classification.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Classification. Linear models and KNN"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-02-26T07:19:48.446164Z",
     "start_time": "2021-02-26T07:19:46.392464Z"
    }
   },
   "outputs": [],
   "source": [
    "import os\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-02-26T07:20:07.052141Z",
     "start_time": "2021-02-26T07:20:06.963694Z"
    }
   },
   "outputs": [],
   "source": [
    "from sklearn.pipeline import Pipeline\n",
    "from sklearn.compose import ColumnTransformer\n",
    "from sklearn.model_selection import train_test_split, cross_validate\n",
    "from sklearn.metrics import accuracy_score#, plot_confusion_matrix\n",
    "from sklearn.linear_model import SGDClassifier, LogisticRegression\n",
    "from sklearn.preprocessing import StandardScaler, OneHotEncoder\n",
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "from sklearn.model_selection import cross_val_score\n",
    "from sklearn.pipeline import Pipeline, make_pipeline\n",
    "from sklearn.preprocessing import StandardScaler, MinMaxScaler, OrdinalEncoder, OneHotEncoder\n",
    "from sklearn.model_selection import GridSearchCV\n",
    "from sklearn.preprocessing import PolynomialFeatures\n",
    "from sklearn.model_selection import RandomizedSearchCV"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Part 1: Titanic survival prediction"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Dataset\n",
    "\n",
    "Read the description here: https://www.kaggle.com/c/titanic/data. Download the dataset and place it in the *data/titanic/* folder in your working directory.\n",
    "You will use train.csv for model training and validation. The test set is used for model testing: once the model is trained, you can predict whether a passenger survived or not for each passenger in the test set, and submit the predictions: https://www.kaggle.com/c/titanic/overview/evaluation.  \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-02-26T07:20:11.392848Z",
     "start_time": "2021-02-26T07:20:11.390422Z"
    }
   },
   "outputs": [],
   "source": [
    "PATH = \"./data/\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-02-26T07:20:13.713016Z",
     "start_time": "2021-02-26T07:20:13.702271Z"
    }
   },
   "outputs": [],
   "source": [
    "data = pd.read_csv(os.path.join(PATH, 'train.csv')).set_index('PassengerId')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-02-26T07:20:14.659217Z",
     "start_time": "2021-02-26T07:20:14.642817Z"
    },
    "scrolled": true
   },
   "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>Survived</th>\n",
       "      <th>Pclass</th>\n",
       "      <th>Name</th>\n",
       "      <th>Sex</th>\n",
       "      <th>Age</th>\n",
       "      <th>SibSp</th>\n",
       "      <th>Parch</th>\n",
       "      <th>Ticket</th>\n",
       "      <th>Fare</th>\n",
       "      <th>Cabin</th>\n",
       "      <th>Embarked</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PassengerId</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>Braund, Mr. Owen Harris</td>\n",
       "      <td>male</td>\n",
       "      <td>22.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>A/5 21171</td>\n",
       "      <td>7.2500</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>Cumings, Mrs. John Bradley (Florence Briggs Th...</td>\n",
       "      <td>female</td>\n",
       "      <td>38.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>PC 17599</td>\n",
       "      <td>71.2833</td>\n",
       "      <td>C85</td>\n",
       "      <td>C</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>Heikkinen, Miss. Laina</td>\n",
       "      <td>female</td>\n",
       "      <td>26.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>STON/O2. 3101282</td>\n",
       "      <td>7.9250</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>Futrelle, Mrs. Jacques Heath (Lily May Peel)</td>\n",
       "      <td>female</td>\n",
       "      <td>35.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>113803</td>\n",
       "      <td>53.1000</td>\n",
       "      <td>C123</td>\n",
       "      <td>S</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>Allen, Mr. William Henry</td>\n",
       "      <td>male</td>\n",
       "      <td>35.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>373450</td>\n",
       "      <td>8.0500</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             Survived  Pclass  \\\n",
       "PassengerId                     \n",
       "1                   0       3   \n",
       "2                   1       1   \n",
       "3                   1       3   \n",
       "4                   1       1   \n",
       "5                   0       3   \n",
       "\n",
       "                                                          Name     Sex   Age  \\\n",
       "PassengerId                                                                    \n",
       "1                                      Braund, Mr. Owen Harris    male  22.0   \n",
       "2            Cumings, Mrs. John Bradley (Florence Briggs Th...  female  38.0   \n",
       "3                                       Heikkinen, Miss. Laina  female  26.0   \n",
       "4                 Futrelle, Mrs. Jacques Heath (Lily May Peel)  female  35.0   \n",
       "5                                     Allen, Mr. William Henry    male  35.0   \n",
       "\n",
       "             SibSp  Parch            Ticket     Fare Cabin Embarked  \n",
       "PassengerId                                                          \n",
       "1                1      0         A/5 21171   7.2500   NaN        S  \n",
       "2                1      0          PC 17599  71.2833   C85        C  \n",
       "3                0      0  STON/O2. 3101282   7.9250   NaN        S  \n",
       "4                1      0            113803  53.1000  C123        S  \n",
       "5                0      0            373450   8.0500   NaN        S  "
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### EDA"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**(0.5 points)** How many females and males are there in the dataset? What about the survived passengers? Is there any relationship between the gender and the survival?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-02-26T07:20:17.101881Z",
     "start_time": "2021-02-26T07:20:17.088146Z"
    }
   },
   "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>cnt</th>\n",
       "      <th>%</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Sex</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>male</th>\n",
       "      <td>577</td>\n",
       "      <td>64.76</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>female</th>\n",
       "      <td>314</td>\n",
       "      <td>35.24</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        cnt      %\n",
       "Sex               \n",
       "male    577  64.76\n",
       "female  314  35.24"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "temp = data['Sex'].value_counts().rename('cnt').to_frame()\n",
    "temp['%'] = (data['Sex'].value_counts(normalize=True) * 100).round(2)\n",
    "temp.index.name = 'Sex'\n",
    "temp"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-02-26T07:20:17.777639Z",
     "start_time": "2021-02-26T07:20:17.767662Z"
    }
   },
   "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>cnt</th>\n",
       "      <th>%</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Survived</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>549</td>\n",
       "      <td>61.62</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>342</td>\n",
       "      <td>38.38</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          cnt      %\n",
       "Survived            \n",
       "0         549  61.62\n",
       "1         342  38.38"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "temp = data['Survived'].value_counts().rename('cnt').to_frame()\n",
    "temp['%'] = (data['Survived'].value_counts(normalize=True) * 100).round(2)\n",
    "temp.index.name = 'Survived'\n",
    "temp"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-02-26T07:20:20.176341Z",
     "start_time": "2021-02-26T07:20:20.023214Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.countplot(data=data, x='Sex', hue='Survived');"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "По грфику видно, что женщин и в абсолютных и в относительных числах выжило больше, чем мужчин."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**(0.5 points)** Plot age distribution of the passengers. What is the average and the median age of survived and deceased passengers? Do age distributions differ for survived and deceased passengers? Why?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-02-24T17:31:08.956507Z",
     "start_time": "2021-02-24T17:31:08.359213Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1224x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(17,4))\n",
    "sns.histplot(data=data, x='Age', kde=True);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-02-26T07:20:25.434241Z",
     "start_time": "2021-02-26T07:20:25.424113Z"
    }
   },
   "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>mean</th>\n",
       "      <th>median</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Survived</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>30.626179</td>\n",
       "      <td>28.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>28.343690</td>\n",
       "      <td>28.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               mean  median\n",
       "Survived                   \n",
       "0         30.626179    28.0\n",
       "1         28.343690    28.0"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.groupby('Survived')['Age'].agg(['mean', 'median'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-02-24T17:31:09.889616Z",
     "start_time": "2021-02-24T17:31:09.441476Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA+QAAAEGCAYAAADsTl+xAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8vihELAAAACXBIWXMAAAsTAAALEwEAmpwYAABqz0lEQVR4nO3dd5hkZZ328e9TOXTOuXtyzgGGOAwZFASRICIqLuYFWddVNyi7q4uuL+q6a0AxI0GCICh5yDAwOeeZzjnnrvC8f1TRMwMDDDPdXR3uz3XVVadOnar6FYeprrueZKy1iIiIiIiIiMjIciS6ABEREREREZGJSIFcREREREREJAEUyEVEREREREQSQIFcREREREREJAEUyEVEREREREQSwJXoAo5FVlaWLSsrS3QZIiIiIiIiIu/LunXrmqy12Ue7b0wE8rKyMtauXZvoMkRERERERETeF2NM+Tvdpy7rIiIiIiIiIgmgQC4iIiIiIiKSAArkIiIiIiIiIgkwJsaQi4iIiIiIyPgUCoWoqqqir68v0aWcEJ/PR1FREW63+5gfo0AuIiIiIiIiCVNVVUVycjJlZWUYYxJdznGx1tLc3ExVVRWTJk065sepy7qIiIiIiIgkTF9fH5mZmWM2jAMYY8jMzHzfrfwK5CIiIiIiIpJQYzmMv+l43oMCuYiIiIiIiEgCKJCLiIiIiIjIqPLtb3+bOXPmMH/+fBYuXMiaNWtO+DkfeeQRbrvttiGoDpKSkobkeTSpm4iMGcUlpVRVViS6jGNSVFxCZUV5ossQERERGXNeffVVHn30UdavX4/X66WpqYmBgYFjemw4HMblOnrMveSSS7jkkkuGstQTpkAuImNGVWUFtz+5K9FlHJNbzpuR6BJERERExqTa2lqysrLwer0AZGVlAVBWVsbatWvJyspi7dq1fOUrX+G5557jW9/6Fvv27WP//v2UlJRw4MAB7rzzTubMmQPAypUr+f73v8/WrVtZu3Yt3/72t5k/fz4HDhzA4XDQ3d3NzJkz2b9/PxUVFXzhC1+gsbGRQCDAL37xC2bOnMmBAwf46Ec/SldXF5deeumQvVd1WRcREREREZFR47zzzqOyspLp06fz+c9/nueff/49H7N9+3aefvpp7r77bq666iruu+8+IBbua2trWbp06eCxqampLFy4cPB5H330Uc4//3zcbjc33ngjP/7xj1m3bh3f//73+fznPw/ATTfdxOc+9zm2bNlCfn7+kL3XYQ3kxpg0Y8z9xpidxpgdxpgVxpgMY8xTxpg98ev04axBRERERERExo6kpCTWrVvHHXfcQXZ2NldddRW/+c1v3vUxl1xyCX6/H4Arr7yS+++/H4D77ruPK6644m3HX3XVVdx7770A3HPPPVx11VV0dXXxyiuv8JGPfISFCxfymc98htraWgBefvllrrnmGgCuu+66oXqrw95l/UfA49baK4wxHiAAfAN4xlp7mzHma8DXgH8a5jpERERERERkjHA6naxcuZKVK1cyb948fvvb3+JyuYhGowBvW+87GAwObhcWFpKZmcnmzZu59957+dnPfva257/kkkv4xje+QUtLC+vWrWPVqlV0d3eTlpbGxo0bj1rTcCzNNmwt5MaYVOAM4E4Aa+2AtbYNuBT4bfyw3wIfGq4aREREREREZGzZtWsXe/bsGby9ceNGSktLKSsrY926dQA88MAD7/ocV111Fd/73vdob29n/vz5b7s/KSmJZcuWcdNNN/GBD3wAp9NJSkoKkyZN4k9/+hMA1lo2bdoEwKmnnso999wDwF133TUk7xOGt8v6JKAR+LUxZoMx5pfGmCCQa62tjR9TB+Qe7cHGmBuNMWuNMWsbGxuHsUwREREREREZLbq6urj++uuZPXs28+fPZ/v27XzrW9/im9/8JjfddBNLly7F6XS+63NcccUV3HPPPVx55ZXveMxVV13FH/7wB6666qrBfXfddRd33nknCxYsYM6cOTz88MMA/OhHP+L//u//mDdvHtXV1UPzRgFjrR2yJzviiY1ZCrwGnGqtXWOM+RHQAXzJWpt22HGt1tp3HUe+dOlSu3bt2mGpU0TGDmPMmJplfbg+X0VERETGkx07djBr1qxElzEkjvZejDHrrLVLj3b8cLaQVwFV1to3V3C/H1gM1Btj8uOF5QMNw1iDiIiIiIiIyKg0bIHcWlsHVBpj3lyM92xgO/AIcH183/XAw8NVg4iIiIiIiMhoNdyzrH8JuCs+w/p+4JPEfgS4zxhzA1AOvHOnfhEREREREZFxalgDubV2I3C0vvJnD+frioiIiIiIiIx2wzmGXERERERERETegQK5iIiIiIiISAIokIuIiIiIiMioVVxSijFmyC7FJaXH9LqPP/44M2bMYOrUqdx2223D8t6Ge1I3ERERERERkeNWVVnB7U/uGrLnu+W8Ge95TCQS4Qtf+AJPPfUURUVFLFu2jEsuuYTZs2cPWR2gFnIRERERERGRI7z++utMnTqVyZMn4/F4uPrqq3n44aFfsVuBXEREREREROQw1dXVFBcXD94uKiqiurp6yF9HgVxEREREREQkARTIRURERERERA5TWFhIZWXl4O2qqioKCwuH/HUUyEVEREREREQOs2zZMvbs2cOBAwcYGBjgnnvu4ZJLLhny19Es6yIiIiIiIjJqFRWXHNPM6O/n+d6Ly+Xif//3fzn//POJRCJ86lOfYs6cOUNWw+DrDPkzioiIiIiIiAyRyoryhLzuRRddxEUXXTSsr6FALiLyLiJRS2vPAJ19Ybr7w/QMROjuD9MXjhCJWqIWolFL1FpcTgdup8HjdJB21g2knnwF4Y5GIh1NhDsbiXS1QDSS6Lf0NkXFJQn7QyciIiIykSmQi4jERa2lsbOfqtZeGrv6ae7qp6V7gKg98jify4HP48RpDA6HwWHAYQx9oTADkSgD4SjJiy7E4fYd8TiHgfSAh8wkD5lJXrKCHnJTfAS9if0oHsouYCIiIiJy7BTIRWRC6+oPs7+xi8qWXipbe+gPRwFI8rrITPJQmhkkK8lDmt9DwOsk4HHicrz3fJi3nDeD2/66g86+EF39YTr7wrT3hmjuHqCuvY/d9V2Dx6b63eSn+shP9VGUHiA94MYYM2zvWURERERGBwVyEZlw+sMR9jZ0sauuk8rWXiAWwKdkJ1Gc4ac4PTAkrdYel4PMJC+ZSd6j1tDcFQvnNe29VLT0sLOuc7CW0swApRkBSjICeN3OE65FREREREYfBXIRmTAaOvpYX9HG3sYuIlFLqt/N8kkZzMhNHvFWaa/LSUGan4I0P4tJx1pLe2+IytZeypu72VPfxbaaDhwGitIDTM1OYnJ2MOHd20VERERk6OibnYiMa9ZaDjb3sL6ilarWXjxOB3MKUpiVl0JuinfUdA03xpAW8JAW8DCvMJVo1FLX0cf+pm72NnTx7K4Gnt0FhWl+ZuQmMy03CZ9azkVERETGNAVyERmXrLUcaOrmlX3NNHcPkOR1cdrULOYWpuB1jf4g63CYwRb0U6dk0tQ1wL7GLvbUx8L587sbmZQdZFZeMqWZQZyO0fHDgoiIiMhQKysporyyesier7S4kIMVVe96zKc+9SkeffRRcnJy2Lp165C99lspkIvIuFPX0cdLe5qobuslPeDm/Nm5TMtNHrOh1RhDdrKX7GQvJ03KoLGznx11neyq62RvQxcBj5M5BSnMLUglxe9OdLkiIiIiQ6q8shr77HeG7PnMqm+85zGf+MQn+OIXv8jHP/7xIXvdo1EgF5Fxo7MvxMt7m9lV34nf7eSsGdnMKUgds0H8aIwx5KT4yEnxcdrULMpbutlW3cHag628cbCVsswA8wpTmZQVHDXd8UVERETGmjPOOIODBw8O++sokIvImGetZVtNBy/uaSJiLcvK0llSmj4muqafCKfDMDkriclZSXT0hdhW3cG2mnb+srmWVL+bRcVpzMpPweN672XaRERERGTkKZCLTGDFJaVUVVYkuowT0t4b4pkd9VS29lKU7uecWbmkTsBu2yk+NyumZLJ8Ugb7GrvYUNHGc7sbeXV/M3MLU1lYnEaSZmgXERERGVX07UxkAquqrOD2J3cluoxjdst5Mwa3rbVsqW7nxT1NOIxh1cwc5hakTPhu2k6HYXpuMtNzk6lt72V9RRvry1vZWNHGrIJklpZmTMgfLERERERGIwVyERlz+sMRntnRwJ6GLkozA5w9M4dkn0LmW+Wn+rl4np/23hDrylvZXtPBtuoOpucls6w0ncwkb6JLFBEREZnQFMhFZExp7OznsS21dPSFOG1qFotL0iZ8q/h7SfW7WTUzh+WTMthQ0cqW6nZ21XUyPTeJkydlJro8ERERkXdVWlx4TDOjv5/ney/XXHMNzz33HE1NTRQVFXHrrbdyww03DFkNbxrWQG6MOQh0AhEgbK1daozJAO4FyoCDwJXW2tbhrENExoek+edx79pKfG4HH15URGG6P9EljSlJXhenT8tmaWkG6yta2VjZxp76LjIv+jIHm7opywomukQRERGRt3mvNcOHw9133z0irzMSU++eZa1daK1dGr/9NeAZa+004Jn4bRGRdxS1lud3NZJ54d9TkObjo8tLFMZPgN/j5NSpWXzy1DIWlaQRmHka59z+PN98eCvNXf2JLk9ERERkwkjEWjiXAr+Nb/8W+FACahCRMWIgHOXRzbVsrGqj440/86GFhQQ8Gm0zFAKeWIt5zc8/zVXLivnDmgpW/vdz/OS5vfSFIokuT0RERGTcG+5AboEnjTHrjDE3xvflWmtr49t1QO7RHmiMudEYs9YYs7axsXGYyxSR0airP8wD66s42NTNyunZtD77SxwaLz7kIt2tfPuyeTxx8+mcNDmD7z2+i1Xff44H11cRjdpElyciIiITgLVj/zvH8byH4Q7kp1lrFwMXAl8wxpxx+J02VvFRq7bW3mGtXWqtXZqdnT3MZYrIaNPc1c+9b1TS2jPABxcUsKA4LdEljXtTc5L55fXLuPvvTiYzycst923ikv97iVf2NiW6NBERERnHfD4fzc3NYzqUW2tpbm7G5/O9r8cNa79Pa211/LrBGPMQsByoN8bkW2trjTH5QMNw1iAiY09DRx8PbazGaQxXLCkiJ/n9fbDJiVkxJZOHv3Aqj2yq4b+f2MVHf7mGVTNz+MZFM5mak5zo8kRERGScKSoqoqqqirHeM9rn81FUVPS+HjNsgdwYEwQc1trO+PZ5wL8DjwDXA7fFrx8erhpEZOypbe/lzxtr8LocXL6okLSAJ9EljX/G8c5LxzndpCy5hKf7r+SZbTV0vPFn2l+5BxvqG9ka44qKS6isKE/Ia4uIiMjwcLvdTJo0KdFlJMRwtpDnAg/Fv+S5gD9aax83xrwB3GeMuQEoB64cxhpEZAypau3hkU01BDwuLl9cSIrPneiSJgYb5fYnd73rIT0DYV7e28x25xUUnnk1Z0zLYmpO0oivAX/LeTNG9PVEREREhtOwBXJr7X5gwVH2NwNnD9frisjYVN7czaOba0nxublscSFJXs2kPpoEPC7OnZ3LnIIUVu9q4K9b6yjJCLByRjbp6sUgIiIiclwSseyZiMgRqlp7+MvmWtICbj68RGF8NCtI83PNshLOnJ5NXXsfd71WwSv7mghFookuTURERGTM0bdeEUmouvY+HtlUQ6rfzeWLivB7nIkuSd6Dw2FYWJzGtJwkXtrbxBsHW9lZ18mZ07OZnBUc8W7sIiIiImOVWshFJGGauvr588ZqAh4Xly0qVBgfY4JeF+fPyePDiwvxOB08urmWRzfX0tUfTnRpIiIiImOCArmIJERrzwAPbajG7XRw2SJ1Ux/LitIDXLO8hFOnZlLe0sPvXy1nS3X7mF5LVERERGQkKJCLyIjr6gvz0IZqrIXLFhWS6tds6mOd02FYWprBtSeVkJPs5dmdDTy4vpq2noFElyYiIiIyaimQi8iIGghHeXhTNX2hCB9aVEBGUDN0jyfpAQ+XLy7k7Jk5NHT184c1FawtbyEaVWu5iIiIyFupj6iIjJhI1PLXLbU0dw9w6YICcpJ9iS5JhoExhrmFqZRlBXluVwMv721mT30X58zKJTvZm+jyREREREYNtZCLyIiw1rJ6VwPlLT2smplDaWYw0SXJMEvyurh4Xj4Xzcujqz/M3W9U8PLeJsJaIk1EREQEUAu5iIyQN8pb2VbTwbKydOYWpCa6HBkhxhim5SRTnB7gxT1NrC1v5UBTN+fOziU3RT0kREREZGJTC7mIDLtddZ28uq+ZmXnJrJicmehyJAF8bifnzs7l0gUF9IUj3Lu2ktf2NxPR2HIRERGZwBTIRWRY1Xf08dSOegrSfJw9KwdjTKJLkgQqywrysZNKmZGbzJoDLdy7tpKmrv5ElyUiIiKSEArkIjJsuvvDPLq5loDHycXz8nE59JEjsdby8+fkcfG8fLr6wtzzeiVvHNRM7CIiIjLxaAy5iAyLSNTy2JZa+kIRrlxaTMCjjxs50tScJArSfKze2cgr+5rZ39jNebNzSddSeCIiIjJBqLlKRIactZbndjVQ297HubO11JW8s4DHxUXz8rhgTh6tPQPc9XoFGypasVat5SIiIjL+qclKRIbclup2tsZnVJ+em5zocmSUM8YwIy+ZonQ/z+xs4IU9TexrjM3Enup3J7o8ERERkWGjFnIRGVK17b08v7uRssyAZlSX9yXodfHB+fmcMyuHxs5+7lpTzuaqNrWWi4iIyLilQC4iQ6ZnIMxft9SR5HVx/pw8zagu75sxhjkFqVx7cgl5qT5W72rkzxtr6OwLJbo0ERERkSGnQC4iQyJqLU9sq6c3FOHi+fn43M5ElyRjWIrPzWULC1k5I5uatl7+sKaCnXUdiS5LREREZEgpkIvIkFizv4WKlh5WTs8mJ9mX6HJkHDDGsKAojWtPKiEz6OGJbfVkfejrtHQPJLo0ERERkSGhQC4iJ+xgUzevH2xhVn4ycwpSEl2OjDNpAQ9XLCni1CmZBKYu57wfvMDT2+sTXZaIiIjICVMgF5ET0tkX4oltdWQleThrRo7GjcuwcBjD0rIMan/7ZbKSPHz6d2v56v2bNLZcRERExjQFchE5btGo5fGtdUSs5aJ5+bid+kiR4RVqPMjDXzyVz6+cwv3rqrjghy/y6r7mRJclIiIiclz07VlEjtuaAy3UtPexamYO6QFPosuRCcLrcvLVC2byp8+egttpuOYXr/Hvf9lOXyiS6NJERERE3hcFchE5LpUtPbx+sIXZ+SnMzNO4cRl5S0rT+etNp/PxFaX86uUDXPw/L7K5qi3RZYmIiIgcs2EP5MYYpzFmgzHm0fjtScaYNcaYvcaYe40xalYTGWN6BsI8sa2O9ICblTOyE12OTGABj4t/v3Quv79hOd39ES77ySv84KndhCLRRJcmIiIi8p5GooX8JmDHYbe/C/zAWjsVaAVuGIEaRGSIWGt5cns9feEoF87VuHEZHU6fls0TN5/BJQsK+NEze7j8J6+wp74z0WWJiIiIvKth/SZtjCkCLgZ+Gb9tgFXA/fFDfgt8aDhrEJGhtaGyjfLmHk6flkV2sjfR5YgMSg24+cFVC/nptYupau3h4h+/xC9f3E80ahNdmoiIiMhRDXfT1g+BrwJv9h3MBNqsteH47Sqg8GgPNMbcaIxZa4xZ29jYOMxlisixaOzs55W9zUzOCjK/MDXR5Ygc1YXz8nnyy2dyxrQs/vOxHVzzi9eobOlJdFkiIiIibzNsgdwY8wGgwVq77ngeb629w1q71Fq7NDtbY1RFEi0cifL4tjq8bgfnzMrVeuMyqmUne/nFx5fyvSvms62mgwt/9CL3vlGBtWotFxERkdFjOFvITwUuMcYcBO4h1lX9R0CaMcYVP6YIqB7GGkRkiLy0t4mW7gHOm52L3+NMdDki78kYw5VLi/nbTacztzCFf3pgC5/+7VoaOvsSXZqIiIgIMIyB3Fr7dWttkbW2DLgaeNZaey2wGrgiftj1wMPDVYOIDI0DTd1sqmpnYXEapZnBRJcj8r4UZwT446dP5l8/MJuX9jZx/g9e4K9bahNdloiIiEhC1iH/J+AWY8xeYmPK70xADSJyjHoGwjy1vZ7MJA+nTslMdDkix8XhMNxw2iQe+/vTKM4I8Pm71vP3d2+gpXsg0aWJiIjIBDYigdxa+5y19gPx7f3W2uXW2qnW2o9Ya/tHogYRef+stTy9o4GBSJQL5uTh0hJnMsZNzUnmgc+dwpfPmc7fttZy7u3P89hmtZaLiIhIYujbtYi8o+21HRxo6ubUKZlkJWmJMxkf3E4HN50zjb986TQK0/184Y/r+dwf1mlsuYiIiIw413sfIiITUUdviOd3N1KU7mdhcVqiyxGJMY6hneHfOEhZdhl/DV/Lo2/spvXpO+je/tyQPHVRcQmVFeVD8lwiIiIyPimQi8jbWGt5cns9BsO5WuJMRhMb5fYndw3507Z0D/D0jnqcH/wKSz/xb6yamUOyz31Cz3nLeTOGqDoREREZr9RlXUTeZmNlG9VtvZwxPYsU/4mFEpGxICPo4YolRZwxLYuq1l7+8FoFW6vbtW65iIiIDCsFchE5Qkv3AC/va2ZSVpDZ+SmJLkdkxDiMYVFJOteeVEJOspdndjbw4PpqWjUTu4iIiAwTBXIRGRSJWp7YVofH6eDsmTnqqi4TUlrAw+WLCzl7Zg6NXf3ctaaC1/Y3E45GE12aiIiIjDMaQy4ig9442EJDZz8Xzcsj6NXHg0xcxhjmFqYyKSvIC3saWXOghd31nayamUNReiDR5YmIiMg4cUwt5MaYU49ln4iMXfUdfbxxsIUZeclMy0lOdDkio0LQ6+LCufl8aGEBkajlgfXVPLW9nt5QJNGliYiIyDhwrE1gPwYWH8M+ERmDwpEoT26rJ+BxsXJ6dqLLeVe33nprokuQCag0M8jHTi7l9QMtrK9o5UBTN6dPy2JmXrKGdoiIiMhxe9dAboxZAZwCZBtjbjnsrhTAOZyFicjIeWV/My09A3xoYQE+9+j+p/3N689OdAnH5JaX/5joEmSIuZ0OTp2axYy8ZJ7d2cCT2+vZXtvByunZZCZ5E12eiIiIjEHv1WXdAyQRC+7Jh106gCuGtzQRGQlVrT1sqGhjXmEqpZnBRJcjMuplJXn5yJIizpqRTWNnP398vYIXdjfSH1Y3dhEREXl/3rWF3Fr7PPC8MeY31tryEapJREaI8fh5ans9qX43p0/LSnQ5kkBjaSjAaKrVuLz4yhawPjqFdftq6T24kVDDgUSXJSIiImPEsY4h9xpj7gDKDn+MtXbVcBQlIiMjfdWn6ewLc8WSItxOrYI4kY2loQCjsdb6vhaea0qmbvoK8uYtYWV2J9/VsAURERF5D8cayP8E/Az4JaA+eSLjwLM760lecD6LS9MpSPMnuhyRMS3XF+bKwlZ2dPp4uSWJe6oyyDj/i7R0D5AR9CS6PBERERmljjWQh621Px3WSkRkxLR2D/BPD2xhoOEAJ581JdHliIwLxsDslD6mBPtZ0xpk/bxzWPnfq/nyudO59qRSPC71QhEREZEjHeu3g78YYz5vjMk3xmS8eRnWykRk2PzLw1tp6xmg6dHbcTkUEkSGktdpOSOri9pff4n5RWnc+pftnPeD5/nbllqstYkuT0REREaRY/0mfj3wj8ArwLr4Ze1wFSUiw+eRTTU8trmWm8+ZTqhRk0+JDJdQcyW/v2E5v/7kMjwuB5+7az1X/OxV1pW3Jro0ERERGSWOKZBbaycd5TJ5uIsTkaFV39HHv/55K4tK0vjMGfonLDLcjDGcNSOHv/796dx2+TwqWnr48E9f4fN3reNgU3eiyxMREZEEO6Yx5MaYjx9tv7X2d0NbjogMF2stX71/M/3hCLdfuRCXZlUXGTEup4Orl5fwwQUF/OLF/dzxwn6e2l7Px04u5e9XTSNdE7+JiIhMSMc6qduyw7Z9wNnAekCBXGSMuPv1Sp7f3citl8xhUlYw0eXIcXJEQ/hDrQRCrfjDbfhDbXgi3biifbii/bgi/bii/TjtQHy7D5cdAAxR4ySKA2tcRI0Da5xEjZMfX+hjRfvD9Bsf/Q4/A8ZHn8NPf/wyYOLbxgtGP+SciKDXxc3nTOejy0v4wdO7+e0rB7l/XRWfOWMynzh1EkneY/2zLCIiIuPBMf3lt9Z+6fDbxpg04J7hKEhEhl55czf/+dh2TpuaxXUnlya6HHkH3nAHqX01pPTVkNpfTUpfDcn99fjDbbEAHmrFG3nvbs4hh5eww0fE4SHs8BJ2eDDWYojisBGMjeKw4dg1EYrmukjufgWPHXjX5w3jpNuZSqcznU5nKl3ONDqdaUdc9ziSYtONy7vKSfHxX5fP55OnTuJ7j+/k+0/u5lcvH+SzZ07mupPL8HuciS5RRERERsDx/hTfDUwaykJEZHhEopav/GkTTofhe1fMx+FQWEokZ6SPzN6DZPXsJbN7H6l9VaT2x0K4L9J1xLF9zmQ6vXn0uNOpSyqg151GjzudXnc6ve40el3p9LjTGXAlEY6H74jxvO9AfMt5M7j9jh9hbARvtA+v7cUb7cVr+2LX8Usg2klSpI3kSBsFAwdJirTjJHJkzcZHiyuXFnfuEdftzgy1rh/F9Nxkfnn9MjZUtHL7U7v5zl938osXD/D5lVO4ZnkJPreCuYiIyHh2rGPI/wK8uVaLE5gF3DdcRYnI0Lnzpf28cbCV269cQEGaP9HlTBgGSO2tIqtnL1nde8jq2UdWz17SeitxEAUgbDy0+wro8BVQkzyfDl8B7d7C2LWvgH5XyojWbI2TPmeQPo5xSIONEoh2D4b05Egr6eEGMkP1TOrbwdzo64OHhoybVlcOza486jwl1HlKaHQXEjHuYXo3Y8uiknR+f8NJvH6ghe8/uYtb/7KdO17YzxdXTeUjS4q1hrmIiMg4dawt5N8/bDsMlFtrq4ahHhEZQrvqOvn+E7s5f04uly0qTHQ545on2kveQDkFAwfJHzjIJ/8pmbT1lwFgMbT7CmkKTGV35jk0BafSFJhKm78Ia8bwmGHjoMeZTI8zmQaK33a3N9pDRqiezHA9GeF6MkL1FPfvZVbvOgAiOGl0F8QDeil17hJaXdkTuiV9+aQM7r3xZF7e28z/e2oX//zQVn763D7+/uxpXLaoELcmYxQRERlXjnUM+fPGmFwOTe62570eY4zxAS8A3vjr3G+t/aYxZhKx8eeZxNYzv87a9xi4KCLv20A4yi33bSTZ5+I7l83DaFzv0LGWtEgTBf0HyI8H8KxwHQaLxdDkyuOerSGmXHUrTcGpNAcmE3IGEl31iOt3BKj1TqLWe+QIp2CkjbyBivilnNk9a1nY/TIQ6/Je6ymjwjudCt90mlz5Ey6gG2M4bVoWp07NZPWuBv7fk7v56v2b+dHTe/jsmZP5yNJidWUXEREZJ461y/qVwH8DzxHrifljY8w/Wmvvf5eH9QOrrLVdxhg38JIx5m/ALcAPrLX3GGN+BtwA/PRE3oSIvN0Pn97NtpoOfn7dEjKTvIkuZ8zzR7oo6d9NSf8uSvt2kRxtB2IBss5Txh7/Amo9ZdR5Shhw+LnlsRu4/aYPJbboUarbmcY+fxr7/PMBMDZKRrie3IEK8gfKKRzYx5kdj0AH9DiCVHqnxwK6d1qCKx9ZxhhWzczlrBk5PLuzgf9dvZd/fXgbP3pmL58+fRIfO7lUs7KLiIiMccf6l/yfgWXW2gYAY0w28DTwjoHcWmuBN2cocscvFlgFfDS+/7fAt1AgFxlSrx9o4afP7+OqpcWcPycv0eWMSU4bJn/gAKV9uyjt30VuKDZKp88EqPBNY413OjWeSTS7cidcC+5Qs8ZBszufZnc+24MnAZAUaaO4fw8lfbsp6d/NjN4NAKz8UhI9rfey3zeHct/02CR245wxhrNn5bJqZg6v7W/hJ8/t5ba/7eQnq/fyiVMn8clTyrSOuYiIyBh1rIHc8WYYj2sG3vMbqDHGSaxb+lTg/4B9QJu1Nhw/pArQwFaRIdTRF+LL926kJCPAv31wdqLLGVP8kS6m9G1lSu8Wigf24rYDRHBQ6ynj5eQLKffNoMFdjFUAH3ZdzjR2BJaxI7AMrCUj3EBJ/y7ad97HhZkbmdfzGiHj4aB3Jvv8c9nvm02/4xgnoxujjDGsmJLJiimZbKxs4yer9/I/z+zhly/u56PLS/j06ZPJS/UlukwRERF5H441kD9ujHkCuDt++yrgr+/1IGttBFgYX7f8IWDmsRZmjLkRuBGgpKTkWB8mMuF96+Ft1HX08afPriCo7qzvKRhpY2rvFqb1bqZwYB8OLO3ODLYFllPunUGVdyoDjuMLObfeeusQVztBGRNbPs2dyy33/pYf/vyHFPXvY0rfFqb2bmVa32aiOKjyTmGvbx77fHPpcqUnuuphtbA4jTs+vpTd9Z389Ll9/PqVg/zmlYN8cEEBN5w2ibmFqYkuUURERI7Bu35bN8ZMBXKttf9ojLkcOC1+16vAXcf6ItbaNmPMamAFkGaMccVbyYuA6nd4zB3AHQBLly61RztGRI70l001PLihmpvOnsbikvEdSE5EariJqb2bmdq7mYJQOQDNrlxeTz6Hvb4FNLoL3vda3kfzzevPPuHnGAm3vPzHRJfwvkSNiwrfDCp8M1idejm5oUqm9m5lSt8WVrU/yKr2B6lzF7MzsIRd/sX0OJMTXfKwmZ6bzA+uWsgt507nVy8f4L43KnloQzUnT87g06dNZtXMHBwOTegoIiIyWr1X89kPga8DWGsfBB4EMMbMi9/3wXd6YHyceSgexv3AucB3gdXAFcRmWr8eePiE3oGIAFDb3ss/P7SFhcVpfGnV1ESXM+r4I13M6F3P7J43BseD17uLeCnlIvb65tPqzk1whXJcjIN6Tyn1nlJeTr2Y9FADU/q2ML13Iyvb/8wZ7Y9Q7p3OzsBS9vrmEnaMzwkOizMCfPODc7j5nOnc+0YFv3n5IJ/+3VomZQX51KllfHhJEQGPesyIiIiMNu/11znXWrvlrTuttVuMMWXv8dh84LfxceQO4D5r7aPGmO3APcaY/wQ2AHceR90icpho1PIP920iHLX88KqFuLRWMRCbmG1S33Zm97xBWd92nESpdxfxfMql7PXPp8OVkegSZYi1unNY6z6btclnkxGqY2bPOmb1ruPC1j8wYDzs9c1nR2AJld7p43IugFS/mxvPmMKnTp3E37bW8cuXDvCvD2/j+0/u5qMnlXDdyaUUpPkTXaaIiIjEvVcgT3uX+971L7q1djOw6Cj79wPL37MyETlmv3r5AK/sa+a7H55HWdb4ntjqPVlLbqiC2T1rmdGzHr/toduRzIakM9keWEazOz/RFcoIaXHn8UrqxbySciGFAweY1bOWab0bmd27li5HCjsDS9gcXEG7KzvRpQ45l9PBBxcU8IH5+ayvaOWXLx7g58/v4+fP7+Pc2blcd3IZp07NxAzB0AwRERE5fu8VyNcaY/7OWvuLw3caYz5NbPZ0EUmwHbUdfO/xXZw3O5crlxYnupyE8UZ7mNP9OnN7XiMzXE8YF/v889geWEa5dzrWOBNdoiSKcVDtnUK1dwqr0y5nUt92ZvWsZXHX8yztWk2Fdxqbg6ewzzeP6Dj7/8QYw5LSDJaUZlDZ0sMfX6/g3jcqeWJbPZOzgnzs5FI+vKSIVL870aWKiIhMSO8VyG8GHjLGXMuhAL4U8ACXDWNdInIM+kIRbr5nI6kBN7d9eP6EbO3KHqhmQfdLzOxdh9uGqHGX8nTaR9jtX0S/Q11z5UgR42avfwF7/QsIRtqZ072GeT2v8YGW39LtSGZb4CS2BE+mw5V54i9mHKPz36TTRXDGafQtvph/b+rmWw+uh4q1/O1/vsacAs3OLiIiMpLeNZBba+uBU4wxZwFz47sfs9Y+O+yVich7+t7ju9hV38lvPrmMjKAn0eWMGIcNM613Ewu7X6Jg4CAh42anfwmbgqfR6ClMdHkyRnQ7U3k95TzeSD6H0v6dzO9+haVdz7Cs6xnKvTPYHFzBft+c4+9dYaPc/uSuoS16iDV09LG5up0tZikX/89LLC5J4+rlJXxgfr4mgRMRERkBx/TX1lq7mtjs6CIySry4p5FfvXyA61eUsnJGTqLLGRFFKYZTOv7K3O5XCUa7aHVm8VzqpWwPLKffEUh0eTJGWePgoG82B32zSQq3MrdnDXO7X+OSll/T7kxnY/B0tgVPHpc9LnJSfJyT4uPxr17IHU9v4a415Xz1/s38x1+2c8nCAq5ZXqI1zUVERIaRfv4WGYOauvr5h/s2MSU7yNcunJXocoZdVvdullb/ni/dlISz82kO+GaxMXg65d7pMA5nypbE6XKl81rKBaxJPpfJfdtY1PUCZ3Y8worOJ9gWWM6GpNPH5SRw0f5ubjhtEp86tYw3DrZyzxsV3L+uirvWVDCnIIWrl5dw6cICUnwaay4iIjKUFMiHWFlJEeWV1Yku45iUFhdysKIq0WXI+/TmEmdtvSF+88nl+D0nNgnVrbfeOkSVDTXLaenN3FS6j3OzGukKO/mfNwZI/dB/DM34XpF3YY2Tff757PPPJ3ugisVdzzO/+xUWdr/Eft8c1iedQZVnKozGMeInwBjD8kkZLJ+UwTc/OIdHNlZz9+uV/Ouft/Ltx7Zz8bwCrl5ezNLS9NE5Pl5ERGSMUSAfYuWV1dhnv5PoMo6JWfWNRJcgx+GXL+3n+d2N/Melc5hdkHLCz/fN688egqqGjrFRpvRtYVnns+SFKuhxJPFy0kVsCp7KP3z7S9x+hcK4jKxGTxFPZFzLi5EPsqD7ZeZ3v8xHmrbS4C5gQ/BMdgYWEzXj789pqt/NdSvK+NjJpWypbueeNyp5ZGMND6yvojQzwOWLirh8cSHFGRouIiIicrzG3zcIkXFsQ0Ur33t8FxfMyeNjJ5cmupwh5bQhZve8wZLO50iPNNLmzOKZtCvYFlhGxEycCetk9OpxpvBqyoW8nnwOM3vWsbjrBc5vu5tTOv/GuqQz2RpYQcjhTXSZQ84Yw/yiNOYXpfEvF8/i8a113L+uih8+s5sfPL2bkydncMWSYi6cm0fQq68VIiIi74f+coqMEe29Ib509wZyU3x8dxwtcea0IeZ1v8qyzmdIinZQ7y7i0dTr2eubj9X4cBmFIsbNtuDJbAucRGn/LpZ1PsPK9oc5ueMpNiadxsbg6fQ6kxJd5rAIeFxcvriIyxcXUdXaw0Prq3lgfRVf+dMm/u3hrVw4N58PLynk5EmZOBzj4zNKRERkOCmQi4wB1lq+8eAWatv7uO8zK0gNjP2JlZw2zJzu11je+TTJ0XaqPFN4PPlaKr3Txt24XBmnjKHcN5Ny30zyBspZ2vkMJ3U+xZKu1WwLnERZ2vj+/7goPcCXzp7GF1dNZV15Kw+sr+LRTbU8sL6KwjQ/H15cyOWLiyjLCia6VBERkVFLgVxkDPjdq+U8tqWWf7pgJktK0xNdzglx2DBze9awrPNpUiJtVHsm8UTKtVSOwwmyZOKo85TyaOanSA/Vs7RrNfO6X2XPl5LYs+tfeKPoEzQHpya6xGFjjGFpWQZLy2ITwT2xrY4H1lfz49V7+Z9n97KsLJ3LFxdx0bx8Uv1j/8dEERGRoaRALjLKra9o5T8f287ZM3P4zBmTE13OcXPYCLN7XuekzqdIibRS4ynjqbSrqfBOVxCXcaPVnctT6VfzasoFtD/4Df7+tBeZ1fQEezNWsqb4BhqSZia6xGHlczu5dGEhly4spK69j4c2VHP/ukq+/uAWvvnINs6dlctliwo5c0Y2bqeGpIiIiCiQi4xiLd0DfPGu9eSm+Lj9yoVjckymsVFm9q5jRcfjpEZaqHWX8nTalZR7ZyiIy7jV5UzjH5/qJ3DzIyyuvYeFNfcwteU59qefypriG6hLnpfoEoddXqqPz62cwmfPnMzmqnYe2lDNI5tqeGxLLRlBDx+cn8/li4uYX5Q6bubEEBEReb8UyEVGqUjUctM9G2jqGuCBz50y9saNW0tZ/w5Oa3+U7HAt9e4ink27goPemQriMmH0u1N5teQzrCu4loW197G45o9cs/lTlKcuZ03xDVSnLk50icPOGMOC4jQWFKfxzxfP4oXdjTy4vpq736jkt6+WMzk7yOWLCvnQokKK0rWEmoiITCwK5CKj1I+e2cOLe5r4r8vnMa8oNdHlvC95A+Wc1v4Xigf20ebM4rH0j7Pbv1BBXCasAVcSrxd/ig0FVzO/9gGW1PyBK7d+hsqUxawpvoHK1GUT4t+H2+ng7Fm5nD0rl/beEH/bUsuDG6r5/pO7+f6TuzlpUgaXLy7kwnn5pPjG2I+QIiIix0GBXGQUWr2rgR8/u4cPLy7i6mXFiS7nmKWHGji14zGm9W2m25HEM6kfZmtwBVHjTHRpIqNCyBlgXdF1bMr/CPPqH2Jp9e+5YtsXqEmex5riGziYdsqECOYAqX43Vy8v4erlJVS29PDnDdU8tKGaf3pgC//28DbOmZ3L5YsKOWO6xpuLiMj4pUAuMsrsb+zi7+/ewMy8FP7zQ3PHxNjKYKSdkzueYG7PGsLGzSvJF7A+aSUhhzfRpYmMSmGnjw0F17A573Lm1P+FZVW/4bLtN1OXNIs1RTewP+OMCRPMAYozDi2htqmqnYfWV8XGm2+uJTPo4YMLCrh8cSHzCjXeXERExhcFcpFRpLMvxN/9bi1up4M7rluC3zO6W5Zd0QGWdK1mWdezOGyYTcFTWJN8Hr3O5ESXJjImRBxeNudfwdbcS5nd8BjLqn/DpTu/QkNwGq8X3cCezLPATJzWYWMMC4vTWFicxj9fPJvndzfy0IYq/rimgt+8cpAp2UEuX1zEhxYVUpjmT3S5IiIiJ0yBXGSUiEYtN9+zkfLmHv7w6ZMozhjFkxtZy4ze9Zze8SjJkTb2+ObzYuoHaXdlJboykTEp6nCzNe9DbMv9ADMbn2B55a/4wK6v0eyfxJriT7E761zscA79MI5R3fJsvEGCM0+jb84q9jV2899P7KKvfDNd21bTs+sl7EBvoks8qqLiEioryhNdhoiIjGIK5CKjxO1P7eaZnQ38+6VzOHlyZqLLeUd5Awc5s+3PFITKqXcX8bf0j1HtnZLoskTGBWtc7Mi5mJ3ZFzCt6RlOqrqTi3b/KysqfsHrRZ9gZ/aFRB3D8KfbRrn9yV1D/7zD4KsfOYMPf+8hdviX4Cudj/MDNzMlO8jMvBRKMwKjannIW86bkegSRERklFMgFxkFHt1cw/+u3svVy4q57uTSRJdzVEnhVk7reIxZvevodiTzRNo1bA8snVDdaUVGijVOdmefx+6sc5ja/BwnVd3J+Xv/nZMrf8kbRdezPecDRByeRJeZEOH2ek6alMnysgzqOvrYWdvJ7vpOdtd34Xc7mZGXzKy8ZLKTvaO61V9ERAQUyEUSbkNFK/9w3yaWlKZz66VzRt0XSFe0n6Vdq1na9SzGWtYkn8sbSWdrwjaRkWAc7M1axd7Ms5jU+jInVd7JOfv+i5Mq72Rt4XVszf0QYacv0VUmhDGG/FQ/+al+zpiezcHmbnbUdrClqp2NlW1kBD3MzEtmZl4yyVpCTURERikFcpEEqmju4dO/XUtuio87rluC1zWKJnGzUWb2rue09kdJjrazy7+IF1M+QKcrI9GViUw8xnAg4zQOpJ9KSfvrnFR5J2cd+H8sr/o16wuuZVPehwm5gomuMmGcDsOU7CSmZCfRF4qwp76LHXUdvLKvmVf2NVOU7mdWXgpTc5LwuNSrR0RERg8FcpEEae8J8cnfvE44avn1J5eRmTR6Wpzz+w9yZvtD5IcqqHMX89eMj1PjnZzoskTEGCrSTqIi7SQK29ezvOrXnF7+Y5ZW/44NBVezMf8q+l0Te5UDn9vJvKJU5hWl0tYzwM66TnbWdfLUjnpW72pgcnaQWXkplIyy8eYiIjIxKZCLJMBAOMpn/rCWypZefn/DcqZkJyW6JCA2Tvz0jkeZ2bueLkcqj6d/lB3+JRonLjIKVacu5qHUxeR1bmV55a84peLnLKn+Axvzr2J9wTX0udMSXWLCpQU8nDw5k5Mmxcab7zhsvHnA42RGbjKz8lPITh49P4iKiMjEMmyB3BhTDPwOyAUscIe19kfGmAzgXqAMOAhcaa1tHa46REYbay1fe2Azr+1v4YdXLeSkUTCj+qH1xJ/BWHgt+TzWJq3SOHGRMaAueS6PzL6d7K5dLK/6Ncurfs3imj+yKe/DrC/4KN3enESXmHBHjjfP4mBTDzvrOthU1caGyjZykr3Mzk9hRl4yPvcoGjokIiLj3nC2kIeBf7DWrjfGJAPrjDFPAZ8AnrHW3maM+RrwNeCfhrEOkVHltr/t5MEN1dxy7nQ+tKgw0eUwvWcDp3f8hZRIK7v8C3kx5YMaJy4yBjUmzeCxmbeR0bM/HsrvZlHtvezIvpB1hdfREpiU6BJHBZfDwdScJKbmJNEbirC7rpNtNR08t7uRF/c2MSU7yJyCVIrT/aNukk0RERl/hi2QW2trgdr4dqcxZgdQCFwKrIwf9lvgORTIZYL46XP7+PkL+/n4ilK+tGpqYoup3cxz1wc4s/V3NLgLeDz9o1R7E1yTiJywlsBkHp/+H7xS8lmWVN/F3IZHmNvwF/ZmnMnawo9TmzI/0SWOGn63kwXFaSwoTqOhs4/tNR3srIt1aU/2uZidn8Ls/BRS/JqlXUREhseIjCE3xpQBi4A1QG48rAPUEevSfrTH3AjcCFBSUjICVYoMrz+uqeC7j+/k0oUFfOuDCVzerLsJnv1PWP9b5uQ4eDrtI2wNnIzVOHGRcaXDV8jqKV/ltZK/Y2HNvSys+xNTW56nKmURaws/zoH0U0EtwINykn3kzPBx2tQs9jV2s622nTUHWlhzoIWSjACz81OYkh3E5dRnpYiIDJ1hD+TGmCTgAeBma23H4SHEWmuNMfZoj7PW3gHcAbB06dKjHiMyVjy6uYZ//vMWzpqRzfc/siAxM/tGQvD6L+C52yDUDSd9lmmX3ca//c8pI1+LiIyYXnc6r5Z+lrVFH2du/Z9ZXP1HPrTjyzQFprCu8GPsyjov0SWOKi6ngxl5yczIS6ajN8T22g6213bw+LY6vC4HM/OSmV2QQk7yxFz/XUREhtawBnJjjJtYGL/LWvtgfHe9MSbfWltrjMkHGoazBpFEW72rgS/fu5Glpen85NoluBPRurL3aXj869C0G6asggtug+wZtPXdNvK1iEhChJwBNhR8lE15VzKj6QmWVv+O8/fcymkHf0zTGR78Ay30ejR/xOFS/O7BWdorW3vZVtPO1poONlW1k53kZU6BJoITEZETM5yzrBvgTmCHtfb2w+56BLgeuC1+/fBw1SCSaKt3NvCZ369jem4yv7x+GX7PCH9pa94HT3wDdj8OGZPhmntg+gXqpioygUUdLnbkXMyO7IsoaX+dRTV38+9ntRBe+wF2ZZ/PhvyraUyakegyRxVjDCUZAUoyAvSFIuyq62R7bXwiuD2xieBmF8TWNtdEcCIi8n4MZwv5qcB1wBZjzMb4vm8QC+L3GWNuAMqBK4exBpGEeXZnPZ/9/Xqm5yXxhxtOInUkJwXq64AX/hte+ym4vHDOrXDy52LbIiIAxlCRdhIVaSfxs2tmce9/fpI5DX9hTsOjVKYsYUPB1ezPOB1r1Pp7ON9hE8E1dvazvaaDHXUd7G6ITQQ3Jz+FWQUppPg0EZyIiLy34Zxl/SXgnX4mPnu4XldkNHhmRz2f+8N6ZuQlx8J4YIS+mEWjsOmP8PSt0N0ACz8GZ/8bJB917kQREQB2N0dZPeWrvFL6WebWP8zC2vu4ZOc/0uYrZEvu5WzL/SC97vRElznqZCd7OXNGNqdOzRycCO61Ay28dqCF0owAgRmn0h+O4HXpRw0RETm6EZllXWQieXp7PZ+7ax2z8lP4/adGMIyXvwpPfB1qNkDRMvjoPVC4ZGReW0TGhX5XCusKr2N9wTVMaX6BRbX3cHr5jzml4qfszTyLzbmXU5W6RMNe3uKtE8Ftq+1ge00H2R/6Oiv+61kuW1TIVcuKmZ6bnOhSRURklFEgFxlCD66v4h/v38ycghR+P1Ld1Jv2wtPfhJ2PQnI+XHYHzPsIOLQ0j4gcH2tc7M1axd6sVWT07Gde3UPMbniMGU1P0eIrYUveZWzP+QB97rRElzrqpPjdrIhPBPcvn/0oF/3HHfzu1YPc+dIBFhancdWyYj64oIAkr76CiYiIArnIkPnFC/v59l93cMqUTH5+3RKSh3v8YHcTPP9dWPsrcHrhrH+GFV8AT3B4X1dEJpSWwGSen/wPvFT6BaY3P8O8uoc48+CPOLX8J+zJXMWWvMuoTlkERj8CHs5hDH0H1vOTa5fQ3NXPQxuqufeNSr7+4Bb+49HtXDwvn6uWFbOkNF0TwYmITGAK5CInyFrLbX/byc9f2M/F8/K5/aoFwzteMNQbm6ztpR/AQBcsvh5Wfl3jxEVkWEWcvtjs7DkXk9m9l3n1f2ZWw2PManqCdm8BO7MvYEf2hbQGyhJd6qiTmeTl06dP5obTJrGhso373qjkL5tq+NO6KqZkB7lyaTGXLy4iO1kTb4qITDQK5CInIBSJ8vUHt3D/uiquO7mUb10yB6djmFo6olHYch888x/QURVbvuycWyFn5vC8noicsFtvvTXRJQyL5uBUnpv8FV4q/SJTm1czq/GvLKv6DSdV/Yq6pNnsyL6IXVnnal3ztzDGsLgkncUl6fzrB2bz2OZa7l1byX/9bSf//cQuzp6Vw1XLijljWjYup3ociIhMBArkIseppXuAz/1hHWsOtHDzOdO46expw9Pt0FrY+zQ8+x9QuwnyF8BlP4VJZwz9a4nIkPrm9WNjUZFbXv7jcT0u7PSxM+dCduZcSHCgiRmNTzCr8a+cdeD7nHngBxxMX8GO7AvZl3EGEadviKse24JeF1cuK+bKZcXsbejkvrVVPLCuiie21ZOb4uWKJUVcubSY0kwNQxIRGc8UyEWOw666Tj79uzeo7+jnB1ct4LJFRUP/ItbCgefh2W9D1euQVqIJ20Rk1Or2ZLG+8FrWF15LZvdeZjX+jZmNjzO59SVCDh8H0k9lT+YqDqafwoArKdHljipTc5L5xkWz+Mp5M3h2Zz33vlHJT5/bx/+t3seKyZlctayYC+bm4XNr+TQRkfFGgVzkfXpqez0337OBoNfFvTeezKKSI9fmLS4ppaqy4oRe47QSJ/9xlpeVZS4q26P854v9/HrDVkLRq4GrT+i5RUSGW3NwKi8Fv8TLpZ+nqH0905qfYWrzc0xvfoawcVORdhJ7M89iX8YZmqn9MB6Xgwvm5nPB3Hxq23u5f20V962r5OZ7N5LysItLFxby4SVFLChK1URwIiLjhAK5yDGKRC3/++xefvjMbuYVpnLHdUvJS317F8yqygpuf3LXcb1Gbuc2Tqn4GWVtr9HtzmB10SfZkncZMy/y8t0TfQNHcct5M4bhWUVEYqxxUpm2jMq0Zaye/I/kd25lavOzTG1ezeTWl4jipCp1EXszz+JA+ml0+AoSXfKokZ/q50tnT+MLZ03ltf3N3Lu2knvXVvL718qZnB3k8kWFfGhRIUXpgUSXKiIiJ0CBXOQYNHT2cfM9G3llXzOXLSrkO5fNw+8Zuq6DOV07WVHxcya3vkSvK5UXyv6eTXkfIawxlyIyTljjpCZlATUpC3ih7GZyuncxtXk1U5ufZdX+/wb+m2Z/GeVpKziYfjJVKYs17hxwOAynTM3ilKlZ/HtviL9tqeXBDdV8/8ndfP/J3Zw8OYPLFxVx4by84V9uU0REhpwCuch7eGlPEzffu4Gu/jDfu2I+H1lSNDRdBa2lsGM9y6t+Q1nba/Q5k3m55HNsyL+KkEuT+IjIOGYMDUkzaUiaySulnyO95yBlba9Q1voq8+sfZHHt3YQdXqpSFnEwfQUH006h1V8KE7ybdqrfzdXLS7h6eQmVLT38eUM1D26o5qsPbOZfH97K+XPyuGxxIadPzdIs7SIiY4QCucg76A9H+OHTe/jZ8/uYlpPEH//uZKbnJh/TY99tqSOD5YKsem4u28dJaa009Hu4tXImd1aW0hGpBL4/RO9ARGRsaA2U0RooY0PBR3FG+ijq2EBZ66uUtr3KygM/AH5AhyeXqtQl3LDITVpvOW2+ktEf0I1jRMZ6ewpmkDTnLB7qPINHNtUQ6Wqle/tzdG19llDjgWN6jqLiEioryoe5UhEReSsFcpGj2FzVxlf+tInd9V1ctbSYb10y5311UT/aUkfuaD+ze95gUdcLpEcaaXdm8EzSh9kWXE7KZA9fHso3cIyOd6kjEZHhEnH6KE9fQXn6CgCS+2opa3uVkrbXKW17jV9e4of1V9DtzqQqZRHVqYuoSllMc2AymFHWKmyjxz2nyPEIR6McbOphZ12QA8mXkbL8MjKDHqbnJTM9J4m0gOcdH6s5RUREEkOBXOQw/eEIP3p6Dz9/YT/ZSV5+/YllnDUz54SeMyXcwoLul5jb/Ro+20utu4THUq5jj38B1mgJGxGRd9Ppy2dL3uVsybscrOVnH53N//7gOxR1bKCwfT0zmp8GoNeVSm3yPOqS51CXNIe6pNn0u1MTXP3IcjkcTM1JYmpOEr2hCLvrO9lV18mr+5p5dV8zOclepucmMy03iRSNNxcRGRUUyEXiXt3XzL8+vJW9DV1cubSIf754Nqn+4/vCYmyESX07mN/9CmX9O7EY9vjnsyF4JrXesqEtXERkojCG3c1RtuZdxta8y8BaUvprKezYQFH7evI7tzCp9WUMFoBWXzF1SbOpj4f0huD0EZ8o7t2GMI0U4w3gySqhJquUhs5MXtrbRLijkYHGckJNFdhQX6JLFBGZsBTIZcKraevl23/dwWObaylK9/PrTy7jrBnH2SrevI9bV3q5oe4/SI620+VIZU3yuWwJnEyXK/29Hy8iIsfOGDp8BXT4CtiRczEAnnAXOV07yevaRl7XNoo6NjCr6QkAIsZJi38STcFpNASnx69nDOta6EcbwpQ4YdpCTezu8rLbk0ZzSjZMWUKhL8SWvi6q23opTPMnukgRkQlFgVwmrL5QhF++uJ//W72PqLXcfM40PnvmFHzu99mNvKcFtj0Em++FyjX8yxkeKtz5rA5+mP2+2eqWLiIyggZcSVSlLaUqbengvmB/I7ld28nr2k52926K2tcxq/Fvg/d3ebIHA3pjYBrNgcm0+kuJOsZft+40d4Tl6T0sT++hecDJ7i4f+7p8ZJxzI6fe9izzClM5f04u58/JY2pO0ohMSiciMpEpkA8148Cs+kaiqzg2o23ymxESikS5b20lP35mL3UdfVwwJ49/vngWxRmBY3+SnhbY+RjseAT2rYZoCLJnwTm3UnzRLXzl/31m+N6AiIi8L93ebPZ7z2R/5pmD+3yhNrK7d8cve8ju3k1p22s4bQSAKE5a/cW0BCbT7J9Ec2AyzYHJtPlLiDjeeXK0sSTTE2FFRjcrMrr56j/9K//z4As8sa1ucI3zyVlBzp+bx/lz8phfmIrDoXAuIjLUFMiHmo1y+x13JrqKY3LLjTckuoQRFYlaHtlUzQ+e2kNFSw+LS9K4/aoFnDIl670fbC0074O9T8Puv8GBF8FGIK0ETvoMzL8S8uaDMdR0JmK+dBEReT/63GlUpi2nMm354D5ndID03oNk9uwns+cAmT37yerey5Tm53AQBQ4P6pNo9k8eN0E93FrD51ZO4XMrp1DX3sdT2+t4fFsdd7ywn58+t4/sZC8rp2dz1swcTpuWpUnhRESGiAK5jHt9oQh/3lDNHS/uZ39jN7PzU/jVJ5Zy1oycd++K19MCFa/CvmdjQbz1YGx/5jQ49SaYfSnkLxj96+CKiMgxiTg8NAWn0xScfsR+Z7Sf9N7ywZCe0bOfzO59TGl+/rCg7qDdV0RLoIxm/yRaApNYku/AHe0n5PAm4u0ct7xUH9etKOO6FWW09Qzw7M4Gnt3ZwBPb6vjTuipcDsPSsnRWzczhrBk56touInICFMjHExvFa/vxRnvxRntx2RBOwjhtCKcND35peNOHZrpgz9Pg9oHLD25/bNsdAFf82ukes4GzrWeAP7xWzm9eKaepq585BSn830cXc+HcvLd3u4tGoWU/1GyIhfDyV6BxR+w+dwAmnQErvghTz4GMSSP/ZkREJGEiDu+7BPWKeEg/QEa8db2s9RWcNswFNyZB7dfocKbT4sql2ZVLizs3tu3Opd8RTNA7OnZpAQ+XLy7i8sVFhCNR1le08ezOBp7b1cB3/rqT7/x1J0Xpfs6akcOpU7NYMTmT1IBaz0VEjpUC+RjhjvaTGmkmNRy7JEdaCUY7CUY6CEY68Ec78dr+waVejsUHrwrAXR9+94OcXvCngT/9KJc08KVBIAP8GRDIPLTteR/jsYeQtZaNlW3c/XoFf9lUS28owsoZ2dx4+mRWTMmM/YLf0wJNu2OXhp1QuxFqN8NAZ+xJPMlQvBzmXQGlp0DhEnCNrdYNEZH3azQsz3WsRm+tfmAOLjOLSf4eiir/ylf+7goyQvVkhOuY37MPtw0NHt3tSB4M5y2uXFrcebS4cuh2pIzKH8NdTgfLJ2WwfFIGX7twJjVtvaze1cDqnY08sL6K379WjsPAvMJUTp2axalTs1hSmv7+J0sVEZlAFMhHGXe0n6xQDVnhWjJDdWSG68gM1RGMdh5x3IDx0O1IoduZQoOnkB5HMv0OP30OP/3Gz4DDT8i4iRg3EVxEjJPIW2b7vv3fv8nGN16DUA+E+2LXoT4I90IofunvhN7WQ5e2ylh47W2FUPc7vxGX77CQnh7fjt9+c/vw+70psdDr8oPz/f9v2do9wCMbKrj79Qp2NvQScMGlZRE+UVzHTPsyrK+C1dWxVvDeliPrzJ0LC66C/IWxLug5s4+rBhGRsWx0Lc/1zm55+Y9jp9YbH+GM5HMP7bBRUiKtZITryQzVkxGOXWb2rMNnD60F3mf8tLjjLeqDQT2XDmfaqJqQtSDNz7UnlXLtSaUMhKNsrGzj5b1NvLKviTte2M9PntuH1+VgaVn6YOv53MJU3M7R8x5ERBJNqSOBjI2SHaomb6CcvFAluQMVZITrccRbuQeMh2ZXHgd8s2hzZdPuzKTdlUmbK5N+EzjhX8831UeheNnxP0G4H3rbYgG3pyV+3XzYduuhffXbYtu9rWCj7/68xhnvMu+LXbt88RZqE3+sBRulKwRPd0/hkb4FvBCeTRgX88x+vuN6lkucr5BU1QdVxEJ+aiGkFMKsD0LW9PhlKqSVgkO/3IuIyAgwDjpcmXS4Mjnom31ov7UEox3xlvR6MsP1ZITqmdy3jXnRNYOHxb4X5NPoLqDJXUCTO7Y94Ej82uEe16HW8y+fO52u/jBvHGjhpb1NvLy3ie89vgsAn9vBouJ0lpWls2xSBotK0kny6uuoiExcw/YJaIz5FfABoMFaOze+LwO4FygDDgJXWmtbh6uG0cZhI+QNlFM4sJ/C/n0UDBzAa/sB6HEkUecuZo9/AQ3uIprcBaPul/C3cXkhOTd2OVbRKPS3x0L7YHBvgYGueCt9X+z6zUv89t8efZje3h46TTLbA4vZFljKrsAiQg4vSQNNzG5+jMKGF7CN+3mo1/LLHktTj6W609LS2wHUA+uH67+EiIjI8TOGbmcq3c5UKjlynLov0j3Ykp4VqiUzVMu03k3M73l18Jh2ZzpN7oJDQd1VQJsrC5vA7xBJXhdnzczhrJk5ADR29rP2YAuvH2zhjYMt/O/qvUSfBafDMDs/haVl6SwviwX0vFRfwuoWERlpw/mT5G+A/wV+d9i+rwHPWGtvM8Z8LX77n4axhoRLDTdR2reL0v6dFPfvGQzgza5cdgaWUO2ZQo2njE5n+qgcLzbkHI5DY9Azp7zn4dGoZXttB1d/v505l36O2vY+LBD0OpmVlcT0vGQKUqdizArgK8Ne/rG45bwZiS5BRETGiT5nkBrnZGq8kw/ttJZgtJ3sUA3ZoRqyQrVkh6qZ1LdjcALXkHEPtqY3uIuo98R+7I+YxEy4lp3s5cJ5+Vw4Lx+Arv4w68tbB0P6H9dU8OuXDwKQm+JlQVEaC4rTWFScxryiVJK1zJqIjFPDFsittS8YY8resvtSYGV8+7fAc4yzQO40UNi/j8l925jct5WMcCMQ+/V6l38x5b4ZVHmm0OdMSnClo1M0atnX2MXrB1t4ZW8zr+xrorUnRPrKTxKKWpZNymByVpCcZK+WWBERkYnJGLqdaXQ7047o+u60ITJC9fGQXkN2uIapvZuZ1/MaEFuardmVR72niAZ3MQ3uIhrdBYQTsH56ktfFGdOzOWN6NgAD4SjbatrZWNnGpso2NlW18+T2+jffLlOyk+IhPZU5BSnMzEshqK7uIjIOjPQnWa61tja+XQe8Y19nY8yNwI0AJSUlI1DaEHj0y9R9JYmspv8lgpNK71Q2BU/joG8Wbc6sidEC/j61dg+wsbKNDRWtbKhsY2NFG539YQDyU32cPSuXU6dm8pHT53PTQ68luFoREZHRK2LcNHqKaPQUHdppLcmRVnJDVeSEqsgZqGRy33bm9rwOQBRDiyuXjEt98NrPYpOb5s0D78g2HHhcDhaVpLOoJH1wX1vPAJur2uMBvY3ndzfwwPoqIPaValJWkNn5KcwpSGV2QQpzClLIStKqKCIytiTsp0VrrTXGvOMaXdbaO4A7AJYuXXrsa3klUqiXv+0Jk37W31HuncmAQ2OgDtfRF2JXXSc7ajvYWNHGhso2DjTFZmp3GJiZl8IlCwtYVJLO4pI0JmUFB1vBI90TZqoBERGRoWMMna4MOl0Z7PXPj+2zlqRoOzkDlbGgPlDFeVNq4PF4p0XjgOxZULg4tvRn4ZKErECSFvAc0YpuraW2vY9tNR1sr+kYbFF/dHPt4GNykr3MzE9hek4S03OTmZabxLTcZE0cJyKj1kh/OtUbY/KttbXGmHygYYRff3hd9jM+fvnPuf2ihYmuJKFCkSgHmrrZUdvBrrpOdtV1srOuk+q23sFjspI8LCpJ5yNLi1hUnM78olR1PRMRERkJxtDlTKPLn8Z+/zwAbvnGDdj2GqjdBDXroXo97HwMNvw+9hiXHwoWxgN6PKinlY5o7z9jDAVpfgrS/Jw7+1Any/aeENtq29keD+o76zpZs7+Z/vChVV0K0/xMy42H9MPCesCj7x4iklgj/Sn0CHA9cFv8+uERfn0ZQh19IfY3drO/sSt23RS/buxmIBL7I+hyGKZkJ7GkNJ1rTy5hZl4yM/JSKEj1aQy4iIjIaJKSH7vMuCB221poPRAL59XrYpc3fgmvxtdMD2QdakF/M6gHMka87NSAm1OmZHHKlKzBfZGopaKlh931neyp72R3fRe76zt5ZW/z4HcUgKJ0P9Nzk5mak8TkrCCTsoJMyg6SnaS5akRkZAznsmd3E5vALcsYUwV8k1gQv88YcwNQDlw5XK8vx+Y9/9gYB67UHFwZRbgzi3BnFOLOKMKVUYgr6dAfXRuNEG6rI9RSTaipglDjQQYaDxJqrmJfNMyTw/w+RERE5AQYxzEFUJcD5uU4WF7oZHlhDScV1jMr+3Ec8cfuaY7wenWU12sivF4dYUNthP7I0JdbVFxCZUX5O97vdJhYuM4Kcv6cvMH94UiU8paeI0L6nvouXtrTdERQT/a6mJQde/zkrCQmZQcHA7t69InIUBrOWdaveYe7zh6u15T3zz77HQA6Qg72dvvY3+1lf7eXffHr8h4PA/bQOqZp7jCTg/1MCfYzOVA7uF0SGMDjsIAPmB6/DB2z6htD+nwiIiJyGBvl9id3va+HtANPAs+Fu8jp2kle1zbyMrZySf52rh2IjUqMGCdNgenUJc+mLmkOdclzaPWXYo3zhMo93iVGXU4HU7KTmJKdxAVzD+2PRC01bb3sb4r1/DvQ1M2Bpm7WHmzlkU012MNmM8pN8caCevZhrepZQYozAridiVv7XUTGJv3EN8H0RAwtAy5aBlykn30j174xib3dPur7D63v6TKWkkA/kwP9rMruiIXv+CXDMww/c4uIiMiYNeBKoiptKVVpSwf3BfsbyOvaHgvpnVuZ2fg4C+oeAKDfGaQhOJO65DnxkD6bLk9uQlejcToMxRkBijMCnBmfRO5NfaEIB5u7OdDYHQ/s3Rxo6uJvW2pp7QkNHueKP0dZZoBJWUlMygpQlhWkLDNIQZofp0Nd4EXk7RTIxyFroSviiAdvJy0h12AI74se+uU2ad45dIXh1MxOpgb7mZoUb+329+PWD7wiIiJynLq9Oezz5rAvc2Vsh42S0VtOXtc2cjtjQX1xzR9x2thSp93uTOqSZg+G9PqkWfS7U9/5BY6xi/1wc/iScWUU4s4owJ1RRHt6AbvSC3ClF+DwHFptx+NyUJoRC+iTs4KDQX1ydpCcZI1XF5nIFMjHuIiFlgEXTQMuGvtdNA64aOp3HxG8fY4oGZ4wU4N9ZHgiZHjCZHjCfPPzn6Qi3mVdREREZNgYBy2BSbQEJrE95wMAOKMDZHXvibeibye3axtTWl8cfEirr5i6pNnUx0N6Q3A6EWc85B5HF/uRZK2luz9CW+8Ad3zna3z9Oz/gQFM3B5u6eX53IwOHzQAf8DgpzXwzqAcoyzzUDT4j6FFYFxnnFMjHkHAUGgfc1Pe5aBhw09TvonnARZTYB7XTWLLiwTvLGybTEybDHcbvtInsBSYiIiLyNhGHh/rkOdQnz2FTfmyfJ9xFbteOeEv6Noo6NjCr6YnY8cZJU2Aq9Ulz+ORCN5nde2kJTDrh8ejDwRhDks9Fks9F16Yn+MZFswbve3O8+sHmWEDfHw/q22s7eHxbHZHooQHryT7XYDg/PKiXZQVJ9buP9tIiMsYokI9SUQutISf1fW7q+t3U98cC+Jvh2++MkOMJU5LWQ7YnTLY3RJo7goYniYiIyPG49dZbE13CWyQBy8nz9LE4tY3FKW0sSWlmccoj/OpSP2y8hgGHn8bgdBqSZtIQnEFD0kxa/JOIOkbRV9z3073e4cSVmosrvQB3RgGd6QU0pBewPr0AZ2oOxhzqARnpaSfUUk24tYZQa03suqWacGstNtR3XKW+1+z1IjL0RtGn1cTWHzHU9rmp6XNT2+ehod81OLu5x0TJ8YVYnNZDrjdEni9Ekiv6Hs8oIiIicuy+ef3oXwhnI7DRRvnp1z/Dj/7n/5HXtZ2c7l3MqX+ERdFeAMLGQ1NwKg3BmTQkzaAhOJOm4BQiDm9iih6i7vXhSJT23hBtvSHaekK09aTQ1pNHa+88ut+ytlzQ4yQt4CEt4CYt4CY94CHN7ybV78b1LjPBH+/s9SJy/BTIE6Qz7KCm101Nn4eaPjdNAy7AYLBke8PMTO4bDN/p7oi6nIuIiIgAGAd7WqLszLmInTkXxXbZCGm9leR07yKnayc53TuZ3vQU8+sfBGLd3ZsDk+MhPdaa3hicTtjpT+Q7eV9cTgeZSV4yk97+w0IoEo2H9AFae2PXbT0h9jd20xs6Mqwn+1yxoO73kB5wDwb3FJ+6wIskggL5CLAWOsJOKnvdVPd6qO7z0BmOjXdymyj5vhAnp3dT4BsgzxfSDOciIiIi74M1TloDZbQGytiVfX58pyWlv4acrl3kdO8kp2snk1tfYm7DX2J3Y2jxl9EQnE5j0gwaA9NoCk6lx52Z0CXYjofb6SA72Ut28tvDen84QltPiNZ4SG+LB/bd9Z30Hza5nDFQ8Hd38Ilfvz44Xj02G3yAwjT/u7asi8jxUyAfJh0hB1W9Hqr6PFT1HgrgfmeEQl+IRandFPhDZHvCGvctIiIiMtSMocNXSIevkL1Zq2L7rCU40Ehu906yu3aR073riInjAHrc6TQFptIYnEZTYCpNwWk0ByYlrsv7CfK6nOSmOMlN8R2x31pLXyhKW+8ArfHW9ee2v0BDx0xeP9BCz8ChlnWXw1CU7qc0MxbQSzNjM8KXZgYpTg/gcSmsixwvBfIh9JdNNWRc8CV+U55Jezj2n9bniFLkH2BJWjdF/gEy1P1cREREJDGModubw35vDvszzhjc7Q21k92zl6zuPWR17yW7Zw/z6x7AHe0HIIqTVn8JTcGpgy3pTcFpdHpyx1xr+puMMfg9TvweP/mpsa779z/yPf768Hex1tLY2c/B5h4ONndT3tzNweYeypu7WV/eSmd/ePB5HAYK0vyUZQYpzQwcus4KUpIRwOcefbPgi4wmCuRD6JcvHSAw41QyPWEWpPZQ5A+R5QmP1c9pERERkQmh351KVeoSqlKXDO4zNkJqXxXZ3fGg3rOXvM5tzGh6avCYPmcSTYMt6VNpDkymOTCZfldKIt7GkDHGkJPiIyfFx/JJGUfcZ62lpXtgMKAffv3YllraekJHHJ+f6jssqMdCenGGn6L0AOkBt9ZZlwlPgXwI/eK6JeSmncEtP/9Foks5NsaBWfWNRFchIiIi8r6N/DJtycBikp3zmJXUyZykTuYkdTAn+SBzkraw0HWo1bi238uurmR2difxd4vdFHRsojkwacwHdYiF9Tcnl1tSmv62+9t6BigfbFk/dP30jnqaugaOODbocVKUfiigF6X7j7j9bmutF5eUUlVZMeTvbzhoOTl5NwrkQygnxQd2DC1HZqPcfsedia7imNxy4w2JLkFERERGkdGwTFsn8BrwmrUkR1rJDNeRGaojM1zPzKQ6TgnX8NkSP2z5NABd7qzBVvSWwCSaA1PiLerJCX0fQyk2a7uHBcVpb7uvsy9EZUsvla09VLX2UtkSu65q7eHVfc10Dxw5I3yKz3VEQC+OB/aiDD/V9U1DspzcSNBycvJuFMhFRERERE6EMXS6Muh0ZXDQN/vQfhvlR1+9kR/8/E4ye/aT2buPzJ79zKv/M+5o3+BhXZ5smv2TBsN6LLCPr6AOkOxzM7vAzeyCt/cUsNbS1hOKBfXWHqoOC+37G7t5fncjfaFDDV8lX76Pnz63j2Sfi2SfiySfi2SfmxRv7DrZ5yLodeHU7MkyyimQy5gx8l3TRERERE6AcVDebvnkj59+cwcwBcNkin29zAx2MjOpk5nBLmYGdzEjaR1B56FW4po+L7u6k9nTk8Tu7kOXugFv/LmGvt7RPKbbEUjFlZobu6RksfK6W+jsC9PVH6a+o/9ta64DBL1Okr3uweD+ZlhP9sZCvN/tHNXvWcY/BXIZM0ZD17RjccvLf0x0CSIiIjKKvNd3mPr45XkbJSXSGu/2Xkemv45ZyQ2cEq7Fa/sHj+83XlpdubS4cmhx5wxut7myiJrj/3p/y8t/HFPdwFf+521H7AtFonT1henoC9HVH6az781LiMbOfvY3dROJ2iMe43SYQ2H9HYK71mCX4aRALiIiIiIyGhgHHa5MOlyZHGDOof3WEox2kBFuID1UT0a4gYxwPUUDe5ndu3bwsCgO2lyZh8K6Kx7W3Tn0OwIJeEMjy+10kB70kB70HPV+ay29ocgRQf3w4F7e1f22cewAfrfzUNf4w7rEvxncAx4nDrWyy3FSIBcRERERGc2ModuZSrczlUrvtCPuckf7SQ83xMJ6uJ6MUGy7tG8HLg6Fy25H0hGt6m2ubFqd2XS4Mk6oVX0sMcYQ8LgIeFzkvsOE95GojYf00BHBvbM/TFtPiIqWHkKRI1vZHYYjgnps+9Bt4/GPwLuTsWpi/OsTERERERmHQg4vDZ5iGjzFR+w3NkpKpIWMcD3p8ZCeEa5nWu8m/D09g8dFcdDhTKfNlY37Qh9za+6m1VdCm7+YDm8BUcfEigtOhyHV737HJdestQyEo3T2x7vG9x3WNb4/RHVbL139Yexhmb3ky39i3jefoCDNT0GaL34d306Nbeel+nCra/yENLH+hYmIiIiITADWOGh3ZdHuyuKAb84R9/kiXaSHm0iLNJIWbiIt3Eh6uJHr5rtJPXD74HFRnHT48mn1FdPmL6Ytft3qK6HTmz/hwjrEWtm9bidet5OsJO9Rj4laS3d/eLA7/B9//B2u/8at1LT3UdPWy8bKNlp7Qm95XshJ9g6G9cI0P/mpvsHtgjQ/6QG3JqAbhybevyIRERERkQmsz5lErTOJWsqO2H/Lv9zATx99lbS+StL7KknrrSStt4K0vkoKOzbhiR5qWY8YJ53ePDq8BbT7Co647vAV0O3OjKXMCchhTLy7upv8VOh4/UFuvfSBI47pGQhTGw/oNW29VLfFtmvbe9le08FT2+sZCEePeIzP7RhsUR9saX/LbZ/bOZJvVYaAArmIiIiIiADQ68mg15NBbcqCI++wlkCohbS+WEhP76skpa+G1L4aJre8SDDUcsThYYeXDm8e7fGAfiiw59PlyaHHk4E1Ezc8BjwupmQnMSU76aj3W2tp7h6gtq2P6nhojwX22O3ndjXS0Nn/tsdlBD1HdIUvTPOTn3aopT07yYtDa7OPKgrkIiIiIiLy7oyhx5NJjyeTmpSFb7vbFekjpb+GlL5aUvurY2G9v5aUvhryurbjD7cfcXwUJ92eTDq9uXR5cujy5NDpjV13eXPp9OTQ7cki6jj6WO7xzhhDVpKXrCQv84pSj3rMQDhKfceRgf3NbvEHm7t5eW/T22aNdzsNuSmHd4V/e0t7sm9i/jdPFAVyERERERE5IWGnj5bAZFoCk496vyfcRUp/Dcn99ST3N5A00EDSQD1J/Y1k9uyjrPUVPNHetz2u251BlyebXnc6Pe6Mwct1892Utr5CrzuDbncGve70CRfePS4HxRkBijOOvqSdtZaOvjC17W/pFt/WS01bH68faKGuo+9ta7Mn+1zxgO4jL9VPdpKHrOTYjwOZwfh20EuK36Ux7UNAgVxERERERIbVgCuJJtd0moLTj36AtXgi3SQP1JMUD+yx4F5PcKCZQKiF9N5ygqEWXNF+zrjMD9tvOuIp+lwp9MSDe68rlX5XCn2uZPrjlz5XyqFr56F9wx7kjWP0BlfjwBlMx5mSjSslG3daHh3BdOpTstmcko0rOQtHIAVj3j4DvI2EiPS0E+luI9rdRqSnLbbd206kt4toXyfRvi6ivZ1E+7uI9nZhw2/vZn+8SosLOVhRNWTPlygJCeTGmAuAHwFO4JfW2tsSUYeIiIiIiIwCxjDgSqLZlURzYMo7H2ct7kgPP7x2Cf/5i3sJhFoJhFpil4GW+O1m0vsq8IU78YY7cEffPQSGHD76XCkMOIOEnD5CjgAhp5+Qw0fIGd92+gk5/Efe7/QTdniJGDcRh+ct14e2PY4ot/1155iY5O6W82Zgn/1O/FYPUEE4Cq0hF00DLpoHXDT1x7ab4tvNAzk0DRQM3jdg33n5No+JkuKOkOaOkHrYdYorQpIrQsAZJckZIeCMxK5dEYLOKEFnmKDz0LbHREm/5N/B2jHx3/XdjHggN8Y4gf8DzgWqgDeMMY9Ya7ePdC0iIiIiIjKGGEPIFWRfq337xHPvwBkdwBvuxBvuxBfuiG1HDm2/ee2J9OCO9OCO9hEcaBrcjl334rSR936xo/jCv6TAK8uJGBcR4yHqcGExRI0TMFjjIIoDa+IX3n4dNQ6IX1vjAAxwqKu5sRawGOzgfmOJb8dvx7cP7T98X+z2BZ8Pwut3xJ/VgrW4gGws2YOLq9v4w+2R226LdUInPtqtn/ZokHaCtNsA7TZAmw3SboO0R4K0h4O09QapswF2EdvfjZ8ox7YWu4swpTffQ3tbC6npme/zjIwuiWghXw7stdbuBzDG3ANcCiiQi4iIiIjIkIo4PIMT0p0IRzQ0GM7dkT48kR5c0X4cNoQzGsJpB+LXIZzRgcHrJ+78Lh/8xJcG9zlsBGMjGCwOGwFrMURx2CiGKOZo14dtO+LXFgPGxK7j3r7PHNoPh91nDsV58+Ztw5aG3cyanRN7nDnsOcybz/Xu2wZDCpBiDMWDrx8C0w50vMvjYzX2WxddUQ891k1X1EtP1E2X9cSuo266o4e2n9hcg99/yQmd09HAWGvf+6ihfEFjrgAusNZ+On77OuAka+0X33LcjcCN8ZszgF0jWuixyQKaEl2EvC86Z2OLztfYovM19uicjS06X2OPztnYovM1toyl81Vqrc0+2h2jdlI3a+0dwB3veWACGWPWWmuXJroOOXY6Z2OLztfYovM19uicjS06X2OPztnYovM1toyX83VsnfSHVjXEezDEFMX3iYiIiIiIiEwYiQjkbwDTjDGTjDEe4GrgkQTUISIiIiIiIpIwI95l3VobNsZ8EXiC2LJnv7LWbhvpOobIqO5SL0elcza26HyNLTpfY4/O2dii8zX26JyNLTpfY8u4OF8jPqmbiIiIiIiIiCSmy7qIiIiIiIjIhKdALiIiIiIiIpIACuTHyRhzgTFmlzFmrzHma4muR97OGPMrY0yDMWbrYfsyjDFPGWP2xK/TE1mjHGKMKTbGrDbGbDfGbDPG3BTfr3M2ChljfMaY140xm+Ln69b4/knGmDXxz8Z745N3yihhjHEaYzYYYx6N39b5GsWMMQeNMVuMMRuNMWvj+/SZOEoZY9KMMfcbY3YaY3YYY1bofI1OxpgZ8X9Xb146jDE363yNbsaYL8e/c2w1xtwd/y4y5v+OKZAfB2OME/g/4EJgNnCNMWZ2YquSo/gNcMFb9n0NeMZaOw14Jn5bRocw8A/W2tnAycAX4v+udM5Gp35glbV2AbAQuMAYczLwXeAH1tqpQCtwQ+JKlKO4Cdhx2G2dr9HvLGvtwsPW2tVn4uj1I+Bxa+1MYAGxf2s6X6OQtXZX/N/VQmAJ0AM8hM7XqGWMKQT+HlhqrZ1LbHLwqxkHf8cUyI/PcmCvtXa/tXYAuAe4NME1yVtYa18AWt6y+1Lgt/Ht3wIfGsma5J1Za2uttevj253EvsgUonM2KtmYrvhNd/xigVXA/fH9Ol+jiDGmCLgY+GX8tkHnayzSZ+IoZIxJBc4A7gSw1g5Ya9vQ+RoLzgb2WWvL0fka7VyA3xjjAgJALePg75gC+fEpBCoPu10V3yejX661tja+XQfkJrIYOTpjTBmwCFiDztmoFe/+vBFoAJ4C9gFt1tpw/BB9No4uPwS+CkTjtzPR+RrtLPCkMWadMebG+D59Jo5Ok4BG4NfxYSG/NMYE0fkaC64G7o5v63yNUtbaauD7QAWxIN4OrGMc/B1TIJcJy8bW/NO6f6OMMSYJeAC42Vrbcfh9Omeji7U2Eu/uV0Ss59DMxFYk78QY8wGgwVq7LtG1yPtymrV2MbEhcl8wxpxx+J36TBxVXMBi4KfW2kVAN2/p7qzzNfrExxtfAvzprffpfI0u8fH8lxL78asACPL2oaljkgL58akGig+7XRTfJ6NfvTEmHyB+3ZDgeuQwxhg3sTB+l7X2wfhunbNRLt4tczWwAkiLdyUDfTaOJqcClxhjDhIbZrWK2HhXna9RLN4ihLW2gdj41uXoM3G0qgKqrLVr4rfvJxbQdb5GtwuB9dba+vhtna/R6xzggLW20VobAh4k9rdtzP8dUyA/Pm8A0+Kz+nmIdXV5JME1ybF5BLg+vn098HACa5HDxMez3gnssNbefthdOmejkDEm2xiTFt/2A+cSG/e/GrgifpjO1yhhrf26tbbIWltG7G/Ws9baa9H5GrWMMUFjTPKb28B5wFb0mTgqWWvrgEpjzIz4rrOB7eh8jXbXcKi7Ouh8jWYVwMnGmED8O+Ob/8bG/N8xE+uNIe+XMeYiYuPxnMCvrLXfTmxF8lbGmLuBlUAWUA98E/gzcB9QApQDV1pr3zrxmySAMeY04EVgC4fGuH6D2DhynbNRxhgzn9jkKU5iP+7eZ639d2PMZGItsBnABuBj1tr+xFUqb2WMWQl8xVr7AZ2v0St+bh6K33QBf7TWftsYk4k+E0clY8xCYpMmeoD9wCeJfz6i8zXqxH/oqgAmW2vb4/v072sUiy+xehWxlXk2AJ8mNmZ8TP8dUyAXERERERERSQB1WRcRERERERFJAAVyERERERERkQRQIBcRERERERFJAAVyERERERERkQRQIBcRERERERFJAAVyERGRCcAY8yFjjDXGzEx0LSIiIhKjQC4iIjIxXAO8FL8WERGRUUCBXEREZJwzxiQBpwE3AFfH9zmMMT8xxuw0xjxljPmrMeaK+H1LjDHPG2PWGWOeMMbkJ7B8ERGRcUuBXEREZPy7FHjcWrsbaDbGLAEuB8qA2cB1wAoAY4wb+DFwhbV2CfAr4NuJKFpERGS8cyW6ABERERl21wA/im/fE7/tAv5krY0CdcaY1fH7ZwBzgaeMMQBOoHZkyxUREZkYFMhFRETGMWNMBrAKmGeMscQCtgUeeqeHANustStGqEQREZEJS13WRURExrcrgN9ba0uttWXW2mLgANACfDg+ljwXWBk/fheQbYwZ7MJujJmTiMJFRETGOwVyERGR8e0a3t4a/gCQB1QB24E/AOuBdmvtALEQ/11jzCZgI3DKiFUrIiIygRhrbaJrEBERkQQwxiRZa7uMMZnA68Cp1tq6RNclIiIyUWgMuYiIyMT1qDEmDfAA/6EwLiIiMrLUQi4iIiIiIiKSABpDLiIiIiIiIpIACuQiIiIiIiIiCaBALiIiIiIiIpIACuQiIiIiIiIiCaBALiIiIiIiIpIA/x98Yutvwu8WXwAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 1224x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(17,4))\n",
    "sns.histplot(data=data, x='Age', hue='Survived', kde=True);"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Форма распределения выживших похожа на таковую для погибших. Но колокол ниже по высоте, поскольку выживших меньше чем погибших. Т.е. можно сказать, что в среднем люди равномерно погибали для каждой возростной группы. Отдельно выделяются дети до 10 лет - у них больше выживших чем погибших. Наверное им отдавали приоритет в спасательных шлюпках. Также практически нет выживших старше 65 лет. но их в целом мало, чтобы делать какие-то выводы."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**(1 point)** Explore \"passenger class\" and \"embarked\" features. What class was \"the safest\"? Is there any relationship between the embarkation port and the survival? Provide the corresponding visualizations."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-02-26T07:20:29.013073Z",
     "start_time": "2021-02-26T07:20:28.807976Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.countplot(data=data, x='Pclass', hue='Survived');"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "В абсолютных числах число выживших примерно одинаковое между классами. Но в относительном исчислении больше всего выживших выше в первом классе (выжевших даже больше чем погибших). Хуже всего дела обстоят в третьем классе - в нем погибших сильно больше выживших. Возможно это связано с тем, что каюты третьего класса распологались на нижних палубах (и добраться до шлюпок им было сложнее), либо же места в шлюпках им выделялись в последнюю очередь."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-02-26T07:20:31.641764Z",
     "start_time": "2021-02-26T07:20:31.486112Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.countplot(data=data, x='Embarked', hue='Survived');"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Embarked это пункт посадки пассажиров. На первый взгляд он не должен влиять на статистику выживаемости. Но между пунктами явно наблюдается дисбаланс: в абсолютных числах больше всего выживших из пункта S, а в относительных из пункта C. Можно предположить, что пункт S - это начальная точка отправления Титаника и там больше всего пассажиров село на борт. В портах С и Q пассажиров добрали. Если посмотреть, на посадку классов, по пунктам посадки, то видно что в C село много пассажиров первого класса, чем и объясняется выживаемость пассажиров из пункта C. Возможно C это какой-то богатый город, где много людей могут себе позволить билет первого класса:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-02-26T07:20:34.042428Z",
     "start_time": "2021-02-26T07:20:33.877894Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.countplot(data=data, x='Embarked', hue='Pclass');"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Modelling"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**(0.5 points)** Find the percentage of missing values for each feature. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-02-26T07:21:04.030522Z",
     "start_time": "2021-02-26T07:21:04.017781Z"
    }
   },
   "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>null_cnt</th>\n",
       "      <th>%</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>column</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Cabin</th>\n",
       "      <td>687</td>\n",
       "      <td>77.10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Age</th>\n",
       "      <td>177</td>\n",
       "      <td>19.87</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Embarked</th>\n",
       "      <td>2</td>\n",
       "      <td>0.22</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Fare</th>\n",
       "      <td>0</td>\n",
       "      <td>0.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Ticket</th>\n",
       "      <td>0</td>\n",
       "      <td>0.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Parch</th>\n",
       "      <td>0</td>\n",
       "      <td>0.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SibSp</th>\n",
       "      <td>0</td>\n",
       "      <td>0.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Sex</th>\n",
       "      <td>0</td>\n",
       "      <td>0.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Name</th>\n",
       "      <td>0</td>\n",
       "      <td>0.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Pclass</th>\n",
       "      <td>0</td>\n",
       "      <td>0.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Survived</th>\n",
       "      <td>0</td>\n",
       "      <td>0.00</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          null_cnt      %\n",
       "column                   \n",
       "Cabin          687  77.10\n",
       "Age            177  19.87\n",
       "Embarked         2   0.22\n",
       "Fare             0   0.00\n",
       "Ticket           0   0.00\n",
       "Parch            0   0.00\n",
       "SibSp            0   0.00\n",
       "Sex              0   0.00\n",
       "Name             0   0.00\n",
       "Pclass           0   0.00\n",
       "Survived         0   0.00"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "temp = data.isna().sum().sort_values(ascending=False).rename('null_cnt').to_frame()\n",
    "temp['%'] = (data.isna().mean()*100).round(2)\n",
    "temp.index.name = 'column'\n",
    "temp"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Ну и сделаем это красиво :)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-02-26T07:21:09.548136Z",
     "start_time": "2021-02-26T07:21:09.000855Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1224x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(17,8))\n",
    "sns.heatmap(data.isnull(), cbar = False).set_title('Null values');"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Think about the ways to handle these missing values for modelling and write your answer below. Which methods would you suggest? What are their advantages and disadvantages?\n",
    "\n",
    "1. Лучше всего использовать модель, которые из коробки могут работать с пропусками (LGBM или Catboost) - они сами с ними разберутся, а потом красиво отобразят в каком-нибудь SHAP.\n",
    "2. Если для линеек и пр. (которые не умеют работать с пропусками по определению), то:\n",
    " - Cabin я бы вовсе удалил, слишком много пропусков.\n",
    " - Age я бы попробовал заполнить с помощью какой либо другой модели. Либо попрбовал один из типовых ипутеров - какой-лучше бы показал себя, тот бы и оставил.\n",
    " - Embarked - медианой - пропусков слишком мало, погоды это не сделет."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**(1.5 points)** Prepare the features and train two models (KNN and Logistic Regression) to predict the survival. Compare the results. Use accuracy as a metric. Don't forget about cross-validation!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-02-24T17:31:18.076613Z",
     "start_time": "2021-02-24T17:31:18.028615Z"
    }
   },
   "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>Pclass</th>\n",
       "      <th>Sex</th>\n",
       "      <th>Age</th>\n",
       "      <th>SibSp</th>\n",
       "      <th>Parch</th>\n",
       "      <th>Fare</th>\n",
       "      <th>Embarked_C</th>\n",
       "      <th>Embarked_Q</th>\n",
       "      <th>Embarked_S</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PassengerId</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>22.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>7.2500</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>38.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>71.2833</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>26.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>7.9250</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>35.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>53.1000</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>35.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>8.0500</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             Pclass  Sex   Age  SibSp  Parch     Fare  Embarked_C  Embarked_Q  \\\n",
       "PassengerId                                                                     \n",
       "1                 3    0  22.0      1      0   7.2500           0           0   \n",
       "2                 1    1  38.0      1      0  71.2833           1           0   \n",
       "3                 3    1  26.0      0      0   7.9250           0           0   \n",
       "4                 1    1  35.0      1      0  53.1000           0           0   \n",
       "5                 3    0  35.0      0      0   8.0500           0           0   \n",
       "\n",
       "             Embarked_S  \n",
       "PassengerId              \n",
       "1                     1  \n",
       "2                     0  \n",
       "3                     1  \n",
       "4                     1  \n",
       "5                     1  "
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y = data['Survived']\n",
    "X = data[['Pclass','Sex','Age','SibSp','Parch','Fare','Embarked']].copy()\n",
    "X['Sex'] = X['Sex'].map({'male':0,'female':1})\n",
    "X = pd.get_dummies(X, columns = ['Embarked'])\n",
    "X = X.fillna(X.median())\n",
    "\n",
    "X.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-02-24T18:21:45.776027Z",
     "start_time": "2021-02-24T18:21:45.715028Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0.79 0.79 0.8 ] ~ 0.79\n"
     ]
    }
   ],
   "source": [
    "pipeline = make_pipeline(StandardScaler(), KNeighborsClassifier())\n",
    "score = cross_val_score(pipeline, X, y, scoring='accuracy', cv=3)\n",
    "score = np.array(score)\n",
    "print(score.round(2), '~', round(score.mean(),2))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-02-24T17:31:19.322002Z",
     "start_time": "2021-02-24T17:31:19.190004Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0.79 0.79 0.79] ~ 0.79\n"
     ]
    }
   ],
   "source": [
    "score = cross_val_score(LogisticRegression(max_iter=500), X, y, scoring='accuracy', cv=3)\n",
    "score = np.array(score)\n",
    "print(score.round(2), '~', round(score.mean(),2))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**(0.5 + X points)** Try more feature engineering and hyperparameter tuning to improve the results. You may use either KNN or Logistic Regression (or both)."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Лучший скор на Кэгле - **0.79186**. Ник на Кэгле - slivka83."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Load the test set and make the predictions. Submit them to kaggle and see the results :)\n",
    "Select the best model, load the test set and make the predictions. Submit them to kaggle.\n",
    "\n",
    "**Note**. X points will depend on your kaggle leaderboard score.\n",
    "$$ f(score) = 0.5, \\ \\ 0.79 \\leq score < 0.81,$$\n",
    "$$ f(score) = 1.0, \\ \\ 0.81 \\leq score < 0.83,$$ \n",
    "$$ f(score) = 2.5, \\ \\ 0.83 \\leq score $$ \n",
    "Your code should generate the output submitted to kaggle. Fix random seeds to make the results reproducible."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-02-24T21:58:55.462483Z",
     "start_time": "2021-02-24T21:58:55.447485Z"
    }
   },
   "source": [
    "решение:\n",
    "https://github.com/slivka83/share/blob/master/titanic-0.79186.ipynb"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Part 2: Porto Seguro’s Safe Driver Prediction"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's build a model that predicts the probability that a driver will initiate an auto insurance claim in the next year.\n",
    "Read the detailed description and download the dataset https://www.kaggle.com/c/porto-seguro-safe-driver-prediction/data. Put the data into *./data/porto/*."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-02-26T07:39:48.693939Z",
     "start_time": "2021-02-26T07:39:45.786003Z"
    }
   },
   "outputs": [],
   "source": [
    "insurance_data = pd.read_csv('porto/train.csv').set_index('id')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-02-26T07:40:00.851764Z",
     "start_time": "2021-02-26T07:40:00.834903Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(595212, 58)\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>target</th>\n",
       "      <th>ps_ind_01</th>\n",
       "      <th>ps_ind_02_cat</th>\n",
       "      <th>ps_ind_03</th>\n",
       "      <th>ps_ind_04_cat</th>\n",
       "      <th>ps_ind_05_cat</th>\n",
       "      <th>ps_ind_06_bin</th>\n",
       "      <th>ps_ind_07_bin</th>\n",
       "      <th>ps_ind_08_bin</th>\n",
       "      <th>ps_ind_09_bin</th>\n",
       "      <th>...</th>\n",
       "      <th>ps_calc_11</th>\n",
       "      <th>ps_calc_12</th>\n",
       "      <th>ps_calc_13</th>\n",
       "      <th>ps_calc_14</th>\n",
       "      <th>ps_calc_15_bin</th>\n",
       "      <th>ps_calc_16_bin</th>\n",
       "      <th>ps_calc_17_bin</th>\n",
       "      <th>ps_calc_18_bin</th>\n",
       "      <th>ps_calc_19_bin</th>\n",
       "      <th>ps_calc_20_bin</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>9</td>\n",
       "      <td>1</td>\n",
       "      <td>5</td>\n",
       "      <td>8</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>7</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>9</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "      <td>4</td>\n",
       "      <td>9</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>4</td>\n",
       "      <td>2</td>\n",
       "      <td>7</td>\n",
       "      <td>7</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>4</td>\n",
       "      <td>9</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 58 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "    target  ps_ind_01  ps_ind_02_cat  ps_ind_03  ps_ind_04_cat  ps_ind_05_cat  \\\n",
       "id                                                                              \n",
       "7        0          2              2          5              1              0   \n",
       "9        0          1              1          7              0              0   \n",
       "13       0          5              4          9              1              0   \n",
       "16       0          0              1          2              0              0   \n",
       "17       0          0              2          0              1              0   \n",
       "\n",
       "    ps_ind_06_bin  ps_ind_07_bin  ps_ind_08_bin  ps_ind_09_bin  ...  \\\n",
       "id                                                              ...   \n",
       "7               0              1              0              0  ...   \n",
       "9               0              0              1              0  ...   \n",
       "13              0              0              1              0  ...   \n",
       "16              1              0              0              0  ...   \n",
       "17              1              0              0              0  ...   \n",
       "\n",
       "    ps_calc_11  ps_calc_12  ps_calc_13  ps_calc_14  ps_calc_15_bin  \\\n",
       "id                                                                   \n",
       "7            9           1           5           8               0   \n",
       "9            3           1           1           9               0   \n",
       "13           4           2           7           7               0   \n",
       "16           2           2           4           9               0   \n",
       "17           3           1           1           3               0   \n",
       "\n",
       "    ps_calc_16_bin  ps_calc_17_bin  ps_calc_18_bin  ps_calc_19_bin  \\\n",
       "id                                                                   \n",
       "7                1               1               0               0   \n",
       "9                1               1               0               1   \n",
       "13               1               1               0               1   \n",
       "16               0               0               0               0   \n",
       "17               0               0               1               1   \n",
       "\n",
       "    ps_calc_20_bin  \n",
       "id                  \n",
       "7                1  \n",
       "9                0  \n",
       "13               0  \n",
       "16               0  \n",
       "17               0  \n",
       "\n",
       "[5 rows x 58 columns]"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print(insurance_data.shape)\n",
    "insurance_data.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's skip EDA for now. We'll use SGDClassifier and build a simple baseline: https://scikit-learn.org/stable/modules/generated/sklearn.linear_model.SGDClassifier.html.\n",
    "Note that loss='log' gives logistic regression."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-02-26T07:41:06.920791Z",
     "start_time": "2021-02-26T07:41:06.836005Z"
    }
   },
   "outputs": [],
   "source": [
    "y = insurance_data['target']\n",
    "X = insurance_data.drop(columns=['target'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-02-26T07:46:35.819593Z",
     "start_time": "2021-02-26T07:46:35.813291Z"
    }
   },
   "outputs": [],
   "source": [
    "numeric_features = [column for column in X if not (column.endswith(\"cat\") or column.endswith(\"bin\"))]\n",
    "numeric_transformer = Pipeline(steps=[('scaler', StandardScaler())])\n",
    "\n",
    "categorical_features = [column for column in X if (column.endswith(\"cat\") or column.endswith(\"bin\"))]\n",
    "categorical_transformer = Pipeline(steps=[('encoder', OneHotEncoder(handle_unknown='ignore'))])\n",
    "\n",
    "preprocessor = ColumnTransformer(\n",
    "    transformers=[\n",
    "        ('num', numeric_transformer, numeric_features),\n",
    "        ('cat', categorical_transformer, categorical_features)])\n",
    "\n",
    "pipeline = Pipeline(steps=[('preprocessor', preprocessor),\n",
    "                           ('classifier', SGDClassifier(loss='log', alpha=0.001, n_jobs=-1, random_state=14))])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-02-26T07:46:37.872266Z",
     "start_time": "2021-02-26T07:46:37.310039Z"
    }
   },
   "outputs": [],
   "source": [
    "X_train, X_valid, y_train, y_valid = train_test_split(X, y, test_size=0.3, random_state=42)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-02-26T07:46:45.180714Z",
     "start_time": "2021-02-26T07:46:40.110715Z"
    }
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/python3/lib/python3.6/site-packages/sklearn/preprocessing/data.py:625: DataConversionWarning: Data with input dtype int64, float64 were all converted to float64 by StandardScaler.\n",
      "  return self.partial_fit(X, y)\n",
      "/opt/python3/lib/python3.6/site-packages/sklearn/base.py:465: DataConversionWarning: Data with input dtype int64, float64 were all converted to float64 by StandardScaler.\n",
      "  return self.fit(X, y, **fit_params).transform(X)\n",
      "/opt/python3/lib/python3.6/site-packages/sklearn/linear_model/stochastic_gradient.py:166: FutureWarning: max_iter and tol parameters have been added in SGDClassifier in 0.19. If both are left unset, they default to max_iter=5 and tol=None. If tol is not None, max_iter defaults to max_iter=1000. From 0.21, default max_iter will be 1000, and default tol will be 1e-3.\n",
      "  FutureWarning)\n"
     ]
    }
   ],
   "source": [
    "clf = pipeline.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We'll calculate the model accuracy:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-02-26T07:46:54.721392Z",
     "start_time": "2021-02-26T07:46:53.493250Z"
    }
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/python3/lib/python3.6/site-packages/sklearn/pipeline.py:451: DataConversionWarning: Data with input dtype int64, float64 were all converted to float64 by StandardScaler.\n",
      "  Xt = transform.transform(Xt)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "0.9631224658945812"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "accuracy_score(y_true=y_valid, y_pred=clf.predict(X_valid))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Wow! Our model gets ~0.963 accuracy! But is it really good?...\n",
    "\n",
    "Let's plot the confusion matrix and analyze the errors."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "disp = plot_confusion_matrix(clf, X_valid, y_valid,\n",
    "                             cmap=plt.cm.Blues,\n",
    "                             normalize=None)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Well, it turns out that our model is completely useless. Let's calculate some basic statistics:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-02-26T07:47:03.704003Z",
     "start_time": "2021-02-26T07:47:01.727872Z"
    },
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>target</th>\n",
       "      <th>ps_ind_01</th>\n",
       "      <th>ps_ind_02_cat</th>\n",
       "      <th>ps_ind_03</th>\n",
       "      <th>ps_ind_04_cat</th>\n",
       "      <th>ps_ind_05_cat</th>\n",
       "      <th>ps_ind_06_bin</th>\n",
       "      <th>ps_ind_07_bin</th>\n",
       "      <th>ps_ind_08_bin</th>\n",
       "      <th>ps_ind_09_bin</th>\n",
       "      <th>...</th>\n",
       "      <th>ps_calc_11</th>\n",
       "      <th>ps_calc_12</th>\n",
       "      <th>ps_calc_13</th>\n",
       "      <th>ps_calc_14</th>\n",
       "      <th>ps_calc_15_bin</th>\n",
       "      <th>ps_calc_16_bin</th>\n",
       "      <th>ps_calc_17_bin</th>\n",
       "      <th>ps_calc_18_bin</th>\n",
       "      <th>ps_calc_19_bin</th>\n",
       "      <th>ps_calc_20_bin</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>595212.000000</td>\n",
       "      <td>595212.000000</td>\n",
       "      <td>595212.000000</td>\n",
       "      <td>595212.000000</td>\n",
       "      <td>595212.000000</td>\n",
       "      <td>595212.000000</td>\n",
       "      <td>595212.000000</td>\n",
       "      <td>595212.000000</td>\n",
       "      <td>595212.000000</td>\n",
       "      <td>595212.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>595212.000000</td>\n",
       "      <td>595212.000000</td>\n",
       "      <td>595212.000000</td>\n",
       "      <td>595212.000000</td>\n",
       "      <td>595212.000000</td>\n",
       "      <td>595212.000000</td>\n",
       "      <td>595212.000000</td>\n",
       "      <td>595212.000000</td>\n",
       "      <td>595212.000000</td>\n",
       "      <td>595212.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>0.036448</td>\n",
       "      <td>1.900378</td>\n",
       "      <td>1.358943</td>\n",
       "      <td>4.423318</td>\n",
       "      <td>0.416794</td>\n",
       "      <td>0.405188</td>\n",
       "      <td>0.393742</td>\n",
       "      <td>0.257033</td>\n",
       "      <td>0.163921</td>\n",
       "      <td>0.185304</td>\n",
       "      <td>...</td>\n",
       "      <td>5.441382</td>\n",
       "      <td>1.441918</td>\n",
       "      <td>2.872288</td>\n",
       "      <td>7.539026</td>\n",
       "      <td>0.122427</td>\n",
       "      <td>0.627840</td>\n",
       "      <td>0.554182</td>\n",
       "      <td>0.287182</td>\n",
       "      <td>0.349024</td>\n",
       "      <td>0.153318</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>0.187401</td>\n",
       "      <td>1.983789</td>\n",
       "      <td>0.664594</td>\n",
       "      <td>2.699902</td>\n",
       "      <td>0.493311</td>\n",
       "      <td>1.350642</td>\n",
       "      <td>0.488579</td>\n",
       "      <td>0.436998</td>\n",
       "      <td>0.370205</td>\n",
       "      <td>0.388544</td>\n",
       "      <td>...</td>\n",
       "      <td>2.332871</td>\n",
       "      <td>1.202963</td>\n",
       "      <td>1.694887</td>\n",
       "      <td>2.746652</td>\n",
       "      <td>0.327779</td>\n",
       "      <td>0.483381</td>\n",
       "      <td>0.497056</td>\n",
       "      <td>0.452447</td>\n",
       "      <td>0.476662</td>\n",
       "      <td>0.360295</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</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>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>2.000000</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>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>6.000000</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>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>4.000000</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>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>7.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>6.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>7.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>9.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>7.000000</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>11.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>6.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>19.000000</td>\n",
       "      <td>10.000000</td>\n",
       "      <td>13.000000</td>\n",
       "      <td>23.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>8 rows × 58 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "              target      ps_ind_01  ps_ind_02_cat      ps_ind_03  \\\n",
       "count  595212.000000  595212.000000  595212.000000  595212.000000   \n",
       "mean        0.036448       1.900378       1.358943       4.423318   \n",
       "std         0.187401       1.983789       0.664594       2.699902   \n",
       "min         0.000000       0.000000      -1.000000       0.000000   \n",
       "25%         0.000000       0.000000       1.000000       2.000000   \n",
       "50%         0.000000       1.000000       1.000000       4.000000   \n",
       "75%         0.000000       3.000000       2.000000       6.000000   \n",
       "max         1.000000       7.000000       4.000000      11.000000   \n",
       "\n",
       "       ps_ind_04_cat  ps_ind_05_cat  ps_ind_06_bin  ps_ind_07_bin  \\\n",
       "count  595212.000000  595212.000000  595212.000000  595212.000000   \n",
       "mean        0.416794       0.405188       0.393742       0.257033   \n",
       "std         0.493311       1.350642       0.488579       0.436998   \n",
       "min        -1.000000      -1.000000       0.000000       0.000000   \n",
       "25%         0.000000       0.000000       0.000000       0.000000   \n",
       "50%         0.000000       0.000000       0.000000       0.000000   \n",
       "75%         1.000000       0.000000       1.000000       1.000000   \n",
       "max         1.000000       6.000000       1.000000       1.000000   \n",
       "\n",
       "       ps_ind_08_bin  ps_ind_09_bin  ...     ps_calc_11     ps_calc_12  \\\n",
       "count  595212.000000  595212.000000  ...  595212.000000  595212.000000   \n",
       "mean        0.163921       0.185304  ...       5.441382       1.441918   \n",
       "std         0.370205       0.388544  ...       2.332871       1.202963   \n",
       "min         0.000000       0.000000  ...       0.000000       0.000000   \n",
       "25%         0.000000       0.000000  ...       4.000000       1.000000   \n",
       "50%         0.000000       0.000000  ...       5.000000       1.000000   \n",
       "75%         0.000000       0.000000  ...       7.000000       2.000000   \n",
       "max         1.000000       1.000000  ...      19.000000      10.000000   \n",
       "\n",
       "          ps_calc_13     ps_calc_14  ps_calc_15_bin  ps_calc_16_bin  \\\n",
       "count  595212.000000  595212.000000   595212.000000   595212.000000   \n",
       "mean        2.872288       7.539026        0.122427        0.627840   \n",
       "std         1.694887       2.746652        0.327779        0.483381   \n",
       "min         0.000000       0.000000        0.000000        0.000000   \n",
       "25%         2.000000       6.000000        0.000000        0.000000   \n",
       "50%         3.000000       7.000000        0.000000        1.000000   \n",
       "75%         4.000000       9.000000        0.000000        1.000000   \n",
       "max        13.000000      23.000000        1.000000        1.000000   \n",
       "\n",
       "       ps_calc_17_bin  ps_calc_18_bin  ps_calc_19_bin  ps_calc_20_bin  \n",
       "count   595212.000000   595212.000000   595212.000000   595212.000000  \n",
       "mean         0.554182        0.287182        0.349024        0.153318  \n",
       "std          0.497056        0.452447        0.476662        0.360295  \n",
       "min          0.000000        0.000000        0.000000        0.000000  \n",
       "25%          0.000000        0.000000        0.000000        0.000000  \n",
       "50%          1.000000        0.000000        0.000000        0.000000  \n",
       "75%          1.000000        1.000000        1.000000        0.000000  \n",
       "max          1.000000        1.000000        1.000000        1.000000  \n",
       "\n",
       "[8 rows x 58 columns]"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "insurance_data.describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As you see, there are less than 4% of positive examples, so we have to deal with a highly imbalanced dataset."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-02-26T07:47:09.835229Z",
     "start_time": "2021-02-26T07:47:09.677093Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.countplot(x='target', data=insurance_data)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Accuracy is not informative here and the Normalized Gini Coefficient will be used instead: https://www.kaggle.com/c/porto-seguro-safe-driver-prediction/overview/evaluation."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-02-26T07:47:33.683094Z",
     "start_time": "2021-02-26T07:47:33.676996Z"
    }
   },
   "outputs": [],
   "source": [
    "# Code for calculating Normalized gini coefficient\n",
    "# https://www.kaggle.com/c/ClaimPredictionChallenge/discussion/703\n",
    "def gini(actual, pred, cmpcol = 0, sortcol = 1):  \n",
    "    assert(len(actual) == len(pred))  \n",
    "    epsilon = 1e-7\n",
    "    values = np.asarray(np.c_[actual, pred, np.arange(len(actual))], dtype=np.float)  \n",
    "    values = values[np.lexsort((values[:, 2], -1 * values[:, 1]))]  \n",
    "    total = values[:, 0].sum() \n",
    "    gini_sum = (values[:, 0].cumsum().sum() + epsilon) / (total + epsilon)  \n",
    "  \n",
    "    gini_sum -= (len(actual) + 1) / 2  \n",
    "    return gini_sum / len(actual)  \n",
    "  \n",
    "def gini_normalized(a, p):  \n",
    "    '''Function to calculate the normalized gini coefficient'''\n",
    "    return gini(a, p) / gini(a, a)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**(1.5 points)** Prove that the **Normalized Gini Coefficient** is equivalent to **2 x AUC - 1** :"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**(0.5 points)** By the way, what other metrics could you suggest for this problem?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's calculate the Normalized Gini Coefficient:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-02-26T07:48:48.154807Z",
     "start_time": "2021-02-26T07:48:47.037392Z"
    }
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/python3/lib/python3.6/site-packages/sklearn/pipeline.py:451: DataConversionWarning: Data with input dtype int64, float64 were all converted to float64 by StandardScaler.\n",
      "  Xt = transform.transform(Xt)\n"
     ]
    }
   ],
   "source": [
    "y_pred = clf.predict_proba(X_valid)[:, 1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-02-26T07:48:49.231725Z",
     "start_time": "2021-02-26T07:48:49.173437Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.25652185561929747"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "gini_normalized(y_valid, y_pred)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**(0.5 points + Y bonus points)** Try different approaches: oversampling / undersampling, careful feature analysis and feature engineering, etc., to get a reasonable pipeline and improve the model quality. Use cross-validation for model evaluation.\n",
    "\n",
    "Select the best model, load the test set and make the predictions. Submit them to kaggle.\n",
    "Y bonus points will be calculated as $\\frac{round(200 * \\max(score - 0.253, 0))}{2}$, where *score* is your kaggle leaderboard score.\n",
    "\n",
    "Note: do not use any classification models which have not been covered in the lessons yet."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Лучший скор - **0.26502**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-03-01T00:07:10.159847Z",
     "start_time": "2021-03-01T00:07:10.155891Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1.0"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "round(200*max(0.26502 - 0.253,0))/2"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.8"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "number_sections": true,
   "sideBar": true,
   "skip_h1_title": false,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": false,
   "toc_position": {
    "height": "calc(100% - 180px)",
    "left": "10px",
    "top": "150px",
    "width": "384px"
   },
   "toc_section_display": true,
   "toc_window_display": true
  },
  "varInspector": {
   "cols": {
    "lenName": 16,
    "lenType": 16,
    "lenVar": 40
   },
   "kernels_config": {
    "python": {
     "delete_cmd_postfix": "",
     "delete_cmd_prefix": "del ",
     "library": "var_list.py",
     "varRefreshCmd": "print(var_dic_list())"
    },
    "r": {
     "delete_cmd_postfix": ") ",
     "delete_cmd_prefix": "rm(",
     "library": "var_list.r",
     "varRefreshCmd": "cat(var_dic_list()) "
    }
   },
   "types_to_exclude": [
    "module",
    "function",
    "builtin_function_or_method",
    "instance",
    "_Feature"
   ],
   "window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}