Data_Analysis_Notebook.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Logistic Regression Analysis: Determinants of Women as Primary Source of Revenue\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "![banner](banner.jpg)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Content Table\n",
    "- [0. Environment Setup](#0.-Environment-Setup)\n",
    "- [1. Data Exploration & Management](#1.-Data-Exploration-&-Management)\n",
    "- [2. Feature Selection](#2.Feature-Selection)\n",
    "  - [A. Quantitative Data](#A.-Quantitative-Data)\n",
    "  - [B. Qualitative Data](#B.-Qualitative-Data)\n",
    "- [3. Final Data Preparation](#3.-Final-Data-Preparation)\n",
    "- [4. Statistical Modeling](#4.-Statistical-Modeling)\n",
    "  - [A. Multicollinearity tests](#A.-Multicollinearity-tests)\n",
    "  - [B. Data Standardization](#B.-Data-Standardization)\n",
    "  - [C. Model creation & fitting](#C.-Model-creation-&-fitting)\n",
    "  - [D. Model Optimization](#D.-Model-Optimization)\n",
    "  - [E. Model testing](#E.-Model-testing)\n",
    "- [5. Conclusion](#5.-Conclusion)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Project Overview\n",
    "\n",
    "This project aims to analyze the determinants for a woman being the primary source of revenue in a household using data from the \"Conditions de Travail 2013\" survey. The analysis will be focused on individuals living in couples in the same household.\n",
    "\n",
    "### Objectives\n",
    "The objective of the project is to analyze, based on the 2013 Working Conditions survey, the determinants for a woman being the primary source of revenue in the household.\n",
    "1. Data Exploration & Management: Explore the dataset and the available data and handle missing data.\n",
    "2. Feature Selection: The dataset presents 541 Columns/Features. A selection needs to be made to be able to conduct further analysis.\n",
    "3. Build and evaluate a logistic regression model to identify significant predictors of a women being the primary source of revenue in the household.\n",
    "\n",
    "### Dataset\n",
    "\n",
    "- **Source**: Conditions de Travail 2013 (Volet \"Actif\")\n",
    "- **Initial Dataset Size**: 33,673 individuals\n",
    "- **Filtered Dataset**: Individuals living in couples at the same household\n",
    "\n",
    "### Dependencies\n",
    "\n",
    "- Python libraries: pandas, numpy, matplotlib, seaborn, statsmodels, scikit-learn, plotly, scipy"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 0. Environment Setup"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Import necessary librairies\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import seaborn as sns\n",
    "import plotly.express as px\n",
    "import statsmodels.api as sm\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "dataset_path = \"Data/individus_ct2013.sas7bdat\""
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 1. Data Exploration & Management"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Warning: column count mismatch (234 + 255 != 541)\n",
      "\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>C1C</th>\n",
       "      <th>C1D</th>\n",
       "      <th>IDENT_MEN</th>\n",
       "      <th>IDENT_IND</th>\n",
       "      <th>NPERS</th>\n",
       "      <th>NACTIFS</th>\n",
       "      <th>TYPMEN5</th>\n",
       "      <th>TYPMEN15</th>\n",
       "      <th>IR</th>\n",
       "      <th>REVMEN</th>\n",
       "      <th>...</th>\n",
       "      <th>lnaisd</th>\n",
       "      <th>lnaisper</th>\n",
       "      <th>lnaismer</th>\n",
       "      <th>natnaisper</th>\n",
       "      <th>natnaismer</th>\n",
       "      <th>nati</th>\n",
       "      <th>revannu_drap</th>\n",
       "      <th>revannu</th>\n",
       "      <th>revmensc</th>\n",
       "      <th>revmenscc_drap</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>0100019101000</td>\n",
       "      <td>010001910100002</td>\n",
       "      <td>2.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2</td>\n",
       "      <td>22</td>\n",
       "      <td>1</td>\n",
       "      <td>2200.0</td>\n",
       "      <td>...</td>\n",
       "      <td>11</td>\n",
       "      <td>11</td>\n",
       "      <td>11</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1500.0</td>\n",
       "      <td>125.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0100083001000</td>\n",
       "      <td>010008300100002</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>5</td>\n",
       "      <td>52</td>\n",
       "      <td>2</td>\n",
       "      <td>2250.0</td>\n",
       "      <td>...</td>\n",
       "      <td>11</td>\n",
       "      <td>11</td>\n",
       "      <td>11</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-2.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-2.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>0100084601000</td>\n",
       "      <td>010008460100002</td>\n",
       "      <td>5.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>4</td>\n",
       "      <td>42</td>\n",
       "      <td>1</td>\n",
       "      <td>2300.0</td>\n",
       "      <td>...</td>\n",
       "      <td>11</td>\n",
       "      <td>11</td>\n",
       "      <td>11</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>3 rows × 541 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   C1C  C1D      IDENT_MEN        IDENT_IND  NPERS  NACTIFS TYPMEN5 TYPMEN15  \\\n",
       "0    2    2  0100019101000  010001910100002    2.0      1.0       2       22   \n",
       "1  NaN  NaN  0100083001000  010008300100002    2.0      2.0       5       52   \n",
       "2    2    2  0100084601000  010008460100002    5.0      2.0       4       42   \n",
       "\n",
       "  IR  REVMEN  ...  lnaisd  lnaisper  lnaismer  natnaisper  natnaismer  nati  \\\n",
       "0  1  2200.0  ...      11        11        11         NaN         NaN   NaN   \n",
       "1  2  2250.0  ...      11        11        11         NaN         NaN   NaN   \n",
       "2  1  2300.0  ...      11        11        11         NaN         NaN   NaN   \n",
       "\n",
       "  revannu_drap  revannu revmensc  revmenscc_drap  \n",
       "0          1.0   1500.0    125.0             1.0  \n",
       "1         -2.0      NaN      NaN            -2.0  \n",
       "2          1.0      0.0      0.0             1.0  \n",
       "\n",
       "[3 rows x 541 columns]"
      ]
     },
     "execution_count": 114,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Load the data from the 'individus_ct2013.sas7bdat' file into a DataFrame named df\n",
    "df = pd.read_sas(dataset_path, format=\"sas7bdat\", encoding=\"latin-1\")\n",
    "\n",
    "# Display the DataFrame\n",
    "df.head(3)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### This analysis will be only based on couples living in the same household. This is represented by the value 1 in the **'COUPLE'** variable."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(24024, 541)"
      ]
     },
     "execution_count": 115,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Filter only persons in couple in the same household for the study\n",
    "df = df[df[\"COUPLE\"] == \"1\"]\n",
    "df.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Creation of the target variable: A binary variable that represents whether a Women is the primary source of revenue in the household"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/tmp/ipykernel_160722/3477372514.py:2: PerformanceWarning:\n",
      "\n",
      "DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
      "\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "target\n",
       "False    14629\n",
       "True      9395\n",
       "Name: count, dtype: int64"
      ]
     },
     "execution_count": 116,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Creat new variable 'target', this target represent women (in couple) who are the primary source of revenue in their household\n",
    "df[\"target\"] = df[[\"SEXE\", \"PRACT\"]].apply(\n",
    "    lambda x: True if x.iloc[0] == \"1\" and x.iloc[1] == \"1\" else False, axis=1\n",
    ")\n",
    "\n",
    "# Counts number of each category NOT PROVIDING (False) vs PROVIDING (True)\n",
    "df[\"target\"].value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### The columns PRACT and SEXE are now represented as needed by the **'target'** variable. 'LIENPREF' is not needed as well"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Dropping useless columns\n",
    "df = df.drop(columns=[\"PRACT\", \"SEXE\", \"LIENPREF\"])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Analysis of the columns types"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "object     462\n",
       "float64     76\n",
       "bool         1\n",
       "Name: count, dtype: int64"
      ]
     },
     "execution_count": 118,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Display the distribution of the target variable and thus the distribution of quantitative and qualitative variables\n",
    "df.dtypes.value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 2.Feature Selection"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Our primary feature selection will be based on the relation of the columns with our target variables. The data set has both **Qualitative** & **Quantitative** data which will be explored seperatly at this stage."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### A. Quantitative Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(24024, 76)"
      ]
     },
     "execution_count": 119,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Get quantitative variables to calculate pearson correlation\n",
    "quantitatives = df.select_dtypes(include=[\"float64\", \"int64\"])\n",
    "quantitatives.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Analysis of the missing values:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "SUPA               23928\n",
       "TPSINT             23777\n",
       "tps_interim        23777\n",
       "SUPH               23336\n",
       "duree_cessation    23280\n",
       "                   ...  \n",
       "AGE                    0\n",
       "ANAIS                  0\n",
       "MNAIS                  0\n",
       "COEFFUC                0\n",
       "revmenscc_drap         0\n",
       "Length: 76, dtype: int64"
      ]
     },
     "execution_count": 120,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "quantitatives.isna().sum().sort_values(ascending=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.plotly.v1+json": {
       "config": {
        "plotlyServerURL": "https://plot.ly"
       },
       "data": [
        {
         "alignmentgroup": "True",
         "hovertemplate": "variable=0<br>index=%{x}<br>value=%{y}<extra></extra>",
         "legendgroup": "0",
         "marker": {
          "color": "#636efa",
          "pattern": {
           "shape": ""
          }
         },
         "name": "0",
         "offsetgroup": "0",
         "orientation": "v",
         "showlegend": true,
         "textposition": "auto",
         "type": "bar",
         "x": [
          "SUPA",
          "TPSINT",
          "tps_interim",
          "SUPH",
          "duree_cessation",
          "RABSP",
          "NBJARR",
          "POURCLIEN",
          "HHTOT",
          "FIN2M",
          "FIN2H",
          "DEBUT2M",
          "DEBUT2H",
          "DEBDETM",
          "debut_contrat",
          "DEBDETA",
          "revannu",
          "DUDET",
          "tps_contrat",
          "DATACCM",
          "DATACCA",
          "AGARRIV",
          "ANARRIV",
          "NBASTR",
          "NBNUIT",
          "TPMAISON",
          "ATMAL",
          "NBDIMA",
          "KMANNU",
          "NBRKM",
          "DAMOIS",
          "COMBIEN",
          "RESTMAIN",
          "NBSAME",
          "NOIK2",
          "INFOH",
          "TOTREVEN",
          "REVMENS",
          "ANETA",
          "revmensc",
          "FINM",
          "FINH",
          "DEBUTM",
          "DEBUTH",
          "WHO",
          "pondqaa",
          "REVMENMN",
          "REVMENMXUC",
          "REVMENMNUC",
          "REVMEN",
          "REVMENUC",
          "REVMENMX",
          "CONGE",
          "HH",
          "AGFINETU",
          "JOURTR",
          "TRAJET",
          "FINETUD",
          "ANFINETU",
          "DATANT",
          "anciennete",
          "CJACTOCCUP",
          "revannu_drap",
          "pondcal",
          "NPERS",
          "TPSINFO",
          "NACTIFS",
          "NBENF18P",
          "NBENF3A17",
          "NBENFM3",
          "NOIK1",
          "AGE",
          "ANAIS",
          "MNAIS",
          "COEFFUC",
          "revmenscc_drap"
         ],
         "xaxis": "x",
         "y": [
          23928,
          23777,
          23777,
          23336,
          23280,
          23107,
          23057,
          23036,
          22948,
          22533,
          22532,
          22530,
          22528,
          22459,
          22452,
          22452,
          22308,
          22243,
          22243,
          22061,
          21996,
          21675,
          21675,
          21349,
          20347,
          17820,
          17496,
          16224,
          15720,
          15720,
          15164,
          14715,
          14562,
          11922,
          8878,
          6523,
          6182,
          4397,
          3447,
          2681,
          2270,
          2265,
          2098,
          2095,
          1727,
          1301,
          1035,
          1035,
          1035,
          1035,
          1035,
          1035,
          761,
          440,
          242,
          190,
          130,
          130,
          130,
          39,
          39,
          4,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0
         ],
         "yaxis": "y"
        }
       ],
       "layout": {
        "barmode": "relative",
        "legend": {
         "title": {
          "text": "variable"
         },
         "tracegroupgap": 0
        },
        "template": {
         "data": {
          "bar": [
           {
            "error_x": {
             "color": "#2a3f5f"
            },
            "error_y": {
             "color": "#2a3f5f"
            },
            "marker": {
             "line": {
              "color": "#E5ECF6",
              "width": 0.5
             },
             "pattern": {
              "fillmode": "overlay",
              "size": 10,
              "solidity": 0.2
             }
            },
            "type": "bar"
           }
          ],
          "barpolar": [
           {
            "marker": {
             "line": {
              "color": "#E5ECF6",
              "width": 0.5
             },
             "pattern": {
              "fillmode": "overlay",
              "size": 10,
              "solidity": 0.2
             }
            },
            "type": "barpolar"
           }
          ],
          "carpet": [
           {
            "aaxis": {
             "endlinecolor": "#2a3f5f",
             "gridcolor": "white",
             "linecolor": "white",
             "minorgridcolor": "white",
             "startlinecolor": "#2a3f5f"
            },
            "baxis": {
             "endlinecolor": "#2a3f5f",
             "gridcolor": "white",
             "linecolor": "white",
             "minorgridcolor": "white",
             "startlinecolor": "#2a3f5f"
            },
            "type": "carpet"
           }
          ],
          "choropleth": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "type": "choropleth"
           }
          ],
          "contour": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "colorscale": [
             [
              0,
              "#0d0887"
             ],
             [
              0.1111111111111111,
              "#46039f"
             ],
             [
              0.2222222222222222,
              "#7201a8"
             ],
             [
              0.3333333333333333,
              "#9c179e"
             ],
             [
              0.4444444444444444,
              "#bd3786"
             ],
             [
              0.5555555555555556,
              "#d8576b"
             ],
             [
              0.6666666666666666,
              "#ed7953"
             ],
             [
              0.7777777777777778,
              "#fb9f3a"
             ],
             [
              0.8888888888888888,
              "#fdca26"
             ],
             [
              1,
              "#f0f921"
             ]
            ],
            "type": "contour"
           }
          ],
          "contourcarpet": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "type": "contourcarpet"
           }
          ],
          "heatmap": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "colorscale": [
             [
              0,
              "#0d0887"
             ],
             [
              0.1111111111111111,
              "#46039f"
             ],
             [
              0.2222222222222222,
              "#7201a8"
             ],
             [
              0.3333333333333333,
              "#9c179e"
             ],
             [
              0.4444444444444444,
              "#bd3786"
             ],
             [
              0.5555555555555556,
              "#d8576b"
             ],
             [
              0.6666666666666666,
              "#ed7953"
             ],
             [
              0.7777777777777778,
              "#fb9f3a"
             ],
             [
              0.8888888888888888,
              "#fdca26"
             ],
             [
              1,
              "#f0f921"
             ]
            ],
            "type": "heatmap"
           }
          ],
          "heatmapgl": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "colorscale": [
             [
              0,
              "#0d0887"
             ],
             [
              0.1111111111111111,
              "#46039f"
             ],
             [
              0.2222222222222222,
              "#7201a8"
             ],
             [
              0.3333333333333333,
              "#9c179e"
             ],
             [
              0.4444444444444444,
              "#bd3786"
             ],
             [
              0.5555555555555556,
              "#d8576b"
             ],
             [
              0.6666666666666666,
              "#ed7953"
             ],
             [
              0.7777777777777778,
              "#fb9f3a"
             ],
             [
              0.8888888888888888,
              "#fdca26"
             ],
             [
              1,
              "#f0f921"
             ]
            ],
            "type": "heatmapgl"
           }
          ],
          "histogram": [
           {
            "marker": {
             "pattern": {
              "fillmode": "overlay",
              "size": 10,
              "solidity": 0.2
             }
            },
            "type": "histogram"
           }
          ],
          "histogram2d": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "colorscale": [
             [
              0,
              "#0d0887"
             ],
             [
              0.1111111111111111,
              "#46039f"
             ],
             [
              0.2222222222222222,
              "#7201a8"
             ],
             [
              0.3333333333333333,
              "#9c179e"
             ],
             [
              0.4444444444444444,
              "#bd3786"
             ],
             [
              0.5555555555555556,
              "#d8576b"
             ],
             [
              0.6666666666666666,
              "#ed7953"
             ],
             [
              0.7777777777777778,
              "#fb9f3a"
             ],
             [
              0.8888888888888888,
              "#fdca26"
             ],
             [
              1,
              "#f0f921"
             ]
            ],
            "type": "histogram2d"
           }
          ],
          "histogram2dcontour": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "colorscale": [
             [
              0,
              "#0d0887"
             ],
             [
              0.1111111111111111,
              "#46039f"
             ],
             [
              0.2222222222222222,
              "#7201a8"
             ],
             [
              0.3333333333333333,
              "#9c179e"
             ],
             [
              0.4444444444444444,
              "#bd3786"
             ],
             [
              0.5555555555555556,
              "#d8576b"
             ],
             [
              0.6666666666666666,
              "#ed7953"
             ],
             [
              0.7777777777777778,
              "#fb9f3a"
             ],
             [
              0.8888888888888888,
              "#fdca26"
             ],
             [
              1,
              "#f0f921"
             ]
            ],
            "type": "histogram2dcontour"
           }
          ],
          "mesh3d": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "type": "mesh3d"
           }
          ],
          "parcoords": [
           {
            "line": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "parcoords"
           }
          ],
          "pie": [
           {
            "automargin": true,
            "type": "pie"
           }
          ],
          "scatter": [
           {
            "fillpattern": {
             "fillmode": "overlay",
             "size": 10,
             "solidity": 0.2
            },
            "type": "scatter"
           }
          ],
          "scatter3d": [
           {
            "line": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "scatter3d"
           }
          ],
          "scattercarpet": [
           {
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "scattercarpet"
           }
          ],
          "scattergeo": [
           {
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "scattergeo"
           }
          ],
          "scattergl": [
           {
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "scattergl"
           }
          ],
          "scattermapbox": [
           {
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "scattermapbox"
           }
          ],
          "scatterpolar": [
           {
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "scatterpolar"
           }
          ],
          "scatterpolargl": [
           {
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "scatterpolargl"
           }
          ],
          "scatterternary": [
           {
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "scatterternary"
           }
          ],
          "surface": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "colorscale": [
             [
              0,
              "#0d0887"
             ],
             [
              0.1111111111111111,
              "#46039f"
             ],
             [
              0.2222222222222222,
              "#7201a8"
             ],
             [
              0.3333333333333333,
              "#9c179e"
             ],
             [
              0.4444444444444444,
              "#bd3786"
             ],
             [
              0.5555555555555556,
              "#d8576b"
             ],
             [
              0.6666666666666666,
              "#ed7953"
             ],
             [
              0.7777777777777778,
              "#fb9f3a"
             ],
             [
              0.8888888888888888,
              "#fdca26"
             ],
             [
              1,
              "#f0f921"
             ]
            ],
            "type": "surface"
           }
          ],
          "table": [
           {
            "cells": {
             "fill": {
              "color": "#EBF0F8"
             },
             "line": {
              "color": "white"
             }
            },
            "header": {
             "fill": {
              "color": "#C8D4E3"
             },
             "line": {
              "color": "white"
             }
            },
            "type": "table"
           }
          ]
         },
         "layout": {
          "annotationdefaults": {
           "arrowcolor": "#2a3f5f",
           "arrowhead": 0,
           "arrowwidth": 1
          },
          "autotypenumbers": "strict",
          "coloraxis": {
           "colorbar": {
            "outlinewidth": 0,
            "ticks": ""
           }
          },
          "colorscale": {
           "diverging": [
            [
             0,
             "#8e0152"
            ],
            [
             0.1,
             "#c51b7d"
            ],
            [
             0.2,
             "#de77ae"
            ],
            [
             0.3,
             "#f1b6da"
            ],
            [
             0.4,
             "#fde0ef"
            ],
            [
             0.5,
             "#f7f7f7"
            ],
            [
             0.6,
             "#e6f5d0"
            ],
            [
             0.7,
             "#b8e186"
            ],
            [
             0.8,
             "#7fbc41"
            ],
            [
             0.9,
             "#4d9221"
            ],
            [
             1,
             "#276419"
            ]
           ],
           "sequential": [
            [
             0,
             "#0d0887"
            ],
            [
             0.1111111111111111,
             "#46039f"
            ],
            [
             0.2222222222222222,
             "#7201a8"
            ],
            [
             0.3333333333333333,
             "#9c179e"
            ],
            [
             0.4444444444444444,
             "#bd3786"
            ],
            [
             0.5555555555555556,
             "#d8576b"
            ],
            [
             0.6666666666666666,
             "#ed7953"
            ],
            [
             0.7777777777777778,
             "#fb9f3a"
            ],
            [
             0.8888888888888888,
             "#fdca26"
            ],
            [
             1,
             "#f0f921"
            ]
           ],
           "sequentialminus": [
            [
             0,
             "#0d0887"
            ],
            [
             0.1111111111111111,
             "#46039f"
            ],
            [
             0.2222222222222222,
             "#7201a8"
            ],
            [
             0.3333333333333333,
             "#9c179e"
            ],
            [
             0.4444444444444444,
             "#bd3786"
            ],
            [
             0.5555555555555556,
             "#d8576b"
            ],
            [
             0.6666666666666666,
             "#ed7953"
            ],
            [
             0.7777777777777778,
             "#fb9f3a"
            ],
            [
             0.8888888888888888,
             "#fdca26"
            ],
            [
             1,
             "#f0f921"
            ]
           ]
          },
          "colorway": [
           "#636efa",
           "#EF553B",
           "#00cc96",
           "#ab63fa",
           "#FFA15A",
           "#19d3f3",
           "#FF6692",
           "#B6E880",
           "#FF97FF",
           "#FECB52"
          ],
          "font": {
           "color": "#2a3f5f"
          },
          "geo": {
           "bgcolor": "white",
           "lakecolor": "white",
           "landcolor": "#E5ECF6",
           "showlakes": true,
           "showland": true,
           "subunitcolor": "white"
          },
          "hoverlabel": {
           "align": "left"
          },
          "hovermode": "closest",
          "mapbox": {
           "style": "light"
          },
          "paper_bgcolor": "white",
          "plot_bgcolor": "#E5ECF6",
          "polar": {
           "angularaxis": {
            "gridcolor": "white",
            "linecolor": "white",
            "ticks": ""
           },
           "bgcolor": "#E5ECF6",
           "radialaxis": {
            "gridcolor": "white",
            "linecolor": "white",
            "ticks": ""
           }
          },
          "scene": {
           "xaxis": {
            "backgroundcolor": "#E5ECF6",
            "gridcolor": "white",
            "gridwidth": 2,
            "linecolor": "white",
            "showbackground": true,
            "ticks": "",
            "zerolinecolor": "white"
           },
           "yaxis": {
            "backgroundcolor": "#E5ECF6",
            "gridcolor": "white",
            "gridwidth": 2,
            "linecolor": "white",
            "showbackground": true,
            "ticks": "",
            "zerolinecolor": "white"
           },
           "zaxis": {
            "backgroundcolor": "#E5ECF6",
            "gridcolor": "white",
            "gridwidth": 2,
            "linecolor": "white",
            "showbackground": true,
            "ticks": "",
            "zerolinecolor": "white"
           }
          },
          "shapedefaults": {
           "line": {
            "color": "#2a3f5f"
           }
          },
          "ternary": {
           "aaxis": {
            "gridcolor": "white",
            "linecolor": "white",
            "ticks": ""
           },
           "baxis": {
            "gridcolor": "white",
            "linecolor": "white",
            "ticks": ""
           },
           "bgcolor": "#E5ECF6",
           "caxis": {
            "gridcolor": "white",
            "linecolor": "white",
            "ticks": ""
           }
          },
          "title": {
           "x": 0.05
          },
          "xaxis": {
           "automargin": true,
           "gridcolor": "white",
           "linecolor": "white",
           "ticks": "",
           "title": {
            "standoff": 15
           },
           "zerolinecolor": "white",
           "zerolinewidth": 2
          },
          "yaxis": {
           "automargin": true,
           "gridcolor": "white",
           "linecolor": "white",
           "ticks": "",
           "title": {
            "standoff": 15
           },
           "zerolinecolor": "white",
           "zerolinewidth": 2
          }
         }
        },
        "title": {
         "text": "Number of missing values in the qualitative variables"
        },
        "xaxis": {
         "anchor": "y",
         "domain": [
          0,
          1
         ],
         "title": {
          "text": "index"
         }
        },
        "yaxis": {
         "anchor": "x",
         "domain": [
          0,
          1
         ],
         "title": {
          "text": "value"
         }
        }
       }
      }
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "px.bar(\n",
    "    quantitatives.isna().sum().sort_values(ascending=False),\n",
    "    title=\"Number of missing values in the qualitative variables\",\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### We chose to fill the missing values with medians to be able to study their correlation. The median was chosen rather then a mean to avoid any bias that could be introduced by outliers."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "quantitatives.fillna(quantitatives.median(), inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Copy the target variable to the quantitatives DataFrame\n",
    "quantitatives[\"target\"] = df[\"target\"]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Correlation Analysis:"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### We use pearson correlation coefficient to analyze the correlation between the variables and our target variable. Scipy was used rather than pandas correlation function so that we can extract and use the pvalue as well."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from scipy.stats import pearsonr\n",
    "\n",
    "# Calculate Correlation of all quantitative variables with target variable\n",
    "quantitative_corr = {\n",
    "    \"Column\": [],\n",
    "    \"Correlation\": [],\n",
    "    \"P-value\": [],\n",
    "}  # Create a dictionary to store the results\n",
    "for col in quantitatives.columns:\n",
    "    corr, pvalue = pearsonr(\n",
    "        quantitatives[col], quantitatives[\"target\"]\n",
    "    )  # Calculate the correlation and p-value using the pearsonr function\n",
    "    quantitative_corr[\"Column\"].append(col)\n",
    "    quantitative_corr[\"Correlation\"].append(corr)\n",
    "    quantitative_corr[\"P-value\"].append(pvalue)\n",
    "# Create a DataFrame from the dictionary and sort it by the correlation and P-value\n",
    "quant_corr = (\n",
    "    pd.DataFrame(quantitative_corr)\n",
    "    .sort_values(by=[\"Correlation\", \"P-value\"], ascending=False, key=(abs))\n",
    "    .reset_index()\n",
    "    .drop(0)\n",
    "    .drop(columns=\"index\")\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "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>Column</th>\n",
       "      <th>Correlation</th>\n",
       "      <th>P-value</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>HH</td>\n",
       "      <td>0.268071</td>\n",
       "      <td>0.000000e+00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>KMANNU</td>\n",
       "      <td>0.205376</td>\n",
       "      <td>4.007308e-227</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>JOURTR</td>\n",
       "      <td>0.190734</td>\n",
       "      <td>1.334825e-195</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>NBRKM</td>\n",
       "      <td>0.181284</td>\n",
       "      <td>1.392872e-176</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>CJACTOCCUP</td>\n",
       "      <td>-0.167778</td>\n",
       "      <td>3.490268e-151</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>72</th>\n",
       "      <td>INFOH</td>\n",
       "      <td>0.002117</td>\n",
       "      <td>7.427827e-01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>73</th>\n",
       "      <td>DATACCM</td>\n",
       "      <td>-0.001456</td>\n",
       "      <td>8.214214e-01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>74</th>\n",
       "      <td>ANARRIV</td>\n",
       "      <td>0.001345</td>\n",
       "      <td>8.348608e-01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75</th>\n",
       "      <td>POURCLIEN</td>\n",
       "      <td>-0.000870</td>\n",
       "      <td>8.927699e-01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>76</th>\n",
       "      <td>RABSP</td>\n",
       "      <td>-0.000859</td>\n",
       "      <td>8.940753e-01</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>76 rows × 3 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        Column  Correlation        P-value\n",
       "1           HH     0.268071   0.000000e+00\n",
       "2       KMANNU     0.205376  4.007308e-227\n",
       "3       JOURTR     0.190734  1.334825e-195\n",
       "4        NBRKM     0.181284  1.392872e-176\n",
       "5   CJACTOCCUP    -0.167778  3.490268e-151\n",
       "..         ...          ...            ...\n",
       "72       INFOH     0.002117   7.427827e-01\n",
       "73     DATACCM    -0.001456   8.214214e-01\n",
       "74     ANARRIV     0.001345   8.348608e-01\n",
       "75   POURCLIEN    -0.000870   8.927699e-01\n",
       "76       RABSP    -0.000859   8.940753e-01\n",
       "\n",
       "[76 rows x 3 columns]"
      ]
     },
     "execution_count": 125,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "quant_corr"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### We apply a filter on the resulting data frame to keep only relevant correlations (**'P-value'<0.05**) and those having a (**correlation coefficient > 0.1**) in absolute value."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "quant_corr_filtered = quant_corr[\n",
    "    (quant_corr[\"P-value\"] < 0.05) & (abs(quant_corr[\"Correlation\"]) > 0.1)\n",
    "]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.plotly.v1+json": {
       "config": {
        "plotlyServerURL": "https://plot.ly"
       },
       "data": [
        {
         "alignmentgroup": "True",
         "hovertemplate": "Column=%{x}<br>Correlation=%{text}<br>color=%{marker.color}<extra></extra>",
         "legendgroup": "",
         "marker": {
          "color": [
           0.2680712606717622,
           0.20537625056206557,
           0.190734414780538,
           0.18128419771979237,
           0.16777843684076418,
           0.11524751640596409,
           0.10437628813955944
          ],
          "coloraxis": "coloraxis",
          "pattern": {
           "shape": ""
          }
         },
         "name": "",
         "offsetgroup": "",
         "orientation": "v",
         "showlegend": false,
         "text": [
          0.2680712606717622,
          0.20537625056206557,
          0.190734414780538,
          0.18128419771979237,
          -0.16777843684076418,
          0.11524751640596409,
          0.10437628813955944
         ],
         "textposition": "inside",
         "texttemplate": "%{text:.2f}",
         "type": "bar",
         "x": [
          "HH",
          "KMANNU",
          "JOURTR",
          "NBRKM",
          "CJACTOCCUP",
          "REVMENS",
          "revmensc"
         ],
         "xaxis": "x",
         "y": [
          0.2680712606717622,
          0.20537625056206557,
          0.190734414780538,
          0.18128419771979237,
          -0.16777843684076418,
          0.11524751640596409,
          0.10437628813955944
         ],
         "yaxis": "y"
        }
       ],
       "layout": {
        "barmode": "relative",
        "coloraxis": {
         "colorbar": {
          "title": {
           "text": "Absolute Correlation"
          }
         },
         "colorscale": [
          [
           0,
           "rgb(247,251,255)"
          ],
          [
           0.125,
           "rgb(222,235,247)"
          ],
          [
           0.25,
           "rgb(198,219,239)"
          ],
          [
           0.375,
           "rgb(158,202,225)"
          ],
          [
           0.5,
           "rgb(107,174,214)"
          ],
          [
           0.625,
           "rgb(66,146,198)"
          ],
          [
           0.75,
           "rgb(33,113,181)"
          ],
          [
           0.875,
           "rgb(8,81,156)"
          ],
          [
           1,
           "rgb(8,48,107)"
          ]
         ]
        },
        "legend": {
         "tracegroupgap": 0
        },
        "template": {
         "data": {
          "bar": [
           {
            "error_x": {
             "color": "#2a3f5f"
            },
            "error_y": {
             "color": "#2a3f5f"
            },
            "marker": {
             "line": {
              "color": "#E5ECF6",
              "width": 0.5
             },
             "pattern": {
              "fillmode": "overlay",
              "size": 10,
              "solidity": 0.2
             }
            },
            "type": "bar"
           }
          ],
          "barpolar": [
           {
            "marker": {
             "line": {
              "color": "#E5ECF6",
              "width": 0.5
             },
             "pattern": {
              "fillmode": "overlay",
              "size": 10,
              "solidity": 0.2
             }
            },
            "type": "barpolar"
           }
          ],
          "carpet": [
           {
            "aaxis": {
             "endlinecolor": "#2a3f5f",
             "gridcolor": "white",
             "linecolor": "white",
             "minorgridcolor": "white",
             "startlinecolor": "#2a3f5f"
            },
            "baxis": {
             "endlinecolor": "#2a3f5f",
             "gridcolor": "white",
             "linecolor": "white",
             "minorgridcolor": "white",
             "startlinecolor": "#2a3f5f"
            },
            "type": "carpet"
           }
          ],
          "choropleth": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "type": "choropleth"
           }
          ],
          "contour": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "colorscale": [
             [
              0,
              "#0d0887"
             ],
             [
              0.1111111111111111,
              "#46039f"
             ],
             [
              0.2222222222222222,
              "#7201a8"
             ],
             [
              0.3333333333333333,
              "#9c179e"
             ],
             [
              0.4444444444444444,
              "#bd3786"
             ],
             [
              0.5555555555555556,
              "#d8576b"
             ],
             [
              0.6666666666666666,
              "#ed7953"
             ],
             [
              0.7777777777777778,
              "#fb9f3a"
             ],
             [
              0.8888888888888888,
              "#fdca26"
             ],
             [
              1,
              "#f0f921"
             ]
            ],
            "type": "contour"
           }
          ],
          "contourcarpet": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "type": "contourcarpet"
           }
          ],
          "heatmap": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "colorscale": [
             [
              0,
              "#0d0887"
             ],
             [
              0.1111111111111111,
              "#46039f"
             ],
             [
              0.2222222222222222,
              "#7201a8"
             ],
             [
              0.3333333333333333,
              "#9c179e"
             ],
             [
              0.4444444444444444,
              "#bd3786"
             ],
             [
              0.5555555555555556,
              "#d8576b"
             ],
             [
              0.6666666666666666,
              "#ed7953"
             ],
             [
              0.7777777777777778,
              "#fb9f3a"
             ],
             [
              0.8888888888888888,
              "#fdca26"
             ],
             [
              1,
              "#f0f921"
             ]
            ],
            "type": "heatmap"
           }
          ],
          "heatmapgl": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "colorscale": [
             [
              0,
              "#0d0887"
             ],
             [
              0.1111111111111111,
              "#46039f"
             ],
             [
              0.2222222222222222,
              "#7201a8"
             ],
             [
              0.3333333333333333,
              "#9c179e"
             ],
             [
              0.4444444444444444,
              "#bd3786"
             ],
             [
              0.5555555555555556,
              "#d8576b"
             ],
             [
              0.6666666666666666,
              "#ed7953"
             ],
             [
              0.7777777777777778,
              "#fb9f3a"
             ],
             [
              0.8888888888888888,
              "#fdca26"
             ],
             [
              1,
              "#f0f921"
             ]
            ],
            "type": "heatmapgl"
           }
          ],
          "histogram": [
           {
            "marker": {
             "pattern": {
              "fillmode": "overlay",
              "size": 10,
              "solidity": 0.2
             }
            },
            "type": "histogram"
           }
          ],
          "histogram2d": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "colorscale": [
             [
              0,
              "#0d0887"
             ],
             [
              0.1111111111111111,
              "#46039f"
             ],
             [
              0.2222222222222222,
              "#7201a8"
             ],
             [
              0.3333333333333333,
              "#9c179e"
             ],
             [
              0.4444444444444444,
              "#bd3786"
             ],
             [
              0.5555555555555556,
              "#d8576b"
             ],
             [
              0.6666666666666666,
              "#ed7953"
             ],
             [
              0.7777777777777778,
              "#fb9f3a"
             ],
             [
              0.8888888888888888,
              "#fdca26"
             ],
             [
              1,
              "#f0f921"
             ]
            ],
            "type": "histogram2d"
           }
          ],
          "histogram2dcontour": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "colorscale": [
             [
              0,
              "#0d0887"
             ],
             [
              0.1111111111111111,
              "#46039f"
             ],
             [
              0.2222222222222222,
              "#7201a8"
             ],
             [
              0.3333333333333333,
              "#9c179e"
             ],
             [
              0.4444444444444444,
              "#bd3786"
             ],
             [
              0.5555555555555556,
              "#d8576b"
             ],
             [
              0.6666666666666666,
              "#ed7953"
             ],
             [
              0.7777777777777778,
              "#fb9f3a"
             ],
             [
              0.8888888888888888,
              "#fdca26"
             ],
             [
              1,
              "#f0f921"
             ]
            ],
            "type": "histogram2dcontour"
           }
          ],
          "mesh3d": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "type": "mesh3d"
           }
          ],
          "parcoords": [
           {
            "line": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "parcoords"
           }
          ],
          "pie": [
           {
            "automargin": true,
            "type": "pie"
           }
          ],
          "scatter": [
           {
            "fillpattern": {
             "fillmode": "overlay",
             "size": 10,
             "solidity": 0.2
            },
            "type": "scatter"
           }
          ],
          "scatter3d": [
           {
            "line": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "scatter3d"
           }
          ],
          "scattercarpet": [
           {
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "scattercarpet"
           }
          ],
          "scattergeo": [
           {
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "scattergeo"
           }
          ],
          "scattergl": [
           {
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "scattergl"
           }
          ],
          "scattermapbox": [
           {
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "scattermapbox"
           }
          ],
          "scatterpolar": [
           {
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "scatterpolar"
           }
          ],
          "scatterpolargl": [
           {
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "scatterpolargl"
           }
          ],
          "scatterternary": [
           {
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "scatterternary"
           }
          ],
          "surface": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "colorscale": [
             [
              0,
              "#0d0887"
             ],
             [
              0.1111111111111111,
              "#46039f"
             ],
             [
              0.2222222222222222,
              "#7201a8"
             ],
             [
              0.3333333333333333,
              "#9c179e"
             ],
             [
              0.4444444444444444,
              "#bd3786"
             ],
             [
              0.5555555555555556,
              "#d8576b"
             ],
             [
              0.6666666666666666,
              "#ed7953"
             ],
             [
              0.7777777777777778,
              "#fb9f3a"
             ],
             [
              0.8888888888888888,
              "#fdca26"
             ],
             [
              1,
              "#f0f921"
             ]
            ],
            "type": "surface"
           }
          ],
          "table": [
           {
            "cells": {
             "fill": {
              "color": "#EBF0F8"
             },
             "line": {
              "color": "white"
             }
            },
            "header": {
             "fill": {
              "color": "#C8D4E3"
             },
             "line": {
              "color": "white"
             }
            },
            "type": "table"
           }
          ]
         },
         "layout": {
          "annotationdefaults": {
           "arrowcolor": "#2a3f5f",
           "arrowhead": 0,
           "arrowwidth": 1
          },
          "autotypenumbers": "strict",
          "coloraxis": {
           "colorbar": {
            "outlinewidth": 0,
            "ticks": ""
           }
          },
          "colorscale": {
           "diverging": [
            [
             0,
             "#8e0152"
            ],
            [
             0.1,
             "#c51b7d"
            ],
            [
             0.2,
             "#de77ae"
            ],
            [
             0.3,
             "#f1b6da"
            ],
            [
             0.4,
             "#fde0ef"
            ],
            [
             0.5,
             "#f7f7f7"
            ],
            [
             0.6,
             "#e6f5d0"
            ],
            [
             0.7,
             "#b8e186"
            ],
            [
             0.8,
             "#7fbc41"
            ],
            [
             0.9,
             "#4d9221"
            ],
            [
             1,
             "#276419"
            ]
           ],
           "sequential": [
            [
             0,
             "#0d0887"
            ],
            [
             0.1111111111111111,
             "#46039f"
            ],
            [
             0.2222222222222222,
             "#7201a8"
            ],
            [
             0.3333333333333333,
             "#9c179e"
            ],
            [
             0.4444444444444444,
             "#bd3786"
            ],
            [
             0.5555555555555556,
             "#d8576b"
            ],
            [
             0.6666666666666666,
             "#ed7953"
            ],
            [
             0.7777777777777778,
             "#fb9f3a"
            ],
            [
             0.8888888888888888,
             "#fdca26"
            ],
            [
             1,
             "#f0f921"
            ]
           ],
           "sequentialminus": [
            [
             0,
             "#0d0887"
            ],
            [
             0.1111111111111111,
             "#46039f"
            ],
            [
             0.2222222222222222,
             "#7201a8"
            ],
            [
             0.3333333333333333,
             "#9c179e"
            ],
            [
             0.4444444444444444,
             "#bd3786"
            ],
            [
             0.5555555555555556,
             "#d8576b"
            ],
            [
             0.6666666666666666,
             "#ed7953"
            ],
            [
             0.7777777777777778,
             "#fb9f3a"
            ],
            [
             0.8888888888888888,
             "#fdca26"
            ],
            [
             1,
             "#f0f921"
            ]
           ]
          },
          "colorway": [
           "#636efa",
           "#EF553B",
           "#00cc96",
           "#ab63fa",
           "#FFA15A",
           "#19d3f3",
           "#FF6692",
           "#B6E880",
           "#FF97FF",
           "#FECB52"
          ],
          "font": {
           "color": "#2a3f5f"
          },
          "geo": {
           "bgcolor": "white",
           "lakecolor": "white",
           "landcolor": "#E5ECF6",
           "showlakes": true,
           "showland": true,
           "subunitcolor": "white"
          },
          "hoverlabel": {
           "align": "left"
          },
          "hovermode": "closest",
          "mapbox": {
           "style": "light"
          },
          "paper_bgcolor": "white",
          "plot_bgcolor": "#E5ECF6",
          "polar": {
           "angularaxis": {
            "gridcolor": "white",
            "linecolor": "white",
            "ticks": ""
           },
           "bgcolor": "#E5ECF6",
           "radialaxis": {
            "gridcolor": "white",
            "linecolor": "white",
            "ticks": ""
           }
          },
          "scene": {
           "xaxis": {
            "backgroundcolor": "#E5ECF6",
            "gridcolor": "white",
            "gridwidth": 2,
            "linecolor": "white",
            "showbackground": true,
            "ticks": "",
            "zerolinecolor": "white"
           },
           "yaxis": {
            "backgroundcolor": "#E5ECF6",
            "gridcolor": "white",
            "gridwidth": 2,
            "linecolor": "white",
            "showbackground": true,
            "ticks": "",
            "zerolinecolor": "white"
           },
           "zaxis": {
            "backgroundcolor": "#E5ECF6",
            "gridcolor": "white",
            "gridwidth": 2,
            "linecolor": "white",
            "showbackground": true,
            "ticks": "",
            "zerolinecolor": "white"
           }
          },
          "shapedefaults": {
           "line": {
            "color": "#2a3f5f"
           }
          },
          "ternary": {
           "aaxis": {
            "gridcolor": "white",
            "linecolor": "white",
            "ticks": ""
           },
           "baxis": {
            "gridcolor": "white",
            "linecolor": "white",
            "ticks": ""
           },
           "bgcolor": "#E5ECF6",
           "caxis": {
            "gridcolor": "white",
            "linecolor": "white",
            "ticks": ""
           }
          },
          "title": {
           "x": 0.05
          },
          "xaxis": {
           "automargin": true,
           "gridcolor": "white",
           "linecolor": "white",
           "ticks": "",
           "title": {
            "standoff": 15
           },
           "zerolinecolor": "white",
           "zerolinewidth": 2
          },
          "yaxis": {
           "automargin": true,
           "gridcolor": "white",
           "linecolor": "white",
           "ticks": "",
           "title": {
            "standoff": 15
           },
           "zerolinecolor": "white",
           "zerolinewidth": 2
          }
         }
        },
        "title": {
         "text": "Pearson Correlation of the quantitative variables with the target variable",
         "x": 0.5
        },
        "uniformtext": {
         "minsize": 8,
         "mode": "hide"
        },
        "xaxis": {
         "anchor": "y",
         "domain": [
          0,
          1
         ],
         "title": {
          "text": "Quantitative Variables"
         }
        },
        "yaxis": {
         "anchor": "x",
         "domain": [
          0,
          1
         ],
         "title": {
          "text": "Pearson Correlation"
         }
        }
       }
      }
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = px.bar(\n",
    "    quant_corr_filtered,\n",
    "    x=\"Column\",\n",
    "    y=\"Correlation\",\n",
    "    color=abs(quant_corr_filtered[\"Correlation\"]),\n",
    "    text=\"Correlation\",\n",
    "    color_continuous_scale=px.colors.sequential.Blues,\n",
    "    title=\"Pearson Correlation of the quantitative variables with the target variable\",\n",
    ")\n",
    "fig.update_traces(texttemplate=\"%{text:.2f}\", textposition=\"inside\")\n",
    "fig.update_layout(\n",
    "    xaxis_title=\"Quantitative Variables\",\n",
    "    yaxis_title=\"Pearson Correlation\",\n",
    "    title_x=0.5,  # Center the title\n",
    "    coloraxis_colorbar=dict(\n",
    "        title=\"Absolute Correlation\"\n",
    "    ),\n",
    "    uniformtext_minsize=8,\n",
    "    uniformtext_mode='hide'\n",
    ")\n",
    "fig.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Using the dictionnary we analyzed the meaning/content of these variables and found recurrent ones that we eliminated"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "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>Column</th>\n",
       "      <th>Correlation</th>\n",
       "      <th>P-value</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>HH</td>\n",
       "      <td>0.268071</td>\n",
       "      <td>0.000000e+00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>KMANNU</td>\n",
       "      <td>0.205376</td>\n",
       "      <td>4.007308e-227</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>REVMENS</td>\n",
       "      <td>0.115248</td>\n",
       "      <td>7.969893e-72</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    Column  Correlation        P-value\n",
       "0       HH     0.268071   0.000000e+00\n",
       "1   KMANNU     0.205376  4.007308e-227\n",
       "2  REVMENS     0.115248   7.969893e-72"
      ]
     },
     "execution_count": 128,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "quant_corr_filtered = (\n",
    "    quant_corr_filtered.set_index(\"Column\") # Set the column as the index \n",
    "    .drop([\"JOURTR\", \"CJACTOCCUP\", \"revmensc\", \"NBRKM\"]) \n",
    "    .reset_index()\n",
    ")\n",
    "quant_corr_filtered"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### We also found that one of the variables **'CJACTOCCUP'** is a qualitative one "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df[\"CJACTOCCUP\"] = df[\"CJACTOCCUP\"].apply(lambda x: str(x)) # Convert the column to string"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### B. Qualitative Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(24024, 463)"
      ]
     },
     "execution_count": 130,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Get qualitative variables to calculate chi² correlation\n",
    "qualitatives = df.select_dtypes(include=[\"object\"])\n",
    "qualitatives.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Copy the target variable to the qualitatives DataFrame\n",
    "qualitatives[\"target\"] = df[\"target\"]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Analysis of the missing values:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "LIEN_20    24024\n",
       "LIEN_19    24024\n",
       "LIEN_18    24023\n",
       "LIEN_16    24021\n",
       "LIEN_17    24021\n",
       "           ...  \n",
       "RISQUAL        0\n",
       "RISFINA        0\n",
       "RISDANG        0\n",
       "RISEMPL        0\n",
       "target         0\n",
       "Length: 464, dtype: int64"
      ]
     },
     "execution_count": 132,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "qualitatives.isna().sum().sort_values(ascending=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 133,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.plotly.v1+json": {
       "config": {
        "plotlyServerURL": "https://plot.ly"
       },
       "data": [
        {
         "alignmentgroup": "True",
         "hovertemplate": "variable=0<br>index=%{x}<br>value=%{y}<extra></extra>",
         "legendgroup": "0",
         "marker": {
          "color": "#636efa",
          "pattern": {
           "shape": ""
          }
         },
         "name": "0",
         "offsetgroup": "0",
         "orientation": "v",
         "showlegend": true,
         "textposition": "auto",
         "type": "bar",
         "x": [
          "LIEN_20",
          "LIEN_19",
          "LIEN_18",
          "LIEN_16",
          "LIEN_17",
          "LIEN_15",
          "LIEN_14",
          "LIEN_13",
          "LIEN_12",
          "LIEN_11",
          "NUMFA_SOURCE",
          "SSECH_SOURCE",
          "IDENT_FA_SOURCE",
          "IDENT_LOG_SOURCE",
          "IDENT_MEN_SOURCE",
          "IDENT_IND_SOURCE",
          "DOUBLONTYPE",
          "LIEN_10",
          "LANG4",
          "LIEN_09",
          "NATIO1NB",
          "INFORM",
          "AIDE2E",
          "AIDFAM",
          "LIEN_08",
          "AFTYPTRAV",
          "NBSALAID",
          "AIDE1E",
          "TRAJET_DRAP",
          "PASTRA",
          "LANG2",
          "TPSINTU",
          "LIEN_07",
          "FORMINIT",
          "ETUDIPL",
          "ACTUACC2",
          "OPA",
          "AIDOM",
          "LIEN_06",
          "RABSPU",
          "REVTRNSA",
          "nati",
          "EREG",
          "ACTUACC1",
          "REVTRSAL",
          "RDET",
          "DUDETU",
          "INDEMNI",
          "REDRISK",
          "SIGN",
          "ARRET",
          "NBSAL",
          "ASTRCOMP",
          "RABS",
          "APPLINSP",
          "CLIENT",
          "SYNDICNS",
          "PREJOUR",
          "NUICOMP",
          "LIEN_05",
          "natnaismer",
          "natnaisper",
          "CHGTINFL",
          "RAISTP",
          "TXTPPB",
          "PRESEM",
          "STATUTEXT",
          "ASSOEVAL",
          "TRANCHRE",
          "DIMCOMP",
          "NBSALB",
          "PACS",
          "OBJATTEIN",
          "OBJMODIF",
          "POLYERR",
          "POLYQUAL",
          "RISKINF3",
          "RISKINF4",
          "RISKINF2",
          "RISKINF1",
          "RISKINF5",
          "B3H",
          "B3G",
          "B3F",
          "B3E",
          "B3D",
          "B3C",
          "B3B",
          "B3A",
          "B2D",
          "B2C",
          "B2A",
          "B2B",
          "NBRKMU",
          "TITPUB",
          "CYCLE",
          "APPLIQ",
          "CHANGOP",
          "TENSION4",
          "CHGTINFO",
          "CHGTCONS",
          "RWDEMAND",
          "LIEN_04",
          "NBSALENTC",
          "AUTRETA",
          "EVACRIT",
          "EPIUTIL",
          "DICOL",
          "MELDOM",
          "DICHEF",
          "INTERACT",
          "CSEDEB",
          "cserdeb",
          "PROFDEB",
          "STATUTA",
          "CHGTCOLL",
          "REMPLAS",
          "LIEN_03",
          "HSUPCOMP",
          "REUNION",
          "INFOHU2",
          "INFOHU1",
          "RWNORMJ",
          "csermer",
          "PROFMER",
          "CSEMER",
          "ITOTREV",
          "STATUTM",
          "PUBLIC1",
          "PUBLIC2",
          "TENSION1",
          "DIFPAYE",
          "DRH",
          "NBSALA",
          "DOCEVAL",
          "FORMSEC",
          "FORTMOD1",
          "FONCTION",
          "TYPEMPLOI",
          "CLASSIF",
          "LIEUW",
          "DISCURP",
          "SECURITE",
          "EVA",
          "POLY",
          "SYNDIC",
          "RWSURV",
          "COMMENT",
          "ACCHEF",
          "VISITUNI",
          "EPI",
          "COLLECT",
          "AIDCHEF",
          "TENSION2",
          "STARK",
          "MEDECIN",
          "RISK",
          "JOINDRE",
          "PRISCONG",
          "CONGEU",
          "DEPNAIS",
          "TCDD",
          "cserper",
          "PROFPER",
          "CSEPER",
          "STATUTP",
          "TINTERIM",
          "C1D",
          "C1C",
          "C1B",
          "C1E",
          "C2",
          "C1A",
          "B6",
          "B5F",
          "B5E",
          "B5D",
          "B5C",
          "B5B",
          "B5A",
          "B4D",
          "B4C",
          "B4B",
          "B4A",
          "B1G",
          "B1I",
          "B1H",
          "B1J",
          "B1F",
          "B1E",
          "B1D",
          "B1C",
          "B1B",
          "B1A",
          "A2F",
          "A2B",
          "A2C",
          "A2D",
          "A2E",
          "A2G",
          "A2L",
          "A2K",
          "A2H",
          "A2I",
          "A2J",
          "A2A",
          "A1L",
          "A1K",
          "A1J",
          "A1I",
          "A1G",
          "A1F",
          "A1E",
          "A1H",
          "A1D",
          "A1C",
          "A1B",
          "A1A",
          "INTRO",
          "FORTMOD6",
          "FORTMOD5",
          "FORTMOD4",
          "FORTMOD3",
          "FRANCE",
          "AIDEXT",
          "CHSCT",
          "AIDAUTR",
          "RESDOM",
          "PAYECOM",
          "FINETUDU",
          "FORTMOD7",
          "FORTMOD2",
          "LIBCONG",
          "ARETMAL",
          "TPP",
          "DEMENA",
          "DEMPRO",
          "NBEMP",
          "TCHOL",
          "TSAN",
          "TCHOC",
          "TINA",
          "TINDEP",
          "ACTIVFIN",
          "AUTSAL",
          "REPQAA",
          "region",
          "NUMENQ_ANON",
          "LE",
          "NUMFA",
          "EC",
          "IDENT_LOG",
          "IDENT_FA",
          "BS",
          "PER1E",
          "CHERCHEUR",
          "MER1E",
          "peun",
          "peun10",
          "SITUA",
          "CJSITUA",
          "TRAREF",
          "cser",
          "CSE",
          "PE",
          "LANG1",
          "SAMEDI",
          "NUIT",
          "PTMATIN",
          "DIMANCHE",
          "SOIR",
          "naf17",
          "fap87",
          "HHTOT_DRAP",
          "naf4",
          "JOURTR_DRAP",
          "FINH_DRAP",
          "fap225",
          "DATACCA_DRAP",
          "NBSAME_DRAP",
          "NBJARR_DRAP",
          "NBDIMA_DRAP",
          "DEBUTH_DRAP",
          "DATACCM_DRAP",
          "DEBUTM_DRAP",
          "RESTMAIN_DRAP",
          "NATIO1NA",
          "congetr",
          "naf38",
          "FIN2H_DRAP",
          "NBASTR_DRAP",
          "COMBIEN_DRAP",
          "DEBUT2M_DRAP",
          "DEBUT2H_DRAP",
          "TPMAISON_DRAP",
          "CONGE_DRAP",
          "ATMAL_DRAP",
          "NBRKM_DRAP",
          "INFOH_DRAP",
          "REVMENS_DRAP",
          "FINM_DRAP",
          "lnaismer",
          "lnaisper",
          "lnaisd",
          "HH_DRAP",
          "DUDET_DRAP",
          "ANETA_DRAP",
          "POURCLIEN_DRAP",
          "TOTREVEN_DRAP",
          "DIPLOME",
          "csei",
          "militaire",
          "CHAMP_CT2005",
          "ELIGEXT",
          "V",
          "typrepqaa",
          "STATUT",
          "FIN2M_DRAP",
          "RGES",
          "ETUDES",
          "SSECH",
          "INTERNET",
          "NAIP",
          "NAIM",
          "LIENMIG",
          "DEBDETM_DRAP",
          "QUAL_SIRET",
          "CHAMP_EMPL",
          "DAMOIS_DRAP",
          "QUAL_EMPLX",
          "DATANT_DRAP",
          "QUAL_ADR",
          "TPSINT_DRAP",
          "QUALITE_VOLET",
          "MERIMMI",
          "TYPEXTFP",
          "PREVIS",
          "NATIO",
          "DEBDETA_DRAP",
          "IMMI",
          "PERIMMI",
          "NBNUIT_DRAP",
          "CORRFORM",
          "ACTIM",
          "SECTOXNO",
          "HSUP",
          "ASTREINTE",
          "MAISON",
          "MISSION",
          "URGFAM1",
          "CVFVP",
          "AIDHID",
          "CWDEBOU",
          "CWPOSTU",
          "CWDEPLA",
          "CWLOURD",
          "CWMVT",
          "CWVIB",
          "CWVUE",
          "CWLETR",
          "CWMINUS",
          "CWVISO",
          "HYGSAL",
          "HYGCHUMI",
          "HYGCOUR",
          "HYGODEUR",
          "HYGCHAUD",
          "HYGFROID",
          "HYGSANI",
          "HYGVUE",
          "HYGPRIV",
          "CONDUITE",
          "CONTROLE",
          "PERIODE",
          "HORVAR",
          "ETAMATRI",
          "IDENT_MEN",
          "IDENT_IND",
          "TYPMEN5",
          "TYPMEN15",
          "IR",
          "CJACTOCCUP",
          "AGEQ",
          "LNAIS",
          "NAIS7",
          "COUPLE",
          "COUPLRP",
          "CONJOINT",
          "ENFANT",
          "PREMOIS",
          "LIEN_01",
          "LIEN_02",
          "TYPOLOG",
          "IPROPLOC",
          "AUTLOG",
          "IPRAN",
          "FAMPRINC",
          "CJENQUETE",
          "ENCADR",
          "REPOS",
          "DETHOR",
          "HORANGT",
          "SECFUPOU",
          "SECINFEC",
          "ACTIP",
          "SECACCID",
          "AIDCOLL",
          "TRAVSEUL",
          "ACCOL",
          "TENSION3",
          "DETRESSE",
          "CALMER",
          "CONTRAD",
          "CORRTAN",
          "CORRINF",
          "CORRCOP",
          "CORRCOL",
          "CORRLOG",
          "CORRMAT",
          "TELEPORT",
          "MICRO1",
          "MICRO2",
          "MEL",
          "INTRANET",
          "CRAINTE",
          "METIER",
          "NOCHOM",
          "TENIR",
          "SOUHAIT",
          "BSANTE",
          "BCHRO",
          "BLIMI",
          "BRECO",
          "COMPUTIL2",
          "COMPUTIL1",
          "SIEG34",
          "DEBORD",
          "SECROUT",
          "ENTENDR",
          "ACCIDT",
          "RWDEP",
          "RWCAD",
          "RWTECH",
          "RWCOLEG",
          "RWNORMH",
          "RWDEM",
          "RWINFO",
          "CHAINE",
          "DEPECH",
          "OBJECTIF",
          "RESTMAI",
          "DELAIS",
          "PUBLIC",
          "INTERUP",
          "INCIDENT",
          "PROCEDUR",
          "REPETE",
          "MONOTON",
          "NOUVELLE",
          "RISQUAL",
          "RISFINA",
          "RISDANG",
          "RISEMPL",
          "target"
         ],
         "xaxis": "x",
         "y": [
          24024,
          24024,
          24023,
          24021,
          24021,
          24020,
          24018,
          24018,
          24017,
          24014,
          24012,
          24012,
          24012,
          24012,
          24012,
          24012,
          24011,
          24004,
          24002,
          23974,
          23960,
          23932,
          23932,
          23932,
          23931,
          23924,
          23924,
          23924,
          23894,
          23863,
          23832,
          23777,
          23764,
          23575,
          23575,
          23549,
          23336,
          23235,
          23203,
          23107,
          23096,
          22926,
          22880,
          22462,
          22271,
          22257,
          22188,
          22163,
          22163,
          22163,
          21987,
          21480,
          21480,
          21471,
          21430,
          21380,
          21380,
          21273,
          20520,
          20520,
          20361,
          20120,
          19989,
          19265,
          19265,
          19084,
          19078,
          17863,
          17755,
          17383,
          17166,
          17004,
          16975,
          16975,
          16797,
          16797,
          16366,
          16366,
          16366,
          16366,
          16366,
          16004,
          16003,
          16003,
          16003,
          16003,
          16003,
          16003,
          16003,
          16003,
          16003,
          16002,
          16002,
          15720,
          15327,
          14123,
          14097,
          14005,
          13990,
          13883,
          13883,
          13705,
          12939,
          12377,
          12152,
          11673,
          11479,
          11458,
          10737,
          9634,
          8595,
          8093,
          8093,
          8093,
          8061,
          8025,
          7575,
          7211,
          7096,
          6719,
          6523,
          6523,
          6218,
          6198,
          6198,
          6198,
          6182,
          5963,
          5663,
          5663,
          5655,
          3737,
          3736,
          3736,
          3433,
          3433,
          2881,
          2665,
          2664,
          2663,
          2645,
          2644,
          2644,
          2644,
          2644,
          2644,
          2644,
          2644,
          2644,
          2644,
          2644,
          2644,
          2644,
          2644,
          2644,
          2644,
          2644,
          2644,
          2568,
          2568,
          2371,
          2352,
          1823,
          1823,
          1823,
          1631,
          1558,
          1378,
          1378,
          1307,
          1307,
          1307,
          1306,
          1306,
          1305,
          1305,
          1305,
          1305,
          1304,
          1303,
          1303,
          1303,
          1303,
          1303,
          1300,
          1300,
          1300,
          1300,
          1299,
          1299,
          1299,
          1298,
          1298,
          1298,
          1296,
          1296,
          1296,
          1296,
          1296,
          1296,
          1296,
          1296,
          1296,
          1296,
          1296,
          1295,
          1294,
          1292,
          1292,
          1292,
          1291,
          1291,
          1291,
          1291,
          1290,
          1289,
          1289,
          1288,
          1180,
          1022,
          1022,
          1022,
          1022,
          789,
          789,
          789,
          789,
          789,
          302,
          253,
          237,
          237,
          187,
          187,
          187,
          133,
          133,
          132,
          131,
          131,
          131,
          131,
          131,
          123,
          100,
          81,
          52,
          52,
          52,
          52,
          52,
          52,
          52,
          52,
          38,
          22,
          18,
          13,
          13,
          4,
          4,
          4,
          3,
          3,
          3,
          3,
          1,
          1,
          1,
          1,
          1,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0
         ],
         "yaxis": "y"
        }
       ],
       "layout": {
        "barmode": "relative",
        "legend": {
         "title": {
          "text": "variable"
         },
         "tracegroupgap": 0
        },
        "showlegend": false,
        "template": {
         "data": {
          "bar": [
           {
            "error_x": {
             "color": "#2a3f5f"
            },
            "error_y": {
             "color": "#2a3f5f"
            },
            "marker": {
             "line": {
              "color": "#E5ECF6",
              "width": 0.5
             },
             "pattern": {
              "fillmode": "overlay",
              "size": 10,
              "solidity": 0.2
             }
            },
            "type": "bar"
           }
          ],
          "barpolar": [
           {
            "marker": {
             "line": {
              "color": "#E5ECF6",
              "width": 0.5
             },
             "pattern": {
              "fillmode": "overlay",
              "size": 10,
              "solidity": 0.2
             }
            },
            "type": "barpolar"
           }
          ],
          "carpet": [
           {
            "aaxis": {
             "endlinecolor": "#2a3f5f",
             "gridcolor": "white",
             "linecolor": "white",
             "minorgridcolor": "white",
             "startlinecolor": "#2a3f5f"
            },
            "baxis": {
             "endlinecolor": "#2a3f5f",
             "gridcolor": "white",
             "linecolor": "white",
             "minorgridcolor": "white",
             "startlinecolor": "#2a3f5f"
            },
            "type": "carpet"
           }
          ],
          "choropleth": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "type": "choropleth"
           }
          ],
          "contour": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "colorscale": [
             [
              0,
              "#0d0887"
             ],
             [
              0.1111111111111111,
              "#46039f"
             ],
             [
              0.2222222222222222,
              "#7201a8"
             ],
             [
              0.3333333333333333,
              "#9c179e"
             ],
             [
              0.4444444444444444,
              "#bd3786"
             ],
             [
              0.5555555555555556,
              "#d8576b"
             ],
             [
              0.6666666666666666,
              "#ed7953"
             ],
             [
              0.7777777777777778,
              "#fb9f3a"
             ],
             [
              0.8888888888888888,
              "#fdca26"
             ],
             [
              1,
              "#f0f921"
             ]
            ],
            "type": "contour"
           }
          ],
          "contourcarpet": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "type": "contourcarpet"
           }
          ],
          "heatmap": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "colorscale": [
             [
              0,
              "#0d0887"
             ],
             [
              0.1111111111111111,
              "#46039f"
             ],
             [
              0.2222222222222222,
              "#7201a8"
             ],
             [
              0.3333333333333333,
              "#9c179e"
             ],
             [
              0.4444444444444444,
              "#bd3786"
             ],
             [
              0.5555555555555556,
              "#d8576b"
             ],
             [
              0.6666666666666666,
              "#ed7953"
             ],
             [
              0.7777777777777778,
              "#fb9f3a"
             ],
             [
              0.8888888888888888,
              "#fdca26"
             ],
             [
              1,
              "#f0f921"
             ]
            ],
            "type": "heatmap"
           }
          ],
          "heatmapgl": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "colorscale": [
             [
              0,
              "#0d0887"
             ],
             [
              0.1111111111111111,
              "#46039f"
             ],
             [
              0.2222222222222222,
              "#7201a8"
             ],
             [
              0.3333333333333333,
              "#9c179e"
             ],
             [
              0.4444444444444444,
              "#bd3786"
             ],
             [
              0.5555555555555556,
              "#d8576b"
             ],
             [
              0.6666666666666666,
              "#ed7953"
             ],
             [
              0.7777777777777778,
              "#fb9f3a"
             ],
             [
              0.8888888888888888,
              "#fdca26"
             ],
             [
              1,
              "#f0f921"
             ]
            ],
            "type": "heatmapgl"
           }
          ],
          "histogram": [
           {
            "marker": {
             "pattern": {
              "fillmode": "overlay",
              "size": 10,
              "solidity": 0.2
             }
            },
            "type": "histogram"
           }
          ],
          "histogram2d": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "colorscale": [
             [
              0,
              "#0d0887"
             ],
             [
              0.1111111111111111,
              "#46039f"
             ],
             [
              0.2222222222222222,
              "#7201a8"
             ],
             [
              0.3333333333333333,
              "#9c179e"
             ],
             [
              0.4444444444444444,
              "#bd3786"
             ],
             [
              0.5555555555555556,
              "#d8576b"
             ],
             [
              0.6666666666666666,
              "#ed7953"
             ],
             [
              0.7777777777777778,
              "#fb9f3a"
             ],
             [
              0.8888888888888888,
              "#fdca26"
             ],
             [
              1,
              "#f0f921"
             ]
            ],
            "type": "histogram2d"
           }
          ],
          "histogram2dcontour": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "colorscale": [
             [
              0,
              "#0d0887"
             ],
             [
              0.1111111111111111,
              "#46039f"
             ],
             [
              0.2222222222222222,
              "#7201a8"
             ],
             [
              0.3333333333333333,
              "#9c179e"
             ],
             [
              0.4444444444444444,
              "#bd3786"
             ],
             [
              0.5555555555555556,
              "#d8576b"
             ],
             [
              0.6666666666666666,
              "#ed7953"
             ],
             [
              0.7777777777777778,
              "#fb9f3a"
             ],
             [
              0.8888888888888888,
              "#fdca26"
             ],
             [
              1,
              "#f0f921"
             ]
            ],
            "type": "histogram2dcontour"
           }
          ],
          "mesh3d": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "type": "mesh3d"
           }
          ],
          "parcoords": [
           {
            "line": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "parcoords"
           }
          ],
          "pie": [
           {
            "automargin": true,
            "type": "pie"
           }
          ],
          "scatter": [
           {
            "fillpattern": {
             "fillmode": "overlay",
             "size": 10,
             "solidity": 0.2
            },
            "type": "scatter"
           }
          ],
          "scatter3d": [
           {
            "line": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "scatter3d"
           }
          ],
          "scattercarpet": [
           {
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "scattercarpet"
           }
          ],
          "scattergeo": [
           {
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "scattergeo"
           }
          ],
          "scattergl": [
           {
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "scattergl"
           }
          ],
          "scattermapbox": [
           {
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "scattermapbox"
           }
          ],
          "scatterpolar": [
           {
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "scatterpolar"
           }
          ],
          "scatterpolargl": [
           {
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "scatterpolargl"
           }
          ],
          "scatterternary": [
           {
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "scatterternary"
           }
          ],
          "surface": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "colorscale": [
             [
              0,
              "#0d0887"
             ],
             [
              0.1111111111111111,
              "#46039f"
             ],
             [
              0.2222222222222222,
              "#7201a8"
             ],
             [
              0.3333333333333333,
              "#9c179e"
             ],
             [
              0.4444444444444444,
              "#bd3786"
             ],
             [
              0.5555555555555556,
              "#d8576b"
             ],
             [
              0.6666666666666666,
              "#ed7953"
             ],
             [
              0.7777777777777778,
              "#fb9f3a"
             ],
             [
              0.8888888888888888,
              "#fdca26"
             ],
             [
              1,
              "#f0f921"
             ]
            ],
            "type": "surface"
           }
          ],
          "table": [
           {
            "cells": {
             "fill": {
              "color": "#EBF0F8"
             },
             "line": {
              "color": "white"
             }
            },
            "header": {
             "fill": {
              "color": "#C8D4E3"
             },
             "line": {
              "color": "white"
             }
            },
            "type": "table"
           }
          ]
         },
         "layout": {
          "annotationdefaults": {
           "arrowcolor": "#2a3f5f",
           "arrowhead": 0,
           "arrowwidth": 1
          },
          "autotypenumbers": "strict",
          "coloraxis": {
           "colorbar": {
            "outlinewidth": 0,
            "ticks": ""
           }
          },
          "colorscale": {
           "diverging": [
            [
             0,
             "#8e0152"
            ],
            [
             0.1,
             "#c51b7d"
            ],
            [
             0.2,
             "#de77ae"
            ],
            [
             0.3,
             "#f1b6da"
            ],
            [
             0.4,
             "#fde0ef"
            ],
            [
             0.5,
             "#f7f7f7"
            ],
            [
             0.6,
             "#e6f5d0"
            ],
            [
             0.7,
             "#b8e186"
            ],
            [
             0.8,
             "#7fbc41"
            ],
            [
             0.9,
             "#4d9221"
            ],
            [
             1,
             "#276419"
            ]
           ],
           "sequential": [
            [
             0,
             "#0d0887"
            ],
            [
             0.1111111111111111,
             "#46039f"
            ],
            [
             0.2222222222222222,
             "#7201a8"
            ],
            [
             0.3333333333333333,
             "#9c179e"
            ],
            [
             0.4444444444444444,
             "#bd3786"
            ],
            [
             0.5555555555555556,
             "#d8576b"
            ],
            [
             0.6666666666666666,
             "#ed7953"
            ],
            [
             0.7777777777777778,
             "#fb9f3a"
            ],
            [
             0.8888888888888888,
             "#fdca26"
            ],
            [
             1,
             "#f0f921"
            ]
           ],
           "sequentialminus": [
            [
             0,
             "#0d0887"
            ],
            [
             0.1111111111111111,
             "#46039f"
            ],
            [
             0.2222222222222222,
             "#7201a8"
            ],
            [
             0.3333333333333333,
             "#9c179e"
            ],
            [
             0.4444444444444444,
             "#bd3786"
            ],
            [
             0.5555555555555556,
             "#d8576b"
            ],
            [
             0.6666666666666666,
             "#ed7953"
            ],
            [
             0.7777777777777778,
             "#fb9f3a"
            ],
            [
             0.8888888888888888,
             "#fdca26"
            ],
            [
             1,
             "#f0f921"
            ]
           ]
          },
          "colorway": [
           "#636efa",
           "#EF553B",
           "#00cc96",
           "#ab63fa",
           "#FFA15A",
           "#19d3f3",
           "#FF6692",
           "#B6E880",
           "#FF97FF",
           "#FECB52"
          ],
          "font": {
           "color": "#2a3f5f"
          },
          "geo": {
           "bgcolor": "white",
           "lakecolor": "white",
           "landcolor": "#E5ECF6",
           "showlakes": true,
           "showland": true,
           "subunitcolor": "white"
          },
          "hoverlabel": {
           "align": "left"
          },
          "hovermode": "closest",
          "mapbox": {
           "style": "light"
          },
          "paper_bgcolor": "white",
          "plot_bgcolor": "#E5ECF6",
          "polar": {
           "angularaxis": {
            "gridcolor": "white",
            "linecolor": "white",
            "ticks": ""
           },
           "bgcolor": "#E5ECF6",
           "radialaxis": {
            "gridcolor": "white",
            "linecolor": "white",
            "ticks": ""
           }
          },
          "scene": {
           "xaxis": {
            "backgroundcolor": "#E5ECF6",
            "gridcolor": "white",
            "gridwidth": 2,
            "linecolor": "white",
            "showbackground": true,
            "ticks": "",
            "zerolinecolor": "white"
           },
           "yaxis": {
            "backgroundcolor": "#E5ECF6",
            "gridcolor": "white",
            "gridwidth": 2,
            "linecolor": "white",
            "showbackground": true,
            "ticks": "",
            "zerolinecolor": "white"
           },
           "zaxis": {
            "backgroundcolor": "#E5ECF6",
            "gridcolor": "white",
            "gridwidth": 2,
            "linecolor": "white",
            "showbackground": true,
            "ticks": "",
            "zerolinecolor": "white"
           }
          },
          "shapedefaults": {
           "line": {
            "color": "#2a3f5f"
           }
          },
          "ternary": {
           "aaxis": {
            "gridcolor": "white",
            "linecolor": "white",
            "ticks": ""
           },
           "baxis": {
            "gridcolor": "white",
            "linecolor": "white",
            "ticks": ""
           },
           "bgcolor": "#E5ECF6",
           "caxis": {
            "gridcolor": "white",
            "linecolor": "white",
            "ticks": ""
           }
          },
          "title": {
           "x": 0.05
          },
          "xaxis": {
           "automargin": true,
           "gridcolor": "white",
           "linecolor": "white",
           "ticks": "",
           "title": {
            "standoff": 15
           },
           "zerolinecolor": "white",
           "zerolinewidth": 2
          },
          "yaxis": {
           "automargin": true,
           "gridcolor": "white",
           "linecolor": "white",
           "ticks": "",
           "title": {
            "standoff": 15
           },
           "zerolinecolor": "white",
           "zerolinewidth": 2
          }
         }
        },
        "title": {
         "text": "Number of Missing Values in the Quantitative Variables",
         "x": 0.5
        },
        "xaxis": {
         "anchor": "y",
         "domain": [
          0,
          1
         ],
         "tickangle": -45,
         "title": {
          "text": "Quantitative Variables"
         }
        },
        "yaxis": {
         "anchor": "x",
         "domain": [
          0,
          1
         ],
         "title": {
          "text": "Number of Missing Values"
         }
        }
       }
      }
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Create the bar chart\n",
    "fig = px.bar(\n",
    "    qualitatives.isna().sum().sort_values(ascending=False),\n",
    "    title=\"Number of Missing Values in the Quantitative Variables\",\n",
    "    labels={\n",
    "        \"Variable\": \"Quantitative Variables\",\n",
    "        \"MissingValues\": \"Number of Missing Values\",\n",
    "    },\n",
    ")\n",
    "\n",
    "# Update layout for better readability\n",
    "fig.update_layout(\n",
    "    xaxis_title=\"Quantitative Variables\",\n",
    "    yaxis_title=\"Number of Missing Values\",\n",
    "    title_x=0.5,  # Center the title\n",
    "    xaxis_tickangle=-45,\n",
    "    showlegend=False,\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### From the graph we decided to drop all columns with more than 2000 missing values as those could negatively impact the analysis and could not be used to calculate the relation to the target variable"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 134,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(24024, 464)"
      ]
     },
     "execution_count": 134,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "qualitatives.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 135,
   "metadata": {},
   "outputs": [],
   "source": [
    "qualitatives = qualitatives.dropna(axis=1, thresh=(len(df) - 2000)).dropna() # Drop columns with more than 2000 missing values then finally drop the rows with missing values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 136,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(18418, 302)"
      ]
     },
     "execution_count": 136,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "qualitatives.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Correlation Analysis:"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### We used the chi2 test to determine whether there is a realtion between the qualitative variables and our target variable for feature selection. The process is generally the same as the one we used for the quantitative variables."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 137,
   "metadata": {},
   "outputs": [],
   "source": [
    "from scipy.stats import chi2_contingency\n",
    "\n",
    "# Calculate Correlation of all qualitative variables with target variable\n",
    "qualitative_corr = {\n",
    "    \"Column\": [],\n",
    "    \"Chi2-stat\": [],\n",
    "    \"P-value\": [],\n",
    "    \"dof\": [],\n",
    "    \"expected\": [],\n",
    "}\n",
    "for col in qualitatives.columns:\n",
    "    contingency_table = pd.crosstab(qualitatives[col], qualitatives[\"target\"]) # Create a contingency table using the crosstab function\n",
    "    cchi2_stat, p_val, dof, expected = chi2_contingency(contingency_table) # Calculate the chi2 statistic, p-value, degrees of freedom and expected values using the chi2_contingency function\n",
    "    qualitative_corr[\"Column\"].append(col)\n",
    "    qualitative_corr[\"Chi2-stat\"].append(cchi2_stat)\n",
    "    qualitative_corr[\"P-value\"].append(p_val)\n",
    "    qualitative_corr[\"dof\"].append(dof)\n",
    "    qualitative_corr[\"expected\"].append(expected)\n",
    "\n",
    "# Create a DataFrame from the dictionary and sort it by the chi2 statistic and P-value\n",
    "quali_corr = (\n",
    "    pd.DataFrame(qualitative_corr)\n",
    "    .sort_values(by=[\"Chi2-stat\", \"P-value\"], ascending=False, key=(abs))\n",
    "    .reset_index()\n",
    "    .drop(0)\n",
    "    .drop(columns=\"index\")[1:]\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 138,
   "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>Column</th>\n",
       "      <th>Chi2-stat</th>\n",
       "      <th>P-value</th>\n",
       "      <th>dof</th>\n",
       "      <th>expected</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>IDENT_FA</td>\n",
       "      <td>11694.082450</td>\n",
       "      <td>1.0</td>\n",
       "      <td>14338</td>\n",
       "      <td>[[0.6172765772613747, 0.38272342273862525], [0...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>IDENT_LOG</td>\n",
       "      <td>11694.082450</td>\n",
       "      <td>1.0</td>\n",
       "      <td>14338</td>\n",
       "      <td>[[0.6172765772613747, 0.38272342273862525], [0...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>IDENT_MEN</td>\n",
       "      <td>11694.082450</td>\n",
       "      <td>1.0</td>\n",
       "      <td>14350</td>\n",
       "      <td>[[0.6172765772613747, 0.38272342273862525], [0...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>PE</td>\n",
       "      <td>6422.827119</td>\n",
       "      <td>0.0</td>\n",
       "      <td>498</td>\n",
       "      <td>[[4.320936040829623, 2.6790639591703767], [0.6...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>peun</td>\n",
       "      <td>6145.467554</td>\n",
       "      <td>0.0</td>\n",
       "      <td>370</td>\n",
       "      <td>[[10.493701813443371, 6.506298186556629], [10....</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>297</th>\n",
       "      <td>REPQAA</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0</td>\n",
       "      <td>[[11369.0, 7049.0]]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>298</th>\n",
       "      <td>TPSINT_DRAP</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>[[11368.38272342274, 7048.617276577262], [0.61...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>299</th>\n",
       "      <td>DATANT_DRAP</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>[[11363.444510804648, 7045.5554891953525], [5....</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>300</th>\n",
       "      <td>EC</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0</td>\n",
       "      <td>[[11369.0, 7049.0]]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>301</th>\n",
       "      <td>typrepqaa</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0</td>\n",
       "      <td>[[11369.0, 7049.0]]</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>300 rows × 5 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "          Column     Chi2-stat  P-value    dof  \\\n",
       "2       IDENT_FA  11694.082450      1.0  14338   \n",
       "3      IDENT_LOG  11694.082450      1.0  14338   \n",
       "4      IDENT_MEN  11694.082450      1.0  14350   \n",
       "5             PE   6422.827119      0.0    498   \n",
       "6           peun   6145.467554      0.0    370   \n",
       "..           ...           ...      ...    ...   \n",
       "297       REPQAA      0.000000      1.0      0   \n",
       "298  TPSINT_DRAP      0.000000      1.0      1   \n",
       "299  DATANT_DRAP      0.000000      1.0      1   \n",
       "300           EC      0.000000      1.0      0   \n",
       "301    typrepqaa      0.000000      1.0      0   \n",
       "\n",
       "                                              expected  \n",
       "2    [[0.6172765772613747, 0.38272342273862525], [0...  \n",
       "3    [[0.6172765772613747, 0.38272342273862525], [0...  \n",
       "4    [[0.6172765772613747, 0.38272342273862525], [0...  \n",
       "5    [[4.320936040829623, 2.6790639591703767], [0.6...  \n",
       "6    [[10.493701813443371, 6.506298186556629], [10....  \n",
       "..                                                 ...  \n",
       "297                                [[11369.0, 7049.0]]  \n",
       "298  [[11368.38272342274, 7048.617276577262], [0.61...  \n",
       "299  [[11363.444510804648, 7045.5554891953525], [5....  \n",
       "300                                [[11369.0, 7049.0]]  \n",
       "301                                [[11369.0, 7049.0]]  \n",
       "\n",
       "[300 rows x 5 columns]"
      ]
     },
     "execution_count": 138,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "quali_corr"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### We apply a filter on the resulting data frame to keep only relevant correlations (**'P-value'<0.05**). Seeing there still way to many variables we select the 50 first with the strongest chi2_stat"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 139,
   "metadata": {},
   "outputs": [],
   "source": [
    "filtered_quali_corr = quali_corr[(quali_corr[\"P-value\"] < 0.05)][:50]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 140,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Create a new column with the number of categories for each qualitative variable \n",
    "filtered_quali_corr[\"Number of categories\"] = [\n",
    "    len(qualitatives[col].unique()) for col in filtered_quali_corr[\"Column\"]\n",
    "]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 141,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.plotly.v1+json": {
       "config": {
        "plotlyServerURL": "https://plot.ly"
       },
       "data": [
        {
         "alignmentgroup": "True",
         "hovertemplate": "Column=%{x}<br>Chi2-stat=%{text}<br>Number of categories=%{marker.color}<extra></extra>",
         "legendgroup": "",
         "marker": {
          "color": [
           499,
           371,
           225,
           88,
           37,
           22,
           85,
           37,
           7,
           11,
           17,
           3,
           8,
           4,
           4,
           4,
           3,
           4,
           8,
           4,
           6,
           4,
           3,
           4,
           4,
           6,
           3,
           10,
           3,
           4,
           6,
           3,
           2,
           10,
           4,
           3,
           4,
           4,
           3,
           3,
           3,
           4,
           6,
           6,
           3,
           3,
           4,
           3,
           4,
           5
          ],
          "coloraxis": "coloraxis",
          "pattern": {
           "shape": ""
          }
         },
         "name": "",
         "offsetgroup": "",
         "orientation": "v",
         "showlegend": false,
         "text": [
          6422.827118741032,
          6145.467553603463,
          5950.75325791626,
          5408.458133150989,
          4798.898222810621,
          3937.053183641214,
          3276.0270551005747,
          2904.826457036068,
          2372.2742345638335,
          2301.6389908646497,
          2224.0576071383903,
          1800.495419050458,
          1605.9310095565324,
          1482.7836649732249,
          1478.4693671376517,
          1359.3410470192684,
          1348.1732432703088,
          1305.848588202065,
          1053.3968831757327,
          1035.8828463472962,
          997.7755211742822,
          956.517347919132,
          859.0643109174291,
          851.0762206082409,
          846.9236195242117,
          811.6856719835341,
          730.5229321694534,
          665.1250860565086,
          625.4769937702029,
          619.4929861756509,
          616.8214907822997,
          573.023809836529,
          491.8604665874022,
          479.30304962512525,
          441.7126192554777,
          421.1811004276563,
          405.7123168895523,
          401.39472717277687,
          400.7602710192553,
          372.95045367902395,
          372.81031855448697,
          372.61439174358924,
          354.0017395715092,
          348.60260909351825,
          347.46970143727276,
          346.91391906326027,
          338.0058042217061,
          328.20697763858465,
          326.6375660179509,
          314.0947248592143
         ],
         "textposition": "inside",
         "texttemplate": "%{text:.2f}",
         "type": "bar",
         "x": [
          "PE",
          "peun",
          "fap225",
          "fap87",
          "CSE",
          "csei",
          "ACTIVFIN",
          "naf38",
          "cser",
          "peun10",
          "naf17",
          "TPP",
          "CJSITUA",
          "SECROUT",
          "naf4",
          "NBRKM_DRAP",
          "CONDUITE",
          "TINA",
          "STATUT",
          "CWVIB",
          "TYPEXTFP",
          "SECFUPOU",
          "TELEPORT",
          "ENCADR",
          "RISFINA",
          "MISSION",
          "ELIGEXT",
          "TYPMEN15",
          "PTMATIN",
          "HYGCHUMI",
          "SSECH",
          "DETRESSE",
          "CJACTOCCUP",
          "DIPLOME",
          "HYGCOUR",
          "NUIT",
          "HYGSAL",
          "SECACCID",
          "SAMEDI",
          "ASTREINTE",
          "NBASTR_DRAP",
          "NBNUIT_DRAP",
          "A2I",
          "A2E",
          "SOIR",
          "PUBLIC",
          "SECINFEC",
          "DEMPRO",
          "DELAIS",
          "INCIDENT"
         ],
         "xaxis": "x",
         "y": [
          6422.827118741032,
          6145.467553603463,
          5950.75325791626,
          5408.458133150989,
          4798.898222810621,
          3937.053183641214,
          3276.0270551005747,
          2904.826457036068,
          2372.2742345638335,
          2301.6389908646497,
          2224.0576071383903,
          1800.495419050458,
          1605.9310095565324,
          1482.7836649732249,
          1478.4693671376517,
          1359.3410470192684,
          1348.1732432703088,
          1305.848588202065,
          1053.3968831757327,
          1035.8828463472962,
          997.7755211742822,
          956.517347919132,
          859.0643109174291,
          851.0762206082409,
          846.9236195242117,
          811.6856719835341,
          730.5229321694534,
          665.1250860565086,
          625.4769937702029,
          619.4929861756509,
          616.8214907822997,
          573.023809836529,
          491.8604665874022,
          479.30304962512525,
          441.7126192554777,
          421.1811004276563,
          405.7123168895523,
          401.39472717277687,
          400.7602710192553,
          372.95045367902395,
          372.81031855448697,
          372.61439174358924,
          354.0017395715092,
          348.60260909351825,
          347.46970143727276,
          346.91391906326027,
          338.0058042217061,
          328.20697763858465,
          326.6375660179509,
          314.0947248592143
         ],
         "yaxis": "y"
        }
       ],
       "layout": {
        "barmode": "relative",
        "coloraxis": {
         "colorbar": {
          "title": {
           "text": "Number of categories"
          }
         },
         "colorscale": [
          [
           0,
           "rgb(8,48,107)"
          ],
          [
           0.125,
           "rgb(8,81,156)"
          ],
          [
           0.25,
           "rgb(33,113,181)"
          ],
          [
           0.375,
           "rgb(66,146,198)"
          ],
          [
           0.5,
           "rgb(107,174,214)"
          ],
          [
           0.625,
           "rgb(158,202,225)"
          ],
          [
           0.75,
           "rgb(198,219,239)"
          ],
          [
           0.875,
           "rgb(222,235,247)"
          ],
          [
           1,
           "rgb(247,251,255)"
          ]
         ]
        },
        "legend": {
         "tracegroupgap": 0
        },
        "template": {
         "data": {
          "bar": [
           {
            "error_x": {
             "color": "#2a3f5f"
            },
            "error_y": {
             "color": "#2a3f5f"
            },
            "marker": {
             "line": {
              "color": "#E5ECF6",
              "width": 0.5
             },
             "pattern": {
              "fillmode": "overlay",
              "size": 10,
              "solidity": 0.2
             }
            },
            "type": "bar"
           }
          ],
          "barpolar": [
           {
            "marker": {
             "line": {
              "color": "#E5ECF6",
              "width": 0.5
             },
             "pattern": {
              "fillmode": "overlay",
              "size": 10,
              "solidity": 0.2
             }
            },
            "type": "barpolar"
           }
          ],
          "carpet": [
           {
            "aaxis": {
             "endlinecolor": "#2a3f5f",
             "gridcolor": "white",
             "linecolor": "white",
             "minorgridcolor": "white",
             "startlinecolor": "#2a3f5f"
            },
            "baxis": {
             "endlinecolor": "#2a3f5f",
             "gridcolor": "white",
             "linecolor": "white",
             "minorgridcolor": "white",
             "startlinecolor": "#2a3f5f"
            },
            "type": "carpet"
           }
          ],
          "choropleth": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "type": "choropleth"
           }
          ],
          "contour": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "colorscale": [
             [
              0,
              "#0d0887"
             ],
             [
              0.1111111111111111,
              "#46039f"
             ],
             [
              0.2222222222222222,
              "#7201a8"
             ],
             [
              0.3333333333333333,
              "#9c179e"
             ],
             [
              0.4444444444444444,
              "#bd3786"
             ],
             [
              0.5555555555555556,
              "#d8576b"
             ],
             [
              0.6666666666666666,
              "#ed7953"
             ],
             [
              0.7777777777777778,
              "#fb9f3a"
             ],
             [
              0.8888888888888888,
              "#fdca26"
             ],
             [
              1,
              "#f0f921"
             ]
            ],
            "type": "contour"
           }
          ],
          "contourcarpet": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "type": "contourcarpet"
           }
          ],
          "heatmap": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "colorscale": [
             [
              0,
              "#0d0887"
             ],
             [
              0.1111111111111111,
              "#46039f"
             ],
             [
              0.2222222222222222,
              "#7201a8"
             ],
             [
              0.3333333333333333,
              "#9c179e"
             ],
             [
              0.4444444444444444,
              "#bd3786"
             ],
             [
              0.5555555555555556,
              "#d8576b"
             ],
             [
              0.6666666666666666,
              "#ed7953"
             ],
             [
              0.7777777777777778,
              "#fb9f3a"
             ],
             [
              0.8888888888888888,
              "#fdca26"
             ],
             [
              1,
              "#f0f921"
             ]
            ],
            "type": "heatmap"
           }
          ],
          "heatmapgl": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "colorscale": [
             [
              0,
              "#0d0887"
             ],
             [
              0.1111111111111111,
              "#46039f"
             ],
             [
              0.2222222222222222,
              "#7201a8"
             ],
             [
              0.3333333333333333,
              "#9c179e"
             ],
             [
              0.4444444444444444,
              "#bd3786"
             ],
             [
              0.5555555555555556,
              "#d8576b"
             ],
             [
              0.6666666666666666,
              "#ed7953"
             ],
             [
              0.7777777777777778,
              "#fb9f3a"
             ],
             [
              0.8888888888888888,
              "#fdca26"
             ],
             [
              1,
              "#f0f921"
             ]
            ],
            "type": "heatmapgl"
           }
          ],
          "histogram": [
           {
            "marker": {
             "pattern": {
              "fillmode": "overlay",
              "size": 10,
              "solidity": 0.2
             }
            },
            "type": "histogram"
           }
          ],
          "histogram2d": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "colorscale": [
             [
              0,
              "#0d0887"
             ],
             [
              0.1111111111111111,
              "#46039f"
             ],
             [
              0.2222222222222222,
              "#7201a8"
             ],
             [
              0.3333333333333333,
              "#9c179e"
             ],
             [
              0.4444444444444444,
              "#bd3786"
             ],
             [
              0.5555555555555556,
              "#d8576b"
             ],
             [
              0.6666666666666666,
              "#ed7953"
             ],
             [
              0.7777777777777778,
              "#fb9f3a"
             ],
             [
              0.8888888888888888,
              "#fdca26"
             ],
             [
              1,
              "#f0f921"
             ]
            ],
            "type": "histogram2d"
           }
          ],
          "histogram2dcontour": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "colorscale": [
             [
              0,
              "#0d0887"
             ],
             [
              0.1111111111111111,
              "#46039f"
             ],
             [
              0.2222222222222222,
              "#7201a8"
             ],
             [
              0.3333333333333333,
              "#9c179e"
             ],
             [
              0.4444444444444444,
              "#bd3786"
             ],
             [
              0.5555555555555556,
              "#d8576b"
             ],
             [
              0.6666666666666666,
              "#ed7953"
             ],
             [
              0.7777777777777778,
              "#fb9f3a"
             ],
             [
              0.8888888888888888,
              "#fdca26"
             ],
             [
              1,
              "#f0f921"
             ]
            ],
            "type": "histogram2dcontour"
           }
          ],
          "mesh3d": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "type": "mesh3d"
           }
          ],
          "parcoords": [
           {
            "line": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "parcoords"
           }
          ],
          "pie": [
           {
            "automargin": true,
            "type": "pie"
           }
          ],
          "scatter": [
           {
            "fillpattern": {
             "fillmode": "overlay",
             "size": 10,
             "solidity": 0.2
            },
            "type": "scatter"
           }
          ],
          "scatter3d": [
           {
            "line": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "scatter3d"
           }
          ],
          "scattercarpet": [
           {
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "scattercarpet"
           }
          ],
          "scattergeo": [
           {
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "scattergeo"
           }
          ],
          "scattergl": [
           {
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "scattergl"
           }
          ],
          "scattermapbox": [
           {
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "scattermapbox"
           }
          ],
          "scatterpolar": [
           {
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "scatterpolar"
           }
          ],
          "scatterpolargl": [
           {
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "scatterpolargl"
           }
          ],
          "scatterternary": [
           {
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "scatterternary"
           }
          ],
          "surface": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "colorscale": [
             [
              0,
              "#0d0887"
             ],
             [
              0.1111111111111111,
              "#46039f"
             ],
             [
              0.2222222222222222,
              "#7201a8"
             ],
             [
              0.3333333333333333,
              "#9c179e"
             ],
             [
              0.4444444444444444,
              "#bd3786"
             ],
             [
              0.5555555555555556,
              "#d8576b"
             ],
             [
              0.6666666666666666,
              "#ed7953"
             ],
             [
              0.7777777777777778,
              "#fb9f3a"
             ],
             [
              0.8888888888888888,
              "#fdca26"
             ],
             [
              1,
              "#f0f921"
             ]
            ],
            "type": "surface"
           }
          ],
          "table": [
           {
            "cells": {
             "fill": {
              "color": "#EBF0F8"
             },
             "line": {
              "color": "white"
             }
            },
            "header": {
             "fill": {
              "color": "#C8D4E3"
             },
             "line": {
              "color": "white"
             }
            },
            "type": "table"
           }
          ]
         },
         "layout": {
          "annotationdefaults": {
           "arrowcolor": "#2a3f5f",
           "arrowhead": 0,
           "arrowwidth": 1
          },
          "autotypenumbers": "strict",
          "coloraxis": {
           "colorbar": {
            "outlinewidth": 0,
            "ticks": ""
           }
          },
          "colorscale": {
           "diverging": [
            [
             0,
             "#8e0152"
            ],
            [
             0.1,
             "#c51b7d"
            ],
            [
             0.2,
             "#de77ae"
            ],
            [
             0.3,
             "#f1b6da"
            ],
            [
             0.4,
             "#fde0ef"
            ],
            [
             0.5,
             "#f7f7f7"
            ],
            [
             0.6,
             "#e6f5d0"
            ],
            [
             0.7,
             "#b8e186"
            ],
            [
             0.8,
             "#7fbc41"
            ],
            [
             0.9,
             "#4d9221"
            ],
            [
             1,
             "#276419"
            ]
           ],
           "sequential": [
            [
             0,
             "#0d0887"
            ],
            [
             0.1111111111111111,
             "#46039f"
            ],
            [
             0.2222222222222222,
             "#7201a8"
            ],
            [
             0.3333333333333333,
             "#9c179e"
            ],
            [
             0.4444444444444444,
             "#bd3786"
            ],
            [
             0.5555555555555556,
             "#d8576b"
            ],
            [
             0.6666666666666666,
             "#ed7953"
            ],
            [
             0.7777777777777778,
             "#fb9f3a"
            ],
            [
             0.8888888888888888,
             "#fdca26"
            ],
            [
             1,
             "#f0f921"
            ]
           ],
           "sequentialminus": [
            [
             0,
             "#0d0887"
            ],
            [
             0.1111111111111111,
             "#46039f"
            ],
            [
             0.2222222222222222,
             "#7201a8"
            ],
            [
             0.3333333333333333,
             "#9c179e"
            ],
            [
             0.4444444444444444,
             "#bd3786"
            ],
            [
             0.5555555555555556,
             "#d8576b"
            ],
            [
             0.6666666666666666,
             "#ed7953"
            ],
            [
             0.7777777777777778,
             "#fb9f3a"
            ],
            [
             0.8888888888888888,
             "#fdca26"
            ],
            [
             1,
             "#f0f921"
            ]
           ]
          },
          "colorway": [
           "#636efa",
           "#EF553B",
           "#00cc96",
           "#ab63fa",
           "#FFA15A",
           "#19d3f3",
           "#FF6692",
           "#B6E880",
           "#FF97FF",
           "#FECB52"
          ],
          "font": {
           "color": "#2a3f5f"
          },
          "geo": {
           "bgcolor": "white",
           "lakecolor": "white",
           "landcolor": "#E5ECF6",
           "showlakes": true,
           "showland": true,
           "subunitcolor": "white"
          },
          "hoverlabel": {
           "align": "left"
          },
          "hovermode": "closest",
          "mapbox": {
           "style": "light"
          },
          "paper_bgcolor": "white",
          "plot_bgcolor": "#E5ECF6",
          "polar": {
           "angularaxis": {
            "gridcolor": "white",
            "linecolor": "white",
            "ticks": ""
           },
           "bgcolor": "#E5ECF6",
           "radialaxis": {
            "gridcolor": "white",
            "linecolor": "white",
            "ticks": ""
           }
          },
          "scene": {
           "xaxis": {
            "backgroundcolor": "#E5ECF6",
            "gridcolor": "white",
            "gridwidth": 2,
            "linecolor": "white",
            "showbackground": true,
            "ticks": "",
            "zerolinecolor": "white"
           },
           "yaxis": {
            "backgroundcolor": "#E5ECF6",
            "gridcolor": "white",
            "gridwidth": 2,
            "linecolor": "white",
            "showbackground": true,
            "ticks": "",
            "zerolinecolor": "white"
           },
           "zaxis": {
            "backgroundcolor": "#E5ECF6",
            "gridcolor": "white",
            "gridwidth": 2,
            "linecolor": "white",
            "showbackground": true,
            "ticks": "",
            "zerolinecolor": "white"
           }
          },
          "shapedefaults": {
           "line": {
            "color": "#2a3f5f"
           }
          },
          "ternary": {
           "aaxis": {
            "gridcolor": "white",
            "linecolor": "white",
            "ticks": ""
           },
           "baxis": {
            "gridcolor": "white",
            "linecolor": "white",
            "ticks": ""
           },
           "bgcolor": "#E5ECF6",
           "caxis": {
            "gridcolor": "white",
            "linecolor": "white",
            "ticks": ""
           }
          },
          "title": {
           "x": 0.05
          },
          "xaxis": {
           "automargin": true,
           "gridcolor": "white",
           "linecolor": "white",
           "ticks": "",
           "title": {
            "standoff": 15
           },
           "zerolinecolor": "white",
           "zerolinewidth": 2
          },
          "yaxis": {
           "automargin": true,
           "gridcolor": "white",
           "linecolor": "white",
           "ticks": "",
           "title": {
            "standoff": 15
           },
           "zerolinecolor": "white",
           "zerolinewidth": 2
          }
         }
        },
        "title": {
         "text": "Chi2 Correlation of the quantitative variables with the target variable",
         "x": 0.5
        },
        "xaxis": {
         "anchor": "y",
         "domain": [
          0,
          1
         ],
         "title": {
          "text": "Column"
         }
        },
        "yaxis": {
         "anchor": "x",
         "domain": [
          0,
          1
         ],
         "title": {
          "text": "Chi2-stat"
         }
        }
       }
      }
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = px.bar(\n",
    "    filtered_quali_corr,\n",
    "    x=\"Column\",\n",
    "    y=\"Chi2-stat\",\n",
    "    text=\"Chi2-stat\",\n",
    "    color=\"Number of categories\",\n",
    "    color_continuous_scale=px.colors.sequential.Blues_r,\n",
    "    title=\"Chi2 Correlation of the quantitative variables with the target variable\",\n",
    ")\n",
    "fig.update_traces(texttemplate=\"%{text:.2f}\", textposition=\"inside\")\n",
    "fig.update_layout(\n",
    "    title_x=0.5,  # Center the title\n",
    ")\n",
    "fig.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### One Hot Encoding:"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Since we use categorical variables that are label encoded, we one hot encoded them to be able to use Linear Regression models without ordinality problems. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 142,
   "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>PE_0000</th>\n",
       "      <th>PE_0003</th>\n",
       "      <th>PE_111A</th>\n",
       "      <th>PE_111B</th>\n",
       "      <th>PE_111C</th>\n",
       "      <th>PE_111D</th>\n",
       "      <th>PE_111E</th>\n",
       "      <th>PE_111F</th>\n",
       "      <th>PE_121A</th>\n",
       "      <th>PE_121B</th>\n",
       "      <th>...</th>\n",
       "      <th>DELAIS_1</th>\n",
       "      <th>DELAIS_2</th>\n",
       "      <th>DELAIS_3</th>\n",
       "      <th>DELAIS_9</th>\n",
       "      <th>INCIDENT_1</th>\n",
       "      <th>INCIDENT_2</th>\n",
       "      <th>INCIDENT_3</th>\n",
       "      <th>INCIDENT_8</th>\n",
       "      <th>INCIDENT_9</th>\n",
       "      <th>target</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>...</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>...</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>...</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>...</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>...</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>33665</th>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>...</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>33666</th>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>...</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>33667</th>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>...</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>33668</th>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>...</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>33669</th>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>...</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>18418 rows × 1572 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       PE_0000  PE_0003  PE_111A  PE_111B  PE_111C  PE_111D  PE_111E  PE_111F  \\\n",
       "9        False    False    False    False    False    False    False    False   \n",
       "11       False    False    False    False    False    False    False    False   \n",
       "15       False    False     True    False    False    False    False    False   \n",
       "21       False    False    False    False    False    False    False    False   \n",
       "26       False    False    False    False    False    False    False    False   \n",
       "...        ...      ...      ...      ...      ...      ...      ...      ...   \n",
       "33665    False    False    False    False    False    False    False    False   \n",
       "33666    False    False    False    False    False    False    False    False   \n",
       "33667    False    False    False    False    False    False    False    False   \n",
       "33668    False    False    False    False    False    False    False    False   \n",
       "33669    False    False    False    False    False    False    False    False   \n",
       "\n",
       "       PE_121A  PE_121B  ...  DELAIS_1  DELAIS_2  DELAIS_3  DELAIS_9  \\\n",
       "9        False    False  ...     False      True     False     False   \n",
       "11       False    False  ...      True     False     False     False   \n",
       "15       False    False  ...     False     False      True     False   \n",
       "21       False    False  ...     False     False      True     False   \n",
       "26       False    False  ...      True     False     False     False   \n",
       "...        ...      ...  ...       ...       ...       ...       ...   \n",
       "33665    False    False  ...     False     False      True     False   \n",
       "33666    False    False  ...      True     False     False     False   \n",
       "33667    False    False  ...     False     False      True     False   \n",
       "33668    False    False  ...      True     False     False     False   \n",
       "33669    False    False  ...     False      True     False     False   \n",
       "\n",
       "       INCIDENT_1  INCIDENT_2  INCIDENT_3  INCIDENT_8  INCIDENT_9  target  \n",
       "9            True       False       False       False       False    True  \n",
       "11           True       False       False       False       False   False  \n",
       "15           True       False       False       False       False    True  \n",
       "21          False       False        True       False       False    True  \n",
       "26           True       False       False       False       False    True  \n",
       "...           ...         ...         ...         ...         ...     ...  \n",
       "33665        True       False       False       False       False   False  \n",
       "33666        True       False       False       False       False    True  \n",
       "33667        True       False       False       False       False    True  \n",
       "33668        True       False       False       False       False    True  \n",
       "33669       False        True       False       False       False   False  \n",
       "\n",
       "[18418 rows x 1572 columns]"
      ]
     },
     "execution_count": 142,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "one_hot_quali = pd.get_dummies(\n",
    "    qualitatives[filtered_quali_corr[\"Column\"].to_list()],\n",
    "    prefix=filtered_quali_corr[\"Column\"].to_list(),\n",
    ")\n",
    "one_hot_quali[\"target\"] = df[\"target\"]\n",
    "one_hot_quali"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Second Correlation Analysis"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### We perform a second relation analysis similar to the previous one on the values of the selected columns "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 143,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Calculate Correlation of all qualitative variables with target variable\n",
    "oh_qualitative_corr = {\n",
    "    \"Column\": [],\n",
    "    \"Chi2-stat\": [],\n",
    "    \"P-value\": [],\n",
    "    \"dof\": [],\n",
    "    \"expected\": [],\n",
    "}\n",
    "for col in one_hot_quali.columns:\n",
    "    contingency_table = pd.crosstab(one_hot_quali[col], one_hot_quali[\"target\"])\n",
    "    cchi2_stat, p_val, dof, expected = chi2_contingency(contingency_table)\n",
    "    oh_qualitative_corr[\"Column\"].append(col)\n",
    "    oh_qualitative_corr[\"P-value\"].append(p_val)\n",
    "    oh_qualitative_corr[\"dof\"].append(dof)\n",
    "    oh_qualitative_corr[\"expected\"].append(expected)\n",
    "    oh_qualitative_corr[\"Chi2-stat\"].append(cchi2_stat)\n",
    "\n",
    "# Create a DataFrame from the dictionary and sort it by the chi2 statistic and P-value\n",
    "one_hot_quali_corr = (\n",
    "    pd.DataFrame(oh_qualitative_corr)\n",
    "    .sort_values(by=[\"Chi2-stat\", \"P-value\"], ascending=False, key=(abs))\n",
    "    .reset_index()\n",
    "    .drop(0)\n",
    "    .drop(columns=\"index\")[1:]\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 144,
   "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>Column</th>\n",
       "      <th>Chi2-stat</th>\n",
       "      <th>P-value</th>\n",
       "      <th>dof</th>\n",
       "      <th>expected</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>TPP_1</td>\n",
       "      <td>1796.127392</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>1</td>\n",
       "      <td>[[2183.3072537734824, 1353.6927462265176], [91...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>SECROUT_1</td>\n",
       "      <td>1479.867368</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>1</td>\n",
       "      <td>[[7437.565479422305, 4611.434520577695], [3931...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>SECROUT_2</td>\n",
       "      <td>1474.633121</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>1</td>\n",
       "      <td>[[3934.5209034640025, 2439.4790965359975], [74...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>cser_5</td>\n",
       "      <td>1377.926440</td>\n",
       "      <td>1.315491e-301</td>\n",
       "      <td>1</td>\n",
       "      <td>[[8042.496525138452, 4986.503474861548], [3326...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>naf4_EV</td>\n",
       "      <td>1361.400263</td>\n",
       "      <td>5.132368e-298</td>\n",
       "      <td>1</td>\n",
       "      <td>[[2291.9479313714846, 1421.0520686285156], [90...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1567</th>\n",
       "      <td>HYGSAL_8</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>1</td>\n",
       "      <td>[[11368.38272342274, 7048.617276577262], [0.61...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1568</th>\n",
       "      <td>HYGSAL_9</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>1</td>\n",
       "      <td>[[11364.67906395917, 7046.32093604083], [4.320...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1569</th>\n",
       "      <td>SECACCID_8</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>1</td>\n",
       "      <td>[[11367.765446845477, 7048.2345531545225], [1....</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1570</th>\n",
       "      <td>SECINFEC_8</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>1</td>\n",
       "      <td>[[11368.38272342274, 7048.617276577262], [0.61...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1571</th>\n",
       "      <td>DEMPRO_9</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>1</td>\n",
       "      <td>[[11367.765446845477, 7048.2345531545225], [1....</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1570 rows × 5 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "          Column    Chi2-stat        P-value  dof  \\\n",
       "2          TPP_1  1796.127392   0.000000e+00    1   \n",
       "3      SECROUT_1  1479.867368   0.000000e+00    1   \n",
       "4      SECROUT_2  1474.633121   0.000000e+00    1   \n",
       "5         cser_5  1377.926440  1.315491e-301    1   \n",
       "6        naf4_EV  1361.400263  5.132368e-298    1   \n",
       "...          ...          ...            ...  ...   \n",
       "1567    HYGSAL_8     0.000000   1.000000e+00    1   \n",
       "1568    HYGSAL_9     0.000000   1.000000e+00    1   \n",
       "1569  SECACCID_8     0.000000   1.000000e+00    1   \n",
       "1570  SECINFEC_8     0.000000   1.000000e+00    1   \n",
       "1571    DEMPRO_9     0.000000   1.000000e+00    1   \n",
       "\n",
       "                                               expected  \n",
       "2     [[2183.3072537734824, 1353.6927462265176], [91...  \n",
       "3     [[7437.565479422305, 4611.434520577695], [3931...  \n",
       "4     [[3934.5209034640025, 2439.4790965359975], [74...  \n",
       "5     [[8042.496525138452, 4986.503474861548], [3326...  \n",
       "6     [[2291.9479313714846, 1421.0520686285156], [90...  \n",
       "...                                                 ...  \n",
       "1567  [[11368.38272342274, 7048.617276577262], [0.61...  \n",
       "1568  [[11364.67906395917, 7046.32093604083], [4.320...  \n",
       "1569  [[11367.765446845477, 7048.2345531545225], [1....  \n",
       "1570  [[11368.38272342274, 7048.617276577262], [0.61...  \n",
       "1571  [[11367.765446845477, 7048.2345531545225], [1....  \n",
       "\n",
       "[1570 rows x 5 columns]"
      ]
     },
     "execution_count": 144,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "one_hot_quali_corr"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### We apply a filter on the resulting data frame to keep only relevant correlations (**'P-value'<0.05**). Seeing there still way to many variables we select the 20 first with the strongest chi2_stat"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 145,
   "metadata": {},
   "outputs": [],
   "source": [
    "filtered_oh_quali_corr = one_hot_quali_corr[(one_hot_quali_corr[\"P-value\"] < 0.05)][:20]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 146,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.plotly.v1+json": {
       "config": {
        "plotlyServerURL": "https://plot.ly"
       },
       "data": [
        {
         "alignmentgroup": "True",
         "hovertemplate": "Column=%{x}<br>Chi2-stat=%{text}<extra></extra>",
         "legendgroup": "",
         "marker": {
          "color": "rgb(8,48,107)",
          "pattern": {
           "shape": ""
          }
         },
         "name": "",
         "offsetgroup": "",
         "orientation": "v",
         "showlegend": false,
         "text": [
          1796.127392336779,
          1479.8673678742953,
          1474.6331212167966,
          1377.9264397278944,
          1361.400262989445,
          1346.9745928291504,
          1346.2130635183094,
          1346.2130635183094,
          1324.6259460197152,
          1307.8621703866074,
          1297.529037381073,
          1240.8966471068613,
          1183.285872692718,
          1149.3917345648022,
          1032.7803757508939,
          1032.6673717198757,
          954.0119561454222,
          953.4759992210195,
          867.0617872504977,
          867.0617872504977
         ],
         "textposition": "inside",
         "texttemplate": "%{text:.2f}",
         "type": "bar",
         "x": [
          "TPP_1",
          "SECROUT_1",
          "SECROUT_2",
          "cser_5",
          "naf4_EV",
          "CONDUITE_2",
          "NBRKM_DRAP_ 0",
          "CONDUITE_1",
          "naf17_OQ",
          "CJSITUA_6",
          "TINA_3",
          "NBRKM_DRAP_ 1",
          "csei_61",
          "cser_6",
          "CWVIB_2",
          "CWVIB_1",
          "SECFUPOU_2",
          "SECFUPOU_1",
          "ACTIVFIN_86",
          "naf38_QA"
         ],
         "xaxis": "x",
         "y": [
          1796.127392336779,
          1479.8673678742953,
          1474.6331212167966,
          1377.9264397278944,
          1361.400262989445,
          1346.9745928291504,
          1346.2130635183094,
          1346.2130635183094,
          1324.6259460197152,
          1307.8621703866074,
          1297.529037381073,
          1240.8966471068613,
          1183.285872692718,
          1149.3917345648022,
          1032.7803757508939,
          1032.6673717198757,
          954.0119561454222,
          953.4759992210195,
          867.0617872504977,
          867.0617872504977
         ],
         "yaxis": "y"
        }
       ],
       "layout": {
        "barmode": "relative",
        "legend": {
         "tracegroupgap": 0
        },
        "template": {
         "data": {
          "bar": [
           {
            "error_x": {
             "color": "#2a3f5f"
            },
            "error_y": {
             "color": "#2a3f5f"
            },
            "marker": {
             "line": {
              "color": "#E5ECF6",
              "width": 0.5
             },
             "pattern": {
              "fillmode": "overlay",
              "size": 10,
              "solidity": 0.2
             }
            },
            "type": "bar"
           }
          ],
          "barpolar": [
           {
            "marker": {
             "line": {
              "color": "#E5ECF6",
              "width": 0.5
             },
             "pattern": {
              "fillmode": "overlay",
              "size": 10,
              "solidity": 0.2
             }
            },
            "type": "barpolar"
           }
          ],
          "carpet": [
           {
            "aaxis": {
             "endlinecolor": "#2a3f5f",
             "gridcolor": "white",
             "linecolor": "white",
             "minorgridcolor": "white",
             "startlinecolor": "#2a3f5f"
            },
            "baxis": {
             "endlinecolor": "#2a3f5f",
             "gridcolor": "white",
             "linecolor": "white",
             "minorgridcolor": "white",
             "startlinecolor": "#2a3f5f"
            },
            "type": "carpet"
           }
          ],
          "choropleth": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "type": "choropleth"
           }
          ],
          "contour": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "colorscale": [
             [
              0,
              "#0d0887"
             ],
             [
              0.1111111111111111,
              "#46039f"
             ],
             [
              0.2222222222222222,
              "#7201a8"
             ],
             [
              0.3333333333333333,
              "#9c179e"
             ],
             [
              0.4444444444444444,
              "#bd3786"
             ],
             [
              0.5555555555555556,
              "#d8576b"
             ],
             [
              0.6666666666666666,
              "#ed7953"
             ],
             [
              0.7777777777777778,
              "#fb9f3a"
             ],
             [
              0.8888888888888888,
              "#fdca26"
             ],
             [
              1,
              "#f0f921"
             ]
            ],
            "type": "contour"
           }
          ],
          "contourcarpet": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "type": "contourcarpet"
           }
          ],
          "heatmap": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "colorscale": [
             [
              0,
              "#0d0887"
             ],
             [
              0.1111111111111111,
              "#46039f"
             ],
             [
              0.2222222222222222,
              "#7201a8"
             ],
             [
              0.3333333333333333,
              "#9c179e"
             ],
             [
              0.4444444444444444,
              "#bd3786"
             ],
             [
              0.5555555555555556,
              "#d8576b"
             ],
             [
              0.6666666666666666,
              "#ed7953"
             ],
             [
              0.7777777777777778,
              "#fb9f3a"
             ],
             [
              0.8888888888888888,
              "#fdca26"
             ],
             [
              1,
              "#f0f921"
             ]
            ],
            "type": "heatmap"
           }
          ],
          "heatmapgl": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "colorscale": [
             [
              0,
              "#0d0887"
             ],
             [
              0.1111111111111111,
              "#46039f"
             ],
             [
              0.2222222222222222,
              "#7201a8"
             ],
             [
              0.3333333333333333,
              "#9c179e"
             ],
             [
              0.4444444444444444,
              "#bd3786"
             ],
             [
              0.5555555555555556,
              "#d8576b"
             ],
             [
              0.6666666666666666,
              "#ed7953"
             ],
             [
              0.7777777777777778,
              "#fb9f3a"
             ],
             [
              0.8888888888888888,
              "#fdca26"
             ],
             [
              1,
              "#f0f921"
             ]
            ],
            "type": "heatmapgl"
           }
          ],
          "histogram": [
           {
            "marker": {
             "pattern": {
              "fillmode": "overlay",
              "size": 10,
              "solidity": 0.2
             }
            },
            "type": "histogram"
           }
          ],
          "histogram2d": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "colorscale": [
             [
              0,
              "#0d0887"
             ],
             [
              0.1111111111111111,
              "#46039f"
             ],
             [
              0.2222222222222222,
              "#7201a8"
             ],
             [
              0.3333333333333333,
              "#9c179e"
             ],
             [
              0.4444444444444444,
              "#bd3786"
             ],
             [
              0.5555555555555556,
              "#d8576b"
             ],
             [
              0.6666666666666666,
              "#ed7953"
             ],
             [
              0.7777777777777778,
              "#fb9f3a"
             ],
             [
              0.8888888888888888,
              "#fdca26"
             ],
             [
              1,
              "#f0f921"
             ]
            ],
            "type": "histogram2d"
           }
          ],
          "histogram2dcontour": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "colorscale": [
             [
              0,
              "#0d0887"
             ],
             [
              0.1111111111111111,
              "#46039f"
             ],
             [
              0.2222222222222222,
              "#7201a8"
             ],
             [
              0.3333333333333333,
              "#9c179e"
             ],
             [
              0.4444444444444444,
              "#bd3786"
             ],
             [
              0.5555555555555556,
              "#d8576b"
             ],
             [
              0.6666666666666666,
              "#ed7953"
             ],
             [
              0.7777777777777778,
              "#fb9f3a"
             ],
             [
              0.8888888888888888,
              "#fdca26"
             ],
             [
              1,
              "#f0f921"
             ]
            ],
            "type": "histogram2dcontour"
           }
          ],
          "mesh3d": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "type": "mesh3d"
           }
          ],
          "parcoords": [
           {
            "line": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "parcoords"
           }
          ],
          "pie": [
           {
            "automargin": true,
            "type": "pie"
           }
          ],
          "scatter": [
           {
            "fillpattern": {
             "fillmode": "overlay",
             "size": 10,
             "solidity": 0.2
            },
            "type": "scatter"
           }
          ],
          "scatter3d": [
           {
            "line": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "scatter3d"
           }
          ],
          "scattercarpet": [
           {
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "scattercarpet"
           }
          ],
          "scattergeo": [
           {
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "scattergeo"
           }
          ],
          "scattergl": [
           {
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "scattergl"
           }
          ],
          "scattermapbox": [
           {
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "scattermapbox"
           }
          ],
          "scatterpolar": [
           {
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "scatterpolar"
           }
          ],
          "scatterpolargl": [
           {
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "scatterpolargl"
           }
          ],
          "scatterternary": [
           {
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "type": "scatterternary"
           }
          ],
          "surface": [
           {
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "colorscale": [
             [
              0,
              "#0d0887"
             ],
             [
              0.1111111111111111,
              "#46039f"
             ],
             [
              0.2222222222222222,
              "#7201a8"
             ],
             [
              0.3333333333333333,
              "#9c179e"
             ],
             [
              0.4444444444444444,
              "#bd3786"
             ],
             [
              0.5555555555555556,
              "#d8576b"
             ],
             [
              0.6666666666666666,
              "#ed7953"
             ],
             [
              0.7777777777777778,
              "#fb9f3a"
             ],
             [
              0.8888888888888888,
              "#fdca26"
             ],
             [
              1,
              "#f0f921"
             ]
            ],
            "type": "surface"
           }
          ],
          "table": [
           {
            "cells": {
             "fill": {
              "color": "#EBF0F8"
             },
             "line": {
              "color": "white"
             }
            },
            "header": {
             "fill": {
              "color": "#C8D4E3"
             },
             "line": {
              "color": "white"
             }
            },
            "type": "table"
           }
          ]
         },
         "layout": {
          "annotationdefaults": {
           "arrowcolor": "#2a3f5f",
           "arrowhead": 0,
           "arrowwidth": 1
          },
          "autotypenumbers": "strict",
          "coloraxis": {
           "colorbar": {
            "outlinewidth": 0,
            "ticks": ""
           }
          },
          "colorscale": {
           "diverging": [
            [
             0,
             "#8e0152"
            ],
            [
             0.1,
             "#c51b7d"
            ],
            [
             0.2,
             "#de77ae"
            ],
            [
             0.3,
             "#f1b6da"
            ],
            [
             0.4,
             "#fde0ef"
            ],
            [
             0.5,
             "#f7f7f7"
            ],
            [
             0.6,
             "#e6f5d0"
            ],
            [
             0.7,
             "#b8e186"
            ],
            [
             0.8,
             "#7fbc41"
            ],
            [
             0.9,
             "#4d9221"
            ],
            [
             1,
             "#276419"
            ]
           ],
           "sequential": [
            [
             0,
             "#0d0887"
            ],
            [
             0.1111111111111111,
             "#46039f"
            ],
            [
             0.2222222222222222,
             "#7201a8"
            ],
            [
             0.3333333333333333,
             "#9c179e"
            ],
            [
             0.4444444444444444,
             "#bd3786"
            ],
            [
             0.5555555555555556,
             "#d8576b"
            ],
            [
             0.6666666666666666,
             "#ed7953"
            ],
            [
             0.7777777777777778,
             "#fb9f3a"
            ],
            [
             0.8888888888888888,
             "#fdca26"
            ],
            [
             1,
             "#f0f921"
            ]
           ],
           "sequentialminus": [
            [
             0,
             "#0d0887"
            ],
            [
             0.1111111111111111,
             "#46039f"
            ],
            [
             0.2222222222222222,
             "#7201a8"
            ],
            [
             0.3333333333333333,
             "#9c179e"
            ],
            [
             0.4444444444444444,
             "#bd3786"
            ],
            [
             0.5555555555555556,
             "#d8576b"
            ],
            [
             0.6666666666666666,
             "#ed7953"
            ],
            [
             0.7777777777777778,
             "#fb9f3a"
            ],
            [
             0.8888888888888888,
             "#fdca26"
            ],
            [
             1,
             "#f0f921"
            ]
           ]
          },
          "colorway": [
           "#636efa",
           "#EF553B",
           "#00cc96",
           "#ab63fa",
           "#FFA15A",
           "#19d3f3",
           "#FF6692",
           "#B6E880",
           "#FF97FF",
           "#FECB52"
          ],
          "font": {
           "color": "#2a3f5f"
          },
          "geo": {
           "bgcolor": "white",
           "lakecolor": "white",
           "landcolor": "#E5ECF6",
           "showlakes": true,
           "showland": true,
           "subunitcolor": "white"
          },
          "hoverlabel": {
           "align": "left"
          },
          "hovermode": "closest",
          "mapbox": {
           "style": "light"
          },
          "paper_bgcolor": "white",
          "plot_bgcolor": "#E5ECF6",
          "polar": {
           "angularaxis": {
            "gridcolor": "white",
            "linecolor": "white",
            "ticks": ""
           },
           "bgcolor": "#E5ECF6",
           "radialaxis": {
            "gridcolor": "white",
            "linecolor": "white",
            "ticks": ""
           }
          },
          "scene": {
           "xaxis": {
            "backgroundcolor": "#E5ECF6",
            "gridcolor": "white",
            "gridwidth": 2,
            "linecolor": "white",
            "showbackground": true,
            "ticks": "",
            "zerolinecolor": "white"
           },
           "yaxis": {
            "backgroundcolor": "#E5ECF6",
            "gridcolor": "white",
            "gridwidth": 2,
            "linecolor": "white",
            "showbackground": true,
            "ticks": "",
            "zerolinecolor": "white"
           },
           "zaxis": {
            "backgroundcolor": "#E5ECF6",
            "gridcolor": "white",
            "gridwidth": 2,
            "linecolor": "white",
            "showbackground": true,
            "ticks": "",
            "zerolinecolor": "white"
           }
          },
          "shapedefaults": {
           "line": {
            "color": "#2a3f5f"
           }
          },
          "ternary": {
           "aaxis": {
            "gridcolor": "white",
            "linecolor": "white",
            "ticks": ""
           },
           "baxis": {
            "gridcolor": "white",
            "linecolor": "white",
            "ticks": ""
           },
           "bgcolor": "#E5ECF6",
           "caxis": {
            "gridcolor": "white",
            "linecolor": "white",
            "ticks": ""
           }
          },
          "title": {
           "x": 0.05
          },
          "xaxis": {
           "automargin": true,
           "gridcolor": "white",
           "linecolor": "white",
           "ticks": "",
           "title": {
            "standoff": 15
           },
           "zerolinecolor": "white",
           "zerolinewidth": 2
          },
          "yaxis": {
           "automargin": true,
           "gridcolor": "white",
           "linecolor": "white",
           "ticks": "",
           "title": {
            "standoff": 15
           },
           "zerolinecolor": "white",
           "zerolinewidth": 2
          }
         }
        },
        "title": {
         "text": "Chi2 Correlation of the quantitative variables with the target variable",
         "x": 0.5
        },
        "xaxis": {
         "anchor": "y",
         "domain": [
          0,
          1
         ],
         "title": {
          "text": "Column"
         }
        },
        "yaxis": {
         "anchor": "x",
         "domain": [
          0,
          1
         ],
         "title": {
          "text": "Chi2-stat"
         }
        }
       }
      }
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = px.bar(\n",
    "    filtered_oh_quali_corr,\n",
    "    x=\"Column\",\n",
    "    y=\"Chi2-stat\",\n",
    "    text=\"Chi2-stat\",\n",
    "    color_discrete_sequence=px.colors.sequential.Blues_r,\n",
    "    title=\"Chi2 Correlation of the quantitative variables with the target variable\",\n",
    ")\n",
    "fig.update_traces(texttemplate=\"%{text:.2f}\", textposition=\"inside\")\n",
    "fig.update_layout(\n",
    "    title_x=0.5,  # Center the title\n",
    ")\n",
    "fig.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 3. Final Data Preparation"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### First we create a dataframe with the selected qualitative and quantitative variables"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 147,
   "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>PE</th>\n",
       "      <th>peun</th>\n",
       "      <th>fap225</th>\n",
       "      <th>fap87</th>\n",
       "      <th>CSE</th>\n",
       "      <th>csei</th>\n",
       "      <th>ACTIVFIN</th>\n",
       "      <th>naf38</th>\n",
       "      <th>cser</th>\n",
       "      <th>peun10</th>\n",
       "      <th>...</th>\n",
       "      <th>SOIR</th>\n",
       "      <th>PUBLIC</th>\n",
       "      <th>SECINFEC</th>\n",
       "      <th>DEMPRO</th>\n",
       "      <th>DELAIS</th>\n",
       "      <th>INCIDENT</th>\n",
       "      <th>HH</th>\n",
       "      <th>KMANNU</th>\n",
       "      <th>REVMENS</th>\n",
       "      <th>target</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>111F</td>\n",
       "      <td>6113</td>\n",
       "      <td>A0Z00</td>\n",
       "      <td>A0Z</td>\n",
       "      <td>11</td>\n",
       "      <td>10</td>\n",
       "      <td>01</td>\n",
       "      <td>AZ</td>\n",
       "      <td>1</td>\n",
       "      <td>6</td>\n",
       "      <td>...</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>66.0</td>\n",
       "      <td>3666.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>111F</td>\n",
       "      <td>6114</td>\n",
       "      <td>A0Z00</td>\n",
       "      <td>A0Z</td>\n",
       "      <td>11</td>\n",
       "      <td>10</td>\n",
       "      <td>NaN</td>\n",
       "      <td>00</td>\n",
       "      <td>1</td>\n",
       "      <td>6</td>\n",
       "      <td>...</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>40.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>111F</td>\n",
       "      <td>6114</td>\n",
       "      <td>A0Z00</td>\n",
       "      <td>A0Z</td>\n",
       "      <td>11</td>\n",
       "      <td>10</td>\n",
       "      <td>01</td>\n",
       "      <td>AZ</td>\n",
       "      <td>1</td>\n",
       "      <td>6</td>\n",
       "      <td>...</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>40.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>111A</td>\n",
       "      <td>6111</td>\n",
       "      <td>A0Z00</td>\n",
       "      <td>A0Z</td>\n",
       "      <td>11</td>\n",
       "      <td>10</td>\n",
       "      <td>01</td>\n",
       "      <td>AZ</td>\n",
       "      <td>1</td>\n",
       "      <td>6</td>\n",
       "      <td>...</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>40.0</td>\n",
       "      <td>1410.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>111A</td>\n",
       "      <td>6111</td>\n",
       "      <td>A0Z00</td>\n",
       "      <td>A0Z</td>\n",
       "      <td>11</td>\n",
       "      <td>10</td>\n",
       "      <td>01</td>\n",
       "      <td>AZ</td>\n",
       "      <td>1</td>\n",
       "      <td>6</td>\n",
       "      <td>...</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>70.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>33667</th>\n",
       "      <td>5300</td>\n",
       "      <td>9</td>\n",
       "      <td>ZZZZZ</td>\n",
       "      <td>ZZZ</td>\n",
       "      <td>53</td>\n",
       "      <td>51</td>\n",
       "      <td>84</td>\n",
       "      <td>OZ</td>\n",
       "      <td>5</td>\n",
       "      <td>9</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>36.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1400.0</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>33668</th>\n",
       "      <td>4000</td>\n",
       "      <td>3115</td>\n",
       "      <td>ZZZZZ</td>\n",
       "      <td>ZZZ</td>\n",
       "      <td>40</td>\n",
       "      <td>40</td>\n",
       "      <td>35</td>\n",
       "      <td>DZ</td>\n",
       "      <td>4</td>\n",
       "      <td>3</td>\n",
       "      <td>...</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>35.0</td>\n",
       "      <td>30000.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>33669</th>\n",
       "      <td>4600</td>\n",
       "      <td>3</td>\n",
       "      <td>ZZZZZ</td>\n",
       "      <td>ZZZ</td>\n",
       "      <td>46</td>\n",
       "      <td>46</td>\n",
       "      <td>78</td>\n",
       "      <td>NZ</td>\n",
       "      <td>4</td>\n",
       "      <td>3</td>\n",
       "      <td>...</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>39.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>33671</th>\n",
       "      <td>2100</td>\n",
       "      <td>7549</td>\n",
       "      <td>ZZZZZ</td>\n",
       "      <td>ZZZ</td>\n",
       "      <td>21</td>\n",
       "      <td>21</td>\n",
       "      <td>43</td>\n",
       "      <td>FZ</td>\n",
       "      <td>2</td>\n",
       "      <td>7</td>\n",
       "      <td>...</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>38.0</td>\n",
       "      <td>11000.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>33672</th>\n",
       "      <td>0003</td>\n",
       "      <td>NNNN</td>\n",
       "      <td>ZZZZZ</td>\n",
       "      <td>ZZZ</td>\n",
       "      <td>00</td>\n",
       "      <td>00</td>\n",
       "      <td>41</td>\n",
       "      <td>FZ</td>\n",
       "      <td>0</td>\n",
       "      <td>N</td>\n",
       "      <td>...</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>8.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>24024 rows × 54 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "         PE  peun fap225 fap87 CSE csei ACTIVFIN naf38 cser peun10  ... SOIR  \\\n",
       "2      111F  6113  A0Z00   A0Z  11   10       01    AZ    1      6  ...    2   \n",
       "3      111F  6114  A0Z00   A0Z  11   10      NaN    00    1      6  ...    3   \n",
       "5      111F  6114  A0Z00   A0Z  11   10       01    AZ    1      6  ...    3   \n",
       "6      111A  6111  A0Z00   A0Z  11   10       01    AZ    1      6  ...    3   \n",
       "7      111A  6111  A0Z00   A0Z  11   10       01    AZ    1      6  ...    3   \n",
       "...     ...   ...    ...   ...  ..  ...      ...   ...  ...    ...  ...  ...   \n",
       "33667  5300     9  ZZZZZ   ZZZ  53   51       84    OZ    5      9  ...    1   \n",
       "33668  4000  3115  ZZZZZ   ZZZ  40   40       35    DZ    4      3  ...    3   \n",
       "33669  4600     3  ZZZZZ   ZZZ  46   46       78    NZ    4      3  ...    3   \n",
       "33671  2100  7549  ZZZZZ   ZZZ  21   21       43    FZ    2      7  ...    3   \n",
       "33672  0003  NNNN  ZZZZZ   ZZZ  00   00       41    FZ    0      N  ...    3   \n",
       "\n",
       "      PUBLIC SECINFEC DEMPRO DELAIS INCIDENT    HH   KMANNU REVMENS target  \n",
       "2          1        1      2      2        3  66.0   3666.0     NaN  False  \n",
       "3          1        1      2      3        1  40.0      NaN     NaN   True  \n",
       "5          1        2      1      3        1  40.0      NaN     NaN   True  \n",
       "6          2        1      2      1        1  40.0   1410.0     NaN  False  \n",
       "7          2        2      2      2        1  70.0      NaN     NaN   True  \n",
       "...      ...      ...    ...    ...      ...   ...      ...     ...    ...  \n",
       "33667      1        2      1      3        1  36.0      NaN  1400.0   True  \n",
       "33668      2        2      2      1        1  35.0  30000.0     NaN   True  \n",
       "33669      2        2      2      2        2  39.0      NaN     NaN  False  \n",
       "33671      1        1      2      1        1  38.0  11000.0     NaN   True  \n",
       "33672      2        2      2      2        1   8.0      NaN     NaN  False  \n",
       "\n",
       "[24024 rows x 54 columns]"
      ]
     },
     "execution_count": 147,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_filtered = df[\n",
    "    filtered_quali_corr[\"Column\"].to_list()\n",
    "    + quant_corr_filtered[\"Column\"].to_list()\n",
    "    + [\"target\"]\n",
    "]\n",
    "df_filtered"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 148,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "23766"
      ]
     },
     "execution_count": 148,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_filtered.isna().sum().sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### We fill the numerical NaNs with a median"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 149,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/tmp/ipykernel_160722/1181171266.py:2: SettingWithCopyWarning:\n",
      "\n",
      "\n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# Select the quantitative variables and fill the missing values with the median\n",
    "df_filtered[df_filtered.select_dtypes([\"int\", \"float\"]).columns] = ( \n",
    "    df_filtered.select_dtypes([\"int\", \"float\"]).fillna(\n",
    "        df_filtered.select_dtypes([\"int\", \"float\"]).median()\n",
    "    )\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 150,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "3209"
      ]
     },
     "execution_count": 150,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_filtered.isna().sum().sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### We find a small number of missing values in the selected qualitative variables"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 151,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "A2I            1296\n",
       "A2E            1296\n",
       "TPP             187\n",
       "DEMPRO          133\n",
       "TINA            131\n",
       "ACTIVFIN        123\n",
       "peun10           13\n",
       "peun             13\n",
       "CJSITUA           4\n",
       "PE                3\n",
       "CSE               3\n",
       "cser              3\n",
       "SOIR              1\n",
       "NUIT              1\n",
       "SAMEDI            1\n",
       "PTMATIN           1\n",
       "NBNUIT_DRAP       0\n",
       "NBASTR_DRAP       0\n",
       "SSECH             0\n",
       "ASTREINTE         0\n",
       "SECINFEC          0\n",
       "SECACCID          0\n",
       "HYGSAL            0\n",
       "PUBLIC            0\n",
       "HYGCOUR           0\n",
       "DIPLOME           0\n",
       "DELAIS            0\n",
       "CJACTOCCUP        0\n",
       "DETRESSE          0\n",
       "MISSION           0\n",
       "HYGCHUMI          0\n",
       "TYPMEN15          0\n",
       "fap225            0\n",
       "fap87             0\n",
       "csei              0\n",
       "naf38             0\n",
       "naf17             0\n",
       "SECROUT           0\n",
       "naf4              0\n",
       "NBRKM_DRAP        0\n",
       "CONDUITE          0\n",
       "STATUT            0\n",
       "CWVIB             0\n",
       "TYPEXTFP          0\n",
       "SECFUPOU          0\n",
       "TELEPORT          0\n",
       "ENCADR            0\n",
       "RISFINA           0\n",
       "ELIGEXT           0\n",
       "INCIDENT          0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 151,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_filtered.select_dtypes(include=[\"object\"]).isna().sum().sort_values(ascending=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 152,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/tmp/ipykernel_160722/571476937.py:1: SettingWithCopyWarning:\n",
      "\n",
      "\n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "\n"
     ]
    }
   ],
   "source": [
    "df_filtered.dropna(inplace=True) # Drop the rows with missing values"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Transform the qualitative variables with a one hot encoding"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 153,
   "metadata": {},
   "outputs": [],
   "source": [
    "one_hot_df = pd.get_dummies(\n",
    "    df_filtered, prefix=df_filtered.select_dtypes(include=[\"object\"]).columns\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 154,
   "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>HH</th>\n",
       "      <th>KMANNU</th>\n",
       "      <th>REVMENS</th>\n",
       "      <th>target</th>\n",
       "      <th>PE_0000</th>\n",
       "      <th>PE_0003</th>\n",
       "      <th>PE_111A</th>\n",
       "      <th>PE_111B</th>\n",
       "      <th>PE_111C</th>\n",
       "      <th>PE_111D</th>\n",
       "      <th>...</th>\n",
       "      <th>DELAIS_1</th>\n",
       "      <th>DELAIS_2</th>\n",
       "      <th>DELAIS_3</th>\n",
       "      <th>DELAIS_8</th>\n",
       "      <th>DELAIS_9</th>\n",
       "      <th>INCIDENT_1</th>\n",
       "      <th>INCIDENT_2</th>\n",
       "      <th>INCIDENT_3</th>\n",
       "      <th>INCIDENT_8</th>\n",
       "      <th>INCIDENT_9</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>66.0</td>\n",
       "      <td>3666.0</td>\n",
       "      <td>1644.0</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>...</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>40.0</td>\n",
       "      <td>1410.0</td>\n",
       "      <td>1644.0</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>...</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>35.0</td>\n",
       "      <td>5500.0</td>\n",
       "      <td>1200.0</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>...</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>5.0</td>\n",
       "      <td>5500.0</td>\n",
       "      <td>1644.0</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>...</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>26.0</td>\n",
       "      <td>9400.0</td>\n",
       "      <td>1644.0</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>...</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>33667</th>\n",
       "      <td>36.0</td>\n",
       "      <td>5500.0</td>\n",
       "      <td>1400.0</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>...</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>33668</th>\n",
       "      <td>35.0</td>\n",
       "      <td>30000.0</td>\n",
       "      <td>1644.0</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>...</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>33669</th>\n",
       "      <td>39.0</td>\n",
       "      <td>5500.0</td>\n",
       "      <td>1644.0</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>...</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>33671</th>\n",
       "      <td>38.0</td>\n",
       "      <td>11000.0</td>\n",
       "      <td>1644.0</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>...</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>33672</th>\n",
       "      <td>8.0</td>\n",
       "      <td>5500.0</td>\n",
       "      <td>1644.0</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>...</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>22306 rows × 1608 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "         HH   KMANNU  REVMENS  target  PE_0000  PE_0003  PE_111A  PE_111B  \\\n",
       "2      66.0   3666.0   1644.0   False    False    False    False    False   \n",
       "6      40.0   1410.0   1644.0   False    False    False     True    False   \n",
       "9      35.0   5500.0   1200.0    True    False    False    False    False   \n",
       "11      5.0   5500.0   1644.0   False    False    False    False    False   \n",
       "15     26.0   9400.0   1644.0    True    False    False     True    False   \n",
       "...     ...      ...      ...     ...      ...      ...      ...      ...   \n",
       "33667  36.0   5500.0   1400.0    True    False    False    False    False   \n",
       "33668  35.0  30000.0   1644.0    True    False    False    False    False   \n",
       "33669  39.0   5500.0   1644.0   False    False    False    False    False   \n",
       "33671  38.0  11000.0   1644.0    True    False    False    False    False   \n",
       "33672   8.0   5500.0   1644.0   False    False     True    False    False   \n",
       "\n",
       "       PE_111C  PE_111D  ...  DELAIS_1  DELAIS_2  DELAIS_3  DELAIS_8  \\\n",
       "2        False    False  ...     False      True     False     False   \n",
       "6        False    False  ...      True     False     False     False   \n",
       "9        False    False  ...     False      True     False     False   \n",
       "11       False    False  ...      True     False     False     False   \n",
       "15       False    False  ...     False     False      True     False   \n",
       "...        ...      ...  ...       ...       ...       ...       ...   \n",
       "33667    False    False  ...     False     False      True     False   \n",
       "33668    False    False  ...      True     False     False     False   \n",
       "33669    False    False  ...     False      True     False     False   \n",
       "33671    False    False  ...      True     False     False     False   \n",
       "33672    False    False  ...     False      True     False     False   \n",
       "\n",
       "       DELAIS_9  INCIDENT_1  INCIDENT_2  INCIDENT_3  INCIDENT_8  INCIDENT_9  \n",
       "2         False       False       False        True       False       False  \n",
       "6         False        True       False       False       False       False  \n",
       "9         False        True       False       False       False       False  \n",
       "11        False        True       False       False       False       False  \n",
       "15        False        True       False       False       False       False  \n",
       "...         ...         ...         ...         ...         ...         ...  \n",
       "33667     False        True       False       False       False       False  \n",
       "33668     False        True       False       False       False       False  \n",
       "33669     False       False        True       False       False       False  \n",
       "33671     False        True       False       False       False       False  \n",
       "33672     False        True       False       False       False       False  \n",
       "\n",
       "[22306 rows x 1608 columns]"
      ]
     },
     "execution_count": 154,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "one_hot_df"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Select the quantitative variables and the previously selected qualitative value variables from the one hot encoded dataframe "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 155,
   "metadata": {},
   "outputs": [],
   "source": [
    "one_hot_df = one_hot_df[\n",
    "    filtered_oh_quali_corr[\"Column\"].to_list()\n",
    "    + quant_corr_filtered[\"Column\"].to_list()\n",
    "    + [\"target\"]\n",
    "]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 4. Statistical modeling"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### A. Multicollinearity tests"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### We analyze the selected variables for collinearity to eliminate highly correlated or associated variables that would badly influence our model. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 156,
   "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>TPP_1</th>\n",
       "      <th>SECROUT_1</th>\n",
       "      <th>SECROUT_2</th>\n",
       "      <th>cser_5</th>\n",
       "      <th>naf4_EV</th>\n",
       "      <th>CONDUITE_2</th>\n",
       "      <th>NBRKM_DRAP_ 0</th>\n",
       "      <th>CONDUITE_1</th>\n",
       "      <th>naf17_OQ</th>\n",
       "      <th>CJSITUA_6</th>\n",
       "      <th>...</th>\n",
       "      <th>CWVIB_2</th>\n",
       "      <th>CWVIB_1</th>\n",
       "      <th>SECFUPOU_2</th>\n",
       "      <th>SECFUPOU_1</th>\n",
       "      <th>ACTIVFIN_86</th>\n",
       "      <th>naf38_QA</th>\n",
       "      <th>HH</th>\n",
       "      <th>KMANNU</th>\n",
       "      <th>REVMENS</th>\n",
       "      <th>target</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>TPP_1</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.125718</td>\n",
       "      <td>-0.125640</td>\n",
       "      <td>-0.167049</td>\n",
       "      <td>-0.143424</td>\n",
       "      <td>-0.127877</td>\n",
       "      <td>-0.128077</td>\n",
       "      <td>0.128077</td>\n",
       "      <td>-0.088461</td>\n",
       "      <td>0.071407</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.122111</td>\n",
       "      <td>0.121923</td>\n",
       "      <td>-0.103316</td>\n",
       "      <td>0.103048</td>\n",
       "      <td>-0.056000</td>\n",
       "      <td>-0.056000</td>\n",
       "      <td>0.529593</td>\n",
       "      <td>0.100230</td>\n",
       "      <td>0.123532</td>\n",
       "      <td>0.308664</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SECROUT_1</th>\n",
       "      <td>0.125718</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.999123</td>\n",
       "      <td>-0.144906</td>\n",
       "      <td>-0.138219</td>\n",
       "      <td>-0.661175</td>\n",
       "      <td>-0.661096</td>\n",
       "      <td>0.661096</td>\n",
       "      <td>-0.119002</td>\n",
       "      <td>0.062976</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.316881</td>\n",
       "      <td>0.317121</td>\n",
       "      <td>-0.238399</td>\n",
       "      <td>0.238491</td>\n",
       "      <td>-0.136021</td>\n",
       "      <td>-0.136021</td>\n",
       "      <td>0.204578</td>\n",
       "      <td>0.303375</td>\n",
       "      <td>0.031349</td>\n",
       "      <td>0.283972</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SECROUT_2</th>\n",
       "      <td>-0.125640</td>\n",
       "      <td>-0.999123</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.144802</td>\n",
       "      <td>0.137748</td>\n",
       "      <td>0.661684</td>\n",
       "      <td>0.661605</td>\n",
       "      <td>-0.661605</td>\n",
       "      <td>0.118607</td>\n",
       "      <td>-0.062752</td>\n",
       "      <td>...</td>\n",
       "      <td>0.316669</td>\n",
       "      <td>-0.316661</td>\n",
       "      <td>0.238813</td>\n",
       "      <td>-0.238701</td>\n",
       "      <td>0.135862</td>\n",
       "      <td>0.135862</td>\n",
       "      <td>-0.204726</td>\n",
       "      <td>-0.303152</td>\n",
       "      <td>-0.031431</td>\n",
       "      <td>-0.283617</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>cser_5</th>\n",
       "      <td>-0.167049</td>\n",
       "      <td>-0.144906</td>\n",
       "      <td>0.144802</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.256628</td>\n",
       "      <td>0.181942</td>\n",
       "      <td>0.181808</td>\n",
       "      <td>-0.181808</td>\n",
       "      <td>0.184649</td>\n",
       "      <td>-0.089359</td>\n",
       "      <td>...</td>\n",
       "      <td>0.114477</td>\n",
       "      <td>-0.114215</td>\n",
       "      <td>0.100931</td>\n",
       "      <td>-0.100576</td>\n",
       "      <td>0.085229</td>\n",
       "      <td>0.085229</td>\n",
       "      <td>-0.190228</td>\n",
       "      <td>-0.124125</td>\n",
       "      <td>-0.112451</td>\n",
       "      <td>-0.294126</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>naf4_EV</th>\n",
       "      <td>-0.143424</td>\n",
       "      <td>-0.138219</td>\n",
       "      <td>0.137748</td>\n",
       "      <td>0.256628</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.103048</td>\n",
       "      <td>0.102672</td>\n",
       "      <td>-0.102672</td>\n",
       "      <td>0.470077</td>\n",
       "      <td>-0.108275</td>\n",
       "      <td>...</td>\n",
       "      <td>0.239039</td>\n",
       "      <td>-0.238803</td>\n",
       "      <td>0.313607</td>\n",
       "      <td>-0.313275</td>\n",
       "      <td>0.232267</td>\n",
       "      <td>0.232267</td>\n",
       "      <td>-0.142649</td>\n",
       "      <td>-0.079932</td>\n",
       "      <td>-0.004528</td>\n",
       "      <td>-0.276469</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CONDUITE_2</th>\n",
       "      <td>-0.127877</td>\n",
       "      <td>-0.661175</td>\n",
       "      <td>0.661684</td>\n",
       "      <td>0.181942</td>\n",
       "      <td>0.103048</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.999434</td>\n",
       "      <td>-0.999434</td>\n",
       "      <td>0.105614</td>\n",
       "      <td>-0.054224</td>\n",
       "      <td>...</td>\n",
       "      <td>0.227721</td>\n",
       "      <td>-0.227952</td>\n",
       "      <td>0.136080</td>\n",
       "      <td>-0.136181</td>\n",
       "      <td>0.142725</td>\n",
       "      <td>0.142725</td>\n",
       "      <td>-0.244954</td>\n",
       "      <td>-0.282261</td>\n",
       "      <td>-0.055904</td>\n",
       "      <td>-0.269212</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>NBRKM_DRAP_ 0</th>\n",
       "      <td>-0.128077</td>\n",
       "      <td>-0.661096</td>\n",
       "      <td>0.661605</td>\n",
       "      <td>0.181808</td>\n",
       "      <td>0.102672</td>\n",
       "      <td>0.999434</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>0.105128</td>\n",
       "      <td>-0.053581</td>\n",
       "      <td>...</td>\n",
       "      <td>0.227756</td>\n",
       "      <td>-0.227986</td>\n",
       "      <td>0.135854</td>\n",
       "      <td>-0.136156</td>\n",
       "      <td>0.142496</td>\n",
       "      <td>0.142496</td>\n",
       "      <td>-0.245055</td>\n",
       "      <td>-0.282421</td>\n",
       "      <td>-0.055949</td>\n",
       "      <td>-0.268930</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CONDUITE_1</th>\n",
       "      <td>0.128077</td>\n",
       "      <td>0.661096</td>\n",
       "      <td>-0.661605</td>\n",
       "      <td>-0.181808</td>\n",
       "      <td>-0.102672</td>\n",
       "      <td>-0.999434</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.105128</td>\n",
       "      <td>0.053581</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.227756</td>\n",
       "      <td>0.227986</td>\n",
       "      <td>-0.135854</td>\n",
       "      <td>0.136156</td>\n",
       "      <td>-0.142496</td>\n",
       "      <td>-0.142496</td>\n",
       "      <td>0.245055</td>\n",
       "      <td>0.282421</td>\n",
       "      <td>0.055949</td>\n",
       "      <td>0.268930</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>naf17_OQ</th>\n",
       "      <td>-0.088461</td>\n",
       "      <td>-0.119002</td>\n",
       "      <td>0.118607</td>\n",
       "      <td>0.184649</td>\n",
       "      <td>0.470077</td>\n",
       "      <td>0.105614</td>\n",
       "      <td>0.105128</td>\n",
       "      <td>-0.105128</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.103762</td>\n",
       "      <td>...</td>\n",
       "      <td>0.154381</td>\n",
       "      <td>-0.154493</td>\n",
       "      <td>0.199522</td>\n",
       "      <td>-0.199632</td>\n",
       "      <td>0.494103</td>\n",
       "      <td>0.494103</td>\n",
       "      <td>-0.142054</td>\n",
       "      <td>-0.162580</td>\n",
       "      <td>0.027168</td>\n",
       "      <td>-0.250588</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CJSITUA_6</th>\n",
       "      <td>0.071407</td>\n",
       "      <td>0.062976</td>\n",
       "      <td>-0.062752</td>\n",
       "      <td>-0.089359</td>\n",
       "      <td>-0.108275</td>\n",
       "      <td>-0.054224</td>\n",
       "      <td>-0.053581</td>\n",
       "      <td>0.053581</td>\n",
       "      <td>-0.103762</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.095907</td>\n",
       "      <td>0.095045</td>\n",
       "      <td>-0.093314</td>\n",
       "      <td>0.092646</td>\n",
       "      <td>-0.062189</td>\n",
       "      <td>-0.062189</td>\n",
       "      <td>0.069512</td>\n",
       "      <td>0.049407</td>\n",
       "      <td>0.033717</td>\n",
       "      <td>0.275295</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TINA_3</th>\n",
       "      <td>0.196354</td>\n",
       "      <td>0.096271</td>\n",
       "      <td>-0.096213</td>\n",
       "      <td>-0.202731</td>\n",
       "      <td>-0.100333</td>\n",
       "      <td>-0.098416</td>\n",
       "      <td>-0.098375</td>\n",
       "      <td>0.098375</td>\n",
       "      <td>-0.081182</td>\n",
       "      <td>0.078146</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.087804</td>\n",
       "      <td>0.087598</td>\n",
       "      <td>-0.085104</td>\n",
       "      <td>0.085069</td>\n",
       "      <td>-0.037632</td>\n",
       "      <td>-0.037632</td>\n",
       "      <td>0.147884</td>\n",
       "      <td>0.076544</td>\n",
       "      <td>0.070832</td>\n",
       "      <td>0.278533</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>NBRKM_DRAP_ 1</th>\n",
       "      <td>0.122732</td>\n",
       "      <td>0.621054</td>\n",
       "      <td>-0.621263</td>\n",
       "      <td>-0.177040</td>\n",
       "      <td>-0.093890</td>\n",
       "      <td>-0.920992</td>\n",
       "      <td>-0.921514</td>\n",
       "      <td>0.921514</td>\n",
       "      <td>-0.102699</td>\n",
       "      <td>0.048808</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.198615</td>\n",
       "      <td>0.199048</td>\n",
       "      <td>-0.109620</td>\n",
       "      <td>0.110079</td>\n",
       "      <td>-0.131120</td>\n",
       "      <td>-0.131120</td>\n",
       "      <td>0.239188</td>\n",
       "      <td>0.306475</td>\n",
       "      <td>0.059176</td>\n",
       "      <td>0.257483</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>csei_61</th>\n",
       "      <td>0.117389</td>\n",
       "      <td>0.138483</td>\n",
       "      <td>-0.138437</td>\n",
       "      <td>-0.234381</td>\n",
       "      <td>-0.261679</td>\n",
       "      <td>-0.057957</td>\n",
       "      <td>-0.057876</td>\n",
       "      <td>0.057876</td>\n",
       "      <td>-0.208711</td>\n",
       "      <td>0.089793</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.289264</td>\n",
       "      <td>0.289184</td>\n",
       "      <td>-0.281121</td>\n",
       "      <td>0.280792</td>\n",
       "      <td>-0.129141</td>\n",
       "      <td>-0.129141</td>\n",
       "      <td>0.001866</td>\n",
       "      <td>0.166733</td>\n",
       "      <td>-0.040603</td>\n",
       "      <td>0.246502</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>cser_6</th>\n",
       "      <td>0.094301</td>\n",
       "      <td>0.124864</td>\n",
       "      <td>-0.124709</td>\n",
       "      <td>-0.289450</td>\n",
       "      <td>-0.320405</td>\n",
       "      <td>-0.023545</td>\n",
       "      <td>-0.023299</td>\n",
       "      <td>0.023299</td>\n",
       "      <td>-0.261268</td>\n",
       "      <td>0.115547</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.320137</td>\n",
       "      <td>0.319842</td>\n",
       "      <td>-0.348591</td>\n",
       "      <td>0.348417</td>\n",
       "      <td>-0.167460</td>\n",
       "      <td>-0.167460</td>\n",
       "      <td>-0.046364</td>\n",
       "      <td>0.121096</td>\n",
       "      <td>-0.067501</td>\n",
       "      <td>0.242515</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CWVIB_2</th>\n",
       "      <td>-0.122111</td>\n",
       "      <td>-0.316881</td>\n",
       "      <td>0.316669</td>\n",
       "      <td>0.114477</td>\n",
       "      <td>0.239039</td>\n",
       "      <td>0.227721</td>\n",
       "      <td>0.227756</td>\n",
       "      <td>-0.227756</td>\n",
       "      <td>0.154381</td>\n",
       "      <td>-0.095907</td>\n",
       "      <td>...</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.999366</td>\n",
       "      <td>0.405977</td>\n",
       "      <td>-0.405828</td>\n",
       "      <td>0.079063</td>\n",
       "      <td>0.079063</td>\n",
       "      <td>-0.117392</td>\n",
       "      <td>-0.224299</td>\n",
       "      <td>0.029367</td>\n",
       "      <td>-0.246877</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CWVIB_1</th>\n",
       "      <td>0.121923</td>\n",
       "      <td>0.317121</td>\n",
       "      <td>-0.316661</td>\n",
       "      <td>-0.114215</td>\n",
       "      <td>-0.238803</td>\n",
       "      <td>-0.227952</td>\n",
       "      <td>-0.227986</td>\n",
       "      <td>0.227986</td>\n",
       "      <td>-0.154493</td>\n",
       "      <td>0.095045</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.999366</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.405937</td>\n",
       "      <td>0.406048</td>\n",
       "      <td>-0.079202</td>\n",
       "      <td>-0.079202</td>\n",
       "      <td>0.117302</td>\n",
       "      <td>0.224500</td>\n",
       "      <td>-0.029521</td>\n",
       "      <td>0.246870</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SECFUPOU_2</th>\n",
       "      <td>-0.103316</td>\n",
       "      <td>-0.238399</td>\n",
       "      <td>0.238813</td>\n",
       "      <td>0.100931</td>\n",
       "      <td>0.313607</td>\n",
       "      <td>0.136080</td>\n",
       "      <td>0.135854</td>\n",
       "      <td>-0.135854</td>\n",
       "      <td>0.199522</td>\n",
       "      <td>-0.093314</td>\n",
       "      <td>...</td>\n",
       "      <td>0.405977</td>\n",
       "      <td>-0.405937</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.999358</td>\n",
       "      <td>0.148113</td>\n",
       "      <td>0.148113</td>\n",
       "      <td>-0.065502</td>\n",
       "      <td>-0.077693</td>\n",
       "      <td>0.048650</td>\n",
       "      <td>-0.234598</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SECFUPOU_1</th>\n",
       "      <td>0.103048</td>\n",
       "      <td>0.238491</td>\n",
       "      <td>-0.238701</td>\n",
       "      <td>-0.100576</td>\n",
       "      <td>-0.313275</td>\n",
       "      <td>-0.136181</td>\n",
       "      <td>-0.136156</td>\n",
       "      <td>0.136156</td>\n",
       "      <td>-0.199632</td>\n",
       "      <td>0.092646</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.405828</td>\n",
       "      <td>0.406048</td>\n",
       "      <td>-0.999358</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.148414</td>\n",
       "      <td>-0.148414</td>\n",
       "      <td>0.065291</td>\n",
       "      <td>0.077845</td>\n",
       "      <td>-0.048612</td>\n",
       "      <td>0.234522</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ACTIVFIN_86</th>\n",
       "      <td>-0.056000</td>\n",
       "      <td>-0.136021</td>\n",
       "      <td>0.135862</td>\n",
       "      <td>0.085229</td>\n",
       "      <td>0.232267</td>\n",
       "      <td>0.142725</td>\n",
       "      <td>0.142496</td>\n",
       "      <td>-0.142496</td>\n",
       "      <td>0.494103</td>\n",
       "      <td>-0.062189</td>\n",
       "      <td>...</td>\n",
       "      <td>0.079063</td>\n",
       "      <td>-0.079202</td>\n",
       "      <td>0.148113</td>\n",
       "      <td>-0.148414</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.050337</td>\n",
       "      <td>-0.068825</td>\n",
       "      <td>0.010289</td>\n",
       "      <td>-0.201527</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>naf38_QA</th>\n",
       "      <td>-0.056000</td>\n",
       "      <td>-0.136021</td>\n",
       "      <td>0.135862</td>\n",
       "      <td>0.085229</td>\n",
       "      <td>0.232267</td>\n",
       "      <td>0.142725</td>\n",
       "      <td>0.142496</td>\n",
       "      <td>-0.142496</td>\n",
       "      <td>0.494103</td>\n",
       "      <td>-0.062189</td>\n",
       "      <td>...</td>\n",
       "      <td>0.079063</td>\n",
       "      <td>-0.079202</td>\n",
       "      <td>0.148113</td>\n",
       "      <td>-0.148414</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.050337</td>\n",
       "      <td>-0.068825</td>\n",
       "      <td>0.010289</td>\n",
       "      <td>-0.201527</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>HH</th>\n",
       "      <td>0.529593</td>\n",
       "      <td>0.204578</td>\n",
       "      <td>-0.204726</td>\n",
       "      <td>-0.190228</td>\n",
       "      <td>-0.142649</td>\n",
       "      <td>-0.244954</td>\n",
       "      <td>-0.245055</td>\n",
       "      <td>0.245055</td>\n",
       "      <td>-0.142054</td>\n",
       "      <td>0.069512</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.117392</td>\n",
       "      <td>0.117302</td>\n",
       "      <td>-0.065502</td>\n",
       "      <td>0.065291</td>\n",
       "      <td>-0.050337</td>\n",
       "      <td>-0.050337</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.214283</td>\n",
       "      <td>0.138920</td>\n",
       "      <td>0.268707</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>KMANNU</th>\n",
       "      <td>0.100230</td>\n",
       "      <td>0.303375</td>\n",
       "      <td>-0.303152</td>\n",
       "      <td>-0.124125</td>\n",
       "      <td>-0.079932</td>\n",
       "      <td>-0.282261</td>\n",
       "      <td>-0.282421</td>\n",
       "      <td>0.282421</td>\n",
       "      <td>-0.162580</td>\n",
       "      <td>0.049407</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.224299</td>\n",
       "      <td>0.224500</td>\n",
       "      <td>-0.077693</td>\n",
       "      <td>0.077845</td>\n",
       "      <td>-0.068825</td>\n",
       "      <td>-0.068825</td>\n",
       "      <td>0.214283</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.026421</td>\n",
       "      <td>0.209452</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>REVMENS</th>\n",
       "      <td>0.123532</td>\n",
       "      <td>0.031349</td>\n",
       "      <td>-0.031431</td>\n",
       "      <td>-0.112451</td>\n",
       "      <td>-0.004528</td>\n",
       "      <td>-0.055904</td>\n",
       "      <td>-0.055949</td>\n",
       "      <td>0.055949</td>\n",
       "      <td>0.027168</td>\n",
       "      <td>0.033717</td>\n",
       "      <td>...</td>\n",
       "      <td>0.029367</td>\n",
       "      <td>-0.029521</td>\n",
       "      <td>0.048650</td>\n",
       "      <td>-0.048612</td>\n",
       "      <td>0.010289</td>\n",
       "      <td>0.010289</td>\n",
       "      <td>0.138920</td>\n",
       "      <td>0.026421</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.114774</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>target</th>\n",
       "      <td>0.308664</td>\n",
       "      <td>0.283972</td>\n",
       "      <td>-0.283617</td>\n",
       "      <td>-0.294126</td>\n",
       "      <td>-0.276469</td>\n",
       "      <td>-0.269212</td>\n",
       "      <td>-0.268930</td>\n",
       "      <td>0.268930</td>\n",
       "      <td>-0.250588</td>\n",
       "      <td>0.275295</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.246877</td>\n",
       "      <td>0.246870</td>\n",
       "      <td>-0.234598</td>\n",
       "      <td>0.234522</td>\n",
       "      <td>-0.201527</td>\n",
       "      <td>-0.201527</td>\n",
       "      <td>0.268707</td>\n",
       "      <td>0.209452</td>\n",
       "      <td>0.114774</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>24 rows × 24 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                  TPP_1  SECROUT_1  SECROUT_2    cser_5   naf4_EV  CONDUITE_2  \\\n",
       "TPP_1          1.000000   0.125718  -0.125640 -0.167049 -0.143424   -0.127877   \n",
       "SECROUT_1      0.125718   1.000000  -0.999123 -0.144906 -0.138219   -0.661175   \n",
       "SECROUT_2     -0.125640  -0.999123   1.000000  0.144802  0.137748    0.661684   \n",
       "cser_5        -0.167049  -0.144906   0.144802  1.000000  0.256628    0.181942   \n",
       "naf4_EV       -0.143424  -0.138219   0.137748  0.256628  1.000000    0.103048   \n",
       "CONDUITE_2    -0.127877  -0.661175   0.661684  0.181942  0.103048    1.000000   \n",
       "NBRKM_DRAP_ 0 -0.128077  -0.661096   0.661605  0.181808  0.102672    0.999434   \n",
       "CONDUITE_1     0.128077   0.661096  -0.661605 -0.181808 -0.102672   -0.999434   \n",
       "naf17_OQ      -0.088461  -0.119002   0.118607  0.184649  0.470077    0.105614   \n",
       "CJSITUA_6      0.071407   0.062976  -0.062752 -0.089359 -0.108275   -0.054224   \n",
       "TINA_3         0.196354   0.096271  -0.096213 -0.202731 -0.100333   -0.098416   \n",
       "NBRKM_DRAP_ 1  0.122732   0.621054  -0.621263 -0.177040 -0.093890   -0.920992   \n",
       "csei_61        0.117389   0.138483  -0.138437 -0.234381 -0.261679   -0.057957   \n",
       "cser_6         0.094301   0.124864  -0.124709 -0.289450 -0.320405   -0.023545   \n",
       "CWVIB_2       -0.122111  -0.316881   0.316669  0.114477  0.239039    0.227721   \n",
       "CWVIB_1        0.121923   0.317121  -0.316661 -0.114215 -0.238803   -0.227952   \n",
       "SECFUPOU_2    -0.103316  -0.238399   0.238813  0.100931  0.313607    0.136080   \n",
       "SECFUPOU_1     0.103048   0.238491  -0.238701 -0.100576 -0.313275   -0.136181   \n",
       "ACTIVFIN_86   -0.056000  -0.136021   0.135862  0.085229  0.232267    0.142725   \n",
       "naf38_QA      -0.056000  -0.136021   0.135862  0.085229  0.232267    0.142725   \n",
       "HH             0.529593   0.204578  -0.204726 -0.190228 -0.142649   -0.244954   \n",
       "KMANNU         0.100230   0.303375  -0.303152 -0.124125 -0.079932   -0.282261   \n",
       "REVMENS        0.123532   0.031349  -0.031431 -0.112451 -0.004528   -0.055904   \n",
       "target         0.308664   0.283972  -0.283617 -0.294126 -0.276469   -0.269212   \n",
       "\n",
       "               NBRKM_DRAP_ 0  CONDUITE_1  naf17_OQ  CJSITUA_6  ...   CWVIB_2  \\\n",
       "TPP_1              -0.128077    0.128077 -0.088461   0.071407  ... -0.122111   \n",
       "SECROUT_1          -0.661096    0.661096 -0.119002   0.062976  ... -0.316881   \n",
       "SECROUT_2           0.661605   -0.661605  0.118607  -0.062752  ...  0.316669   \n",
       "cser_5              0.181808   -0.181808  0.184649  -0.089359  ...  0.114477   \n",
       "naf4_EV             0.102672   -0.102672  0.470077  -0.108275  ...  0.239039   \n",
       "CONDUITE_2          0.999434   -0.999434  0.105614  -0.054224  ...  0.227721   \n",
       "NBRKM_DRAP_ 0       1.000000   -1.000000  0.105128  -0.053581  ...  0.227756   \n",
       "CONDUITE_1         -1.000000    1.000000 -0.105128   0.053581  ... -0.227756   \n",
       "naf17_OQ            0.105128   -0.105128  1.000000  -0.103762  ...  0.154381   \n",
       "CJSITUA_6          -0.053581    0.053581 -0.103762   1.000000  ... -0.095907   \n",
       "TINA_3             -0.098375    0.098375 -0.081182   0.078146  ... -0.087804   \n",
       "NBRKM_DRAP_ 1      -0.921514    0.921514 -0.102699   0.048808  ... -0.198615   \n",
       "csei_61            -0.057876    0.057876 -0.208711   0.089793  ... -0.289264   \n",
       "cser_6             -0.023299    0.023299 -0.261268   0.115547  ... -0.320137   \n",
       "CWVIB_2             0.227756   -0.227756  0.154381  -0.095907  ...  1.000000   \n",
       "CWVIB_1            -0.227986    0.227986 -0.154493   0.095045  ... -0.999366   \n",
       "SECFUPOU_2          0.135854   -0.135854  0.199522  -0.093314  ...  0.405977   \n",
       "SECFUPOU_1         -0.136156    0.136156 -0.199632   0.092646  ... -0.405828   \n",
       "ACTIVFIN_86         0.142496   -0.142496  0.494103  -0.062189  ...  0.079063   \n",
       "naf38_QA            0.142496   -0.142496  0.494103  -0.062189  ...  0.079063   \n",
       "HH                 -0.245055    0.245055 -0.142054   0.069512  ... -0.117392   \n",
       "KMANNU             -0.282421    0.282421 -0.162580   0.049407  ... -0.224299   \n",
       "REVMENS            -0.055949    0.055949  0.027168   0.033717  ...  0.029367   \n",
       "target             -0.268930    0.268930 -0.250588   0.275295  ... -0.246877   \n",
       "\n",
       "                CWVIB_1  SECFUPOU_2  SECFUPOU_1  ACTIVFIN_86  naf38_QA  \\\n",
       "TPP_1          0.121923   -0.103316    0.103048    -0.056000 -0.056000   \n",
       "SECROUT_1      0.317121   -0.238399    0.238491    -0.136021 -0.136021   \n",
       "SECROUT_2     -0.316661    0.238813   -0.238701     0.135862  0.135862   \n",
       "cser_5        -0.114215    0.100931   -0.100576     0.085229  0.085229   \n",
       "naf4_EV       -0.238803    0.313607   -0.313275     0.232267  0.232267   \n",
       "CONDUITE_2    -0.227952    0.136080   -0.136181     0.142725  0.142725   \n",
       "NBRKM_DRAP_ 0 -0.227986    0.135854   -0.136156     0.142496  0.142496   \n",
       "CONDUITE_1     0.227986   -0.135854    0.136156    -0.142496 -0.142496   \n",
       "naf17_OQ      -0.154493    0.199522   -0.199632     0.494103  0.494103   \n",
       "CJSITUA_6      0.095045   -0.093314    0.092646    -0.062189 -0.062189   \n",
       "TINA_3         0.087598   -0.085104    0.085069    -0.037632 -0.037632   \n",
       "NBRKM_DRAP_ 1  0.199048   -0.109620    0.110079    -0.131120 -0.131120   \n",
       "csei_61        0.289184   -0.281121    0.280792    -0.129141 -0.129141   \n",
       "cser_6         0.319842   -0.348591    0.348417    -0.167460 -0.167460   \n",
       "CWVIB_2       -0.999366    0.405977   -0.405828     0.079063  0.079063   \n",
       "CWVIB_1        1.000000   -0.405937    0.406048    -0.079202 -0.079202   \n",
       "SECFUPOU_2    -0.405937    1.000000   -0.999358     0.148113  0.148113   \n",
       "SECFUPOU_1     0.406048   -0.999358    1.000000    -0.148414 -0.148414   \n",
       "ACTIVFIN_86   -0.079202    0.148113   -0.148414     1.000000  1.000000   \n",
       "naf38_QA      -0.079202    0.148113   -0.148414     1.000000  1.000000   \n",
       "HH             0.117302   -0.065502    0.065291    -0.050337 -0.050337   \n",
       "KMANNU         0.224500   -0.077693    0.077845    -0.068825 -0.068825   \n",
       "REVMENS       -0.029521    0.048650   -0.048612     0.010289  0.010289   \n",
       "target         0.246870   -0.234598    0.234522    -0.201527 -0.201527   \n",
       "\n",
       "                     HH    KMANNU   REVMENS    target  \n",
       "TPP_1          0.529593  0.100230  0.123532  0.308664  \n",
       "SECROUT_1      0.204578  0.303375  0.031349  0.283972  \n",
       "SECROUT_2     -0.204726 -0.303152 -0.031431 -0.283617  \n",
       "cser_5        -0.190228 -0.124125 -0.112451 -0.294126  \n",
       "naf4_EV       -0.142649 -0.079932 -0.004528 -0.276469  \n",
       "CONDUITE_2    -0.244954 -0.282261 -0.055904 -0.269212  \n",
       "NBRKM_DRAP_ 0 -0.245055 -0.282421 -0.055949 -0.268930  \n",
       "CONDUITE_1     0.245055  0.282421  0.055949  0.268930  \n",
       "naf17_OQ      -0.142054 -0.162580  0.027168 -0.250588  \n",
       "CJSITUA_6      0.069512  0.049407  0.033717  0.275295  \n",
       "TINA_3         0.147884  0.076544  0.070832  0.278533  \n",
       "NBRKM_DRAP_ 1  0.239188  0.306475  0.059176  0.257483  \n",
       "csei_61        0.001866  0.166733 -0.040603  0.246502  \n",
       "cser_6        -0.046364  0.121096 -0.067501  0.242515  \n",
       "CWVIB_2       -0.117392 -0.224299  0.029367 -0.246877  \n",
       "CWVIB_1        0.117302  0.224500 -0.029521  0.246870  \n",
       "SECFUPOU_2    -0.065502 -0.077693  0.048650 -0.234598  \n",
       "SECFUPOU_1     0.065291  0.077845 -0.048612  0.234522  \n",
       "ACTIVFIN_86   -0.050337 -0.068825  0.010289 -0.201527  \n",
       "naf38_QA      -0.050337 -0.068825  0.010289 -0.201527  \n",
       "HH             1.000000  0.214283  0.138920  0.268707  \n",
       "KMANNU         0.214283  1.000000  0.026421  0.209452  \n",
       "REVMENS        0.138920  0.026421  1.000000  0.114774  \n",
       "target         0.268707  0.209452  0.114774  1.000000  \n",
       "\n",
       "[24 rows x 24 columns]"
      ]
     },
     "execution_count": 156,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Calculate the correlation matrix\n",
    "corr_matrix = one_hot_df.corr()\n",
    "corr_matrix"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Due to the still high number of columns we need further processing to find collinear variables"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 157,
   "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>TPP_1</th>\n",
       "      <th>SECROUT_1</th>\n",
       "      <th>SECROUT_2</th>\n",
       "      <th>cser_5</th>\n",
       "      <th>naf4_EV</th>\n",
       "      <th>CONDUITE_2</th>\n",
       "      <th>NBRKM_DRAP_ 0</th>\n",
       "      <th>CONDUITE_1</th>\n",
       "      <th>naf17_OQ</th>\n",
       "      <th>CJSITUA_6</th>\n",
       "      <th>...</th>\n",
       "      <th>CWVIB_2</th>\n",
       "      <th>CWVIB_1</th>\n",
       "      <th>SECFUPOU_2</th>\n",
       "      <th>SECFUPOU_1</th>\n",
       "      <th>ACTIVFIN_86</th>\n",
       "      <th>naf38_QA</th>\n",
       "      <th>HH</th>\n",
       "      <th>KMANNU</th>\n",
       "      <th>REVMENS</th>\n",
       "      <th>target</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>TPP_1</th>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SECROUT_1</th>\n",
       "      <td>NaN</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.999123</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SECROUT_2</th>\n",
       "      <td>NaN</td>\n",
       "      <td>-0.999123</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>cser_5</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>naf4_EV</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CONDUITE_2</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.999434</td>\n",
       "      <td>-0.999434</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>NBRKM_DRAP_ 0</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.999434</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CONDUITE_1</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-0.999434</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>naf17_OQ</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CJSITUA_6</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TINA_3</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>NBRKM_DRAP_ 1</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-0.920992</td>\n",
       "      <td>-0.921514</td>\n",
       "      <td>0.921514</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>csei_61</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>cser_6</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CWVIB_2</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.999366</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CWVIB_1</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.999366</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SECFUPOU_2</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.999358</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SECFUPOU_1</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-0.999358</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ACTIVFIN_86</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>naf38_QA</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>HH</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>KMANNU</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>REVMENS</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>target</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>24 rows × 24 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "               TPP_1  SECROUT_1  SECROUT_2  cser_5  naf4_EV  CONDUITE_2  \\\n",
       "TPP_1            1.0        NaN        NaN     NaN      NaN         NaN   \n",
       "SECROUT_1        NaN   1.000000  -0.999123     NaN      NaN         NaN   \n",
       "SECROUT_2        NaN  -0.999123   1.000000     NaN      NaN         NaN   \n",
       "cser_5           NaN        NaN        NaN     1.0      NaN         NaN   \n",
       "naf4_EV          NaN        NaN        NaN     NaN      1.0         NaN   \n",
       "CONDUITE_2       NaN        NaN        NaN     NaN      NaN    1.000000   \n",
       "NBRKM_DRAP_ 0    NaN        NaN        NaN     NaN      NaN    0.999434   \n",
       "CONDUITE_1       NaN        NaN        NaN     NaN      NaN   -0.999434   \n",
       "naf17_OQ         NaN        NaN        NaN     NaN      NaN         NaN   \n",
       "CJSITUA_6        NaN        NaN        NaN     NaN      NaN         NaN   \n",
       "TINA_3           NaN        NaN        NaN     NaN      NaN         NaN   \n",
       "NBRKM_DRAP_ 1    NaN        NaN        NaN     NaN      NaN   -0.920992   \n",
       "csei_61          NaN        NaN        NaN     NaN      NaN         NaN   \n",
       "cser_6           NaN        NaN        NaN     NaN      NaN         NaN   \n",
       "CWVIB_2          NaN        NaN        NaN     NaN      NaN         NaN   \n",
       "CWVIB_1          NaN        NaN        NaN     NaN      NaN         NaN   \n",
       "SECFUPOU_2       NaN        NaN        NaN     NaN      NaN         NaN   \n",
       "SECFUPOU_1       NaN        NaN        NaN     NaN      NaN         NaN   \n",
       "ACTIVFIN_86      NaN        NaN        NaN     NaN      NaN         NaN   \n",
       "naf38_QA         NaN        NaN        NaN     NaN      NaN         NaN   \n",
       "HH               NaN        NaN        NaN     NaN      NaN         NaN   \n",
       "KMANNU           NaN        NaN        NaN     NaN      NaN         NaN   \n",
       "REVMENS          NaN        NaN        NaN     NaN      NaN         NaN   \n",
       "target           NaN        NaN        NaN     NaN      NaN         NaN   \n",
       "\n",
       "               NBRKM_DRAP_ 0  CONDUITE_1  naf17_OQ  CJSITUA_6  ...   CWVIB_2  \\\n",
       "TPP_1                    NaN         NaN       NaN        NaN  ...       NaN   \n",
       "SECROUT_1                NaN         NaN       NaN        NaN  ...       NaN   \n",
       "SECROUT_2                NaN         NaN       NaN        NaN  ...       NaN   \n",
       "cser_5                   NaN         NaN       NaN        NaN  ...       NaN   \n",
       "naf4_EV                  NaN         NaN       NaN        NaN  ...       NaN   \n",
       "CONDUITE_2          0.999434   -0.999434       NaN        NaN  ...       NaN   \n",
       "NBRKM_DRAP_ 0       1.000000   -1.000000       NaN        NaN  ...       NaN   \n",
       "CONDUITE_1         -1.000000    1.000000       NaN        NaN  ...       NaN   \n",
       "naf17_OQ                 NaN         NaN       1.0        NaN  ...       NaN   \n",
       "CJSITUA_6                NaN         NaN       NaN        1.0  ...       NaN   \n",
       "TINA_3                   NaN         NaN       NaN        NaN  ...       NaN   \n",
       "NBRKM_DRAP_ 1      -0.921514    0.921514       NaN        NaN  ...       NaN   \n",
       "csei_61                  NaN         NaN       NaN        NaN  ...       NaN   \n",
       "cser_6                   NaN         NaN       NaN        NaN  ...       NaN   \n",
       "CWVIB_2                  NaN         NaN       NaN        NaN  ...  1.000000   \n",
       "CWVIB_1                  NaN         NaN       NaN        NaN  ... -0.999366   \n",
       "SECFUPOU_2               NaN         NaN       NaN        NaN  ...       NaN   \n",
       "SECFUPOU_1               NaN         NaN       NaN        NaN  ...       NaN   \n",
       "ACTIVFIN_86              NaN         NaN       NaN        NaN  ...       NaN   \n",
       "naf38_QA                 NaN         NaN       NaN        NaN  ...       NaN   \n",
       "HH                       NaN         NaN       NaN        NaN  ...       NaN   \n",
       "KMANNU                   NaN         NaN       NaN        NaN  ...       NaN   \n",
       "REVMENS                  NaN         NaN       NaN        NaN  ...       NaN   \n",
       "target                   NaN         NaN       NaN        NaN  ...       NaN   \n",
       "\n",
       "                CWVIB_1  SECFUPOU_2  SECFUPOU_1  ACTIVFIN_86  naf38_QA   HH  \\\n",
       "TPP_1               NaN         NaN         NaN          NaN       NaN  NaN   \n",
       "SECROUT_1           NaN         NaN         NaN          NaN       NaN  NaN   \n",
       "SECROUT_2           NaN         NaN         NaN          NaN       NaN  NaN   \n",
       "cser_5              NaN         NaN         NaN          NaN       NaN  NaN   \n",
       "naf4_EV             NaN         NaN         NaN          NaN       NaN  NaN   \n",
       "CONDUITE_2          NaN         NaN         NaN          NaN       NaN  NaN   \n",
       "NBRKM_DRAP_ 0       NaN         NaN         NaN          NaN       NaN  NaN   \n",
       "CONDUITE_1          NaN         NaN         NaN          NaN       NaN  NaN   \n",
       "naf17_OQ            NaN         NaN         NaN          NaN       NaN  NaN   \n",
       "CJSITUA_6           NaN         NaN         NaN          NaN       NaN  NaN   \n",
       "TINA_3              NaN         NaN         NaN          NaN       NaN  NaN   \n",
       "NBRKM_DRAP_ 1       NaN         NaN         NaN          NaN       NaN  NaN   \n",
       "csei_61             NaN         NaN         NaN          NaN       NaN  NaN   \n",
       "cser_6              NaN         NaN         NaN          NaN       NaN  NaN   \n",
       "CWVIB_2       -0.999366         NaN         NaN          NaN       NaN  NaN   \n",
       "CWVIB_1        1.000000         NaN         NaN          NaN       NaN  NaN   \n",
       "SECFUPOU_2          NaN    1.000000   -0.999358          NaN       NaN  NaN   \n",
       "SECFUPOU_1          NaN   -0.999358    1.000000          NaN       NaN  NaN   \n",
       "ACTIVFIN_86         NaN         NaN         NaN          1.0       1.0  NaN   \n",
       "naf38_QA            NaN         NaN         NaN          1.0       1.0  NaN   \n",
       "HH                  NaN         NaN         NaN          NaN       NaN  1.0   \n",
       "KMANNU              NaN         NaN         NaN          NaN       NaN  NaN   \n",
       "REVMENS             NaN         NaN         NaN          NaN       NaN  NaN   \n",
       "target              NaN         NaN         NaN          NaN       NaN  NaN   \n",
       "\n",
       "               KMANNU  REVMENS  target  \n",
       "TPP_1             NaN      NaN     NaN  \n",
       "SECROUT_1         NaN      NaN     NaN  \n",
       "SECROUT_2         NaN      NaN     NaN  \n",
       "cser_5            NaN      NaN     NaN  \n",
       "naf4_EV           NaN      NaN     NaN  \n",
       "CONDUITE_2        NaN      NaN     NaN  \n",
       "NBRKM_DRAP_ 0     NaN      NaN     NaN  \n",
       "CONDUITE_1        NaN      NaN     NaN  \n",
       "naf17_OQ          NaN      NaN     NaN  \n",
       "CJSITUA_6         NaN      NaN     NaN  \n",
       "TINA_3            NaN      NaN     NaN  \n",
       "NBRKM_DRAP_ 1     NaN      NaN     NaN  \n",
       "csei_61           NaN      NaN     NaN  \n",
       "cser_6            NaN      NaN     NaN  \n",
       "CWVIB_2           NaN      NaN     NaN  \n",
       "CWVIB_1           NaN      NaN     NaN  \n",
       "SECFUPOU_2        NaN      NaN     NaN  \n",
       "SECFUPOU_1        NaN      NaN     NaN  \n",
       "ACTIVFIN_86       NaN      NaN     NaN  \n",
       "naf38_QA          NaN      NaN     NaN  \n",
       "HH                NaN      NaN     NaN  \n",
       "KMANNU            1.0      NaN     NaN  \n",
       "REVMENS           NaN      1.0     NaN  \n",
       "target            NaN      NaN     1.0  \n",
       "\n",
       "[24 rows x 24 columns]"
      ]
     },
     "execution_count": 157,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Filter the correlation matrix to keep only the values above 0.8\n",
    "filtered_corr_matrix = corr_matrix[abs(corr_matrix) >= 0.8]\n",
    "filtered_corr_matrix"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 158,
   "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>TPP_1</th>\n",
       "      <th>SECROUT_1</th>\n",
       "      <th>SECROUT_2</th>\n",
       "      <th>cser_5</th>\n",
       "      <th>naf4_EV</th>\n",
       "      <th>CONDUITE_2</th>\n",
       "      <th>NBRKM_DRAP_ 0</th>\n",
       "      <th>CONDUITE_1</th>\n",
       "      <th>naf17_OQ</th>\n",
       "      <th>CJSITUA_6</th>\n",
       "      <th>...</th>\n",
       "      <th>CWVIB_2</th>\n",
       "      <th>CWVIB_1</th>\n",
       "      <th>SECFUPOU_2</th>\n",
       "      <th>SECFUPOU_1</th>\n",
       "      <th>ACTIVFIN_86</th>\n",
       "      <th>naf38_QA</th>\n",
       "      <th>HH</th>\n",
       "      <th>KMANNU</th>\n",
       "      <th>REVMENS</th>\n",
       "      <th>target</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>TPP_1</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SECROUT_1</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-0.999123</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SECROUT_2</th>\n",
       "      <td>NaN</td>\n",
       "      <td>-0.999123</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>cser_5</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>naf4_EV</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CONDUITE_2</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.999434</td>\n",
       "      <td>-0.999434</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>NBRKM_DRAP_ 0</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.999434</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CONDUITE_1</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-0.999434</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>naf17_OQ</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CJSITUA_6</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TINA_3</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>NBRKM_DRAP_ 1</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-0.920992</td>\n",
       "      <td>-0.921514</td>\n",
       "      <td>0.921514</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>csei_61</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>cser_6</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CWVIB_2</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-0.999366</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CWVIB_1</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.999366</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SECFUPOU_2</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-0.999358</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SECFUPOU_1</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-0.999358</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ACTIVFIN_86</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>naf38_QA</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>HH</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>KMANNU</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>REVMENS</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>target</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>24 rows × 24 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "               TPP_1  SECROUT_1  SECROUT_2  cser_5  naf4_EV  CONDUITE_2  \\\n",
       "TPP_1            NaN        NaN        NaN     NaN      NaN         NaN   \n",
       "SECROUT_1        NaN        NaN  -0.999123     NaN      NaN         NaN   \n",
       "SECROUT_2        NaN  -0.999123        NaN     NaN      NaN         NaN   \n",
       "cser_5           NaN        NaN        NaN     NaN      NaN         NaN   \n",
       "naf4_EV          NaN        NaN        NaN     NaN      NaN         NaN   \n",
       "CONDUITE_2       NaN        NaN        NaN     NaN      NaN         NaN   \n",
       "NBRKM_DRAP_ 0    NaN        NaN        NaN     NaN      NaN    0.999434   \n",
       "CONDUITE_1       NaN        NaN        NaN     NaN      NaN   -0.999434   \n",
       "naf17_OQ         NaN        NaN        NaN     NaN      NaN         NaN   \n",
       "CJSITUA_6        NaN        NaN        NaN     NaN      NaN         NaN   \n",
       "TINA_3           NaN        NaN        NaN     NaN      NaN         NaN   \n",
       "NBRKM_DRAP_ 1    NaN        NaN        NaN     NaN      NaN   -0.920992   \n",
       "csei_61          NaN        NaN        NaN     NaN      NaN         NaN   \n",
       "cser_6           NaN        NaN        NaN     NaN      NaN         NaN   \n",
       "CWVIB_2          NaN        NaN        NaN     NaN      NaN         NaN   \n",
       "CWVIB_1          NaN        NaN        NaN     NaN      NaN         NaN   \n",
       "SECFUPOU_2       NaN        NaN        NaN     NaN      NaN         NaN   \n",
       "SECFUPOU_1       NaN        NaN        NaN     NaN      NaN         NaN   \n",
       "ACTIVFIN_86      NaN        NaN        NaN     NaN      NaN         NaN   \n",
       "naf38_QA         NaN        NaN        NaN     NaN      NaN         NaN   \n",
       "HH               NaN        NaN        NaN     NaN      NaN         NaN   \n",
       "KMANNU           NaN        NaN        NaN     NaN      NaN         NaN   \n",
       "REVMENS          NaN        NaN        NaN     NaN      NaN         NaN   \n",
       "target           NaN        NaN        NaN     NaN      NaN         NaN   \n",
       "\n",
       "               NBRKM_DRAP_ 0  CONDUITE_1  naf17_OQ  CJSITUA_6  ...   CWVIB_2  \\\n",
       "TPP_1                    NaN         NaN       NaN        NaN  ...       NaN   \n",
       "SECROUT_1                NaN         NaN       NaN        NaN  ...       NaN   \n",
       "SECROUT_2                NaN         NaN       NaN        NaN  ...       NaN   \n",
       "cser_5                   NaN         NaN       NaN        NaN  ...       NaN   \n",
       "naf4_EV                  NaN         NaN       NaN        NaN  ...       NaN   \n",
       "CONDUITE_2          0.999434   -0.999434       NaN        NaN  ...       NaN   \n",
       "NBRKM_DRAP_ 0            NaN   -1.000000       NaN        NaN  ...       NaN   \n",
       "CONDUITE_1         -1.000000         NaN       NaN        NaN  ...       NaN   \n",
       "naf17_OQ                 NaN         NaN       NaN        NaN  ...       NaN   \n",
       "CJSITUA_6                NaN         NaN       NaN        NaN  ...       NaN   \n",
       "TINA_3                   NaN         NaN       NaN        NaN  ...       NaN   \n",
       "NBRKM_DRAP_ 1      -0.921514    0.921514       NaN        NaN  ...       NaN   \n",
       "csei_61                  NaN         NaN       NaN        NaN  ...       NaN   \n",
       "cser_6                   NaN         NaN       NaN        NaN  ...       NaN   \n",
       "CWVIB_2                  NaN         NaN       NaN        NaN  ...       NaN   \n",
       "CWVIB_1                  NaN         NaN       NaN        NaN  ... -0.999366   \n",
       "SECFUPOU_2               NaN         NaN       NaN        NaN  ...       NaN   \n",
       "SECFUPOU_1               NaN         NaN       NaN        NaN  ...       NaN   \n",
       "ACTIVFIN_86              NaN         NaN       NaN        NaN  ...       NaN   \n",
       "naf38_QA                 NaN         NaN       NaN        NaN  ...       NaN   \n",
       "HH                       NaN         NaN       NaN        NaN  ...       NaN   \n",
       "KMANNU                   NaN         NaN       NaN        NaN  ...       NaN   \n",
       "REVMENS                  NaN         NaN       NaN        NaN  ...       NaN   \n",
       "target                   NaN         NaN       NaN        NaN  ...       NaN   \n",
       "\n",
       "                CWVIB_1  SECFUPOU_2  SECFUPOU_1  ACTIVFIN_86  naf38_QA  HH  \\\n",
       "TPP_1               NaN         NaN         NaN          NaN       NaN NaN   \n",
       "SECROUT_1           NaN         NaN         NaN          NaN       NaN NaN   \n",
       "SECROUT_2           NaN         NaN         NaN          NaN       NaN NaN   \n",
       "cser_5              NaN         NaN         NaN          NaN       NaN NaN   \n",
       "naf4_EV             NaN         NaN         NaN          NaN       NaN NaN   \n",
       "CONDUITE_2          NaN         NaN         NaN          NaN       NaN NaN   \n",
       "NBRKM_DRAP_ 0       NaN         NaN         NaN          NaN       NaN NaN   \n",
       "CONDUITE_1          NaN         NaN         NaN          NaN       NaN NaN   \n",
       "naf17_OQ            NaN         NaN         NaN          NaN       NaN NaN   \n",
       "CJSITUA_6           NaN         NaN         NaN          NaN       NaN NaN   \n",
       "TINA_3              NaN         NaN         NaN          NaN       NaN NaN   \n",
       "NBRKM_DRAP_ 1       NaN         NaN         NaN          NaN       NaN NaN   \n",
       "csei_61             NaN         NaN         NaN          NaN       NaN NaN   \n",
       "cser_6              NaN         NaN         NaN          NaN       NaN NaN   \n",
       "CWVIB_2       -0.999366         NaN         NaN          NaN       NaN NaN   \n",
       "CWVIB_1             NaN         NaN         NaN          NaN       NaN NaN   \n",
       "SECFUPOU_2          NaN         NaN   -0.999358          NaN       NaN NaN   \n",
       "SECFUPOU_1          NaN   -0.999358         NaN          NaN       NaN NaN   \n",
       "ACTIVFIN_86         NaN         NaN         NaN          NaN       1.0 NaN   \n",
       "naf38_QA            NaN         NaN         NaN          1.0       NaN NaN   \n",
       "HH                  NaN         NaN         NaN          NaN       NaN NaN   \n",
       "KMANNU              NaN         NaN         NaN          NaN       NaN NaN   \n",
       "REVMENS             NaN         NaN         NaN          NaN       NaN NaN   \n",
       "target              NaN         NaN         NaN          NaN       NaN NaN   \n",
       "\n",
       "               KMANNU  REVMENS  target  \n",
       "TPP_1             NaN      NaN     NaN  \n",
       "SECROUT_1         NaN      NaN     NaN  \n",
       "SECROUT_2         NaN      NaN     NaN  \n",
       "cser_5            NaN      NaN     NaN  \n",
       "naf4_EV           NaN      NaN     NaN  \n",
       "CONDUITE_2        NaN      NaN     NaN  \n",
       "NBRKM_DRAP_ 0     NaN      NaN     NaN  \n",
       "CONDUITE_1        NaN      NaN     NaN  \n",
       "naf17_OQ          NaN      NaN     NaN  \n",
       "CJSITUA_6         NaN      NaN     NaN  \n",
       "TINA_3            NaN      NaN     NaN  \n",
       "NBRKM_DRAP_ 1     NaN      NaN     NaN  \n",
       "csei_61           NaN      NaN     NaN  \n",
       "cser_6            NaN      NaN     NaN  \n",
       "CWVIB_2           NaN      NaN     NaN  \n",
       "CWVIB_1           NaN      NaN     NaN  \n",
       "SECFUPOU_2        NaN      NaN     NaN  \n",
       "SECFUPOU_1        NaN      NaN     NaN  \n",
       "ACTIVFIN_86       NaN      NaN     NaN  \n",
       "naf38_QA          NaN      NaN     NaN  \n",
       "HH                NaN      NaN     NaN  \n",
       "KMANNU            NaN      NaN     NaN  \n",
       "REVMENS           NaN      NaN     NaN  \n",
       "target            NaN      NaN     NaN  \n",
       "\n",
       "[24 rows x 24 columns]"
      ]
     },
     "execution_count": 158,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Set the diagonal values to NaN to avoid visualzing the variables correlation with itself\n",
    "np.fill_diagonal(filtered_corr_matrix.values, np.nan)\n",
    "filtered_corr_matrix"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 159,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Drop rows and columns with all NaN values to avoid visualizing them\n",
    "filtered_corr_matrix.dropna(thresh=1, inplace=True)\n",
    "filtered_corr_matrix.dropna(thresh=1, axis=1, inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 160,
   "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>SECROUT_1</th>\n",
       "      <th>SECROUT_2</th>\n",
       "      <th>CONDUITE_2</th>\n",
       "      <th>NBRKM_DRAP_ 0</th>\n",
       "      <th>CONDUITE_1</th>\n",
       "      <th>NBRKM_DRAP_ 1</th>\n",
       "      <th>csei_61</th>\n",
       "      <th>cser_6</th>\n",
       "      <th>CWVIB_2</th>\n",
       "      <th>CWVIB_1</th>\n",
       "      <th>SECFUPOU_2</th>\n",
       "      <th>SECFUPOU_1</th>\n",
       "      <th>ACTIVFIN_86</th>\n",
       "      <th>naf38_QA</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>SECROUT_1</th>\n",
       "      <td>NaN</td>\n",
       "      <td>-0.999123</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SECROUT_2</th>\n",
       "      <td>-0.999123</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CONDUITE_2</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.999434</td>\n",
       "      <td>-0.999434</td>\n",
       "      <td>-0.920992</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>NBRKM_DRAP_ 0</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.999434</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>-0.921514</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CONDUITE_1</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-0.999434</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.921514</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>NBRKM_DRAP_ 1</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-0.920992</td>\n",
       "      <td>-0.921514</td>\n",
       "      <td>0.921514</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>csei_61</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.809746</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>cser_6</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.809746</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CWVIB_2</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-0.999366</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CWVIB_1</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-0.999366</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SECFUPOU_2</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-0.999358</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SECFUPOU_1</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-0.999358</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ACTIVFIN_86</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>naf38_QA</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               SECROUT_1  SECROUT_2  CONDUITE_2  NBRKM_DRAP_ 0  CONDUITE_1  \\\n",
       "SECROUT_1            NaN  -0.999123         NaN            NaN         NaN   \n",
       "SECROUT_2      -0.999123        NaN         NaN            NaN         NaN   \n",
       "CONDUITE_2           NaN        NaN         NaN       0.999434   -0.999434   \n",
       "NBRKM_DRAP_ 0        NaN        NaN    0.999434            NaN   -1.000000   \n",
       "CONDUITE_1           NaN        NaN   -0.999434      -1.000000         NaN   \n",
       "NBRKM_DRAP_ 1        NaN        NaN   -0.920992      -0.921514    0.921514   \n",
       "csei_61              NaN        NaN         NaN            NaN         NaN   \n",
       "cser_6               NaN        NaN         NaN            NaN         NaN   \n",
       "CWVIB_2              NaN        NaN         NaN            NaN         NaN   \n",
       "CWVIB_1              NaN        NaN         NaN            NaN         NaN   \n",
       "SECFUPOU_2           NaN        NaN         NaN            NaN         NaN   \n",
       "SECFUPOU_1           NaN        NaN         NaN            NaN         NaN   \n",
       "ACTIVFIN_86          NaN        NaN         NaN            NaN         NaN   \n",
       "naf38_QA             NaN        NaN         NaN            NaN         NaN   \n",
       "\n",
       "               NBRKM_DRAP_ 1   csei_61    cser_6   CWVIB_2   CWVIB_1  \\\n",
       "SECROUT_1                NaN       NaN       NaN       NaN       NaN   \n",
       "SECROUT_2                NaN       NaN       NaN       NaN       NaN   \n",
       "CONDUITE_2         -0.920992       NaN       NaN       NaN       NaN   \n",
       "NBRKM_DRAP_ 0      -0.921514       NaN       NaN       NaN       NaN   \n",
       "CONDUITE_1          0.921514       NaN       NaN       NaN       NaN   \n",
       "NBRKM_DRAP_ 1            NaN       NaN       NaN       NaN       NaN   \n",
       "csei_61                  NaN       NaN  0.809746       NaN       NaN   \n",
       "cser_6                   NaN  0.809746       NaN       NaN       NaN   \n",
       "CWVIB_2                  NaN       NaN       NaN       NaN -0.999366   \n",
       "CWVIB_1                  NaN       NaN       NaN -0.999366       NaN   \n",
       "SECFUPOU_2               NaN       NaN       NaN       NaN       NaN   \n",
       "SECFUPOU_1               NaN       NaN       NaN       NaN       NaN   \n",
       "ACTIVFIN_86              NaN       NaN       NaN       NaN       NaN   \n",
       "naf38_QA                 NaN       NaN       NaN       NaN       NaN   \n",
       "\n",
       "               SECFUPOU_2  SECFUPOU_1  ACTIVFIN_86  naf38_QA  \n",
       "SECROUT_1             NaN         NaN          NaN       NaN  \n",
       "SECROUT_2             NaN         NaN          NaN       NaN  \n",
       "CONDUITE_2            NaN         NaN          NaN       NaN  \n",
       "NBRKM_DRAP_ 0         NaN         NaN          NaN       NaN  \n",
       "CONDUITE_1            NaN         NaN          NaN       NaN  \n",
       "NBRKM_DRAP_ 1         NaN         NaN          NaN       NaN  \n",
       "csei_61               NaN         NaN          NaN       NaN  \n",
       "cser_6                NaN         NaN          NaN       NaN  \n",
       "CWVIB_2               NaN         NaN          NaN       NaN  \n",
       "CWVIB_1               NaN         NaN          NaN       NaN  \n",
       "SECFUPOU_2            NaN   -0.999358          NaN       NaN  \n",
       "SECFUPOU_1      -0.999358         NaN          NaN       NaN  \n",
       "ACTIVFIN_86           NaN         NaN          NaN       1.0  \n",
       "naf38_QA              NaN         NaN          1.0       NaN  "
      ]
     },
     "execution_count": 160,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "filtered_corr_matrix"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 161,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABy8AAAfHCAYAAABlrCFOAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjkuMCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy80BEi2AAAACXBIWXMAAA9hAAAPYQGoP6dpAAEAAElEQVR4nOzdeZjVZcE//veZAYZNxpXFHUVQc98Ql9DEkHJBsySfEknNJTPlSQVNcenJNRXNLVFEc2lxJS0tlyzFzAXNfUNcQUUW2YU5vz/6Od8mQOfAjHPivF7Xda783Pd9PvM+w7muvHxz359CsVgsBgAAAAAAAKCFVbV0AAAAAAAAAIBEeQkAAAAAAACUCeUlAAAAAAAAUBaUlwAAAAAAAEBZUF4CAAAAAAAAZUF5CQAAAAAAAJQF5SUAAAAAAABQFpSXAAAAAAAAQFlQXgIAAAAAAABlQXkJAAAt6Nprr02hUMgbb7zRZPd84403UigUcu211zbZPf/b7bLLLtlll11aOgYAAADwOZSXAAAsd1577bUcfvjhWW+99dK2bdt06tQpO+64Y0aOHJk5c+a0dLwmc+ONN+aiiy5q6RgNHHzwwSkUCunUqdNif9evvPJKCoVCCoVCzj///JLv/+677+a0007L+PHjmyDt0isUCjn66KMXO/dpIf344483288vl98DAAAANLVWLR0AAACa0l133ZVvfvObqampyUEHHZRNNtkk8+fPz9/+9rccf/zxee655/LLX/6ypWM2iRtvvDHPPvtsjj322Abj66yzTubMmZPWrVu3SK5WrVpl9uzZGTt2bL71rW81mLvhhhvStm3bzJ07d6nu/e677+b000/Puuuumy222KLR77v33nuX6ueVq6X9PQAAAEC5U14CALDcmDBhQgYNGpR11lkn999/f7p161Y/94Mf/CCvvvpq7rrrrmX+OcViMXPnzk27du0WmZs7d27atGmTqqqWO+SkUCikbdu2Lfbza2pqsuOOO+amm25apLy88cYb8/Wvfz233HLLF5Jl9uzZad++fdq0afOF/DwAAABg2Tg2FgCA5ca5556bmTNn5uqrr25QXH6qR48e+dGPflR/vWDBgpx55plZf/31U1NTk3XXXTcnnXRS5s2b1+B96667bvbcc8/cc8892WabbdKuXbtceeWVefDBB1MoFHLzzTfnJz/5SdZYY420b98+M2bMSJL8/e9/zx577JHa2tq0b98+ffv2zcMPP/y5n+OOO+7I17/+9ay++uqpqanJ+uuvnzPPPDMLFy6sX7PLLrvkrrvuysSJE+uPYV133XWTLPmZl/fff3923nnndOjQISuuuGL22WefvPDCCw3WnHbaaSkUCnn11Vdz8MEHZ8UVV0xtbW2GDBmS2bNnf272Tx144IH5wx/+kGnTptWP/eMf/8grr7ySAw88cJH1H330UX784x9n0003TceOHdOpU6cMGDAgTz/9dP2aBx98MNtuu22SZMiQIfWf+9PPucsuu2STTTbJE088kS9/+ctp3759TjrppPq5f3/m5eDBg9O2bdtFPn///v2z0kor5d133230Z22sF198Mfvvv39WXnnltG3bNttss03uvPPOZvs9PPPMM+nbt2/at2+fHj165He/+12S5C9/+Ut69+6ddu3apVevXvnzn//cIMPEiRNz1FFHpVevXmnXrl1WWWWVfPOb31zkuayfHo/70EMP5fDDD88qq6ySTp065aCDDsrUqVOb+LcHAABApbDzEgCA5cbYsWOz3nrrZYcddmjU+kMPPTRjxozJ/vvvn//93//N3//+95x11ll54YUXcttttzVY+9JLL+Xb3/52Dj/88Bx22GHp1atX/dyZZ56ZNm3a5Mc//nHmzZuXNm3a5P7778+AAQOy9dZbZ8SIEamqqsro0aPzla98JX/961+z3XbbLTHXtddem44dO2bo0KHp2LFj7r///px66qmZMWNGzjvvvCTJySefnOnTp+ftt9/OhRdemCTp2LHjEu/55z//OQMGDMh6662X0047LXPmzMkll1ySHXfcMU8++WR98fmpb33rW+nevXvOOuusPPnkkxk1alQ6d+6cc845p1G/2/322y9HHHFEbr311nzve99L8q9dlxtuuGG22mqrRda//vrruf322/PNb34z3bt3z+TJk3PllVemb9++ef7557P66qtno402yhlnnJFTTz013//+97PzzjsnSYM/7ylTpmTAgAEZNGhQvvOd76RLly6LzTdy5Mjcf//9GTx4cMaNG5fq6upceeWVuffee3P99ddn9dVX/9zPOHfu3Hz44YeLjM+cOXORseeeey477rhj1lhjjQwbNiwdOnTIb37zmwwcODC33HJL9t133yb9PUydOjV77rlnBg0alG9+85u5/PLLM2jQoNxwww059thjc8QRR+TAAw/Meeedl/333z9vvfVWVlhhhST/KpkfeeSRDBo0KGuuuWbeeOONXH755dlll13y/PPPp3379g0+29FHH50VV1wxp512Wl566aVcfvnlmThxYn25DwAAACUpAgDAcmD69OnFJMV99tmnUevHjx9fTFI89NBDG4z/+Mc/LiYp3n///fVj66yzTjFJ8Y9//GODtQ888EAxSXG99dYrzp49u368rq6uuMEGGxT79+9frKurqx+fPXt2sXv37sXdd9+9fmz06NHFJMUJEyY0WPefDj/88GL79u2Lc+fOrR/7+te/XlxnnXUWWTthwoRikuLo0aPrx7bYYoti586di1OmTKkfe/rpp4tVVVXFgw46qH5sxIgRxSTF733vew3uue+++xZXWWWVRX7Wfxo8eHCxQ4cOxWKxWNx///2Lu+22W7FYLBYXLlxY7Nq1a/H000+vz3feeefVv2/u3LnFhQsXLvI5ampqimeccUb92D/+8Y9FPtun+vbtW0xSvOKKKxY717dv3wZj99xzTzFJ8ac//Wnx9ddfL3bs2LE4cODAz/2MxWKxmORzX//4xz/q1++2227FTTfdtMGfX11dXXGHHXYobrDBBs3ye7jxxhvrx1588cVikmJVVVXx0UcfXeR38O/3Wdz3b9y4ccUkxeuuu65+7NPv7tZbb12cP39+/fi5555bTFK84447lvTrAwAAgCVybCwAAMuFT49q/XT32Oe5++67kyRDhw5tMP6///u/SbLIszG7d++e/v37L/ZegwcPbvD8y/Hjx9cfjzplypR8+OGH+fDDDzNr1qzstttueeihh1JXV7fEbP9+r48//jgffvhhdt5558yePTsvvvhioz7fv3vvvfcyfvz4HHzwwVl55ZXrxzfbbLPsvvvu9b+Lf3fEEUc0uN55550zZcqU+t9zYxx44IF58MEHM2nSpNx///2ZNGnSYo+MTf71nMxPnxO6cOHCTJkyJR07dkyvXr3y5JNPNvpn1tTUZMiQIY1a+9WvfjWHH354zjjjjOy3335p27Ztrrzyykb/rH322Sd/+tOfFnkdf/zxDdZ99NFHuf/++/Otb32r/s/zww8/zJQpU9K/f/+88soreeedd+rzN8XvoWPHjhk0aFD9da9evbLiiitmo402Su/evevHP/3n119/vX7s379/n3zySaZMmZIePXpkxRVXXGyG73//+2ndunX99ZFHHplWrVot9nsFAAAAn8exsQAALBc6deqU5F9lX2NMnDgxVVVV6dGjR4Pxrl27ZsUVV8zEiRMbjHfv3n2J9/rPuVdeeSXJv0rNJZk+fXpWWmmlxc4999xz+clPfpL7779/kbJw+vTpS7znknz6Wf79qNtPbbTRRrnnnnsya9asdOjQoX587bXXbrDu06xTp06t/11/nq997WtZYYUV8utf/zrjx4/Ptttumx49eizy7MQkqaury8iRI3PZZZdlwoQJDZ7vucoqqzTq5yXJGmuskTZt2jR6/fnnn5877rgj48ePz4033pjOnTs3+r1rrrlm+vXrt8j422+/3eD61VdfTbFYzCmnnJJTTjllsfd6//33s8YaazTZ72HNNddc5MjW2trarLXWWouMJWnwjMo5c+bkrLPOyujRo/POO++kWCzWzy3u+7fBBhs0uO7YsWO6deu22D9nAAAA+DzKSwAAlgudOnXK6quvnmeffbak9zX2mXz/vhvt8+Y+3VV53nnnZYsttljse5b0fMpp06alb9++6dSpU84444ysv/76adu2bZ588smceOKJn7ljsylVV1cvdvzfi6zPU1NTk/322y9jxozJ66+/ntNOO22Ja3/2s5/llFNOyfe+972ceeaZWXnllVNVVZVjjz22pM/8WX9Oi/PUU0/l/fffT5L885//zLe//e2S3t8Yn+b/8Y9/vMTdu5+W6E31e1jSn19j/lx/+MMfZvTo0Tn22GPTp0+f1NbWplAoZNCgQV/Y9w8AAIDKpbwEAGC5seeee+aXv/xlxo0blz59+nzm2nXWWSd1dXV55ZVXstFGG9WPT548OdOmTcs666yz1DnWX3/9JP8qVBe3M++zPPjgg5kyZUpuvfXWfPnLX64fnzBhwiJrG1u8fvpZXnrppUXmXnzxxay66qoNdl02pQMPPDDXXHNNqqqqGhxj+p9+97vfZdddd83VV1/dYHzatGlZddVV668b+5kbY9asWRkyZEg23njj7LDDDjn33HOz7777Ztttt22yn5Ek6623XpKkdevWn/t9aInfw+IyDB48OD//+c/rx+bOnZtp06Ytdv0rr7ySXXfdtf565syZee+99/K1r32t2TICAACw/PLMSwAAlhsnnHBCOnTokEMPPTSTJ09eZP61117LyJEjk6S+WLnooosarLnggguSJF//+teXOsfWW2+d9ddfP+eff35mzpy5yPwHH3ywxPd+ujPu33fCzZ8/P5dddtkiazt06NCoY2S7deuWLbbYImPGjGlQQD377LO59957m7Vk2nXXXXPmmWfmF7/4Rbp27brEddXV1Yvs6vztb39b/yzIT31asi6pSCvFiSeemDfffDNjxozJBRdckHXXXTeDBw/OvHnzlvne/65z587ZZZddcuWVV+a9995bZP7fvw8t8Xv4T4vLcMkllzQ4wvbf/fKXv8wnn3xSf3355ZdnwYIFGTBgQJNnAwAAYPln5yUAAMuN9ddfPzfeeGMOOOCAbLTRRjnooIOyySabZP78+XnkkUfy29/+NgcffHCSZPPNN8/gwYPzy1/+sv6o1sceeyxjxozJwIEDG+wkK1VVVVVGjRqVAQMG5Etf+lKGDBmSNdZYI++8804eeOCBdOrUKWPHjl3se3fYYYestNJKGTx4cI455pgUCoVcf/31iz2udeutt86vf/3rDB06NNtuu206duyYvfbaa7H3Pe+88zJgwID06dMnhxxySObMmZNLLrkktbW1n3mc67KqqqrKT37yk89dt+eee+aMM87IkCFDssMOO+Sf//xnbrjhhvpdi59af/31s+KKK+aKK67ICiuskA4dOqR3796f+UzSxbn//vtz2WWXZcSIEdlqq62SJKNHj84uu+ySU045Jeeee25J9/s8l156aXbaaadsuummOeyww7Leeutl8uTJGTduXN5+++08/fTTSb7438Pi7Lnnnrn++utTW1ubjTfeOOPGjcuf//znJT5zc/78+dltt93yrW99Ky+99FIuu+yy7LTTTtl7772XOQsAAACVR3kJAMByZe+9984zzzyT8847L3fccUcuv/zy1NTUZLPNNsvPf/7zHHbYYfVrR40alfXWWy/XXnttbrvttnTt2jXDhw/PiBEjljnHLrvsknHjxtXvOpw5c2a6du2a3r175/DDD1/i+1ZZZZX8/ve/z//+7//mJz/5SVZaaaV85zvfyW677bbI8xKPOuqojB8/PqNHj86FF16YddZZZ4nlZb9+/fLHP/4xI0aMyKmnnprWrVunb9++Oeecc5qk8FpWJ510UmbNmpUbb7wxv/71r7PVVlvlrrvuyrBhwxqsa926dcaMGZPhw4fniCOOyIIFCzJ69OiSPsPHH3+c733ve9lyyy1z8skn14/vvPPO+dGPfpSf//zn2W+//bL99ts32efbeOON8/jjj+f000/PtddemylTpqRz587Zcsstc+qpp9av+yJ/D0sycuTIVFdX54YbbsjcuXOz44475s9//vMSn9f5i1/8IjfccENOPfXUfPLJJ/n2t7+diy++uFmPtgUAAGD5VSgu7q9wAwAAwGe49tprM2TIkPzjH//INtts09JxAAAAWE545iUAAAAAAABQFpSXAAAAAAAAQFlQXgIAAAAAAABlQXkJAABAyQ4++OAUi0XPuwQAACgDDz30UPbaa6+svvrqKRQKuf322z/3PQ8++GC22mqr1NTUpEePHrn22msXWXPppZdm3XXXTdu2bdO7d+889thjTR/+PygvAQAAAAAA4L/YrFmzsvnmm+fSSy9t1PoJEybk61//enbdddeMHz8+xx57bA499NDcc8899Wt+/etfZ+jQoRkxYkSefPLJbL755unfv3/ef//95voYSZJCsVgsNutPAAAAAAAAAL4QhUIht912WwYOHLjENSeeeGLuuuuuPPvss/VjgwYNyrRp0/LHP/4xSdK7d+9su+22+cUvfpEkqaury1prrZUf/vCHGTZsWLPlt/MSAAAAAAAAysy8efMyY8aMBq958+Y1yb3HjRuXfv36NRjr379/xo0blySZP39+nnjiiQZrqqqq0q9fv/o1zaVVs96dirDTXn9p6Qgshb+N7dvSEQAAAAAAKGN3te7V0hEq2j9O/nZOP/30BmMjRozIaaedtsz3njRpUrp06dJgrEuXLpkxY0bmzJmTqVOnZuHChYtd8+KLLy7zz/8syksAAAAAAAAoM8OHD8/QoUMbjNXU1LRQmi+O8hIAAAAAAADKTE1NTbOVlV27ds3kyZMbjE2ePDmdOnVKu3btUl1dnerq6sWu6dq1a7Nk+pRnXgIAAAAAAEAF6dOnT+67774GY3/605/Sp0+fJEmbNm2y9dZbN1hTV1eX++67r35Nc1FeAgAAAAAAwH+xmTNnZvz48Rk/fnySZMKECRk/fnzefPPNJP86gvaggw6qX3/EEUfk9ddfzwknnJAXX3wxl112WX7zm9/kuOOOq18zdOjQXHXVVRkzZkxeeOGFHHnkkZk1a1aGDBnSrJ/FsbEAAAAAAADwX+zxxx/PrrvuWn/96bMyBw8enGuvvTbvvfdefZGZJN27d89dd92V4447LiNHjsyaa66ZUaNGpX///vVrDjjggHzwwQc59dRTM2nSpGyxxRb54x//mC5dujTrZykUi8Vis/4Elns77fWXlo7AUvjb2L4tHQEAAAAAgDJ2V+teLR2hon39k5daOkKLcGwsAAAAAAAAUBYcGwsAAAAAAMAiCq0LLR2BCmTnJQAAAAAAAFAWlJcAAAAAAABAWVBeAgAAAAAAAGVBeQkAAAAAAACUBeUlAAAAAAAAUBaUlwAAAAAAAEBZaNXSAQAAAAAAACg/Va0KLR2BCmTnJQAAAAAAAFAWlJcAAAAAAABAWVBeAgAAAAAAAGVBeQkAAAAAAACUBeUlAAAAAAAAUBaUlwAAAAAAAEBZUF4CAAAAAAAAZaFVSwcAAAAAAACg/BRa2wPHF8+3DgAAAAAAACgLyksAAAAAAACgLCgvAQAAAAAAgLKgvAQAAAAAAADKgvISAAAAAAAAKAvKSwAAAAAAAKAstGrpAAAAAAAAAJSfqlaFlo5ABWrSnZdPP/10qqurm/KWAAAAAAAAQIVo8mNji8ViU98SAAAAAAAAqAAlHRu73377feb89OnTUyjYQgwAAAAAAACUrqTycuzYsdl9993TpUuXxc4vXLiwSUIBAAAAAAAAlaek8nKjjTbKN77xjRxyyCGLnR8/fnx+//vfN0kwAAAAAAAAoLKU9MzLrbfeOk8++eQS52tqarL22msvcygAAAAAAACg8pS08/KKK674zKNhN9poo0yYMGGZQwEAAAAAAACVp6TysqampqSbn3322TniiCOy4oorlvQ+AAAAAAAAWlahdaGlI1CBSjo2tlQ/+9nP8tFHHzXnjwAAAAAAAACWE81aXhaLxea8PQAAAAAAALAcadbyEgAAAAAAAKCxlJcAAAAAAABAWVBeAgAAAAAAAGVBeQkAAAAAAACUhValLL7uuutywAEHpKamplHrd95557Rr126pggEAAAAAANByqloVWjoCFaiknZdDhgzJ9OnTG73+7rvvTrdu3UoOBQAAAAAAAFSeksrLYrHYXDkAAAAAAACAClfyMy8LBVuEAQAAAAAAgKZX0jMvk2S33XZLq1af/bYnn3xyqQMBAAAAAAAAlank8rJ///7p2LFjc2QBAAAAAAAAKljJ5eXxxx+fzp07N0cWAAAAAAAAoIKV9MxLz7sEAAAAAAAAmktJOy+LxWJz5QAAAAAAAKCMFFrb1MYXr6SdlxMmTMhqq63WXFkAAAAAAACAClbSzsvjjjtuseO1tbXp2bNnDj30UOUmAAAAAAAAsFRK2nlZW1u72Ne0adNy1VVXpVevXnn22WebKysAAAAAAACwHCtp5+Xo0aOXOFdXV5fDDjssw4cPz9ixY5c5GAAAAAAAAFBZStp5+Zk3qqrKMccckyeeeKKpbgkAAAAAAABUkCYrL5OkQ4cOmT17dlPeEgAAAAAAAKgQJR0b+3n+9Kc/pWfPnk15SwAAAAAAAFpAVatCS0egApVUXt55552LHZ8+fXqeeOKJjBo1KqNGjWqSYAAAAAAAAEBlKam8HDhw4GLHV1hhhfTq1SujRo3KoEGDmiIXAAAAAAAAUGFKKi/r6uqaKweUnS/3WTUDB3RLr/VXSG2n1jn4mMfz6oRZLR0LAAAAAABguVXV0gGgXLVrW5Vnnp+Ry8e83tJRAAAAAAAAKkLJ5eWCBQty3nnnZauttkrHjh3TsWPHbLXVVjn//PPzySefNEdGaBH3PPB+rr15Yh4fP7WlowAAAAAAAFSEko6NnTNnTnbfffeMGzcu/fr1y5e//OUkyQsvvJATTzwxd955Z+699960bdu2WcICAAAAAAAAy6+Sysuzzz47b731Vp566qlsttlmDeaefvrp7L333jn77LNz2mmnNWVGAAAAAAAAoAKUVF7efPPNueCCCxYpLpNk8803z/nnn5+TTz75c8vLefPmZd68eQ3GampqUlNTU0ocaDK79+2c43/Qs/76x6f9M888P70FEwEAAAAAQMsqVBdaOgIVqKTycuLEidluu+2WOL/99tvnzTff/Nz7nHXWWTn99NMbjI0YMcKOTVrM3x6bkudffrz++oMp81swDQAAAAAAQGUqqbzs1KlT3n///ay11lqLnZ80aVJWWGGFz73P8OHDM3To0AZjdl3SkubMWZh35ixs6RgAAAAAAAAVraTyctddd83Pfvaz3HLLLYudP/vss7Prrrt+7n0cEct/gxU6tkqX1Wqy6sr/+q6uvUb7JMlHU+fno2mftGQ0AAAAAACA5VJJ5eWIESPSu3fvbL/99hk6dGg23HDDFIvFvPDCC7nwwgvz/PPP59FHH22urPCF2qn3Kjn52A3rr884ceMkyTU3vpFrbprYUrEAAAAAAACWWyWVlxtvvHH+9Kc/5ZBDDsmgQYNSKPzrQa3FYjEbbrhh7r333nzpS19qlqDwRfvDfZPzh/smt3QMAAAAAACAilFSeZkk22+/fZ577rmMHz8+L7/8cpKkZ8+e2WKLLZo6GwAAAAAAAFBBSi4vP7XFFlsstrB8/PHHs8022yxLJgAAAAAAAFpYVXWhpSNQgaqW5k0zZ87MnDlzGoyNHz8+e+21V3r37t0kwQAAAAAAAIDKUlJ5+dZbb6VPnz6pra1NbW1thg4dmtmzZ+eggw5K796906FDhzzyyCPNlRUAAAAAAABYjpV0bOzxxx+fuXPnZuTIkbn11lszcuTI/PWvf03v3r3z2muvZc0112yunAAAAAAAAMByrqTy8qGHHsqtt96a7bffPt/61rfStWvX/M///E+OPfbYZooHAAAAAAAAVIqSjo2dPHlyunfvniTp3Llz2rdvnwEDBjRLMAAAAAAAAKCylFReJklVVVWDf27Tpk2TBgIAAAAAAAAqU0nHxhaLxfTs2TOFQiFJMnPmzGy55ZYNCs0k+eijj5ouIQAAAAAAAFARSiovR48e3Vw5AAAAAAAAKCOFqkJLR6AClVReDh48uLlyAAAAAAAAABWupGdeTp06NZdccklmzJixyNz06dOXOAcAAAAAAADweUoqL3/xi1/koYceSqdOnRaZq62tzV//+tdccsklTRYOAAAAAAAAqBwllZe33HJLjjjiiCXOH3744fnd7363zKEAAAAAAACAylNSefnaa69lgw02WOL8BhtskNdee22ZQwEAAAAAAACVp6Tysrq6Ou++++4S5999991UVZV0SwAAAAAAAIAkSatSFm+55Za5/fbbs/322y92/rbbbsuWW27ZJMEAAAAAAABoOYVqG9b44pVUXh599NEZNGhQ1lxzzRx55JGprq5OkixcuDCXXXZZLrzwwtx4443NEhQAAAAAAABYvpVUXn7jG9/ICSeckGOOOSYnn3xy1ltvvSTJ66+/npkzZ+b444/P/vvv3yxBAQAAAAAAgOVbSeVlkvzf//1f9tlnn9xwww159dVXUywW07dv3xx44IHZbrvtmiMjAAAAAAAAUAFKLi+TZLvttlNUAgAAAAAAAE2qpCetnnvuuZkzZ0799cMPP5x58+bVX3/88cc56qijmi4dAAAAAAAAUDFKKi+HDx+ejz/+uP56wIABeeedd+qvZ8+enSuvvLLp0gEAAAAAAAAVo6TyslgsfuY1AAAAAAAAwNJaqmdeAgAAAAAAsHyrqi60dAQqUEk7LwEAAAAAAACaS8k7L0eNGpWOHTsmSRYsWJBrr702q666apI0eB4mAAAAAAAAQClKKi/XXnvtXHXVVfXXXbt2zfXXX7/IGgAAAAAAAIBSlVRevvHGG80UAwAAAAAAAKh0JT3zcty4cfn973/fYOy6665L9+7d07lz53z/+9/PvHnzmjQgAAAAAAAAUBlKKi9PP/30PPfcc/XX//znP3PIIYekX79+GTZsWMaOHZuzzjqryUMCAAAAAAAAy7+Sjo19+umn89Of/rT++uabb07v3r3rn4O51lprZcSIETnttNOaNCQAAAAAAABfrEJVoaUjUIFK2nk5derUdOnSpf76L3/5SwYMGFB/ve222+att95qunQAAAAAAABAxSipvOzSpUsmTJiQJJk/f36efPLJbL/99vXzH3/8cVq3bt20CQEAAAAAAICKUFJ5+bWvfS3Dhg3LX//61wwfPjzt27fPzjvvXD//zDPPZP3112/ykAAAAAAAAMDyr6RnXp555pnZb7/90rdv33Ts2DFjxoxJmzZt6uevueaafPWrX23ykAAAAAAAAMDyr6TyctVVV81DDz2U6dOnp2PHjqmurm4w/9vf/jYdO3Zs0oAAAAAAAABAZSipvPxUbW3tYsdXXnnlZQoDAAAAAAAAVK6SnnkJAAAAAAAA0FyWauclAAAAAAAAy7eq6kJLR6AC2XkJAAAAAAAAlAXlJQAAAAAAAFAWlJcAAAAAAABAWVBeAgAAAAAAAGVBeQkAAAAAAACUBeUlAAAAAAAAUBZatXQAAAAAAAAAyk+hutDSEahAdl4CAAAAAAAAZUF5CQAAAAAAAJQF5SUAAAAAAABQFpSXAAAAAAAAQFlQXgIAAAAAAABlQXkJAAAAAAAAlAXlJQAAAAAAAFAWWrV0AAAAAAAAAMpPocoeOL54vnUAAAAAAABAWVBeAgAAAAAAAGVBeQkAAAAAAACUBeUlAAAAAAAAUBaUlwAAAAAAAEBZUF4CAAAAAAAAZaFVSwcAAAAAAACg/BSqCi0dgQpk5yUAAAAAAABQFpSXAAAAAAAAQFkom2Njd9rrLy0dgaX0t7F9WzoCAAAAAAAAywE7LwEAAAAAAICyoLwEAAAAAAAAyoLyEgAAAAAAACgLyksAAAAAAACgLLRq6QAAAAAAAACUn6rqQktHoALZeQkAAAAAAACUBeUlAAAAAAAAUBaUlwAAAAAAAEBZUF4CAAAAAAAAZUF5CQAAAAAAAJQF5SUAAAAAAABQFpSXAAAAAAAAQFlo1dIBAAAAAAAAKD+FqkJLR6AC2XkJAAAAAAAAlAXlJQAAAAAAAFAWlJcAAAAAAABAWVBeAgAAAAAAAGVBeQkAAAAAAACUBeUlAAAAAAAAUBZatXQAAAAAAAAAyk+hyh44vni+dQAAAAAAAEBZUF4CAAAAAAAAZUF5CQAAAAAAAJQF5SUAAAAAAABQFpSXAAAAAAAAQFlQXgIAAAAAAABlQXkJAAAAAAAAlIVWLR0AAAAAAACA8lOoKrR0BCqQnZcAAAAAAABAWVBeAgAAAAAAAGWh5PLy7rvvzqGHHpoTTjghL774YoO5qVOn5itf+UqThQMAAAAAAAAqR0nl5Y033pi99947kyZNyrhx47LlllvmhhtuqJ+fP39+/vKXvzR5SAAAAAAAAGD516qUxeedd14uuOCCHHPMMUmS3/zmN/ne976XuXPn5pBDDmmWgAAAAAAAAEBlKKm8fOWVV7LXXnvVX3/rW9/Kaqutlr333juffPJJ9t133yYPCAAAAAAAAFSGksrLTp06ZfLkyenevXv92K677prf//732XPPPfP22283eUAAAAAAAAC+eFXVhZaOQAUq6ZmX2223Xf7whz8sMt63b9+MHTs2F110UVPlAgAAAAAAACpMSeXlcccdl7Zt2y52bpdddsnYsWNz0EEHNUkwAAAAAAAAoLKUVF727ds3w4cPX+L8rrvumtGjR9dfn3322Zk2bdpShwMAAAAAAAAqR0nlZal+9rOf5aOPPmrOHwEAAAAAAAAsJ5q1vCwWi815ewAAAAAAAGA50qzlJQAAAAAAAEBjKS8BAAAAAACAstCqpQMAAAAAAABQfgpVhZaOQAWy8xIAAAAAAAAoCyWVl9ddd13mzZvX6PU777xz2rVrV3IoAAAAAAAAoPKUVF4OGTIk06dPb/T6u+++O926dSs5FAAAAAAAAFB5Siovi8Vic+UAAAAAAAAAKlzJz7wsFDycFQAAAAAAAGh6rUp9w2677ZZWrT77bU8++eRSBwIAAAAAAAAqU8nlZf/+/dOxY8fmyAIAAAAAAECZKFSVfIAnLLOSy8vjjz8+nTt3bo4sAAAAAAAAQAUrqTL3vEsAAAAAAACguZRUXhaLxebKAQAAAAAAAFS4ksrLCRMmZLXVVmuuLAAAAAAAAEAFK+mZl8cdd9xix2tra9OzZ88ceuihyk0AAAAAAABgqZS087K2tnaxr2nTpuWqq65Kr1698uyzzzZXVgAAAAAAAGA5VtLOy9GjRy9xrq6uLocddliGDx+esWPHLnMwAAAAAAAAoLKUVF5+lqqqqhxzzDEZMGBAU90SAAAAAACAFlKoKrR0BCpQScfGfp4OHTpk9uzZTXlLAAAAAAAAoEI0aXn5pz/9KT179mzKWwIAAAAAAAAVoqRjY++8887Fjk+fPj1PPPFERo0alVGjRjVJMAAAAAAAAKCylFReDhw4cLHjK6ywQnr16pVRo0Zl0KBBTZELAAAAAAAAqDAllZd1dXXNlWO59eU+q2bggG7ptf4Kqe3UOgcf83henTCrpWMBAAAAAABA2WnSZ16yqHZtq/LM8zNy+ZjXWzoKAAAAAAAAlLWSdl4myYIFC3LhhRfmpptuyssvv5wk6dmzZw488MD86Ec/SuvWrZs85H+zex54P0nStXNNCycBAAAAAABovEJVoaUjUIFKKi/nzJmT3XffPePGjUu/fv3y5S9/OUnywgsv5MQTT8ydd96Ze++9N23btm2WsAAAAAAAAMDyq6Ty8uyzz85bb72Vp556KptttlmDuaeffjp77713zj777Jx22mmfeZ958+Zl3rx5DcbqFs5PVXWbUuIAAAAAAAAAy5GSnnl5880354ILLlikuEySzTffPOeff35uvPHGz73PWWedldra2gavt1+9oZQoZWn3vp1z7292qn9ttnFtS0cCAAAAAACA/xol7bycOHFitttuuyXOb7/99nnzzTc/9z7Dhw/P0KFDG4ztMejvpUQpS397bEqef/nx+usPpsxvwTQAAAAAAADw36Wk8rJTp055//33s9Zaay12ftKkSVlhhRU+9z41NTWpqalpMLY8HBk7Z87CvDNnYUvHAAAAAAAAgP9KJR0bu+uuu+ZnP/vZEufPPvvs7LrrrsscanmyQsdW6dG9Q9Zdq0OSZO012qdH9w5ZecXWLZwMAAAAAAAAyktJOy9HjBiR3r17Z/vtt8/QoUOz4YYbplgs5oUXXsiFF16Y559/Po8++mhzZf2vtFPvVXLysRvWX59x4sZJkmtufCPX3DSxpWIBAAAAAABA2SmpvNx4443zpz/9KYccckgGDRqUQqGQJCkWi9lwww1z77335ktf+lKzBP1v9Yf7JucP901u6RgAAAAAAAAlKVQVWjoCFaik8jJJtt9++zz33HMZP358Xn755SRJz549s8UWWzR1NgAAAAAAAKCClFxefmqLLbZYbGH5+OOPZ5tttlmWTAAAAAAAAEAFqlqaN82cOTNz5sxpMDZ+/Pjstdde6d27d5MEAwAAAAAAACpLSeXlW2+9lT59+qS2tja1tbUZOnRoZs+enYMOOii9e/dOhw4d8sgjjzRXVgAAAAAAAGA5VtKxsccff3zmzp2bkSNH5tZbb83IkSPz17/+Nb17985rr72WNddcs7lyAgAAAAAAAMu5ksrLhx56KLfeemu23377fOtb30rXrl3zP//zPzn22GObKR4AAAAAAABQKUoqLydPnpzu3bsnSTp37pz27dtnwIABzRIMAAAAAACAllOoKunpg9AkSv7WVf3bF7Wqqipt2rRp0kAAAAAAAABAZSpp52WxWEzPnj1TKBSSJDNnzsyWW27ZoNBMko8++qjpEgIAAAAAAAAVoaTycvTo0c2VAwAAAAAAAKhwJZWXgwcPbq4cAAAAAAAAQIUr6ZmXU6dOzSWXXJIZM2YsMjd9+vQlzgEAAAAAAAB8npLKy1/84hd56KGH0qlTp0Xmamtr89e//jWXXHJJk4UDAAAAAAAAKkdJ5eUtt9ySI444Yonzhx9+eH73u98tcygAAAAAAACg8pT0zMvXXnstG2ywwRLnN9hgg7z22mvLHAoAAAAAAICWVVVdaOkIVKCSdl5WV1fn3XffXeL8u+++m6qqkm4JAAAAAAAAkKTE8nLLLbfM7bffvsT52267LVtuueWyZgIAAAAAAAAqUEnHxh599NEZNGhQ1lxzzRx55JGprq5OkixcuDCXXXZZLrzwwtx4443NEhQAAAAAAABYvpVUXn7jG9/ICSeckGOOOSYnn3xy1ltvvSTJ66+/npkzZ+b444/P/vvv3yxBAQAAAAAAgOVbSeVlkvzf//1f9tlnn9xwww159dVXUywW07dv3xx44IHZbrvtmiMjAAAAAAAAUAFKLi+TZLvttlNUAgAAAAAAAE2qqpTF5557bubMmVN//fDDD2fevHn11x9//HGOOuqopksHAAAAAABAiyhUFbxa8FWpSiovhw8fno8//rj+esCAAXnnnXfqr2fPnp0rr7yy6dIBAAAAAAAAFaOk8rJYLH7mNQAAAAAAAMDSKqm8BAAAAAAAAMrTpZdemnXXXTdt27ZN796989hjjy1x7S677JJCobDI6+tf/3r9moMPPniR+T322KNZP0OrZr07AAAAAAAA0Ox+/etfZ+jQobniiivSu3fvXHTRRenfv39eeumldO7ceZH1t956a+bPn19/PWXKlGy++eb55je/2WDdHnvskdGjR9df19TUNN+HyFKUl6NGjUrHjh2TJAsWLMi1116bVVddNUkaPA8TAAAAAAAA+GJccMEFOeywwzJkyJAkyRVXXJG77ror11xzTYYNG7bI+pVXXrnB9c0335z27dsvUl7W1NSka9euzRf8P5RUXq699tq56qqr6q+7du2a66+/fpE1AAAAAAAAwBdj/vz5eeKJJzJ8+PD6saqqqvTr1y/jxo1r1D2uvvrqDBo0KB06dGgw/uCDD6Zz585ZaaWV8pWvfCU//elPs8oqqzRp/n9XUnn5xhtvNFMMAAAAAAAA4FPz5s3LvHnzGozV1NQs9tjWDz/8MAsXLkyXLl0ajHfp0iUvvvji5/6sxx57LM8++2yuvvrqBuN77LFH9ttvv3Tv3j2vvfZaTjrppAwYMCDjxo1LdXX1Unyqz1dVyuJx48bl97//fYOx6667Lt27d0/nzp3z/e9/f5FfIgAAAAAAAP99ClVVXi34Ouuss1JbW9vgddZZZzXLn/XVV1+dTTfdNNttt12D8UGDBmXvvffOpptumoEDB+b3v/99/vGPf+TBBx9slhxJieXl6aefnueee67++p///GcOOeSQ9OvXL8OGDcvYsWOb7ZcGAAAAAAAAlWL48OGZPn16g9e/Hwv771ZdddVUV1dn8uTJDcYnT578uc+rnDVrVm6++eYccsghn5tpvfXWy6qrrppXX3218R+kRCWVl08//XR22223+uubb745vXv3zlVXXZWhQ4fm4osvzm9+85smDwkAAAAAAACVpKamJp06dWrwWtyRsUnSpk2bbL311rnvvvvqx+rq6nLfffelT58+n/lzfvvb32bevHn5zne+87mZ3n777UyZMiXdunUr7cOUoKTycurUqQ3Oyv3LX/6SAQMG1F9vu+22eeutt5ouHQAAAAAAAPC5hg4dmquuuipjxozJCy+8kCOPPDKzZs3KkCFDkiQHHXTQYnduXn311Rk4cGBWWWWVBuMzZ87M8ccfn0cffTRvvPFG7rvvvuyzzz7p0aNH+vfv32yfo1Upi7t06ZIJEyZkrbXWyvz58/Pkk0/m9NNPr5//+OOP07p16yYPCQAAAAAAACzZAQcckA8++CCnnnpqJk2alC222CJ//OMf6zcmvvnmm6mqariv8aWXXsrf/va33HvvvYvcr7q6Os8880zGjBmTadOmZfXVV89Xv/rVnHnmmUvcAdoUSiovv/a1r2XYsGE555xzcvvtt6d9+/bZeeed6+efeeaZrL/++k0eEgAAAAAAAPhsRx99dI4++ujFzj344IOLjPXq1SvFYnGx69u1a5d77rmnKeM1Sknl5Zlnnpn99tsvffv2TceOHTNmzJi0adOmfv6aa67JV7/61SYPCQAAAAAAACz/SiovV1111Tz00EOZPn16OnbsmOrq6gbzv/3tb9OxY8cmDQgAAAAAAMAXr1BVaOkIVKCSystP1dbWLnZ85ZVXXqYwAAAAAAAAQOWq+vwlAAAAAAAAAM1PeQkAAAAAAACUBeUlAAAAAAAAUBaUlwAAAAAAAEBZUF4CAAAAAAAAZUF5CQAAAAAAAJSFVi0dAAAAAAAAgPJTqCq0dAQqkJ2XAAAAAAAAQFlQXgIAAAAAAABlQXkJAAAAAAAAlAXlJQAAAAAAAFAWlJcAAAAAAABAWVBeAgAAAAAAAGWhVUsHAAAAAAAAoPwUquyB44vnWwcAAAAAAACUBeUlAAAAAAAAUBaUlwAAAAAAAEBZUF4CAAAAAAAAZUF5CQAAAAAAAJQF5SUAAAAAAABQFpSXAAAAAAAAQFlo1dIBAAAAAAAAKD+FqkJLR6AC2XkJAAAAAAAAlAXlJQAAAAAAAFAWlJcAAAAAAABAWSibZ17+bWzflo4A8F/hrta9WjoCS+msPX7Z0hFYCh1Wrm3pCCyle8Zs0dIRAAAAACiRnZcAAAAAAABAWVBeAgAAAAAAAGWhbI6NBQAAAAAAoHwUquyB44vnWwcAAAAAAACUBeUlAAAAAAAAUBaUlwAAAAAAAEBZUF4CAAAAAAAAZUF5CQAAAAAAAJQF5SUAAAAAAABQFpSXAAAAAAAAQFlo1dIBAAAAAAAAKEOFQksnoALZeQkAAAAAAACUBeUlAAAAAAAAUBaUlwAAAAAAAEBZUF4CAAAAAAAAZUF5CQAAAAAAAJQF5SUAAAAAAABQFlq1dAAAAAAAAADKT6Gq0NIRqEB2XgIAAAAAAABlQXkJAAAAAAAAlAXlJQAAAAAAAFAWlJcAAAAAAABAWVBeAgAAAAAAAGVBeQkAAAAAAACUBeUlAAAAAAAAUBZatXQAAAAAAAAAyk+hyh44vni+dQAAAAAAAEBZKLm8fO+99/KrX/0qd999d+bPn99gbtasWTnjjDOaLBwAAAAAAABQOUoqL//xj39k4403zg9+8IPsv//++dKXvpTnnnuufn7mzJk5/fTTmzwkAAAAAAAAsPwrqbw86aSTsu+++2bq1KmZPHlydt999/Tt2zdPPfVUc+UDAAAAAAAAKkSrUhY/8cQTufTSS1NVVZUVVlghl112WdZee+3stttuueeee7L22ms3V04AAAAAAABgOVdSeZkkc+fObXA9bNiwtGrVKl/96ldzzTXXNFkwAAAAAAAAoLKUVF5usskmeeSRR7LZZps1GP/xj3+curq6fPvb327ScAAAAAAAALSMQlWhpSNQgUp65uVBBx2Uhx9+eLFzJ5xwQk4//XRHxwIAAAAAAABLpaTy8tBDD83111+/xPkTTzwxEyZMqL9++OGHM2/evKVPBwAAAAAAAFSMksrLUg0YMCDvvPNOc/4IAAAAAAAAYDnRrOVlsVhsztsDAAAAAAAAy5FmLS8BAAAAAAAAGkt5CQAAAAAAAJQF5SUAAAAAAABQFlo1580LhUJz3h4AAAAAAIBmUqiyB44vXrN+64rFYnPeHgAAAAAAAFiOlFRevv/++585v2DBgjz22GP11x9//HHWW2+9pUsGAAAAAAAAVJSSystu3bo1KDA33XTTvPXWW/XXU6ZMSZ8+fZouHQAAAAAAAFAxSiov//MY2DfeeCOffPLJZ64BAAAAAAAAaIwmf+ZloVBo6lsCAAAAAAAAFaDJy0sAAAAAAACApdGqlMWFQiEff/xx2rZtm2KxmEKhkJkzZ2bGjBlJUv+/AAAAAAAA/HcrVDltky9eSeVlsVhMz549G1xvueWWDa4dGwsAAAAAAAAsjZLKywceeKC5cgAAAAAAAAAVrqTycuLEiTnggANSU1PTXHkAAAAAAACAClVVyuIhQ4Zk+vTpzZUFAAAAAAAAqGAllZfFYrG5cgAAAAAAAAAVrqTyMkkKhUJz5AAAAAAAAAAqXEnPvEyS3XbbLa1affbbnnzyyaUOBAAAAAAAAFSmksvL/v37p2PHjs2RBQAAAAAAgDJRqHIaJ1+8ksvL448/Pp07d26OLAAAAAAAAEAFK+mZl553CQAAAAAAADSXksrLYrHYXDkAAAAAAACACldSeTlhwoSsttpqzZUFAAAAAAAAqGAlPfNy5MiRjVp3wQUXLFUYAAAAAAAAoHKVVF4+9dRTn7vGczEBAAAAAACApVFSefnAAw80Vw4AKtTKO22T9f73kNRutUnart45j3/jqEy+876WjkUJvtxn1Qwc0C291l8htZ1a5+BjHs+rE2a1dKyKdtC+XbPHLqukY/vqPP/KrFw85q28O3n+Ete3a1uVwft1yw5b12bFTq3y2sQ5ufyGt/PyhDlJkurq5OBvdMu2m3VKt85tMmt2XZ56/uNc/Zt389G0BV/UxwIAAACgApT0zEsAaGrVHdpnxjMv5dljTm/pKCyldm2r8szzM3L5mNdbOgpJvvW1ztln99VyybVv5UdnvJy58+rysx+vn9atl3w6xnHfWytbbdIx5/5yYo44+cU88ezHOfuEHlllpdZJkpo2VemxTvvceOfk/ODUl3PGJROyZteanH7sel/UxwIAAABaQlWVV0u+KlRJOy/322+/xY7X1tamZ8+eOfTQQ7Paaqs1STAAKsMH9zyUD+55qKVjsAzueeD9JEnXzjUtnIQkGdh/tdw0dlLGPTUjSXLuLyfm1xdvkh22qs1f/j5tkfVtWhey0zYr5rSRE/LsS//aMfur2ydl+y07Zc+vrJIxt0zK7Dl1GX7eaw3ed+n1b+eS03pltZVb54OPPmn2zwUAAABAZSiptq2trV3sa9q0abnqqqvSq1evPPvss82VFQCAz9B1tTZZZcXWefK5mfVjs+fU5cXXZ2ejHh0W+57q6kKqqwuZ/0ldg/F58+vypQ06LvFndWhXnbq6YmbNXtg04QEAAAAgJe68HD169BLn6urqcthhh2X48OEZO3bsMgcDAKA0K9f+61/tpk1vuBNy2oxP6uf+05y5dXn+lVk5cO+uefPdNzJt+oLs0melbNSjQ96dPG+x72ndupBDDlg9Dz46NbPn1i12DQAAAAAsjZLKy89SVVWVY445JgMGDPjctfPmzcu8eQ3/Y1hNTU1qahw3BwDlbPe+nXP8D3rWX//4tH/mmeent2CiyrZrn5Xyo4PXrL8+5YKle+7oub+cmKGHrJ2bRm6ShQuLeXXi7Dz46NRssG77RdZWVycn/2DdJMklY95eqp8HAAAAAEvSZOVlknTo0CGzZ8/+3HVnnXVWTj/99AZjI0aMyGmnndaUcQCAJva3x6bk+Zcfr7/+YMr8FkzDo09Nz0uvzaq/bt36X08EWLG2dT6avqB+fMVOrfPam3OWeJ/33p+f4896NTVtqtKhXVU+mr4gJx21Tt57v+FfNvu0uOyySpuccPardl0CAAAA0OSatLz805/+lJ49e37uuuHDh2fo0KENxuy6BIDyN2fOwrwzxzMOy8WcuXWZM7dhgTxl2ifZcuOOef3/Lyvbt63Khuu1z+/v//Bz7zdvfl3mza9Lx/bV2XqTThn1m3fr5z4tLtfoUpMTzn41H8/yPQAAAACg6ZVUXt55552LHZ8+fXqeeOKJjBo1KqNGjfrc+zgiFoBPVXdonw491q6/bt99zXTafMPM/2h65r71Xgsmo7FW6NgqXVaryaor/+v/29de419HjX40dX4+mvbJZ72VZnD7PR/k23t3yTuT52XSB/MzeL9umTLtkzzy5P873vfsE9bPI09Oz51//lehufUmK6RQSN56b17W6NImhx6wRt56b27u/euUJP8qLk85unt6rNMup174eqqqClnp/3+G5sczF2bBwuIX/0EBAACAZlcoFFo6AhWopPJy4MCBix1fYYUV0qtXr4waNSqDBg1qilwAVIjarTdJn/uur7/e+PyTkiRvXXdrnjlkeEvFogQ79V4lJx+7Yf31GSdunCS55sY3cs1NE1sqVsX6zd3vp21NVX508Frp2L46z70yKyef/3o++eT/FYzdOtekU8f/96+BHdpXZ8g3u2XVlVrn41kL8/Dj0zL6d+9l4f+/uXLVldqkz1a1SZLLf7phg593/Fmv5pkXZzb/BwMAAACgIhSKxaK/Kg/wX+Su1r1aOgJL6aw9ftnSEVgKHVaubekILKV7xmzR0hEAAADgv9oHPxnS0hEq2mo/Hd3SEVpEVUsHAAAAAAAAAEhKPDY2Serq6nLttdfm1ltvzRtvvJFCoZDu3btn//33z3e/+13nHwMAAAAAAABLpaSdl8ViMXvvvXcOPfTQvPPOO9l0003zpS99KRMnTszBBx+cfffdt7lyAgAAAAAAAMu5knZeXnvttXnooYdy3333Zdddd20wd//992fgwIG57rrrctBBBzVpSAAAAAAAAGD5V9LOy5tuuiknnXTSIsVlknzlK1/JsGHDcsMNNzRZOAAAAAAAAKBylLTz8plnnsm55567xPkBAwbk4osvXuZQAAAAAAAAtKxCVUl74KBJlPSt++ijj9KlS5clznfp0iVTp05d5lAAAAAAAABA5SmpvFy4cGFatVryZs3q6uosWLBgmUMBAAAAAAAAlaekY2OLxWIOPvjg1NTULHZ+3rx5TRIKAAAAAAAAqDwllZeDBw/+3DUHHXTQUocBAAAAAAAAKldJ5eXo0aObKwcAAAAAAABQ4Up65iUAAAAAAABAcylp5+X3vve9z11TKBRy9dVXL3UgAAAAAAAAWl6hqtDSEahAJZWXU6dOXeLcwoUL8+c//znz5s1TXgIAAAAAAAAlK6m8vO222xY7fscdd+Skk05KTU1NTj311CYJBgAAAAAAAFSWZXrm5cMPP5ydd945Bx54YPbcc8+8/vrrGTZsWFNlAwAAAAAAACrIUpWXzz//fPbaa6/ssssu6dmzZ1566aWcc845WWmllZo6HwAAAAAAAFAhSiov33rrrQwZMiSbb755WrVqlWeeeSZXX3111lxzzebKBwAAAAAAAFSIkp552atXrxQKhQwdOjQ77rhjXnnllbzyyiuLrNt7772bLCAAAAAAAABQGUoqL+fOnZskOe+883Leeectdk2hUMjChQuXPRkAAAAAAABQUUoqL+vq6porBwAAAAAAAOWkqqSnD0KT8K0DAAAAAAAAysJSlZe//e1vs99++2WTTTbJJptskv322y+/+93vmjobAAAAAAAAUEFKKi/r6upywAEH5IADDsjzzz+fHj16pEePHnnuuedywAEHZNCgQSkWi82VFQAAAAAAAFiOlfTMy5EjR+bPf/5z7rzzzuy5554N5u68884MGTIkI0eOzLHHHtuUGQEAAAAAAIAKUNLOy9GjR+e8885bpLhMkr333jvnnnturrnmmiYLBwAAAAAAAFSOksrLV155Jf369VvifL9+/fLKK68scygAAAAAAACg8pR0bGy7du0ybdq0rL322oudnzFjRtq2bdskwQAAAAAAAGg5hapCS0egApW087JPnz65/PLLlzh/6aWXpk+fPsscCgAAAAAAAKg8Je28PPnkk7PLLrtkypQp+fGPf5wNN9wwxWIxL7zwQn7+85/njjvuyAMPPNBcWQEAAAAAAIDlWEnl5Q477JBf//rX+f73v59bbrmlwdxKK62Um266KTvuuGOTBgQAAAAAAAAqQ0nlZZLsu+++6d+/f+6555688sorSZKePXvmq1/9atq3b9/kAQEAAAAAAIDKUFJ5ef/99+foo4/Oo48+mn333bfB3PTp0/OlL30pV1xxRXbeeecmDQkAAAAAAAAs/6pKWXzRRRflsMMOS6dOnRaZq62tzeGHH54LLrigycIBAAAAAAAAlaOk8vLpp5/OHnvsscT5r371q3niiSeWORQAAAAAAABQeUo6Nnby5Mlp3br1km/WqlU++OCDZQ4FAAAAAABAyyoUStoDB02ipG/dGmuskWeffXaJ888880y6deu2zKEAAAAAAACAylNSefm1r30tp5xySubOnbvI3Jw5czJixIjsueeeTRYOAAAAAAAAqBwlHRv7k5/8JLfeemt69uyZo48+Or169UqSvPjii7n00kuzcOHCnHzyyc0SFAAAAAAAAFi+lVRedunSJY888kiOPPLIDB8+PMViMUlSKBTSv3//XHrppenSpUuzBAUAAAAAAACWbyWVl0myzjrr5O67787UqVPz6quvplgsZoMNNshKK63UHPkAAAAAAACAClFyefmplVZaKdtuu21TZgEAAAAAAAAq2FKXlwAAAAAAACzHqgotnYAKVNXSAQAAAAAAAAAS5SUAAAAAAABQJpSXAAAAAAAAQFlQXgIAAAAAAABlQXkJAAAAAAAAlAXlJQAAAAAAAFAWlJcAAAAAAABAWWjV0gEAAAAAAAAoP4Uqe+D44vnWAQAAAAAAAGVBeQkAAAAAAACUBeUlAAAAAAAAUBaUlwAAAAAAAEBZUF4CAAAAAAAAZUF5CQAAAAAAAJSFVi0dAAAAAAAAgPJTqCq0dAQqkJ2XAAAAAAAAQFlQXgIAAAAAAABlQXkJAAAAAAAAlAXlJQAAAAAAAFAWlJcAAAAAAABAWVBeAgAAAAAAAGWhVUsHAFrGXa17tXQEltLXP3mppSOwlM7a6y8tHQEAAAAAoKwpLwEAAAAAAFhUwQGefPF86wAAAAAAAICyoLwEAAAAAAAAyoLyEgAAAAAAACgLyksAAAAAAACgLCgvAQAAAAAAgLKgvAQAAAAAAADKQquWDgAAAAAAAED5KVQVWjoCFcjOSwAAAAAAAKAsKC8BAAAAAACAsqC8BAAAAAAAAMpCyc+8fOyxxzJu3LhMmjQpSdK1a9f06dMn2223XZOHAwAAAAAAACpHo8vL999/P9/4xjfy8MMPZ+21106XLl2SJJMnT85xxx2XHXfcMbfccks6d+7cbGEBAAAAAACA5Vejj4096qijsnDhwrzwwgt544038ve//z1///vf88Ybb+SFF15IXV1dfvCDHzRnVgAAAAAAAGA51uidl/fcc08eeuih9OrVa5G5Xr165eKLL84uu+zSlNkAAAAAAACACtLo8rKmpiYzZsxY4vzHH3+cmpqaJgkFAAAAAABAC6tq9AGe0GQa/a074IADMnjw4Nx2220NSswZM2bktttuy5AhQ/Ltb3+7WUICAAAAAAAAy79G77y84IILUldXl0GDBmXBggVp06ZNkmT+/Plp1apVDjnkkJx//vnNFhQAAAAAAABYvpV0bOzll1+ec845J0888UQmTZqUJOnatWu23nrrdOrUqdlCAgAAAAAAAMu/RpeXn+rUqVN23XXX5sgCAAAAAAAAVDBPWgUAAAAAAADKgvISAAAAAAAAKAslHxsLAAAAAADA8q9QKLR0BCpQs+28PPvsszNt2rTmuj0AAAAAAACwnGm28vJnP/tZPvroo+a6PQAAAAAAALCcabbyslgsNtetAQAAAAAAgOVQs5WXAAAAAAAAAKVQXgIAAAAAAABlQXkJAAAAAAAAlAXlJQAAAAAAAFAWWjXXjXfeeee0a9euuW4PAAAAAABAc6qyB44vXqO/dXV1dTnnnHOy4447Ztttt82wYcMyZ86cJa6/++67061btyYJCQAAAAAAACz/Gl1e/t///V9OOumkdOzYMWussUZGjhyZH/zgB82ZDQAAAAAAAKggjS4vr7vuulx22WW55557cvvtt2fs2LG54YYbUldX15z5AAAAAAAAgArR6PLyzTffzNe+9rX66379+qVQKOTdd99tlmAAAAAAAABAZWl0eblgwYK0bdu2wVjr1q3zySefNHkoAAAAAAAAoPK0auzCYrGYgw8+ODU1NfVjc+fOzRFHHJEOHTrUj916661NmxAAAAAAAACoCI0uLwcPHrzI2He+850mDQMAAAAAAEB5KFQVWjoCFajR5eXo0aObMwcAAAAAAABQ4Rr9zMvPUiwW84c//CH7779/U9wOAAAAAAAAqEDLVF5OmDAhp5xyStZee+3su+++mTt3blPlAgAAAAAAACpMo4+N/dS8efPyu9/9LldffXX+9re/ZeHChTn//PNzyCGHpFOnTs2REQAAAAAAAKgAjd55+cQTT+Soo45K165dc9FFF2XgwIF56623UlVVlf79+ysuAQAAAAAAgGXS6J2XvXv3zg9/+MM8+uij6dWrV3NmAgAAAAAAACpQo8vL3XbbLVdffXXef//9fPe7303//v1TKBSaMxsAAAAAAABQQRpdXt5zzz156623Mnr06Bx55JGZM2dODjjggCRRYgIAAAAAACxvCo1++iA0mZK+dWuttVZOPfXUTJgwIddff30++OCDtGrVKvvss09OOumkPPHEE82VEwAAAAAAAFjOLXVlvvvuu+fGG2/Mu+++mx/+8If5wx/+kO22264pswEAAAAAAAAVZJn3+6600kr54Q9/mKeeeipnnXVWU2QCAAAAAAAAKlBJ5eWCBQvy7LPP5uWXX24wfscdd2SLLbbIKaec0qThAAAAAAAAgMrR6PLy2WefTY8ePbL55ptno402yn777ZfJkyenb9+++d73vpc99tgjr732WnNmBQAAAAAAAJZjrRq78MQTT0yPHj3yi1/8IjfddFNuuummvPDCCznkkEPyxz/+Me3atWvOnAAAAAAAAMByrtHl5T/+8Y/ce++92WKLLbLzzjvnpptuykknnZTvfve7zZkPAAAAAACAllBVaOkEVKBGHxv74YcfZvXVV0+S1NbWpkOHDtl+++2bLRgAAAAAAABQWRq987JQKOTjjz9O27ZtUywWUygUMmfOnMyYMaPBuk6dOjV5SAAAAAAAAGD51+jyslgspmfPng2ut9xyywbXhUIhCxcubNqEAAAAAAAAQEVodHn5wAMPNGcOAAAAAAAAoMI1urzs27dvc+YAAAAAAAAAKlyjy8tPTZ8+PX/605/yxhtvpFAopHv37unXr59nXQIAAAAAAADLpKTy8le/+lWOPvrozJgxo8F4bW1trrjiihxwwAFNGg4AAAAAAACoHFWNXfjkk09myJAhGThwYJ566qnMmTMns2fPzuOPP5699tor3/3ud/P00083Z1YAAAAAAAC+IIVClVcLvipVo3deXnLJJRk4cGCuvfbaBuNbbbVVrrvuusyePTsjR47MNddc09QZAQAAAAAAgArQ6PLy4YcfzmWXXbbE+SOOOCJHHXVUk4QCWBor77RN1vvfQ1K71SZpu3rnPP6NozL5zvtaOhYs977cZ9UMHNAtvdZfIbWdWufgYx7PqxNmtXSsinbQvl2zxy6rpGP76jz/yqxcPOatvDt5/hLXt2tblcH7dcsOW9dmxU6t8trEObn8hrfz8oQ5SZLq6uTgb3TLtpt1SrfObTJrdl2eev7jXP2bd/PRtAVf1McCAAAAoAI0es/pu+++m549ey5xvmfPnnnnnXeaJBTA0qju0D4znnkpzx5zektHgYrSrm1Vnnl+Ri4f83pLRyHJt77WOfvsvlouufat/OiMlzN3Xl1+9uP107p1YYnvOe57a2WrTTrm3F9OzBEnv5gnnv04Z5/QI6us1DpJUtOmKj3WaZ8b75ycH5z6cs64ZELW7FqT049d74v6WAAAAAA0wqWXXpp11103bdu2Te/evfPYY48tce21116bQqHQ4NW2bdsGa4rFYk499dR069Yt7dq1S79+/fLKK68062dodHk5e/bsRQL/u5qamsydO7dJQgEsjQ/ueSgvj7gok+/4c0tHgYpyzwPv59qbJ+bx8VNbOgpJBvZfLTeNnZRxT83IhLfm5txfTswqK7bODlvVLnZ9m9aF7LTNihn16/fy7Euz8u778/Or2yfl3ffnZc+vrJIkmT2nLsPPey0PPTYtb0+alxdfm51Lr387Pbu3z2ort/4iPx4AAAAAS/DrX/86Q4cOzYgRI/Lkk09m8803T//+/fP+++8v8T2dOnXKe++9V/+aOHFig/lzzz03F198ca644or8/e9/T4cOHdK/f/9m7QQbfWxsktxzzz2prV38f/iaNm1aU+QBAGApdV2tTVZZsXWefG5m/djsOXV58fXZ2ahHh/zl79MWeU91dSHV1YXM/6Suwfi8+XX50gYdl/izOrSrTl1dMbNmL2yy/AAAAAAsvQsuuCCHHXZYhgwZkiS54oorctddd+Waa67JsGHDFvueQqGQrl27LnauWCzmoosuyk9+8pPss88+SZLrrrsuXbp0ye23355BgwY1y+coqbwcPHjwZ84XCks+juzfzZs3L/PmzWswVlNTk5qamlLiAADwb1au/de/2k2b/kmD8WkzPqmf+09z5tbl+Vdm5cC9u+bNd9/ItOkLskuflbJRjw55d/K8xb6ndetCDjlg9Tz46NTMnlu32DUAAAAALJtS+rT58+fniSeeyPDhw+vHqqqq0q9fv4wbN26JP2PmzJlZZ511UldXl6222io/+9nP8qUvfSlJMmHChEyaNCn9+vWrX19bW5vevXtn3LhxzVZeNvrY2Lq6us99LVzYuL95f9ZZZ6W2trbB66yzzlrqDwEAfDF279s59/5mp/rXZhsv/kQGvhi79lkpt1+5af2rurpxf5HsP537y4kpFJKbRm6S31+9eQbuvmoefHRqisVF11ZXJyf/YN0kySVj3l6G9AAAAEDZqyp4teCrlD7tww8/zMKFC9OlS5cG4126dMmkSZMW+55evXrlmmuuyR133JFf/epXqauryw477JC33/7Xf/P59H2l3LMplLTzshRf//rXM2rUqHTr1m2RueHDh2fo0KENxuy6BIDy97fHpuT5lx+vv/5gyvwWTMOjT03PS6/Nqr9u3fpffy9txdrW+Wj6gvrxFTu1zmtvzlnifd57f36OP+vV1LSpSod2Vflo+oKcdNQ6ee/9hn+z79PisssqbXLC2a/adQkAAADQjJq7T+vTp0/69OlTf73DDjtko402ypVXXpkzzzyzyX5OqZqtvHzooYcyZ87i/yOZI2IB4L/TnDkL884czzgsF3Pm1mXO3IYF8pRpn2TLjTvm9f+/rGzftiobrtc+v7//w8+937z5dZk3vy4d21dn6006ZdRv3q2f+7S4XKNLTU44+9V8PMv3AAAAAKA5ldKnrbrqqqmurs7kyZMbjE+ePHmJz7T8T61bt86WW26ZV199NUnq3zd58uQGmxUnT56cLbbYolH3XBqNPjYWoNxVd2ifTptvmE6bb5gkad99zXTafMO0XWvRHeBA01mhY6v06N4h667VIUmy9hrt06N7h6y8YusWTlaZbr/ng3x77y7ZfstOWXfNtjn+++tkyrRP8siT0+vXnH3C+tm736r111tvskK22XSFdFm1Tbb6UsecO6xH3npvbu7965Qk/youTzm6e3qu2z7nXDExVVWFrFTbKivVtkqrpTyqFgAAAICm06ZNm2y99da577776sfq6upy3333Ndhd+VkWLlyYf/7zn/VFZffu3dO1a9cG95wxY0b+/ve/N/qeS6PZdl4CfNFqt94kfe67vv564/NPSpK8dd2teeaQ4Ut6G7CMduq9Sk4+dsP66zNO3DhJcs2Nb+Samya2VKyK9Zu730/bmqr86OC10rF9dZ57ZVZOPv/1fPLJ/3uAZbfONenU8f/9a2CH9tUZ8s1uWXWl1vl41sI8/Pi0jP7de/n0cearrtQmfbb61/NNL//phg1+3vFnvZpnXpzZ/B8MAAAAgM80dOjQDB48ONtss0222267XHTRRZk1a1aGDBmSJDnooIOyxhpr1D8384wzzsj222+fHj16ZNq0aTnvvPMyceLEHHrooUmSQqGQY489Nj/96U+zwQYbpHv37jnllFOy+uqrZ+DAgc32OZSXwHLjo4cey12te7V0DKg4f7hvcv5w3+TPX8gX5rrbJuW625b80PTBP36+wfVDj03LQ49NW+L6yR/OT//B45soHQAAAADN4YADDsgHH3yQU089NZMmTcoWW2yRP/7xj+nSpUuS5M0330xV1f87lHXq1Kk57LDDMmnSpKy00krZeuut88gjj2TjjTeuX3PCCSdk1qxZ+f73v59p06Zlp512yh//+Me0bdu22T5HoVgsFj9/WelWWGGFPP3001lvvfWa4/bAMlLy/ff6+icvtXQEltJOe/2lpSOwFDqsXNvSEVhK94zZoqUjAAAAwH+1WVf9pKUjVLQOh/20pSO0CM+8BAAAAAAAAMpCsx0be9JJJ2XllVdurtsDAAAAAADQjApV9sDxxWt0efnQQw81at2Xv/zlJMnw4cOXLhEAAAAAAABQkRpdXu6yyy4pFApJkiU9JrNQKGThwoVNkwwAAAAAAACoKI0uL1daaaWssMIKOfjgg/Pd7343q666anPmAgAAAAAAACpMow8rfu+993LOOedk3Lhx2XTTTXPIIYfkkUceSadOnVJbW1v/AgAAAAAAAFgajS4v27RpkwMOOCD33HNPXnzxxWy22WY5+uijs9Zaa+Xkk0/OggULmjMnAAAAAAAAsJxrdHn579Zee+2ceuqp+fOf/5yePXvm7LPPzowZM5o6GwAAAAAAAFBBSi4v582blxtvvDH9+vXLJptsklVXXTV33XVXVl555ebIBwAAAAAAQEsoFLxa8lWhWjV24WOPPZbRo0fn5ptvzrrrrpshQ4bkN7/5jdISAAAAAAAAaBKNLi+33377rL322jnmmGOy9dZbJ0n+9re/LbJu7733brp0AAAAAAAAQMVodHmZJG+++WbOPPPMJc4XCoUsXLhwmUMBAAAAAAAAlafR5WVdXV1z5gAAAAAAAAAqXFVjFz7//POfu+a8885bpjAAAAAAAABA5Wp0edm/f/+8+eabS5w///zzc/LJJzdJKAAAAAAAAKDyNLq83GmnndKvX7988MEHi8z9/Oc/z0knnZTrrruuScMBAAAAAAAAlaPRz7y8/vrrs9dee6V///558MEH06lTpyTJhRdemGHDhmXMmDEZNGhQswUFAAAAAADgC1TV6D1w0GQa/a1r1apVbr311rRr1y577rln5s6dm4suuijHH398Ro8enQMPPLA5cwIAAAAAAADLuUbvvEySdu3a5e67707fvn2z9dZb5+WXX87o0aPzne98p7nyAQAAAAAAABWi0eXlnXfeWf/PRx55ZH70ox9l4MCBqa2tbTC39957N21CAAAAAAAAoCI0urwcOHDgImO33HJLbrnllvrrQqGQhQsXNkkwAAAAAAAAoLI0urysq6trzhwAAAAAAABAhatqypvNmTOnKW8HAAAAAAAAVJAmKS/nzZuXn//85+nevXtT3A4AAAAAAICWVih4teSrQjW6vJw3b16GDx+ebbbZJjvssENuv/32JMno0aPTvXv3XHTRRTnuuOOaKycAAAAAAACwnGv0My9PPfXUXHnllenXr18eeeSRfPOb38yQIUPy6KOP5oILLsg3v/nNVFdXN2dWAAAAAAAAYDnW6PLyt7/9ba677rrsvffeefbZZ7PZZptlwYIFefrpp1Oo4K2rAAAAAAAAQNNo9LGxb7/9drbeeuskySabbJKampocd9xxiksAAAAAAACgSTS6vFy4cGHatGlTf92qVat07NixWUIBAAAAAAAAlafRx8YWi8UcfPDBqampSZLMnTs3RxxxRDp06NBg3a233tq0CQEAAAAAAICK0OjycvDgwQ2uv/Od7zR5GAAAAAAAAKByNbq8HD16dHPmAAAAAAAAoIwUqhr99EFoMr51AAAAAAAAQFkoqbx84IEH8vOf/zwPP/xwkuTKK6/M2muvndVWWy2HHXZY5syZ0ywhAQAAAAAAgOVfo4+Nveqqq3LkkUeme/fuOfnkkzNixIj83//9X7773e+mqqoqv/rVr7LKKqvk7LPPbs68AAAAAAAAwHKq0TsvR44cmQsvvDCvvPJKbr/99px66qm59NJLc/nll+fSSy/NqFGj8rvf/a45swIAAAAAAADLsUaXl6+//nr23nvvJMkee+yRQqGQ7bbbrn6+d+/eeeutt5o+IQAAAAAAAFARGl1ezp07N+3atau/rqmpSU1NTYPrBQsWNG06AAAAAAAAoGI0+pmXhUIhH3/8cdq2bZtisZhCoZCZM2dmxowZSVL/vwAAAAAAAABLo9HlZbFYTM+ePRtcb7nllg2uC4VC06YDAAAAAACgZRQafYAnNJlGl5cPPPBAc+YAAAAAAAAAKlyjy8u+ffs2Zw4AAAAAAACgwtnvCwAAAAAAAJSFRu+8rKqq+txnWhYKhSxYsGCZQwEAAAAAAACVp9Hl5W233bbEuXHjxuXiiy9OXV1dk4QCAAAAAAAAKk+jy8t99tlnkbGXXnopw4YNy9ixY/M///M/OeOMM5o0HAAAAAAAAFA5Gl1e/rt33303I0aMyJgxY9K/f/+MHz8+m2yySVNnAwAAAAAAoKVUffbjBKE5VJWyePr06TnxxBPTo0ePPPfcc7nvvvsyduxYxSUAAAAAAACwzBq98/Lcc8/NOeeck65du+amm25a7DGyAAAAAAAAAEur0eXlsGHD0q5du/To0SNjxozJmDFjFrvu1ltvbbJwAAAAAAAAQOVodHl50EEHpVBwtjEAAAAAAADQPBpdXl577bXNGAMAAAAAAACodFUtHQAAAAAAAAAgUV4CAAAAAAAAZaLRx8YCAAAAAABQOQoFe+D44vnWAQAAAAAAAGVBeQkAAAAAAACUBeUlAAAAAAAAUBaUlwAAAAAAAEBZUF4CAAAAAAAAZUF5CQAAAAAAAJSFVi0dAAAAAAAAgDJUVWjpBFQg5SVUqLP2+GVLR2ApnbXXX1o6Akvpb2P7tnQElsLfNt+qpSOw1J5s6QAAAAAAlMixsQAAAAAAAEBZUF4CAAAAAAAAZUF5CQAAAAAAAJQF5SUAAAAAAABQFpSXAAAAAAAAQFlQXgIAAAAAAABloVVLBwAAAAAAAKAMFeyB44vnWwcAAAAAAACUBeUlAAAAAAAAUBaUlwAAAAAAAEBZUF4CAAAAAAAAZUF5CQAAAAAAAJQF5SUAAAAAAABQFlq1dAAAAAAAAADKUKHQ0gmoQHZeAgAAAAAAAGVBeQkAAAAAAACUBeUlAAAAAAAAUBaUlwAAAAAAAEBZUF4CAAAAAAAAZUF5CQAAAAAAAJQF5SUAAAAAAABQFlq1dAAAAAAAAADKUJU9cHzxfOsAAAAAAACAsqC8BAAAAAAAAMqC8hIAAAAAAAAoC8pLAAAAAAAAoCwoLwEAAAAAAICy0KTl5WuvvZavfOUrTXlLAAAAAAAAoEK0asqbzZw5M3/5y1+a8pYAAAAAAAC0hIIDPPnilVReXnzxxZ85/8477yxTGAAAAAAAAKBylVReHnvssenWrVvatGmz2Pn58+c3SSgAAAAAAOD/Y+/Oo62sC/3xvzccPKDIYVDGnAHnIVHR1IsDV0TNUEvNHEAjh7SMckCvWpYXNU2tLCtTM4dMU9P7K41wKNGyxCHnCUVRBJlBOMBh//7oG91zAT37cA57y3691nqW7s/z2c9+P+29XC3efD4PQPUpqbzcaKONcskll+Twww9f4fmnnnoqAwYMaJFgAAAAAAAAQHUpabPiAQMG5Iknnljp+UKhkGKxuMqhAAAAAAAAgOpT0srLCy+8MB988MFKz2+11VaZOHHiKocCAAAAAAAAqk9JKy+32mqr7LTTTis9365du2y00UbLXo8fPz719fXNTwcAAAAAAABUjZLKy1INHTo0kydPbs2PAAAAAAAAANYQJW0bWyrPvwQAAAAAAPiYalModwKqUKuuvAQAAAAAAABoKuUlAAAAAAAAUBGUlwAAAAAAAEBFaNXyslCwFzIAAAAAAADQNK1aXhaLxda8PAAAAAAAALAGKam8nDp16oeeX7JkSR5//PFlr+fOnZtNN920eckAAAAAAACAqlJSedmrV69GBea2226bt956a9nr6dOnZ7fddmu5dAAAAAAAAJRHoY2jnEeVKunO/+82sG+88UYWL178oXMAAAAAAAAAmqLFa9tCodDSlwQAAAAAAACqQPWuOQUAAAAAAAAqSk0pkwuFQubOnZv27dunWCymUChk3rx5mTNnTpIs+ycAAAAAAABAqUoqL4vFYvr379/o9Sc/+clGr20bCwAAAAAAADRHSeXlgw8+2Fo5AAAAAAAAgCpXUnn55ptv5ogjjkhtbW1r5QEAAAAAAACqVJtSJo8YMSKzZ89urSwAAAAAAABUikLBUc6jSpVUXhaLxdbKAQAAAAAAAFS5ksrLJClUcdMLAAAAAAAAtJ6SnnmZJPvuu29qaj78bRMmTGh2IAAAAAAAAKA6lVxeDhkyJB07dmyNLAAAAAAAAEAVK7m8POOMM9K9e/fWyAIAAAAAAABUsZKeeel5lwAAAAAAAEBrKWnlZbFYbK0cAAAAAAAAVJI2Ja2BgxZR0q9u4sSJWX/99VsrCwAAAAAAAFDFSlp5edVVVzVp3ve+971mhQEAAAAAAACqV0nl5ZNPPvmRczwXEwAAAAAAAGiOksrLBx98sLVyALSI/9htvQwb2iubb7Zu6jq1y/Cv/D2vTpxf7lg0ge8OVl2vIw5Pn+OOzVrrdcv8l1/OaxdfmnnPPrfCuYWamnzihBHp/umDUtu9exa88WYmXvn9zHr00WVzPnH8iHTbd5902GTjLK2vz9ynns4bV34/C958c3XdEgAAAABVxpNWgTVKh/Zt8szzc/LjX7xe7iiUyHcHq2a9Iftlk2+MyqSf/DRPHnlU5r/0Srb58dVp17XLCudvdOop6fnZw/L6xZfmiUM+m3dvvyNbXnFZ1tli82Vz6nYakHdv+3WeOea4PHfiySnU1GTra36UNh3ar67bAgAAAKDKlLTy8tBDD13heF1dXfr3758vfvGLWX/99VskGEBz3P/g1CRJz+61ZU5CqXx3sGr6HPOFTLnzrkz97T1Jkle/c1G6/Mce6THsM3n7uhuWm7/+gQfm7Wt/npmPjE+STLn9jnTedWD6HHtMXj7nv5Ikz51yaqP3vHz+Bdn1oQfSccutMmfChNa9IQAAAACqUkkrL+vq6lZ4zJo1Kz/72c+y+eab59lnn22trAAArEChpiYdt9wys/7y138PFouZ9Ze/Zt3ttlvhe9qs1S5LF9U3GltaX59OO+yw0s+p6bhukmTJnNmrnBkAAAAAVqSklZfXX3/9Ss8tXbo0I0eOzOjRo3PvvfeucjAAAJqmXZfOKdTUZPH0GY3GF0+fkbU32XiF75n56GPpfczRmf3EhCx86+10HrhLuu2zdwpt2674QwqFbHrmNzL7ySfzwauvtfAdAAAAABWpUCh3AqpQSeXlh2nTpk2+8pWvZOjQoR85t76+PvX1jf+mf21tbWprbRUINN1/DuqeM77cf9nrb3zzH3nmeauBPg58d1B+r1/63fQ7/7wMuPvOpFjMgrffznu/vTc9hh28wvmbnXN21t5sszwz/PjVnBQAAACAatJi5WWSrLPOOvnggw8+ct6YMWPyrW99q9HYBRdckG9+85stGQdYwz3y+PQ8//Lfl72eNn1RGdNQCt8dtKzFM2eluGRJ2nXr2mi8XbeuWfT+9BW+Z8nMWXnha19PYa210q5zXRZNnZaNT/9KFk6evNzcTUefla7/sWeeOf6LWTR1aqvcAwAAAAAkLVxejh07Nv379//IeaNHj86oUaMajVl1CZRqwYKGTF7QUO4YNIPvDlpWccmSzHvhhXQeuEtmPPjQPwcLhXQeuEve/dVtH/7eRYuyaOq0FGpq0m3fffP+H8Y2Or/p6LPSbZ+9848TRqZ+8jutdAcAAAAA8E8llZf33HPPCsdnz56dJ554Itdee22uvfbaj7yOLWKB1rJux5r0WL8263X9539jNuyzdpJkxsxFmTFrcTmj8RF8d7BqJv/y5vT/9rcy77nnM/fZ59L76KPStkOHvHf3P///W//vXJj6qVPz5vd/mCTpuO02qe3ePfNefCm13btnw5NPTKFNIW/fcMOya252ztlZf+jQPH/619Iw/4O069YtSdIwb16W/p9HAAAAAABASyipvBw2bNgKx9ddd91svvnmufbaa3PkkUe2RC6AZtljYLece/oWy15feNZWSZLrbnkj1936Zrli0QS+O1g179//h7Tr0iUbnnJy1lqvW+a/9FKePeXULJ4xI0lS27NnikuXLpvfZq21stGXT0n7T/RJwwcfZOYj4/Pyuf+Vhrnzls3pdcThSZLtrmv8l9NePu+CTL3n3tVwVwAAAABUm0KxWCyWOwSw+u3x6YfLHQGqziP3Dip3BJrhke13LHcEmmmPpyeUOwIAAAB8rC38/64pd4Sq1v7Ak8odoSxa9JmXAAAAAAAArCEKbcqdgCpUcnm5dOnS3HDDDbnzzjvzxhtvpFAoZJNNNslnP/vZHHPMMSkUCq2REwAAAAAAAFjDlVSZF4vFHHzwwfniF7+YyZMnZ9ttt83WW2+dN998M8OHD88hhxzSWjkBAAAAAACANVxJKy9vuOGG/OlPf8q4ceOy9957Nzr3wAMPZNiwYbnxxhtz7LHHtmhIAAAAAAAAYM1X0srLW2+9Neecc85yxWWS7LPPPjn77LNz8803t1g4AAAAAAAAoHqUVF4+88wz2X///Vd6fujQoXn66adXORQAAAAAAABQfUoqL2fMmJEePXqs9HyPHj0yc+bMVQ4FAAAAAAAAVJ+SysuGhobU1Kz8MZlt27bNkiVLVjkUAAAAAAAAUH1W3kSuQLFYzPDhw1NbW7vC8/X19S0SCgAAAAAAgDJrU9IaOGgRJZWXxx133EfOOfbYY5sdBgAAAAAAAKheJZWX119/fWvlAAAAAAAAAKqc9b4AAAAAAABARShp5eXxxx//kXMKhUJ+/vOfNzsQAAAAAAAAUJ1KKi9nzpy50nMNDQ354x//mPr6euUlAAAAAAAAULKSysu77rprheO//e1vc84556S2tjbnn39+iwQDAAAAAAAAqktJ5eX/NX78+Jx99tmZMGFCTj311Jx99tnp0qVLS2UDAAAAAACgXAqFciegCrVpzpuef/75fPrTn85ee+2V/v3756WXXsoll1yiuAQAAAAAAACaraTy8q233sqIESOy/fbbp6amJs8880x+/vOf5xOf+ERr5QMAAAAAAACqREnbxm6++eYpFAoZNWpUdt9997zyyit55ZVXlpt38MEHt1hAAAAAAAAAoDqUVF4uXLgwSfLd73433/3ud1c4p1AopKGhYdWTAQAAAAAAAFWlpPJy6dKlrZUDAAAAAAAAqHIlPfMSAAAAAAAAoLU0q7y8/fbbc+ihh2abbbbJNttsk0MPPTR33HFHS2cDAAAAAAAAqkjJ28Z+/vOfz+23357+/ftniy22SJI899xzOeKII/K5z30ut956awqFQquEBQAAAAAAYDUp2MCT1a+k8vKqq67KH//4x9xzzz056KCDGp275557MmLEiFx11VU5/fTTWzIjAAAAAAAAUAVKqsyvv/76fPe7312uuEySgw8+OJdeemmuu+66FgsHAAAAAAAAVI+SystXXnklgwcPXun5wYMH55VXXlnlUAAAAAAAAED1Kam87NChQ2bNmrXS83PmzEn79u1XNRMAAAAAAABQhUoqL3fbbbf8+Mc/Xun5q6++OrvtttsqhwIAAAAAAACqT00pk88999zstddemT59er7xjW9kiy22SLFYzAsvvJDLL788v/3tb/Pggw+2VlYAAAAAAABgDVZSefmpT30qt912W770pS/lN7/5TaNzXbp0ya233prdd9+9RQMCAAAAAABQBoVCuRNQhUoqL5PkkEMOyZAhQ3L//ffnlVdeSZL0798/++23X9Zee+0WDwgAAAAAAABUh5LKywceeCCnnnpq/vKXv+SQQw5pdG727NnZeuutc80112TPPfds0ZAAAAAAAADAmq9NKZOvvPLKjBw5Mp06dVruXF1dXU488cR873vfa7FwAAAAAAAAQPUoqbx8+umns//++6/0/H777ZcnnnhilUMBAAAAAAAA1aek8vK9995Lu3btVnq+pqYm06ZNW+VQAAAAAAAAQPUpqbzs06dPnn322ZWef+aZZ9KrV69VDgUAAAAAAABUn5LKywMOOCDnnXdeFi5cuNy5BQsW5IILLshBBx3UYuEAAAAAAACA6lFTyuT/+q//yp133pn+/fvn1FNPzeabb54kefHFF3P11VenoaEh5557bqsEBQAAAAAAYDVqU9IaOGgRJZWXPXr0yKOPPpqTTz45o0ePTrFYTJIUCoUMGTIkV199dXr06NEqQQEAAAAAAIA1W0nlZZJstNFG+d3vfpeZM2fm1VdfTbFYTL9+/dKlS5fWyAcAAAAAAABUiZLLy3/p0qVLdt5555bMAgAAAAAAAFQxmxUDAAAAAAAAFUF5CQAAAAAAAFQE5SUAAAAAAABQEZr9zEsAAAAAAADWXMVCodwRqEJWXgIAAAAAAAAVQXkJAAAAAAAAVATlJQAAAAAAAFARlJcAAAAAAABARVBeAgAAAAAAABVBeQkAAAAAAABUBOUlAAAAAAAAUBFqyh0AAAAAAACAClSwBo7Vz68OAAAAAAAAqAjKSwAAAAAAAKAiKC8BAAAAAACAiqC8BAAAAAAAACqC8hIAAAAAAACoCMpLAAAAAAAAoCLUlDsAAAAAAAAAFahgDRyrn18dAAAAAAAAUBGUlwAAAAAAAEBFUF4CAAAAAAAAFcEzL6FKrdO1rtwRoOo8sv2O5Y5AM+zx9IRyRwAAAACAqmHlJQAAAAAAAFARlJcAAAAAAABARVBeAgAAAAAAABXBMy8BAAAAAABYTrFQKHcEqpCVlwAAAAAAAEBFUF4CAAAAAAAAFUF5CQAAAAAAAFQE5SUAAAAAAABQEZSXAAAAAAAAQEVQXgIAAAAAAAAVoabcAQAAAAAAAKhABWvgWP1a7Fc3c+bM3HjjjS11OQAAAAAAAKDKtFh5OWnSpIwYMaKlLgcAAAAAAABUmSZvGztnzpwPPT937txVDgMAAAAAAABUryaXl507d06hUFjp+WKx+KHnAQAAAAAAAD5Mk8vLddddN+eee24GDhy4wvOvvPJKTjzxxBYLBgAAAAAAAFSXJpeXO+64Y5Jk0KBBKzzfuXPnFIvFlkkFAAAAAAAAVJ02TZ141FFHpX379is937Nnz1xwwQUtEgoAAAAAAACoPk1eeTly5MgPPd+jRw/lJQAAAAAAwJqiUCh3AqpQk1deAgAAAAAAALQm5SUAAAAAAABQEZSXAAAAAAAAQEVQXgIAAAAAAAAVodXKy4svvjizZs1qrcsDAAAAAAAAa5hWKy//+7//OzNmzGitywMAAAAAAABrmJrWunCxWGytSwMAAAAAANDa2nj6IKufXx0AAAAAAABQEZSXAAAAAAAAQEVQXgIAAAAAAAAVQXkJAAAAAAAAa4Crr746G2+8cdq3b5+BAwfm8ccfX+ncn/3sZ9lzzz3TpUuXdOnSJYMHD15u/vDhw1MoFBod+++/f6veQ6uVl3vuuWc6dOjQWpcHAAAAAAAA/p/bbrsto0aNygUXXJAJEyZk++23z5AhQzJ16tQVzn/ooYfy+c9/Pg8++GAee+yxbLDBBtlvv/0yefLkRvP233//vPvuu8uOW2+9tVXvo8nl5dKlS3PJJZdk9913z84775yzzz47CxYsWOn83/3ud+nVq1eLhAQAAAAAAABW7nvf+15GjhyZESNGZKuttso111yTtddeO9ddd90K599888055ZRTssMOO2SLLbbItddem6VLl2bcuHGN5tXW1qZnz57Lji5durTqfTS5vLzoootyzjnnpGPHjunTp0+uuuqqfPnLX27NbAAAAAAAAFCV6uvrM2fOnEZHfX39CucuWrQoTzzxRAYPHrxsrE2bNhk8eHAee+yxJn3eBx98kMWLF6dr166Nxh966KF07949m2++eU4++eRMnz69+TfVBE0uL2+88cb86Ec/yv3335+777479957b26++eYsXbq0NfMBAAAAAABQBsVCwVHGY8yYMamrq2t0jBkzZoXf1fvvv5+Ghob06NGj0XiPHj0yZcqUJn3fZ511Vnr37t2oAN1///1z4403Zty4cbnkkkvy8MMPZ+jQoWloaGj+D+sj1DR14qRJk3LAAQcsez148OAUCoW88847+cQnPtEq4QAAAAAAAKAajR49OqNGjWo0Vltb2yqfdfHFF+dXv/pVHnroobRv337Z+JFHHrns37fddttst9122WyzzfLQQw9l3333bZUsTV55uWTJkkZhk6Rdu3ZZvHhxi4cCAAAAAACAalZbW5tOnTo1OlZWXq633npp27Zt3nvvvUbj7733Xnr27Pmhn3PZZZfl4osvzh/+8Idst912Hzp30003zXrrrZdXX321tJspQZNXXhaLxQwfPrzR/ygLFy7MSSedlHXWWWfZ2J133tmyCQEAAAAAAICVWmuttTJgwICMGzcuw4YNS5IsXbo048aNy6mnnrrS91166aW56KKLcv/992ennXb6yM95++23M3369PTq1auloi+nyeXlcccdt9zY0Ucf3aJhAAAAAAAAgNKNGjUqxx13XHbaaafssssuufLKKzN//vyMGDEiSXLsscemT58+y56beckll+T888/PLbfcko033njZszE7duyYjh07Zt68efnWt76Vww47LD179sxrr72WM888M3379s2QIUNa7T6aXF5ef/31rRYCAAAAAAAAaL4jjjgi06ZNy/nnn58pU6Zkhx12yH333ZcePXokSSZNmpQ2bf79RMkf//jHWbRoUT772c82us4FF1yQb37zm2nbtm2eeeaZ/OIXv8isWbPSu3fv7Lfffvn2t7/das/eTJJCsVgsrupFisVi7rvvvvz85z/PHXfc0RK5gFY25Linyh0Bqs55Tx1f7gg0wx5PTyh3BAAAAICymP+oRwWW0zqfOrTcEcqizUdPWbmJEyfmvPPOy4YbbphDDjkkCxcubKlcAAAAAAAAQJVp8rax/1JfX5877rgjP//5z/PII4+koaEhl112WU444YR06tSpNTICAAAAAACwuhVWaQ0cNEuTf3VPPPFETjnllPTs2TNXXnllhg0blrfeeitt2rTJkCFDFJcAAAAAAADAKmnyysuBAwfmtNNOy1/+8pdsvvnmrZkJAAAAAAAAqEJNLi/33Xff/PznP8/UqVNzzDHHZMiQISkUCq2ZDQAAAAAAAKgiTd429v77789zzz2XzTffPCeffHJ69eqVr371q0mixAQAAAAAAABWWUlPWt1ggw1y/vnnZ+LEifnlL3+ZadOmpaamJp/5zGdyzjnn5IknnmitnAAAAAAAAMAarqTy8n/7z//8z9xyyy155513ctppp+X3v/99dtlll5bMBgAAAAAAAFSRJj/zcmW6dOmS0047LaeddlouvfTSlsgEAAAAAABAmRULzV4DB81W0q9uyZIlefbZZ/Pyyy83Gv/tb3+bHXbYIeedd16LhgMAAAAAAACqR5PLy2effTZ9+/bN9ttvny233DKHHnpo3nvvvQwaNCjHH3989t9//7z22mutmRUAAAAAAABYgzV529izzjorffv2zQ9/+MPceuutufXWW/PCCy/khBNOyH333ZcOHTq0Zk4AAAAAAABgDdfk8vJvf/tb/vCHP2SHHXbInnvumVtvvTXnnHNOjjnmmNbMBwAAAAAAAFSJJm8b+/7776d3795Jkrq6uqyzzjrZddddWy0YAAAAAAAAUF2avPKyUChk7ty5ad++fYrFYgqFQhYsWJA5c+Y0mtepU6cWDwkAAAAAAACs+ZpcXhaLxfTv37/R609+8pONXhcKhTQ0NLRsQgAAAAAAAKAqNLm8fPDBB1szBwAAAAAAAJWkUCh3AqpQk8vLQYMGtWYOAAAAAAAAoMo1ubz8l9mzZ2fs2LF54403UigUsskmm2Tw4MGedQkAAAAAAACskpLKy5tuuimnnnpq5syZ02i8rq4u11xzTY444ogWDQcAAAAAAABUjzZNnThhwoSMGDEiw4YNy5NPPpkFCxbkgw8+yN///vd8+tOfzjHHHJOnn366NbMCAAAAAAAAa7Amr7z8wQ9+kGHDhuWGG25oNL7jjjvmxhtvzAcffJCrrroq1113XUtnBAAAAAAAAKpAk8vL8ePH50c/+tFKz5900kk55ZRTWiQUwP917CE9s/9e3dJx7bZ5/pX5+f4v3so77y1a6fwO7dvkuEN75VMD6tK5U01ee3NBfnzz23l54oIkSdu2yfDDemXn7TqlV/e1Mv+DpXny+bn5+a/fyYxZS1bXbVUF393HU68jDk+f447NWut1y/yXX85rF1+aec8+t8K5hZqafOKEEen+6YNS2717FrzxZiZe+f3MevTRZXM+cfyIdNt3n3TYZOMsra/P3KeezhtXfj8L3nxzdd0SAAAAAPAx0ORtY9955530799/pef79++fyZMnt0gogP/t8AO65zP/uX5+cMNb+eqFL2dh/dL89zc2S7t2hZW+52vHb5Adt+mYS3/6Zk4698U88ezcXHxm33Tr0i5JUrtWm/TdaO3ccs97+fL5L+fCH0zMJ3rW5lunb7q6bqsq+O4+ntYbsl82+caoTPrJT/PkkUdl/kuvZJsfX512XbuscP5Gp56Snp89LK9ffGmeOOSzeff2O7LlFZdlnS02XzanbqcBefe2X+eZY47LcyeenEJNTba+5kdp06H96rotAAAAAEpULLRxlPGoVk2+8w8++CDt26/8Dxhra2uzcOHCFgkF8L8NG7J+br13Sh57ck4mvrUwl/70zXTr3C6f2rFuhfPXalfIHjt1zrW3vZtnX5qfd6Yuyk13T8k7U+tz0D7dkiQfLFia0d99LX96fFbenlKfF1/7IFf/8u3032TtrN+13eq8vTWa7+7jqc8xX8iUO+/K1N/ekwWvT8yr37koDQsXpsewz6xw/voHHpi3r70uMx8Zn/rJkzPl9jsy85Hx6XPsMcvmPHfKqZl6z7354LXXM//lV/Ly+Rekfe9e6bjlVqvrtgAAAACAj4EmbxubJPfff3/q6lb8B86zZs1qiTwAjfRcf61069wuE56bt2zsgwVL8+LrH2TLvuvk4b/OWu49bdsW0rZtIYsWL200Xr9oabbu13Gln7VOh7ZZurSY+R80tFj+aua7+3gq1NSk45Zb5q2fX//vwWIxs/7y16y73XYrfE+btdpl6aL6RmNL6+vTaYcdVvo5NR3XTZIsmTN7lTMDAAAAAGuOksrL44477kPPFwor3wbwf6uvr099feM/5KytrU1tbW0pcYAq0LXun/+ZmjV7caPxWXMWLzv3fy1YuDTPvzI/Rx3cM5PeeSOzZi/JXrt1yZZ918k779Wv8D3t2hVywhG989BfZuaDhUtXOIfS+O4+ntp16ZxCTU0WT5/RaHzx9BlZe5ONV/iemY8+lt7HHJ3ZT0zIwrfeTueBu6TbPnun0Lbtij+kUMimZ34js598Mh+8+loL3wEAAAAA8HHW5G1jly5d+pFHQ0PTVryMGTMmdXV1jY4xY8Y0+yaANcfeu3XJ3T/ZdtnRtm3T/lLE/3XpT99MoZDcetU2+Z+fb59h/7leHvrLzBSLy89t2zY598sbJ0l+8Iu3VyF9dfPdVa/XL/1uFr45KQPuvjO7//2v2XT0WXnvt/emuHTFZfJm55ydtTfbLC+dOXo1JwUAAAAAKl1JKy9LceCBB+baa69Nr169ljs3evTojBo1qtGYVZdAkvzlydl56bX5y163a/fPv2PRua5dZsxesmy8c6d2eW3SgpVe592pi3LGmFdTu1abrNOhTWbMXpJzTtko705tvHrvX+VXj25r5cyLX7VybxX47tYMi2fOSnHJkrTr1rXReLtuXbPo/ekrfM+SmbPywte+nsJaa6Vd57osmjotG5/+lSycPHm5uZuOPitd/2PPPHP8F7No6tRWuQcAAAAA4OOrySsvS/WnP/0pCxas+A+na2tr06lTp0aH8hJI/rlt6DtTFy073py8MNNnLc4nt/r38w7Xbt8mW2y6dl54df6HXOmf6hctzYzZS9Jx7bYZsE2nPPbknGXn/lV+9elRm7MvfTVz53te4qrw3a0ZikuWZN4LL6TzwF3+PVgopPPAXTL3mWc+/L2LFmXR1Gkp1NSk2777ZsaDDzc6v+nos9Jtn73zj5Enpn7yO60RHwAAAAD4mGu1lZcALeXu+6fl8wf3yOT36jNl2qIcd2ivTJ+1OI9OmL1szsVnbpZHJ8zOPX98P0kyYJt1Uygkb71bnz491soXj+iTt95dmD/8+Z8rx9q2Tc47dZP03ahDzr/i9bRpU0iX//ccxrnzGrKkYQV7lFIy393H0+Rf3pz+3/5W5j33fOY++1x6H31U2nbokPfuvidJ0v87F6Z+6tS8+f0fJkk6brtNart3z7wXX0pt9+7Z8OQTU2hTyNs33LDsmpudc3bWHzo0z5/+tTTM/yDtunVLkjTMm5el9St+nikAAAAAUH2Ul0DF+/XvpqZ9bZt8dfgG6bh22zz3yvyce9nrWbz43yVVr+616dTx3/9JW2ftthnxuV5Zr0u7zJ3fkPF/n5Xr73g3/3o073pd1spuO9YlSX78nS0afd4ZY17NMy/Oa/0bqwK+u4+n9+//Q9p16ZINTzk5a63XLfNfeinPnnJqFs+YkSSp7dmz0fMs26y1Vjb68ilp/4k+afjgg8x8ZHxePve/0jD3399FryMOT5Jsd921jT7r5fMuyNR77l0NdwUAAABAyQqFciegChWKxWKrLFFZd9118/TTT2fTTTdtjcsDq2jIcU+VOwJUnfOeOr7cEWiGPZ6eUO4IAAAAAGUx92+/K3eEqrbuzgeUO0JZtNozLwEAAAAAAABKobwEAAAAAAAAKkKrlZfnnHNOunbt2lqXBwAAAAAAANYwNU2d+Kc//alJ8/7jP/4jSTJ69OjmJQIAAAAAAACqUpPLy7322iuFQiFJUiwWVzinUCikoaGhZZIBAAAAAAAAVaXJ5WWXLl2y7rrrZvjw4TnmmGOy3nrrtWYuAAAAAAAAyqnQak8fhJVq8q/u3XffzSWXXJLHHnss2267bU444YQ8+uij6dSpU+rq6pYdAAAAAAAAAM3R5PJyrbXWyhFHHJH7778/L774Yrbbbruceuqp2WCDDXLuuedmyZIlrZkTAAAAAAAAWMM1a73vhhtumPPPPz9//OMf079//1x88cWZM2dOS2cDAAAAAAAAqkjJ5WV9fX1uueWWDB48ONtss03WW2+9/H//3/+Xrl27tkY+AAAAAAAAoErUNHXi448/nuuvvz6/+tWvsvHGG2fEiBH59a9/rbQEAAAAAAAAWkSTy8tdd901G264Yb7yla9kwIABSZJHHnlkuXkHH3xwy6UDAAAAAAAAqkaTy8skmTRpUr797W+v9HyhUEhDQ8MqhwIAAAAAAACqT5PLy6VLl7ZmDgAAAAAAACpIsVAodwSqUJumTnz++ec/cs53v/vdVQoDAAAAAAAAVK8ml5dDhgzJpEmTVnr+sssuy7nnntsioQAAAAAAAIDq0+Tyco899sjgwYMzbdq05c5dfvnlOeecc3LjjTe2aDgAAAAAAACgejS5vPzlL3+ZzTbbLEOGDMmcOXOWjV9xxRU5++yzc8MNN+TII49slZAAAAAAAADAmq/J5WVNTU3uvPPOdOjQIQcddFAWLlyYK6+8MmeccUauv/76HHXUUa2ZEwAAAAAAAFjD1ZQyuUOHDvnd736XQYMGZcCAAXn55Zdz/fXX5+ijj26tfAAAAAAAAECVaHJ5ec899yz795NPPjlf/epXM2zYsNTV1TU6d/DBB7dsQgAAAAAAAFa/QpM38IQW0+TyctiwYcuN/eY3v8lvfvObZa8LhUIaGhpaJBgAAAAAAABQXZpcXi5durQ1cwAAAAAAAABVrkXX+y5YsKAlLwcAAAAAAABUkRYpL+vr63P55Zdnk002aYnLAQAAAAAAAFWoyeVlfX19Ro8enZ122imf+tSncvfddydJrr/++myyySa58sor87Wvfa21cgIAAAAAAABruCY/8/L888/PT37ykwwePDiPPvpoPve5z2XEiBH5y1/+ku9973v53Oc+l7Zt27ZmVgAAAAAAAGAN1uTy8vbbb8+NN96Ygw8+OM8++2y22267LFmyJE8//XQKhUJrZgQAAAAAAACqQJPLy7fffjsDBgxIkmyzzTapra3N1772NcUlAAAAAADAGqgYHRCrX5OfednQ0JC11lpr2euampp07NixVUIBAAAAAAAA1afJKy+LxWKGDx+e2traJMnChQtz0kknZZ111mk0784772zZhAAAAAAAAEBVaHJ5edxxxzV6ffTRR7d4GAAAAAAAAKB6Nbm8vP7661szBwAAAAAAAFDlmvzMSwAAAAAAAIDWVFJ5+eCDD+byyy/P+PHjkyQ/+clPsuGGG2b99dfPyJEjs2DBglYJCQAAAAAAAKz5mrxt7M9+9rOcfPLJ2WSTTXLuuefmggsuyEUXXZRjjjkmbdq0yU033ZRu3brl4osvbs28AAAAAAAArAbFgg08Wf2a/Ku76qqrcsUVV+SVV17J3XffnfPPPz9XX311fvzjH+fqq6/OtddemzvuuKM1swIAAAAAAABrsCaXl6+//noOPvjgJMn++++fQqGQXXbZZdn5gQMH5q233mr5hAAAAAAAAEBVaHJ5uXDhwnTo0GHZ69ra2tTW1jZ6vWTJkpZNBwAAAAAAAFSNJj/zslAoZO7cuWnfvn2KxWIKhULmzZuXOXPmJMmyfwIAAAAAAAA0R5PLy2KxmP79+zd6/clPfrLR60Kh0LLpAAAAAAAAgKrR5PLywQcfbM0cAAAAAAAAQJVrcnk5aNCg1swBAAAAAAAAVLkml5cAAAAAAABUkUKbciegCjW5vGzTps1HPtOyUChkyZIlqxwKAAAAAAAAqD5NLi/vuuuulZ577LHH8v3vfz9Lly5tkVAAAAAAAABA9WlyefmZz3xmubGXXnopZ599du6999584QtfyIUXXtii4QAAAAAAAIDq0azNit95552MHDky2267bZYsWZKnnnoqv/jFL7LRRhu1dD4AAAAAAACgSpRUXs6ePTtnnXVW+vbtm+eeey7jxo3Lvffem2222aa18gEAAAAAAABVosnbxl566aW55JJL0rNnz9x6660r3EYWAAAAAAAAoLmaXF6effbZ6dChQ/r27Ztf/OIX+cUvfrHCeXfeeWeLhQMAAAAAAKA8ioVCuSNQhZpcXh577LEp+JECAAAAAAAAraTJ5eUNN9zQijEAAAAAAACAatem3AEAAAAAAAAAEuUlAAAAAAAAUCGUlwAAAAAAAEBFUF4CAAAAAAAAFUF5CQAAAAAAAFSEmnIHAAAAAAAAoPIUC9bAsfr51QEAAAAAAAAVQXkJAAAAAAAAVATlJQAAAAAAAFARlJcAAAAAAABARSgUi8ViuUMAAAD8y/OH7FvuCDTTVneNK3cEAACgBc145s/ljlDVum63Z7kjlIWVlwAAAAAAAEBFqCl3AAAAAAAAACpQoVDuBFQhKy8BAAAAAACAiqC8BAAAAAAAACqC8hIAAAAAAACoCMpLAAAAAAAAoCIoLwEAAAAAAICKoLwEAAAAAAAAKoLyEgAAAAAAAKgINeUOAAAAAAAAQOUpFqyBY/XzqwMAAAAAAAAqgvISAAAAAAAAqAjKSwAAAAAAAKAiKC8BAAAAAACAiqC8BAAAAAAAACqC8hIAAAAAAACoCDXlDgAAAAAAAEDlKaZQ7ghUISsvAQAAAAAAgIqgvAQAAAAAAAAqgvISAAAAAAAAqAjKSwAAAAAAAKAiKC8BAAAAAACAiqC8BAAAAAAAACqC8hIAAAAAAACoCDXlDgAAAAAAAEDlKRasgWP186sDAAAAAAAAKoLyEgAAAAAAAKgIyksAAAAAAACgIigvAQAAAAAAgIqgvAQAAAAAAAAqgvISAAAAAAAAqAg15Q4AAAAAAABABSoUyp2AKmTlJQAAAAAAAFARlJcAAAAAAABARVBeAgAAAAAAABVBeQkAAAAAAABUBOUlAAAAAAAAUBGUlwAAAAAAAEBFUF4CAAAAAAAAFaGm3AEAAAAAAACoPEVr4CgDvzoAAAAAAACgIigvAQAAAAAAgIqgvAQAAAAAAAAqgvISAAAAAAAAqAjKSwAAAAAAAKAitGh5+dZbb+X4449vyUsCAAAAAAAAVaJFy8sZM2bkF7/4RUteEgAAAAAAgDIoFgqOMh7VqqaUyffcc8+Hnn/99ddXKQwAAAAAAABQvUoqL4cNG5ZCoZBisbjSOYUqboIBAAAAAACA5itp29hevXrlzjvvzNKlS1d4TJgwobVyAgAAAAAAAGu4ksrLAQMG5Iknnljp+Y9alQkAAAAAAACwMiVtG3vGGWdk/vz5Kz3ft2/fPPjgg6scCgAAAAAAAKg+JZWXe+6554eeX2eddTJo0KBVCgQAAAAAAABUp5K2jV2ZJUuWtMRlAAAAAAAAgCpWUnl533335R//+EeSZOnSpfn2t7+dPn36pLa2Np/4xCdy8cUXe+YlAAAAAADAGqBYaOMo41GtSto29vTTT8/PfvazJMkll1ySq666Kueee2623HLLvPTSSxkzZkwKhULOOuusVgkLAAAAAAAArLlKKi/feOONbLTRRkmSW265JT/+8Y/zuc99Lkmy//77p2/fvjn99NOVlwAAAAAAAEDJSlpz2rVr17zzzjtJkmnTpqVv376Nzvfv3z+TJ09uuXQAAAAAAABA1SipvDzkkENy0UUXpaGhIZ/5zGfyox/9qNEzLn/wgx9khx12aOmMAAAAAAAAQBUoadvY//7v/87gwYOzxRZbZLfddsvtt9+esWPHpn///nn11VczY8aM3H///a2VFQAAAAAAAFiDlbTysq6uLo8++mi+/vWvZ/r06dl4441TW1ubRYsW5fOf/3yeffbZDBw4sLWyAgAAAAAAAGuwQvF/7/vawsaPH5+ddtoptbW1rfURAADAGub5Q/YtdwSaaau7xpU7AgAA0ILeffGpckeoar222KHcEcqipG1jSzV06NA89dRT2XTTTVvzYwAAAMqiy9DPpNuww1PTuWvq33gt7177gyx85aWVzu960KHpsv/Babde9zTMnZ05j/4pU2+6NsXFi5Mka2+1bboNOyLtN+uXdl3Xy1tjzs/cx8evrtsBAACAsitp29hSteKiTgAAgLLqtPte6THipEy77ca8/vWTsvCN17LR+ZekbV3nFc/fc590P2Zkpt12Y147bUTe+eFl6bTHXul+9BeXzWnTvkMWvvFapvz0+6vpLgAAAKCytOrKSwAAgDVVt4M/m1ljf5fZD9yfJHn3mivTccCu6bzv/pl+56+Wm7/2FltnwYvPZs6fH0iSLJ72Xub8+cF06L/FsjnzJjyeeRMeXz03AAAAABWoVVdeAgAArJFqatJ+s/6Z//SEf48Vi5n/zISsvflWK3zLBy8+l/ab9U/7fpsnSdr16JWOA3bJvCeUlQAAAPAvVl4CAACUqGbduhTats2S2TMbjS+ZNTO1fTZY4Xvm/PmB1HSqyyYXXZUUCinU1GTGfffk/d/csjoiAwAAwMdCq5aXhUJhheP19fWpr69vNFZbW5va2trWjAMAAFA2a2+9fdY77Ki8+9PvZ8HLL2StXr3T84QvZ8nnjs77t99U7ngAAABQEVp129hisbjC8TFjxqSurq7RMWbMmNaMAgAA0GKWzJ2dYkNDauq6NBqv6dwlS2bNWOF7uh81IrMeHptZf/xd6idNzNy/js/Um6/Leod9PlnJX/wEAACAatOqKy/nzp27wvHRo0dn1KhRjcasugQAAD42lizJwtdezjrbfTJzHx//z7FCIets+8nM+P3dK3xLobY2Wdr4L3gWGxqWvTcr+cufAAAA5VIstOoaOFihJpeXO+64Y8aNG5cuXbrkk5/85Eq3hE2SCRMmfOi1bBELAAB83E2/5470/spZWfDay1nwyovpdtBhadO+fWaNuz9J0vsrZ2XJjPcz9aafJ0nm/e2xdD34s1k48dX/t21sn3Q/akTm/u2xZOnSJEmhffus1bPPss9o16NnajfeLA3z5mbJ+1NX/00CAADAatbk8vIzn/nMssJx2LBhrZUHAADgY2HO+IfStlNd1j9yeGq6dEn9xNcy6cKz0zB7ZpKk3frdG62mnHb7TSkWi+l+1IjUdF0vDXNmZe7f/7Ks3EySDpttno2/871lr3sef0qSZNYD9+edH1y6mu4MAAAAyqdQXNmDKQEAAMrg+UP2LXcEmmmru8aVOwIAANCC3nnpmXJHqGq9N9+u3BHKotmbFc+aNSvXXnttRo8enRkzZiT553axkydPbrFwAAAAAAAAQPVo8rax/9szzzyTwYMHp66uLm+88UZGjhyZrl275s4778ykSZNy4403tnROAAAAAAAAYA3XrJWXo0aNyvDhw/PKK6+kffv2y8YPOOCA/OlPf2qxcAAAAAAAAED1aNbKy7/97W/5yU9+stx4nz59MmXKlFUOBQAAAAAAQHkVC4VyR6AKNWvlZW1tbebMmbPc+Msvv5z1119/lUMBAAAAAAAA1adZ5eXBBx+cCy+8MIsXL06SFAqFTJo0KWeddVYOO+ywFg0IAAAAAAAAVIdmlZeXX3555s2bl+7du2fBggUZNGhQ+vbtm3XXXTcXXXRRS2cEAAAAAAAAqkCznnlZV1eXsWPHZvz48Xn66aczb9687Ljjjhk8eHBL5wMAAAAAAACqRLPKy3/Zfffds/vuuydJZs2a1RJ5AAAAAAAAgCrVrG1jL7nkktx2223LXh9++OHp1q1b+vTpk6effrrFwgEAAAAAAADVo1nl5TXXXJMNNtggSTJ27NiMHTs2v//97zN06NCcccYZLRoQAAAAAAAAqA7N2jZ2ypQpy8rL//mf/8nhhx+e/fbbLxtvvHEGDhzYogEBAAAAAABY/YoplDsCVahZKy+7dOmSt956K0ly3333ZfDgwUmSYrGYhoaGlksHAAAAAAAAVI1mrbw89NBDc9RRR6Vfv36ZPn16hg4dmiR58skn07dv3xYNCAAAAAAAAFSHZpWXV1xxRTbZZJNMmjQpl156aTp27Jgkeffdd/PlL3+5RQMCAAAAAAAA1aFZ5eVll12WHj165Otf/3qj8bq6ukybNq1FggEAAAAAAADVpVnPvPzJT36SLbbYYrnxrbfeOtdcc80qhwIAAAAAAACqT7PKyylTpqRXr17Lja+//vp59913VzkUAAAAAAAAUH2aVV5usMEGGT9+/HLj48ePT+/evVc5FAAAAAAAAFB9mvXMy5EjR+b000/P4sWLs88++yRJxo0blzPPPHO552ACAAAAAADw8VMsNGsNHKySZpWXZ5xxRqZPn55TTjklixYtSpK0b98+Z511VkaPHt2iAQEAAAAAAIDqUCgWi8XmvnnevHl54YUX0qFDh/Tr1y+1tbUtmQ0AAKhCzx+yb7kj0Exb3TWu3BEAAIAW9NYrz5c7QlXboN9W5Y5QFs1aefkvHTt2zM4779xSWQAAAAAAAIAqZrNiAAAAAAAAoCIoLwEAAAAAAICKoLwEAAAAAAAAKsIqPfMSAAAAAACANVMxhXJHoApZeQkAAAAAAABUBOUlAAAAAAAAUBGUlwAAAAAAAEBFUF4CAAAAAAAAFUF5CQAAAAAAAFQE5SUAAAAAAABQEZSXAAAAAAAAQEWoKXcAAAAAAAAAKk+xYA0cq59fHQAAAAAAAFARlJcAAAAAAABARVBeAgAAAAAAABVBeQkAAAAAAABUBOUlAAAAAAAAUBGUlwAAAAAAAEBFqCl3AAAAAAAAACpPMYVyR6AKWXkJAAAAAAAAVATlJQAAAAAAAFARlJcAAAAAAABARVBeAgAAAAAAABVBeQkAAAAAAABUBOUlAAAAAAAAUBGUlwAAAAAAAEBFqCl3AAAAAAAAACpPsWANHKufXx0AAAAAAABQEZSXAAAAAAAAQEVQXgIAAAAAAAAVQXkJAAAAAAAAVATlJQAAAAAAAFARlJcAAAAAAABARagpdwAAAAAAAAAqTzGFckegCll5CQAAAAAAAFQE5SUAAAAAAABQEZSXAAAAAAAAsAa4+uqrs/HGG6d9+/YZOHBgHn/88Q+df/vtt2eLLbZI+/bts+222+Z3v/tdo/PFYjHnn39+evXqlQ4dOmTw4MF55ZVXWvMWlJcAAAAAAADwcXfbbbdl1KhRueCCCzJhwoRsv/32GTJkSKZOnbrC+Y8++mg+//nP54QTTsiTTz6ZYcOGZdiwYXn22WeXzbn00kvz/e9/P9dcc03++te/Zp111smQIUOycOHCVruPQrFYLLba1QEAAEr0/CH7ljsCzbTVXePKHQEAAGhBr7/2WrkjVLVNN9uspPkDBw7MzjvvnB/+8IdJkqVLl2aDDTbIaaedlrPPPnu5+UcccUTmz5+f//mf/1k2tuuuu2aHHXbINddck2KxmN69e+frX/96vvGNbyRJZs+enR49euSGG27IkUceuQp3t3JWXgIAAAAAAMDH2KJFi/LEE09k8ODBy8batGmTwYMH57HHHlvhex577LFG85NkyJAhy+ZPnDgxU6ZMaTSnrq4uAwcOXOk1W0JNq10ZAAAAAAAAaJb6+vrU19c3GqutrU1tbe1yc99///00NDSkR48ejcZ79OiRF198cYXXnzJlygrnT5kyZdn5f42tbE5rUF4CALBGsvXox5etRwEAACpDsVAod4SqNmbMmHzrW99qNHbBBRfkm9/8ZnkCrSbKSwAAAAAAAKgwo0ePzqhRoxqNrWjVZZKst956adu2bd57771G4++991569uy5wvf07NnzQ+f/65/vvfdeevXq1WjODjvsUNK9lMIzLwEAAAAAAKDC1NbWplOnTo2OlZWXa621VgYMGJBx4/69m9HSpUszbty47Lbbbit8z2677dZofpKMHTt22fxNNtkkPXv2bDRnzpw5+etf/7rSa7YEKy8BAAAAAADgY27UqFE57rjjstNOO2WXXXbJlVdemfnz52fEiBFJkmOPPTZ9+vTJmDFjkiRf/epXM2jQoFx++eU58MAD86tf/Sp///vf89Of/jRJUigUcvrpp+c73/lO+vXrl0022STnnXdeevfunWHDhrXafSgvAQAAAAAA4GPuiCOOyLRp03L++ednypQp2WGHHXLfffelR48eSZJJkyalTZt/b8r6qU99Krfcckv+67/+K+ecc0769euXu+++O9tss82yOWeeeWbmz5+fL33pS5k1a1b22GOP3HfffWnfvn2r3UehWCwWW+3qAABQJs8fsm+5I9BMW9017qMnAQAA0Opee/31ckeoapttumm5I5SFZ14CAAAAAAAAFcG2sQAAAAAAACynWCyUOwJVyMpLAAAAAAAAoCIoLwEAAAAAAICKoLwEAAAAAAAAKoLyEgAAAAAAAKgIyksAAAAAAACgIigvAQAAAAAAgIqgvAQAAAAAAAAqQk25AwAAAAAAAFB5itbAUQZ+dQAAAAAAAEBFUF4CAAAAAAAAFUF5CQAAAAAAAFQE5SUAAAAAAABQEZSXAAAAAAAAQEVQXgIAAAAAAAAVoabcAQAAAAAAAKg8xRTKHYEqZOUlAAAAAAAAUBGUlwAAAAAAAEBFUF4CAAAAAAAAFUF5CQAAAAAAAFQE5SUAAAAAAABQEZSXAAAAAAAAQEVQXgIAAAAAAAAVoabcAQAAAAAAAKg8xRTKHYEqZOUlAAAAAAAAUBGUlwAAAAAAAEBFUF4CAAAAAAAAFUF5CQAAAAAAAFQE5SUAAAAAAABQEZSXAAAAAAAAQEWoKXcAAAAAAAAAKk8xhXJHoApZeQkAAAAAAABUBOUlAAAAAAAAUBGUlwAAAAAAAEBFUF4CAAAAAAAAFUF5CQAAAAAAAFQE5SUAAAAAAABQEZSXAAAAAAAAQEWoKXcAAAAAAAAAKk8xhXJHoApZeQkAAAAAAABUBOUlAAAAAAAAUBGUlwAAAAAAAEBFKKm8nDBhQiZOnLjs9S9/+cvsvvvu2WCDDbLHHnvkV7/6VYsHBAAAAAAAAKpDSeXliBEj8tprryVJrr322px44onZaaedcu6552bnnXfOyJEjc91117VKUAAAAAAAAGDNVlPK5FdeeSX9+vVLkvzoRz/KVVddlZEjRy47v/POO+eiiy7K8ccf37IpAQAAAAAAgDVeSeXl2muvnffffz8bbbRRJk+enF122aXR+YEDBzbaVhYAAAAAAICPp2KxUO4IVKGSto0dOnRofvzjHydJBg0alDvuuKPR+V//+tfp27dvy6UDAAAAAAAAqkZJKy8vueSS7L777hk0aFB22mmnXH755XnooYey5ZZb5qWXXspf/vKX3HXXXa2VFQAAAAAAAFiDlbTysnfv3nnyySez22675b777kuxWMzjjz+eP/zhD/nEJz6R8ePH54ADDmitrAAAAAAAAMAarFAsFoutdfG33347vXv3Tps2JXWkAACwyp4/ZN9yR6CZtrprXLkjAAAAkOT5V98pd4SqtlXf3uWOUBat2iputdVWeeONN1rzIwAAAAAAAIA1RKuWl624qBMAAAAAAABYw9jPFQAAAAAAAKgINeUOAAAAAAAAQOUpplDuCFQhKy8BAAAAAACAitCq5WWhoJEHAAAAAAAAmqZVt40tFouteXkAAFgjdBn6mXQbdnhqOndN/Ruv5d1rf5CFr7y00vldDzo0XfY/OO3W656GubMz59E/ZepN16a4eHGSZO2ttk23YUek/Wb90q7renlrzPmZ+/j41XU7AAAAAM1W8srLxYsXp6amJs8+++xHzn3++eez0UYbNSsYAABUg06775UeI07KtNtuzOtfPykL33gtG51/SdrWdV7x/D33SfdjRmbabTfmtdNG5J0fXpZOe+yV7kd/cdmcNu07ZOEbr2XKT7+/mu4CAAAAoGWUvPKyXbt22XDDDdPQ0PCRczfYYINmhQIAgGrR7eDPZtbY32X2A/cnSd695sp0HLBrOu+7f6bf+avl5q+9xdZZ8OKzmfPnB5Iki6e9lzl/fjAd+m+xbM68CY9n3oTHV88NAAAAALSgZj3z8txzz80555yTGTNmtHQeAACoHjU1ab9Z/8x/esK/x4rFzH9mQtbefKsVvuWDF59L+836p32/zZMk7Xr0SscBu2TeE8pKAAAA4OOvWc+8/OEPf5hXX301vXv3zkYbbZR11lmn0fkJEyas5J3/VF9fn/r6+kZjtbW1qa2tbU4cAAD4WKpZty6Ftm2zZPbMRuNLZs1MbZ8V72Iy588PpKZTXTa56KqkUEihpiYz7rsn7//mltURGQAAgCpSTKHcEahCzSovhw0btkofOmbMmHzrW99qNHbBBRfkm9/85ipdFwAA1nRrb7191jvsqLz70+9nwcsvZK1evdPzhC9nyeeOzvu331TueAAAAACrpFnl5QUXXLBKHzp69OiMGjWq0ZhVlwAAVJslc2en2NCQmroujcZrOnfJklkrfkRD96NGZNbDYzPrj79LktRPmpg27Tuk18lfy/t33JwUi62eGwAAAKC1NOuZl0kya9asXHvttRk9evSyZ19OmDAhkydP/sj31tbWplOnTo0O5SUAAFVnyZIsfO3lrLPdJ/89VihknW0/mQ9een6FbynU1iZLGxeUxYaGZe8FAAAA+Dhr1srLZ555JoMHD05dXV3eeOONjBw5Ml27ds2dd96ZSZMm5cYbb2zpnAAAsEaafs8d6f2Vs7LgtZez4JUX0+2gw9KmffvMGnd/kqT3V87KkhnvZ+pNP0+SzPvbY+l68GezcOKr/2/b2D7pftSIzP3bY8nSpUmSQvv2Watnn2Wf0a5Hz9RuvFka5s3Nkvenrv6bBAAAAGiiZpWXo0aNyvDhw3PppZdm3XXXXTZ+wAEH5KijjmqxcAAAsKabM/6htO1Ul/WPHJ6aLl1SP/G1TLrw7DTMnpkkabd+90ZbwU67/aYUi8V0P2pEarqul4Y5szL3739ZVm4mSYfNNs/G3/nestc9jz8lSTLrgfvzzg8uXU13BgAAAFC6QrFY+kNx6urqMmHChGy22WZZd9118/TTT2fTTTfNm2++mc033zwLFy5sjawAANBkzx+yb7kj0Exb3TWu3BEAAABI8uyrU8odoapt07dnuSOURbOeeVlbW5s5c+YsN/7yyy9n/fXXX+VQAAAAAAAAQPVpVnl58MEH58ILL8zixYuTJIVCIZMmTcpZZ52Vww47rEUDAgAAAAAAsPoVU3CU8ahWzSovL7/88sybNy/du3fPggULMmjQoPTt2zfrrrtuLrroopbOCAAAAAAAAFSBmua8qa6uLmPHjs348ePz9NNPZ968edlxxx0zePDgls4HAAAAAAAAVIlmlZf/svvuu2f33XdPksyaNasl8gAAAAAAAABVqlnbxl5yySW57bbblr0+/PDD061bt/Tp0ydPP/10i4UDAAAAAAAAqkezystrrrkmG2ywQZJk7NixGTt2bH7/+99n6NChOeOMM1o0IAAAAAAAAFAdmrVt7JQpU5aVl//zP/+Tww8/PPvtt1823njjDBw4sEUDAgAAAAAAANWhWeVlly5d8tZbb2WDDTbIfffdl+985ztJkmKxmIaGhhYNCAAAAAAAwOpXLBbKHYEq1Kzy8tBDD81RRx2Vfv36Zfr06Rk6dGiS5Mknn0zfvn1bNCAAAAAAAABQHZpVXl5xxRXZZJNNMmnSpFx66aXp2LFjkuTdd9/Nl7/85RYNCAAAAAAAAFSHZpWXl112WXr06JGvf/3rjcbr6uoybdq0FgkGAAAAAAAAVJc2zXnTT37yk2yxxRbLjW+99da55pprVjkUAAAAAAAAUH2aVV5OmTIlvXr1Wm58/fXXz7vvvrvKoQAAAAAAAIDq06zycoMNNsj48eOXGx8/fnx69+69yqEAAAAAAACA6tOsZ16OHDkyp59+ehYvXpx99tknSTJu3LiceeaZyz0HEwAAAAAAAKApmlVennHGGZk+fXpOOeWULFq0KEnSvn37nHXWWRk9enSLBgQAAAAAAGD1W5pCuSNQhQrFYrHY3DfPmzcvL7zwQjp06JB+/fqltra2JbMBAECzPX/IvuWOQDNtdde4ckcAAAAgyVOvTCt3hKq2Q7/1yx2hLJq18vJfOnbsmJ133rmlsgAAAAAAAABVrE25AwAAAAAAAAAkyksAAAAAAACgQigvAQAAAAAAgIqgvAQAAAAAAAAqQk25AwAAAAAAAFB5iimUOwJVyMpLAAAAAAAAoCIoLwEAAAAAAICKoLwEAAAAAAAAKoLyEgAAAAAAAKgIyksAAAAAAACgIigvAQAAAAAAgIqgvAQAAAAAAAAqQk25AwAAAAAAAFB5isVCuSNQhay8BAAAAAAAACqC8hIAAAAAAACoCMpLAAAAAAAAoCIoLwEAAAAAAICKoLwEAAAAAAAAKoLyEgAAAAAAAKgINeUOAAAAAAAAQOUpplDuCFQhKy8BAAAAAACAiqC8BAAAAAAAACqC8hIAAAAAAACoCMpLAAAAAAAAoCIoLwEAAAAAAICKoLwEAAAAAAAAKoLyEgAAAAAAAKgINeUOAAAAAAAAQOUpFgvljkAVsvISAAAAAAAAqAjKSwAAAAAAAKAiKC8BAAAAAACAiqC8BAAAAAAAACqC8hIAAAAAAACoCMpLAAAAAAAAoCLUlDsAAAAAAAAAlaeYQrkjUIWsvAQAAAAAAAAqgvISAAAAAAAAqAjKSwAAAAAAAKAiKC8BAAAAAACAiqC8BAAAAAAAACqC8hIAAAAAAACoCMpLAAAAAAAAoCLUlDsAAAAAAAAAladYLJQ7AlXIyksAAAAAAACgIlh5CQDAGmmru8aVOwLAx8Ien3643BFopkfuHVTuCAAA0OKsvAQAAAAAAAAqgvISAAAAAAAAqAjKSwAAAAAAAKAiKC8BAAAAAACAiqC8BAAAAAAAACpCTbkDAAAAAAAAUHmWljsAVcnKSwAAAAAAAKAiKC8BAAAAAACAiqC8BAAAAAAAACqC8hIAAAAAAACoCMpLAAAAAAAAoCIoLwEAAAAAAICKUFPuAAAAAAAAAFSeYrFQ7ghUISsvAQAAAAAAgIqgvAQAAAAAAAAqgvISAAAAAAAAqAjKSwAAAAAAAKAiKC8BAAAAAACAiqC8BAAAAAAAACqC8hIAAAAAAACoCDXlDgAAAAAAAEDlKaZQ7ghUISsvAQAAAAAAgIqgvAQAAAAAAAAqgvISAAAAAAAAqAjKSwAAAAAAAKAiKC8BAAAAAACAiqC8BAAAAAAAACpCTbkDAAAAAAAAUHmKxUK5I1CFrLwEAAAAAAAAKoLyEgAAAAAAAKgIyksAAAAAAACgIigvAQAAAAAAgIqgvAQAAAAAAAAqgvISAAAAAAAAqAjKSwAAAAAAAKAi1JQ7AAAAAAAAAJWnmEK5I1CFrLwEAAAAAAAAKoLyEgAAAAAAAKgIyksAAAAAAACgIigvAQAAAAAAgIpQUnm5ePHinHnmmenbt2922WWXXHfddY3Ov/fee2nbtm2LBgQAAAAAAACqQ0nl5UUXXZQbb7wxJ510Uvbbb7+MGjUqJ554YqM5xWKxRQMCAAAAAAAA1aGmlMk333xzrr322hx00EFJkuHDh2fo0KEZMWLEslWYhUKh5VMCAAAAAACwWi21Xo0yKGnl5eTJk7PNNtsse923b9889NBDefTRR3PMMcekoaGhxQMCAAAAAAAA1aGk8rJnz5557bXXGo316dMnDz74YP72t79l+PDhLZkNAAAAAAAAqCIllZf77LNPbrnlluXGe/funQceeCATJ05ssWAAAAAAAABAdSnpmZfnnXdeXnzxxRWe69OnTx5++OGMHTu2RYIBAAAAAAAA1aWklZcbbbRRhgwZstLzvXv3znHHHbfs9YEHHph33323+ekAAAAAAACAqlFSeVmqP/3pT1mwYEFrfgQAAAAAAACwhmjV8hIAAAAAAACgqUp65iUAAAAAAADVoZhCuSNQhay8BAAAAAAAACqC8hIAAAAAAACoCMpLAAAAAAAAoCK0anl5zjnnpGvXrq35EQAAAAAAAMAaoqY5b5o+fXq6deuWJHnrrbfys5/9LAsWLMjBBx+cPffcc9m80aNHt0xKAAAAAAAAYI1X0srLf/zjH9l4443TvXv3bLHFFnnqqaey884754orrshPf/rT7L333rn77rtbKSoAAAAAAACwJiupvDzzzDOz7bbb5k9/+lP22muvHHTQQTnwwAMze/bszJw5MyeeeGIuvvji1soKAAAAAADAalIsFhxlPKpVoVgsFps6eb311ssDDzyQ7bbbLvPmzUunTp3yt7/9LQMGDEiSvPjii9l1110za9as1soLAAAAtKA9Pv1wuSPQTI/cO6jcEQCANdxDzy4od4Sqttc2HcodoSxKWnk5Y8aM9OzZM0nSsWPHrLPOOunSpcuy8126dMncuXNbNiEAAAAAAABQFUoqL5OkUCh86GsAAAAAAACA5qgp9Q3Dhw9PbW1tkmThwoU56aSTss466yRJ6uvrWzYdAAAAAAAAUDVKKi+PO+64Rq+PPvro5eYce+yxq5YIAAAAYBX9x27rZdjQXtl8s3VT16ldhn/l73l14vxyxwIAAD5CSeXl9ddf31o5AAAAAFpMh/Zt8szzc/LAI9Ny9mmblzsOAADQRCVvGwsAAABQ6e5/cGqSpGf32jInAQAASlFSeXnooYc2ad6dd97ZrDAAAAAAAABUhmKx3AmoRiWVl506dUqhUGitLAAAAAAAAEAVK6m8vOGGG1rkQ+vr61NfX99orLa2NrW1tnIBAAAASvOfg7rnjC/3X/b6G9/8R555fnYZEwEAAM3VppTJn/3sZ3PfffeluIrrhMeMGZO6urpGx5gxY1bpmgAAAEB1euTx6Rnx1b8vO158dW65IwEAAM1U0srLmTNn5sADD0zv3r0zYsSIDB8+PJtuumnJHzp69OiMGjWq0ZhVlwAAAEBzLFjQkMkLGsodAwAAaAElrbwcN25cXn/99Zxwwgm56aab0q9fv+yzzz655ZZbltsG9sPU1tamU6dOjQ7lJQAAANBS1u1Yk76brJONN1gnSbJhn7XTd5N10rVzuzInAwAAPkyhuAp7wD7wwAO57rrrctddd6W2tjaf//znc/zxx2fAgAEtmREAAABoJXt8+uFyR2gVQ/ftkXNP32K58etueSPX3fpmGRK1vEfuHVTuCADAGu7Bfywod4Sqtve2HcodoSxWqbz8l7lz5+aWW27JOeeck9mzZ2fJkiUtkQ0AAABoZWtqeVkNlJcAQGsb94+F5Y5Q1fbdtn25I5RFSc+8XJGJEyfmhhtuyA033JDZs2dn8ODBLZELAAAAAAAAqDIlPfPyXxYuXJibbrop++yzT/r165cbb7wxJ5xwQiZOnJj77ruvpTMCAAAAAAAAVaCklZePP/54rrvuutx2221ZuHBhDjnkkNx3333Zd999UygUWisjAAAAAAAAUAVKKi933XXXbL/99vn2t7+dL3zhC+nSpUtr5QIAAAAAAACqTEnl5d///vfsuOOOrZUFAAAAAAAAqGIlPfPy0EMPzfHHH59f/vKXeeutt1orEwAAAAAAAFCFSlp5OXz48Dz00EO59dZbs2jRomyyySbZe++9s88++2TvvfdOz549WysnAAAAAAAAsIYrqbz85je/mSSpr6/P+PHj8/DDD+ehhx7KL3/5yyxevDj9+/fPPvvsk6uvvro1sgIAAAAAALCaFIuFckegChWKxWJxVS8yc+bMXH755fnBD36QefPmpaGhoSWyAQAAAK1sj08/XO4INNMj9w4qdwQAYA33x2fqyx2hqg3errbcEcqipJWX/7Jo0aI89thjeeihh/LQQw/lr3/9a/r06ZPPfvazGTTI/3EGAAAAAAAASldSeXnhhRcuKys32mij/Md//Ee+9KUv5eabb07v3r1bKyMAAAAAAABQBUp+5uWGG26Yyy+/PJ/73OfSrVu31soFAAAAAAAAVJk2pUz+/e9/nyOPPDI33HBDevfunW233TannXZa7rjjjkybNq21MgIAAAAAAABVoFAsFovNeePcuXPz5z//OQ8//HAefPDBPP300+nbt2/23nvv/PCHP2zpnAAAAEAr2OPTD5c7As30yL2Dyh0BAFjD/fGZ+nJHqGqDt6std4SyaHZ5+S8NDQ15/PHHc8899+RHP/pR5s2bl4aGhpbKBwAAALQi5eXHl/ISAGhtY59WXpbTf25fneVlSc+8TJKlS5fm73//ex588ME89NBDGT9+fObPn59PfOITOeSQQ7L33nu3Rk4AAAAAAABgDVdSeTl06NA8+uijmTt3bnr37p299947V1xxRfbee+9suummrZURAAAAAAAAqAIllZedO3fOZZddlr322iv9+vVrrUwAAAAAAABAFWpTyuSRI0fmiiuuSI8ePZY7N3v27Gy99db585//3GLhAAAAAAAAgOpRUnl51VVX5Utf+lI6deq03Lm6urqceOKJ+d73vtdi4QAAAAAAAIDqUVJ5+eSTT2bIkCErPb/ffvvliSeeWOVQAAAAAAAAQPUpqbycOnVq2rVrt9LzNTU1mTZt2iqHAgAAAAAAAKpPTSmT+/Tpk2effTZ9+/Zd4flnnnkmvXr1apFgAAAAAAAAlE8xhXJHoAqVtPLygAMOyHnnnZeFCxcud27BggW54IILctBBB7VYOAAAAAAAAKB6FIrFYrGpk997773suOOOadu2bU499dRsvvnmSZIXX3wxV199dRoaGjJhwoT06NGj1QIDAAAALWePTz9c7gg00yP3Dip3BABgDfeHpxeVO0JV22/7tcodoSxKWnnZo0ePPProo9lmm20yevToHHLIITnkkENyzjnnZJtttskjjzyiuAQAAAAAAIAKNWPGjHzhC19Ip06d0rlz55xwwgmZN2/eh84/7bTTsvnmm6dDhw7ZcMMN85WvfCWzZ89uNK9QKCx3/OpXvyo5X0nPvEySjTbaKL/73e8yc+bMvPrqqykWi+nXr1+6dOlS8ocDAAAAAAAAq88XvvCFvPvuuxk7dmwWL16cESNG5Etf+lJuueWWFc5/55138s477+Syyy7LVlttlTfffDMnnXRS3nnnndxxxx2N5l5//fXZf//9l73u3LlzyflK2jYWAAAAWLPYNvbjy7axAEBrs21sebXGtrEvvPBCttpqq/ztb3/LTjvtlCS57777csABB+Ttt99O7969m3Sd22+/PUcffXTmz5+fmpp/rpUsFAq56667MmzYsFXKWNK2sQAAAAAAAEDrq6+vz5w5cxod9fX1q3TNxx57LJ07d15WXCbJ4MGD06ZNm/z1r39t8nVmz56dTp06LSsu/+XLX/5y1ltvveyyyy657rrr0pw1lMpLAAAAAAAAlrO06CjnMWbMmNTV1TU6xowZs0rf6ZQpU9K9e/dGYzU1NenatWumTJnSpGu8//77+fa3v50vfelLjcYvvPDC/PrXv87YsWNz2GGH5ZRTTskPfvCDkjOW/MxLAAAAAAAAoHWNHj06o0aNajRWW1u7wrlnn312Lrnkkg+93gsvvLDKmebMmZMDDzwwW221Vb75zW82Onfeeect+/dPfvKTmT9/fr773e/mK1/5SkmfobwEAAAAAACAClNbW7vSsvL/+vrXv57hw4d/6JxNN900PXv2zNSpUxuNL1myJDNmzEjPnj0/9P1z587N/vvvn3XXXTd33XVX2rVr96HzBw4cmG9/+9upr69v8n0kyksAAAAAAAD4WFt//fWz/vrrf+S83XbbLbNmzcoTTzyRAQMGJEkeeOCBLF26NAMHDlzp++bMmZMhQ4aktrY299xzT9q3b/+Rn/XUU0+lS5cuJRWXifISAAAAAAAAqsKWW26Z/fffPyNHjsw111yTxYsX59RTT82RRx6Z3r17J0kmT56cfffdNzfeeGN22WWXzJkzJ/vtt18++OCD3HTTTZkzZ07mzJmT5J+ladu2bXPvvffmvffey6677pr27dtn7Nix+e///u984xvfKDmj8hIAAAAAAACqxM0335xTTz01++67b9q0aZPDDjss3//+95edX7x4cV566aV88MEHSZIJEybkr3/9a5Kkb9++ja41ceLEbLzxxmnXrl2uvvrqfO1rX0uxWEzfvn3zve99LyNHjiw5X6FYLBZX4f4AAACAj7E9Pv1wuSPQTI/cO6jcEQCANdx9Ty0qd4Sqtv8Oa5U7Qlm0KXcAAAAAAAAAgMS2sQAAAAAAAKxAsVgodwSqkJWXAAAAAAAAQEVQXgIAAAAAAAAVQXkJAAAAAAAAVATlJQAAAAAAAFARlJcAAAAAAABARVBeAgAAAAAAABWhptwBAAAAAAAAqDzFYrkTUI2svAQAAAAAAAAqgvISAAAAAAAAqAjKSwAAAAAAAKAiKC8BAAAAAACAiqC8BAAAAAAAACqC8hIAAAAAAACoCMpLAAAAAAAAoCLUlDsAAAAAAAAAlWdpCuWOQBWy8hIAAAAAAACoCMpLAAAAAAAAoCIoLwEAAAAAAICKoLwEAAAAAAAAKoLyEgAAAAAAAKgIyksAAAAAAACgItSUOwAAAAAAAACVp1gsdwKqkZWXAAAAAAAAQEVQXgIAAAAAAAAVQXkJAAAAAAAAVATlJQAAAAAAAFARasodAAAAgDXDHp9+uNwRaIZH7h1U7ggAAADLWHkJAAAAAAAAVATlJQAAAAAAAFARbBsLAAAAAADAcorFQrkjUIWsvAQAAAAAAAAqgvISAAAAAAAAqAjKSwAAAAAAAKAiKC8BAAAAAACAiqC8BAAAAAAAACqC8hIAAAAAAACoCDXlDgAAAAAAAEDlWVosdwKqkZWXAAAAAAAAQEVQXgL/P3t3HmbXfP8B/HNnRiabSYbsIpFIIkUiqSUoIoRYShG1pSWkYldiqZTIoqToSlUtjdiCiqVo7UKK2NISRVNLYk3wk11kssz9/dHHPJ0mYc7kXvdk7uv1PPd5nPP93u99zxwTiXe+5wAAAAAAAKSC8hIAAAAAAABIBeUlAAAAAAAAkArKSwAAAAAAACAVlJcAAAAAAABAKigvAQAAAAAAgFQoK3QAAAAAAAAA0iebLXQCipGdlwAAAAAAAEAqKC8BAAAAAACAVFBeAgAAAAAAAKmgvAQAAAAAAABSQXkJAAAAAAAApILyEgAAAAAAAEiFskIHAAAAAAAAIH2ykSl0BIqQnZcAAAAAAABAKigvAQAAAAAAgFRQXgIAAAAAAACpoLwEAAAAAAAAUkF5CQAAAAAAAKSC8hIAAAAAAABIBeUlAAAAAAAAkAplhQ4AAAAAAABA+lRnC52AYmTnJQAAAAAAAJAKyksAAAAAAAAgFZSXAAAAAAAAQCooLwEAAAAAAIBUUF4CAAAAAAAAqaC8BAAAAAAAAFKhLJeLff755zF9+vTYbbfdcrksAAAAAAAA37BsttAJKEY53Xn51ltvxYABA3K5JAAAAAAAAFAk3DYWAAAAAAAASIVEt43daKONvnJ81apV6xQGAAAAAAAAKF6Jysuqqqo46aSTolevXmscf/fdd2Ps2LE5CQYAAAAAAAAUl0TlZZ8+fWLTTTeNY445Zo3jr7zyivISAAAAAAAAqJdEz7zcf//9Y8GCBWsd32ijjeLoo49e10wAAAAAAABAEcpks9lsoUMAAACw/tvlgKcKHYF6ePr+/oWOAABASt35XHWhIxS17++YaA9ig5HXr3r//fePOXPm5PMjAAAAAAAAyINs1quQr2KV1/Jy6tSp8cUXX+TzIwAAAAAAAIAGojj3mwIAAAAAAACpo7wEAAAAAAAAUkF5CQAAAAAAAKSC8hIAAAAAAABIBeUlAAAAAAAAkApl+Vz8pz/9aWy00Ub5/AgAAAAAAADyoDqbKXQEilC9ysvPPvssNt5444iIeP/99+O6666LL774Ig488MDYdddda+aNHDkyNykBAAAAAACABi/RbWNfffXV2GyzzaJNmzbRs2fPePnll2P77bePX//613HttdfGgAED4t57781TVAAAAAAAAKAhS1RennvuudGrV6+YOnVq7L777vHd73439t9//1i4cGHMnz8/TjjhhPj5z3+er6wAAAAAAABAA5botrEvvvhiPPHEE9G7d+/YZptt4tprr42TTz45Skr+04GedtppseOOO+YlKAAAAAAAANCwJdp5OW/evGjXrl1ERDRv3jyaNWsWlZWVNeOVlZWxePHi3CYEAAAAAAAAikKi8jIiIpPJfOUxAAAAAAAAQH0kum1sRMTQoUOjvLw8IiKWLVsWJ554YjRr1iwiIqqqqnKbDgAAAAAAACgaicrLY445ptbxD37wg9XmHH300euWCAAAAFJkt51axUH7to8tNt8wWlRsEENPfynemvV5oWMBAEDeZbOFTkAxSlRe3nDDDfnKAQAAAKnUpHFJzHh9UTzx9Kdx3mlbFDoOAABAg5b4trEAAABQTB6e8klERLRrU17gJAAAAA1fovLykEMOqdO8u+++u15hAAAAAAAAgOKVqLysqKiITCazzh9aVVUVVVVVtc6Vl5dHebm/xQoAAAAAAADFKlF5OXHixJx86Pjx42Ps2LG1zo0ePTrGjBmTk/UBAACgPvbq3ybOOaVHzfHZY16NGa8vLGAiAACA4pKovDz00EPjRz/6UQwaNGiddmCOHDkyRowYUeucXZcAAAAU2tMvfBav//ulmuNPP1tewDQAAADFJ1F5OX/+/Nh///2jQ4cOceyxx8bQoUOja9euiT/ULWIBAABIoy++WBUffrGq0DEAAACKVkmSyY8//ni88847MWzYsLjllluie/fusccee8SkSZNWe4YlAAAANAQbNi+Lbl2axWabNouIiE6bNI1uXZrFRi03KHAyAADIr2zWq5CvYpXJZuv/5T/xxBMxYcKEuOeee6K8vDyOPPLIOO6442LbbbfNZUYAAADWA7sc8FShI+TFvnu2jfPP6Lna+QmTZseE294tQKLcevr+/oWOAABASk16uogbtBQ4apf6P8JxfbZO5eWXFi9eHJMmTYqf/vSnsXDhwli5cmUusgEAALAeaajlZUOnvAQAYG2Ul4VVrOVlomdersmsWbNi4sSJMXHixFi4cGEMHDgwF7kAAAAAAACAIpPomZdfWrZsWdxyyy2xxx57RPfu3eOmm26KYcOGxaxZs+Khhx7KdUYAAAAAAACgCCTaefnCCy/EhAkT4o477ohly5bFwQcfHA899FDsueeekckU59ZVAAAAAAAAIDcSlZc77rhjbLPNNnHRRRfFkCFDorKyMl+5AAAAAAAAgCKTqLx86aWX4tvf/na+sgAAAAAAAJAS1dlCJ6AYJXrm5SGHHBLHHXdc3HzzzfH+++/nKxMAAAAAAABQhBLtvBw6dGg8+eSTcdttt8Xy5cujS5cuMWDAgNhjjz1iwIAB0a5du3zlBAAAAAAAABq4ROXlmDFjIiKiqqoqnnnmmXjqqafiySefjJtvvjlWrFgRPXr0iD322COuuuqqfGQFAAAAAAAAGrBMNptd5zsWz58/P375y1/GlVdeGUuWLIlVq1blIhsAAADrkV0OeKrQEaiHp+/vX+gIAACk1C1/89DLQvrBrplCRyiIRDsvv7R8+fKYNm1aPPnkk/Hkk0/G888/H5tsskkceuih0b+/P/QAAAAAAAAAySUqL8eNG1dTVnbu3Dl22223GD58eNx6663RoUOHfGUEAAAAAAAAikDiZ1526tQpfvnLX8b3v//92HjjjfOVCwAAAAAAACgyicrLBx98MKZMmRITJ06MH//4x9GjR4/Yfffdo3///tG/f/9o3bp1vnICAAAAAADwDcpmi/OZixRWSZLJgwYNip///Ofx3HPPxf/93//FpZdeGk2bNo3LLrssOnbsGFtttVWceuqp+coKAAAAAAAANGCZbDabXZcFVq1aFS+88ELcd9998fvf/z6WLFkSq1atylU+AAAA1hO7HPBUoSNQD0/f37/QEQAASKmbpxY6QXH74W6FTlAYiW4bGxFRXV0dL730UkyZMiWefPLJeOaZZ+Lzzz+Pjh07xsEHHxwDBgzIR04AAAAAAACggUtUXu67777x7LPPxuLFi6NDhw4xYMCA+PWvfx0DBgyIrl275isjAAAAAAAAUAQSlZctW7aMX/ziF7H77rtH9+7d85UJAAAAAAAAKEIlSSYff/zx8etf/zratm272tjChQtjq622ir/97W85CwcAAAAAAAAUj0Q7L3/729/G8OHDo6KiYrWxFi1axAknnBC/+tWvYtddd81ZQAAAAAAAAL552WyhE1CMEu28/Mc//hGDBg1a6/jee+8d06dPX+dQAAAAAAAAQPFJVF5+8sknscEGG6x1vKysLD799NN1DgUAAAAAAAAUn0Tl5SabbBL//Oc/1zo+Y8aMaN++/TqHAgAAAAAAAIpPovJyv/32i1GjRsWyZctWG/viiy9i9OjR8d3vfjdn4QAAAAAAAIDiUZZk8gUXXBB333139OjRI0499dTYYostIiLiX//6V1x11VWxatWqOP/88/MSFAAAAAAAAGjYEpWXbdu2jWeffTZOOumkGDlyZGSz2YiIyGQyMWjQoLjqqquibdu2eQkKAAAAAAAANGyJysuIiM6dO8df//rXmD9/frz11luRzWaje/fuUVlZmY98AAAAAAAAQJFIXF5+qbKyMrbffvtcZgEAAAAAACAlqrOFTkAxKil0AAAAAAAAAIAI5SUAAAAAAACQEspLAAAAAAAAIBWUlwAAAAAAAEAqKC8BAAAAAACAVFBeAgAAAAAAAKlQVugAAAAAAAAApE82W+gEFCM7LwEAAAAAAIBUUF4CAAAAAAAAqaC8BAAAAAAAAFJBeQkAAAAAAACkgvISAAAAAAAASAXlJQAAAAAAAJAKyksAAAAAAAAgFcoKHQAAAAAAAID0yWYLnYBiZOclAAAAAAAAkArKSwAAAAAAACAVlJcAAAAAAABAKigvAQAAAAAAgFRQXgIAAAAAAACpoLwEAAAAAAAAUqGs0AEAAAAAAABIn+psoRNQjOy8BAAAAAAAAFJBeQkAAAAAAACkgvISAAAAAAAASAXlJQAAAAAAAJAKyksAAAAAAAAgFZSXAAAAAAAAQCooLwEAAAAAAIBUKCt0AAAAAAAAANInmy10AoqRnZcAAAAAAABAKigvAQAAAAAAgFRQXgIAAAAAAACpoLwEAAAAAAAAUkF5CQAAAAAAAKRCWaEDAAAA0DA8fX//QkcAWC/scsBThY5APfjvHAB8M5SXAAAAAAAArKa6utAJKEZuGwsAAAAAAACkgvISAAAAAAAASAXlJQAAAAAAAJAKyksAAAAAAAAgFZSXAAAAAAAAQCooLwEAAAAAAIBUUF4CAAAAAAAAqVBW6AAAAAAAAACkTzZb6AQUIzsvAQAAAAAAgFRQXgIAAAAAAACpoLwEAAAAAAAAUkF5CQAAAAAAAKSC8hIAAAAAAABIBeUlAAAAAAAAkAplhQ4AAAAAAABA+mSzhU5AMbLzEgAAAAAAAEgF5SUAAAAAAACQCspLAAAAAAAAIBWUlwAAAAAAAEAqKC8BAAAAAACAVFBeAgAAAAAAAKmgvAQAAAAAAABSoazQAQAAAAAAAEif6myhE1CM7LwEAAAAAAAAUqFe5WV1dfVaz7/33nvrFAgAAAAAAAAoTonKy0WLFsVhhx0WzZo1i7Zt28aFF14Yq1atqhn/9NNPo0uXLjkPCQAAAAAAADR8iZ55OWrUqHjllVfi5ptvjgULFsTPfvaz+Pvf/x533313NGrUKCIislk3QAYAAAAAAACSS7Tz8t57741rrrkmDj300PjRj34UL730Unz66adxwAEHRFVVVUREZDKZvAQFAAAAAAAAGrZE5eWnn34anTt3rjlu1apVPPbYY7F48eLYb7/9YunSpTkPCAAAAAAAABSHRLeN7dSpU7zxxhu1nmu54YYbxiOPPBJ77713HHzwwTkPCAAAAAAAwDfPowILrTjvdppo5+Xee+8dN9xww2rnmzdvHg8//HA0btw4Z8EAAAAAAACA4pJo5+XYsWPjo48+WuPYhhtuGI8++mj8/e9/z0kwAAAAAAAAoLgk2nlZWVkZW2211VrHN9xww+jfv3/Nca9eveL999+vfzoAAAAAAACgaCQqL5OaPXt2rFixIp8fAQAAAAAAADQQeS0vAQAAAAAAAOpKeQkAAAAAAACkgvISAAAAAAAASIWyQgcAAAAAAAAgfbLZQiegGNl5CQAAAAAAAKRCXsvLa665Jtq2bZvPjwAAAAAAAAAaiES3jR0xYsQaz7do0SJ69OgRhxxySJSXl9ecP+qoo9YtHQAAAAAAAFA0EpWX//jHP9Z4fsGCBfHWW2/FqFGj4oknnohOnTrlJBwAAAAAAABQPBKVl1OmTFnr2KJFi2LIkCFx3nnnxaRJk9Y5GAAAAAAAAFBccvbMy4qKihg1alQ888wzuVoSAAAAAAAAKCKJdl5+nVatWsW8efNyuSQAAAAAAAAFUF1d6AQUo5ztvIyIeO6552LzzTfP5ZIAAAAAAABAkUi083LGjBlrPL9w4cKYPn16XHLJJTF69OicBAMAAAAAAACKS6Lysk+fPpHJZCKbza421qpVqxgxYkScfPLJOQsHAAAAAAAAFI9E5eWsWbPWeL6ioiIqKytzEggAAAAAAAAoTonKy86dO+crBwAAAAAAAFDkSurzpjvvvDMOOeSQ2HrrrWPrrbeOQw45JCZPnpzrbAAAAAAAAEARSVReVldXx+GHHx6HH354vP7669GtW7fo1q1bvPbaa3H44YfHEUccscbnYQIAAAAAAAB8nUS3jf3tb38bjz32WNx3333x3e9+t9bYfffdF8cee2z89re/jTPOOCOXGQEAAAAAAPiG2a9GISTaeXnDDTfE5ZdfvlpxGRFx4IEHxmWXXRYTJkzIWTgAAAAAAACgeCQqL998880YOHDgWscHDhwYb7755jqHAgAAAAAAAIpPovKySZMmsWDBgrWOL1q0KBo3bryumQAAAAAAAIAilKi83GmnneLqq69e6/hVV10VO+200zqHAgAAAAAAAIpPWZLJ559/fuy+++7x2Wefxdlnnx09e/aMbDYbb7zxRvzyl7+MP//5zzFlypR8ZQUAAAAAAAAasETl5c477xx33HFHDB8+PO66665aY5WVlXHbbbfFd77znZwGBAAAAAAAAIpDovIyIuLggw+OQYMGxcMPPxxvvvlmRET06NEj9t5772jatGnOAwIAAAAAAPDNq84WOgHFKHF5mc1m48MPP4wePXrEAQccEGVliZcAAAAAAAAAWE1JksmzZs2K3r17R8+ePaN3796x+eabx0svvZSvbAAAAADAemK3nVrFr8b1ir/cunM8fX//6NalWaEjAQDroUTl5TnnnBMrV66MW265JSZPnhwdO3aM4cOH5ysbAAAAALCeaNK4JGa8viiuvvGdQkcBANZjicrLp59+Oq677ro48sgj4+CDD47JkyfHK6+8Ep9//nm+8gEAAAAA64GHp3wSE29/N156eX6howAAX2HevHkxZMiQqKioiJYtW8awYcNiyZIlX/me3XffPTKZTK3XiSeeWGvOe++9F/vvv380bdo02rRpU7MpMqlED6z85JNPonv37jXH7du3jyZNmsQnn3wSXbp0SfzhAAAAAAAAwDdnyJAhMWfOnHj00UdjxYoVceyxx8bw4cNj0qRJX/m+448/PsaNG1dz3LRp05p/XrVqVey///7Rrl27ePbZZ2POnDlx9NFHxwYbbBCXXHJJonyJystMJhNLliyJJk2a1JwrKSmJxYsXx6JFi2rOVVRUJAoBAAAAAAAA5Ncbb7wRDz30ULz44oux3XbbRUTElVdeGfvtt1/84he/iA4dOqz1vU2bNo127dqtceyRRx6J119/PR577LFo27Zt9OnTJy666KL4yU9+EmPGjIlGjRrVOWOi28Zms9no0aNHVFZW1ryWLFkSffv2jcrKymjZsmVUVlZ+7TpVVVWxaNGiWq+qqqokUQAAAACAAtmrf5t45E+71Lx6b9mi0JEAoMHJR582bdq0aNmyZU1xGRExcODAKCkpieeff/4r33vrrbdGq1atYuutt46RI0fG0qVLa63bq1evaNu2bc25QYMGxaJFi+K1115LlDHRzsspU6YkWnxtxo8fH2PHjq11bvTo0TFmzJicrA8AAAAA5M/TL3wWr//7pZrjTz9bXsA0AORLNlvoBMUtH33a3Llzo02bNrXOlZWVxUYbbRRz585d6/uOOuqo6Ny5c3To0CFmzJgRP/nJT2LmzJlx991316z738VlRNQcf9W6a5KovOzfv3+ixddm5MiRMWLEiFrnysvLc7I2AAAAAJBfX3yxKj78YlWhYwBAg5akTzvvvPPi0ksv/cr13njjjXpnGT58eM0/9+rVK9q3bx977rlnvP3227H55pvXe901SVRezpgxY43nW7RoEZ06dYpMJlOndcrLy5WVAAAAANCAbNi8LNq2Lo9WG/3n//t12qRpRETMm7885i1YUchoALBeStKnnXXWWTF06NCvnNO1a9do165dfPLJJ7XOr1y5MubNm7fW51muSb9+/SIi4q233orNN9882rVrFy+88EKtOR9//HFERKJ1IxKWl3369IlMJhPZ/9knnMlkonHjxnHGGWfEuHHjorS0NFEIAAAAAGD9tku/jeP8M3rWHI/7yZYRETFh0uyYcNu7hYoFAEWhdevW0bp166+dt9NOO8WCBQti+vTpse2220ZExBNPPBHV1dU1hWRdvPzyyxER0b59+5p1L7744vjkk09qbkv76KOPRkVFRWy55ZaJvpZE5eWsWbPWeP7LL3LUqFFRWVkZZ599dqIQAAAAAMD67cHHP44HH/+40DEAgK/wrW99K/bZZ584/vjj4w9/+EOsWLEiTj311DjiiCOiQ4cOERHx4Ycfxp577hk33XRT7LDDDvH222/HpEmTYr/99ouNN944ZsyYEWeeeWbstttu0bt374iI2HvvvWPLLbeMH/7wh3HZZZfF3Llz44ILLohTTjkl8d1YE5WXnTt3Xuv5bbbZJioqKmLs2LHKSwAAAAAAAEihW2+9NU499dTYc889o6SkJAYPHhxXXHFFzfiKFSti5syZsXTp0oiIaNSoUTz22GPxm9/8Jj7//PPYdNNNY/DgwXHBBRfUvKe0tDQeeOCBOOmkk2KnnXaKZs2axTHHHBPjxo1LnC+T/d97wK6DWbNmRa9evWLJkiW5WhIAAAAAGpRdDniq0BGoh6fv71/oCADfuF/em7MKiXo466BMoSMURKKdl19n7ty5dbqfLgAAAAAAAOmWrVZeFlZxlpcluVro008/jVGjRsWAAQNytSQAAAAAAABQRBLtvOzbt29kMqu3vAsXLowPPvggtthii7jllltyFg4AAAAAAAAoHonKy4MOOmiN5ysqKmKLLbaIQYMGRWlpaS5yAQAAAAAAAEUmUXk5evTofOUAAAAAAAAAilyiZ15OmDAhqqqq8pUFAAAAAAAAKGKJysvjjz8+Fi5cWHPcoUOHmD17dq4zAQAAAAAAAEUoUXmZzWZrHS9evDiqq6tzGggAAAAAAAAoTomeeQkAAAAAAEBxqM5+/RzItUQ7LzOZTGQymbUeAwAAAAAAANRXop2X2Ww2evToUVNYLlmyJPr27RslJbU70Hnz5uUuIQAAAAAAAFAUEpWXN9xwQ75yAAAAAAAAAEUuUXl5zDHH5CsHAAAAAAAAUOQSPfNy/vz5ceWVV8aiRYtWG1u4cOFaxwAAAAAAAAC+TqLy8ne/+11MnTo1KioqVhtr0aJF/O1vf4srr7wyZ+EAAAAAAACA4pGovLzrrrvixBNPXOv4CSecEJMnT17nUAAAAAAAABRWNutVyFexSlRevv3229G9e/e1jnfv3j3efvvtdQ4FAAAAAAAAFJ9E5WVpaWl89NFHax3/6KOPoqQk0ZIAAAAAAAAAEZGwvOzbt2/ce++9ax2/5557om/fvuuaCQAAAAAAAChCZUkmn3rqqXHEEUdEx44d46STTorS0tKIiFi1alX8/ve/j1//+tcxadKkvAQFAAAAAAAAGrZE5eXgwYPj3HPPjdNPPz3OP//86Nq1a0REvPPOO7FkyZI455xz4tBDD81LUAAAAAAAAKBhS1ReRkRcfPHF8b3vfS9uvfXWeOuttyKbzUb//v3jqKOOih122CEfGQEAAAAAAIAikLi8jIjYYYcdFJUAAAAAAABATpUkmXzZZZfFF198UXP8zDPPRFVVVc3x4sWL4+STT85dOgAAAAAAAAqiujrrVcBXsUpUXo4cOTIWL15cc7zvvvvGhx9+WHO8dOnSuOaaa3KXDgAAAAAAACgaicrLbDb7lccAAAAAAAAA9ZWovAQAAAAAAADIF+UlAAAAAAAAkAplSd9w/fXXR/PmzSMiYuXKlTFx4sRo1apVRESt52ECAAAAAAAAJJGovOzUqVNcd911Ncft2rWLm2++ebU5AAAAAAAAAEklKi9nz56dpxgAAAAAAACkSTZb6AQUo0TPvJw2bVo88MADtc7ddNNN0aVLl2jTpk0MHz48qqqqchoQAAAAAAAAKA6JysuxY8fGa6+9VnP86quvxrBhw2LgwIFx3nnnxf333x/jx4/PeUgAAAAAAACg4UtUXr7yyiux55571hzffvvt0a9fv7juuutixIgRccUVV8Sf/vSnnIcEAAAAAAAAGr5E5eX8+fOjbdu2NcdPPfVU7LvvvjXH22+/fbz//vu5SwcAAAAAAAAUjUTlZdu2bWPWrFkREbF8+fL4+9//HjvuuGPN+OLFi2ODDTbIbUIAAAAAAACgKCQqL/fbb78477zz4m9/+1uMHDkymjZtGrvuumvN+IwZM2LzzTfPeUgAAAAAAACg4StLMvmiiy6KQw45JPr37x/NmzePG2+8MRo1alQzPmHChNh7771zHhIAAAAAAABo+BKVl61atYqpU6fGwoULo3nz5lFaWlpr/M4774zmzZvnNCAAAAAAAADfvGy20AkoRonKyy+1aNFijec32mijdQoDAAAAAAAAFK9Ez7wEAAAAAAAAyBflJQAAAAAAAJAKyksAAAAAAAAgFZSXAAAAAAAAQCooLwEAAAAAAIBUUF4CAAAAAAAAqVBW6AAAAAAAAACkT3U2W+gIFCE7LwEAAAAAAIBUUF4CAAAAAAAAqaC8BAAAAAAAAFJBeQkAAAAAAACkgvISAAAAAAAASAXlJQAAAAAAAJAKZYUOAAAAAAAAQPpkqwudgGJk5yUAAAAAAACQCspLAAAAAAAAIBWUlwAAAAAAAEAqKC8BAAAAAACAVFBeAgAAAAAAAKmgvAQAAAAAAABSQXkJAAAAAAAApEJZoQMAAAAAAACQPtlsttARKELKSwAAAID10C4HPFXoCNTT0/f3L3QEAIDUcttYAAAAAAAAIBWUlwAAAAAAAEAqKC8BAAAAAACAVFBeAgAAAAAAAKmgvAQAAAAAAABSoazQAQAAAAAAAEif6upCJ6AY2XkJAAAAAAAApILyEgAAAAAAAEgF5SUAAAAAAACQCspLAAAAAAAAIBWUlwAAAAAAAEAqKC8BAAAAAACAVFBeAgAAAAAAAKlQVugAAAAAAAAApE82my10BIqQnZcAAAAAAABAKigvAQAAAAAAgFRQXgIAAAAAAACpoLwEAAAAAAAAUkF5CQAAAAAAAKSC8hIAAAAAAABIhbJCBwAAAAAAACB9qrOFTkAxsvMSAAAAAAAASAXlJQAAAAAAAJAKyksAAAAAAAAgFZSXAAAAAAAAQCrktLycP39+3HTTTblcEgAAAAAAACgSOS0v33vvvTj22GNzuSQAAAAAAABQJMqSTF60aNFXji9evHidwgAAAAAAAADFK1F52bJly8hkMmsdz2azXzkOAAAAAADA+iFbnS10BIpQovJyww03jPPPPz/69eu3xvE333wzTjjhhJwEAwAAAAAAAIpLovLy29/+dkRE9O/ff43jLVu2jGxWCw8AAAAAAAAkV5Jk8lFHHRWNGzde63i7du1i9OjR6xwKAAAAAAAAKD6Jdl4ef/zxXznetm1b5SUAAAAAAABQL4l2XibVq1eveP/99/P5EQAAAAAAAEADkdfycvbs2bFixYp8fgQAAAAAAADQQCS6bSwAAAAAAADFIZstdAKKUV53XgIAAAAAAADUlfISAAAAAAAASAXlJQAAAAAAAJAKyksAAAAAAAAgFfJaXl5zzTXRtm3bfH4EAAAAAAAA0ECUJZk8YsSINZ5v0aJF9OjRIw455JAoLy+vOX/UUUetWzoAAAAAAACgaCQqL//xj3+s8fyCBQvirbfeilGjRsUTTzwRnTp1ykk4AAAAAAAAoHgkKi+nTJmy1rFFixbFkCFD4rzzzotJkyatczAAAAAAAAAKp7o6W+gIFKGcPfOyoqIiRo0aFc8880yulgQAAAAAAACKSM7Ky4iIVq1axbx583K5JAAAAAAAAFAkclpePvfcc7H55pvnckkAAAAAAACgSCR65uWMGTPWeH7hwoUxffr0uOSSS2L06NE5CQYAAAAAAAAUl0TlZZ8+fSKTyUQ2u/oDWlu1ahUjRoyIk08+OWfhAAAAAAAAgOKRqLycNWvWGs9XVFREZWVlTgIBAAAAAAAAxSlRedm5c+d85QAAAAAAACBF1nQnTsi3kvq86c4774xDDjkktt5669h6663jkEMOicmTJ+c6GwAAAAAAAFBEEpWX1dXVcfjhh8fhhx8er7/+enTr1i26desWr732Whx++OFxxBFHaOEBAAAAAACAekl029jf/va38dhjj8V9990X3/3ud2uN3XfffXHsscfGb3/72zjjjDNymREAAAAAAAAoAol2Xt5www1x+eWXr1ZcRkQceOCBcdlll8WECRNyFg4AAAAAAAAoHonKyzfffDMGDhy41vGBAwfGm2++uc6hAAAAAAAAgOKTqLxs0qRJLFiwYK3jixYtisaNG69rJgAAAAAAAKAIJSovd9ppp7j66qvXOn7VVVfFTjvttM6hAAAAAAAAgOJTlmTy+eefH7vvvnt89tlncfbZZ0fPnj0jm83GG2+8Eb/85S/jz3/+c0yZMiVfWQEAAAAAAPiGZKsLnYBilKi83HnnneOOO+6I4cOHx1133VVrrLKyMm677bb4zne+k9OAAAAAAAAAQHFIVF5GRBx88MExaNCgePjhh+PNN9+MiIgePXrE3nvvHU2bNs15QAAAAAAAAKA4JC4vs9lsfPjhh9GjR4844IADoqws8RIAAAAAAAAAqylJMnnWrFnRu3fv6NmzZ/Tu3Ts233zzeOmll/KVDQAAAIAis9tOreJX43rFX27dOZ6+v39069Ks0JEAAPgGJSovzznnnFi5cmXccsstMXny5OjYsWMMHz48X9kAAAAAKDJNGpfEjNcXxdU3vlPoKAAAFECie74+/fTTMXny5Nhll10iImLHHXeMjh07xueffx7NmvlbcAAAAACsm4enfBIREe3alBc4CQAAhZCovPzkk0+ie/fuNcft27ePJk2axCeffBJdunTJeTgAAAAAAAAKozqbLXQEilCi8jKTycSSJUuiSZMmNedKSkpi8eLFsWjRoppzFRUVX7lOVVVVVFVV1TpXXl4e5eX+Rh0AAAAAAAAUq0TPvMxms9GjR4+orKyseS1ZsiT69u0blZWV0bJly6isrPzadcaPHx8tWrSo9Ro/fny9vwgAAAAA1j979W8Tj/xpl5pX7y1bFDoSAAAFlmjn5ZQpU3LyoSNHjowRI0bUOmfXJQAAAEBxefqFz+L1f79Uc/zpZ8sLmAYAgDRIVF72798/Jx/qFrEAAAAAfPHFqvjwi1WFjgEAQIokKi9nzJixxvMtWrSITp06RSaTyUkoAAAAAIrThs3Lom3r8mi10X/+4nunTZpGRMS8+ctj3oIVhYwGAMA3IFF52adPn8hkMpHNZmudz2Qy0bhx4zjjjDNi3LhxUVpamtOQAAAAABSHXfptHOef0bPmeNxPtoyIiAmTZseE294tVCwAAL4hicrLWbNmrfH8ggULYvr06TFq1KiorKyMs88+OyfhAAAAACguDz7+cTz4+MeFjgEAQIEkKi87d+681vPbbLNNVFRUxNixY5WXAAAAAAAA67n/vRMnfBNKcrnYtttuu9bdmQAAAAAAAABfJafl5dy5c6N169a5XBIAAAAAAAAoEjkrLz/99NMYNWpUDBgwIFdLAgAAAAAAAEUk0TMv+/btG5lMZrXzCxcujA8++CC22GKLuOWWW3IWDgAAAAAAACgeicrLgw46aI3nKyoqYosttohBgwZFaWlpLnIBAAAAAAAARSZReTl69Oh85QAAAAAAAACKXKLycsKECTFkyJAoLy/PVx4AAAAAAABSoLo6W+gIFKGSJJOPP/74WLhwYc1xhw4dYvbs2bnOBAAAAAAAABShROVlNlu7YV+8eHFUV1fnNBAAAAAAAABQnBKVlwAAAAAAAAD5kqi8zGQykclk1noMAAAAAAAAUF9lSSZns9no0aNHTWG5ZMmS6Nu3b5SU1O5A582bl7uEAAAAAAAAQFFIVF7ecMMN+coBAAAAAAAAFLlE5eUxxxyTrxwAAAAAAABAkUv0zMv58+fHlVdeGYsWLVptbOHChWsdAwAAAAAAYP2SzXoV8lWsEpWXv/vd72Lq1KlRUVGx2liLFi3ib3/7W1x55ZU5CwcAAAAAAAAUj0Tl5V133RUnnnjiWsdPOOGEmDx58jqHAgAAAAAAAIpPovLy7bffju7du691vHv37vH222+vcygAAAAAAACg+CQqL0tLS+Ojjz5a6/hHH30UJSWJlgQAAAAAAACIiITlZd++fePee+9d6/g999wTffv2XddMAAAAAAAAQBEqSzL51FNPjSOOOCI6duwYJ510UpSWlkZExKpVq+L3v/99/PrXv45JkyblJSgAAAAAAADQsCUqLwcPHhznnntunH766XH++edH165dIyLinXfeiSVLlsQ555wThx56aF6CAgAAAAAA8M3JVmcLHYEilKi8jIi4+OKL43vf+17ceuut8dZbb0U2m43+/fvHUUcdFTvssEM+MgIAAAAAAABFIHF5GRGxww47KCoBAAAAAACAnCpJMvmyyy6LL774oub4mWeeiaqqqprjxYsXx8knn5y7dAAAAAAAAEDRSFRejhw5MhYvXlxzvO+++8aHH35Yc7x06dK45pprcpcOAAAAAAAAKBqJystsNvuVxwAAAAAAAAD1lai8BAAAAAAAAMgX5SUAAAAAAACQCmVJ33D99ddH8+bNIyJi5cqVMXHixGjVqlVERK3nYQIAAAAAALD+qvb4QAogUXnZqVOnuO6662qO27VrFzfffPNqcwAAAAAAAACSSlRezp49O08xAAAAAAAAgGKX6JmX06ZNiwceeKDWuZtuuim6dOkSbdq0ieHDh0dVVVVOAwIAAAAAAADFIVF5OXbs2Hjttddqjl999dUYNmxYDBw4MM4777y4//77Y/z48TkPCQAAAAAAADR8icrLV155Jfbcc8+a49tvvz369esX1113XYwYMSKuuOKK+NOf/pTzkAAAAAAAAEDDl6i8nD9/frRt27bm+Kmnnop999235nj77beP999/P3fpAAAAAAAAgKJRlmRy27ZtY9asWbHpppvG8uXL4+9//3uMHTu2Znzx4sWxwQYb5DwkAAAAAAAA36xsdbbQEShCiXZe7rfffnHeeefF3/72txg5cmQ0bdo0dt1115rxGTNmxOabb57zkAAAAAAAAEDDl2jn5UUXXRSHHHJI9O/fP5o3bx433nhjNGrUqGZ8woQJsffee+c8JAAAAAAAANDwJSovW7VqFVOnTo2FCxdG8+bNo7S0tNb4nXfeGc2bN89pQAAAAAAAAKA4JCovv9SiRYs1nt9oo43WKQwAAAAAAABQvBI98xIAAAAAAAAgX5SXAAAAAAAAQCooLwEAAAAAAIBUqNczLwEAAAAAAGjYstXZQkegCNl5CQAAAAAAAKSC8hIAAAAAAABIBeUlAAAAAAAAkArKSwAAAAAAACAVlJcAAAAAAABAKigvAQAAAAAAgFQoK3QAAAAAAAAA0qc6W+gEFCM7LwEAAAAAAIBUUF4CAAAAAAAAqaC8BAAAAAAAAFJBeQkAAAAAAACkgvISAAAAAAAASAXlJQAAAAAAAJAKyksAAAAAAAAgFcoKHQAAAAAAAID0yVZnCx2BImTnJQAAAAAAAJAKyksAAAAAAAAgFdw2FgAAAGA99PT9/QsdAWC98JcNtih0BOpp/xUzCx0BKAA7LwEAAAAAAIBUUF4CAAAAAAAAqaC8BAAAAAAAAFLBMy8BAAAAAABYTTabLXQEipCdlwAAAAAAAEAqKC8BAAAAAACAVFBeAgAAAAAAAKmgvAQAAAAAAABSQXkJAAAAAAAApILyEgAAAAAAAEgF5SUAAAAAAACQCmWFDgAAAAAAAED6VFdnCx2BImTnJQAAAAAAAJAKyksAAAAAAAAgFZSXAAAAAAAAQCooLwEAAAAAAIBUUF4CAAAAAAAAqaC8BAAAAAAAAFKhrNABAAAAAAAASJ9sNlvoCBQhOy8BAAAAAACAVFBeAgAAAAAAAKmgvAQAAAAAAABSQXkJAAAAAAAApILyEgAAAAAAAEgF5SUAAAAAAACQCspLAAAAAAAAIBXK1uXNVVVV8cEHH0THjh2jvLw8V5kAAAAAAAAosGx1ttARKEJ13nk5ceLEmDZtWkRELFu2LIYNGxbNmjWLHj16RPPmzePEE0+MqqqqvAUFAAAAAAAAGrY6l5fjxo2LkpL/TB81alQ88cQTceedd8Zrr70WkydPjilTpsSoUaPyFhQAAAAAAABo2OpcXn700UfRvn37iIi477774uqrr46DDz44evbsGd/73vfiuuuuizvuuCNvQQEAAAAAAIB1M2/evBgyZEhUVFREy5YtY9iwYbFkyZK1zp89e3ZkMpk1vu68886aeWsav/322xPnq3N52a5du3j77bcjIuLzzz+PVq1a1Rpv3bp1fPbZZ4kDAAAAAAAAAN+MIUOGxGuvvRaPPvpoPPDAAzF16tQYPnz4WudvuummMWfOnFqvsWPHRvPmzWPfffetNfeGG26oNe+ggw5KnK/O5eWQIUPi/PPPjwULFsQPf/jDGDduXE0Lu3Tp0hgzZkx85zvfSRwAAAAAAAAAyL833ngjHnroobj++uujX79+scsuu8SVV14Zt99+e3z00UdrfE9paWm0a9eu1uuee+6Jww47LJo3b15rbsuWLWvNa9y4ceKMdS4vR48eHa1bt46uXbvG9OnT49FHH422bdtGjx49ok2bNvHcc8/FlVdemTgAAAAAAAAAkH/Tpk2Lli1bxnbbbVdzbuDAgVFSUhLPP/98ndaYPn16vPzyyzFs2LDVxk455ZRo1apV7LDDDjFhwoTIZrOJM5bVdWKjRo3iz3/+czz00ENx//33R2lpaVRXV0f79u3jO9/5Thx11FHRrFmzxAEAAAAAAACA2qqqqqKqqqrWufLy8igvL6/3mnPnzo02bdrUOldWVhYbbbRRzJ07t05r/PGPf4xvfetbsfPOO9c6P27cuNhjjz2iadOm8cgjj8TJJ58cS5YsidNPPz1RxjqXl1/aZ599Yp999kn6NgAAAAAAANYj2erku+bInfHjx8fYsWNrnRs9enSMGTNmtbnnnXdeXHrppV+53htvvLHOmb744ouYNGlSjBo1arWx/z7Xt2/f+Pzzz+Pyyy/Pf3m5atWqKC0trTl+4YUXorq6Ovr27btOTS8AAAAAAADwHyNHjowRI0bUOre2Lu6ss86KoUOHfuV6Xbt2jXbt2sUnn3xS6/zKlStj3rx50a5du6/NNHny5Fi6dGkcffTRXzu3X79+cdFFF0VVVVWiDrHO5eW7774bgwcPjpdffjn22muvuOOOO2Lw4MHx+OOPR0REly5d4sEHH4wePXrU+cMBAAAAAACA1SW5RWzr1q2jdevWXztvp512igULFsT06dNj2223jYiIJ554Iqqrq6Nfv35f+/4//vGPceCBB9bps15++eWorKxMvPmxpK4TzzrrrGjevHnce++9UVFREfvtt1+sXLky3n///fjwww+je/fu8ZOf/CTRhwMAAAAAAADfjG9961uxzz77xPHHHx8vvPBCPPPMM3HqqafGEUccER06dIiIiA8//DB69uwZL7zwQq33vvXWWzF16tT40Y9+tNq6999/f1x//fXxz3/+M9566624+uqr45JLLonTTjstccY677ycOnVqPPLII9GnT5/Yddddo7KyMqZOnRqbbLJJRERccsklsd9++yUOAAAAAAAAAHwzbr311jj11FNjzz33jJKSkhg8eHBcccUVNeMrVqyImTNnxtKlS2u9b8KECdGxY8fYe++9V1tzgw02iKuuuirOPPPMyGaz0a1bt/jVr34Vxx9/fOJ8mWw2W6enrVZUVMQrr7wSXbp0ierq6igvL4+XXnopttlmm4j4T9v67W9/OxYtWpQ4BAAAAAAA5MNfNtii0BGop/1XzCx0hKJ33NhPvn4SeTNhdJtCRyiIOt82dquttooJEyZERMSNN94YG2+8cdx+++0147fddpvnXQIAAAAAAAD1Vufbxo4ZMyYOOuiguOyyy6KkpCQefvjhOP744+OJJ56IkpKSePHFF2PSpEn5zAoAAAAAAMA3pLpuN++EnKpzeTlo0KB44403Yvr06bHtttvGZpttFlOnTo2rrroqli5dGpdcckkMGDAgn1kBAAAAAACABqzO5WVExGabbRabbbZZzXHbtm1j3Lhxa5z785//PE488cRo2bLluuQDAAAAAAAAikSdn3mZ1CWXXBLz5s3L1/IAAAAAAABAA5O38jLrPsgAAAAAAABAAnkrLwEAAAAAAACSUF4CAAAAAAAAqaC8BAAAAAAAAFKhrNABAAAAAAAASJ9sdbbQEShCedt5ueuuu0aTJk3ytTwAAAAAAADQwORt5+Vf//rXfC0NAAAAAAAANEB1Li9LSkoik8l85ZxMJhMrV65c51AAAAAAAABA8alzeXnPPfesdWzatGlxxRVXRHV1dU5CAQAAAAAAAMWnzuXl9773vdXOzZw5M84777y4//77Y8iQITFu3LichgMAAAAAAACKR0l93vTRRx/F8ccfH7169YqVK1fGyy+/HDfeeGN07tw51/kAAAAAAACAIlHnnZcREQsXLoxLLrkkrrzyyujTp088/vjjseuuu+YrGwAAAAAAAAWSzWYLHYEiVOfy8rLLLotLL7002rVrF7fddtsabyMLAAAAAAAAUF+ZbB1r85KSkmjSpEkMHDgwSktL1zrv7rvvzlk4AAAAAABYF3/ZYItCR6Ce9l8xs9ARit7Ro+YUOkJRu+mi9oWOUBB13nl59NFHRyaTyWcWAAAAAAAAoIjVubycOHFiHmMAAAAAAAAAxa6k0AEAAAAAAAAAIhLsvDzkkEPqNM8zLwEAAAAAIH822mW76HrWsGjx7a2jcYc28dLgk+Pj+x4vdCyAnKhzedmiRYt85gAAAAAAAOqgtFnTWDRjZrw/8a7YbvJVhY4DkFN1Li9vuOGGfOYAAAAAAADq4NOHp8anD08tdAyKQHV1ttARKEJ1fublO++8E9msf0kBAAAAAACA/Khzedm9e/f49NNPa44PP/zw+Pjjj/MSCgAAAAAAACg+dS4v/3fX5V//+tf4/PPP6/WhVVVVsWjRolqvqqqqeq0FAAAAAAAANAx1Li9zafz48dGiRYtar/HjxxciCgAAAAAAAJASZXWdmMlkIpPJrHauPkaOHBkjRoyoda68vLxeawEAAAAAAAANQ53Ly2w2G0OHDq0pGZctWxYnnnhiNGvWrNa8u++++2vXKi8vV1YCAAAAAEA9lDZrGs26dao5btqlY1Rs0zOWz1sYy96fU8BkAOuuzuXl0UcfXWun5Q9+8IO8BAIAAAAAANauxbZbx06P31xzvOUvfhoREe/fdHfMGDayULFogLLV2UJHoAjVubycOHFiHmMAAAAAAAB1MW/qC/GXDbYodAyAvCip68RDDz00HnroochmtewAAAAAAABA7tW5vJw/f37sv//+0alTp7jwwgvjnXfeyWcuAAAAAAAAoMjUubx8/PHH45133olhw4bFLbfcEt27d4899tgjJk2aFFVVVfnMCAAAAAAAABSBOpeXERGdO3eOMWPGxDvvvBOPPvpodOjQIY4//vho3759nHLKKTF9+vR85QQAAAAAAAAauETl5X/bY4894pZbbom5c+fG+PHj4/bbb49+/frlMhsAAAAAAABQRMrW5c2zZs2KiRMnxsSJE2PhwoUxcODAXOUCAAAAAAAAikzi8nLZsmUxefLkmDBhQkydOjU23XTTGDZsWBx77LGx6aab5iMjAAAAAAAA37BsNlvoCBShOpeXL7zwQkyYMCHuuOOOWLZsWRx88MHx0EMPxZ577hmZTCafGQEAAAAAAIAiUOfycscdd4xtttkmLrroohgyZEhUVlbmMxcAAAAAAABQZOpcXr700kvx7W9/O59ZAAAAAAAAgCJW5/KyrKwsZsyY8bXzevfuvU6BAAAAAAAAgOJU5/KyT58+kclk1vhw1i/PZzKZWLVqVU4DAgAAAAAAAMWhzuXlrFmz8pkDAAAAAAAAKHJ1Li9vvPHGOPvss6Np06b5zAMAAAAAAEAKZKurCx2BIlRS14ljx46NJUuW5DMLAAAAAAAAUMTqXF6u6VmXAAAAAAAAALlS5/IyIiKTyeQrBwAAAAAAAFDk6vzMy4iIHj16fG2BOW/evHUKBAAAAAAAABSnROXl2LFjo0WLFvnKAgAAAAAAABSxROXlEUccEW3atMlXFgAAAAAAAKCI1fmZl553CQAAAAAAAORTnXdeZrPZfOYAAAAAAAAgRaqrdUN88+q883L58uUxd+7c+OKLL1YbW7p0acyYMSOqq6tzGg4AAAAAAAAoHnUuL2+55ZY47rjjolGjRquNNWrUKI477riYNGlSTsMBAAAAAAAAxaPO5eX1118fZ599dpSWlq42VlZWFueee25ce+21OQ0HAAAAAAAAFI86l5f//ve/Y8cdd1zr+Pbbbx9vvPFGTkIBAAAAAAAAxafO5eXnn38eixYtWuv44sWLY+nSpTkJBQAAAAAAABSfOpeX3bt3j2effXat408//XR07949J6EAAAAAAACA4lNW14lHHXVUXHDBBbHzzjtH7969a4298sorceGFF8a5556b84AAAAAAAAB887LZbKEjUITqXF6eeeaZ8eCDD8a2224bAwcOjJ49e0ZExL/+9a947LHHYuedd44zzzwzb0EBAAAAAACAhq3Ot43dYIMN4pFHHomLL7445syZE9dee21cc801MWfOnLj44ovjsccei5kzZ+YzKwAAAAAAANCA1bm8jPhPgXnuuefGyy+/HJ9//nksXbo0pk6dGi1btoxddtklttlmm3zlBAAAAAAAABq4ROXlf5s6dWocc8wx0aFDh/jFL34RAwYMiOeeey6X2QAAAAAAAIAiUudnXkZEzJ07NyZOnBh//OMfY9GiRXHYYYdFVVVV3HvvvbHlllvmKyMAAAAAAABQBOq88/KAAw6ILbbYImbMmBG/+c1v4qOPPoorr7wyn9kAAAAAAACAIlLnnZcPPvhgnH766XHSSSdF9+7d85kJAAAAAAAAKEJ1Li+ffvrp+OMf/xjbbrttfOtb34of/vCHccQRR+QzGwAAAAAAAAWSrc4WOgJFqM63jd1xxx3juuuuizlz5sQJJ5wQt99+e3To0CGqq6vj0UcfjcWLF+czJwAAAAAAANDA1bm8/FKzZs3iuOOOi6effjpeffXVOOuss+LnP/95tGnTJg488MB8ZAQAAAAAAACKQOLy8r9tscUWcdlll8UHH3wQt912W64yAQAAAAAAAEVoncrLL5WWlsZBBx0U9913Xy6WAwAAAAAAAIpQTspLAAAAAAAAgHWlvAQAAAAAAABSoazQAQAAAAAAAEifbHW20BEoQnZeAgAAAAAAAKmgvAQAAAAAAABSQXkJAAAAAAAApILyEgAAAAAAAEgF5SUAAAAAAACQCspLAAAAAAAAIBWUlwAAAAAAAEAqlBU6AAAAAAAAAOlTna0udASKkJ2XAAAAAAAAQCooLwEAAAAAAIBUUF4CAAAAAAAAqaC8BAAAAAAAAFJBeQkAAAAAAACkgvISAAAAAAAASIWyQgcAAAAAAAAgfbLV2UJHoAjZeQkAAAAAAACkgvISAAAAAAAASAXlJQAAAAAAAJAKyksAAAAAAAAgFZSXAAAAAAAAQCqUFToAAAAAAEDa/WWDLQodgXraf8XMQkcAIAE7LwEAAAAAAIBUsPMSAAAAAACA1WSrs4WOQBGy8xIAAAAAAABIBeUlAAAAAAAAkArKSwAAAAAAACAVlJcAAAAAAABAKigvAQAAAAAAgFRQXgIAAAAAAACpUFboAAAAAAAAAKRPNpstdASKkJ2XAAAAAAAAQCooLwEAAAAAAIBUUF4CAAAAAAAAqaC8BAAAAAAAAFJBeQkAAAAAAACkgvISAAAAAAAASAXlJQAAAAAAAJAKZYUOAAAAAAAAQPpUV1cXOgJFyM5LAAAAAAAAIBWUlwAAAAAAAEAqKC8BAAAAAACAVFBeAgAAAAAAAKmgvAQAAAAAAABSQXkJAAAAAAAApEJZoQMAAAAAAACQPtnqbKEjUITsvAQAAAAAAABSQXkJAAAAAAAApILyEgAAAAAAAEgF5SUAAAAAAACQCspLAAAAAAAAIBWUlwAAAAAAAEAqKC8BAAAAAACAVCgrdAAAAAAAAADSJ5utLnQEipCdlwAAAAAAAEAqKC8BAAAAAACAVFBeAgAAAAAAAKmgvAQAAAAAAABSIefl5bx583K9JAAAAAAAAFAEclZePvLII3HYYYfFJptskqslAQAAAAAAgCJSti5vfvfdd2PChAlx4403xvz582PfffeNm266KVfZAAAAAAAAKJBsdbbQEShCicvL5cuXx9133x3XX399PPPMMzFw4MD44IMP4h//+Ef06tUrHxkBAAAAAACAIpDotrGnnXZadOjQIX7729/GwQcfHB988EHcf//9kclkorS0NF8ZAQAAAAAAgCKQaOfl1VdfHT/5yU/ivPPOiw033DBfmQAAAAAAAIAilGjn5c033xwvvPBCtG/fPg4//PB44IEHYtWqVfnKBgAAAAAAABSRROXlkUceGY8++mi8+uqr0bNnzzjllFOiXbt2UV1dHa+//nq+MgIAAAAAAABFIFF5+aUuXbrE2LFjY/bs2XHLLbfE4MGD4wc/+EF07NgxTj/99FxnBAAAAAAAAIpAomde/q9MJhODBg2KQYMGxbx58+Kmm26KG264IVfZAAAAAAAAgCKSyWaz2fq++f/+7/+iUaNGUVFRkctMAAAAAACp8pcNtih0BOpp/xUzCx0B1lv7Dp1R6AhF7cGJvQsdoSAS3zZ2wYIFccopp0SrVq2ibdu2UVlZGe3atYuRI0fG0qVL85ERAAAAAAAAKAKJbhs7b9682GmnneLDDz+MIUOGxLe+9a2IiHj99dfjyiuvjEcffTSefvrpmDFjRjz33HOefwkAAAAAAADUWaLycty4cdGoUaN4++23o23btquN7b333vHDH/4wHnnkkbjiiityGhQAAAAAAABo2BKVl/fee29cc801qxWXERHt2rWLyy67LPbbb78YPXp0HHPMMTkLCQAAAAAAADR8iZ55OWfOnNhqq63WOr711ltHSUlJjB49ep2DAQAAAAAAAMUlUXnZqlWrmD179lrHZ82aFW3atFnXTAAAAAAAAEARSnTb2EGDBsX5558fjz76aDRq1KjWWFVVVYwaNSr22WefnAYEAAAAAADgm1edrS50BIpQovJy3Lhxsd1220X37t3jlFNOiZ49e0Y2m4033ngjfv/730dVVVXcdNNN+coKAAAAAAAANGCJysuOHTvGtGnT4uSTT46RI0dGNpuNiIhMJhN77bVX/O53v4tOnTrlJSgAAAAAAADQsCUqLyMiunTpEg8++GDMnz8/3nzzzYiI6NatW2y00Uarzf3ggw+iQ4cOUVKS6NGaAAAAAAAAQBFKXF5+qbKyMnbYYYevnLPlllvGyy+/HF27dq3vxwAAAAAAAABFIq9bIr+8rSwAAAAAAADA13E/VwAAAAAAACAVlJcAAAAAAABAKtT7mZcAAAAAAAA0XNlqjwfkm5fXnZeZTCafywMAAAAAUE8b7bJdbHfP1bHnu3+L/VfMjLYH7lnoSACQ3/Iym9XIAwAAAACkUWmzprFoxsz45+ljCx0FAGrk7Lax2Wx2tZ2Wr7/+enTo0CFXHwEAAAAAQI58+vDU+PThqYWOAQC1JNp5WVVVFWeffXbstttucemll0ZExM9+9rNo3rx5bLjhhnHUUUfFokWLauZvuummUVpamtvEAAAAAAAAQIOUaOflyJEj44477ogjjzwybrzxxnjvvffigQceiGuuuSZKSkriwgsvjAsuuCCuuOKKr1ynqqoqqqqqap0rLy+P8vLy5F8BAAAAAAAA0CAk2nk5efLkuPHGG+MXv/hFPPDAA/GHP/whrrjiihgyZEgceeSR8fvf/z7uu+++r11n/Pjx0aJFi1qv8ePH1/uLAAAAAAAAANZ/iXZe/t///V/06NEjIiK6du0apaWl0a1bt5rx7t27x6effvq164wcOTJGjBhR65xdlwAAAAAAAOmRra4udASKUKKdl506dYpp06ZFRMSLL74YmUwmXnjhhZrx559/PjbZZJOvXae8vDwqKipqvZSXAAAAAAAAUNwS7bw88cQTY+jQoXH99dfH9OnT4xe/+EX89Kc/jX/9619RUlISV199dZx11ln5ygoAAAAAQI6UNmsazbp1qjlu2qVjVGzTM5bPWxjL3p9TwGQAFLNE5eUZZ5wRbdq0iWnTpsVxxx0XRx55ZPTq1SsuvPDCWLp0aZx55plx/vnn5ysrAAAAAAA50mLbrWOnx2+uOd7yFz+NiIj3b7o7ZgwbWahYABS5TDabzRY6BAAAAABAmv1lgy0KHYF62n/FzEJHgPXWXkOmFzpCUXv01m0LHaEgEj3zcm2y2WysWrUqF0sBAAAAAAAARSpRebly5cq44IILon///jF69OiIiLj88sujefPm0bRp0zjmmGNi+fLleQkKAAAAAAAASYwPwgAAMp5JREFUNGyJnnk5duzYuP7662PIkCExefLk+OSTT+Ivf/lLXHvttbFq1ar46U9/Gr/5zW/i3HPPzVdeAAAAAAAAoIFKVF5OmjQprr/++vjud78bJ510UmyxxRYxadKkOPzwwyMionHjxnHRRRcpLwEAAAAAANZz2epsoSNQhBLdNvajjz6KbbbZJiIiunXrFo0aNao5jojYfvvt4913381tQgAAAAAAAKAoJCovW7RoEQsWLKg5/va3vx0bbrhhzXFVVVVkMpmchQMAAAAAAACKR6Lycsstt4y///3vNcfPPPNMbLLJJjXHr776anTv3j136QAAAAAAAICikeiZl3/4wx9igw02WOv4ihUrPO8SAAAAAAAAqJc677wcMWJEbLLJJtGlS5eYOnVqrFy5crU5Rx11VBx22GE5DQgAAAAAAAAUhzqXl1deeWUsWbIkIiIGDBgQ8+bNy1soAAAAAAAAoPjU+baxm222WVxxxRWx9957RzabjWnTpkVlZeUa5+622245CwgAAAAAAAAUhzqXl5dffnmceOKJMX78+MhkMnHwwQevcV4mk4lVq1blLCAAAAAAAADfvGy2utARKEJ1Li8POuigOOigg2LJkiVRUVERM2fOjDZt2uQzGwAAAAAAAFBE6lxefql58+YxZcqU6NKlS5SVJX47AAAAAAAAwBrVq33s379/zT8vW7Ysli9fXmu8oqJi3VIBAAAAAAAARaekPm9aunRpnHrqqdGmTZto1qxZVFZW1noBAAAAAAAAJFWv8vKcc86JJ554Iq6++uooLy+P66+/PsaOHRsdOnSIm266KdcZAQAAAAAAgCJQr9vG3n///XHTTTfF7rvvHscee2zsuuuu0a1bt+jcuXPceuutMWTIkFznBAAAAAAAABq4epWX8+bNi65du0bEf55vOW/evIiI2GWXXeKkk07KXToAAAAAAAAKoro6W+gIFKF63Ta2a9euMWvWrIiI6NmzZ/zpT3+KiP/syGzZsmXOwgEAAAAAAADFo17l5bHHHhuvvPJKREScd955cdVVV0Xjxo3jzDPPjHPOOSenAQEAAAAAAIDiUK/bxp555pk1/zxw4MD417/+FdOnT49u3bpF7969cxYOAAAAAAAAKB71Ki8jIh5//PF4/PHH45NPPonq6upaYxMmTFjnYAAAAAAAAEBxqVd5OXbs2Bg3blxst9120b59+8hkMrnOBQAAAAAAABSZepWXf/jDH2LixInxwx/+MNd5AAAAAAAAgCJVUp83LV++PHbeeedcZwEAAAAAAACKWL12Xv7oRz+KSZMmxahRo3KdBwAAAAAAgBTIVlcXOgJFqF7l5bJly+Laa6+Nxx57LHr37h0bbLBBrfFf/epXOQkHAAAAAAAAFI96lZczZsyIPn36RETEP//5z1pjmUxmnUMBAAAAAAAAxade5eWUKVNynQMAAAAAAAAociWFDgAAAAAAAAAQobwEAAAAAAAAUkJ5CQAAAAAAAKRCvZ55CQAAAAAAQMOWrc4WOgJFyM5LAAAAAAAAIBWUlwAAAAAAAEAqKC8BAAAAAACAVFBeAgAAAAAAAKmgvAQAAAAAAABSQXkJAAAAAAAApILyEgAAAAAAAEiFskIHAAAAAAAAIH2y2epCR6AI2XkJAAAAAAAApILyEgAAAAAAAEgF5SUAAAAAAACQCspLAAAAAAAAIBWUlwAAAAAAAEAqKC8BAAAAAACAVCgrdAAAAAAAAADSJ1udLXQEipCdlwAAAAAAAEAqKC8BAAAAAACAVFBeAgAAAAAAAKmgvAQAAAAAAABSQXkJAAAAAAAApILyEgAAAAAAAEgF5SUAAAAAAACQCmWFDgAAAAAAAED6ZKurCx2BImTnJQAAAAAAAJAKyksAAAAAAAAgFZSXAAAAAAAAUCQuvvji2HnnnaNp06bRsmXLOr0nm83GhRdeGO3bt48mTZrEwIED480336w1Z968eTFkyJCoqKiIli1bxrBhw2LJkiWJ8ykvAQAAAAAAoEgsX748vv/978dJJ51U5/dcdtllccUVV8Qf/vCHeP7556NZs2YxaNCgWLZsWc2cIUOGxGuvvRaPPvpoPPDAAzF16tQYPnx44nyZbDabTfwuAAAAAIAi8pcNtih0BOpp/xUzCx0B1lu7HPBUoSMUtafv75/X9SdOnBhnnHFGLFiw4CvnZbPZ6NChQ5x11llx9tlnR0TEwoULo23btjFx4sQ44ogj4o033ogtt9wyXnzxxdhuu+0iIuKhhx6K/fbbLz744IPo0KFDnXPZeQkAAAAAAAApU1VVFYsWLar1qqqq+sZzzJo1K+bOnRsDBw6sOdeiRYvo169fTJs2LSIipk2bFi1btqwpLiMiBg4cGCUlJfH8888n+ryy3MRmbaqqqmL8+PExcuTIKC8vL3QcEnDt1l+u3frJdVt/uXbrL9du/eXarZ9ct/WXa7f+cu3WT67b+quhX7uGvHuvoV+7hsy1I9/yvfOPrzZmzJgYO3ZsrXOjR4+OMWPGfKM55s6dGxERbdu2rXW+bdu2NWNz586NNm3a1BovKyuLjTbaqGZOXbltbJ4tWrQoWrRoEQsXLoyKiopCxyEB12795dqtn1y39Zdrt/5y7dZfrt36yXVbf7l26y/Xbv3kuq2/XLv1l2u3/nLtoGGrqqpabadleXn5Gv+ywnnnnReXXnrpV673xhtvRM+ePWuO63rb2GeffTa+853vxEcffRTt27evOX/YYYdFJpOJO+64Iy655JK48cYbY+bM2n/Zp02bNjF27NhEz9e08xIAAAAAAABSZm1F5ZqcddZZMXTo0K+c07Vr13rlaNeuXUREfPzxx7XKy48//jj69OlTM+eTTz6p9b6VK1fGvHnzat5fV8pLAAAAAAAAWI+1bt06WrdunZe1u3TpEu3atYvHH3+8pqxctGhRPP/88zU7KnfaaadYsGBBTJ8+PbbddtuIiHjiiSeiuro6+vXrl+jzSnKaHgAAAAAAAEit9957L15++eV47733YtWqVfHyyy/Hyy+/HEuWLKmZ07Nnz7jnnnsiIiKTycQZZ5wRP/vZz+K+++6LV199NY4++ujo0KFDHHTQQRER8a1vfSv22WefOP744+OFF16IZ555Jk499dQ44ogjokOHDony2XmZZ+Xl5TF69GgPS14PuXbrL9du/eS6rb9cu/WXa7f+cu3WT67b+su1W3+5dusn12395dqtv1y79ZdrB9THhRdeGDfeeGPNcd++fSMiYsqUKbH77rtHRMTMmTNj4cKFNXPOPffc+Pzzz2P48OGxYMGC2GWXXeKhhx6Kxo0b18y59dZb49RTT40999wzSkpKYvDgwXHFFVckzpfJZrPZen5tAAAAAAAAADnjtrEAAAAAAABAKigvAQAAAAAAgFRQXgIAAAAAAACpoLwEAID1xJgxY6JPnz6FjgEAAOTQvffeG926dYvS0tI444wzCh0HoOAaTHn56aefxkknnRSdOnWK8vLyaNeuXQwaNCieeeaZiIjYbLPNIpPJrPb6+c9/Xmudu+66K3bfffdo0aJFNG/ePHr37h3jxo2LefPmRUTExIkTa95bUlIS7du3j8MPPzzee++91TK99tprcdhhh0Xr1q2jvLw8evToERdeeGEsXbq01rxMJhP33nvvau8fOnRoHHTQQTF79uw1Zv/v18SJE7/y+7Ns2bIYOnRo9OrVK8rKyuKggw6q+zc3z1y7iV/5/XnyySfje9/7XrRv3z6aNWsWffr0iVtvvTXBdzi35s6dG6eddlp07do1ysvLY9NNN40DDjggHn/88Zo5zz77bOy3335RWVkZjRs3jl69esWvfvWrWLVqVa21MplMNG7cON59991a5w866KAYOnRozfHQoUNrvl8bbLBBtG3bNvbaa6+YMGFCVFdXr7bmV12TL+2+++5xxhln1PkaPfnkk2sdnzt37td+36677rrYddddo7KyMiorK2PgwIHxwgsvfO371vR1rOnf/3vvvTcymUzN8f/mbdKkSWy11VZx7bXXrnG9L7+3Xbp0iXPPPTeWLVtWa97/fl9XrFgRRx55ZGyyySbxz3/+s2ZOJpOJ5557rtZ7q6qqYuONN45MJhNPPvlknb7O/87erFmz6N69ewwdOjSmT59ea97/fp2tW7eO/fbbL1599dU1rjto0KAoLS2NF198cbWx//5eNGrUKLp16xbjxo2LlStX1inz/1q2bFmccsopsfHGG0fz5s1j8ODB8fHHHydex89c/X7mXnvttRg8eHDNf0N+85vffO171sTP3PrzM3fttdfG7rvvHhUVFZHJZGLBggX1Wiftzj777Fo//19nwYIFccopp0T79u1rfk/z17/+tWZ86tSpccABB0SHDh3W+vNMunz44Yfxgx/8IDbeeONo0qRJ9OrVK1566aVCx/rGfdV/H4844ojYZ599as1/6KGHIpPJxJgxY2qdHzNmTHTq1CmmT5++xl9Tv7TnnnvGIYccEhGr/zfuv389y2QysfHGG8c+++wTM2bMqNPXMnv27Bg2bFh06dIlmjRpEptvvnmMHj06li9fXvdvyHqkIV27iIiLL744dt5552jatGm0bNmyzu8rlEL8+fu/X9dff31ErP0v43z5e8WXX345Ilb/vUfbtm1j8ODB8c4779R6X11+P/y/a/+3L3+v+nXmzZsXp512WmyxxRbRpEmT6NSpU5x++umxcOHCr33vunLt1u3aRRT+94vTpk2L0tLS2H///VcbW758eVx22WWxzTbbRNOmTaNVq1bxne98J2644YZYsWLF1/4ZasyYMbW+T0l/bVzTmm+99VbN+Jp+7fy6PyN9nYcffjh23HHH2HDDDaN169YxePDgmD17dq05VVVVcf7550fnzp2jvLw8Nttss5gwYUKdP2N9csIJJ8Shhx4a77//flx00UUxc+bMGDBgQLRt2zYaN24cXbt2jQsuuCBWrFhR632/+c1van5N2nTTTePMM89c7c+XX+X999+P4447Ljp06BCNGjWKzp07x49//OP47LPP1jj/tttui9LS0jjllFPW6esF+DoNprwcPHhw/OMf/4gbb7wx/v3vf8d9990Xu+++e61faMeNGxdz5syp9TrttNNqxs8///w4/PDDY/vtt48HH3ww/vnPf8Yvf/nLeOWVV+Lmm2+umVdRURFz5syJDz/8MO66666YOXNmfP/736+V57nnnot+/frF8uXL4y9/+Uv8+9//josvvjgmTpwYe+21V6I/iG666aa1Mp911lmx1VZb1Tp3+OGHf+Uaq1atiiZNmsTpp58eAwcOrPNnfxNcu6++ds8++2z07t077rrrrpgxY0Yce+yxcfTRR8cDDzxQ5xy5Mnv27Nh2223jiSeeiMsvvzxeffXVeOihh2LAgAE1v2m55557on///tGxY8eYMmVK/Otf/4of//jH8bOf/SyOOOKIyGaztdbMZDJx4YUXfu1n77PPPjFnzpyYPXt2PPjggzFgwID48Y9/HN/97nfr/T+6I5Jfo5kzZ67272KbNm2+9nOefPLJOPLII2PKlCkxbdq02HTTTWPvvfeODz/8MHHmxo0bx6WXXhrz58//2rlf5n399dfjhBNOiJNOOmm1/+n95ff2nXfeiV//+tdxzTXXxOjRo9e65tKlS+PAAw+MF198MZ5++unYeuuta8Y23XTTuOGGG2rNv+eee6J58+YJv8qIG264IebMmROvvfZaXHXVVbFkyZLo169f3HTTTWv9Oh9++OGoqqqK/ffff7Wf1ffeey+effbZOPXUU9f6h50vvxdvvvlmnHXWWTFmzJi4/PLLE2ePiDjzzDPj/vvvjzvvvDOeeuqp+Oijj2r+cFhXfubq/zO3dOnS6Nq1a/z85z+Pdu3a1TtvhJ+59eVnbunSpbHPPvvET3/603q9f33RvHnz2Hjjjes0d/ny5bHXXnvF7NmzY/LkyTFz5sy47rrrYpNNNqmZ8/nnn8c222wTV111Vb4iU0d1+T3m/Pnz/7+9+46K4mr/AP6lrbQFwYJ0DVLEgiJvLBgRS7CgvsZCEpUNWBIl1Z8tRg82CPKKGhNjIiJgBAQLeUmxoaBYDhoEkiiCCMpRMCaCCDaQfX5/eJiXdXdhF1FWfD7n8AdT7tyZZ+6de3dm7sDT0xN6eno4cOAALl68iIiICJiZmb2AHGqOpq6P3t7eOHXqlMz1Ki0tDba2tnIPdaSlpcHb2xv9+/eHm5ubwvrq6tWrSEtLw6xZs5Tmqb4+Kysrw9GjR6GrqwtfX1+V9ufSpUuQSqX4/vvvceHCBWzcuBHfffddm6zP2lrsgCdld+rUqZg3b57K67Sm1uh/N/ybPn16s/Kdn5+P0tJS7NmzBxcuXMD48eOFm1vqtoefRWlpKUpLS7F+/Xr8+eefiImJwcGDBxs9x1oKx+7ZtXZ7MSoqCh999BFOnDiB0tJSYXpNTQ18fHwQFhaGuXPn4vTp0zh79iyCgoLw9ddf48KFCzKx2LRpk1yMFi5cKLMtdevGhnVh/V+3bt2U7os6fSRFiouLMXHiRAwfPhw5OTk4dOgQ/vnnH7k+87Rp03D06FFERUUhPz8fCQkJcHZ2btY2NVl1dTVu3boFHx8fWFlZQSwWQ09PD/7+/jh8+DDy8/OxadMmREZGyvQd4+PjsXTpUgQHByMvLw9RUVFITExU+RwvKiqCh4cHLl++jISEBBQWFuK7777D0aNHMWjQIOGhhoaioqKwePFiJCQkqHWTlDHG1EZtQEVFBQGg9PR0pcvY29vTxo0blc7PzMwkALRp0yal2yAiio6OJlNTU5l5mzdvJgBUWVlJRERSqZRcXV3Jw8OD6urqZJbNyckhLS0tCgsLE6YBoOTkZLltSiQSmjhxotz04OBgcnNzU7ovTVGWbmvg2DXP2LFjKSAg4JnTUdeYMWPI2tqaqqur5eZVVFRQdXU1dejQgd566y25+SkpKQSAdu/eLUwDQAsXLiRtbW36448/hOkTJ04kiUQi/K/seB49epQAUGRkpEyaqsTEy8uLPvnkE7nllMUoLS2NAAjn07N6/PgxicViio2NVWs9iURCvr6+5OLiQosWLRKmJycnU8MqXVl+HRwcKDw8XCa9p4/tW2+9Rf369ZOZVn9cKyoqaPDgwdSnTx8qKyuTW2b58uVkYmJC9+/fF6aPGjWKVqxYQQAoLS1Npf1UFkd/f38Si8VUXl6udD/rz7Xc3FyZdVeuXElvv/025eXlkampqUwelR2LUaNG0cCBA1XKc0N37twhPT092rNnjzAtLy+PANCZM2dUTofLXMuUuaauI43hMvdylLmGWrq+bq66ujpat24dOTg4kEgkIltbW1q7di09evSIgoKCqEuXLtSuXTuys7Oj0NBQYb2KigqaNWsWdezYkcRiMXl7e1NOTo4wX522xNatW+m1116jmpoalZZXdh686p53LCMjI6lr166kpaXVZF6WLFlCQ4YMeS77+TJp6vqYn58vd819/fXXacuWLaSvr08PHjwgIqIHDx5Qu3btKDo6moie9A1MTEzo3r17MmkGBweTlZUVPX78mIjk6y9F9VlGRgYBoFu3bjVrH8PDw6lbt27NWleTteXYKepvaprW7n83pOx6VlxcTAAoOzubiBRf1+Pi4ggAXbp0Sa328NNpN6SsraqKpKQkEolEVFtb26z1VcGxa9nYtUZ7saqqioyNjenSpUvk5+dHISEhwrx169aRtrY2nT9/Xm69mpoauTpTWYyePk7NrRufpqjuVKWP1Jg9e/aQrq6uzG9vKSkppKWlJbRdDxw4QKampnT79m2V0mxNXl5e9NFHH9GiRYvIzMyMLCwsKDg4WJgfERFBvXr1IkNDQ7KxsaF58+ZRVVUVEf3vfGz4p6wf99lnn8m0BYOCgmj48OEyyyxYsIA8PT1Vyvfo0aPJxsZGrq9WVlZGhoaG9MEHH8hMLyoqIgMDA7pz5w4NGDCA4uLiVNoOY4w1R5t489LY2BjGxsb48ccf8ejRo2alERcXB2NjY8yfP1/hfGXDv9y6dQvJycnQ0dGBjo4OACAnJwcXL17EggULoK0te4jd3NwwcuRIJCQkNCufbQ3HrnkqKythbm7+QrdZXl6OgwcPIigoCEZGRnLz27dvj8OHD+P27dtyT/wBwPjx4+Hk5CR3/Dw9PeHr64ulS5eqnafhw4fDzc0N+/fvV3vd1nb//n3U1tY2K446OjoIDQ3F119/jevXr6u0DhHh4MGDKCkpwYABA5Qu9+eff+L06dMQiURy827evAkvLy8AwPHjxxW+zda/f3907doV+/btA/DkzasTJ05g5syZKuWzKZ999hmqqqpw5MgRhfMrKyuxe/duAJDZByJCdHQ0ZsyYARcXF3Tv3h179+5tcnsGBgbNGrItKysLtbW1Mm+6u7i4wM7ODmfOnFEpDS5zmoPLnOaXOU30+eefIywsDCtWrMDFixcRHx8PCwsLbN68GSkpKUhKSkJ+fj7i4uLQtWtXYb2pU6fi1q1bOHDgALKysuDu7o4RI0YofOq5KSkpKRg0aBCCgoJgYWGBXr16ITQ0VG5Iada45xnLwsJC7Nu3D/v371c4FN7TUlJS4OHhgalTp6Jz587o168fIiMjn8Neay5Vro9OTk6wsrJCWloaAKCqqgrnz5/H1KlT0bVrV+FafPr0aTx69Aje3t4AgOnTp+PRo0cy9RURITY2Fu+9957QZ2hKdXU1du3ahe7du6v8pvTTWqO9/7y9KrHTZK3Z/25JBgYGAJ68rdac9nBLq6yshImJCXR1dZ/bNjh2L7+kpCS4uLjA2dkZM2bMwI4dO4Q3S+Pi4jBy5Ej069dPbj09PT2FdaYqWqpuVKQ5faSG+vfvD21tbURHR6Ourg6VlZX44YcfMHLkSOjp6QH4X7snPDwc1tbWcHJywsKFC/HgwYNm5/t5io2NhZGRETIzMxEeHo7Vq1cL/ShtbW1s3rwZFy5cQGxsLI4dO4bFixcDAAYPHoz8/HwAT4Z1Lisrw+DBg+XSLywsxMGDB4U+Yv26WVlZwmeJioqK8Ouvv2Ls2LFN5re8vByHDh3C/PnzhbJZr0uXLpg+fToSExNl3oCOjo7GuHHjYGpqihkzZiAqKkrNo8QYY6prEzcvdXV1ERMTg9jYWLRv3x6enp5YtmyZ3HcqlixZIjT46v8yMjIAAJcvX8Zrr70mXCAbU1lZCWNjYxgZGcHCwgJpaWkyHbCCggIAQI8ePRSu36NHD2GZVx3HTn1JSUk4d+4cAgICXuh2CwsLQURwcXFRukxTx8/FxUXh8fvyyy9x8OBBIabqcHFxkfsmwvNkY2Mjcx727NmzWeksWbIEVlZWzR7GedKkSejbt2+jQ00C/8uvSCTCuHHjEBwcjKFDh8os8/PPP8PY2Fj4vsitW7ewaNEiubQ++eQT1NTU4MiRI412bAMDA4WhaWJiYjB27Fh06tRJ/Z1UoP78ezrm9fvZvn17xMfHY8KECTLnampqKu7fvw8fHx8AaLKRTURITU3FoUOHMHz4cLXzefPmTYhEIrnjZGFhodL3GgEuc/Vaqsw9Ky5zV2Wma1qZ0zRVVVX46quvEB4eDolEAgcHBwwZMgSzZ89GSUkJHB0dMWTIENjb22PIkCF45513AAAnT57E2bNnsWfPHnh4eMDR0RHr169H+/btVbr5+7SioiLs3bsXdXV1+PXXX7FixQpERERg7dq1Lb3LbdbzjmVNTQ127tyJfv36oU+fPk3mp6ioCFu3boWjoyMOHTqEefPm4eOPP0ZsbOxzOwaaRpXrIwB4e3sLw4xmZGTAyckJnTp1wtChQ4Xp6enp6NatG+zt7QEA5ubmmDRpkswQe2lpabh69WqTbe/6ut3Y2BhisRgpKSlITEyUeyBS1X38+uuv8f7776u9riZ7FWKn6Vqr/13/96xD+QNAWVkZ1q9fD2trazg7Oze7PdxS/vnnH6xZswZz5859btsAOHZtQVRUFGbMmAHgyRCtlZWVOH78OIAnsWmqbmwOderGhnWhsbGx3CeWFFG1j6RIt27dcPjwYSxbtgzt2rVD+/btcf36dSQlJQnLFBUV4eTJk/jzzz+RnJyMTZs2Ye/evUpvwLe2Pn36IDg4GI6OjvD394eHh4fwCZFPP/0U3t7e6Nq1K4YPH461a9cK+yoSiYRPo5ibm6NLly4yD4YOHjwY+vr6cHR0xBtvvIHVq1cL8959912sXr0aQ4YMgZ6eHhwcHDBs2DCVho29fPkyiKjR30ArKirw999/AwCkUiliYmKE8/jtt9/GyZMnUVxc3IyjxRhjTWszrfHJkyejtLQUKSkpGD16NNLT0+Hu7o6YmBhhmUWLFiEnJ0fmz8PDAwDUGkdfLBYjJycHv/32GyIiIuDu7o6QkBC55dRJ81XGsVNdWloaAgICEBkZ+cJ/wFfnmKh7/FxdXeHv79+sN8GISK0Pwj+rjIwMmfPw119/VTuNsLAw7N69G8nJydDX1292XtatW4fY2Fjk5eWplN/t27cjNDQUW7dulVnG29sbOTk5yMzMhEQiQUBAACZPniyXlq+vLwoKCvD99983mq8ZM2bgzJkzKCoqQkxMDAIDA5u3gwrUn1tPxzwjIwNZWVmIiYmBk5MTvvvuO5n5O3bsgJ+fn/A09DvvvINTp07hypUrMss1vKk0ZswY+Pn5YeXKlS2Wf3VwmXuiJcpcS+Ey9z9tscy1pLy8PDx69AgjRoyQm/fee+8hJycHzs7O+Pjjj3H48GFhXm5uLqqrq9GhQweZH4+Ki4vljp0qpFIpOnfujG3btqF///7w8/PDF198IRcvptzzjqW9vb1aDxtIpVK4u7sjNDQU/fr1w9y5czFnzpxXKqaqXvOGDRuGU6dOoba2Funp6Rg2bBgAwMvLS+YGWP2be/UCAwNx4sQJIU47duyAl5cXunfv3uj26uv2nJwcnD17Fj4+PhgzZgyuXbum1v7duHEDo0ePxtSpUzFnzhy11tV0bT12L4vW6H/X/50+fbrZ+baxsYGRkRGsrKxw79497Nu3T27Uhxft7t27GDduHFxdXV9I+4Vj9/LKz8/H2bNnhYecdHV14efnJzzc9zyPgap1Y8O6MCcnB5s3b1YpfVX6SIrcvHkTc+bMgUQiwblz53D8+HGIRCJMmTJFOB5SqRRaWlqIi4vD66+/jrFjx2LDhg2IjY3VyLcvn34QzdLSErdu3QLw5OHOESNGwNraGmKxGDNnzsTt27dx//79JtNNTEzE+fPnER8fj19++QXr168X5qWnpyM0NBTffvstzp8/j/379+OXX37BmjVrVM53U+dffXk9cuQI7t27J7zV2bFjR4waNUrhd1UZY6wlPL8xLVqBvr4+Ro0ahVGjRmHFihWYPXs2goOD8d577wF4Uqkq67g4OTnh5MmTqK2tbfIpNG1tbSGdHj164MqVK5g3b57wcXMnJycAT37sUDTkQ15enrAM8KRRWFlZKbfcnTt3YGpq2vSOtwEcu6YdP34c48ePx8aNG+Hv79+iaavC0dERWlpauHTpktJlGh4/RUNc5OXlwdXVVeG6q1atgpOTE3788Ue18pWXlyfzEfnnHZNu3bo903A669evR1hYGFJTU1V6w6IxQ4cOhY+PDz7//HOhrDytYX579uyJzMxMhISEYN68ecIyRkZGQrnYsWMH3NzcEBUVhVmzZsmkNXPmTEyYMAGBgYEgIixYsEDhNjt06ABfX1/MmjULDx8+xJgxY1BVVfVM+1qvvkPUMOYN99PZ2Rm3bt2Cn58fTpw4AeDJUCjJycmora2VuYlUV1eHHTt2yDzA4O3tja1bt0IkEsHKyqrZQz916dIFNTU1uHPnjsz58tdff6n81DKXuSeetcy1JC5z8vupKWVO0zw97FJD7u7uKC4uxoEDB5Camopp06Zh5MiR2Lt3L6qrq2FpaSn8ON9Qc8qBpaUl9PT0ZIYE69GjB27evImamhqFwxUzWc87luoOA2dpaSlXr/fo0UMYOvpVoMr1EXhSv9y7dw/nzp1DWlqa8Ia7l5cXAgMDUV5ejszMTLm3G0eMGAE7OzvExMRg0aJF2L9/f5MPkQCydTsAbN++HaampoiMjFT5befS0lJ4e3tj8ODB2LZtm0rrvEzacuxeNq3R/36aiYmJ0jYkALl2ZEZGBkxMTNC5c2eIxWKZ/ACqtYdNTEwAoEXarlVVVRg9ejTEYjGSk5NVepuxJXDsXs7fzqKiovD48WNYWVkJ04gI7dq1wzfffAMnJ6cm68bmUrVufLouVJUqfSRFtmzZAlNTU4SHhwvTdu3aBVtbW2RmZmLgwIGwtLSEtbW1THx79OgBIsL169fh6Oiodn6fp6fLlZaWFqRSKa5evQpfX1/MmzcPISEhMDc3x8mTJzFr1izU1NTA0NCw0XRtbW0BPHkIua6uDnPnzsX//d//QUdHBytWrMDMmTMxe/ZsAEDv3r1x7949zJ07F1988UWjowh0794dWlpayMvLw6RJk+Tm5+XloVOnTkLbNSoqCuXl5TLtY6lUit9//x2rVq1qkyMWMMZaV5uuVVxdXXHv3j2Vln333XdRXV2Nb7/9VuH8+kaYIkuXLhWeggGAvn37wsXFBRs3boRUKpVZNjc3F6mpqcLTVgDg7OyMrKwsmeXq6uqQm5src6PsVcKxk5Weno5x48Zh3bp1z304GmXMzc3h4+ODLVu2KIzNnTt38Oabb8Lc3BwRERFy81NSUnD58mWZ49eQra0tPvzwQyxbtkzlb3EdO3YMf/zxh8wbS5pcnsLDw7FmzRocPHhQePr1WYWFheGnn35S+TuKOjo6jT6hqK2tjWXLlmH58uUKl5NIJIiJicHixYtlnvZ7WmBgINLT0+Hv7/9M39F42qZNm2BiYtLocLtBQUHCsDLAk++H2NjYIDc3V+ZJ0oiICMTExMicb/UdNjs7u2e6idK/f3/o6ekJQ8QAT562LSkpwaBBg1RKg8ucZuIyJ08TypymcXR0hIGBgUwd0JCJiQn8/PwQGRmJxMRE7Nu3D+Xl5XB3d8fNmzehq6uL7t27y/x17NhR7Xx4enqisLBQpk1TUFAAS0tLvnGpIk2JZT1PT0/hm0j1CgoKhKEzXwWqXB8BwMHBAba2tkhJSUFOTo7wfShra2tYW1sjIiICNTU1cm/vaWtrIyAgALGxsYiPjxfeAlGXlpYWtLW1VX4z5MaNGxg2bBj69++P6OjoNvkDYFuNXVvwovrfDTk7O+P69ev466+/ZKafP38e+vr6sLOzk5nerVs3ODg4yNz8AqBWe9jc3BwdO3aUa7vevXsXhYWFKrdd7969izfffBMikQgpKSnPNJrOs+LYqRe71vD48WPs3LkTERERMm3j3NxcWFlZISEhAe+++y5SU1ORnZ0tt35tba3KMVakperGxqjbRwKA+/fvy13r6vsx9W1XT09PlJaWorq6WlimoKAA2trasLGxaYGcvxhZWVmQSqWIiIjAwIED4eTkhNLS0malJZVKUVtbKxyjxo5jU29UdujQAaNGjcK3334rd827efMm4uLihBvSt2/fxn//+1/s3r1b5jzOzs5GRUWFzAgkjDHWYqgN+Oeff8jb25t++OEHys3NpaKiIkpKSiILCwsKDAwkIiJ7e3tavXo1lZWVyfxVVlYK6SxevJh0dHRo0aJFdPr0abp69SqlpqbSlClTaNOmTUREFB0dTaampnJ5mDZtGo0bN074/9SpU2RoaEj//ve/KTMzk65du0ZJSUlka2tLgwcPpocPHwrLxsfHk4GBAW3ZsoUKCgooOzubAgMDydTUlG7evCm3reDgYHJzc1P7OF24cIGys7Np/PjxNGzYMMrOzqbs7Gy102lJHLumHTt2jAwNDenzzz+X2f/bt2+rlU5LuHLlCnXp0oVcXV1p7969VFBQQBcvXqSvvvqKXFxciIhoz549pKOjQ3PmzKHc3FwqLi6m7du3k5mZGU2ZMoWkUqmQHgBKTk4W/r99+zaZmpqSvr4+SSQSYbpEIqHRo0dTWVkZXb9+nbKysigkJISMjY3J19eXHj9+LCyraky8vLzok08+kdtHZTFKS0sjAJSfny93LtbU1DR57MLCwkgkEtHevXtl1q2qqmpy3YYkEglNnDhRZtrMmTNJX1+fGlbpT+f36tWrlJSURGKxmAICAhpNr7a2lqytrek///mPMO3pWO3atYt0dHQoPDxc4TJSqZT+/vtvevToERERVVRUEABKS0tTaT8BUHR0tJD3w4cP0+TJk0lHR4fi4uLk9rOiokJm/cWLF1Pv3r1JKpWSm5sbLVmyRG4bd+7cIZFIRD///LPSY/EsPvjgA7Kzs6Njx47Rb7/9RoMGDaJBgwaplQaXueaXuUePHgnXOUtLS1q4cCFlZ2fT5cuXm1y3IS5zL0+ZKysro+zsbIqMjCQAdOLECcrOzm6V6yUR0cqVK8nMzIxiY2OpsLCQzpw5Q9u3b6eIiAiKj4+nvLw8ys/Pp1mzZlGXLl2orq6OpFIpDRkyhNzc3OjQoUNUXFxMp06domXLltG5c+eISL22RElJCYnFYvrwww8pPz+ffv75Z+rcuTOtXbtWWKaqqkooKwBow4YNlJ2dTdeuXXseh+WlpAmxrHf27FnS1dWlkJAQunz5MsXFxZGhoSHt2rXrOey55lLl+khE5O/vT2KxWGYaEVFAQACJxWJycnJSmP61a9dIW1ubzMzM6IMPPpCb/3T91fC6WVZWRhcvXqT58+eTlpaWSvXw9evXqXv37jRixAi6fv26zDWvrWlrsavfZnZ2Nq1atYqMjY2FOlXddv6LoAn973q1tbXUs2dP8vb2plOnTtGVK1doz549ZGlpKdOOUNb2aEid9nBoaCh16NCBdu3aRYWFhZSZmUm+vr7UtWtXun//fpPHsLKykgYMGEC9e/emwsJCmWPUsI3c0jh2zx47otZpLyYnJ5NIJKI7d+7IzVu8eDF5eHjQw4cP6Y033iAzMzP65ptvKCcnh65cuUKJiYnk7u4u99udshgVFxcTALnl1a0bm5qvah+pMUePHiUtLS1atWoVFRQUUFZWFvn4+JC9vb0Qz6qqKrKxsaEpU6bQhQsX6Pjx4+To6EizZ89WaRsvkqL+7sSJE0kikVBOTg4BoE2bNtGVK1do586dZG1tLVM+FPXfdu3aRYmJiXTx4kXhfLCysqLp06cLywQHB5NYLKaEhAQqKiqiw4cPk4ODA02bNk2lfBcUFFDHjh3pjTfeoOPHj1NJSQkdOHCAevXqRX379hWuZRs3biRLS0uZMllv2rRpNGXKFPUOGGOMqaBN3Lx8+PAhLV26lNzd3cnU1JQMDQ3J2dmZli9fLlzw7O3tCYDc3/vvvy+TVmJiIg0dOpTEYjEZGRlRnz59aPXq1cLFRFkD4cyZMwSAMjMzhWm///47TZ48mczNzUlPT48cHBxo+fLldO/ePbn14+LiqH///iQWi8nCwoLGjh1Lubm5Cve3uTcvlR2D1sSxa5pEIlG4/15eXmql01JKS0spKCiI7O3tSSQSkbW1NU2YMEGmgXXixAny8fEhExMTEolE1LNnT1q/fr1cZ+7pH+eJnnRIAMjdSKnfb11dXerUqRONHDmSduzYQXV1dXJ5VCUmzb2RoujvzJkzTR43ZedxcHBwk+s2pKiTUFxcTCKRSOGNlIbHrVu3brRw4UKqrq5uND0ioi+//JI6deokLKsoVvHx8aSjo0NhYWFKl6nXnBsp9X/6+vrk4OBAEomEsrKyZJZT1iEuKSkhXV1dCgsLIwB09uxZhdsZM2YMTZo0iYha/kbKgwcPaP78+WRmZkaGhoY0adKkZv0IyWWueWWuvuP+rHUnl7mXp8wFBwcrjHl0dHSLbUMddXV1tHbtWrK3tyc9PT2ys7Oj0NBQ2rZtG/Xt25eMjIzIxMSERowYQefPnxfWu3v3Ln300UdkZWVFenp6ZGtrS9OnT6eSkhJhP9VpS5w+fZoGDBhA7dq1o9dee41CQkJk6gZlZa1hnfCq05RY1vvpp5+oV69e1K5dO3JxcaFt27a11K6+VFS5PkZHRxMAuR9qY2JiFPYnGnrzzTeV1meKfsRtWH7EYjH961//or1796q0L/X51LS+2vPSlmKnKI36P1WvwS+SJvS/G7px4wZJJBKys7MjAwMDcnV1pbCwMJkH1VS5AUakenv48ePHtHnzZurduzcZGhqSjY0N+fn5UXFxsUrHsLE2qqppNAfH7tljR9Q67UVfX18aO3aswnmZmZkEgHJzc+nhw4f05ZdfUu/evUlfX5/Mzc3J09OTYmJiqLa2VmY9dW9eEqlXNzY1X9U+UlMSEhKoX79+ZGRkRJ06daIJEyZQXl6ezDJ5eXk0cuRIMjAwIBsbG1qwYIHKN6tfpMZuXhIRbdiwgSwtLcnAwIB8fHxo586dMuVDUf9t9+7d5O7uTsbGxmRkZESurq4UGhpKDx48EJapra2llStXkoODA+nr65OtrS3Nnz+/yXLXUHFxMUkkErKwsCAtLS0CQG+99ZbMb6C9e/em+fPnK1w/MTGRRCIR/f333ypvkzHGVKFF9Ap8GZsxxhhjjDHGGGOMMcYYY0oFBwdjw4YNOHLkCAYOHNja2WGMvcL45iVjjDHGGGOMMcYYY4wxxhAdHY3Kykp8/PHHbfI73IyxlwPXPm3EmDFjYGxsrPAvNDS0tbPHGsGxaxuUxdDY2BgZGRmtnT2NEhoaqvRYjRkzprWzpxCXU83DZU51XObaLi4HbUtJSUmjMS0pKWntLLJmehnrYfYEx+7VExcXpzTmPXv2bO3ssUZw7DQLt+db37O0LQMCAvDpp5/yjUvGWKviNy/biBs3buDBgwcK55mbm8Pc3PwF54ipimPXNhQWFiqdZ21tDQMDgxeYG81WXl6O8vJyhfMMDAxgbW39gnPUNC6nmofLnOq4zLVdXA7alsePH+Pq1atK53ft2hW6urovLkOsxbyM9TB7gmP36qmqqsJff/2lcJ6enh7s7e1fcI6Yqjh2moXb862P25aMsZcd37xkjDHGGGOMMcYYY4wxxhhjjGkEfvebMcYYY4wxxhhjjDHGGGOMMaYR+OYlY4wxxhhjjDHGGGOMMcYYY0wj8M1LxhhjjDHGGGOMMcYYY4wxxphG4JuXjDHGGGOMMcYYY4wxxhhjjDGNwDcvGWOMMcYYY4wxxhhjjDHGGGMagW9eMsYYY4wxxhhjjDHGGGOMMcY0At+8ZIwxxhhjjDHGGGOMMcYYY4xpBL55yRhjjDHGGGOMMcYYY4wxxhjTCP8PBs9p2RMl6oAAAAAASUVORK5CYII=",
      "text/plain": [
       "<Figure size 2500x2500 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Create a heatmap of the filtered correlation matrix\n",
    "plt.figure(figsize=(25, 25))\n",
    "sns.heatmap(filtered_corr_matrix, annot=True, cmap=\"coolwarm\", center=0)\n",
    "plt.title(\"Correlation Matrix Heatmap\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Using the dictionnary we eliminate highly correlated variables."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 162,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/tmp/ipykernel_160722/847262101.py:1: SettingWithCopyWarning:\n",
      "\n",
      "\n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "\n"
     ]
    }
   ],
   "source": [
    "one_hot_df.drop(\n",
    "    columns=[\n",
    "        \"SECROUT_1\",\n",
    "        \"CONDUITE_2\",\n",
    "        \"NBRKM_DRAP_ 0\",\n",
    "        \"NBRKM_DRAP_ 1\",\n",
    "        \"CWVIB_1\",\n",
    "        \"SECFUPOU_2\",\n",
    "        \"ACTIVFIN_86\",\n",
    "        \"csei_61\",\n",
    "    ],\n",
    "    inplace=True,\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### B. Data Standardization"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### The quantiative data has a high variance which is why we apply a standardization to have a mean of 0 and a standard deviation of 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 163,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/tmp/ipykernel_160722/3478200850.py:5: SettingWithCopyWarning:\n",
      "\n",
      "\n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "\n"
     ]
    }
   ],
   "source": [
    "def standard_columns(column):\n",
    "    return (column - column.mean()) / column.std() \n",
    "\n",
    "# Standardize the quantitative variables\n",
    "one_hot_df[one_hot_df.select_dtypes([\"int\", \"float\"]).columns] = one_hot_df[\n",
    "    one_hot_df.select_dtypes([\"int\", \"float\"]).columns \n",
    "].apply(standard_columns)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### C. Model creation & fitting"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 164,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The final list of columns to be used in the model: ['TPP_1', 'SECROUT_2', 'cser_5', 'naf4_EV', 'CONDUITE_1', 'naf17_OQ', 'CJSITUA_6', 'TINA_3', 'cser_6', 'CWVIB_2', 'SECFUPOU_1', 'naf38_QA', 'HH', 'KMANNU', 'REVMENS', 'target']\n"
     ]
    }
   ],
   "source": [
    "print(f\"The final list of columns to be used in the model: {one_hot_df.columns.tolist()}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### We will use a Logistic Regression model more known as 'logi' through the statsmodels library"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 165,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Optimization terminated successfully.\n",
      "         Current function value: 0.461927\n",
      "         Iterations 7\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>Logit Regression Results</caption>\n",
       "<tr>\n",
       "  <th>Dep. Variable:</th>           <td>y</td>        <th>  No. Observations:  </th>  <td> 22306</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>                 <td>Logit</td>      <th>  Df Residuals:      </th>  <td> 22291</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Method:</th>                 <td>MLE</td>       <th>  Df Model:          </th>  <td>    14</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>            <td>Sun, 07 Jul 2024</td> <th>  Pseudo R-squ.:     </th>  <td>0.3080</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>                <td>16:51:25</td>     <th>  Log-Likelihood:    </th> <td> -10304.</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>converged:</th>             <td>True</td>       <th>  LL-Null:           </th> <td> -14889.</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Covariance Type:</th>     <td>nonrobust</td>    <th>  LLR p-value:       </th>  <td> 0.000</td> \n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "   <td></td>      <th>coef</th>     <th>std err</th>      <th>z</th>      <th>P>|z|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>x1</th>  <td>    0.7776</td> <td>    0.053</td> <td>   14.616</td> <td> 0.000</td> <td>    0.673</td> <td>    0.882</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>x2</th>  <td>   -0.8112</td> <td>    0.045</td> <td>  -18.215</td> <td> 0.000</td> <td>   -0.898</td> <td>   -0.724</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>x3</th>  <td>   -0.6498</td> <td>    0.046</td> <td>  -14.083</td> <td> 0.000</td> <td>   -0.740</td> <td>   -0.559</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>x4</th>  <td>   -0.6687</td> <td>    0.046</td> <td>  -14.599</td> <td> 0.000</td> <td>   -0.758</td> <td>   -0.579</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>x5</th>  <td>    0.0232</td> <td>    0.043</td> <td>    0.538</td> <td> 0.590</td> <td>   -0.061</td> <td>    0.108</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>x6</th>  <td>   -0.3162</td> <td>    0.044</td> <td>   -7.200</td> <td> 0.000</td> <td>   -0.402</td> <td>   -0.230</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>x7</th>  <td>    2.9232</td> <td>    0.118</td> <td>   24.679</td> <td> 0.000</td> <td>    2.691</td> <td>    3.155</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>x8</th>  <td>    0.9319</td> <td>    0.047</td> <td>   19.830</td> <td> 0.000</td> <td>    0.840</td> <td>    1.024</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>x9</th>  <td>    0.4454</td> <td>    0.052</td> <td>    8.528</td> <td> 0.000</td> <td>    0.343</td> <td>    0.548</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>x10</th> <td>   -0.8480</td> <td>    0.046</td> <td>  -18.246</td> <td> 0.000</td> <td>   -0.939</td> <td>   -0.757</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>x11</th> <td>    0.1334</td> <td>    0.041</td> <td>    3.264</td> <td> 0.001</td> <td>    0.053</td> <td>    0.213</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>x12</th> <td>   -0.7799</td> <td>    0.060</td> <td>  -12.975</td> <td> 0.000</td> <td>   -0.898</td> <td>   -0.662</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>x13</th> <td>    0.1335</td> <td>    0.021</td> <td>    6.320</td> <td> 0.000</td> <td>    0.092</td> <td>    0.175</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>x14</th> <td>    0.1506</td> <td>    0.023</td> <td>    6.659</td> <td> 0.000</td> <td>    0.106</td> <td>    0.195</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>x15</th> <td>    1.1901</td> <td>    0.053</td> <td>   22.318</td> <td> 0.000</td> <td>    1.086</td> <td>    1.295</td>\n",
       "</tr>\n",
       "</table>"
      ],
      "text/latex": [
       "\\begin{center}\n",
       "\\begin{tabular}{lclc}\n",
       "\\toprule\n",
       "\\textbf{Dep. Variable:}   &        y         & \\textbf{  No. Observations:  } &    22306    \\\\\n",
       "\\textbf{Model:}           &      Logit       & \\textbf{  Df Residuals:      } &    22291    \\\\\n",
       "\\textbf{Method:}          &       MLE        & \\textbf{  Df Model:          } &       14    \\\\\n",
       "\\textbf{Date:}            & Sun, 07 Jul 2024 & \\textbf{  Pseudo R-squ.:     } &   0.3080    \\\\\n",
       "\\textbf{Time:}            &     16:51:25     & \\textbf{  Log-Likelihood:    } &   -10304.   \\\\\n",
       "\\textbf{converged:}       &       True       & \\textbf{  LL-Null:           } &   -14889.   \\\\\n",
       "\\textbf{Covariance Type:} &    nonrobust     & \\textbf{  LLR p-value:       } &    0.000    \\\\\n",
       "\\bottomrule\n",
       "\\end{tabular}\n",
       "\\begin{tabular}{lcccccc}\n",
       "             & \\textbf{coef} & \\textbf{std err} & \\textbf{z} & \\textbf{P$> |$z$|$} & \\textbf{[0.025} & \\textbf{0.975]}  \\\\\n",
       "\\midrule\n",
       "\\textbf{x1}  &       0.7776  &        0.053     &    14.616  &         0.000        &        0.673    &        0.882     \\\\\n",
       "\\textbf{x2}  &      -0.8112  &        0.045     &   -18.215  &         0.000        &       -0.898    &       -0.724     \\\\\n",
       "\\textbf{x3}  &      -0.6498  &        0.046     &   -14.083  &         0.000        &       -0.740    &       -0.559     \\\\\n",
       "\\textbf{x4}  &      -0.6687  &        0.046     &   -14.599  &         0.000        &       -0.758    &       -0.579     \\\\\n",
       "\\textbf{x5}  &       0.0232  &        0.043     &     0.538  &         0.590        &       -0.061    &        0.108     \\\\\n",
       "\\textbf{x6}  &      -0.3162  &        0.044     &    -7.200  &         0.000        &       -0.402    &       -0.230     \\\\\n",
       "\\textbf{x7}  &       2.9232  &        0.118     &    24.679  &         0.000        &        2.691    &        3.155     \\\\\n",
       "\\textbf{x8}  &       0.9319  &        0.047     &    19.830  &         0.000        &        0.840    &        1.024     \\\\\n",
       "\\textbf{x9}  &       0.4454  &        0.052     &     8.528  &         0.000        &        0.343    &        0.548     \\\\\n",
       "\\textbf{x10} &      -0.8480  &        0.046     &   -18.246  &         0.000        &       -0.939    &       -0.757     \\\\\n",
       "\\textbf{x11} &       0.1334  &        0.041     &     3.264  &         0.001        &        0.053    &        0.213     \\\\\n",
       "\\textbf{x12} &      -0.7799  &        0.060     &   -12.975  &         0.000        &       -0.898    &       -0.662     \\\\\n",
       "\\textbf{x13} &       0.1335  &        0.021     &     6.320  &         0.000        &        0.092    &        0.175     \\\\\n",
       "\\textbf{x14} &       0.1506  &        0.023     &     6.659  &         0.000        &        0.106    &        0.195     \\\\\n",
       "\\textbf{x15} &       1.1901  &        0.053     &    22.318  &         0.000        &        1.086    &        1.295     \\\\\n",
       "\\bottomrule\n",
       "\\end{tabular}\n",
       "%\\caption{Logit Regression Results}\n",
       "\\end{center}"
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.summary.Summary'>\n",
       "\"\"\"\n",
       "                           Logit Regression Results                           \n",
       "==============================================================================\n",
       "Dep. Variable:                      y   No. Observations:                22306\n",
       "Model:                          Logit   Df Residuals:                    22291\n",
       "Method:                           MLE   Df Model:                           14\n",
       "Date:                Sun, 07 Jul 2024   Pseudo R-squ.:                  0.3080\n",
       "Time:                        16:51:25   Log-Likelihood:                -10304.\n",
       "converged:                       True   LL-Null:                       -14889.\n",
       "Covariance Type:            nonrobust   LLR p-value:                     0.000\n",
       "==============================================================================\n",
       "                 coef    std err          z      P>|z|      [0.025      0.975]\n",
       "------------------------------------------------------------------------------\n",
       "x1             0.7776      0.053     14.616      0.000       0.673       0.882\n",
       "x2            -0.8112      0.045    -18.215      0.000      -0.898      -0.724\n",
       "x3            -0.6498      0.046    -14.083      0.000      -0.740      -0.559\n",
       "x4            -0.6687      0.046    -14.599      0.000      -0.758      -0.579\n",
       "x5             0.0232      0.043      0.538      0.590      -0.061       0.108\n",
       "x6            -0.3162      0.044     -7.200      0.000      -0.402      -0.230\n",
       "x7             2.9232      0.118     24.679      0.000       2.691       3.155\n",
       "x8             0.9319      0.047     19.830      0.000       0.840       1.024\n",
       "x9             0.4454      0.052      8.528      0.000       0.343       0.548\n",
       "x10           -0.8480      0.046    -18.246      0.000      -0.939      -0.757\n",
       "x11            0.1334      0.041      3.264      0.001       0.053       0.213\n",
       "x12           -0.7799      0.060    -12.975      0.000      -0.898      -0.662\n",
       "x13            0.1335      0.021      6.320      0.000       0.092       0.175\n",
       "x14            0.1506      0.023      6.659      0.000       0.106       0.195\n",
       "x15            1.1901      0.053     22.318      0.000       1.086       1.295\n",
       "==============================================================================\n",
       "\"\"\""
      ]
     },
     "execution_count": 165,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "col_target = \"target\"\n",
    "col_features = one_hot_df.columns.tolist()[:-1] # Get all the columns except the target column\n",
    "\n",
    "X = np.array(one_hot_df[col_features]).astype(float) # Convert the features to a numpy array\n",
    "y = np.array(one_hot_df[col_target]) # Convert the target to a numpy array\n",
    "\n",
    "model = sm.Logit(y, X) # Create a logistic regression model\n",
    "result = model.fit() # Fit the model\n",
    "\n",
    "summary = result.summary() # Get the summary of the model\n",
    "summary"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### From the model summary we can see that the coefficient of the 5th variable: **'CONDUITE_1'** has a **p-value** of **0.59** meaning it is highly insignificant which is why we decided to drop it. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 166,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/tmp/ipykernel_160722/3206857883.py:1: SettingWithCopyWarning:\n",
      "\n",
      "\n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "\n"
     ]
    }
   ],
   "source": [
    "one_hot_df.drop(columns=[\"CONDUITE_1\"], inplace=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### We'll also try adding a constant to the model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 167,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Optimization terminated successfully.\n",
      "         Current function value: 0.453739\n",
      "         Iterations 7\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>Logit Regression Results</caption>\n",
       "<tr>\n",
       "  <th>Dep. Variable:</th>           <td>y</td>        <th>  No. Observations:  </th>  <td> 22306</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>                 <td>Logit</td>      <th>  Df Residuals:      </th>  <td> 22291</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Method:</th>                 <td>MLE</td>       <th>  Df Model:          </th>  <td>    14</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>            <td>Sun, 07 Jul 2024</td> <th>  Pseudo R-squ.:     </th>  <td>0.3202</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>                <td>16:51:26</td>     <th>  Log-Likelihood:    </th> <td> -10121.</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>converged:</th>             <td>True</td>       <th>  LL-Null:           </th> <td> -14889.</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Covariance Type:</th>     <td>nonrobust</td>    <th>  LLR p-value:       </th>  <td> 0.000</td> \n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "    <td></td>       <th>coef</th>     <th>std err</th>      <th>z</th>      <th>P>|z|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>const</th> <td>   -1.8818</td> <td>    0.102</td> <td>  -18.423</td> <td> 0.000</td> <td>   -2.082</td> <td>   -1.682</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>x1</th>    <td>    1.4514</td> <td>    0.069</td> <td>   21.085</td> <td> 0.000</td> <td>    1.316</td> <td>    1.586</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>x2</th>    <td>   -0.7051</td> <td>    0.039</td> <td>  -18.221</td> <td> 0.000</td> <td>   -0.781</td> <td>   -0.629</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>x3</th>    <td>   -0.5729</td> <td>    0.047</td> <td>  -12.237</td> <td> 0.000</td> <td>   -0.665</td> <td>   -0.481</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>x4</th>    <td>   -0.3539</td> <td>    0.049</td> <td>   -7.295</td> <td> 0.000</td> <td>   -0.449</td> <td>   -0.259</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>x5</th>    <td>   -0.3076</td> <td>    0.045</td> <td>   -6.903</td> <td> 0.000</td> <td>   -0.395</td> <td>   -0.220</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>x6</th>    <td>    3.0443</td> <td>    0.122</td> <td>   25.002</td> <td> 0.000</td> <td>    2.806</td> <td>    3.283</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>x7</th>    <td>    1.3918</td> <td>    0.056</td> <td>   24.936</td> <td> 0.000</td> <td>    1.282</td> <td>    1.501</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>x8</th>    <td>    0.6340</td> <td>    0.053</td> <td>   11.898</td> <td> 0.000</td> <td>    0.530</td> <td>    0.738</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>x9</th>    <td>   -0.3864</td> <td>    0.052</td> <td>   -7.411</td> <td> 0.000</td> <td>   -0.489</td> <td>   -0.284</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>x10</th>   <td>    0.3868</td> <td>    0.043</td> <td>    9.082</td> <td> 0.000</td> <td>    0.303</td> <td>    0.470</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>x11</th>   <td>   -0.7463</td> <td>    0.060</td> <td>  -12.368</td> <td> 0.000</td> <td>   -0.865</td> <td>   -0.628</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>x12</th>   <td>    0.0930</td> <td>    0.021</td> <td>    4.353</td> <td> 0.000</td> <td>    0.051</td> <td>    0.135</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>x13</th>   <td>    0.2008</td> <td>    0.023</td> <td>    8.817</td> <td> 0.000</td> <td>    0.156</td> <td>    0.245</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>x14</th>   <td>    1.1230</td> <td>    0.053</td> <td>   21.190</td> <td> 0.000</td> <td>    1.019</td> <td>    1.227</td>\n",
       "</tr>\n",
       "</table>"
      ],
      "text/latex": [
       "\\begin{center}\n",
       "\\begin{tabular}{lclc}\n",
       "\\toprule\n",
       "\\textbf{Dep. Variable:}   &        y         & \\textbf{  No. Observations:  } &    22306    \\\\\n",
       "\\textbf{Model:}           &      Logit       & \\textbf{  Df Residuals:      } &    22291    \\\\\n",
       "\\textbf{Method:}          &       MLE        & \\textbf{  Df Model:          } &       14    \\\\\n",
       "\\textbf{Date:}            & Sun, 07 Jul 2024 & \\textbf{  Pseudo R-squ.:     } &   0.3202    \\\\\n",
       "\\textbf{Time:}            &     16:51:26     & \\textbf{  Log-Likelihood:    } &   -10121.   \\\\\n",
       "\\textbf{converged:}       &       True       & \\textbf{  LL-Null:           } &   -14889.   \\\\\n",
       "\\textbf{Covariance Type:} &    nonrobust     & \\textbf{  LLR p-value:       } &    0.000    \\\\\n",
       "\\bottomrule\n",
       "\\end{tabular}\n",
       "\\begin{tabular}{lcccccc}\n",
       "               & \\textbf{coef} & \\textbf{std err} & \\textbf{z} & \\textbf{P$> |$z$|$} & \\textbf{[0.025} & \\textbf{0.975]}  \\\\\n",
       "\\midrule\n",
       "\\textbf{const} &      -1.8818  &        0.102     &   -18.423  &         0.000        &       -2.082    &       -1.682     \\\\\n",
       "\\textbf{x1}    &       1.4514  &        0.069     &    21.085  &         0.000        &        1.316    &        1.586     \\\\\n",
       "\\textbf{x2}    &      -0.7051  &        0.039     &   -18.221  &         0.000        &       -0.781    &       -0.629     \\\\\n",
       "\\textbf{x3}    &      -0.5729  &        0.047     &   -12.237  &         0.000        &       -0.665    &       -0.481     \\\\\n",
       "\\textbf{x4}    &      -0.3539  &        0.049     &    -7.295  &         0.000        &       -0.449    &       -0.259     \\\\\n",
       "\\textbf{x5}    &      -0.3076  &        0.045     &    -6.903  &         0.000        &       -0.395    &       -0.220     \\\\\n",
       "\\textbf{x6}    &       3.0443  &        0.122     &    25.002  &         0.000        &        2.806    &        3.283     \\\\\n",
       "\\textbf{x7}    &       1.3918  &        0.056     &    24.936  &         0.000        &        1.282    &        1.501     \\\\\n",
       "\\textbf{x8}    &       0.6340  &        0.053     &    11.898  &         0.000        &        0.530    &        0.738     \\\\\n",
       "\\textbf{x9}    &      -0.3864  &        0.052     &    -7.411  &         0.000        &       -0.489    &       -0.284     \\\\\n",
       "\\textbf{x10}   &       0.3868  &        0.043     &     9.082  &         0.000        &        0.303    &        0.470     \\\\\n",
       "\\textbf{x11}   &      -0.7463  &        0.060     &   -12.368  &         0.000        &       -0.865    &       -0.628     \\\\\n",
       "\\textbf{x12}   &       0.0930  &        0.021     &     4.353  &         0.000        &        0.051    &        0.135     \\\\\n",
       "\\textbf{x13}   &       0.2008  &        0.023     &     8.817  &         0.000        &        0.156    &        0.245     \\\\\n",
       "\\textbf{x14}   &       1.1230  &        0.053     &    21.190  &         0.000        &        1.019    &        1.227     \\\\\n",
       "\\bottomrule\n",
       "\\end{tabular}\n",
       "%\\caption{Logit Regression Results}\n",
       "\\end{center}"
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.summary.Summary'>\n",
       "\"\"\"\n",
       "                           Logit Regression Results                           \n",
       "==============================================================================\n",
       "Dep. Variable:                      y   No. Observations:                22306\n",
       "Model:                          Logit   Df Residuals:                    22291\n",
       "Method:                           MLE   Df Model:                           14\n",
       "Date:                Sun, 07 Jul 2024   Pseudo R-squ.:                  0.3202\n",
       "Time:                        16:51:26   Log-Likelihood:                -10121.\n",
       "converged:                       True   LL-Null:                       -14889.\n",
       "Covariance Type:            nonrobust   LLR p-value:                     0.000\n",
       "==============================================================================\n",
       "                 coef    std err          z      P>|z|      [0.025      0.975]\n",
       "------------------------------------------------------------------------------\n",
       "const         -1.8818      0.102    -18.423      0.000      -2.082      -1.682\n",
       "x1             1.4514      0.069     21.085      0.000       1.316       1.586\n",
       "x2            -0.7051      0.039    -18.221      0.000      -0.781      -0.629\n",
       "x3            -0.5729      0.047    -12.237      0.000      -0.665      -0.481\n",
       "x4            -0.3539      0.049     -7.295      0.000      -0.449      -0.259\n",
       "x5            -0.3076      0.045     -6.903      0.000      -0.395      -0.220\n",
       "x6             3.0443      0.122     25.002      0.000       2.806       3.283\n",
       "x7             1.3918      0.056     24.936      0.000       1.282       1.501\n",
       "x8             0.6340      0.053     11.898      0.000       0.530       0.738\n",
       "x9            -0.3864      0.052     -7.411      0.000      -0.489      -0.284\n",
       "x10            0.3868      0.043      9.082      0.000       0.303       0.470\n",
       "x11           -0.7463      0.060    -12.368      0.000      -0.865      -0.628\n",
       "x12            0.0930      0.021      4.353      0.000       0.051       0.135\n",
       "x13            0.2008      0.023      8.817      0.000       0.156       0.245\n",
       "x14            1.1230      0.053     21.190      0.000       1.019       1.227\n",
       "==============================================================================\n",
       "\"\"\""
      ]
     },
     "execution_count": 167,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "col_target = \"target\"\n",
    "col_features = one_hot_df.columns.tolist()[:-1]\n",
    "\n",
    "X = np.array(one_hot_df[col_features]).astype(float)\n",
    "y = np.array(one_hot_df[col_target])\n",
    "X = sm.add_constant(X)\n",
    "model = sm.Logit(y, X)\n",
    "result = model.fit()\n",
    "\n",
    "summary = result.summary()\n",
    "summary"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### **Results Analysis**: \n",
    "- The logistic regression model has a **Pseudo R-squared of 0.32** meaning it explains through the 14 provided variables **32%** of the variance in our target variable.\n",
    "- The **LLR p-value** and the **p-values** of the coeficients are all < 0.05 meaning that our model is **highly statistically significant**.\n",
    "- We can see that **Log-Likelihood > LL-Null** suggesting that the chosen variables **significantly improve the model's ability to predict the outcome variable**."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### D. Model Optimization"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### We aim to streamline our model by reducing the number of variables while maintaining high performance. Our approach involves systematically testing various combinations of selected variables to strike an optimal balance between model complexity and predictive accuracy. This process will help identify the most influential predictors while ensuring the model remains robust and effective in its predictions."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 168,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Processing combinations:   0%|          | 0/16383 [00:00<?, ?it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Processing combinations: 100%|██████████| 16383/16383 [11:17<00:00, 24.17it/s]\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Features</th>\n",
       "      <th>Pseudo R-squared</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>(TPP_1,)</td>\n",
       "      <td>0.084428</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>(SECROUT_2,)</td>\n",
       "      <td>0.059901</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>(cser_5,)</td>\n",
       "      <td>0.070238</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>(naf4_EV,)</td>\n",
       "      <td>0.056070</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>(naf17_OQ,)</td>\n",
       "      <td>0.048079</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16378</th>\n",
       "      <td>(TPP_1, SECROUT_2, cser_5, naf17_OQ, CJSITUA_6...</td>\n",
       "      <td>0.318455</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16379</th>\n",
       "      <td>(TPP_1, SECROUT_2, naf4_EV, naf17_OQ, CJSITUA_...</td>\n",
       "      <td>0.315089</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16380</th>\n",
       "      <td>(TPP_1, cser_5, naf4_EV, naf17_OQ, CJSITUA_6, ...</td>\n",
       "      <td>0.309135</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16381</th>\n",
       "      <td>(SECROUT_2, cser_5, naf4_EV, naf17_OQ, CJSITUA...</td>\n",
       "      <td>0.302536</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16382</th>\n",
       "      <td>(TPP_1, SECROUT_2, cser_5, naf4_EV, naf17_OQ, ...</td>\n",
       "      <td>0.320244</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>16383 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                Features  Pseudo R-squared\n",
       "0                                               (TPP_1,)          0.084428\n",
       "1                                           (SECROUT_2,)          0.059901\n",
       "2                                              (cser_5,)          0.070238\n",
       "3                                             (naf4_EV,)          0.056070\n",
       "4                                            (naf17_OQ,)          0.048079\n",
       "...                                                  ...               ...\n",
       "16378  (TPP_1, SECROUT_2, cser_5, naf17_OQ, CJSITUA_6...          0.318455\n",
       "16379  (TPP_1, SECROUT_2, naf4_EV, naf17_OQ, CJSITUA_...          0.315089\n",
       "16380  (TPP_1, cser_5, naf4_EV, naf17_OQ, CJSITUA_6, ...          0.309135\n",
       "16381  (SECROUT_2, cser_5, naf4_EV, naf17_OQ, CJSITUA...          0.302536\n",
       "16382  (TPP_1, SECROUT_2, cser_5, naf4_EV, naf17_OQ, ...          0.320244\n",
       "\n",
       "[16383 rows x 2 columns]"
      ]
     },
     "execution_count": 168,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import itertools\n",
    "from joblib import Parallel, delayed\n",
    "from tqdm import tqdm\n",
    "\n",
    "col_target = \"target\"\n",
    "col_features = one_hot_df.columns.tolist()[:-1]\n",
    "y_np = np.array(one_hot_df[col_target]) # Convert the target to a numpy array\n",
    "\n",
    "\n",
    "def fit_model(subset_features):\n",
    "    \"\"\"\n",
    "    Fits a logistic regression model and returns pseudo R-squared.\n",
    "\n",
    "    Args:\n",
    "    - features (list): Feature names.\n",
    "\n",
    "    Returns:\n",
    "    - tuple: (features, pseudo R-squared).\n",
    "    \"\"\"\n",
    "\n",
    "    X_subset = np.array(one_hot_df[list(subset_features)]).astype(float) # Convert the features to a numpy array\n",
    "    X_subset = sm.add_constant(X_subset)\n",
    "    model = sm.Logit(y_np, X_subset)\n",
    "    result = model.fit(disp=False)\n",
    "    return subset_features, result.prsquared\n",
    "\n",
    "\n",
    "# Generate all possible combinations of col_features\n",
    "all_combinations = []\n",
    "for i in range(1, len(col_features) + 1):\n",
    "    for subset_features in itertools.combinations(col_features, i):\n",
    "        all_combinations.append(subset_features) # Append the subset to the list\n",
    "\n",
    "# Parallel processing with tqdm progress bar\n",
    "num_cores = 4 # Number of cores to use for parallel processing\n",
    "results = Parallel(n_jobs=num_cores)(\n",
    "    delayed(fit_model)(subset)\n",
    "    for subset in tqdm(all_combinations, desc=\"Processing combinations\")\n",
    ")\n",
    "\n",
    "# Extract results\n",
    "combinations, r_squared_values = zip(*results)\n",
    "\n",
    "# Create a DataFrame to display the results\n",
    "results_df = pd.DataFrame(\n",
    "    {\"Features\": combinations, \"Pseudo R-squared\": r_squared_values}\n",
    ")\n",
    "\n",
    "results_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 169,
   "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>Features</th>\n",
       "      <th>Pseudo R-squared</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>16382</th>\n",
       "      <td>(TPP_1, SECROUT_2, cser_5, naf4_EV, naf17_OQ, ...</td>\n",
       "      <td>0.320244</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16370</th>\n",
       "      <td>(TPP_1, SECROUT_2, cser_5, naf4_EV, naf17_OQ, ...</td>\n",
       "      <td>0.319610</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16377</th>\n",
       "      <td>(TPP_1, SECROUT_2, cser_5, naf4_EV, CJSITUA_6,...</td>\n",
       "      <td>0.318637</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16378</th>\n",
       "      <td>(TPP_1, SECROUT_2, cser_5, naf17_OQ, CJSITUA_6...</td>\n",
       "      <td>0.318455</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16373</th>\n",
       "      <td>(TPP_1, SECROUT_2, cser_5, naf4_EV, naf17_OQ, ...</td>\n",
       "      <td>0.318402</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16324</th>\n",
       "      <td>(TPP_1, SECROUT_2, cser_5, naf17_OQ, CJSITUA_6...</td>\n",
       "      <td>0.317755</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16315</th>\n",
       "      <td>(TPP_1, SECROUT_2, cser_5, naf4_EV, CJSITUA_6,...</td>\n",
       "      <td>0.317705</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16289</th>\n",
       "      <td>(TPP_1, SECROUT_2, cser_5, naf4_EV, naf17_OQ, ...</td>\n",
       "      <td>0.317676</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16372</th>\n",
       "      <td>(TPP_1, SECROUT_2, cser_5, naf4_EV, naf17_OQ, ...</td>\n",
       "      <td>0.317498</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16369</th>\n",
       "      <td>(TPP_1, SECROUT_2, cser_5, naf4_EV, naf17_OQ, ...</td>\n",
       "      <td>0.317233</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                Features  Pseudo R-squared\n",
       "16382  (TPP_1, SECROUT_2, cser_5, naf4_EV, naf17_OQ, ...          0.320244\n",
       "16370  (TPP_1, SECROUT_2, cser_5, naf4_EV, naf17_OQ, ...          0.319610\n",
       "16377  (TPP_1, SECROUT_2, cser_5, naf4_EV, CJSITUA_6,...          0.318637\n",
       "16378  (TPP_1, SECROUT_2, cser_5, naf17_OQ, CJSITUA_6...          0.318455\n",
       "16373  (TPP_1, SECROUT_2, cser_5, naf4_EV, naf17_OQ, ...          0.318402\n",
       "16324  (TPP_1, SECROUT_2, cser_5, naf17_OQ, CJSITUA_6...          0.317755\n",
       "16315  (TPP_1, SECROUT_2, cser_5, naf4_EV, CJSITUA_6,...          0.317705\n",
       "16289  (TPP_1, SECROUT_2, cser_5, naf4_EV, naf17_OQ, ...          0.317676\n",
       "16372  (TPP_1, SECROUT_2, cser_5, naf4_EV, naf17_OQ, ...          0.317498\n",
       "16369  (TPP_1, SECROUT_2, cser_5, naf4_EV, naf17_OQ, ...          0.317233"
      ]
     },
     "execution_count": 169,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "results_df = results_df.sort_values(by=\"Pseudo R-squared\", ascending=False)\n",
    "results_df.head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 170,
   "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>Features</th>\n",
       "      <th>Pseudo R-squared</th>\n",
       "      <th>Num Features</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>16382</th>\n",
       "      <td>(TPP_1, SECROUT_2, cser_5, naf4_EV, naf17_OQ, ...</td>\n",
       "      <td>0.320244</td>\n",
       "      <td>14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16370</th>\n",
       "      <td>(TPP_1, SECROUT_2, cser_5, naf4_EV, naf17_OQ, ...</td>\n",
       "      <td>0.319610</td>\n",
       "      <td>13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16377</th>\n",
       "      <td>(TPP_1, SECROUT_2, cser_5, naf4_EV, CJSITUA_6,...</td>\n",
       "      <td>0.318637</td>\n",
       "      <td>13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16378</th>\n",
       "      <td>(TPP_1, SECROUT_2, cser_5, naf17_OQ, CJSITUA_6...</td>\n",
       "      <td>0.318455</td>\n",
       "      <td>13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16373</th>\n",
       "      <td>(TPP_1, SECROUT_2, cser_5, naf4_EV, naf17_OQ, ...</td>\n",
       "      <td>0.318402</td>\n",
       "      <td>13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>(CWVIB_2,)</td>\n",
       "      <td>0.044554</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>(cser_6,)</td>\n",
       "      <td>0.042976</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>(SECFUPOU_1,)</td>\n",
       "      <td>0.040610</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>(KMANNU,)</td>\n",
       "      <td>0.038618</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>(naf38_QA,)</td>\n",
       "      <td>0.033592</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>16383 rows × 3 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                Features  Pseudo R-squared  \\\n",
       "16382  (TPP_1, SECROUT_2, cser_5, naf4_EV, naf17_OQ, ...          0.320244   \n",
       "16370  (TPP_1, SECROUT_2, cser_5, naf4_EV, naf17_OQ, ...          0.319610   \n",
       "16377  (TPP_1, SECROUT_2, cser_5, naf4_EV, CJSITUA_6,...          0.318637   \n",
       "16378  (TPP_1, SECROUT_2, cser_5, naf17_OQ, CJSITUA_6...          0.318455   \n",
       "16373  (TPP_1, SECROUT_2, cser_5, naf4_EV, naf17_OQ, ...          0.318402   \n",
       "...                                                  ...               ...   \n",
       "8                                             (CWVIB_2,)          0.044554   \n",
       "7                                              (cser_6,)          0.042976   \n",
       "9                                          (SECFUPOU_1,)          0.040610   \n",
       "12                                             (KMANNU,)          0.038618   \n",
       "10                                           (naf38_QA,)          0.033592   \n",
       "\n",
       "       Num Features  \n",
       "16382            14  \n",
       "16370            13  \n",
       "16377            13  \n",
       "16378            13  \n",
       "16373            13  \n",
       "...             ...  \n",
       "8                 1  \n",
       "7                 1  \n",
       "9                 1  \n",
       "12                1  \n",
       "10                1  \n",
       "\n",
       "[16383 rows x 3 columns]"
      ]
     },
     "execution_count": 170,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Add a Number of Features Variable\n",
    "results_df[\"Num Features\"] = results_df[\"Features\"].apply(len)\n",
    "results_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 171,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot the results\n",
    "plt.figure(figsize=(10, 6))\n",
    "plt.scatter(\n",
    "    results_df[\"Num Features\"],\n",
    "    results_df[\"Pseudo R-squared\"],\n",
    "    alpha=0.6, # Set the transparency of the points for visibility due to the overlapping points\n",
    "    edgecolors=\"w\",\n",
    "    linewidth=0.5,\n",
    ")\n",
    "plt.xlabel(\"Number of Features\")\n",
    "plt.ylabel(\"Pseudo R-squared\")\n",
    "plt.title(\"Pseudo R-squared vs. Number of Features\")\n",
    "plt.grid(True)\n",
    "\n",
    "# Highlight the best model\n",
    "best_model_idx = results_df[\"Pseudo R-squared\"].idxmax()\n",
    "best_model_r2 = results_df.loc[best_model_idx, \"Pseudo R-squared\"]\n",
    "best_model_features = results_df.loc[best_model_idx, \"Num Features\"]\n",
    "plt.scatter(\n",
    "    best_model_features,\n",
    "    best_model_r2,\n",
    "    color=\"red\",\n",
    "    s=100,\n",
    "    edgecolors=\"k\",\n",
    "    label=f\"Best Model\\n(R²={best_model_r2:.3f}, Features={best_model_features})\",\n",
    ")\n",
    "\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Using the above plot we can see that even with **less features**, there are combinations that offer a similarly high R²"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 172,
   "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>Features</th>\n",
       "      <th>Pseudo R-squared</th>\n",
       "      <th>Num Features</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>(TPP_1,)</td>\n",
       "      <td>0.084428</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>(TPP_1, CJSITUA_6)</td>\n",
       "      <td>0.140544</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>119</th>\n",
       "      <td>(TPP_1, cser_5, CJSITUA_6)</td>\n",
       "      <td>0.190021</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>471</th>\n",
       "      <td>(TPP_1, SECROUT_2, cser_5, CJSITUA_6)</td>\n",
       "      <td>0.227992</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1489</th>\n",
       "      <td>(TPP_1, SECROUT_2, cser_5, CJSITUA_6, TINA_3)</td>\n",
       "      <td>0.258208</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3517</th>\n",
       "      <td>(TPP_1, SECROUT_2, cser_5, naf17_OQ, CJSITUA_6...</td>\n",
       "      <td>0.276181</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7020</th>\n",
       "      <td>(TPP_1, SECROUT_2, naf17_OQ, CJSITUA_6, TINA_3...</td>\n",
       "      <td>0.294518</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10122</th>\n",
       "      <td>(TPP_1, SECROUT_2, cser_5, naf17_OQ, CJSITUA_6...</td>\n",
       "      <td>0.300685</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13170</th>\n",
       "      <td>(TPP_1, SECROUT_2, cser_5, naf17_OQ, CJSITUA_6...</td>\n",
       "      <td>0.306971</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15134</th>\n",
       "      <td>(TPP_1, SECROUT_2, cser_5, naf17_OQ, CJSITUA_6...</td>\n",
       "      <td>0.311476</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16009</th>\n",
       "      <td>(TPP_1, SECROUT_2, cser_5, naf4_EV, CJSITUA_6,...</td>\n",
       "      <td>0.315782</td>\n",
       "      <td>11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16324</th>\n",
       "      <td>(TPP_1, SECROUT_2, cser_5, naf17_OQ, CJSITUA_6...</td>\n",
       "      <td>0.317755</td>\n",
       "      <td>12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16370</th>\n",
       "      <td>(TPP_1, SECROUT_2, cser_5, naf4_EV, naf17_OQ, ...</td>\n",
       "      <td>0.319610</td>\n",
       "      <td>13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16382</th>\n",
       "      <td>(TPP_1, SECROUT_2, cser_5, naf4_EV, naf17_OQ, ...</td>\n",
       "      <td>0.320244</td>\n",
       "      <td>14</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                Features  Pseudo R-squared  \\\n",
       "0                                               (TPP_1,)          0.084428   \n",
       "18                                    (TPP_1, CJSITUA_6)          0.140544   \n",
       "119                           (TPP_1, cser_5, CJSITUA_6)          0.190021   \n",
       "471                (TPP_1, SECROUT_2, cser_5, CJSITUA_6)          0.227992   \n",
       "1489       (TPP_1, SECROUT_2, cser_5, CJSITUA_6, TINA_3)          0.258208   \n",
       "3517   (TPP_1, SECROUT_2, cser_5, naf17_OQ, CJSITUA_6...          0.276181   \n",
       "7020   (TPP_1, SECROUT_2, naf17_OQ, CJSITUA_6, TINA_3...          0.294518   \n",
       "10122  (TPP_1, SECROUT_2, cser_5, naf17_OQ, CJSITUA_6...          0.300685   \n",
       "13170  (TPP_1, SECROUT_2, cser_5, naf17_OQ, CJSITUA_6...          0.306971   \n",
       "15134  (TPP_1, SECROUT_2, cser_5, naf17_OQ, CJSITUA_6...          0.311476   \n",
       "16009  (TPP_1, SECROUT_2, cser_5, naf4_EV, CJSITUA_6,...          0.315782   \n",
       "16324  (TPP_1, SECROUT_2, cser_5, naf17_OQ, CJSITUA_6...          0.317755   \n",
       "16370  (TPP_1, SECROUT_2, cser_5, naf4_EV, naf17_OQ, ...          0.319610   \n",
       "16382  (TPP_1, SECROUT_2, cser_5, naf4_EV, naf17_OQ, ...          0.320244   \n",
       "\n",
       "       Num Features  \n",
       "0                 1  \n",
       "18                2  \n",
       "119               3  \n",
       "471               4  \n",
       "1489              5  \n",
       "3517              6  \n",
       "7020              7  \n",
       "10122             8  \n",
       "13170             9  \n",
       "15134            10  \n",
       "16009            11  \n",
       "16324            12  \n",
       "16370            13  \n",
       "16382            14  "
      ]
     },
     "execution_count": 172,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Extract the best models for each number of features\n",
    "best_models = results_df.loc[\n",
    "    results_df.groupby(\"Num Features\")[\"Pseudo R-squared\"].idxmax()\n",
    "]\n",
    "best_models"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 174,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plotting the results\n",
    "plt.figure(figsize=(10, 6))\n",
    "plt.plot(\n",
    "    best_models[\"Num Features\"],\n",
    "    best_models[\"Pseudo R-squared\"],\n",
    "    marker=\"o\",\n",
    "    linestyle=\"-\",\n",
    "    color=\"b\",\n",
    ")\n",
    "plt.title(\"Number of Variables vs. Pseudo R-squared\")\n",
    "plt.xlabel(\"Number of Variables\")\n",
    "plt.ylabel(\"Pseudo R-squared\")\n",
    "plt.grid(True)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### From the following graph we can choose a number of variables that also offers a performant model. We chose the **7** variables listed below"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 175,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['TPP_1', 'SECROUT_2', 'naf17_OQ', 'CJSITUA_6', 'TINA_3', 'cser_6', 'REVMENS']"
      ]
     },
     "execution_count": 175,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "best_features = list(best_models.iloc[6][\"Features\"])\n",
    "best_features"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### We retrain a model with the final chosen combination of variables"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 176,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Optimization terminated successfully.\n",
      "         Current function value: 0.470935\n",
      "         Iterations 7\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>Logit Regression Results</caption>\n",
       "<tr>\n",
       "  <th>Dep. Variable:</th>        <td>target</td>      <th>  No. Observations:  </th>  <td> 17844</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>                 <td>Logit</td>      <th>  Df Residuals:      </th>  <td> 17836</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Method:</th>                 <td>MLE</td>       <th>  Df Model:          </th>  <td>     7</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>            <td>Sun, 07 Jul 2024</td> <th>  Pseudo R-squ.:     </th>  <td>0.2945</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>                <td>17:03:35</td>     <th>  Log-Likelihood:    </th> <td> -8403.4</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>converged:</th>             <td>True</td>       <th>  LL-Null:           </th> <td> -11910.</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Covariance Type:</th>     <td>nonrobust</td>    <th>  LLR p-value:       </th>  <td> 0.000</td> \n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "      <td></td>         <th>coef</th>     <th>std err</th>      <th>z</th>      <th>P>|z|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>const</th>     <td>   -2.4526</td> <td>    0.091</td> <td>  -26.926</td> <td> 0.000</td> <td>   -2.631</td> <td>   -2.274</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>TPP_1</th>     <td>    1.6284</td> <td>    0.071</td> <td>   23.075</td> <td> 0.000</td> <td>    1.490</td> <td>    1.767</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>SECROUT_2</th> <td>   -1.0821</td> <td>    0.039</td> <td>  -27.899</td> <td> 0.000</td> <td>   -1.158</td> <td>   -1.006</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>naf17_OQ</th>  <td>   -0.8505</td> <td>    0.040</td> <td>  -21.376</td> <td> 0.000</td> <td>   -0.928</td> <td>   -0.772</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>CJSITUA_6</th> <td>    3.0820</td> <td>    0.135</td> <td>   22.867</td> <td> 0.000</td> <td>    2.818</td> <td>    3.346</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>TINA_3</th>    <td>    1.4998</td> <td>    0.061</td> <td>   24.394</td> <td> 0.000</td> <td>    1.379</td> <td>    1.620</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>cser_6</th>    <td>    1.1193</td> <td>    0.052</td> <td>   21.421</td> <td> 0.000</td> <td>    1.017</td> <td>    1.222</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>REVMENS</th>   <td>    1.3330</td> <td>    0.057</td> <td>   23.295</td> <td> 0.000</td> <td>    1.221</td> <td>    1.445</td>\n",
       "</tr>\n",
       "</table>"
      ],
      "text/latex": [
       "\\begin{center}\n",
       "\\begin{tabular}{lclc}\n",
       "\\toprule\n",
       "\\textbf{Dep. Variable:}   &      target      & \\textbf{  No. Observations:  } &    17844    \\\\\n",
       "\\textbf{Model:}           &      Logit       & \\textbf{  Df Residuals:      } &    17836    \\\\\n",
       "\\textbf{Method:}          &       MLE        & \\textbf{  Df Model:          } &        7    \\\\\n",
       "\\textbf{Date:}            & Sun, 07 Jul 2024 & \\textbf{  Pseudo R-squ.:     } &   0.2945    \\\\\n",
       "\\textbf{Time:}            &     17:03:35     & \\textbf{  Log-Likelihood:    } &   -8403.4   \\\\\n",
       "\\textbf{converged:}       &       True       & \\textbf{  LL-Null:           } &   -11910.   \\\\\n",
       "\\textbf{Covariance Type:} &    nonrobust     & \\textbf{  LLR p-value:       } &    0.000    \\\\\n",
       "\\bottomrule\n",
       "\\end{tabular}\n",
       "\\begin{tabular}{lcccccc}\n",
       "                    & \\textbf{coef} & \\textbf{std err} & \\textbf{z} & \\textbf{P$> |$z$|$} & \\textbf{[0.025} & \\textbf{0.975]}  \\\\\n",
       "\\midrule\n",
       "\\textbf{const}      &      -2.4526  &        0.091     &   -26.926  &         0.000        &       -2.631    &       -2.274     \\\\\n",
       "\\textbf{TPP\\_1}     &       1.6284  &        0.071     &    23.075  &         0.000        &        1.490    &        1.767     \\\\\n",
       "\\textbf{SECROUT\\_2} &      -1.0821  &        0.039     &   -27.899  &         0.000        &       -1.158    &       -1.006     \\\\\n",
       "\\textbf{naf17\\_OQ}  &      -0.8505  &        0.040     &   -21.376  &         0.000        &       -0.928    &       -0.772     \\\\\n",
       "\\textbf{CJSITUA\\_6} &       3.0820  &        0.135     &    22.867  &         0.000        &        2.818    &        3.346     \\\\\n",
       "\\textbf{TINA\\_3}    &       1.4998  &        0.061     &    24.394  &         0.000        &        1.379    &        1.620     \\\\\n",
       "\\textbf{cser\\_6}    &       1.1193  &        0.052     &    21.421  &         0.000        &        1.017    &        1.222     \\\\\n",
       "\\textbf{REVMENS}    &       1.3330  &        0.057     &    23.295  &         0.000        &        1.221    &        1.445     \\\\\n",
       "\\bottomrule\n",
       "\\end{tabular}\n",
       "%\\caption{Logit Regression Results}\n",
       "\\end{center}"
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.summary.Summary'>\n",
       "\"\"\"\n",
       "                           Logit Regression Results                           \n",
       "==============================================================================\n",
       "Dep. Variable:                 target   No. Observations:                17844\n",
       "Model:                          Logit   Df Residuals:                    17836\n",
       "Method:                           MLE   Df Model:                            7\n",
       "Date:                Sun, 07 Jul 2024   Pseudo R-squ.:                  0.2945\n",
       "Time:                        17:03:35   Log-Likelihood:                -8403.4\n",
       "converged:                       True   LL-Null:                       -11910.\n",
       "Covariance Type:            nonrobust   LLR p-value:                     0.000\n",
       "==============================================================================\n",
       "                 coef    std err          z      P>|z|      [0.025      0.975]\n",
       "------------------------------------------------------------------------------\n",
       "const         -2.4526      0.091    -26.926      0.000      -2.631      -2.274\n",
       "TPP_1          1.6284      0.071     23.075      0.000       1.490       1.767\n",
       "SECROUT_2     -1.0821      0.039    -27.899      0.000      -1.158      -1.006\n",
       "naf17_OQ      -0.8505      0.040    -21.376      0.000      -0.928      -0.772\n",
       "CJSITUA_6      3.0820      0.135     22.867      0.000       2.818       3.346\n",
       "TINA_3         1.4998      0.061     24.394      0.000       1.379       1.620\n",
       "cser_6         1.1193      0.052     21.421      0.000       1.017       1.222\n",
       "REVMENS        1.3330      0.057     23.295      0.000       1.221       1.445\n",
       "==============================================================================\n",
       "\"\"\""
      ]
     },
     "execution_count": 176,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "import statsmodels.api as sm\n",
    "\n",
    "col_target = \"target\"\n",
    "col_features = best_features\n",
    "X = one_hot_df[col_features].astype(float)\n",
    "X = sm.add_constant(X)\n",
    "X_train, X_test, y_train, y_test = train_test_split(\n",
    "    X, one_hot_df[col_target], test_size=0.2, random_state=42\n",
    ")\n",
    "model = sm.Logit(y_train, X_train)\n",
    "result = model.fit()\n",
    "summary = result.summary()\n",
    "summary"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### The model is still performant and significant"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### E. Model testing\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 177,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Import the necessary libraries for model testing\n",
    "from sklearn.metrics import (\n",
    "    confusion_matrix,\n",
    "    roc_auc_score,\n",
    "    roc_curve,\n",
    "    precision_score,\n",
    "    recall_score,\n",
    "    f1_score,\n",
    ")\n",
    "from sklearn.calibration import calibration_curve"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 0- Predictions:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 178,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Make the predictions using the model \n",
    "y_pred_proba = [] # Create an empty list to store the predicted probabilities\n",
    "for i in range(0, len(X_test)):\n",
    "    y_pred_proba.append(result.predict(X_test.iloc[i].values))\n",
    "y_pred = (np.array(y_pred_proba) > 0.5).astype(int) # Convert the predicted probabilities to binary values"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 1- Precision, Recall & F1 Metrics:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 179,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Precision: 0.7225063938618926\n",
      "Recall: 0.653556969346443\n",
      "F1 Score: 0.6863042818098998\n"
     ]
    }
   ],
   "source": [
    "# Calculate precision, recall, and F1 score metrics for the model\n",
    "precision = precision_score(y_test, y_pred)\n",
    "recall = recall_score(y_test, y_pred)\n",
    "f1 = f1_score(y_test, y_pred)\n",
    "\n",
    "print(\"Precision:\", precision)\n",
    "print(\"Recall:\", recall)\n",
    "print(\"F1 Score:\", f1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 2- Confusion Matrix:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 180,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x600 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "# Confusion Matrix\n",
    "conf_matrix = confusion_matrix(y_test, y_pred)\n",
    "\n",
    "# Plot the confusion matrix\n",
    "plt.figure(figsize=(8, 6))\n",
    "sns.heatmap(\n",
    "    conf_matrix,\n",
    "    annot=True,\n",
    "    fmt=\"d\",\n",
    "    cmap=\"Blues\",\n",
    "    xticklabels=[\"Predicted 0\", \"Predicted 1\"],\n",
    "    yticklabels=[\"Actual 0\", \"Actual 1\"],\n",
    ")\n",
    "plt.xlabel(\"Predicted\")\n",
    "plt.ylabel(\"Actual\")\n",
    "plt.title(\"Confusion Matrix for logit model\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "It looks like the model is better at predicting negative instances (Predicted 0) than positive instances (Predicted 1).\n",
    "This could also be caused by a classes imbalance which needs further processing.\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 3- ROC Curve & ROC-AUC metric:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 181,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ROC AUC Score: 0.85\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# ROC AUC Score\n",
    "roc_auc = roc_auc_score(y_test, y_pred_proba)\n",
    "print(f\"ROC AUC Score: {roc_auc:.2f}\")\n",
    "\n",
    "# ROC curve\n",
    "fpr, tpr, thresholds = roc_curve(y_test, y_pred_proba)\n",
    "# Plotting ROC curve\n",
    "plt.figure()\n",
    "plt.plot(fpr, tpr, color=\"darkorange\", lw=2, label=\"ROC curve (area = %0.2f)\" % roc_auc)\n",
    "plt.plot([0, 1], [0, 1], color=\"navy\", lw=2, linestyle=\"--\")\n",
    "plt.xlim([0.0, 1.0])\n",
    "plt.ylim([0.0, 1.05])\n",
    "plt.xlabel(\"False Positive Rate\")\n",
    "plt.ylabel(\"True Positive Rate\")\n",
    "plt.title(\"Receiver Operating Characteristic (ROC)\")\n",
    "plt.legend(loc=\"lower right\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "With a ROC AUC of 0.85 we can say that the model performs well in distinguishing between the two classes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 182,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Calculate the calibration curve\n",
    "prob_true, prob_pred = calibration_curve(y_test, y_pred_proba, n_bins=26)\n",
    "\n",
    "# Plot the calibration curve\n",
    "plt.plot(prob_pred, prob_true, marker=\"o\", label=\"Reliability\")\n",
    "plt.plot([0, 1], [0, 1], linestyle=\"--\", label=\"Perfectly Calibrated\")\n",
    "plt.xlabel(\"Predicted Probability\")\n",
    "plt.title(\"Calibration Curve\")\n",
    "plt.ylabel(\"True Fraction\")\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The graph shows that the model’s predictions are well calibrated, meaning that the predicted probability of an event closely matches the actual probability of the event. The reliability is high throughout, which means the model is consistently good at making these predictions excpet some minor disturbances at predictions with probabilities between 0.6 & 0.8."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 5. Conclusion"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### After multiple levels of feature selection, model fitting & optimization we found the following variables to be the determinants for a woman being the primary source of revenue in the household:\n",
    "### **Work Characteristics:**\n",
    "\n",
    "- **Full-time Work (TPP_1)**: Women who work full-time (TPP_1) are more likely to assume the role of primary source of revenue, as indicated by its positive coefficient of (1.62).\n",
    "- **Workplace Risk (SECROUT_2)**: The absence of high-risk workplace conditions (SECROUT_2) correlates negatively with the likelihood of a woman being the primary source of revenue (-1.08).\n",
    "\n",
    "### **Occupational Factors:**\n",
    "\n",
    "- **Industry Sector (naf17_OQ)**: Engaging in administrative activities, education, or health sectors (naf17_OQ) corresponds to a decrease in the likelihood (-0.85) of a woman being the primary source of revenue.\n",
    "- **Socio-Professional Category (cser_6)**: Occupying blue-collar positions (cser_6), such as 'Ouvrier', is associated with a higher probability (1.12) of being the primary source of revenue.\n",
    "\n",
    "### **Family and Career Dynamics:**\n",
    "\n",
    "- **Spouse's Work Situation (CJSITUA_6)**: Having an inactive spouse (CJSITUA_6), such as one who is a homemaker or retired, significantly increases the odds (3.08) of a woman being the primary source of revenue.\n",
    "- **Career Breaks (TINA_3)**: Never having taken extended breaks from work (TINA_3) positively influences (1.5) the likelihood of being the primary source of revenue.\n",
    "\n",
    "### **Income Considerations:**\n",
    "\n",
    "- **Monthly Income (REVMENS)**: Higher net monthly income (REVMENS) positively impacts (1.33) the likelihood of a woman being the primary source of revenue.\n",
    "\n",
    "These variables collectively provide insights into the socioeconomic factors influencing the likelihood of a woman being the primary source of revenue in a household, highlighting both occupational and household dynamics impacting income and employment roles.\n",
    "\n"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": ".venv",
   "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": "3.11.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}