{"cells": [{"cell_type": "markdown", "metadata": {}, "source": ["# 2016 - Une solution \u00e0 la comp\u00e9tition de machine learning 2A\n", "\n", "Ce notebook a \u00e9t\u00e9 propos\u00e9 par un \u00e9tudiant pour la comp\u00e9tition organis\u00e9e pour ce cours : [classification binaire](http://www.xavierdupre.fr/app/ensae_teaching_cs/helpsphinx/all_notebooks.html#annee-2016)."]}, {"cell_type": "code", "execution_count": 1, "metadata": {}, "outputs": [{"data": {"text/plain": ["['ensae_competition_test_X.txt', 'ensae_competition_train.txt']"]}, "execution_count": 2, "metadata": {}, "output_type": "execute_result"}], "source": ["from pyensae.datasource import download_data\n", "download_data(\"ensae_competition_2016.zip\",\n", "              url=\"https://github.com/sdpython/ensae_teaching_cs/raw/master/_doc/competitions/2016_ENSAE_2A/\")"]}, {"cell_type": "code", "execution_count": 2, "metadata": {"collapsed": true}, "outputs": [], "source": ["# packages\n", "import pandas as pd\n", "import numpy as np\n", "from sklearn import svm, linear_model, datasets, metrics\n", "import seaborn as sns\n", "import matplotlib.pyplot as plt\n", "%matplotlib inline\n", "from statsmodels.nonparametric.kde import KDEUnivariate\n", "from statsmodels.nonparametric import smoothers_lowess"]}, {"cell_type": "code", "execution_count": 3, "metadata": {}, "outputs": [{"data": {"text/html": ["<div>\n", "<style>\n", "    .dataframe thead tr:only-child th {\n", "        text-align: right;\n", "    }\n", "\n", "    .dataframe thead th {\n", "        text-align: left;\n", "    }\n", "\n", "    .dataframe tbody tr th {\n", "        vertical-align: top;\n", "    }\n", "</style>\n", "<table border=\"1\" class=\"dataframe\">\n", "  <thead>\n", "    <tr>\n", "      <th></th>\n", "      <th>X1</th>\n", "      <th>X2</th>\n", "      <th>X3</th>\n", "      <th>X4</th>\n", "      <th>X5</th>\n", "      <th>X6</th>\n", "      <th>X7</th>\n", "      <th>X8</th>\n", "      <th>X9</th>\n", "      <th>X10</th>\n", "      <th>...</th>\n", "      <th>X15</th>\n", "      <th>X16</th>\n", "      <th>X17</th>\n", "      <th>X18</th>\n", "      <th>X19</th>\n", "      <th>X20</th>\n", "      <th>X21</th>\n", "      <th>X22</th>\n", "      <th>X23</th>\n", "      <th>Y</th>\n", "    </tr>\n", "    <tr>\n", "      <th>ID</th>\n", "      <th>LIMIT_BAL</th>\n", "      <th>SEX</th>\n", "      <th>EDUCATION</th>\n", "      <th>MARRIAGE</th>\n", "      <th>AGE</th>\n", "      <th>PAY_0</th>\n", "      <th>PAY_2</th>\n", "      <th>PAY_3</th>\n", "      <th>PAY_4</th>\n", "      <th>PAY_5</th>\n", "      <th>...</th>\n", "      <th>BILL_AMT4</th>\n", "      <th>BILL_AMT5</th>\n", "      <th>BILL_AMT6</th>\n", "      <th>PAY_AMT1</th>\n", "      <th>PAY_AMT2</th>\n", "      <th>PAY_AMT3</th>\n", "      <th>PAY_AMT4</th>\n", "      <th>PAY_AMT5</th>\n", "      <th>PAY_AMT6</th>\n", "      <th>default payment next month</th>\n", "    </tr>\n", "  </thead>\n", "  <tbody>\n", "    <tr>\n", "      <th>0</th>\n", "      <td>180000</td>\n", "      <td>1</td>\n", "      <td>2</td>\n", "      <td>1</td>\n", "      <td>47</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>...</td>\n", "      <td>99694</td>\n", "      <td>65977</td>\n", "      <td>67415</td>\n", "      <td>3700</td>\n", "      <td>3700</td>\n", "      <td>4100</td>\n", "      <td>2360</td>\n", "      <td>2500</td>\n", "      <td>2618</td>\n", "      <td>0</td>\n", "    </tr>\n", "    <tr>\n", "      <th>1</th>\n", "      <td>110000</td>\n", "      <td>2</td>\n", "      <td>2</td>\n", "      <td>1</td>\n", "      <td>35</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>...</td>\n", "      <td>4869</td>\n", "      <td>4966</td>\n", "      <td>5070</td>\n", "      <td>1053</td>\n", "      <td>1073</td>\n", "      <td>1081</td>\n", "      <td>178</td>\n", "      <td>184</td>\n", "      <td>185</td>\n", "      <td>1</td>\n", "    </tr>\n", "    <tr>\n", "      <th>2</th>\n", "      <td>70000</td>\n", "      <td>2</td>\n", "      <td>2</td>\n", "      <td>2</td>\n", "      <td>22</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>...</td>\n", "      <td>69927</td>\n", "      <td>50579</td>\n", "      <td>49483</td>\n", "      <td>2501</td>\n", "      <td>3001</td>\n", "      <td>2608</td>\n", "      <td>1777</td>\n", "      <td>1792</td>\n", "      <td>1793</td>\n", "      <td>1</td>\n", "    </tr>\n", "    <tr>\n", "      <th>3</th>\n", "      <td>200000</td>\n", "      <td>2</td>\n", "      <td>1</td>\n", "      <td>2</td>\n", "      <td>27</td>\n", "      <td>-2</td>\n", "      <td>-2</td>\n", "      <td>-2</td>\n", "      <td>-2</td>\n", "      <td>-2</td>\n", "      <td>...</td>\n", "      <td>1665</td>\n", "      <td>3370</td>\n", "      <td>-36</td>\n", "      <td>5610</td>\n", "      <td>15616</td>\n", "      <td>1673</td>\n", "      <td>3385</td>\n", "      <td>0</td>\n", "      <td>95456</td>\n", "      <td>0</td>\n", "    </tr>\n", "    <tr>\n", "      <th>4</th>\n", "      <td>370000</td>\n", "      <td>2</td>\n", "      <td>1</td>\n", "      <td>1</td>\n", "      <td>39</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>...</td>\n", "      <td>48216</td>\n", "      <td>47675</td>\n", "      <td>48074</td>\n", "      <td>2157</td>\n", "      <td>2000</td>\n", "      <td>1668</td>\n", "      <td>2000</td>\n", "      <td>3000</td>\n", "      <td>1000</td>\n", "      <td>0</td>\n", "    </tr>\n", "    <tr>\n", "      <th>5</th>\n", "      <td>260000</td>\n", "      <td>2</td>\n", "      <td>1</td>\n", "      <td>1</td>\n", "      <td>29</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>-2</td>\n", "      <td>-2</td>\n", "      <td>...</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>3090</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>141516</td>\n", "      <td>0</td>\n", "    </tr>\n", "    <tr>\n", "      <th>6</th>\n", "      <td>90000</td>\n", "      <td>2</td>\n", "      <td>1</td>\n", "      <td>1</td>\n", "      <td>43</td>\n", "      <td>-1</td>\n", "      <td>-1</td>\n", "      <td>2</td>\n", "      <td>-1</td>\n", "      <td>-1</td>\n", "      <td>...</td>\n", "      <td>7660</td>\n", "      <td>21175</td>\n", "      <td>4009</td>\n", "      <td>4367</td>\n", "      <td>9</td>\n", "      <td>7660</td>\n", "      <td>21175</td>\n", "      <td>4009</td>\n", "      <td>7452</td>\n", "      <td>0</td>\n", "    </tr>\n", "    <tr>\n", "      <th>7</th>\n", "      <td>220000</td>\n", "      <td>2</td>\n", "      <td>1</td>\n", "      <td>1</td>\n", "      <td>43</td>\n", "      <td>-1</td>\n", "      <td>3</td>\n", "      <td>2</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>...</td>\n", "      <td>1090</td>\n", "      <td>1090</td>\n", "      <td>0</td>\n", "      <td>167</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>1</td>\n", "    </tr>\n", "    <tr>\n", "      <th>8</th>\n", "      <td>50000</td>\n", "      <td>1</td>\n", "      <td>2</td>\n", "      <td>1</td>\n", "      <td>35</td>\n", "      <td>1</td>\n", "      <td>2</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>...</td>\n", "      <td>21260</td>\n", "      <td>70</td>\n", "      <td>29575</td>\n", "      <td>0</td>\n", "      <td>2052</td>\n", "      <td>1800</td>\n", "      <td>0</td>\n", "      <td>29935</td>\n", "      <td>1200</td>\n", "      <td>1</td>\n", "    </tr>\n", "    <tr>\n", "      <th>9</th>\n", "      <td>50000</td>\n", "      <td>2</td>\n", "      <td>3</td>\n", "      <td>2</td>\n", "      <td>40</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>...</td>\n", "      <td>8292</td>\n", "      <td>8465</td>\n", "      <td>8650</td>\n", "      <td>1271</td>\n", "      <td>1130</td>\n", "      <td>1000</td>\n", "      <td>307</td>\n", "      <td>325</td>\n", "      <td>436</td>\n", "      <td>0</td>\n", "    </tr>\n", "  </tbody>\n", "</table>\n", "<p>10 rows \u00d7 24 columns</p>\n", "</div>"], "text/plain": ["          X1  X2        X3       X4  X5    X6    X7    X8    X9   X10  \\\n", "ID LIMIT_BAL SEX EDUCATION MARRIAGE AGE PAY_0 PAY_2 PAY_3 PAY_4 PAY_5   \n", "0     180000   1         2        1  47     0     0     0     0     0   \n", "1     110000   2         2        1  35     0     0     0     0     0   \n", "2      70000   2         2        2  22     0     0     0     0     0   \n", "3     200000   2         1        2  27    -2    -2    -2    -2    -2   \n", "4     370000   2         1        1  39     0     0     0     0     0   \n", "5     260000   2         1        1  29     0     0     0    -2    -2   \n", "6      90000   2         1        1  43    -1    -1     2    -1    -1   \n", "7     220000   2         1        1  43    -1     3     2     0     0   \n", "8      50000   1         2        1  35     1     2     0     0     0   \n", "9      50000   2         3        2  40     0     0     0     0     0   \n", "\n", "              ...                   X15       X16       X17      X18      X19  \\\n", "ID            ...             BILL_AMT4 BILL_AMT5 BILL_AMT6 PAY_AMT1 PAY_AMT2   \n", "0             ...                 99694     65977     67415     3700     3700   \n", "1             ...                  4869      4966      5070     1053     1073   \n", "2             ...                 69927     50579     49483     2501     3001   \n", "3             ...                  1665      3370       -36     5610    15616   \n", "4             ...                 48216     47675     48074     2157     2000   \n", "5             ...                     0         0         0     3090        0   \n", "6             ...                  7660     21175      4009     4367        9   \n", "7             ...                  1090      1090         0      167        0   \n", "8             ...                 21260        70     29575        0     2052   \n", "9             ...                  8292      8465      8650     1271     1130   \n", "\n", "        X20      X21      X22      X23                          Y  \n", "ID PAY_AMT3 PAY_AMT4 PAY_AMT5 PAY_AMT6 default payment next month  \n", "0      4100     2360     2500     2618                          0  \n", "1      1081      178      184      185                          1  \n", "2      2608     1777     1792     1793                          1  \n", "3      1673     3385        0    95456                          0  \n", "4      1668     2000     3000     1000                          0  \n", "5         0        0        0   141516                          0  \n", "6      7660    21175     4009     7452                          0  \n", "7         0        0        0        0                          1  \n", "8      1800        0    29935     1200                          1  \n", "9      1000      307      325      436                          0  \n", "\n", "[10 rows x 24 columns]"]}, "execution_count": 4, "metadata": {}, "output_type": "execute_result"}], "source": ["# dataframe\n", "# df = pd.read_excel(\"default_of_credit_card_clients.xls\", header=[0, 1], encoding=\"utf8\", index_col=0, engine='openpyxl')\n", "df = pd.read_csv(\"ensae_competition_train.txt\", header=[0, 1], encoding=\"utf8\", index_col=0, sep=\"\\t\")\n", "df.head(10)"]}, {"cell_type": "code", "execution_count": 4, "metadata": {}, "outputs": [{"data": {"text/plain": ["MultiIndex(levels=[['X1', 'X10', 'X11', 'X12', 'X13', 'X14', 'X15', 'X16', 'X17', 'X18', 'X19', 'X2', 'X20', 'X21', 'X22', 'X23', 'X3', 'X4', 'X5', 'X6', 'X7', 'X8', 'X9', 'Y'], ['AGE', 'BILL_AMT1', 'BILL_AMT2', 'BILL_AMT3', 'BILL_AMT4', 'BILL_AMT5', 'BILL_AMT6', 'EDUCATION', 'LIMIT_BAL', 'MARRIAGE', 'PAY_0', 'PAY_2', 'PAY_3', 'PAY_4', 'PAY_5', 'PAY_6', 'PAY_AMT1', 'PAY_AMT2', 'PAY_AMT3', 'PAY_AMT4', 'PAY_AMT5', 'PAY_AMT6', 'SEX', 'default payment next month']],\n", "           labels=[[0, 11, 16, 17, 18, 19, 20, 21, 22, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 12, 13, 14, 15, 23], [8, 22, 7, 9, 0, 10, 11, 12, 13, 14, 15, 1, 2, 3, 4, 5, 6, 16, 17, 18, 19, 20, 21, 23]],\n", "           names=[None, 'ID'])"]}, "execution_count": 5, "metadata": {}, "output_type": "execute_result"}], "source": ["df.columns"]}, {"cell_type": "code", "execution_count": 5, "metadata": {"collapsed": true}, "outputs": [], "source": ["# Retrait 2\u00e8me ligne header\n", "\n", "df1 = df.copy()\n", "df1.columns = df1.columns.droplevel(-1)"]}, {"cell_type": "code", "execution_count": 6, "metadata": {}, "outputs": [{"data": {"text/plain": ["Index(['X1', 'X2', 'X3', 'X4', 'X5', 'X6', 'X7', 'X8', 'X9', 'X10', 'X11',\n", "       'X12', 'X13', 'X14', 'X15', 'X16', 'X17', 'X18', 'X19', 'X20', 'X21',\n", "       'X22', 'X23', 'Y'],\n", "      dtype='object')"]}, "execution_count": 7, "metadata": {}, "output_type": "execute_result"}], "source": ["df1.columns"]}, {"cell_type": "code", "execution_count": 7, "metadata": {}, "outputs": [{"data": {"text/plain": ["<matplotlib.text.Text at 0x2a5062587f0>"]}, "execution_count": 8, "metadata": {}, "output_type": "execute_result"}, {"data": {"image/png": "iVBORw0KGgoAAAANSUhEUgAAAsoAAACdCAYAAACtgl4mAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nFEVRmuH/A8bTWUyYoiGiAAAAAElFTkSuQmCC\n", "text/plain": ["<matplotlib.figure.Figure at 0x2a50588ac18>"]}, "metadata": {}, "output_type": "display_data"}], "source": ["# statistiques descriptives\n", "\n", "# param\u00e8tres des graphes\n", "fig = plt.figure(figsize=(12, 6)) \n", "alpha=alpha_scatterplot = 0.2 \n", "alpha_bar_chart = 0.55\n", "\n", "'''graphs - the history of past payment'''\n", "\n", "# September 2005\n", "plt.subplot2grid((3,6),(0,0))\n", "plt.scatter(df1.Y, df1.X6, alpha=alpha_scatterplot)\n", "# axe x\n", "plt.xlabel(\"Default\")\n", "# axe y\n", "plt.ylabel(\"Payment delay\")\n", "# grid - titre                        \n", "plt.grid(b=True, which='major', axis='y')  \n", "plt.title(\"September 2005\")\n", "\n", "# August 2005\n", "plt.subplot2grid((3,6),(0,1))\n", "plt.scatter(df1.Y, df1.X7, alpha=alpha_scatterplot)\n", "# axe x\n", "plt.xlabel(\"Default\")\n", "# axe y\n", "plt.ylabel(\"Payment delay\")\n", "# grid - titre                         \n", "plt.grid(b=True, which='major', axis='y')  \n", "plt.title(\"August 2005\")\n", "\n", "# July 2005\n", "plt.subplot2grid((3,6),(0,2))\n", "plt.scatter(df1.Y, df1.X8, alpha=alpha_scatterplot)\n", "# axe x\n", "plt.xlabel(\"Default\")\n", "# axe y\n", "plt.ylabel(\"Payment delay\")\n", "# grid - titre                         \n", "plt.grid(b=True, which='major', axis='y')  \n", "plt.title(\"July 2005\")\n", "\n", "# May 2005\n", "plt.subplot2grid((3,6),(0,3))\n", "plt.scatter(df1.Y, df1.X9, alpha=alpha_scatterplot)\n", "# axe x\n", "plt.xlabel(\"Default\")\n", "# axe y\n", "plt.ylabel(\"Payment delay\")\n", "# grid - titre                         \n", "plt.grid(b=True, which='major', axis='y')  \n", "plt.title(\"May 2005\")\n", "\n", "# April 2005\n", "plt.subplot2grid((3,6),(0,4))\n", "plt.scatter(df1.Y, df1.X10, alpha=alpha_scatterplot)\n", "# axe x\n", "plt.xlabel(\"Default\")\n", "# axe y\n", "plt.ylabel(\"Payment delay\")\n", "# grid - titre             \n", "plt.grid(b=True, which='major', axis='y')  \n", "plt.title(\"April 2005\")\n", "\n", "# March 2005\n", "plt.subplot2grid((3,6),(0,5))\n", "plt.scatter(df1.Y, df1.X11, alpha=alpha_scatterplot)\n", "# axe x\n", "plt.xlabel(\"Default\")\n", "# axe y\n", "plt.ylabel(\"Payment delay\")\n", "# grid - titre                          \n", "plt.grid(b=True, which='major', axis='y')  \n", "plt.title(\"March 2005\")"]}, {"cell_type": "code", "execution_count": 8, "metadata": {}, "outputs": [{"data": {"text/plain": ["<matplotlib.legend.Legend at 0x2a5065c50b8>"]}, "execution_count": 9, "metadata": {}, "output_type": "execute_result"}, {"data": {"image/png": "iVBORw0KGgoAAAANSUhEUgAAAvwAAAEQCAYAAADMNqDrAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAAAAAElFTkSuQmCC\n", "text/plain": ["<matplotlib.figure.Figure at 0x2a505801390>"]}, "metadata": {}, "output_type": "display_data"}], "source": ["fig = plt.figure(figsize=(12, 6)) \n", "alpha=alpha_scatterplot = 0.2 \n", "alpha_bar_chart = 0.55\n", "\n", "'''Graphs - bill statement'''\n", "\n", "# personnes pas en d\u00e9faut de paiement\n", "ax1 = plt.subplot2grid((3,6),(1,0), colspan=3)\n", "# kernel density\n", "df1.X12[df1.Y == 0].plot(kind='kde')    \n", "df1.X13[df1.Y == 0].plot(kind='kde')\n", "df1.X14[df1.Y == 0].plot(kind='kde')\n", "df1.X15[df1.Y == 0].plot(kind='kde')\n", "df1.X16[df1.Y == 0].plot(kind='kde')\n", "df1.X17[df1.Y == 0].plot(kind='kde')\n", "# axes\n", "plt.xlabel(\"Bill statement\")    \n", "plt.title(\"People distribution, no default\")\n", "# limites\n", "ax1.set_xlim(0, 200000)\n", "# l\u00e9gende\n", "plt.legend(('September','August','July','May','April','March'),loc='best')\n", "\n", "# personnes en d\u00e9faut de paiement\n", "ax2 = plt.subplot2grid((3,6),(1,3), colspan=3)\n", "# kernel density\n", "df1.X12[df1.Y == 1].plot(kind='kde')    \n", "df1.X13[df1.Y == 1].plot(kind='kde')\n", "df1.X14[df1.Y == 1].plot(kind='kde')\n", "df1.X15[df1.Y == 1].plot(kind='kde')\n", "df1.X16[df1.Y == 1].plot(kind='kde')\n", "df1.X17[df1.Y == 1].plot(kind='kde')\n", "# axes\n", "plt.xlabel(\"Bill statement\")    \n", "plt.title(\"People distribution, default\")\n", "# limites\n", "ax2.set_xlim(0, 200000)\n", "# l\u00e9gende\n", "plt.legend(('September','August','July','May','April','March'),loc='best')\n", "\n", "\n", "'''Graphs - amount of bill payed'''\n", "\n", "# personnes pas en d\u00e9faut de paiement\n", "ax1 = plt.subplot2grid((3,6),(2,0), colspan=3)\n", "# kernel density\n", "df1.X18[df1.Y == 0].plot(kind='kde')    \n", "df1.X19[df1.Y == 0].plot(kind='kde')\n", "df1.X20[df1.Y == 0].plot(kind='kde')\n", "df1.X21[df1.Y == 0].plot(kind='kde')\n", "df1.X22[df1.Y == 0].plot(kind='kde')\n", "df1.X23[df1.Y == 0].plot(kind='kde')\n", "# axes\n", "plt.xlabel(\"Amount of bill payed\")    \n", "plt.title(\"People distribution, no default\")\n", "# limites\n", "ax1.set_xlim(0, 25000)\n", "# l\u00e9gende\n", "plt.legend(('September','August','July','May','April','March'),loc='best')\n", "\n", "# personnes en d\u00e9faut de paiement\n", "ax2 = plt.subplot2grid((3,6),(2,3), colspan=3)\n", "# kernel density\n", "df1.X18[df1.Y == 1].plot(kind='kde')    \n", "df1.X19[df1.Y == 1].plot(kind='kde')\n", "df1.X20[df1.Y == 1].plot(kind='kde')\n", "df1.X21[df1.Y == 1].plot(kind='kde')\n", "df1.X22[df1.Y == 1].plot(kind='kde')\n", "df1.X23[df1.Y == 1].plot(kind='kde')\n", "# axes\n", "plt.xlabel(\"Amount of bill payed\")    \n", "plt.title(\"People distribution, default\")\n", "# limites\n", "ax2.set_xlim(0, 25000)\n", "# l\u00e9gende\n", "plt.legend(('September','August','July','May','April','March'),loc='best')"]}, {"cell_type": "code", "execution_count": 9, "metadata": {}, "outputs": [{"data": {"text/plain": ["<matplotlib.axes._subplots.AxesSubplot at 0x2a506461588>"]}, "execution_count": 10, "metadata": {}, "output_type": "execute_result"}, {"data": {"image/png": "iVBORw0KGgoAAAANSUhEUgAAAl4AAAIGCAYAAACF/IEcAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAIABJREFUeJzs3Xt0VPW9///XTkKGkFCoPxVFhRBUqLiQEjlUXd96vory\nqx5PrdVIYKAk4hDD2KlaCwq1nhS8HA3+ZMSmQQ1htFih1va0sYK9pNV66cXUojWSHOItRKUghCSQ\ny+zfHyyjFLbk8tl7hj3PR9es1WzG17wnmZm8896fvbdl27YtAAAAuC4t0QUAAACkChovAAAAj9B4\nAQAAeITGCwAAwCM0XgAAAB6h8QIAAPBIhlvBXTv+12jeA1NvM5r3057tRvPGDxlpNE+SZneY/fHc\noLeM5klSR0+n0bwhaelG8zLTzL/EMy2zmRmW2ecsSRtPMfs3VbzHMpqXlm7+LDZuZJpk22a/h5L0\nwXvDjeZ1dpt/LWZldhnN6+4x+9rOSI8bzXPD91wo8adv/4/50AEw3Sv8qyHH5rmaPxBMvAAAADzi\n2sQLAADgM8V7El2B55h4AQAAeISJFwAASAw7+dfYmcbECwAAwCNMvAAAQGLEmXgBAADAJUy8AABA\nQtis8Tqy1tZWN+oAAADwvX43XpFIxI06AABAqonH3b0lIcddjTNmzNCoUaMO2mbbthoaGlwvCgAA\nwI8cG6+cnBxVVlYqOzv7oO1FRUWuFwUAAFIAa7w+MXPmTGVkHNyXtbW1KT8/3/WiAAAA/Mix8bIs\nSwsXLlRtba3a29tVUVGh0tJS5eUl35W+AQDAUSje4+4tCTnuaiwpKVFhYaGKi4vV2NioUCik6upq\nL2sDAADwFceJVywWUzgcVjAYVE1NjVpaWhSJRFRfX+9lfQAAwK/suLu3JOQ48QoEAqqqqupd51VW\nVqampiZFo1GVl5d7ViAAAIBJtm1r2bJlamxs1GWXXaY5c+Yccp9Vq1bpT3/6kySpoaFBL7zwgpYs\nWaLGxkYNHTpUJSUlOu+88/r92I6NV0FBwSHbcnNzaboAAIAZCTrXVl1dnbKzs7V+/XrNmzdPV155\npQKBwEH3+eY3vylJeu+993Tffff1br/33ns1duzYAT8212oEAAApZcuWLcrPz5dlWZowYYK2bdvm\neN9NmzbpwgsvlHTgwMPFixcrHA5r586dA3ps167V+MDU24zmhf9aZjQvbri+3bKN5knSj7P2G827\nSLlG8ySpx/DztmQZzXPjLwuzFUoZxhOl1e8n9/fR/DOW0ky/dsy/pY0LWGafc3yI+SfdY2cZzUs3\n/GJMs82/GrsNfy5OPBpejAOUqGs17t27VyeffLLuueceZWdnf+blEH/3u9/pwQcflCQtWbJEI0aM\nUE1NjdasWaPFixf3+7GZeAEAgMRI0CWDcnJy1NnZqZtvvlnt7e0aPnz4Ye/3/vvvKysrq/dk8iNG\njJAknX/++dq6deuAnjKNFwAASClnnHGGXnnlFUnSG2+8oXHjxqmjo0PNzc0H3e/ZZ5/t3c0oqXcy\n9tprr2nMmDEDemwaLwAAkBgJOp3E1KlTtXv3bs2aNUsXX3yxAoGAXn311UN2HW7evFkXXHBB79c3\n3nijCgsLtXr1aoVCoQE9ZdfWeAEAACQjy7J05513HrRt+vTpmj59+kHb1q5de9DXa9asGfRj03gB\nAIDESNLL+riJXY0AAAAeYeIFAAASI0kv6+MmJl4AAAAeYeIFAAASI0GXDEokx4nXc889p2AwqKKi\nItXU1PRuX7RokSeFAQAA+I1j47V69WqtWbNGlZWVam5u1uLFi9XR0aE9e/Z4WR8AAPCrBJ3HK5Ec\ndzV2dXUpK+vANbYWLFiguro6hcNh7dixw7PiAAAA/MRx4jVv3jy1tLT0fj1lyhQtX778M6/gDQAA\n0GcJulZjIjk2Xs3NzVqyZIlqa2vV3t6uiooK3XLLLVq5cqWX9QEAAPiG467GkpISFRYWqri4WI2N\njQqFQoecOh8AAGCgbJsz1/eKxWIKh8MKBoOqqalRS0uLIpGI6uvrvawPAADANxwnXoFAQFVVVcrI\nOHCXsrIyNTU1KRqNqry83LMCAQCATyXpkYducmy8CgoKDtmWm5tL0wUAADBAnLkeAAAkRpIeeegm\n1xqvn/ZsN5oXn3qb0bxv/rXMaN7s/BuM5knSDZ1mfzwttvlLcwZs22het2UZzXNj2WaX4RpN50nS\nPw2/s01/NLrzczGbF5fZ17Yb9lhmfzLpRtMOSP6l08n/cx7dw2WV/YSJFwAASIwUXONFGw0AAOAR\nJl4AACAx4sm/M9o0Gi8AAJAY7GoEAACAW5h4AQCAxEjB00kw8QIAAPAIEy8AAJAYrPECAACAWxwn\nXs8++6weeughjRkzRhdeeKEefPBBBQIBzZkzR1/96le9rBEAAPhRCq7xcmy8HnroIa1bt0579uzR\n1772Nf3qV79SIBDQ7NmzabwAAAAGwLHxsixLmZmZOvbYYzV37lxlZ2dLkoYMGeJZcQAAwMdScOLl\nuMZr1qxZ2r9/vyQpFApJktra2nTOOed4UxkAAIDPODZe27dv18KFC1VbW6v29nZVVFSotLRUeXl5\nXtYHAAB8yrZ7XL0lI8ddjSUlJSosLFRxcbEaGxsVCoVUXV3tZW0AAAC+4jjxisViCofDCgaDqqmp\nUUtLiyKRiOrr672sDwAA+FU87u4tCTlOvAKBgKqqqpSRceAuZWVlampqUjQaVXl5uWcFAgAA+IVj\n41VQUHDIttzcXJouAABgBmeuBwAAgFu4ViMAAEiMJF2H5SbXGq/xQ0Yazdst22je7PwbjOb96C/3\nGc2TpKzR/8do3l0n/F+jeZKUbjgvw+yPOWV9kJZ6H2ammf4OurF7YZhtGc1Lxbef6c8wyfxr5/10\n3s9+wsQLAAAkBmu8AAAA4BYmXgAAIDFY4wUAAOARdjUCAADALUy8AABAYqTgrkYmXgAAAB5h4gUA\nABKDiddn+/vf/+5WHQAAAL7nOPG6//77D/ratm3V1NTo0ksvVSQScb0wAADgcyl4VKNj4/Xhhx/q\n3XffVUFBgY477jjZtq3nn39e5557rpf1AQAA+IZj47V8+XK9//77isVi2rdvnwoKCjRy5EhNmzbN\ny/oAAIBfscbrE//zP/+jUaNG6dvf/rYWLVqkX/3qV8rKytLq1au9rA8AAMA3HBuv9957T/Pnz1dt\nba0CgYAyMzO1e/dujRs3zsv6AACAX9lxd29JyHFXY0lJiQoLC1VcXKzGxkaFQiFVV1d7WRsAAICv\nOE68YrGYwuGwgsGgampq1NLSokgkovr6ei/rAwAAfhWPu3tLQo4Tr0AgoKqqKmVkHLhLWVmZmpqa\nFI1GVV5e7lmBAAAAfuHYeBUUFByyLTc3l6YLAACYkaTrsNzEtRoBAAA84tq1Gmd3mI3+cdZ+o3k3\ndJqtL2v0/zGaJ0kdzX8wmnf72cuM5klSt2yjeWbTjg6WC5mnd6e7kGrO0fA37tFQY7vhP53deM6m\nf8l0G85zY/pg+vuYY7vxKZEkknQdlpuYeAEAAHjEtYkXAADAZ2LiBQAAALcw8QIAAIlhp97KXhov\nAACQGOxqBAAAgFuYeAEAgMRg4gUAAAC3MPECAACJwSWDPvHII49Ikl577TUFg8He28svv+xZcQAA\nAH7iOPH63e9+p+LiYt1111268847dcopp6i1tVXXXnutHn/8cS9rBAAAfsQar0988MEH2rhxoz76\n6COdcsopkqThw4d7VhgAAIDfOE68vvrVryotLU3FxcXau3evcnJytHfvXtXV1XlZHwAA8KsUPIGq\n48TLsiz97Gc/0zHHHKO0tDRVVFRo0aJFWrlypZf1AQAA+IbjxKukpESFhYUqLi5WY2OjQqGQqqur\nvawNAAD4GWu8PhGLxRQOhxUMBlVTU6OWlhZFIhHV19d7WR8AAIBvOE68AoGAqqqqlJFx4C5lZWVq\nampSNBpVeXm5ZwUCAACfSsGJl2PjVVBQcMi23Nxcmi4AAIAB4sz1AAAgMVLwzPWuNV436C2jeRcp\n12hei232MpV3nfB/jeZJ0u1nLzOb9+flRvMkqWfrS0bz7MYtZvNa9xjNkyS1dRiNs/e2Gc2TpC2V\n7Ubz0iyzh3z32JbRPDeYrtCNg+bXZZj9CM8w/qzNOxpqNG3J+O2JLgEGMfECAAAJYcc5jxcAAABc\nwsQLAAAkRgoe1cjECwAAwCNMvAAAQGJwVCMAAIBHWFwPAAAAtzDxAgAAicHiegAAALilTxOv7u5u\n/fOf/9TnP/95ZWZmul0TAABIBUy8PvHxxbCfeuopBYNB/fd//7euueYarV271qvaAAAAfMVx4vW3\nv/1NkrRx40b96Ec/UlragR5t9uzZmj9/vifFAQAAH7NT76hGx8br1FNP1erVqzVhwgTdf//9mjJl\nihoaGjRy5Egv6wMAAPANx12NkydP1plnnqlhw4Zp165devnll3X88cdr0qRJXtYHAAD8Kh5395aE\nHBuvlpYWPfLIIzr77LO1ZMkSjRgxQk8++aTGjRvnZX0AAAC+4birsaSkRIWFhSouLlZjY6NCoZCq\nq6u9rA0AAPgZZ67/RCwWUzgcVjAYVE1NjVpaWhSJRFRfX+9lfQAAAEbZtq2lS5dq1qxZeuyxxw57\nn2g0qv/8z//U3Llz9dRTT0mS3n77bc2ePVuFhYV6++23B/TYjhOvQCCgqqoqZWQcuEtZWZmampoU\njUZ7TzUBAAAwYAm6SHZdXZ2ys7O1fv16zZs3T1deeaUCgcAh91uyZInOPffc3q8feugh3XbbbZKk\nyspKLV++vN+P7TjxKigo6G26Ppabm0vTBQAAjmpbtmxRfn6+LMvShAkTtG3btsPe795779U111yj\nd955R5K0bds2TZw4URMnTtRbb701oMfmkkEAACAx4ra7Nwd79+5VZmam7rnnHmVnZ6u1tfWQ+8yb\nN09PPvmkwuGw7rnnngPlxuN68cUX9etf/1rxAR416dpFsjt6Oo3m9cjsAryA4ZO2pRtNO6Db8HPu\n2fqS0TxJSj9tutE8e/QEo3k92+qM5kmS9uw0Gmft2WU0T5JG5LxiPNOk7u7U+5vPti0XMs1+hKfL\nfI2mEwNHwbzAdIVtO7hUn2k5OTnq7OzUzTffrBUrVmj48OGH3GfEiBGSpC9+8Yu9jVdaWpq+9KUv\nSdKAr+ST/K9gAADgS3Y87urNyRlnnKFXXjnwB+obb7yhcePGqaOjQ83Nzb33+XgK9s477/Q2YWPH\njlV9fb3q6+s1ZsyYAT1n1yZeAAAAyWjq1KnauHGjZs2apUsvvVSBQEAvvfSSHnjgAcViMUnSXXfd\npcbGRlmWpdtvv12StGDBAt1yyy2SpDvvvHNAj03jBQAAEiNB5/GyLOuQxmn69OmaPv2T5TMrVqw4\n5L/Lzc3V+vXrB/XY7GoEAADwCBMvAACQGAk6j1ciMfECAADwCBMvAACQGCl4rUYaLwAAkBgDPAnp\n0cxxV+OePXu8rAMAAMD3HCdeF154oc466yyde+65uuiii3TKKad4WRcAAPA7djV+YuLEiVq9erV+\n//vf6/7771dzc7OmT5+uGTNmaNKkSV7WCAAA4AufucYrEAjooosu0kUXXaSuri798Y9/1I9+9KPD\nnlQMAACgXzidxCfOOeecg74eMmSIzj77bJ144omuFwUAAOBHjo1XWlqa5s+fr9raWrW3t6uiokKL\nFi1SXl6el/UBAAC/itvu3pKQ467GkpISFRYWqri4WI2NjQqFQlq7dq2HpQEAAPiL48QrFospHA4r\nGAyqpqZGLS0tikQiqq+v97I+AADgU3Y87uotGTlOvAKBgKqqqpSRceAuZWVlampqUjQaVXl5uWcF\nAgAA+IVj41VQUHDIttzcXJouAABgRpKuw3ITF8kGAADwiGvXahySlm40z5JlNK/bMpuX4ULTbjrS\nbtxiOFGyR08wmmdljzSal37qNKN5khR/93WjeXa6+bfhvv1mM4cGuo3mpVmp91euG6tNxsYzjeZt\nt8z+nCXJ7G8CKW74kzHN8O8WN+zvGJLoEtzDxAsAAABucW3iBQAA8Jk4cz0AAADcwsQLAAAkBmu8\nAAAA4BYmXgAAICFsJl4AAABwCxMvAACQGCk48epz47Vv3z7t27dPI0eaPcElAABIUUl6IWs3OTZe\njz76qH72s59p6NChuuyyy1RTU6P09HR94Qtf0Le//W0vawQAAPAFx8brl7/8pTZs2KCuri5deuml\neuaZZ2RZlubMmeNlfQAAwK/Y1fipf8jI0MaNG2XbtkaMGKHa2loNGzbMy9oAAAB8xbHx+vrXv66P\nPvpIeXl5qqio0Nq1a9Xd3a1zzz3Xy/oAAIBfpeDEy/F0Ei0tLfrtb38r27aVlZWl7Oxsvf766xo3\nbpyX9QEAAPiG48SrpKREhYWFKi4uVmNjo0KhkKqrq72sDQAA+JhtM/HqFYvFFA6HFQwGVVNTo5aW\nFkUiEdXX13tZHwAAgG84TrwCgYCqqqqUkXHgLmVlZWpqalI0GlV5eblnBQIAAJ9KwTVejo1XQUHB\nIdtyc3NpugAAAAaISwYBAIDESMGJFxfJBgAA8IhrE6/MNLPRpjvEHsN5RwO7dY/xzJ5tdUbz0k+d\nZjTPGpptNE8yX2NP1htG8yTJtn9rNK+72+w7MDPQbTTvaGDblvHMrg6z04ITbfO/EpqtLqN5AcO/\nDdyYPqTL7M+6qyvdaF4ysZl4AQAAwC2s8QIAAInBxAsAAABuYeIFAAASI57oArzHxAsAAMAjTLwA\nAEBCcFQjAAAAXMPECwAAJEYKTryO2Hjt3btXe/fu1ciRIzV06FAvagIAAKkgBRfXOzZeTz/9tB5+\n+GFZlqW33npLX/jCF5SZmanrr79ekydP9rJGAAAAX3Bc41VdXa0f/ehH2rBhgzZv3qwRI0bovvvu\n09133+1lfQAAwKfsuO3qLRk5TryGDh2qP/7xjxozZoxef/11SVJOTo4yMlgWBgAAMBCOXdQVV1yh\n119/Xb/97W81ZswY3XHHHWpra9O0aWYvEAwAAFJUCq7xctzV2NzcrJdeekkXXHCBCgsL9eijj6q0\ntFR5eXle1gcAAOAbjhOvkpISFRYWqri4WI2NjQqFQqqurvayNgAA4GPJug7LTY4Tr1gspnA4rGAw\nqJqaGrW0tCgSiai+vt7L+gAAAHzDceIVCARUVVXVu5i+rKxMTU1NikajKi8v96xAAADgUym4xsux\n8SooKDhkW25uLk0XAADAAHFuCAAAkBA2Ey9zMi2z0ZbRNKnLMp14FGjrMJ+5Z6fRuPi7rxvNSz81\n+U9/kn7SxESXcERpltkFsLZt/v0XGNptPNOkeI/555xm+C3txjLnk+0hRvM+sHqM5gWM/3aR0gx/\nI3tceL8gcZh4AQCAxEjBiZfjUY0AAAAwi4kXAABIiFRc48XECwAAwCNMvAAAQGIw8QIAAIBbmHgB\nAICEYI0XAAAAXMPECwAAJEQqTrwcG69XX31Vq1evVkdHh2zblmVZysrK0nXXXacpU6Z4WSMAAPAh\nGq9PueOOO1RRUaGRI0dqw4YNuuqqq7Rr1y5dd911evzxx72sEQAAwBcc13gNGzZMf/3rX7V7926d\neeaZ2r17t+rq6pSVleVlfQAAwK9sy91bEnKceK1cuVJPPPGEnnrqKe3du1c5OTmaNGmSVq5c6WV9\nAAAAvuHYeI0cOVKhUOiQ7a2tra4WBAAAUkMqrvHq9+kkIpGIG3UAAAD4nuPEa8aMGRo1apRs25Yk\nWZYl27bV0NDgWXEAAMC/7HhyrsNyk2PjlZOTo8rKSmVnZx+0vaioyPWiAAAA/Mix8Zo5c6YyMg7+\n57a2NuXn57teFAAA8D/WeH2KZVlauHChamtr1d7eroqKCpWWliovL8/L+gAAAHzDceJVUlKiwsJC\nFRcXq7GxUaFQSNXV1X0PttKNFNibJ7P7gbus5N+vbLpCe2+b4UTJ2rPLaJ6dbvYqVj1ZbxjNk6T0\nkyYazzTNsuxEl+A52/BTzhzWYzQv3m3+Mydtj9nMuMy/bizDn2Qnxc3+btmeZvbnLEnphn+/7O/0\n79X97CQ915abHCdesVhM4XBYwWBQNTU1amlpUSQSUX19vZf1AQAA+IZjGx0IBFRVVdW7zqusrExN\nTU2KRqMqLy/3rEAAAOBPqbjGy7HxKigoOGRbbm4uTRcAAMAA+XfHMQAASGqpeB6vfp+5HgAAAAPD\nxAsAACSE6aORjwZMvAAAADzCxAsAACQEa7wAAADgmn43XgsXLnSjDgAAkGLsuOXqLRk57mqcM2fO\nIdts21ZDQ4OrBQEAAPiVY+O1Z88e/eQnP1FmZuZB24uKilwvCgAA+F8qHtXo2HitWLFCXV1dhzRe\ny5Ytc70oAADgf8m6O9BNjo3X5MmTD7v9+OOPd60YAAAAP+v34vpIJOJGHQAAIMXYtuXqLRk5Trxm\nzJihUaNGHbSNxfUAAOBoZ9u2li1bpsbGRl122WWHPaCwqqpKNTU1GjJkiJYuXapJkyZpyZIlamxs\n1NChQ1VSUqLzzjuv34/t2Hjl5OSosrJS2dnZB21ncT0AADDBjifmcevq6pSdna3169dr3rx5uvLK\nKxUIBA66z8UXX6yioiI1NzdrxYoVWr16tSTp3nvv1dixYwf82I67GmfOnKmMjIP7sra2NuXn5w/4\nwQAAABJty5Ytys/Pl2VZmjBhgrZt23bIfU466SRJUlpaWm9TZlmWFi9erHA4rJ07dw7osR0bL8uy\ntHDhQtXW1qq9vV0VFRUqLS1VXl7egB4IAADg0+K25erNyd69e5WZmal77rlH2dnZam1tdbxvZWWl\ngsGgJGnJkiV6/PHHdckll2jNmjUDes6OjVdJSYnuv/9+rVq1Sueee67i8biqq6t1ySWXDOiBAAAA\nkkFOTo46Ozt18803q729XcOHDz/s/Z555hkNGzZMU6dOlSSNGDFCknT++edr69atA3psxzVesVhM\nmzZtUjAY1PTp01VRUaFIJKLS0lJNmDDhiMEbTzF7GcjV75s9OuGfhi8P/kGa+R3Vp3enG83bUtlu\nNE+SRuS8YjRv336zPxjb/q3RPDdYlvkzCE76y/9nPBPJ5+xJtxjN251m+INR0jHxTqN5Qyyzn7Vd\ntvlLFg9NM/ucn87MMponSdONJw5Moo48POOMM7R582bNnDlTb7zxhsaNG6eOjg7t2rVLo0ePliQ1\nNjbqqaee0gMPPND737W2tmr48OF67bXXNGbMmAE9tuO7LBAIqKqqqnedV1lZmZqamhSNRlVeXj6g\nBwMAAEi0qVOnauPGjZo1a5YuvfRSBQIBvfTSS3rggQcUi8UkSdFoVO+9957mz5+vcePGqaysTDfe\neGPvbsq77757QI9t2bY7J+x/d/oFRvNWv3+C0bwT4mb/ynFj4pXXbbbGM+NuTLz2Gc0zP/FKzvO4\nfBoTLwzU746KiVeX0byjY+LVYzRvU+ZQo3mSdPtbjxnPHIg3Tnd3+dLEN2tczR8I8684AAAAHJb5\nP28AAAD6IBUvks3ECwAAwCNMvAAAQELY8eRfh2saEy8AAACPMPECAAAJ8Vlnl/crJl4AAAAecWy8\n/vGPfygcDuumm25SXV1d7/bvfe97nhQGAAD8zbYtV2/JyHFX44oVK3TvvffKsixVVlbqN7/5jW64\n4YbDXsEbAAAAR+bYeHV2duqEEw6cLf673/2uNm3apEWLFmnPnj2eFQcAAPwrFc/j5dh4XXLJJdq+\nfbtOPPFESdLFF1+s8ePHH3SxSAAAgIFKxcX1jo3X/PnzD9k2fvx4lZWVuVkPAACAb/X7qMZIJOJG\nHQAAIMWwuP5TZsyYoVGjRh20zbZtNTQ0uF4UAACAHzk2Xjk5OaqsrFR2dvZB24uKilwvCgAA+F8q\nLq533NU4c+ZMZWQc3Je1tbUpPz/f9aIAAAD8yLHxsixLCxcuVG1trdrb21VRUaHS0lLl5eV5WR8A\nAPCpuG25ektGjrsaS0pKVFhYqOLiYjU2NioUCqm6urrPwfEes0/Y9LWN4obzjgZpVvLPdIcGuo3m\ndXebvyrW0fB9RGqYOGaH0byX3j3BaJ4k9cjs7wLbTjea58bvAtOLus0+YySa42+lWCymcDisYDCo\nmpoatbS0KBKJqL6+3sv6AACAT3FU46cEAgFVVVX1rvMqKytTU1OTotGoysvLPSsQAADALxwbr4KC\ngkO25ebm0nQBAAAjknUdlpvML4ABAADAYTlOvAAAANyUiocqMfECAADwCBMvAACQEKzxAgAAgGuY\neAEAgIRI1nNtucmx8aqrq1NFRYXGjRunGTNmqLy8XOnp6SopKdF5553nZY0AAAC+4Nh43X333Vq1\napV27typUCikDRs2aNiwYbrmmmtovAAAwKCl4uX7HBuvtLQ0HXfccTruuOM0c+ZMHX/88ZKkIUOG\neFYcAACAnzg2XjNmzFBnZ6cyMzN16623SpL27dun008/3bPiAACAf9mGL6J+NHBsvIqKig7ZNnTo\nUN1www2uFgQAAFJDPAXPoNrv00lEIhE36gAAAPC9z9zVOGrUqIO22bathoYG14sCAAD+F2dX4ydy\ncnJUWVmp7Ozsg7YfbhckAAAAjsyx8Zo5c6YyMg7+57a2NuXn57teFAAA8L9UXFzvuMbLsiwtXLhQ\ntbW1am9vV0VFhUpLS5WXl+dlfQAAAL7hOPEqKSlRYWGhiouL1djYqFAopOrq6j4Hp6WbPVTBdE/c\nYzjv8/E07Uozeyo40yeW63Hh0gzd3WYv95lmmX3dZAa6jeZJqXmJCySn/W1mr/r271PeNZonSS//\n9USjeQGZ/YwYZpn+bSBZhj/H/HzgXyqeQNXxt2YsFlM4HFYwGFRNTY1aWloUiURUX1/vZX1HDdNN\nFwAA8B/HP5cCgYCqqqp613mVlZWpqalJ0WhU5eXlnhUIAAD8KRXXeDk2XgUFBYdsy83NpekCAAAY\nILMLBAAAAPooFRfpmF0ZDQAAAEdMvAAAQEIw8QIAAIBrmHgBAICESMWjGpl4AQAAeKRfjdf27dvd\nqgMAAKRwp4/zAAAgAElEQVSYuOXuLRk57mrcuHHjQV/btq1169bpG9/4hq688krXCwMAAPAbx8br\nl7/8pbq7u3XZZZcpMzNTtn3galEfn8keAABgMOKs8fpEVVWVbr75Zm3ZskXbt2/X+eefrxNOOEGX\nX365l/UBAAD4xmeOryZPnqzJkyeroaFB9913X+/UCwAAYLBSsavo037DU089Vd///vclSa2trRo+\nfLirRQEAAP/jBKp9EIlE3KgDAADA9xwnXjNmzNCoUaMO2mbbthoaGlwvCgAA+F/cSr3F9Y6NV05O\njiorK5WdnX3Q9qKiIteLAgAA8CPHxmvmzJmHnDqira1N+fn5rhcFAAD8LxUX1zuu8bIsSwsXLlRt\nba3a29tVUVGh0tJS5eXleVkfAACAbzhOvEpKSlRYWKji4mI1NjYqFAqpurq6z8Fp6Wb72DTDJ1nr\nSr3dyjAkMLTbaB5nacFAdXWlG83b/sbnjOZJ0pTx7xvNe7Vx1JHv1A+Wbf6SxQHDx+r5+dcVRzV+\nSiwWUzgcVjAYVE1NjVpaWhSJRFRfX+9lfQAAAL7hOPEKBAKqqqrqXedVVlampqYmRaNRlZeXe1Yg\nAADwp2S9kLWbHBuvgoKCQ7bl5ubSdAEAAAwQV7wGAAAJwUWyAQAA4BomXgAAICFS8aBuJl4AAAAe\nYeIFAAASIhWPamTiBQAA4BHHxuvnP/+5JOmdd95RJBLRnDlztGjRIm3dutWz4gAAgH/FXb4lI8fG\na+PGjZKk//qv/1JRUZEee+wxLVmyRLfffrtXtQEAAPiK4xqv3bt364UXXtDOnTs1ZcoUSdIpp5yi\nnp4ez4oDAAD+xVGNn3LRRRfpL3/5iy644AK1trZKkvbu3atTTz3Vs+IAAAD8xHHiFQ6HD9mWk5Oj\nxYsXu1oQAABIDRzV2AeRSMSNOgAAQIpJxcX1jhOvGTNmaNSoUbJtW5Z1oCW1bVsNDQ2eFQcAAOAn\njo1XTk6OKisrlZ2dfdD2oqIi14sCAAD+l6xTKTc57mqcOXOmMjIO7sva2tqUn5/velEAAAB+5Nh4\nWZalhQsXqra2Vu3t7aqoqFBpaany8vK8rA8AAPiUbbl7S0aOuxpLSkpUWFio4uJiNTY2KhQKqbq6\n2svaDpJm+GQfccNnD3FjXGo6M0lfgykvcxjnxsPA7O80e7nd/V3mL987ZEeW0bzJ4983mvdq4yij\neZKUZpv9/cK1/fzF8ecZi8UUDocVDAZVU1OjlpYWRSIR1dfXe1kfAADwKY5q/JRAIKCqqqredV5l\nZWVqampSNBpVeXm5ZwUCAAD4hWPjVVBQcMi23Nxcmi4AAGBEsk6l3MSuYwAAAI+YX0kJAADQB1wk\nGwAAAK5h4gUAABKCi2QDAADANUy8AABAQnBU46d8fJb6P/zhDyosLNTcuXM1e/ZsPf30054VBwAA\n4CeOE69f//rX+sY3vqHKyko98sgjysrKUnd3t4LBoL7yla94WSMAAPChVJx4OTZexx13nJ566imd\neeaZ+sUvfqGzzjpLjY2NCgQCXtYHAADgG46N1x133KGNGzfq/fff15tvvqk//OEPmjhxou677z4v\n6wMAAD6Viufx+sxrNc6ZM0dz5sw5aHtra6vrRQEAAPhRv08nEYlE3KgDAACkmLjl7s2JbdtaunSp\nZs2apccee+yw99m1a5eKiop09dVXq66uTpL09ttva/bs2SosLNTbb789oOfsOPGaMWOGRo0aJds+\nMAi0LEu2bauhoWFADwQAAJAM6urqlJ2drfXr12vevHm68sorD1nD/sQTT2jOnDmaOnWqlixZosrK\nSj300EO67bbbJEmVlZVavnx5vx/bsfHKyclRZWWlsrOzD9peVFTU7wcBAAD4V4k6qnHLli3Kz8+X\nZVmaMGGCtm3bpokTJx5yn4KCAn3+859XR0eHJB10v7feemtAj+3YeM2cOVMZGQf/c1tbm/Lz8wf0\nQAAAAJ+WqMX1e/fu1cknn6x77rlH2dnZh12/vnfvXjU3N+vZZ5/t3fsXj8f14osvqq2tTfH4wNpG\nxzVelmVp4cKFqq2tVXt7uyoqKlRaWqq8vLwBPRAAAEAyyMnJUWdnp26++Wa1t7dr+PDhh73P6NGj\nddVVV8myDiwYS0tL05e+9CVdeOGFSksb2FUXHf+rkpIS3X///Vq1apXOPfdcxeNxVVdX65JLLhnQ\nAwEAAHxaXLarNydnnHGGXnnlFUnSG2+8oXHjxqmjo0PNzc2H3GfXrl0aOnSoJGns2LGqr69XfX29\nxowZM6Dn7LirMRaLadOmTQoGg5o+fboqKioUiURUWlqqCRMmHDHYtlPrkuNHw9XG3Rjpmv45m97f\n78brMN5j+Dl3p9Z7BeZ095j95Olx4f3S1ZVuNK9jT6bRvLNOe99oniS91HCi8UyYNXXqVG3cuFGz\nZs3SpZdeqkAgoJdeekkPPPCAYrGYJOnqq6/WjTfeqB/+8IdavHixJGnBggW65ZZbJEl33nnngB77\nM8/jVVVV1bvOq6ysTE1NTYpGoyovLx/QgwEAAHwsUYvrLcs6pHGaPn26pk+f3vv1Mccco7Vr1x50\nn9zcXK1fv35Qj+3YeBUUFByyLTc3l6YLAABggBwbLwAAADel4iWDjoalSQAAAL7AxAsAACREotZ4\nJRITLwAAAI8w8QIAAAnxWRey9ismXgAAAB5h4gUAABLis84u71eOjdd5552n8847TxdeeKG+/OUv\nKysry8u6AAAAfMex8crLy9O3vvUtbdq0Sdddd52GDRumCy64QBdccIGOOeYYL2sEAAA+lHrzriOs\n8Ro9erTmz5+vtWvX6vbbb9f+/fv17W9/26vaAAAAfMVx4vVv//ZvB319/PHHa86cOZozZ47rRQEA\nAP/jPF6fcv311x92e2trq2vFAAAA+Fm/TycRiUTcqAMAAKSYuGxXb8nIcVfjjBkzNGrUqIO22bat\nhoYG14sCAADwI8fGKycnR5WVlcrOzj5oe1FRketFAQAA/0vOmZS7HBuvmTNnKiPj4H9ua2tTfn6+\n60UBAAD/Y3H9p1iWpYULF6q2tlbt7e2qqKhQaWmp8vLyvKwPAADANxwnXiUlJSosLFRxcbEaGxsV\nCoVUXV3d5+AP3htupMCPBSyzV9LcY5nts4fZ5q/02W74SprrMsxfIcq2zWaOjWcazevqMD/ITusw\nnLfH/Gvn7Em3GM2bOGaH0bz9beZfi11d6Ubz9nearbG7x/ylcc+qW2k0b+v0wx/NPhht+8y+p99p\nzz7ynfph6IfmZy4npO8zmrfJ8u/V/ZJ1AbybHD8JYrGYwuGwgsGgampq1NLSokgkovr6ei/rAwAA\n8A3HNjoQCKiqqqp3nVdZWZmampoUjUZVXl7uWYEAAMCfUm/e9RmNV0FBwSHbcnNzaboAAAAGyL87\njgEAQFLjqEYAAAC4hokXAABICDsFV3kx8QIAAPAIEy8AAJAQrPECAACAaxwnXps3b1ZVVZWGDx+u\nOXPmqLKyUh0dHfra176mYDDoZY0AAMCHUvHM9Y6N18MPP6xHH31U7e3t+trXvqann35aQ4YMUWFh\nIY0XAADAAHzmGq8///nP6u7u1rBhw/T2229r2LBhsu3U604BAIB5qdhRODZe3//+9/XjH/9Yp512\nmqLRqFatWqXu7m7deuutXtYHAADgG46N12mnnaZly5b1fr1y5UpJUmtrq/tVAQAA30vFNV79Pqox\nEom4UQcAAIDvOU68ZsyYoVGjRsm2bVmWJUmybVsNDQ2eFQcAAPwrFc/j5dh45eTkqLKyUtnZ2Qdt\nLyoqcr0oAAAAP3JsvGbOnKmMjIP/ua2tTfn5+a4XBQAA/I9rNX6KZVlauHChamtr1d7eroqKCpWW\nliovL8/L+gAAgE/FXb4lI8eJV0lJiQoLC1VcXKzGxkaFQiFVV1f3ObizO91IgR+LDzHbFZutzp1z\nkZh+0WTIMpwopRvO3G51G8070TZ/OVLTP2s3jurZnWb2eb/07glG8/59yrtG8yRp+xufM5q3v8vs\n97DHNv/+2zr9eqN5p70UNZonSX+Z/G2jeV2W2SvdHZe+32ieJP2zO2A0b6jRNCSa4ys4FospHA4r\nGAyqpqZGLS0tikQiqq+v97I+AADgU7bL/0tGjn/SBQIBVVVV9a7zKisrU1NTk6LRqMrLyz0rEAAA\nwC8cG6+CgoJDtuXm5tJ0AQAAI5J1HZabzO4sBwAAgCPzK48BAAD6IG4n5zosNzHxAgAA8AgTLwAA\nkBCpN+9i4gUAAOAZJl4AACAh3DiBdLJzbLz+8pe/KBqNavv27WpubtbUqVN17LHH6lvf+pZOOeUU\nL2sEAADwBcddjffcc49WrlypZ555Rr/4xS80evRo3XjjjVq2bJmX9QEAAJ9KxTPXOzZeaWlp2rlz\npzo7O7Vz507t2LFDJ510kuwUPPQTAADABMddjd///vf18MMP64MPPtDYsWNVVlYmSbr+erMXZQUA\nAKkpFc9c79h4jR8/Xnfcccch2ydOnOhqQQAAAH7V79NJRCIRN+oAAAApJi7b1Vsycpx4zZgxQ6NG\njZJt27IsS5Jk27YaGho8Kw4AAMBPHBuvnJwcVVZWKjs7+6DtRUVFrhcFAAD8L1mPPHST467GmTNn\nKiPj4L6sra1N+fn5rhcFAADgR46Nl2VZWrhwoWpra9Xe3q6KigqVlpYqLy/Py/oAAIBPxV2+JSPH\nXY0lJSUqLCxUcXGxGhsbFQqFVF1d7WVtAADAx1Lx3KCOjVcsFtOmTZsUDAY1ffp0VVRUKBKJqLS0\nVBMmTDhicFZml9FCe+wss3lG09xxNFxI0zKcl244r9ky+zqUpJPtIUbzLOPfRemYeKfRvB7DNb78\n1xON5knSlPHvG80bssPsZ05Xl+lXt9S2L9No3l8mf9toniTlv3qv0bwXzlxsNK8r3u+D+49omGX6\nN4z51w4Sx/F3eyAQUFVVVe86r7KyMjU1NSkajaq8vNyzAgEAgD8l6ykf3OTYeBUUFByyLTc3l6YL\nAABggI6GvVkAAMCHknUBvJvM79wGAADAYTHxAgAACcEJVAEAAOAaJl4AACAhUvGoRiZeAAAAHmHi\nBQAAEoIz13/Kq6++qtWrV6ujo0O2bcuyLGVlZem6667TlClTvKwRAADAFxwbrzvuuEMVFRUaOXKk\nNmzYoKuuukq7du3Sddddp8cff9zLGgEAgA9xHq9PGTZsmP76179q9+7dOvPMM7V7927V1dUpK8vs\n9csAAABShePEa+XKlXriiSf01FNPae/evcrJydGkSZO0cuVKL+sDAAA+lYrn8XJsvEaOHKlQKORl\nLQAAAL7muKvxueeeUzAYVFFRkWpqanq3L1q0yJPCAACAv8Vlu3pLRo6N1+rVq7VmzRpVVlaqublZ\nixcvVkdHh/bs2eNlfQAAAL7huKuxq6urdyH9ggULVFdXp3A4rB07dnhWHAAA8K9UPI+X48Rr3rx5\namlp6f16ypQpWr58ubZt2+ZJYQAAAH7jOPFqbm7WkiVLVFRUpGnTpmndunV68cUXOaoRAAAYkazr\nsNzk2HiVlJSosLBQxcXFamxsVCgU0tq1a/sc3N1j9jKQ6Sl4Vcluw3kZsgwnSgHDl/s0/SY0XZ8k\nfWD1GM07KZ5uNE+ShlhmT0to22ZrDLjwYftq4yijeZPHv280r2NPptE8SXqnPdtoXpdl/v3ywpmL\njeads+Vuo3mm65OkLjsFf2GhzxxfHbFYTOFwWMFgUDU1NWppaVEkElF9fb2X9QEAAJ+yXf5fMnKc\neAUCAVVVVSkj48BdysrK1NTUpGg0qvLycs8KBAAA/hRPwcX1jo1XQUHBIdtyc3NpugAAAAbIsfEC\nAABwU+rNuz5jjRcAAADMYuIFAAASIhVPJ8HECwAAwCNMvAAAQEIw8QIAAIBr+t14LVy40I06AABA\nirFt29VbMnLc1ThnzpxDttm2rYaGBlcLAgAA8CvHxmvPnj36yU9+oszMg68vVlRU5HpRAADA/1jj\n9SkrVqxQV1fXIduXLVvmakEAAAB+5Tjxmjx58mG3jx8/3rViAABA6kiWC1nv379fN910kz788EMt\nWLBAF1100WHvd9ddd6murk6ZmZm64447dPLJJ2vu3LmybVuWZWnZsmWaMGHCZz6W48TrueeeUzAY\nVFFRkWpqanq3L1q0aIBPCwAAIPls3rxZ06ZNUywW09q1ax3vN3fuXD3++OO6/vrrtW7dut7ta9eu\nVSwWO2LTJX1G47V69WqtWbNGlZWVam5u1uLFi9XR0aE9e/b079kAAAAcRrIc1bhlyxbl5+crMzNT\nn/vc59Ta2nrY+5100kmSJMuyFAgEev///PnztWTJEu3bt++Ij+XYeHV1dSkrK0tDhgzRggULVFhY\nqHA4rB07dvT5iQAAACS7vXv3qqOjQ5WVlcrOznZsvD5WXV2tgoICSVI0GtWjjz6qvLw8bdy48YiP\n5dh4zZs3Ty0tLb1fT5kyRcuXL9e2bdv6+jwAAAAcxWW7enNSWVmpuXPn9t42b96srKwshUIhtbe3\na/jw4Y7/7SOPPKJzzjlHp5xyiiRpxIgRkqR///d/19atW4/4nB0X1zc3N2vJkiUqKirStGnTtG7d\nOr344otauXLlEUMBAACSVSgUUigU6v365z//uV555RWdfvrp2r17d2/j1draqo6ODh1//PGSpJdf\nfllbt27VnXfe2fvftra2avjw4Xrttdc0ZsyYIz62Y+NVUlKiwsJCFRcXq7GxUaFQ6DMXnAEAAPRH\nspxd/uKLL9ZNN92kX/ziF7rmmmt6t2/evFkvv/yy7rrrLknSPffcI+nAIvtp06bpm9/8pubOnath\nw4bpc5/7nO69994jPpZlOzzrWCymTZs26YorrtD06dNVUVGh3bt3q7S0tE+r9l/NvaxPT7avfmmP\nMJq3x4obzcuxLaN5kjTMcOY7ad1G8yQpkOSX+0zu6g7ocuFw6v+3o8doXo/MvhaHWWbrk6T9ttmf\ndqfhV89Zp71vNE+S/rH1OKN5wzMOPXfjYHXFzX4fuw3/nM/ZcrfRPEl6ftJio3k/z0o3midJK5se\nN545EF884TxX819ped7V/IFwnHgFAgFVVVUpI+PAXcrKytTU1KRoNKry8nLPCgQAAP6Uimeud2y8\nPl6t/2m5ubk0XQAAwIhkOYGql46GPTEAAAC+4DjxAgAAcFM8SRbXe4mJFwAAgEeYeAEAgIRgjRcA\nAABcw8QLAAAkBGu8AAAA4BrHxusf//iHwuGwbrrpJtXV1fVu/973vudJYQAAwN9sl/+XjBx3Na5Y\nsUL33nuvLMtSZWWlfvOb3+iGG27Qtm3bvKwPAADANxwbr87OTp1wwgmSpO9+97vatGmTFi1apD17\n9nhWHAAA8C/WeH3KJZdcou3bt/d+/fGVu8eNG+dJYQAAAH7jOPGaP3/+IdvGjx+v++67z816AABA\nikjWdVhucpx4PffccwoGgyoqKlJNTU3v9kWLFnlSGAAAgN84Nl6rV6/WmjVrVFlZqebmZi1evFgd\nHR2s8QIAAEbEbdvVWzJy3NXY1dWlrKwsSdKCBQtUV1encDisHTt2eFYcAACAnzhOvObNm6eWlpbe\nr6dMmaLly5dzOgkAAGBEKp7Hy7Hxam5u1pIlS1RbW6v29nZVVFTolltu0cqVK72sDwAAwDccdzWW\nlJSosLBQxcXFamxsVCgU0tq1az0s7WDdSdq5fix+lGSaluzXnEqXZTwzzfBLMd0yX+PQtE6jebZt\ntkbLMv9+Dhh+x6QZXh/yUsOJRvMk6YT0fUbz/tkdMJonScOsHqN5XbbZT53nJy02midJ5712t9G8\nX5y91GheMrHto+E3nVmOr+BYLKZwOKxgMKiamhq1tLQoEomovr7ey/oAAAB8w3HiFQgEVFVVpYyM\nA3cpKytTU1OTotGoysvLPSsQAAD4UzzJ92a5wbHxKigoOGRbbm4uTRcAADDCTtJTPrgp2ZfoAAAA\n+IbjxAsAAMBNqbirkYkXAACAR5h4AQCAhGCNFwAAAFzDxAsAACREsl7I2k2OjVddXZ0qKio0btw4\nzZgxQ+Xl5UpPT1dJSYnOO+88L2sEAADwBcfG6+6779aqVau0c+dOhUIhbdiwQcOGDdM111xD4wUA\nAAYtWS9k7SbHxistLU3HHXecjjvuOM2cOVPHH3+8JGnIkCGeFQcAAOAnjo3XjBkz1NnZqczMTN16\n662SpH379un000/3rDgAAOBfqXhUo2PjVVRUdMi2oUOH6rbbbnO1IAAAAL9yPJ3Ec889p2AwqKKi\nItXU1PRuX7RokSeFAQAAf4vLdvWWjBwbr9WrV2vNmjWqrKxUc3OzFi9erI6ODu3Zs8fL+gAAAHzD\ncVdjV1eXsrKyJEkLFixQXV2dwuGwduzY4VlxAADAv1JxjZfjxGvevHlqaWnp/XrKlClavny5tm3b\n5klhAAAAfuM48WpubtaSJUtUVFSkadOmad26dXrxxRe1cuVKL+sDAAA+lYpnrrfsz5jz7d69W8XF\nxWpsbFQoFFJpaWmfg7825jIjBX5sYtpwo3mje8xepvL99LjRPEn6f+Jma5xz2jtG8ySpbUem0bz9\nHWbPE9fVlW40T5J64pbRvP1d5q/c9fSQLKN5pr+LbnzUmv2pHB0Xsn3P6jKaN/SoeNbJL83wq/HO\nP68wmidJQ47NM545EMcMP83V/J2tW13NHwjHd1ksFlM4HFYwGFRNTY1aWloUiURUX1/vZX0AAMCn\nbNt29ZaMHP/UDgQCqqqqUkbGgbuUlZWpqalJ0WhU5eXlnhUIAADgF46NV0FBwSHbcnNzaboAAIAR\nyXquLTeZX1wCAADQB8m6O9BNrKQEAADwCBMvAACQEKl4OgkmXgAAAB5h4gUAABLCTsHF9Y4Trw0b\nNqi7u9vLWgAAAHzNsfFqa2vTvHnz9Nhjj6mzs9PLmgAAQAqI27art2TkuKtx/vz5mjt3rh566CFd\ncMEFGjt2rGzblmVZeuyxx7ysEQAAwBccG6+///3vWrt2rdLT0/XII4/o9NNP97IuAADgc6l4Hi/H\nxmv9+vW6/vrrlZub62E5AAAA/uXYeN1xxx1e1gEAAFIMRzUCAADANZzHCwAAJEQqrvFi4gUAAOAR\nJl4AACAhmHgBAADANUy8AABAQqTevEuy7FSc8wEAACQAuxoBAAA8QuMFAADgERovAAAAj9B4AQAA\neITGCwAAwCOenk7ixz/+sfbv36958+bpBz/4gWzb1jPPPKPW1lb95je/MZI5evRorVu3TkOGDFFp\naam+/OUvDyrvxBNP1OOPP654PK5QKKQZM2YMur6vfvWrWrp0qY499ljdcMMN/co7XKZt2/rZz36m\n448/Xueff74WLFgwqLwTTzxRf/vb39TQ0KDJkyfr5ptvHnSN8XhcL774oiTpH//4h5599lmNHDly\nUDVu3LhR3d3duvTSSzV37txB1Td69Gg9+eST2rdvn77xjW/okksuGVDO4V7Tu3bt0o033qj29nbd\ncsstmjJlyqAz//a3v2np0qUaPny41q9fP+i8qqoq1dTUaMiQIVq6dKkmTZo0qLxHHnlEzz77rOLx\nuL7zne9o6tSpg67xY/Pnz9ell16qq666alB50WhUmzdv1ogRI/T1r39dl19++aBr3Ldvn2699VZ9\n+OGHuvDCCzV//vxB5a1atUp/+tOfJEkNDQ164YUXBpX35ptv6rbbblM8Htc111yjmTNnDvo5NzY2\natmyZerp6dF3vvMdnX322X3OO9zndX/eL33J6897pa+Zg3m/HC6vv++Xw7nllltUWFioyZMnq6qq\nStnZ2SooKOh3Djxge6izs9OeM2eOvWvXLnvu3Ll2W1ub3draas+aNctY5nvvvWfH43G7ra3Nnj17\n9qDz2tvbbdu27X379tlXX331oPN6enrs7du32/PmzbNXrlzZ77zDZW7YsMF+4oknBpR1uLzf/va3\ndjQaHXDe4TJ7enps27btjo4Ou7i4eNB5Dz/8cO9z/vrXvz7ovGg0av/617+29+3bZ19xxRUDzjnc\na7qiosLevHmz/c9//tO+9tprjWS2trba+/bt69N7py957777rm3btv3ee+/ZpaWlg87bt2+fbdu2\n3dLSYodCISM12rZt19XV2bNnzz7i670veatWrbKff/75I9bWn8xHH33UfvLJJ43lfezdd9+1b7rp\npkHnlZWV2c8//7y9f/9+Y6+dJUuW2G+88Yb94Ycf2vPmzetX3uE+r/vzfulLXn/eK33NHMz75XB5\n/X2/HM7//u//2jfccIPd1dVlz5o1y96/f/+AcuA+T3c1DhkyRFdccYWKi4tVVFSkYcOGKScnx2jm\n6NGjZVmWMjIylJHR/4Hev+ZlZWVJklpbW3XyyScPOi8tLU2xWGxQf4kcLnPdunUKBoPasmXLoPP+\n/Oc/a+fOnZozZ45++tOfGqtRkn7/+9/rnHPOGXTe+PHjJUldXV0aOnTooPM++OADjR07VoFAQJZl\nac+ePQPKOdxresuWLcrPz9cxxxyjjo4OI5k5OTkKBALGajzppJMkSWlpaUfM7UvexxkfffSRcnNz\njdQoST/5yU902WWXGcu79957dc011+idd94xkvnnP/9Zb775poLBoH7/+98bqVGSNm3apAsvvHDQ\neaeeeqp6enq0b9++Pn329iXz4/fOscceq+3bt/cr73Cf1/15v/Qlrz/vlb5mDub9cri8/r5fDmfc\nuHFKS0vTQw89pEsvvVSZmZkDyoH7PF/jNX36dDU0NGjatGmuZlZXV3/mroj+5FVWVuryyy8/4gdf\nX/J27typjo4OHXvssQPKOlzmRRddpJ///Oe68847deeddw46b/fu3crJyVF1dbUef/xxdXZ2Djrz\nY88884wuuuiiQedNmjRJP/7xj/WVr3xF//Ef/zHovDFjxuj111/X7t279dZbb2nv3r0DyjmcvXv3\nqtu2bNMAAAShSURBVLm5WRs2bOjzdclMv0/6mldZWalgMGgkb8WKFVqwYEGff95Hyty6datGjRrV\n518oR8qbN2+ennzySYXDYd1zzz1GMnfv3q3c3Fz98Ic/1IMPPjjovI/97ne/69OyiSPlfelLX9Ld\nd9+tyy+/vE8/575kfvzeeeutt9Tc3DygvE9/Xvf3/WL6878/mYN5v/xrXn/fL4dz7bXXKhaLDeq5\nw32eN14//OEPFQqFFIvFXMt85ZVX9Oabbw74F/K/5oVCIf3qV7/SD37wg0HnxWIxXX311QPKccoc\nPny4LMvSKaecMuALjn46Lzs7W2eddZYyMjJ08skna+fOnYPOlKTOzk699957Gjt27KDz1q5dq9LS\nUj3zzDN6+umn1d7ePqi8goICPfPMM1q6dKkmTpzYr/VnR3pN5+TkaPTo0brqqqtkWZaRzP7qS94z\nzzyjYcOG9Wl9SV/yli5dqo0bN6q8vNxIjdXV1SosLOxTVl/yRowYIUn64he/qB07dhjJ/Pi9k52d\n3aeJe1++j++//76ysrKUnZ096LxVq1apvLxcNTU1qqysPGJeXzKvvfZarV69Wg8++KC++MUv9jvv\nXz+v+/t+Mf3539fMwbxfDpfX3/fL4UyYMEFjxozp14QP3vO08WpqatJHH32kcDisF154oV9Thb5m\nNjc3KxqN6vbbbzda49ChQwc0uv3XvK1bt+q////27t8ltTCMA/iDgkFNDS3+ARauDv4HbU354xRH\nziLoEkEQhBC0hTnqquEfEDXk5tSQOohylhpUgggXRSEHQeF7h4sOF7n3nN6XAxe+n0Vw+PIcPM/L\ng+979O5OCoWC1Go1sW1bOfP7+1tEZP2qmndwcCD9fl9Efi/6u7u7ypmz2UxeX18lGo1qqXGxWMjO\nzo74/X7x+/0yn8+V8nw+n5RKJbm9vZVAICDb29s/ytl0T4fDYel0OjKZTBxti+ruEyd5/X5fnp6e\nHD3s4SRvtUW0+ox0ZA6HQ7m8vJT7+3upVqvy9fWllLfql8/Pz/UQplrj/v6+DAYDWS6XslwulfNE\nROr1uqNv253krfpma2vL0ba3k8xgMCjlclnOzs5kb2/PVd6m9dpNv+he/51mqvTLpjy3/UL/OS8P\nlF1cXODt7Q0A8Pz8jGKxCMuyEIlEYFkWptOpcmYoFMLh4SFM00Q2m9VS48nJCY6Pj/H4+KglDwCa\nzeaPD9dvuuZ4PI54PI5Go6Glxkwmg2QyiUqloqXGYrGIq6sr2LatLc8wDBiGgXw+ryUvlUrBMAx0\nu12lnD/v6fF4DMuykEgk0G63tWT2er31e+l0Wjnv/PwcR0dHME0T19fXynk3Nzc4PT1FLBbDy8uL\nlmteeXh4+Ofheid5uVwOyWQShmHg/f1dS42j0QipVAqJRAK1Wk3LNVuWhdFopKW+TqezXiuq1aqW\nzFarBdM0YZomPj4+XOVtWq/d9IuTPDe94jRTpV825bntl79ReViNvME/ySYiIiLyCH9AlYiIiMgj\nHLyIiIiIPMLBi4iIiMgjHLyIiIiIPMLBi4iIiMgjHLyIiIiIPMLBi4iIiMgjHLyIiIiIPPILH4Vl\nfRaKXdAAAAAASUVORK5CYII=\n", "text/plain": ["<matplotlib.figure.Figure at 0x2a50644c048>"]}, "metadata": {}, "output_type": "display_data"}], "source": ["# Matrice des corr\u00e9lations\n", "\n", "sns.set(context=\"paper\", font=\"monospace\")\n", "corrmat = df1.corr()\n", "\n", "# atplotlib figure\n", "f, ax = plt.subplots(figsize=(12, 9))\n", "\n", "# Draw the heatmap using seaborn\n", "sns.heatmap(corrmat, vmax=.8, square=True)"]}, {"cell_type": "code", "execution_count": 10, "metadata": {}, "outputs": [{"data": {"text/html": ["<div>\n", "<style>\n", "    .dataframe thead tr:only-child th {\n", "        text-align: right;\n", "    }\n", "\n", "    .dataframe thead th {\n", "        text-align: left;\n", "    }\n", "\n", "    .dataframe tbody tr th {\n", "        vertical-align: top;\n", "    }\n", "</style>\n", "<table border=\"1\" class=\"dataframe\">\n", "  <thead>\n", "    <tr style=\"text-align: right;\">\n", "      <th></th>\n", "      <th>X1</th>\n", "      <th>X2</th>\n", "      <th>X3</th>\n", "      <th>X4</th>\n", "      <th>X5</th>\n", "      <th>X6</th>\n", "      <th>X7</th>\n", "      <th>X8</th>\n", "      <th>X9</th>\n", "      <th>X10</th>\n", "      <th>...</th>\n", "      <th>X22</th>\n", "      <th>X23</th>\n", "      <th>Y</th>\n", "      <th>TotalDelay</th>\n", "      <th>TotalPayment</th>\n", "      <th>PartMay</th>\n", "      <th>PartJune</th>\n", "      <th>PartJuly</th>\n", "      <th>PartAugust</th>\n", "      <th>PartSeptember</th>\n", "    </tr>\n", "  </thead>\n", "  <tbody>\n", "    <tr>\n", "      <th>0</th>\n", "      <td>180000</td>\n", "      <td>1</td>\n", "      <td>2</td>\n", "      <td>1</td>\n", "      <td>47</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>...</td>\n", "      <td>2500</td>\n", "      <td>2618</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>71798</td>\n", "      <td>0.962902</td>\n", "      <td>0.964215</td>\n", "      <td>0.958865</td>\n", "      <td>0.961978</td>\n", "      <td>0.961112</td>\n", "    </tr>\n", "    <tr>\n", "      <th>1</th>\n", "      <td>110000</td>\n", "      <td>2</td>\n", "      <td>2</td>\n", "      <td>1</td>\n", "      <td>35</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>...</td>\n", "      <td>184</td>\n", "      <td>185</td>\n", "      <td>1</td>\n", "      <td>0</td>\n", "      <td>17094</td>\n", "      <td>0.963518</td>\n", "      <td>0.963962</td>\n", "      <td>0.777823</td>\n", "      <td>0.721906</td>\n", "      <td>0.850305</td>\n", "    </tr>\n", "    <tr>\n", "      <th>2</th>\n", "      <td>70000</td>\n", "      <td>2</td>\n", "      <td>2</td>\n", "      <td>2</td>\n", "      <td>22</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>...</td>\n", "      <td>1792</td>\n", "      <td>1793</td>\n", "      <td>1</td>\n", "      <td>0</td>\n", "      <td>51151</td>\n", "      <td>0.963766</td>\n", "      <td>0.964848</td>\n", "      <td>0.962690</td>\n", "      <td>0.956517</td>\n", "      <td>0.962942</td>\n", "    </tr>\n", "    <tr>\n", "      <th>3</th>\n", "      <td>200000</td>\n", "      <td>2</td>\n", "      <td>1</td>\n", "      <td>2</td>\n", "      <td>27</td>\n", "      <td>-2</td>\n", "      <td>-2</td>\n", "      <td>-2</td>\n", "      <td>-2</td>\n", "      <td>-2</td>\n", "      <td>...</td>\n", "      <td>0</td>\n", "      <td>95456</td>\n", "      <td>0</td>\n", "      <td>-126</td>\n", "      <td>224043</td>\n", "      <td>1.028571</td>\n", "      <td>-0.004450</td>\n", "      <td>-0.004802</td>\n", "      <td>-0.004502</td>\n", "      <td>-0.009899</td>\n", "    </tr>\n", "    <tr>\n", "      <th>4</th>\n", "      <td>370000</td>\n", "      <td>2</td>\n", "      <td>1</td>\n", "      <td>1</td>\n", "      <td>39</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>...</td>\n", "      <td>3000</td>\n", "      <td>1000</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>42614</td>\n", "      <td>0.937577</td>\n", "      <td>0.958029</td>\n", "      <td>0.965386</td>\n", "      <td>0.959245</td>\n", "      <td>0.968143</td>\n", "    </tr>\n", "    <tr>\n", "      <th>5</th>\n", "      <td>260000</td>\n", "      <td>2</td>\n", "      <td>1</td>\n", "      <td>1</td>\n", "      <td>29</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>-2</td>\n", "      <td>-2</td>\n", "      <td>...</td>\n", "      <td>0</td>\n", "      <td>141516</td>\n", "      <td>0</td>\n", "      <td>-14</td>\n", "      <td>160056</td>\n", "      <td>0.000000</td>\n", "      <td>0.000000</td>\n", "      <td>0.000000</td>\n", "      <td>0.000000</td>\n", "      <td>0.942813</td>\n", "    </tr>\n", "    <tr>\n", "      <th>6</th>\n", "      <td>90000</td>\n", "      <td>2</td>\n", "      <td>1</td>\n", "      <td>1</td>\n", "      <td>43</td>\n", "      <td>-1</td>\n", "      <td>-1</td>\n", "      <td>2</td>\n", "      <td>-1</td>\n", "      <td>-1</td>\n", "      <td>...</td>\n", "      <td>4009</td>\n", "      <td>7452</td>\n", "      <td>0</td>\n", "      <td>-39</td>\n", "      <td>135882</td>\n", "      <td>0.000000</td>\n", "      <td>0.000000</td>\n", "      <td>0.000000</td>\n", "      <td>0.997711</td>\n", "      <td>0.393333</td>\n", "    </tr>\n", "    <tr>\n", "      <th>7</th>\n", "      <td>220000</td>\n", "      <td>2</td>\n", "      <td>1</td>\n", "      <td>1</td>\n", "      <td>43</td>\n", "      <td>-1</td>\n", "      <td>3</td>\n", "      <td>2</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>...</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>1</td>\n", "      <td>32</td>\n", "      <td>1002</td>\n", "      <td>0.000000</td>\n", "      <td>0.999083</td>\n", "      <td>0.999083</td>\n", "      <td>0.999083</td>\n", "      <td>0.866455</td>\n", "    </tr>\n", "    <tr>\n", "      <th>8</th>\n", "      <td>50000</td>\n", "      <td>1</td>\n", "      <td>2</td>\n", "      <td>1</td>\n", "      <td>35</td>\n", "      <td>1</td>\n", "      <td>2</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>...</td>\n", "      <td>29935</td>\n", "      <td>1200</td>\n", "      <td>1</td>\n", "      <td>63</td>\n", "      <td>78530</td>\n", "      <td>-0.012172</td>\n", "      <td>0.985915</td>\n", "      <td>0.915291</td>\n", "      <td>0.956269</td>\n", "      <td>0.999979</td>\n", "    </tr>\n", "    <tr>\n", "      <th>9</th>\n", "      <td>50000</td>\n", "      <td>2</td>\n", "      <td>3</td>\n", "      <td>2</td>\n", "      <td>40</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>...</td>\n", "      <td>325</td>\n", "      <td>436</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>19283</td>\n", "      <td>0.962316</td>\n", "      <td>0.963619</td>\n", "      <td>0.879296</td>\n", "      <td>0.847636</td>\n", "      <td>0.806216</td>\n", "    </tr>\n", "    <tr>\n", "      <th>10</th>\n", "      <td>130000</td>\n", "      <td>1</td>\n", "      <td>2</td>\n", "      <td>1</td>\n", "      <td>24</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>2</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>...</td>\n", "      <td>0</td>\n", "      <td>780</td>\n", "      <td>0</td>\n", "      <td>16</td>\n", "      <td>38480</td>\n", "      <td>0.997442</td>\n", "      <td>0.999978</td>\n", "      <td>0.959249</td>\n", "      <td>0.999979</td>\n", "      <td>0.899013</td>\n", "    </tr>\n", "    <tr>\n", "      <th>11</th>\n", "      <td>200000</td>\n", "      <td>2</td>\n", "      <td>1</td>\n", "      <td>2</td>\n", "      <td>25</td>\n", "      <td>-1</td>\n", "      <td>-1</td>\n", "      <td>-1</td>\n", "      <td>-1</td>\n", "      <td>-1</td>\n", "      <td>...</td>\n", "      <td>4970</td>\n", "      <td>8888</td>\n", "      <td>0</td>\n", "      <td>-63</td>\n", "      <td>82748</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>12</th>\n", "      <td>230000</td>\n", "      <td>2</td>\n", "      <td>2</td>\n", "      <td>1</td>\n", "      <td>38</td>\n", "      <td>-2</td>\n", "      <td>-2</td>\n", "      <td>-2</td>\n", "      <td>-2</td>\n", "      <td>-2</td>\n", "      <td>...</td>\n", "      <td>2132</td>\n", "      <td>2204</td>\n", "      <td>0</td>\n", "      <td>-126</td>\n", "      <td>111453</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>13</th>\n", "      <td>90000</td>\n", "      <td>2</td>\n", "      <td>1</td>\n", "      <td>2</td>\n", "      <td>29</td>\n", "      <td>-2</td>\n", "      <td>-2</td>\n", "      <td>-2</td>\n", "      <td>-2</td>\n", "      <td>-2</td>\n", "      <td>...</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>-126</td>\n", "      <td>0</td>\n", "      <td>1.004184</td>\n", "      <td>1.004184</td>\n", "      <td>1.004184</td>\n", "      <td>1.004184</td>\n", "      <td>1.004184</td>\n", "    </tr>\n", "    <tr>\n", "      <th>14</th>\n", "      <td>230000</td>\n", "      <td>1</td>\n", "      <td>3</td>\n", "      <td>2</td>\n", "      <td>37</td>\n", "      <td>-1</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>-1</td>\n", "      <td>...</td>\n", "      <td>5003</td>\n", "      <td>3016</td>\n", "      <td>0</td>\n", "      <td>-34</td>\n", "      <td>287273</td>\n", "      <td>0.887873</td>\n", "      <td>-0.000340</td>\n", "      <td>0.769071</td>\n", "      <td>0.760863</td>\n", "      <td>0.952064</td>\n", "    </tr>\n", "    <tr>\n", "      <th>15</th>\n", "      <td>130000</td>\n", "      <td>1</td>\n", "      <td>2</td>\n", "      <td>2</td>\n", "      <td>33</td>\n", "      <td>2</td>\n", "      <td>2</td>\n", "      <td>-1</td>\n", "      <td>-1</td>\n", "      <td>-2</td>\n", "      <td>...</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>78</td>\n", "      <td>3578</td>\n", "      <td>0.000000</td>\n", "      <td>0.000000</td>\n", "      <td>0.000000</td>\n", "      <td>0.000000</td>\n", "      <td>0.993003</td>\n", "    </tr>\n", "    <tr>\n", "      <th>16</th>\n", "      <td>90000</td>\n", "      <td>2</td>\n", "      <td>2</td>\n", "      <td>1</td>\n", "      <td>35</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>...</td>\n", "      <td>4000</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>2</td>\n", "      <td>91108</td>\n", "      <td>0.954684</td>\n", "      <td>0.883510</td>\n", "      <td>0.952754</td>\n", "      <td>0.953646</td>\n", "      <td>0.952596</td>\n", "    </tr>\n", "    <tr>\n", "      <th>17</th>\n", "      <td>10000</td>\n", "      <td>2</td>\n", "      <td>2</td>\n", "      <td>1</td>\n", "      <td>37</td>\n", "      <td>-1</td>\n", "      <td>4</td>\n", "      <td>3</td>\n", "      <td>2</td>\n", "      <td>2</td>\n", "      <td>...</td>\n", "      <td>0</td>\n", "      <td>36</td>\n", "      <td>0</td>\n", "      <td>70</td>\n", "      <td>3236</td>\n", "      <td>0.999550</td>\n", "      <td>0.848221</td>\n", "      <td>0.800478</td>\n", "      <td>0.999590</td>\n", "      <td>0.999652</td>\n", "    </tr>\n", "    <tr>\n", "      <th>18</th>\n", "      <td>80000</td>\n", "      <td>1</td>\n", "      <td>3</td>\n", "      <td>1</td>\n", "      <td>36</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>...</td>\n", "      <td>3000</td>\n", "      <td>6200</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>73411</td>\n", "      <td>0.961214</td>\n", "      <td>0.926935</td>\n", "      <td>0.962670</td>\n", "      <td>0.963936</td>\n", "      <td>0.962893</td>\n", "    </tr>\n", "    <tr>\n", "      <th>19</th>\n", "      <td>320000</td>\n", "      <td>2</td>\n", "      <td>1</td>\n", "      <td>1</td>\n", "      <td>36</td>\n", "      <td>-1</td>\n", "      <td>2</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>...</td>\n", "      <td>5000</td>\n", "      <td>11906</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>96906</td>\n", "      <td>0.792507</td>\n", "      <td>0.755381</td>\n", "      <td>0.703680</td>\n", "      <td>0.400943</td>\n", "      <td>0.999851</td>\n", "    </tr>\n", "  </tbody>\n", "</table>\n", "<p>20 rows \u00d7 31 columns</p>\n", "</div>"], "text/plain": ["        X1  X2  X3  X4  X5  X6  X7  X8  X9  X10      ...          X22     X23  \\\n", "0   180000   1   2   1  47   0   0   0   0    0      ...         2500    2618   \n", "1   110000   2   2   1  35   0   0   0   0    0      ...          184     185   \n", "2    70000   2   2   2  22   0   0   0   0    0      ...         1792    1793   \n", "3   200000   2   1   2  27  -2  -2  -2  -2   -2      ...            0   95456   \n", "4   370000   2   1   1  39   0   0   0   0    0      ...         3000    1000   \n", "5   260000   2   1   1  29   0   0   0  -2   -2      ...            0  141516   \n", "6    90000   2   1   1  43  -1  -1   2  -1   -1      ...         4009    7452   \n", "7   220000   2   1   1  43  -1   3   2   0    0      ...            0       0   \n", "8    50000   1   2   1  35   1   2   0   0    0      ...        29935    1200   \n", "9    50000   2   3   2  40   0   0   0   0    0      ...          325     436   \n", "10  130000   1   2   1  24   0   0   2   0    0      ...            0     780   \n", "11  200000   2   1   2  25  -1  -1  -1  -1   -1      ...         4970    8888   \n", "12  230000   2   2   1  38  -2  -2  -2  -2   -2      ...         2132    2204   \n", "13   90000   2   1   2  29  -2  -2  -2  -2   -2      ...            0       0   \n", "14  230000   1   3   2  37  -1   0   0   0   -1      ...         5003    3016   \n", "15  130000   1   2   2  33   2   2  -1  -1   -2      ...            0       0   \n", "16   90000   2   2   1  35   0   0   0   0    0      ...         4000       0   \n", "17   10000   2   2   1  37  -1   4   3   2    2      ...            0      36   \n", "18   80000   1   3   1  36   0   0   0   0    0      ...         3000    6200   \n", "19  320000   2   1   1  36  -1   2   0   0    0      ...         5000   11906   \n", "\n", "    Y  TotalDelay  TotalPayment   PartMay  PartJune  PartJuly  PartAugust  \\\n", "0   0           0         71798  0.962902  0.964215  0.958865    0.961978   \n", "1   1           0         17094  0.963518  0.963962  0.777823    0.721906   \n", "2   1           0         51151  0.963766  0.964848  0.962690    0.956517   \n", "3   0        -126        224043  1.028571 -0.004450 -0.004802   -0.004502   \n", "4   0           0         42614  0.937577  0.958029  0.965386    0.959245   \n", "5   0         -14        160056  0.000000  0.000000  0.000000    0.000000   \n", "6   0         -39        135882  0.000000  0.000000  0.000000    0.997711   \n", "7   1          32          1002  0.000000  0.999083  0.999083    0.999083   \n", "8   1          63         78530 -0.012172  0.985915  0.915291    0.956269   \n", "9   0           0         19283  0.962316  0.963619  0.879296    0.847636   \n", "10  0          16         38480  0.997442  0.999978  0.959249    0.999979   \n", "11  0         -63         82748  0.000000  0.000000  0.000000    0.000000   \n", "12  0        -126        111453  0.000000  0.000000  0.000000    0.000000   \n", "13  0        -126             0  1.004184  1.004184  1.004184    1.004184   \n", "14  0         -34        287273  0.887873 -0.000340  0.769071    0.760863   \n", "15  0          78          3578  0.000000  0.000000  0.000000    0.000000   \n", "16  0           2         91108  0.954684  0.883510  0.952754    0.953646   \n", "17  0          70          3236  0.999550  0.848221  0.800478    0.999590   \n", "18  0           0         73411  0.961214  0.926935  0.962670    0.963936   \n", "19  0           0         96906  0.792507  0.755381  0.703680    0.400943   \n", "\n", "    PartSeptember  \n", "0        0.961112  \n", "1        0.850305  \n", "2        0.962942  \n", "3       -0.009899  \n", "4        0.968143  \n", "5        0.942813  \n", "6        0.393333  \n", "7        0.866455  \n", "8        0.999979  \n", "9        0.806216  \n", "10       0.899013  \n", "11       0.000000  \n", "12       0.000000  \n", "13       1.004184  \n", "14       0.952064  \n", "15       0.993003  \n", "16       0.952596  \n", "17       0.999652  \n", "18       0.962893  \n", "19       0.999851  \n", "\n", "[20 rows x 31 columns]"]}, "execution_count": 11, "metadata": {}, "output_type": "execute_result"}], "source": ["# on modifie les colonnes (cr\u00e9ation de variables d'int\u00e9r\u00eat)\n", "\n", "df1['TotalDelay'] = df1.X11 + 2*df1.X10 + 4*df1.X9 + 8*df1.X8 + 16*df1.X7 + 32*df1.X6\n", "df1['TotalPayment'] = df1.X23 + 2*df1.X22 + 3*df1.X21 + 4*df1.X20 + 5*df1.X19 + 6*df1.X18\n", "df1['PartMay'] = -(df1.X22 - df1.X17)/(df1.X17 + 1)\n", "df1['PartJune'] = -(df1.X21 - df1.X16)/(df1.X16 + 1)\n", "df1['PartJuly'] = -(df1.X20 - df1.X15)/(df1.X15 + 1)\n", "df1['PartAugust'] = -(df1.X19 - df1.X14)/(df1.X14 + 1)\n", "df1['PartSeptember'] = -(df1.X18 - df1.X13)/(df1.X13 + 1)\n", "df1.head(20)"]}, {"cell_type": "code", "execution_count": 11, "metadata": {}, "outputs": [{"data": {"text/plain": ["<matplotlib.axes._subplots.AxesSubplot at 0x2a5067c70f0>"]}, "execution_count": 12, "metadata": {}, "output_type": "execute_result"}, {"data": {"image/png": "iVBORw0KGgoAAAANSUhEUgAAApQAAAI8CAYAAABPgDu/AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAAAAAElFTkSuQmCC\n", "text/plain": ["<matplotlib.figure.Figure at 0x2a50644c240>"]}, "metadata": {}, "output_type": "display_data"}], "source": ["# Matrice des corr\u00e9lations\n", "\n", "sns.set(context=\"paper\", font=\"monospace\")\n", "corrmat = df1.corr()\n", "\n", "# matplotlib figure\n", "f, ax = plt.subplots(figsize=(12, 9))\n", "\n", "# Draw the heatmap using seaborn\n", "sns.heatmap(corrmat, vmax=.8, square=True)"]}, {"cell_type": "code", "execution_count": 12, "metadata": {}, "outputs": [{"data": {"text/html": ["<div>\n", "<style>\n", "    .dataframe thead tr:only-child th {\n", "        text-align: right;\n", "    }\n", "\n", "    .dataframe thead th {\n", "        text-align: left;\n", "    }\n", "\n", "    .dataframe tbody tr th {\n", "        vertical-align: top;\n", "    }\n", "</style>\n", "<table border=\"1\" class=\"dataframe\">\n", "  <thead>\n", "    <tr style=\"text-align: right;\">\n", "      <th></th>\n", "      <th>X1</th>\n", "      <th>X2</th>\n", "      <th>X3</th>\n", "      <th>X4</th>\n", "      <th>X5</th>\n", "      <th>X6</th>\n", "      <th>X12</th>\n", "      <th>Y</th>\n", "      <th>TotalDelay</th>\n", "      <th>TotalPayment</th>\n", "      <th>PartMay</th>\n", "      <th>PartJune</th>\n", "      <th>PartJuly</th>\n", "      <th>PartAugust</th>\n", "      <th>PartSeptember</th>\n", "    </tr>\n", "  </thead>\n", "  <tbody>\n", "    <tr>\n", "      <th>0</th>\n", "      <td>180000</td>\n", "      <td>1</td>\n", "      <td>2</td>\n", "      <td>1</td>\n", "      <td>47</td>\n", "      <td>0</td>\n", "      <td>179253</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>71798</td>\n", "      <td>0.962902</td>\n", "      <td>0.964215</td>\n", "      <td>0.958865</td>\n", "      <td>0.961978</td>\n", "      <td>0.961112</td>\n", "    </tr>\n", "    <tr>\n", "      <th>1</th>\n", "      <td>110000</td>\n", "      <td>2</td>\n", "      <td>2</td>\n", "      <td>1</td>\n", "      <td>35</td>\n", "      <td>0</td>\n", "      <td>6137</td>\n", "      <td>1</td>\n", "      <td>0</td>\n", "      <td>17094</td>\n", "      <td>0.963518</td>\n", "      <td>0.963962</td>\n", "      <td>0.777823</td>\n", "      <td>0.721906</td>\n", "      <td>0.850305</td>\n", "    </tr>\n", "    <tr>\n", "      <th>2</th>\n", "      <td>70000</td>\n", "      <td>2</td>\n", "      <td>2</td>\n", "      <td>2</td>\n", "      <td>22</td>\n", "      <td>0</td>\n", "      <td>66505</td>\n", "      <td>1</td>\n", "      <td>0</td>\n", "      <td>51151</td>\n", "      <td>0.963766</td>\n", "      <td>0.964848</td>\n", "      <td>0.962690</td>\n", "      <td>0.956517</td>\n", "      <td>0.962942</td>\n", "    </tr>\n", "    <tr>\n", "      <th>3</th>\n", "      <td>200000</td>\n", "      <td>2</td>\n", "      <td>1</td>\n", "      <td>2</td>\n", "      <td>27</td>\n", "      <td>-2</td>\n", "      <td>4941</td>\n", "      <td>0</td>\n", "      <td>-126</td>\n", "      <td>224043</td>\n", "      <td>1.028571</td>\n", "      <td>-0.004450</td>\n", "      <td>-0.004802</td>\n", "      <td>-0.004502</td>\n", "      <td>-0.009899</td>\n", "    </tr>\n", "    <tr>\n", "      <th>4</th>\n", "      <td>370000</td>\n", "      <td>2</td>\n", "      <td>1</td>\n", "      <td>1</td>\n", "      <td>39</td>\n", "      <td>0</td>\n", "      <td>141552</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>42614</td>\n", "      <td>0.937577</td>\n", "      <td>0.958029</td>\n", "      <td>0.965386</td>\n", "      <td>0.959245</td>\n", "      <td>0.968143</td>\n", "    </tr>\n", "    <tr>\n", "      <th>5</th>\n", "      <td>260000</td>\n", "      <td>2</td>\n", "      <td>1</td>\n", "      <td>1</td>\n", "      <td>29</td>\n", "      <td>0</td>\n", "      <td>71864</td>\n", "      <td>0</td>\n", "      <td>-14</td>\n", "      <td>160056</td>\n", "      <td>0.000000</td>\n", "      <td>0.000000</td>\n", "      <td>0.000000</td>\n", "      <td>0.000000</td>\n", "      <td>0.942813</td>\n", "    </tr>\n", "    <tr>\n", "      <th>6</th>\n", "      <td>90000</td>\n", "      <td>2</td>\n", "      <td>1</td>\n", "      <td>1</td>\n", "      <td>43</td>\n", "      <td>-1</td>\n", "      <td>16139</td>\n", "      <td>0</td>\n", "      <td>-39</td>\n", "      <td>135882</td>\n", "      <td>0.000000</td>\n", "      <td>0.000000</td>\n", "      <td>0.000000</td>\n", "      <td>0.997711</td>\n", "      <td>0.393333</td>\n", "    </tr>\n", "    <tr>\n", "      <th>7</th>\n", "      <td>220000</td>\n", "      <td>2</td>\n", "      <td>1</td>\n", "      <td>1</td>\n", "      <td>43</td>\n", "      <td>-1</td>\n", "      <td>1090</td>\n", "      <td>1</td>\n", "      <td>32</td>\n", "      <td>1002</td>\n", "      <td>0.000000</td>\n", "      <td>0.999083</td>\n", "      <td>0.999083</td>\n", "      <td>0.999083</td>\n", "      <td>0.866455</td>\n", "    </tr>\n", "    <tr>\n", "      <th>8</th>\n", "      <td>50000</td>\n", "      <td>1</td>\n", "      <td>2</td>\n", "      <td>1</td>\n", "      <td>35</td>\n", "      <td>1</td>\n", "      <td>48047</td>\n", "      <td>1</td>\n", "      <td>63</td>\n", "      <td>78530</td>\n", "      <td>-0.012172</td>\n", "      <td>0.985915</td>\n", "      <td>0.915291</td>\n", "      <td>0.956269</td>\n", "      <td>0.999979</td>\n", "    </tr>\n", "    <tr>\n", "      <th>9</th>\n", "      <td>50000</td>\n", "      <td>2</td>\n", "      <td>3</td>\n", "      <td>2</td>\n", "      <td>40</td>\n", "      <td>0</td>\n", "      <td>5538</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>19283</td>\n", "      <td>0.962316</td>\n", "      <td>0.963619</td>\n", "      <td>0.879296</td>\n", "      <td>0.847636</td>\n", "      <td>0.806216</td>\n", "    </tr>\n", "    <tr>\n", "      <th>10</th>\n", "      <td>130000</td>\n", "      <td>1</td>\n", "      <td>2</td>\n", "      <td>1</td>\n", "      <td>24</td>\n", "      <td>0</td>\n", "      <td>46113</td>\n", "      <td>0</td>\n", "      <td>16</td>\n", "      <td>38480</td>\n", "      <td>0.997442</td>\n", "      <td>0.999978</td>\n", "      <td>0.959249</td>\n", "      <td>0.999979</td>\n", "      <td>0.899013</td>\n", "    </tr>\n", "    <tr>\n", "      <th>11</th>\n", "      <td>200000</td>\n", "      <td>2</td>\n", "      <td>1</td>\n", "      <td>2</td>\n", "      <td>25</td>\n", "      <td>-1</td>\n", "      <td>8926</td>\n", "      <td>0</td>\n", "      <td>-63</td>\n", "      <td>82748</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>12</th>\n", "      <td>230000</td>\n", "      <td>2</td>\n", "      <td>2</td>\n", "      <td>1</td>\n", "      <td>38</td>\n", "      <td>-2</td>\n", "      <td>12696</td>\n", "      <td>0</td>\n", "      <td>-126</td>\n", "      <td>111453</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>13</th>\n", "      <td>90000</td>\n", "      <td>2</td>\n", "      <td>1</td>\n", "      <td>2</td>\n", "      <td>29</td>\n", "      <td>-2</td>\n", "      <td>-240</td>\n", "      <td>0</td>\n", "      <td>-126</td>\n", "      <td>0</td>\n", "      <td>1.004184</td>\n", "      <td>1.004184</td>\n", "      <td>1.004184</td>\n", "      <td>1.004184</td>\n", "      <td>1.004184</td>\n", "    </tr>\n", "    <tr>\n", "      <th>14</th>\n", "      <td>230000</td>\n", "      <td>1</td>\n", "      <td>3</td>\n", "      <td>2</td>\n", "      <td>37</td>\n", "      <td>-1</td>\n", "      <td>36571</td>\n", "      <td>0</td>\n", "      <td>-34</td>\n", "      <td>287273</td>\n", "      <td>0.887873</td>\n", "      <td>-0.000340</td>\n", "      <td>0.769071</td>\n", "      <td>0.760863</td>\n", "      <td>0.952064</td>\n", "    </tr>\n", "    <tr>\n", "      <th>15</th>\n", "      <td>130000</td>\n", "      <td>1</td>\n", "      <td>2</td>\n", "      <td>2</td>\n", "      <td>33</td>\n", "      <td>2</td>\n", "      <td>2183</td>\n", "      <td>0</td>\n", "      <td>78</td>\n", "      <td>3578</td>\n", "      <td>0.000000</td>\n", "      <td>0.000000</td>\n", "      <td>0.000000</td>\n", "      <td>0.000000</td>\n", "      <td>0.993003</td>\n", "    </tr>\n", "    <tr>\n", "      <th>16</th>\n", "      <td>90000</td>\n", "      <td>2</td>\n", "      <td>2</td>\n", "      <td>1</td>\n", "      <td>35</td>\n", "      <td>0</td>\n", "      <td>72112</td>\n", "      <td>0</td>\n", "      <td>2</td>\n", "      <td>91108</td>\n", "      <td>0.954684</td>\n", "      <td>0.883510</td>\n", "      <td>0.952754</td>\n", "      <td>0.953646</td>\n", "      <td>0.952596</td>\n", "    </tr>\n", "    <tr>\n", "      <th>17</th>\n", "      <td>10000</td>\n", "      <td>2</td>\n", "      <td>2</td>\n", "      <td>1</td>\n", "      <td>37</td>\n", "      <td>-1</td>\n", "      <td>3305</td>\n", "      <td>0</td>\n", "      <td>70</td>\n", "      <td>3236</td>\n", "      <td>0.999550</td>\n", "      <td>0.848221</td>\n", "      <td>0.800478</td>\n", "      <td>0.999590</td>\n", "      <td>0.999652</td>\n", "    </tr>\n", "    <tr>\n", "      <th>18</th>\n", "      <td>80000</td>\n", "      <td>1</td>\n", "      <td>3</td>\n", "      <td>1</td>\n", "      <td>36</td>\n", "      <td>0</td>\n", "      <td>81066</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>73411</td>\n", "      <td>0.961214</td>\n", "      <td>0.926935</td>\n", "      <td>0.962670</td>\n", "      <td>0.963936</td>\n", "      <td>0.962893</td>\n", "    </tr>\n", "    <tr>\n", "      <th>19</th>\n", "      <td>320000</td>\n", "      <td>2</td>\n", "      <td>1</td>\n", "      <td>1</td>\n", "      <td>36</td>\n", "      <td>-1</td>\n", "      <td>7868</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>96906</td>\n", "      <td>0.792507</td>\n", "      <td>0.755381</td>\n", "      <td>0.703680</td>\n", "      <td>0.400943</td>\n", "      <td>0.999851</td>\n", "    </tr>\n", "  </tbody>\n", "</table>\n", "</div>"], "text/plain": ["        X1  X2  X3  X4  X5  X6     X12  Y  TotalDelay  TotalPayment   PartMay  \\\n", "0   180000   1   2   1  47   0  179253  0           0         71798  0.962902   \n", "1   110000   2   2   1  35   0    6137  1           0         17094  0.963518   \n", "2    70000   2   2   2  22   0   66505  1           0         51151  0.963766   \n", "3   200000   2   1   2  27  -2    4941  0        -126        224043  1.028571   \n", "4   370000   2   1   1  39   0  141552  0           0         42614  0.937577   \n", "5   260000   2   1   1  29   0   71864  0         -14        160056  0.000000   \n", "6    90000   2   1   1  43  -1   16139  0         -39        135882  0.000000   \n", "7   220000   2   1   1  43  -1    1090  1          32          1002  0.000000   \n", "8    50000   1   2   1  35   1   48047  1          63         78530 -0.012172   \n", "9    50000   2   3   2  40   0    5538  0           0         19283  0.962316   \n", "10  130000   1   2   1  24   0   46113  0          16         38480  0.997442   \n", "11  200000   2   1   2  25  -1    8926  0         -63         82748  0.000000   \n", "12  230000   2   2   1  38  -2   12696  0        -126        111453  0.000000   \n", "13   90000   2   1   2  29  -2    -240  0        -126             0  1.004184   \n", "14  230000   1   3   2  37  -1   36571  0         -34        287273  0.887873   \n", "15  130000   1   2   2  33   2    2183  0          78          3578  0.000000   \n", "16   90000   2   2   1  35   0   72112  0           2         91108  0.954684   \n", "17   10000   2   2   1  37  -1    3305  0          70          3236  0.999550   \n", "18   80000   1   3   1  36   0   81066  0           0         73411  0.961214   \n", "19  320000   2   1   1  36  -1    7868  0           0         96906  0.792507   \n", "\n", "    PartJune  PartJuly  PartAugust  PartSeptember  \n", "0   0.964215  0.958865    0.961978       0.961112  \n", "1   0.963962  0.777823    0.721906       0.850305  \n", "2   0.964848  0.962690    0.956517       0.962942  \n", "3  -0.004450 -0.004802   -0.004502      -0.009899  \n", "4   0.958029  0.965386    0.959245       0.968143  \n", "5   0.000000  0.000000    0.000000       0.942813  \n", "6   0.000000  0.000000    0.997711       0.393333  \n", "7   0.999083  0.999083    0.999083       0.866455  \n", "8   0.985915  0.915291    0.956269       0.999979  \n", "9   0.963619  0.879296    0.847636       0.806216  \n", "10  0.999978  0.959249    0.999979       0.899013  \n", "11  0.000000  0.000000    0.000000       0.000000  \n", "12  0.000000  0.000000    0.000000       0.000000  \n", "13  1.004184  1.004184    1.004184       1.004184  \n", "14 -0.000340  0.769071    0.760863       0.952064  \n", "15  0.000000  0.000000    0.000000       0.993003  \n", "16  0.883510  0.952754    0.953646       0.952596  \n", "17  0.848221  0.800478    0.999590       0.999652  \n", "18  0.926935  0.962670    0.963936       0.962893  \n", "19  0.755381  0.703680    0.400943       0.999851  "]}, "execution_count": 13, "metadata": {}, "output_type": "execute_result"}], "source": ["# drop some columns\n", "\n", "df1 = df1.drop(['X'+str(n) for n in range(7,12)] + ['X'+str(n) for n in range(13,24)], axis=1)\n", "df1.head(20)"]}, {"cell_type": "code", "execution_count": 13, "metadata": {}, "outputs": [{"data": {"text/plain": ["<matplotlib.colorbar.Colorbar at 0x2a506a1a0b8>"]}, "execution_count": 14, "metadata": {}, "output_type": "execute_result"}, {"data": {"image/png": "iVBORw0KGgoAAAANSUhEUgAAAXEAAAEFCAYAAAAVNNdzAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nnDEAAAAASUVORK5CYII=\n", "text/plain": ["<matplotlib.figure.Figure at 0x2a505a1d630>"]}, "metadata": {}, "output_type": "display_data"}], "source": ["from sklearn.decomposition import PCA\n", "from numpy import inf\n", "pca = PCA(n_components=2, svd_solver='randomized')\n", "dfpca = df1.values\n", "dfpca[dfpca == -inf] = 0\n", "y = dfpca[:, 7]\n", "proj = pca.fit_transform(dfpca[:, :7 + 8:])\n", "plt.scatter(proj[:, 0], proj[:, 1], c=y)\n", "plt.colorbar()"]}, {"cell_type": "code", "execution_count": 14, "metadata": {}, "outputs": [{"name": "stdout", "output_type": "stream", "text": ["       X1  X2  X3  X4  X5  X6     X12  Y  TotalDelay  TotalPayment   PartMay  \\\n", "0  180000   1   2   1  47   0  179253  0           0         71798  0.962902   \n", "1  110000   2   2   1  35   0    6137  1           0         17094  0.963518   \n", "2   70000   2   2   2  22   0   66505  1           0         51151  0.963766   \n", "3  200000   2   1   2  27  -2    4941  0        -126        224043  1.028571   \n", "4  370000   2   1   1  39   0  141552  0           0         42614  0.937577   \n", "\n", "   PartJune  PartJuly  PartAugust  PartSeptember  \n", "0  0.964215  0.958865    0.961978       0.961112  \n", "1  0.963962  0.777823    0.721906       0.850305  \n", "2  0.964848  0.962690    0.956517       0.962942  \n", "3 -0.004450 -0.004802   -0.004502      -0.009899  \n", "4  0.958029  0.965386    0.959245       0.968143  \n"]}], "source": ["# training/crossval set\n", "\n", "X = df1.values\n", "X[X==-inf] = 0\n", "print(df1.head())\n", "# training set\n", "X_train = X[:, :]\n", "Y_train = X[:, 7].ravel()\n", "X_train = np.delete(X_train, 7, axis=1)\n", "# expected result\n", "expected = X[20000:, 7].ravel()\n", "# cross-validation data set\n", "X_cross = X[20000:, :]\n", "X_cross = np.delete(X_cross, 7, axis=1)"]}, {"cell_type": "code", "execution_count": 15, "metadata": {}, "outputs": [{"name": "stdout", "output_type": "stream", "text": ["[[ 209 1732]\n", " [  26  533]]\n"]}], "source": ["from sklearn.naive_bayes import GaussianNB\n", "\n", "# train the model\n", "GNB = GaussianNB()\n", "GNB.fit(X_train, Y_train)\n", "\n", "# use the model to predict the labels of the test data\n", "predicted = GNB.predict(X_cross)\n", "\n", "print(metrics.confusion_matrix(expected, predicted))"]}, {"cell_type": "code", "execution_count": 16, "metadata": {"scrolled": true}, "outputs": [{"name": "stdout", "output_type": "stream", "text": ["[[1848   93]\n", " [ 352  207]]\n"]}], "source": ["from sklearn.ensemble import GradientBoostingClassifier\n", "\n", "GBR = GradientBoostingClassifier()\n", "GBR.fit(X_train,Y_train)\n", "predicted = GBR.predict(X_cross)\n", "print(metrics.confusion_matrix(expected, predicted))"]}, {"cell_type": "code", "execution_count": 17, "metadata": {}, "outputs": [{"name": "stdout", "output_type": "stream", "text": ["[[1864   77]\n", " [ 368  191]]\n"]}], "source": ["from sklearn.neighbors import KNeighborsClassifier\n", "\n", "KNC = KNeighborsClassifier(5)\n", "KNC.fit(X_train, Y_train)\n", "predicted = KNC.predict(X_cross)\n", "print(metrics.confusion_matrix(expected, predicted))\n", "pred = KNC.predict_proba(X_train)"]}, {"cell_type": "code", "execution_count": 18, "metadata": {"scrolled": true}, "outputs": [{"name": "stdout", "output_type": "stream", "text": ["[[ 1.   0. ]\n", " [ 0.6  0.4]\n", " [ 0.6  0.4]\n", " [ 1.   0. ]\n", " [ 1.   0. ]\n", " [ 0.6  0.4]\n", " [ 1.   0. ]\n", " [ 0.4  0.6]\n", " [ 0.4  0.6]\n", " [ 0.8  0.2]]\n", "[ 0.  1.  1.  0.  0.  0.  0.  1.  1.  0.]\n"]}], "source": ["print(pred[:10])\n", "print(Y_train[:10])"]}, {"cell_type": "code", "execution_count": 19, "metadata": {}, "outputs": [{"name": "stdout", "output_type": "stream", "text": ["Fitting 3 folds for each of 5 candidates, totalling 15 fits\n"]}, {"name": "stderr", "output_type": "stream", "text": ["[Parallel(n_jobs=1)]: Done  15 out of  15 | elapsed:   27.2s finished\n"]}, {"data": {"text/plain": ["MLPClassifier(activation='relu', alpha=0.0001, batch_size='auto', beta_1=0.9,\n", "       beta_2=0.999, early_stopping=False, epsilon=1e-08,\n", "       hidden_layer_sizes=(20,), learning_rate='constant',\n", "       learning_rate_init=0.001, max_iter=200, momentum=0.9,\n", "       nesterovs_momentum=True, power_t=0.5, random_state=None,\n", "       shuffle=True, solver='adam', tol=0.0001, validation_fraction=0.1,\n", "       verbose=False, warm_start=False)"]}, "execution_count": 20, "metadata": {}, "output_type": "execute_result"}], "source": ["# neural network\n", "\n", "from sklearn.neural_network import MLPClassifier\n", "from sklearn.metrics import r2_score\n", "from sklearn.model_selection import GridSearchCV\n", "\n", "# optimisation - choix du nombre de couches\n", "param_grid = [\n", "  {'hidden_layer_sizes': [(nb,) for nb in range(20,50,10)]}, \n", "  {'alpha': [a/100 for a in range(0,40,20)]}\n", " ]\n", "\n", "neural2 = GridSearchCV(MLPClassifier(), param_grid, verbose=1)\n", "neural2.fit(X_train, Y_train)\n", "neural2.best_estimator_"]}, {"cell_type": "code", "execution_count": 20, "metadata": {"scrolled": true}, "outputs": [{"name": "stdout", "output_type": "stream", "text": ["[[1780  161]\n", " [ 378  181]]\n"]}, {"data": {"text/plain": ["array([[  1.00000000e+000,   4.54194423e-294],\n", "       [  1.00000000e+000,   1.14673239e-101],\n", "       [  1.00000000e+000,   1.13397258e-051],\n", "       [  1.00000000e+000,   1.89540529e-117],\n", "       [  1.00000000e+000,   8.01811448e-032],\n", "       [  1.00000000e+000,   1.14003085e-160],\n", "       [  1.00000000e+000,   1.02562443e-115],\n", "       [  1.00000000e+000,   9.41507727e-017],\n", "       [  1.00000000e+000,   3.16744761e-026],\n", "       [  6.95744980e-001,   3.04255020e-001]])"]}, "execution_count": 21, "metadata": {}, "output_type": "execute_result"}], "source": ["neural = MLPClassifier(activation='relu', alpha=0.0001, batch_size='auto', beta_1=0.9,\n", "       beta_2=0.999, early_stopping=False, epsilon=1e-08,\n", "       hidden_layer_sizes=(170,), learning_rate='constant',\n", "       learning_rate_init=0.001, max_iter=200, momentum=0.9,\n", "       nesterovs_momentum=True, power_t=0.5, random_state=None,\n", "       shuffle=True, solver='adam', tol=0.0001, validation_fraction=0.1,\n", "       verbose=False, warm_start=False)\n", "neural.fit(X_train, Y_train)\n", "predicted = neural.predict(X_cross)\n", "print(metrics.confusion_matrix(expected, predicted))\n", "neural.predict_proba(X_cross[:10])"]}, {"cell_type": "code", "execution_count": 21, "metadata": {"collapsed": true}, "outputs": [], "source": ["if_you_have_time = False\n", "if if_you_have_time:\n", "    from sklearn.gaussian_process import GaussianProcessClassifier\n", "\n", "    GPC = GaussianProcessClassifier()\n", "    GPC.fit(X_train, Y_train)\n", "    predicted = GPC.predict(X_cross)\n", "    print(metrics.confusion_matrix(expected, predicted))\n", "    GPC.predict_proba(X_cross)"]}, {"cell_type": "code", "execution_count": 22, "metadata": {}, "outputs": [{"name": "stdout", "output_type": "stream", "text": ["[[1928   13]\n", " [  65  494]]\n", "[[ 1.   0. ]\n", " [ 1.   0. ]\n", " [ 0.   1. ]\n", " ..., \n", " [ 1.   0. ]\n", " [ 1.   0. ]\n", " [ 0.8  0.2]]\n", "[ 0.  0.  1. ...,  0.  0.  0.]\n"]}], "source": ["from sklearn.ensemble import RandomForestClassifier\n", "\n", "RFC = RandomForestClassifier(5)\n", "RFC.fit(X_train, Y_train)\n", "predicted = RFC.predict(X_cross)\n", "print(metrics.confusion_matrix(expected, predicted))\n", "print(RFC.predict_proba(X_cross))\n", "print(expected)"]}, {"cell_type": "code", "execution_count": 23, "metadata": {"collapsed": true}, "outputs": [], "source": ["if if_you_have_time:\n", "    from sklearn.svm import SVC\n", "\n", "    SVC = SVC(probability = True)\n", "    SVC.fit(X_train, Y_train)\n", "    predicted = SVC.predict(X_cross)\n", "    print(metrics.confusion_matrix(expected, predicted))\n", "    print(SVC.predict_proba(X_cross))\n", "    print(expected)"]}, {"cell_type": "code", "execution_count": 24, "metadata": {}, "outputs": [{"name": "stdout", "output_type": "stream", "text": ["[[1861   80]\n", " [ 413  146]]\n", "[[ 0.85603222  0.14396778]\n", " [ 0.72020052  0.27979948]\n", " [ 0.69524708  0.30475292]\n", " ..., \n", " [ 0.69884026  0.30115974]\n", " [ 0.85309221  0.14690779]\n", " [ 0.75099513  0.24900487]]\n", "[ 0.  1.  1. ...,  0.  0.  0.]\n"]}], "source": ["from sklearn.linear_model import LogisticRegression\n", "\n", "LR = LogisticRegression()\n", "LR.fit(X_train, Y_train)\n", "predicted = LR.predict(X_cross)\n", "print(metrics.confusion_matrix(expected, predicted))\n", "print(LR.predict_proba(X_train))\n", "print(Y_train)"]}, {"cell_type": "code", "execution_count": 25, "metadata": {"collapsed": true}, "outputs": [], "source": ["#--------------#\n", "#\u00a0mod\u00e8le final #\n", "#--------------#\n", "\n", "# dataframe\n", "dfend = pd.read_csv(\"ensae_competition_test_X.txt\", header=[0, 1], sep='\\t', encoding=\"utf8\", index_col=0)\n", "dfend.columns = dfend.columns.droplevel(-1)\n", "\n", "# modifications colonnes\n", "dfend['TotalDelay'] = dfend.X11 + 2*dfend.X10 + 4*dfend.X9 + 8*dfend.X8 + 16*dfend.X7 + 32*dfend.X6\n", "dfend['TotalPayment'] = dfend.X23 + 2*dfend.X22 + 3*dfend.X21 + 4*dfend.X20 + 5*dfend.X19 + 6*dfend.X18\n", "dfend['PartMay'] = -(dfend.X22 - dfend.X17)/(dfend.X17 + 1)\n", "dfend['PartJune'] = -(dfend.X21 - dfend.X16)/(dfend.X16 + 1)\n", "dfend['PartJuly'] = -(dfend.X20 - dfend.X15)/(dfend.X15 + 1)\n", "dfend['PartAugust'] = -(dfend.X19 - dfend.X14)/(dfend.X14 + 1)\n", "dfend['PartSeptember'] = -(dfend.X18 - dfend.X13)/(dfend.X13 + 1)\n", "dfend = dfend.drop(['X'+str(n) for n in range(7,12)] + ['X'+str(n) for n in range(13,24)], axis=1)\n", "\n", "# dataset as array\n", "X = dfend.values\n", "X[X==-inf] = 0"]}, {"cell_type": "code", "execution_count": 26, "metadata": {"collapsed": true}, "outputs": [], "source": ["# pr\u00e9dictions\n", "\n", "# r\u00e9seau de neuronnes\n", "l = neural.predict(X)\n", "text_file = open('answerN.txt','w')\n", "for e in l:\n", "    text_file.write(str(int(e)) + '\\n')\n", "text_file.close()"]}, {"cell_type": "code", "execution_count": 27, "metadata": {"collapsed": true}, "outputs": [], "source": ["# random forest\n", "l = RFC.predict(X)\n", "text_file = open('answerRF.txt','w')\n", "for e in l:\n", "    text_file.write(str(int(e)) + '\\n')\n", "text_file.close()"]}, {"cell_type": "code", "execution_count": 28, "metadata": {"collapsed": true}, "outputs": [], "source": []}], "metadata": {"anaconda-cloud": {}, "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": "3.6.4"}}, "nbformat": 4, "nbformat_minor": 2}