V27Vo988wzuuqqq/S73/1OBw8eVHZ2tvr06SNJcrvduuSSS/Tmm2/K6XRq1KhRGj58uP7yl7/o4osv1pIlSyRJ06ZN0+LFi/Xkk09Kkr7++mstXLhQJ0+e1NChQ7Vt2zZdddVVevzxx/Xaa6/ppz/9qd5++23NnTtXTz31lC//ypUr9cUXX2jZsmWKjo7WkiVLVFRUpD/84Q+aM2eO5s6dq/79+2vz5s364IMPKBIBQBj97W9/U+/evesViM668847m72e119/XZ999pnKy8sVGxurxx57TGvWrNHXX3+tgwcPqry8XFFRUXriiSc0Z84cFRcXa9asWdq4caMSEhK0cuVKORwOX78yduxY3X777XK5XHrwwQd1xRVXaNiwYcHcdADA/+f06dPKysqSJB09elQ1NTVasGBBvWWeeeYZdevWTatXr5bH49FDDz2kV155RXfffbcefvhhvfTSS+rXr592796te++9V2VlZdq4caOqqqr01ltvqba2Vvfee2+jRaLTp09r5cqVuummm3xtffr00W9+8xtJ0pgxY/z2DQkJCdq2bZvWrFkjm82m5557Trt27VJdXZ3fdn/93bma+3sL7QNFIljCuR/SXq9Xl1xyiZ577jl9++23+slPfqIuXbpIkt577z3t3btXOTk5vscePXpU33//vTZt2qQVK1ZIkhISEpSenl7vOX75y19Kkq699lq53W6dPHlS6enpSkxM1MKFC7V3715t27ZN119/ve8xt912my666CJ16dJFvXr10pEjR/TRRx+pT58++ulPfypJSktLU1pamvbt2+d73LvvvqtPP/1Uw4cPl3TmOuez/5nOyMjQxIkTlZqaqptvvlkPPvhgUPclAKBphmHUmz5+/Lhyc3MlyfdPgeaorKxUVlaWOnToIEm+L/UjRozQpEmTFBMTI0kaPXq0JkyYoKioKKWnpysnJ0e33nqrBg4cqNTUVJ08eVIffvihjhw5ohdeeMGX4/PPP6dIBAAh0vBys8rKSk2YMEFFRUW+tr/85S964403ZLPZFBsbq5ycHP3pT3/ST37yE11xxRXq16+fpDOFnX/5l3/Rtm3btHXrVmVmZio2NlaxsbGy2+3atWuXb51//OMftWrVKklnfvfccMMNeuyxx3zzf/7zn0tSk33DwIEDFRUVpbvvvlsDBw7UkCFDdN111+no0aN+28/9neJPc39vdevWLaB9jchCkQiW0PBD+qzly5erU6dOvum6ujplZWXp8ccf900fPHhQP/jBDxQdHV3vi/9FF9W/eV9cXJwkyWazSTrzI+H111/X0qVLlZubK7vdrm7dutX7ED37xf/s4wzDUHR0tG8dZ9eza9cu3wfr2VxnL2mTzpzJdOTIEUnSpEmTNHz4cG3ZskXLly/XK6+8orKyshbuMQBAoK677jrt2bNHhw8f1iWXXKIuXbr4+qDf/va3Onz4sG/Zs5/9Z3k8Ht/f0dH1v0YdOnRIdXV1qqurq9dP1NXV+R43d+5cffHFF6qsrNTvf/97lZeXa+bMmTIMQ2+++aY6duwoSfruu+98/RYAIPR+8Ytf6IorrvB9Dkvy+3leW1srr9dbr10685ugtrb2vPVGRUXVm244JlFDZ3/71NXVNdo3dO7cWeXl5froo4/0/vvv69FHH9WvfvUr5ebm+m2/9dZbG+3Lzn3Os8/b2O8ttA8XXXgRwDoGDhyot956SwcPHpQkvfHGG7rvvvskSampqb5iy+HDh/U///M/5314N7R582bddddduvvuu/XjH/9YGzdulNfrbfIx/fr105dffqndu3dLkt555x3fh+i5OcvKynT8+HFJ0gsvvKCCggLV1tbq9ttv16lTpzRq1CgVFxdr165dcrvdLd8ZAICAXHbZZRozZoweeeQR7d+/39f+zTff6KOPPqr3T4bu3btr//79+vbbb2UYht566y3fvJSUFFVUVMjtdquurk5PPfWU3nrrLd1yyy1644035PF4VFdXp8WLF+vmm2/Wd999p9TUVHXr1k1jx47Vo48+qk8//VRdunRR//799eqrr0o68x/bUaNG6Z133gnfTgGAdm7Pnj365ptvdOzYMV/bwIEDtWjRIhmGIbfbraVLl+oXv/iF+vfvr6+++ko7duyQJO3evVsffvihbrzxRt1yyy1auXKlXC6XXC6X1qxZE1CepvqGd999V2PHjtX111+v/Px83XnnnXI6nY22X3zxxfJ4PPr73/8uSfX6soaa+r2F9oEziRBRBg4cqAcffFD333+/bDabunTpovnz58tms2nq1KkqKirynRH0T//0T/XOBPLn/vvv17Rp03zFpf79++uLL75o8jE//OEPNXfuXP3nf/6nvF6vunTpotLS0nrL3H333aqurtbIkSNls9nUs2dPlZSUKDo6WoWFhfr1r3/tOyNp1qxZio2Nbd2OAQC0yKRJk7Rq1SpNnjxZp06d0rFjx/SDH/xAw4YNU25urjZu3ChJuuqqq5STk6Phw4crPj5et956qz799FNJUk5Ojr755htlZ2fLMAzdeOONGj16tGpra/Xss8/qzjvvVG1tra677jo9+eSTuvjii/XQQw9p7Nix6tChg6KiovTMM89IOnOG0YwZM2S32+V2u5WZmak77rjDtP0DAG3ducNdSGfOmJk+fXq9W9EXFRXpmWeekd1ul8fj0S233KLx48crNjZWL7zwgmbMmKHTp0/LZrNp9uzZ+vGPf6wrrrhCX3/9tTIzM9WtWzf16tUr4IyN9Q1er1d/+ctflJmZqU6dOukHP/iBZsyYoZ49e/pt79q1qx5//HE9+OCD6t69+3nDcpyrqd9baB9sRsML84EItXjxYv30pz/V9ddfL7fbrXvuuUf5+flKTU01OxoAAAAAAJbHmURoM6666irNmDHDN/ZDeno6BSIAAAAAAJqJM4kAAAAAAADAwNUAAAAAAACw8OVmdXV1OnHihGJiYhgkC0CbYBiGPB6POnfuXO/uSQgd+hIAbQ19SfjRlwBoa5rqSyxbJDpx4sQF7zIFAJHo6quvVteuXc2O0S7QlwBoq+hLwoe+BEBb5a8vsWyRKCYmRtKZ0C29PbjT6VRSUlIoYgVNJGSUIiMnGYMnEnJGQkbJf063260vvvjC9/mG0GuqL7HisWS1TOS5MKtlsloeyXqZIj0PfUn4nduXfPHFF5Y6fkLFau+TUGgP2yi1j+1sD9soBXc7m+pLLFskOnsqZ2xsrOLi4lr8+EAeE26RkFGKjJxkDJ5IyBkJGaXGc3KqevhcqC+x4rFktUzkuTCrZbJaHsl6mdpCHvqS8Dm3L5Gsd/yESnvYzvawjVL72M72sI1S8LfTX1/ChcwAAAAAAACgSAQAAAAAAACKRAAAAAAAABBFIgAAAAAAAIgiEQAAAAAAAESRCAAAAAAAAKJIBAAAAAAAAFEkAgAAAAAAgCgSye3xtqgdAABYH/07AMAK6I8QaaLNDmC22Jgo2SeXn9e++vksE9IAAIBgoH8HAFgB/REiTbs/kwgAAAAAAADNLBIdP35cmZmZ2rdvnySpsrJSdrtdaWlpKi0t9S23c+dOZWdna8iQIXriiSdUW1sbmtQAgDbjhRde0LBhw5SRkaFXX31VkjR16lSlpaUpKytLWVlZ2rBhg8kpAQAAgLbvgkWiTz75RKNGjVJVVZUk6fTp0yosLNSLL76oNWvWyOl0atOmTZKkxx9/XNOmTdP69etlGIaWLl0a0vAAgMi2bds2vf/++1q1apX+/Oc/a+HChfrqq6/kdDq1aNEilZeXq7y8XIMHDzY7KgAAANDmXbBItHTpUhUXFyshIUGStGPHDvXq1UuJiYmKjo6W3W7XunXr9M033+j06dPq37+/JCk7O1vr1q0LbXoAQES78cYb9dprryk6OlrffvutvF6vOnTooP3796uwsFB2u13z5s1TXV2d2VEBAACANu+CA1fPnDmz3vTBgwcVHx/vm05ISFB1dfV57fHx8aqurm51QKfTGdDjHA5Hs5ZLTk5u9ToCFer1B0sk5CRj8ERCzkjIKEVOTrPFxMRo3rx5euWVV5Senq7a2loNGDBAxcXF6tq1q/Ly8lRWVqaRI0cG/ByN9SVWfI2slilS84Szf4/UfRROVstEHgAA/Gvx3c3q6upks9l804ZhyGazNdreWklJSYqLi2vRYxwOR5NfDpsrGOtoTLAyhlok5CRj8ERCzkjIKPnP6XK5Ai58t3UPP/ywHnzwQY0fP15bt27VggULfPNGjx6tlStXtqpI5K8vseKxZLVMbTVPMLepre6jYLJapkjPQ1/SuGeffVaHDx9WSUmJdu7cqSeeeEInTpzQz3/+cz399NOKjm73N3YGgAtq8d3NevTooZqaGt90TU2NEhISzms/dOiQ7xI1AAD8+fLLL7Vz505JUseOHZWWlqY1a9Zo/fr1vmUMw+CLPQCgSVu3btWKFSt804yVCgCBaXGRqF+/ftqzZ4/27t0rr9eriooKDRo0SJdffrni4uJ8p8uWl5dr0KBBQQ8MAGg79u3bp6KiIrndbrndbr3zzju64YYbNGvWLB05ckQej0dLlixh4GoAQKO+//57lZaWavz48ZLEWKkA0Aot/tdsXFycSkpKlJ+fL5fLpdTUVKWnp0uS5s6dq6KiIh0/flzXXnutxowZE/TAAIC2IzU1VTt27NCdd96pqKgopaWlaeLEibrkkks0atQo1dbWKi0tTZmZmWZHBQBY1LRp0zRp0iQdOHBA0vljqAZ7rNT2MoZUe9jOcGyjmWPghvt5zNQetlEKz3Y2u0i0ceNG398pKSlatWrVecv07dtXZWVlwUkGAGgX8vPzlZ+fX68tNzdXubm5JiUCAESKZcuWqWfPnkpJSdHy5cslNT6GamslJSXJ6XRaakyrULHa2F2hYIVtDMfzW2E7Q609bKMU3O1sanw7BnkAAAAAEJHWrFmjmpoaZWVl6ciRIzp58qRsNhtjpQJAgCgSAQAAAIhIr776qu/v5cuXa9u2bZo9e7YyMzN9/3VnrFQAaD6KRAAAAADaFMZKBYDAUCQCAAAAEPGys7OVnZ0tibFSASBQF5kdAAAAAAAAq3F7vC1qDzd/OaySDZGLM4kAAAAAAGggNiZK9snl57Wvfj7LhDTn85fPKtkQuTiTCAAAAAAAABSJAAAAAAAAQJEIAAAAAAAAokgEAAAAAAAAUSQCAAAAAACAKBIBAAAAAABAFIkAAAAAAAAgikQAAAAAAAAQRSIAAAAAACzB7fE2qw0IlWizAwAAAAAAACk2Jkr2yeX12lY/n2VSGrRHnEkEAAAAAAAAikQAAAAAAACgSAQAAAAAAABRJAIAAAAAwLIYuBrhxMDVAAAAAABYlL/BrCUGtEZocCYRAMBUL7zwgoYNG6aMjAy9+uqrkqTKykrZ7XalpaWptLTU5IQAAKvz15dMnTpVaWlpysrKUlZWljZs2GBySgCwPs4kAgCYZtu2bXr//fe1atUq1dbWatiwYUpJSVFhYaEWLlyonj17Ki8vT5s2bVJqaqrZcQEAFuSvL0lNTZXT6dSiRYuUkJBgdkQAiBicSQQAMM2NN96o1157TdHR0fr222/l9Xp19OhR9erVS4mJiYqOjpbdbte6devMjgoAsCh/fUmHDh20f/9+FRYWym63a968eaqrqzM7KgBYHmcSAQBMFRMTo3nz5umVV15Renq6Dh48qPj4eN/8hIQEVVdXt+o5nE6n33aHw9Gq9YaC1TJZKU/fa671237i5Gl9vvOzem3JycmNrifY22SlfSRZL49kvUzkaXsa9iW1tbUaMGCAiouL1bVrV+Xl5amsrEwjR44M+DnO9iXt5fVqD9t5oW0MRl/S2DrcHq9iY6KatY6WaLjes8/vr69sS9rD8SqFZzspEgEATPfwww/rwQcf1Pjx41VVVSWbzeabZxhGvelAJCUlKS4url6bw+Fo8sufGayWyWp5JDU6cGdLcgZzm6y2j6yWR7JepkjP43K5Gi18t3fn9iVbt27VggULfPNGjx6tlStXtqpIlJSUJKfTaanjJ1Ss9j4JhdZuY2v3T6gGo25qvW31NW0Px6sU3O1sqi/hcjMAgGm+/PJL7dy5U5LUsWNHpaWl6YMPPlBNTY1vmZqaGsaTAAA0yl9fsmbNGq1fv963jGEYio7m/+MAcCEUiQAAptm3b5+Kiorkdrvldrv1zjvvKCcnR3v27NHevXvl9XpVUVGhQYMGmR0VAGBR/vqSG264QbNmzdKRI0fk8Xi0ZMkSDR482OyoAGB5lNMBAKZJTU3Vjh07dOeddyoqKkppaWnKyMhQ9+7dlZ+fL5fLpdTUVKWnp5sdFQBgUf76kokTJ+qSSy7RqFGjVFtbq7S0NGVmZpodFQAsjyIRAMBU+fn5ys/Pr9eWkpKiVatWmZQIABBp/PUlubm5ys3NNSkR2qNQDUYNhBNFIgAA0G409gWeL/YAgNbyN2h0aweiBsKNIhEAAGg3QnU3GQAAgLaAgasBAAAAAADQuiJReXm5MjIylJGRoWeffVaStHPnTmVnZ2vIkCF64oknVFtbG5SgAAAAAAAACJ2Ai0SnTp3SzJkztXDhQpWXl2v79u2qrKzU448/rmnTpmn9+vUyDENLly4NZl4AAAAAAACEQMBFIq/Xq7q6Op06dUq1tbWqra1VdHS0Tp8+rf79+0uSsrOztW7duqCFBQAAAAAAQGgEPHB1ly5d9Mgjj2jo0KHq2LGjbrjhBsXExCg+Pt63THx8vKqrq1sV0Ol0BvQ4h8PRrOWSk5NbvY5AhXr9wRIJOckYPJGQMxIySpGTEwAAAACkVhSJPv/8c/35z3/Wu+++q65du+rXv/61tmzZIpvN5lvGMIx604FISkpSXFxcix7jcDiaLP40VzDW0ZhgZQy1SMhJxuCJhJyRkFHyn9PlcgVc+AYAAACAUAv4crPNmzcrJSVFl156qWJjY5Wdna0PPvhANTU1vmUOHTqkhISEoARtib7XXHtem9vjDXsOAAAAAACASBHwmUR9+/bVc889p5MnT6pjx47auHGjbrzxRq1fv973H/Ty8nINGjQomHmbpXOnDrJPLq/Xtvr5rLDnAAAAAAAAiBQBF4kGDhyo//3f/1V2drZiYmL0s5/9TOPGjdPgwYNVVFSk48eP69prr9WYMWOCmRcAAAAAAAAhEHCRSJLGjRuncePG1Wvr27evysrKWhUKAAAAAAAA4RXwmEQAAAAAAABoOygSAQAAAAAAgCIRAAAAAAAAKBIFhdvjbVYbAAAAAACAVbVq4GqcERsTJfvk8nptq5/PMikNAAAAAABAy3EmEQAAAICI9sILL2jYsGHKyMjQq6++KkmqrKyU3W5XWlqaSktLTU4IAJGBM4kAAAAARKxt27bp/fff16pVq1RbW6thw4YpJSVFhYWFWrhwoXr27Km8vDxt2rRJqampZscFAEvjTCIAAAAAEevGG2/Ua6+9pujoaH377bfyer06evSoevXqpcTEREVHR8tut2vdunVmRwUAy+NMIgAAAAARLSYmRvPmzdMrr7yi9PR0HTx4UPHx8b75CQkJqq6ubtVzOJ1OSZLD4WjVeiJFW93Ovtdcq86dOig5Oble+4mTp/X5zs/qtTVc5lz+9k9Ty1tBW31Npba9becKx3ZSJAIAmGr+/Plau3atJCk1NVUFBQWaOnWqHA6HOnbsKEmaOHGiBg8ebGZMAIDFPfzww3rwwQc1fvx4VVVVyWaz+eYZhlFvOhBJSUlyOp2WLwQEg8PhaNPb2fCmQ9KZGw+1ZJsjcf9EYubmaOvH61nB3E6Xy+UrfDdEkQgAYJrKykpt3rxZK1askM1m0wMPPKANGzbI6XRq0aJFSkhIMDsiAMDivvzyS7ndbl1zzTXq2LGj0tLStG7dOkVFRfmWqampoU8BgGZgTCIAgGni4+M1ZcoUxcbGKiYmRr1799b+/fu1f/9+FRYWym63a968eaqrqzM7KgDAovbt26eioiK53W653W698847ysnJ0Z49e7R37155vV5VVFRo0KBBZkcFAMvjTCIAgGn69Onj+7uqqkpr167V4sWLtW3bNhUXF6tr167Ky8tTWVmZRo4cGfDzNHY6rRWvX7daJivlacnYEIGcjh3otgbyuLNjYpzL33gY4coTalbLRJ62JTU1VTt27NCdd96pqKgopaWlKSMjQ927d1d+fr5cLpdSU1OVnp5udlQAsDyKRAAA0+3evVt5eXkqKCjQlVdeqQULFvjmjR49WitXrmxVkSgpKUlxcXH12qx4/brVMlktT1OCkTPQwlKgz91wTIyWjocR7DyhYrVMkZ6nqXEk2rP8/Hzl5+fXa0tJSdGqVatMSoRI4/Z4FRsTdeEFgTaOIhEAwFQOh0MPP/ywCgsLlZGRoV27dqmqqkpDhgyRdGaw0ehouisAABA6sTFRfov3QHvDmEQAANMcOHBAEyZM0Ny5c5WRkSHpTFFo1qxZOnLkiDwej5YsWcKdzQAAAIAw4F+zAADTvPzyy3K5XCopKfG15eTkaNy4cRo1apRqa2uVlpamzMxME1MCAAAA7QNFIgCAaYqKilRUVOR3Xm5ubpjTAAAAAO0bl5sBAAAAAACAIhEAAAAAAAAoEgEAAAAAAEAUiQAAAAAAACCKRAAAAAAAABBFIgAAAAAAms3t8ZodAQiZaLMDAAAAtAVuj1exMVEXbAMARLbYmCjZJ5ef1776+SwT0gDBRZEIAAAgCPz9aOAHAwAAiCRcbgYAAAAAAACKRAAAAAAAAKBIBAAAAAAAAFEkAgAAAAAAgBi4GgAARDjuIAZg/vz5Wrt2rSQpNTVVBQUFmjp1qhwOhzp27ChJmjhxogYPHmxmTACwPIpEAAAgonFXMaB9q6ys1ObNm7VixQrZbDY98MAD2rBhg5xOpxYtWqSEhASzIwJAxOByMwAAAAARKz4+XlOmTFFsbKxiYmLUu3dv7d+/X/v371dhYaHsdrvmzZunuro6s6MCgOW16kyijRs3av78+Tp16pRuvvlmFRUVqbKyUrNnz5bL5dLQoUM1adKkYGUFAAAAgHr69Onj+7uqqkpr167V4sWLtW3bNhUXF6tr167Ky8tTWVmZRo4cGfDzOJ1OSZLD4Wh15kjQVrczOTnZ7AimaOzS7BMnT+vznZ+ZkCi42urx2lA4tjPgItE//vEPFRcXa9myZbr00kt13333adOmTSouLtbChQvVs2dP5eXladOmTUpNTQ1mZgAAAACoZ/fu3crLy1NBQYGuvPJKLViwwDdv9OjRWrlyZauKRElJSXI6ne2iyOBwONrFdrYn/i7Nls5cnh3pr3V7OV6DuZ0ul8tX+G4o4MvNNmzYoGHDhqlHjx6KiYlRaWmpOnbsqF69eikxMVHR0dGy2+1at25dwMEBAAAA4EIcDofGjh2ryZMn66677tKuXbu0fv1633zDMBQdzXCsAHAhAX9S7t27VzExMRo/frwOHDigW2+9VX369FF8fLxvmYSEBFVXV7cqYGPVraY0Vl6Owd8AACAASURBVF3zd2pWU5W45p7K1ZLnC2T9ZouEnGQMnkjIGQkZpcjJCZjJ3+nvLo9XcdytDEAzHThwQBMmTFBpaalSUlIknSkKzZo1SwMGDFCnTp20ZMkS3XXXXSYnBQDrC7hI5PV6tX37di1cuFCdOnXSQw89pA4dOshms/mWMQyj3nQgkpKSFBcX16p1nNXSU7NaeyrXhQpQkXBKXCTkJGPwRELOSMgo+c/Z1GmdQHvV2J3JGjslHgAaevnll+VyuVRSUuJry8nJ0bhx4zRq1CjV1tYqLS1NmZmZJqYEgMgQcJHohz/8oVJSUtS9e3dJ0r/+679q3bp1ior6///zV1NTwy0nAQAAAIRMUVGRioqK/M7Lzc0NcxoAiGwBj0l02223afPmzTp69Ki8Xq/++te/Kj09XXv27NHevXvl9XpVUVGhQYMGBTMvAKCNmT9/vjIyMpSRkaE5c+ZIkiorK2W325WWlqbS0lKTEwIAAADtQ8BnEvXr108PPPCA7rnnHnk8Ht18880aNWqUrrzySuXn58vlcik1NVXp6enBzAsAaEMqKyu1efNmrVixQjabTQ888IAqKio0d+5c7pQJAAAAhFmrhvgfMWKERowYUa8tJSVFq1atalUoAED7EB8frylTpig2NlaS1Lt3b1VVVfnulCnJd6dMikQAAABAaHEfSACAafr06eP7u6qqSmvXrtW9994btjtlWvEOdFbLFEievtdcq86dOtRrc7lrFRdr7a8dge77s4/zN6i+v7u3hSJDsNcRbFbLRB4AAPyz9re1dsTfl8iWfrEEgEi1e/du5eXlqaCgQFFRUaqqqvLNC9WdMq14pzyrZWpNnki8Y1kg23qhfeTv7m1S49vd2tffaseQZL1MkZ6HO2UCAEKJIpFFNHYLYABo6xwOhx5++GEVFhYqIyND27ZtU01NjW8+d8oEAAAAwiPgu5sBANBaBw4c0IQJEzR37lxlZGRIOnNjBO6UCQAAAIQfZxIBAEzz8ssvy+VyqaSkxNeWk5OjkpIS7pQJAAAAhBlFojaEcY0ARJqioiIVFRX5ncedMgEAAIDwokjUhjCuEQAAAAAACBRjEgEAAAAAAIAiEQAAAAAAACgSAQAAAAAAQBSJAAAAAAAAIIpEAAAAAAAAEEUiAAAAuT3eZrU1bE9OTg5ZpmBqzrYAAABEmx0AAADAbLExUbJPLq/Xtvr5rGYv29TyVhCJmQEAQPhxJhEAAACAiDZ//nxlZGQoIyNDc+bMkSRVVlbKbrcrLS1NpaWlJicEgMhAkQgAAABAxKqsrNTmzZu1YsUKrVy5Up999pkqKipUWFioF198UWvWrJHT6dSmTZvMjgoAlkeRCAAAAEDEio+P15QpUxQbG6uYmBj17t1bVVVV6tWrlxITExUdHS273a5169aZHRUALI8xiQAAAABErD59+vj+rqqq0tq1a3XvvfcqPj7e156QkKDq6upWPY/T6ZQkORyOVq0nUrTV7YyUGw6Ei9vjVWxMVL02l7tWcbHnlwpOnDytz3d+Fq5oLdJWj9eGwrGdFIkAAAAARLzdu3crLy9PBQUFioqKUlVVlW+eYRiy2WytWn9SUpKcTme7KDI4HI52sZ1o/MYNjd3swIrHRXs5XoO5nS6Xy1f4bojLzQAAAABENIfDobFjx2ry5Mm666671KNHD9XU1Pjm19TUKCEhwcSEABAZKBIBAAAAiFgHDhzQhAkTNHfuXGVkZEiS+vXrpz179mjv3r3yer2qqKjQoEGDTE4KANbH5WYAAAAAItbLL78sl8ulkpISX1tOTo5KSkqUn58vl8ul1NRUpaenm5gSACIDRSIAAAAAEauoqEhFRUV+561atSrMaQAgsnG5GQAAAAAAACgSAQCAprk93ha1Izz87f++11xrQhIAANBWcLkZAABokr/b40pnboUL8zR222IAAIBAcSYRAAAAAAAAKBIBAMx3/PhxZWZmat++fZKkqVOnKi0tTVlZWcrKytKGDRtMTggAACIJl0QDgeFyM9Tj9ngVGxPlm05OTj6vDQCC6ZNPPlFRUZGqqqp8bU6nU4sWLVJCQoJ5wQAAQMTiUmkgMBSJUA/jGwAIt6VLl6q4uFgFBQWSpFOnTmn//v0qLCxUdXW1Bg8erIkTJ+qiizj5FQAAAAglikQAAFPNnDmz3vShQ4c0YMAAFRcXq2vXrsrLy1NZWZlGjhwZ8HM4nU6/7Q6HI+B1horVMjkcDiUnJ/ud19bPNDVj+1ry+jf2uvhbR2PLtvQ5A2XF49pKrJYHANB+USQCAFhKYmKiFixY4JsePXq0Vq5c2aoiUVJSkuLi4uq1NVX8MIvVMl0oT1s/ld+M7QvG69/SdYT6mIu04zrcWprH5XI1WvgGAKC1OHcfAGApu3bt0vr1633ThmEoOpr/aQAAAACh1uoi0bPPPqspU6ZIknbu3Kns7GwNGTJETzzxhGpra1sdEADQvhiGoVmzZunIkSPyeDxasmSJBg8ebHYsAAAAoM1rVZFo69atWrFihW/68ccf17Rp07R+/XoZhqGlS5e2OiAAoH3p27evxo0bp1GjRikjI0PXXHONMjMzzY4FAAAAtHkBn7///fffq7S0VOPHj9fnn3+ub775RqdPn1b//v0lSdnZ2Zo3b57uueeeoIUFALRdGzdu9P2dm5ur3NxcE9MAAAAA7U/ARaJp06Zp0qRJOnDggCTp4MGDio+P982Pj49XdXV1qwMGMjBfuO/20ZLna2x+oOsIRo5gr8MMVs8nRUZGKTJyRkJGKXJyAgAAAIAUYJFo2bJl6tmzp1JSUrR8+XJJUl1dnWw2m28ZwzDqTQfK3x1pAhXuu31cqADVnPWbcZeTUK0jVKx2lxJ/IiGjFBk5IyGj5D8nd6QBAAAAYGUBFYnWrFmjmpoaZWVl6ciRIzp58qRsNptqamp8yxw6dEgJCQlBCwoAAAAAAMzh9ngVGxN1XrvL41Vcg/bGloX1BVQkevXVV31/L1++XNu2bdPs2bOVmZnp++95eXm5Bg0aFLSgAAAAAADAHLExUbJPLj+vffXzWee1r34+K1yxEGSturtZQ3PnztXs2bOVnp6ukydPasyYMcFcPQAAAAD4dfz4cWVmZmrfvn2SpKlTpyotLU1ZWVnKysrShg0bTE4IANYX8MDVZ2VnZys7O1vSmdsWl5WVtToUAAAAADTXJ598oqKiIlVVVfnanE6nFi1axBAYANACQT2TCAAAAMHl9ngt8XzhzgG0xNKlS1VcXOwrCJ06dUr79+9XYWGh7Ha75s2bp7q6OpNTAoD1tfpMIgAAAIROU2NAtIXnA4Jh5syZ9aYPHTqkAQMGqLi4WF27dlVeXp7Kyso0cuTIgJ/j7B1KHQ5Hq7JGikjfzki4G25bF85jKNKP1+YKx3ZSJAIAAADQpiQmJmrBggW+6dGjR2vlypWtKhIlJSXJ6XS2i+LD2ZsRAa0RrmOovRyvwdxOl8vlK3w3xOVmAAAAANqUXbt2af369b5pwzAUHc3/xwHgQigSAQAAAGhTDMPQrFmzdOTIEXk8Hi1ZskSDBw82OxYAWB7ldAAAAABtSt++fTVu3DiNGjVKtbW1SktLU2ZmptmxAMDyKBIBAAAft8er2JgoSQz6GW7n7nsraiyf1XOjfdm4caPv79zcXOXm5pqYBgAiD0UiAADg4+/OVtzVKjysflcxq+cDAACtx5hEAAAAAAAAoEgEAAAAAAAAikQAAAAAAAAQRSJYkNvjbVF7qNYBAAAAAEB7wsDVsJxgDIzJ4JoAgPaIO40BAIDWoEgEAADQRvBPEgAA0BpcbgYAAAAAAACKRAguxgICEIjjx48rMzNT+/btkyRVVlbKbrcrLS1NpaWlJqcDAAAA2gcuN0NQcZo7gJb65JNPVFRUpKqqKknS6dOnVVhYqIULF6pnz57Ky8vTpk2blJqaam5QAAAAoI3jTCIAgKmWLl2q4uJiJSQkSJJ27NihXr16KTExUdHR0bLb7Vq3bp3JKQEAAIC2jzOJAACmmjlzZr3pgwcPKj4+3jedkJCg6urqVj2H0+n02+5wOFq13lAwO1NycrKpz4/wa3jMBXIMXOi4PTu/7zXXqnOnDvXmnTh5Wp/v/KzFz9kaZr/PGrJaHgBA+0WRCABgKXV1dbLZbL5pwzDqTQciKSlJcXFx9docDoflCiJWzIS2LxjHXFPraHhcN7wsffXzWWE97q32PmtpHpfL1WjhGwCA1uJyMwCApfTo0UM1NTW+6ZqaGt+laAAAAABChyIRAMBS+vXrpz179mjv3r3yer2qqKjQoEGDzI4FAAAsijspA8HD5WYAAEuJi4tTSUmJ8vPz5XK5lJqaqvT0dLNjAQAAi/J3h2XurgwEhiIRAMASNm7c6Ps7JSVFq1atMjENACDSHD9+XDk5Ofrv//5v/ehHP1JlZaVmz54tl8uloUOHatKkSWZHBADL43IzAACAdioYl2j4W4dVLv3wlyM5OblF+Rpb1irbiDM++eQTjRo1SlVVVZKk06dPq7CwUC+++KLWrFkjp9OpTZs2mRsSACIAZxIBAAC0U8G4RMPKl3n4yya1LF8w1oHQW7p0qYqLi1VQUCBJ2rFjh3r16qXExERJkt1u17p165SammpmTACwPIpEAAAAACLazJkz600fPHhQ8fHxvumEhARVV1eHOxYARByKRAAAAADalLq6OtlsNt+0YRj1pgPhdDolSQ6Ho1XriRTh2s6+11yrzp06nNd+4uRpfb7zs2atIzk5OdixEAThfK/wvgweikRAI9wer2JjoprdDgAAAGvo0aOHampqfNM1NTVKSEho1TqTkpLkdDrbRUHC4XCEdTsbu6SzPezrtixcr1+4j1ezBHM7XS6Xr/DdEEUioBGMQQAAABCZ+vXrpz179mjv3r360Y9+pIqKCg0fPtzsWABgeRSJAAAAYDmc0YvWiIuLU0lJifLz8+VyuZSamqr09HSzYwGA5VEkAgAAgOVwRi8CsXHjRt/fKSkpWrVqlYlpACDyXGR2AAAAAAAAzuX2eM9rc/lpAxBcrTqTaP78+Vq7dq0kKTU1VQUFBaqsrNTs2bPlcrk0dOhQTZo0KShBAQAAAADtg7+zCVc/n8UZhkCIBXwmUWVlpTZv3qwVK1Zo5cqV+uyzz1RRUaHCwkK9+OKLWrNmjZxOpzZt2hTMvAAAAAAAAAiBgItE8fHxmjJlimJjYxUTE6PevXurqqpKvXr1UmJioqKjo2W327Vu3bpg5gUAAAAAAEAIBHy5WZ8+fXx/V1VVae3atbr33nsVHx/va09ISFB1dXWrAjqdzhY/Jjk52W+7w+Fo9rKNLd/a52tsfqDrCEaOYK4jlPuzqXU0bDcrR1MCeYwZIiFnJGSUIicnAJiJu5UBAGAdrb672e7du5WXl6eCggJFRUWpqqrKN88wDNlstlatPykpSXFxca1MeUZTP/qDsXxLHu9wOJq1/tZmaA/raO6+DHWOprQ0o1kiIWckZJT853S5XAEVvgGgLeMuZgAAWEer7m7mcDg0duxYTZ48WXfddZd69Oihmpoa3/yamholJCS0OiQAAAAAAABCK+Ai0YEDBzRhwgTNnTtXGRkZkqR+/fppz5492rt3r7xeryoqKjRo0KCghQUAAAAAAEBoBHy52csvvyyXy6WSkhJfW05OjkpKSpSfny+Xy6XU1FSlp6cHJSgAAAAAAABCJ+AiUVFRkYqKivzOW7VqVcCBAAAAAAAAEH6tGpMIQNPcHq+k8we6PtsOAGbhcwiRyt+xy/EMANbS2Ocyn9fW1+q7mwFoHHdsAWBVfD4hUvk7djluAcBa+J4RuTiTCAAAAAAAAJxJBACwptGjR+u7775TdPSZrmr69Onq16+fyakAAACAtosiEQDAcgzDUFVVld59911fkQgAAABAaHG5GQDAcr766itJ0v3336877rhDixYtMjkRAAAA0Pbx71kAgOUcPXpUKSkpevLJJ+XxeDRmzBj9+Mc/1s033xzQ+pxOp992h8PRmpghEexMfa+5Vp07dQjqOoHmOPdYPvt3w7t9Bovb41VsTFSrlj9x8rQ+3/nZecs2lbnh+7Wx95vLXau42Ppfu899Pit+FgEA2ieKRAAAy7n++ut1/fXX+6ZHjBihTZs2BVwkSkpKUlxcXL02h8MRsh+sgQpVJu4uAjOcPZbD8V5r6V10GrtDWktz+lu+sRyNPV9L94/L5Wq08I3zMb4dALQMRSIAgOVs375dHo9HKSkpks6MUcTYRACAlmB8OwBoOcYkAgBYzrFjxzRnzhy5XC4dP35cK1as0ODBg82OBQCIIIxvBwAtR0kdAGA5t912mz755BPdeeedqqur0z333FPv8jMAAC4kVOPbtZcxpMK1nVa79BuhF4pji/dl8FAkAgBY0qOPPqpHH33U7BgAgAgVivHtnE5nuyhqWHHcPrQdwT622svxGsztbGp8Oy43AwCgjXB7vGZHACTVPxbbwxd3WNP27du1detW3zTj2wHAhfEpCQBAG9HYHZuAcGvp3caAUDh27JjmzZunN998Ux6PRytWrNDTTz9tdiwAsDSKRAAAAADaHMa3A4CWo0gEAAAAoE1ifDvAWtwer2Jjos5rd3m8imvQ3tiyCC2KRAAAAAAAIOSauhyZS+atgYGrgXagscFsGeQWAAAAAHAWZxIB7QADiAJtC6dfA8HH+woAAIpEAABEHAq/QPDxvgIAgMvNAAAAAAABaMmQBgxzgGBpeCwlJye3+PjiGG0cZxIBFtfY6e/hPi3e3/Nxaj4AAED71ZIz8DhbD8Hi71hq6XEUjHW0VRSJAIuzSofKBykAAAAAtG1cbgYAAAAAAACKRAAAhFNj17u7PF4lJyc3a1kAAAAgFLjcDACAMGrqElIu6QQAAICZOJMIAAAAAABYSjDOqG7JOpo627u1644knEkEAAAAAAAsJRg38AnWHfiscCOhcOFMIgAAAAAAAFAkAgAAAAAAAEUiAAD88nedeUuvPW/ttept9Vp3oC3g/QkAaIsYkwgAAD/8XZfe0mvPW7uOYFyLDyA0gvEZgbbP7fEqNibqgm2hXgfQ1rWH90Tfa649ry0U202RCEDY8CUHAAC0J1b4hwPQHrSHf6x17tQhLJ8FISkSrV69Wi+99JJqa2t13333KTc3NxRPAyDC8CUHLUFfAgBoLfoSAGiZoBeJqqurVVpaquXLlys2NlY5OTm66aabdNVVVwX7qQC0Q+E6zbKpdXP2U+jRlwAAWou+BABaLuhFosrKSg0YMEDdunWTJA0ZMkTr1q3TxIkTW7QewzAkSW63O6Ac3TrX/wHncrmaveyFlm/t8zW2TCDrCEaOYK8jFPvzQuvwN8+MHG1hHS15r5hxfEVH2TR6WkW9tv/nicFyuWpbnKW5Rk9bG9DzNdy2s59nZz/f0Lhw9SUXOv5Cdcw39n5szbJtaR2RmDkY64jEzMFYRyRmDsY6zv08aclnC31J84WqLwnH74RQrKOlmvscLfmeymdL6NcRiZmDsQ6zModbsD4LmupLbEaQe5jf/e53OnnypCZNmiRJWrZsmXbs2KEZM2a0aD3Hjh3TF198EcxoAGAJV199tbp27Wp2DEujLwGAptGXXBh9CQA0zV9fEvQzierq6mSz2XzThmHUm26uzp076+qrr1ZMTExAjwcAqzEMQx6PR507dzY7iuXRlwCAf/QlzUdfAgD+NdWXBL1I1KNHD23fvt03XVNTo4SEhBav56KLLuK/IwDanA4dOpgdISLQlwBA4+hLmoe+BAAa11hfclGwn+gXv/iFtm7dqu+++06nTp3S22+/rUGDBgX7aQAAbRh9CQCgtehLAKDlgn4m0WWXXaZJkyZpzJgx8ng8GjFihK677rpgPw0AoA2jLwEAtBZ9CQC0XNAHrgYAAAAAAEDkCfrlZgAAAAAAAIg8FIkAAAAAAABAkQgAAAAAAAAUiQAAAAAAACCKRAAAAAAAABBFIgAAAAAAAIgiEQAAAAAAAESRCAAAAAAAAIrgItHq1as1bNgwpaWlafHixefN37lzp7KzszVkyBA98cQTqq2tNSGlNH/+fGVkZCgjI0Nz5szxO/+2225TVlaWsrKy/G5LqI0ePVoZGRm+DJ988km9+ZWVlbLb7UpLS1NpaWnY80nSsmXLfPmysrKUnJys6dOn11vGzH15/PhxZWZmat++fZKat8/279+v3Nxcpaen66GHHtKJEyfCmnHJkiXKzMyU3W7X1KlT5Xa7z3vMihUrNHDgQN8+Dcfr3zDn1KlTlZaW5suwYcOG8x4T7vf7uRk3bdpU79gcMGCA8vLyznuMGfsS53v22Wc1ZcoUSeb3Exs3blR2draGDh2qZ555RpK5n7fl5eW+/urZZ5+VZM4+au7naTizNffzM1yZGuY5a9GiRRo9erRvOpz9TMNMH3/8sUaOHKmMjAw99thjpu+jzZs364477lBmZqYKCgp8ecKxj/x9F7TCcY0La8lr1xZYqY8MNqv1uaFilb48FKz4/SAULPGdw4hA//d//2fcdtttxuHDh40TJ04Ydrvd2L17d71lMjIyjI8//tgwDMOYOnWqsXjx4rDn3LJli/Fv//ZvhsvlMtxutzFmzBjj7bffrrdMXl6e8dFHH4U921l1dXXGwIEDDY/H43f+qVOnjNTUVOPrr782PB6Pcf/99xvvvfdemFPW98UXXxiDBw82vv3223rtZu3Lv/3tb0ZmZqZx7bXXGv/4xz+avc/GjRtnVFRUGIZhGPPnzzfmzJkTtoxfffWVMXjwYOPYsWNGXV2dUVBQYLz66qvnPW769OnG6tWrQ5brQjkNwzAyMzON6urqJh8Xzve7v4xnHTx40PjlL39p7Nmz57zHhXtf4nyVlZXGTTfdZPznf/6nYRjm9hNff/21MXDgQOPAgQOG2+02Ro0aZbz33numfd6ePHnSuOGGG4xvv/3W8Hg8xogRI4wtW7aEfR+15PM0XNla8vkZjkyNfQbt3r3buOWWW4x7773X1xaufqZhpmPHjhk333yzsXPnTsMwDGPSpEm+fWHWPho0aJDx97//3TAMw8jPzzeWLl1qGEbo95G/74KrV682/bjGhbX0tYt0Vuojg81qfW6oWKUvDwUrfj8IBat854jIM4kqKys1YMAAdevWTZ06ddKQIUO0bt063/xvvvlGp0+fVv/+/SVJ2dnZ9eaHS3x8vKZMmaLY2FjFxMSod+/e2r9/f71lnE6nfve738lut2v69OlyuVxhzfjVV19Jku6//37dcccdWrRoUb35O3bsUK9evZSYmKjo6GjZ7XZT9uW5nnrqKU2aNEndu3ev127Wvly6dKmKi4uVkJAgqXn7zOPx6MMPP9SQIUMkhf4YbZgxNjZWxcXF6tKli2w2m66++urzjk1J+vTTT7VixQrZ7Xb9+te/1pEjR0KW0V/OU6dOaf/+/SosLJTdbte8efNUV1dX7zHhfr83zHiuOXPmKCcnR//8z/983rxw70vU9/3336u0tFTjx4+XZH4/sWHDBg0bNkw9evRQTEyMSktL1bFjR9M+b71er+rq6nTq1CnV1taqtrZW0dHRYd9Hzf08Defr19zPz3Bl8vcZ5Ha7NW3aND388MO+tnD2Mw0zbdmyRf3791ffvn0lSUVFRRo8eLCp+8jr9er48ePyer1yuVyKi4sLyz7y912wqqrK9OMaF9aS1y7SWa2PDDar9bmhYpW+PBSs+P0gFKzynSMii0QHDx5UfHy8bzohIUHV1dWNzo+Pj683P1z69Onje+Gqqqq0du1apaam+uafOHFC11xzjR5//HGtWLFCR48e1YsvvhjWjEePHlVKSooWLFigP/7xj3rzzTe1ZcsW3/wL7etwq6ys1OnTpzV06NB67Wbuy5kzZ+rnP/+5b7o5++zw4cPq0qWLoqOjJYX+GG2Y8fLLL9fNN98sSfruu++0ePFi/fKXvzzvcfHx8fqP//gPrVq1Sj179jzvEr9Q5zx06JAGDBigWbNmaenSpdq+fbvKysrqPSbc7/eGGc+qqqrStm3bNGbMGL+PC/e+RH3Tpk3TpEmTdPHFF0syv5/Yu3evvF6vxo8fr6ysLL3++uumft526dJFjzzyiIYOHarU1FRdfvnliomJCfs+au7naThfv+Z+foYrk7/PoOeff17Dhw9XYmKiry2c/UzDTHv37lWnTp00adIkZWVl6be//a0uvvhiU/fRU089pdGjR+uWW27R4cOHlZ6eHpZ95O+7oM1mM/24xoW15LWLdFbrI4PNan1uqFilLw8FK34/CAWrfOeIyCJRXV2dbDabb9owjHrTF5ofbrt379b999+vgoKCemcYdO7cWX/4wx/Uu3dvRUdH6/7779emTZvCmu3666/XnDlz1LVrV3Xv3l0jRoyol8Fq+/LNN9/Uv//7v5/XboV9eVZz9pm/NjP2a3V1te677z4NHz5cN91003nzFyxYoOTkZNlsNj3wwAP661//GtZ8iYmJWrBggRISEtSxY0eNHj36vNfVKsfokiVLdM899yg2NtbvfLP3ZXu2bNky9ezZUykpKb42s48br9errVu3atasWVqyZIl27Nihf/zjH6Zl+vzzz/XnP/9Z7777rv7617/qoosu0pYtW0x/bzX2Opn9+knnf36alWnLli06cOCAhg8fXq/dzH7G6/Vq8+bNeuyxx7R8+XKdOnVKv//9703bRzU1NZo7d64qKiq0efNm9evXT7Nnzw7rPjr3u2BiYqJlj2ucrzmvXSSzYh8ZbFbrc0PFqn15KFj5+0EohPs7R0QWiXr06KGamhrfdE1NTb1TihvOP3TokN9LQ8LB4XBo7Nixmjx5su6666568/bv31/vrAjDMHz/zQqX7du3a+vWrY1muNC+Die3260PP/xQt99++3nzrLAvz2rOPuvevbuOHTsmr9fb6DKh9uWXXyonJ0d3W3iqWwAAIABJREFU3XWXJkyYcN78Y8eO6Y9//KNv2jAMRUVFhTGhtGvXLq1fv75ehoavq1Xe7++8846GDRvmd54V9mV7tmbNGm3ZskVZWVmaN2+eNm7cqLKyMlOPmx/+8IdKSUlR9+7d1aFDh/+XvXuPjqo6+zj+C5NJuLa+2ERYSrEiCAVBGotEKREvISEZgzHagBIrRcCFQbJ8QYyxVLwkIjaV10qrBd4l+C5ECySkIeICpWKwyLQljVXLUhIBYwiihHCZmUzm/YPF1JALmczlzOX7WYu1mD2Tc56zz57ZM8/Zex/dcsstqqysNOzzdteuXUpMTNTFF1+smJgYZWZm6q9//avh762OPk+Nft+39/lpVExlZWXav3+/MjIyVFBQoOrqai1YsMDQfuYHP/iBxowZo0GDBslkMik1NVVVVVWG1dHevXs1bNgw/fCHP1SPHj101113ac+ePQGro/O/CwZru0ZbXT13oSwY+0hfC7Y+11+CtS/3h0j6HDXiO0dIJomuv/567d69W8eOHdPp06e1bds2TZw40f38pZdeqtjYWFmtVklnV3n/7vOBUldXp3nz5mn58uVKS0tr83zPnj313HPP6eDBg3K5XHrttdd06623BjTGEydOaNmyZbLZbGpqatKmTZtaxTBmzBgdOHDAPUyzrKzMkLqUziYMLr/8cvXu3bvNc8FQl+d0pc7MZrOuvfZalZeXS5I2b94c0HptamrSL3/5Sz300EOaOXNmu6/p3bu3/vjHP7rvdrdu3bqA16nL5dIzzzyj48ePy+Fw6PXXX28TQzC8348dO6YzZ860murxXcFQl5FszZo1KisrU0lJiebPn6+bbrpJhYWFhrabSZMmadeuXWpsbJTT6dR7772nlJQUwz5vhw8frsrKSp06dUoul0s7duzQuHHjDH9vdfR5auT7vqPPT6NiKiws1NatW1VSUqKnnnpKo0aN0m9/+1tD+5kJEyboo48+Ul1dnSTpnXfe0ciRIw2ro2HDhqmqqkpHjx6VdDapf/XVVwekjtr7LhiM7RpteXLuQlkw9pG+Fmx9rr8Ea1/uD5HyOWrUdw5jhlp46ZJLLlFeXp5ycnLkcDiUlZWl0aNH6/7779f8+fN19dVXa/ny5SooKFBTU5NGjhzZ4Toh/rRq1SrZbDYVFRW5y7Kzs7Vjxw53nEuXLtUDDzwgh8Ohn/zkJ+1OpfKnSZMmad++fZo6dapaWlo0ffp0jR07VhkZGXr55Zd1ySWXqKioSLm5ubLZbEpKSlJKSkpAYzzn4MGDGjBgQKuy755zo+vynNjY2A7r7LHHHtNNN92km2++WUuWLNHixYu1cuVKDRw4UL/5zW8CFuObb76po0ePas2aNVqzZo0k6aabbtJDDz3UKsbf/va3+vWvf60zZ87o8ssvd9/+NVCGDx+u2bNna9q0aWpublZycrLS09MlKaje74cOHWrTNiUFVV2iLSPbzZgxYzRr1ixNnz5dDodDN9xwg6ZNm6YrrrjCkM/bCRMm6F//+pcyMzNlNpt19dVXa/bs2br11lsNfW919nlq1Pnr7PPT6M+i8xnVz5xbd23u3Lmy2WwaMWKEHnnkEUnGnLchQ4booYceUk5OjkwmkwYPHuxeF87fddTRd8Fga9doy9NzF27CqS0GW5/rL8Hal/tDMH4/8AejvnNEuVwul8+2BgAAAAAAgJAUktPNAAAAAAAA4FskiQAAAAAAAECSCAAAAAAAACSJAAAAAAAAIJJEAAAAAAAAEEkiAAAAAAAAiCQRAAAAAAAARJIIAAAAAAAAIkkEAAAAAAAAkSQCAAAAAACASBIBAAAAAABAJIkAAAAAAAAgkkQAAAAAAAAQSSIAAAAAAACIJBEAAAAAAABEkggAAAAAAAAiSQQAAAAAAACRJAIAAAAAAIBIEgEAAAAAAEAkiQAAAAAAACCSRAAAAAAAABBJIgAAAAAAAIgkEQAAAAAAAESSCAAAAAAAACJJBAAAAAAAAJEkAgAAAAAAgEgSAQAAAAAAQCSJAAAAAAAAIJJEAAAAAAAAEEkiAAAAAAAAiCQRAAAAAAAARJIIAAAAAAAAIkkEA/zjH//QjBkzZLFYlJ6erlmzZmn//v3661//qvT09Dav/+c//6n58+d3us3NmzcrIyNDGRkZGjdunH72s5+5H+/du1czZsxQRUVFm7+rr69XdnZ2p9v+n//5Hy1dutSzgwQAhJSbbrpJ//znP326zcWLF2vVqlWSpIyMDDU2Nvp0+wCA0HHo0CGNGDHC/RvFYrHozjvvlNVq1aFDh3TVVVfpnnvuafN3ixcv1lVXXaVjx45J8k9/BXxXtNEBILLY7XbNmTNHq1ev1siRIyVJJSUluv/++1VYWNju31x99dVasWJFp9udOnWqpk6dKunsB+nQoUP1y1/+8oLxXHLJJVq/fr2HRwEAgGdKSkqMDgEAYLCePXu26g/Ky8v16KOPavXq1YqNjdWBAwd0+PBhXXrppZKkU6dO6W9/+5tR4SJCkSRCQJ0+fVonTpzQqVOn3GW33Xab+vbtK6fT6S7bu3ev/vu//1u/+c1v5HA49OSTT6qsrEyLFy9W37599emnn+qrr77SVVddpWeffVZ9+vS54L63b9+uVatW6ejRo0pMTNRTTz2lL7/8UhaLRX//+9/V3Nys5557Tu+++65MJpPGjh2rJUuWtNrG//7v/2rjxo1atWqV1q9fr8OHD6uhoUGHDx/WJZdcoueee07x8fGqr6/X0qVLVVdXJ4fDobS0NM2dO1fNzc168skn9be//U1ms1mXXXaZCgsLFRsb2255V44LAOA7V199tWbPnq33339fR44c0axZszR9+nQ1NDTokUce0TfffCNJSkpK0oIFC7Rx40a99dZb+sMf/iBJbR6fc9VVV2n37t1699139fbbb6tHjx6qra1Vz5499eyzz2rIkCEBP1YAgLG+/fZbxcXFSZJMJpNSU1O1ZcsWzZ07V5K0bds23XzzzVq9erWRYSLCMN0MAfX9739fCxcu1KxZs3TzzTdr4cKF+tOf/qTrr79eZrNZkvTBBx/o0Ucf1e9//3v95Cc/abON6upqrVq1SuXl5Tp8+HC708jac/LkSa1fv17l5eX6y1/+0iYr/3//93/66KOPVFJSorKyMp08eVLl5eXu51955RVVVFRo3bp17g/zvXv36oUXXlBFRYV69erlHpW0cOFC3XHHHdq4caPefPNNVVZWqry8XP/4xz+0Z88elZaWauPGjRo0aJA+/fTTDssBAIFlt9v1X//1X1q/fr1WrFihwsJC2Ww2bdiwQZdddpk2bdqk1157TbW1tTpx4kS39vHhhx/q8ccfV1lZmcaMGaOXX37Zx0cBAAhGZ86ccU83mzRpkp555hnNnj3b/fzUqVNbjTTavHmzbr/9diNCRQRjJBEC7r777tOdd96pDz/8UB9++KFeeeUVvfLKK1q4cKG++uorzZ07V9OmTdPw4cPb/fuf/exniomJkSQNGzZMx48f79J+p0yZIpPJpF69eunyyy/X119/rQEDBrifr6ysVEZGhnr27ClJ+u1vfyvp7JpE27ZtU0NDg37/+9/re9/7nvtvxo0bp759+0qSfvzjH+v48eM6deqUPvzwQx0/flwvvPCCpLNDRT/55BNNmDBBJpNJd955pyZMmKDJkydr9OjRamxsbLccABB4N998syRp5MiRstvtOnXqlH72s59p9uzZqqur0/XXX6+HH35Y/fr169b2R44c6e5/fvzjH+vtt9/2WewAgOB1/nSzyspKzZs3T6WlpZKkUaNGyWQyqbq6WhdffLFOnjypYcOGGRUuIhQjiRBQVqtVf/zjH9W3b19NmjRJixYt0p///GdFRUWpublZJpNJq1ev1qZNm7Rv3752t3EuiSNJUVFRcrlcXdp3dPR/cqLt/d13n5eko0eP6siRI5KkwYMHa8WKFXriiSdaLTzaXiwtLS1yuVxav369SkpKVFJSotdff11z5szR9773PZWUlOiRRx6RyWTSggUL9Nprr3VYDgAIvNjYWElnP9clyeVyafTo0dq+fbt+/vOf6/Dhw7rzzjtVXV3dpj9xOBwX3H53+zEAQHi5/vrr9cMf/lBbt251l912220qLS1VSUmJMjIyDIwOkYokEQKqf//+Wrlypfbu3esua2hoUFNTk3tO7k9+8hM98sgjWrRokU6fPh2w2BITE1VWVia73a6Wlhb9+te/1p///GdJZ9eSmDx5shITE/XEE090up2+ffvqmmuu0Zo1ayRJjY2NmjZtmrZv36533nlHv/jFLzR27Fjl5uZq6tSpqq6u7rAcABAcli9frpdeekm33HKLHnvsMV155ZXav3+/+vfvr/3798tms8nhcOitt94yOlQAQIg4t1D1rbfe6i7LyMhQRUWFysvL273zM+BvTDdDQP3oRz/S7373OxUXF+urr75SbGys+vXrp2eeecZ95VaSbr/9dr311lsqKirSlClTAhJbdna2Dh8+rMzMTLlcLo0bN04zZszQypUr3a/Jz89Xenp6q7WK2rN8+XI9+eSTslgsstvtSk9P12233San06m//OUvSk9PV+/evfX9739fTz75pAYOHNhuOQAgONx7771avHix0tPTFRMTo6uuukppaWnq0aOHfvrTnyo1NVVxcXG67rrrWFMOANCuc2sSndPS0qKlS5e612aVzt59eciQIerXr58uuugiI8JEhItyMcYZAAAAAAAg4jHdDAAAAAAAACSJAAAAAAAAQJIIAAAAAAAACuKFq1taWnTy5EmZzWb3LWgBIJS5XC45HA716dNHPXqQow8E+hIA4Ya+JPDoSwCEm876kqBNEp08eVL//ve/jQ4DAHxu2LBh6tevn9FhRAT6EgDhir4kcOhLAISr9vqSoE0SnbsN4LBhwxQTE2NwNL5TXV2tUaNGGR1GwEXicXPMkaOrx2232/Xvf/+71W1OI90bb7yhdevWuR8fOnRIGRkZuuWWW1RYWCibzabU1FTl5eV1a/u+7EsitX2fj3qgDs6hHoypA/qSwAvk7xLeV61RH21RJ61RH211pU4660uCNkl0bihnTEyMYmNjDY7Gt8LteLoqEo+bY44cnhw3Q9X/484779Sdd94pSdq/f7/mzZun+++/X9OmTdPatWs1cOBAzZkzRzt37lRSUpLH2/d1XxKp7ft81AN1cA71YFwd0JcETqB/l/C+ao36aIs6aY36aKurddJeX8JEZgBAUPj1r3+tvLw8HTx4UIMHD9agQYMUHR0ti8WiiooKo8MDAAAAwl7QjiQCAESOyspKnTlzRqmpqSorK1NcXJz7ufj4eNXX13u1/erqam9DlCRZrVafbCfUUQ/UwTnUA3UAAAgvJIkAAIZbv3697rvvPkln7yLz3aGvLpfL62kVo0aN8noostVqVUJCglfbCAfUA3VwDvVgTB3YbDafJb4BADgf080AAIay2+368MMPddNNN0mSBgwYoIaGBvfzDQ0Nio+PNyo8AAAAIGKQJAIAGOrTTz/V5Zdfrt69e0uSxowZowMHDqi2tlZOp1NlZWWaOHGiwVECAAAA4Y/pZgAAQx08eFADBgxwP46NjVVRUZFyc3Nls9mUlJSklJQUAyMEAAAAIgNJIgCAoaZMmaIpU6a0KktMTFRpaalBEQEAAACRielmAAAAAAAAIEkUTuwOZ5fKAADhiX4AABAu7A5nm7sH0qcB/sd0szASYzbJ8nBJq7Itz2cYFA0AINDoBwAA4YI+DTAGI4kAAAAAAABAkggAAAAAAAAkiQAAAAAAACCSRAAAAAAAABBJIgAAAAAAAIgkEQAAAAAAAESSCAAAAAAAACJJBAAAAAAAAJEkAgAAAAAAgEgSAQAAAAAAQCSJAAAAAAAAIJJEAAAAAAAfsjucXSoLJqEYM+AP0UYHAAAAAAAIHzFmkywPl7Qq2/J8hkHRdE0oxgz4AyOJAAAAAAAAQJIIAAAAQGh74YUXNGXKFKWlpWnNmjWSpMrKSlksFiUnJ6u4uNjgCAEgNDDdDAAAAEDI2rNnjz744AOVlpaqublZU6ZMUWJiovLz87V27VoNHDhQc+bM0c6dO5WUlGR0uAAQ1BhJBAAAACBkjRs3Tq+++qqio6P19ddfy+l0qrGxUYMHD9agQYMUHR0ti8WiiooKo0MFgKDHSKIIZXc4FWM2dbkcAAAACFZms1krVqzQ6tWrlZKSoiNHjiguLs79fHx8vOrr673aR3V1tbdhdonVag3IfvwpISGh3XJPjs3bbQwfMVJ9evdsU37y1Bl98vFHPt9fIAVjTEaiPtrypk68ShLNmDFDx44dU3T02c0sXbpUX3zxhVauXKnm5mbde++9uvvuu73ZBfykvdX7JVbwBwAAQGiaP3++7r//fs2dO1c1NTWKiopyP+dyuVo97o5Ro0YpNjbW2zA7ZbVaO0xWhANfHJsn2+jo944n2wi28xHubcRT1EdbXakTm83WYeK720kil8ulmpoavfPOO+4kUX19vfLy8rRx40bFxMQoOztb1113na688sru7gYAAAAAOvTZZ5/JbrdrxIgR6tWrl5KTk1VRUSGT6T+j4xsaGhQfH29glAAQGrq9JtHnn38uSZo5c6Zuu+02rVu3TpWVlRo/frwuuugi9e7dW5MnT2buLwCgUzt27FBmZqZSU1P11FNPSeKONACArjt06JAKCgpkt9tlt9u1fft2ZWdn68CBA6qtrZXT6VRZWZkmTpxodKgAEPS6PZKosbFRiYmJevzxx+VwOJSTk6PU1NQ2c3+rqqq8CjBQc38DyV9zJj2ZR9vZ8DN/xReJc0U55sgRqcftrYMHD2rJkiV64403dPHFF+vee+/Vzp07tWTJEu5IAwDokqSkJFVVVWnq1KkymUxKTk5WWlqa+vfvr9zcXNlsNiUlJSklJcXoUAEg6HU7STR27FiNHTvW/TgrK0uFhYV64IEH3GWhMvc3kIyYM+np/vwRXyTOFeWYI0dXj7uzub+R6u2339aUKVM0YMAASVJxcbFqa2vdd6SR5L4jDUkiAEBHcnNzlZub26osMTFRpaWlBkUEAKGp20mivXv3yuFwKDExUdLZhNCll16qhoYG92uY+wsA6Extba3MZrPmzp2ruro63XjjjRo6dGjQ3pEm2EeMBerOLMFeD4FAHZxFPVAHAIDw0u0k0YkTJ7RixQqtX79eDodDmzZt0nPPPaeFCxfq2LFj6tWrl7Zt26Ynn3zSl/ECAMKI0+nU3r17tXbtWvXu3VsPPPCAevbsGZR3pAnlkXK+jDuU68FXqIOzqAdj6oBRqUDg2B1OxZhNXS4HwkG3k0STJk3Svn37NHXqVLW0tGj69OlKSEhQXl6ecnJy5HA4lJWVpdGjR/syXogPJQDh4wc/+IESExPVv39/SdItt9zCHWkAAEBQiDGbZHm4pE35luczDIgGCIxuJ4kkacGCBVqwYEGrMovFIovF4lVQ6BwfVgDCxaRJk/TII4+osbFRffr00XvvvaeUlBS9/PLLqq2t1WWXXaaysjLdcccdRocKAAAAhD2vkkQAAHhjzJgxmjVrlqZPny6Hw6EbbrhB06ZN0xVXXMEdaQAAAIAAI0kEADBUVlaWsrKyWpVxRxoAAAAg8HoYHQAAAAAAAACMR5IIAAAAAAAAJIkAAAAAAABAkggAAAAAAAAiSQQAAAAAAACRJAIAAAAAAIBIEgEAAAAAAEAkiQAAAAAAACCSRAAAAAAAABBJIgAAAAAAAIgkEQAAAAAAAESSCAAAAAAAACJJBAAAAAAAAJEkAgAgaNkdTo/KAQAAAG9EGx0AAABoX4zZJMvDJW3KtzyfYUA0AACEPrvDqRizyegwgKBFkggAAAAAEBG4AAN0julmAAAAAAAAIEkEAAAAAAAAppsBAAAACHEvvviitm7dKklKSkrSokWL9Oijj8pqtapXr16SpAcffFC33nqrkWECQNAjSQQAAAAgZFVWVmrXrl3atGmToqKiNGvWLL399tuqrq7WunXrFB8fb3SIABAymG4GAAAAIGTFxcVp8eLFiomJkdls1pAhQ/Tll1/qyy+/VH5+viwWi1asWKGWlhajQwWAoMdIIgAAAAAha+jQoe7/19TUaOvWrXrttde0Z88eLVmyRP369dOcOXP05ptv6q677ur2fqqrq30R7gVZrdaA7MefEhIS2i335Ng82cbwESPVp3fPLm/bF4w8T+HQRnyJ+mjLmzohSQQAAAAg5O3fv19z5szRokWLdMUVV+h3v/ud+7kZM2Zo8+bNXiWJRo0apdjYWF+E2iGr1dphciQc+OLYOtrG+be19/ct7Y06T+HeRjxFfbTVlTqx2WwdJr5JEgEADDVjxgwdO3ZM0dFnu6SlS5fqiy++0MqVK9Xc3Kx7771Xd999t8FRAgCCmdVq1fz585Wfn6+0tDR9+umnqqmp0eTJkyVJLpfL3c8AADrGJyUAwDAul0s1NTV655133F/e6+vrlZeXp40bNyomJkbZ2dm67rrrdOWVVxocLQAgGNXV1WnevHkqLi5WYmKipLP9yzPPPKPx48erd+/eev3113X77bcbHCkABD+SRAAAw3z++eeSpJkzZ+rbb7/VXXfdpT59+mj8+PG66KKLJEmTJ09WRUWFHnzwQSNDBQAEqVWrVslms6moqMhdlp2drdmzZ2vatGlqbm5WcnKy0tPTDYwSAEIDSSIAgGEaGxuVmJioxx9/XA6HQzk5OUpNTVVcXJz7NfHx8aqqqvJqP75abDTQCyN2NJ/c7nAqxmzq8nZ8HTcLRFIH51AP1EEwKCgoUEFBQbvPMV0ZADzjdZLo2Wef1TfffKOioiJ9/PHHeuyxx3Ty5Elde+21euKJJ5j7CwDo0NixYzV27Fj346ysLBUWFuqBBx5wl7lcLkVFRXm1H18sNhpMCyPGmE1tFuiUOl6k05dxB1M9GIU6OIt6MKYOOltsFAhmHV3g8PTCBwD/8iqDs3v3bm3atEk33nijJGnhwoV66qmndM011yg/P18bNmzQ9OnTfREnACAM7d27Vw6Ho9UaEpdeeqkaGhrcr2loaFB8fLxRIQIAAB/w9AIHAGP06O4ffvvttyouLtbcuXMlSYcPH9aZM2d0zTXXSJIyMzNVUVHhmygBAGHpxIkTWrZsmWw2m5qamrRp0yY999xz2r17t44dO6bTp09r27ZtmjhxotGhAgAAAGGv2yOJfvWrXykvL091dXWSpCNHjrRaQyIuLk719fVeBxiOw2nbm7s+fMRI9ends1XZyVNn9MnHH7V5rafDmtvbX2fb8Nfc+kics88xR45IPW5vTZo0Sfv27dPUqVPV0tKi6dOnKyEhQXl5ecrJyZHD4VBWVpZGjx5tdKgAAABA2OtWkuiNN97QwIEDlZiYqI0bN0qSWlpaWq0Z4Ys1JCTfrCMRTDqbu37+8Mstz2f4ZJ67p9vwx9z6SFy3gGOOHF09btaRaN+CBQu0YMGCVmUWi0UWi8WgiAAAAIDI1K0kUXl5uRoaGpSRkaHjx4/r1KlTioqKarWGxNGjR1lDAgCALmLhTgAAABitW0miNWvWuP+/ceNG7dmzR4WFhUpPT3dfUS8pKWENCQAAuqi9BT1ZzBMAAACB5NP70y9fvlwFBQVqamrSyJEjlZOT48vNAwAAAADCiCcjaUNx1G1HMYfisSAyeJ0kyszMVGZmpiRp+PDhevPNN70OCgAAAAAQ/tobSSu1P5rWk9cGi1CMGZGth9EBAAAAAAAAwHgkiQAAAAAAAECSCAAAAAAAACSJAAAAAAAAIJJEAAAAAAAAEEkiAAAAAAAAiCQRAAAAAAAARJIIAAAAAAAAIkkEAAAAAAAAkSQCAAAAAACASBKhC+wOp0flAAAAAAAg9EQbHQCCX4zZJMvDJW3KtzyfYUA0AAAAALxldzgVYzZ1uRxAZCBJBAAAAAARhgvBANrDdDMAAAAAAACQJAIAAAAQ2l588UWlpaUpLS1Ny5YtkyRVVlbKYrEoOTlZxcXFBkcIAKGBJBEAAACAkFVZWaldu3Zp06ZN2rx5sz766COVlZUpPz9fL730ksrLy1VdXa2dO3caHSoABD2SRAAAAABCVlxcnBYvXqyYmBiZzWYNGTJENTU1Gjx4sAYNGqTo6GhZLBZVVFQYHSoABD2SRAAAAABC1tChQ3XNNddIkmpqarR161ZFRUUpLi7O/Zr4+HjV19cbFSIAhAzubgYAMNyzzz6rb775RkVFRfr444/12GOP6eTJk7r22mv1xBNPKDqa7goA0Ln9+/drzpw5WrRokUwmk2pqatzPuVwuRUVFebX96upqLyPsGqvVGpD9JCQk+C2GzrYdLrpaR/6o50C1kVBBfbTlTZ3wrRsAYKjdu3dr06ZNuvHGGyVJCxcu1FNPPaVrrrlG+fn52rBhg6ZPn25skACAoGa1WjV//nzl5+crLS1Ne/bsUUNDg/v5hoYGxcfHe7WPUaNGKTY21ttQO2W1WoMiwRIMMQQ7X9RRd7YRLG0kWFAfbXWlTmw2W4eJb6abAQAM8+2336q4uFhz586VJB0+fFhnzpxxTxvIzMxkDQkAQKfq6uo0b948LV++XGlpaZKkMWPG6MCBA6qtrZXT6VRZWZkmTpxocKQAEPwYSQQAMMyvfvUr5eXlqa6uTpJ05MiRVmtIxMXF+WQNCV9NEfDncGZ/XgVrL+7hI0aqT++ebcpPnjqjTz7+yOPtRRrq4CzqgToIBqtWrZLNZlNRUZG7LDs7W0VFRcrNzZXNZlNSUpJSUlIMjBIAQgNJIgCAId544w0NHDhQiYmJ2rhxoySppaWl1ZoRvlhDQvLNFIFQHs7cUdyWh0valG15PuOC6yeEaj34CnVwFvVgTB10NkUgUhUUFKigoKDd50pLSwMcDQCENpJEAABDlJeXq6GhQRkZGTp+/LhOnTqlqKhamQlOAAAgAElEQVSoVmtIHD161Os1JAAAAAB0DUkiAIAh1qxZ4/7/xo0btWfPHhUWFio9Pd19db6kpIQ1JAAAMJjd4VSM2dTl8nDX0XHbHE7FRmB9ILyQJApivvjQjdQPbgCha/ny5SooKFBTU5NGjhypnJwco0PyqUB/LtMPAAC8FWM2dThFORJ1Vh/nl0dqHSF0kSQKYr74MOYDHUAoyMzMVGZmpiRp+PDhevPNNw2OyH8C/blMPwAAAICu6mF0AAAAAAAAADAeSSIAAAAAAACQJPIFu8PZpTIAAAAAAIBg5dWaRC+88ILeeustRUVFKSsrS/fdd58qKytVWFgom82m1NRU5eXl+SrWoNXeeg+s9QAAAAAAAEJJt0cS7dmzRx988IFKS0v1pz/9SWvXrtUnn3yi/Px8vfTSSyovL1d1dbV27tzpy3gBAIAfMToWACKbJ5/59A++Rz8Mo3V7JNG4ceP06quvKjo6WvX19XI6nWpsbNTgwYM1aNAgSZLFYlFFRYWSkpJ8FjAAAPAfRscCQGTzpB/gDpq+Rz8Mo3k13cxsNmvFihVavXq1UlJSdOTIEcXFxbmfj4+PV319vVcBVldXe/X3gZCQkNBuudVq7XJ5R9swwvnxdRZbR8fo7WvDBcccOSL1uAEAAACED6+SRJI0f/583X///Zo7d65qamoUFRXlfs7lcrV63B2jRo1SbGyst2Eaor3kitVqDaqEUHs8ia+rrw2F4/Y1jjlydPW4bTZbSCS+4Tt2h1MxZpPRYQAAAABd0u0k0WeffSa73a4RI0aoV69eSk5OVkVFhUym/3wZbmhoUHx8vE8CBQAg1DBkHAAAAKGk2wtXHzp0SAUFBbLb7bLb7dq+fbuys7N14MAB1dbWyul0qqysTBMnTvRlvAAAAAAAAPCDbo8kSkpKUlVVlaZOnSqTyaTk5GSlpaWpf//+ys3Nlc1mU1JSklJSUnwZL/zM26kRHf398BEjvQkLAAAAAAD4mVdrEuXm5io3N7dVWWJiokpLS70KCsbxdmoEdzgAAAAAACA0dXu6GQAAAAAAAMIHSSIAAAAAAACQJAIAAAAAAABJIgAAAAAAAIgkEQAAAAAAAESSCAAAAACAgLI7nEaHALQr2ugAAAAAAACIJDFmkywPl7Qp3/J8hgHRAP/BSCIAAAAAAACQJAIAAAAQ+pqampSenq5Dhw5Jkh599FElJycrIyNDGRkZevvttw2OEACCH9PNAmz4iJFGhwAAAACElX379qmgoEA1NTXusurqaq1bt07x8fHGBQYAIYYkUYD16d2TuacAAACAD23YsEFLlizRokWLJEmnT5/Wl19+qfz8fNXX1+vWW2/Vgw8+qB49mEgBAJ0hSQQAMNQLL7ygt956S1FRUcrKytJ9992nyspKFRYWymazKTU1VXl5eUaHCQAIYk8//XSrx0ePHtX48eO1ZMkS9evXT3PmzNGbb76pu+66q9v7qK6u9jbMLrFarQHZT0JCQkD2A9/4brsIVBsJFdRHW97UCUkiAIBh9uzZow8++EClpaVqbm7WlClTlJiYqPz8fK1du1YDBw7UnDlztHPnTiUlJRkdLgAgRAwaNEi/+93v3I9nzJihzZs3e5UkGjVqlGJjY30RXoesVivJG7TrXLugjbRGfbTVlTqx2WwdJr4ZbwkAMMy4ceP06quvKjo6Wl9//bWcTqcaGxs1ePBgDRo0SNHR0bJYLKqoqDA6VABACPn000/11ltvuR+7XC5FR3N9HAAuhE9KAIChzGazVqxYodWrVyslJUVHjhxRXFyc+/n4+HjV19d7tQ9fTRHwdOhuOF3ZYph7a9TBWdQDdRCsXC6XnnnmGY0fP169e/fW66+/rttvv93osAAg6JEkAgAYbv78+br//vs1d+5c1dTUKCoqyv2cy+Vq9bg7fDFFINKHMzPM/T+og7OoB2PqoLMpAviP4cOHa/bs2Zo2bZqam5uVnJys9PR0o8MCgKBHkggAYJjPPvtMdrtdI0aMUK9evZScnKyKigqZTCb3axoaGrh9MQCgS3bs2OH+/9133627777bwGgAIPSwJhG6ze5wGh0CgBB36NAhFRQUyG63y263a/v27crOztaBAwdUW1srp9OpsrIyTZw40ehQAQAAgLDHSCJ0W4zZJMvDJa3KtjyfYVA0AEJRUlKSqqqqNHXqVJlMJiUnJystLU39+/dXbm6ubDabkpKSlJKSYnSoAAAAQNgjSQQAMFRubq5yc3NblSUmJqq0tNSgiAAAAIDIxHQzAAAAAAAAkCQCAAAAAAAASSIAAAAAAACIJBEAAAAAAABEkggAAAAAAAAiSeQ3dofT6BAAAAAAAAC6LNroAMJVjNkky8Mlbcq3PJ9hQDQAAAAAAACdYyQRAAAAAAAASBIBAAAAAADAyyTRiy++qLS0NKWlpWnZsmWSpMrKSlksFiUnJ6u4uNgnQQIAAAAAAMC/up0kqqys1K5du7Rp0yZt3rxZH330kcrKypSfn6+XXnpJ5eXlqq6u1s6dO30ZLwAAAAAAAPyg20miuLg4LV68WDExMTKbzRoyZIhqamo0ePBgDRo0SNHR0bJYLKqoqPBlvAAAAAAAAPCDbt/dbOjQoe7/19TUaOvWrbrnnnsUFxfnLo+Pj1d9fb1XAVZXV3v19740fMRI9end0+gwQpbVajU6hIDjmCNHpB43AAAAgPDR7STROfv379ecOXO0aNEimUwm1dTUuJ9zuVyKioryavujRo1SbGysl1H6Dre1776EhASjQwgoq9XKMUeIrh63zWYLqsQ30FV2h1MxZpOk1p/l3y0HAABA6PMqSWS1WjV//nzl5+crLS1Ne/bsUUNDg/v5hoYGxcfHex0kAAAwTozZxEUSAACACNDtNYnq6uo0b948LV++XGlpaZKkMWPG6MCBA6qtrZXT6VRZWZkmTpzos2ABAAAAAADgH90eSbRq1SrZbDYVFRW5y7Kzs1VUVKTc3FzZbDYlJSUpJSXFJ4ECAAAAAADAf7qdJCooKFBBQUG7z5WWlnY7IAAAAABAxzpaE87mcCr2vPL2yhAe2msHHbUNT8sRubxeuBoAAAAAEDidrRV3fnl7ZefKEdraawcdnVfWF0RXdXtNIgAAAAAAAIQPkkQAAAAAAAAgSQQAAAAg9DU1NSk9PV2HDh2SJFVWVspisSg5OVnFxcUGRwcAoYEkEQAAAICQtm/fPk2bNk01NTWSpDNnzig/P18vvfSSysvLVV1drZ07dxobJACEAJJEAABDvfjii0pLS1NaWpqWLVsmiau/ocLucLZbbmunvKPXAoAvbNiwQUuWLFF8fLwkqaqqSoMHD9agQYMUHR0ti8WiiooKg6MEgODH3c0AAIaprKzUrl27tGnTJkVFRWnWrFkqKyvT8uXLtXbtWg0cOFBz5szRzp07lZSUZHS4OI+nd9cBAH95+umnWz0+cuSI4uLi3I/j4+NVX1/v1T6qq6u9+vuuslqtF3xNQkJCACJBMPluuzj3/47aQXttqKPX2h1OxZhNrcpOnjqjTz7+qLuhBlxX3jORxps6IUkEADBMXFycFi9erJiYGEnSkCFDVFNT4776K8l99ZckEQCgq1paWhQVFeV+7HK5Wj3ujlGjRik2Ntbb0DpltVpJAKFd59pFV9qIJ22ovQs+W57PCJl2yHumra7Uic1m6zDxTZIIAGCYoUOHuv9fU1OjrVu36p577gnaq7+eXpXhS0tr4XSlL5yOxRvUA3UQrAYMGKCGhgb344aGBvdUNABAx0gSAQAMt3//fs2ZM0eLFi2SyWRyLzwqBc/VX65UeS9c6o+2cBb1YEwddHb1F/8xZswYHThwQLW1tbrssstUVlamO+64w+iwACDosXA1AqKjBUtZyBSA1WrVL37xCz388MO6/fbbufoLAPBabGysioqKlJubqylTpuiKK65QSkqK0WEBQNBjJBECorPFTQFErrq6Os2bN0/FxcVKTEyUxNVfAED37dixw/3/xMRElZaWGhgNAIQekkQAAMOsWrVKNptNRUVF7rLs7Gz31V+bzaakpCSu/gIAAAABQJIIAGCYgoICFRQUtPscV38BAACCn93hVIzZdMEyhAaSRAAAAAAAoFvaW1qEZUVCFwtXAwAAAAAAgCQRAAAAAAAASBIBAAAD2R3OLpUBAADA/1iTCAAAGIZ1DAAAAIIHI4kAAAAAIEgxuhJAIDGSCAAAAACCFCMuAQQSI4lgqPaujNg6uFrCVRQAAAAAAPyHkUQwVEdXRs4vO1cOAAAAAAD8g5FEAAAAAAAAIEkEAAAAAAAAkkQdrnPD+jcAAAAAACCSRPyaRO2tiSOx/g0AAAAA37M7nIoxm9otBwCjRXySCAAAAAAChYvUAIJZxE83AwAAAAAAAEmiDjHcM/iwfhQAAAAAAP7j9XSzpqYmZWdn6/e//70uu+wyVVZWqrCwUDabTampqcrLy/NFnAHX3jBQhoAai6G5ABAZOluvo71yAAAA+IZXSaJ9+/apoKBANTU1kqQzZ84oPz9fa9eu1cCBAzVnzhzt3LlTSUlJvogVAABEAC4KAAAAGMOr6WYbNmzQkiVLFB8fL0mqqqrS4MGDNWjQIEVHR8tisaiiosIngQIAAAAAgI6xFAe85dVIoqeffrrV4yNHjiguLs79OD4+XvX19d7sQtXV1V79/YUkJCT4dfsIDKvVanQI7QrWuPwpEo9ZitzjBrqKqWIAAPgfo3HhLa/XJPqulpYWRUVFuR+7XK5Wj7tj1KhRio2N9TY0hLlgTPZZrdagjMufIvGYpa4ft81m83viGwhWfGkFAAAIfj5NEg0YMEANDQ3uxw0NDe6paAAAAAAQSDNmzNCxY8cUHX32Z8/SpUs1ZswYg6MCgODl1ZpE5xszZowOHDig2tpaOZ1OlZWVaeLEib7cBQAgDDU1NSk9PV2HDh2SJFVWVspisSg5OVnFxcUGR3dhzP8HgODjcrlUU1OjkpIS9z8SRADQOZ+OJIqNjVVRUZFyc3Nls9mUlJSklJQUX+4CABBmwuFOmUylAoDg8/nnn0uSZs6cqW+//VZ33XWX7rnnHoOjAoDg5pMk0Y4dO9z/T0xMVGlpqS82CwCIAOfulLlo0SJJre+UKcl9p8xgThIBAIJPY2OjEhMT9fjjj8vhcCgnJ0c/+tGPdMMNNxgdGgAELZ+OJAIAwFOBuFMmACDyjB07VmPHjnU/zsrK0s6dO7udJPLVzScudLOL8++YGok3BUFb320X5/7vr7bR0R1JbfZmxcZ0PYUQqLv/cpfhtrypE5JEAICg4o87Zfrqi31HHS5f4AMnWL4IBkscRqMeqINgtnfvXjkcDiUmJko625+cW8C6OwJ112X6FLTnXLsIxN2EO5tG78n0+kC05Ui9u3JnulInnd11mSQRACCo+ONOmb74Ys+XkOAQDOeAtnAW9WBMHXT2xR6tnThxQitWrND69evlcDi0adMmPfHEE0aHBQBBjSQRACCofPdOmZdddpnKysp0xx13GB0WACDETJo0Sfv27dPUqVPV0tKi6dOnt5p+BgBoiyQRACCocKdMAICvLFiwQAsWLDA6DAAIGSSJAABBgTtlAgAAAMbqYXQAAAAAABDp7A5nxK/zhfbZHU73/0OljXw35q6UI3gwkggAAAAADNbZHaUQ2UKxbYRizDiLkUQAAAAAAAAgSYTQ196QRYYxAgAAAADgGaabIeS1N5SRYYwAEH7sDqdizKY25TaHU7HnlXf0WgAAAHSMJBEAAAgJna1vwMUCAAAA7zHdDAAAAAAAACSJAADoKtY7AwB4gn4D6BrWmQ0eTDcDAKCLuJ0rAMATrJ0JdA3vleDBSCIAAAAAAACQJAIAAAAAAECYJomYzwgAAAAAAOCZsFyTiPmMAAAAAAAAngnLkUQAAAAAACB0dTQbiFlC/hWWI4kAAABCjd3hVIzZ1OVyAADCGXeVNUbEJIn4ghVZ+KINAAg1fBkGAABGi5gkEV+8IgvnGwAAAAAAz7AmEQAACDvhtI6BL44lnOoDAAD4T8SMJAIAAJEjnEaU+uJYwqk+AACA/zCSCBGFK6kAAAAAALSPkUSIKFxJBQC0p70bG9gcTsW2c7OD4SNGBiosAABwnu/22QkJCW3KPN3GOR31+/68+VF72/bkJkz+iI0kEQAAiHjtXUTY8nwGFxYAAAgyHfXZvthGoPt9T47FF8fdFUw3A9T+dDObB1PT2itLSEjwaHobU96CE+cKAAAAQKTwy0iiLVu2aOXKlWpubta9996ru+++2x+7AXzG20yyp9PYApUFhvc4V8Yxui85f/juueHMCG2+GJbtyTDwQMfhz+M7fxh+d4f4exuHP4f9+0ugpgigLaP7EgAINT5PEtXX16u4uFgbN25UTEyMsrOzdd111+nKK6/09a4AAGEqGPoSEoThKVjuFObPofLe6uz4AjkMP5zWEeTzxBjB0JcAQKjxeZKosrJS48eP10UXXSRJmjx5sioqKvTggw96tB2XyyVJstvt3Yrjoj7nLUJls7Up87TcX68Np22EYszd2UZ7PHlte6/v7LWhJFyO47u6cq66ctznPs/Ofb6hY/QlxuwvWLYRijGfK/eEv2Lu6v46e70/j9sTgd5fd3Q1Hl/1+/QlXUdfYsz+gmUboRizL7YRCjF3xNvj9pQvYvYFb/ryjl57oZg760uiXD7uYf7whz/o1KlTysvLkyS98cYbqqqq0pNPPunRdk6cOKF///vfvgwNAILCsGHD1K9fP6PDCGr0JQDQOfqSC6MvAYDOtdeX+HwkUUtLi6KiotyPXS5Xq8dd1adPHw0bNkxms7lbfw8AwcblcsnhcKhPnz5GhxL06EsAoH30JV1HXwIA7eusL/F5kmjAgAHau3ev+3FDQ4Pi4+M93k6PHj24OgIg7PTs2dPoEEICfQkAdIy+pGvoSwCgYx31JT18vaPrr79eu3fv1rFjx3T69Glt27ZNEydO9PVuAABhjL4EAOAt+hIA8JzPRxJdcsklysvLU05OjhwOh7KysjR69Ghf7wYAEMboSwAA3qIvAQDP+XzhagAAAAAAAIQen083AwAAAAAAQOghSQQAAAAAAACSRAAAAAAAACBJBAAAAAAAAJEkAgAAAAAAgEgSAQAAAAAAQCSJAAAAAAAAIJJEAAAAAAAAEEkiv3vxxReVlpamtLQ0LVu2TJJUWVkpi8Wi5ORkFRcXGxyh/zz77LNavHixJOnjjz9WZmamJk+erMcee0zNzc0GR+dbO3bsUGZmplJTU/XUU09JiozzXFJS4m7fzz77rKTwPddNTU1KT0/XoUOHJHV8fsP1+CPdli1bNGXKFCUnJ+u1114zOhy/8qTfCvf23pV+7Msvv9Tdd9+tlJQUPfDAAzp58qSRIfuUJ31bOLcFT/q6cG4P8L1I/p3QkRdeeEFTpkxRWlqa1qxZI4k6kSLrd9WFzJgxQ2lpacrIyFBGRob27dsXUd/TzueX36Eu+M3777/v+vnPf+6y2Wwuu93uysnJcW3ZssWVlJTk+uKLL1wOh8M1c+ZM17vvvmt0qD5XWVnpuu6661yPPPKIy+VyudLS0lx///vfXS6Xy/Xoo4+6XnvtNSPD86kvvvjCNWHCBFddXZ3Lbre7pk2b5nr33XfD/jyfOnXK9dOf/tT19ddfuxwOhysrK8v1/vvvh+W5/sc//uFKT093jRw50nXw4EHX6dOnOzy/4Xj8ke6rr75yTZo0yfXNN9+4Tp486bJYLK79+/cbHZZfeNpvhXN772o/Nnv2bFdZWZnL5XK5XnzxRdeyZcuMCdjHPO3bwrUteNrXhWt7gO9F8u+Ejvz1r391ZWdnuxwOh+v06dOuSZMmuT7++OOIrhOXK7J+V11IS0uLa8KECS6Hw+Eui6Tvaefz1+9QRhL5UVxcnBYvXqyYmBiZzWYNGTJENTU1Gjx4sAYNGqTo6GhZLBZVVFQYHapPffvttyouLtbcuXMlSYcPH9aZM2d0zTXXSJIyMzPD6pjffvttTZkyRQMGDJDZbFZxcbF69eoV9ufZ6XSqpaVFp0+fVnNzs5qbmxUdHR2W53rDhg1asmSJ4uPjJUlVVVXtnt9wb+uRqrKyUuPHj9dFF12k3r17a/LkyWF7Xj3pt8K5vXe1H3M4HPrwww81efLkVuXhwJO+LZzbgid9XTi3B/hepP5O6My4ceP06quvKjo6Wl9//bWcTqcaGxsjuk4i7XfVhXz++eeSpJkzZ+q2227TunXrIup72vn89Ts02k/xQtLQoUPd/6+pqdHWrVt1zz33KC4uzl0eHx+v+vp6I8Lzm1/96lfKy8tTXV2dJOnIkSOtjjkuLi6sjrm2tlZms1lz585VXV2dbrzxRg0dOjTsz3Pfvn310EMPKTU1Vb169dJPf/pTmc3msDzXTz/9dKvH57fpc+c33Nt6pGrvfFdVVRkYkf940m+Fc3vvaj/2zTffqG/fvoqOjm5VHg486dvCuS140teFc3uA70Xq74QLMZvNWrFihVavXq2UlJQOv3NFikj7XXUhjY2NSkxM1OOPPy6Hw6GcnBylpqZGzPe08/nrdygjiQJg//79mjlzphYtWqRBgwYpKirK/ZzL5Wr1ONS98cYbGjhwoBITE91lLS0tYX3MTqdTu3fv1jPPPKPXX39dVVVVOnjwYFgfsyR98skn+tOf/qR33nlH7733nnr06KH3338/7I9b6rhNh3tbj1SReF670m+Fa7140o+1d8zhUAeSZ31buLYFybO+LpzbA/wnkn4ndNX8+fO1e/du1dXVqaamJmLrJBJ/V13I2LFjtWzZMvXr10/9+/dXVlaWVqxYEbF14q/foYwk8jOr1ar58+crPz9faWlp2rNnjxoaGtzPNzQ0uKewhIPy8nI1NDQoIyNDx48f16lTpxQVFdXqmI8ePRpWx/yDH/xAiYmJ6t+/vyTplltuUUVFhUwmk/s14XaeJWnXrl1KTEzUxRdfLOnscNdVq1aF9bk+Z8CAAe2+j88vD9fjjzQDBgzQ3r173Y/D8f38XV3tt8K1vXvSj/Xv318nTpyQ0+mUyWQKq7bhSd8Wrm1B8qyvC+f2AP+ItN8JF/LZZ5/JbrdrxIgR6tWrl5KTkyPiO3VHIvF31YXs3btXDofDnThzuVy69NJLI/Z946/foYwk8qO6ujrNmzdPy5cvV1pamiRpzJgxOnDggGpra+V0OlVWVqaJEycaHKnvrFmzRmVlZSopKdH8+fN10003qbCwULGxsbJarZLO3iUknI550qRJ2rVrlxobG+V0OvXee+8pJSUlrM+zJA0fPlyVlZU6deqUXC6XduzYoXHjxoX1uT6no/fxpZdeGhHHH2muv/567d69W8eOHdPp06e1bdu2sD2vnvRb4drePenHzGazrr32WpWXl0uSNm/eHBZ1IHnWt4VrW5A86+vCuT3A9yLxd8KFHDp0SAUFBbLb7bLb7dq+fbuys7Mjtk4i8XfVhZw4cULLli2TzWZTU1OTNm3apOeeey5ivqedz1+/QxlJ5EerVq2SzWZTUVGRuyw7O1tFRUXKzc2VzWZTUlKSUlJSDIwyMJYvX66CggI1NTVp5MiRysnJMToknxkzZoxmzZql6dOny+Fw6IYbbtC0adN0xRVXhPV5njBhgv71r38pMzNTZrNZV199tWbPnq1bb701bM/1ObGxsR2+j8O5rUeqSy65RHl5ecrJyZHD4VBWVpZGjx5tdFh+4Wm/FUntvaNjXbJkiRYvXqyVK1dq4MCB+s1vfmNwpL7had8Wrm3B074uXNsDfI/fCW0lJSWpqqpKU6dOlclkUnJystLS0tS/f/+IrZP2hOvnbVdMmjRJ+/bt09SpU9XS0qLp06crISEhYr6nnc9fv0OjXC6Xy08xAwAAAAAAIEQw3QwAAAAAAAAkiQAAAAAAAECSCAAAAAAAACJJBAAAAAAAAJEkAgAAAAAAgEgSAQAAAAAAQCSJAAAAAAAAIJJEAAAAAAAAEEkiAAAAAAAAiCQRAAAAAAAARJIIAAAAAAAAIkkEAAAAAAAAkSQCAAAAAACASBIBAAAAAABAJIkAAAAAAAAgkkQAAAAAAAAQSSIAAAAAAACIJBEAAAAAAABEkggAAAAAAAAiSQQAAAAAAACRJAIAAAAAAIBIEgEAAAAAAEAkiQAAAAAAACCSRAAAAAAAABBJIgAAAAAAAIgkEQAAAAAAAESSCAAAAAAAACJJBAAAAAAAAJEkAgAAAAAAgEgSAQAAAAAAQCSJAAD/397dx0VZ5/sff48w4G3HLCZ8mMfKNDpS2lIaxUJ6UkHiUMQW2kpmpbalxTHNo2StZd7E5upx7dZsI8uUUJBjaBvGUXErOa1GWbatkJnhuN0YKsMwXL8/fDg/URCYGZi71/Px2Mdyfa9rruvzmWuaa/xc3+/3AgAAAABRJAIAAAAAAICkUG8HgI717bffauTIkRo4cKAkqaGhQd26dVNmZqbGjBmjpUuXql+/frr11lub3Ud+fr42b96sF198sU3HXr58uaKionTzzTe7FPv48eN18OBB9ejRQyaTSXa7XdHR0XryySfVpUuXVu+nuLhYq1evVm5ubqvy9ZRZs2Zpx44d6tWrV6P2l156SRdddJFHjzVx4kTl5OSoV69euv/++/XYY4/p8ssv9+gxACDY2O12DR8+XFFRUXrllVe8HQ4AAIDHUSQKQp07d1ZBQYFz+eDBg5owYYJCQkL08MMPt9txP/zwQ7cLFTNnzlRiYqIkyTAMPfzww1q2bJkee+wxl/bXnvk2ZcKECbr33nvb/Tg7duxw/v3yyy+3+/EAIBi89957ioqKUkVFhb7++mv179/f2yEBAAB4FEUiqE+fPpo2bZpWrlyprVu3asCAAbr33nuVl5ent99+W3a7XT///LPuv/9+jRs3TpJktVp177336vDhw+rTp4+eeuopRURE6HseyxIAACAASURBVJdfftH8+fO1b98+2e12xcbGaubMmXr77bdVUVGhxYsXKyQkRAkJCcrJydHHH38sh8Ohf/u3f1N2dra6d++uN998U2vWrJHZbFZ4eLjmzZvXZHHJZDJp2LBh+t///V9J0tdff6358+frp59+ksPh0Pjx45Weni5JWrp0qTZu3KiePXuqX79+zn3MmjXLmW9paalycnLUqVMnXXnllSorK9Obb76pjz76SHl5eTpx4oS6d++u3NxcrVu3Tm+99ZYaGhrUs2dPPf744+rfv7/q6uqazetcTo/jzOURI0botttu086dO3Xo0CGlpqbqkUcekSTl5eVp1apV6tSpk84//3wtWrRIy5YtkyTdfffdeumll3TXXXdp6dKluuqqq/T2228rNzdXnTp10oUXXqjHH39cl156qWbNmqXu3bvryy+/1Pfff68rrrhCixYtUrdu3dz8dAFA4Hjrrbc0ZswY/eu//qv+/Oc/a968eZJO9gjNy8tTt27ddO211+r9999XSUmJy9cEAAAAb2FOIkiSoqKitG/fPufysWPHtG7dOr300kvasGGDlixZomeffda5fv/+/Zo7d642btyogQMHav78+ZKkZ555RoMGDVJ+fr42bNigH3/8UatWrdJdd92l6OhozZw5UyNHjtRLL72kkJAQ5efnq7CwUBaLRTk5OXI4HHrmmWf0yiuv6J133tEdd9yh8vLyJmP++eef9e6772rYsGGqr6/XtGnTNH36dOXn5+uNN97Qq6++qr/97W/6y1/+oi1btmjDhg1as2aNampqztrXjz/+qJkzZ+rZZ59VQUGBhg0bpurqauf6v//978rNzVVubq4++ugjbdiwQatXr9aGDRt033336aGHHpKkZvM65bXXXlNqaqrzf+vWrWvV+Tl+/LizePbqq6/qwIED+uKLL5STk6NXXnlFGzdu1IgRI/T8889rwYIFkqQ///nP6t27t3MfO3fu1CuvvKLXX39dhYWFuuWWW/Tggw/KMAxJUkVFhVauXKlNmzbp4MGDKi4ublVsABAM/v73v+uTTz5RYmKibr31VhUUFOjHH3/Utm3blJ+fr7y8POXn5+vYsWPO17R0TQAAAPA19CSCpJO9cjp37uxc7tatm1544QWVlpaqsrJSX3zxhY4fP+5cf8MNNzh75KSnpzt77HzwwQf69NNPlZeXJ0mqra1t8ngffPCBfvnlF5WVlUk6Oc/DBRdcoJCQECUmJiojI0M33XST4uLilJCQ4Hzd4sWL9fzzzzsLG8OHD1dmZqYqKyv1zTffaPbs2c5ta2tr9fnnn+vrr7/WyJEjnXdub7/9duXm5jaKZ9euXerfv7+ioqIkSbfddpuefvpp5/orrrjC+foPPvhAVVVVysjIcK4/evSofvrpp2bzOsXV4Wb//u//Lkm66KKLdMEFF+jnn3/Wxx9/rLi4OGchaMKECefcx7Zt2zRmzBjnnEhpaWmaP3++vv32W0nSr3/9a4WFhUmSBg4cqJ9//rnNcQJAoHrrrbc0fPhwnX/++Tr//PN18cUXa+3atbJarUpMTNR5550nSbrrrrv017/+VVLz1zoAAABfRZEIkqRPP/3UOZm1JH3//fe68847dccddygmJkaJiYnaunWrc31ISIjz74aGBoWGhjr/Xrp0qXOehqNHj8pkMp11vIaGBs2ePdtZADp27JhsNpskKScnR/v27VNZWZleeuklFRQUaOnSpZIaz0l0OofDoR49ejSaa+nIkSPq0aOHFi9e7CwqnRn76W2nbyNJnTr9/452Xbt2bRR7amqqZsyY4Vw+fPiw/uVf/uWceZ2LyWRqdHy73d5ofXh4+FnbhoSENHpva2trdfDgwWbnyGhoaDirzTAM1dfXS1KjIuGZ8QBAMDt+/LgKCgoUFhamESNGSJJqamr0xhtvKDk5udlrjKvXBAAAAG9huBm0f/9+rVixQhMnTnS2VVRUqFevXvrd736nuLg4Z4HI4XBIOjkJ9XfffSdJWrNmjeLj4yVJcXFxeu2112QYhurq6vTAAw/ojTfekHTyh/OpgkRcXJxWr16turo6NTQ06PHHH9dzzz2nH374QQkJCerZs6cmTJigRx55RJ9++mmLOVx66aWNJuQ+dOiQbrnlFlVUVCg+Pl7FxcU6evSoGhoaGhWSTvnVr37l7DElSZs3b262wBUXF6f/+Z//0eHDhyWdvLt89913nzOvlpx//vmqqKiQJFVXV+ujjz5q8TXDhg3Tzp07nXGsWbPGOSTw9Pf6lF//+tfatGmTfvjhB0nSO++8c9YcTQCAs52a027btm0qKSlRSUmJ/vKXv+j48eMaNGiQtmzZol9++UWSnD1pJdevCQAAAN5CT6IgVFtbq9TUVEkne8uEh4frP//zP3XTTTc556G58cYblZeXp8TERJlMJg0dOlS9evVSVVWVpJPDkWbPnq0jR47osssuc07eOWfOHM2fP18pKSmy2+264YYbdN9990mSRowYoeeee052u12/+93vtGjRIt12221yOBy68sornZMnP/DAA5owYYI6d+6skJCQRsO+mhMWFqYVK1Zo/vz5euWVV1RfX6+HH35YMTExkqQvv/xSt99+u8477zxFRUXpxx9/bPT6nj176rnnntNjjz2mTp06KTo6WqGhoerSpctZx4qLi9P999+viRMnymQyqXv37lq+fLlMJlOzebVk/PjxevTRRzV69GhdfPHFuv7661t8zRVXXKEZM2Y439+IiAg988wzkqTExESNHz9e//3f/+3c/sYbb9SECRN09913q6GhQb169dKLL77YqMcUAOBsb731lu65555GvYTOO+88jR8/Xq+99pruuOMO3XnnnercubMGDBjgvHa4ek0AAADwFpPBmBJANTU1WrFihaZOnaouXbros88+0+TJk7Vt27YmexMBACCdHK79ySefKDMzU5K0atUq7d69W3/84x+9HBkAAEDb0ZMIkNS9e3eZzWalp6crNDRUoaGh+uMf/0iBCABwTpdeeqlefvllrV27ViaTSb1799ZTTz3l7bAAAABcQk8iAAAAAAAAMHE1AAAAAAAAfHi4WUNDg44dOyaz2cyQHwABwTAM2e12devWjQnDOwjXEgCBhmsJAKA9+WyR6NixY9q3b5+3wwAAjxs4cKB69Ojh7TCCAtcSAIGKawkAoD34bJHIbDZLOnkBDAsLkyRVVFQoOjram2F1KPINbMGWrxR8OZ+Zb11dnfbt2+f8fkP7a+pa0hR//mwSu3cQu3cQO9cSAED7cqtIVFBQoJdeekmSFB8fr8cee0x79+7VnDlzdOzYMV177bX6/e9/r9DQth/m1LCAsLAwhYeHO9tP/zsYkG9gC7Z8peDLual8GfbUcZq7ljTFnz+bxO4dxO4dxH4S1xIAQHtweSDziRMnNH/+fOXm5qqgoEC7du1SWVmZZsyYoblz52rz5s0yDENr1671ZLwAAAAAAABoBy4XiRwOhxoaGnTixAnV19ervr5eoaGhqq2t1ZAhQyRJaWlpKi4u9liwAAAAAAAAaB8uDzfr3r27Hn74YSUlJalLly667rrrZDabFRER4dwmIiJC1dXVHgkUAAAAAAAA7cflItEXX3yhd955R1u3blWPHj306KOPaseOHY3GRxuG4fZ46YqKikbL5eXlbu3P35BvYAu2fKXgyznY8gUAAADgv1wuEm3fvl2xsbG64IILJJ0cWrZy5UpZrVbnNkeOHJHFYnErwOjoaOckf+Xl5YqJiXFrf/6EfANbsOUrBV/OZ+Zrs9nOKnwDAAAAgK9weU6iqKgolZWV6fjx4zIMQyUlJRo6dKjCw8Odd84LCgoUHx/vsWABAAAAAADQPlzuSRQXF6fPP/9caWlpMpvNuuqqqzRp0iSNHDlS2dnZqqmp0aBBg5SZmenJeAEAAAAAANAOXC4SSdKkSZM0adKkRm1RUVHKy8tzKygAQOCpqalRRkaGXnjhBV188cUqKyvTggULZLPZlJSUpKysLEnS3r17NWfOHB07dkzXXnutfv/73ys01K3LFQAAAIBWcHm4WaCrszta1QYAaNnu3bs1duxYVVZWSpJqa2s1e/ZsrVixQps2bVJFRYVKS0slSTNmzNDcuXO1efNmGYahtWvXeiVmrgMAAAAINtyabUaYOUQp0wsatW38Q6qXogEA/7Z27Vo98cQTmjlzpiRpz5496tevn/r27StJSklJUXFxsS6//HLV1tZqyJAhkk4+FGHZsmUaN25ch8fMdQAAAADBhiIRAKDdzZ8/v9Hy4cOHFRER4Vy2WCyqrq4+qz0iIkLV1dVuH781T5U79dCFU5p7Et+Z2/kCX4yptYjdO4jdO/w5dgBAcKBIBADocA0NDTKZTM5lwzBkMpmabXdXdHS0wsPDm11fXl7ebFHoTK3drqO0JXZfQ+zeQeze4anYbTZbqwrfAAC4gjmJAAAdLjIyUlar1blstVplsVjOaj9y5IgsFos3QgQAAACCDkUiAECHGzx4sPbv36+qqio5HA4VFRUpPj5effr0UXh4uHNIRkFBgeLj470cLQAAABAcGG4GAOhw4eHhWrhwoaZOnSqbzaaEhAQlJiZKknJycpSdna2amhoNGjRImZmZXo4WAAAACA4UiQAAHaakpMT5d2xsrAoLC8/aJioqSnl5eR0ZFgAAAAAx3AwAAAAAAACiSAQAAAAAAABRJAIAAAAAAIAoEgEAAAAAAEAUiQAAAAAAACCKRAAAAAAAABBFIgAAAAAAAIgiEQAAAAAAAESRCAAAAAAAAKJIBAAAAAAAAFEkAgAAAAAAgKRQV1+4bt06vfHGG87lb7/9Vqmpqbr55pu1YMEC2Ww2JSUlKSsryyOBAgAAAAAAoP24XCT6zW9+o9/85jeSpK+++koPPvig7r//fo0dO1a5ubnq3bu3Jk+erNLSUiUkJHgsYAAAAAAAAHieR4abPfnkk8rKytKBAwfUr18/9e3bV6GhoUpJSVFxcbEnDgEAAAAAAIB25HJPolPKyspUW1urpKQkFRUVKSIiwrnOYrGourrarf1XVFQ0Wi4vL3drf60VExPTZHtHHd9bx/M28g18wZZzsOULAAAAwH+5XSRas2aN7rnnHklSQ0ODTCaTc51hGI2WXREdHa3w8HBJJ/+x1VzxpqN05PF9Id+ORL6BL9hyPjNfm812VuEbAAAAAHyFW8PN6urq9PHHH2vEiBGSpMjISFmtVud6q9Uqi8XiXoQAAAAAAABod24Vib788ktdcskl6tq1qyRp8ODB2r9/v6qqquRwOFRUVKT4+HiPBAoAAAAAAID249ZwswMHDigyMtK5HB4eroULF2rq1Kmy2WxKSEhQYmKi20ECAAAAAACgfblVJBozZozGjBnTqC02NlaFhYVuBQUAAAAAAICO5dZwMwAAAAAAAAQGikQAAAAAAACgSAQAAAAAAACKRAAAAAAAABBFIgAAAAAAAIgiEQAAAAAAAESRCAAAAAAAAKJIBAAAAAAAAFEkAgAAAAAAgCgSAQAAAAAAQBSJAAAAAAAAIIpEAAAAAAAAEEUiAAAAAAAAiCIRAAAAAAAARJEIAOBlBQUFSk5OVnJyshYtWiRJ2rt3r9LS0jR69GjNmTNH9fX1Xo4SAAAACHwUiQAAXnPixAnNnz9fubm5Kigo0K5du1RWVqYZM2Zo7ty52rx5swzD0Nq1a70dKgAAABDwKBIBALzG4XCooaFBJ06cUH19verr6xUaGqra2loNGTJEkpSWlqbi4mIvRwoAAAAEvlBvBwAACF7du3fXww8/rKSkJHXp0kXXXXedzGazIiIinNtERESourrareNUVFS0uE15eXmj5ZiYmFZt5wt8MabWInbvIHbv8OfYAQDBgSIRAMBrvvjiC73zzjvaunWrevTooUcffVQ7duyQyWRybmMYRqNlV0RHRys8PLzZ9eXl5c0Whc7U2u06Slti9zXE7h3E7h2eit1ms7Wq8A0AgCvcGm5WUlKitLQ0JSUl6emnn5YklZWVKSUlRaNGjdKSJUs8EiQAIDBt375dsbGxuuCCCxQWFqa0tDR9+OGHslqtzm2OHDkii8XixSgBAACA4OBykejAgQN64okntGLFChUWFurzzz9XaWmpZs+erRUrVmjTpk2qqKhQaWmpJ+MFAASQqKgolZWV6fjx4zIMQyUlJRo6dKjCw8OdwzIKCgoUHx/v5UgBAACAwOfycLP33ntPY8aMUWRkpCRpyZIlqqqqUr9+/dS3b19JUkpKioqLi5WQkOCZaAEAASUuLk6ff/650tLSZDabddVVV2nSpEkaOXKksrOzVVNTo0GDBikzM9PboQIAAAABz+UiUVVVlcxms6ZMmaJDhw7ppptu0oABAxpNNmqxWDw+2WhHTfjnKxOWBtsEh+Qb+IIt52DL1xWTJk3SpEmTGrVFRUUpLy/PSxEBAAAAwcnlIpHD4dCuXbuUm5urrl276oEHHlDnzp3bdbJRX5issCOP7wv5diTyDXzBlvOZ+TLZKAAAAABf5nKR6MILL1RsbKx69eolSbr55ptVXFyskJAQ5zZWq5XJRgEAAAAAAPyAyxNXDx8+XNu3b9fRo0flcDi0bds2JSYmav/+/aqqqpLD4VBRURGTjQIAAAAAAPgBl3sSDR48WPfdd5/GjRsnu92uG2+8UWPHjtVll12mqVOnymazKSEhQYmJiZ6MFwAAAAAAAO3A5SKRJKWnpys9Pb1RW2xsrAoLC90KCgAAAAAAAB3L5eFmAAAAAAAACBwUiQAAAAAAAECRCAAAAAAAABSJAAAAAAAAIIpEAAAAAAAAEEUiAAAAAAAAiCIRAAAAAAAARJEIAAAAAAAAokgEAAAAAAAAUSQCAAAAAACAKBIBAAAAAABAFIkAAAAAAAAgikQAAAAAAAAQRSIAAAAAAACIIhEAAAAAAABEkQgAAAAAAACiSAQAAAAAAABRJAIAAAAAAICkUHdePH78eP3www8KDT25m3nz5umbb77R888/r/r6et1999266667PBIoAAAAAAAA2o/LRSLDMFRZWamtW7c6i0TV1dXKyspSfn6+wsLClJGRoWHDhunyyy/3WMAAAAAAAADwPJeLRP/4xz8kSRMnTtRPP/2kO+64Q926ddP111+vnj17SpJGjx6t4uJiPfTQQ56JFgAAAAAAAO3C5TmJjh49qtjYWP3pT3/Sa6+9pjVr1ui7775TRESEcxuLxaLq6mqPBAoAAAAAAID243JPomuuuUbXXHONczk9PV0LFizQAw884GwzDEMmk8mtACsqKhotl5eXu7W/1oqJiWmyvaOO763jeRv5Br5gyznY8gUAAADgv1wuEu3atUt2u12xsbGSThaE+vTpI6vV6tzGarXKYrG4FWB0dLTCw8MlnfzHVnPFm47Skcf3hXw7EvkGvmDL+cx8bTbbWYVvAAAAAPAVLg83++WXX7R48WLZbDbV1NRo/fr1evbZZ7Vz50798MMPOnHihLZs2aL4+HhPxgsAAAAAAIB24HJPouHDh2v37t269dZb1dDQoHHjxikmJkZZWVnKzMyU3W5Xenq6rr76ak/GCwAAAAAAgHbgcpFIkh555BE98sgjjdpSUlKUkpLiVlAAAAAAAADoWC4PN/M3dXZHm9rbax8AAAAAAAC+yK2eRP4kzByilOkFZ7Vv/ENqh+4DANBYSUmJli9frhMnTujGG29Udna2ysrKtGDBAtlsNiUlJSkrK8vbYQIAAAABL2h6EgEAfM+BAwf0xBNPaMWKFSosLNTnn3+u0tJSzZ49WytWrNCmTZtUUVGh0tJSb4cKAAAABDyKRAAAr3nvvfc0ZswYRUZGymw2a8mSJerSpYv69eunvn37KjQ0VCkpKSouLvZ2qAAAAEDAC5rhZgAA31NVVSWz2awpU6bo0KFDuummmzRgwABFREQ4t7FYLKqurvZilAAAAEBwoEgEAPAah8OhXbt2KTc3V127dtUDDzygzp07y2QyObcxDKPRsisqKipa3Ka8vLzRckxMTKu28wW+GFNrEbt3ELt3+HPsAIDgQJEIAOA1F154oWJjY9WrVy9J0s0336zi4mKFhIQ4t7FarbJYLG4dJzo6WuHh4c2uLy8vb7YodKbWbtdR2hK7ryF27yB27/BU7DabrVWFbwAAXMGcRAAArxk+fLi2b9+uo0ePyuFwaNu2bUpMTNT+/ftVVVUlh8OhoqIixcfHeztUAAAAIOAFfU+iOrtDYeaQljcEAHjc4MGDdd9992ncuHGy2+268cYbNXbsWF122WWaOnWqbDabEhISlJiY6O1QAQAAgIAX9EWiMHOIUqYXnNW+8Q+pXogGAIJPenq60tPTG7XFxsaqsLDQSxEBAAAAwYnhZgAAAAAAAKBIBAAAAAAAAIpEAAAAAAAAEEUiAAAAAAAAiCIRAAAAAAAARJEIAAAAAAAAokgEAAAAAAAAUSQCAAAAAACAKBIBAAAAAABAHigSLVq0SLNmzZIk7d27V2lpaRo9erTmzJmj+vp6twMEAAAAAABA+3OrSLRz506tX7/euTxjxgzNnTtXmzdvlmEYWrt2rdsBAgAAAAAAoP25XCT66aeftGTJEk2ZMkWSdPDgQdXW1mrIkCGSpLS0NBUXF3smSgAAAAAAALSrUFdfOHfuXGVlZenQoUOSpMOHDysiIsK5PiIiQtXV1W4HWFFR0Wi5vLzcpf3ExMS4Hcu5uBqXt/brq8g38AVbzsGWLwAAAAD/5VKRaN26derdu7diY2OVn58vSWpoaJDJZHJuYxhGo2VXRUdHKzw8XNLJf2y1d7HHVe0Rly/n2x7IN/AFW85n5muz2c4qfAMAAACAr3CpSLRp0yZZrValpqbq559/1vHjx2UymWS1Wp3bHDlyRBaLxWOBAgAAAAAAoP24VCRatWqV8+/8/Hx99NFHWrBggW655RbnnfOCggLFx8d7LFBfVmd3KMwc0mIbAAAAAACAr3J5TqKm5OTkKDs7WzU1NRo0aJAyMzM9uXufFWYOUcr0gkZtG/+Q6qVoAAD+iBsOAAAA8Da3i0RpaWlKS0uTJEVFRSkvL8/toAAACDbccAAAAIC3dfJ2AAAAAAAAAPA+ikQAAAAAAACgSAQAAAAAAACKRAAAAAAAABBFIgAAAAAAAIgiUbupszva1A4AAAAAAOBNod4OIFA19ShjiccZAwAAAAAA30RPIgAAAAAAAFAkAgAAAAAAAEUiAAAAAAAAiCIRAAAAAAAARJEIAAAAAAAAokgEAAAAAAAAUSQCAAAAAACAKBIBAAAAAABAFIkAAAAAAAAgikQdrs7uaFM7AASDRYsWadasWZKkvXv3Ki0tTaNHj9acOXNUX1/v5egAAACA4BDq7QCCTZg5RCnTC85q3/iHVC9EAwDet3PnTq1fv1433XSTJGnGjBl6+umnNWTIEM2ePVtr167VuHHjvBskAAAAEAToSQQA8JqffvpJS5Ys0ZQpUyRJBw8eVG1trYYMGSJJSktLU3FxsTdDBAAAAIKGWz2Jli5dqs2bN8tkMik9PV333HOPysrKtGDBAtlsNiUlJSkrK8tTsQIAAszcuXOVlZWlQ4cOSZIOHz6siIgI5/qIiAhVV1e7fZyKiooWtykvL2+0HBMT06rtPMWd47VXTB2B2L2D2L3Dn2MHAAQHl4tEH330kf7617+qsLBQ9fX1GjNmjGJjYzV79mzl5uaqd+/emjx5skpLS5WQkODJmAEAAWDdunXq3bu3YmNjlZ+fL0lqaGiQyWRybmMYRqNlV0VHRys8PLzZ9eXl5c0Wac7U2u08paXjtSV2X0Ps3kHs3uGp2G02W6sK3wAAuMLlItHQoUP1+uuvKzQ0VNXV1XI4HDp69Kj69eunvn37SpJSUlJUXFxMkQgAcJZNmzbJarUqNTVVP//8s44fPy6TySSr1erc5siRI7JYLF6MEgAAAAgebg03M5vNWrZsmV599VUlJiaeNUzAYrG4PUzgzDslrnbT9Ye7Tk3lFmzdksk38AVbzsGWb1usWrXK+Xd+fr4++ugjLViwQLfccovzjntBQYHi4+O9GCUAAAAQPNx+utm0adN0//33a8qUKaqsrPT4MIHThwj4cxfj1jgzt0DP90zkG/iCLecz82WIQOvk5OQoOztbNTU1GjRokDIzM70dEgAAABAUXC4Sff3116qrq9OVV16pLl26aNSoUSouLlZISIhzG6vVyjABAECL0tLSlJaWJkmKiopSXl6elyPyjDq7Q2HmkBbbAAAAAF/QydUXfvvtt8rOzlZdXZ3q6ur0/vvvKyMjQ/v371dVVZUcDoeKiooYJuCGqCsHNdleZ3d0cCQAAFeEmUOUMr2g0f8oEAEAAMBXudyTKCEhQXv27NGtt96qkJAQjRo1SsnJyerVq5emTp0qm82mhIQEJSYmejLeoNKta2elTC84q33jH1K9EA0AAAAAAAhkbs1JNHXqVE2dOrVRW2xsrAoLC90KCgAAAAAAAB3L5eFmAAAAAAAACBwUiQAAAAAAAECRCAAAAAAAABSJAAAAAAAAIIpEAAC4rc7u8HYIAAAAgNvceroZAACQwswhSplecFb7xj+keiEaAAAAwDX0JAIAAAAAAABFIgAAAAAAAFAkAgAAAAAAgCgSAQAAAAAAQBSJfEZbnozT3LYdvQ8AAAAAABA4eLqZj2jqyTjNPRXHE0/R4Uk8AAAAAADgdPQkAgAAAAAAAEUiAAAAAAAAUCQCAAAAAACAKBIBAAAAAABAAVok4gldAAAAAAAAbROQTzdry5PCAAAAAAAA4GZPouXLlys5OVnJyclavHixJKmsrEwpKSkaNWqUlixZ4pEgAQAAAAAA0L5cLhKVlZVp+/btWr9+vTZs2KDPPvtMRUVFmj17tlasWKFNmzapoqJCpaWlnowXAAAAAAAA7cDlIlFERIRmzZqlsLAwmc1mwNseDQAAEqJJREFU9e/fX5WVlerXr5/69u2r0NBQpaSkqLi42JPxAgAAAAAAoB24PCfRgAEDnH9XVlbq3Xff1W9/+1tFREQ42y0Wi6qrq90KsKKiotFyeXl5i6+JiYlx65j+7Mz3J+rKQerWtbNb++hI3jy2NwRbvlLw5Rxs+QIAAADwX25PXP3VV19p8uTJmjlzpkJCQlRZWelcZxiGTCaTW/uPjo5WeHi4pJP/2ArmAlBrNPX+nDmJt3Tuiby99R4H2/kNtnyl4Mv5zHxtNttZhW8AAAAA8BVuTVxdXl6uCRMmaPr06brtttsUGRkpq9XqXG+1WmWxWNwOEgAAAAAAAO3L5SLRoUOH9OCDDyonJ0fJycmSpMGDB2v//v2qqqqSw+FQUVGR4uPjPRYsAAAAAAAA2ofLw81Wrlwpm82mhQsXOtsyMjK0cOFCTZ06VTabTQkJCUpMTPRIoAAAAAAAAGg/LheJsrOzlZ2d3eS6wsJClwMCAMBX1dkdCjOHeH0faFlz7zPvPwAAQPPcnrgaAIBgEWYOafPDANpjH2gZ7zMAAEDbuTVxNQAAAAAAAAIDRSIAAAAAAABQJAIAwFfV2R1taveVfQMAAMA/MScRAMCrli9frnfffVeSlJCQoJkzZ6qsrEwLFiyQzWZTUlKSsrKyvByld7TnvDrM2QMAAIAz0ZMIAOA1ZWVl2r59u9avX68NGzbos88+U1FRkWbPnq0VK1Zo06ZNqqioUGlpqbdDBQAAAAIeRaIAwhABAP4mIiJCs2bNUlhYmMxms/r376/Kykr169dPffv2VWhoqFJSUlRcXOztUAEAAICAx3CzANLU0AGGDQDwZQMGDHD+XVlZqXfffVe//e1vFRER4Wy3WCyqrq526zgVFRUtblNeXt5oOSYmxq1jtrfT4z0z9tY4V36u7M9V7XWsjsivI98nTyN27/Dn2AEAwYEiEQDA67766itNnjxZM2fOVEhIiCorK53rDMOQyWRya//R0dEKDw9vdn15ebnPF4XOdCre9oi9o94Lb73vnjimP35mTiF27/BU7DabrVWFbwAAXMFwMwCAV5WXl2vChAmaPn26brvtNkVGRspqtTrXW61WWSwWL0YIAAAABAeKRAAArzl06JAefPBB5eTkKDk5WZI0ePBg7d+/X1VVVXI4HCoqKlJ8fLyXIwUAAAACH8PN0CHq7A6FmUNa3Q4gOKxcuVI2m00LFy50tmVkZGjhwoWaOnWqbDabEhISlJiY6MUoAQAAgOBAkQgdoqlJtSUm1gaCXXZ2trKzs5tcV1hY2MHRAAAAAMGN4WYAAKDd1dkdZ7XFxMQ02Q4AAADvoCcRAABod/QoBQAA8H30JAIAAAAAAABFIgAA/M3pQ7RiYmKcf9uaGLrlj8O5movZH3MBAADwJww3Q4t4MhkA+JZzDd06s90fh3MxNA0AAMA7KBKhRfxYBwC0F244AAAA+A63i0Q1NTXKyMjQCy+8oIsvvlhlZWVasGCBbDabkpKSlJWV5Yk44YOa+mHPj30AQFs0dSOCmxAAAADe4VaRaPfu3crOzlZlZaUkqba2VrNnz1Zubq569+6tyZMnq7S0VAkJCZ6IFT6GH/YAAPx/3DwBAAD+zq0i0dq1a/XEE09o5syZkqQ9e/aoX79+6tu3ryQpJSVFxcXFFIkAAEDA4+YJAADwd24ViebPn99o+fDhw4qIiHAuWywWVVdXu3MIVVRUNFouLy9v8TWnP+kFbXfme9zW97Opc3SufZy+fWvObyAJtnyl4Ms52PJFYGquN4zN7lA4DzYAAAAIGB6duLqhoUEmk8m5bBhGo2VXREdHKzw8XNLJf2xRAGp/7r7HbX39qe2D7fwGW75S8OV8Zr42m+2swjfgD9ryNLVT7QAAAPA/nTy5s8jISFmtVuey1WqVxWLx5CEAAEAb1NkdQXFMAAAAuM+jPYkGDx6s/fv3q6qqShdffLGKiop0++23e/IQaGcMEQCAwHKuXkAdeUx6FwEAAPg+jxaJwsPDtXDhQk2dOlU2m00JCQlKTEz05CHQzvhhDwDwVc3dyOAGBwAAgGd4pEhUUlLi/Ds2NlaFhYWe2C0AAICTN3pFAQAABBOPzkkEAACAjtHU3E/MBwUAANzh0eFmAAAA6BgMEQcAAJ5GTyJ4FHcwAQAAAADwT/QkgkcxXwQAAAAAAP6JnkQAAAAAAACgSAT/x8SdAOA5fH8CAAAEL4abwe8xcScAeA7fqQAAAMGLnkQAAABtUGd3KCYmxtthAAAAeBw9iRCQ6uwOhZlDWt0OAEBr8ZAGAAAQqCgSISDxAx4AAAAAgLZhuBkAAAAAAAAoEgEAADQlkJ70dnoup8+n1JYcm9s2kN4nAACCHcPNAAAAmhBIQ5c9kUsgvR8AAKBp9CQCAAB+zRO9YWwd3BumqTjokQMAALyNnkTwOc09gcxmdyjczSeTeeKpZ01ty1PTAMB7murh0lzvlnP1hmntPjyhLTEDAAB0FIpE8Dlt+QF/qt0T+3ZnH/ywBwA0xRM3ETr6RkRbj9eWmyfcaAEAwLdRJALk/o/WtvZ+aq8hBZ7oKQUA8Bx/nAuorcdztycXN1oAAPAdFIkAuf+jtT17P7UFk4oCAAAAAFzFxNUAAAA+wBO9TH1l8uv2yiUmJsZncgQAIBC1S0+ijRs36vnnn1d9fb3uvvtu3XXXXe1xGASA04dBxcTEeDmaxnxliFZ7DiHz5XkkfGXoHPNneA/XEgQbfxya1p5x+EouAAAEE48Xiaqrq7VkyRLl5+crLCxMGRkZGjZsmC6//HJPHwoBwJfnJvCVH6ftGYcvzyPhy++/r3xGAxnXEqBj+ELR2xdiaG91dsdZN8OCIW8AgP/xeJGorKxM119/vXr27ClJGj16tIqLi/XQQw+1aT+GYUiS6urqGrXbbLZWvb5nt8YXXZvNdlZbW9vba9tA2oc/xuyJfbhyvNP/vyXn2oc72za1vae2bU5bXtPWXNqLO3mfvu2p77NT329oXntfS5rS1Hn1x++WQNiHP8bsiX14I2ajoV7j577bqO2VOSM7NI6mYmgpDn/U1Ptss9W3eT9cSwAA7clkePgK8+KLL+r48ePKysqSJK1bt0579uzRU0891ab9/PLLL9q3b58nQwMAnzBw4ED16NHD22H4NK4lAHBuXEsAAO3B4z2JGhoaZDKZnMuGYTRabq1u3bpp4MCBMpvNLr0eAHyNYRiy2+3q1q2bt0PxeVxLAKBpXEsAAO3J40WiyMhI7dq1y7lstVplsVjavJ9OnTpxdwRAwOncubO3Q/ALXEsAoHlcSwAA7aWTp3d4ww03aOfOnfrhhx904sQJbdmyRfHx8Z4+DAAggHEtAQAAADqex3sSXXTRRcrKylJmZqbsdrvS09N19dVXe/owAIAAxrUEAAAA6Hgen7gaAAAAAAAA/sfjw80AAAAAAADgfygSAQAAAAAAgCIRAAAAAAAAKBIBAAAAAABAFIkAAAAAAAAgikQAAAAAAAAQRSIAAAAAAACIIhEAAAAAAADkg0WijRs3asyYMRo1apRWr1591vq9e/cqLS1No0eP1pw5c1RfX++FKD2rpZyXL1+u4cOHKzU1VampqU1u429qamp0yy236Ntvvz1rXSCe43PlG2jnd/ny5UpOTlZycrIWL1581vpAPL8t5Rxo59jfNXW+ysrKlJKSolGjRmnJkiVejrB5S5cu1ZgxY5ScnKxVq1ZJ8p/YT1m0aJFmzZolyX++D8aPH6/k5GTnf8O7d+9u8drtK0pKSpSWlqakpCQ9/fTTkvzjM7Nu3Trn+52amqqYmBjNmzfPL2KXpIKCAuf3zKJFiyT5z+cdABDkDB/y/fffG8OHDzd+/PFH49ixY0ZKSorx1VdfNdomOTnZ+OSTTwzDMIz/+q//MlavXu2NUD2mNTlPnjzZ+L//+z8vReh5f/vb34xbbrnFGDRokHHgwIGz1gfaOW4p30A6vzt27DDuvPNOw2azGXV1dUZmZqaxZcuWRtsE2vltTc6BdI79XVPna+PGjUZCQoLxzTffGHa73Zg4caLxwQcfeDvUs3z44YdGRkaGYbfbjRMnThjDhw839u7d6xexn1JWVmYMGzbMeOyxxwzD8I/vg4aGBiMuLs6w2+3OttZcu33BN998Y8TFxRmHDh0y6urqjLFjxxoffPCBX31mDMMw9u3bZ4wcOdL47rvv/CL248ePG9ddd53xz3/+07Db7UZ6erqxY8cOv/i8AwDgUz2JysrKdP3116tnz57q2rWrRo8ereLiYuf6gwcPqra2VkOGDJEkpaWlNVrvj1rKWZIqKir04osvKiUlRfPmzZPNZvNStJ6xdu1aPfHEE7JYLGetC8RzfK58pcA6vxEREZo1a5bCwsJkNpvVv39/fffdd871gXh+W8pZCqxz7O+aOl+VlZXq16+f+vbtq9DQUKWkpPjk53Lo0KF6/fXXFRoaqn/+859yOBw6evSoX8QuST/99JOWLFmiKVOmSPKf74N//OMfkqSJEyfqP/7jP/TGG2+06trtC9577z2NGTNGkZGRMpvNWrJkibp06eI3n5lTnnzySWVlZenAgQN+EbvD4VBDQ4NOnDih+vp61dfXKzQ01C8+7wAA+FSR6PDhw4qIiHAuWywWVVdXN7s+IiKi0Xp/1FLOx44d05VXXqkZM2Zo/fr1Onr0qFasWOGNUD1m/vz5uvbaa5tcF4jn+Fz5Btr5HTBggPMHcGVlpd59910lJCQ41wfi+W0p50A7x/6uqfNlMpnO+T3sS8xms5YtW6bk5GTFxsa2eA3xJXPnzlVWVpbOO+88Sf7zfXD06FHFxsbqT3/6k1577TWtWbNG3333nV+871VVVXI4HJoyZYpSU1P15ptv+tVnRjp5M622tlZJSUl+E3v37t318MMPKykpSQkJCerTp4/MZrNffN4BAPCpIlFDQ4NMJpNz2TCMRsstrfdHLeXUrVs3vfzyy+rfv79CQ0M1ceJElZaWeiPUDhGI5/hcAvX8fvXVV5o4caJmzpypSy65xNkeyOe3uZwD9Rz7u9PPV9++ff3qczlt2jTt3LlThw4dUmVlpV/Evm7dOvXu3VuxsbHONn/5Prjmmmu0ePFi9ejRQ7169VJ6erqWLVvmF7E7HA7t3LlTzzzzjN5++23t2bNHBw4c8IvYT1mzZo3uueceSf7zmfniiy/0zjvvaOvWrdq2bZs6deqkHTt2+EXsAAD4VJEoMjJSVqvVuWy1WhsN0Tlz/ZEjR5odwuMvWsr5u+++U15ennPZMAyFhoZ2aIwdKRDP8bkE4vktLy/XhAkTNH36dN12222N1gXq+T1XzoF4jv3dmeerpe9hX/H1119r7969kqQuXbpo1KhR+vDDD/0i9k2bNmnHjh1KTU3VsmXLVFJSory8PL/4Pti1a5d27tzpXDYMQ3369PGL9/3CCy9UbGysevXqpc6dO+vmm29WWVmZX8QuSXV1dfr44481YsQISS3/ZvIV27dvV2xsrC644AKFhYUpLS3trP9WffXzDgCATxWJbrjhBu3cuVM//PCDTpw4oS1btig+Pt65vk+fPgoPD1d5ebmkk0+OOH29P2op586dO+vZZ5/VgQMHZBiGVq9erZEjR3ox4vYViOf4XALt/B46dEgPPvigcnJylJycfNb6QDy/LeUcaOfY3zV1vgYPHqz9+/c7h+YUFRX55Ofy22+/VXZ2turq6lRXV6f3339fGRkZfhH7qlWrVFRUpIKCAk2bNk0jRozQggUL/OL74JdfftHixYtls9lUU1Oj9evX69lnnz3ntdtXDB8+XNu3b9fRo0flcDi0bds2JSYm+sVnRpK+/PJLXXLJJeratask//lvNSoqSmVlZTp+/LgMw1BJSYmGDh3qF593AAB86nb2RRddpKysLGVmZsputys9PV1XX3217r//fk2bNk1XXXWVcnJylJ2drZqaGg0aNEiZmZneDtstrcl53rx5euCBB2S32/WrX/3K2e06kATyOW5KoJ7flStXymazaeHChc62jIwMlZSUBOz5bU3OgXSO/V1z52vhwoWaOnWqbDabEhISlJiY6MUom5aQkKA9e/bo1ltvVUhIiEaNGqXk5GT16tXL52Nvjj98HwwfPly7d+/WrbfeqoaGBo0bN04xMTFNXrt9zeDBg3Xfffdp3LhxstvtuvHGGzV27FhddtllfvGZOXDggCIjI53L4eHhfvHfalxcnD7//HOlpaXJbDbrqquu0qRJkzRy5Eif/7wDAGAyDMPwdhAAAAAAAADwLp8abgYAAAAAAADvoEgEAAAAAAAAikQAAAAAAACgSAQAAAAAAABRJAIAAAAAAIAoEgEAAAAAAEAUiQAAAAAAACDp/wGukLBI/z51ygAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 1440x1080 with 9 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "temp = imputer.transform(diabetes_in)\n",
    "\n",
    "diabetes_inputed = pd.DataFrame(temp, columns=diabetes_in.columns)\n",
    "\n",
    "diabetes_inputed.hist(bins=50, figsize=(20, 15))\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "After applying a inputer that used a logistic regression to resolve with the values with 0, the data looks much better."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "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>Pregnancies</th>\n",
       "      <th>Glucose</th>\n",
       "      <th>BloodPressure</th>\n",
       "      <th>SkinThickness</th>\n",
       "      <th>Insulin</th>\n",
       "      <th>BMI</th>\n",
       "      <th>DiabetesPedigreeFunction</th>\n",
       "      <th>Age</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.614986</td>\n",
       "      <td>1.460390</td>\n",
       "      <td>0.119633</td>\n",
       "      <td>-0.171435</td>\n",
       "      <td>1.079669</td>\n",
       "      <td>-0.863342</td>\n",
       "      <td>-0.513290</td>\n",
       "      <td>2.715500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2.374415</td>\n",
       "      <td>-0.691978</td>\n",
       "      <td>0.943900</td>\n",
       "      <td>0.399656</td>\n",
       "      <td>-0.504260</td>\n",
       "      <td>-0.358499</td>\n",
       "      <td>0.054719</td>\n",
       "      <td>1.040666</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1.201462</td>\n",
       "      <td>-0.431085</td>\n",
       "      <td>-0.210074</td>\n",
       "      <td>-0.217194</td>\n",
       "      <td>-0.348935</td>\n",
       "      <td>-0.284258</td>\n",
       "      <td>1.496351</td>\n",
       "      <td>-0.047977</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>-0.851206</td>\n",
       "      <td>-0.822425</td>\n",
       "      <td>4.076116</td>\n",
       "      <td>-1.232505</td>\n",
       "      <td>-0.975099</td>\n",
       "      <td>-1.486971</td>\n",
       "      <td>-0.812730</td>\n",
       "      <td>-0.550427</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>-0.851206</td>\n",
       "      <td>-1.115929</td>\n",
       "      <td>-1.034341</td>\n",
       "      <td>0.807934</td>\n",
       "      <td>-0.802723</td>\n",
       "      <td>0.710580</td>\n",
       "      <td>0.119546</td>\n",
       "      <td>-0.969136</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>571</th>\n",
       "      <td>2.081177</td>\n",
       "      <td>1.101662</td>\n",
       "      <td>0.284487</td>\n",
       "      <td>-0.110691</td>\n",
       "      <td>-0.056565</td>\n",
       "      <td>0.131495</td>\n",
       "      <td>2.724980</td>\n",
       "      <td>1.459374</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>572</th>\n",
       "      <td>0.614986</td>\n",
       "      <td>-1.181153</td>\n",
       "      <td>0.449340</td>\n",
       "      <td>-0.121571</td>\n",
       "      <td>-0.775738</td>\n",
       "      <td>-0.180319</td>\n",
       "      <td>-0.272503</td>\n",
       "      <td>0.705699</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>573</th>\n",
       "      <td>-1.144444</td>\n",
       "      <td>-0.626755</td>\n",
       "      <td>-0.704634</td>\n",
       "      <td>1.726559</td>\n",
       "      <td>-0.772877</td>\n",
       "      <td>1.215422</td>\n",
       "      <td>0.079415</td>\n",
       "      <td>-1.052878</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>574</th>\n",
       "      <td>1.787939</td>\n",
       "      <td>0.384206</td>\n",
       "      <td>-0.374927</td>\n",
       "      <td>-0.488084</td>\n",
       "      <td>0.137989</td>\n",
       "      <td>-0.803949</td>\n",
       "      <td>-0.695423</td>\n",
       "      <td>0.203248</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>575</th>\n",
       "      <td>-0.851206</td>\n",
       "      <td>-0.855036</td>\n",
       "      <td>-0.539781</td>\n",
       "      <td>-1.641732</td>\n",
       "      <td>-1.170828</td>\n",
       "      <td>-1.902724</td>\n",
       "      <td>-0.420680</td>\n",
       "      <td>-0.717911</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>576 rows × 8 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     Pregnancies   Glucose  BloodPressure  SkinThickness   Insulin       BMI  \\\n",
       "0       0.614986  1.460390       0.119633      -0.171435  1.079669 -0.863342   \n",
       "1       2.374415 -0.691978       0.943900       0.399656 -0.504260 -0.358499   \n",
       "2       1.201462 -0.431085      -0.210074      -0.217194 -0.348935 -0.284258   \n",
       "3      -0.851206 -0.822425       4.076116      -1.232505 -0.975099 -1.486971   \n",
       "4      -0.851206 -1.115929      -1.034341       0.807934 -0.802723  0.710580   \n",
       "..           ...       ...            ...            ...       ...       ...   \n",
       "571     2.081177  1.101662       0.284487      -0.110691 -0.056565  0.131495   \n",
       "572     0.614986 -1.181153       0.449340      -0.121571 -0.775738 -0.180319   \n",
       "573    -1.144444 -0.626755      -0.704634       1.726559 -0.772877  1.215422   \n",
       "574     1.787939  0.384206      -0.374927      -0.488084  0.137989 -0.803949   \n",
       "575    -0.851206 -0.855036      -0.539781      -1.641732 -1.170828 -1.902724   \n",
       "\n",
       "     DiabetesPedigreeFunction       Age  \n",
       "0                   -0.513290  2.715500  \n",
       "1                    0.054719  1.040666  \n",
       "2                    1.496351 -0.047977  \n",
       "3                   -0.812730 -0.550427  \n",
       "4                    0.119546 -0.969136  \n",
       "..                        ...       ...  \n",
       "571                  2.724980  1.459374  \n",
       "572                 -0.272503  0.705699  \n",
       "573                  0.079415 -1.052878  \n",
       "574                 -0.695423  0.203248  \n",
       "575                 -0.420680 -0.717911  \n",
       "\n",
       "[576 rows x 8 columns]"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.preprocessing import StandardScaler\n",
    "\n",
    "scaler = StandardScaler()\n",
    "\n",
    "scaler.fit(diabetes_inputed)\n",
    "\n",
    "temp = scaler.transform(diabetes_inputed)\n",
    "\n",
    "diabetes_train = pd.DataFrame(temp, columns=diabetes_inputed.columns)\n",
    "diabetes_train"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Binary Classifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "SGDClassifier(max_iter=500, random_state=42)"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.linear_model import SGDClassifier\n",
    "\n",
    "# Existe aleatoriedade dentro do SGDClassifier, por isso o argumento\n",
    "# random_state=RANDOM_SEED.\n",
    "sgd_clf = SGDClassifier(\n",
    "    max_iter=500,\n",
    "    tol=1e-3,\n",
    "    random_state=RANDOM_SEED,\n",
    ")\n",
    "sgd_clf.fit(diabetes_train, diabetes_labels)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Cross Validation"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Before we test the performance of the Model \"SGD Classifier\" we should know the proportions of the target variable. The reason for this is because if the data has a proportion of 80% of people who have diabetes and 20% who haven´t, and the Classifier has an accuracy of 80%, the model would be as bad as one who just guess constanly \"user with diabetes\"."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    0.668403\n",
       "1    0.331597\n",
       "Name: Outcome, dtype: float64"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "diabetes_labels.value_counts(True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0.63793103 0.77586207 0.79310345 0.77586207 0.74137931 0.70689655\n",
      " 0.59649123 0.77192982 0.78947368 0.77192982]\n"
     ]
    }
   ],
   "source": [
    "import time\n",
    "from sklearn.model_selection import cross_val_score\n",
    "\n",
    "t1 = time.process_time()\n",
    "Binary_score = cross_val_score(\n",
    "               sgd_clf,\n",
    "               diabetes_train,\n",
    "               diabetes_labels,\n",
    "               cv=10,\n",
    "               scoring=\"accuracy\",\n",
    "               n_jobs=-1,\n",
    "               )\n",
    "t2 = time.process_time()\n",
    "\n",
    "print(Binary_score)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Here we see that the model had an accuracy around 73%, which may seem ok (but not good enough). But, when it compares the porportion of people who has diabetes in the data, we realize that this model was almost as bad as one who keeps only guessing \"people who don't have diabetes\". Therefore, we need to improve."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Hyperparameters adjusting"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "It is possible to further improve the performance of the SGDC Classifier by using GridSearch, which combines thousands of possible parameters of the model and finds the best of them. For that, it is necessary to pass a grid of parameters, it is necessary to know the SGDC classifier parameters."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "dict_keys(['alpha', 'average', 'class_weight', 'early_stopping', 'epsilon', 'eta0', 'fit_intercept', 'l1_ratio', 'learning_rate', 'loss', 'max_iter', 'n_iter_no_change', 'n_jobs', 'penalty', 'power_t', 'random_state', 'shuffle', 'tol', 'validation_fraction', 'verbose', 'warm_start'])"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sgd_clf.get_params().keys()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "# train across 5 folds, that's a total of (6+4)*5=50 rounds of training.\n",
    "param_grid = {\n",
    "    'alpha': [1e-4, 1e-3, 1e-2, 1e-1, 1e0, 1e1, 1e2, 1e3], # learning rate\n",
    "    'loss': ['log'], # logistic regression,\n",
    "    'penalty': ['l2'],\n",
    "    'n_jobs': [-1]\n",
    "}\n",
    "\n",
    "sgd_clf_grid = GridSearchCV(\n",
    "    sgd_clf,  # Modelo\n",
    "    param_grid,  # Grid\n",
    "    cv=5,  # Partições de C.V.\n",
    "    scoring='neg_mean_squared_error',\n",
    "    return_train_score=True,\n",
    "    n_jobs=-1,\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now, the best SGDC classifier was obtained (that is, the model with the best parameters). With that said, it is necessary to train the model again."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "GridSearchCV(cv=5, estimator=SGDClassifier(max_iter=500, random_state=42),\n",
       "             n_jobs=-1,\n",
       "             param_grid={'alpha': [0.0001, 0.001, 0.01, 0.1, 1.0, 10.0, 100.0,\n",
       "                                   1000.0],\n",
       "                         'loss': ['log'], 'n_jobs': [-1], 'penalty': ['l2']},\n",
       "             return_train_score=True, scoring='neg_mean_squared_error')"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sgd_clf_grid.fit(diabetes_train, diabetes_labels)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'alpha': 0.1, 'loss': 'log', 'n_jobs': -1, 'penalty': 'l2'}"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sgd_clf_grid.best_params_"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "After using Grid Search, it is possible to calculate once again the model accuracy, and a higher performance can be observed."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0.74137931 0.81034483 0.77586207 0.77586207 0.77586207 0.86206897\n",
      " 0.77192982 0.75438596 0.75438596 0.77192982]\n"
     ]
    }
   ],
   "source": [
    "t1 = time.process_time()\n",
    "Binary_score_grid = cross_val_score(\n",
    "    sgd_clf_grid,\n",
    "    diabetes_train,\n",
    "    diabetes_labels,\n",
    "    cv=10,\n",
    "    scoring=\"accuracy\",\n",
    "    n_jobs=-1,\n",
    ")\n",
    "t2 = time.process_time()\n",
    "\n",
    "print(Binary_score_grid)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Feature importance"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Lastly, it is also interesting to analyze the features that impacted the most on the prediction of the model. For that, the graphic below was plotted and Insulin was the feature that had the most importance, followed by Glucose and Blood Pressure, the latter having a negative coefficient value, however the magnitude is what matters, and the negative coefficient only indicates that the dependent variable and independent variable have a strong negative correlation. That makes sense, since people with lower Blood Pressure tend to not have diabetes, it is something more predictable. However, when this value gets higher, more outliers can appear, and it becomes more difficult to predict whether a person will have diabetes or not. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\matfs\\Anaconda3\\lib\\site-packages\\yellowbrick\\model_selection\\importances.py:199: YellowbrickWarning: detected multi-dimensional feature importances but stack=False, using mean to aggregate them.\n",
      "  YellowbrickWarning,\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAaQAAAEUCAYAAABkhkJAAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy8QZhcZAAAgAElEQVR4nO3dd3QU9f7/8WcSQkKLgEqxIKIGhIhAUEooJl6BGEKAhC4CggpSLwgE6UgoUi6olCBeUATp3fK7UiIgkSYISEckhBJKKEkk2c3u5/cHh/0SEgSkZCSvxzmck9mZnc97PjvMa+czs7tuxhiDiIhINnPP7gJERERAgSQiIhahQBIREUtQIImIiCUokERExBIUSCIiYgm5sruAnKp06dL4+vri7v5/7wn8/PyIior6W+vbuXMnCxcuZNiwYXerxExKly5NbGwshQsXvmdtZGXBggXYbDZatWp1X9u9E8nJyXTo0IGkpCS6d+9OnTp1ADh06BC9evUC4OLFiyQlJfHEE08A0KhRI9q2bXvHbe/du5ehQ4dy+fJl3N3d6dmzJzVr1sy03Pvvv8+mTZsyvZ7//e9/efjhh2+7XWMM7dq14+OPP8bHx+dv138//Prrr8yYMYMJEybclfUlJCQQFRXFkSNHAMiTJw+dOnUiMDDQtcymTZuYNm0a8fHxAOTNm5e33nqL0NBQAP7zn/8wb948ihYtCkB6ejplypShe/fulChRwrWeVatWMXPmTJKSkkhPT6d06dL07duXokWLsmDBAtauXcvkyZPvynYBnDx5kl69ejFnzhwuXbrE22+/TUpKCu3bt2fBggXMmTPnrrWFkWzh6+trzp07d9fWt2jRIvPOO+/ctfVl5W7XfKv69u1rpk+fft/bvRObN282//rXv/5ymXv1mgUHB5s1a9YYY4zZs2ePqVSpkrHb7ZmW69Wrl5kxY8Zda9dutxtfX19z8eLFu7bOf4q33nrLfPnll67p/fv3G39/f3PkyBFjjDFr1qwxtWvXNtu2bXMtExcXZ1577TWzatUqY4wx48ePN8OHD8+w3oULF5ratWub5ORkY4wxS5YsMSEhISYuLs4YY4zD4TCTJk0yderUMTabzcyfP9906tTpnm3nxo0bTd26de/Z+nWGZEGHDx8mKiqKCxcu4HA4aN26NRERETidTkaMGMGvv/5KSkoKxhiGDx/OY489xscff0xSUhL9+vWjYcOGfPjhh6xcuRK48s7s6vQnn3zCjh07OH36NKVLl2bs2LFMmTKF//3vfzidTh5//HEGDx7sepeWlfj4eNq0aUNAQAC7d+/G4XDQrVs35s2bx++//46fnx/jx4/nxIkTtG7dmpo1a/Lrr79ijGHQoEFUrlwZu93OqFGjiI2NxcPDg/Lly9OvXz/y589PUFAQ5cuXZ//+/fTs2ZM1a9bw008/4e3tTd26dRk0aBDnzp3jzJkzPP7440yYMIGHH36YoKAgGjVqRGxsLCdPniQsLIwePXoAsHDhQmbMmIG7uzuFChVi9OjRFC9enDVr1jBlyhTsdjve3t707duXihUrcvjwYfr374/NZsMYQ0RERJZnaKtWreLTTz/F6XSSL18+1zZ88MEHJCQkEBYWxrx58/D29r6l195mszFy5Eg2b96Mu7s7FSpUIDIyknz58lGrVi0aNmzIhg0bSE5Opn379jRr1izTOpYvX06uXFf+a8fFxeHj44OHh8cttX9tHR999BHbtm3D4XBQrlw5+vfvT/78+Vm1ahXTp0/HZrORmJhIeHg4Xbt2pV+/fgC0atWK6dOn06RJE6Kjo3n++ecBqFWrFtHR0eTNm5d27dpRokQJTp48yZw5czhy5Ajjxo0jNTUVd3d3unXrRu3atUlISKBv375cvHgRgKCgILp27Zqh1vT0dMqVK8eWLVvw8fHJMO3u7k6/fv2Ii4vD3d2dF154gaFDhxIbG8vo0aNZtmwZ77//PoUKFWLfvn2cPHmSsmXLMnr0aPLkycOaNWsYP348Hh4elC1blvXr17Nw4UKKFSuWoYbTp09z+fJlnE4n7u7u+Pr6MnnyZPLnzw/ARx99xIABA6hUqZLrOU8++STDhw/HZrPd8HUIDw9n2bJlfPvtt0RERDBhwgRGjRrFk08+CYC7uzsdO3bkiSeeIC0tLcNzt23bxvjx40lLS+PMmTPUqlWLDz/8ELvdzrBhw9ixYweenp6UKFGCkSNHkitXriwfP336NOHh4cydO5dBgwa59usJEybQpEkTtm7dijGGyZMns2rVKpxOJ08++SSDBw/m0UcfpUWLFjzyyCMcPnyYVq1a/fVIxz2LOvlLvr6+pn79+qZBgwauf2fPnjV2u928/vrrZvfu3cYYYy5dumSCg4PN9u3bzS+//GK6du1qHA6HMcaY6Oho8+677xpjMr7b/vnnn01ISIirrWunP/74Y1O3bl3XO+YlS5aYHj16uKbnzp1rOnTocMOaz507Z44dO2Z8fX1d7+wGDRpkAgMDTVJSkklNTTUBAQFm27ZtruWWL19ujDEmJibGBAQEGJvNZiZOnGi6dOlibDabcTgcJjIy0gwcONAYY0xgYKD59NNPXe1ee4Y0c+ZMEx0dbYwxxul0mg4dOpjPP//c9bxRo0YZY4w5deqUeeGFF0xcXJzZu3evqVKlijlx4oQxxpgZM2aYgQMHmiNHjpj69eubxMREY4wxBw4cMAEBASYlJcX069fP1c7p06dNjx49XP1+1aFDh0z16tVd71Y3btxoAgICTFJSUqbXICtZnSGNHz/edO/e3djtdpOenm769Oljhg4daowxpmbNmmbw4MHGGGOOHz9uXn75ZXPw4MEs1+1wOExgYKApU6aMmTNnTpbL9OrVy9SoUSPDPrhw4UJjjDETJkwwY8aMMU6n0xhjzOjRo82HH35oHA6HadmypWubT5w4YcqUKWMuXLiQ6QypZs2aZs+ePa72rk7/8ccfxtfX1/zyyy/GGGMSExNNnTp1zPHjx40xxpw8edLUrFnTnDx50kycONG1/cnJyaZbt24mKSkpw3Zc3+610wsXLnT1sd1uN/369TNxcXHmp59+Mg0aNHD1Q8uWLU1aWppJS0szoaGhZsmSJebs2bPmpZdeMvv37zfGGDN//nzj6+trTp48makvN2zYYAICAkyVKlVMp06dzOeff25OnTpljDHm3LlzxtfX11y+fDnL1+GqrM6QjDEmKirKfPjhh+bMmTPG19fXpKWl3XAd154hdevWzWzZssUYY0xSUpJ56aWXzN69e01sbGyGfXPUqFFmx44dN3z8jz/+MP7+/sYYk6Hfrn18wYIFpmfPnq7jyFdffeU6NjVv3tz1f/tmdIaUjb744otM4/eHDh0iLi6ODz74wPVYamoqe/bsoWXLljz00EPMnTuXY8eOsWnTJvLly3fb7VaoUMH1Dnrt2rXs2rWL8PBwAJxOJ5cvX77pOjw9PQkKCgKgRIkSVKxY0fVusEiRIly8eJEiRYrw0EMPucbIa9eujYeHB/v372fdunX8+9//xtPTE4DWrVvTuXNn1/orV66cZbtt2rRh69atzJgxgz/++IODBw/y4osvuua/+uqrABQtWpSHH36YixcvsmXLFmrUqEHx4sUBXNdpZs+ezenTpzNct3FzcyMuLo7XXnuNvn37snPnTqpVq8aAAQMyXO8D+Pnnn6latarr3Wq1atUoXLgwu3fvxs3N7aZ9mJV169bRt29f1+vTqlUrevbs6Zr/xhtvAPDYY48REBDATz/9xLPPPptpPe7u7qxZs4a4uDhatmzJs88+y0svvZRpufbt22d53SomJoY///yT9evXA2C32ylSpAju7u5ER0cTExPDsmXLOHToEMYYUlNTb2tf9PT0dL1uv/zyC2fOnKFTp04Z6j9w4AC1atXi3XffJT4+nurVq9OnTx/XfnYrXnrpJSZOnMibb75J9erVad++PU8++STHjh3LsFytWrXInTs3AM899xwXL15k8+bNrmu9AE2aNLnhNd6AgABiYmLYsWMHW7ZscZ05z5o1y7XfXatbt24cPXrU1a8zZ8684Ta4ubmRJ08e1/5nbvHb3saMGcOPP/7IlClT+P3330lLSyMlJYUyZcrgcDho0qQJNWrUIDg4mPLly7tGZK5//OjRozdtKyYmhj179mQ4jlx75ufv739LNSuQLMbhcFCgQAGWLVvmeuzs2bMUKFCAmJgYoqKiaNeuHa+++iqlSpVi+fLlmdbh5uaWYae12+0Z5ufNm9f1t9PppEOHDrRs2RK4MlRzdXjkr3h6emY46F4NlutdP1TkdDrx8PDA6XRmeL7T6cxQ57U1XmvMmDHs3LmT8PBwqlSpQnp6eoZt9fLycv19tR88PDwytJWamsrx48dxOp1Uq1Ytw4XtkydPUqRIEcqUKcP/+3//j40bNxIbG8ukSZNYvHhxhqGa67cBrhws0tPTb9gfN+NwODKs8+r6rrq2P6/25bVsNhs//PADr7/+Om5ubpQoUYKqVauyZ8+eLAPpr+oYNGgQAQEBwJWbNOx2O8nJyTRq1Ii6devi7+9PeHg4P/zwQ5YHyb/aD729vV0HWKfTia+vL3PnznXNT0hIoHDhwnh6erJ69WpiY2P5+eefiYiI4L///a9rGPBqO1f76vp2SpQowQ8//MCmTZv4+eefadOmDVFRUZlen6z2m6tvCq7fpuudOXOGyZMnM3DgQCpXrkzlypXp1KkTkZGRLF26lP79+1OyZEk2b95MrVq1APj4448B2LhxI6NHj860zmvt2rWLFi1aULhwYUqUKMGOHTuoUqVKhmW6detGly5dXNPGGJo3b46fnx81a9YkJCSE7du3Y4yhYMGCrFixgl9++YWff/6ZHj168O6779KsWbMsH69atepf1gdX9peOHTvStGlTANLS0rh06ZJr/q2+WdFt3xbz9NNP4+3t7QqkkydPUr9+fXbv3s1PP/1EYGAgLVu2xM/Pj1WrVuFwOIArB6qrB67ChQtz4sQJzp07hzGGb7755obt1ahRg4ULF5KcnAzAxIkT6dOnz13bnsTERNatWwfAmjVr8PT0xNfXl5o1a/L1119jt9txOp3Mnj3bdfC73rXbtmHDBtq0aUPDhg15+OGH2bhxo6sPbqRKlSrExsZy+vRpAObOncuYMWOoVq0aP/30E4cPHwbgxx9/pEGDBqSmptKrVy++/fZbQkJCGDx4MPnz5ycuLi7DeqtVq8aGDRtc77avXru69oztdl3tl/T0dFe/VK9e3TV/6dKlwJXreLGxsZnunvP09GTcuHF89913AJw6dYqtW7feVhjBlf1i1qxZ2O12HA4HH3zwARMmTODIkSNcvnyZ7t27ExgYSGxsLOnp6TgcDlfwX7sf7t69G7hy4E1MTMyyravX7LZt2wbAb7/9Rt26dTl37hyjR4/ms88+47XXXmPAgAE8/fTTHDx4MMPzPTw8eOihh1xtXb12CjBr1iwGDhxIzZo16dOnjyucb4W/vz+HDh1ytfftt9/y559/ZgqlggULsm7dOmbNmuUKxcuXLxMXF0e5cuUAiIyMdF2fuSopKYmYmJgbXt8zxjBv3jxOnTpF3bp1AejcuTPDhw937XMOh4NPP/2UgwcPUrJkSddzz58/z759++jduzevvfYax48fJz4+HqfTyapVq2jfvj2VKlWiW7duhIaGsmvXrhs+fitq1KjB/PnzXceR//znP65rirdDZ0gWkzt3biZPnkxUVBTTp08nPT2d7t274+/vT8GCBenVqxehoaGkp6cTEBDguhmhQoUKTJo0iS5duvDpp5/SvHlzwsPDefTRR3nllVduuGM1adKEhIQEmjZtipubG8WLF2fUqFF3bXu8vLxYtmwZY8eOxdvbm0mTJuHh4UGnTp0YPXo0DRs2JD09nfLlyzNw4MAs11GrVi1XTZ07d+ajjz5i4sSJeHp6UqlSpUxBcb3SpUvTu3dvOnToAMCjjz7KiBEjKFq0KMOGDaNnz56ud8RTpkwhX758vPfee/Tv35958+bh4eHBv/71r0wH9WeffZbBgwfTpUsXHA4H3t7eTJ06lQIFCvzt/urSpQujRo0iLCyM9PR0KlSokOE/dlxcHI0aNcJmszF48GCeeuqpDM93c3Nj8uTJDBs2jOjoaNzd3YmMjKRs2bK3VUfXrl1dr8/Vmxr69OmDt7e3azjH09OTMmXKUKpUKeLi4nj88cepU6cOLVq0YPLkyfTu3ZuhQ4cye/ZsXnjhhQxnNdd65JFH+Pjjjxk5cqTrJpKxY8dSrFgx2rZtS2RkJPXr18fT05OyZctSr169TOsYMGAAAwcOpGDBggQEBLiGwhs1asSWLVsICQnB29ubxx9/nFatWrnC668ULlyYMWPG8P777+Ph4YGfnx/u7u6ZblDx9PRkxowZjB07li+++MI1vBYeHk7Dhg0BCAwMpGDBgkydOpUTJ064tvOVV17JcIv2ihUr2Lx5M25ubjidTkqVKsWsWbNcw4lX19e9e3ecTiepqan4+fkxc+ZM1zJXa2/fvj1hYWHkyZOH4sWLU7FiRY4ePUqjRo1Yt24d9evXJ2/evBQsWJDhw4dTpEiRLB+/foQlKy1atOD06dOum2wee+wxRowYcdPnXc/N3OqApMhtio+PJzQ0lO3bt2d3KQ+Eq3ep3ejALnfXpUuXiI6OpmvXrnh7e7Nz5066dOnCjz/++LevEcpf0xmSiEgWfHx8XGc6uXLlwtPTkwkTJiiM7iGdIYmIiCXopgYREbEEBZKIiFiCriHdR06nk5SUlEyf4RERyQmMMdjtdvLly5fpg+agQLqvUlJSOHDgQHaXISKSrXx9fbP8eIQC6T66+ulwX1/fDJ8ZyCl2796Nn59fdpeR7dQPV6gfrshJ/WCz2Thw4MANv8lEgXQfXR2my507d4avKslJcup2X0/9cIX64Yqc1g83umShmxpERMQSFEgiImIJCiQREbEEBZKIiFiCAklERCxBgSQiIpagQBK5z8qU089HwK3/rPWD7p/aDzbHzX8n6Xbpc0gi91k+77w0ndcpu8sQuSPzm0256+vUGZKIiFiCAklERCxBgSQiIpagQBIREUtQIImIiCUokERExBIUSCIiYgkKJBERsQQFkoiIWIICSURELEGBJCIilqBAEhERS1AgiYiIJSiQRETEEhRIIiJiCQokERGxBAWSiIhYggJJREQsQYEkIiKWoEASERFLUCCJiIglKJBERMQSHohAKl269F1dX2RkJIsXLyYhIYG33377rq5bRESy9kAE0r1StGhRPvvss+wuQ0QkR3igAmnTpk289dZbvPfee9StW5du3bphs9lITk7mnXfeoXHjxjRu3JjVq1cD0Lp1azZt2gRAfHw8QUFBGdZ37WORkZEMHz6cFi1aEBQUxKJFi+7vxomIPOByZXcBd9v27dv57rvvKFKkCE2bNmXDhg1cvHiRxx9/nGnTprF3716WL1/Oq6++etvrPnXqFHPmzOHAgQO8+eabhIeH34MtEBHJmR64QHruuecoVqwYAM888wwXL16kYsWKjB8/noSEBF555RU6d+78t9YdEBCAm5sbvr6+XLhw4W6WLSKS4z1QQ3YAXl5err/d3NwwxlCyZEm+++47QkND2bp1KxERETidTgCMMQCkp6ff8rrd3NzuQeUiIjnbAxdIWfnqq6/45JNPCA4OZvDgwSQmJpKcnEyhQoU4dOgQAKtWrcrmKkVEcrYcEUgNGzbkyJEjhIaG0qpVK3r37o2Pjw8dOnRgzpw5NGrUiNTU1OwuU0QkR3MzV8es5J5LS0tj9+7d+Pn5ZRhazCm2bduGv79/dpdhCU3ndcruEkTuyPxmU277OTc7BuaIMyQREbE+BZKIiFiCAklERCxBgSQiIpagQBIREUtQIImIiCUokERExBIUSCIiYgkKJBERsQQFkoiIWIICSURELEGBJCIilqBAEhERS1AgiYiIJSiQRETEEhRIIiJiCQokERGxBAWSiIhYggJJREQsQYEkIiKWoEASERFLUCCJiIgl5MruAkRympTUP5nfbEp2lyFyR2wOO7k9PO/qOnWGJHKf7fttb3aXYAnbtm3L7hIs4Z/aD3c7jECBJCIiFqFAEhERS1AgiYiIJSiQRETEEhRIIiJiCQokERGxBAWSiIhYggJJREQsQYEkIiKWoEASERFLUCCJiIglKJBERMQSFEgi99kLZctkdwmW4O/vn90lWEJ294Mz3Zat7V9LPz8hcp/lzpOP36PCs7sMEQBK9V+U3SW46AxJREQsQYEkIiKWoEASERFLUCCJiIglKJBERMQSFEgiImIJCiQREbEEBZKIiFiCAklERCxBgSQiIpagQBIREUtQIImIiCUokERExBIUSCIiYgkKJBERsQQFkoiIWIICSURELEGBJCIilqBAEhERS1AgiYiIJSiQRETEEhRIIiJiCbmyu4BbkZ6ezmeffcby5ctxc3PD4XDQqFEj3n33Xfr168fLL79M48aNs7tMERG5A/+IQBo6dChnz55l3rx5+Pj4kJycTOfOnSlQoEB2lyYiIneJ5YfsTp06xfLlyxk1ahQ+Pj4A5M+fn0GDBvHII4+4louPjycoKMg1/cknn/DJJ58AsGLFCl5//XVCQkKIjIzEbrdz+fJlevXqRf369QkNDWXp0qUA7Nu3j6ZNm9K4cWNatGjBH3/8AcC6deuIiIigYcOGdOnShfPnz9+nHhARyRksf4a0c+dOnnnmGR566KEMjz/zzDM888wzrF279i+fn5CQwMiRI1m8eDHFihWjd+/e/Pjjj/zyyy8UKlSIlStXkpiYSJMmTShTpgxffPEF7dq1Izg4mCVLlrBjxw58fHwYN24cX375JQ899BBz585l7NixREVF3ctNFxHJUSwfSABubm6uv7///numTJmC0+kkd+7cPPfcc3/53O3bt1OpUiWKFSsGwJgxYwCYPHkyI0aMAKBw4cK8+uqrbN68mdq1azNs2DDWr19PUFAQgYGBrFu3jpMnT/Lmm28C4HQ6MwWkiIjcGcsHkp+fH4cPHyY5OZn8+fNTr1496tWrR3x8vCsg4EpoGWNc0+np6eTKlYtcuXJlCLTExESADMtenXY4HNSrV4+KFSuydu1aZs6cSUxMDK+88gqVKlVi6tSpAKSlpZGSknIvN1tEJMex/DWkxx57jAYNGtC3b18uXboEXAmbmJgY3N3/r3wfHx8uXLhAYmIiNpuN9evXA/DCCy+wY8cOzpw5A8CIESNYvXo1VatWZeHChcCVkFq9ejUvv/wyPXr0YNeuXTRv3pzu3buzZ88eXnzxRXbs2MGRI0eAK2dXH3300f3sBhGRB57lz5AAhgwZwowZM3jzzTdxOBykpKRQpUoVPvvsM6KjowEoUKAAHTp0ICIigmLFivHCCy8AULRoUfr370/79u1xOp1UqFCBxo0bc/nyZYYMGUJoaCgOh4OOHTtSrlw5OnbsSP/+/Zk0aRKenp4MGTKERx99lBEjRtCjRw+cTidFixZ1Df2JiMjd4WauH7uSeyYtLY3du3fj5+eHl5dXdpdz323btg1/f//sLsMSfo8Kz+4SRAAo1X/RfWvrZsdAyw/ZiYhIzqBAEhERS1AgiYiIJSiQRETEEhRIIiJiCQokERGxBAWSiIhYggJJREQsQYEkIiKWoEASERFLUCCJiIglKJBERMQSFEgiImIJCiQREbEEBZKIiFiCAklERCxBgSQiIpagQBIREUtQIImIiCUokERExBIUSCIiYgkKJBERsYRc2V2ASE5ju5xCqf6LsrsMEQCc6Tbcc+XO7jIAnSGJ3He79uzL7hIsYdu2bdldgiVkdz9YJYxAgSQiIhahQBIREUtQIImIiCUokERExBIUSCIiYgkKJBERsQQFkoiIWIICSURELEGBJCIilqBAEhERS1AgiYiIJSiQ/kFsdkd2l3BH/P39s7sESyjzfLnsLkHEkvRt3/8guT09CO21LLvLkDu0YlxYdpcgYkk6QxIREUtQIImIiCUokERExBIUSCIiYgkKJBERsQQFkoiIWIICSURELEGBJCIilqBAEhERS1AgiYiIJSiQRETEEhRIIiJiCQokERGxBAWSiIhYggJJREQsQYEkIiKWoEASERFLUCCJiIglKJBERMQSFEgiImIJCiQREbEEBZKIiFhCrpstsGnTJjp27EiJEiUwxmC322nevDlt2rQhKCiIL7/8kieeeOKOivjkk08A6Nq1K0FBQXh7e+Pp6Ykxhly5ctGnTx+qVq16R22IiIi13TSQAPz8/Jg1axYAycnJhISEEBAQcM+KmjZtmivk1qxZw/vvv8+GDRvuWXsiIpL9bimQrpWWloaHhwcFChRwPeZ0OhkxYgSxsbG4ubnRoEED3nnnHQCmTp3K8uXL8fDwICAggN69e+Ph4cH06dOZP38+hQoVwsfHh/Lly2fZXpUqVThz5gznz59n9OjRXLhwgaNHj9K7d28eeeQRRo4cSWpqKoUKFWLo0KE8+eSTzJgxgyVLluDu7k758uUZNmwY+/btY9CgQaSnp+Pl5cXIkSMpWbIkpUuXZv/+/QAsXryYzZs3M2rUKIKCgihfvjx79+5lzpw5rF+/ni+++AKn00m5cuUYPHgwXl5ef6fPRUQkC7cUSLt37yYsLAyn00lcXBzBwcEUKVLENf/rr7/m5MmTLF++HJvNRuvWrfH19cXNzY01a9awaNEiPD096dq1K3PnzqV8+fIsWrSIJUuW4ObmRrNmzW4YSCtXrqRkyZIUKlQIgIIFCzJ16lRsNhsRERFMnTqVxx57jPXr1zNw4EA+//xzoqOjWb9+PR4eHvTv35+EhAS++OIL2rVrR3BwMEuWLGHHjh2ULFnyL7e7Vq1aTJgwgYMHDzJ//nzmzp2Ll5cX48aN4/PPP+e99967xW4WEZGb+VtDdh06dGDatGmu+Zs2baJRo0Z4eHiQJ08eQkNDiY2Nxd3dnZCQEPLkyQNAeHg4S5cuJTU1ldq1a5MvXz4A6tWrh9PpdK3vnXfewdPTE7vdTvHixZkwYYJr3tXg+uOPPzh27BidOnVyzUtOTsbDw4OKFSsSERHBq6++Srt27ShatOauoj4AAA2PSURBVCi1a9dm2LBhrF+/nqCgIAIDA2+63S+++KJr+44ePUrTpk0BsNvtlC1b9la6TkREbtFtD9nlz5+f4OBgNm7c6Hrs2jABMMbgcDiyfH56ejpubm4YY/6viFy5sNlsrulrryFdz9vb29XmE088wbJlywBwOBycPXsWgMmTJ7Njxw7WrVtHhw4dGDt2LPXq1aNixYqsXbuWmTNnEhMTw/Dhw131urm5kZ6enqGtq0NyDoeD4OBgBgwYAEBKSsoNt09ERP6e277t2+FwsHnz5gxnCFWrVmXp0qU4HA4uX77MihUrqFKlClWrVuWbb74hNTWV9PR0Fi1aRNWqValWrRpr164lKSmJtLQ0fvjhh9suvFSpUly8eJGtW7cCsGjRIt5//30SExN5/fXX8fX1pXv37gQEBLB//3569OjBrl27aN68Od27d2fPnj0AFCpUiIMHD2KMYc2aNVm2VaVKFX744QfOnTuHMYYhQ4bwxRdf3HbNIiJyY7d1DenqWUTp0qV5++23XWcnzZo1448//iAsLAy73U5oaCivvfYaAHv37iU8PJz09HRq1KjBG2+8Qa5cuWjTpg0RERH4+Pjw2GOP3XbhuXPnZuLEiURFRZGWlkb+/PkZPXo0hQsXplmzZkRERJAnTx6efvppwsPDeemll+jfvz+TJk3C09OTIUOGANCrVy86duzII488gr+/P+fPn8/UVpkyZejSpQtt2rTB6XTy/PPPu27aEBGRu8PNXDt2JvdUWloau3fvxs/P72/foRfaa9ldrkrutxXjwrK7BEvYtm0b/v7+2V1GtstJ/XCzY6C+qUFERCxBgSQiIpagQBIREUtQIImIiCUokERExBIUSCIiYgkKJBERsQQFkoiIWIICSURELEGBJCIilqBAEhERS1AgiYiIJSiQRETEEhRIIiJiCQokERGxBAWSiIhYggJJREQsQYEkIiKWoEASERFLUCCJiIglKJBERMQSFEgiImIJubK7ALl1NruDFePCsrsMuUMpf6aSL693dpchYjk6Q/oHye3pkd0l3JFt27ZldwmWsG/vb9ldgoglKZBERMQSFEgiImIJCiQREbEEBZKIiFiCAklERCxBgSQiIpagQBIREUvQB2PvI2MMADabLZsryT5paWnZXYIlqB+uUD9ckVP64eqx7+qx8Hpu5kZz5K5LSkriwIED2V2GiEi28vX1pUCBApkeVyDdR06nk5SUFDw9PXFzc8vuckRE7itjDHa7nXz58uHunvmKkQJJREQsQTc1iIiIJSiQRETEEhRIIiJiCQokERGxBAWSiIhYggJJREQsQYEkIiKWoEASERFLUCCJiIglKJBERMQSFEhyz5w4cYJWrVpRr149OnXqREpKSqZljh8/TsWKFQkLCyMsLIz27dtnQ6X3xooVK3j99depU6cOs2fPzjR/7969NG7cmLp169K/f3/S09Ozocp772b98OmnnxIYGOjaB7Ja5kGRnJxM/fr1iY+PzzQvp+wPf8mI3CPvvPOOWblypTHGmE8//dR89NFHmZb5/vvvzcCBA+93affcqVOnTGBgoDl//rxJSUkxoaGh5uDBgxmWCQkJMdu3bzfGGNOvXz8ze/bs7Cj1nrqVfnj33XfNL7/8kk0V3j87duww9evXN+XKlTPHjh3LND8n7A83ozMkuSfsdjtbtmyhbt26ADRu3Jjvv/8+03K7du3iwIEDhIWF8eabb7J///77Xeo9sXHjRqpWrUrBggXJmzcvdevWzbD9x48fJzU1lQoVKgA37p9/upv1A8Du3buJjo4mNDSUYcOGPbC/DTR//nwGDx5MkSJFMs3LKfvDzSiQ5J44f/48+fPnJ1euK78B+eijj5KQkJBpOS8vLxo0aMCSJUto3749nTt3fiB+wPD06dM8+uijrukiRYpk2P7r59+of/7pbtYPKSkpPP/88/Tu3ZslS5Zw6dIlJk+enB2l3nNRUVFUrlw5y3k5ZX+4Gf1irNyx7777jpEjR2Z47Kmnnsr0m09Z/QZU165dXX/Xrl2bcePG8fvvv1OmTJl7U+x94nQ6M2yvMSbD9M3mPyhutp358uXjs88+c02/9dZbfPDBB/z73/++r3Vmt5yyP9yMAknuWHBwMMHBwRkes9vtVKlSBYfDgYeHB2fOnMlyqGLWrFnUr1+fQoUKAVf+I149q/onK1asGFu3bnVNX7/9xYoV48yZM67ps2fPZtk//3Q364cTJ06wceNGIiIigAfn9b9dOWV/uBkN2ck94enpSeXKlfn2228BWLp0KbVq1cq03JYtW1i4cCEAmzdvxul0UqpUqfta671QvXp1YmNjSUxM5PLly/zvf//LsP2PP/44Xl5ebNu2DYBly5Zl2T//dDfrB29vb8aMGcOxY8cwxjB79mxee+21bKw4e+SU/eGmsvGGCnnAxcfHmzfeeMMEBwebt956y1y4cMEYY8ycOXPMhAkTjDFX7sJq27atCQkJMY0bNzZ79+7NzpLvquXLl5uQkBBTp04dM23aNGOMMR06dDA7d+40xhizd+9eEx4eburWrWt69uxp0tLSsrPce+Zm/fD999+75kdGRj6w/XBVYGCg6y67nLg//BX9hLmIiFiChuxERMQSFEgiImIJCiQREbEEBZKIiFiCAklERCxBgSRiQRs2bCAwMJCIiAiSkpJo3rw5ISEhrFixgubNm//lc7/++mumTZv2t9s+duxYhm/QuBPx8fFUrFjxrqxLHnw57yPRIv8A33zzDU2aNOG9995jy5YtnDt3jh9++AGA0NDQv3xuixYt7qjtEydOcOTIkTtah8jfoUASuUsWLlzIjBkzcHd3p1ChQowePZrixYszb948Zs2ahbu7O4888ggDBw7k6aefxmazMXbsWLZs2YLD4aBs2bIMGDCAuXPnsnr1ary8vPjtt984cOAACQkJhIWFMX78eCIiIti+fTvp6emMGTOGmJgYPDw8qFixIoMHDyY6Oprz588zaNAgEhISGDZsGCdPnsRutxMSEkLHjh2Jj4+nbdu21K5dm19//ZVLly7Ru3dvgoKCGDBgAAkJCbRv357PP//ctX1HjhyhefPmrF+/nty5c+NwOHjllVeYOXMmSUlJjBkzBpvNxpkzZ6hevTojRozI0D+ffPKJq67rp5OSkoiKiuLAgQPY7XaqVatGnz59cuTXCOVkGrITuQv27dvH2LFjmT59OitWrCAoKIgpU6YQGxvL9OnT+fLLL1m+fDn169enc+fOGGOYNm0aHh4eLF68mOXLl1OkSBHGjh1Lhw4dCAoKom3btkyaNInhw4dTokQJli1bhpeXl6vNOXPm8Ntvv7Fs2TJWrlxJSkqK66uarurduzfh4eEsXryYhQsXsnHjRtcyx44do0aNGixcuJBevXoxYsQIPDw8XO1dG0YATz/9NM899xxr1qwBrgwrPvHEEzzzzDN8+eWXdOvWjQULFvDNN9+wZs0adu/efcv9N2LECMqVK8fixYtZunQp58+fZ8aMGX/35ZB/KL39ELkLYmNjqVGjBsWLFwegbdu2AHz00Ue8/vrrFC5cGLjyOzdRUVHEx8cTExNDUlISGzduBK58Ie3DDz98y21u3LiRsLAwvL29AZgwYQJw5cwD4M8//2TLli1cvHiRiRMnuh7bt28f5cuXx9PTk9q1awNQtmxZLly4cNM2IyIiWLJkCfXq1WPx4sU0bdoUgFGjRrFu3TqmTp3K77//TlpaGn/++ScFCxa8pW2JiYlh165dru81TE1NveV+kAeHAknkLvDw8MjwcwGpqakcP34cp9OZaVljDOnp6TidTj744ANXKKSkpNzWj9NdP5x19uzZDO05nU6MMcydO5c8efIAkJiYiJeXF+fPn8fT0xN39yuDJLf6UwfBwcGMGjWKw4cPs2XLFkaNGgXAG2+8QenSpalZsybBwcH8+uuvXP+tZG5ubhkes9vtGWqdOHEizzzzDACXLl3KkT+/kNNpyE7kLqhSpQqxsbGcPn0agLlz5zJmzBhq1qzJt99+S2JiIgCLFi2iYMGCPPXUU9SoUYPZs2djs9lwOp0MHDiQ8ePH33Kb1apVY+XKla7nDxkyhG+++cY1P3/+/FSoUME19HXp0iVatGjB6tWr/3K9Hh4eGcLiWl5eXoSEhBAZGUmdOnXIkycPly5dYteuXbz//vvUqVOHU6dOERcXlymMCxUqxG+//YYxhuTkZNauXeuaV6NGDWbOnIkxBpvNRqdOnfjqq69uuS/kwaAzJJG7oHTp0vTu3ZsOHToAV37xc8SIERQtWpS2bdvSpk0bnE4nhQsXJjo6Gnd3d9577z1Gjx5No0aNcDgcPP/880RGRt5ym82bN+f48eM0btwYYwwvv/wyrVu3ZsqUKa5lxo4dy4cffkhoaCg2m4369evToEED4uPjb7jeZ599Fi8vLyIiIliwYEGmM5UmTZrw1VdfMWTIEAB8fHx45513aNSoEXnz5qVo0aJUqlSJo0eP8uSTT7qe16BBA9avX0+dOnUoWrQoL7/8suuMqX///kRFRREaGordbqd69equvpScQ9/2LSIilqAhOxERsQQFkoiIWIICSURELEGBJCIilqBAEhERS1AgiYiIJSiQRETEEhRIIiJiCf8fTxMAV/fPRcEAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x135c2148f08>"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "viz = FeatureImportances(sgd_clf, relative=False, topn=3)\n",
    "viz.fit(diabetes_train, diabetes_labels)\n",
    "viz.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Also, by plotting the importance of the all features, the difference of importance between the three features discussed previously and the reamining features is quite notable."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\matfs\\Anaconda3\\lib\\site-packages\\yellowbrick\\model_selection\\importances.py:199: YellowbrickWarning: detected multi-dimensional feature importances but stack=False, using mean to aggregate them.\n",
      "  YellowbrickWarning,\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x135c28bb908>"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "viz = FeatureImportances(sgd_clf, relative=False)\n",
    "viz.fit(diabetes_train, diabetes_labels)\n",
    "viz.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Confusion Matrix"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Although the accuracy wasn´t really good, it is important to understand in which cases the model has missed. With the confusion matrix it was learned that there was a lot of false negatives. This is unacceptable in this scneario, since the model should never overlook people with diabetes because it is a health situation and people can be harmed if they are not properly diagnosed. Therefore, the model should focous on having a high Recall."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import cross_val_predict\n",
    "\n",
    "y_train_pred = cross_val_predict(sgd_clf_grid, diabetes_train, diabetes_labels, cv=3, n_jobs=-1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[349,  36],\n",
       "       [ 87, 104]], dtype=int64)"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.metrics import confusion_matrix\n",
    "\n",
    "mat = confusion_matrix(diabetes_labels, y_train_pred)\n",
    "mat"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Precision Recall"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.7428571428571429\n",
      "0.5445026178010471\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import precision_score, recall_score\n",
    "\n",
    "print(precision_score(diabetes_labels, y_train_pred))\n",
    "print(recall_score(diabetes_labels, y_train_pred))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As we can see, the Recall is actually low, and now it is important to adjust this value to make it more accpetable."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_scores = cross_val_predict(\n",
    "    sgd_clf_grid,\n",
    "    diabetes_train,\n",
    "    diabetes_labels,\n",
    "    cv=3,\n",
    "    method=\"decision_function\",\n",
    "    n_jobs=-1,\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.metrics import precision_recall_curve\n",
    "\n",
    "precisions, recalls, thresholds = precision_recall_curve(diabetes_labels, y_scores)\n",
    "\n",
    "plt.figure(figsize=(8, 4))\n",
    "\n",
    "plt.plot(thresholds, precisions[:-1], \"b--\", label=\"Precision\", linewidth=2)\n",
    "plt.plot(thresholds, recalls[:-1], \"g-\", label=\"Recall\", linewidth=2)\n",
    "\n",
    "plt.xlabel(\"Threshold\", fontsize=16)\n",
    "plt.legend(loc=\"upper left\", fontsize=16)\n",
    "plt.xlim([-5, 5])\n",
    "plt.ylim([0, 1])\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "With this graph it is possible to discover a value of Threshold that increases the Recall, which is -0.75. With this, it is possible to adjust the results for the aim of decreasing false negatives"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_train_pred_90 = (y_scores > -0.75)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Precision: 0.5892116182572614\n",
      "Recall: 0.743455497382199\n"
     ]
    }
   ],
   "source": [
    "print('Precision: {}'.format(precision_score(diabetes_labels, y_train_pred_90)))\n",
    "print('Recall: {}'.format(recall_score(diabetes_labels, y_train_pred_90)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(8, 6))\n",
    "\n",
    "plt.plot(recalls, precisions, \"b-\", linewidth=2)\n",
    "\n",
    "plt.xlabel(\"Recall\", fontsize=16)\n",
    "plt.ylabel(\"Precision\", fontsize=16)\n",
    "plt.axis([0, 1, 0, 1])\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "It is possible to see that when the Recall is increased, the Precision has the opposite reaction. In the current scneario, this is not a problem, because the goal is to just increase the Recall."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### ROC Curve"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.metrics import roc_curve\n",
    "\n",
    "fpr, tpr, thresholds = roc_curve(diabetes_labels, y_scores)\n",
    "\n",
    "plt.figure(figsize=(8, 6))\n",
    "plt.plot(fpr, tpr, linewidth=2)\n",
    "plt.plot([0, 1], [0, 1], 'k--')\n",
    "plt.axis([0, 1, 0, 1])\n",
    "plt.xlabel('False Positive Rate', fontsize=16)\n",
    "plt.ylabel('True Positive Rate', fontsize=16)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "For the diabetes problem, the ideal format of this curve is when the area below the curve is high thus making the curve closer to the upper left corner. With the SGD classifier, the curve obtained does not match the ideal shape. This concept is known as ROC curve and what is desired in the present study is a high sensitivity, or true positive rate, meaning a high recall."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Random Forest Classifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0.79310345 0.79310345 0.77586207 0.81034483 0.81034483 0.86206897\n",
      " 0.73684211 0.70175439 0.71929825 0.75438596]\n"
     ]
    }
   ],
   "source": [
    "from sklearn.ensemble import RandomForestClassifier\n",
    "\n",
    "forest_clf = RandomForestClassifier(n_estimators=100, random_state=42)\n",
    "\n",
    "\n",
    "t1 = time.process_time()\n",
    "Rand_Forrest_Score = cross_val_score(\n",
    "    forest_clf,\n",
    "    diabetes_train,\n",
    "    diabetes_labels,\n",
    "    cv=10,\n",
    "    scoring=\"accuracy\",\n",
    "    n_jobs=-1,\n",
    ")\n",
    "t2 = time.process_time()\n",
    "\n",
    "print(Rand_Forrest_Score)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Through the cross val score, it seems that this classifier has a better perfomance than the SDG."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Adjusting Hyperparameters "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Similarly to the SGDC model, here the hyperparameters of the model will be adjusted."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "dict_keys(['bootstrap', 'ccp_alpha', 'class_weight', 'criterion', 'max_depth', 'max_features', 'max_leaf_nodes', 'max_samples', 'min_impurity_decrease', 'min_impurity_split', 'min_samples_leaf', 'min_samples_split', 'min_weight_fraction_leaf', 'n_estimators', 'n_jobs', 'oob_score', 'random_state', 'verbose', 'warm_start'])"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "forest_clf.get_params().keys()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [],
   "source": [
    "param_grid = {\n",
    "    'bootstrap': [True],\n",
    "    'max_depth': [80, 90, 100, 110],\n",
    "    'max_features': [2, 3],\n",
    "    'min_samples_leaf': [3, 4, 5],\n",
    "    'min_samples_split': [8, 10, 12],\n",
    "}\n",
    "\n",
    "forest_clf_grid = GridSearchCV(\n",
    "    forest_clf,  # Modelo\n",
    "    param_grid,  # Grid\n",
    "    cv=5,  # Partições de C.V.\n",
    "    scoring='neg_mean_squared_error',\n",
    "    return_train_score=True,\n",
    "    n_jobs=-1,\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "GridSearchCV(cv=5, estimator=RandomForestClassifier(random_state=42), n_jobs=-1,\n",
       "             param_grid={'bootstrap': [True], 'max_depth': [80, 90, 100, 110],\n",
       "                         'max_features': [2, 3], 'min_samples_leaf': [3, 4, 5],\n",
       "                         'min_samples_split': [8, 10, 12]},\n",
       "             return_train_score=True, scoring='neg_mean_squared_error')"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "forest_clf_grid.fit(diabetes_train, diabetes_labels)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'bootstrap': True,\n",
       " 'max_depth': 80,\n",
       " 'max_features': 3,\n",
       " 'min_samples_leaf': 4,\n",
       " 'min_samples_split': 8}"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "forest_clf_grid.best_params_"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now, with the hyperparameters adjusted, there is a slight improvement to the model's performance."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0.79310345 0.77586207 0.77586207 0.79310345 0.79310345 0.86206897\n",
      " 0.75438596 0.68421053 0.75438596 0.75438596]\n"
     ]
    }
   ],
   "source": [
    "t1 = time.process_time()\n",
    "Random_Forrest_grid_Score = cross_val_score(\n",
    "    forest_clf_grid,\n",
    "    diabetes_train,\n",
    "    diabetes_labels,\n",
    "    cv=10,\n",
    "    scoring=\"accuracy\",\n",
    "    n_jobs=-1,\n",
    ")\n",
    "t2 = time.process_time()\n",
    "\n",
    "print(Random_Forrest_grid_Score)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Feature Importances"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "By plotting the features importance graphs, there are some noticable differences:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x135c28337c8>"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "viz = FeatureImportances(forest_clf, relative=False, topn=3)\n",
    "viz.fit(diabetes_train, diabetes_labels)\n",
    "viz.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x135c243b048>"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "viz = FeatureImportances(forest_clf, relative=False)\n",
    "viz.fit(diabetes_train, diabetes_labels)\n",
    "viz.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- Although Glucose still is one of the most important features, now the 3rd best is no longer Blood pressure, it is BMI. In fact, Blood Pressure is the least important feature, indicating the difference of impact in comparison to the previous model.\n",
    "\n",
    "- Blood Pressure now doesn't have a negative coefficent, which means that for the Random Forest higher values are more important to the prediction than lower values, contrary to what happened in SGDC Classifier.\n",
    "\n",
    "- Unlike previsously, now age has a bigger impact on the prediction. This corroborates what the hypothesis made in the exploratory analysis, that age was important to determine if a person will have diabetes."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [],
   "source": [
    "# O \"score\" vai ser a probabilidade de que a amostra seja da classe positiva.\n",
    "y_probas_forest = cross_val_predict(\n",
    "    forest_clf_grid,\n",
    "    diabetes_train,\n",
    "    diabetes_labels,\n",
    "    cv=3,\n",
    "    method=\"predict_proba\",\n",
    "    n_jobs=-1,\n",
    ")\n",
    "\n",
    "# Gambiarra para desviar do bug #9589 introduzido no Scikit-Learn 0.19.0:\n",
    "y_scores_forest = y_probas_forest[:, 1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fpr_forest, tpr_forest, thresholds_forest = roc_curve(\n",
    "    diabetes_labels,\n",
    "    y_scores_forest,\n",
    ")\n",
    "\n",
    "plt.figure(figsize=(8, 6))\n",
    "plt.plot(fpr, tpr, \"r:\", linewidth=2, label=\"SGD\")\n",
    "plt.plot(fpr_forest, tpr_forest, linewidth=2, label=\"Random Forest\")\n",
    "plt.plot([0, 1], [0, 1], 'k--')\n",
    "plt.axis([0, 1, 0, 1])\n",
    "plt.xlabel('False Positive Rate', fontsize=16)\n",
    "plt.ylabel('True Positive Rate', fontsize=16)\n",
    "plt.legend(loc=\"lower right\", fontsize=16)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "SGD: 0.840103\n",
      "RandomForest: 0.850316\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import roc_auc_score\n",
    "\n",
    "print('SGD: {:4f}'.format(roc_auc_score(diabetes_labels, y_scores)))\n",
    "print('RandomForest: {:4f}'.format(roc_auc_score(diabetes_labels, y_scores_forest)))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As it can be seen, the Random Forest Classifier has a better performance because the area under the curve is higher."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Logistic Regression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0.75862069 0.77586207 0.77586207 0.75862069 0.74137931 0.86206897\n",
      " 0.73684211 0.73684211 0.73684211 0.73684211]\n"
     ]
    }
   ],
   "source": [
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.model_selection import cross_val_predict\n",
    "\n",
    "clf_log = LogisticRegression(random_state=RANDOM_SEED)\n",
    "\n",
    "\n",
    "t1 = time.process_time()\n",
    "Log_Score = cross_val_score(\n",
    "    clf_log,\n",
    "    diabetes_train,\n",
    "    diabetes_labels,\n",
    "    cv=10,\n",
    "    scoring=\"accuracy\",\n",
    "    n_jobs=-1,\n",
    ")\n",
    "t2 = time.process_time()\n",
    "\n",
    "print(Log_Score)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.7619782214156079"
      ]
     },
     "execution_count": 67,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Log_Score.mean()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Adjusting Hyperparameters"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "dict_keys(['C', 'class_weight', 'dual', 'fit_intercept', 'intercept_scaling', 'l1_ratio', 'max_iter', 'multi_class', 'n_jobs', 'penalty', 'random_state', 'solver', 'tol', 'verbose', 'warm_start'])"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "clf_log.get_params().keys()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [],
   "source": [
    "param_grid = [\n",
    "    {\n",
    "    'penalty' : ['l1', 'l2'],\n",
    "    'C' : np.logspace(-4, 4, 20),\n",
    "    'solver' : ['liblinear']\n",
    "    },\n",
    "]\n",
    "\n",
    "logistic_clf_grid = GridSearchCV(\n",
    "    clf_log,  # Modelo\n",
    "    param_grid,  # Grid\n",
    "    cv=5,  # Partições de C.V.\n",
    "    scoring='neg_mean_squared_error',\n",
    "    return_train_score=True,\n",
    "    n_jobs=-1,\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "GridSearchCV(cv=5, estimator=LogisticRegression(random_state=42), n_jobs=-1,\n",
       "             param_grid=[{'C': array([1.00000000e-04, 2.63665090e-04, 6.95192796e-04, 1.83298071e-03,\n",
       "       4.83293024e-03, 1.27427499e-02, 3.35981829e-02, 8.85866790e-02,\n",
       "       2.33572147e-01, 6.15848211e-01, 1.62377674e+00, 4.28133240e+00,\n",
       "       1.12883789e+01, 2.97635144e+01, 7.84759970e+01, 2.06913808e+02,\n",
       "       5.45559478e+02, 1.43844989e+03, 3.79269019e+03, 1.00000000e+04]),\n",
       "                          'penalty': ['l1', 'l2'], 'solver': ['liblinear']}],\n",
       "             return_train_score=True, scoring='neg_mean_squared_error')"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "logistic_clf_grid.fit(diabetes_train, diabetes_labels)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'C': 0.03359818286283781, 'penalty': 'l2', 'solver': 'liblinear'}"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "logistic_clf_grid.best_params_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0.75862069 0.77586207 0.75862069 0.79310345 0.74137931 0.86206897\n",
      " 0.73684211 0.73684211 0.73684211 0.77192982]\n"
     ]
    }
   ],
   "source": [
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.model_selection import cross_val_predict\n",
    "\n",
    "clf_log = LogisticRegression(random_state=RANDOM_SEED)\n",
    "\n",
    "\n",
    "t1 = time.process_time()\n",
    "Log_Score = cross_val_score(\n",
    "    logistic_clf_grid,\n",
    "    diabetes_train,\n",
    "    diabetes_labels,\n",
    "    cv=10,\n",
    "    scoring=\"accuracy\",\n",
    "    n_jobs=-1,\n",
    ")\n",
    "t2 = time.process_time()\n",
    "\n",
    "print(Log_Score)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.7672111312764669"
      ]
     },
     "execution_count": 69,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Log_Score.mean()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Features Importance"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "After analysing the importance of features in the logistic regression model, it is noticable that glucose still remains as the most important feature, and pregnancies also becomes the 2nd most important (like in the SGDC classifier). BMI, like the Random Forest model, also becomes one of the top three most important features. The higher the parameter BMI is, the more likely a person is to develop diabtes, according to a study (https://vitagene.com/blog/does-obesity-cause-type-2-diabetes/), therefore it makes sense that this would be one of the most important features."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\matfs\\Anaconda3\\lib\\site-packages\\yellowbrick\\model_selection\\importances.py:199: YellowbrickWarning: detected multi-dimensional feature importances but stack=False, using mean to aggregate them.\n",
      "  YellowbrickWarning,\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x135c2841dc8>"
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "viz = FeatureImportances(clf_log, relative=False, topn=3)\n",
    "viz.fit(diabetes_train, diabetes_labels)\n",
    "viz.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\matfs\\Anaconda3\\lib\\site-packages\\yellowbrick\\model_selection\\importances.py:199: YellowbrickWarning: detected multi-dimensional feature importances but stack=False, using mean to aggregate them.\n",
      "  YellowbrickWarning,\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAcEAAAEUCAYAAABTSI10AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy8QZhcZAAAgAElEQVR4nOzdZ1RUV/v38e9Q7WJviSVGsSCRaAKKJaJRkSYltoQYjXobe2KIIqjE2DE2YizRR43RqLegaNTYiQVEY4mS2DsWLFgQBYaZ/bzg79xgRQMOMNdnLddy5szZ+zdnzszF3nPmHI1SSiGEEEKYIDNjBxBCCCGMRYqgEEIIkyVFUAghhMmSIiiEEMJkSREUQghhsqQICiGEMFlSBE2Era0tHh4eeHl5Gf4FBQW9cntHjhxh1KhROZjwSba2tiQmJuZqH0/z3//+l6VLl772fv+N+/fv06VLF9zc3Ni8eXOWZcnJyQwePBh3d3c6dOjAggULntpGWFgYTk5OWfYRLy8vjh49+sq5goODiYuLe+X1X5eEhAS6dOmSY+35+/vz+++/51h7Xl5e3Lt375nLk5KS+PTTT7P9+Pj4eOrWrZvldf7www/x9/fn0qVLOZY7J/3666/Mmzcvx9u1yPEWRZ61ePFiSpcunSNtnT59moSEhBxpK685cOAAtWrVMnaMl3Ls2DFu3brFli1bnlj2//7f/6NQoUL89ttv3L9/Hzc3N9577z3s7e2feGyHDh1y9I+b6OhoOnfunGPt5ZYKFSqwfPlyY8d4psjIyOcuv3v3bpY/Vl70eIBChQpleZxSirFjxzJt2jSmTp366mFzSdeuXXOlXSmCgjNnzjBu3Dju3LmDTqfD398fPz8/9Ho948eP56+//iI5OdnwJqlcuTIzZ84kKSmJwMBAOnbsyHfffcdvv/0GQGxsrOF2WFgYhw8f5vr169ja2jJlyhRmz57N5s2b0ev1VKlShdGjR1OhQoVn5ouPj6d79+44OzsTFxeHTqdj0KBBrFixgrNnz2JnZ8fUqVO5cuUK/v7+NG/enL/++gulFKNGjaJx48ZotVomTpxITEwM5ubm2NvbExgYSLFixXBxccHe3p4TJ07w1VdfsX37dvbs2UOhQoVo164do0aN4tatW9y4cYMqVaowffp0ypQpg4uLC97e3sTExHD16lW8vLwYMmQIAKtWrWLhwoWYmZlRqlQpJk2aRKVKldi+fTuzZ89Gq9VSqFAhhg0bhoODA2fOnCEoKIi0tDSUUvj5+fHxxx8/sS22bt3KDz/8gF6vp2jRoobnMGLECBISEvDy8mLFihUUKlTIsI5OpyM5OZn09HRSU1PR6/VYWVm99H7yrNft8OHDhIaGkpaWxo0bN2jatCnjx49n2rRpXL9+na+//prJkyczZcoUPv74Y9q3bw9kjJYe3bazs6N169YcP36cKVOmUKRIkafuk8nJyQQGBnLhwgXMzMyoX78+Y8aMwcws66SWi4sLM2bMoEGDBllu161bl++++46DBw9iaWnJG2+8wYQJE7h9+zYeHh4cOnSIsLAwLl++zI0bN7h8+TIVKlQgNDSU8uXLc+TIEUJCQtBqtVStWpUrV64wfPhwHB0ds70dV6xYwZIlSzAzM6Ns2bKMHDmSGjVqkJiYSGBgIBcvXsTGxoZy5cpRq1YtBg4ciK2tLTExMeh0OoYNG8bt27cBaNmyJUOGDCEwMJCUlBS8vLyIiIigXr16xMTEULp0aebOncvq1auxsLCgWrVqTJw48am5UlNTuX79OmXLlgUgLS2NKVOmsH//fnQ6HfXq1SM4OJhixYo9czsAjBs3jiJFipCcnEx4eDi7d+9+qX3+WfeHhYVx+/ZtRo0axalTpxgzZgx37txBo9HQs2dPOnbsSGxsLNOmTePNN9/k1KlTpKen8+2339KoUaNnvyBKmITatWsrd3d35enpafh38+ZNpdVqVYcOHVRcXJxSSql79+4pV1dXdejQIXXw4EE1cOBApdPplFJKzZ07V/3nP/9RSikVHh6u+vTpo5RSau/evcrNzc3QV+bbM2fOVO3atVNarVYppdTq1avVkCFDDLeXL1+uevXq9czMt27dUpcuXVK1a9dWW7duVUopNWrUKNWqVSuVlJSkUlJSlLOzszpw4IDhcWvXrlVKKRUVFaWcnZ1VWlqamjFjhhowYIBKS0tTOp1ODR8+XI0cOVIppVSrVq3UDz/8YOh32LBhav78+UoppRYtWqTmzp2rlFJKr9erXr16qQULFhjWmzhxolJKqWvXrqkGDRqoixcvqmPHjilHR0d15coVpZRSCxcuVCNHjlTnzp1T7u7uKjExUSml1MmTJ5Wzs7NKTk5WgYGBhn6uX7+uhgwZYtjuj5w+fVo1bdpUXbx4USmlVHR0tHJ2dlZJSUlPvAaZJSUlqY4dOyonJydlZ2enJkyY8NTHzZw5Uzk6OmbZR8LCwl74un355Zdq7969Siml7t+/rxwdHdXRo0cN2+jIkSNKKaU++eQTtXHjRkN/mW/Xrl1brV69WimlnrtPrl69WvXs2VMppVR6eroKCgpS58+ff+K5ZO438+39+/er9u3bK71er5RSavLkyYZ9p2HDhobt0Lp1a5WUlKSUUuo///mPmjFjhtJqtapFixYqKipKKaVUTEyMsrW1NTz3zB5/ro9ER0erNm3aqFu3bimlMt5Hrq6uSq/Xqy+//FJNnjxZKaVUQkKCcnZ2VjNnzjRsn1u3bqkffvjBsN8mJyerIUOGqHv37mXJn/nxW7duVW3btlV37txRSik1fvx49eOPP6pLly6pOnXqKE9PT+Xu7q6aNGmi2rdvr6ZOnaru37+vlFIqLCxMTZw40bCtvv/+ezV69Ojnboe9e/eqOnXqqPj4eKWUeqV9/ln3z5w5U3377bdKq9Wq1q1bq02bNimlMt57zZs3VwcPHlR79+5VdevWVf/8849SSqkFCxaojz/++InXITMZCZqQp02Hnj59mosXLzJixAjDfSkpKfzzzz9069aNkiVLsnz5ci5dukRsbCxFixZ96X4bNmyIhUXGrrZjxw6OHj2Kr68vAHq9nocPH76wDUtLS1xcXACoWrUqDg4OFCtWDIDy5ctz9+5dypcvT8mSJfHw8AAy/ko2NzfnxIkT7Ny5ky+//BJLS0sgYxTSv39/Q/uNGzd+ar/du3fnzz//ZOHChZw/f55Tp07xzjvvGJa3bt0ayJhOK1OmDHfv3mX//v00a9aMSpUqAfDZZ58BsHTpUq5fv264DaDRaLh48SIffvghw4YN48iRIzRp0oTg4OAnRjd79+7FycmJN998E4AmTZpQunRp4uLi0Gg0z9x2Y8aMwdnZma+++oqbN2/So0cPHBwcaNeu3ROPfdZ06PNet4kTJ7Jz507mzJnD2bNnSU1N5cGDB8/M8yyPXoPz588/c59s3rw506ZNw9/fn6ZNm9K9e3eqVauW7T5q166Nubk5H330Ec2aNaNdu3bY29sTHx+f5XHvv/++Yf+qV68ed+/e5eTJk0DGfgXg5OT00tPmu3btokOHDob3oY+PD+PGjSM+Pp4//viD1atXAxn79KMRc2bNmzenT58+XL16laZNmzJ06FCKFy/O3bt3n9pfTEwM7du3p2TJkgAEBgYCGbMrmadDd+3aRUBAAK1atTK8x6OiokhKSiI6OhoArVZLmTJlXrgdKlWqRJUqVQDYs2fPS+/zL3ovnD9/ntTUVNq2bQtkvPfatm3Lrl27cHR0pHLlytStWxfIeO0ebdNnkSJo4nQ6HcWLF8/y3cDNmzcpXrw4UVFRjBs3jh49etC6dWveeust1q5d+0QbGo0GlekUtFqtNsvyIkWKGP6v1+vp1asX3bp1AzKmXJ71Bs7M0tIyywf9o2L2OHNz8yy39Xo95ubm6PX6LOvr9fosOTNnzCw0NJQjR47g6+uLo6Mj6enpWZ6rtbW14f+PtoO5uXmWvlJSUrh8+TJ6vZ4mTZowffp0w7KrV69Svnx56tSpw6ZNm4iOjiYmJoZZs2YRERFBxYoVs2R+vNgppUhPT3/m9gDYsmULa9euxczMzPDhGhsb+9Qi+CzPe90++eQTbG1tad68Oa6uroap6KfJzn7yvH3S2tqaLVu2EBsby969e+nRowdjxowx/IH0rL7S0tIAKFGiBJGRkRw8eJC9e/cyZMgQPv/8c8MH+iOZp5Mzv66PP6/H97cX0ev1T82Znp6OhYVFlvYf/yMIwN7enm3bthETE8PevXv56KOP+Omnn7CxsXlqf4/vi/fu3XvqATPNmzenR48eDB48mPXr11OsWDH0ej0jRowwbJvk5GRSU1O5cePGc7fD4+/3l93nW7Vq9dT7H9HpdM98H8DTX7vnkaNDTVyNGjWy/EV49epV3N3diYuLY8+ePbRq1Ypu3bphZ2fH1q1b0el0QMZO/2inK126NFeuXOHWrVsopVi/fv0z+2vWrBmrVq3i/v37AMyYMYNvvvkmx55PYmIiO3fuBGD79u1YWlpSu3Ztmjdvzq+//opWq0Wv17N06VKcnZ2f2kbm57Z79266d+9Ox44dKVOmDNHR0YZt8CyOjo7ExMRw/fp1AJYvX05oaChNmjRhz549nDlzBoA//vgDT09PUlJSGDp0KBs2bMDNzY3Ro0dTrFgxLl68mKXdJk2asHv3bsPRe4++i8w8Mn2aevXqsXHjRgAePHjArl27XrjO4571ut27d4+jR4/y9ddf07ZtW65du8bFixcNH/aP7yePjhQ9ffo0J06ceGpfz9snly1bRmBgIM2aNSMgIIBmzZrxzz//PNFG5r5iY2O5ceMGkDGi/eyzz3BwcGDgwIF07Ngx20ev1qxZEysrK8P+deTIEU6ePPncUfjjmjdvzoYNGwxHPYeHh2NjY0O1atVo2bIlq1atAuD27dts3br1ibanTJnCjz/+SJs2bQgKCuLtt9/m1KlTWFhYoNPpnvjAb9q0KVu2bDG8bmFhYSxatOip2Xr27EnRokWZOXMmkPGaL126lLS0NPR6PSNHjmTq1KkvtR1eZZ9/0XvhrbfewsLCwnAUdEJCAps2baJp06bZfh0yk5GgibOysuLHH39k3LhxzJ8/n/T0dAYPHkyjRo2wsbFh6NCheHh4kJ6ejrOzs+HAiIYNGzJr1iwGDBjADz/8QJcuXfD19aVcuXJ88MEHzzys/qOPPiIhIYFOnTqh0WioVKnSM7+ofxXW1tZERkYyZcoUChUqxKxZszA3N+eLL75g0qRJdOzYkfT0dOzt7Rk5cuRT22jRooUhU//+/Zk8eTIzZszA0tKSd99994ni9DhbW1sCAgLo1asXAOXKlWP8+PFUqFCBMWPG8NVXX6GUwsLCgtmzZ1O0aFH69etHUFAQK1aswNzcnDZt2vDee+9lafftt99m9OjRDBgwAJ1OR6FChZgzZw7Fixd/bp5JkyYxZswY1qxZg5mZGa6urnh5eWV3kwLPft1KlChBnz598Pb2pkiRIlSoUIF3332XCxcu0KRJEz788EMCAgIICQnhiy++YPjw4fzxxx+89dZbz5yCft4+WbduXfbt20eHDh0oXLgwlSpVwt/f/4k2vv76a0JCQlixYgX169enfv36QMZru3PnTtzd3SlSpAglS5bku+++y9Y2sLCwICwsjNGjRzN16lSqV69O2bJls4w8Mvvmm28M048A3bp1IyAggM8++4zu3buj1+sNB66YmZkRGBhIcHAwHh4e2NjYULly5Sfa7t69O8OHD8fd3R0rKytsbW1xc3MzHOzl5uaW5ec9LVu25PTp04YjK99++22+++477ty580ReS0tLRo4cSa9evfDz86Nfv35MmjQJb29vdDoddevWZfjw4c/dDo9/tfH222+/9D5fpkyZp94fGxtryPnjjz8yduxYwsLC0Ol09O/fHycnJ8NjXspzvzEUIh95/OAAIXLaxIkT1Y0bN5RSSl25ckW999576u7duznS9i+//KIOHjyolFIqNTVV+fr6Gg4+yWtyczu8bjISFEKIbKpSpQqfffaZ4fu7sWPHUqJEiRxp+9Eo7dH31e3bt3/iu8q8Ije3w+umUUouqiuEEMI0yYExQgghTJYUQSGEECZLvhMUOU6v15OcnPzEb/uEEKZNKYVWq6Vo0aJP/R2kMUgRFDkuOTnZcFYJIYR4XO3atV/4057XRYqgyHGPzl5Su3btVzpRc06Li4vDzs7O2DFeKL/khPyTNb/khPyT9d/kTEtL4+TJk889w9HrJkVQ5LhHU6BWVlZZTitmTHklx4vkl5yQf7Lml5yQf7L+25x56WuSvDEpK4QQQhiBFEEhhBAmS4qgEEIIkyVFUAghhMmSIiiEEMJkSREUQghhsuQE2iLHpaamGn5LlBcO+X748CGFCxc2dgwhCqT09HQsLLL3a7u89tkA8jtBYQIKFy7M999/b+wYQhRIQ4cONXaEf0WmQ4UQQpgsKYJCCCFMlhRBIYQQJkuKoBBCCJMlRVAIIYTJkqND85D09HR++ukn1q5di0ajQafT4e3tzX/+8x8CAwN5//338fHxMXZMIYQoMKQI5iHffvstN2/eZMWKFZQoUYL79+/Tv3//PHPxSSGEKGhkOjSPuHbtGmvXrmXixImUKFECgGLFijFq1CjKli1reFx8fDwuLi6G22FhYYSFhQGwbt06OnTogJubG8OHD0er1fLw4UOGDh2Ku7s7Hh4erFmzBoDjx4/TqVMnfHx86Nq1K+fPnwdg586d+Pn50bFjRwYMGMDt27df0xYQQojXT0aCecSRI0eoWbMmJUuWzHJ/zZo1qVmzJjt27Hju+gkJCUyYMIGIiAgqVqxIQEAAf/zxBwcPHqRUqVL89ttvJCYm8tFHH1GnTh0WL15Mjx49cHV1ZfXq1Rw+fJgSJUrw/fff8/PPP1OyZEmWL1/OlClTGDduXG4+dSGEMBopgnlI5qst//7778yePRu9Xo+VlRW1atV67rqHDh3i3XffpWLFigCEhoYC8OOPPzJ+/HgASpcuTevWrdm3bx8tW7ZkzJgx7Nq1CxcXF1q1asXOnTu5evUqn376KQB6vf6JoiyEEAWJFME8ws7OjjNnznD//n2KFStG+/btad++PfHx8YaiBBmFMvPpXh+dt8/CwiJLEU1MTATg8VPDKqXQ6XS0b98eBwcHduzYwaJFi4iKiuKDDz7g3XffZc6cOUDGef6Sk5Nz82kLIYRRyXeCeUTlypXx9PRk2LBh3Lt3D8gocFFRUZiZ/e9lKlGiBHfu3CExMZG0tDR27doFQIMGDTh8+DA3btwAYPz48Wzbtg0nJydWrVoFZBTGbdu28f777zNkyBCOHj1Kly5dGDx4MP/88w/vvPMOhw8f5ty5c0DGKHLy5MmvczMIIcRrJSPBPCQkJISFCxfy6aefotPpSE5OxtHRkZ9++om5c+cCULx4cXr16oWfnx8VK1akQYMGAFSoUIGgoCA+//xz9Ho9DRs2xMfHh4cPHxISEoKHhwc6nY6+fftSv359+vbtS1BQELNmzcLS0pKQkBDKlSvH+PHjGTJkCHq9ngoVKhimVYUQoiCSSymJHJcXL5ciV5EQIne8zFUk8uJng0yHCiGEMFlSBIUQQpgsKYJCCCFMlhRBIYQQJkuODhUF3qNTxwkhct6j3yrnVzISFAXeP//8Y+wI2XLgwAFjR8i2/JI1v+SE/JP18Zz5uQCCFEEhhBAmTIqgEEIIkyVFUAghhMmSIigKvAb17IwdIVsaNWpk7AjZlptZ9en6XGtbiMfl7280hcgGq8LWnJqy29gxRDbV+rqZsSMIEyIjQSGEECZLiqAQQgiTJUVQCCGEyZIiKIQQwmTJgTGvID4+nvbt21OzZk00Gg1arZby5cszYcIEKlasaOx4z+Xl5UVkZKSxYwghRJ4gI8FXVL58eSIjI1mzZg3r16/H1taWyZMnGzvWC0kBFEKI/5GRYA5xdHRk6tSpuLi4YG9vz7Fjx1i2bBm7du1i8eLF6PV66tevz+jRo7G2tmbDhg3MnDmTIkWKULduXXQ6HRMnTsTFxQVPT092797Nw4cPmTRpEnZ2duzbt49p06aRkpLCvXv3CAwMpE2bNgwfPpxixYrx999/k5CQQP/+/fH19eXOnTsEBQVx9uxZrKysGD58OE2aNMHW1pYTJ06QnJzMmDFjOHXqFDqdjt69e+Pu7s7x48cZNWoU6enpWFtbM2HCBKpXr27szSuEELlCRoI5QKvVsmnTJho2bAhAixYt2LRpE4mJiaxcuZLly5cTGRlJmTJlWLBgAYmJiYwfP57FixezatUq7t69m6U9GxsbVq1aRZcuXZg7dy4Av/zyC2PHjmX16tWMHTuWGTNmGB5/7do1li1bxuzZsw2j0RkzZlC1alU2btzI5MmTmT59epY+Zs+eTf369YmIiGDp0qXMmTOHS5cusXjxYnr06EFERASdOnXi8OHDubnphBDCqGQk+IquX7+Ol5cXAGlpadjb2zN06FD27NnDO++8A0BsbCwXLlygU6dOQEaxrFevHn/++ScODg5UqFABgI4dO7J161ZD282bNwegVq1abN68GYDQ0FB27NjB77//zl9//UVycrLh8c7Ozmg0GmrXrs2dO3cA2L9/P1OmTAHA1taWFStWZMkfHR1NSkoK4eHhADx48IBTp07RsmVLxowZw65du3BxcaFVq1Y5u+GEECIPkSL4ih59J/g01tbWAOh0OlxdXQkODgYgOTkZnU7Hvn370OuffWqoR+trNBrDfd26dcPR0RFHR0eaNGnC119//dzHW1hYZLl95swZatSoYbit1+sJDQ2lfv36ANy8eZOSJUtiaWmJg4MDO3bsYNGiRURFRTF27NjsbRQhhMhnZDo0Fzk6OrJlyxZu3bqFUoqQkBAWL17Mu+++y9GjR7l+/TpKKTZs2JClYD3uzp07nD9/nsGDB9OiRQu2bduGTqd7bt+NGzdm/fr1QEYB7N27d5Y+nJyc+PXXX4GMUa2npydXr15lyJAhHD16lC5dujB48OB8cy0+IYR4FTISzEV16tRhwIABdO/eHb1eT926denTpw/W1tYEBwfTs2dPrKyseOONNyhRosQz27GxscHPzw83NzcsLCxwcnIiJSWFBw8ePHOdQYMGERwcjKenJxYWFkyePDlLERwwYAAhISG4u7uj0+kICAigatWq9O3bl6CgIGbNmoWlpSUhISE5uUmEECJP0SillLFDmJrbt2+zZMkSBgwYgJmZGWPHjqVatWr4+/sbO1qOSE1NJS4uDjs7O8NUrbHJCbTzj5w8gfaBAwfyzdU58kvWf5MzL342yEjQCGxsbLh37x7u7u6Ym5tTv359w8EzQgghXh8pgkag0WgMB8sIIYQwHjkwRgghhMmSIiiEEMJkSREUQghhsuQ7QVHgpT1MzdEjDkXu0qfrMbOQv8/F6yF7mijwjv4TZ+wI2XLgwAFjR8i23MwqBVC8TrK3CSGEMFlSBIUQQpgsKYJCCCFMlhRBUeDVq1vX2BGyJT+cMuuRx7Oma7VGSiLEvyNHh4oCr3CRIvwQ2MPYMQq0ARMWGjuCEK9ERoJCCCFMlhRBIYQQJkuKoBBCCJMlRVAIIYTJkiJYQMXGxuLg4ICXlxeenp64urqyePFiAPz9/WnUqBFpaWlZ1vHy8jJc2DciIoLhw4e/9txCCPE6ydGhBZidnR1LliwB4P79+7i5ueHs7AxAsWLF2L17Ny4uLgCcPXuW69evU6JECaPlFUKI101GgiYiNTUVc3NzihcvDkDbtm3ZtGmTYfmGDRto166dseIJIYRRSBEswOLi4vDy8sLDwwMXFxfef/99ypcvD0CLFi3Yt28f2v/7kXNUVBStWrUyZlwhhHjtZDq0AHt8OrRXr17MmzcPACsrKxo1akR0dDSVKlXizTffpFChQsaMK4QQr50UQRNRrFgxXF1diY6ONtzXvn17Nm3aRIUKFejQoYMR0wkhhHHIdKiJ0Ol07Nu3j3r16hnua9GiBbGxsezcuZMWLVoYMZ0QQhiHjAQLsEffCWo0GtLT07G1taV37978+eefQMaU6LvvvguAtbW1MaMKIYRRSBEsoBwdHTl06NBTlz36nhAgNDQ0yzqOjo4A+Pj44OPjk7shhRDCyGQ6VAghhMmSIiiEEMJkSREUQghhsqQICiGEMFlyYIwo8B4+eCBXPs9l6VotFpaWxo4hxEuTkaAo8P45dszYEbLlwIEDxo6QbY9nlQIo8ispgkIIIUyWFEEhhBAmS4qgEEIIkyVFUBR49erWN3aEbGnUqNFr7zNdq3vtfQqRl8jRoaLAK1ykEGOG/mbsGHnSqO/djR1BCKOSkaAQQgiTJUVQCCGEyZIiKIQQwmRJERRCCGGyXlgE4+PjsbOzw8vLCy8vL9q1a0dgYCA3b97k6NGjBAUFPXf94cOHExERke1ASUlJ9O/fP9uPzywsLAxnZ+csWadNm/bSbYSFhQHg5eX1SjmyKzY2FgcHB0PeR/90upw7Yi8wMJDLly8D0Lt3bxISEnKsbSGEyO+ydXRo+fLliYyMBEApxdSpUxk0aBDLli2jQYMGORro7t27HPsXp7nq0qULAwcOBODBgwd06NCBxo0b07x585du69Fzzk12dnZZLnKb02JjYw1/VPz000+51o8QQuRHLz0dqtFoGDhwIKdOneLnn3/G398fgH379tG1a1e8vb1p3bo1W7duNawTFRWFj48PHh4ebNiwAQCdTseECRPw9vbG09OTRYsWATB27FiuX79u+OBes2YN3t7eeHl5MWLECFJTU9FqtQQEBNCxY0c6duzIypUrn5q1SJEi2Nvbc+rUKQDmzZtn6G/y5MkopQCYP38+bdu2pXPnzhw5csSwvq2tLZAxOv3iiy9wc3Ojb9++dOzYkfj4eCIiIvD398fDw4OpU6dy8+ZN+vXrh4+PD76+vkRHRwOQnJzMsGHD8PHxwcvLi99+e/Hh+v7+/sTGxgIZo3EXFxcgY2Q9duxYunbtiouLC+Hh4QDcuXOH/v374+rqipeXFzExMcybN4/r16/Tp08fbt++jYuLC/Hx8ej1esaOHYubmxvu7u7MmzcPyCiYPXv2pF+/fovdbAgAACAASURBVLRr145BgwaRlpb2wqxCCJFfvdLvBK2srKhWrRply5Y13PfLL78wduxYatasSUxMDOPHj6dNmzYAPHz4kJUrV3Lr1i18fX157733DEVy9erVpKWl8fnnn2NnZ0dwcDCffvops2bN4tSpU6xcuZLly5djbW3N999/z4IFC2jcuDF3795lzZo1JCQk8P3339OpU6cncl6+fJmDBw/SvXt3du7cSVxcHKtWrUKj0RAQEMDatWt56623CA8PZ/Xq1Wg0Gjp37oy9vX2WdmbNmkWNGjWYPXs2R48epXPnzoZlCQkJbNiwAQsLC7788kt8fX1p3bo1169fp1u3bqxZs4Y5c+ZQv359Jk2axP379+nSpQvvvPMOAHFxcVmmXT///HM8PT2fu/2vXbvGsmXLOHnyJJ9++im+vr7MmDGDqlWrMmvWLE6cOMGoUaNYsWIFy5cvZ968eZQqVcqw/q+//srVq1dZu3YtaWlp+Pv7U7t2bQoXLsyhQ4fYuHEj5cuXp1OnTuzevdtQgIUQoqB55R/LazQaChUqZLgdGhrKjh07+P333/nrr79ITk42LPP29sbCwoIKFSrQsGFD/vrrL2JiYjh27Bh79+4FMqYuT5w4QcWKFQ3rxcbGcuHCBUOB02q11KtXj65du3Lu3Dk+//xzWrRowTfffGNYZ/ny5WzduhW9Xo+5uTl9+/alUaNGTJo0iSNHjuDj4wNASkoKlStX5ubNm7Rs2ZKiRYsC0L59e/R6fZbnumfPHqZMmQJAgwYNqF27tmFZvXr1sLDI2IzR0dGcPXuWmTNnApCens6lS5eIjo4mJSXFMGp78OABp06domjRoq80Hers7IxGo6F27drcuXMHgP379xsy2trasmLFimeuHxsbi7e3N+bm5hQuXBgPDw9iYmJwcXGhVq1ahtegZs2a3L1796WyCSFEfvJKRTAtLY1z585x69Ytw33dunXD0dERR0dHmjRpwtdff21YZm5ubvi/Xq/H0tISnU5HQEAAbdu2BSAxMZGiRYty48YNw2N1Oh2urq4EBwcDGdOKOp2OEiVKsH79evbs2cMff/yBt7c369evB7J+J5iZTqeje/fu9OjRA4B79+5hbm7OihUrDNOiABYWFk9MAZqbm2d5TGaZ/xDQ6/UsXrwYGxsbAK5fv06ZMmXQ6/WEhoZSv37G6btu3rxJyZIlOXjw4DO3MWDoMz09Pcv91tbWQMYfIplzZ7595swZatSo8dR2Hy/ySinDwTiP2n7U/rOetxBCFAQv/Z2gXq8nLCyMd955h6pVqwIZ30edP3+ewYMH06JFC7Zt25blCMf169ejlOLy5cvExcXRoEEDnJycWLlyJVqtluTkZLp168bhw4exsLAwfOg7OjqyZcsWbt26hVKKkJAQFi9ezLZt2wgICOCDDz4gODiYIkWKcPXq1efmdnJyIjIykuTkZNLT0+nfvz+bNm2iSZMm7Nixg6SkJFJTU9myZcsT6zZp0oR169YBcOLECU6dOpWl4GTuY9myZQCcPn0aDw8PHj58iJOTE7/++iuQURg9PT1fmLdUqVKcPn0aIMv3q8/SuHFjwx8CZ86coXfv3mg0GszNzZ842tTJyYk1a9ag0+l4+PAh69atw9HR8YV9CCFEQZOtkeD169cN31vp9Xrq1q3L1KlTOX78OAA2Njb4+fnh5uaGhYUFTk5OpKSk8ODBAyDjABUfHx/S09MZM2YMpUuXpkuXLly4cAFvb2/S09Px8fHB0dERrVZL5cqV8ff3Z8mSJQwYMIDu3bsb+u3Tpw9mZmZs3rwZNzc3rK2t8fT0xNbWls2bNz/zObi4uHD8+HE6deqETqejefPmeHt7o9Fo6N69O35+fpQoUYLKlSs/sW7//v0JDAzEw8ODqlWrUrZs2SwjwEeCg4MZNWoUHh4eAEyePJlixYoxYMAAQkJCcHd3N4yAq1at+txC2KtXL4YPH054eDitW7d+4Ws0aNAggoOD8fT0xMLCgsmTJ6PRaPjggw/o06cP8+fPNzy2c+fOnD9/Hi8vL7RaLR4eHnz44YeGA3GEEMJUaJTMd71QZGQkb7zxBo0aNeLKlSt88sknbN26FTMzOdfA06SmphIXF4ednV2W6VVjkhNoP92rnkD7wIEDRrnqxcvKLzkh/2T9Nznz4meDXEUiG9566y1Gjx6NXq/HzMyMMWPGSAEUQogCQIpgNjRo0OClznojhBAif5DhjBBCCJMlRVAIIYTJkiIohBDCZMl3gqLAe/gg5ZWPgizo0rU6LCzNX/xAIQooGQmKAu+fY38bO0K2HDhw4LX3KQVQmDopgkIIIUyWFEEhhBAmS4qgEEIIkyVFUBR4dnXqGDtCtrzoVFR6ucCxEDlOjg4VBZ510aLs8fI1dox/zTky3NgRhChwZCQohBDCZEkRFEIIYbKkCAohhDBZUgSFEEKYLCmCj/n999/x8fHB09MTDw8PwxXZXVxciI+Pz/LYbdu2MWPGjGe2deLECby8vPDy8uL999/ngw8+wMvLi48++ggAW1vbp67Xu3dvEhISntnu07IIIYR4eXJ0aCYJCQlMmjSJiIgISpUqRXJyMv7+/tSoUeOpj2/dujWtW7d+Znu2trZERkYCMHz4cN5//318fHxemOOnn356tScghBDipchIMJPbt2+j1WpJSUkBoGjRokycOJG3337b8Jhz587Rtm1bDh8+TEREBMOHDwcyRmfTp0/Hz88PNzc34uListXnqFGj8PT0xNPTkwsXLhjaio+PJzU1lREjRtCuXTvc3d3ZsGFDlnUfz/Lll1/Ss2dPPvzwQ0JCQgyPmzdvHt7e3nh6ejJ58mSUUty/f58+ffrg4+ODj48P27ZtA2DhwoV4enrSsWNHRo0a9crbUggh8gMpgpnUqVOH1q1b06ZNG/z8/AgNDUWv11OtWjUArl27xoABAxg/fjwNGzZ8Yn0bGxtWrVpFly5dmDt3brb6bNq0KWvXrsXZ2Znly5dnWbZkyRIePHjAxo0bWbhwIbNmzSLt/34w/bQshw4dYubMmaxdu5YdO3Zw4sQJdu7cSVxcHKtWrWLNmjUkJCSwdu1atmzZQpUqVYiIiGDcuHH8+eef6HQ65s6dS3h4OBEREWi12udOywohRH4nRfAx3377Ldu3b6dr165cuXKFTp06sXnzZgAGDx7Mm2++SePGjZ+6bvPmzQGoVasWd+7cyVZ/bdq0AeDtt99+Yp39+/fj4eGBmZkZ5cqVY/369VhZWT0zi4ODA8WKFaNw4cK8+eab3L17l5iYGI4cOYKPjw/e3t7ExcVx+vRpHBwc2Lp1K/369ePo0aP0798fc3NzHBwc8PPz44cffqBHjx5UqFDh5TagEELkI1IEM4mKimLDhg1UqFABX19fpk2bRnBwMKtWrQIgKCiIS5cuERUV9dT1ra2tAdBoNNnu08LCwrCOUuqJZZnbunDhgmEk+LQsj/rP3J5Op6N79+5ERkYSGRnJf//7X/r27Uv16tXZuHEjHh4e/Pnnn/j5+aHX6/nxxx8JCQlBKUWvXr3Yt29ftp+LEELkN1IEMylUqBDff/+94chLpRTHjh2jbt26ANjb2xMSEsKYMWN48OBBrud577332LBhA0opbt26xSeffGIogtnN4uTkRGRkJMnJyaSnp9O/f382bdrEL7/8QlhYGK6urowePZrExETu3LlDhw4dqF27NoMHD8bZ2ZkTJ07k+vMUQghjkaNDM3FycmLAgAH07dsXrVYLZExx9u/fn3Xr1gEZhcnR0ZHp06dTJ5dPzNytWzfGjh2Lp6cnACNHjqRYsWKG5dnJ4uLiwvHjx+nUqRM6nY7mzZvj7e1NcnIyX331FR4eHpibmxMQEEDp0qXp3Lkzfn5+FC5cmBo1auDrm//PuSmEEM+iUY/PwQnxL6WmphIXF4ednV2WKVpjkhNo56wDBw688KoXeUF+yQn5J+u/yZkXPxtkOlQIIYTJkiIohBDCZEkRFEIIYbKkCAohhDBZcnSoKPBSk5Pz1EElr0qflobZ/50sQQiRM2QkKAq8uOPHjR0hWw4cOPDc5VIAhch5UgSFEEKYLCmCQgghTJYUQSGEECZLiqAo8OrUr/ta+knTaV9LP0KInCNHh4oCr2ihInRa8UWu97Oy8+xc70MIkbNkJCiEEMJkSREUQghhsqQICiGEMFlSBIUQQpgsKYJCCCFMlhRBE3Dy5ElsbW3ZtGmTsaMIIUSeIkXQBISHh9O+fXtWrFhh7ChCCJGnyO8ECzitVsu6detYunQpXbp04eLFi1StWpXY2FjGjh2Lubk5DRs25MyZMyxZsoQLFy4QEhLCnTt3KFSoECNHjqRevXrGfhpCCJErZCRYwP3xxx9UrlyZGjVq0KZNG1asWIFWq+Wbb74hNDSUNWvWYGHxv7+Fhg0bRkBAAKtXr+a7777jyy+/NGJ6IYTIXVIEC7jw8HDc3d0B6NChAxERERw7dowyZcpQp04dAPz8/ABITk4mLi6OwMBAvLy8GDp0KA8ePOD27dtGyy+EELlJpkMLsFu3brFr1y7+/vtvfv75Z5RS3Lt3j507d6LX6594vF6vx8rKisjISMN9165dw8bG5nXGFkKI10ZGggVYZGQkTk5O7Ny5k+3bt7Njxw769u3L7t27uXfvHidOnABg3bp1ABQvXpzq1asbiuCePXv4+OOPjZZfCCFym4wEC7DVq1c/8Z3exx9/zPz581mwYAHDhg3DzMyMGjVqUKhQIQBCQ0MJCQlh/vz5WFpaMm3aNDQajTHiCyFErpMiWIA9GuFlVrp0aQ4dOsSUKVNYtmwZRYoUYeHChSQkJABQs2ZNlixZ8rqjCiGEUUgRNEFmZmbY2Njg5+eHpaUlVapUYdy4ccaOJYQQr50UQRPVp08f+vTpY+wYQghhVHJgjBBCCJMlRVAIIYTJkulQUeAlpzxgZefZud5Pmk6LlbllrvcjhMg5MhIUBd7xv4+9ln6kAAqR/0gRFEIIYbKkCAohhDBZUgSFEEKYLCmCosBrUK9OrrWtT0/LtbaFELlPjg4VBZ5V4aKcHeebK22/FRSeK+0KIV4PGQkKIYQwWVIEhRBCmCwpgkIIIUyWFEEhhBAmS4pgPmJra5uj7Q0fPpyIiAgSEhLo3bt3jrYthBD5gRRBQYUKFfjpp5+MHUMIIV47KYL5UGxsLD179qRfv360a9eOQYMGkZaWxv379+nTpw8+Pj74+Piwbds2APz9/YmNjQUgPj4eFxeXLO1lvm/48OGMHTuWrl274uLiQni4/ARACFFwye8E86lDhw6xceNGypcvT6dOndi9ezd3796lSpUqzJs3j2PHjrF27Vpat2790m1fu3aNZcuWcfLkST799FN8fXPnN3ZCCGFsUgTzqVq1alGxYkUAatasyd27d3FwcGDq1KkkJCTwwQcf0L9//1dq29nZGY1GQ+3atblz505OxhZCiDxFpkPzKWtra8P/NRoNSimqV6/Oxo0b8fDw4M8//8TPzw+9Xg+AUgqA9PT0bLet0WhyIbkQQuQdUgQLkF9++YWwsDBcXV0ZPXo0iYmJ3L9/n1KlSnH69GkAtm7dauSUQgiRd0gRLEA6duzIuXPn8PDw4OOPPyYgIIASJUrQq1cvli1bhre3NykpKcaOKYQQeYZGPZonEyKHpKamEhcXh52dXZZpW2PKDyfQPnDgAI0aNcqx9nJTfsmaX3JC/sn6b3Lmxc8GGQkKIYQwWVIEhRBCmCwpgkIIIUyWFEEhhBAmS4qgEEIIkyVnjBEFXtrD5Bw9ijMzfXoaZhZWudK2ECL3yUhQFHhH/zmea21LARQif5MiKIQQwmRJERRCCGGypAgKIYQwWVIERZ6QptXlWtt16tbPtbaFEPmbHB0q8gQrS3M8hkbmStvrvvfKlXaFEPmfjASFEEKYLCmCQgghTJYUQSGEECZLiqAQQgiTJQfG/J/Y2Fj69u1L1apVUUqh1Wrp0qUL3bt3x8XFhZ9//pk33njjX/URFhYGwMCBA3FxcaFQoUJYWlqilMLCwoJvvvkGJyennHg6QgghskGKYCZ2dnYsWbIEgPv37+Pm5oazs3Ou9Tdv3jxDYd2+fTtff/01u3fvzrX+hBBCZCVF8BlSU1MxNzenePHihvv0ej3jx48nJiYGjUaDp6cnffr0AWDOnDmsXbsWc3NznJ2dCQgIwNzcnPnz57Ny5UpKlSpFiRIlsLe3f2p/jo6O3Lhxg9u3bzNp0iTu3LnDhQsXCAgIoGzZskyYMIGUlBRKlSrFt99+y5tvvsnChQtZvXo1ZmZm2NvbM2bMGI4fP86oUaNIT0/H2tqaCRMmUL16dWxtbTlx4gQAERER7Nu3j4kTJ+Li4oK9vT3Hjh1j2bJl7Nq1i8WLF6PX66lfvz6jR4/G2to69ze4EEIYgRTBTOLi4vDy8kKv13Px4kVcXV0pX768Yfmvv/7K1atXWbt2LWlpafj7+1O7dm00Gg3bt28nPDwcS0tLBg4cyPLly7G3tyc8PJzVq1ej0Wjo3LnzM4vgb7/9RvXq1SlVqhQANjY2zJkzh7S0NPz8/JgzZw6VK1dm165djBw5kgULFjB37lx27dqFubk5QUFBJCQksHjxYnr06IGrqyurV6/m8OHDVK9e/bnPu0WLFkyfPp1Tp06xcuVKli9fjrW1Nd9//z0LFiygX79+ObaNhRAiL5EimMnj06G9evVi3rx5huWxsbF4e3tjbm5O4cKF8fDwICYmBjMzM9zc3ChcuDAAvr6+rFmzhpSUFFq2bEnRokUBaN++PXq93tBenz59sLS0RKvVUqlSJaZPn25Y9qhYnj9/nkuXLvHFF18Ylt2/fx9zc3McHBzw8/OjdevW9OjRgwoVKtCyZUvGjBnDrl27cHFxoVWrVi983u+8847h+V24cIFOnToBoNVqqVev3ittSyGEyA+kCD5DsWLFcHV1JTo62nBf5gIGoJRCp3v66b7S09PRaDQopQz3WVhYkJaWZrid+TvBxxUqVMjQ5xtvvEFkZMbZVHQ6HTdv3gTgxx9/5PDhw+zcuZNevXoxZcoU2rdvj4ODAzt27GDRokVERUUxduxYQ16NRkN6enqWvh5Nd+p0OlxdXQkODgYgOTn5mc9PCCEKAvmJxDPodDr27duXZSTk5OTEmjVr0Ol0PHz4kHXr1uHo6IiTkxPr168nJSWF9PR0wsPDcXJyokmTJuzYsYOkpCRSU1PZsmXLS+d46623uHv3Ln/++ScA4eHhfP311yQmJtKhQwdq167N4MGDcXZ25sSJEwwZMoSjR4/SpUsXBg8ezD///ANAqVKlOHXqFEoptm/f/tS+HB0d2bJlC7du3UIpRUhICIsXL36FrSeEEPmDjAQzefSd4KPRkq2tLb179zaMwjp37sz58+fx8vJCq9Xi4eHBhx9+CMCxY8fw9fUlPT2dZs2a8cknn2BhYUH37t3x8/OjRIkSVK5c+aUzWVlZMWPGDMaNG0dqairFihVj0qRJlC5dms6dO+Pn50fhwoWpUaMGvr6+vPfeewQFBTFr1iwsLS0JCQkBYOjQofTt25eyZcvSqFEjbt++/URfderUYcCAAXTv3h29Xk/dunUNB/4IIURBpFGZ5+uEyAGpqanExcVhZ2f3UkeWmvoJtA8cOECjRo2MHSNb8kvW/JIT8k/Wf5PzVT8bcpNMhwohhDBZUgSFEEKYLCmCQgghTJYUQSGEECZLjg4VeUKaVpdrB7AkP0ihaJFCudK2ECJ/k5GgyBOsLM1zre3jx/7OtbaFEPmbFEEhhBAmS4qgEEIIkyVFUAghhMmSIiiEEMJkSREUQghhsqQICiGEMFlSBIUQQpgs+bG8yHGPLkyS+QLCxpaammrsCNmSX3JC/smaX3JC/sn6qjkffSbkpYsXyaWURI5LSkri5MmTxo4hhMijateuTfHixY0dA5AiKHKBXq8nOTkZS0tLNBqNseMIIfIIpRRarZaiRYtiZpY3vo2TIiiEEMJk5Y1SLIQQQhiBFEEhhBAmS4qgEEIIkyVFUAghhMmSIiiEEMJkSREUQghhsqQICiGEMFlSBIUQQpgsKYJCCCFMlhRBIYQQJkuKoCgw1q1bR4cOHWjbti1Lly59YvmxY8fw8fGhXbt2BAUFkZ6eboSUL865detWvLy88PT0pF+/fty9e9cIKTO8KOsjUVFRuLi4vMZkWb0o59mzZ/H398fT05PPP//caNv0RTn//vtvfH198fT05D//+Q/37t0zQsr/uX//Pu7u7sTHxz+xLK+8n/41JUQBcO3aNdWqVSt1+/ZtlZycrDw8PNSpU6eyPMbNzU0dOnRIKaVUYGCgWrp0aZ7LmZSUpJydndW1a9eUUkpNnz5dfffdd689Z3ayPnLjxg3Vvn171apVKyOkfHFOvV6v2rZtq/744w+llFKhoaFq8uTJeS6nUkp17dpVRUVFKaWUmjBhgpo6deprz/nI4cOHlbu7u6pfv766dOnSE8vzwvspJ8hIUBQI0dHRODk5YWNjQ5EiRWjXrh2///67Yfnly5dJSUmhYcOGAPj4+GRZnldyarVaRo8eTYUKFQCwtbXl6tWrrz1ndrI+EhwczIABA4yQMMOLcv79998UKVKEFi1aANC3b18+/vjjPJcT/ncFFoCHDx9SqFCh157zkZUrVzJ69GjKly//xLK88n7KCVIERYFw/fp1ypUrZ7hdvnx5EhISnrm8XLlyWZa/Li/KWapUKT788EMAUlJSmDdvHm3atHntOeHFWQF+/vln6tWrxzvvvPO64xm8KOfFixcpW7YsI0aMwNvbm9GjR1OkSJE8lxNg+PDhBAcH06xZM6Kjo+nSpcvrjmkwbtw4Gjdu/NRleeX9lBOkCIoCQa/XZ7l2oVIqy+0XLX9dspsjKSmJPn36UKdOHby9vV9nRIMXZT158iSbN2+mX79+xohn8KKc6enp7Nu3j65du7J69WrefPNNJk6cmOdypqSkEBQUxKJFi9i9ezfdunVj2LBhrz1nduSV91NOkCIoCoSKFSty48YNw+0bN25kmcZ5fPnNmzefOs2T216UEzL+yu7WrRu2traMGzfudUc0eFHW33//nRs3buDr60ufPn0MufNaznLlylGtWjUaNGgAgLu7O0eOHMlzOU+ePIm1tTX29vYAdO7cmX379r32nNmRV95POUGKoCgQmjZtSkxMDImJiTx8+JDNmzcbvgMCqFKlCtbW1hw4cACAyMjILMvzSk6dTkffvn1xdXUlKCjIqH9dvyjroEGD2LRpE5GRkcybN4/y5cuzbNmyPJfTwcGBxMREjh8/DsD27dupX79+nstZrVo1rl27xtmzZwHYtm2boXDnNXnl/ZQjjHhQjhA5au3atcrNzU21bdtWzZs3TymlVK9evdSRI0eUUkodO3ZM+fr6qnbt2qmvvvpKpaam5rmcmzdvVra2tsrT09Pwb8SIEUbJ+aKsmV26dMloR4cq9eKchw8fVr6+vqpDhw6qZ8+e6ubNm3kyZ1RUlPLw8FDu7u6qe/fu6uLFi0bJmVmrVq0MR4fmxffTv6VRSiljF2IhhBDCGGQ6VAghhMmSIiiEEMJkSREUQghhsqQICiGEMFlSBIUQQpgsKYJCCAB2795Nq1at8PPzIykpiS5duuDm5sa6deteePquX3/9lXnz5r1y35cuXWLgwIGvvH5m8fHxODg45EhbouCzMHYAIUTesH79ej766CP69evH/v37uXXrFlu2bAHAw8Pjuet27dr1X/V95coVzp0796/aEOJVSBEUIh9btWoVCxcuxMzMjFKlSjFp0iQqVarEihUrWLJkCWZmZpQtW5aRI0dSo0YN0tLSmDJlCvv370en01GvXj2Cg4NZvnw527Ztw9ramr///puTJ0+SkJCAl5cXU6dOxc/Pj0OHDpGenk5oaChRUVGYm5vj4ODA6NGjmTt3Lrdv32bUqFEkJCQwZswYrl69ilarxc3Njb59+xIfH89nn31Gy5Yt+euvv7h37x4BAQG4uLgQHBxMQkICn3/+OQsWLDA8v3PnztGlSxd27dqFlZUVOp2ODz74gEWLFpGUlERoaChpaWncuHGDpk2bMn78+CzbJywszJDr8dtJSUmMGzeOkydPotVqadKkCd988w0WFvKxaEpkOlSIfOr48eNMmTKF+fPns27dOlxcXJg9ezYxMTHMnz+fn3/+mbVr1+Lu7k7//v1RSjFv3jzMzc2JiIhg7dq1lC9fnilTptCrVy9cXFz47LPPmDVrFmPHjqVq1apERkZibW1t6HPZsmX8/fffREZG8ttvv5GcnMyGDRuy5AoICMDX15eIiAhWrVpFdHS04TGXLl2iWbNmrFq1iqFDhzJ+/HjMzc0N/WUugAA1atSgVq1abN++HciYsn3jjTeoWbMmP//8M4MGDeK///0v69evZ/v27cTFxWV7+40fP5769esTERHBmjVruH37NgsXLnzVl0PkU/InjxD5VExMDM2aNaNSpUoAfPbZZwBMnjyZDh06ULp0aSDjWm/jxo0jPj6eqKgokpKSiI6OBjKuX1imTJls9xkdHY2Xl5fhOnfTp08HMkZYAA8ePGD//v3cvXuXGTNmGO47fvw49vb2WFpa0rJlSwDq1avHnTt3Xtinn58fq1evpn379kRERNCpUycAJk6cyM6dO5kzZw5nz54lNTWVBw8eYGNjk63nEhUVxdGjR1m1ahWQcRUHYXqkCAqRT5mbmz9xKZ7Lly+j1+ufeKxSivT0dPR6PSNGjDAUouTkZFJTU7Pd5+NThTdv3szSn16vRynF8uXLKVy4MACJiYlYW1tz+/ZtLC0tMTPLmIDK7snBXV1dmThxImfOnGH//v2GyyB98skn2Nra0rx5c1xdXfnrr794/CyQGo0my31arTZL1hkzZlCzZk0A7t27l28vByRenUyHCpFPOTo6EhMTw/Xr1wFYuGf8LAAAAedJREFUvnw5oaGhNG/enA0bNpCYmAhAeHg4NjY2VPv/7d0xq+JAFAXggxZBEUGbVPIaQezEIqBYp9EEIgpaiBYiaK0QEEEQBUFsxc4iVrZaioU/QfwBioJ1sIqYbLEQdnnsYrub87UzdwamOXOnma8v5HI5rNdrWJYF27YxGAwwn88/3jOTyWC73br1w+EQu93OHQ+FQkilUu6zommaqFar2O/3f13X7/f/FlC/EgQB+Xweuq5DlmUEAgGYponT6YRutwtZlvF4PHC9Xr9dACKRCM7nMxzHwfP5xOFwcMdyuRxWqxUcx4FlWWi32zAM4+OzoP8DO0Gif1QikUCv10Oz2QTw89+8yWQCURTRaDRQr9dh2zai0SiWyyV8Ph86nQ6m0yk0TcP7/UYymYSu6x/vWalUcL/fUSwW4TgOJElCrVbDYrFw58xmM4xGIyiKAsuyUCgUoKoqbrfbH9eNx+MQBAGlUgmbzeZbR1Yul2EYBobDIQAgHA6j1WpB0zQEg0GIooh0Oo3L5YJYLObWqaqK4/EIWZYhiiIkSXI7w36/j/F4DEVR8Hq9kM1m3bMk7+AvEkRE5Fl8DiUiIs9iCBIRkWcxBImIyLMYgkRE5FkMQSIi8iyGIBEReRZDkIiIPIshSEREnvUDBLZOpvJkn0gAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x135c2409ec8>"
      ]
     },
     "execution_count": 54,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "viz = FeatureImportances(clf_log, relative=False)\n",
    "viz.fit(diabetes_train, diabetes_labels)\n",
    "viz.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [],
   "source": [
    "# O \"score\" vai ser a probabilidade de que a amostra seja da classe positiva.\n",
    "y_probas_logistic = cross_val_predict(\n",
    "    logistic_clf_grid,\n",
    "    diabetes_train,\n",
    "    diabetes_labels,\n",
    "    cv=3,\n",
    "    method=\"predict_proba\",\n",
    "    n_jobs=-1,\n",
    ")\n",
    "\n",
    "# Gambiarra para desviar do bug #9589 introduzido no Scikit-Learn 0.19.0:\n",
    "y_scores_logistic = y_probas_forest[:, 1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0.75862069, 0.77586207, 0.75862069, 0.79310345, 0.74137931,\n",
       "       0.86206897, 0.73684211, 0.73684211, 0.73684211, 0.77192982])"
      ]
     },
     "execution_count": 56,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Log_Score_Grid = cross_val_score(\n",
    "    logistic_clf_grid,\n",
    "    diabetes_train,\n",
    "    diabetes_labels,\n",
    "    cv=10,\n",
    "    scoring=\"accuracy\",\n",
    "    n_jobs=-1,\n",
    ")\n",
    "Log_Score_Grid"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "After using the Logistic Regression it is possible to conclude that its performance is better than the SGD classifier, with an accuracy of 76%. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# SVM"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "SVC(C=100000.0, degree=1, kernel='linear', random_state=42)"
      ]
     },
     "execution_count": 57,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.svm import LinearSVC, SVC\n",
    "\n",
    "clf_svm= SVC(C=1e5, kernel='linear', degree=1,  random_state=RANDOM_SEED)\n",
    "clf_svm.fit(diabetes_train, diabetes_labels)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 100 folds for each of 192 candidates, totalling 19200 fits\n"
     ]
    },
    {
     "ename": "KeyboardInterrupt",
     "evalue": "",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mKeyboardInterrupt\u001b[0m                         Traceback (most recent call last)",
      "\u001b[1;32m<timed exec>\u001b[0m in \u001b[0;36m<module>\u001b[1;34m\u001b[0m\n",
      "\u001b[1;32m~\\Anaconda3\\lib\\site-packages\\sklearn\\utils\\validation.py\u001b[0m in \u001b[0;36minner_f\u001b[1;34m(*args, **kwargs)\u001b[0m\n\u001b[0;32m     61\u001b[0m             \u001b[0mextra_args\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mlen\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0margs\u001b[0m\u001b[1;33m)\u001b[0m \u001b[1;33m-\u001b[0m \u001b[0mlen\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mall_args\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     62\u001b[0m             \u001b[1;32mif\u001b[0m \u001b[0mextra_args\u001b[0m \u001b[1;33m<=\u001b[0m \u001b[1;36m0\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m---> 63\u001b[1;33m                 \u001b[1;32mreturn\u001b[0m \u001b[0mf\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m*\u001b[0m\u001b[0margs\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;33m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m     64\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     65\u001b[0m             \u001b[1;31m# extra_args > 0\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m~\\Anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_search.py\u001b[0m in \u001b[0;36mfit\u001b[1;34m(self, X, y, groups, **fit_params)\u001b[0m\n\u001b[0;32m    839\u001b[0m                 \u001b[1;32mreturn\u001b[0m \u001b[0mresults\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    840\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 841\u001b[1;33m             \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_run_search\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mevaluate_candidates\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m    842\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    843\u001b[0m             \u001b[1;31m# multimetric is determined here because in the case of a callable\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m~\\Anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_search.py\u001b[0m in \u001b[0;36m_run_search\u001b[1;34m(self, evaluate_candidates)\u001b[0m\n\u001b[0;32m   1294\u001b[0m     \u001b[1;32mdef\u001b[0m \u001b[0m_run_search\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mself\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mevaluate_candidates\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   1295\u001b[0m         \u001b[1;34m\"\"\"Search all candidates in param_grid\"\"\"\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m-> 1296\u001b[1;33m         \u001b[0mevaluate_candidates\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mParameterGrid\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mparam_grid\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m   1297\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   1298\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m~\\Anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_search.py\u001b[0m in \u001b[0;36mevaluate_candidates\u001b[1;34m(candidate_params, cv, more_results)\u001b[0m\n\u001b[0;32m    807\u001b[0m                                    (split_idx, (train, test)) in product(\n\u001b[0;32m    808\u001b[0m                                    \u001b[0menumerate\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mcandidate_params\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 809\u001b[1;33m                                    enumerate(cv.split(X, y, groups))))\n\u001b[0m\u001b[0;32m    810\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    811\u001b[0m                 \u001b[1;32mif\u001b[0m \u001b[0mlen\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mout\u001b[0m\u001b[1;33m)\u001b[0m \u001b[1;33m<\u001b[0m \u001b[1;36m1\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m~\\Anaconda3\\lib\\site-packages\\joblib\\parallel.py\u001b[0m in \u001b[0;36m__call__\u001b[1;34m(self, iterable)\u001b[0m\n\u001b[0;32m    932\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    933\u001b[0m             \u001b[1;32mwith\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_backend\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mretrieval_context\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 934\u001b[1;33m                 \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mretrieve\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m    935\u001b[0m             \u001b[1;31m# Make sure that we get a last message telling us we are done\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    936\u001b[0m             \u001b[0melapsed_time\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mtime\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mtime\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m \u001b[1;33m-\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_start_time\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m~\\Anaconda3\\lib\\site-packages\\joblib\\parallel.py\u001b[0m in \u001b[0;36mretrieve\u001b[1;34m(self)\u001b[0m\n\u001b[0;32m    831\u001b[0m             \u001b[1;32mtry\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    832\u001b[0m                 \u001b[1;32mif\u001b[0m \u001b[0mgetattr\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_backend\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;34m'supports_timeout'\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;32mFalse\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 833\u001b[1;33m                     \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_output\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mextend\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mjob\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mget\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mtimeout\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mtimeout\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m    834\u001b[0m                 \u001b[1;32melse\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    835\u001b[0m                     \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_output\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mextend\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mjob\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mget\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m~\\Anaconda3\\lib\\site-packages\\joblib\\_parallel_backends.py\u001b[0m in \u001b[0;36mwrap_future_result\u001b[1;34m(future, timeout)\u001b[0m\n\u001b[0;32m    519\u001b[0m         AsyncResults.get from multiprocessing.\"\"\"\n\u001b[0;32m    520\u001b[0m         \u001b[1;32mtry\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 521\u001b[1;33m             \u001b[1;32mreturn\u001b[0m \u001b[0mfuture\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mresult\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mtimeout\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mtimeout\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m    522\u001b[0m         \u001b[1;32mexcept\u001b[0m \u001b[0mLokyTimeoutError\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    523\u001b[0m             \u001b[1;32mraise\u001b[0m \u001b[0mTimeoutError\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m~\\Anaconda3\\lib\\concurrent\\futures\\_base.py\u001b[0m in \u001b[0;36mresult\u001b[1;34m(self, timeout)\u001b[0m\n\u001b[0;32m    428\u001b[0m                 \u001b[1;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m__get_result\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    429\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 430\u001b[1;33m             \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_condition\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mwait\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mtimeout\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m    431\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    432\u001b[0m             \u001b[1;32mif\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_state\u001b[0m \u001b[1;32min\u001b[0m \u001b[1;33m[\u001b[0m\u001b[0mCANCELLED\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mCANCELLED_AND_NOTIFIED\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m~\\Anaconda3\\lib\\threading.py\u001b[0m in \u001b[0;36mwait\u001b[1;34m(self, timeout)\u001b[0m\n\u001b[0;32m    294\u001b[0m         \u001b[1;32mtry\u001b[0m\u001b[1;33m:\u001b[0m    \u001b[1;31m# restore state no matter what (e.g., KeyboardInterrupt)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    295\u001b[0m             \u001b[1;32mif\u001b[0m \u001b[0mtimeout\u001b[0m \u001b[1;32mis\u001b[0m \u001b[1;32mNone\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 296\u001b[1;33m                 \u001b[0mwaiter\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0macquire\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m    297\u001b[0m                 \u001b[0mgotit\u001b[0m \u001b[1;33m=\u001b[0m \u001b[1;32mTrue\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    298\u001b[0m             \u001b[1;32melse\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;31mKeyboardInterrupt\u001b[0m: "
     ]
    }
   ],
   "source": [
    "%%time\n",
    "from sklearn.model_selection import GridSearchCV, ShuffleSplit\n",
    "from pprint import pprint\n",
    "\n",
    "grid = GridSearchCV(\n",
    "    clf_svm,\n",
    "    {\n",
    "        'C': [10.0**k for k in np.arange(-3, 5, 0.5)], 'kernel': ('linear', 'rbf'),\n",
    "        'degree': [1, 2, 3, 4, 5, 6]\n",
    "    },\n",
    "    scoring='accuracy',\n",
    "    cv=ShuffleSplit(\n",
    "        n_splits=100,\n",
    "        test_size=0.33,\n",
    "        random_state=RANDOM_SEED,\n",
    "    ),\n",
    "    n_jobs=-1,\n",
    "    verbose=3,\n",
    ")\n",
    "# len(clf_svm.get_params().keys())\n",
    "grid.fit(diabetes_train, diabetes_labels)\n",
    "\n",
    "print(grid.best_params_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "C = 0.00100: mean_accuracy = 0.667, stddev_accuracy = 0.026\n",
      "C = 0.00100: mean_accuracy = 0.667, stddev_accuracy = 0.026\n",
      "C = 0.00100: mean_accuracy = 0.667, stddev_accuracy = 0.026\n",
      "C = 0.00100: mean_accuracy = 0.667, stddev_accuracy = 0.026\n",
      "C = 0.00100: mean_accuracy = 0.667, stddev_accuracy = 0.026\n",
      "C = 0.00100: mean_accuracy = 0.667, stddev_accuracy = 0.026\n",
      "C = 0.00100: mean_accuracy = 0.667, stddev_accuracy = 0.026\n",
      "C = 0.00100: mean_accuracy = 0.667, stddev_accuracy = 0.026\n",
      "C = 0.00100: mean_accuracy = 0.667, stddev_accuracy = 0.026\n",
      "C = 0.00100: mean_accuracy = 0.667, stddev_accuracy = 0.026\n",
      "C = 0.00100: mean_accuracy = 0.667, stddev_accuracy = 0.026\n",
      "C = 0.00100: mean_accuracy = 0.667, stddev_accuracy = 0.026\n",
      "C = 0.10000: mean_accuracy = 0.771, stddev_accuracy = 0.023\n",
      "C = 0.10000: mean_accuracy = 0.720, stddev_accuracy = 0.042\n",
      "C = 0.10000: mean_accuracy = 0.771, stddev_accuracy = 0.023\n",
      "C = 0.10000: mean_accuracy = 0.720, stddev_accuracy = 0.042\n",
      "C = 0.10000: mean_accuracy = 0.771, stddev_accuracy = 0.023\n",
      "C = 0.10000: mean_accuracy = 0.720, stddev_accuracy = 0.042\n",
      "C = 0.10000: mean_accuracy = 0.771, stddev_accuracy = 0.023\n",
      "C = 0.10000: mean_accuracy = 0.720, stddev_accuracy = 0.042\n",
      "C = 0.10000: mean_accuracy = 0.771, stddev_accuracy = 0.023\n",
      "C = 0.10000: mean_accuracy = 0.720, stddev_accuracy = 0.042\n",
      "C = 0.10000: mean_accuracy = 0.771, stddev_accuracy = 0.023\n",
      "C = 0.10000: mean_accuracy = 0.720, stddev_accuracy = 0.042\n",
      "C = 10.00000: mean_accuracy = 0.769, stddev_accuracy = 0.023\n",
      "C = 10.00000: mean_accuracy = 0.742, stddev_accuracy = 0.024\n",
      "C = 10.00000: mean_accuracy = 0.769, stddev_accuracy = 0.023\n",
      "C = 10.00000: mean_accuracy = 0.742, stddev_accuracy = 0.024\n",
      "C = 10.00000: mean_accuracy = 0.769, stddev_accuracy = 0.023\n",
      "C = 10.00000: mean_accuracy = 0.742, stddev_accuracy = 0.024\n",
      "C = 10.00000: mean_accuracy = 0.769, stddev_accuracy = 0.023\n",
      "C = 10.00000: mean_accuracy = 0.742, stddev_accuracy = 0.024\n",
      "C = 10.00000: mean_accuracy = 0.769, stddev_accuracy = 0.023\n",
      "C = 10.00000: mean_accuracy = 0.742, stddev_accuracy = 0.024\n",
      "C = 10.00000: mean_accuracy = 0.769, stddev_accuracy = 0.023\n",
      "C = 10.00000: mean_accuracy = 0.742, stddev_accuracy = 0.024\n",
      "C = 1000.00000: mean_accuracy = 0.769, stddev_accuracy = 0.023\n",
      "C = 1000.00000: mean_accuracy = 0.696, stddev_accuracy = 0.029\n",
      "C = 1000.00000: mean_accuracy = 0.769, stddev_accuracy = 0.023\n",
      "C = 1000.00000: mean_accuracy = 0.696, stddev_accuracy = 0.029\n",
      "C = 1000.00000: mean_accuracy = 0.769, stddev_accuracy = 0.023\n",
      "C = 1000.00000: mean_accuracy = 0.696, stddev_accuracy = 0.029\n",
      "C = 1000.00000: mean_accuracy = 0.769, stddev_accuracy = 0.023\n",
      "C = 1000.00000: mean_accuracy = 0.696, stddev_accuracy = 0.029\n",
      "C = 1000.00000: mean_accuracy = 0.769, stddev_accuracy = 0.023\n",
      "C = 1000.00000: mean_accuracy = 0.696, stddev_accuracy = 0.029\n",
      "C = 1000.00000: mean_accuracy = 0.769, stddev_accuracy = 0.023\n",
      "C = 1000.00000: mean_accuracy = 0.696, stddev_accuracy = 0.029\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "P = np.array([p['C'] for p in grid.cv_results_['params']])\n",
    "M = grid.cv_results_['mean_test_score']\n",
    "S = grid.cv_results_['std_test_score']\n",
    "\n",
    "for p, m, s in zip(P, M, S):\n",
    "    print(f'C = {p:.5f}: mean_accuracy = {m:.3f}, stddev_accuracy = {s:.3f}')\n",
    "\n",
    "plt.figure(figsize=(8, 6))\n",
    "plt.errorbar(P, M, S, capsize=4)\n",
    "plt.semilogx()\n",
    "plt.title('Accuracy from CV', fontsize=20)\n",
    "plt.xlabel(r\"$C$\", fontsize=20)\n",
    "plt.ylabel(r\"$accuracy$\", fontsize=20)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.7250130890052354"
      ]
     },
     "execution_count": 60,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#General accuracy of SVM\n",
    "M.mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [],
   "source": [
    "import random\n",
    "M_set = set(M)\n",
    "M_list = list(M_set)\n",
    "M_list.append(M[random.randint(0,len(M))])\n",
    "M_list.append(M[random.randint(0,len(M))])\n",
    "M_list.append(M[random.randint(0,len(M))])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "10"
      ]
     },
     "execution_count": 62,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(M_list)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(8, 6))\n",
    "plt.errorbar(P, M, S, capsize=4)\n",
    "plt.semilogx()\n",
    "plt.title('SVM C parameter', fontsize=20)\n",
    "plt.xlabel(r\"$\\gamma$\", fontsize=20)\n",
    "plt.ylabel(r\"$y$\", fontsize=20, rotation=0)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The idea of SVM algorithm is to trace a curve that can separate when a patient has diabets. However, sometimes the samples are mixed and a curve that completely separates those values is not possible. To calculate the best curve possible, the C parameter used in the algorithm associate penalties to the values that are miscalculated. The higher this parameter is, the less regularization the model will execute, and, hence, overfitting will occur."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This model also has another parameter degree. This determines the degree of the curve that will be used to separate the elements. Whilst creating the model, the grid search was used with this parameter (and also with C), and the outcome was that the model had a better performance when degree was equal to 1. Therefore, the curve obtained is linear."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Finally, it is discoverded that the general accuracy of the Supporting Vector Machine is the diabetes scenario has a 72%  of accuracy."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Features Importance"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "def f_importances(coef, names):\n",
    "    imp = coef\n",
    "    imp,names = zip(*sorted(zip(imp,names)))\n",
    "    plt.barh(range(len(names)), imp, align='center')\n",
    "    plt.yticks(range(len(names)), names)\n",
    "    plt.show()\n",
    "\n",
    "f_importances(clf_svm.coef_[0], diabetes_train.columns.tolist())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "After analysing the importance of features in the Supporting Vector Machine model, it is noticable that glucose still remains as the most important feature, and pregnancies also becomes the 2nd most important (similary to the previous models)."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Conclusion"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAnkAAAE1CAYAAACbe/YrAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy8QZhcZAAAgAElEQVR4nO3dfXzO9f///9vOzMYy9t5JqaRoaCYZlmYi8R5mYQs5/VpOesfYu5Zlq1DLSZaT3gjphCJCm0nI2buEZEXG4h3v3m/e1WGaMtvMtuP1+8PP8WnNyY6Z45jD/Xq5uFz2Onm+no/XcTx37O51crycDMMwEBERERGH4mzvAkRERESk6inkiYiIiDgghTwRERERB6SQJyIiIuKAFPJEREREHJCrvQuoTsxmM/n5+bi5ueHk5GTvckREREQuyzAMiouLqVWrFs7O5Y/bKeT9QX5+PkeOHLF3GSIiIiIVdu+99+Ll5VVuvkLeH7i5uQEXXqwaNWrYuZrqJSsri6CgIHuXITcAjRWxhsaLVJTGSnnnz5/nyJEjlvzyZwp5f3DxFG2NGjVwd3e3czXVj14TqSiNFbGGxotUlMbKpV3uEjPdeCEiIiLigBTyRERERByQQp6IiIiIA1LIExEREXFACnkiIiIiDkghT0RERMQBKeTJFS1fvpygoCDatGlDUFAQy5cvt3dJUk1prIg1NF5Erj99T55c1vLly0lKSmLx4sV4enpSUFBAbGwsAP3797dzdVKdaKyINTReRGxDR/LkslJSUli8eDEdO3bE1dWVjh07snjxYlJSUuxdmlQzGitiDY0XEdtQyJPLys7OJiwsrMy8sLAwsrOz7VSRVFcaK2INjRcR21DIk8tq2rQpO3bsKDNvx44dNG3a1E4VSXWlsSLW0HgRsQ2FPLmspKQkYmNj2bZtGyUlJWzbto3Y2FiSkpLsXZpUMxorYg2NFxHb0I0XclkXL4AeM2YM2dnZNG3alJSUFF0YLeVorIg1NF5EbMPJMAzD3kVUF0VFRWRlZREUFIS7u7u9y6lWMjMzadWqlb3LkBuAxopYQ+NFKkpjpbyr5RadrhURERFxQAp5IiIiIg5IIU9ERETEASnkiYiIiDgghTwRERERB6SQJyIiIuKAFPJEREREHJBdQl5GRgbdunWjS5cufPDBB+WWHzx4kD59+tCzZ09GjhzJmTNnANizZw9t27YlKiqKqKgonn/+eQDOnDnDiBEjiIiIYMCAAeTk5Nh0f0RERESqG5uHPJPJxMyZM1m2bBlpaWmsWLGCH374ocw6KSkpxMXFsXbtWho2bMjixYsByMrKYtiwYaSnp5Oens6UKVMAmDVrFiEhIXz66afExMSQkpJi690SERERqVZsHvJ27txJaGgo3t7eeHp60rVrVzZs2FBmHbPZTH5+PgCFhYXUrFkTgAMHDrBjxw4iIyMZNWoUP//8MwDbt28nMjISgB49evD5559TXFxsw70SERERqV5sHvJOnjyJr6+vZdrPzw+TyVRmncTERJKTkwkLC2Pnzp3069cPAC8vLwYNGkRGRgYdOnQgPj6+3DZdXV2pXbs2ubm5NtojERERkerH1dYdms1mnJycLNOGYZSZPnfuHElJSbz77rsEBwfzzjvvMH78eBYuXMjkyZMt6/Xv35/U1FTy8vLK9WEYBs7Olc+vWVlZlW7ryDIzM+1dgtwgNFbEGhovUlEaK9axecgLCAhg7969lumcnBz8/Pws00eOHMHd3Z3g4GAA+vbty+zZszGbzSxYsIARI0bg4uJiWd/FxQU/Pz9OnTpFQEAAJSUl5Ofn4+3tXekaL/eg35uZHgwtFaWxItbQeJGK0lgpr6io6IoHpmx+urZdu3bs2rWL3NxcCgsL2bRpE+Hh4ZblDRo04JdffuHYsWMAbNmyhebNm+Ps7Mxnn33Gxo0bAUhLS6NFixZ4enrSoUMH0tLSAFi/fj0hISG4ubnZetdEREREqg2bH8nz9/cnPj6ewYMHU1xcTHR0NMHBwQwfPpy4uDiaN2/OlClTGDduHIZh4OPjw6uvvgrAtGnTeOGFF5g7dy716tVj+vTpAIwdO5bExES6d++Ol5cXM2bMsPVuiYiIiFQrToZhGPYuorq4eNhTp2vL02FyqSiNFbGGxotUlMZKeVfLLXrihYiIiIgDUsgTERERcUAKeSIiIiIOSCFPRERExAEp5ImIiIg4IIU8EREREQekkCciIiLigBTyRERERByQQp6IiIiIA1LIExEREXFACnkiIiIiDkghT0RERMQBKeSJiIiIOCCFPBEREREHpJAnIiIi4oAU8kREREQckEKeiIiIiANSyBMRERFxQAp5IiIiIg5IIU9ERETEASnkiYiIiDgghTwRERERB2SXkJeRkUG3bt3o0qULH3zwQbnlBw8epE+fPvTs2ZORI0dy5swZAI4ePcqAAQOIioqib9++ZGdnA/C///2Pli1bEhUVRVRUFLGxsTbdHxEREZHqxuYhz2QyMXPmTJYtW0ZaWhorVqzghx9+KLNOSkoKcXFxrF27loYNG7J48WIAkpOTGT58OOnp6YwbN47x48cDkJWVRWRkJOnp6aSnp1vWFxEREblZ2Tzk7dy5k9DQULy9vfH09KRr165s2LChzDpms5n8/HwACgsLqVmzJgAxMTG0b98egMDAQH7++WcADhw4wJEjR4iKimLw4MEcPnzYhnskIiIiUv3YPOSdPHkSX19fy7Sfnx8mk6nMOomJiSQnJxMWFsbOnTvp168fAL1798bFxQWAOXPm0LlzZwDc3d3p2bMnH3/8MbGxsTz99NOcP3/eRnskIiIiUv242rpDs9mMk5OTZdowjDLT586dIykpiXfffZfg4GDeeecdxo8fz8KFCy3rT58+nf3797NkyRIAxowZY2nfoUMHUlNTOXbsGE2aNKlUjVlZWZVq5+gyMzPtXYLcIDRWxBoaL1JRGivWsXnICwgIYO/evZbpnJwc/Pz8LNNHjhzB3d2d4OBgAPr27cvs2bMBKCkpYfz48ZhMJpYsWYKXlxcAS5cupUePHtStWxe4EARdXSu/a0FBQbi7u1e6vSPKzMykVatW9i5DbgAaK2INjRepKI2V8oqKiq54YMrmp2vbtWvHrl27yM3NpbCwkE2bNhEeHm5Z3qBBA3755ReOHTsGwJYtW2jevDkA06ZN4+zZs7z99tuWgAfw9ddfs2rVKgD27NmD2Wzm7rvvtuFeiYiIiFQvNj+S5+/vT3x8PIMHD6a4uJjo6GiCg4MZPnw4cXFxNG/enClTpjBu3DgMw8DHx4dXX32V3NxcPvjgA26//XZiYmIs20tPTycpKYnExETS09Nxd3cnNTUVZ2d9BaCIiIjcvGwe8gAiIyOJjIwsM2/RokWWnzt06ECHDh3KtTt06NAlt+fv788777xTtUWKiIiI3MB0uEtERETEASnkiYiIiDgghTwRERERB6SQJyIiIuKAFPJEREREHJBCnoiIiIgDUsgTERERcUAKeSIiIiIOSCFPRERExAEp5ImIiIg4IIU8EREREQekkCciIja3fPlygoKCaNOmDUFBQSxfvtzeJYk4HFd7FyAiIjeX5cuXk5SUxOLFi/H09KSgoIDY2FgA+vfvb+fqRByHjuSJiIhNpaSksHjxYjp27IirqysdO3Zk8eLFpKSk2Ls0EYeikCciIjaVnZ1NWFhYmXlhYWFkZ2fbqSIRx6SQJyIiNtW0aVN27NhRZt6OHTto2rSpnSoScUwKeSIiYlNJSUnExsaybds2SkpK2LZtG7GxsSQlJdm7NBGHohsvRETEpi7eXDFmzBiys7Np2rQpKSkpuulCpIop5ImIiM3179+f/v37k5mZSatWrexdjohD0ulaEREREQekkCciIiLigBTyRERERByQQp6IiIiIA7JLyMvIyKBbt2506dKFDz74oNzygwcP0qdPH3r27MnIkSM5c+YMAGfOnGHEiBFEREQwYMAAcnJyADh//jwJCQlERETQq1cvjh49atP9EREREalubB7yTCYTM2fOZNmyZaSlpbFixQp++OGHMuukpKQQFxfH2rVradiwIYsXLwZg1qxZhISE8OmnnxITE2N5BM7SpUvx8PDg008/ZcKECTz//PO23i0RERGRasXmIW/nzp2Ehobi7e2Np6cnXbt2ZcOGDWXWMZvN5OfnA1BYWEjNmjUB2L59O5GRkQD06NGDzz//nOLiYrZv307Pnj0BaN26Nbm5ufz000823CsRERGR6sXmIe/kyZP4+vpapv38/DCZTGXWSUxMJDk5mbCwMHbu3Em/fv3KtXV1daV27drk5uaW26avry+//PKLDfZGREREpHqy+Zchm81mnJycLNOGYZSZPnfuHElJSbz77rsEBwfzzjvvMH78eBYuXFhuW4Zh4OzsXG4bF+dXVlZWVqXbOrLMzEx7lyA3CI0VsYbGi1SUxop1bB7yAgIC2Lt3r2U6JycHPz8/y/SRI0dwd3cnODgYgL59+zJ79mzgwlG/U6dOERAQQElJCfn5+Xh7e+Pv78/Jkye58847ATh16lSZbVorKCgId3f3Srd3RPpWeqkojRWxhsaLVJTGSnlFRUVXPDBl89O17dq1Y9euXeTm5lJYWMimTZsIDw+3LG/QoAG//PILx44dA2DLli00b94cgA4dOpCWlgbA+vXrCQkJwc3NjQ4dOpCeng7A3r17cXd357bbbrPxnomIiIhUHzY/kufv7098fDyDBw+muLiY6OhogoODGT58OHFxcTRv3pwpU6Ywbtw4DMPAx8eHV199FYCxY8eSmJhI9+7d8fLyYsaMGQAMGjSIF198ke7du1OjRg2mT59u690SERERqVacDMMw7F1EdXHxsKdO15anw+RSURorYg2NF6kojZXyrpZbrnokb8CAARXqyN3dnbffftv6CkVERESkyl015B04cIBJkyZdcR3DMCxfTCwiIiIi9nfVkNeyZUt69ep11Q2tW7euSgoSERERkWt31btr33vvvQptSKdqRURERKoPm99dK/bRqVMnDh8+bJe+AwMD2bp1q136vln1T17P2cLiSrc/9OnLnPv95yqsqOJq1rmVZhEvVLp9bQ83lr/SrQorEhG5MVkV8n7++We+//57zp49S+3atWnSpAm33nrr9apNqtC1hqzIZ9LJSI2qomrkejtbWHxt79c1tLX3HXCRz6TbrW8RkeqkQiGvoKCAhIQEtmzZwi233IKnpycFBQXk5eXxyCOP8Nprr+Hh4XG9axURERGRCqrQEy9SU1M5d+4cW7duZc+ePWzfvp09e/awefNmCgsLSU1Nvd51ioiIiIgVKhTyNm/ezKRJk8o9Kqx+/fpMnDiRTZs2XZfiRERERKRyKhTyzpw5w+23337JZXfccQd5eXlVWpSIiIiIXJsKXZN37tw5Hn744UsuMwyDoqKiqqxJRERERK5RhUJeRb8rT0RERESqhwqFvDZt2pSb9/nnn3P06FGaNWtG27Ztq7wwEREREam8Cl2TFxcXx0cffWSZ/sc//kFCQgLffPMN8fHxvP/++9etQBERERGxXoVCXmZmJhEREQCcP3+eJUuWMHv2bN544w3efvttlixZcl2LFBERERHrVCjknT9/nlq1agGwf/9+nJycCA0NBaBJkyacOnXq+lUoIiIiIlarUMi77777WLJkCWfPnmXZsmWEh4dblplMJmrXrn3dChQRERER61Uo5CUnJ7Ns2TJCQkLYv38/cXFxlmUrVqygdevW161AEREREbFehe6ubdSoERs3buT06dN4e3vj5ORkWTZ69GicnSuUFUVERETERioU8tLS0srNc3JyolatWjRp0uSyT8MQEREREfuoUMhbuXLlJeeXlpZy9OhRJk2aRPfu3au0MBERERGpvAqFvGXLll122Xfffcezzz6rkCciIiJSjVzzxXRBQUH6ChURERGRaqZCR/LMZvNl569evZrGjRtb1WlGRgbz58+npKSEIUOGMGDAAMuy7OxsEhMTLdO5ubnUqVOH9957j2HDhlnm5+Xlcfr0ab799lv27NnDmDFjCAgIAKBZs2ZMmTLFqppEREREHEmFQl6zZs3K3FH7R/fccw8zZsyocIcmk4mZM2eyZs0aatSoQb9+/Wjbti2NGjUCoGnTpqSnpwNQWFhITEwMEydOxMfHxzLfbDYzZMgQ4uPjAcjKymLYsGGMHDmywnWIiIiIOLIKhbwtW7aUm+fs7IyHhwfe3t5Wdbhz505CQ0Mt7bp27cqGDRsYPXp0uXUXLFhA69atCQkJKTN/9erVeHh4EBkZCcCBAwc4deoU69ato379+rz00kvceuutVtUl4kjcg3bw+IoN9ivgh7fs1rV7UG0gym7932j6J6/nbGFxpdsf+vRlzv3+cxVWVHE169xKs4gXKt2+tocby1/pVoUViVQvFQp59evXr7IOT548ia+vr2Xaz8+P7777rtx6eXl5rFy5koyMjDLzS0tLefPNN5k3b55lnpeXFxEREXTp0oXly5cTHx/Phx9+WOkas7KyKt3WkWVmZtq7BKmgoqwwJj5xc3610cRlJzRWrXC2sPjaxsoTC6quGBvTWLnx6P2yzlVD3qxZsxg3btxVNzRnzpwyT8K4HLPZXObUr2EYlzwVvHbtWjp37oyPj0+Z+V988QV33XUXgYGBlnmTJ0+2/Ny/f39SU1PJy8vDy8vrqvVcSlBQEO7u7pVqe71c6/+2q8LEZSfs0q/+t10Jy07QqlUru3SdmZlpt74Bu+77DcnOr5ddx4vGyg3F7p8t1VBRUdEVD0xdNeS999579OnT56odLV26tEIhLyAggL1791qmc3Jy8PPzK7fe5s2bL3mN3ebNm+nW7f/+4JvNZhYsWMCIESNwcXGxzP/jz47gbGExGan2OwVlz1+uyGfS7dKvyM3A7qf2wW6n93VqXxzdVUNeYWEhXbp0wTCMK65X0SNf7dq144033iA3NxcPDw82bdrEyy+/XGYdwzA4ePAgLVu2LNd+3759DB8+3DLt7OzMZ599RoMGDejWrRtpaWm0aNECT0/PCtUjInIzK8oK038gRRzUVUPe999/X6Ud+vv7Ex8fz+DBgykuLiY6Oprg4GCGDx9OXFwczZs3Jzc3Fzc3t0sGx+PHj1u+KuWiadOm8cILLzB37lzq1avH9OnTq7RmERERkRtNhW68qGqRkZGWO2MvWrRokeVnHx8fvvzyy0u23b9/f7l5jRs3vqYbLUREREQcjV1CnoiIiNx8OnXqxOHDh+3Sd2BgIFu3brVL3/aikCciIiI2cS0hK/KZdLteP3ojsjrkjRkzhv79+9OuXbvrUY+IiIhUU/b+Oi973ixzI36dl9Uh79FHH2XevHlMmjSJxx9/nD59+lj91AsRERG58djz67zs/T15N+Ld2M7WNujZsyfvv/8+8+bN49dffyU6OpqEhIRLPrVCREREROzD6pAHF76A+D//+Q8//vgjpaWl+Pj4MHHiRObMmVPV9YmIiIhIJVh9unbmzJmsWbOGO+64gyeeeILZs2fj5uZGQUEBHTt2rNBTL0RERETk+rI65OXm5rJo0SKaNGlSZr6npyepqalVVpiIiIiIVJ7Vp2uffvppyxcPHzt2jL/97W/k5OQAEBYWVrXViYiIiEilWB3yEhMTufvuuwGoX78+bdq0YcKECVVemIiIiIhUntUh7/Tp0wwePBgAd3d3hg4dajmSJyIiIiLVg9Uhr7S0FJPJZJk+deoUhmFUaVEiIiIicm2svvFi6NChPPbYY7Rv3x4nJyd27tzJc889dz1qExEREZFKsjrkRUdHExQUxO7du3FxcSE2NpZ77733etQmIiIiIpVkdcgDaNKkSbmvUBERERGR6sPqkPftt9+ycOFCCgoKMAwDs9nMiRMn2L59+3UoTy5yD9rB4ys22LeIH96yS7fuQbUB+zwrUeRmYPdnci47YZdua3u42aVfEVuxOuQlJycTFRXFxo0b6devH1u2bKFLly7Xozb5g6KsMLs9FBrs+2Bou/8BEnFg9vxcgQu/3/auQcRRWR3ynJycGDFiBKdPn+buu+8mMjKSPn36XI/aRERERKSSrP4KlVq1agFw55138q9//YuaNWvi7Gz1ZkRERETkOrL6SF7z5s0ZN24cY8eOZeTIkfz444+4ulbq/g0RERERuU6sTmfPPfcchw4domHDhkyYMIGdO3eSmpp6PWoTkWtwLdcyHvr0Zc79/nMVVlNxNevcSrOIFyrdXhfTi4hcYHXIi4mJIT39wh+Phx9+mIcffriqaxKRa3TNF7JfQ3t73qQjIiL/x+qQ5+HhwS+//EJAQEClO83IyGD+/PmUlJQwZMgQBgwYYFmWnZ1NYmKiZTo3N5c6deqwbt06Pv74Y1JTU/Hx8QEuhMz4+Hh++uknEhIS+PXXX2nYsCEzZsywXDsoIiIicjOyOuQVFhbyyCOPEBAQgKenp2V+RkZGhdqbTCZmzpzJmjVrqFGjBv369aNt27Y0atQIgKZNm1qOFBYWFhITE8PEiRMByMrKIjExkR49epTZ5qRJk3jiiSfo3r07c+fOZd68eSQkJFi7ayIiIiIOw+qQl5SUVG6ek5NThdvv3LmT0NBQvL29AejatSsbNmxg9OjR5dZdsGABrVu3JiQkBIADBw7w448/smDBAgIDA3nhhRfw9PTk66+/Zu7cuQD07t2bgQMHKuSJiIjITc3qkPfyyy9fcn5Fj+SdPHkSX19fy7Sfnx/fffddufXy8vJYuXJlme36+voybNgwHnjgAV5//XUmT57M+PHjqV27tuUOX19fX0wmkzW7JCIiIuJwrA55L7zwf3e9FRcX88knn3DHHXdUuL3ZbC5z5M8wjEseCVy7di2dO3e2XH8HWI7WATz55JM8+uijPPfcc+XaW3Nk8VKysrKuqf31kpmZedP2b+99F+vo/RJraLzcWG7mvwX27t9aVoe8Nm3alJlu164d/fr146mnnqpQ+4CAAPbu3WuZzsnJwc/Pr9x6mzdvZuTIkZbpvLw8Vq9ezdChQ4EL4dDFxYV69eqRl5dHaWkpLi4ul92eNYKCgnB3d7+mbVS5ZSfseseiXe+YtPO+i3V0d61YRb/fNxY7vl92/2yphmO1qKjoigemrvlRFadPn+bkyZMVXr9du3bs2rWL3NxcCgsL2bRpE+Hh4WXWMQyDgwcP0rJlS8s8T09P3nrrLfbv3w/A+++/z6OPPoqbmxshISGsX78egLS0tHLbExEREbnZWH0kLzIyssz0Tz/9RN++fSvc3t/fn/j4eAYPHkxxcTHR0dEEBwczfPhw4uLiaN68Obm5ubi5uZU5mubi4sKsWbOYOHEi586d46677mL69OkAvPTSSyQmJjJ//nxuvfVWXn/9dWt3S0RERMShXNM1eU5OTtSrV4977rnHqm1ERkaWC4uLFi2y/Ozj48OXX35Zrl1ISAgff/xxufn169dn6dKlVtUgIiIi4sisPl175513sn79etq0aYOPjw+pqamcOnXqetQmIiIiIpVkdchLTEzk7rvvBi4cQWvTpg3PP/98lRcmIiIiIpVndcg7ffo0gwcPBsDd3Z2hQ4eSk5NT5YWJiIiISOVZHfJKS0vLfNnwqVOnMAyjSosSERERkWtj9Y0XQ4cO5bHHHqN9+/Y4OTmxc+dOnnvuuetRm4iIiIhUktUhLzo6mqCgIHbv3o2LiwtPPvkkjRs3vh61iYiISDXiHrSDx1dssF8BP7xlt67dg2oDUXbrvzKsDnkmk4kPP/yQiRMncuzYMWbMmMGkSZPKPI9WRERuDp06deLw4cPXtI36H/6tUu0CAwPZunXrNfUt1inKCiMj1T5Bx95PvIh8Jt1ufVeW1SFv/PjxdOrUCfi/u2snTJhQ5nvuRETk5nCtIcvef7hFHJnurhURERFxQLq7VkRERMQBVcndtQkJCdejNvkTu18PsOyEXbqt7eFml35FRERuZNd8d21sbCz33nvv9ahN/sBeF7peFPlMut1rEBERkYqzOuQBNGnShCZNmlBaWsqGDRtISkrio48+qurapArpDjgREZGbS6VC3u+//86KFSv44IMPKCgoYODAgVVdl1Qx3QEnIiJyc7Eq5B07doz33nuPtWvXUr9+fc6dO8fWrVvx8vK6XvWJiIiISCVU+O7aESNGMHDgQNzc3FiyZAnr1q2jVq1aCngiIiIi1VCFQ96hQ4e47777aNy4MQ0aNADAycnpuhUmIiIiIpVX4ZC3fft2evXqxbp16wgLCyMuLo6ioqLrWZuIiIiIVFKFQ56rqyvdunVj6dKlrFmzBj8/P4qKiujSpQvLly+/njWKiIiIiJWsfuIFQKNGjUhOTubzzz8nNjaWlStXVnVdIiIiInINKhXyLvLw8KBv3758/PHHVVWPiIiIiFSBawp5IiIiIlI9KeSJiIiIOKBKPfHiWmVkZDB//nxKSkoYMmQIAwYMsCzLzs4mMTHRMp2bm0udOnVYt24dmZmZTJkyheLiYry9vXn11VepX78+e/bsYcyYMQQEBADQrFkzpkyZYvP9EhEREakubB7yTCYTM2fOZM2aNdSoUYN+/frRtm1bGjVqBEDTpk1JT08HoLCwkJiYGCZOnAhAQkIC8+bNo0mTJqxatYpXXnmF+fPnk5WVxbBhwxg5cqStd0dERESkWrL56dqdO3cSGhqKt7c3np6edO3alQ0bNlxy3QULFtC6dWtCQkI4f/48Y8eOpUmTJsCFh97//PPPABw4cIAdO3YQGRnJqFGjLPNFREREblY2D3knT57E19fXMu3n54fJZCq3Xl5eHitXrmT06NEA1KhRg6ioKADMZjP/+Mc/6Ny5MwBeXl4MGjSIjIwMOnToQHx8vA32RERERKT6svnpWrPZXOZxaIZhXPLxaGvXrqVz5874+PiUmX/+/HkSExMpKSmxnJ6dPHmyZXn//v1JTU0lLy+v0s/VzcrKqlQ7R5eZmWnvEuQGobEi1tB4ubHY8/2y91ixd//WsnnICwgIYO/evZbpnJwc/Pz8yq23efPmctfY5efn89RTT+Ht7c38+fNxc3PDbDazYMECRowYgYuLi2XdP/5sraCgINzd3Svd3hFlZmUZsPUAABthSURBVGbSqlUre5chNwCNFbGGxssNZtkJu71fdh8rdtz3yykqKrrigSmbn65t164du3btIjc3l8LCQjZt2kR4eHiZdQzD4ODBg7Rs2bLM/ISEBBo0aMCsWbOoUaMGAM7Oznz22Wds3LgRgLS0NFq0aIGnp6dtdkhERESkGrL5kTx/f3/i4+MZPHgwxcXFREdHExwczPDhw4mLi6N58+bk5ubi5uZW5mjaoUOH2LJlC40aNaJXr17Ahev5Fi1axLRp03jhhReYO3cu9erVY/r06bbeLREREZFqxS7fkxcZGUlkZGSZeYsWLbL87OPjw5dffllmebNmzTh8+PAlt9e4cWM+/PDDqi9URERE5AalJ16IiIiIOCCFPBEREREHpJAnIiIi4oAU8kREREQckEKeiIiIiANSyBMRERFxQAp5IiIiIg5IIU9ERETEASnkiYiIiDgghTwRERERB6SQJyIiIuKAFPJEREREHJBCnoiIiIgDUsgTERERcUAKeSIiIiIOSCFPRERExAEp5ImIiIg4IIU8EREREQekkCciIiLigBTyRERERByQQp6IiIiIA1LIExEREXFArvboNCMjg/nz51NSUsKQIUMYMGCAZVl2djaJiYmW6dzcXOrUqcO6dev46aefSEhI4Ndff6Vhw4bMmDGDWrVqcebMGZ599lmOHz9OvXr1mDVrFr6+vvbYNREREZFqweZH8kwmEzNnzmTZsmWkpaWxYsUKfvjhB8vypk2bkp6eTnp6Oh9++CF16tRh4sSJAEyaNIknnniCDRs2EBQUxLx58wCYNWsWISEhfPrpp8TExJCSkmLr3RIRERGpVmx+JG/nzp2Ehobi7e0NQNeuXdmwYQOjR48ut+6CBQto3bo1ISEhFBcX8/XXXzN37lwAevfuzcCBA0lISGD79u188MEHAPTo0YPJkydTXFyMm5ub7XZMRETkJhD5TLr9Ol92wm5d1/a48TKFzUPeyZMny5xK9fPz47vvviu3Xl5eHitXriQjIwOA06dPU7t2bVxdL5Ts6+uLyWQqt01XV1dq165Nbm4u/v7+13t3REREbhoZqVF26zvymXS79n8jsnnIM5vNODk5WaYNwygzfdHatWvp3LkzPj4+l13vUu0uruvsXPkz0VlZWZVu68gyMzPtXYLcIDRWxBoaL1JRGivWsXnICwgIYO/evZbpnJwc/Pz8yq23efNmRo4caZmuV68eeXl5lJaW4uLiUqadn58fp06dIiAggJKSEvLz8y2ngysjKCgId3f3Srd3RJmZmbRq1creZcgNQGNFrKHxIhW27ITGyp8UFRVd8cCUzW+8aNeuHbt27SI3N5fCwkI2bdpEeHh4mXUMw+DgwYO0bNnSMs/NzY2QkBDWr18PQFpamqVdhw4dSEtLA2D9+vWEhIToejwRERG5qdk85Pn7+xMfH8/gwYN57LHH6NGjB8HBwQwfPpwDBw4AF742xc3NrdzRtJdeeomVK1fSrVs39u7dy7hx4wAYO3Ys+/bto3v37ixbtowXX3zR1rslIiIiUq3Y5XvyIiMjiYyMLDNv0aJFlp99fHz48ssvy7WrX78+S5cuLTff29ubN998s+oLFREREblB6YkXIiIiIg5IIU9ERETEASnkiYiIiDgghTwRERERB6SQJyIiIuKAFPJEREREHJBCnoiIiIgDUsgTERERcUAKeSIiIiIOSCFPRERExAEp5ImIiIg4IIU8EREREQekkCciVWL58uUEBQXRpk0bgoKCWL58ub1LEhG5qbnauwARufEtX76cpKQkFi9ejKenJwUFBcTGxgLQv39/O1cnInJz0pE8EblmKSkpLF68mI4dO+Lq6krHjh1ZvHgxKSkp9i5NROSmpZAnItcsOzubsLCwMvPCwsLIzs62U0UiIqKQJyLXrGnTpuzYsaPMvB07dtC0aVM7VSQiIgp5InLNkpKSiI2NZdu2bZSUlLBt2zZiY2NJSkqyd2kiIjct3XghItfs4s0VY8aMITs7m6ZNm5KSkqKbLkRE7EghT0SqRP/+/enfvz+ZmZm0atXK3uWIiNz0dLpWRERExAEp5ImIiIg4IIU8EREREQdkl2vyMjIymD9/PiUlJQwZMoQBAwaUWX7s2DFeeuklfv/9d3x9fXn99dcpKSlh2LBhlnXy8vI4ffo03377LXv27GHMmDEEBAQA0KxZM6ZMmWLTfRIRERGpTmwe8kwmEzNnzmTNmjXUqFGDfv360bZtWxo1agSAYRg89dRTJCUlER4ezowZM1i4cCEJCQmkp6cDYDabGTJkCPHx8QBkZWUxbNgwRo4caevdEREREamWbH66dufOnYSGhuLt7Y2npyddu3Zlw4YNluUHDx7E09OT8PBwAEaNGlXuSN/q1avx8PAgMjISgAMHDrBjxw4iIyMZNWoUP//8s+12SERERKQasnnIO3nyJL6+vpZpPz8/TCaTZfq///0vf/nLX5gwYQK9evXipZdewtPT07K8tLSUN998k2eeecYyz8vLi0GDBpGRkUGHDh0sR/hEREREblY2P11rNptxcnKyTBuGUWa6pKSEPXv28P7779O8eXNmzZrF1KlTmTp1KgBffPEFd911F4GBgZY2kydPtvzcv39/UlNTycvLw8vLq1I1ZmVlVaqdo8vMzLR3CXKD0FgRa2i8SEVprFjH5iEvICCAvXv3WqZzcnLw8/OzTPv6+tKgQQOaN28OQI8ePYiLi7Ms37x5M926dbNMm81mFixYwIgRI3BxcbHM/+PP1goKCsLd3b3S7R2RvuBWKkpjRayh8SIVtuyExsqfFBUVXfHAlM1P17Zr145du3aRm5tLYWEhmzZtslx/B9CyZUtyc3P5/vvvAdi6dSv33XefZfm+ffsICQmxTDs7O/PZZ5+xceNGANLS0mjRokWZU7wiIiIiNxubH8nz9/cnPj6ewYMHU1xcTHR0NMHBwQwfPpy4uDiaN2/O3LlzSU5OprCwkICAAKZPn25pf/z4cctXpVw0bdo0XnjhBebOnUu9evXKrC8iIiJyM7LL9+RFRkZa7oy9aNGiRZafW7RowapVqy7Zdv/+/eXmNW7cmA8//LBqixQRERG5gemJFyIiIiIOSCFPRERExAEp5ImIiIg4IIU8EREREQekkCciIiLigOxyd62IiIjcfDp16sThw4cr3b7+h3+rdNvAwEC2bt1a6fY3IoU8ERERsYlrCVl6Oor1dLpWRERExAEp5ImIiIg4IIU8EREREQekkCciIiLigBTyRERERByQQp6IiIiIA1LIExEREXFACnkiIiIiDkghT0RERMQBKeSJiIiIOCA91uwPDMMA4Pz583aupHoqKiqydwlyg9BYEWtovEhFaayUdTGvXMwvf+ZkXG7JTSgvL48jR47YuwwRERGRCrv33nvx8vIqN18h7w/MZjP5+fm4ubnh5ORk73JERERELsswDIqLi6lVqxbOzuWvwFPIExEREXFAuvFCRERExAEp5ImIiIg4IIU8EREREQekkCciIiLigBTyRERERByQQp6IiIiIA1LIExEREXFACnkiIiIiDkghT0RERMQBKeSJiIiIOCCFPBv56quvGDRoUJVsKyoq6orL/9jP1db9o06dOtGtWzeioqKIioqiU6dOxMXFUVBQUOlaq5LJZGL48OH2LsPuTpw4QVBQkOV9ioyMpFOnTsyZM6dKtr9mzRoSExOrZFt/3GabNm0sNUdFRREbG1ulfVzJypUrWbdunc36q06++uorWrZsSVRUFD179iQiIoL33nvPsnz48OGYTCab1ZOWlkafPn0sY3fJkiWWZZ06deLEiRNV0s/y5ctZvnw5cOH9b9++PdOmTbP5/srVbdiwgd69e9OzZ08iIyN56623+Oijjy75GfH888+zZMkS1qxZQ2BgYLnf63fffZfAwMAqG0c3Old7FyDWS09Pv+LyPXv2VHjdP1u4cCG33347AOfPn+eJJ54gLS2NJ554wvpCq5i/vz+LFi2ydxnVgp+fX5n31mQy0bVrV7p3784999xjx8our1OnTkydOtUufX/zzTe0adPGLn1XB0FBQSxduhSAs2fP0r17dx566CEaNWpk09+pFStW8OGHH7JgwQL8/Pw4c+YMw4YNw8PDg5iYmCrtq3///paf161bx5QpUwgLC6vSPuTamUwmpk2bxpo1a6hbty75+fkMGjSIv/3tb+zbt49ff/0VHx8fAAoLC9m2bRvPPfcc27ZtIyAggI0bN9KjRw/L9j777DNuueUWe+1OtaOQVw28+eabrF27FhcXFx566CESEhJwcXFhyZIlvP/++3h5eXH33Xdz5513MmbMGAIDAzl8+DC7du3itddeA6BOnTqkpqYyb948AGJiYvjoo48s6/72228kJSVx7NgxatSoQWJiIg8++OAV68rLyyMvLw9vb28APv/8c+bMmUNJSQm33347L7/8MnXr1uWrr77ilVdewcXFhfvvv5+jR4+ydOlSBg0aRJ06dfjXv/7FrFmzyMnJuWT7adOm8eWXX+Ls7Eznzp0ZPXr0JfetoKCAwYMHs3XrVk6dOkVSUhI//fQTrq6uxMfHEx4ezhtvvIHJZOI///kP//vf/4iJieGpp566ju9e9ZCTk4NhGNSqVYuSkhImTpzIv/71L06dOkVgYCCvv/46p06dYvTo0TRu3Jjs7Gx8fHyYPXs23t7epKWlMX/+fGrXrk39+vXx9PQEYN++faSkpFBUVETdunWZPHkyDRo0YNCgQTRr1ozMzEyKiop49tlnWbJkCUePHmXo0KEMHTq0wrVfqY8/jp//9//+H0FBQeTk5LBq1SreeecdPv30U0pLSwkLCyMhIYH8/Hz+/ve/c+rUKQCefvppPDw82Lp1K7t378bX15f27dtfj7fghlFUVISLiwteXl7AhfC9ZMkS9uzZwxdffMHvv//O8ePHeeihh5g4ceIVx9OTTz5J3bp1qVmzJkVFRTz99NM89NBDGIZB165dWbp0Kf7+/pa+58+fz6uvvoqfnx8At9xyC9OmTePs2bNlajx79iwTJkzAZDJx8uRJHnzwQVJSUjCZTDz77LMUFBTg7OxMcnIy999//yU/Q9544w0AnJycOHDgAJMmTSI5OZlJkyaxZMkSbr31VqZPn86ePXsoLS2ld+/eDB06lK+++orXXnsNs9lM48aNmTZtmo3emZvT6dOnKS4u5ty5cwDUqlWLqVOn4u7uTufOnVm/fr3l7NTmzZsJDQ2lbt26ALRu3ZrMzEwKCgrw9PTkp59+olatWpaxLYAhNrF7925j4MCB5eZv377diImJMQoKCozi4mJj1KhRxvvvv29kZ2cbXbp0MfLy8oxz584ZMTExxpw5cwzDMIx7773XMAzDGDhwoLF//37DMAxj4cKFxhdffFFm+R9/njhxojF16lTDMAzj+++/Nx5//PFytXTs2NGIiIgwevToYTz44INGr169jCVLlhhms9n49ddfjZ49exq//fabYRiGsXz5cmPChAnG+fPnjfDwcCM7O9swDMN4+eWXLfs5cOBAS82Xa3/ixAmjW7duhmEYRkFBgTF27Fjj3Llzl9y348ePGx07djQMwzDi4uKMt99+2zAMw/jvf/9rPPTQQ0ZOTo4xZ84cIzo62igqKjJOnTpl3H///cbvv/9u7dtVrR0/fty47777jJ49expdu3Y12rRpY8TGxhqff/65YRiGsWfPHmPixImGYRhGaWmpMXDgQGPDhg3G8ePHjcDAQOPgwYOGYRjG6NGjjSVLlhi//PKL5fUrLi42hg0bZowfP94oKioyOnbsaHkf1q9fb/Tu3dswjAvvbUpKimEYhvHGG28YnTt3NgoKCowTJ04YISEh5WpevXq10bp1a6Nnz56Wf7t27bpqHxfHj2FcGMu7d+82DMMw/vnPfxpjxowxSkpKjNLSUuPvf/+7kZaWZqxZs8ay74cOHbKM+fHjxxurV6+uwnfhxrF7927j/vvvN3r27Gn06NHDCA4ONsaPH2+YzWbDMC783h8/ftxYvXq10aFDByMvL88oKCgwwsPDje+///6K4+nee+81jh8/bhiGYaxatcpISEgwDOPCGIyNjS1Tx6+//mrce++9xunTpy9b68VaMjIyjHnz5hmGYRhFRUVG586djQMHDhhvvPGGsWjRIsMwLoyBt95667KfIXPmzLGMn4EDB1rGzsU+li1bZrz66quWPgYOHGh8/fXXxu7du41WrVoZZ86cufYXXyrkxRdfNJo1a2b06dPHmD59uuXvyVdffVXmb9WwYcMsf+dWr15tjB8/3khOTjbWr19vGIZhLFq0yFizZo3lPRbD0JE8O9u9ezfdu3fHw8MDgD59+pCWlsb58+fp2LEjtWvXBqB79+6cOXOmTNtHHnmE0aNH07lzZx555BEeeuihy/bz9ddfM2PGDAACAwNZsWLFJde7eLp248aNTJ06lb/+9a84OTmxf/9+fv75ZwYPHgyA2WymTp06HDlyBB8fH5o0aQJAdHQ0KSkplu0FBwcDXLa9v78/7u7u9OvXj44dO/Lss8/i7u5+yX374zUWu3fv5pVXXgHgjjvuoEWLFuzfvx+Atm3bUqNGDXx8fPD29iYvL8/hDt9fPF1rNpuZOnUqR48etbz/rVu3xtvbmw8++IBjx47x448/Wq6r9PHxoVmzZgA0btyY33//nW+//ZaWLVvyl7/8BYDIyEh2797Njz/+yC233GJ5DyMiInjxxRfJy8sDIDw8HIDbbruNFi1a4OHhQf369cuN04sudbr2yJEjV+zj4vyLWrRoAcCuXbv47rvv6N27NwDnzp3jtttuo0+fPrz++uuYTCYefvhhnn766cq+xA7lz6drn3zySRYuXMjIkSPLrNeyZUvLZ84dd9zB77//Tps2ba44ni5e3hEREcHMmTMpKCjg448/trw3Fzk7X7gE3N3d/ar19ujRg++++453332XY8eO8dtvv1FQUMCDDz7ImDFjyM7OpkOHDgwcOBAXF5dLfoZcza5du8jOzmb37t0AFBQUcPjwYRo1akTDhg11NMiGJk2axN/+9jd27NjBjh07ePzxx5kxYwaPPvoop0+f5vjx49SsWZMff/yRdu3alWkbERHBypUriYiIYPPmzSxatMhyFFd044Xdmc3mcvNKSkpwdna+5LI/Gjp0KEuXLuXOO+/ktddeY/78+Zdd19XVFScnJ8v00aNHr7j9rl270r59eyZMmABAaWkpDzzwAOnp6aSnp7Nq1SrmzJmDi4vLFbdTs2bNK7Z3dXXlo48+YuzYsfz222/069ePf//731fdN8Mwyk2XlpYCZf+IODk5lVvXkTg7O/Pcc89hMplYvHgxAFu2bOHZZ5+lZs2a9O7dm9atW1teg0u9Nn9+jVxdL/zf71Lv6x9fZzc3t3JtrHW1Pi6On4v+OJ6GDBliGU8fffQRo0aN4q677uLTTz8lMjKSvXv3Eh0dfdXfo5tN7dq1iYiI4Jtvvim37FLj40rj6Y/vj6enJ+Hh4WzcuJHdu3fzyCOPlNm2t7c3d9xxB1lZWWXm79mzx/If0IuWLl3K9OnTqVevHgMHDuSee+7BMAxatWrFJ598QlhYGOvXr2fUqFGX/Qy5mtLSUhISEixjaMWKFURHR5fbL7m+tm/fzvr16/H396dPnz7MnDmT5ORkVq1ahZOTE4899hjr1q1j3bp1REVFWf6zcFHbtm05cOAAR44coW7dugrnf6KQZ2ehoaF88sknnDt3jpKSElavXk1oaCgPPvgg//znPzl79iznz59n06ZNZUIaXLjuLj8/33IN1KFDhwBwcXGhpKSkzLohISF88sknwIWAN3z48HLb+7OxY8eSmZnJ9u3badGiBfv27bN8eM6bN4/p06dz9913c+bMGQ4fPgxARkbGJbd1ufaHDh1i4MCBtG7dmvHjx3PPPffw73//+7L79sfXbdWqVQAcP36cb775hvvvv/+qr7cjcnV15bnnnmPevHnk5OSwa9cuIiIi6NOnD7fccgtfffWVJTRdSqtWrdi3bx8mkwmz2cz69esBuPvuu/ntt9/47rvvAFi/fj233Xab5RrNqlDZPkJDQ0lPTyc/P5+SkhKefvppNm7cyPvvv88bb7xBREQEL730Erm5uZw9exYXF5crvgY3k9LSUvbs2WM5ons11oyni3+k27dvf8mjabGxsUydOpWcnBwAcnNzmTp1Kg0aNCiz3pdffknfvn3p2bMnRUVFfP/995jNZqZPn87atWvp1asXL774IocOHbrsZ8jVhIaGsnLlSoqLi8nPz+eJJ55g3759FXpNpOrUrFmT1NRUy5kawzDIzs6madOmAPTq1YvPPvvMcgfun128lv3FF1+kW7duNq39RqDTtTa0d+9eWrZsaZmOjIxk8uTJZGdn06dPH0pKSggLC2PgwIG4uroyePBg+vbti6enJ3Xr1i33ofn3v/+dxMREXF1d8fT0tJy+fOSRR4iKimLNmjWWdePi4khOTqZnz564uroyffr0q4Y8Hx8fhg8fbvlgffXVVxk3bhxmsxl/f39ee+01atSowfTp0xk/fjzOzs40bNjwkv8L9vX1vWT7unXrcv/999OjRw88PDx44IEHCA8Px8PD45L7dlFSUhIvvviiZR9feeUVy8XcN6Pw8HBatmzJ7NmzGTRoEM8++yyffPIJbm5uPPDAA1f8OoG//OUvJCcnM3ToUDw8PGjUqBEANWrUYObMmbz88ssUFhZSp04dZs6cWaV1V7aPTp068f333/P4449TWlpK+/bt6dWrl+XGi8jISFxcXEhISOCWW26hXbt2vP7663h5efHXv/61SvfhRpCVlUVUVBROTk6UlJQQGBhY4a8jiomJqfB4atWqFU5OTvTp0+eSy/v3709JSQnDhg2zHCns27dvuTtrhwwZwsSJE1m4cCG1a9emZcuWnDhxgkGDBvHMM8+wZs0aXFxcmDZtGs2aNbvkZ8jBgwevuF/9+vXjP//5D7169aKkpITevXvTtm1bvvrqqwq9LlI1QkNDGT16NKNGjaK4uBiA9u3bWy61uPXWW6lbty5ms9lyacCfRUREkJ6eTqdOnWxW943CyXDkc1k3sH//+9/885//tNyl+NRTTxETE1PtBrHZbGbGjBmMHj0aT09P3nnnHUwmU5V/z5qIVG+GYXDkyBHGjx9PWlqavcsREXQkr9qqX78+Bw4coEePHjg5OREWFkbHjh3tXVY5zs7OeHt7Ex0djZubG/Xr1y9z44WI3Bzee+893nrrLWbPnm3vUkTk/6cjeSIiIiIOSDdeiIiIiDgghTwRERERB6SQJyIiIuKAFPJEREREHJBCnoiIiIgDUsgTERERcUD/H+PyGV0N2bDdAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 720x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "pd.DataFrame({\n",
    "    'Logistic Regression': Log_Score_Grid,\n",
    "    'Random Forrest': Random_Forrest_grid_Score,\n",
    "    'Binary Classifier': Binary_score_grid,\n",
    "    'SVM': M_list,\n",
    "}).plot.box(\n",
    "    xlabel='Regressor',\n",
    "    ylabel=r'Accuracy $[\\mathtt{USD}]$',\n",
    "    figsize=(10, 5),\n",
    ");"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Analyzing the general results of accuracy of each model, it is possible to understand that all of them had a similar performance. But analyzing the details it is possible to conclude that Random Forrest had the best general accuracy. Moreover, it is intersting to point out that a specific outlier of Logistic Regression, Random Forrest and Binary Classifier had aproximalary the same accuracy. This may have been due to the fact that all of them had the same cut of the cros val and that section of the database impact the best it could to predict if semeone had or not diabetes."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy = 0.7077922077922078\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import accuracy_score\n",
    "\n",
    "final_model = forest_clf_grid.best_estimator_\n",
    "\n",
    "X_test = strat_test_set.drop('Outcome', axis=1)\n",
    "y_test = strat_test_set['Outcome'].copy()\n",
    "\n",
    "X_test_prepared = imputer.transform(X_test)\n",
    "X_test_prepared = scaler.transform(X_test_prepared)\n",
    "final_predictions = final_model.predict(X_test_prepared)\n",
    "\n",
    "final_accuracy = accuracy_score(y_test, final_predictions)\n",
    "\n",
    "print(f'Accuracy = {final_accuracy}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    0.61039\n",
       "1    0.38961\n",
       "Name: Outcome, dtype: float64"
      ]
     },
     "execution_count": 66,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_test.value_counts(True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**In Conclusion**, the Random Forrest Classifier perfomed a marvelous job, since it had an accuracy of 70,77% in the test set. This shows that not only it doesn't overfit or underfit but also acts much better then a model which just guess that everybody doesn't have diabetes. This is because in the cell above, it is learned that there were 61,03% of people who didn't have diabetes. Therefore a model that only guess, would score an 61,03% of accuracy and the Random Forrest one was 10% better."
   ]
  },
  {
   "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": 5
}