Notebook-MRF.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "7c1b1b2d-174f-43f2-831d-9a7f65d7a2d7",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import matplotlib.cm as cm\n",
    "import seaborn as sns\n",
    "import sklearn as sklearn\n",
    "from sklearn.preprocessing import LabelEncoder\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.feature_selection import RFE\n",
    "from sklearn.model_selection import train_test_split, GridSearchCV\n",
    "from sklearn.metrics import accuracy_score, precision_score, recall_score, f1_score, roc_auc_score\n",
    "from sklearn.model_selection import StratifiedKFold\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.tree import DecisionTreeClassifier\n",
    "from sklearn.ensemble import RandomForestClassifier, GradientBoostingClassifier\n",
    "from sklearn.svm import SVC"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "751c8510-e559-48a5-a6f6-b7c59d33af29",
   "metadata": {},
   "outputs": [],
   "source": [
    "import warnings\n",
    "warnings.filterwarnings('ignore')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c14e51b7-5c34-490b-8e8d-fd653a506c44",
   "metadata": {},
   "source": [
    "## I. Data Processing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "ad509bc8-9c55-48a0-84a2-4820c2769ef3",
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.read_csv(\"diabetic_data.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "984e10d2-1ac5-45ac-8e3b-1759f75b62ce",
   "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>encounter_id</th>\n",
       "      <th>patient_nbr</th>\n",
       "      <th>race</th>\n",
       "      <th>gender</th>\n",
       "      <th>age</th>\n",
       "      <th>weight</th>\n",
       "      <th>admission_type_id</th>\n",
       "      <th>discharge_disposition_id</th>\n",
       "      <th>admission_source_id</th>\n",
       "      <th>time_in_hospital</th>\n",
       "      <th>...</th>\n",
       "      <th>citoglipton</th>\n",
       "      <th>insulin</th>\n",
       "      <th>glyburide-metformin</th>\n",
       "      <th>glipizide-metformin</th>\n",
       "      <th>glimepiride-pioglitazone</th>\n",
       "      <th>metformin-rosiglitazone</th>\n",
       "      <th>metformin-pioglitazone</th>\n",
       "      <th>change</th>\n",
       "      <th>diabetesMed</th>\n",
       "      <th>readmitted</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2278392</td>\n",
       "      <td>8222157</td>\n",
       "      <td>Caucasian</td>\n",
       "      <td>Female</td>\n",
       "      <td>[0-10)</td>\n",
       "      <td>Unknown</td>\n",
       "      <td>6</td>\n",
       "      <td>25</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>NO</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>149190</td>\n",
       "      <td>55629189</td>\n",
       "      <td>Caucasian</td>\n",
       "      <td>Female</td>\n",
       "      <td>[10-20)</td>\n",
       "      <td>Unknown</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>7</td>\n",
       "      <td>3</td>\n",
       "      <td>...</td>\n",
       "      <td>No</td>\n",
       "      <td>Up</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>Ch</td>\n",
       "      <td>Yes</td>\n",
       "      <td>&gt;30</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>64410</td>\n",
       "      <td>86047875</td>\n",
       "      <td>AfricanAmerican</td>\n",
       "      <td>Female</td>\n",
       "      <td>[20-30)</td>\n",
       "      <td>Unknown</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>7</td>\n",
       "      <td>2</td>\n",
       "      <td>...</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>Yes</td>\n",
       "      <td>NO</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>500364</td>\n",
       "      <td>82442376</td>\n",
       "      <td>Caucasian</td>\n",
       "      <td>Male</td>\n",
       "      <td>[30-40)</td>\n",
       "      <td>Unknown</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>7</td>\n",
       "      <td>2</td>\n",
       "      <td>...</td>\n",
       "      <td>No</td>\n",
       "      <td>Up</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>Ch</td>\n",
       "      <td>Yes</td>\n",
       "      <td>NO</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>16680</td>\n",
       "      <td>42519267</td>\n",
       "      <td>Caucasian</td>\n",
       "      <td>Male</td>\n",
       "      <td>[40-50)</td>\n",
       "      <td>Unknown</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>7</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>No</td>\n",
       "      <td>Steady</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>Ch</td>\n",
       "      <td>Yes</td>\n",
       "      <td>NO</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 50 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   encounter_id  patient_nbr             race  gender      age   weight  \\\n",
       "0       2278392      8222157        Caucasian  Female   [0-10)  Unknown   \n",
       "1        149190     55629189        Caucasian  Female  [10-20)  Unknown   \n",
       "2         64410     86047875  AfricanAmerican  Female  [20-30)  Unknown   \n",
       "3        500364     82442376        Caucasian    Male  [30-40)  Unknown   \n",
       "4         16680     42519267        Caucasian    Male  [40-50)  Unknown   \n",
       "\n",
       "   admission_type_id  discharge_disposition_id  admission_source_id  \\\n",
       "0                  6                        25                    1   \n",
       "1                  1                         1                    7   \n",
       "2                  1                         1                    7   \n",
       "3                  1                         1                    7   \n",
       "4                  1                         1                    7   \n",
       "\n",
       "   time_in_hospital  ... citoglipton insulin  glyburide-metformin  \\\n",
       "0                 1  ...          No      No                   No   \n",
       "1                 3  ...          No      Up                   No   \n",
       "2                 2  ...          No      No                   No   \n",
       "3                 2  ...          No      Up                   No   \n",
       "4                 1  ...          No  Steady                   No   \n",
       "\n",
       "   glipizide-metformin  glimepiride-pioglitazone  metformin-rosiglitazone  \\\n",
       "0                   No                        No                       No   \n",
       "1                   No                        No                       No   \n",
       "2                   No                        No                       No   \n",
       "3                   No                        No                       No   \n",
       "4                   No                        No                       No   \n",
       "\n",
       "   metformin-pioglitazone  change diabetesMed readmitted  \n",
       "0                      No      No          No         NO  \n",
       "1                      No      Ch         Yes        >30  \n",
       "2                      No      No         Yes         NO  \n",
       "3                      No      Ch         Yes         NO  \n",
       "4                      No      Ch         Yes         NO  \n",
       "\n",
       "[5 rows x 50 columns]"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Handle unique values in columns\n",
    "unique_value_columns = ['weight', 'payer_code', 'medical_specialty', 'diag_1', 'diag_2', 'diag_3']\n",
    "\n",
    "for column in unique_value_columns:\n",
    "    df[column] = df[column].astype(str)  # Convert the column to string type\n",
    "    df[column] = df[column].replace('?', pd.NA)  # Replace '?' with NaN\n",
    "    df[column] = df[column].fillna('Unknown')  # Fill NaN with 'Unknown'\n",
    "\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "b458c6e4-e87b-4130-a474-afab9bd3185a",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Preprocess the age feature\n",
    "age_mapping = {\n",
    "    '[0-10)': 5,\n",
    "    '[10-20)': 15,\n",
    "    '[20-30)': 25,\n",
    "    '[30-40)': 35,\n",
    "    '[40-50)': 45,\n",
    "    '[50-60)': 55,\n",
    "    '[60-70)': 65,\n",
    "    '[70-80)': 75,   \n",
    "    '[80-90)': 85,\n",
    "    '[90-100)': 95\n",
    "}\n",
    "df['age'] = df['age'].map(age_mapping)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "52955ab6-2658-4192-b22f-4c9620921061",
   "metadata": {},
   "outputs": [],
   "source": [
    "df = df.dropna()  # Remove rows with missing values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "53f8df3e-57aa-4ab0-81b1-8f27eb2b29c1",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Identify categorical columns\n",
    "nominal_cols = ['race']\n",
    "ordinal_cols = ['gender', 'max_glu_serum', 'A1Cresult', 'metformin', 'repaglinide', 'nateglinide', 'chlorpropamide', 'glimepiride', 'acetohexamide',\n",
    "               'glipizide', 'glyburide', 'tolbutamide', 'tolbutamide', 'pioglitazone', 'rosiglitazone', 'acarbose', 'miglitol', 'troglitazone',\n",
    "               'tolazamide', 'examide', 'citoglipton', 'insulin', 'glyburide-metformin', 'glipizide-metformin', 'glimepiride-pioglitazone', 'metformin-rosiglitazone',\n",
    "               'metformin-pioglitazone', 'change', 'diabetesMed'] + unique_value_columns\n",
    "\n",
    "# Applying one-hot encoding to 'race'\n",
    "df = pd.get_dummies(df, columns=nominal_cols)\n",
    "\n",
    "# Apply Label Encoding to other (naturally) ordinal columns\n",
    "label_encoders = {}\n",
    "for column in ordinal_cols:\n",
    "    label_encoders[column] = LabelEncoder()\n",
    "    df[column] = label_encoders[column].fit_transform(df[column])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "977326e8-4d17-4937-a765-9fa5a0c28381",
   "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>encounter_id</th>\n",
       "      <th>patient_nbr</th>\n",
       "      <th>gender</th>\n",
       "      <th>age</th>\n",
       "      <th>weight</th>\n",
       "      <th>admission_type_id</th>\n",
       "      <th>discharge_disposition_id</th>\n",
       "      <th>admission_source_id</th>\n",
       "      <th>time_in_hospital</th>\n",
       "      <th>payer_code</th>\n",
       "      <th>...</th>\n",
       "      <th>metformin-pioglitazone</th>\n",
       "      <th>change</th>\n",
       "      <th>diabetesMed</th>\n",
       "      <th>readmitted</th>\n",
       "      <th>race_?</th>\n",
       "      <th>race_AfricanAmerican</th>\n",
       "      <th>race_Asian</th>\n",
       "      <th>race_Caucasian</th>\n",
       "      <th>race_Hispanic</th>\n",
       "      <th>race_Other</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>162</th>\n",
       "      <td>2521974</td>\n",
       "      <td>80499960</td>\n",
       "      <td>1</td>\n",
       "      <td>85</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>3</td>\n",
       "      <td>7</td>\n",
       "      <td>5</td>\n",
       "      <td>4</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>&lt;30</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>192</th>\n",
       "      <td>2595612</td>\n",
       "      <td>89193870</td>\n",
       "      <td>0</td>\n",
       "      <td>45</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>1</td>\n",
       "      <td>7</td>\n",
       "      <td>2</td>\n",
       "      <td>4</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>&gt;30</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>460</th>\n",
       "      <td>4084524</td>\n",
       "      <td>76959585</td>\n",
       "      <td>0</td>\n",
       "      <td>75</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>1</td>\n",
       "      <td>7</td>\n",
       "      <td>10</td>\n",
       "      <td>4</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>&gt;30</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>593</th>\n",
       "      <td>4804968</td>\n",
       "      <td>114960726</td>\n",
       "      <td>0</td>\n",
       "      <td>55</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>1</td>\n",
       "      <td>7</td>\n",
       "      <td>2</td>\n",
       "      <td>4</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>NO</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>696</th>\n",
       "      <td>5319174</td>\n",
       "      <td>87847776</td>\n",
       "      <td>1</td>\n",
       "      <td>75</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>6</td>\n",
       "      <td>7</td>\n",
       "      <td>11</td>\n",
       "      <td>4</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>NO</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",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 55 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     encounter_id  patient_nbr  gender  age  weight  admission_type_id  \\\n",
       "162       2521974     80499960       1   85       0                  6   \n",
       "192       2595612     89193870       0   45       0                  6   \n",
       "460       4084524     76959585       0   75       0                  6   \n",
       "593       4804968    114960726       0   55       0                  6   \n",
       "696       5319174     87847776       1   75       0                  6   \n",
       "\n",
       "     discharge_disposition_id  admission_source_id  time_in_hospital  \\\n",
       "162                         3                    7                 5   \n",
       "192                         1                    7                 2   \n",
       "460                         1                    7                10   \n",
       "593                         1                    7                 2   \n",
       "696                         6                    7                11   \n",
       "\n",
       "     payer_code  ...  metformin-pioglitazone  change  diabetesMed  readmitted  \\\n",
       "162           4  ...                       0       1            0         <30   \n",
       "192           4  ...                       0       1            0         >30   \n",
       "460           4  ...                       0       0            1         >30   \n",
       "593           4  ...                       0       1            1          NO   \n",
       "696           4  ...                       0       1            1          NO   \n",
       "\n",
       "     race_?  race_AfricanAmerican  race_Asian  race_Caucasian  race_Hispanic  \\\n",
       "162   False                 False       False            True          False   \n",
       "192   False                 False       False            True          False   \n",
       "460   False                  True       False           False          False   \n",
       "593   False                 False       False            True          False   \n",
       "696   False                 False       False           False          False   \n",
       "\n",
       "     race_Other  \n",
       "162       False  \n",
       "192       False  \n",
       "460       False  \n",
       "593       False  \n",
       "696        True  \n",
       "\n",
       "[5 rows x 55 columns]"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "62be85e3-9565-4049-b559-03663ab5a1cb",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Convert the target variable to binary\n",
    "df['readmitted'] = df['readmitted'].apply(lambda x: 1 if x in ['<30', '>30'] else 0)\n",
    "\n",
    "# Split the data into features (X) and target variable (y)\n",
    "X = df.drop(['encounter_id', 'patient_nbr', 'readmitted', 'weight'], axis=1)\n",
    "y = df['readmitted']"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e61782f1-063b-4320-87e4-7e4cbe7f9cbf",
   "metadata": {},
   "source": [
    "## II. Feature Selection"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "8518064f-f43f-4377-bd91-7d98d61b36be",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Selected Features: ['age', 'time_in_hospital', 'num_lab_procedures', 'num_medications', 'number_inpatient', 'diag_1', 'diag_2', 'diag_3', 'number_diagnoses']\n",
      "Accuracy: 0.75\n",
      "Precision: 0.7333333333333333\n",
      "Recall: 0.9166666666666666\n",
      "F1-score: 0.8148148148148148\n",
      "AUC-ROC: 0.7957175925925927\n",
      "Cross-validation Accuracy: 0.5974576271186441\n"
     ]
    }
   ],
   "source": [
    "# Split the data into training and testing sets\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)\n",
    "\n",
    "# Create a Random Forest classifier\n",
    "rf = RandomForestClassifier(n_estimators=100, random_state=42)\n",
    "\n",
    "# Perform RFE with stability selection\n",
    "n_features_to_select = 10  # Number of features to select\n",
    "n_iterations = 50  # Number of iterations for stability selection\n",
    "rfe_scores = []\n",
    "\n",
    "for i in range(n_iterations):\n",
    "    # Create an RFE object with the Random Forest classifier\n",
    "    rfe = RFE(estimator=rf, n_features_to_select=n_features_to_select, step=1)\n",
    "    \n",
    "    # Fit RFE on a random subset of the training data\n",
    "    train_indices = np.random.choice(X_train.shape[0], size=int(0.8 * X_train.shape[0]), replace=False)\n",
    "    rfe.fit(X_train.iloc[train_indices], y_train.iloc[train_indices])\n",
    "    \n",
    "    # Store the selected features\n",
    "    selected_features = X_train.columns[rfe.support_]\n",
    "    rfe_scores.append(selected_features)\n",
    "\n",
    "# Calculate the stability scores for each feature\n",
    "stability_scores = {}\n",
    "for feature in X_train.columns:\n",
    "    stability_scores[feature] = sum(feature in selection for selection in rfe_scores) / n_iterations\n",
    "\n",
    "# Select the features with stability scores above a threshold\n",
    "stability_threshold = 0.8\n",
    "selected_features = [feature for feature, score in stability_scores.items() if score >= stability_threshold]\n",
    "\n",
    "# Fit the Random Forest classifier on the selected features\n",
    "rf.fit(X_train[selected_features], y_train)\n",
    "\n",
    "# Evaluate the model on the testing set\n",
    "y_pred = rf.predict(X_test[selected_features])\n",
    "accuracy = accuracy_score(y_test, y_pred)\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",
    "auc_roc = roc_auc_score(y_test, rf.predict_proba(X_test[selected_features])[:, 1])\n",
    "\n",
    "print(\"Selected Features:\", selected_features)\n",
    "print(\"Accuracy:\", accuracy)\n",
    "print(\"Precision:\", precision)\n",
    "print(\"Recall:\", recall)\n",
    "print(\"F1-score:\", f1)\n",
    "print(\"AUC-ROC:\", auc_roc)\n",
    "\n",
    "# Perform cross-validation to assess the stability of the selected features\n",
    "cv = StratifiedKFold(n_splits=5)\n",
    "cv_scores = []\n",
    "\n",
    "for train_index, test_index in cv.split(X, y):\n",
    "    X_train_cv, X_test_cv = X.iloc[train_index], X.iloc[test_index]\n",
    "    y_train_cv, y_test_cv = y.iloc[train_index], y.iloc[test_index]\n",
    "    \n",
    "    rf.fit(X_train_cv[selected_features], y_train_cv)\n",
    "    y_pred_cv = rf.predict(X_test_cv[selected_features])\n",
    "    cv_scores.append(accuracy_score(y_test_cv, y_pred_cv))\n",
    "\n",
    "print(\"Cross-validation Accuracy:\", np.mean(cv_scores))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "3da58916-b678-489e-a8ce-d6103eb5ade2",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA90AAAJOCAYAAACqS2TfAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjguMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/H5lhTAAAACXBIWXMAAA9hAAAPYQGoP6dpAACvoUlEQVR4nOzde3zO9f/H8ee1sY2xOTVj1uZ8Ps9ZJlZIoeRQZA45ZcLKOcfSEmlCFqGIHKNIookO5CwhVIRocz5sY2N7//7w2/Xd1aZsdrlsHvfb7bpxva/353O9Pu9d++zzvD4nizHGCAAAAAAAZDonRxcAAAAAAEB2RegGAAAAAMBOCN0AAAAAANgJoRsAAAAAADshdAMAAAAAYCeEbgAAAAAA7ITQDQAAAACAnRC6AQAAAACwE0I3AAAAAAB2QugGAAD3RHR0tJ599lkVLFhQFotF4eHh9+y9N23aJIvFok2bNt2z98wsXbt2lb+/v6PLAABkEKEbABzoo48+ksViSfMxbNgwu7znli1bNHbsWF26dMku878byeOxc+dOR5eSYe+//74++ugjR5dxXxo0aJC+/vprDR8+XAsWLFDz5s1v2zcmJkZjxoxRpUqV5O7uroIFC6patWoaMGCATp8+fQ+rvjMHDx7U2LFj9eeffzqshsaNG992fXLo0CG7vCefdwD4bzkcXQAAQBo/fryKFy9u01apUiW7vNeWLVs0btw4de3aVfny5bPLezzI3n//fRUqVEhdu3Z1dCn3nY0bN6p169Z69dVX/7XfjRs31KhRIx06dEjBwcHq37+/YmJidODAAS1atEhPP/20ihYteo+qvjMHDx7UuHHj1LhxY4fulS5WrJjCwsJStdtrvPi8A8B/I3QDwH2gRYsWCggIcHQZdyU2Nlbu7u6OLsNh4uLilDt3bkeXcV87c+bMHX3Rs2rVKu3Zs0cLFy7U888/b/Pa9evXlZCQYKcKsz5PT0917tzZ0WXcFWOMrl+/rly5cjm6FADIFBxeDgBZwFdffaVHHnlE7u7uyps3r1q2bKkDBw7Y9Nm3b5+6du2qEiVKyM3NTd7e3urevbvOnz9v7TN27FgNHjxYklS8eHHroad//vmn/vzzT1ksljQPFbVYLBo7dqzNfCwWiw4ePKjnn39e+fPnV8OGDa2vf/LJJ6pZs6Zy5cqlAgUKqGPHjjp58mSGlr1r167KkyePTpw4oSeffFJ58uSRj4+PZsyYIUn65Zdf1KRJE7m7u8vPz0+LFi2ymT75kPXvvvtOvXv3VsGCBeXh4aEuXbro4sWLqd7v/fffV8WKFeXq6qqiRYuqX79+qQ7Fb9y4sSpVqqRdu3apUaNGyp07t0aMGCF/f38dOHBAmzdvto5t48aNJUkXLlzQq6++qsqVKytPnjzy8PBQixYt9PPPP9vMO/nc46VLl2rChAkqVqyY3Nzc1LRpU/3++++p6t22bZueeOIJ5c+fX+7u7qpSpYqmTp1q0+fQoUN69tlnVaBAAbm5uSkgIEBffPGFTZ8bN25o3LhxKl26tNzc3FSwYEE1bNhQGzZs+M+f0dGjR9WuXTsVKFBAuXPnVt26dfXll1+m+hkYYzRjxgzr2NzOH3/8IUlq0KBBqtfc3Nzk4eGR7uW7nW3btql58+by9PRU7ty5FRgYqB9//DFVv1OnTqlHjx4qWrSoXF1dVbx4cfXt21cJCQn66KOP1K5dO0nSo48+al2+lOeP38nvsHTrC4dKlSrJzc1NlSpV0sqVK+9oOe5UfHy8xowZo1KlSsnV1VW+vr4aMmSI4uPjbfrNmzdPTZo0kZeXl1xdXVWhQgXNnDnTps+/fd6T1xH/lPxZSHkYvr+/v5588kl9/fXXCggIUK5cufTBBx9Iki5duqSBAwfK19dXrq6uKlWqlCZOnKikpCSb+S5evFg1a9ZU3rx55eHhocqVK6f6PQAAR2FPNwDcBy5fvqxz587ZtBUqVEiStGDBAgUHB6tZs2aaOHGi4uLiNHPmTDVs2FB79uyxHsq6YcMGHT16VN26dZO3t7cOHDigWbNm6cCBA/rpp59ksVj0zDPP6MiRI/r000/17rvvWt/joYce0tmzZ9Ndd7t27VS6dGm9+eabMsZIkiZMmKBRo0apffv2evHFF3X27FlNmzZNjRo10p49ezJ0SHtiYqJatGihRo0a6e2339bChQsVEhIid3d3jRw5Up06ddIzzzyjiIgIdenSRfXq1Ut1uH5ISIjy5cunsWPH6vDhw5o5c6aOHz9uDbnSraAwbtw4BQUFqW/fvtZ+O3bs0I8//qicOXNa53f+/Hm1aNFCHTt2VOfOnVW4cGE1btxY/fv3V548eTRy5EhJUuHChSXdCqarVq1Su3btVLx4cUVHR+uDDz5QYGCgDh48mOrw37feektOTk569dVXdfnyZb399tvq1KmTtm3bZu2zYcMGPfnkkypSpIgGDBggb29v/frrr1qzZo0GDBggSTpw4IAaNGggHx8fDRs2TO7u7lq6dKnatGmjFStW6Omnn7Yue1hYmF588UXVrl1bV65c0c6dO7V792499thjt/3ZREdHq379+oqLi9PLL7+sggUL6uOPP1arVq20fPlyPf3002rUqJEWLFigF154QY899pi6dOnyrz9vPz8/SdL8+fP12muv/WtAv9PlS8vGjRvVokUL1axZU2PGjJGTk5M1bH7//feqXbu2JOn06dOqXbu2Ll26pF69eqlcuXI6deqUli9frri4ODVq1Egvv/yy3nvvPY0YMULly5eXJOu/d/o7vH79erVt21YVKlRQWFiYzp8/r27duqlYsWL/Ol4pJSYmplqXuLm5KU+ePEpKSlKrVq30ww8/qFevXipfvrx++eUXvfvuuzpy5IhWrVplnWbmzJmqWLGiWrVqpRw5cmj16tV66aWXlJSUpH79+kmSwsPDb/t5T6/Dhw/rueeeU+/evdWzZ0+VLVtWcXFxCgwM1KlTp9S7d289/PDD2rJli4YPH66///7beiG+DRs26LnnnlPTpk01ceJESdKvv/6qH3/80fp7AAAOZQAADjNv3jwjKc2HMcZcvXrV5MuXz/Ts2dNmuqioKOPp6WnTHhcXl2r+n376qZFkvvvuO2vbpEmTjCRz7Ngxm77Hjh0zksy8efNSzUeSGTNmjPX5mDFjjCTz3HPP2fT7888/jbOzs5kwYYJN+y+//GJy5MiRqv1247Fjxw5rW3BwsJFk3nzzTWvbxYsXTa5cuYzFYjGLFy+2th86dChVrcnzrFmzpklISLC2v/3220aS+fzzz40xxpw5c8a4uLiYxx9/3CQmJlr7TZ8+3Ugyc+fOtbYFBgYaSSYiIiLVMlSsWNEEBgamar9+/brNfI25Neaurq5m/Pjx1rZvv/3WSDLly5c38fHx1vapU6caSeaXX34xxhhz8+ZNU7x4cePn52cuXrxoM9+kpCTr/5s2bWoqV65srl+/bvN6/fr1TenSpa1tVatWNS1btkxV938ZOHCgkWS+//57a9vVq1dN8eLFjb+/v80ySzL9+vX7z3nGxcWZsmXLGknGz8/PdO3a1cyZM8dER0en6nuny5c8rt9++621T+nSpU2zZs1sxisuLs4UL17cPPbYY9a2Ll26GCcnJ5vPZcr3MsaYZcuW2cw/5Vjc6e9wtWrVTJEiRcylS5esbevXr7eOw39J/lz+8xEcHGyMMWbBggXGycnJ5mdljDERERFGkvnxxx9txuGfmjVrZkqUKGHTdrvPe/I64p+Sfx9Trn/8/PyMJLNu3Tqbvq+//rpxd3c3R44csWkfNmyYcXZ2NidOnDDGGDNgwADj4eFhbt68mXpQAOA+wOHlAHAfmDFjhjZs2GDzkG7twbl06ZKee+45nTt3zvpwdnZWnTp19O2331rnkfL8x+vXr+vcuXOqW7euJGn37t12qbtPnz42zz/77DMlJSWpffv2NvV6e3urdOnSNvWm14svvmj9f758+VS2bFm5u7urffv21vayZcsqX758Onr0aKrpe/XqZbOnum/fvsqRI4fWrl0rSfrmm2+UkJCggQMHysnpf38ee/bsKQ8PD5vDpSXJ1dVV3bp1u+P6XV1drfNNTEzU+fPnlSdPHpUtWzbNn0+3bt3k4uJiff7II49IknXZ9uzZo2PHjmngwIGpjh5I3jN84cIFbdy4Ue3bt9fVq1etP4/z58+rWbNm+u2333Tq1ClJt8b0wIED+u233+54mSRp7dq1ql27ts3pBXny5FGvXr30559/6uDBg+man3Trs7xt2zbrqRAfffSRevTooSJFiqh///7WQ6HTs3z/tHfvXv322296/vnndf78eeu0sbGxatq0qb777jslJSUpKSlJq1at0lNPPZXmdRf+bS+8dOe/w3///bf27t2r4OBgeXp6Wqd/7LHHVKFChTseO39//1TrkiFDhkiSli1bpvLly6tcuXI2tTRp0kSSbrs+ST4SJzAwUEePHtXly5fvuJ47Vbx4cTVr1symbdmyZXrkkUeUP39+m3qDgoKUmJio7777TtKtz25sbOwdnQoBAI7A4eUAcB+oXbt2mhv0yQEoeaP4n1Ke23rhwgWNGzdOixcv1pkzZ2z62WMjWVKqQ7h/++03GWNUunTpNPunDL3p4ebmpoceesimzdPTU8WKFUsVejw9PdM8V/ufNeXJk0dFihSxnlt6/PhxSbeCe0ouLi4qUaKE9fVkPj4+NqH4vyQlJWnq1Kl6//33dezYMSUmJlpfK1iwYKr+Dz/8sM3z/PnzS5J12ZLPe/63q9z//vvvMsZo1KhRGjVqVJp9zpw5Ix8fH40fP16tW7dWmTJlVKlSJTVv3lwvvPCCqlSp8q/Ldfz4cdWpUydVe/Kh1cePH8/Qlfg9PT319ttv6+2339bx48cVGRmpyZMna/r06fL09NQbb7yRruX7p+TfreDg4NvWcPnyZSUkJOjKlSsZvpvAnf4OJ3++0vrdud0XM2lxd3dXUFDQbWv59ddfU/0uJUu53vjxxx81ZswYbd26VXFxcTb9Ll++bPPFQGb457okud59+/b9Z70vvfSSli5dqhYtWsjHx0ePP/642rdv/6+3pAOAe4nQDQD3seSLBS1YsEDe3t6pXs+R43+r8fbt22vLli0aPHiwqlWrZj2Hs3nz5qkuOpSW2+2xSxkO/+mfVxdOSkqSxWLRV199JWdn51T98+TJ8591pCWtef1bu/n/88vtKb1XVn7zzTc1atQode/eXa+//roKFCggJycnDRw4MM2fT2YsW/J8X3311VR7EZOVKlVKktSoUSP98ccf+vzzz7V+/Xp9+OGHevfddxUREWFzlIEj+Pn5qXv37nr66adVokQJLVy4UG+88Ua6lu+fkqedNGmSqlWrlmafPHny6MKFC3dVe3p+h+0tKSlJlStX1pQpU9J83dfXV9KtL3SaNm2qcuXKacqUKfL19ZWLi4vWrl2rd9991y7rk7R+n5KSkvTYY49Z99T/U5kyZSRJXl5e2rt3r77++mt99dVX+uqrrzRv3jx16dJFH3/88X/WCgD2RugGgPtYyZIlJd3aqLzd3ivp1t7PyMhIjRs3TqNHj7a2p3Wo8O02hpP3pP7zSt3/3MP7X/UaY1S8eHHrBvH94rffftOjjz5qfR4TE6O///5bTzzxhKT/Xbzr8OHDKlGihLVfQkKCjh079q/jn9Ltxnf58uV69NFHNWfOHJv2S5cuWS9olx7Jn439+/fftrbk5ciZM+cd1V+gQAF169ZN3bp1U0xMjBo1aqSxY8f+a+j28/PT4cOHU7UfOnTI+npmyZ8/v0qWLKn9+/dLSv/ypZQ8fh4eHv867UMPPSQPDw/re97O7X7ud/o7nDxOaf3OpjW+GVGyZEn9/PPPatq06b8eFr969WrFx8friy++sDniIq3TQ+5kfZLy9If0rk9iYmLu6Gfr4uKip556Sk899ZSSkpL00ksv6YMPPtCoUaNu+8ULANwrnNMNAPexZs2aycPDQ2+++aZu3LiR6vXkK44n7xX9517Q5Kv7ppR8L+1/hmsPDw8VKlTIep5ksvfff/+O633mmWfk7OyscePGparFGGNz+7J7bdasWTZjOHPmTN28eVMtWrSQJAUFBcnFxUXvvfeeTe1z5szR5cuX1bJlyzt6H3d391RjK936Gf1zTJYtW3bbc47/S40aNVS8eHGFh4ener/k9/Hy8lLjxo31wQcf6O+//041j5RXrP/nzyZPnjwqVapUqltJ/dMTTzyh7du3a+vWrda22NhYzZo1S/7+/uk6HznZzz//nOoK3NKtwHbw4EHrKQDpWb5/qlmzpkqWLKnJkycrJibmttM6OTmpTZs2Wr16tXbu3JmqX/JY3+736k5/h4sUKaJq1arp448/tjkdZMOGDRk6Lz4t7du316lTpzR79uxUr127dk2xsbGS0l6fXL58WfPmzUs13e0+78lfNqRcn8TGxqZrz3P79u21detWff3116leu3Tpkm7evCkp9WfXycnJelrEf31+AeBeYE83ANzHPDw8NHPmTL3wwguqUaOGOnbsqIceekgnTpzQl19+qQYNGmj69Ony8PCw3k7rxo0b8vHx0fr163Xs2LFU86xZs6YkaeTIkerYsaNy5sypp556Su7u7nrxxRf11ltv6cUXX1RAQIC+++47HTly5I7rLVmypN544w0NHz5cf/75p9q0aaO8efPq2LFjWrlypXr16qVXX30108YnPRISEtS0aVO1b99ehw8f1vvvv6+GDRuqVatWkm7t0Rw+fLjGjRun5s2bq1WrVtZ+tWrVUufOne/ofWrWrKmZM2fqjTfeUKlSpeTl5aUmTZroySef1Pjx49WtWzfVr19fv/zyixYuXGizVz09nJycNHPmTD311FOqVq2aunXrpiJFiujQoUM6cOCANajMmDFDDRs2VOXKldWzZ0+VKFFC0dHR2rp1q/766y/rfcIrVKigxo0bq2bNmipQoIB27typ5cuXKyQk5F/rGDZsmD799FO1aNFCL7/8sgoUKKCPP/5Yx44d04oVK2wuSnenNmzYoDFjxqhVq1aqW7eu8uTJo6NHj2ru3LmKj4+3uWf8nS5fWuP34YcfqkWLFqpYsaK6desmHx8fnTp1St9++608PDy0evVqSbdODVi/fr0CAwOtt9r6+++/tWzZMv3www/Kly+fqlWrJmdnZ02cOFGXL1+Wq6ur9T7Xd/I7LElhYWFq2bKlGjZsqO7du+vChQuaNm2aKlasmOYXA+n1wgsvaOnSperTp4++/fZbNWjQQImJiTp06JCWLl1qvU/2448/bt1z3Lt3b8XExGj27Nny8vJK9eXG7T7vjz/+uB5++GH16NFDgwcPlrOzs+bOnWtd9jsxePBgffHFF3ryySfVtWtX1axZU7Gxsfrll1+0fPly/fnnnypUqJBefPFFXbhwQU2aNFGxYsV0/PhxTZs2TdWqVbNeWwAAHMoRl0wHANyS1i2y0vLtt9+aZs2aGU9PT+Pm5mZKlixpunbtanbu3Gnt89dff5mnn37a5MuXz3h6epp27dqZ06dPp7qFljG3bsXj4+NjnJycbG7fExcXZ3r06GE8PT1N3rx5Tfv27c2ZM2due8uws2fPplnvihUrTMOGDY27u7txd3c35cqVM/369TOHDx9O93gEBwcbd3f3VH0DAwNNxYoVU7X7+fnZ3PoqeZ6bN282vXr1Mvnz5zd58uQxnTp1MufPn081/fTp0025cuVMzpw5TeHChU3fvn1T3ZLrdu9tzK1bQbVs2dLkzZvXSLLeTun69evmlVdeMUWKFDG5cuUyDRo0MFu3bjWBgYE2t1xKvrXVsmXLbOZ7u1u6/fDDD+axxx4zefPmNe7u7qZKlSpm2rRpNn3++OMP06VLF+Pt7W1y5sxpfHx8zJNPPmmWL19u7fPGG2+Y2rVrm3z58plcuXKZcuXKmQkTJtjcZu12/vjjD/Pss8+afPnyGTc3N1O7dm2zZs2aVP10h7cMO3r0qBk9erSpW7eu8fLyMjly5DAPPfSQadmypdm4cWOa7/9fy/fPW4Yl27Nnj3nmmWdMwYIFjaurq/Hz8zPt27c3kZGRNv2OHz9uunTpYh566CHj6upqSpQoYfr162dzW7fZs2ebEiVKGGdn51TvdSe/w8bc+t0pX768cXV1NRUqVDCfffaZCQ4OvuNbht3uc5ksISHBTJw40VSsWNG4urqa/Pnzm5o1a5px48aZy5cvW/t98cUXpkqVKsbNzc34+/ubiRMnmrlz56a63dftPu/GGLNr1y5Tp04d4+LiYh5++GEzZcqU294y7Ha3q7t69aoZPny4KVWqlHFxcTGFChUy9evXN5MnT7Z+NpcvX24ef/xx4+XlZX2v3r17m7///vs/xwwA7gWLMffgajMAADjIRx99pG7dumnHjh1pXiEeAADAnjinGwAAAAAAOyF0AwAAAABgJ4RuAAAAAADshHO6AQAAAACwE/Z0AwAAAABgJ4RuAAAAAADsJIejC7gfJSUl6fTp08qbN68sFoujywEAAAAA3GeMMbp69aqKFi0qJ6fb788mdKfh9OnT8vX1dXQZAAAAAID73MmTJ1WsWLHbvk7oTkPevHkl3Ro8Dw8PB1cDAAAAALjfXLlyRb6+vtb8eDuE7jQkH1Lu4eFB6AYAAAAA3NZ/nZLMhdQAAAAAALATQjcAAAAAAHZC6AYAAAAAwE4I3QAAAAAA2AmhGwAAAAAAOyF0AwAAAABgJ4RuAAAAAADshNANAAAAAICdELoBAAAAALATQjcAAAAAAHZC6AYAAAAAwE4I3QAAAAAA2AmhGwAAAAAAOyF0AwAAAABgJ4RuAAAAAADshNANAAAAAICdELoBAAAAALCTHI4uAHfnMad2ji4hy9iQtMzRJQAAAAB4wLCnGwAAAAAAOyF0AwAAAABgJ4RuAAAAAADshNANAAAAAICdELoBAAAAALATQjcAAAAAAHZC6AYAAAAAwE4I3QAAAAAA2AmhGwAAAAAAOyF0AwAAAABgJ4RuAAAAAADshNANAAAAAICdELoBAAAAALATh4fuGTNmyN/fX25ubqpTp462b99+274HDhxQ27Zt5e/vL4vFovDw8DT7nTp1Sp07d1bBggWVK1cuVa5cWTt37rTTEgAAAAAAkDaHhu4lS5YoNDRUY8aM0e7du1W1alU1a9ZMZ86cSbN/XFycSpQoobfeekve3t5p9rl48aIaNGignDlz6quvvtLBgwf1zjvvKH/+/PZcFAAAAAAAUrEYY4yj3rxOnTqqVauWpk+fLklKSkqSr6+v+vfvr2HDhv3rtP7+/ho4cKAGDhxo0z5s2DD9+OOP+v777zNc15UrV+Tp6anLly/Lw8Mjw/O5Fx5zaufoErKMDUnLHF0CAAAAgGziTnOjw/Z0JyQkaNeuXQoKCvpfMU5OCgoK0tatWzM83y+++EIBAQFq166dvLy8VL16dc2ePTszSgYAAAAAIF0cFrrPnTunxMREFS5c2Ka9cOHCioqKyvB8jx49qpkzZ6p06dL6+uuv1bdvX7388sv6+OOPbztNfHy8rly5YvMAAAAAAOBu5XB0AZktKSlJAQEBevPNNyVJ1atX1/79+xUREaHg4OA0pwkLC9O4cePuZZkAAAAAgAeAw/Z0FypUSM7OzoqOjrZpj46Ovu1F0u5EkSJFVKFCBZu28uXL68SJE7edZvjw4bp8+bL1cfLkyQy/PwAAAAAAyRwWul1cXFSzZk1FRkZa25KSkhQZGal69epleL4NGjTQ4cOHbdqOHDkiPz+/207j6uoqDw8PmwcAAAAAAHfLoYeXh4aGKjg4WAEBAapdu7bCw8MVGxurbt26SZK6dOkiHx8fhYWFSbp18bWDBw9a/3/q1Cnt3btXefLkUalSpSRJgwYNUv369fXmm2+qffv22r59u2bNmqVZs2Y5ZiEBAAAAAA8sh4buDh066OzZsxo9erSioqJUrVo1rVu3znpxtRMnTsjJ6X8740+fPq3q1atbn0+ePFmTJ09WYGCgNm3aJEmqVauWVq5cqeHDh2v8+PEqXry4wsPD1alTp3u6bAAAAAAAOPQ+3fcr7tOdPXGfbgAAAACZ5b6/TzcAAAAAANkdoRsAAAAAADshdAMAAAAAYCeEbgAAAAAA7ITQDQAAAACAnRC6AQAAAACwE0I3AAAAAAB2QugGAAAAAMBOCN0AAAAAANgJoRsAAAAAADshdAMAAAAAYCeEbgAAAAAA7ITQDQAAAACAnRC6AQAAAACwE0I3AAAAAAB2QugGAAAAAMBOCN0AAAAAANgJoRsAAAAAADshdAMAAAAAYCeEbgAAAAAA7ITQDQAAAACAnRC6AQAAAACwE0I3AAAAAAB2QugGAAAAAMBOCN0AAAAAANgJoRsAAAAAADshdAMAAAAAYCeEbgAAAAAA7ITQDQAAAACAnRC6AQAAAACwE0I3AAAAAAB2QugGAAAAAMBOCN0AAAAAANgJoRsAAAAAADshdAMAAAAAYCf3ReieMWOG/P395ebmpjp16mj79u237XvgwAG1bdtW/v7+slgsCg8P/9d5v/XWW7JYLBo4cGDmFg0AAAAAwH/I4egClixZotDQUEVERKhOnToKDw9Xs2bNdPjwYXl5eaXqHxcXpxIlSqhdu3YaNGjQv857x44d+uCDD1SlShV7lY8HVPMqrzm6hCxj3b43HF0CAAAA4DAO39M9ZcoU9ezZU926dVOFChUUERGh3Llza+7cuWn2r1WrliZNmqSOHTvK1dX1tvONiYlRp06dNHv2bOXPn99e5QMAAAAAcFsODd0JCQnatWuXgoKCrG1OTk4KCgrS1q1b72re/fr1U8uWLW3mDQAAAADAveTQw8vPnTunxMREFS5c2Ka9cOHCOnToUIbnu3jxYu3evVs7duy4o/7x8fGKj4+3Pr9y5UqG3xsAAAAAgGQOP7w8s508eVIDBgzQwoUL5ebmdkfThIWFydPT0/rw9fW1c5UAAAAAgAeBQ0N3oUKF5OzsrOjoaJv26OhoeXt7Z2ieu3bt0pkzZ1SjRg3lyJFDOXLk0ObNm/Xee+8pR44cSkxMTDXN8OHDdfnyZevj5MmTGXpvAAAAAABScmjodnFxUc2aNRUZGWltS0pKUmRkpOrVq5eheTZt2lS//PKL9u7da30EBASoU6dO2rt3r5ydnVNN4+rqKg8PD5sHAAAAAAB3y+G3DAsNDVVwcLACAgJUu3ZthYeHKzY2Vt26dZMkdenSRT4+PgoLC5N06+JrBw8etP7/1KlT2rt3r/LkyaNSpUopb968qlSpks17uLu7q2DBgqnaAQAAAACwJ4eH7g4dOujs2bMaPXq0oqKiVK1aNa1bt856cbUTJ07Iyel/O+RPnz6t6tWrW59PnjxZkydPVmBgoDZt2nSvywdwjzRsO9nRJWQpP6x41dElAAAAQPdB6JakkJAQhYSEpPnaP4O0v7+/jDHpmj9hHAAAAADgCNnu6uUAAAAAANwvCN0AAAAAANgJoRsAAAAAADshdAMAAAAAYCeEbgAAAAAA7OS+uHo5AOD+VPmVdx1dQpbyyzuDHF0CAAC4z7CnGwAAAAAAOyF0AwAAAABgJ4RuAAAAAADshNANAAAAAICdELoBAAAAALATQjcAAAAAAHZC6AYAAAAAwE4I3QAAAAAA2AmhGwAAAAAAOyF0AwAAAABgJ4RuAAAAAADshNANAAAAAICdELoBAAAAALATQjcAAAAAAHZC6AYAAAAAwE4I3QAAAAAA2AmhGwAAAAAAOyF0AwAAAABgJ4RuAAAAAADsJIejCwAAALZKvPeOo0vIUo6+/IqjSwAA4LbY0w0AAAAAgJ0QugEAAAAAsBNCNwAAAAAAdkLoBgAAAADATgjdAAAAAADYCaEbAAAAAAA7IXQDAAAAAGAnhG4AAAAAAOyE0A0AAAAAgJ3cF6F7xowZ8vf3l5ubm+rUqaPt27fftu+BAwfUtm1b+fv7y2KxKDw8PFWfsLAw1apVS3nz5pWXl5fatGmjw4cP23EJAAAAAABIzeGhe8mSJQoNDdWYMWO0e/duVa1aVc2aNdOZM2fS7B8XF6cSJUrorbfekre3d5p9Nm/erH79+umnn37Shg0bdOPGDT3++OOKjY2156IAAAAAAGAjh6MLmDJlinr27Klu3bpJkiIiIvTll19q7ty5GjZsWKr+tWrVUq1atSQpzdclad26dTbPP/roI3l5eWnXrl1q1KhRJi8BAAAAAABpc+ie7oSEBO3atUtBQUHWNicnJwUFBWnr1q2Z9j6XL1+WJBUoUCDT5gkAAAAAwH9x6J7uc+fOKTExUYULF7ZpL1y4sA4dOpQp75GUlKSBAweqQYMGqlSpUpp94uPjFR8fb31+5cqVTHlvAAAAAMCDzeHndNtbv379tH//fi1evPi2fcLCwuTp6Wl9+Pr63sMKAQAAAADZlUNDd6FCheTs7Kzo6Gib9ujo6NteJC09QkJCtGbNGn377bcqVqzYbfsNHz5cly9ftj5Onjx51+8NAAAAAIBDQ7eLi4tq1qypyMhIa1tSUpIiIyNVr169DM/XGKOQkBCtXLlSGzduVPHixf+1v6urqzw8PGweAAAAAADcLYdfvTw0NFTBwcEKCAhQ7dq1FR4ertjYWOvVzLt06SIfHx+FhYVJunXxtYMHD1r/f+rUKe3du1d58uRRqVKlJN06pHzRokX6/PPPlTdvXkVFRUmSPD09lStXLgcsJQAAAADgQeTw0N2hQwedPXtWo0ePVlRUlKpVq6Z169ZZL6524sQJOTn9b4f86dOnVb16devzyZMna/LkyQoMDNSmTZskSTNnzpQkNW7c2Oa95s2bp65du9p1eQAAAAAASObw0C3dOvc6JCQkzdeSg3Qyf39/GWP+dX7/9ToAAAAAAPdChs7pXrBggRo0aKCiRYvq+PHjkqTw8HB9/vnnmVocAAAAAABZWbpD98yZMxUaGqonnnhCly5dUmJioiQpX758Cg8Pz+z6AAAAAADIstIduqdNm6bZs2dr5MiRcnZ2trYHBATol19+ydTiAAAAAADIytIduo8dO2ZzIbNkrq6uio2NzZSiAAAAAADIDtIduosXL669e/emal+3bp3Kly+fGTUBAAAAAJAtpPvq5aGhoerXr5+uX78uY4y2b9+uTz/9VGFhYfrwww/tUSMAAAAAAFlSukP3iy++qFy5cum1115TXFycnn/+eRUtWlRTp05Vx44d7VEjAAAAAABZUobu092pUyd16tRJcXFxiomJkZeXV2bXBQAAAABAlpfu0H3s2DHdvHlTpUuXVu7cuZU7d25J0m+//aacOXPK398/s2sEAAAAACBLSveF1Lp27aotW7akat+2bZu6du2aGTUBAAAAAJAtpDt079mzRw0aNEjVXrdu3TSvag4AAAAAwIMq3aHbYrHo6tWrqdovX76sxMTETCkKAAAAAIDsIN2hu1GjRgoLC7MJ2ImJiQoLC1PDhg0ztTgAAAAAALKydF9IbeLEiWrUqJHKli2rRx55RJL0/fff68qVK9q4cWOmFwgAAAAAQFaV7j3dFSpU0L59+9S+fXudOXNGV69eVZcuXXTo0CFVqlTJHjUCAAAAAJAlZeg+3UWLFtWbb76Z2bUAAAAAAJCtZCh0X7p0Sdu3b9eZM2eUlJRk81qXLl0ypTAAAAAAALK6dIfu1atXq1OnToqJiZGHh4csFov1NYvFQugGAAAAAOD/pfuc7ldeeUXdu3dXTEyMLl26pIsXL1ofFy5csEeNAAAAAABkSekO3adOndLLL7+s3Llz26MeAAAAAACyjXSH7mbNmmnnzp32qAUAAAAAgGwl3ed0t2zZUoMHD9bBgwdVuXJl5cyZ0+b1Vq1aZVpxAAAAAABkZekO3T179pQkjR8/PtVrFotFiYmJd18VAAAAAADZQLpD9z9vEQYAAAAAANKW7nO6AQAAAADAnUn3nm5Jio2N1ebNm3XixAklJCTYvPbyyy9nSmEAAAAAAGR16Q7de/bs0RNPPKG4uDjFxsaqQIECOnfunHLnzi0vLy9CNwAAAAAA/y/dh5cPGjRITz31lC5evKhcuXLpp59+0vHjx1WzZk1NnjzZHjUCAAAAAJAlpTt07927V6+88oqcnJzk7Oys+Ph4+fr66u2339aIESPsUSMAAAAAAFlSukN3zpw55eR0azIvLy+dOHFCkuTp6amTJ09mbnUAAAAAAGRh6T6nu3r16tqxY4dKly6twMBAjR49WufOndOCBQtUqVIle9QIAAAAAECWlO493W+++aaKFCkiSZowYYLy58+vvn376uzZs/rggw8yvUAAAAAAALKqdO/pDggIsP7fy8tL69aty9SCAAAAAADILtK9p7tJkya6dOlSqvYrV66oSZMmmVETAAAAAADZQrpD96ZNm5SQkJCq/fr16/r+++8zpSgAAAAAALKDOw7d+/bt0759+yRJBw8etD7ft2+f9uzZozlz5sjHxydDRcyYMUP+/v5yc3NTnTp1tH379tv2PXDggNq2bSt/f39ZLBaFh4ff9TwBAAAAALCHOz6nu1q1arJYLLJYLGkeRp4rVy5NmzYt3QUsWbJEoaGhioiIUJ06dRQeHq5mzZrp8OHD8vLyStU/Li5OJUqUULt27TRo0KBMmScAAAAAAPZwx3u6jx07pj/++EPGGG3fvl3Hjh2zPk6dOqUrV66oe/fu6S5gypQp6tmzp7p166YKFSooIiJCuXPn1ty5c9PsX6tWLU2aNEkdO3aUq6trpswTAAAAAAB7uOM93X5+frpx44aCg4NVsGBB+fn53fWbJyQkaNeuXRo+fLi1zcnJSUFBQdq6det9M08AAAAAADIiXRdSy5kzp1auXJlpb37u3DklJiaqcOHCNu2FCxdWVFTUPZtnfHy8rly5YvMAAAAAAOBupfvq5a1bt9aqVavsUIrjhIWFydPT0/rw9fV1dEkAAAAAgGzgjg8vT1a6dGmNHz9eP/74o2rWrCl3d3eb119++eU7nlehQoXk7Oys6Ohom/bo6Gh5e3unt7QMz3P48OEKDQ21Pr9y5QrBGwAAAABw19IduufMmaN8+fJp165d2rVrl81rFoslXaHbxcVFNWvWVGRkpNq0aSNJSkpKUmRkpEJCQtJbWobn6erqetuLsgEAgAdD+ZXjHF1ClvLr02McXQIAZAnpDt3Hjh3L1AJCQ0MVHBysgIAA1a5dW+Hh4YqNjVW3bt0kSV26dJGPj4/CwsIk3bpQ2sGDB63/P3XqlPbu3as8efKoVKlSdzRPAAAAAADuhXSH7pSMMZJu7eHOqA4dOujs2bMaPXq0oqKiVK1aNa1bt856IbQTJ07Iyel/p56fPn1a1atXtz6fPHmyJk+erMDAQG3atOmO5gkAAAAAwL2QodA9f/58TZo0Sb/99pskqUyZMho8eLBeeOGFDBUREhJy20O/k4N0Mn9/f2vYz+g8AQAAAAC4F9IduqdMmaJRo0YpJCREDRo0kCT98MMP6tOnj86dO6dBgwZlepEAAAAAAGRF6Q7d06ZN08yZM9WlSxdrW6tWrVSxYkWNHTuW0A0AAAAAwP9L9326//77b9WvXz9Ve/369fX3339nSlEAAAAAAGQH6Q7dpUqV0tKlS1O1L1myRKVLl86UogAAAAAAyA7SfXj5uHHj1KFDB3333XfWc7p//PFHRUZGphnGAQAAAAB4UKV7T3fbtm21bds2FSpUSKtWrdKqVatUqFAhbd++XU8//bQ9agQAAAAAIEvK0C3DatasqU8++SSzawEAAAAAIFvJUOhOTEzUypUr9euvv0qSKlSooNatWytHjgzNDgAAAACAbCndKfnAgQNq1aqVoqKiVLZsWUnSxIkT9dBDD2n16tWqVKlSphcJAAAAAEBWlO5zul988UVVrFhRf/31l3bv3q3du3fr5MmTqlKlinr16mWPGgEAAAAAyJLSvad779692rlzp/Lnz29ty58/vyZMmKBatWplanEAAAAAAGRl6d7TXaZMGUVHR6dqP3PmjEqVKpUpRQEAAAAAkB2kO3SHhYXp5Zdf1vLly/XXX3/pr7/+0vLlyzVw4EBNnDhRV65csT4AAAAAAHiQpfvw8ieffFKS1L59e1ksFkmSMUaS9NRTT1mfWywWJSYmZladAAAAAABkOekO3d9++6096gAAAAAAINtJd+gODAy0Rx0AAAAAAGQ76Q7dknT9+nXt27dPZ86cUVJSks1rrVq1ypTCAAAAAADI6tIdutetW6cuXbro3LlzqV7jPG4AAAAAAP4n3Vcv79+/v9q1a6e///5bSUlJNg8CNwAAAAAA/5Pu0B0dHa3Q0FAVLlzYHvUAAAAAAJBtpDt0P/vss9q0aZMdSgEAAAAAIHtJ9znd06dPV7t27fT999+rcuXKypkzp83rL7/8cqYVBwAAAABAVpbu0P3pp59q/fr1cnNz06ZNm2SxWKyvWSwWQjcAAAAAAP8v3aF75MiRGjdunIYNGyYnp3QfnQ4AAAAAwAMj3ak5ISFBHTp0IHADAAAAAPAf0p2cg4ODtWTJEnvUAgAAAABAtpLuw8sTExP19ttv6+uvv1aVKlVSXUhtypQpmVYcAAAAAABZWbpD9y+//KLq1atLkvbv32/zWsqLqgEAAAAA8KBLd+j+9ttv7VEHAAAAAADZDldDAwAAAADATu54T/czzzxzR/0+++yzDBcDAAAAAEB2cseh29PT0551AAAAAACQ7dxx6J43b5496wAAAAAAINvhnG4AAAAAAOyE0A0AAAAAgJ3cF6F7xowZ8vf3l5ubm+rUqaPt27f/a/9ly5apXLlycnNzU+XKlbV27Vqb12NiYhQSEqJixYopV65cqlChgiIiIuy5CAAAAAAApOLw0L1kyRKFhoZqzJgx2r17t6pWrapmzZrpzJkzafbfsmWLnnvuOfXo0UN79uxRmzZt1KZNG+3fv9/aJzQ0VOvWrdMnn3yiX3/9VQMHDlRISIi++OKLe7VYAAAAAAA4PnRPmTJFPXv2VLdu3ax7pHPnzq25c+em2X/q1Klq3ry5Bg8erPLly+v1119XjRo1NH36dGufLVu2KDg4WI0bN5a/v7969eqlqlWr/ucedAAAAAAAMlOGQveCBQvUoEEDFS1aVMePH5ckhYeH6/PPP0/XfBISErRr1y4FBQX9ryAnJwUFBWnr1q1pTrN161ab/pLUrFkzm/7169fXF198oVOnTskYo2+//VZHjhzR448/nq76AAAAAAC4G+kO3TNnzlRoaKieeOIJXbp0SYmJiZKkfPnyKTw8PF3zOnfunBITE1W4cGGb9sKFCysqKirNaaKiov6z/7Rp01ShQgUVK1ZMLi4uat68uWbMmKFGjRqlOc/4+HhduXLF5gEAAAAAwN1Kd+ieNm2aZs+erZEjR8rZ2dnaHhAQoF9++SVTi8uoadOm6aefftIXX3yhXbt26Z133lG/fv30zTffpNk/LCxMnp6e1oevr+89rhgAAAAAkB3lSO8Ex44dU/Xq1VO1u7q6KjY2Nl3zKlSokJydnRUdHW3THh0dLW9v7zSn8fb2/tf+165d04gRI7Ry5Uq1bNlSklSlShXt3btXkydPTnVouiQNHz5coaGh1udXrlwheAMAAAAA7lq693QXL15ce/fuTdW+bt06lS9fPl3zcnFxUc2aNRUZGWltS0pKUmRkpOrVq5fmNPXq1bPpL0kbNmyw9r9x44Zu3LghJyfbRXN2dlZSUlKa83R1dZWHh4fNAwAAAACAu5XuPd2hoaHq16+frl+/LmOMtm/frk8//VRhYWH68MMP011AaGiogoODFRAQoNq1ays8PFyxsbHq1q2bJKlLly7y8fFRWFiYJGnAgAEKDAzUO++8o5YtW2rx4sXauXOnZs2aJUny8PBQYGCgBg8erFy5csnPz0+bN2/W/PnzNWXKlHTXBwAAAABARqU7dL/44ovKlSuXXnvtNcXFxen5559X0aJFNXXqVHXs2DHdBXTo0EFnz57V6NGjFRUVpWrVqmndunXWi6WdOHHCZq91/fr1tWjRIr322msaMWKESpcurVWrVqlSpUrWPosXL9bw4cPVqVMnXbhwQX5+fpowYYL69OmT7voAAAAAAMiodIXumzdvatGiRWrWrJk6deqkuLg4xcTEyMvL666KCAkJUUhISJqvbdq0KVVbu3bt1K5du9vOz9vbW/PmzburmgAAAAAAuFvpOqc7R44c6tOnj65fvy5Jyp07910HbgAAAAAAsqt0X0itdu3a2rNnjz1qAQAAAAAgW0n3Od0vvfSSXnnlFf3111+qWbOm3N3dbV6vUqVKphUHAAAAAEBWlu7QnXyxtJdfftnaZrFYZIyRxWJRYmJi5lUHAAAAAEAWlu7QfezYMXvUAQAAAABAtpPu0O3n52ePOgAAAAAAyHbSHbrnz5//r6936dIlw8UAAAAAAJCdpDt0DxgwwOb5jRs3FBcXJxcXF+XOnZvQDQAAAADA/0v3LcMuXrxo84iJidHhw4fVsGFDffrpp/aoEQAAAACALCnde7rTUrp0ab311lvq3LmzDh06lBmzBAAAwAMgeHsPR5eQpXxce46jSwCQTune0307OXLk0OnTpzNrdgAAAAAAZHnp3tP9xRdf2Dw3xujvv//W9OnT1aBBg0wrDAAAAACArC7dobtNmzY2zy0Wix566CE1adJE77zzTmbVBQAAAABAlpfu0J2UlGSPOgAAAAAAyHbSfU73+PHjFRcXl6r92rVrGj9+fKYUBQAAAABAdpDu0D1u3DjFxMSkao+Li9O4ceMypSgAAAAAALKDdIduY4wsFkuq9p9//lkFChTIlKIAAAAAAMgO7vic7vz588tischisahMmTI2wTsxMVExMTHq06ePXYoEAAAAACAruuPQHR4eLmOMunfvrnHjxsnT09P6mouLi/z9/VWvXj27FAkAAAAAQFZ0x6E7ODhYklS8eHHVr19fOXPmtFtRAAAAAABkB+m+ZVhgYKD1/9evX1dCQoLN6x4eHndfFQAAAAAA2UC6L6QWFxenkJAQeXl5yd3dXfnz57d5AAAAAACAW9IdugcPHqyNGzdq5syZcnV11Ycffqhx48apaNGimj9/vj1qBAAAAAAgS0r34eWrV6/W/Pnz1bhxY3Xr1k2PPPKISpUqJT8/Py1cuFCdOnWyR50AAAAAAGQ56d7TfeHCBZUoUULSrfO3L1y4IElq2LChvvvuu8ytDgAAAACALCzdobtEiRI6duyYJKlcuXJaunSppFt7wPPly5epxQEAAAAAkJWlO3R369ZNP//8syRp2LBhmjFjhtzc3DRo0CANHjw40wsEAAAAACCrSvc53YMGDbL+PygoSIcOHdKuXbtUqlQpValSJVOLAwAAAAAgK0t36E7p+vXr8vPzk5+fX2bVAwAAAABAtpHuw8sTExP1+uuvy8fHR3ny5NHRo0clSaNGjdKcOXMyvUAAAAAAALKqdIfuCRMm6KOPPtLbb78tFxcXa3ulSpX04YcfZmpxAAAAAABkZekO3fPnz9esWbPUqVMnOTs7W9urVq2qQ4cOZWpxAAAAAABkZekO3adOnVKpUqVStSclJenGjRuZUhQAAAAAANlBukN3hQoV9P3336dqX758uapXr54pRQEAAAAAkB2k++rlo0ePVnBwsE6dOqWkpCR99tlnOnz4sObPn681a9bYo0YAAAAAmeizP9hZlh7PlNzj6BKQhaV7T3fr1q21evVqffPNN3J3d9fo0aP166+/avXq1XrssccyVMSMGTPk7+8vNzc31alTR9u3b//X/suWLVO5cuXk5uamypUra+3atan6/Prrr2rVqpU8PT3l7u6uWrVq6cSJExmqDwAAAACAjLjj0H306FEZYyRJjzzyiDZs2KAzZ84oLi5OP/zwgx5//PEMFbBkyRKFhoZqzJgx2r17t6pWrapmzZrpzJkzafbfsmWLnnvuOfXo0UN79uxRmzZt1KZNG+3fv9/a548//lDDhg1Vrlw5bdq0Sfv27dOoUaPk5uaWoRoBAAAAAMiIOw7dpUuX1tmzZ63PO3TooOjo6LsuYMqUKerZs6e6deumChUqKCIiQrlz59bcuXPT7D916lQ1b95cgwcPVvny5fX666+rRo0amj59urXPyJEj9cQTT+jtt99W9erVVbJkSbVq1UpeXl53XS8AAAAAAHfqjkN38l7uZGvXrlVsbOxdvXlCQoJ27dqloKCg/xXk5KSgoCBt3bo1zWm2bt1q01+SmjVrZu2flJSkL7/8UmXKlFGzZs3k5eWlOnXqaNWqVXdVKwAAAAAA6ZXuc7oz07lz55SYmKjChQvbtBcuXFhRUVFpThMVFfWv/c+cOaOYmBi99dZbat68udavX6+nn35azzzzjDZv3pzmPOPj43XlyhWbBwAAAAAAd+uOr15usVhksVhStd1vkpKSJN264NugQYMkSdWqVdOWLVsUERGhwMDAVNOEhYVp3Lhx97ROAAAAAED2d8eh2xijrl27ytXVVZJ0/fp19enTR+7u7jb9Pvvsszt+80KFCsnZ2TnVueHR0dHy9vZOcxpvb+9/7V+oUCHlyJFDFSpUsOlTvnx5/fDDD2nOc/jw4QoNDbU+v3Llinx9fe94OQAAAAAASMsdH14eHBwsLy8veXp6ytPTU507d1bRokWtz5Mf6eHi4qKaNWsqMjLS2paUlKTIyEjVq1cvzWnq1atn01+SNmzYYO3v4uKiWrVq6fDhwzZ9jhw5Ij8/vzTn6erqKg8PD5sHAAAAAAB36473dM+bN88uBYSGhio4OFgBAQGqXbu2wsPDFRsbq27dukmSunTpIh8fH4WFhUmSBgwYoMDAQL3zzjtq2bKlFi9erJ07d2rWrFnWeQ4ePFgdOnRQo0aN9Oijj2rdunVavXq1Nm3aZJdlAAAAAAAgLXccuu2lQ4cOOnv2rEaPHq2oqChVq1ZN69ats14s7cSJE3Jy+t8O+fr162vRokV67bXXNGLECJUuXVqrVq1SpUqVrH2efvppRUREKCwsTC+//LLKli2rFStWqGHDhvd8+QAAAAAADy6Hh25JCgkJUUhISJqvpbV3ul27dmrXrt2/zrN79+7q3r17ZpQHAAAAAECGOPSWYQAAAAAAZGeEbgAAAAAA7ITQDQAAAACAndwX53QDAAAAQHaXFFXG0SVkKU7eRxxdQqZgTzcAAAAAAHZC6AYAAAAAwE4I3QAAAAAA2AmhGwAAAAAAOyF0AwAAAABgJ4RuAAAAAADshNANAAAAAICdELoBAAAAALATQjcAAAAAAHZC6AYAAAAAwE4I3QAAAAAA2AmhGwAAAAAAOyF0AwAAAABgJ4RuAAAAAADshNANAAAAAICdELoBAAAAALATQjcAAAAAAHZC6AYAAAAAwE4I3QAAAAAA2AmhGwAAAAAAOyF0AwAAAABgJ4RuAAAAAADshNANAAAAAICdELoBAAAAALATQjcAAAAAAHZC6AYAAAAAwE4I3QAAAAAA2AmhGwAAAAAAOyF0AwAAAABgJ4RuAAAAAADshNANAAAAAICd3Behe8aMGfL395ebm5vq1Kmj7du3/2v/ZcuWqVy5cnJzc1PlypW1du3a2/bt06ePLBaLwsPDM7lqAAAAAAD+ncND95IlSxQaGqoxY8Zo9+7dqlq1qpo1a6YzZ86k2X/Lli167rnn1KNHD+3Zs0dt2rRRmzZttH///lR9V65cqZ9++klFixa192IAAAAAAJCKw0P3lClT1LNnT3Xr1k0VKlRQRESEcufOrblz56bZf+rUqWrevLkGDx6s8uXL6/XXX1eNGjU0ffp0m36nTp1S//79tXDhQuXMmfNeLAoAAAAAADYcGroTEhK0a9cuBQUFWducnJwUFBSkrVu3pjnN1q1bbfpLUrNmzWz6JyUl6YUXXtDgwYNVsWJF+xQPAAAAAMB/yOHINz937pwSExNVuHBhm/bChQvr0KFDaU4TFRWVZv+oqCjr84kTJypHjhx6+eWX76iO+Ph4xcfHW59fuXLlThcBAAAAAIDbcvjh5Zlt165dmjp1qj766CNZLJY7miYsLEyenp7Wh6+vr52rBAAAAAA8CBwaugsVKiRnZ2dFR0fbtEdHR8vb2zvNaby9vf+1//fff68zZ87o4YcfVo4cOZQjRw4dP35cr7zyivz9/dOc5/Dhw3X58mXr4+TJk3e/cAAAAACAB55DQ7eLi4tq1qypyMhIa1tSUpIiIyNVr169NKepV6+eTX9J2rBhg7X/Cy+8oH379mnv3r3WR9GiRTV48GB9/fXXac7T1dVVHh4eNg8AAAAAAO6WQ8/plqTQ0FAFBwcrICBAtWvXVnh4uGJjY9WtWzdJUpcuXeTj46OwsDBJ0oABAxQYGKh33nlHLVu21OLFi7Vz507NmjVLklSwYEEVLFjQ5j1y5swpb29vlS1b9t4uHAAAAADggebw0N2hQwedPXtWo0ePVlRUlKpVq6Z169ZZL5Z24sQJOTn9b4d8/fr1tWjRIr322msaMWKESpcurVWrVqlSpUqOWgQAAAAAANLk8NAtSSEhIQoJCUnztU2bNqVqa9eundq1a3fH8//zzz8zWBkAAAAAABmX7a5eDgAAAADA/YLQDQAAAACAnRC6AQAAAACwE0I3AAAAAAB2QugGAAAAAMBOCN0AAAAAANgJoRsAAAAAADshdAMAAAAAYCeEbgAAAAAA7ITQDQAAAACAnRC6AQAAAACwE0I3AAAAAAB2QugGAAAAAMBOCN0AAAAAANgJoRsAAAAAADshdAMAAAAAYCeEbgAAAAAA7ITQDQAAAACAnRC6AQAAAACwE0I3AAAAAAB2QugGAAAAAMBOCN0AAAAAANgJoRsAAAAAADshdAMAAAAAYCeEbgAAAAAA7ITQDQAAAACAnRC6AQAAAACwE0I3AAAAAAB2QugGAAAAAMBOCN0AAAAAANgJoRsAAAAAADshdAMAAAAAYCeEbgAAAAAA7ITQDQAAAACAnRC6AQAAAACwk/sidM+YMUP+/v5yc3NTnTp1tH379n/tv2zZMpUrV05ubm6qXLmy1q5da33txo0bGjp0qCpXrix3d3cVLVpUXbp00enTp+29GAAAAAAA2HB46F6yZIlCQ0M1ZswY7d69W1WrVlWzZs105syZNPtv2bJFzz33nHr06KE9e/aoTZs2atOmjfbv3y9JiouL0+7duzVq1Cjt3r1bn332mQ4fPqxWrVrdy8UCAAAAAMDxoXvKlCnq2bOnunXrpgoVKigiIkK5c+fW3Llz0+w/depUNW/eXIMHD1b58uX1+uuvq0aNGpo+fbokydPTUxs2bFD79u1VtmxZ1a1bV9OnT9euXbt04sSJe7loAAAAAIAHnENDd0JCgnbt2qWgoCBrm5OTk4KCgrR169Y0p9m6datNf0lq1qzZbftL0uXLl2WxWJQvX75MqRsAAAAAgDuRw5Fvfu7cOSUmJqpw4cI27YULF9ahQ4fSnCYqKirN/lFRUWn2v379uoYOHarnnntOHh4eafaJj49XfHy89fmVK1fSsxgAAAAAAKTJ4YeX29ONGzfUvn17GWM0c+bM2/YLCwuTp6en9eHr63sPqwQAAAAAZFcODd2FChWSs7OzoqOjbdqjo6Pl7e2d5jTe3t531D85cB8/flwbNmy47V5uSRo+fLguX75sfZw8eTKDSwQAAAAAwP84NHS7uLioZs2aioyMtLYlJSUpMjJS9erVS3OaevXq2fSXpA0bNtj0Tw7cv/32m7755hsVLFjwX+twdXWVh4eHzQMAAAAAgLvl0HO6JSk0NFTBwcEKCAhQ7dq1FR4ertjYWHXr1k2S1KVLF/n4+CgsLEySNGDAAAUGBuqdd95Ry5YttXjxYu3cuVOzZs2SdCtwP/vss9q9e7fWrFmjxMRE6/neBQoUkIuLi2MWFAAAAADwwHF46O7QoYPOnj2r0aNHKyoqStWqVdO6deusF0s7ceKEnJz+t0O+fv36WrRokV577TWNGDFCpUuX1qpVq1SpUiVJ0qlTp/TFF19IkqpVq2bzXt9++60aN258T5YLAAAAAACHh25JCgkJUUhISJqvbdq0KVVbu3bt1K5duzT7+/v7yxiTmeUBAAAAAJAh2frq5QAAAAAAOBKhGwAAAAAAOyF0AwAAAABgJ4RuAAAAAADshNANAAAAAICdELoBAAAAALATQjcAAAAAAHZC6AYAAAAAwE4I3QAAAAAA2AmhGwAAAAAAOyF0AwAAAABgJ4RuAAAAAADshNANAAAAAICdELoBAAAAALATQjcAAAAAAHZC6AYAAAAAwE4I3QAAAAAA2AmhGwAAAAAAOyF0AwAAAABgJ4RuAAAAAADshNANAAAAAICdELoBAAAAALATQjcAAAAAAHZC6AYAAAAAwE4I3QAAAAAA2AmhGwAAAAAAOyF0AwAAAABgJ4RuAAAAAADshNANAAAAAICdELoBAAAAALATQjcAAAAAAHZC6AYAAAAAwE4I3QAAAAAA2AmhGwAAAAAAOyF0AwAAAABgJ/dF6J4xY4b8/f3l5uamOnXqaPv27f/af9myZSpXrpzc3NxUuXJlrV271uZ1Y4xGjx6tIkWKKFeuXAoKCtJvv/1mz0UAAAAAACAVh4fuJUuWKDQ0VGPGjNHu3btVtWpVNWvWTGfOnEmz/5YtW/Tcc8+pR48e2rNnj9q0aaM2bdpo//791j5vv/223nvvPUVERGjbtm1yd3dXs2bNdP369Xu1WAAAAAAAOD50T5kyRT179lS3bt1UoUIFRUREKHfu3Jo7d26a/adOnarmzZtr8ODBKl++vF5//XXVqFFD06dPl3RrL3d4eLhee+01tW7dWlWqVNH8+fN1+vRprVq16h4uGQAAAADgQefQ0J2QkKBdu3YpKCjI2ubk5KSgoCBt3bo1zWm2bt1q01+SmjVrZu1/7NgxRUVF2fTx9PRUnTp1bjtPAAAAAADsIYcj3/zcuXNKTExU4cKFbdoLFy6sQ4cOpTlNVFRUmv2joqKsrye33a7PP8XHxys+Pt76/PLly5KkK1eupGNpHOOmueHoErKMzPx53kyM/+9OkJR5437zBqeHpEdmjXtiPOOeHpk17kmcDpUumfZ5j2Pc0yOzxj0hJiFT5vOgyKxxj7uamCnzeVBk2vqdcU8Xp9z3dx5L/lwYY/61n0ND9/0iLCxM48aNS9Xu6+vrgGpgL56eno4u4YHk6TnZ0SU8kDw9Rzm6hAeS54wRji7hgeQ59DVHl/BA8tRbji7hgbRYnzi6hAcU25GOkTXG/erVq/+aNRwaugsVKiRnZ2dFR0fbtEdHR8vb2zvNaby9vf+1f/K/0dHRKlKkiE2fatWqpTnP4cOHKzQ01Po8KSlJFy5cUMGCBWWxWNK9XA+yK1euyNfXVydPnpSHh4ejy3lgMO6Owbg7BuPuGIy7YzDujsG4Owbj7hiMe8YZY3T16lUVLVr0X/s5NHS7uLioZs2aioyMVJs2bSTdCryRkZEKCQlJc5p69eopMjJSAwcOtLZt2LBB9erVkyQVL15c3t7eioyMtIbsK1euaNu2berbt2+a83R1dZWrq6tNW758+e5q2R50Hh4e/NI6AOPuGIy7YzDujsG4Owbj7hiMu2Mw7o7BuGfMnRxN6/DDy0NDQxUcHKyAgADVrl1b4eHhio2NVbdu3SRJXbp0kY+Pj8LCwiRJAwYMUGBgoN555x21bNlSixcv1s6dOzVr1ixJksVi0cCBA/XGG2+odOnSKl68uEaNGqWiRYtagz0AAAAAAPeCw0N3hw4ddPbsWY0ePVpRUVGqVq2a1q1bZ70Q2okTJ+Tk9L+LrNevX1+LFi3Sa6+9phEjRqh06dJatWqVKlWqZO0zZMgQxcbGqlevXrp06ZIaNmyodevWyc3N7Z4vHwAAAADgweXw0C1JISEhtz2cfNOmTana2rVrp3bt2t12fhaLRePHj9f48eMzq0TcIVdXV40ZMybV4fqwL8bdMRh3x2DcHYNxdwzG3TEYd8dg3B2Dcbc/i/mv65sDAAAAAIAMcfrvLgAAAAAAICMI3QAAAAAA2AmhGwAAAAAAOyF0AwAAAABgJ4Ru/CeutQcAQObj7ysAPBgI3fhPFovF5jkbCfcG4wzgXkhKSnJ0CQ+U/v37a8qUKZJu/X1lXX9v8XnHg4J1y/3lvrhPN+5f69ev1+bNmxUXF6e6deuqQ4cOqUI4Ml9SUpKcnG59J3b9+nW5ubk5uKIHR8qxj4uLU+7cuR1cUfb3008/6ejRo4qJiVGjRo1Urlw5R5f0QDh69Kgefvhh5ciRQ8YY1u33wN9//62YmBjNnj1befLkUa9evazBm/G3v5TrdyA7S/lZv3r1qvLmzevgisCaB7c1Z84cdezYUSdPntR3332n9957T1u2bHF0WdleyhXlW2+9pZdeekkXLlxwcFUPhpRjP2PGDE2aNEl///23g6vK3ubNm6c2bdpo1qxZeuWVV9S/f399//33ji4r21u0aJEef/xxTZ8+XTdu3JDFYmEP4D1QpEgRjRgxQs2bN9eUKVP0wQcfSGKP972wefNmfffdd5Kknj17KjQ01MEVPXh++OEHffvtt1q/fr2jS8nWUm7LTJw4URMmTNDvv//u4KrAnm6kadmyZRoyZIg+/PBDPfPMMzp9+rTq1aunuLg4m358a5z5ksdzyJAhWrhwoUaMGKGYmBgVKFDAwZVlfynHfv78+Zo0aZISEhIcXFX29fnnn2vw4MH64IMP1KpVK0VHR6tFixb64osv9Mgjjzi6vGzrm2++0bBhw+Tu7q6VK1cqR44c6t27t3LmzMk6/R4oXbq0XnrpJRlj9O6770qSevfuzR5vOzHGKCYmRn369FGxYsUUERGhdevWWQM47o3hw4dr2bJlyps3r06dOqVHHnlEYWFhKlOmjKNLy3ZSbsssWLBAEyZMSLWnm3XNvUfoRioXLlzQl19+qcGDB+uZZ56RJBUtWlR+fn765JNPtHDhQvn5+Wns2LFycnLiF9cOPvvsM82fP19r1qxRQECAJOnatWu6fPmy8uXLx+HmdjRv3jwtXLhQX375pWrWrGltP3PmjLy8vBxYWfZy/vx5LVu2TC+99JLatm2rmzdvqlixYurVq5cWLFigGzduKGfOnI4uM9u5ceOGfvjhBzVp0kTDhg1TWFiYFi5cKEkEbztK/juZPLalS5dW3759JYngbWcWi0V58+bV1q1bVbFiRUVHR+uDDz5QlSpVJBE+7oX33ntPc+bM0dq1axUQEKBp06ZpwIABeuWVVwjddrJ48WLNnz9f69evt37WY2JidOnSJXl5ecnFxYXP/j3GX1WkkidPHvXt21cdO3a0trVu3Vq//fabfHx8lCtXLs2fP9+6wcAvbOY7ceKEqlevroCAAP3888+aNGmSqlWrppo1ayosLEwxMTGOLjHb+vXXX1WvXj3VrFlThw8f1gcffKAaNWqobt261nCCu5cjRw4VKFBAgYGB1ueSlD9/fp05c0Y3btxwZHnZVs6cOdWtWzd1795d5cqV04wZM1S6dGktXLhQH3zwgW7cuJEqcHPY891JSkqy/p2Mj4+3HsZftmxZ9erVS82bN9e7777LoeZ2lJCQoDNnzsjb21slS5bUihUrtHHjRklKdWoFY5/5Dhw4oKFDhyogIEBLly7V6NGj9f7776t+/fq6fv26o8vLls6fP6/atWurSpUqOnTokN555x1VrVpVbdq00cCBA3X9+nW23+8x9nQjFRcXFwUEBMjZ2VnSrXNwjh8/ru+++06lS5eWJI0aNUpr1qzR+fPnVbBgQUeWm+Ulf9OY8hvHhx9+WF9//bV69uypyMhI1a1bVwMHDtTFixf11ltvKTg4WHny5HFw5VlfWt/yFilSRKtXr1afPn20fft2lSpVSs2aNVNiYqJ69uypRx99VEWLFnVQxdmHp6enxowZY11/JO8BfPjhh1WwYEGboznWr1+vxx9/3FGlZjt+fn7y8/OTdOtL1hkzZqhfv35auHChLBaL+vTpo7i4OM2ePVuhoaFsmN2FlEcNvPfee4qMjFRiYqKqVKmiCRMmqEKFCurVq5ckKTw8XE5OTurZsydjnglSjr2Li4vKlCmjXbt26fTp02rRooXeeustSVKTJk1svmhi7DNXQkKCtm3bpho1amjr1q3q0aOHJk2apD59+ujmzZsaPXq06tWrp6efftrRpWZZaW3LJCUl6euvv1afPn30zTffKCAgQH369NG1a9f06aef6q+//lKpUqUcVPGDidCNNCUHbklq2LChtmzZoty5c1v/iOXPn1/58+fnaoh3KeVGwYkTJ6yHwT3zzDOaM2eOPvvsM40cOVJBQUHy8/PT6dOntWrVqlTn1iP9Uo79hQsX5O7uLldXVz311FO6cOGCvvnmG3Xv3l1BQUEqV66cvvnmG23btk2urq4Orjz7SA7cxhjrzyIxMVHXrl2zPm/RooVy5sypxx57jI1hO0hKSlLevHk1ffp0hYSEaNGiRbp69arWrFmjM2fOaODAgRxqfheSx27YsGH66KOPNGDAACUkJGjRokX69ddf9dlnn1mDt5OTk4YMGSIvLy+1bt3awZVnbSnXKYsXL9bvv/+uSpUqqWbNmvL19dWKFSvUtm1bTZ48WTdu3FBQUJCaNGmixo0ba9y4cQ6uPnu4ePGi8ufPLxcXF3Xr1k2zZs3S/v37FRERoW7dukm6dbjzvn37lD9/fgdXm3Wl3JaJjo7W1atXVapUKfXv3183b97U9u3bNXz4cDVt2lT+/v46dOiQPvvsM44mcwQD/IukpCSbf40xJi4uzjzxxBMmJCTEUWVlCynHdOzYsaZq1aqmbNmyxs/Pz3z66afGGGMSExOt/167ds00b97cNG7c2NqOuzdu3DgTGBhoKlSoYD7++GNz6dIlY4wxsbGxxphbP6f4+HjTsmVL07JlS5ufGzLfqlWrTLFixczly5fNk08+aUqWLGkSEhKMMYaxt5ObN28aY4yJiYkxHTp0MBaLxVSvXt067qxv7s6SJUtMuXLlzE8//WSMMeazzz4z7u7upmDBgqZx48bWz/XevXtNeHi49eeBjEm5nhg8eLB56KGHTMWKFU2ZMmXMs88+aw4cOGCMMea3334ztWrVMpUrVzZly5Y1lSpVMvHx8Y4qO1tZsGCB8fPzM4cOHTLGGPPDDz+YBg0amLp165odO3YYY4w5efKkeeKJJ0zdunX5zGdQys/6mDFjTM2aNU3hwoVNgwYNzIcffmgSEhKs6++kpCRz7do188QTT5igoCDW6w5A6MYdS0hIMEePHjVPPPGEqV69urlx44Yxhg3huzV+/Hjz0EMPmbVr15pLly6Z5s2bm4ceesgcOXLEGGPMtWvXzOzZs02jRo1MjRo12BC+SynHbebMmaZgwYImPDzcdOjQwRQrVswMHTrUHD161BhjzJUrV8yyZctM06ZNTdWqVRn7e+Dbb781VapUMY0bNzalS5e2jnny+gb2kZSUZGJjY039+vVNnTp1rOPNuN+9Tz/91IwYMcIYY8zq1atNgQIFzHvvvWeWLFliXF1dTevWrVOtUwghd2/fvn3m2WefNbt27TJJSUlmwYIFJigoyDRr1szs37/fGGPM8ePHzZw5c8z06dP5zGeSFStWmOnTpxuLxWIeeeQR8/vvv1vbAwMDjbe3tylXrpypXr26qVOnjnUdz2c+415//XXj5eVlVq5caS5fvmzq1KljSpcubf2cx8XFmQkTJpjHHnvMVKtWjW0ZByF0444kJiaahQsXmqeffto88sgjrCTvQsqjBy5dumSaNm1q3bO9atUqky9fPvP+++9b+1y/ft18/PHHJjQ0lI2CTLRv3z7Tv39/88UXX1jb3nnnHVOuXDkzZMgQ8+eff5ozZ86YESNGmL59+zL298jGjRuNxWKx2RhjzO+NV155xeTLl49xvwuHDx+2/v+jjz4yf/31lzHGmBMnTphLly6ZOnXqmAkTJhhjjDl16pQpVaqUsVgspnfv3sYYvsTOLJ9++qlp1KiRadmypbl27Zq1fenSpaZp06amefPm1j3eKcecbZq7M3ToUFO0aFEzZcoU069fP1OqVClToUIF88cffxhjjDl06JD58ssvTXh4uFm9erV1vFnXZExSUpI5e/asadCggVmyZIkxxpjIyEiTJ08eM2vWLGPM/z7T77//PtsyDkbofkClPNzkTv31119m1apVrCTvQspvFc+dO2euXbtmvLy8zPHjx83GjRtNnjx5zMyZM40xtw5vHjt2rDl79qzNPNgouHsbNmwwefLkMQ899JBZuXKlzWtTpkwx5cqVM0OHDjVRUVHm+vXr1t8Txj59MrKeOXr0qOnduzcbBnchI+OejPV7xm3fvt1Ur17dzJ4924SGhhqLxWLdy2eMMT///LPx9fU1e/fuNcYY8+eff5rnnnvOREZGsscpk4WFhZkKFSoYX19fc/78eZvXli5dah5//HETEBBgjh8/7qAKs58DBw4Yb29vmy+yjx49aqpXr24qVqxofvvttzSn4+/q3Tlz5oypWrWquXr1qlm7dq3NdmRcXJyZM2eOOXHihM00jLljcHWUB1TyRRfOnj0r6dbFi/6Lj4+PWrduLWdnZxljrLf4wZ0xKS7sMmTIEPXt21dubm569NFH1a9fPz311FOaOnWq+vTpI+nWxb02bNigb775xmY+KS9yh4wJCgrSoEGDFBMTo++++876eyBJgwYNUp8+fRQREaEVK1bI1dXVenV5xj590rueMcaoePHiioiIUI4cOZSUlMR6JgMysn5PvqgO6/f0O3/+vCTJy8tLdevW1ejRozVnzhwdPHhQJUuW1M2bN62vu7m56Z133tGOHTvUq1cvXblyRY0bN5aTk9Md/ZyQWsrbfSUbNmyYBg0aJHd3d7388ss6ffq09bV27dqpU6dOatCggYoVK3YvS83Wrl27pvj4eOtdbpKSklS8eHHNnz9fp0+fVs+ePfXHH39YX0vG39U7l3Lcrly5IkkqUKCAjDF67rnn9Nxzz2nKlCnW7cjTp09r/vz52r17t818GHMHcWTix72X8tv0devWGVdXV3Pw4EFjzL9/85VyuujoaPsVmE2l3OO0adMmU7NmTbNlyxZjjDHh4eHG19fXtGnTxtrn6tWrpkWLFqZJkyZ8I3mX/m0P0quvvmoefvhhM2XKlFRHFCxevJixz6DMWM9ERUXZr8BsKqPjnnL9dObMGfsVmA3169fP9OzZ0zr2U6dONbly5TKVK1c2H3zwgbVfYmKiiY+PN7NmzTKlS5c2fn5+pkGDBpxbeZdSjtuGDRvMqlWrzOLFi61tERERpkGDBiY4ONicPn36P+eBjEtISDC+vr5m8ODBNu0XL140tWvXNp6enqZ69erWdk6lSJ+Un9Pw8HAzZMgQ67V/Fi9ebHx8fEzLli2tfWJjY80TTzxhmjZtyrbMfYI93Q+QlLcVWLRokTZv3qyEhAQ99dRT2r9/v5ydndP8pt2k2EM7Y8YMvfLKK7p48eI9rT2rS77V0YoVKzR37lwFBASoXr16kqQ+ffqodevWOnr0qOrUqaPOnTvrscce0+nTp7Vu3brb/lzw31J+5ufPn69Bgwbptdde0/z58yVJkyZN0jPPPKOpU6dqwYIF1j1WktShQwfGPgMyaz3z6quvsp5Jh7sZ9+T104wZMxQaGsq4p8PTTz+tGTNmyMnJSdeuXVObNm20fv16NWrUSHPnztXUqVMl3Tr6wMXFRV27dtVPP/2kNWvW6LvvvlPOnDl18+ZNbsuWQcnjNnToUOv9n0NCQtS4cWPt2LFDvXv3VocOHfTHH39o5MiR+uuvv247D6TfN998o1WrVmnlypXKmTOnXnrpJW3evFnvvPOOtY+rq6vKlCmjNWvW6OzZsxoxYoQk7oeeXsmf08GDB+vNN99U1apV5eLiIkkKDAxU9+7dtWXLFrVo0UKdOnVSixYtdPLkSX311Vdsy9wvHBz64QCvvvqq8fPzM++8847p37+/qVq1qilatKj5+eefjTG2e0RSfhP5wQcfmNy5c1sv1oD/ljx+iYmJJi4uzrRo0cK4u7ubRx55xKbf9evXzYoVK0zfvn1N3759zcSJEzmnNRMl3zamXbt2pkGDBqZw4cImODjY+npoaKgpUaKEGT9+vPWWYbg7rGccg3G/N/65l+7jjz82tWrVMidPnjTG3LodVa9evUydOnXMtGnTrP0mTpxos8eVvax3LyIiwhQuXNjs3r3bGGPM3LlzjcViMRs2bLD2mT59uilTpox54403HFVmtjNs2DDj4+Njqlevbtzc3Ey/fv3MN998YwYOHGgqVKhgnn76aTNp0iTTsGFDU6tWLXPt2jXz+OOPm549ezq69CxrwYIFxsfHx+zatcvaduPGDXPu3DmTlJRk1q9fb55++mnTp08fM2HCBLYj7zOE7gfMwYMHTfHixc2XX35pbdu6datp2bKl8fHxsV7N8+bNmzYbAxEREcbDw8OsWLHintecHSRvZJ0/f950797d+Pr6milTpvznipBDgjLmn4fzFylSxHz33XfGmP/dBix//vzWKwYbY0yvXr1M27ZtOeQtE7CecQzG3XEWLlxoGjZsaIKCgqwXLfr9999N7969TUBAgOnZs6dp2bKl8fb2Zr1+F3bv3p3qi4qBAwea4cOHG2NuXbXc09PTegeQq1evWvstW7aMsc8kEydONEWKFDHbtm0zxhgzbdo0Y7FYTPfu3c2mTZvMokWLTMOGDU2jRo1M27Ztrfc/b9mypRk2bJgxhsPLM2LcuHGmefPmxphbV4J/7733TMWKFU2RIkVMeHh4mtPwmb9/ELofMLt27TKurq7mhx9+sGnfuHGjKVSokClevLj1vn7Jv6gRERHG09PTLF++/J7Xmx0sXrzYVK9e3frH6fz586Zjx46mfv36ZubMmdYNCL6JvHvt27c3Fy5csGlbtGiRKVmypImLi7O2Xb9+3Xz44YembNmyZufOndb2lLdzQ8axnnEMxv3euN3e6RUrVphGjRqZRx991Bq8jx07ZsaPH28ef/xx07ZtW87hvgt9+/Y1zZo1s2m7efOmadq0qZk8ebLZuXOnzZWbb968ad544w3z8ccfp5oGGXfq1CkTHBxsPXd+xYoVJn/+/Oa1114zHh4e5vnnnzd//vlnquleffVVU7hwYet5yPh3aW2HzJo1y5QpU8Z06tTJVKpUyXTs2NGMHj3ahIWFmRw5cpjDhw/f1d0rYF+E7gdE8i/fpUuXTMOGDc3IkSNNTEyM9fWbN2+aJk2amHLlypkyZcpYb3Mya9YskydPHjbI7sLy5cvNY489ZoKCgsz27duNMcacPXvWdOjQwTRo0MB88MEHbARkghMnTphOnTpZv1FPtnXrVuPt7W02btxo075v3z6TN29es379ept2/lBlHOsZx2Dc752UYXnjxo1m/fr1NuuWzz//PFXwvnbtmrl586b158QXrBmX/KXF8ePHreM4d+5cU6xYMePs7GwTsK9cuWIef/xxM3r0aIfUml1du3bNfPbZZ+bixYtmx44dxt/f30ydOtUYY8zkyZONxWIxgYGB1s//3r17Tf/+/U3x4sWtpwDg36Vcz1y/ft260yAqKsq89dZbJigoyERERFhvw/bjjz+a+vXr3/Zigbg/cPWIbOqft9BIvmCFp6enatWqpa+++kqLFy9WQkKCJCkmJkb58+fXsGHDVKhQIS1dulTSrYvsLFiwQG3btr23C5AFmVtfYqVqb9u2rQYMGKCcOXNq6NCh2rFjhwoVKqQZM2bIz89PkydP1urVqx1Qcfbi6+urTz75RC4uLnr//fcVFRUl6dZtekqVKqUFCxZo79691v4PPfSQihcvnmo+XNzlzrGecQzG3XGSL2b0yiuvqG3bturZs6eaN2+u9u3b69dff1WrVq00aNAgGWPUvXt3HT9+XG5ubnJ2drbeepDbsaVffHy8JClnzpz65JNPVL58eX3//fcyxuiRRx5R3bp1VaZMGfn6+kqSjh07pg4dOujixYsaNWqUI0vPdtzc3PTkk08qX758+uabb1SxYkUFBwdLunXRtM6dOytXrlzy8fGRJFWtWlWtWrXS5s2bVb16dUeWniWkvCjmxIkT1aFDB1WpUkWvvfaa/vzzTw0dOlRffvmlevfurZIlS+ratWsKCwtT3rx5VbhwYQdXj39jMWmlBGRpKX9h58yZo23btsnJyUmBgYF67rnnJEmdO3fWgQMH5O/vr9q1a2vNmjVycnLS999/r0cffVRFihTRokWLHLkYWZL5/ysBr1q1SpUqVVKpUqWsr61evVozZ87U9evXNWXKFFWrVk1nzpzRjBkzNHr0aO6beBdMiiswX758WbVr11Z8fLy2bdumwoUL6/PPP9eIESNUtmxZPfroo6pQoYImTpyo8+fPa/v27Yx9BrCecQzG3TFSrmN+/fVXtW7dWp988om8vb0VFRWl9u3bq3z58po9e7aKFSum5cuXa9y4cWrSpIn1Cua4e1FRUfL29lbDhg31999/a968eWrUqJE2b96sadOmad26dfLy8lLevHmVJ08ebdq0STlz5lRiYiLr+UyU/PvQvXt3/fbbb/ryyy/l4uKidu3aqXPnzurQoYMk6caNG8qZM6eDq82aRo4cqYiICA0dOlSnT5/Wrl27lJCQoKFDh+qZZ55RbGysVq9erQ8//NC6LZMzZ06bvxG4zzhi9zrujSFDhhgfHx/TtWtXExISYnLkyGE9BMgYY9577z3ToUMHU69ePfP888+ba9euGWOMadOmjRk1apRJTEzkUNs70LlzZ5v7Uu7cudNUrlzZPPvss+bo0aM2fZcuXWq8vLxMUFCQ9T7dyTjEPGNSjvH8+fPN1atXzZEjR0zDhg1NiRIlrPd7Xrt2rXnhhRdM/vz5TfXq1c1jjz1mPVSRsc841jOOwbg7xuTJk03Xrl1Nr169jDH/O7T/yJEjplChQiYkJMTad9OmTaxb7tLatWuth4f379/fPPvss9bXAgMDTbFixczmzZuNMbeul/LDDz+Yjz76yERGRlrHnsP57Wfr1q0mZ86cplKlSqZ06dKmcuXKjHcmOHTokKlYsaL5+uuvrW3bt283PXr0MI888og5cOCAiYqKMq+//roZNGgQVynPIgjd2cT169dtnn/88cemePHi1ot3rVixwlgsFmOxWMzYsWNt+iZvjN24ccOMGDHCFCxY0Bw6dOjeFJ7FXb161YwZM8bkz5/f5lYk77//vnn00UdNhw4dzB9//GEzTd26dY2fn5/p37+/MYZziO/G999/b+rWrWtWrFhhBg4caCwWizl+/Lgx5tZGcN26dW2C97Vr18zZs2fNqVOnOL8yA1jPOAbj7jgp18+XL182vXr1Mq6uriYoKMj6evLPZ86cOcbHx8ecOnXKZh4E74yJjY01I0eONCVLljSNGjUyefPmtV6BP1ly8N60aZP1S9SUGHv727Vrlxk5ciS3Os1ER44cMQUKFLC5E4UxxmzZssX4+/ublStXGmNuXbcgGZ/1+x+hOxt47rnnzJo1a6zPr1+/bt566y3z3nvvGWOMWbNmjfH09DTvvfeeeeutt4zFYrG+luzo0aPm2WefNSVKlOBCF+l0/vx5M2nSJFOwYEGbDd6IiAjTqFEj07FjR2sQPHv2rAkODjaffvopV6/NBH/99Zdp3bq18fX1NZ6enmbfvn3GGNu9T3Xr1jWlSpUy0dHRqabnZ3DnWM84BuPuOCnXDxcvXjTG3LqA1+DBg43FYjELFiyw6b9gwQJTsWJFa1/cvcuXL5v69esbi8ViXnrpJWt78pdJxhjTuHFjU7x4cbNu3TrW6fcBAnf6pLXj5Y8//jAVK1Y006ZNS3VUUkBAgBkyZMh/zgP3H0J3NjBq1CjrH6Dkb3pPnTplfv/9d3Py5ElToUIF88477xhjjPnpp5+Mq6ursVgsZu7cuTbz+fHHH82xY8fuae1ZWco/LJGRkaZfv37GYrGYSZMmWdsjIiJM48aNTY0aNcybb75pGjdubJo0aWLdMGADIeOSx+7tt982uXPnNjVq1DBLly5N1e/IkSOmQYMGJm/evKluJ4Y7x3rGMRh3x0i5bp4wYYLp06eP+fXXX40xt4J3SEiIcXZ2Nh9++KE5efKkOX36tHn88cdNUFAQG8CZJCkpyZw7d84MGjTI9O7d21SsWNGMGTPG+npsbKz1/1WqVDFt2rRxQJVAxqVcz5w9e9bmvvKDBg0y7u7uZs2aNdbtzUuXLpkaNWpY70OPrIXQnYUNHTrUzJs3z/p8xowZZtasWTa3itm0aZOpXLmy9TYCv/zyi+nZs6dZvXo130ZmkiFDhphatWqZDh06GF9fX+Pi4mKzx3v16tWma9eupkaNGqZ9+/bcp/Uu/XODdvPmzebHH380bdu2NY0aNUp1T1ZjjDl8+LDp2bMnh19lAOsZx2Dc7w9Dhgwx3t7e5qOPPjJ///23tf2vv/4yL730krFYLKZAgQImJCTEBAYGWg81Z/2eMbcbt6ioKDN69GhTrlw5m7+vN2/etJ4+xJgjqxo7dqypXLmyqV+/vhkwYIC1vWvXriZ37tymZ8+eZujQoaZp06amUqVKrN+zKO5bkUVdunRJ27Zt09atW5WYmKgePXpo/fr12r9/v3Lnzq1nnnlGuXLlUo4cObR//35t2LBBjRo10rBhw5QrVy61bNlSFotFN2/e5PYld+Hzzz/XzJkz9dVXX6lu3bo6ceKE5s2bpylTpsjZ2VmvvfaannzySbVs2VJXr15V3rx5Gfe7kPKqnPHx8XJ1dVWjRo0kSUWLFtWAAQM0Z84cOTs7q1OnTpKkKVOmqHfv3po1a5YkcRXbdGA94xiM+/1h5cqVmj9/vr766itVq1ZNknTx4kVduHBBPj4+mjhxonLnzq3Zs2erQoUKmjZtmiQpISFBLi4uDqw8a0q5fo+MjNTff/+tfPny6ZFHHlHhwoXVvXt3SdKyZcuUkJCgsWPH6sknn9TDDz+s2bNny8nJifU7soR/3oVi+vTpGjNmjI4dO6bly5fryJEjWrt2rebNm6dKlSpp586dOnr0qEqWLKmvvvpKOXLk4LOeFTk69SP9kvf0RUdHm2effdYEBgaaZcuWGWNufStWrlw5M3/+fOsekSFDhhiLxWJKlixpqlevbt3TyiFwd2/q1KmmWrVqNm2nT582ISEhxmKx2FxNOBnjnjEp92K8++675vnnnzeNGjUyc+bMMSdOnDDGGPPnn3+aNm3amIYNG5rQ0FDTsmVLU6BAAfZwZwDrGcdg3O8fH330kQkKCjKJiYlm//79ZsKECaZ48eKmXLlypnPnzubq1avm+PHjZuDAgcbDw8MsX77c0SVnWSk/r0OHDjUlS5Y0pUqVMg0bNjSPPvqoOXfunDHm1jr+zTffNF5eXsbf399UqVIlzQuoAVnBV199ZebOnWtdd8THx5uvvvrKeHl5mebNm1v7Xbt2zWYbiD3dWROhOwtKGSC2bNliAgMDTc2aNc3nn39ujDHmhRdeMGXLljULFiyw/jHatWuX2bhxI7fQyGRr1qwxRYsWNbt27bJpj4yMNDlz5jQWi8V8+OGHDqouexo2bJgpVKiQCQsLM3369DE1atQwXbt2tV4l/sSJE6Zfv36mWbNmplWrVhzOn0GsZxyDcXeM5NCXMvx98sknxmKxmM6dO5tixYqZTp06menTp5upU6ea4sWLm/379xtjbp3jHRoaaiwWi/nss88cUn92MWnSJFOkSBGzdetWY8yt8+ktFoupWrWq9WKY586dMwcPHjTLli3jM48s6+effza5c+c2Li4uZsWKFdb2GzdumK+++sp4e3ubli1bppqOL1SzLkJ3FhYaGmpat25tateubfLmzWtKlChh/cVN3jD75JNPbC7MYAy3FciI2wW2o0ePmoCAABMSEmJzG569e/eajh07muXLlzPemWjRokWmVKlSZufOncYYYzZs2GCcnZ1NxYoVzXPPPWe9UFRsbKyJi4vjtmCZgPWMYzDu907K9fuFCxfM+fPnrc/nzJljevfubT7++GPz119/GWOMOXnypKlSpYrZvn27td/Ro0fN8OHDuR3bXTh58qRp3bq1da/f2rVrTZ48ecyrr75qqlatamrUqGHzs0nGZx5ZwT/D8vnz582cOXNM0aJFzfPPP2/z2o0bN8zXX39tLBaLefXVV+9lmbAjQncW9fHHH5v8+fObXbt2mXPnzplTp06Zxx57zAQEBJhVq1YZY4wJDg42+fPnN1999ZWDq83aUm6QTZ8+3YSEhJigoCCzZMkSc/XqVbNq1SpTpkwZ06VLF/Ppp5+aPXv2mObNm5v27dsT+u7SP/9ILV++3IwaNcoYY8yqVatM/vz5TUREhHn33XdN3rx5zQsvvGCOHDnyr/PAnWM94xiM+72Tcv3w+uuvm1q1apkyZcqYBg0amB9//NGm782bN01sbKx54oknTOPGjVN9Gct6/u6tXr3aHD9+3OzYscP4+vqamTNnGmOMGTNmjLFYLKZIkSLckg1ZTsr1zJQpU6xHyVy8eNHMnTvXFChQwPTq1ctmmoSEBPPTTz/xpVI2QujOokaPHm0aNGhgc/++v/76y9SuXdsUL17cumH2+uuv8wubSYYMGWIeeugh6+1j/P39TdeuXY0xt/bAPv3008bNzc2UK1fOBAQEcG7lXUq5QZu8d+PmzZsmOjranD171tSuXdu8/fbbxphbe7ZLlixpHn74YTN69GiH1JsdsZ5xDMb93hs7dqwpWLCg+eCDD8yCBQtMUFCQ8fX1NQsXLjTx8fHm2rVrZvz48aZJkyamRo0anLZyl/5r3N5++23z7LPPmri4OGPMrSMOOnToYAYPHsxnHllKys/68ePHTbly5Yyfn5/57bffjDG3bgM2d+5cU7hwYdO7d+8058FnPnsgdGcxyRtgYWFhJiAgwPoHKXkD4JtvvjHu7u6mbNmyJjIy0jodv7B359tvvzWlSpUyO3bsMMYYs3HjRpMjRw4zf/58a58bN26Y33//3ezfv9+6kmXPR8ak/CP11ltvmUGDBplt27ZZ2/bt22d8fX3Nd999Z4y5dS/ujh07mo8++oiN4EzAesYxGHfHiI6ONlWrVk11u8HOnTsbX19f6/UiFixYYAYOHGhdr7N+z5iU6+jFixebiRMnmnnz5pm9e/da2/v372/8/PxMfHy8uXHjhnn66afN+PHjra/zmUdWM3LkSPPYY4+Zhg0bGmdnZ1O0aFHr6SgXL1408+bNM0WLFjXt27d3cKWwFydHXz0d6WOxWCRJTz31lPbu3au3335bkpQzZ05Jt26j1LRpU7Vt21aNGze2TsdtBe5ObGysChYsqICAAC1ZskStW7fWe++9pxdeeEFXr17V5s2blZCQoJIlS6pixYpycnJSUlISt+vJoORbaQwdOlSTJ09WrVq15Ovra9PnoYce0hdffKHvv/9egwYNUkJCgrp06WIde2Qc6xnHYNwdIzExUefPn1eBAgUkSdevX5ckLViwQB4eHnr33XclSZ07d9a7775rvV0P6/f0M8ZY1+/Dhw9Xjx49tGbNGr366qvq1auXJk6cKEnq0qWL8uXLpxIlSqhWrVo6dOiQhg8fbp0Hn3lkJREREQoPD9e4ceO0ePFiRUZGqmzZsmrUqJEOHz6sfPnyqU2bNhoxYoTi4uLYhsmm+IuRRVWsWFGzZ89Wr169FBMTo/bt26tAgQKaMWOGqlSpogkTJkjinsQZkfL+icmuXr0qZ2dnff311+rVq5fCwsLUt29fSdLGjRu1bt06lS5dWrlz57ZO8895IH0iIyO1dOlSrVmzRnXq1LF5rXLlynryySf12Wef6dNPP5W/v7++/fZbWSwWm4063B3WM47BuNtPWuv3IkWKqGDBgpo/f76efPJJubm5We+1XbZs2TTnw7hnTPIXS7/88os2bdqkDRs2qF69ejp+/LimTZumJUuWKF++fOrdu7fef/99bdy4UU5OThoyZAj3JkaWdejQIbVp00b16tWTJPn4+GjWrFl64YUX9Pjjj2vjxo0qWbKkXnjhBb300kuyWCxprquQtVmMMcbRRSDjVqxYoZdeekkuLi6Sbu3927Ztm3LmzCljjPUPHO5MypXcvHnzVLBgQbVq1UoJCQmqVq2aDh06pHnz5ik4OFjSrT0izz77rDw9PfXJJ58w3hk0ZcoUtW3bVn5+fta2pUuXauzYsdq8ebMKFSpkDdQpx/ivv/7ShQsXVKlSJTk5OenmzZvsfbID1jOOwbhnrpSB7ffff1fevHnl4eGhXLlyadWqVXrllVf01FNPKTw83DpNvXr11KRJE+sXHbh7YWFh2rZtm4wxWrx4sXLlyiVJOnHihMaNG6e//vpLX375Zap1OYEbWVX//v21fv16HT582Kb9/fffV0hIiHx8fLRlyxb5+voqOZaxfs9+CN3ZwOnTp3Xq1CnFxsbqkUcekbOzM+HjLg0ZMkSLFy/W888/r9DQUHl5eembb75R7969VaJECQ0ZMkTnz5/XvHnzdPr0ae3Zs0c5cuRgQzgD1q9fr0mTJmndunU2G1Th4eF68803debMGUnSjRs3rGFjw4YNyp8/v2rVqmXtzwaZfbGecQzG/e69//77qlOnjmrWrCnp1mHNX3zxhaKiotS1a1d17txZ1atX1/Tp0/XWW2+pWLFiqlChgg4fPqyLFy9q3759jHcm+vjjj9WtWzflz59f3333nSpWrGh97dtvv1XTpk21c+dO1ahRw4FVAul3u73TkZGReuWVV9SxY0cNHDhQbm5ukqSvvvpKq1at0smTJ3XhwgVt3LjR5ohJZC+E7myI8HF3Zs2apZEjR2r9+vWqUqWKdSyvX7+uHTt2aMiQIfr7779VuHBhlShRQvPnz1fOnDkZ9wxI/pIiOVCvW7dOvr6+qlixoqKiolSnTh0FBgZq/vz51mmuXr2q9u3b66mnntJLL73kwOofbHzeHYNxT5/NmzdbD+EcPHiwjhw5or59+2rGjBk6cOCAVq9eLS8vL7322muqVauWdu7cqXfffVcuLi4qUKCAJk6cyGHNd+F2IWTVqlV65pln1LdvX40cOVJFixaVJB04cEDPPPOMlixZomrVqt3jaoGMS/lZ//nnn+Xk5GQ9RSUmJkbDhw/X3r171ahRIw0aNEjXrl1Tv379VLJkSTVq1Ej9+/fX559/bv1yENkPoRv4h379+snV1VVTpkyx7lH6556lP//8U/nz55eHh4csFgt7njJg5MiRKlSokAYOHCiLxaJffvlFtWvXVteuXTVo0CCVKVNGH3zwgd59912VKVNGY8aM0enTpxUREaFTp05p586djDmA/7RgwQK99957ql+/vnLkyKHy5cvrxRdflHRrT9PkyZPl7u6uIUOGqGHDhqmmZ/2eMSlDyObNmxUTE6NKlSqpaNGiypkzpxYuXKgXXnhBzz//vJ599lkVLVpU48aN0+nTp7Vr1y7OZ0WWkfIox9GjR2vx4sUyxujcuXOaPHmyevTooatXr+qNN97Q119/rQMHDsjf31+urq7av3+/Dh48qCeffFIrV65U1apVHbw0sBf+igApGGN05MgReXh4SJL1kPEcOXLo2rVr2r9/v2rVqiV/f3+badggS58rV65o27ZtSkhIkIeHh7p3767KlStr1qxZGjVqlJycnDR8+HB1795dRYsW1ejRo9WsWTMVLlxYfn5+2rFjB3ufAPyr5ND3wgsvKDExUTNmzNDvv/+ukSNHWvu0aNFCFotFkyZN0pQpUxQTE6PmzZvbzIf1e/qlvKDlq6++qk8//VSXL19WhQoV1KFDB7300kvq1KmTnJyc1KlTJy1atEhdu3ZV/vz59fnnn8vJyYn1O7KElIF7/PjxmjVrlhYtWqQaNWrolVdeUc+ePXXu3DkNHTpUr7/+uoYMGaKNGzfqoYce0iOPPCJJ+vDDD1WwYEH5+Pg4clFgb/fo1mTAfSet+znfvHnTjBgxwtSqVcvs2bPHps/Ro0fNo48+arZv334vy8x2ku9FfO7cOdOhQwfz6KOPmvfff9/a/sknnxgfHx/Tt29fc+zYMet0+/fvN6dOnbL24x65AP5L8vrCGGOWLl1qSpcubZo0aWL27Nlj0+/rr782VapUMUOHDr3HFWY/Kf9ubt682dSuXdts2bLFHDlyxPTs2dPUrVvXjBkzxnof+pUrVxqLxWJGjx5tzpw5k2oewP1o2bJlNtshv/zyi2nWrJlZu3atMcaYVatWmfz585t27doZJycn8/bbb5uEhASbeWzZssX079/feHp6plonIfshdOOBlPIP+sGDB82vv/5qTpw4YYwx5s8//zRFihQxTz31lNm0aZO5du2aOXnypHnqqadMo0aNzM2bNx1VdraQcux/+ukn07hxY1OnTh0zZ86cVMG7X79+5uDBg/86DwD4NymD96JFi0yNGjVMjx49zM8//2zT76effmLdcheOHz9u83zFihWmR48e5tVXX7W2xcTEmAEDBpg6deqYsWPHWoP3/PnzjcViMUOHDjVRUVH3tG4gvcaPH286d+5ss744efKkmTFjhomPjzebNm0yRYsWNdOnTzfGGNOuXTuTI0cOM3bsWJtp1qxZY1544QWzf//+e74MuPc4pxsPHJPiUKARI0ZoyZIlSkhI0NWrVxUaGqrXXntNf/75p9q0aaOkpCSdOXNGvr6+kqSffvqJi6ZlktDQUP3xxx+Kjo7WoUOH5OXlpcGDB+vFF1+UxWLRwoULNWLECDVq1Ehvvvmm9WcAAOmVcr2/YMEChYeHq1q1aho4cKAqV65s05f1e/qFhITIzc1NkydPVlJSkq5du6ZnnnlGP/zwgwIDA7V27Vpr37i4OI0cOVLbt29X3bp19eabb8rV1VVLly5Vx44dNXr0aI0ePZpzunHfio2NlYuLi3LmzKkdO3aoSpUqcnV11eXLl+Xp6am+ffsqPj5eERERcnFx0YABA7R9+3blyJFD3333nc1dbuLi4rhi+QOCE5XwwEle2b3zzjuaPXu2Fi1apHz58mnHjh0aPny4oqOjNWPGDG3YsEG//vqr9u/fr4cfflgtW7bkdj2ZZMGCBfroo4/0zTffyN/fXzdv3lRwcLDmzJkjJycnde/eXf/X3p2H5Zj9fwB/P6mIJGnTF5GlKZJkkLGN0STLxDQzJvsSZfkqWyNJlrEvUVlaVEL2LW3DWCajbCGmyVKSGRGyTZJU5/eHb/fveYQpW8r7dV1dM933uZ/Oc3uu85zPOef+nIEDByI3NxdRUVF8zomI3opMJpMC78GDB0Mmk8HHxwdeXl5YsmQJGjduLJVlwF12NjY20rPw9+7dg7a2NjZu3IjJkyfj+PHjWLNmDZycnKCkpITq1atj3rx5GDduHP755x/p+/SHH35AlSpVYGJiwoCbPkoLFixA165dYWVlBQDYu3cvfvrpJzg7O2PcuHGoVasWcnNzcf78ebRu3Rqqqqp49uwZrl+/jnnz5qFbt24AoLAXNwPuTwdnuumTUZxUp6ioCEVFRejbty/Mzc0xb948qcyOHTswYMAArFmzBiNHjizxGpwBeTfmzJmD2NhYHD16FEpKSpDJZLh16xa+/fZb3L59G9OnT8ewYcMUOl6v2nqGiKi05Ge8/f39ceLECQQFBbFteUPy9xN4PqC6YcMGLF++HC1atMDt27cxfvx43Lx5E0OGDJFWMgHPt+FUVVWFkpISCgoKUKVKFYXXIvqYJCQkYMyYMWjYsCFmzZqFVq1a4eHDhxg7diwyMjLQv39/ODk5QVVVFYsXL8a0adPg4OCAlJQUFBYWIjExUUrOy8/5p4nfMvRJEHKZVC9dugRlZWVkZmZKAfSzZ89QVFSE7777Ds7OzggICMDTp09RWFio8DoMuN9O8f2sVq0a8vLy8PTpU2mfbn19fcyfPx+3bt3C8uXLsXfvXgD/PyLMTjERvUxxGyE/h/Cq+YTiGW8AcHJyQnBwsDQYS2X3YvDw5MkTPHnyBHPnzkVycjJ0dXXh6+sLfX19hIWFITg4WLr/1apVk+69srIyAxH6qFlZWcHT0xM5OTnw8vLCyZMnUatWLaxZswZGRkYIDw/H2rVrkZ+fDzc3Nyxfvhz5+fmwsrKStjgtLCzk5/wTxl4sVXryo4oTJkxA27ZtAQDdu3dHYGAgrl69ChUVFanTVbt2bdSuXRtVq1ZlkP2WXuzIFt/PPn36IDk5GT///DMAQEVFBQDw9OlTdO/eHX379oWdnR2Akp06IqJiRUVFUhuRnZ2NZ8+eAXjebrw4aFpMJpOhoKBA4RgH9d6N0aNHw8nJCbdv34aXlxeSk5Ohp6cHPz8//Oc//8GSJUsQGRmpcA3vPX3s8vPzAQD29vZwcHCQBpaSkpKgoaEBPz8/NG3aFOHh4QgICEB+fj5cXV2xYcMGrFq1CioqKtJqDvp0saWjSq+4Q3blyhU8evRI+sIfPnw4WrVqhR9//BFpaWlQVlZGfn4+EhISoKenV55VrhTkl4OHhITAzc0NQ4YMQWxsLExMTLB+/XosW7YMrq6uOHHiBC5fvgwfHx80a9YMP//8M2efiOhfFbcxc+fOhbW1NWxsbDBz5kwAzwf5XhZ4CyGk54jXrVuHX3755cNVuBIrbq8HDRqEESNGIDs7WyHw9vb2xvfff4+ePXuWc02JSk8IAVVVVQDAwoULceTIEWRkZCAqKgozZszA6dOnpcDb2NgYmzdvxtKlS1FQUIBq1apJr8FcQMRnuumTsHnzZnh5eaF27dqIioqCtrY2AODgwYNYvnw5Dh8+jJYtWyI3NxdFRUU4e/YsVFRU+OzNO+Dm5obNmzfj66+/hrq6Onx9feHt7Q0nJyccOHAATk5O0uivrq6ulCGe956IXkV+UG/dunX46aefMGvWLCQmJiIpKQmNGzfG9u3bASjm4pBvVwIDA+Hk5ITdu3dLK2vo7cj/u2zYsAEhISHQ1tbG9OnT0apVK6kc86NQRePt7Q0vLy/s3LkThoaGiImJwebNm6Grq4tZs2ahdevWePToEQYOHAgDAwOsXbuWfRhSwKCbPgkhISEICgpCSkoKLl26BB0dHenc/fv3sW/fPty4cQO1atXC6NGjoayszCzl70BMTIzUqbW0tER8fDw6duyIjRs3YsCAAQCAmzdvIjMzE7m5uejQoQMzxBPRa8kHdr/88gvOnz8PIyMj2NvbIy8vDzt27MCiRYvQrFkz7Ny5EwBKJOry9/eHm5sbQkND0a9fv3J7LxVNaRJaypfZuHEjFi1ahG+++Qbz5s3jYCpVOEIIFBYW4vvvv0f9+vXh4+Mjndu0aRNmzZoFExMTzJ07F+bm5sjNzZXyFfDzTvLYq6VK52WdgqFDh0JDQwMzZszAd999hx07dkBHRwdCCNSuXRtDhgxRKF9YWMig7x24e/cu2rRpA0tLS2zbtg0jR47E6tWrMWDAADx48AD37t2DkZER6tatK13De09EL9O3b194e3ujUaNGAIDjx49jzJgxuH//PrZu3QrgeXIue3t7yGQyLFq0CD/88AO2bdum0KYUB9zBwcEMuMtAPiFpamoqmjRp8tJyxY8GKSkpYdCgQdDR0UH37t0BMEcHVTwymQzKysrQ0NBAVlaWwiqNgQMH4tSpUwgMDMS9e/cQEBAAU1NTANxxhUrip4EqFflGLi4uDkeOHEF8fDyUlJRgb2+Pn3/+GQUFBRgyZAju3r0rZc5+EZe9vRtPnz7FrVu3sG3bNowaNQqLFy+Gs7MzAGDfvn2YNm0aHjx4oHAN7z0RvSg1NRUNGjTAf/7zH+lYw4YN4ezsDDU1NYSFhUnH1dTUYG9vD3d3d/z222/w8PCQznl7e2PatGkICQmBvb39B30PFZl8wrrx48fDxsYG2dnZrywvn5PDxsYGVapUYY4OqtBMTU1x9OhRnDp1SuF4o0aN0KZNG3Tr1g2fffaZdJwBN72Iy8upUpo8eTJCQ0Ohrq6OW7duoX///vD09ETTpk2xfft2+Pj4QENDA8HBwUya9g68akQ3JSUFo0ePxokTJzBv3jxMnToVAJCbmwsHBwdoa2sjKCiIsx9EVGorV65E165dYW5ujuzsbISGhiIoKAjW1tYKSz+fPHmCo0eP4quvvkKVKlXw+PFj9O3bF8OHD5ceb6GyuX37NiZPnoxRo0ahc+fOry0rv7T22rVrMDAwkBJSEVVEPXr0wJ9//ol169bB1NQUWlpacHBwQOfOnTFx4kTIZDLOcNMrMeimSkH+y/2PP/6AnZ0dwsPDoauri+vXr2PAgAFo06YN/P39oaenh61bt2L27Nno2bMnli1bVs61r9jk731YWBgyMzOhr6+PYcOGAQCWL1+OoKAgfPHFFxg9ejSys7OxcuVKZGZmIjExEcrKynzuiYhKJSsrC99//z0uX76MQ4cOwdTUFHfu3MH69euxfv16dOvWDStXrixxXXGeiLy8PCmjMJVNYGAgZs+ejUaNGmH79u3Q19d/ZVn5Nt3Hxwdr1qzBoUOHFB4lIqoo5JeU29nZ4ezZswAAdXV1FBYWIjk5mX0Z+lcMuqlSWb58OS5dugRlZWWsWrVKagAvXLiAzp07w9HREUuWLEFRUREOHz6Mrl27cjnzO+Ll5YWlS5eiffv2OHz4MOzt7eHn5wc9PT0sXrwY0dHROHbsGCwtLaGrq4udO3dCRUWFWWyJqEzOnTuHWbNm4fTp0/jll1/QvHlz3LlzB2FhYdiwYQPMzc2xfv368q5mpVJUVIQ9e/Zg4cKFSE9PR2pqKmrVqvXSpJfygYe/vz/c3d2xatUqODg4lEfVif5VaWan5T/XsbGxyMrKQn5+PoYPHw5lZWX2ZehfMeimSuPBgweYOHEiNm3ahO7duyM6OhpFRUV49uwZqlatitWrV2PRokU4efKkwpJyNpRvpvhLSgiBvLw8DB48GJMmTYKVlRX++OMPdOrUCVZWVggJCYG+vj6Kiorwxx9/oF69eqhduzZkMhmzlBNRqcl3es+dOwdPT0+cPXtWIfBetWoV0tLSsH79ei7xfAsvC0KePn2Kw4cPw8nJCfXr18fvv/8OAArt+IsBd3HCOj4/Tx8r+c/s6xIEAq/uL7IvQ6XBbySqNDQ1NTF9+nSMHTsWsbGx2LVrF5SUlFC1alUAz7Paampqonr16grXMeAuO/kO2eXLl3Hx4kUYGBigcePGkMlkMDMzw4kTJ5CQkABHR0dcvXoVSkpKaNmyJbS0tKTnnvglRUSlJb9ss1WrVpg7dy4sLCxga2uL5ORk6OjowNXVFWFhYQqJvKhs5Nv3AwcOYMOGDdi9ezfu3r2LHj16IDAwEHfu3JEykhdvsQlAYQ90Btz0sStrgsDi/uKLbQv7MlQanOmmSufatWuYP38+QkNDERoaiq5du0JZWRmDBw8G8HxZEJ+5eTemTp2K7du348GDBygoKMCmTZtgZ2cnnb906RK++OILGBsbY9u2bQqZh4mI3oT8zFRSUhK8vLwQFRWFy5cvS9uJ8dnKt/fTTz8hPDwcTZs2xc2bN6GtrY1p06ahZ8+eiImJwZQpU1CvXj3s379f4brNmzdj4MCB2LlzJ7dkowqBCQLpQ+BMN1U6DRs2xPTp0zFs2DAMGjQIrVq1wpw5c5Cbm4uIiAhplpXKTn6MLiYmBtHR0Vi6dCnWrVsHHR0dBAQEID4+XipjbGyM3377Derq6kygQ0TvhHwwbW5uDg8PD0yaNAkNGjR4aRkqu9DQUGzcuBHbt2/HoUOHMGrUKJw+fVqaGfz666+xfPlynD59Gi4uLgrXdu/eHTExMQy4qUIIDAxE69atce3aNTRr1uy1ZV9MEGhra/vamXEieZzppkorPT0dvr6+CAoKwooVKzBixAgAQH5+Pkcl31JUVBT27t0LIyMjTJs2DQBw4cIF9O/fH02aNMG0adPQoUOHEtdxKw0iepV3MTvNZyvfjUmTJuHp06dYtWoVdu7ciREjRmDRokVwdnbG48eP8c8//0BXVxcnTpxA27ZtpWW3zJFCFQkTBNKHxN4vVVqNGjXC2LFjpQRf0dHRAAAVFZVyrlnFlpWVBS8vL4SFheHy5cvScTMzM2zduhWpqalYsmQJDh8+XOJaBtxE9DLyz1ZmZmbi4sWLKCwslFYlvWp1UmFhofT/jx8/ZsD9lorvZ25uLlq0aIH4+HgMGzZMCrgLCwuxefNmREZGQklJCVZWVqhSpYp0HQNu+pi92I4oKSmhV69emDNnDqpXr45evXoBUMxTALw8QWBgYCADbioT9oCpQnhZh0u+s/UqTZo0wZQpUzBo0CD07t0bUVFRXHb4lvT09BAWFoauXbvi1KlT2Llzp3TOzMwM27ZtQ1xcHGJjY8uxlkRUUQghpAE5T09P2NnZoV27dujTpw/mzp2LvLy8lw7YCSGkIC8sLAwrVqxAXl7eB617RSaEKPHdWnw/TU1NMW7cOHTp0gVBQUFwdnYG8HxgY8uWLcjIyHjpdUQfKyYIpPLG5eX00ZNfDv7XX3+hsLAQDRs2lM6/akmi/DK3kydPYt++fRg0aBCMjY0/SL0rq+L7nZycDBcXF6ioqMDJyQl9+/aVyly9ehWGhobsiBFRqc2fPx8rVqzA+vXr0a5dOzg4OODSpUvYvXs3LCwsFMrKt/sBAQEYN24c9uzZI81UUdn8/vvvePLkCbS1tdGqVSvIZDKMGjUK4eHhOHLkCAwNDZGTk4OxY8ciOzsbCQkJXFVAFRITBFK5EUQfqfnz5yv8Pn36dNGoUSOhq6srLCwsxIYNG8S9e/deem1RUZH0/8HBwWLq1Kni7t2777W+n5Li+5uUlCS++uor0aNHD7F3794S5QoKCj501YiogikqKhJ3794VXbp0EVu2bBFCCPHrr7+KGjVqiMDAQCGEEE+fPhWFhYVCCCH9Vwgh1q5dK2rVqiV27tz54SteQXl6eoqAgADpdxcXF1G3bl2hqakpTExMhLW1tcjPzxc3btwQ9vb2QlVVVRgaGgoLCwvxxRdfiPz8fCEE23eqeEJCQoSBgYFISEgQQgixbNkyUa1aNRERESGEEOLZs2ciJiZG1K5dW0yYMEHh2tu3b4vY2NgPXmeqPBh000cpOTlZyGQy8c033wghhNi8ebPQ0dER4eHh4tChQ6J///7CzMxMLFy4UDx69EjhWvmA29/fXygrK4s9e/Z80Pp/CuQDb2tra9GmTRsRFxdXzrUiooron3/+Ee3atRM3btwQERERQl1dXaxZs0YIIcSTJ09ESEiIOHfunEL7Xhxw79ixo7yqXeHcvn1bWFlZiS+//FJs3rxZHDhwQJiamopjx46Jixcvil27dokWLVqI1q1bS0H1/v37xd69e8WRI0ekAY9nz56V59sgeiMTJ04UY8eOFUIIsWPHDqGhoSG1Mzk5OeLmzZuisLBQxMfHKwwqcYCJ3gUuL6ePkhACR48ehYODA9q2bQtbW1sIIeDk5CSVmTJlCiIiIrBu3Tp06tRJSsTzYrKLkJAQfPvtt+X1Vio18b8lnomJiQgLC4O3tzeTpRHRa71sF4Pc3Fy0a9cOhoaGiI+Px4IFC6T2PjU1Fc7Ozhg/frz0GEtQUBBcXV2xfv16PltZSsXtdUZGBiZMmICioiLUq1cPSkpKWLVqlVTm3LlzGDRoEDp06IDAwMASr8NdKKiiKX7c0NnZGebm5jA3N4eNjQ2WLFkiJQgMCQkBADg6Opa4juhdYKtJHyWZTIbOnTtj8+bNOHPmDJydnZGZmQkAUmKLpUuXok6dOvD19ZWukQ+4f/rpJwQHBzPgLoOy7l8uk8kghIClpSVWrlwJJSWlUiW4I6JPk3zAlp6ejpycHDx58gTVq1fHokWLcPLkSVhZWcHJyQkFBQXIycmBi4sLhBDo06cPgOfJvA4fPoyNGzcy4C4DmUyGoqIiGBoaYuXKlQCA8PBwhV0oZDIZLCwsYG9vj+TkZOTm5pZ4HQbc9DETTBBIHym2nPRR69ixI8LCwmBsbIwDBw7gyZMnUFZWlhrUdu3aSUFeccAdHByMKVOmYN26deyQlYF8Zzg1NRUPHz6UOlyvC8ZfXCzDLykiepXiNsbDwwM9e/aEhYUFFi5ciCtXrqBnz55wc3NDTEwMbGxs8O2336JXr174+++/ERsbiypVqqCgoAA1atTAunXrFJI30usVt+HF35MNGzbEqlWr0K1bNyQnJyMgIEChvLGxMR48eICcnJwPXleityGTyaR25vfff8eBAwdw9uxZCCEwYcIEjBw5EqqqqjAyMsLt27dx9epV/PDDD3j48CG8vLzKufZUmXF5OX00XrVkLT8/HwkJCXBwcIC5uTk2bNiAGjVqQEVFBR07doSJiQlCQkIgnucowNSpU9GlSxd888035fAuKj4PDw+Eh4ejWrVq6NKlC9zc3GBkZPTSfx8hl0HY19cXOTk5cHd3L49qE1EFsWfPHri6umLlypU4fPgwzpw5A11dXSxcuBBNmjRBfHw8goKCoKGhgQYNGmDChAnS9j3MmF12Lw6oamtrQ0lJCRoaGrh27Rr++9//4sGDB+jXrx/GjBmDO3fuYPjw4VBVVUV0dDS32aQKYebMmahfvz5GjRoFAHB1dcW2bdvw5MkT1K1bF/Xq1UNUVBTu3LmDCRMmYN++fahbty60tLRQvXp1HD58GCoqKlxSTu8Ng276KMh3CsLDw3H16lX8888/GD9+POrXrw8hBOLi4jBgwABUq1YNRkZG0NHRwfnz53H27FmoqKhIryFesYUY/buoqCiMHz8evr6+OH78OE6dOoXc3FwEBwejadOmCv9O4oUte1xcXBASEoIff/yxPN8CEX1kXhyw27ZtG9LS0qQBuo0bN2LdunXQ0tLCnDlz0Lx58xLXsCP89uQHVDt37owpU6agadOmSEtLw6RJk3DgwAHUq1cPLVq0wJMnTxAREaHw3Ur0sbpz5w7s7OxQrVo1jB49Gtra2nBxcUFgYCDq1KmDP//8EzNnzoSqqipOnjyJKlWqSKsna9WqhU6dOkFJSYkDe/ReMeimj8q0adMQHh6OFi1a4NmzZ0hMTMSuXbvQtWtXAEBcXBxcXV3x559/4syZMzA2NpaWHLKhLLsXO1O7d+9GSkoKpk+fDgCIjIyEr68vHj9+jJCQECnwBv5/mWhxwrrQ0FDuXUlEr+Tv749r167h+vXrsLCwwJQpU6RzmzZtwrp161CnTh14eHigVatWAMBB1HfkVQOqAQEBMDExQUZGBiZPnozjx49jwYIFGDhwIIMQqhCYIJAqjA+UJZ3olYq3gFmzZo34z3/+IxITE4UQQkRERAiZTCbq1KkjoqKihBDP92fdv3+/6N2790v3bKXSk996x9fXV0ydOlX069dPeHl5KZSLjIwUNjY2olOnTiI5OVnhnL+/v9DQ0OCWPURUgnzb7O7uLjQ1NcUXX3whdHR0RN26dcWFCxcUyoeHhwszMzPh7u7+oata6bz4vbhr1y4xb9486fd9+/aJr7/+WnTo0EH8+eefQgghLl26JCZPnszvVqpwij+r6enponfv3kJDQ0N07969RDlPT09hZWUlHj9+/KGrSCQ4pEPlYu7cudiyZQuA50kv7t+/jxs3bmD+/Plo3bo1IiIiMHDgQKxatQq2trYYOnQofv31VygpKcHa2hr79u2TMmVzZLLsirdXA4AZM2bAy8sLp06dQlJSEnx8fJCWliaV7dWrF1xcXJCbmwsfHx/p+MqVKzFx4kSEhIQwYR0RKRBCSG3z9evXUVBQgAMHDuD3339HWFgYWrZsiVGjRiE5OVm6xsHBAUuXLsXcuXPLq9qVgvy99/Pzg5ubGzZs2ID8/HypTO/evTFhwgTUrFkTTk5OuHDhApo1a4alS5dCSUmJs35UITBBIFUo5R3106fn2rVrwtzcXNja2oq9e/dKxxMSEkRGRoZISUkRzZo1Ez4+PkIIIaKjo4VMJhMymUzEx8eXV7UrpVu3bonJkyeLkydPCiGEOH78uOjevbto2LChSE1NVSgbHx+vMAPy008/ifDw8A9eZyL6eAUEBCj8vnXrVlGlShVhZmYm0tLSpOMxMTHC1tZWtG/fvsQKGiGEKCgoeO91rYzkZ6c9PDyElpaW6Nq1qzAyMhK1a9cu0a5HR0cLS0tL4eTkJIRQXAFF9DGT/6xfuXJF3L9/Xzx8+FAI8f8z3h07dhTLli0Tubm5IiMjQ3Tr1k306NGDn3MqFxzGpA/O0NAQGzZswLNnz7B69Wrs3r0bANC+fXs0aNAAFy9ehJaWlrS/do0aNeDi4oIlS5bg888/L8+qVypbtmxB3bp1ERsbCzU1NQDPt2BbuHAhjI2NYW1tjfT0dKm8lZWVwuqChQsXwsHBobyqT0QfmcjISKxZswaFhYXSVoL169dHv379cOXKFTx69Egq26NHD7i4uEBbWxt9+vRRaGsAbj34popnp7OyspCXl4fY2FgcPnwY4eHhsLS0RPfu3RVWMtna2sLX1xerV68GAD4/TxWG/PaD1tbWsLKywtSpU3HlyhU0bNgQK1asgJaWFmbMmAFzc3O4urpCVVUVERER0p71RB8Sg2764IQQMDMzw7Jly1BQUAB/f3/s2bNHOn/79m2cPn0aDx8+xI0bN7B06VI8fvwYkydPlraNobJ78QumQ4cO+PHHH3H58mXcv39fOm5paYl58+bhs88+Q/PmzZGZmalwHTvDRPQy1tbWOH36NKpUqYIjR44AeD5Y5+7ujk6dOsHW1haXLl2SytvY2MDR0RH9+vVDgwYNyqnWlc+bDqgyCKGKJioqCuHh4fD19YW9vT2uXbuGESNGICUlBY0bN4aPjw969uyJ3Nxc9OvXD1FRUVBRUUFBQQEfn6APr7yn2unTVLy0JykpSXz11VfCxsZG7N69WzpvbW0tZDKZMDIyEmZmZiI/P7+calr5xMTEiPv37wshhPjrr79E7969hY6OjkhJSVEoFx8fLyZOnMhlnkT0r+SXa544cULIZDIxbdo06VhiYqLo2bOnqF+/vrh48eJLX4NtzZt5MeFZRkaGcHBwECoqKiIuLk7h3OnTp4Wtra1QU1MTN27c+JDVJHprTBBIFRm3DKNyI/63zcP58+cxadIkKCsrw9nZGX379gUAbN++HVWrVkWvXr24Ldg7cvXqVTRp0gQjR47EsmXLoKGhgczMTDg6OuL06dM4evQojI2NS1zHPXKJ6FUOHz6MzMxMDBw4EGPGjMGzZ8/w+eefw9XVFZMmTcK8efMAAImJifDy8kJycjIiIyPRvHnzcq555RIbG4v27dtDU1MTf//9N8aMGYMTJ04gLi4On332mVQuISEB27dvx5IlS9iuU4Uh5LYP9PPzw/Xr15GamoqWLVti1qxZUrmoqCj4+voiNzcXq1atgpmZmXSOCQKpPDHopnL1b4F3MQZ9705MTAy+++47DB48GIsXL5YC71GjRuHs2bPYv38/WrRoUd7VJKKPnBACOTk5sLe3R35+PjQ0NBAXF4f4+HiYmppi3bp1cHZ2hpubmxR4nzlzBmPHjoWBgQF27dpVzu+g8uCAKlVm8sHyjBkzsGbNGrRs2RLXr1/H/fv3cerUKTRu3FgqHxMTA09PT7Rp0wZr165VCNiJyk25zbET/Y/8UnNra2vRpk0b8dtvv5VzrSqn4nsdExMjVFVVhZOTk3jw4IEQQogbN26Itm3bil69epVnFYmogsnOzhbGxsZCJpOJBQsWSMefPHkigoKChLKysvDw8JCOX7x4kUs834Po6GhRvXp14eTkJGVxvnHjhujZs+dL90UnqmjedMcVoo8BZ7rpoyD+NwqZmJiIsLAweHt7cwnQO7JgwQLk5ORg7ty5UFJSku51TEwM7OzsMHbsWMycORNaWlq4e/cutLS0eO+JqNQePHiAgQMHIicnB1WrVsWQIUMwaNAgAMCTJ0+wefNmjBs3Do6OjvD19ZWu41LPd6e4XY+NjYWdnR2GDx+ORYsWoVatWsjMzES/fv2go6ODyMjI8q4q0RvZsmULBgwYAFNTU2zZskVakZeYmAgPDw9cvnwZBw8eRKNGjRSuYztDHwt+Cum9KGsWVJlMBiEELC0tsXLlSmlrKiq7F++9mpoaFixYgIULF6KoqEjaKsPW1hYTJ06Ej48PJk+ejMePH0NbW5tZbImoTDQ1NREVFYWtW7dCRUUFwcHB2LRpE4Dn7c/gwYMxZ84c/PHHH5Af52dH+O0sWLAAHh4eUrsuhECPHj2wZ88eBAcHw8vLC/fu3YOBgQGioqIQERFR3lUmKrV3teMK2xn6WPCTSO+c/KhiamoqHj58iNzcXOncq7y46ILPmZWd/L1PS0vD7du34erqio0bN8LT0xPz58+X9tkGoLBHbvHWMgC/pIio7PT19eHn54fq1atj/fr1CAkJQWFhIWxtbZGVlYVDhw5JwSGVHQdU6VNS3A+JjY3FgwcP0KBBAyxevBg2Njawt7fHxYsXpbKWlpbw9PSEs7Mz9PT0yqvKRK/F5eX03nh4eCA8PBzVqlVDly5d4ObmBiMjo5cu9RFySS58fX2Rk5MDd3f38qh2pTB9+nTs3bsXWVlZGDFiBIYOHYrk5GQMGDAAnp6eGD58OPT09ODg4IARI0agd+/eALgMi4jeXnp6OqZMmYKUlBTk5eWhRo0aSExMhKqqKhMavaEXB1Rr1qwJXV1dhIeHY/DgwZg9ezbc3d2lweolS5bg999/x8OHD3Ho0CG261QhMUEgVSYMuum9iIqKwvjx4+Hr64vjx4/j1KlTyM3NRXBwMJo2barQgZDvhAUEBMDFxQUhISH48ccfy/MtVCjy93P79u2YOHEi/Pz8cP78eURHR8PAwADTp0/HzZs3YW9vD0NDQxQVFaFGjRo4c+YMlJWV2Rkmonfm5s2bSExMRFZWFoYOHQplZWVu+/gOcECVPjXccYUqCwbd9E68+IW+e/dupKSkYPr06QCAyMhI+Pr64vHjxwgJCZECb+D/lxD5+/vDzc0NoaGh6Nev34d/E5VAXFwcdu7cCXNzc4wYMQIAsG/fPnh7e6NmzZrw9vZGQUEBDh48iIKCAowZMwbKysocFSai94ptzJvhgCp9ypggkCoTBt301uS/0P38/HD9+nWkpqaiZcuWmDVrllQuKioKvr6+yM3Nxdq1a2FqaiqdCwgIwNSpUxEcHAx7e/sP/RYqhVu3bqFjx464c+cOZs+eDVdXV+lcZGQkli1bBk1NTbi5ucHKyko6x84wEdHHjQOq9KngjitUWfFTSm+lOHkLAMyYMQNeXl44deoUkpKS4OPjg7S0NKlsr1694OLigtzcXPj4+EjHV65ciYkTJyIkJIQB91vQ19fHrl27oK+vj+joaFy4cEE617t3b0ydOhWpqaklMtiyQ0ZE9PG6desWRowYgdDQUDx69Eg63qdPH0yaNAmPHj3C5MmTkZ2djTFjxuC///0vA26qMJggkD4VDLrprRSPLmZlZSEvLw+xsbE4fPgwwsPDYWlpie7duysE3ra2tvD19cXq1asBPG9sb968iaCgIHz77bfl8h4qk5YtW2Lbtm24e/cufH19kZycLJ3r2bMn/P398fPPP5djDYmIqCw4oEqVFXdcoU8Jl5fTW9uyZQsGDBgAU1NTbNmyRUpokZiYCA8PD1y+fBkHDx5Eo0aNFK7jKPz7c/bsWTg6OsLS0hKurq4KS/kB3nsiooomKSkJw4cPR5s2beDi4oLmzZtL5+Lj49GuXTu261QhMUEgfQoYdFOZvdjIXb9+HdOmTcOOHTtw8OBBdOrUSTqXmJgIT09PHDlyBKmpqTAwMCiPKn+Szp49CycnJxgaGmLx4sUlBj2IiKhi4YAqVQZMEEifIgbd9MZiY2PRvn17aGpq4u+//8aYMWNw4sQJxMXF4bPPPpPKJSQkYPv27ViyZAk7Ax/YyZMnsXbtWgQFBXE0mIioEuCAKlUWTBBInxIG3fRGrl69iiZNmmDkyJFYtmyZtG+io6MjTp8+jaNHj8LY2LjEdWwoP7zi0WAuwyIiqhw4oEoVHXdcoU8NW2p6I0ZGRoiKikJ4eDjc3Nzw6NEjGBgYICgoCJ9//jm+/PJL/PHHHyWuY0P54clkMggh2DEjIqok2rZti3Xr1jFzM1VYTBBInxrOdNMbKZ49jY2NhZ2dHYYPH45FixahVq1ayMzMRL9+/aCjo4PIyMjyrioREVGlxOdaqaJjgkD6VDDoplJbsGABcnJyMHfuXCgpKUlf9jExMbCzs8PYsWMxc+ZMaGlp4e7du9DS0uLsKhERERG9EhME0qeAERG90otL1tTU1LBgwQIsXLgQRUVF0nPCtra2mDhxInx8fDB58mQ8fvwY2traXPZGRERERK9lYWGBoKAgnDt3Dl5eXkhPT1c4z4CbKgPl8q4AfZzkk26lpaWhZs2acHV1ha6uLgYPHoyioiK4u7tLDaG2tjb69OmD9PR0qKmpSa/DmW4iIiIieh0LCwv4+flh7dq1MDQ0LO/qEL1zXF5OrzV9+nTs3bsXWVlZGDFiBIYOHYrk5GQMGDAAnp6eGD58OPT09ODg4IARI0agd+/eAEru5U1ERERE9DrccYUqKwbdpEC+kdu+fTsmTpwIPz8/nD9/HtHR0TAwMMD06dNx8+ZN2Nvbw9DQEEVFRahRowbOnDkDZWVlJnYhIiIiojfCfiRVRgy66aXi4uKwc+dOmJubY8SIEQCAffv2wdvbGzVr1oS3tzcKCgpw8OBBFBQUYMyYMVBWVmayCyIiIiIiIjkMuqmEW7duoWPHjrhz5w5mz54NV1dX6VxkZCSWLVsGTU1NuLm5wcrKSjrHgJuIiIiIiEgRH5agEvT19bFr1y7o6+sjOjoaFy5ckM717t0bU6dORWpqKiIiIhSuY8BNRERERESkiDPd9EpJSUkYPnw42rRpAxcXFzRv3lw6Fx8fj3bt2jHQJiIiIiIieg0G3fRaZ8+ehaOjIywtLeHq6gpTU1OF81xSTkRERERE9GoMuulfnT17Fk5OTjA0NMTixYvRqFGj8q4SERERERFRhcBnuulfWVhYwM/PDzVr1oShoWF5V4eIiIiIiKjC4Ew3lVrxvonye3kTERERERHRqzHopjIpDryJiIiIiIjo33G6ksqEATcREREREVHpMegmIiIiIiIiek8YdBMRERERERG9Jwy6iYiIiIiIiN4TBt1ERERERERE7wmDbiIiIiIiIqL3hEE3ERFROZk1axZatWpV3tV4qY+5bkRERBUJg24iIqI3cOfOHYwZMwYNGjRA1apVoa+vDxsbGxw7dqzc6vQhA+Vr165BJpOV+Bk0aNA7+xsNGzbEihUr3tnrERERlQfl8q4AERFRRWRvb4/8/HysX78eRkZGyMrKwsGDB5GdnV3eVfugfv31VzRv3lz6XU1NrRxr83L5+flQVVUt72oQEdEnijPdREREZfTgwQMcPXoUixYtwpdffglDQ0O0bdsW7u7u+OabbxTKOTo6QkdHBxoaGujWrRuSkpJe+9pBQUEwMTFBtWrV8Nlnn2H16tUK5//++284ODhAS0sLNWrUQJs2bXDixAmEhoZi9uzZSEpKkmadQ0NDS12PhQsXQk9PDzVr1sTIkSORl5dXqntRp04d6OvrSz+1atUq1d9MS0uDnZ0d9PT0oK6ujs8//xy//vqrdL5r167IyMjAxIkTpfcDvHw2f8WKFWjYsKH0+7Bhw9C3b1/MmzcPBgYGMDY2BgD89ddf+OGHH6CpqQktLS3Y2dnh2rVr0nVHjhxB27ZtUaNGDWhqauKLL75ARkZGqe4DERHRqzDoJiIiKiN1dXWoq6tjz549ePr06SvLff/997h9+zZiYmKQmJiI1q1b46uvvsK9e/deWn7Tpk2YOXMm5s2bh5SUFMyfPx+enp5Yv349ACAnJwddunTBjRs3EBERgaSkJLi5uaGoqAj9+/fH5MmT0bx5c9y8eRM3b95E//79S1WPbdu2YdasWZg/fz5Onz6NunXrlgj2y+rf/mZOTg569uyJgwcP4uzZs+jRowf69OmD69evAwB27dqFevXqYc6cOdL7KYuDBw/i0qVLOHDgACIjI/Hs2TPY2NigZs2aOHr0KI4dOwZ1dXX06NED+fn5KCgoQN++fdGlSxecP38eCQkJGD16tBTsExERvTFBREREZbZjxw5Ru3ZtUa1aNdGhQwfh7u4ukpKSpPNHjx4VGhoaIi8vT+G6xo0bC39/fyGEEF5eXsLc3FzhXHh4uEL5uXPnCisrKyGEEP7+/qJmzZoiOzv7pXV68fVKWw8rKysxduxYhfPt2rUr8Vry0tPTBQChpqYmatSoIf2cOXOmVH/zZZo3by58fX2l3w0NDYW3t/e/vkdvb29haGgo/T506FChp6cnnj59Kh3bsGGDMDY2FkVFRdKxp0+fCjU1NfHLL7+I7OxsAUAcOXLklfUjIiJ6E5zpJiIiegP29vbIzMxEREQEevTogSNHjqB169bSku6kpCTk5OSgTp060sy4uro60tPTkZaWVuL1Hj9+jLS0NIwcOVKh/M8//yyVP3fuHCwsLKClpVXqepamHikpKWjXrp3CdVZWVqV6/a1bt+LcuXPSj6mpaan+Zk5ODqZMmQITExNoampCXV0dKSkp0kz32zIzM1N4jjspKQmpqamoWbOmVB8tLS3k5eUhLS0NWlpaGDZsGGxsbNCnTx+sXLmyzLPrREREL8NEakRERG+oWrVqsLa2hrW1NTw9PeHo6AgvLy8MGzYMOTk5qFu3Lo4cOVLiOk1NzRLHcnJyAACBgYElAuAqVaoAeLMkZWWtR1nVr18fTZo0KfPfnDJlCg4cOIClS5eiSZMmUFNTw3fffYf8/PzX/j0lJSUIIRSOPXv2rES5GjVqlKiTpaUlNm3aVKKsjo4OACAkJAQTJkxAbGwstm7dihkzZuDAgQNo3779a+tERET0Ogy6iYiI3hFTU1Ps2bMHANC6dWvcunULysrKCkm+XkVPTw8GBga4evUqBg4c+NIyLVu2RFBQEO7du/fS2W5VVVUUFhYqHCtNPUxMTHDixAkMGTJEOnb8+PF/rfOrlOZvHjt2DMOGDUO/fv0APA+K5ZOaver96Ojo4NatWxBCSM9bnzt3rlR12rp1K3R1daGhofHKchYWFrCwsIC7uzusrKwQHh7OoJuIiN4Kl5cTERGVUXZ2Nrp164aNGzfi/PnzSE9Px/bt27F48WLY2dkBALp37w4rKyv07dsX+/fvx7Vr1xAfHw8PDw+cPn36pa87e/ZsLFiwAD4+Prh8+TIuXLiAkJAQLF++HADg4OAAfX199O3bF8eOHcPVq1exc+dOJCQkAHi+r3V6ejrOnTuHu3fv4unTp6Wqh4uLC4KDgxESEoLLly/Dy8sLycnJb3x/SvM3mzZtil27duHcuXNISkrCgAEDUFRUpPA6DRs2RFxcHG7cuIG7d+8CeJ7V/M6dO1i8eDHS0tKwatUqxMTE/GudBg4cCG1tbdjZ2eHo0aNIT0/HkSNHMGHCBPz9999IT0+Hu7s7EhISkJGRgf379+PKlSswMTF54/tAREQEMOgmIiIqM3V1dbRr1w7e3t7o3LkzWrRoAU9PT4waNQp+fn4AAJlMhujoaHTu3BnDhw9Hs2bN8OOPPyIjIwN6enovfV1HR0cEBQUhJCQEZmZm6NKlC0JDQ9GoUSMAz2d+9+/fD11dXfTs2RNmZmZYuHChtPzc3t4ePXr0wJdffgkdHR1s3ry5VPXo378/PD094ebmBktLS2RkZGDMmDFvfH9K8zeXL1+O2rVro0OHDujTpw9sbGzQunVrhdeZM2cOrl27hsaNG0tLwE1MTLB69WqsWrUK5ubmOHnyJKZMmfKvdapevTri4uLQoEEDfPvttzAxMZG2RtPQ0ED16tVx8eJF2Nvbo1mzZhg9ejTGjRsHJyenN74PREREACATLz4YRURERERERETvBGe6iYiIiIiIiN4TBt1ERERERERE7wmDbiIiIiIiIqL3hEE3ERERERER0XvCoJuIiIiIiIjoPWHQTURERERERPSeMOgmIiIiIiIiek8YdBMRERERERG9Jwy6iYiIiIiIiN4TBt1ERERERERE7wmDbiIiIiIiIqL3hEE3ERERERER0Xvyf0r9PpOlqpHyAAAAAElFTkSuQmCC",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Bar Plot for Stability Scores\n",
    "plt.figure(figsize=(10, 6))\n",
    "sorted_scores = sorted(stability_scores.items(), key=lambda x: x[1], reverse=True)\n",
    "features, scores = zip(*sorted_scores)\n",
    "colors = cm.viridis(np.linspace(0, 1, len(features)))\n",
    "plt.bar(features, scores, color=colors)\n",
    "plt.axhline(y=stability_threshold, color='r', linestyle='--', label='Stability Threshold')\n",
    "plt.xticks(rotation=45, ha='right')\n",
    "plt.xlabel('Features')\n",
    "plt.ylabel('Stability Score')\n",
    "plt.title('Feature Stability Scores')\n",
    "plt.legend()\n",
    "plt.tight_layout()\n",
    "plt.show()\n",
    "\n",
    "# Feature Importance Plot\n",
    "plt.figure(figsize=(10, 6))\n",
    "importances = rf.feature_importances_\n",
    "sorted_indices = np.argsort(importances)[::-1]\n",
    "colors = cm.viridis(np.linspace(0, 1, len(selected_features)))\n",
    "plt.bar(range(len(selected_features)), importances[sorted_indices], color=colors)\n",
    "plt.xticks(range(len(selected_features)), [selected_features[i] for i in sorted_indices], rotation=45, ha='right')\n",
    "plt.xlabel('Selected Features')\n",
    "plt.ylabel('Feature Importance')\n",
    "plt.title('Feature Importances of Selected Features')\n",
    "plt.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "7f3b7856-3c70-4532-9b80-9bbba7a9faa3",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 600x400 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot the cross-validation accuracy scores\n",
    "plt.figure(figsize=(6, 4))\n",
    "sns.boxplot(data=cv_scores, orient='h')\n",
    "plt.xlabel('Cross-validation Accuracy', fontsize=12)\n",
    "plt.title('Distribution of Cross-validation Accuracy Scores', fontsize=14)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a98da76f-6813-456a-8b12-2d68aadedd14",
   "metadata": {},
   "source": [
    "## III. Model Development"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c554ce1b-41cc-4aa6-aa33-064930ce876b",
   "metadata": {},
   "source": [
    "## A. Traditional Model Selection"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "ede3b91d-0c90-4218-bb7c-e21d0a967fcd",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Evaluating Logistic Regression...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/tramngo/Desktop/src/pandas_env/lib/python3.9/site-packages/sklearn/linear_model/_logistic.py:469: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  n_iter_i = _check_optimize_result(\n",
      "/Users/tramngo/Desktop/src/pandas_env/lib/python3.9/site-packages/sklearn/linear_model/_logistic.py:469: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  n_iter_i = _check_optimize_result(\n",
      "/Users/tramngo/Desktop/src/pandas_env/lib/python3.9/site-packages/sklearn/linear_model/_logistic.py:469: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  n_iter_i = _check_optimize_result(\n",
      "/Users/tramngo/Desktop/src/pandas_env/lib/python3.9/site-packages/sklearn/linear_model/_logistic.py:469: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  n_iter_i = _check_optimize_result(\n",
      "/Users/tramngo/Desktop/src/pandas_env/lib/python3.9/site-packages/sklearn/linear_model/_logistic.py:469: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  n_iter_i = _check_optimize_result(\n",
      "/Users/tramngo/Desktop/src/pandas_env/lib/python3.9/site-packages/sklearn/linear_model/_logistic.py:469: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  n_iter_i = _check_optimize_result(\n",
      "/Users/tramngo/Desktop/src/pandas_env/lib/python3.9/site-packages/sklearn/linear_model/_logistic.py:469: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  n_iter_i = _check_optimize_result(\n",
      "/Users/tramngo/Desktop/src/pandas_env/lib/python3.9/site-packages/sklearn/linear_model/_logistic.py:469: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  n_iter_i = _check_optimize_result(\n",
      "/Users/tramngo/Desktop/src/pandas_env/lib/python3.9/site-packages/sklearn/linear_model/_logistic.py:469: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  n_iter_i = _check_optimize_result(\n",
      "/Users/tramngo/Desktop/src/pandas_env/lib/python3.9/site-packages/sklearn/linear_model/_logistic.py:469: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  n_iter_i = _check_optimize_result(\n",
      "/Users/tramngo/Desktop/src/pandas_env/lib/python3.9/site-packages/sklearn/linear_model/_logistic.py:469: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  n_iter_i = _check_optimize_result(\n",
      "/Users/tramngo/Desktop/src/pandas_env/lib/python3.9/site-packages/sklearn/linear_model/_logistic.py:469: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  n_iter_i = _check_optimize_result(\n",
      "/Users/tramngo/Desktop/src/pandas_env/lib/python3.9/site-packages/sklearn/linear_model/_logistic.py:469: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  n_iter_i = _check_optimize_result(\n",
      "/Users/tramngo/Desktop/src/pandas_env/lib/python3.9/site-packages/sklearn/linear_model/_logistic.py:469: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  n_iter_i = _check_optimize_result(\n",
      "/Users/tramngo/Desktop/src/pandas_env/lib/python3.9/site-packages/sklearn/linear_model/_logistic.py:469: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  n_iter_i = _check_optimize_result(\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best parameters: {'C': 0.1}\n",
      "Accuracy: 0.6333\n",
      "Precision: 0.6522\n",
      "Recall: 0.8333\n",
      "F1-score: 0.7317\n",
      "AUC-ROC: 0.7164\n",
      "\n",
      "Evaluating Decision Tree...\n",
      "Best parameters: {'max_depth': 3, 'min_samples_split': 5}\n",
      "Accuracy: 0.5500\n",
      "Precision: 0.6452\n",
      "Recall: 0.5556\n",
      "F1-score: 0.5970\n",
      "AUC-ROC: 0.5822\n",
      "\n",
      "Evaluating Random Forest...\n",
      "Best parameters: {'max_depth': 5, 'n_estimators': 200}\n",
      "Accuracy: 0.6500\n",
      "Precision: 0.6471\n",
      "Recall: 0.9167\n",
      "F1-score: 0.7586\n",
      "AUC-ROC: 0.6933\n",
      "\n",
      "Evaluating Gradient Boosting...\n",
      "Best parameters: {'learning_rate': 0.01, 'n_estimators': 100}\n",
      "Accuracy: 0.6667\n",
      "Precision: 0.6538\n",
      "Recall: 0.9444\n",
      "F1-score: 0.7727\n",
      "AUC-ROC: 0.7384\n",
      "\n",
      "Evaluating SVM...\n",
      "Best parameters: {'C': 0.1, 'kernel': 'linear'}\n",
      "Accuracy: 0.6500\n",
      "Precision: 0.6667\n",
      "Recall: 0.8333\n",
      "F1-score: 0.7407\n",
      "AUC-ROC: 0.6701\n",
      "\n",
      "Best model: GradientBoostingClassifier(learning_rate=0.01)\n"
     ]
    }
   ],
   "source": [
    "# Define the models to evaluate\n",
    "models = [\n",
    "    ('Logistic Regression', LogisticRegression()),\n",
    "    ('Decision Tree', DecisionTreeClassifier()),\n",
    "    ('Random Forest', RandomForestClassifier()),\n",
    "    ('Gradient Boosting', GradientBoostingClassifier()),\n",
    "    ('SVM', SVC(probability=True))\n",
    "]\n",
    "\n",
    "# Define the hyperparameters to tune for each model\n",
    "hyperparameters = {\n",
    "    'Logistic Regression': {'C': [0.1, 1, 10]},\n",
    "    'Decision Tree': {'max_depth': [3, 5, 7], 'min_samples_split': [2, 5, 10]},\n",
    "    'Random Forest': {'n_estimators': [50, 100, 200], 'max_depth': [3, 5, 7]},\n",
    "    'Gradient Boosting': {'n_estimators': [50, 100, 200], 'learning_rate': [0.01, 0.1, 1]},\n",
    "    'SVM': {'C': [0.1, 1, 10], 'kernel': ['linear', 'rbf']}\n",
    "}\n",
    "\n",
    "# Evaluate each model using cross-validation and grid search\n",
    "best_model = None\n",
    "best_score = 0\n",
    "\n",
    "# For plotting purposes\n",
    "results = []\n",
    "\n",
    "for name, model in models:\n",
    "    print(f\"Evaluating {name}...\")\n",
    "    \n",
    "    # Perform grid search to find the best hyperparameters\n",
    "    grid_search = GridSearchCV(model, hyperparameters[name], cv=5, scoring='roc_auc', n_jobs=-1)\n",
    "    grid_search.fit(X_train, y_train)\n",
    "    \n",
    "    # Evaluate the model on the testing set\n",
    "    y_pred = grid_search.predict(X_test)\n",
    "    accuracy = accuracy_score(y_test, y_pred)\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",
    "    auc_roc = roc_auc_score(y_test, grid_search.predict_proba(X_test)[:, 1])\n",
    "\n",
    "    # Plot the evaluation metrics for the current model\n",
    "    results.append([accuracy, precision, recall, f1, auc_roc])\n",
    "    \n",
    "    print(f\"Best parameters: {grid_search.best_params_}\")\n",
    "    print(f\"Accuracy: {accuracy:.4f}\")\n",
    "    print(f\"Precision: {precision:.4f}\")\n",
    "    print(f\"Recall: {recall:.4f}\")\n",
    "    print(f\"F1-score: {f1:.4f}\")\n",
    "    print(f\"AUC-ROC: {auc_roc:.4f}\")\n",
    "    print()\n",
    "    \n",
    "    # Update the best model if the current model has a higher AUC-ROC score\n",
    "    if auc_roc > best_score:\n",
    "        best_model = grid_search.best_estimator_\n",
    "        best_score = auc_roc\n",
    "\n",
    "print(f\"Best model: {best_model}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "24912be5-4aa1-4a5c-9956-47f00a239b22",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "model_names = [name for name, _ in models]\n",
    "accuracy_scores = [result[0] for result in results]\n",
    "precision_scores = [result[1] for result in results]\n",
    "recall_scores = [result[2] for result in results]\n",
    "\n",
    "# Create a dataset for plotting\n",
    "data = pd.DataFrame({\n",
    "    'Model': model_names * 3,\n",
    "    'Metric': ['Accuracy'] * len(model_names) + ['Precision'] * len(model_names) + ['Recall'] * len(model_names),\n",
    "    'Score': accuracy_scores + precision_scores + recall_scores\n",
    "})\n",
    "\n",
    "# Set the figure size\n",
    "plt.figure(figsize=(10, 6))\n",
    "\n",
    "# Create the bar plot using Seaborn\n",
    "sns.barplot(x='Model', y='Score', hue='Metric', data=data, palette='rocket')\n",
    "\n",
    "# Set the title and axis labels\n",
    "plt.title('Accuracy, Precision, and Recall Scores', fontsize=16)\n",
    "plt.xlabel('Model', fontsize=14)\n",
    "plt.ylabel('Score', fontsize=14)\n",
    "\n",
    "# Rotate the x-axis labels for better visibility\n",
    "plt.xticks(rotation=45, ha='right')\n",
    "\n",
    "# Adjust the legend position\n",
    "plt.legend(bbox_to_anchor=(1.02, 1), loc='upper left', borderaxespad=0)\n",
    "\n",
    "# Show the plot\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "df6300cc-ff2b-450a-82a2-0d76bf50a3a8",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Extract the AUC-ROC scores from the results\n",
    "auc_roc_scores = [result[-1] for result in results]\n",
    "\n",
    "# Create a dataset for plotting\n",
    "data = pd.DataFrame({\n",
    "    'Model': model_names,\n",
    "    'AUC-ROC Score': auc_roc_scores\n",
    "})\n",
    "\n",
    "# Set the figure size\n",
    "plt.figure(figsize=(10, 6))\n",
    "\n",
    "# Create the bar plot using Seaborn\n",
    "sns.barplot(x='Model', y='AUC-ROC Score', data=data, palette='rocket')\n",
    "\n",
    "# Set the title and axis labels\n",
    "plt.title('AUC-ROC Scores for Different Models', fontsize=16)\n",
    "plt.xlabel('Model', fontsize=14)\n",
    "plt.ylabel('AUC-ROC Score', fontsize=14)\n",
    "\n",
    "# Rotate the x-axis labels for better visibility\n",
    "plt.xticks(rotation=45, ha='right')\n",
    "\n",
    "# Show the plot\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "c49c19cb-4545-42d6-8524-90263471b520",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(GradientBoostingClassifier(learning_rate=0.01), 0.7384259259259259)"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "best_model, best_score"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "76b0b295-58dc-45d7-a4a9-8075332315a9",
   "metadata": {},
   "source": [
    "## B. Markov Model"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1c0e589e-bb4f-4f20-9f44-45a99203df96",
   "metadata": {},
   "source": [
    "### Step 1: Calculate count & probability for each count => Define potential values for each variable's potential function"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "ef89842e-0778-4687-a7c5-a5d64106cc9d",
   "metadata": {},
   "outputs": [],
   "source": [
    "def calculate_variable_probabilities(variable):\n",
    "    value_counts = variable.value_counts()\n",
    "    total_count = variable.count()\n",
    "    \n",
    "    # Create a dictionary to store the count and probability of each unique value\n",
    "    probabilities = {}\n",
    "    \n",
    "    # Iterate over each unique value and calculate its probability\n",
    "    for value, count in value_counts.items():\n",
    "        probability = count / total_count\n",
    "        probabilities[value] = {'count': count, 'probability': probability}\n",
    "    \n",
    "    return probabilities"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "df7cd613-5acf-451f-a723-fdb4e212cb01",
   "metadata": {},
   "outputs": [],
   "source": [
    "X_selected = X[selected_features]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "5c2355ce-9ae8-4c49-9e2e-9be296aaa03e",
   "metadata": {},
   "outputs": [],
   "source": [
    "X_selected.to_csv(\"X_selected.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "9247c6cf-8b0f-4995-99ea-e174bc9324d4",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "age:\n",
      " {55: {'count': 66, 'probability': 0.2214765100671141}, 65: {'count': 62, 'probability': 0.2080536912751678}, 75: {'count': 61, 'probability': 0.20469798657718122}, 85: {'count': 42, 'probability': 0.14093959731543623}, 45: {'count': 40, 'probability': 0.1342281879194631}, 35: {'count': 17, 'probability': 0.05704697986577181}, 25: {'count': 6, 'probability': 0.020134228187919462}, 15: {'count': 4, 'probability': 0.013422818791946308}}\n",
      "time_in_hospital:\n",
      " {4: {'count': 51, 'probability': 0.17114093959731544}, 6: {'count': 37, 'probability': 0.12416107382550336}, 2: {'count': 36, 'probability': 0.12080536912751678}, 5: {'count': 34, 'probability': 0.11409395973154363}, 3: {'count': 34, 'probability': 0.11409395973154363}, 7: {'count': 23, 'probability': 0.07718120805369127}, 1: {'count': 17, 'probability': 0.05704697986577181}, 9: {'count': 15, 'probability': 0.050335570469798654}, 8: {'count': 15, 'probability': 0.050335570469798654}, 10: {'count': 14, 'probability': 0.04697986577181208}, 12: {'count': 8, 'probability': 0.026845637583892617}, 11: {'count': 6, 'probability': 0.020134228187919462}, 14: {'count': 5, 'probability': 0.016778523489932886}, 13: {'count': 3, 'probability': 0.010067114093959731}}\n",
      "num_lab_procedures:\n",
      " {72: {'count': 10, 'probability': 0.03355704697986577}, 58: {'count': 10, 'probability': 0.03355704697986577}, 65: {'count': 10, 'probability': 0.03355704697986577}, 63: {'count': 9, 'probability': 0.030201342281879196}, 56: {'count': 9, 'probability': 0.030201342281879196}, 49: {'count': 9, 'probability': 0.030201342281879196}, 57: {'count': 8, 'probability': 0.026845637583892617}, 68: {'count': 8, 'probability': 0.026845637583892617}, 67: {'count': 8, 'probability': 0.026845637583892617}, 54: {'count': 8, 'probability': 0.026845637583892617}, 77: {'count': 8, 'probability': 0.026845637583892617}, 51: {'count': 7, 'probability': 0.02348993288590604}, 79: {'count': 7, 'probability': 0.02348993288590604}, 75: {'count': 7, 'probability': 0.02348993288590604}, 53: {'count': 7, 'probability': 0.02348993288590604}, 70: {'count': 7, 'probability': 0.02348993288590604}, 60: {'count': 7, 'probability': 0.02348993288590604}, 52: {'count': 7, 'probability': 0.02348993288590604}, 69: {'count': 7, 'probability': 0.02348993288590604}, 61: {'count': 7, 'probability': 0.02348993288590604}, 55: {'count': 6, 'probability': 0.020134228187919462}, 64: {'count': 6, 'probability': 0.020134228187919462}, 43: {'count': 6, 'probability': 0.020134228187919462}, 71: {'count': 6, 'probability': 0.020134228187919462}, 62: {'count': 6, 'probability': 0.020134228187919462}, 80: {'count': 5, 'probability': 0.016778523489932886}, 59: {'count': 5, 'probability': 0.016778523489932886}, 46: {'count': 5, 'probability': 0.016778523489932886}, 50: {'count': 5, 'probability': 0.016778523489932886}, 81: {'count': 5, 'probability': 0.016778523489932886}, 76: {'count': 5, 'probability': 0.016778523489932886}, 66: {'count': 5, 'probability': 0.016778523489932886}, 74: {'count': 5, 'probability': 0.016778523489932886}, 84: {'count': 4, 'probability': 0.013422818791946308}, 73: {'count': 4, 'probability': 0.013422818791946308}, 45: {'count': 4, 'probability': 0.013422818791946308}, 48: {'count': 4, 'probability': 0.013422818791946308}, 78: {'count': 4, 'probability': 0.013422818791946308}, 41: {'count': 4, 'probability': 0.013422818791946308}, 88: {'count': 3, 'probability': 0.010067114093959731}, 47: {'count': 3, 'probability': 0.010067114093959731}, 85: {'count': 2, 'probability': 0.006711409395973154}, 44: {'count': 2, 'probability': 0.006711409395973154}, 40: {'count': 2, 'probability': 0.006711409395973154}, 93: {'count': 2, 'probability': 0.006711409395973154}, 42: {'count': 2, 'probability': 0.006711409395973154}, 39: {'count': 2, 'probability': 0.006711409395973154}, 87: {'count': 2, 'probability': 0.006711409395973154}, 91: {'count': 2, 'probability': 0.006711409395973154}, 38: {'count': 2, 'probability': 0.006711409395973154}, 96: {'count': 2, 'probability': 0.006711409395973154}, 98: {'count': 2, 'probability': 0.006711409395973154}, 82: {'count': 2, 'probability': 0.006711409395973154}, 83: {'count': 2, 'probability': 0.006711409395973154}, 95: {'count': 2, 'probability': 0.006711409395973154}, 31: {'count': 1, 'probability': 0.003355704697986577}, 97: {'count': 1, 'probability': 0.003355704697986577}, 86: {'count': 1, 'probability': 0.003355704697986577}, 106: {'count': 1, 'probability': 0.003355704697986577}, 102: {'count': 1, 'probability': 0.003355704697986577}, 105: {'count': 1, 'probability': 0.003355704697986577}, 32: {'count': 1, 'probability': 0.003355704697986577}, 36: {'count': 1, 'probability': 0.003355704697986577}, 94: {'count': 1, 'probability': 0.003355704697986577}, 90: {'count': 1, 'probability': 0.003355704697986577}}\n",
      "num_medications:\n",
      " {11: {'count': 25, 'probability': 0.08389261744966443}, 13: {'count': 22, 'probability': 0.0738255033557047}, 16: {'count': 21, 'probability': 0.07046979865771812}, 19: {'count': 20, 'probability': 0.06711409395973154}, 14: {'count': 16, 'probability': 0.053691275167785234}, 12: {'count': 15, 'probability': 0.050335570469798654}, 9: {'count': 15, 'probability': 0.050335570469798654}, 6: {'count': 14, 'probability': 0.04697986577181208}, 17: {'count': 13, 'probability': 0.0436241610738255}, 3: {'count': 11, 'probability': 0.03691275167785235}, 8: {'count': 11, 'probability': 0.03691275167785235}, 15: {'count': 10, 'probability': 0.03355704697986577}, 20: {'count': 10, 'probability': 0.03355704697986577}, 10: {'count': 9, 'probability': 0.030201342281879196}, 7: {'count': 8, 'probability': 0.026845637583892617}, 5: {'count': 8, 'probability': 0.026845637583892617}, 21: {'count': 8, 'probability': 0.026845637583892617}, 26: {'count': 7, 'probability': 0.02348993288590604}, 4: {'count': 6, 'probability': 0.020134228187919462}, 25: {'count': 6, 'probability': 0.020134228187919462}, 23: {'count': 5, 'probability': 0.016778523489932886}, 22: {'count': 5, 'probability': 0.016778523489932886}, 34: {'count': 4, 'probability': 0.013422818791946308}, 18: {'count': 4, 'probability': 0.013422818791946308}, 28: {'count': 4, 'probability': 0.013422818791946308}, 2: {'count': 3, 'probability': 0.010067114093959731}, 27: {'count': 3, 'probability': 0.010067114093959731}, 24: {'count': 3, 'probability': 0.010067114093959731}, 30: {'count': 2, 'probability': 0.006711409395973154}, 1: {'count': 2, 'probability': 0.006711409395973154}, 35: {'count': 2, 'probability': 0.006711409395973154}, 32: {'count': 2, 'probability': 0.006711409395973154}, 29: {'count': 2, 'probability': 0.006711409395973154}, 33: {'count': 1, 'probability': 0.003355704697986577}, 31: {'count': 1, 'probability': 0.003355704697986577}}\n",
      "number_inpatient:\n",
      " {0: {'count': 204, 'probability': 0.6845637583892618}, 1: {'count': 43, 'probability': 0.14429530201342283}, 2: {'count': 26, 'probability': 0.087248322147651}, 3: {'count': 15, 'probability': 0.050335570469798654}, 4: {'count': 4, 'probability': 0.013422818791946308}, 6: {'count': 2, 'probability': 0.006711409395973154}, 7: {'count': 2, 'probability': 0.006711409395973154}, 5: {'count': 1, 'probability': 0.003355704697986577}, 9: {'count': 1, 'probability': 0.003355704697986577}}\n",
      "diag_1:\n",
      " {47: {'count': 21, 'probability': 0.07046979865771812}, 35: {'count': 19, 'probability': 0.06375838926174497}, 72: {'count': 19, 'probability': 0.06375838926174497}, 32: {'count': 14, 'probability': 0.04697986577181208}, 82: {'count': 12, 'probability': 0.040268456375838924}, 4: {'count': 12, 'probability': 0.040268456375838924}, 46: {'count': 10, 'probability': 0.03355704697986577}, 48: {'count': 9, 'probability': 0.030201342281879196}, 26: {'count': 9, 'probability': 0.030201342281879196}, 37: {'count': 9, 'probability': 0.030201342281879196}, 38: {'count': 8, 'probability': 0.026845637583892617}, 9: {'count': 7, 'probability': 0.02348993288590604}, 13: {'count': 7, 'probability': 0.02348993288590604}, 11: {'count': 6, 'probability': 0.020134228187919462}, 30: {'count': 5, 'probability': 0.016778523489932886}, 65: {'count': 5, 'probability': 0.016778523489932886}, 34: {'count': 5, 'probability': 0.016778523489932886}, 8: {'count': 4, 'probability': 0.013422818791946308}, 70: {'count': 4, 'probability': 0.013422818791946308}, 68: {'count': 4, 'probability': 0.013422818791946308}, 27: {'count': 4, 'probability': 0.013422818791946308}, 59: {'count': 3, 'probability': 0.010067114093959731}, 15: {'count': 3, 'probability': 0.010067114093959731}, 44: {'count': 3, 'probability': 0.010067114093959731}, 5: {'count': 3, 'probability': 0.010067114093959731}, 21: {'count': 3, 'probability': 0.010067114093959731}, 57: {'count': 3, 'probability': 0.010067114093959731}, 85: {'count': 2, 'probability': 0.006711409395973154}, 23: {'count': 2, 'probability': 0.006711409395973154}, 7: {'count': 2, 'probability': 0.006711409395973154}, 54: {'count': 2, 'probability': 0.006711409395973154}, 41: {'count': 2, 'probability': 0.006711409395973154}, 10: {'count': 2, 'probability': 0.006711409395973154}, 36: {'count': 2, 'probability': 0.006711409395973154}, 53: {'count': 2, 'probability': 0.006711409395973154}, 64: {'count': 2, 'probability': 0.006711409395973154}, 89: {'count': 2, 'probability': 0.006711409395973154}, 62: {'count': 2, 'probability': 0.006711409395973154}, 66: {'count': 2, 'probability': 0.006711409395973154}, 73: {'count': 2, 'probability': 0.006711409395973154}, 18: {'count': 2, 'probability': 0.006711409395973154}, 14: {'count': 2, 'probability': 0.006711409395973154}, 1: {'count': 2, 'probability': 0.006711409395973154}, 51: {'count': 2, 'probability': 0.006711409395973154}, 67: {'count': 2, 'probability': 0.006711409395973154}, 3: {'count': 2, 'probability': 0.006711409395973154}, 75: {'count': 2, 'probability': 0.006711409395973154}, 76: {'count': 2, 'probability': 0.006711409395973154}, 6: {'count': 2, 'probability': 0.006711409395973154}, 87: {'count': 1, 'probability': 0.003355704697986577}, 63: {'count': 1, 'probability': 0.003355704697986577}, 49: {'count': 1, 'probability': 0.003355704697986577}, 33: {'count': 1, 'probability': 0.003355704697986577}, 52: {'count': 1, 'probability': 0.003355704697986577}, 2: {'count': 1, 'probability': 0.003355704697986577}, 90: {'count': 1, 'probability': 0.003355704697986577}, 45: {'count': 1, 'probability': 0.003355704697986577}, 71: {'count': 1, 'probability': 0.003355704697986577}, 80: {'count': 1, 'probability': 0.003355704697986577}, 83: {'count': 1, 'probability': 0.003355704697986577}, 61: {'count': 1, 'probability': 0.003355704697986577}, 58: {'count': 1, 'probability': 0.003355704697986577}, 56: {'count': 1, 'probability': 0.003355704697986577}, 91: {'count': 1, 'probability': 0.003355704697986577}, 78: {'count': 1, 'probability': 0.003355704697986577}, 29: {'count': 1, 'probability': 0.003355704697986577}, 84: {'count': 1, 'probability': 0.003355704697986577}, 79: {'count': 1, 'probability': 0.003355704697986577}, 86: {'count': 1, 'probability': 0.003355704697986577}, 20: {'count': 1, 'probability': 0.003355704697986577}, 42: {'count': 1, 'probability': 0.003355704697986577}, 25: {'count': 1, 'probability': 0.003355704697986577}, 55: {'count': 1, 'probability': 0.003355704697986577}, 39: {'count': 1, 'probability': 0.003355704697986577}, 50: {'count': 1, 'probability': 0.003355704697986577}, 74: {'count': 1, 'probability': 0.003355704697986577}, 77: {'count': 1, 'probability': 0.003355704697986577}, 12: {'count': 1, 'probability': 0.003355704697986577}, 0: {'count': 1, 'probability': 0.003355704697986577}, 28: {'count': 1, 'probability': 0.003355704697986577}, 81: {'count': 1, 'probability': 0.003355704697986577}, 22: {'count': 1, 'probability': 0.003355704697986577}, 43: {'count': 1, 'probability': 0.003355704697986577}, 40: {'count': 1, 'probability': 0.003355704697986577}, 60: {'count': 1, 'probability': 0.003355704697986577}, 69: {'count': 1, 'probability': 0.003355704697986577}, 88: {'count': 1, 'probability': 0.003355704697986577}, 24: {'count': 1, 'probability': 0.003355704697986577}, 19: {'count': 1, 'probability': 0.003355704697986577}, 17: {'count': 1, 'probability': 0.003355704697986577}, 16: {'count': 1, 'probability': 0.003355704697986577}, 31: {'count': 1, 'probability': 0.003355704697986577}}\n",
      "diag_2:\n",
      " {6: {'count': 23, 'probability': 0.07718120805369127}, 8: {'count': 20, 'probability': 0.06711409395973154}, 22: {'count': 17, 'probability': 0.05704697986577181}, 46: {'count': 11, 'probability': 0.03691275167785235}, 80: {'count': 10, 'probability': 0.03355704697986577}, 66: {'count': 10, 'probability': 0.03355704697986577}, 53: {'count': 10, 'probability': 0.03355704697986577}, 41: {'count': 10, 'probability': 0.03355704697986577}, 83: {'count': 9, 'probability': 0.030201342281879196}, 7: {'count': 7, 'probability': 0.02348993288590604}, 64: {'count': 7, 'probability': 0.02348993288590604}, 48: {'count': 7, 'probability': 0.02348993288590604}, 98: {'count': 7, 'probability': 0.02348993288590604}, 27: {'count': 5, 'probability': 0.016778523489932886}, 43: {'count': 5, 'probability': 0.016778523489932886}, 16: {'count': 5, 'probability': 0.016778523489932886}, 51: {'count': 4, 'probability': 0.013422818791946308}, 52: {'count': 4, 'probability': 0.013422818791946308}, 63: {'count': 4, 'probability': 0.013422818791946308}, 82: {'count': 4, 'probability': 0.013422818791946308}, 44: {'count': 4, 'probability': 0.013422818791946308}, 21: {'count': 3, 'probability': 0.010067114093959731}, 19: {'count': 3, 'probability': 0.010067114093959731}, 38: {'count': 3, 'probability': 0.010067114093959731}, 9: {'count': 3, 'probability': 0.010067114093959731}, 88: {'count': 3, 'probability': 0.010067114093959731}, 25: {'count': 2, 'probability': 0.006711409395973154}, 31: {'count': 2, 'probability': 0.006711409395973154}, 36: {'count': 2, 'probability': 0.006711409395973154}, 45: {'count': 2, 'probability': 0.006711409395973154}, 69: {'count': 2, 'probability': 0.006711409395973154}, 29: {'count': 2, 'probability': 0.006711409395973154}, 89: {'count': 2, 'probability': 0.006711409395973154}, 91: {'count': 2, 'probability': 0.006711409395973154}, 68: {'count': 2, 'probability': 0.006711409395973154}, 72: {'count': 2, 'probability': 0.006711409395973154}, 67: {'count': 2, 'probability': 0.006711409395973154}, 54: {'count': 2, 'probability': 0.006711409395973154}, 26: {'count': 2, 'probability': 0.006711409395973154}, 77: {'count': 2, 'probability': 0.006711409395973154}, 13: {'count': 2, 'probability': 0.006711409395973154}, 42: {'count': 2, 'probability': 0.006711409395973154}, 65: {'count': 2, 'probability': 0.006711409395973154}, 62: {'count': 2, 'probability': 0.006711409395973154}, 58: {'count': 2, 'probability': 0.006711409395973154}, 47: {'count': 2, 'probability': 0.006711409395973154}, 20: {'count': 2, 'probability': 0.006711409395973154}, 79: {'count': 1, 'probability': 0.003355704697986577}, 34: {'count': 1, 'probability': 0.003355704697986577}, 78: {'count': 1, 'probability': 0.003355704697986577}, 56: {'count': 1, 'probability': 0.003355704697986577}, 84: {'count': 1, 'probability': 0.003355704697986577}, 11: {'count': 1, 'probability': 0.003355704697986577}, 12: {'count': 1, 'probability': 0.003355704697986577}, 94: {'count': 1, 'probability': 0.003355704697986577}, 101: {'count': 1, 'probability': 0.003355704697986577}, 86: {'count': 1, 'probability': 0.003355704697986577}, 87: {'count': 1, 'probability': 0.003355704697986577}, 104: {'count': 1, 'probability': 0.003355704697986577}, 96: {'count': 1, 'probability': 0.003355704697986577}, 61: {'count': 1, 'probability': 0.003355704697986577}, 30: {'count': 1, 'probability': 0.003355704697986577}, 1: {'count': 1, 'probability': 0.003355704697986577}, 0: {'count': 1, 'probability': 0.003355704697986577}, 76: {'count': 1, 'probability': 0.003355704697986577}, 99: {'count': 1, 'probability': 0.003355704697986577}, 37: {'count': 1, 'probability': 0.003355704697986577}, 57: {'count': 1, 'probability': 0.003355704697986577}, 55: {'count': 1, 'probability': 0.003355704697986577}, 75: {'count': 1, 'probability': 0.003355704697986577}, 15: {'count': 1, 'probability': 0.003355704697986577}, 97: {'count': 1, 'probability': 0.003355704697986577}, 102: {'count': 1, 'probability': 0.003355704697986577}, 93: {'count': 1, 'probability': 0.003355704697986577}, 5: {'count': 1, 'probability': 0.003355704697986577}, 103: {'count': 1, 'probability': 0.003355704697986577}, 39: {'count': 1, 'probability': 0.003355704697986577}, 2: {'count': 1, 'probability': 0.003355704697986577}, 85: {'count': 1, 'probability': 0.003355704697986577}, 70: {'count': 1, 'probability': 0.003355704697986577}, 40: {'count': 1, 'probability': 0.003355704697986577}, 35: {'count': 1, 'probability': 0.003355704697986577}, 4: {'count': 1, 'probability': 0.003355704697986577}, 33: {'count': 1, 'probability': 0.003355704697986577}, 23: {'count': 1, 'probability': 0.003355704697986577}, 18: {'count': 1, 'probability': 0.003355704697986577}, 74: {'count': 1, 'probability': 0.003355704697986577}, 60: {'count': 1, 'probability': 0.003355704697986577}, 49: {'count': 1, 'probability': 0.003355704697986577}, 81: {'count': 1, 'probability': 0.003355704697986577}, 100: {'count': 1, 'probability': 0.003355704697986577}, 32: {'count': 1, 'probability': 0.003355704697986577}, 90: {'count': 1, 'probability': 0.003355704697986577}, 92: {'count': 1, 'probability': 0.003355704697986577}, 3: {'count': 1, 'probability': 0.003355704697986577}, 28: {'count': 1, 'probability': 0.003355704697986577}, 59: {'count': 1, 'probability': 0.003355704697986577}, 24: {'count': 1, 'probability': 0.003355704697986577}, 73: {'count': 1, 'probability': 0.003355704697986577}, 50: {'count': 1, 'probability': 0.003355704697986577}, 10: {'count': 1, 'probability': 0.003355704697986577}, 71: {'count': 1, 'probability': 0.003355704697986577}, 17: {'count': 1, 'probability': 0.003355704697986577}, 14: {'count': 1, 'probability': 0.003355704697986577}, 95: {'count': 1, 'probability': 0.003355704697986577}}\n",
      "diag_3:\n",
      " {5: {'count': 26, 'probability': 0.087248322147651}, 37: {'count': 20, 'probability': 0.06711409395973154}, 20: {'count': 16, 'probability': 0.053691275167785234}, 7: {'count': 15, 'probability': 0.050335570469798654}, 42: {'count': 13, 'probability': 0.0436241610738255}, 18: {'count': 11, 'probability': 0.03691275167785235}, 48: {'count': 9, 'probability': 0.030201342281879196}, 76: {'count': 7, 'probability': 0.02348993288590604}, 8: {'count': 7, 'probability': 0.02348993288590604}, 79: {'count': 6, 'probability': 0.020134228187919462}, 45: {'count': 5, 'probability': 0.016778523489932886}, 47: {'count': 5, 'probability': 0.016778523489932886}, 80: {'count': 5, 'probability': 0.016778523489932886}, 61: {'count': 4, 'probability': 0.013422818791946308}, 22: {'count': 4, 'probability': 0.013422818791946308}, 16: {'count': 4, 'probability': 0.013422818791946308}, 17: {'count': 4, 'probability': 0.013422818791946308}, 21: {'count': 4, 'probability': 0.013422818791946308}, 84: {'count': 4, 'probability': 0.013422818791946308}, 86: {'count': 4, 'probability': 0.013422818791946308}, 40: {'count': 4, 'probability': 0.013422818791946308}, 23: {'count': 4, 'probability': 0.013422818791946308}, 35: {'count': 3, 'probability': 0.010067114093959731}, 63: {'count': 3, 'probability': 0.010067114093959731}, 102: {'count': 3, 'probability': 0.010067114093959731}, 73: {'count': 3, 'probability': 0.010067114093959731}, 6: {'count': 3, 'probability': 0.010067114093959731}, 25: {'count': 3, 'probability': 0.010067114093959731}, 39: {'count': 3, 'probability': 0.010067114093959731}, 96: {'count': 3, 'probability': 0.010067114093959731}, 58: {'count': 3, 'probability': 0.010067114093959731}, 82: {'count': 2, 'probability': 0.006711409395973154}, 31: {'count': 2, 'probability': 0.006711409395973154}, 62: {'count': 2, 'probability': 0.006711409395973154}, 90: {'count': 2, 'probability': 0.006711409395973154}, 49: {'count': 2, 'probability': 0.006711409395973154}, 57: {'count': 2, 'probability': 0.006711409395973154}, 28: {'count': 2, 'probability': 0.006711409395973154}, 104: {'count': 2, 'probability': 0.006711409395973154}, 44: {'count': 2, 'probability': 0.006711409395973154}, 89: {'count': 2, 'probability': 0.006711409395973154}, 56: {'count': 2, 'probability': 0.006711409395973154}, 27: {'count': 2, 'probability': 0.006711409395973154}, 51: {'count': 2, 'probability': 0.006711409395973154}, 32: {'count': 2, 'probability': 0.006711409395973154}, 72: {'count': 2, 'probability': 0.006711409395973154}, 70: {'count': 2, 'probability': 0.006711409395973154}, 60: {'count': 1, 'probability': 0.003355704697986577}, 103: {'count': 1, 'probability': 0.003355704697986577}, 24: {'count': 1, 'probability': 0.003355704697986577}, 34: {'count': 1, 'probability': 0.003355704697986577}, 29: {'count': 1, 'probability': 0.003355704697986577}, 43: {'count': 1, 'probability': 0.003355704697986577}, 67: {'count': 1, 'probability': 0.003355704697986577}, 93: {'count': 1, 'probability': 0.003355704697986577}, 66: {'count': 1, 'probability': 0.003355704697986577}, 68: {'count': 1, 'probability': 0.003355704697986577}, 2: {'count': 1, 'probability': 0.003355704697986577}, 1: {'count': 1, 'probability': 0.003355704697986577}, 77: {'count': 1, 'probability': 0.003355704697986577}, 36: {'count': 1, 'probability': 0.003355704697986577}, 9: {'count': 1, 'probability': 0.003355704697986577}, 85: {'count': 1, 'probability': 0.003355704697986577}, 33: {'count': 1, 'probability': 0.003355704697986577}, 71: {'count': 1, 'probability': 0.003355704697986577}, 87: {'count': 1, 'probability': 0.003355704697986577}, 46: {'count': 1, 'probability': 0.003355704697986577}, 55: {'count': 1, 'probability': 0.003355704697986577}, 95: {'count': 1, 'probability': 0.003355704697986577}, 0: {'count': 1, 'probability': 0.003355704697986577}, 19: {'count': 1, 'probability': 0.003355704697986577}, 4: {'count': 1, 'probability': 0.003355704697986577}, 59: {'count': 1, 'probability': 0.003355704697986577}, 41: {'count': 1, 'probability': 0.003355704697986577}, 99: {'count': 1, 'probability': 0.003355704697986577}, 65: {'count': 1, 'probability': 0.003355704697986577}, 30: {'count': 1, 'probability': 0.003355704697986577}, 92: {'count': 1, 'probability': 0.003355704697986577}, 81: {'count': 1, 'probability': 0.003355704697986577}, 38: {'count': 1, 'probability': 0.003355704697986577}, 52: {'count': 1, 'probability': 0.003355704697986577}, 69: {'count': 1, 'probability': 0.003355704697986577}, 91: {'count': 1, 'probability': 0.003355704697986577}, 54: {'count': 1, 'probability': 0.003355704697986577}, 78: {'count': 1, 'probability': 0.003355704697986577}, 94: {'count': 1, 'probability': 0.003355704697986577}, 15: {'count': 1, 'probability': 0.003355704697986577}, 74: {'count': 1, 'probability': 0.003355704697986577}, 64: {'count': 1, 'probability': 0.003355704697986577}, 11: {'count': 1, 'probability': 0.003355704697986577}, 97: {'count': 1, 'probability': 0.003355704697986577}, 98: {'count': 1, 'probability': 0.003355704697986577}, 83: {'count': 1, 'probability': 0.003355704697986577}, 75: {'count': 1, 'probability': 0.003355704697986577}, 14: {'count': 1, 'probability': 0.003355704697986577}, 13: {'count': 1, 'probability': 0.003355704697986577}, 100: {'count': 1, 'probability': 0.003355704697986577}, 101: {'count': 1, 'probability': 0.003355704697986577}, 26: {'count': 1, 'probability': 0.003355704697986577}, 88: {'count': 1, 'probability': 0.003355704697986577}, 50: {'count': 1, 'probability': 0.003355704697986577}, 53: {'count': 1, 'probability': 0.003355704697986577}, 10: {'count': 1, 'probability': 0.003355704697986577}, 12: {'count': 1, 'probability': 0.003355704697986577}, 3: {'count': 1, 'probability': 0.003355704697986577}}\n",
      "number_diagnoses:\n",
      " {6: {'count': 147, 'probability': 0.49328859060402686}, 5: {'count': 68, 'probability': 0.22818791946308725}, 9: {'count': 40, 'probability': 0.1342281879194631}, 3: {'count': 18, 'probability': 0.06040268456375839}, 4: {'count': 13, 'probability': 0.0436241610738255}, 8: {'count': 6, 'probability': 0.020134228187919462}, 7: {'count': 3, 'probability': 0.010067114093959731}, 2: {'count': 2, 'probability': 0.006711409395973154}, 1: {'count': 1, 'probability': 0.003355704697986577}}\n"
     ]
    }
   ],
   "source": [
    "for feature in selected_features:\n",
    "    prob = calculate_variable_probabilities(X_selected[feature])\n",
    "    print(f\"{feature}:\\n {prob}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0bbc15ab-6660-4185-99dc-58c8fb25498e",
   "metadata": {},
   "source": [
    "### Step 2: Construct Potential functions for each variable"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "30ddf6c1-4095-45c7-9960-28b77641ec33",
   "metadata": {},
   "outputs": [],
   "source": [
    "from pgmpy.factors.discrete import DiscreteFactor\n",
    "\n",
    "def construct_potential_functions(df, variables):\n",
    "    potential_functions = {}\n",
    "    \n",
    "    for variable in variables:\n",
    "        # Calculate probabilities for the variable\n",
    "        probabilities = calculate_variable_probabilities(df[variable])\n",
    "        \n",
    "        # Extract the unique values and their corresponding probabilities\n",
    "        values = list(probabilities.keys())\n",
    "        probs = [probabilities[value]['probability'] for value in values]\n",
    "        \n",
    "        # Create a DiscreteFactor for the variable\n",
    "        factor = DiscreteFactor([variable], [len(values)], probs)\n",
    "        \n",
    "        potential_functions[variable] = factor\n",
    "    \n",
    "    return potential_functions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "25b981b9-ff25-4a2e-b0b2-6249c56bfc5f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Potential Function for age:\n",
      "+--------+------------+\n",
      "| age    |   phi(age) |\n",
      "+========+============+\n",
      "| age(0) |     0.2215 |\n",
      "+--------+------------+\n",
      "| age(1) |     0.2081 |\n",
      "+--------+------------+\n",
      "| age(2) |     0.2047 |\n",
      "+--------+------------+\n",
      "| age(3) |     0.1409 |\n",
      "+--------+------------+\n",
      "| age(4) |     0.1342 |\n",
      "+--------+------------+\n",
      "| age(5) |     0.0570 |\n",
      "+--------+------------+\n",
      "| age(6) |     0.0201 |\n",
      "+--------+------------+\n",
      "| age(7) |     0.0134 |\n",
      "+--------+------------+\n",
      "\n",
      "Potential Function for time_in_hospital:\n",
      "+----------------------+-------------------------+\n",
      "| time_in_hospital     |   phi(time_in_hospital) |\n",
      "+======================+=========================+\n",
      "| time_in_hospital(0)  |                  0.1711 |\n",
      "+----------------------+-------------------------+\n",
      "| time_in_hospital(1)  |                  0.1242 |\n",
      "+----------------------+-------------------------+\n",
      "| time_in_hospital(2)  |                  0.1208 |\n",
      "+----------------------+-------------------------+\n",
      "| time_in_hospital(3)  |                  0.1141 |\n",
      "+----------------------+-------------------------+\n",
      "| time_in_hospital(4)  |                  0.1141 |\n",
      "+----------------------+-------------------------+\n",
      "| time_in_hospital(5)  |                  0.0772 |\n",
      "+----------------------+-------------------------+\n",
      "| time_in_hospital(6)  |                  0.0570 |\n",
      "+----------------------+-------------------------+\n",
      "| time_in_hospital(7)  |                  0.0503 |\n",
      "+----------------------+-------------------------+\n",
      "| time_in_hospital(8)  |                  0.0503 |\n",
      "+----------------------+-------------------------+\n",
      "| time_in_hospital(9)  |                  0.0470 |\n",
      "+----------------------+-------------------------+\n",
      "| time_in_hospital(10) |                  0.0268 |\n",
      "+----------------------+-------------------------+\n",
      "| time_in_hospital(11) |                  0.0201 |\n",
      "+----------------------+-------------------------+\n",
      "| time_in_hospital(12) |                  0.0168 |\n",
      "+----------------------+-------------------------+\n",
      "| time_in_hospital(13) |                  0.0101 |\n",
      "+----------------------+-------------------------+\n",
      "\n",
      "Potential Function for num_lab_procedures:\n",
      "+------------------------+---------------------------+\n",
      "| num_lab_procedures     |   phi(num_lab_procedures) |\n",
      "+========================+===========================+\n",
      "| num_lab_procedures(0)  |                    0.0336 |\n",
      "+------------------------+---------------------------+\n",
      "| num_lab_procedures(1)  |                    0.0336 |\n",
      "+------------------------+---------------------------+\n",
      "| num_lab_procedures(2)  |                    0.0336 |\n",
      "+------------------------+---------------------------+\n",
      "| num_lab_procedures(3)  |                    0.0302 |\n",
      "+------------------------+---------------------------+\n",
      "| num_lab_procedures(4)  |                    0.0302 |\n",
      "+------------------------+---------------------------+\n",
      "| num_lab_procedures(5)  |                    0.0302 |\n",
      "+------------------------+---------------------------+\n",
      "| num_lab_procedures(6)  |                    0.0268 |\n",
      "+------------------------+---------------------------+\n",
      "| num_lab_procedures(7)  |                    0.0268 |\n",
      "+------------------------+---------------------------+\n",
      "| num_lab_procedures(8)  |                    0.0268 |\n",
      "+------------------------+---------------------------+\n",
      "| num_lab_procedures(9)  |                    0.0268 |\n",
      "+------------------------+---------------------------+\n",
      "| num_lab_procedures(10) |                    0.0268 |\n",
      "+------------------------+---------------------------+\n",
      "| num_lab_procedures(11) |                    0.0235 |\n",
      "+------------------------+---------------------------+\n",
      "| num_lab_procedures(12) |                    0.0235 |\n",
      "+------------------------+---------------------------+\n",
      "| num_lab_procedures(13) |                    0.0235 |\n",
      "+------------------------+---------------------------+\n",
      "| num_lab_procedures(14) |                    0.0235 |\n",
      "+------------------------+---------------------------+\n",
      "| num_lab_procedures(15) |                    0.0235 |\n",
      "+------------------------+---------------------------+\n",
      "| num_lab_procedures(16) |                    0.0235 |\n",
      "+------------------------+---------------------------+\n",
      "| num_lab_procedures(17) |                    0.0235 |\n",
      "+------------------------+---------------------------+\n",
      "| num_lab_procedures(18) |                    0.0235 |\n",
      "+------------------------+---------------------------+\n",
      "| num_lab_procedures(19) |                    0.0235 |\n",
      "+------------------------+---------------------------+\n",
      "| num_lab_procedures(20) |                    0.0201 |\n",
      "+------------------------+---------------------------+\n",
      "| num_lab_procedures(21) |                    0.0201 |\n",
      "+------------------------+---------------------------+\n",
      "| num_lab_procedures(22) |                    0.0201 |\n",
      "+------------------------+---------------------------+\n",
      "| num_lab_procedures(23) |                    0.0201 |\n",
      "+------------------------+---------------------------+\n",
      "| num_lab_procedures(24) |                    0.0201 |\n",
      "+------------------------+---------------------------+\n",
      "| num_lab_procedures(25) |                    0.0168 |\n",
      "+------------------------+---------------------------+\n",
      "| num_lab_procedures(26) |                    0.0168 |\n",
      "+------------------------+---------------------------+\n",
      "| num_lab_procedures(27) |                    0.0168 |\n",
      "+------------------------+---------------------------+\n",
      "| num_lab_procedures(28) |                    0.0168 |\n",
      "+------------------------+---------------------------+\n",
      "| num_lab_procedures(29) |                    0.0168 |\n",
      "+------------------------+---------------------------+\n",
      "| num_lab_procedures(30) |                    0.0168 |\n",
      "+------------------------+---------------------------+\n",
      "| num_lab_procedures(31) |                    0.0168 |\n",
      "+------------------------+---------------------------+\n",
      "| num_lab_procedures(32) |                    0.0168 |\n",
      "+------------------------+---------------------------+\n",
      "| num_lab_procedures(33) |                    0.0134 |\n",
      "+------------------------+---------------------------+\n",
      "| num_lab_procedures(34) |                    0.0134 |\n",
      "+------------------------+---------------------------+\n",
      "| num_lab_procedures(35) |                    0.0134 |\n",
      "+------------------------+---------------------------+\n",
      "| num_lab_procedures(36) |                    0.0134 |\n",
      "+------------------------+---------------------------+\n",
      "| num_lab_procedures(37) |                    0.0134 |\n",
      "+------------------------+---------------------------+\n",
      "| num_lab_procedures(38) |                    0.0134 |\n",
      "+------------------------+---------------------------+\n",
      "| num_lab_procedures(39) |                    0.0101 |\n",
      "+------------------------+---------------------------+\n",
      "| num_lab_procedures(40) |                    0.0101 |\n",
      "+------------------------+---------------------------+\n",
      "| num_lab_procedures(41) |                    0.0067 |\n",
      "+------------------------+---------------------------+\n",
      "| num_lab_procedures(42) |                    0.0067 |\n",
      "+------------------------+---------------------------+\n",
      "| num_lab_procedures(43) |                    0.0067 |\n",
      "+------------------------+---------------------------+\n",
      "| num_lab_procedures(44) |                    0.0067 |\n",
      "+------------------------+---------------------------+\n",
      "| num_lab_procedures(45) |                    0.0067 |\n",
      "+------------------------+---------------------------+\n",
      "| num_lab_procedures(46) |                    0.0067 |\n",
      "+------------------------+---------------------------+\n",
      "| num_lab_procedures(47) |                    0.0067 |\n",
      "+------------------------+---------------------------+\n",
      "| num_lab_procedures(48) |                    0.0067 |\n",
      "+------------------------+---------------------------+\n",
      "| num_lab_procedures(49) |                    0.0067 |\n",
      "+------------------------+---------------------------+\n",
      "| num_lab_procedures(50) |                    0.0067 |\n",
      "+------------------------+---------------------------+\n",
      "| num_lab_procedures(51) |                    0.0067 |\n",
      "+------------------------+---------------------------+\n",
      "| num_lab_procedures(52) |                    0.0067 |\n",
      "+------------------------+---------------------------+\n",
      "| num_lab_procedures(53) |                    0.0067 |\n",
      "+------------------------+---------------------------+\n",
      "| num_lab_procedures(54) |                    0.0067 |\n",
      "+------------------------+---------------------------+\n",
      "| num_lab_procedures(55) |                    0.0034 |\n",
      "+------------------------+---------------------------+\n",
      "| num_lab_procedures(56) |                    0.0034 |\n",
      "+------------------------+---------------------------+\n",
      "| num_lab_procedures(57) |                    0.0034 |\n",
      "+------------------------+---------------------------+\n",
      "| num_lab_procedures(58) |                    0.0034 |\n",
      "+------------------------+---------------------------+\n",
      "| num_lab_procedures(59) |                    0.0034 |\n",
      "+------------------------+---------------------------+\n",
      "| num_lab_procedures(60) |                    0.0034 |\n",
      "+------------------------+---------------------------+\n",
      "| num_lab_procedures(61) |                    0.0034 |\n",
      "+------------------------+---------------------------+\n",
      "| num_lab_procedures(62) |                    0.0034 |\n",
      "+------------------------+---------------------------+\n",
      "| num_lab_procedures(63) |                    0.0034 |\n",
      "+------------------------+---------------------------+\n",
      "| num_lab_procedures(64) |                    0.0034 |\n",
      "+------------------------+---------------------------+\n",
      "\n",
      "Potential Function for num_medications:\n",
      "+---------------------+------------------------+\n",
      "| num_medications     |   phi(num_medications) |\n",
      "+=====================+========================+\n",
      "| num_medications(0)  |                 0.0839 |\n",
      "+---------------------+------------------------+\n",
      "| num_medications(1)  |                 0.0738 |\n",
      "+---------------------+------------------------+\n",
      "| num_medications(2)  |                 0.0705 |\n",
      "+---------------------+------------------------+\n",
      "| num_medications(3)  |                 0.0671 |\n",
      "+---------------------+------------------------+\n",
      "| num_medications(4)  |                 0.0537 |\n",
      "+---------------------+------------------------+\n",
      "| num_medications(5)  |                 0.0503 |\n",
      "+---------------------+------------------------+\n",
      "| num_medications(6)  |                 0.0503 |\n",
      "+---------------------+------------------------+\n",
      "| num_medications(7)  |                 0.0470 |\n",
      "+---------------------+------------------------+\n",
      "| num_medications(8)  |                 0.0436 |\n",
      "+---------------------+------------------------+\n",
      "| num_medications(9)  |                 0.0369 |\n",
      "+---------------------+------------------------+\n",
      "| num_medications(10) |                 0.0369 |\n",
      "+---------------------+------------------------+\n",
      "| num_medications(11) |                 0.0336 |\n",
      "+---------------------+------------------------+\n",
      "| num_medications(12) |                 0.0336 |\n",
      "+---------------------+------------------------+\n",
      "| num_medications(13) |                 0.0302 |\n",
      "+---------------------+------------------------+\n",
      "| num_medications(14) |                 0.0268 |\n",
      "+---------------------+------------------------+\n",
      "| num_medications(15) |                 0.0268 |\n",
      "+---------------------+------------------------+\n",
      "| num_medications(16) |                 0.0268 |\n",
      "+---------------------+------------------------+\n",
      "| num_medications(17) |                 0.0235 |\n",
      "+---------------------+------------------------+\n",
      "| num_medications(18) |                 0.0201 |\n",
      "+---------------------+------------------------+\n",
      "| num_medications(19) |                 0.0201 |\n",
      "+---------------------+------------------------+\n",
      "| num_medications(20) |                 0.0168 |\n",
      "+---------------------+------------------------+\n",
      "| num_medications(21) |                 0.0168 |\n",
      "+---------------------+------------------------+\n",
      "| num_medications(22) |                 0.0134 |\n",
      "+---------------------+------------------------+\n",
      "| num_medications(23) |                 0.0134 |\n",
      "+---------------------+------------------------+\n",
      "| num_medications(24) |                 0.0134 |\n",
      "+---------------------+------------------------+\n",
      "| num_medications(25) |                 0.0101 |\n",
      "+---------------------+------------------------+\n",
      "| num_medications(26) |                 0.0101 |\n",
      "+---------------------+------------------------+\n",
      "| num_medications(27) |                 0.0101 |\n",
      "+---------------------+------------------------+\n",
      "| num_medications(28) |                 0.0067 |\n",
      "+---------------------+------------------------+\n",
      "| num_medications(29) |                 0.0067 |\n",
      "+---------------------+------------------------+\n",
      "| num_medications(30) |                 0.0067 |\n",
      "+---------------------+------------------------+\n",
      "| num_medications(31) |                 0.0067 |\n",
      "+---------------------+------------------------+\n",
      "| num_medications(32) |                 0.0067 |\n",
      "+---------------------+------------------------+\n",
      "| num_medications(33) |                 0.0034 |\n",
      "+---------------------+------------------------+\n",
      "| num_medications(34) |                 0.0034 |\n",
      "+---------------------+------------------------+\n",
      "\n",
      "Potential Function for number_inpatient:\n",
      "+---------------------+-------------------------+\n",
      "| number_inpatient    |   phi(number_inpatient) |\n",
      "+=====================+=========================+\n",
      "| number_inpatient(0) |                  0.6846 |\n",
      "+---------------------+-------------------------+\n",
      "| number_inpatient(1) |                  0.1443 |\n",
      "+---------------------+-------------------------+\n",
      "| number_inpatient(2) |                  0.0872 |\n",
      "+---------------------+-------------------------+\n",
      "| number_inpatient(3) |                  0.0503 |\n",
      "+---------------------+-------------------------+\n",
      "| number_inpatient(4) |                  0.0134 |\n",
      "+---------------------+-------------------------+\n",
      "| number_inpatient(5) |                  0.0067 |\n",
      "+---------------------+-------------------------+\n",
      "| number_inpatient(6) |                  0.0067 |\n",
      "+---------------------+-------------------------+\n",
      "| number_inpatient(7) |                  0.0034 |\n",
      "+---------------------+-------------------------+\n",
      "| number_inpatient(8) |                  0.0034 |\n",
      "+---------------------+-------------------------+\n",
      "\n",
      "Potential Function for diag_1:\n",
      "+------------+---------------+\n",
      "| diag_1     |   phi(diag_1) |\n",
      "+============+===============+\n",
      "| diag_1(0)  |        0.0705 |\n",
      "+------------+---------------+\n",
      "| diag_1(1)  |        0.0638 |\n",
      "+------------+---------------+\n",
      "| diag_1(2)  |        0.0638 |\n",
      "+------------+---------------+\n",
      "| diag_1(3)  |        0.0470 |\n",
      "+------------+---------------+\n",
      "| diag_1(4)  |        0.0403 |\n",
      "+------------+---------------+\n",
      "| diag_1(5)  |        0.0403 |\n",
      "+------------+---------------+\n",
      "| diag_1(6)  |        0.0336 |\n",
      "+------------+---------------+\n",
      "| diag_1(7)  |        0.0302 |\n",
      "+------------+---------------+\n",
      "| diag_1(8)  |        0.0302 |\n",
      "+------------+---------------+\n",
      "| diag_1(9)  |        0.0302 |\n",
      "+------------+---------------+\n",
      "| diag_1(10) |        0.0268 |\n",
      "+------------+---------------+\n",
      "| diag_1(11) |        0.0235 |\n",
      "+------------+---------------+\n",
      "| diag_1(12) |        0.0235 |\n",
      "+------------+---------------+\n",
      "| diag_1(13) |        0.0201 |\n",
      "+------------+---------------+\n",
      "| diag_1(14) |        0.0168 |\n",
      "+------------+---------------+\n",
      "| diag_1(15) |        0.0168 |\n",
      "+------------+---------------+\n",
      "| diag_1(16) |        0.0168 |\n",
      "+------------+---------------+\n",
      "| diag_1(17) |        0.0134 |\n",
      "+------------+---------------+\n",
      "| diag_1(18) |        0.0134 |\n",
      "+------------+---------------+\n",
      "| diag_1(19) |        0.0134 |\n",
      "+------------+---------------+\n",
      "| diag_1(20) |        0.0134 |\n",
      "+------------+---------------+\n",
      "| diag_1(21) |        0.0101 |\n",
      "+------------+---------------+\n",
      "| diag_1(22) |        0.0101 |\n",
      "+------------+---------------+\n",
      "| diag_1(23) |        0.0101 |\n",
      "+------------+---------------+\n",
      "| diag_1(24) |        0.0101 |\n",
      "+------------+---------------+\n",
      "| diag_1(25) |        0.0101 |\n",
      "+------------+---------------+\n",
      "| diag_1(26) |        0.0101 |\n",
      "+------------+---------------+\n",
      "| diag_1(27) |        0.0067 |\n",
      "+------------+---------------+\n",
      "| diag_1(28) |        0.0067 |\n",
      "+------------+---------------+\n",
      "| diag_1(29) |        0.0067 |\n",
      "+------------+---------------+\n",
      "| diag_1(30) |        0.0067 |\n",
      "+------------+---------------+\n",
      "| diag_1(31) |        0.0067 |\n",
      "+------------+---------------+\n",
      "| diag_1(32) |        0.0067 |\n",
      "+------------+---------------+\n",
      "| diag_1(33) |        0.0067 |\n",
      "+------------+---------------+\n",
      "| diag_1(34) |        0.0067 |\n",
      "+------------+---------------+\n",
      "| diag_1(35) |        0.0067 |\n",
      "+------------+---------------+\n",
      "| diag_1(36) |        0.0067 |\n",
      "+------------+---------------+\n",
      "| diag_1(37) |        0.0067 |\n",
      "+------------+---------------+\n",
      "| diag_1(38) |        0.0067 |\n",
      "+------------+---------------+\n",
      "| diag_1(39) |        0.0067 |\n",
      "+------------+---------------+\n",
      "| diag_1(40) |        0.0067 |\n",
      "+------------+---------------+\n",
      "| diag_1(41) |        0.0067 |\n",
      "+------------+---------------+\n",
      "| diag_1(42) |        0.0067 |\n",
      "+------------+---------------+\n",
      "| diag_1(43) |        0.0067 |\n",
      "+------------+---------------+\n",
      "| diag_1(44) |        0.0067 |\n",
      "+------------+---------------+\n",
      "| diag_1(45) |        0.0067 |\n",
      "+------------+---------------+\n",
      "| diag_1(46) |        0.0067 |\n",
      "+------------+---------------+\n",
      "| diag_1(47) |        0.0067 |\n",
      "+------------+---------------+\n",
      "| diag_1(48) |        0.0067 |\n",
      "+------------+---------------+\n",
      "| diag_1(49) |        0.0034 |\n",
      "+------------+---------------+\n",
      "| diag_1(50) |        0.0034 |\n",
      "+------------+---------------+\n",
      "| diag_1(51) |        0.0034 |\n",
      "+------------+---------------+\n",
      "| diag_1(52) |        0.0034 |\n",
      "+------------+---------------+\n",
      "| diag_1(53) |        0.0034 |\n",
      "+------------+---------------+\n",
      "| diag_1(54) |        0.0034 |\n",
      "+------------+---------------+\n",
      "| diag_1(55) |        0.0034 |\n",
      "+------------+---------------+\n",
      "| diag_1(56) |        0.0034 |\n",
      "+------------+---------------+\n",
      "| diag_1(57) |        0.0034 |\n",
      "+------------+---------------+\n",
      "| diag_1(58) |        0.0034 |\n",
      "+------------+---------------+\n",
      "| diag_1(59) |        0.0034 |\n",
      "+------------+---------------+\n",
      "| diag_1(60) |        0.0034 |\n",
      "+------------+---------------+\n",
      "| diag_1(61) |        0.0034 |\n",
      "+------------+---------------+\n",
      "| diag_1(62) |        0.0034 |\n",
      "+------------+---------------+\n",
      "| diag_1(63) |        0.0034 |\n",
      "+------------+---------------+\n",
      "| diag_1(64) |        0.0034 |\n",
      "+------------+---------------+\n",
      "| diag_1(65) |        0.0034 |\n",
      "+------------+---------------+\n",
      "| diag_1(66) |        0.0034 |\n",
      "+------------+---------------+\n",
      "| diag_1(67) |        0.0034 |\n",
      "+------------+---------------+\n",
      "| diag_1(68) |        0.0034 |\n",
      "+------------+---------------+\n",
      "| diag_1(69) |        0.0034 |\n",
      "+------------+---------------+\n",
      "| diag_1(70) |        0.0034 |\n",
      "+------------+---------------+\n",
      "| diag_1(71) |        0.0034 |\n",
      "+------------+---------------+\n",
      "| diag_1(72) |        0.0034 |\n",
      "+------------+---------------+\n",
      "| diag_1(73) |        0.0034 |\n",
      "+------------+---------------+\n",
      "| diag_1(74) |        0.0034 |\n",
      "+------------+---------------+\n",
      "| diag_1(75) |        0.0034 |\n",
      "+------------+---------------+\n",
      "| diag_1(76) |        0.0034 |\n",
      "+------------+---------------+\n",
      "| diag_1(77) |        0.0034 |\n",
      "+------------+---------------+\n",
      "| diag_1(78) |        0.0034 |\n",
      "+------------+---------------+\n",
      "| diag_1(79) |        0.0034 |\n",
      "+------------+---------------+\n",
      "| diag_1(80) |        0.0034 |\n",
      "+------------+---------------+\n",
      "| diag_1(81) |        0.0034 |\n",
      "+------------+---------------+\n",
      "| diag_1(82) |        0.0034 |\n",
      "+------------+---------------+\n",
      "| diag_1(83) |        0.0034 |\n",
      "+------------+---------------+\n",
      "| diag_1(84) |        0.0034 |\n",
      "+------------+---------------+\n",
      "| diag_1(85) |        0.0034 |\n",
      "+------------+---------------+\n",
      "| diag_1(86) |        0.0034 |\n",
      "+------------+---------------+\n",
      "| diag_1(87) |        0.0034 |\n",
      "+------------+---------------+\n",
      "| diag_1(88) |        0.0034 |\n",
      "+------------+---------------+\n",
      "| diag_1(89) |        0.0034 |\n",
      "+------------+---------------+\n",
      "| diag_1(90) |        0.0034 |\n",
      "+------------+---------------+\n",
      "| diag_1(91) |        0.0034 |\n",
      "+------------+---------------+\n",
      "\n",
      "Potential Function for diag_2:\n",
      "+-------------+---------------+\n",
      "| diag_2      |   phi(diag_2) |\n",
      "+=============+===============+\n",
      "| diag_2(0)   |        0.0772 |\n",
      "+-------------+---------------+\n",
      "| diag_2(1)   |        0.0671 |\n",
      "+-------------+---------------+\n",
      "| diag_2(2)   |        0.0570 |\n",
      "+-------------+---------------+\n",
      "| diag_2(3)   |        0.0369 |\n",
      "+-------------+---------------+\n",
      "| diag_2(4)   |        0.0336 |\n",
      "+-------------+---------------+\n",
      "| diag_2(5)   |        0.0336 |\n",
      "+-------------+---------------+\n",
      "| diag_2(6)   |        0.0336 |\n",
      "+-------------+---------------+\n",
      "| diag_2(7)   |        0.0336 |\n",
      "+-------------+---------------+\n",
      "| diag_2(8)   |        0.0302 |\n",
      "+-------------+---------------+\n",
      "| diag_2(9)   |        0.0235 |\n",
      "+-------------+---------------+\n",
      "| diag_2(10)  |        0.0235 |\n",
      "+-------------+---------------+\n",
      "| diag_2(11)  |        0.0235 |\n",
      "+-------------+---------------+\n",
      "| diag_2(12)  |        0.0235 |\n",
      "+-------------+---------------+\n",
      "| diag_2(13)  |        0.0168 |\n",
      "+-------------+---------------+\n",
      "| diag_2(14)  |        0.0168 |\n",
      "+-------------+---------------+\n",
      "| diag_2(15)  |        0.0168 |\n",
      "+-------------+---------------+\n",
      "| diag_2(16)  |        0.0134 |\n",
      "+-------------+---------------+\n",
      "| diag_2(17)  |        0.0134 |\n",
      "+-------------+---------------+\n",
      "| diag_2(18)  |        0.0134 |\n",
      "+-------------+---------------+\n",
      "| diag_2(19)  |        0.0134 |\n",
      "+-------------+---------------+\n",
      "| diag_2(20)  |        0.0134 |\n",
      "+-------------+---------------+\n",
      "| diag_2(21)  |        0.0101 |\n",
      "+-------------+---------------+\n",
      "| diag_2(22)  |        0.0101 |\n",
      "+-------------+---------------+\n",
      "| diag_2(23)  |        0.0101 |\n",
      "+-------------+---------------+\n",
      "| diag_2(24)  |        0.0101 |\n",
      "+-------------+---------------+\n",
      "| diag_2(25)  |        0.0101 |\n",
      "+-------------+---------------+\n",
      "| diag_2(26)  |        0.0067 |\n",
      "+-------------+---------------+\n",
      "| diag_2(27)  |        0.0067 |\n",
      "+-------------+---------------+\n",
      "| diag_2(28)  |        0.0067 |\n",
      "+-------------+---------------+\n",
      "| diag_2(29)  |        0.0067 |\n",
      "+-------------+---------------+\n",
      "| diag_2(30)  |        0.0067 |\n",
      "+-------------+---------------+\n",
      "| diag_2(31)  |        0.0067 |\n",
      "+-------------+---------------+\n",
      "| diag_2(32)  |        0.0067 |\n",
      "+-------------+---------------+\n",
      "| diag_2(33)  |        0.0067 |\n",
      "+-------------+---------------+\n",
      "| diag_2(34)  |        0.0067 |\n",
      "+-------------+---------------+\n",
      "| diag_2(35)  |        0.0067 |\n",
      "+-------------+---------------+\n",
      "| diag_2(36)  |        0.0067 |\n",
      "+-------------+---------------+\n",
      "| diag_2(37)  |        0.0067 |\n",
      "+-------------+---------------+\n",
      "| diag_2(38)  |        0.0067 |\n",
      "+-------------+---------------+\n",
      "| diag_2(39)  |        0.0067 |\n",
      "+-------------+---------------+\n",
      "| diag_2(40)  |        0.0067 |\n",
      "+-------------+---------------+\n",
      "| diag_2(41)  |        0.0067 |\n",
      "+-------------+---------------+\n",
      "| diag_2(42)  |        0.0067 |\n",
      "+-------------+---------------+\n",
      "| diag_2(43)  |        0.0067 |\n",
      "+-------------+---------------+\n",
      "| diag_2(44)  |        0.0067 |\n",
      "+-------------+---------------+\n",
      "| diag_2(45)  |        0.0067 |\n",
      "+-------------+---------------+\n",
      "| diag_2(46)  |        0.0067 |\n",
      "+-------------+---------------+\n",
      "| diag_2(47)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_2(48)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_2(49)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_2(50)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_2(51)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_2(52)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_2(53)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_2(54)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_2(55)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_2(56)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_2(57)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_2(58)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_2(59)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_2(60)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_2(61)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_2(62)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_2(63)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_2(64)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_2(65)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_2(66)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_2(67)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_2(68)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_2(69)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_2(70)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_2(71)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_2(72)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_2(73)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_2(74)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_2(75)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_2(76)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_2(77)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_2(78)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_2(79)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_2(80)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_2(81)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_2(82)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_2(83)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_2(84)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_2(85)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_2(86)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_2(87)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_2(88)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_2(89)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_2(90)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_2(91)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_2(92)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_2(93)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_2(94)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_2(95)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_2(96)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_2(97)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_2(98)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_2(99)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_2(100) |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_2(101) |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_2(102) |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_2(103) |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_2(104) |        0.0034 |\n",
      "+-------------+---------------+\n",
      "\n",
      "Potential Function for diag_3:\n",
      "+-------------+---------------+\n",
      "| diag_3      |   phi(diag_3) |\n",
      "+=============+===============+\n",
      "| diag_3(0)   |        0.0872 |\n",
      "+-------------+---------------+\n",
      "| diag_3(1)   |        0.0671 |\n",
      "+-------------+---------------+\n",
      "| diag_3(2)   |        0.0537 |\n",
      "+-------------+---------------+\n",
      "| diag_3(3)   |        0.0503 |\n",
      "+-------------+---------------+\n",
      "| diag_3(4)   |        0.0436 |\n",
      "+-------------+---------------+\n",
      "| diag_3(5)   |        0.0369 |\n",
      "+-------------+---------------+\n",
      "| diag_3(6)   |        0.0302 |\n",
      "+-------------+---------------+\n",
      "| diag_3(7)   |        0.0235 |\n",
      "+-------------+---------------+\n",
      "| diag_3(8)   |        0.0235 |\n",
      "+-------------+---------------+\n",
      "| diag_3(9)   |        0.0201 |\n",
      "+-------------+---------------+\n",
      "| diag_3(10)  |        0.0168 |\n",
      "+-------------+---------------+\n",
      "| diag_3(11)  |        0.0168 |\n",
      "+-------------+---------------+\n",
      "| diag_3(12)  |        0.0168 |\n",
      "+-------------+---------------+\n",
      "| diag_3(13)  |        0.0134 |\n",
      "+-------------+---------------+\n",
      "| diag_3(14)  |        0.0134 |\n",
      "+-------------+---------------+\n",
      "| diag_3(15)  |        0.0134 |\n",
      "+-------------+---------------+\n",
      "| diag_3(16)  |        0.0134 |\n",
      "+-------------+---------------+\n",
      "| diag_3(17)  |        0.0134 |\n",
      "+-------------+---------------+\n",
      "| diag_3(18)  |        0.0134 |\n",
      "+-------------+---------------+\n",
      "| diag_3(19)  |        0.0134 |\n",
      "+-------------+---------------+\n",
      "| diag_3(20)  |        0.0134 |\n",
      "+-------------+---------------+\n",
      "| diag_3(21)  |        0.0134 |\n",
      "+-------------+---------------+\n",
      "| diag_3(22)  |        0.0101 |\n",
      "+-------------+---------------+\n",
      "| diag_3(23)  |        0.0101 |\n",
      "+-------------+---------------+\n",
      "| diag_3(24)  |        0.0101 |\n",
      "+-------------+---------------+\n",
      "| diag_3(25)  |        0.0101 |\n",
      "+-------------+---------------+\n",
      "| diag_3(26)  |        0.0101 |\n",
      "+-------------+---------------+\n",
      "| diag_3(27)  |        0.0101 |\n",
      "+-------------+---------------+\n",
      "| diag_3(28)  |        0.0101 |\n",
      "+-------------+---------------+\n",
      "| diag_3(29)  |        0.0101 |\n",
      "+-------------+---------------+\n",
      "| diag_3(30)  |        0.0101 |\n",
      "+-------------+---------------+\n",
      "| diag_3(31)  |        0.0067 |\n",
      "+-------------+---------------+\n",
      "| diag_3(32)  |        0.0067 |\n",
      "+-------------+---------------+\n",
      "| diag_3(33)  |        0.0067 |\n",
      "+-------------+---------------+\n",
      "| diag_3(34)  |        0.0067 |\n",
      "+-------------+---------------+\n",
      "| diag_3(35)  |        0.0067 |\n",
      "+-------------+---------------+\n",
      "| diag_3(36)  |        0.0067 |\n",
      "+-------------+---------------+\n",
      "| diag_3(37)  |        0.0067 |\n",
      "+-------------+---------------+\n",
      "| diag_3(38)  |        0.0067 |\n",
      "+-------------+---------------+\n",
      "| diag_3(39)  |        0.0067 |\n",
      "+-------------+---------------+\n",
      "| diag_3(40)  |        0.0067 |\n",
      "+-------------+---------------+\n",
      "| diag_3(41)  |        0.0067 |\n",
      "+-------------+---------------+\n",
      "| diag_3(42)  |        0.0067 |\n",
      "+-------------+---------------+\n",
      "| diag_3(43)  |        0.0067 |\n",
      "+-------------+---------------+\n",
      "| diag_3(44)  |        0.0067 |\n",
      "+-------------+---------------+\n",
      "| diag_3(45)  |        0.0067 |\n",
      "+-------------+---------------+\n",
      "| diag_3(46)  |        0.0067 |\n",
      "+-------------+---------------+\n",
      "| diag_3(47)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_3(48)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_3(49)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_3(50)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_3(51)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_3(52)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_3(53)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_3(54)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_3(55)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_3(56)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_3(57)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_3(58)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_3(59)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_3(60)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_3(61)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_3(62)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_3(63)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_3(64)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_3(65)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_3(66)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_3(67)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_3(68)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_3(69)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_3(70)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_3(71)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_3(72)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_3(73)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_3(74)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_3(75)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_3(76)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_3(77)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_3(78)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_3(79)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_3(80)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_3(81)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_3(82)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_3(83)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_3(84)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_3(85)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_3(86)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_3(87)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_3(88)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_3(89)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_3(90)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_3(91)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_3(92)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_3(93)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_3(94)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_3(95)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_3(96)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_3(97)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_3(98)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_3(99)  |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_3(100) |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_3(101) |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_3(102) |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_3(103) |        0.0034 |\n",
      "+-------------+---------------+\n",
      "| diag_3(104) |        0.0034 |\n",
      "+-------------+---------------+\n",
      "\n",
      "Potential Function for number_diagnoses:\n",
      "+---------------------+-------------------------+\n",
      "| number_diagnoses    |   phi(number_diagnoses) |\n",
      "+=====================+=========================+\n",
      "| number_diagnoses(0) |                  0.4933 |\n",
      "+---------------------+-------------------------+\n",
      "| number_diagnoses(1) |                  0.2282 |\n",
      "+---------------------+-------------------------+\n",
      "| number_diagnoses(2) |                  0.1342 |\n",
      "+---------------------+-------------------------+\n",
      "| number_diagnoses(3) |                  0.0604 |\n",
      "+---------------------+-------------------------+\n",
      "| number_diagnoses(4) |                  0.0436 |\n",
      "+---------------------+-------------------------+\n",
      "| number_diagnoses(5) |                  0.0201 |\n",
      "+---------------------+-------------------------+\n",
      "| number_diagnoses(6) |                  0.0101 |\n",
      "+---------------------+-------------------------+\n",
      "| number_diagnoses(7) |                  0.0067 |\n",
      "+---------------------+-------------------------+\n",
      "| number_diagnoses(8) |                  0.0034 |\n",
      "+---------------------+-------------------------+\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# Construct the potential functions for each variable\n",
    "potential_functions = construct_potential_functions(X_selected, selected_features)\n",
    "\n",
    "# Print the potential functions\n",
    "for variable, factor in potential_functions.items():\n",
    "    print(f\"Potential Function for {variable}:\")\n",
    "    print(factor)\n",
    "    print()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7ea70391-c4df-4a4c-96de-5cc16e58cf6c",
   "metadata": {},
   "source": [
    "### Step 3: Study the pairwise relationship of each pair of variables => Needed for Pairwise Potential function"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bf8951f8-0f9b-49a6-8cbd-ee07a188aee0",
   "metadata": {},
   "source": [
    "#### 1. Correlation Analysis"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "36c4fd0b-824a-4436-bb83-70d2383ea9eb",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Correlation Matrix:\n",
      "                         age  time_in_hospital  num_lab_procedures  \\\n",
      "age                 1.000000          0.154621           -0.072163   \n",
      "time_in_hospital    0.154621          1.000000            0.282633   \n",
      "num_lab_procedures -0.072163          0.282633            1.000000   \n",
      "num_medications     0.058245          0.474992            0.327564   \n",
      "number_inpatient    0.118071          0.134915           -0.043921   \n",
      "diag_1             -0.006717         -0.002848           -0.107023   \n",
      "diag_2              0.029984          0.054079            0.030975   \n",
      "diag_3              0.028224          0.121292           -0.028415   \n",
      "number_diagnoses    0.049904          0.214374            0.182942   \n",
      "readmitted          0.075893          0.199749           -0.064353   \n",
      "\n",
      "                    num_medications  number_inpatient    diag_1    diag_2  \\\n",
      "age                        0.058245          0.118071 -0.006717  0.029984   \n",
      "time_in_hospital           0.474992          0.134915 -0.002848  0.054079   \n",
      "num_lab_procedures         0.327564         -0.043921 -0.107023  0.030975   \n",
      "num_medications            1.000000          0.273319  0.065383  0.064043   \n",
      "number_inpatient           0.273319          1.000000  0.037870  0.016089   \n",
      "diag_1                     0.065383          0.037870  1.000000 -0.188255   \n",
      "diag_2                     0.064043          0.016089 -0.188255  1.000000   \n",
      "diag_3                     0.101079          0.077026 -0.043373 -0.063792   \n",
      "number_diagnoses           0.502164          0.239503 -0.014297  0.144984   \n",
      "readmitted                 0.242722          0.285953 -0.028542 -0.009831   \n",
      "\n",
      "                      diag_3  number_diagnoses  readmitted  \n",
      "age                 0.028224          0.049904    0.075893  \n",
      "time_in_hospital    0.121292          0.214374    0.199749  \n",
      "num_lab_procedures -0.028415          0.182942   -0.064353  \n",
      "num_medications     0.101079          0.502164    0.242722  \n",
      "number_inpatient    0.077026          0.239503    0.285953  \n",
      "diag_1             -0.043373         -0.014297   -0.028542  \n",
      "diag_2             -0.063792          0.144984   -0.009831  \n",
      "diag_3              1.000000         -0.021659    0.089243  \n",
      "number_diagnoses   -0.021659          1.000000    0.101144  \n",
      "readmitted          0.089243          0.101144    1.000000  \n"
     ]
    }
   ],
   "source": [
    "training_data = pd.concat([X_selected, y], axis=1)\n",
    "\n",
    "correlation_matrix = training_data.corr(method=\"spearman\")\n",
    "\n",
    "# Print the correlation matrix\n",
    "print(\"Correlation Matrix:\")\n",
    "print(correlation_matrix)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8cb6c01c-45d5-417b-af4d-c8d8a7820b2c",
   "metadata": {},
   "source": [
    "From the correlation analysis (using both Peason and Spearman scores), if we use 0.3 as a threshold for medium to strong correlation, there are a few noticeable pairs: \n",
    "\n",
    "- `num_medications` and `number_diagnoses`\n",
    "- `time_in_hospital` and `num_medications`\n",
    "- `num_lab_procedures` and `num_medications`"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "72084619-f2cf-434b-bb43-dd769231fffe",
   "metadata": {},
   "source": [
    "### Step 4: Implement the Markov model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "68aa0ac4-9549-4585-afc6-c893d5556d5b",
   "metadata": {},
   "outputs": [],
   "source": [
    "from pgmpy.models import MarkovNetwork\n",
    "from pgmpy.factors.discrete import DiscreteFactor\n",
    "from pgmpy.inference import VariableElimination\n",
    "\n",
    "# Create an instance of the MarkovModel\n",
    "model = MarkovNetwork()\n",
    "\n",
    "# Add nodes (variables) to the model\n",
    "variables = selected_features + ['readmitted']\n",
    "model.add_nodes_from(variables)\n",
    "\n",
    "# Add the potential functions (factors) to the model\n",
    "for variable, factor in potential_functions.items():\n",
    "    model.add_factors(factor)\n",
    "\n",
    "# Add edges based on correlation analysis or domain knowledge\n",
    "model.add_edge('number_inpatient', 'readmitted')\n",
    "model.add_edge('num_medications', 'readmitted')\n",
    "model.add_edge('time_in_hospital', 'readmitted')\n",
    "model.add_edge('time_in_hospital', 'num_medications')\n",
    "model.add_edge('num_lab_procedures', 'num_medications')\n",
    "model.add_edge('num_medications', 'number_diagnoses')\n",
    "\n",
    "# Create a factor for the target variable 'readmitted'\n",
    "readmitted_factor = DiscreteFactor(['readmitted'], [2], [1 - y.mean(), y.mean()])\n",
    "model.add_factors(readmitted_factor)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "23c90f82-3ce4-4261-8e3e-87e5acb683ed",
   "metadata": {},
   "source": [
    "### Step 5: Make inference with the model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "2252eeb7-41b2-427f-9d11-8b9e498e7c3d",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Inference Results:\n",
      "+---------------+-------------------+\n",
      "| readmitted    |   phi(readmitted) |\n",
      "+===============+===================+\n",
      "| readmitted(0) |            0.0006 |\n",
      "+---------------+-------------------+\n",
      "| readmitted(1) |            0.0008 |\n",
      "+---------------+-------------------+\n"
     ]
    }
   ],
   "source": [
    "# Create an instance of the VariableElimination class\n",
    "inference = VariableElimination(model)\n",
    "\n",
    "# Specify the query variable(s) and evidence (if any)\n",
    "query_variables = ['readmitted']\n",
    "evidence = {'num_medications': 5, 'time_in_hospital': 10}\n",
    "\n",
    "# Perform inference\n",
    "query_result = inference.query(query_variables, evidence)\n",
    "\n",
    "# Print the inference results\n",
    "print(\"Inference Results:\")\n",
    "print(query_result)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2d5a91cb-611b-4aa6-b503-90bdc569e0ab",
   "metadata": {},
   "source": [
    "We can see that based on the provided evidence ({'num_medications': 5, 'time_in_hospital': 10}), the model suggests that the probability of readmission is slightly higher (0.08%) compared to non-readmission (0.06%).\n",
    "\n",
    "However, the estimated probabilities are quite low (less than 1% for both cases). This could indicate that the model's predictive power is limited based on the given evidence and the current structure of the MRF model."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bcd621ea-a769-41e5-aa68-71d7bcc0e980",
   "metadata": {},
   "source": [
    "### Step 6: Model Evaluation"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e98e4677-328d-43c1-b36a-4afa976ffd02",
   "metadata": {},
   "source": [
    "### Preprocess `X_test` to get it ready for evaluation "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "c199c936-090a-4bbf-a8f0-036a2ac82acc",
   "metadata": {},
   "outputs": [],
   "source": [
    "def map_values_to_states(row, potential_functions):\n",
    "    mapped_row = {}\n",
    "    for variable, value in row.items():\n",
    "        if variable in potential_functions:\n",
    "            factor = potential_functions[variable]\n",
    "            unique_values = factor.state_names[variable]\n",
    "            if value in unique_values:\n",
    "                mapped_value = unique_values.index(value)\n",
    "                mapped_row[variable] = mapped_value\n",
    "            else:\n",
    "                # Handle unknown or unexpected values\n",
    "                closest_value = min(unique_values, key=lambda x: abs(x - value) if isinstance(x, (int, float)) else float('inf'))\n",
    "                mapped_value = unique_values.index(closest_value)\n",
    "                mapped_row[variable] = mapped_value\n",
    "        else:\n",
    "            mapped_row[variable] = value\n",
    "    return mapped_row"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "ccfa93cc-ac4f-4d89-9bff-279d3fb4111a",
   "metadata": {},
   "outputs": [],
   "source": [
    "X_test_selected = X_test[selected_features]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "efde38f7-2d6c-476a-8354-02d8566d357d",
   "metadata": {},
   "outputs": [],
   "source": [
    "for _, row in X_test_selected.iterrows():\n",
    "    # Map the feature values to their corresponding state values\n",
    "    mapped_row = map_values_to_states(row, potential_functions)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "id": "e2fee8b5-bfc6-417c-8d14-9e8188a1b02c",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Markov Network Model Evaluation:\n",
      "Accuracy: 0.6000\n",
      "Precision: 0.6000\n",
      "Recall: 1.0000\n",
      "F1-score: 0.7500\n",
      "AUC-ROC: 0.2928\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# Create an instance of the MarkovNetwork\n",
    "model = MarkovNetwork()\n",
    "\n",
    "# Add nodes (variables) to the model\n",
    "variables = selected_features + ['readmitted']\n",
    "model.add_nodes_from(variables)\n",
    "\n",
    "# Add the potential functions (factors) to the model\n",
    "for variable, factor in potential_functions.items():\n",
    "    model.add_factors(factor)\n",
    "\n",
    "# Add edges based on correlation analysis or domain knowledge\n",
    "model.add_edge('number_inpatient', 'readmitted')\n",
    "model.add_edge('num_medications', 'readmitted')\n",
    "model.add_edge('time_in_hospital', 'readmitted')\n",
    "model.add_edge('time_in_hospital', 'num_medications')\n",
    "model.add_edge('num_lab_procedures', 'num_medications')\n",
    "model.add_edge('num_medications', 'number_diagnoses')\n",
    "\n",
    "# Create a factor for the target variable 'readmitted'\n",
    "readmitted_factor = DiscreteFactor(['readmitted'], [2], [1 - y.mean(), y.mean()])\n",
    "# readmitted_factor = DiscreteFactor(['readmitted'], [2], [0.5, 0.5])\n",
    "model.add_factors(readmitted_factor)\n",
    "\n",
    "# Create an instance of the VariableElimination class\n",
    "inference = VariableElimination(model)\n",
    "\n",
    "# Evaluate the Markov network model on the testing set\n",
    "y_pred_markov = []\n",
    "y_pred_proba_markov = []\n",
    "\n",
    "for _, row in X_test_selected.iterrows():\n",
    "    # Map the feature values to their corresponding state values\n",
    "    mapped_row = map_values_to_states(row, potential_functions)\n",
    "    evidence = mapped_row\n",
    "    query_result = inference.query(['readmitted'], evidence)\n",
    "    predicted_class = query_result.values.ravel().argmax()\n",
    "    predicted_proba = query_result.values.ravel()\n",
    "    y_pred_markov.append(predicted_class)\n",
    "    y_pred_proba_markov.append(predicted_proba)\n",
    "\n",
    "y_pred_proba_markov = np.array(y_pred_proba_markov)\n",
    "\n",
    "# Calculate the evaluation metrics\n",
    "accuracy_markov = accuracy_score(y_test, y_pred_markov)\n",
    "precision_markov = precision_score(y_test, y_pred_markov)\n",
    "recall_markov = recall_score(y_test, y_pred_markov)\n",
    "f1_markov = f1_score(y_test, y_pred_markov)\n",
    "auc_roc_markov = roc_auc_score(y_test, y_pred_proba_markov[:, 1])\n",
    "\n",
    "print(\"Markov Network Model Evaluation:\")\n",
    "print(f\"Accuracy: {accuracy_markov:.4f}\")\n",
    "print(f\"Precision: {precision_markov:.4f}\")\n",
    "print(f\"Recall: {recall_markov:.4f}\")\n",
    "print(f\"F1-score: {f1_markov:.4f}\")\n",
    "print(f\"AUC-ROC: {auc_roc_markov:.4f}\")\n",
    "print()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "3f8d0a5f-4860-428d-83ec-3807046d525b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Define the model names and their respective scores\n",
    "model_names = ['Markov Network Model', 'Gradient Boosting']\n",
    "accuracy_scores = [0.6000, 0.6667]\n",
    "precision_scores = [0.6000, 0.6538]\n",
    "recall_scores = [1.0000, 0.9444]\n",
    "f1_scores = [0.7500, 0.7727]\n",
    "auc_roc_scores = [0.2928, 0.7384]\n",
    "\n",
    "# Create a dataset for plotting\n",
    "data = {\n",
    "    'Model': model_names * 5,\n",
    "    'Metric': ['Accuracy'] * 2 + ['Precision'] * 2 + ['Recall'] * 2 + ['F1-score'] * 2 + ['AUC-ROC'] * 2,\n",
    "    'Score': accuracy_scores + precision_scores + recall_scores + f1_scores + auc_roc_scores\n",
    "}\n",
    "\n",
    "# Set the figure size\n",
    "plt.figure(figsize=(10, 6))\n",
    "\n",
    "# Create the bar plot using Seaborn\n",
    "sns.barplot(x='Model', y='Score', hue='Metric', data=data, palette='rocket')\n",
    "\n",
    "# Set the title and axis labels\n",
    "plt.title('Model Evaluation Metrics', fontsize=16)\n",
    "plt.xlabel('Model', fontsize=14)\n",
    "plt.ylabel('Score', fontsize=14)\n",
    "\n",
    "# Rotate the x-axis labels for better visibility\n",
    "plt.xticks(rotation=45, ha='right')\n",
    "\n",
    "# Adjust the legend position\n",
    "plt.legend(bbox_to_anchor=(1.02, 1), loc='upper left', borderaxespad=0)\n",
    "\n",
    "# Show the plot\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ee0c0197-d267-4ca7-8b9d-abe8935fd3e7",
   "metadata": {},
   "source": [
    "## IV. Model Comparision"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c970c455-0aed-4230-a1f2-c06959c1ca4f",
   "metadata": {},
   "source": [
    "#### 1. Accuracy:\n",
    "\n",
    "Gradient Boosting Classifier: 0.6667 (66.67%)\n",
    "\n",
    "Markov Network Model: 0.6 (60%) \n",
    "\n",
    "The Gradient Boosting Classifier has a higher accuracy, correctly predicting the target variable for a larger proportion of instances compared to the Markov Network Model.\n",
    "\n",
    "#### 2. Precision:\n",
    "   \n",
    "Gradient Boosting Classifier: 0.6538 (65.38%)\n",
    "\n",
    "Markov Network Model: 0.6 (60%) \n",
    "\n",
    "The Gradient Boosting Classifier has a higher precision, indicating that among the instances it predicted as positive (readmitted), a larger proportion of them are actually positive compared to the Markov Network Model.\n",
    "\n",
    "#### 3. Recall:\n",
    "\n",
    "Gradient Boosting Classifier: 0.9444 (94.44%)\n",
    "\n",
    "Markov Network Model: 1.0000 (100.00%) \n",
    "\n",
    "The Markov Network Model has a perfect recall of 1.0000, meaning it correctly identifies all the actual positive instances. In contrast, the Gradient Boosting Classifier has a recall of 0.7429, indicating that it misses some positive instances.\n",
    "\n",
    "#### 4. F1-score:\n",
    "\n",
    "Gradient Boosting Classifier: 0.7727 (77.27%)\n",
    "\n",
    "Markov Network Model: 0.75 (75%) \n",
    "\n",
    "The Markov Network Model has a slightly higher F1-score, which is the harmonic mean of precision and recall. This suggests that the Markov Network Model has a better balance between precision and recall compared to the Gradient Boosting Classifier.\n",
    "\n",
    "#### 5. AUC-ROC:\n",
    "\n",
    "Gradient Boosting Classifier: 0.7384 (73.84%)\n",
    "\n",
    "Markov Network Model: 0.2928 (29.28%) \n",
    "\n",
    "The Gradient Boosting Classifier has a significantly higher AUC-ROC score, indicating that it has a better ability to distinguish between positive and negative instances across different probability thresholds. The Markov Network Model's AUC-ROC score of 0.2928 suggests that it performs worse than random guessing (AUC-ROC of 0.5). The Markov Model is likely predicting a high number of false positives. In other words, while the model correctly identifies all the actual positive instances (perfect recall), it also incorrectly classifies many negative instances as positive, resulting in a high false positive rate.\n",
    "\n",
    "Overall, the Gradient Boosting Classifier outperforms the Markov Network Model in terms of accuracy, precision, and AUC-ROC. It has a better overall predictive performance and a stronger ability to discriminate between positive and negative instances.\n",
    "\n",
    "However, the Markov Network Model has a perfect recall, meaning it captures all the positive instances, but at the cost of lower precision. The Markov Network Model has an F1 score of 0.7500, indicating a relatively good balance between precision and recall."
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.9.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}