5_freq_sev.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 5. Claim Frequency and Severity Modelling"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Usually, Poisson GLM is used to model claim frequency. We now investigate the possibility to use decision trees to model claim counts using a Poisson loss function."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Description of ```tree.DecisionTreeRegressor``` parametrization:\n",
    "* criterion: the function to measure the quality of a split.\n",
    "* splitter: the strategy used to choose the split at each node. \n",
    "* max_depth: the maximum depth of the tree. If None, then nodes are expanded until all leaves are pure or until all leaves contain less than min_samples_split samples.\n",
    "* min_samples_split (default =2): minimum number of samples required to split an internal node.\n",
    "* min_samples_leaf (default =1): The minimum number of samples required to be at a leaf node.\n",
    "* min_weight_fraction_leaf (default =0.0): The minimum weighted fraction of the sum total of weights (of all the input samples) required to be at a leaf node.\n",
    "* max_features(default =None, but auto is valid choice): The number of features to consider when looking for the best split.\n",
    "* ccp_alpha: Complexity parameter used for Minimal Cost-Complexity Pruning"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import os"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>nclaims</th>\n",
       "      <th>amount</th>\n",
       "      <th>avg</th>\n",
       "      <th>exp</th>\n",
       "      <th>coverage</th>\n",
       "      <th>fuel</th>\n",
       "      <th>use</th>\n",
       "      <th>fleet</th>\n",
       "      <th>sex</th>\n",
       "      <th>ageph</th>\n",
       "      <th>bm</th>\n",
       "      <th>agec</th>\n",
       "      <th>power</th>\n",
       "      <th>pc</th>\n",
       "      <th>town</th>\n",
       "      <th>long</th>\n",
       "      <th>lat</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>1618.001036</td>\n",
       "      <td>1618.001036</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>TPL</td>\n",
       "      <td>gasoline</td>\n",
       "      <td>private</td>\n",
       "      <td>N</td>\n",
       "      <td>male</td>\n",
       "      <td>50</td>\n",
       "      <td>5</td>\n",
       "      <td>12</td>\n",
       "      <td>77</td>\n",
       "      <td>1000</td>\n",
       "      <td>BRUSSEL</td>\n",
       "      <td>4.355223</td>\n",
       "      <td>50.845386</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>PO</td>\n",
       "      <td>gasoline</td>\n",
       "      <td>private</td>\n",
       "      <td>N</td>\n",
       "      <td>female</td>\n",
       "      <td>64</td>\n",
       "      <td>5</td>\n",
       "      <td>3</td>\n",
       "      <td>66</td>\n",
       "      <td>1000</td>\n",
       "      <td>BRUSSEL</td>\n",
       "      <td>4.355223</td>\n",
       "      <td>50.845386</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>TPL</td>\n",
       "      <td>diesel</td>\n",
       "      <td>private</td>\n",
       "      <td>N</td>\n",
       "      <td>male</td>\n",
       "      <td>60</td>\n",
       "      <td>0</td>\n",
       "      <td>10</td>\n",
       "      <td>70</td>\n",
       "      <td>1000</td>\n",
       "      <td>BRUSSEL</td>\n",
       "      <td>4.355223</td>\n",
       "      <td>50.845386</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>TPL</td>\n",
       "      <td>gasoline</td>\n",
       "      <td>private</td>\n",
       "      <td>N</td>\n",
       "      <td>male</td>\n",
       "      <td>77</td>\n",
       "      <td>0</td>\n",
       "      <td>15</td>\n",
       "      <td>57</td>\n",
       "      <td>1000</td>\n",
       "      <td>BRUSSEL</td>\n",
       "      <td>4.355223</td>\n",
       "      <td>50.845386</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>155.974606</td>\n",
       "      <td>155.974606</td>\n",
       "      <td>0.046575</td>\n",
       "      <td>TPL</td>\n",
       "      <td>gasoline</td>\n",
       "      <td>private</td>\n",
       "      <td>N</td>\n",
       "      <td>female</td>\n",
       "      <td>28</td>\n",
       "      <td>9</td>\n",
       "      <td>7</td>\n",
       "      <td>70</td>\n",
       "      <td>1000</td>\n",
       "      <td>BRUSSEL</td>\n",
       "      <td>4.355223</td>\n",
       "      <td>50.845386</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   nclaims       amount          avg       exp coverage      fuel      use  \\\n",
       "0        1  1618.001036  1618.001036  1.000000      TPL  gasoline  private   \n",
       "1        0     0.000000          NaN  1.000000       PO  gasoline  private   \n",
       "2        0     0.000000          NaN  1.000000      TPL    diesel  private   \n",
       "3        0     0.000000          NaN  1.000000      TPL  gasoline  private   \n",
       "4        1   155.974606   155.974606  0.046575      TPL  gasoline  private   \n",
       "\n",
       "  fleet     sex  ageph  bm  agec  power    pc     town      long        lat  \n",
       "0     N    male     50   5    12     77  1000  BRUSSEL  4.355223  50.845386  \n",
       "1     N  female     64   5     3     66  1000  BRUSSEL  4.355223  50.845386  \n",
       "2     N    male     60   0    10     70  1000  BRUSSEL  4.355223  50.845386  \n",
       "3     N    male     77   0    15     57  1000  BRUSSEL  4.355223  50.845386  \n",
       "4     N  female     28   9     7     70  1000  BRUSSEL  4.355223  50.845386  "
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = pd.read_csv(r'.\\data\\PC_data.txt', sep = '\\s+')\n",
    "df.columns = map(str.lower, df.columns)\n",
    "df = df.drop('id', axis = 1)\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "# converting type(str) columns to categorical\n",
    "df['coverage'] = df['coverage'].astype('category').cat.codes\n",
    "df['fleet'] = df['fleet'].astype('category').cat.codes\n",
    "df['sex'] = df['sex'].astype('category').cat.codes\n",
    "df['town'] = df['town'].astype('category').cat.codes\n",
    "df['fuel'] = df['fuel'].astype('category').cat.codes\n",
    "df['use'] = df['use'].astype('category').cat.codes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "exp = df['exp']\n",
    "y = np.array(df['nclaims']/df['exp'])\n",
    "x = np.array(df[['bm','ageph','agec','power','coverage','fuel','sex','fleet','use']])\n",
    "col_names = ['ageph','agec','bm','power','coverage','fuel','sex','fleet','use']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Since there is no convenient way to implement the exposure in the Poisson regression tree. Therefore, we divide the response by the exposure and we fit regression tree minimizing MSE and setting weights = exposure term."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn import metrics\n",
    "from sklearn import tree"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "rgf_dt = tree.DecisionTreeRegressor(criterion=\"mse\", max_depth = 3,\n",
    "                                      min_samples_split = 10,\n",
    "                                      min_samples_leaf = 10,\n",
    "                                      ccp_alpha = 0, splitter=\"best\").fit(x,y,exp)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can provide a stylized representation of the Regression Tree via ```export_text()``` built-in method:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "|--- ageph <= 6.50\n",
      "|   |--- ageph <= 1.50\n",
      "|   |   |--- agec <= 55.50\n",
      "|   |   |   |--- value: [0.12]\n",
      "|   |   |--- agec >  55.50\n",
      "|   |   |   |--- value: [0.09]\n",
      "|   |--- ageph >  1.50\n",
      "|   |   |--- agec <= 57.50\n",
      "|   |   |   |--- value: [0.15]\n",
      "|   |   |--- agec >  57.50\n",
      "|   |   |   |--- value: [0.12]\n",
      "|--- ageph >  6.50\n",
      "|   |--- ageph <= 10.50\n",
      "|   |   |--- agec <= 26.50\n",
      "|   |   |   |--- value: [0.24]\n",
      "|   |   |--- agec >  26.50\n",
      "|   |   |   |--- value: [0.19]\n",
      "|   |--- ageph >  10.50\n",
      "|   |   |--- bm <= 6.50\n",
      "|   |   |   |--- value: [0.23]\n",
      "|   |   |--- bm >  6.50\n",
      "|   |   |   |--- value: [0.29]\n",
      "\n"
     ]
    }
   ],
   "source": [
    "text_representation = tree.export_text(rgf_dt, feature_names = col_names)\n",
    "print(text_representation)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can provide a graphic representation of the Regression Tree via ```plot_treet()``` built-in method:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x864 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "%matplotlib inline\n",
    "plt.figure(figsize=(12,12))\n",
    "_ = tree.plot_tree(rgf_dt, \n",
    "                   feature_names = col_names,\n",
    "                   filled = True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "It is possible to retrieve all samples the fell into the nodes. In order to do so, we implement a custom function to provide a new stylized representation of the Regression Tree retaining more information and non-rounded leaf values:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 837,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The binary tree structure has 7 nodes and has the following tree structure:\n",
      "node=0 test node: go to node 1 if X[:, 0] <= 18.5 else to node 2.\n",
      "\t*node=1 leaf node. Value here is: [[1.08180202]]\n",
      "\tnode=2 test node: go to node 3 if X[:, 1] <= 32.5 else to node 6.\n",
      "\t\tnode=3 test node: go to node 4 if X[:, 0] <= 91.5 else to node 5.\n",
      "\t\t\t*node=4 leaf node. Value here is: [[0.13938908]]\n",
      "\t\t\t*node=5 leaf node. Value here is: [[0.05205362]]\n",
      "\t\t*node=6 leaf node. Value here is: [[0.01212021]]\n"
     ]
    }
   ],
   "source": [
    "n_nodes = rgf_dt.tree_.node_count\n",
    "children_left = rgf_dt.tree_.children_left\n",
    "children_right = rgf_dt.tree_.children_right\n",
    "feature = rgf_dt.tree_.feature\n",
    "threshold = rgf_dt.tree_.threshold\n",
    "value = rgf_dt.tree_.value\n",
    "\n",
    "# The tree structure can be traversed to compute various properties such\n",
    "# as the depth of each node and whether or not it is a leaf.\n",
    "node_depth = np.zeros(shape=n_nodes, dtype=np.int64)\n",
    "is_leaves = np.zeros(shape=n_nodes, dtype=bool)\n",
    "stack = [(0, -1)]  # seed is the root node id and its parent depth\n",
    "while len(stack) > 0:\n",
    "    node_id, parent_depth = stack.pop()\n",
    "    node_depth[node_id] = parent_depth + 1\n",
    "\n",
    "    # If we have a test node\n",
    "    if (children_left[node_id] != children_right[node_id]):\n",
    "        stack.append((children_left[node_id], parent_depth + 1))\n",
    "        stack.append((children_right[node_id], parent_depth + 1))\n",
    "    else:\n",
    "        is_leaves[node_id] = True\n",
    "\n",
    "\n",
    "print(\"The binary tree structure has %s nodes and has \"\n",
    "      \"the following tree structure:\"\n",
    "      % n_nodes)\n",
    "for i in range(n_nodes):\n",
    "    if is_leaves[i]:\n",
    "        print(\"%s*node=%s leaf node. Value here is: %s\" % (node_depth[i] * \"\\t\", i, value[i]))\n",
    "    else:\n",
    "        print(\"%snode=%s test node: go to node %s if X[:, %s] <= %s else to \"\n",
    "              \"node %s.\"\n",
    "              % (node_depth[i] * \"\\t\",\n",
    "                 i,\n",
    "                 children_left[i],\n",
    "                 feature[i],\n",
    "                 threshold[i],\n",
    "                 children_right[i],\n",
    "                 ))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We subset the dataset to verify whether the prediction in a leaf node is what you would expect:\n",
    "* take the node ```bm <= 1.5``` and ```ageph > 55.50```  \n",
    "* Calculate the expected claim frequency as ```sum(nclaims)/sum(expo)```\n",
    "* Compare it with the correspondent leaf prediction"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 680,
   "metadata": {},
   "outputs": [],
   "source": [
    "# subsetting the dataset\n",
    "sub_df = df.loc[(df.bm<=1.5)&(df.ageph>55.50)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 681,
   "metadata": {},
   "outputs": [],
   "source": [
    "# calculate the expected claim frequency (rounding by 8)\n",
    "exp_cf = np.round(np.sum(sub_df.nclaims)/np.sum(sub_df.exp),8)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 682,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Expected claim frequency: 0.08898655, Predicted claim frequency in leaf: 0.08898655\n"
     ]
    }
   ],
   "source": [
    "# compare the expected claim frequency and the predicted claim frequency in the leaf\n",
    "print(f'Expected claim frequency: {exp_cf}, Predicted claim frequency in leaf: {np.round(float(rgf_dt.tree_.value[4]),8)}')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 5.1 Pruning Trees with cost complexity"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The DecisionTreeClassifier provides parameters such as ```min_samples_leaf``` and ```max_depth``` to prevent a tree from overfiting. Cost complexity pruning provides another option to control the size of a tree."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "this pruning technique is parameterized by the cost complexity parameter, ```ccp_alpha```. Greater values of ```ccp_alpha``` increase the number of nodes pruned. Here we only show the effect of ccp_alpha on regularizing the trees and how to choose a ccp_alpha based on validation scores."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Minimal cost complexity pruning recursively finds the node with the “weakest link”. The weakest link is characterized by an effective alpha, where the nodes with the smallest effective alpha are pruned first. To get an idea of what values of ```ccp_alpha``` could be appropriate, scikit-learn provides ```DecisionTreeRegressor.cost_complexity_pruning_path``` that returns the effective alphas and the corresponding total leaf impurities at each step of the pruning process. As alpha increases, more of the tree is pruned, which increases the total impurity of its leaves."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 817,
   "metadata": {},
   "outputs": [],
   "source": [
    "path = rgf_dt.cost_complexity_pruning_path(x, y, exp)\n",
    "ccp_alphas, impurities = path.ccp_alphas, path.impurities"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 818,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'Total Impurity vs effective alpha for training set')"
      ]
     },
     "execution_count": 818,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAZUAAAEWCAYAAACufwpNAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjQuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8rg+JYAAAACXBIWXMAAAsTAAALEwEAmpwYAAAvEUlEQVR4nO3de7xU1X338c/X4w2NBEVQQFFMqIoabycY22hNrFF5zAPSmGgu2taqNFo1ebxgYxrTXNRo2iSNLaFVS5N6q1eaGAkl0TaJRjDITULBOxcFUYoEwvX3/LHWyDDMnLMPzJzDnPN9v17zmj1r7bVm7dlz5nf2WnuvrYjAzMysHnbq6gaYmVn34aBiZmZ146BiZmZ146BiZmZ146BiZmZ146BiZmZ146DSDUgKSe/t6na0R9JJkuZ1dTs6QtJXJb0h6bX8+mxJr0paJenYOr5Pl3w2Hfnu1PN7JukPJM3Pn+OoetS5vSR9StJP6r1uTyNfp9I4klaVvdwDWAtszK8viYh/q1LmFOAHEXFAB94ngKERsaBK3uO5vn8u3vLOIekl4M8j4j+7ui3VSDoQ+B/goIhYmtOeBz4fEY9sZ90191ln6kg76tlmSVOAiRHx7e2tK9f3L8DCiLi+HvU1G0kHAy8Cu0TEhq5sy85d+ebdXUS8q7S8o/+ANpqknbv6y74NDgKWlwJKWdqcLmpPd7LNn+O2fJea9PvXnCLCj054AC8Bf5SXdwO+BSzOj2/ltD2BNcAmYFV+DASGA08CK4AlwHeBXcvqDuC9Nd73cVIwAzgFWAhcAyzNdY0CRpD+I38T+KuysjcA9wP3Am8DvwaOrvW+wL8AX614r2uB14Dvl9Jy/vfzdq7J23kN8CPgLyvaPxMYVWW7HgMuq0ibAYwGBPxd3sb/zXUcWePzeTdwe/4sFgFfBVqAP6rYF3fn5wB+Czyfyw8EHgCWkf5TvLys7hbgr4Dn8+f3DHAg8F9l9awCPlHx2YwF7q9o57eB77TV5hrbV/i7k/ffOGBybu8TpKO08nXHAPOBt4Db2Nzb8R7gp8By4A3g34A+Ndr0fMW+3y1/jhNJ38EFwEVVvoc/AFaSv89l+RcD64F1ub7/KPubuzbv/7Wkf6LHlu2P54Czy+r5E+DnBbe3I+u2AN/Mn8uLwGV5/Z1rfD7X5v36NjAPODWn71TW/uXAfcA+Oe+VXGfpd+PELvut66o37mkPtgwqfwM8BfQH+gG/BL6S804h/7iUlT0e+ED+ozgYmAtcWZbfkaCyAfhrYBfgItKP4V3AXsARwO+AQ/L6N+Q/1o/l9a9i8yH2Vu/L1kFlA3Az6UejV+W2lX8m+fXHgV+VvT46//HsWmW7zgd+UfZ6GOmHczfgdNIPeB9SgDkcGFDj83kY+B4poPcHniZ1TdbaF+U/wjvl9/lrYFfgEOAF4PScfzUwCzg0t+NooG+Nz+6d9yL9F78a6J1ft5ACwgfaa3OV7Sv83cn7723g5Pw5fputfzh/mD/XwaTvzhk5773AablcP1Lg/FaRv4f8+gngH4DdgWNy3aUf0xtI38NR+TPvVaW+fyF/9yre41lSIO+V084hBbCdSMH8t6XvBtUDRa3t7ci6Y0gB7ABgb+A/qRFUSN+VV4GB+fXBwHvy8pWk340D8uf8PeDusvVqBqpO/a3r6gb0lAdbBpXngRFleacDL+XlU6j4IatS15XAQ2WvOxJU1pD/qyUFkgBOKFv/GfKRQf5jfqosbyfSj9tJ1d6XrYPKOmD3svwtto2tf1h2I/2nOjS/vhX4hxrbtRfpB+Gg/PprwB15+cOkI68PADu18TnuR/oPtldZ2nnAz2rtC7b8ET4BeKUi/zrgzrw8DxhZ471rBpX8+ufA+Xn5NDYfGbXZ5gLfw5rfnbz/7inLexdpDPDAsnU/WJZ/HzC2xvuMAqYX/Hs4ML/PXmX5NwL/UvY9/K92tuud717Fe/xZO+WeLe0jqgeKqtvbwXV/SlnQJx0F1woq7yUdYf8R+Z+3sry55ECbXw8gBdvSPww7RFDx2V9dYyDwctnrl3NaVZJ+T9IPJb0maSXwdWDfbXzv5RFROllgTX5+vSx/DenHpOTV0kJEbCJ1adVsa4VlEfG7og2LiLWkP8ZPS9qJ9GP5/Rrrvk3qLjs3J51L6nIhIn5K6ua5DXhd0nhJvatUcxDpCGyJpBWSVpD+++tfsMkHAQNLZXP5vyL98EP6sXy+YF2V7iJtP8An8+sOt3kbvjvl+3sVKciX7+/XypZXk78rkvpLukfSovw+P2jnfcoNBN7M+7TkZWBQtXZ10BblJJ0v6dmyz+7IdtpZdXs7uO7AinbU3JZIJ0FcSQqkS/NnWvr8DwIeKmv7XFIw3q9KVV3GQaVrLCZ9QUoG5zRI/21U+kfgN6T/4HuTfrjU0BZudmBpIf/QH8Dmtq4mndVWsn9F2Wrb0l7+BOBTwKnA6oh4so3ydwPnSTqR1L32s3cqjvhORBxP6tL7PVJXVKVXSf/17xsRffKjd0Qc0U67y8u/WFa2T0TsFREjyvLfU7CuSv8OnCLpAOBsNgeVjra5o9+d8v39LmAfNu/vttxI2p/vy+/z6Xbep9xiYB9Je5WlDSaNK5Rsy3dpi3RJBwH/RBrT6BsRfYDZHWjntlpC+rspObDWigARcVdEfJD0GxGkLmRI+/7Miu/b7hGxiPY/n07joNI17gaul9RP0r6kPvkf5LzXgb6S3l22/l6kAcpVkg4D/qIT23q8pNGSdib9B7WW1K8Lqevgk5JaJJ0B/GEH636dNA7xjhxENpEGNqsepZR5lPSH9zfAvflICknvl3SCpF1IXWS/Y/Op3OXvtQT4CfBNSb0l7STpPZKKbsfTwEpJ10rqlT+HIyW9P+f/M/AVSUOVvE9S31rbXtG2ZaSuyztJgWvuNra5o9+dEZI+KGlX4CukMa4iRwl7kQaIV0gaRPUgXlWu/5fAjZJ2l/Q+4ELykWdBbX6e2Z6kH99lAJL+lHSk0mj3AVdIGiSpD2kgvipJh0r6sKTdSN/bNWz+7o4DvpaDI/n3Y2TOW0b6u2nvM2g4B5Wu8VVgGumslFmks6q+ChARvyEFnRfyYe5A0gD5J0mDqP9EOhurszxCGtB8C/gMMDoi1ue8K4CPkgbIP0UaQO6IG0nBdYWkq8rS/xU4is2BtqrcXfYgqf/5rrKs3qTP6S1SN8py0vhMNeeTBtmfy+vfT+qrblfuRvwoaWD5RdLZPf9MOjsL4G9JPyg/If2w3046ooLUvTEhb/vHa7zFXVW2raNt7uh35y7gS6Rur+NJ+7WILwPHkc62+xFpv3TEeaRxgcXAQ8CXImJyB8rfDgzLn+fD1VaIiOdI/6w8SQpCRwG/6GA7t8U/kb4DM4HppH+GNlDlHx3SuOJNpO/Sa6Ruzb/Ked8mnSH3E0lvk/65OwEgIlaTxhV/kT+DDzRsa9rhix+tJkk3kAZxP93J73s+cHHuArBO0tMvIOwsks4ExkXEQe2u3IR8pGI7FEl7AJ8Fxnd1W8zqIXeNjpC0c+4a/BLpaKxbclCxHYak00l9w6+zdZePWbMSqXvwLVL311zSOGq35O4vMzOrGx+pmJlZ3fToCSX33XffOPjgg7u6GWZmTeWZZ555IyL6Vcvr0UHl4IMPZtq0aV3dDDOzpiLp5Vp57v4yM7O6aWhQkXSGpHmSFkgaWyX/MElPSlpbcfEbkq6QNFvSHElXlqXvI2my0l3jJkvaO6fvKulOSbMkzVC62ZWZmXWihgUVSS2kCf3OJE1Lfp6kYRWrvQlcTsXVzpKOJE3LPpw0XfhZkobm7LHAlIgYCkzJr8nrExFHkWZ1/Waeq8rMzDpJI390hwMLIuKFiFgH3AOMLF8hIpZGxFTS9M3lDidNub460t3aniBNqkeuY0JenkCaYhtS4JpSqpc0dUhrPTfIzMza1sigMogtp3heyJZTWbdlNnCypL75CusRbJ7Zc788qV5pcr3SlN8zgJH5qtUhpHmLtpoNVNLFkqZJmrZs2bIOb5SZmdXWyLO/qk0nXehKy4iYK+lm0m1NV5ECRnv3l76DdIQzjTSJ4C+rlYmI8eQpQFpbW33lp5n1KA9PX8Qtk+axeMUaBvbpxdWnH8qoY4v+v9++RgaVhWx5pFB+H452RcTtpJlHkfT1XB+kmy4NiIglkgaQ7pJG7ib7XKm8pF+S7hdtZmakgHLdg7NYsz5NkLxoxRque3AWQN0CSyO7v6YCQyUNyfdmOJc0bXMhkvrn58HAaNJ08OQ6LsjLF5CmZkfSHpL2zMunARvyVNdmZgbcMmneOwGlZM36jdwyaV7d3qNhRyoRsUHSZcAkoIV0//A5ksbk/HGS9id1V/UGNuVTh4dFxErggXxDo/XApRHxVq76JuA+SRcCrwDn5PT+wCRJm0h3jPtMo7bNzKwZLV6xpkPp26KhV9RHxKOkG9KUp40rW36NLW+zWb7eSTXSl5NuNVuZ/hJw6HY018ysWxvYpxeLqgSQgX16VVl72/g6DjOzHuLq0w+l1y4tW6T12qWFq0+v3//jPXruLzOznqQ0GH/N/TNZt3ETg5rs7C8zM9vBjDp2EHc//QoA915yYt3rd/eXmZnVjYOKmZnVjYOKmZnVjYOKmZnVjYOKmZnVjYOKmZnVjYOKmZnVjYOKmZnVjYOKmZnVjYOKmZnVjYOKmZnVjYOKmZnVjYOKmZnVjYOKmZnVTUODiqQzJM2TtEDS2Cr5h0l6UtJaSVdV5F0habakOfk2w6X0fSRNljQ/P++d03eRNEHSLElzJV3XyG0zM7OtNSyoSGoBbgPOBIYB50kaVrHam8DlwK0VZY8ELgKGA0cDZ0kamrPHAlMiYigwJb+GdK/63SLiKOB44BJJB9d7u8zMrLZGHqkMBxZExAsRsQ64BxhZvkJELI2IqcD6irKHA09FxOqI2AA8AZyd80YCE/LyBGBUqTpgT0k7A72AdcDK+m6SmZm1pZFBZRDwatnrhTmtiNnAyZL6StoDGAEcmPP2i4glAPm5f06/H/gtsAR4Bbg1It6srFjSxZKmSZq2bNmyjm6TmZm1oZFBRVXSokjBiJgL3AxMBh4DZgAb2ik2HNgIDASGAP9P0iFV6h4fEa0R0dqvX78izTEzs4IaGVQWsvnoAuAAYHHRwhFxe0QcFxEnk8Ze5ues1yUNAMjPS3P6J4HHImJ9RCwFfgG0buc2mJlZBzQyqEwFhkoaImlX4FxgYtHCkvrn58HAaODunDURuCAvXwA8kpdfAT6sZE/gA8BvtnsrzMyssJ0bVXFEbJB0GTAJaAHuiIg5ksbk/HGS9gemAb2BTfnU4WERsRJ4QFJf0iD+pRHxVq76JuA+SReSAsk5Of024E7SeIyAOyNiZqO2z8zMttawoAIQEY8Cj1akjStbfo3ULVat7Ek10pcDp1ZJX8XmAGNmZl3AV9SbmVndOKiYmVndOKiYmVndOKiYmVndOKiYmVndOKiYmVndOKiYmVndOKiYmVndOKiYmVndOKiYmVndOKiYmVndOKiYmVndOKiYmVndOKiYmVndOKiYmVndtBtUJL1H0m55+RRJl0vq0/CWmZlZ0ylypPIAsFHSe4HbgSHAXQ1tlZmZNaUiQWVTRGwAzga+FRGfAwYUqVzSGZLmSVogaWyV/MMkPSlpraSrKvKukDRb0px8m+FS+j6SJkuan5/3zumfkvRs2WOTpGOKtNPMzOqjSFBZL+k84ALghzltl/YKSWoh3Tf+TGAYcJ6kYRWrvQlcDtxaUfZI4CJgOHA0cJakoTl7LDAlIoYCU/JrIuLfIuKYiDgG+AzwUkQ8W2D7zMysTooElT8FTgS+FhEvShoC/KBAueHAgoh4ISLWAfcAI8tXiIilETEVWF9R9nDgqYhYnY+SniAdKZHrmJCXJwCjqrz3ecDdBdpoZmZ11G5QiYjngGuBX+fXL0bETQXqHgS8WvZ6YU4rYjZwsqS+kvYARgAH5rz9ImJJbssSoH+V8p+gRlCRdLGkaZKmLVu2rGBzzMysiCJnf30UeBZ4LL8+RtLEAnWrSloUaVREzAVuBibn950BbChSVtIJwOqImF2j7vER0RoRrf369StSpZmZFVSk++sGUlfWCoA8TjGkQLmFbD66ADgAWFy0YRFxe0QcFxEnk8Ze5ues1yUNAMjPSyuKnou7vszMukSRoLIhIv63Iq3IEcdUYKikIZJ2Jf3YFznCAUBS//w8GBjN5kAxkXTSAPn5kbIyOwHnkMZvzMysk+1cYJ3Zkj4JtOQzsC4HftleoYjYIOkyYBLQAtwREXMkjcn54yTtD0wDegOb8qnDwyJiJfCApL6kQfxLI+KtXPVNwH2SLgReIQWRkpOBhRHxQoHtMjOzOisSVP4S+AKwlnTR4yTgq0Uqj4hHgUcr0saVLb9G6harVvakGunLgVNr5D0OfKBI28zMrP6KBJVDI+ILpMBiZmZWU5Exlb+V9BtJX5F0RMNbZGZmTavIdSofAk4BlgHjJc2SdH2jG2ZmZs2n0NT3EfFaRHwHGEO6ZuWvG9koMzNrTkUufjxc0g2SZgPfJZ35VXVw3czMerYiA/V3kq4R+UhEFL540czMep52g0pE+BRdMzMrpN2gki94vJE0ff3upfSIOKSB7TIzsyZUZKD+TuAfSRM6fgj4V+D7jWyUmZk1pyJBpVdETAEUES9HxA3AhxvbLDMza0ZFBup/lydqnJ/n8lpE9XuYmJlZD1fkSOVKYA/SRJLHA59m8yzBZmZm7yhy9tdUAEkREX/a+CaZmVmzKnLx44mSngPm5tdHS/qHhrfMzMyaTpHur28BpwPLASJiBum+JWZmZlsoOvfXqxVJGxvQFjMza3JFzv56VdLvA5FvC3w5uSvMzKzZPTx9EbdMmsfiFWsY2KcXV59+KKOOHdTVzWpaRY5UxgCXAoOAhcAx+XW7JJ0haZ6kBZLGVsk/TNKTktZKuqoi7wpJsyXNybcZLqXvI2mypPn5ee+yvPfl+ubkKfp3x8yshoenL+K6B2exaMUaAli0Yg3XPTiLh6cv6uqmNa0iZ3+9AXyqoxVLagFuA04jBaOpkiZGxHNlq71JOvIZVVH2SOAiYDiwDnhM0o8iYj4wFpgSETflQDUWuFbSzsAPgM9ExIyy+9ubmVV1y6R5rFm/ZW/+mvUbueb+mdz99Ctd1KrGe27JSoYN6N2QumsGFUl/D0St/Ii4vJ26hwMLIuKFXN89wEjgnaASEUuBpZL+T0XZw4GnImJ1LvsEcDbwjVzHKXm9CcDjwLXAR4CZ+USC0r3szcxqWrxiTdX0dRs3dXJLOtewAb0ZeUxjuvjaOlKZtp11DwLKB/gXAicULDsb+Fo+2lgDjChrz34RsQQgIpZIKl3d/3ukcZ9JQD/gnoj4RmXFki4GLgYYPHhwx7bIzLqVgX16sahKYBnUpxf3XnJiF7So+dUMKhExYTvrVrVqixSMiLmSbgYmA6uAGaQJLduyM/BB4P3AamCKpGfyvGXldY8HxgO0trYWao+ZdU9Xn34o1z04a4susF67tHD16Yd2YauaW6FTirfRQuDAstcHAIVv8hURt0fEcRFxMmnsZX7Oel3SAID8vLTs/Z6IiDdyt9mjwHHbuQ1m1o2NOnYQN44+il1b0k/hoD69uHH0UT77azs0MqhMBYZKGpJPRT4XmFi0cKlbS9JgYDTp7pPkOkpzj10APJKXJwHvk7RHHrT/Q8rGb8zMqhl17CCOHdyHE4bswy/GftgBZTvVDCq5+wlJ52xLxRGxAbiM9GM/F7gvIuZIGiNpTK57f0kLgc8D10taKKl0SsIDeXqY/wAujYi3cvpNwGmS5pPOLLspv99bwN+SgtmzwK8j4kfb0nYzM9s2bQ3Uj5B0PXAd8O/bUnlEPErqhipPG1e2/BqpW6xa2ZNqpC8HTq2R9wPSacVmZtYF2goqjwFvAHtKWkkaeI/Sc0Q05iRnMzNrWjW7vyLi6oh4N/CjiOgdEXuVP3diG83MrEkUuaJ+pKT9SKfqAvwqIpY1tllmZtaMitxP5RzgaeAc4OPA05I+1uiGmZlZ8ykyS/H1wPvzlCpI6gf8J3B/IxtmZmbNp8h1KjuVAkq2vGA5MzPrYYocqTyW59MqXXz4CSpOEzYzM4NiA/VXSxpNmldLwPiIeKjhLTMzs6ZT5EiFiHgQeLDBbTEzsybnsREzM6sbBxUzM6ubItepnCXJwcfMzNpVJFicC8yX9A1Jhze6QWZm1rzaDSoR8WngWOB54E5JT0q6WNJeDW+dmZk1lULdWhGxEngAuAcYAJwN/FrSXzawbWZm1mSKjKn8X0kPAT8FdgGGR8SZwNHAVQ1un5mZNZEi16l8DPi7iPiv8sSIWC3pzxrTLDMza0ZFur+WVAaU0q2GI2JKWwUlnSFpnqQFksZWyT8sj9GslXRVRd4VkmZLmiPpyrL0fSRNljQ/P++d0w+WtEbSs/kxDjMz61RFgsppVdLObK+QpBbgtrzuMOA8ScMqVnsTuBy4taLskcBFwHBSN9tZkobm7LHAlIgYCkzJr0uej4hj8mNMu1tmZmZ1VTOoSPoLSbOAwyTNLHu8CMwsUPdwYEFEvBAR60iD/CPLV4iIpRExFVhfUfZw4KmIWB0RG4AnSCcHkOuYkJcnAKMKtMXMzDpBW0cqdwEfBR7Jz6XH8fk04/YMAl4te70wpxUxGzhZUl9JewAjgANz3n4RsQQgP/cvKzdE0nRJT0g6qVrF+XToaZKmLVvmG1iamdVTWwP1EREvSbq0MkPSPhHxZjt1q1qdRRoVEXPzuM1kYBUwA9jQTrElwOCIWC7peOBhSUfk06HL6x4PjAdobW0t1B4zMyumvSMVgGeAafn5mbLX7VnI5qMLgAOAxUUbFhG3R8RxEXEyaexlfs56XdIAgPy8NK+/NiKW5+VnSBdr/l7R9zMzs+1XM6hExFmSBPxhRBwSEUPKHocUqHsqMFTSEEm7kqZ7mVi0YZL65+fBwGg23yRsInBBXr6A1D2HpH755AAkHQIMBV4o+n5mZrb92rxOJSIiX/h4fEcrjogNki4DJgEtwB0RMUfSmJw/TtL+pKOe3sCmfOrwsNIV/JL6kgbxL42It3LVNwH3SboQeAU4J6efDPyNpA3ARmBMgS46MzOroyIXPz4l6f35LK0OiYhHqbj1cESMK1t+jdQtVq1s1YH23MV1apX0B0hTyZiZWRcpElQ+BFwi6WXgt6QB+IiI9zW0ZWZm1nSKBJV2L3Q0MzODYkHFp92amVkhRYLKj0iBRcDuwBBgHnBEA9tlZmZNqN2gEhFHlb+WdBxwScNaZGZmTavD956PiF8D729AW8zMrMm1e6Qi6fNlL3cCjgM8aZaZmW2lyJhK+b3oN5DGWHw9iJmZbaXImMqXAST1Ti/j7Ya3yszMmlKRe9S35vuqzARmSZqRZwE2MzPbQpHurzuAz0bEfwNI+iBwJ+Ar6s3MbAtFzv56uxRQACLi54C7wMzMbCtFjlSelvQ90tTzAXwCeDxfr1I6xdjMzKxQUDkmP3+pIv33SUHmw/VskJmZNa8iZ399qDMaYmZmza/IxY99gPOBg8vXj4jLG9YqMzNrSkW6vx4FngJmAZsa2xwzM2tmRc7+2j0iPh8Rd0bEhNKjSOWSzpA0T9ICSWOr5B8m6UlJayVdVZF3haTZkubk2wyX0veRNFnS/Py8d0W5wZJWVdZnZmaNVySofF/SRZIG5B/0fSTt014hSS3AbaSbfA0DzpM0rGK1N4HLgVsryh4JXAQMB44GzpI0NGePBaZExFBgSn5d7u+AHxfYLjMzq7MiQWUdcAvwJPBMfkwrUG44sCAiXoiIdcA9wMjyFSJiaURMBdZXlD0ceCoiVkfEBuAJ4OycNxIoHSlNAEaVCkkaBbwAzCnQPjMzq7MiQeXzwHsj4uCIGJIfhxQoNwh4tez1wpxWxGzgZEl9Je0BjAAOzHn7RcQSgPzcH0DSnsC1wJcLvoeZmdVZkYH6OcDqbahbVdIK3Zo4IuZKuhmYDKwCZpBmSG7Ll4G/i4hVUrW3zo2SLgYuBhg8eHCR5piZWUFFgspG4FlJPwPWlhILnFK8kM1HFwAHAIuLNiwibgduB5D09VwfwOuSBkTEEkkDgKU5/QTgY5K+AfQBNkn6XUR8t6Le8cB4gNbW1kJBzszMiikSVB7Oj46aCgyVNARYBJwLfLJoYUn9I2KppMHAaODEnDURuAC4KT8/AhARJ5WVvQFYVRlQrGd4ePoibpk0j8Ur1jCwTy+uPv1QRh1btOfVzLZHkSvqC50+XKXcBkmXAZOAFuCOiJgjaUzOHydpf9Kgf2/SkcWVwLCIWAk8IKkvaRD/0oh4K1d9E3CfpAuBV4BztqV91j09PH0R1z04izXrNwKwaMUarntwFoADi1knUET1HiBJ90XEx/O9VLZaKSKafur71tbWmDatyIls1iz+4KafsmjFmq3Sd23ZiWMH9+n8BllTeG7JSoYN6M29l5zY/sqGpGciorVaXltHKlfk57Pq3ySzxlhcJaAArNvoySCstmEDejPyGB/J1kPNoFJ22u7Lndccs+0zsE+vqkcqg/r08n+hZp2gyHUqZk3j6tMPpdcuLVuk9dqlhatPP7SLWmTWsxQ5+8usaZQG46+5fybrNm5ikM/+MutUDirW7Yw6dhB3P/0KgLu8zDpZzaBS66wv0pXy0R3O/jIzs/pq60jFZ32ZmVmHtHX2l8/6MjOzDmn37C9JH5A0Nd/4ap2kjZJWdkbjzMysuRQ5pfi7wHnAfKAX8OfA3zeyUWZm1pwKnf0VEQsktUTERuBOSb9scLvMzKwJFQkqqyXtSpr+/hvAEmDPxjbLzMyaUZHur8/k9S4Dfku6R8roRjbKzMyaU5GgMioifhcRKyPiyxHxeXy6sZmZVVEkqFxQJe1P6twOMzPrBtq6ov480p0ah0iaWJbVG1je6IaZmVnzaWug/pekQfl9gW+Wpb8NzGxko8zMrDnV7P6KiJcj4vGIOBH4DbBXfiyMiA1FKpd0hqR5khZIGlsl/zBJT0paK+mqirwrJM2WNCffZriUvo+kyZLm5+e9c/pwSc/mxwxJZxf6BMzMrG6KXFF/DvA06V7wHwd+JeljBcq1ALcBZwLDgPMkDatY7U3gcuDWirJHAhcBw4GjgbMkDc3ZY4EpETEUmJJfA8wGWiPiGOAM4HuSPAuzmVknKjJQfz3w/oi4ICLOJ/3Qf7FAueHAgoh4ISLWAfcAI8tXiIilETEVWF9R9nDgqYhYnY+KngBKRx4jgQl5eQIwKte1uuwIaneqz7BsZmYNVCSo7BQRS8teLy9YbhDwatnrhTmtiNnAyZL6StoDGEG6PgZgv7JbHS8B+pcKSTpB0hxgFjCmaDedmZnVR5HuocckTQLuzq8/Afy4QDlVSSt09BARcyXdDEwGVgEzgHYDRET8CjhC0uHABEk/jojfbdEo6WLgYoDBgwcXaY6ZmRXUblCJiKsljQY+SAoU4yPioQJ1L2Tz0QXAAcDiog2LiNuB2wEkfT3XB/C6pAERsUTSAGBplbJzJf0WOBKYVpE3HhgP0Nra6i6yOnt4+iJumTSPxSvWMNC38jXrcYoM1N8cEQ9GxOcj4nMR8VA+imjPVGCopCF57rBzgYntlCl/3/75eTBpWpjSkdJENl+QeQHwSF5vSGlgXtJBwKHAS0Xfz7bfw9MXcd2Ds1i0Yg0BLFqxhusenMXD0xd1ddPMrJMU6f46Dbi2Iu3MKmlbiIgNki4DJgEtwB0RMUfSmJw/TtL+pCOJ3sCmfOrwsIhYCTwgqS9pEP/SiHgrV30TcJ+kC4FXSGelQTqSGitpPbAJ+GxEvFFg+6xObpk0jzXrN26Rtmb9Rq65f+Y794zvLM8tWcmwAb079T3NrO0r6v8C+CxwiKTyix33An5RpPKIeBR4tCJtXNnya6RusWplT6qRvhw4tUr694HvF2mXNcbiFWuqpq/buKmTWwLDBvRm5DHudjPrbG0dqdxFGpC/kc3XggC8HRFvNrRV1pQG9unFoiqBZVCfXtx7yYld0CIz62xtXVH/vxHxUkScl6+uLz0cUKyqq08/lF67tGyR1muXFq4+/dAuapGZdTZfcW51UzrL65r7Z7Ju4yYG+ewvsx7HQcXqatSxg94ZlHeXl1nPU+TKeDMzs0IcVMzMrG4cVMzMrG4cVMzMrG4cVMzMrG4cVMzMrG4cVMzMrG4cVMzMrG4cVMzMrG58Rb0BvrmWmdWHg4q9c3Ot0r1QSjfXAhxYzKxDHFSs7jfX8g2yzHouj6lY3W+u5RtkmfVcDT1SkXQG8G3S7YT/OSJuqsg/DLgTOA74QkTcWpZ3BXARIOCfIuJbOX0f4F7gYNI96D8eEW9JOo10q+FdgXXA1RHx00ZuX3fhm2uZWb007EhFUgtwG+l+9sOA8yQNq1jtTeBy4NaKskeSAspw4GjgLElDc/ZYYEpEDAWmsPmulG8AH42Io4AL8K2FC/PNtcysXhrZ/TUcWBARL0TEOuAeYGT5ChGxNCKmAusryh4OPBURqyNiA/AEcHbOGwlMyMsTgFG5rukRsTinzwF2l7RbnbepWxp17CBuHH0Uu7akr8OgPr24cfRRHqQ3sw5rZPfXIODVstcLgRMKlp0NfE1SX2ANMAKYlvP2i4glABGxRFL/KuX/GJgeEWu3qeU9kG+uZWb10MigoippUaRgRMyVdDMwGVgFzAA2FHpT6QjgZuAjNfIvBi4GGDx4cJEqzcysoEZ2fy0EDix7fQCwuMa6W4mI2yPiuIg4mTT2Mj9nvS5pAEB+XloqI+kA4CHg/Ih4vka94yOiNSJa+/Xr16ENMjOztjUyqEwFhkoaImlX4FxgYtHCpW4tSYOB0cDdOWsiaSCe/PxIXq8P8CPguoj4RT02wMzMOqZh3V8RsUHSZcAk0inFd0TEHEljcv44SfuTxkp6A5skXQkMi4iVwAN5TGU9cGlEvJWrvgm4T9KFwCvAOTn9MuC9wBclfTGnfSQi3jmSMTOzxmrodSoR8SjwaEXauLLl10jdYtXKnlQjfTlwapX0rwJf3Z729hSe58vMGsXTtPQwbc3zZWa2vTxNSw/T1jxfzy1Z2UWtMrPuwkGlh2lrni/P2WVm28vdXz2M5/kys0bykUoP43m+zKyRfKTSw5TO8rrm/pms27iJQT77y8zqyEGlB/I8X2bWKO7+MjOzunFQMTOzunFQMTOzunFQMTOzuvFA/Tbw3FlmZtU5qHRQW3NnObCYWU/noNJBbc2dVTpNtxk8t2Qlwwb07upmmFk34zGVDmpr7qxm4nm+zKwRfKTSQZ47y8ysNh+pdJDnzjIzq62hQUXSGZLmSVogaWyV/MMkPSlpraSrKvKukDRb0px8m+FS+j6SJkuan5/3zul9Jf1M0ipJ323UNo06dhA3jj6KQX16IdIRyo2jj/IgvZkZDez+ktQC3AacBiwEpkqaGBHPla32JnA5MKqi7JHARcBwYB3wmKQfRcR8YCwwJSJuyoFqLHAt8Dvgi8CR+dEwo44d5CBiZlZFI49UhgMLIuKFiFgH3AOMLF8hIpZGxFRgfUXZw4GnImJ1RGwAngDOznkjgQl5eQI5IEXEbyPi56TgYmZmXaCRQWUQ8GrZ64U5rYjZwMm5S2sPYARwYM7bLyKWAOTn/nVqr5mZbadGnv2lKmlRpGBEzJV0MzAZWAXMADbUpVHSxcDFAIMHD65HlWZmljXySGUhm48uAA4AFhctHBG3R8RxEXEyaexlfs56XdIAgPy8tCONiojxEdEaEa39+vXrSFEzM2tHI4PKVGCopCGSdgXOBSYWLSypf34eDIwG7s5ZE4EL8vIFwCN1a7GZmW0XRRTqkdq2yqURwLeAFuCOiPiapDEAETFO0v7ANKA3sInU1TUsIlZK+m+gL2kQ//MRMSXX2Re4DxgMvAKcExFv5ryXcl27AiuAj1ScbVbZvmXAy9uxifsCb2xH+R2Vt6v5dNdt667bBc29bQdFRNWunoYGle5O0rSIaO3qdtSbt6v5dNdt667bBd1323xFvZmZ1Y2DipmZ1Y2DyvYZ39UNaBBvV/PprtvWXbcLuum2eUzFzMzqxkcqZmZWNw4qZmZWNz0qqBSYil+SvpPzZ0o6rr2ytabiz3nX5fXnSTq9LP14SbNy3nckVZvSplm37fGc9mx+bNfcbJ25XW3dPqHZ91k729bM++w0Sc/kffOMpA+XlanrPtuBtquu+6vuIqJHPEgXYD4PHEK6OHIG6ULL8nVGAD8mzVv2AeBX7ZUFvgGMzctjgZvz8rC83m7AkFy+Jec9DZyY3+fHwJndaNseB1qbdJ/tCXwQGAN8t+J9mn2ftbVtzbzPjgUG5uUjgUWN2Gc72HbVbX814tGTjlTanYo/v/7XSJ4C+ijNL9ZW2apT8ef0eyJibUS8CCwAhuf6ekfEk5G+If9Kxf1kmnXbtnMbuny7osbtE7rDPqu1bQ3Q2ds1PSJKcwrOAXaXtFsD9tkOsV3b0f5O05OCSpGp+Gut01bZWlPxt1XXwnba0VE7yraV3JkPy7+4nV0Onb1dbbWj2fdZe7rDPvtjYHpErKX++2xH2a6Seu2vuutJQaXIVPy11tmWafzrWVd7dpRtA/hURBwFnJQfn2mnrm19n/bWqefn3B32WVuafp9JOgK4GbikA+3oiB1lu6C++6vuelJQKTIVf6112ipbayr+tuo6oJ12dNSOsm1ExKL8/DZwF9vXLdbZ29VWO5p9n9XU7PtM0gHAQ8D5EfF82XvUc5/tKNtV7/1Vf40etNlRHqQbkr1AGlguDZYdUbHO/2HLgban2ysL3MKWA23fyMtHsOVg9gtsHsyemusvDSCO6A7bluvaN6+zC3A/MKZZtquszj9h68Hspt5ntbat2fcZ0Cev98dV2lK3fbajbFe991cjHl3egE7d2HR2xv+QzsT4Qk4bU9op+ctwW86fRdkZFtXK5vS+wBTSTcSmAPuU5X0hrz+PsjNPgFbSLZOfB75Lntmg2beNdIbRM8BM0uDit8mBtIm26yXSTeFWkf7DLJ2l0x322Vbb1uz7DLge+C3wbNmjfyP22Y6wXY3YX/V+eJoWMzOrm540pmJmZg3moGJmZnXjoGJmZnXjoGJmZnXjoGJmZnXjoGLWBknnSJor6Wf59d15BtrPdbCePpI+W/Z6oKT7693eivdcVY91zDrCpxSbtUHSY6SZY38maX/SzLMHbUM9BwM/jIgj693GNt5zVUS8a3vXMesIH6mYAZI+LenpPEnf9yS1SPpr0nTx4yTdAvwE6J/XOUnSeyQ9lu938d+SDst17SfpIUkz8uP3gZuA9+Syt0g6WNLsvP6v8hxPpbY8nu8FsqekOyRNlTRdUuWsuEh6l6Qpkn6d771RbZ1TJP1XbtNzksZJ2qks/2u5nU9J2i+nfTS3a7qk/yylm7Wrq6++9MOPrn4AhwP/AeySX/8Dab4lKLt3BXAwMLus3BRgaF4+AfhpXr4XuDIvtwDvrlL2ndfA54Av5+UBwP/k5a8Dn87LfUhXZO9Z0fadSVO8A+xLug1BqQdiVX4+hTTl/SG5PZOBj+W8AD6al78BXJ+X9y6r58+Bb3b1fvKjOR47b1MkMuteTgWOB6bmWcR70c5EjJLeBfw+8O9lM4+X7nfxYeB8gIjYCPyvyu6aWcV9pB/6LwEfB/49p38E+L+SrsqvdwcGA3PLmwJ8XdLJwCbSlOr7Aa9VvMfTEfFCbvvdpCOw+4F1wA/zOs8Ap+XlA4B78ySHuwIvttF+s3c4qJilH+YJEXFdB8rsBKyIiGO2980jYpGk5ZLeB3yCLadv/+OImNdG8U8B/YDjI2K9pJdIwWert6nxen1ElJY3svk34e+Bv42IiZJOAW4ovkXWk3lMxSx1Y32sdK9vpfuGtzkYHxErgRclnZPLSNLRZfX9RU5vkdQbeBvYq40q7wGuAd4dEbNy2iTgL0s3YZJ0bJVy7waW5oDyIaBWu4dLGpLHUj4B/Lyt7cv1LsrLF7Szrtk7HFSsx4uI50izwv5E0kxSV9SAAkU/BVwoaQZpxtjSIPkVwIckzSJ1KR0REcuBX0ianQf9K90PnEvqCiv5Cml685l5UP8rVcr9G9AqaVpuz29qtPVJ0skCs0ldWQ+1s203kLr2/ht4o511zd7hU4rNurncfXVVRJzVxU2xHsBHKmZmVjc+UjEzs7rxkYqZmdWNg4qZmdWNg4qZmdWNg4qZmdWNg4qZmdXN/wcIQkRfI9+wIQAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "%matplotlib inline\n",
    "fig, ax = plt.subplots()\n",
    "ax.plot(ccp_alphas[:-1], impurities[:-1], marker='o', drawstyle=\"steps-post\")\n",
    "ax.set_xlabel(\"effective alpha\")\n",
    "ax.set_ylabel(\"total impurity of leaves\")\n",
    "ax.set_title(\"Total Impurity vs effective alpha for training set\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We train a decision tree using the effective alphas."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 819,
   "metadata": {},
   "outputs": [],
   "source": [
    "clfs = []\n",
    "for ccp_alpha in ccp_alphas:\n",
    "    clf = tree.DecisionTreeRegressor(criterion=\"mse\", max_depth = 30,\n",
    "                                      min_samples_split = 2000,\n",
    "                                      min_samples_leaf = 1000,\n",
    "                                      ccp_alpha = ccp_alpha, splitter=\"best\").fit(x,y,exp)\n",
    "    clfs.append(clf)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 808,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of nodes in the last tree is: 7 with ccp_alpha: 0.0002247868330777364\n"
     ]
    }
   ],
   "source": [
    "print(f\"Number of nodes in the last tree is: {clfs[-1].tree_.node_count} with ccp_alpha: {ccp_alphas[-1]}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We remove the last element in clfs and ccp_alphas, because it is the trivial tree with only one node. Here we show that the number of nodes and tree depth decreases as alpha increases."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 809,
   "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": [
    "clfs = clfs[:-1]\n",
    "ccp_alphas = ccp_alphas[:-1]\n",
    "\n",
    "node_counts = [clf.tree_.node_count for clf in clfs]\n",
    "depth = [clf.tree_.max_depth for clf in clfs]\n",
    "\n",
    "fig, ax = plt.subplots(2, 1)\n",
    "ax[0].plot(ccp_alphas, node_counts, marker='o', drawstyle=\"steps-post\")\n",
    "ax[0].set_xlabel(\"alpha\")\n",
    "ax[0].set_ylabel(\"number of nodes\")\n",
    "ax[0].set_title(\"Number of nodes vs alpha\")\n",
    "ax[1].plot(ccp_alphas, depth, marker='o', drawstyle=\"steps-post\")\n",
    "ax[1].set_xlabel(\"alpha\")\n",
    "ax[1].set_ylabel(\"depth of tree\")\n",
    "ax[1].set_title(\"Depth vs alpha\")\n",
    "fig.tight_layout()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 811,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.metrics import mean_absolute_error, mean_squared_error, r2_score\n",
    "from sklearn.model_selection import cross_val_score\n",
    "\n",
    "alpha_loop_values = []\n",
    "for ccp_alpha in ccp_alphas:\n",
    "    regr_dt = tree.DecisionTreeRegressor(criterion=\"mse\", max_depth = 30,\n",
    "                                      min_samples_split = 2000,\n",
    "                                      min_samples_leaf = 1000,\n",
    "                                      ccp_alpha = ccp_alpha, splitter = 'best').fit(x, y, exp)\n",
    "    \n",
    "    predictions = regr_dt.predict(x)\n",
    "    cvs = np.mean(cross_val_score(regr_dt, x, y, cv = 5, scoring='neg_mean_squared_error'))\n",
    "    node_counts = regr_dt.tree_.node_count\n",
    "    depth = regr_dt.tree_.max_depth\n",
    "    \n",
    "    alpha_loop_values.append([ccp_alpha, node_counts, depth,\n",
    "                              cvs])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 812,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>alpha</th>\n",
       "      <th>node_counts</th>\n",
       "      <th>depth</th>\n",
       "      <th>cvs</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>247</td>\n",
       "      <td>13</td>\n",
       "      <td>0.950788</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.000044</td>\n",
       "      <td>13</td>\n",
       "      <td>3</td>\n",
       "      <td>0.950614</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.000049</td>\n",
       "      <td>13</td>\n",
       "      <td>3</td>\n",
       "      <td>0.950610</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.000065</td>\n",
       "      <td>11</td>\n",
       "      <td>3</td>\n",
       "      <td>0.950628</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.000096</td>\n",
       "      <td>9</td>\n",
       "      <td>3</td>\n",
       "      <td>0.950697</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      alpha  node_counts  depth       cvs\n",
       "0  0.000000          247     13  0.950788\n",
       "1  0.000044           13      3  0.950614\n",
       "2  0.000049           13      3  0.950610\n",
       "3  0.000065           11      3  0.950628\n",
       "4  0.000096            9      3  0.950697"
      ]
     },
     "execution_count": 812,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "alpha_results = pd.DataFrame(alpha_loop_values, \n",
    "                         columns=['alpha', 'node_counts', 'depth','cvs'])\n",
    "\n",
    "# taking absolute value of cvs\n",
    "alpha_results['cvs'] = np.abs(alpha_results.cvs)\n",
    "alpha_results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 813,
   "metadata": {},
   "outputs": [],
   "source": [
    "alpha_results = alpha_results.loc[(alpha_results['cvs']==np.min(alpha_results['cvs']))]\n",
    "best_alpha = alpha_results.loc[(alpha_results['depth']==np.max(alpha_results['depth']))]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 815,
   "metadata": {},
   "outputs": [],
   "source": [
    "regr_dt = tree.DecisionTreeRegressor(criterion=\"mse\", max_depth = int(best_alpha.depth),\n",
    "                                      min_samples_split = 2000,\n",
    "                                      min_samples_leaf = 1000,\n",
    "                                      ccp_alpha = float(best_alpha.alpha), splitter = 'best').fit(x, y, exp)\n",
    "y_deep_tree = regr_dt.predict(x)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 816,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "|--- bm <= 6.50\n",
      "|   |--- bm <= 1.50\n",
      "|   |   |--- ageph <= 55.50\n",
      "|   |   |   |--- value: [0.12]\n",
      "|   |   |--- ageph >  55.50\n",
      "|   |   |   |--- value: [0.09]\n",
      "|   |--- bm >  1.50\n",
      "|   |   |--- ageph <= 57.50\n",
      "|   |   |   |--- value: [0.15]\n",
      "|   |   |--- ageph >  57.50\n",
      "|   |   |   |--- value: [0.12]\n",
      "|--- bm >  6.50\n",
      "|   |--- bm <= 10.50\n",
      "|   |   |--- ageph <= 26.50\n",
      "|   |   |   |--- value: [0.24]\n",
      "|   |   |--- ageph >  26.50\n",
      "|   |   |   |--- value: [0.19]\n",
      "|   |--- bm >  10.50\n",
      "|   |   |--- value: [0.27]\n",
      "\n"
     ]
    }
   ],
   "source": [
    "text_representation = tree.export_text(regr_dt, feature_names = col_names)\n",
    "print(text_representation)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}