AI_project.ipynb

{
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/FrancescoMonaco/BiasMedical/blob/main/AI_project.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "q76RfSVX9KVd"
      },
      "source": [
        "#Setup\n",
        "\n",
        "---\n",
        "Base imports, definition of basic models and utility functions\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 1,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "FVgvsdja4FAW",
        "outputId": "c89131b3-7dbd-4b58-d9f4-a375dfa4d0b5",
        "cellView": "form"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Looking in indexes: https://pypi.org/simple, https://us-python.pkg.dev/colab-wheels/public/simple/\n",
            "Requirement already satisfied: imbalanced-learn in /usr/local/lib/python3.10/dist-packages (0.10.1)\n",
            "Requirement already satisfied: numpy>=1.17.3 in /usr/local/lib/python3.10/dist-packages (from imbalanced-learn) (1.22.4)\n",
            "Requirement already satisfied: scipy>=1.3.2 in /usr/local/lib/python3.10/dist-packages (from imbalanced-learn) (1.10.1)\n",
            "Requirement already satisfied: scikit-learn>=1.0.2 in /usr/local/lib/python3.10/dist-packages (from imbalanced-learn) (1.2.2)\n",
            "Requirement already satisfied: joblib>=1.1.1 in /usr/local/lib/python3.10/dist-packages (from imbalanced-learn) (1.2.0)\n",
            "Requirement already satisfied: threadpoolctl>=2.0.0 in /usr/local/lib/python3.10/dist-packages (from imbalanced-learn) (3.1.0)\n",
            "Looking in indexes: https://pypi.org/simple, https://us-python.pkg.dev/colab-wheels/public/simple/\n",
            "Collecting fairlearn\n",
            "  Downloading fairlearn-0.8.0-py3-none-any.whl (235 kB)\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m235.0/235.0 kB\u001b[0m \u001b[31m5.7 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hRequirement already satisfied: numpy>=1.17.2 in /usr/local/lib/python3.10/dist-packages (from fairlearn) (1.22.4)\n",
            "Requirement already satisfied: pandas>=0.25.1 in /usr/local/lib/python3.10/dist-packages (from fairlearn) (1.5.3)\n",
            "Requirement already satisfied: scikit-learn>=0.22.1 in /usr/local/lib/python3.10/dist-packages (from fairlearn) (1.2.2)\n",
            "Requirement already satisfied: scipy>=1.4.1 in /usr/local/lib/python3.10/dist-packages (from fairlearn) (1.10.1)\n",
            "Requirement already satisfied: python-dateutil>=2.8.1 in /usr/local/lib/python3.10/dist-packages (from pandas>=0.25.1->fairlearn) (2.8.2)\n",
            "Requirement already satisfied: pytz>=2020.1 in /usr/local/lib/python3.10/dist-packages (from pandas>=0.25.1->fairlearn) (2022.7.1)\n",
            "Requirement already satisfied: joblib>=1.1.1 in /usr/local/lib/python3.10/dist-packages (from scikit-learn>=0.22.1->fairlearn) (1.2.0)\n",
            "Requirement already satisfied: threadpoolctl>=2.0.0 in /usr/local/lib/python3.10/dist-packages (from scikit-learn>=0.22.1->fairlearn) (3.1.0)\n",
            "Requirement already satisfied: six>=1.5 in /usr/local/lib/python3.10/dist-packages (from python-dateutil>=2.8.1->pandas>=0.25.1->fairlearn) (1.16.0)\n",
            "Installing collected packages: fairlearn\n",
            "Successfully installed fairlearn-0.8.0\n"
          ]
        }
      ],
      "source": [
        "#@title Libraries installation and base imports\n",
        "!sudo pip install imbalanced-learn\n",
        "!sudo pip install fairlearn\n",
        "# Base import\n",
        "import pandas as pd\n",
        "import numpy as np\n",
        "import matplotlib.pyplot as plt\n",
        "# Model import\n",
        "from sklearn.linear_model import LassoCV, LinearRegression\n",
        "from sklearn.metrics import roc_auc_score, mean_squared_error, accuracy_score\n",
        "from sklearn.model_selection import train_test_split, KFold\n",
        "from sklearn.preprocessing import MinMaxScaler\n",
        "# Support import\n",
        "import seaborn as sns\n",
        "import imblearn\n",
        "import os\n",
        "import requests\n",
        "from io import StringIO\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 2,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "2O_V-R0k9bAE",
        "outputId": "9aafcb57-ac48-468e-abcd-36e2afa759a4",
        "cellView": "form"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "From now on index 0 will indicate white people, index 1 will indicate black people\n",
            "Dataset distribution:\n",
            "0    43202\n",
            "1     5582\n"
          ]
        }
      ],
      "source": [
        "#@title Dataset load and base processing\n",
        "# Load the dataset\n",
        "url = \"https://gitlab.com/labsysmed/dissecting-bias/-/raw/master/data/data_new.csv\"\n",
        "df = pd.read_csv(StringIO(requests.get(url).text))\n",
        "\n",
        "# Extract race feauture\n",
        "race_mapping = {'white': 0, 'black': 1}\n",
        "df['race'] = df['race'].map(race_mapping)\n",
        "\n",
        "race = df['race']\n",
        "# Substitute NaN values in order to use the data\n",
        "df.fillna(0, inplace=True)\n",
        "print(\"From now on index 0 will indicate white people, index 1 will indicate black people\\n\",\"Dataset distribution:\\n\", df['race'].value_counts().to_string(dtype=False), sep='')"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 3,
      "metadata": {
        "id": "ej185boMK0pf"
      },
      "outputs": [],
      "source": [
        "#@title X, y extraction\n",
        "\n",
        "X = df.drop(['cost_t', 'gagne_sum_t', 'cost_avoidable_t'], axis = 1)\n",
        "\n",
        "# Extract target variables (y1, y2, y3) from DataFrame\n",
        "y1 = df['cost_t']\n",
        "y2 = df['gagne_sum_t']\n",
        "y3 = df['cost_avoidable_t']\n",
        "y = df[['cost_t', 'gagne_sum_t', 'cost_avoidable_t']]"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 4,
      "metadata": {
        "id": "_deNqyEI8AAT"
      },
      "outputs": [],
      "source": [
        "#@title Function for plotting num of comorbidities and spread of people at various risk levels\n",
        "def plot_comorbidities(df, tit):\n",
        "  \"\"\"Function for plotting num of comorbidities and spread of people at various risk levels\n",
        "\n",
        "  Parameters\n",
        "  ------\n",
        "  df: pd.DataFrame\n",
        "    Dataframe where the data is stored\n",
        "  tit: str\n",
        "    String to use as title of the plot\n",
        "\n",
        "  Returns\n",
        "  ------\n",
        "  void\n",
        "  \"\"\"\n",
        "  \n",
        "  columns_to_sum = ['alcohol_elixhauser_tm1','anemia_elixhauser_tm1',\n",
        "              'arrhythmia_elixhauser_tm1','arthritis_elixhauser_tm1','bloodlossanemia_elixhauser_tm1','coagulopathy_elixhauser_tm1',\n",
        "              'compdiabetes_elixhauser_tm1','depression_elixhauser_tm1','drugabuse_elixhauser_tm1','electrolytes_elixhauser_tm1',\n",
        "              'hypertension_elixhauser_tm1','hypothyroid_elixhauser_tm1','liver_elixhauser_tm1','neurodegen_elixhauser_tm1','obesity_elixhauser_tm1',\n",
        "              'paralysis_elixhauser_tm1','psychosis_elixhauser_tm1','pulmcirc_elixhauser_tm1','pvd_elixhauser_tm1','renal_elixhauser_tm1','uncompdiabetes_elixhauser_tm1',\n",
        "              'valvulardz_elixhauser_tm1','wtloss_elixhauser_tm1','cerebrovasculardz_romano_tm1','chf_romano_tm1','dementia_romano_tm1','hemiplegia_romano_tm1',\n",
        "              'hivaids_romano_tm1','metastatic_romano_tm1','myocardialinfarct_romano_tm1','pulmonarydz_romano_tm1','tumor_romano_tm1','ulcer_romano_tm1']\n",
        "\n",
        "  sum_series = df.loc[:,columns_to_sum].sum(axis=1)\n",
        "  print(\"max:\",sum_series.max())\n",
        "\n",
        "  # Create a new DataFrame 'sum_df' from the sum Series\n",
        "  sum_df = pd.DataFrame({'Sum': sum_series})\n",
        "  df_c = df.copy()\n",
        "  df_c.loc[:,'sum'] = df.loc[:,'gagne_sum_tm1']\n",
        "\n",
        "  # Scale the raw Risk scores to 0-100 \n",
        "  scaler = MinMaxScaler(feature_range=(0, 100))\n",
        "\n",
        "\n",
        "  df_plot1 = df_c.loc[df['race'] == 0]\n",
        "  df_plot2 = df_c.loc[df['race'] == 1]\n",
        "\n",
        "  scaler.fit(df_plot1['risk_score_t'].values.reshape(-1, 1))\n",
        "  df_plot1.loc[:,'risk_score_t_scaled'] = scaler.transform(df_plot1['risk_score_t'].values.reshape(-1, 1)).astype(int)\n",
        "  scaler.fit(df_plot2['risk_score_t'].values.reshape(-1, 1))\n",
        "  df_plot2.loc[:,'risk_score_t_scaled'] = scaler.transform(df_plot2['risk_score_t'].values.reshape(-1, 1)).astype(int)\n",
        "\n",
        " # Print the percentage at high risk\n",
        "  print(\"*-------------------------------------*\\n\", \\\n",
        "        \"Percentage of White People in the high risk section: \",\\\n",
        "        df_plot1[df_plot1.risk_score_t_scaled>75].shape[0]/(df_plot1[df_plot1.risk_score_t_scaled>75].shape[0]+df_plot2[df_plot2.risk_score_t_scaled>75].shape[0]),\\\n",
        "        \"%\\nPercentage of Black People in the high risk section: \",\\\n",
        "        df_plot2[df_plot2.risk_score_t_scaled>75].shape[0]/(df_plot1[df_plot1.risk_score_t_scaled>75].shape[0]+df_plot2[df_plot2.risk_score_t_scaled>75].shape[0]),\\\n",
        "        \"%\\n\", \"*-------------------------------------*\\n\",sep='')\n",
        "  print(\"*-------------------------------------*\\n\", \\\n",
        "        \"Num of White People in the high risk section: \",\\\n",
        "        df_plot1[df_plot1.risk_score_t_scaled>75].shape[0],\\\n",
        "        \"\\nNum of Black People in the high risk section: \",\\\n",
        "        df_plot2[df_plot2.risk_score_t_scaled>75].shape[0],\\\n",
        "        \"\\n\", \"*-------------------------------------*\\n\",sep='')\n",
        "\n",
        " # Create the plots\n",
        "  counts_1 = df_plot1['risk_score_t_scaled'].value_counts() \n",
        "  counts_2 = df_plot2['risk_score_t_scaled'].value_counts()\n",
        "\n",
        "  print(\"mean:\",df_plot1['risk_score_t_scaled'].mean())\n",
        "\n",
        "  if tit == 'New Label model' or tit == 'Decorrelate model': # A polynomial of degree 3 acts strangely in this context\n",
        "    coefficients1 = np.polyfit(df_plot1['risk_score_t_scaled'], df_plot1['sum'], deg=4)\n",
        "  else:\n",
        "    coefficients1 = np.polyfit(df_plot1['risk_score_t_scaled'], df_plot1['sum'], deg=3)\n",
        "  \n",
        "  polynomial1 = np.poly1d(coefficients1)\n",
        "\n",
        "  coefficients2 = np.polyfit(df_plot2['risk_score_t_scaled'], df_plot2['sum'], deg=3)\n",
        "  polynomial2 = np.poly1d(coefficients2)\n",
        "\n",
        "  x_count1 = counts_1.index.to_numpy()\n",
        "  y_count1 = counts_1.values\n",
        "  y_count1 = (y_count1-np.min(y_count1))/(np.max(y_count1)-np.min(y_count1))\n",
        "\n",
        "  x_count2 = counts_2.index.to_numpy()\n",
        "  y_count2 = counts_2.values\n",
        "  y_count2 = (y_count2-np.min(y_count2))/(np.max(y_count2)-np.min(y_count2))\n",
        "\n",
        "  #coeffsum1 = np.polyfit(x_count1, y_count1, deg=4)\n",
        "  #coeffsum2 = np.polyfit(x_count2, y_count2, deg=4)\n",
        "\n",
        "  #poly3 = np.poly1d(coeffsum1)\n",
        "  #poly4 = np.poly1d(coeffsum2)\n",
        "\n",
        "  # Generate x values for the curve\n",
        "  x_vals = np.linspace(0, 100, num=100)\n",
        "  x_vals2 = np.linspace(10,100, num= 90)\n",
        "\n",
        "  # Evaluate the polynomial at x values to get y values for the curve\n",
        "  y_vals1 = polynomial1(x_vals)\n",
        "  y_vals2 = polynomial2(x_vals)\n",
        "  y_vals3 = np.interp(x_vals2, x_count1, y_count1)#poly3(x_vals2)\n",
        "  y_vals4 = np.interp(x_vals2, x_count2, y_count2)#poly4(x_vals2)\n",
        "\n",
        "  # Create a plot\n",
        "  plt.rcParams['figure.figsize'] = (12, 6)\n",
        "  f, (ax1, ax2) = plt.subplots(1, 2)\n",
        "    # Comorbidity plot\n",
        "  ax1.plot(x_vals, y_vals1, label='White People')\n",
        "  ax1.plot(x_vals, y_vals2, label='Black People')\n",
        "  ax1.set_xlabel('risk_score_t')\n",
        "  ax1.set_ylabel('comorbidities')\n",
        "  ax1.set_title('Risk score vs Num of Comorbidities')\n",
        "  ax1.legend()\n",
        "  ax1.grid(True)\n",
        "   # Number plot\n",
        "  ax2.plot(x_vals2, y_vals3, label='White People')\n",
        "  ax2.plot(x_vals2, y_vals4, label='Black People')\n",
        "  ax2.set_xlabel('risk_score_t')\n",
        "  ax2.set_ylabel('% of people in bin')\n",
        "  ax2.set_title('% of people vs Risk Score')\n",
        "  ax2.legend()\n",
        "  ax2.grid(True)\n",
        "\n",
        "  f.suptitle(tit, fontsize=16)\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 5,
      "metadata": {
        "id": "1C89ZuZicY1m"
      },
      "outputs": [],
      "source": [
        "#@title Lasso model \n",
        "def train_lasso(X, y1, race, y2=None, y3=None):\n",
        "  \"\"\"Function to train a lasso regression\n",
        "\n",
        "  Parameters\n",
        "  ------\n",
        "  X: pd.DataFrame\n",
        "    features\n",
        "  y1: pd.DataFrame\n",
        "    target values\n",
        "  race: pd.DataFrame\n",
        "    sensible feature\n",
        "  y2: pd.DataFrame, optional\n",
        "    second target values\n",
        "  y3: pd.DataFrame, optional\n",
        "    third target values\n",
        "    \n",
        "  Returns\n",
        "  ------\n",
        "  pd.DataFrame\n",
        "    Mean Squared Error of the training\n",
        "  LassoCV\n",
        "    Trained model\n",
        "  \"\"\"\n",
        "  if y2 is None and y3 is None:\n",
        "    X_train, X_val, y1_train, y1_val, race_train, race_val = train_test_split(X, y1, race, test_size=0.2, random_state=42)\n",
        "    # Initialize LassoCV model\n",
        "    lasso_model = LassoCV(cv=5) \n",
        "\n",
        "    # Train the LassoCV model\n",
        "    lasso_model.fit(X_train, y1_train)\n",
        "    y1_pred = lasso_model.predict(X_val)\n",
        "\n",
        "    # Calculate mean squared error\n",
        "    mse_1 = mean_squared_error(y1_val, y1_pred)\n",
        "    # Create a DataFrame to store the MSE values\n",
        "    mse_df = pd.DataFrame({'Target Variable': ['y1'],\n",
        "                          'MSE': [mse_1]})\n",
        "\n",
        "    # Filter rows based on 'race'\n",
        "    race_black_mse_1 = mean_squared_error(y1_val[race_val == 1], y1_pred[race_val == 1])\n",
        "    race_white_mse_1 = mean_squared_error(y1_val[race_val == 0], y1_pred[race_val == 0])\n",
        "\n",
        "    mse_df = pd.DataFrame({'Target Variable': ['Cost_tot'],\n",
        "                          'MSE': [mse_1],\n",
        "                          'Race: Black MSE': [race_black_mse_1],\n",
        "                          'Race: White MSE': [race_white_mse_1]})\n",
        "    return mse_df, lasso_model\n",
        "\n",
        "  else:\n",
        "    X_train, X_val, y1_train, y1_val, y2_train, y2_val, y3_train, y3_val, race_train, race_val = train_test_split(X, y1, y2, y3, race, test_size=0.2, random_state=42)\n",
        "    # Initialize LassoCV model\n",
        "    lasso_model = LassoCV(cv=5) \n",
        "\n",
        "    # Train the LassoCV model for each target variable\n",
        "    lasso_model.fit(X_train, y1_train)\n",
        "    y1_pred = lasso_model.predict(X_val)\n",
        "\n",
        "    lasso_model.fit(X_train, y2_train)\n",
        "    y2_pred = lasso_model.predict(X_val)\n",
        "\n",
        "    lasso_model.fit(X_train, y3_train)\n",
        "    y3_pred = lasso_model.predict(X_val)\n",
        "\n",
        "    # Calculate mean squared error for each target variable\n",
        "    mse_1 = mean_squared_error(y1_val, y1_pred)\n",
        "    mse_2 = mean_squared_error(y2_val, y2_pred)\n",
        "    mse_3 = mean_squared_error(y3_val, y3_pred)\n",
        "    # Create a DataFrame to store the MSE values\n",
        "    mse_df = pd.DataFrame({'Target Variable': ['y1', 'y2', 'y3'],\n",
        "                          'MSE': [mse_1, mse_2, mse_3]})\n",
        "\n",
        "    # Filter rows based on 'race'\n",
        "    race_black_mse_1 = mean_squared_error(y1_val[race_val == 1], y1_pred[race_val == 1])\n",
        "    race_white_mse_1 = mean_squared_error(y1_val[race_val == 0], y1_pred[race_val == 0])\n",
        "\n",
        "    race_black_mse_2 = mean_squared_error(y2_val[race_val == 1], y2_pred[race_val == 1])\n",
        "    race_white_mse_2 = mean_squared_error(y2_val[race_val == 0], y2_pred[race_val == 0])\n",
        "\n",
        "    race_black_mse_3 = mean_squared_error(y3_val[race_val == 1], y3_pred[race_val == 1])\n",
        "    race_white_mse_3 = mean_squared_error(y3_val[race_val == 0], y3_pred[race_val == 0])\n",
        "\n",
        "    mse_df = pd.DataFrame({'Target Variable': ['Cost_tot', 'Gagne_sum', 'Cost_avoid'],\n",
        "                          'MSE': [mse_1, mse_2, mse_3],\n",
        "                          'Race: Black MSE': [race_black_mse_1, race_black_mse_2, race_black_mse_3],\n",
        "                          'Race: White MSE': [race_white_mse_1, race_white_mse_2, race_white_mse_3]})\n",
        "    return mse_df, lasso_model"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 6,
      "metadata": {
        "id": "QcKlcQCBF73W"
      },
      "outputs": [],
      "source": [
        "#@title Names of the main columns of the dataset\n",
        "col_names = ['program_enrolled_t','bps_mean_t','ghba1c_mean_t','hct_mean_t','cre_mean_t','ldl_mean_t','dem_female','dem_age_band_18-24_tm1','dem_age_band_25-34_tm1',\n",
        "             'dem_age_band_35-44_tm1','dem_age_band_45-54_tm1','dem_age_band_55-64_tm1','dem_age_band_65-74_tm1','dem_age_band_75+_tm1','alcohol_elixhauser_tm1','anemia_elixhauser_tm1',\n",
        "             'arrhythmia_elixhauser_tm1','arthritis_elixhauser_tm1','bloodlossanemia_elixhauser_tm1','coagulopathy_elixhauser_tm1',\n",
        "             'compdiabetes_elixhauser_tm1','depression_elixhauser_tm1','drugabuse_elixhauser_tm1','electrolytes_elixhauser_tm1',\n",
        "             'hypertension_elixhauser_tm1','hypothyroid_elixhauser_tm1','liver_elixhauser_tm1','neurodegen_elixhauser_tm1','obesity_elixhauser_tm1',\n",
        "             'paralysis_elixhauser_tm1','psychosis_elixhauser_tm1','pulmcirc_elixhauser_tm1','pvd_elixhauser_tm1','renal_elixhauser_tm1','uncompdiabetes_elixhauser_tm1',\n",
        "             'valvulardz_elixhauser_tm1','wtloss_elixhauser_tm1','cerebrovasculardz_romano_tm1','chf_romano_tm1','dementia_romano_tm1','hemiplegia_romano_tm1',\n",
        "             'hivaids_romano_tm1','metastatic_romano_tm1','myocardialinfarct_romano_tm1','pulmonarydz_romano_tm1','tumor_romano_tm1','ulcer_romano_tm1',\n",
        "             'cost_dialysis_tm1','cost_emergency_tm1','cost_home_health_tm1','cost_ip_medical_tm1','cost_ip_surgical_tm1','cost_laboratory_tm1',\n",
        "             'cost_op_primary_care_tm1','cost_op_specialists_tm1','cost_op_surgery_tm1','cost_other_tm1','cost_pharmacy_tm1','cost_physical_therapy_tm1',\n",
        "             'cost_radiology_tm1','lasix_dose_count_tm1','lasix_min_daily_dose_tm1','lasix_mean_daily_dose_tm1','lasix_max_daily_dose_tm1','cre_tests_tm1',\n",
        "             'crp_tests_tm1','esr_tests_tm1','ghba1c_tests_tm1','hct_tests_tm1','ldl_tests_tm1','nt_bnp_tests_tm1','sodium_tests_tm1','trig_tests_tm1','cre_min-low_tm1',\n",
        "             'cre_min-high_tm1','cre_min-normal_tm1','cre_mean-low_tm1','cre_mean-high_tm1','cre_mean-normal_tm1','cre_max-low_tm1','cre_max-high_tm1','cre_max-normal_tm1',\n",
        "             'crp_min-low_tm1','crp_min-high_tm1','crp_min-normal_tm1','crp_mean-low_tm1','crp_mean-high_tm1','crp_mean-normal_tm1','crp_max-low_tm1','crp_max-high_tm1',\n",
        "             'crp_max-normal_tm1','esr_min-low_tm1','esr_min-high_tm1','esr_min-normal_tm1','esr_mean-low_tm1','esr_mean-high_tm1','esr_mean-normal_tm1','esr_max-low_tm1',\n",
        "             'esr_max-high_tm1','esr_max-normal_tm1','ghba1c_min-low_tm1','ghba1c_min-high_tm1','ghba1c_min-normal_tm1','ghba1c_mean-low_tm1','ghba1c_mean-high_tm1',\n",
        "             'ghba1c_mean-normal_tm1','ghba1c_max-low_tm1','ghba1c_max-high_tm1','ghba1c_max-normal_tm1','hct_min-low_tm1','hct_min-high_tm1','hct_min-normal_tm1',\n",
        "             'hct_mean-low_tm1','hct_mean-high_tm1','hct_mean-normal_tm1','hct_max-low_tm1','hct_max-high_tm1','hct_max-normal_tm1','ldl_min-low_tm1','ldl_min-high_tm1',\n",
        "             'ldl_min-normal_tm1','ldl-mean-low_tm1','ldl-mean-high_tm1','ldl-mean-normal_tm1','ldl_max-low_tm1','ldl_max-high_tm1','ldl_max-normal_tm1',\n",
        "             'nt_bnp_min-low_tm1','nt_bnp_min-high_tm1','nt_bnp_min-normal_tm1','nt_bnp_mean-low_tm1','nt_bnp_mean-high_tm1','nt_bnp_mean-normal_tm1',\n",
        "             'nt_bnp_max-low_tm1','nt_bnp_max-high_tm1','nt_bnp_max-normal_tm1','sodium_min-low_tm1','sodium_min-high_tm1','sodium_min-normal_tm1',\n",
        "             'sodium_mean-low_tm1','sodium_mean-high_tm1','sodium_mean-normal_tm1','sodium_max-low_tm1','sodium_max-high_tm1','sodium_max-normal_tm1',\n",
        "             'trig_min-low_tm1','trig_min-high_tm1','trig_min-normal_tm1',\n",
        "             'trig_mean-low_tm1','trig_mean-high_tm1','trig_mean-normal_tm1','trig_max-low_tm1','trig_max-high_tm1','trig_max-normal_tm1','gagne_sum_tm1']"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "IeXpDqAxBmB7"
      },
      "source": [
        "#Base Model\n",
        "\n",
        "---\n",
        "The base model mimics the one used in common health applications produced by insurance companies in the USA to manage health in populations.\n",
        "The model is based on the one showed in Obermeyer, Z., Powers, B., Vogeli, C., Mullainathan, S., *Dissecting racial bias in an algorithm used to manage the health of populations*, Science 366, https://escholarship.org/content/qt6h92v832/qt6h92v832.pdf, pp.447-453, 2019.\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 7,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "5WuVvVFGBvPX",
        "outputId": "1281e6cd-e91f-4494-8feb-f192ba936d12"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Target Variable          MSE  Race: Black MSE  Race: White MSE\n",
            "       Cost_tot 2.656967e+08     4.160072e+08     2.459493e+08\n",
            "      Gagne_sum 2.784019e+00     4.464028e+00     2.563304e+00\n",
            "     Cost_avoid 1.386470e+08     1.617675e+08     1.356095e+08\n"
          ]
        }
      ],
      "source": [
        "#@title Training\n",
        "mse_df, _ = train_lasso(X, y1, race, y2, y3)\n",
        "\n",
        "# Print\n",
        "print(mse_df.to_string(index=False))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 8,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "3V1c-W8y0eyH",
        "outputId": "c95e2f0c-6af1-4f66-ca1e-3d55d39f94d6"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "max: 18\n",
            "*-------------------------------------*\n",
            "Percentage of White People in the high risk section: 0.8125%\n",
            "Percentage of Black People in the high risk section: 0.1875%\n",
            "*-------------------------------------*\n",
            "\n",
            "*-------------------------------------*\n",
            "Num of White People in the high risk section: 13\n",
            "Num of Black People in the high risk section: 3\n",
            "*-------------------------------------*\n",
            "\n",
            "mean: 3.7695477061247162\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "<ipython-input-4-df564a953dd6>:41: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  df_plot1.loc[:,'risk_score_t_scaled'] = scaler.transform(df_plot1['risk_score_t'].values.reshape(-1, 1)).astype(int)\n",
            "<ipython-input-4-df564a953dd6>:43: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  df_plot2.loc[:,'risk_score_t_scaled'] = scaler.transform(df_plot2['risk_score_t'].values.reshape(-1, 1)).astype(int)\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1200x600 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ],
      "source": [
        "plot_comorbidities(df, 'Base model')"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "UAkCD6jjQ4Cg"
      },
      "source": [
        "#Correlation Remover\n",
        "\n",
        "---\n",
        "The following approach consists in using a library specilized in fairness to pre-process the data and remove the correlation between the sensitive features and the others.\n",
        "\n",
        "Bird, S., Dudík, M., Edgar, R., Horn, B., Lutz, R., Milan, V., Sameki, M., Wallach, H., Walker, K., *Fairlearn: A toolkit for assessing and improving fairness in AI*, Techreport MSR-TR-2020-32, Microsoft, https://www.microsoft.com/en-us/research/publication/fairlearn-a-toolkit-for-assessing-and-improving-fairness-in-ai/, 2020.\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 9,
      "metadata": {
        "id": "16npjt2IOuSj",
        "cellView": "form"
      },
      "outputs": [],
      "source": [
        "#@title Clean the data\n",
        "# Load the dataset\n",
        "url = \"https://gitlab.com/labsysmed/dissecting-bias/-/raw/master/data/data_new.csv\"\n",
        "df = pd.read_csv(StringIO(requests.get(url).text))\n",
        "\n",
        "# Extract race feauture\n",
        "race_mapping = {'white': 0, 'black': 1}\n",
        "df['race'] = df['race'].map(race_mapping)\n",
        "\n",
        "race = df['race']\n",
        "# Substitute NaN values in order to use the data\n",
        "df.fillna(0, inplace=True)\n",
        "\n",
        "\n",
        "X = df.drop(['cost_t', 'gagne_sum_t', 'cost_avoidable_t'], axis = 1)\n",
        "\n",
        "# Extract target variables (y1, y2, y3) from DataFrame\n",
        "y1 = df['cost_t']\n",
        "y2 = df['gagne_sum_t']\n",
        "y3 = df['cost_avoidable_t']\n",
        "y = df[['cost_t', 'gagne_sum_t', 'cost_avoidable_t']]"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 10,
      "metadata": {
        "id": "fToTEfgyVHNm"
      },
      "outputs": [],
      "source": [
        "#@title Util function to produce heatmaps\n",
        "# As shown on an example on FairLearn\n",
        "def plot_heatmap(df, title, axis, n = 0):\n",
        "  \"\"\"Function for plotting the correlation between features\n",
        "\n",
        "  Parameters\n",
        "  ------\n",
        "  df: pd.DataFrame\n",
        "    Dataframe where the data is stored\n",
        "  title: str\n",
        "    String to use as title of the plot\n",
        "  axis: plt.axis\n",
        "    Axis where to plot the result\n",
        "  n: int, optional (default is 0)\n",
        "    number of the plot (to manage where the y labels \n",
        "                  and the colorbar should be printed)\n",
        "\n",
        "  Returns\n",
        "  ------\n",
        "  void\n",
        "  \"\"\"\n",
        "  \n",
        "  data = df[['cost_t', 'cost_op_surgery_tm1','cost_other_tm1',\n",
        "             'cost_pharmacy_tm1','cost_physical_therapy_tm1',\n",
        "             'cost_dialysis_tm1','cost_emergency_tm1','cost_home_health_tm1',\n",
        "             'risk_score_t']]\n",
        "  if n == 2:\n",
        "    dataplot = sns.heatmap(data.corr(), cmap='cividis', ax = axis, cbar = False, yticklabels = False)\n",
        "    #dataplot.set_ylabel('Column Names')\n",
        "    dataplot.set_xlabel('')\n",
        "    #dataplot.set_yticklabels(data.columns, rotation=0)\n",
        "    axis.set_title(title)\n",
        "  elif n == 3:\n",
        "    dataplot = sns.heatmap(data.corr(), cmap='cividis', ax=axis)\n",
        "    dataplot.set_ylabel('')\n",
        "    dataplot.set_xlabel('')\n",
        "    dataplot.set_yticks([])\n",
        "    axis.set_title(title)\n",
        "  else: \n",
        "    dataplot = sns.heatmap(data.corr(), cmap='cividis', ax = axis, cbar = False)\n",
        "    dataplot.set_xlabel('')\n",
        "    axis.set_title(title)\n",
        "  "
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 11,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "lAdgkQ53Q3nJ",
        "outputId": "e53a4ed8-52c7-49ef-b7b1-b0ce0e9d6925",
        "cellView": "form"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Without alpha\n",
            "Target Variable          MSE  Race: Black MSE  Race: White MSE\n",
            "       Cost_tot 3.233641e+02     6.350033e+02     2.824217e+02\n",
            "      Gagne_sum 3.099202e+00     4.941010e+00     2.857230e+00\n",
            "     Cost_avoid 1.131832e+06     1.993187e+06     1.018669e+06\n",
            "With alpha\n",
            "Target Variable          MSE  Race: Black MSE  Race: White MSE\n",
            "       Cost_tot 3.395141e+07     7.556353e+07     2.848451e+07\n",
            "      Gagne_sum 3.266279e+00     5.497961e+00     2.973086e+00\n",
            "     Cost_avoid 6.484305e+06     1.428402e+07     5.459597e+06\n"
          ]
        }
      ],
      "source": [
        "#@title Model\n",
        "from fairlearn.preprocessing import CorrelationRemover\n",
        "\n",
        "# Retrieve all the data\n",
        "var = ['risk_score_t', 'race', 'cost_t' , 'gagne_sum_t', 'cost_avoidable_t', 'cost_dialysis_tm1',\n",
        "                 'cost_emergency_tm1','cost_home_health_tm1',\n",
        "                 'cost_ip_medical_tm1','cost_ip_surgical_tm1','cost_laboratory_tm1',\n",
        "             'cost_op_primary_care_tm1','cost_op_specialists_tm1','cost_op_surgery_tm1',\n",
        "             'cost_other_tm1','cost_pharmacy_tm1','cost_physical_therapy_tm1','cost_radiology_tm1']\n",
        "\n",
        "var2 = ['program_enrolled_t','bps_mean_t','ghba1c_mean_t','hct_mean_t','cre_mean_t','ldl_mean_t','dem_female','dem_age_band_18-24_tm1','dem_age_band_25-34_tm1',\n",
        "             'dem_age_band_35-44_tm1','dem_age_band_45-54_tm1','dem_age_band_55-64_tm1','dem_age_band_65-74_tm1','dem_age_band_75+_tm1','alcohol_elixhauser_tm1','anemia_elixhauser_tm1',\n",
        "             'arrhythmia_elixhauser_tm1','arthritis_elixhauser_tm1','bloodlossanemia_elixhauser_tm1','coagulopathy_elixhauser_tm1',\n",
        "             'compdiabetes_elixhauser_tm1','depression_elixhauser_tm1','drugabuse_elixhauser_tm1','electrolytes_elixhauser_tm1',\n",
        "             'hypertension_elixhauser_tm1','hypothyroid_elixhauser_tm1','liver_elixhauser_tm1','neurodegen_elixhauser_tm1','obesity_elixhauser_tm1',\n",
        "             'paralysis_elixhauser_tm1','psychosis_elixhauser_tm1','pulmcirc_elixhauser_tm1','pvd_elixhauser_tm1','renal_elixhauser_tm1','uncompdiabetes_elixhauser_tm1',\n",
        "             'valvulardz_elixhauser_tm1','wtloss_elixhauser_tm1','cerebrovasculardz_romano_tm1','chf_romano_tm1','dementia_romano_tm1','hemiplegia_romano_tm1',\n",
        "             'hivaids_romano_tm1','metastatic_romano_tm1','myocardialinfarct_romano_tm1','pulmonarydz_romano_tm1','tumor_romano_tm1','ulcer_romano_tm1',\n",
        "             'lasix_dose_count_tm1','lasix_min_daily_dose_tm1','lasix_mean_daily_dose_tm1','lasix_max_daily_dose_tm1','cre_tests_tm1',\n",
        "             'crp_tests_tm1','esr_tests_tm1','ghba1c_tests_tm1','hct_tests_tm1','ldl_tests_tm1','nt_bnp_tests_tm1','sodium_tests_tm1','trig_tests_tm1','cre_min-low_tm1',\n",
        "             'cre_min-high_tm1','cre_min-normal_tm1','cre_mean-low_tm1','cre_mean-high_tm1','cre_mean-normal_tm1','cre_max-low_tm1','cre_max-high_tm1','cre_max-normal_tm1',\n",
        "             'crp_min-low_tm1','crp_min-high_tm1','crp_min-normal_tm1','crp_mean-low_tm1','crp_mean-high_tm1','crp_mean-normal_tm1','crp_max-low_tm1','crp_max-high_tm1',\n",
        "             'crp_max-normal_tm1','esr_min-low_tm1','esr_min-high_tm1','esr_min-normal_tm1','esr_mean-low_tm1','esr_mean-high_tm1','esr_mean-normal_tm1','esr_max-low_tm1',\n",
        "             'esr_max-high_tm1','esr_max-normal_tm1','ghba1c_min-low_tm1','ghba1c_min-high_tm1','ghba1c_min-normal_tm1','ghba1c_mean-low_tm1','ghba1c_mean-high_tm1',\n",
        "             'ghba1c_mean-normal_tm1','ghba1c_max-low_tm1','ghba1c_max-high_tm1','ghba1c_max-normal_tm1','hct_min-low_tm1','hct_min-high_tm1','hct_min-normal_tm1',\n",
        "             'hct_mean-low_tm1','hct_mean-high_tm1','hct_mean-normal_tm1','hct_max-low_tm1','hct_max-high_tm1','hct_max-normal_tm1','ldl_min-low_tm1','ldl_min-high_tm1',\n",
        "             'ldl_min-normal_tm1','ldl-mean-low_tm1','ldl-mean-high_tm1','ldl-mean-normal_tm1','ldl_max-low_tm1','ldl_max-high_tm1','ldl_max-normal_tm1',\n",
        "             'nt_bnp_min-low_tm1','nt_bnp_min-high_tm1','nt_bnp_min-normal_tm1','nt_bnp_mean-low_tm1','nt_bnp_mean-high_tm1','nt_bnp_mean-normal_tm1',\n",
        "             'nt_bnp_max-low_tm1','nt_bnp_max-high_tm1','nt_bnp_max-normal_tm1','sodium_min-low_tm1','sodium_min-high_tm1','sodium_min-normal_tm1',\n",
        "             'sodium_mean-low_tm1','sodium_mean-high_tm1','sodium_mean-normal_tm1','sodium_max-low_tm1','sodium_max-high_tm1','sodium_max-normal_tm1',\n",
        "             'trig_min-low_tm1','trig_min-high_tm1','trig_min-normal_tm1',\n",
        "             'trig_mean-low_tm1','trig_mean-high_tm1','trig_mean-normal_tm1','trig_max-low_tm1','trig_max-high_tm1','trig_max-normal_tm1','gagne_sum_tm1']\n",
        "X_toProcess = df[var]\n",
        "\n",
        "# Extract target variables (y1, y2, y3) from DataFrame\n",
        "y1 = df['cost_t']\n",
        "y2 = df['gagne_sum_t']\n",
        "y3 = df['cost_avoidable_t']\n",
        "\n",
        "# No alpha version\n",
        "cr = CorrelationRemover(sensitive_feature_ids=[\"race\",'risk_score_t', 'cost_t'])\n",
        "X_cr = cr.fit_transform(X_toProcess)\n",
        "X_cr = pd.DataFrame(X_cr, columns=var[3:])\n",
        "X_cr[\"race\"] = X[\"race\"]\n",
        "X_cr[\"risk_score_t\"] = X_toProcess[\"risk_score_t\"]\n",
        "X_cr['cost_t'] = X_toProcess['cost_t']\n",
        "X_cr = pd.concat([X_cr, df[var2]], axis=1)\n",
        "\n",
        "mse_df, model_correlation = train_lasso(X_cr, y1, race, y2, y3)\n",
        "# Print\n",
        "print(\"Without alpha\\n\", mse_df.to_string(index=False), sep='')\n",
        "\n",
        "# Alpha version\n",
        "cr_alpha = CorrelationRemover(sensitive_feature_ids=['race','risk_score_t'], alpha=0.75)\n",
        "X_toProcess = df[var]\n",
        "X_cr_alpha = cr_alpha.fit_transform(X_toProcess)\n",
        "X_cr_alpha = pd.DataFrame(X_cr_alpha, columns=var[2:])\n",
        "X_cr_alpha[\"race\"] = X[\"race\"]\n",
        "X_cr_alpha[\"risk_score_t\"] = X_toProcess[\"risk_score_t\"]\n",
        "X_cr_alpha = pd.concat([X_cr_alpha, df[var2]], axis=1)\n",
        "mse_df, _ = train_lasso(X_cr_alpha, y1, race, y2, y3)\n",
        "\n",
        "# Print\n",
        "print(\"With alpha\\n\", mse_df.to_string(index=False), sep='')"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 12,
      "metadata": {
        "id": "Tmrt_y0m0lb9",
        "cellView": "form"
      },
      "outputs": [],
      "source": [
        "#@title Calculate risk\n",
        "y_r = X_cr['risk_score_t']\n",
        "\n",
        "_, model_r = train_lasso(X_cr, y_r, race)\n",
        "\n",
        "risk = model_r.predict(X_cr)\n",
        "\n",
        "X_cr['risk_score_t'] = risk"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 13,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 0
        },
        "id": "wAjoPciZh-KS",
        "outputId": "56b53251-29fa-419a-926b-71a7084f69df"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "max: 18\n",
            "*-------------------------------------*\n",
            "Percentage of White People in the high risk section: 0.6428571428571429%\n",
            "Percentage of Black People in the high risk section: 0.35714285714285715%\n",
            "*-------------------------------------*\n",
            "\n",
            "*-------------------------------------*\n",
            "Num of White People in the high risk section: 9\n",
            "Num of Black People in the high risk section: 5\n",
            "*-------------------------------------*\n",
            "\n",
            "mean: 0.9648395907596871\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "<ipython-input-4-df564a953dd6>:41: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  df_plot1.loc[:,'risk_score_t_scaled'] = scaler.transform(df_plot1['risk_score_t'].values.reshape(-1, 1)).astype(int)\n",
            "<ipython-input-4-df564a953dd6>:43: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  df_plot2.loc[:,'risk_score_t_scaled'] = scaler.transform(df_plot2['risk_score_t'].values.reshape(-1, 1)).astype(int)\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1200x600 with 2 Axes>"
            ],
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAA9wAAAJJCAYAAABYjD9ZAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/bCgiHAAAACXBIWXMAAA9hAAAPYQGoP6dpAAEAAElEQVR4nOzdd3gUVRfA4d9uei8kkARCEiD0DoLU0HuTXpSiiIUiUlQsNCkiRVAQVKRIUYqCiPQSlN5C7wgJNaEmpJed749l92NJIYENk3Le5+HJZubOzJmbZe+emTv3ahRFURBCCCGEEEIIIYRZadUOQAghhBBCCCGEyIsk4RZCCCGEEEIIIbKBJNxCCCGEEEIIIUQ2kIRbCCGEEEIIIYTIBpJwCyGEEEIIIYQQ2UASbiGEEEIIIYQQIhtIwi2EEEIIIYQQQmQDSbiFEEIIIYQQQohsIAm3EEIIIYQQQgiRDSThFkIIkWn+/v5oNBrjP61Wi5OTE0WKFKFhw4aMGDGCgwcPqh1mrmWoV/FyLFq0CI1GQ9++fc2yv759+6LRaFi0aJFZ9ieEECL3s1Q7ACGEELlPnTp1KFGiBABxcXHcvXuXkJAQgoODmT59OkFBQSxYsIBixYqpHGn+dfXqVQICAvDz8+Pq1atqhyOEEELkS5JwCyGEyLL+/funuiuoKAobN25k6NCh7Nq1i9q1a7Nv3z4CAgLUCVIIIYQQQmXSpVwIIYRZaDQaWrVqxcGDBwkMDCQ8PJz+/furHZYQQgghhGok4RZCCGFWrq6uzJw5E4AdO3Zw5MiRVGWSk5OZP38+DRo0wN3dHRsbGwICAnjvvfe4du1auvu+ceMGI0eOpEKFCjg5OeHg4EDJkiXp27cve/fuTVX++vXrDB48mMDAQGxtbXFxcaFOnTr88MMPpKSkpCr/5DO99+/fZ+jQoRQvXhwbGxsaNGgAwNixY9FoNIwdO5awsDDeeustfH19sbKySnXXf/Xq1bRo0QJPT0+sra0pXLgwr7/+OmfOnMl8hQJnzpxhzJgx1KlTh8KFC2NtbU2BAgVo0qQJK1euTFW+b9++xp4FoaGhJs/dp/WM+JEjR+jVqxdFixbFxsYGd3d3mjdvzoYNG7IUJ0CDBg3QaDQEBwezf/9+WrduTYECBXByciIoKIh///3XWHbTpk00btwYNzc3HB0dadq0KUePHk133+fOnaNfv374+fkZ42zcuHGadWCQnJzMzJkzqVChAra2tnh6etKpUydOnjz5zHO5cOEC77zzDsWLFze+f+rXr8/SpUuzVilCCCHyLelSLoQQwuxatmyJu7s79+/fZ+vWrVSrVs247tGjR7Rr147g4GAcHR2pVq0anp6enDx5knnz5rFq1Sq2bt1KlSpVTPa5fft2OnfuzMOHDylYsCCNGzfG2tqaq1evsnz5cgBq165tLH/o0CFatGjB/fv3KVq0KB06dCAyMpLg4GD27t3LmjVrWLduHdbW1qniv3v3LtWrV+fhw4fUq1ePatWqpSp38eJFqlSpgrW1NXXq1EFRFDw8PAB9kterVy9WrlyJjY0N1apVo3Dhwly4cIFly5bxxx9/8Mcff9CiRYtM1eeMGTP4+eefKV26NBUqVMDV1ZWwsDB27tzJ9u3b2b9/PzNmzDCWr1u3LtHR0fz+++84ODjQuXPndPc9a9Yshg0bhk6no3LlytSsWZPbt28THBzMli1bGDduHKNHj85UnE/6+++/jYlu06ZNOX/+PP/88w9NmzZlx44dhISEMGTIEF599VWaNWvGsWPH2LZtG0FBQYSEhBjHCHhyf507dyY+Pp5SpUrRsWNHIiIi2LVrFzt27GDz5s38/PPPJtvodDq6dOnC2rVrsba2pkGDBri5uXHgwAFq1KjBm2++mW78q1atonfv3sTHx1O6dGlatWpFZGQkBw4c4I033mDHjh0sWLAgy/UihBAin1GEEEKITPLz81MAZeHChc8s26RJEwVQXn/9dZPlPXv2VAClTZs2Snh4uMm6b775RgGUwMBAJTk52bg8LCxMcXFxUQDlk08+URISEky2Cw8PV/7991/j7/Hx8cZY3333XSUxMdG47vLly4q/v78CKJ9++qnJfhYuXKgACqA0btxYiYyMTHVeY8aMMZZ5/fXXlfj4+FRlPv30UwVQatasqfz3338m61atWqVYWFgobm5uyoMHD0zWGfb7tODgYOXy5cuplp87d04pUqSIAigHDhwwWXflyhUFUPz8/FJtZ7Bp0yZFo9EoHh4eyq5du0zWnThxwrjv4ODgdPfxtKCgIAVQNBqNsmTJEpN1w4YNUwClVKlSiqOjo7Jt2zbjuuTkZKVTp04KoPTv399ku9u3bxv//hMmTFB0Op1x3aFDhxQ3NzcFUH788UeT7WbPnq0ASqFChZQzZ84YlyclJSnvvfeesb779OmT6txtbGwUW1tb5ffffzdZd/XqVaVChQoKoCxevNhkXZ8+fTL9/0MIIUT+IAm3EEKITMtKwt29e3cFUFq2bGlcdubMGUWj0Sg+Pj5KVFRUmtu1atVKAZS//vrLuGzo0KEKoLRt2zZTcS5ZskQBFB8fnzQT4tWrVyuA4uTkpMTFxRmXGxJuKyurNBNcRfl/wu3u7q48fPgw1fp79+4pdnZ2iq2trXL9+vU09/H+++8rgPLdd9+ZLE8v4c7IDz/8oADKyJEjTZZnJuGuWbOmAiirV69Oc/3KlSsVQOnUqVOm4zEk3F26dEm17t69e8ZzfDpeRVGUI0eOKIASEBBgsvzLL79UAKVatWppHnPatGnGCzVPKlGihAIoc+fOTbVNXFyc4uXllWbC3a1bNwVQpk2blubxDh48mGY8knALIYR4mjzDLYQQIlvodDoAk2eGN2zYgKIotGzZEicnpzS3Mzwr/eQz2Zs2bQJgwIABmTp2cHAwAN27d8fGxibV+o4dO+Lm5sajR4/SfMa8SpUqz5zSrEmTJri4uKRavnPnTuLi4ozPW6clrXN8lujoaFatWsWnn37KgAED6Nu3L3379uX3338H4Pz585neF+i7zR88eBA7Ozvatm1rtjgNWrVqlWqZu7s7BQoUSHd9YGAgADdv3jRZbvh79unTJ81jvfXWW4C+m79h2xs3bnDp0iUAXn/99VTb2Nra0rVr11TLdTodGzduBKBbt25pHq969eo4OjoSEhJCfHx8mmWEEEIIkGe4hRBCZJO7d+8C+iTL4L///gPg559/TvW87dPu3LljfB0aGgpA6dKlM3XsGzduAKQ7JZlGoyEgIIAHDx4Yyz7J39//mcdIr4zhHLdv357mAGVPevIcM/LXX3/Rr18/7t27l26ZqKioTO3L4MqVKyiKQlxcXJoXJZ4nzicVLVo0zeWOjo7cu3cvzfWGizAJCQkmy5/193R1dTWOGXD9+nV8fHy4fv06AB4eHjg6Oqa5XVr7u3fvnrEufX1909zu6fLpXVgRQgghJOEWQghhdoqiEBISAkCFChWMyw13vStXrkylSpUy3EfNmjWzL8BnsLOze+4yhnMsUaIEderUyXAfmbmAcOPGDbp160ZcXBwfffQRvXr1wt/fH0dHR7RaLVu2bKF58+YoivLMfaUVp6OjI506dcrStpmh1Wbcie5Z69ViqBdI/476k551sUIIIUT+Jgm3EEIIs9uwYQMPHjwAoFmzZsblhjuGderUYfbs2ZneX9GiRTl//jznzp1LNXp1Wgx3HA13m9Ny5coVk7LmYjjHUqVKsWjRohfe319//UVcXByvvfYaU6ZMSbX+4sWLz7VfQ5wajYYFCxbk2AQY9H+jc+fOpfv3jIyM5P79+8ayT/68e/cu0dHRad7lvnr1aqplHh4e2NnZERcXx7Rp04wjzwshhBDPI+e2rkIIIXKlyMhIPvzwQwCaNm1K5cqVjetatmwJwLp167L07Kth+qyffvopU+UNzx6vWLEizeOsWbOGBw8e4OTkZDJlmTkYpisLDg4mIiLihfdnSCT9/PxSrVMUxTgl2tMM05glJyenud7Hx4eKFSvy6NEj4zPyOZXh77l48eI01xum5woMDDQm2kWKFDE+h59WHSUkJLBq1apUyy0sLGjatClAhvN7CyGEEJkhCbcQQgizUBSFjRs3UqNGDS5evIi3t3eqBLlKlSp06tSJa9eu0bFjxzTvMMbExLBs2TLCw8ONy4YNG4aTkxPr1q3j888/JykpyWSbiIgIdu/ebfy9S5cuFC1alJs3bzJs2DCTpPPKlSsMHz4cgMGDB2Nra2uO0zcqVKgQgwcPJiYmhrZt23Ly5MlUZRISEli3bh3nzp175v7KlCkDwOrVq7l165ZxeUpKCqNHj053QDNPT0+sra25ffu2MWl/2oQJEwDo168ff/31V6r1iqJw4MABtmzZ8sw4s9Pbb7+Ns7MzR48eZdKkSSbd50NCQoznMXLkSJPthg4dCsDYsWNN6jolJYURI0akGpzNYMyYMVhbWzNy5EgWL15s0s3c4NSpU/zxxx8vempCCCHyOOlSLoQQIsvmz59vHDk6ISGBu3fvcvToUWNi16BBAxYsWJDmXdmFCxfy8OFDNm7cSKlSpahUqRIBAQEoisLVq1c5fvw4iYmJnD17lkKFCgH6LuWrV6+mc+fOTJw4kfnz51OrVi2srKwIDQ0lJCSEnj17UrduXUD/XO3q1atp0aIFc+fOZcOGDbz66qs8evSIHTt2EB8fT/PmzRkzZky21M9XX33FrVu3WL58ufF59WLFimFpacn169c5duwYMTExbNy48ZnPcbdt25Zq1apx5MgRSpYsSVBQEA4ODhw4cICbN2/y8ccfp9nV3MrKinbt2rF69WoqV65M3bp1sbe3B/R/P8O+Z82axfDhw2nXrh0lSpSgVKlSuLi4cOfOHY4fP05ERAQff/yxyaMBL1uhQoVYtmwZXbp04bPPPmPJkiVUqVKFiIgIdu3aRXJyMv369ePtt9822W7gwIFs3bqVv/76i0qVKtGwYUPc3Nw4cOAAt27d4r333mPu3Lmpjle1alWWLl1qHAn+888/p2zZsnh6enL//n1OnjzJ9evX6datGx07dnxZ1SCEECI3Um1CMiGEELmOYR7uJ/85ODgoPj4+SlBQkDJ8+HDl4MGDz9xPSkqKsnz5cqVVq1ZKoUKFFCsrK6VAgQJK+fLllX79+ilr1qxREhMTU20XGhqqfPDBB0qpUqUUW1tbxdHRUSlZsqTy5ptvKvv27UtVPiwsTBk4cKBSrFgxxdraWnFyclJq1aqlzJ07V0lKSkpV3jAP99PzMj/JMA/3mDFjnnmeGzZsUDp27KgULlxYsbKyUlxdXZUyZcoo3bt3V5YvX67ExMSYlCedebgfPXqkfPrpp8bzLliwoNKhQwfl8OHDys6dOxVACQoKSrXdvXv3lHfeeUcpWrSoYmVlle7+T548qQwYMEAJDAxUbG1tFXt7e6VYsWJK8+bNlW+//Va5cePGM8/VwDAP986dO9Ncb3gPXblyJc316cWoKPp53Pv06aMUKVLEWJ8NGzZUfvvtt3TjSUpKUqZPn66ULVtWsbGxUQoUKKC0b99eOXbs2DP/3leuXFE+/PBDpXz58oqDg4Nia2ur+Pn5KQ0aNFC++uor5dKlSyblZR5uIYQQT9MoShaHNRVCCCGEEEIIIcQzyTPcQgghhBBCCCFENpCEWwghhBBCCCGEyAaScAshhBBCCCGEENlAEm4hhBBCCCGEECIbSMIthBBCCCGEEEJkA0m4hRBCCCGEEEKIbCAJtxBCCCGEEEIIkQ0k4RZCCCGEEEIIIbKBJNxCCCGEEEIIIUQ2kIRbCCGEEEIIIYTIBpJwCyGEEEIIIYQQ2UASbiGEEEIIIYQQIhtIwi2EEEIIIYQQQmQDSbiFEEIIIYQQQohsIAm3EEIIIYQQQgiRDSThFkIIIYQQQgghsoEk3EIIIYQQQgghRDaQhFsIIYQQQgghhMgGknALIYQQQgghhBDZQBJuIYQQQgghhBAiG0jCLYQQQgghhBBCZANJuPOhBg0a0KBBgyxtM3bsWDQaDXfv3s2eoESelpyczEcffYSvry9arZYOHTqoHZIqNBoNgwYNema5RYsWodFouHr16jPL+vv707dvX+PvwcHBaDQagoODn7nt1atX0Wg0LFq06JllhRAipzt06BC1a9fGwcEBjUbDsWPH1A7phTz9+Z6XZKWtelJm21EhchJJuPMAw5dzwz9LS0sKFy5M3759uXHjhtrhiUwyXNQoVKgQsbGxqdb7+/vTpk0bFSJ7cQsWLGDq1Kl07tyZxYsX8+GHHz5zmzVr1tCyZUs8PDywtrbGx8eHrl27smPHjpcQcd6yfPlyZs6cqXYYQog87saNG7Ru3RpnZ2fKli3LX3/9larMH3/8QcGCBYmMjDTrsZOSkujSpQv379/nm2++YcmSJfj5+Zn1GCJj/v7+Jt9HHRwcqFGjBr/88ovaoQGQmJjIrFmzqFKlCs7Ozri6ulKuXDkGDBjAuXPn1A5P5GGWagcgzGf8+PEEBAQQHx/P/v37WbRoEbt37+bUqVPY2toay23ZskXFKMWzREREMHfuXIYPH652KGazY8cOChcuzDfffPPMsoqi8Oabb7Jo0SKqVKnCsGHD8PLy4tatW6xZs4bGjRuzZ88eateu/RIiV8cbb7xB9+7dsbGxyfK29evXJy4uDmtra+Oy5cuXc+rUKYYOHWpS1s/Pj7i4OKysrF40ZCGEoE+fPty4cYMpU6awZ88eunTpwrlz5/D39wcgPj6eESNGMGHCBFxcXMx67MuXLxMaGspPP/1E//79zbpvkXmVK1c2fn+5desW8+fPp0+fPiQkJPD2228by6XVVmW3Tp06sXHjRnr06MHbb79NUlIS586dY/369dSuXZvSpUu/tFhE/iIJdx7SsmVLqlevDkD//v3x8PBgypQprFu3jq5duxrLvcwPt5wiJiYGBwcHtcPIlMqVKzN16lTef/997Ozs1A7HLCIiInB1dc1U2enTp7No0SKGDh3KjBkz0Gg0xnWfffYZS5YswdIy93x0KYpCfHx8lv6WFhYWWFhYPNfxtFqtyQW2jGg0mkyXFUKIjMTFxbFjxw6Cg4OpX78+7777Lnv37mXz5s288847AEybNg0XF5dsSYgjIiIAMt3WiOxRuHBhXn/9dePvffv2pVixYnzzzTcmCXdW2ipzOHToEOvXr2fixIl8+umnJutmz57Nw4cPX1os8fHxWFtbo9VKR+P8Qv7SeVi9evUA/VXfJ6X1DPd3331HuXLlsLe3x83NjerVq7N8+fIM9x8aGkqJEiUoX7484eHh6ZZ79OgRQ4cOxd/fHxsbGwoWLEjTpk05evSoSbkDBw7QqlUr3NzccHBwoGLFisyaNcukzI4dO6hXrx4ODg64urrSvn17zp49a1LG0DX7zJkz9OzZEzc3N+rWrWtcv3TpUqpVq4adnR3u7u50796da9euZXiuq1evRqPRsGvXrlTrfvjhBzQaDadOnQLg9u3b9OvXjyJFimBjY4O3tzft27fP1PO4AKNHjyY8PJy5c+dmWC6955/Sei63b9++ODo6EhYWRps2bXB0dKRw4cLMmTMHgJMnT9KoUSMcHBzw8/N75t/eICYmhuHDh+Pr64uNjQ2lSpVi2rRpKIpiEsvOnTs5ffq0sZtZes9sxcXFMXnyZEqXLs20adNMkm2DN954gxo1ahh//++//+jSpQvu7u7Y29vz6quv8vfff6dZVytXrmTcuHEULlwYJycnOnfuTGRkJAkJCQwdOpSCBQvi6OhIv379SEhIMNlHcnIyX375JcWLF8fGxgZ/f38+/fTTVOUMXf83b95M9erVsbOz44cffjAps2zZMkqVKoWtrS3VqlXjn3/+MVmf1jPciqIwYcIEihQpgr29PQ0bNuT06dOp6ufp90WDBg34+++/CQ0NNda/4W5Tes9wnzt3js6dO+Pu7o6trS3Vq1dn3bp1JmWSkpIYN24cgYGB2NraUqBAAerWrcvWrVtTxSSEyPvi4+NRFAU3NzdAf0HP1dXV+IjUjRs3+Oqrr5g1a1aWE41ntf19+/YlKCgIgC5duqDRaDIcq8bwGfvPP//wzjvvUKBAAZydnenduzcPHjxIVX7jxo3G4zs5OdG6des0P3+z8h3l3LlzdO3aFWdnZwoUKMAHH3xAfHz8M+vi4cOHDB061NjulihRgilTpqDT6TLcrk2bNhQrVizNdbVq1TLesAHYunUrdevWxdXVFUdHR0qVKpUqSc0sT09PSpcuneq7aFrfYS5evEinTp3w8vLC1taWIkWK0L1792c+fjBhwgS0Wi3fffddumUMx69Tp06qdRYWFhQoUMBk2Y0bN3jrrbfw8fHBxsaGgIAA3nvvPRITE41lsvL947fffuPzzz+ncOHC2NvbExUVBei/+7Zo0QIXFxfs7e0JCgpiz549GZ6vyH1yz20ikWWGL+uGxi89P/30E0OGDKFz587GD/wTJ05w4MABevbsmeY2ly9fplGjRri7u7N161Y8PDzS3f+7777L6tWrGTRoEGXLluXevXvs3r2bs2fPUrVqVUD/4d6mTRu8vb354IMP8PLy4uzZs6xfv54PPvgAgG3bttGyZUuKFSvG2LFjiYuL47vvvqNOnTocPXrUmEQYdOnShcDAQCZNmmRMACdOnMgXX3xB165d6d+/P3fu3OG7776jfv36hISEpHtlvHXr1jg6OrJy5Upjo26wYsUKypUrR/ny5QF9l6XTp08zePBg/P39iYiIYOvWrYSFhaWKMS316tWjUaNGfP3117z33ntmu8udkpJCy5YtqV+/Pl9//TXLli1j0KBBODg48Nlnn9GrVy86duzIvHnz6N27N7Vq1SIgICDd/SmKQrt27di5cydvvfUWlStXZvPmzYwcOZIbN27wzTff4OnpyZIlS5g4cSLR0dFMnjwZgDJlyqS5z927d3P//n2GDh2aqTu84eHh1K5dm9jYWIYMGUKBAgVYvHgx7dq1Y/Xq1bz22msm5SdPnoydnR2ffPIJly5d4rvvvsPKygqtVsuDBw8YO3as8XGMgIAARo8ebdy2f//+LF68mM6dOzN8+HAOHDjA5MmTOXv2LGvWrDE5zvnz5+nRowfvvPMOb7/9NqVKlTKu27VrFytWrGDIkCHY2Njw/fff06JFCw4ePGh8D6Vl9OjRTJgwgVatWtGqVSuOHj1Ks2bNTBr/tHz22WdERkZy/fp1Y5d+R0fHdMufPn2aOnXqULhwYT755BMcHBxYuXIlHTp04PfffzfW6dixY5k8eTL9+/enRo0aREVFcfjwYY4ePUrTpk0zjEkIkfe4ublRvHhxJk2axKRJk9i7dy/Hjh0zJkEfffSRsQ3Kisy0/e+88w6FCxdm0qRJDBkyhFdeeYVChQo9c9+DBg3C1dWVsWPHcv78eebOnUtoaKgxSQJYsmQJffr0oXnz5kyZMoXY2Fjmzp1L3bp1CQkJMbbrWf2O0rVrV/z9/Zk8eTL79+/n22+/5cGDBxk+7xwbG0tQUBA3btzgnXfeoWjRouzdu5dRo0Zx69atDMfq6NatG7179+bQoUO88sorxuWhoaHs37+fqVOnAvo2oE2bNlSsWJHx48djY2PDpUuXnjsJTE5O5vr168/8LpqYmEjz5s1JSEhg8ODBeHl5cePGDdavX8/Dhw/TfQTh888/Z9KkSfzwww8md9CfZnief9myZdSpUyfDnnI3b96kRo0aPHz4kAEDBlC6dGlu3LjB6tWriY2NxdraOsvfP7788kusra0ZMWIECQkJWFtbs2PHDlq2bEm1atUYM2YMWq2WhQsX0qhRI/7991+Tmwsil1NErrdw4UIFULZt26bcuXNHuXbtmrJ69WrF09NTsbGxUa5du2ZSPigoSAkKCjL+3r59e6VcuXIZHmPMmDEKoNy5c0c5e/as4uPjo7zyyivK/fv3nxmfi4uLMnDgwHTXJycnKwEBAYqfn5/y4MEDk3U6nc74unLlykrBggWVe/fuGZcdP35c0Wq1Su/evVPF2qNHD5N9Xb16VbGwsFAmTpxosvzkyZOKpaVlquVP69Gjh1KwYEElOTnZuOzWrVuKVqtVxo8fryiKojx48EABlKlTp2a4r7Q8Wce7du1SAGXGjBnG9X5+fkrr1q2Nv+/cuVMBlJ07d5rs58qVKwqgLFy40LisT58+CqBMmjTJuOzBgweKnZ2dotFolN9++824/Ny5cwqgjBkzJsN4165dqwDKhAkTTJZ37txZ0Wg0yqVLl4zLgoKCnvkeUxRFmTVrlgIoa9aseWZZRVGUoUOHKoDy77//Gpc9evRICQgIUPz9/ZWUlBRFUf5fV+XLl1cSExONZXv06KFoNBqlZcuWJvutVauW4ufnZ/z92LFjCqD079/fpNyIESMUQNmxY4dxmZ+fnwIomzZtShUvoADK4cOHjctCQ0MVW1tb5bXXXjMuM/yfvnLliqIoihIREaFYW1srrVu3Nvk/8emnnyqA0qdPH+OytN4XrVu3Njkfg7TeK40bN1YqVKigxMfHG5fpdDqldu3aSmBgoHFZpUqVTN6PQgixfft2xc3NzfhZN3ToUEVRFGXPnj2KnZ2dcvXq1SzvM7Ntv+Gzb9WqVc/cp+Eztlq1aiZtwtdff60Ayp9//qkoir49cXV1Vd5++22T7W/fvq24uLiYLM/qd5R27dqZ7PP9999XAOX48ePGZX5+fiaf719++aXi4OCgXLhwwWTbTz75RLGwsFDCwsLSPefIyEjFxsZGGT58uMnyr7/+WtFoNEpoaKiiKIryzTffGL+LZJWfn5/SrFkz5c6dO8qdO3eUkydPKm+88YYCpPoe+HRbFRISkqm/35P7Gj58uKLVapVFixY9MzadTqcEBQUpgFKoUCGlR48eypw5c4zn/aTevXsrWq1WOXToUJr7UZSsf/8oVqyYEhsba7KfwMBApXnz5ibtemxsrBIQEKA0bdr0meckcg/pUp6HNGnSBE9PT3x9fencuTMODg6sW7eOIkWKZLidq6sr169f59ChQ888xqlTpwgKCsLf359t27Y984qlYf8HDhzg5s2baa4PCQnhypUrDB06NNUdZsMV5lu3bnHs2DH69u2Lu7u7cX3FihVp2rQpGzZsSLXfd9991+T3P/74A51OR9euXbl7967xn5eXF4GBgezcuTPD8+jWrRsREREm3Z9Wr16NTqejW7duANjZ2WFtbU1wcHCa3dIyq379+jRs2JCvv/6auLi4597P0558bs7V1ZVSpUrh4OBg8ox/qVKlcHV15b///stwXxs2bMDCwoIhQ4aYLB8+fDiKorBx48Ysx2foYuXk5JSp8hs2bKBGjRomjww4OjoyYMAArl69ypkzZ0zK9+7d22SAsJo1axoHaXtSzZo1uXbtGsnJycbjAAwbNsyknGFgmKe7kAUEBNC8efM0Y65VqxbVqlUz/l60aFHat2/P5s2bSUlJSXObbdu2kZiYyODBg0262T89CNqLun//Pjt27KBr1648evTI+H/k3r17NG/enIsXLxpnPnB1deX06dNcvHjRrDEIIXKvRo0aERYWxv79+wkLC+Obb75Bp9MxZMgQhg8fjp+fH3PnzqV06dKUKlWKefPmZbi/52n7s2LAgAEmbcJ7772HpaWlcb9bt27l4cOH9OjRw+R7g4WFBTVr1jR+b3ieOAcOHGjy++DBgwEyPKdVq1ZRr1493NzcTOJp0qQJKSkpqR5PepKzszMtW7Zk5cqVxl5/oO+l9+qrr1K0aFHg/8/A//nnn8/spp6WLVu24OnpiaenJxUqVGDJkiX069fPeAc9PYY72Js3b05zppYnKYrCoEGDmDVrFkuXLqVPnz7PjEuj0bB582YmTJiAm5sbv/76KwMHDsTPz49u3boZn+HW6XSsXbuWtm3bmnSzf3I/kPXvH3369DHpsXjs2DEuXrxIz549uXfvnvFvGRMTQ+PGjfnnn3+eq/5FziQJdx4yZ84ctm7dyurVq2nVqhV3797N1CjHH3/8MY6OjtSoUYPAwEAGDhyYbtehtm3b4uTkxObNm3F2ds5UXF9//TWnTp3C19eXGjVqMHbsWJNkzvBcTUbdaUNDQwFMuuYalClTxvgh9aSnu0NfvHgRRVEIDAw0NgaGf2fPnjUOuJIewzM2K1asMC5bsWIFlStXpmTJkgDY2NgwZcoUNm7cSKFChYzdt2/fvp3hvtMyduxYbt++/cwvJJlla2uLp6enyTIXFxeKFCmS6llpFxeXZ14wCA0NxcfHJ1VybOgubvibZYXhPfXo0aNMlQ8NDU33PZFWDIYvFAaGBt7X1zfVcp1OZ3xuLDQ0FK1WS4kSJUzKeXl54erqmuo4GXXFDwwMTLWsZMmSxMbGcufOnTS3Mez/6W09PT0zddErsy5duoSiKHzxxRep/o+MGTMG+P/AROPHj+fhw4eULFmSChUqMHLkSE6cOGG2WIQQuZOjoyM1a9Y0fq4uXLiQ27dv88knn7Bt2zZGjhzJV199xddff83w4cMzvNj9PG1/Vjz9mero6Ii3t7fxkTzDBcVGjRql+kzcsmWL8fPweeJ8+tjFixdHq9VmON7LxYsX2bRpU6pYmjRpAvDM7zHdunXj2rVr7Nu3D9B//zpy5IjxpoGhTJ06dejfvz+FChWie/furFy5MtPJX82aNdm6dSubNm1i2rRpuLq68uDBg2cO2BsQEMCwYcOYP38+Hh4eNG/enDlz5qT5/PYvv/zCnDlz+O677+jRo0em4gL9d7TPPvuMs2fPcvPmTX799VdeffVVVq5caZzb+86dO0RFRWX4nRSy/v0jre+koE/En/57zp8/n4SEBLNPnSfUI89w5yE1atQwXo3r0KEDdevWpWfPnpw/fz7DZzbLlCnD+fPnWb9+PZs2beL333/n+++/Z/To0YwbN86kbKdOnVi8eDHLli0zjjr6LF27dqVevXqsWbOGLVu2MHXqVKZMmcIff/xBy5Ytn/+En+HpZ591Oh0ajYaNGzem+XxwRnUE+g/qDh06sGbNGr7//nvCw8PZs2cPkyZNMik3dOhQ2rZty9q1a9m8eTNffPEFkydPZseOHVSpUiXT8devX58GDRrw9ddfp7pbD6Q5oBiQ7l3S9J6JTm/5k1fAXxbDlBwnT56kQ4cOZt//i9ZBenX+tNw6urzhC9WIESPSvUNvuOhQv359Ll++zJ9//smWLVuYP38+33zzDfPmzZMpeYQQgL7X0meffca0adNwcHDg119/pXPnzsbP986dO7Ns2TIaNmyobqDpMHwmLlmyBC8vr1TrzTljRmbaF51OR9OmTfnoo4/SXG+4+J+etm3bYm9vz8qVK6lduzYrV65Eq9XSpUsXYxk7Ozv++ecfdu7cyd9//82mTZtYsWIFjRo1YsuWLc8cX8XDw8N4AaB58+aULl2aNm3aMGvWrFS9xJ42ffp0+vbta2xXhgwZYnzG/cnemnXq1OHYsWPMnj2brl27mvQqyCxvb2+6d+9Op06dKFeuHCtXrkw1gKg5pfWdFGDq1KlUrlw5zW2e9b1U5B5yhzuPsrCwYPLkydy8eZPZs2c/s7yDgwPdunVj4cKFhIWF0bp1ayZOnJhqxMypU6fy1ltv8f7772d6JGvQf7C9//77rF27litXrlCgQAEmTpwI6K/qAsZRvtNiGOzi/PnzqdadO3cODw+PZ077Vbx4cRRFISAggCZNmqT69+qrrz7zPLp168bdu3fZvn07q1atQlEUkyvDTx5r+PDhbNmyhVOnTpGYmMj06dOfuf+nGe5yPz3KNfx/MLynp7J4njvLz8PPz4+bN2+muht97tw54/qsqlu3rrGrV3oXDp6OIb33xPPGkN5xdDpdqu7T4eHhPHz4MEvHSasL9oULF7C3t0/VA+HJ46e17Z07dzL16EJmLxQYRrC1srJK8/9IkyZNTHo0uLu7069fP3799VeuXbtGxYoVGTt2bKaOJYTI+8aPH09AQAC9evUC9INR+fj4GNf7+PgYH1NJizna/ow8/ZkaHR3NrVu3jAOcGb6fFCxYMM3PQ8NI6M8T59PHvnTpEjqdLsPBVYsXL050dHS6n89P9+J6moODA23atGHVqlXodDpWrFhBvXr1TP4moJ+yq3HjxsyYMYMzZ84wceJEduzY8cxH79LSunVrgoKCmDRpUqZ6I1SoUIHPP/+cf/75h3///ZcbN26k6ulXokQJtmzZws2bN2nRokWme8WlxcrKiooVK5KUlMTdu3fx9PTE2dk5w++k8OLfPwzvLWdn53T/nk8+7iByN0m487AGDRpQo0YNZs6cmeFUE/fu3TP53dramrJly6IoCklJSSbrNBoNP/74I507d6ZPnz6ppgp6WkpKSqouMQULFsTHx8c4nVLVqlUJCAhg5syZqZJHwx1Gb29vKleuzOLFi03KnDp1ii1bttCqVasM4wDo2LEjFhYWjBs3LtWdS0VRUtVDWpo0aYK7uzsrVqxgxYoV1KhRw6SbUGxsbKq6Ll68OE5OTqmmj8qMoKAgGjRowJQpU1Lt18/PDwsLi1TPbH3//fdZPs7zaNWqFSkpKaku6HzzzTdoNJrn6r1gb2/Pxx9/zNmzZ/n444/TvMu+dOlSDh48aIzh4MGDxu5xoJ+q7Mcff8Tf35+yZctmOYa0GN5fT48AO2PGDED/hSKz9u3bZzIl3rVr1/jzzz9p1qxZuncODA3vd999Z1InGY1I+yQHB4dMdU0rWLAgDRo04IcffuDWrVup1j/Z5f3p/y+Ojo6UKFHiud7nQoi858KFC8yePZtZs2YZL/oVKlTImJAAnD17Ns07xwbmaPsz8uOPP5p8z5k7dy7JycnG9qt58+Y4OzszadKkVN+H4P+fic8Tp2FaTgPDaO4ZtZ1du3Zl3759bN68OdW6hw8fGscdyUi3bt24efMm8+fP5/jx46luGty/fz/VNoY7sM/7+f7xxx9z7949fvrpp3TLREVFpYq/QoUKaLXaNI9bsWJFNmzYwNmzZ2nbtu0zx7u5ePEiYWFhqZY/fPiQffv24ebmhqenJ1qtlg4dOvDXX39x+PDhVOUNbfCLfv+oVq0axYsXZ9q0aURHR6dan94jZiJ3ki7ledzIkSPp0qULixYtSrNbMkCzZs3w8vKiTp06FCpUiLNnzzJ79mxat26d5uBVWq2WpUuX0qFDB7p27cqGDRto1KhRmvt+9OgRRYoUoXPnzlSqVAlHR0e2bdvGoUOHjHd8tVotc+fOpW3btlSuXJl+/frh7e3NuXPnOH36tLFhmTp1Ki1btqRWrVq89dZbxik3XFxcMnVXrXjx4kyYMIFRo0Zx9epVOnTogJOTE1euXGHNmjUMGDCAESNGZLgPKysrOnbsyG+//UZMTAzTpk0zWX/hwgUaN25M165dKVu2LJaWlqxZs4bw8HC6d+/+zBjTMmbMmDS727m4uNClSxe+++47NBoNxYsXZ/369c98hstc2rZtS8OGDfnss8+4evUqlSpVYsuWLfz5558MHTrUePU2q0aOHMnp06eZPn06O3fupHPnznh5eXH79m3Wrl3LwYMH2bt3LwCffPIJv/76Ky1btmTIkCG4u7uzePFirly5wu+//57luV7TU6lSJfr06cOPP/7Iw4cPCQoK4uDBgyxevJgOHTpkqTtk+fLlad68ucm0YECqxzee5OnpyYgRI5g8eTJt2rShVatWhISEsHHjxgyn5DOoVq0aK1asYNiwYbzyyis4OjrStm3bNMvOmTOHunXrUqFCBd5++22KFStGeHg4+/bt4/r16xw/fhyAsmXL0qBBA6pVq4a7uzuHDx82Tv8nhBAffvgh3bp1M5naqHPnzrRv3944p/Nff/3F+vXrM9zPi7b9GUlMTDS22efPn+f777+nbt26tGvXDtDffZw7dy5vvPEGVatWpXv37nh6ehIWFsbff/9NnTp1jBedsxrnlStXaNeuHS1atGDfvn0sXbqUnj17UqlSpXTjHTlyJOvWraNNmzb07duXatWqERMTw8mTJ1m9ejVXr159ZpvQqlUrnJycGDFiBBYWFnTq1Mlk/fjx4/nnn39o3bo1fn5+RERE8P3331OkSBGTAcKyomXLlpQvX54ZM2YwcODANO/c7tixg0GDBtGlSxdKlixJcnIyS5YsSTNGg1dffZU///yTVq1a0blzZ9auXZvuXeHjx4/Ts2dPWrZsSb169XB3d+fGjRssXryYmzdvMnPmTONF70mTJrFlyxaCgoIYMGAAZcqU4datW6xatYrdu3fj6ur6wt8/tFot8+fPp2XLlpQrV45+/fpRuHBhbty4wc6dO3F2duavv/7KYk2LHOulj4suzM4wvUVa0xekpKQoxYsXV4oXL26czurpacF++OEHpX79+kqBAgUUGxsbpXjx4srIkSOVyMhIY5knp6wyiI2NVYKCghRHR0dl//79acaWkJCgjBw5UqlUqZLi5OSkODg4KJUqVVK+//77VGV3796tNG3a1FiuYsWKynfffWdSZtu2bUqdOnUUOzs7xdnZWWnbtq1y5swZkzJpxfqk33//Xalbt67i4OCgODg4KKVLl1YGDhyonD9/Ps3yT9u6dasCKBqNJtWUa3fv3lUGDhyolC5dWnFwcFBcXFyUmjVrKitXrnzmfjOK2zCVxdPTMN25c0fp1KmTYm9vr7i5uSnvvPOOcurUqTSnBXNwcEhzv2lN1/X0FGTpefTokfLhhx8qPj4+ipWVlRIYGKhMnTrVZIqLjI6TkdWrVyvNmjVT3N3dFUtLS8Xb21vp1q2bEhwcbFLu8uXLSufOnRVXV1fF1tZWqVGjhrJ+/XqTMulNF5Pe/520/hZJSUnKuHHjlICAAMXKykrx9fVVRo0aZTJ9lqJkXHc8ns5k6dKlSmBgoGJjY6NUqVIl1dRuT08Lpij6/8vjxo1TvL29FTs7O6VBgwbKqVOnUk0bk9a0YNHR0UrPnj0VV1dXBTBOEZbWtGCGOu3du7fi5eWlWFlZKYULF1batGmjrF692lhmwoQJSo0aNRRXV1fFzs5OKV26tDJx4kSTKXaEEPnT33//rTg6Oio3b95MtW7y5MmKj4+P4u3trUyZMiVT+8tM2/8804Lt2rVLGTBggOLm5qY4OjoqvXr1MpnW68l9N2/eXHFxcVFsbW2V4sWLK3379jWZ4jGzcRralzNnziidO3dWnJycFDc3N2XQoEFKXFycSdmnP98VRd/ujho1SilRooRibW2teHh4KLVr11amTZuW6c/fXr16KYDSpEmTVOu2b9+utG/fXvHx8VGsra0VHx8fpUePHqmmIktLRu3fokWLTNqbp9uq//77T3nzzTeV4sWLK7a2toq7u7vSsGFDZdu2bSb7gdRTjP3555+KpaWl0q1bN+N0XE8LDw9XvvrqKyUoKEjx9vZWLC0tFTc3N6VRo0YmbZtBaGio0rt3b+MUu8WKFVMGDhyoJCQkGMu8yPcPg5CQEKVjx47G7+B+fn5K165dle3bt6dZXuROGkVRYWQkIYQQQgghVLBo0SL69evHoUOH0pz6KTuNHTuWcePGcefOnUz1UBJC5H7yDLcQQgghhBBCCJENJOEWQgghhBBCCCGygSTcQgghhBBCCCFENpBnuIUQQgghhBBCiGwgd7iFEEIIIYQQQohsIAm3EEIIIYQQQgiRDSzVDuBF6HQ6bt68iZOTExqNRu1whBBCCBRF4dGjR/j4+KDVynXtFyVtvRBCiJwmK219rk64b968ia+vr9phCCGEEKlcu3aNIkWKqB1GridtvRBCiJwqM219rk64nZycAP2JOjs7v9C+kpKS2LJlC82aNcPKysoc4eV5UmdZJ3WWdVJnWSd1lnXmrLOoqCh8fX2NbZR4MeZs63MT+X+cPaRezU/qNHtIvZqfWm19rk64DV3LnJ2dzZJw29vb4+zsLG/qTJI6yzqps6yTOss6qbOsy446k+7P5mHOtj43kf/H2UPq1fykTrOH1Kv5qdXWy8NlQgghhBBCCCFENpCEWwghhBBCCCGEyAaScAshhBBCCCGEENkgVz/DLYQQ5qQoCsnJyaSkpKgdygtJSkrC0tKS+Pj4XH8uL0tW6szCwgJLS0t5RlsIIXKh3NLWS1tufmq19ZJwCyEEkJiYyK1bt4iNjVU7lBemKApeXl5cu3ZNksJMymqd2dvb4+3tjbW19UuITgghhDnkprZe2nLzU6utl4RbCJHv6XQ6rly5goWFBT4+PlhbW+fqxk2n0xEdHY2joyNarTw5lBmZrTNFUUhMTOTOnTtcuXKFwMBAqWMhhMgFcltbL225+anV1kvCLYTI9xITE9HpdPj6+mJvb692OC9Mp9ORmJiIra2tNNKZlJU6s7Ozw8rKitDQUOM2Qgghcrbc1tZLW25+arX18tcTQojHpEETmSXvFSGEyJ3k81tklrneK/KOE0IIIYQQQgghsoEk3EIIkQ9pNBrWrl2b7vrg4GA0Gg0PHz58aTG9THn9/IQQQghp63PG+amacD969IihQ4fi5+eHnZ0dtWvX5tChQ2qGJIQQuca8efNwcnIiOTnZuCw6OhobGxvatGljUtbQ6Fy+fDlT+65duza3bt3CxcUFgEWLFuHq6mqWuP39/dFoNGg0GhwcHKhatSqrVq0yy76FEEKIvCS9tt7KyooGDRqYlJW2PmdSNeHu378/W7duZcmSJZw8eZJmzZrRpEkTbty4oWZYQgiRKzRs2JDo6GgOHz5sXPbvv//i5eXFkSNHiI+PNy7fuXMnRYsWpXjx4pnat7W1NV5eXtk2guv48eO5desWISEhvPLKK3Tr1o29e/dmy7GEEEKI3Cqjtv7AgQPS1ucCqiXccXFx/P7773z99dfUr1+fEiVKMHbsWEqUKMHcuXPVCksIIXKNUqVK4e3tTXBwsHFZcHAw7dq1w8/Pj/3795ssb9iwocn2d+/e5bXXXsPe3p7AwEDWrVtnUt7QDSs4OJh+/foRGRlpvFo9duxYABISEhgxYgSFCxfGwcGBmjVrmsSTHicnJ7y8vChZsiRz5szBzs6Ov/76C4Br167RtWtXXF1dcXd3p3379ly9etW4rU6nY/z48RQpUgQbGxsqV67Mpk2bjOuvXr2KRqPht99+o3bt2tja2lK+fHl27dqVYUz79u0jKCgIOzs7fH19GTJkCDExMc88FyGEECK7pNfWt2/fnoCAAGnrc0Fbr1rCnZycTEpKSqoh1u3s7Ni9e7dKUQkhhJ6iKMQmJr/0f4qiZCnOhg0bsnPnTuPvO3fuJCgoiNq1axsbw7i4OA4cOJCqER43bhxdu3blxIkTtGrVil69enH//v1Ux6hduzYzZ87E2dmZW7ducevWLUaMGAHAoEGD2LdvH7/99hsnTpygS5cutGjRgosXL2b6HCwtLbGysiIxMZGkpCSaN2+Ok5MT//77L3v27MHR0ZEWLVqQmJgIwKxZs5g+fTrTpk3jxIkTNG/enHbt2qU65siRIxk+fDghISHUqlWLtm3bcu/evTRjuHz5Ml26dKFjx46cOHGCFStWsHv3bgYNGpTp8xBCCJG7qNXWZ7W9T6utb9CgAUFBQcbl0tbn3LZetXm4nZycqFWrFl9++SVlypShUKFC/Prrr+zbt48SJUqkuU1CQgIJCQnG36OiogBISkoiKSnpheIxbP+i+8lPpM6yTuos615GnSUlJaEoCjqdDp1OB0BsYjLlx27NtmOm59TYpthbZ/6jOSgoiGHDhpGYmEhcXBwhISHUr1+fqKgofvnlF3Q6HXv27CEhIYGgoCDj+QH06dOHbt26ATBhwgS+/fZb9u/fT4sWLYzldDodlpaWODk5odFoKFiwoHH7q1evsnDhQq5evYqPjw8Aw4YNY9OmTSxYsICJEyemG7ehvhMTE5kxYwaRkZE0aNCAX3/9FZ1Ox48//mjs4vbzzz/j7u7Ojh07aNasGdOmTeOjjz6ia9euAEyePJmdO3fyzTffMHv2bGPsAwcO5LXXXgNgzpw5bNq0ifnz5zNy5EiT89PpdHz11Vd07tyZIUOGoNFoKF68ODNnzqRhw4bMmTMn1cVhnU6HoigkJSVhYWFhsk7+fwshRO4Ql5RC2dGbVTn2mfHNM93eN2zYkKFDh5KcnGxs64OCgkhKSmLevHmA/s5tQkJCqoS7b9++9OjRA4BJkybx7bffcvDgQVq0aGFSztraGhcXFzQaDV5eXsblYWFhLFy4kLCwMGNbP2LECDZt2sTChQuZNGnSM+NPTExk+vTpREZG0qhRI1asWIFOp2P+/PnGtn7hwoW4uroSHBxsbOs//vhjunfvDsCUKVPYuXMnM2fOZM6cOcZ9Dxo0iE6dOgEwd+5cNm3axM8//8xHH32UKg5DW//BBx+g1WoJDAzk22+/JSgoiLlz577QXNsZUS3hBliyZAlvvvkmhQsXxsLCgqpVq9KjRw+OHDmSZvnJkyczbty4VMu3bNlitgnst259+V+wczups6yTOsu67KwzS0tLvLy8iI6ONl5ZjUtMybbjZeRR1COSrS2eXfCx6tWrExMTQ3BwMA8fPqREiRLY2tpSp04dBg0aREREBFu2bMHf3x9XV1fjhUqAEiVKmPzu5OREWFgYUVFRxMbG6uN59AitVkt8fDyKopiUP3DgACkpKZQuXdokpoSEBJydnU3KPkmn0/HJJ5/wxRdfEB8fj4ODA2PGjKFevXp88cUXXLp0yTiAi0F8fDynT5+mbNmy3Lx5k8qVK5vsv3r16pw6dYqoqCiio6MBqFChgkmZSpUqceLEiTTPLyQkhNOnT7N69WpjecNFgZMnT1KqVCmTeAwXOP755x+TgWwA476FEEIIc2jQoAExMTEcOnSIBw8eULJkSTw9PQkKCqJfv37Ex8cTHBxMsWLFKFq0qMm2FStWNL52cHDA2dmZiIiITB/75MmTpKSkULJkSZPlCQkJFChQIMNtP/74Yz7//HPi4+NxdHTkq6++onXr1owcOZJLly7h5ORkUj4+Pp7Lly8TFRXFzZs3qVOnjsn6OnXqcPz4cZNltWrVMr62tLSkevXqnD17Ns14Tpw4wYkTJ9Js669cuUKZMmUyPJ/npWrCXbx4cXbt2kVMTAxRUVF4e3vTrVs3ihUrlmb5UaNGMWzYMOPvUVFR+Pr60qxZM5ydnV8olqSkJLZu3UrTpk2xsrJ6oX3lF/m6zuIewsNQNHH3IfYemrgHkJIAFtagtUKxsAZ7dxTnIuDiC7YuoNHk7zp7Ti+jzuLj47l27RqOjo7Gq5tOisKpsU2z5XgZsbOyyNLgJZUrV6ZIkSLGRrhBgwbGBszX15dTp06xb98+GjdunOpz0tnZ2WSZVqvF2toaZ2dn40VMJycnnJ2dsbW1RaPRmJTX6XRYWFhw6NChVHd5HR0d0/1c1mq1jBgxgj59+uDo6EihQoWM55yUlES1atVYsmRJqu08PT2N5ezt7U32b21tjaWlJc7Ozjg6OgL//2JhYOjOltb5xcXF0bdvX4YNG5aq/osWLYq1tbXJsvj4eOzs7Khfv36qK+LpXWgQOcDtk3B5J7j5Q9l2akcjhFCZnZUFZ8Y3V+3YmVWiRAmKFCnCzp07efDgAUFBQQD4+Pjg6+vL3r172blzJ40aNUq17dPfnTQajUlvt2eJjo7GwsKCI0eOpNnWZ2TkyJH07ds3VVsfHR1NtWrVWLZsWaptPD09Mx1bVkVHR9O3b1+GDx+OVmv6ZPXTFyrMSdWE28DBwQEHBwcePHjA5s2b+frrr9MsZ2Njg42NTarlVlZWZvsibs595Rd5vs4SY+HqbgjbB+GnIfwURGVxJH1rJyhYBq13FQrf12IVXQYrz0DIplEh86LsfJ+lpKSg0WjQarUmH8COFplvDNXUsGFDdu3axYMHDxg5cqSxQatXrx6bN2/m4MGDvPfee6kal6fP98llhuWG17a2tqSkpJiUr1atGikpKdy9e5d69eplKWZPT89UV8sN+1y5ciVeXl7pJuw+Pj7s27fPpNvc3r17qVGjhknsBw8eNE6ZkpyczNGjRxk0aFCa51elShXOnz9PYGBgqjpJi1arRaPRpPm+zNOfh7ld2H7Y+gWUaSsJtxACjUaTpce41NSwYUOCg4ONbb1B/fr12bhxo7GtfxHW1takpJj28KtSpQopKSlERERkua338PBI81HhqlWrsmLFCgoWLJhhW79nzx7jxQWAPXv2UKNGDZNy+/fvp379+oC+rT9y5Ei6z2Qb2voSJUpkqq03F1XfYZs3b0ZRFEqVKsWlS5cYOXIkpUuXpl+/fmqGJQQ8Coez6+DCZrj6LyTHpy7jWAjsPcDeHewLgKUNpCRBSqL+X8wdeHgNYu9C4iO4fhCL6wepDvD9XHD1g1It9f+K1gZL69THECITGjZsyMCBA0lKSjJpmOrXr8+QIUNITExM9UxXVvn7+xMdHc327dupVKkS9vb2lCxZkl69etG7d2+mT59OlSpVuHPnDtu3b6dixYq0bt06y8fp1asXU6dOpX379sbRSUNDQ/njjz/46KOPKFKkCCNHjmTMmDEUL16cypUrs3DhQo4dO5bqSvmcOXMIDAykTJkyfPPNNzx48IA333wzzeN+9NFH1K5dm8GDB/P222/j4ODAmTNn2Lp1K7Nnz36uOhM5kOPjMQiiM9+dUgghcoL02vqgoCAGDRokbX0ObutVTbgjIyMZNWoU169fx93dnU6dOjFx4kS5OyDUodPBfzvgyCI4vxF0TzyX6VwEijcA78pQqDwUKqvvJp4ZibHwMAxuHSfl2iGizmzHNf4amoehcGCe/p+NC5TrAFX7QOGqcudbZEnDhg2Ji4ujdOnSFCpUyNhVLCgoiEePHhmnFHkRtWvX5t1336Vbt27cu3ePMWPGMHbsWBYuXMiECRMYPnw4N27cwMPDg1dffZU2bdo813Hs7e35559/+Pjjj+nYsSOPHj2icOHCJl3ihwwZQmRkJMOHDyciIoKyZcuybt06AgMDTfb11Vdf8dVXX3Hs2DFKlCjBunXr8PDwSPO4FStWZP369UyePJl69eqhKArFixc3Dion8ggHSbiFELnT0229gbT1Ob+t1yhZnYMmB4mKisLFxYXIyEizPMO9YcMGWrVqJQl/JuWZOktOgKO/wN7v4GHo/5cXeQVKt4HAZlCwjFmSYGOdNQnCKmw3XNgI5zfp74IbFCwHVXtDlV5g45T+zvKJl/E+i4+P58qVKwQEBGTbCJUvk06nIyoqCmdn55faZSqnuHr1KgEBAYSEhFC5cuVMbZPVOsvoPWPOtkmYuT7vXYbvqoKVA3x20zwBZpM808bmMFKv5pdb6jS3tfX5vS1/ltzU1ueOhxaEyA4pSXD8V9g1FSLD9MtsXKBSd6jWV38XO7tYO0CZNvp/uhQI3QshS+DMnxBxGjZ9DMGToMYAqPkuOKR9pU4IIUQWGLqUJ8VAQjTYZDzgjxBCCPGiJOEW+dOFzbDpE7j/n/53Ry+oPwIq9wJr80wxl2laCwiop//XcgqcXA0HfoB7F+GfqbB3tv4CQL3h4Jh9IzcKIUSeZ+0IVvaQFAsxEZJwCyGEyHbSP0HkL4/CYVVfWN5Vn2zbe0DzSfDBMajx9stPtp9m56aPY+AB6PqL/pnx5Dg4MBe+raxPwBNljl8hMuLv74+iKJnuYpafzJkzB39/f2xtbalZsyYHDx7MsPyqVasoXbo0tra2VKhQgQ0bNpisVxSF0aNH4+3tjZ2dHU2aNOHixYvG9cHBwWg0mjT/HTp0KFvOMUMaDTg8vnAZfeflH18IIYRZ5Ka2XhJukT8oin4wtDmvwOk1oLGA2oPhg+NQayBY2akdoSmtBZRtDwOC4Y014FMFEqNhxwT4rhqELNMP8iaEEJm0YsUKhg0bxpgxYzh69CiVKlWiefPmRESkPYDY3r176dGjB2+99RYhISF06NCBDh06cOrUKWOZr7/+mm+//ZZ58+Zx4MABHBwcaN68OfHx+pkdateuza1bt0z+9e/fn4CAAKpXr/5SzjsVQ7fyGBk4TQghRPaThFvkfXEP4Lde8NcHEB+pv2s8YCc0m5DzuxNqNFC8EfTfAZ1+Bpei8Ogm/Pk+LGoFEefUjlAIkUvMmDGDt99+m379+lG2bFnmzZuHvb09CxYsSLP8rFmzaNGiBSNHjqRMmTJ8+eWXVK1a1Th1iqIozJw5k88//5z27dtTsWJFfvnlF27evMnatWsB/ZyuXl5exn8FChTgzz//pF+/fsb54l86x8ej+0aHq3N8IYQQ+Yo8wy3ythtH9F3IH4aBhTU0Hg013wOLXPbW12qhQmf9qOkHf4DgKRC2D+bVhbpDod4IsMr5I24KIdSRmJjIkSNHGDVqlHGZVqulSZMm7Nu3L81t9u3bx7Bhw0yWNW/e3JhMX7lyhdu3b9OkSRPjehcXF2rWrMm+ffvo3r17qn2uW7eOe/fu0a9fv3RjTUhIICEhwfh7VFQUoB8JOSkp6dkn+wxauwJYAClRt9GZYX/ZxXCu5jhn8X9Sr+aXW+o0KSkJRVHQ6XTG6TNzMsNEUoaYxYvLap3qdDoURSEpKQkLCwuTdVl5v+eyrEOITFIUOPgTbP4UdEng6gddF+u7ZudmVrZQ5wMo9xpsGAkXNumf6z69Bjr+CIWrqR2hECIHunv3LikpKSZztwIUKlSIc+fS7ilz+/btNMvfvn3buN6wLL0yT/v5559p3rw5RYoUSTfWyZMnM27cuFTLt2zZgr39i4+zUepWJKWBsDOHORG94Znl1bZ161a1Q8iTpF7NL6fXqaWlJV5eXkRHR5OYmKh2OJn26NEjtUPIczJbp4mJicTFxfHPP/+QnJxssi42NvNjKknCLfKelGTY+BEc/ln/e+k20H4O2LmqGpZZuRaFHr/B2XWw4SO4dwnmN4UGn0DdYbnvDr4QIs+7fv06mzdvZuXKlRmWGzVqlMmd9aioKHx9fWnWrJlZ5jXXHrkNm9biV8COIq1avfD+sktSUhJbt26ladOmOXpu49xG6tX8ckudxsfHc+3aNRwdHXPFPNyKovDo0SOcnJzUewQnj8lqncbHx2NnZ0f9+vXTnIc7s+RbuchbEmNg9Zv6O79ooNmXUGuQ/lnovEaj0Q+s5l8P/h6mv8u9cyJc3AodfwD3YmpHKITIITw8PLCwsCA83PS55fDwcLy8vNLcxsvLK8Pyhp/h4eF4e3ublElr1NiFCxdSoEAB2rVrl2GsNjY22NjYpFpuZWVlni/zLj4AaGPvoM3ByYGB2c5bmJB6Nb+cXqcpKSloNBq0Wi1abc4fxsrQ5dkQs3hxWa1TrVaLRqNJ872dlfe6/PVE3hEdAYta65NtS1v9tFq1B+fNZPtJ9u7QeSG89iPYOMP1gzCvPpxZp3ZkIoe4evUqGo2GY8eOmW2fGo3G+CxvbtKgQQOGDh2qdhgvnbW1NdWqVWP79u3GZTqdju3bt1OrVq00t6lVq5ZJedB3GTWUDwgIwMvLy6RMVFQUBw4cSLVPRVFYuHAhvXv3Vv8LucPjUcqjZZRyIUTeIW39/+W0tl4SbpE3PLwGPzeFmyFg5w59/oKyGd9FyVM0GqjUDd7bA0VrQeIjWPkGbP4MUnL2ICbixfTt29dkbuMCBQrQsmVLk6mbcoJFixYZY9RqtRQpUoR+/fqlOyWVML9hw4bx008/sXjxYs6ePct7771HTEyMcQCz3r17mwyq9sEHH7Bp0yamT5/OuXPnGDt2LIcPH2bQoEGA/ovY0KFDmTBhAuvWrePkyZP07t0bHx8fOnToYHLsHTt2cOXKFfr37//Szjddjo/n4Y65ox/vQwghcri02voWLVpw4sQJtUMzIW192iThFrlf5HX9ne0HV8HNH/pvA98aakelDtei+osNtQfrf983Gxa3hahb6sYlslWLFi2Mcxxv374dS0vLNEeIVpuzszO3bt3i+vXr/PTTT2zcuJE33nhD7bDyjW7dujFt2jRGjx5N5cqVOXbsGJs2bTIOehYWFsatW///rKhduzbLly/nxx9/pFKlSqxevZq1a9dSvnx5Y5mPPvqIwYMHM2DAAF555RWio6PZtGlTqmfdfv75Z2rXrk3p0qVfzslmxHCHOykWEqPVjUUIITIprba+TZs2aoeVirT1qUnCLXI3Q7L9MBTcAqDvBihQXO2o1GVhpZ9jvNtSfRfzsH3wYwP9FGkiT7KxsTHOc1y5cmU+/vhjbty4wZ07d9Isn5KSwltvvUVAQAB2dnaUKlWKWbNmpSq3YMECypUrh42NDd7e3sY7m2kZM2YM3t7eGV5t12g0eHl54ePjQ8uWLRkyZAjbtm0jLi4OgPnz51OmTBlsbW0pXbo033//vcn2J0+epFGjRtjZ2VGgQAEGDBhAdPT/E6a+ffvSoUMHxo0bh6enJ87Ozrz77rsZjkabkJDAiBEj8PX1pXDhwtSqVYvg4OB0y+d2gwYNIjQ0lISEBA4cOEDNmjWN64KDg1m0aJFJ+S5dunD+/HkSEhI4deoUrZ4aZEyj0TB+/Hhu375NfHw827Zto2TJkqmOu3z5cvbs2ZMt55RlNo5g5aB/Ld3KhRC5xNNt/SeffMK1a9ekrc8Fbb0MmiZyr8gbsKjN/+9s910PLoXVjirnKNMWCpaFFa9DxBlY2Ao6fA/lO6kdWe6gKPo7YC+blf0LjTsQHR3NsmXLKFasGAUKFEizjE6no0iRIqxatYoCBQqwd+9eBgwYgLe3N127dgVg7ty5DBs2jK+++oqWLVsSGRmZZsKkKApDhgxh/fr1/Pvvv5QoUSLTsdrZ2aHT6UhOTmbZsmWMHj2a2bNnU6VKFUJCQnj77bdxcHCgT58+xMTE0Lx5c2rVqsWhQ4eIiIigf//+DBo0yCRJ3L59O7a2tgQHB3P16lX69etHgQIFmDhxYpoxDBo0iDNnzrB8+XKcnZ3Ztm0bLVq04OTJkwQGBmb6XEQu41gQHlzRJ9z5/SKtEPmZWm09vFB7Hx0dzdKlSylRooS09bmgrZeEW+ROMffgl3b6L0xu/tD3b3BJf17XfKtAcXhzM/zeHy5u1o/gfueCfvqwvD6Y3ItKioVJPi//uJ/eBGuHLG2yfv16HB0dAYiJicHb25tff/013RE4raysTOY5DggIYN++faxcudLYCE+YMIHhw4fzwQcfGMu98sorJvtJTk7m9ddfJyQkhN27d1O4cOYveF28eJF58+ZRvXp1nJycGDNmDNOnT6djx47GmM6cOcMPP/xAnz59WL58OfHx8fzyyy84OOjrZ/bs2bRt25YpU6YYu0VbW1uzYMEC7O3tKVeuHOPHj2fkyJF8+eWXqeojLCyMhQsXEhYWhpeXF1FRUQwfPpzNmzezcOFCJk2alOnzEbmMIeGOkTvcQuRrarX1kOX2Pq22fv369dLW54K2XhJukfskxcGv3fVzT7v4Qp/1kmxnxNYZevwKW0frn+ne9ZW+7jp8D5app94RuU/Dhg2ZO3cuAA8ePGDOnDl06dKFAwcOEBAQkOY2c+bMYcGCBYSFhREXF0diYqJxKqeIiAhu3rxJ48aNMzzuhx9+iI2NDfv378fDw+OZcUZGRuLo6IhOpyM+Pp66desyf/58YmJiuHz5Mm+99RZvv/22sXxycjIuLi4AnD17lkqVKhkbYIA6deqg0+k4f/68sRGuVKkS9vb2xjK1atUiOjqaa9eu4efnZxLPyZMnSUlJSdUFOiEhId07BiKPcJSRyoUQucvTbf33339Py5YtOXjwYKr2zUDa+pzR1kvCLXIXXYr+bu31g2DrCq//Dq6+akeV82ktoPlE8CwN64fCqdX6EXq7LdUn5CI1K3v91Wc1jptFDg4OJt27fvrpJ9zc3Jg/f36a3at+++03RowYwfTp06lVqxZOTk5MnTqVAwcOAPruX5nRtGlTfv31VzZv3kyvXr2eWd7JyYmjR4+i1Wrx9vY2Hscw1/NPP/1k8kwxgIWFRaZieR7R0dFYWFhw5MgRNBoN0dHRODo6otVqjXcRRB4lU4MJIUC9tt5w7Cx4uq2fP38+Li4u/PTTT0yYMCFVeWnr9XJCWy8Jt8g9FAU2fwrn1oOFtf6urWcptaPKXaq+Ac4+sLI3XNkFi1pBr9/BqZDakeU8Gk2Wu3bnFIbpOAwDlDxtz5491K5dm/fff9+47PLly8bXTk5O+Pv7s337dho2bJjucdq1a0fbtm3p2bMnFhYWzxwZXavVpvncV6FChfDx8eG///5LtzEvU6YMixYtIiYmxnjle8+ePWi1WkqV+v/nwPHjx4mLizM28Pv378fR0RFf39QX5qpUqUJKSgoRERHUqVOHqKgonJ2d0+2eJ/IQwx1u6VIuRP4mbb209S+BfKsQucf+7+HAPP3r1+aBX21148mtSjTWDzDn4Am3T8LPTeDe5WdvJ3KshIQEbt++ze3btzl79ixDhgwhOjo63elCAgMDOXz4MJs3b+bChQt88cUXHDp0yKTM2LFjmT59Ot9++y0XL17k6NGjfPfdd6n29dprr7FkyRL69evH6tWrn/scxo0bx+TJk/n222+5cOECJ0+eZOHChcyYMQOAXr16YWtrS58+fTh16hQ7d+5k8ODBvPHGG8YuZgCJiYm89dZbnDlzhg0bNjBmzBgGDRqUZsNasmRJevXqRe/evfnjjz8IDQ3l4MGDTJ48mb///vu5z0XkAtKlXAiRyzzd1g8ePJjo6Gjatm2bZnlp6/VyQlsvd7hF7vBfMGz5XP+66Zcy0vaL8qkCb22BJR31AwctaAG9/4RCZdWOTDyHTZs24e3tDeivWJcuXZpFixbRoEGDNMu/8847hISE0K1bNzQaDT169OD9999n48aNxjJ9+vQhPj6eb775hhEjRuDh4UHnzp3T3F/nzp3R6XS88cYbaLVa42AoWdG/f3/s7e2ZOnUqI0eOxMHBgQoVKjB06FAA7O3t2bx5Mx988AGvvPIK9vb2dOrUydhIGzRu3JjAwEDq169PQkICPXr0YOzYseked+HChUyYMIGRI0dy48YNPDw8ePXVV3Pk3KbCjKRLuRAil0mrrV+1apW09bmgrdcoiqK8lCNlg6ioKFxcXIiMjMTZ+cWeQ01KSmLDhg20atUKKysrM0WYt720Ont4DX4Mgth7UPl1aD87146wnePeZ9ER+qQ7/CTYucEba/TJeA7yMuosPj6eK1euEBAQgK2tbbYc42XS6XT5snt03759efjwIWvXrs3ytlmts4zeM+Zsm0Q21ee1Q/rePS5F4cOT5tmnmeW49iKPkHo1v9xSp7mtrc+vbfmz5Ma2Xv56ImdLioeVb+iTbe9K0Hpark22cyTHgtBnHRSuBnEPYHE7CNuvdlRCCJG9HD31P2Mi9OODCCGEENlEEm6Rs20cCTdD9Hdfuy4Bq8yNqCiywN4d3lgLRWtDQhQseQ2u/Kt2VEIIkX0MXcqT4/Wfe0IIIUQ2kWe4Rc519Bf9PzTQ6WdwS3uOQWEGts76KdZW9ILLO2B5V+i1Cvzrqh2ZEJm2aNEitUMQuYW1PVg7QeIjiL4Dti5qRySEECITcmNbL3e4Rc505wJs+Ej/utHn+pG1Rfaytofuv0LxxpAUC8u6QuhetaMSQojs8WS3ciGEECKbSMItcp7kRPijPyTHQbEGUHeY2hHlH1a20H05FGsISTGwrIs80y2EyJscH08xEx2ubhxCCCHyNEm4Rc4TPAluHdc/t91hHsjIjC+XlS30+BUCgiAxGpZ2hutH1I7qpcjFkzaIl0zeK3mAw+M73NF31I1DCPFSyee3yCxzvVckkxE5y9XdsHum/nXbb8HZW9Vw8i0rO+jxG/jX0z/juLQjhJ9WO6psY5jGJDY2VuVIRG5heK/k5ClwxDPIHW4h8hVp60VWmautl0HTRM4R9wD+eAdQoMrrULad2hHlb9b2+qR7SQe4fkg/enm/jVCguNqRmZ2FhQWurq5EROif5bS3t0eTi6ef0+l0JCYmEh8fL3N3ZlJm60xRFGJjY4mIiMDV1RULC4uXGKUwK8fHI5XLM9xC5Au5ra2Xttz81GrrJeEWOcfGTyDqOrgFQIspakcjAGwc9aOVL2oD4afglw7w5iZwKax2ZGbn5eUFYGyIczNFUYiLi8POzi5Hf5nISbJaZ66ursb3jMilDAl3dO7/Py+EyJzc1NZLW25+arX1knCLnOHiNjjxG6CBjj/qEz2RM9i5wRtrYEELuH9Zf8e73yZwKKB2ZGal0Wjw9vamYMGCJCUlqR3OC0lKSuKff/6hfv360uU5k7JSZ1ZWVnJnOy9wkIRbiPwmN7X10pabn1ptvSTcQn0J0bD+Q/3rV98D3xrqxiNScywIvf/UJ913L8DyLtDnL7B2UDsys7OwsMj1yZSFhQXJycnY2tpKI51JUmf5kLFLuQyaJkR+kxvaemmXzE+tOpUHAoT6dkyAyDBwKQoNP1M7GpEeV1944w/9He8bR2BVX0jJ2VeHhRAiXcYu5eEgoxYLIYTIJpJwC3VdOwQH5ulft50pXclzOs9S0HMlWNrBxS2wboh8URVC5E6GLuUpiRAfqW4sQggh8ixJuIV6khNh3WBAgUo9oERjtSMSmeFbA7osAo0FHF8O28epHZEQQmSdlS3YuOhfy3PcQgghsokk3EI9e7+FO2fB3gOaT1I7GpEVpVroeyQA7P4GDi9QNRwhhHgujp76nzI1mBBCiGwiCbdQR+R1+He6/nWLyWDvrm48Iuuq9oYGn+pf/z0CLm5VNx4hhMgqx0L6n9Hh6sYhhBAiz5KEW6hjyxeQFAtFa0OFLmpHI55X0EdQqScoKfpB1G6fVDsiIYTIPIfHd7ijZaRyIYQQ2UMSbvHyXfkXTv8BGi20+hoyMfG8yKE0Gmg7CwLqQ2I0LOsKkTfUjkoIITLnyZHKhRBCiGwgCbd4uVKSYePH+tfV3wSvCurGI16cpTV0XQKepeHRTVjeTT+3uhBC5HTGubjlGW4hhBDZQxJu8XIdXgARp/VzOcuc23mHnat+ujAHTwg/CX8MAJ1O7aiEECJjhqnBpEu5EEKIbCIJt3h5Yu7Bzgn6142+kIHS8ho3P+i+HCxs4PzfsGO82hEJIUTGZNA0IYQQ2UwSbvHy7JoC8ZH6buTV+qodjcgOvjWg/Wz9693fwLFf1Y1HCCEyYpwWTO5wCyGEyB6qJtwpKSl88cUXBAQEYGdnR/Hixfnyyy9RFEXNsER2uP/f/+dqbj4JtBbqxiOyT8WuUG+E/vVfQyDsgLrxCCFEeox3uCNAvnsIIYTIBpZqHnzKlCnMnTuXxYsXU65cOQ4fPky/fv1wcXFhyJAhaoYmzG3HBNAlQYkm+hGtRd7W8DO4ex7O/gUresGAYHAponZUQghhyjAtmC4J4h7Io05CCCHMTtU73Hv37qV9+/a0bt0af39/OnfuTLNmzTh48KCaYQlzuxkCp34HNNBkrNrRiJdBq4XXftA/PhBzB37rCUlxakclhBCmLG3A1kX/OlpGKhdCCGF+qt7hrl27Nj/++CMXLlygZMmSHD9+nN27dzNjxow0yyckJJCQkGD8PSoqCoCkpCSSkpJeKBbD9i+6n/wkU3WmKFhsGY0W0FXoQkqB0pCP6zhfvc801tD5FywXNEFz6zi6PweR0m5uluddz1d1ZiZSZ1lnzjqTes9lHAvpxxeJiQBKqx2NEEKIPEbVhPuTTz4hKiqK0qVLY2FhQUpKChMnTqRXr15plp88eTLjxo1LtXzLli3Y29ubJaatW7eaZT/5SUZ15hl1ktpX/yFFY8n2lJrEbdjwEiPLufLT+6yAzwBqX5qC9tRqzty35HKhVs+1n/xUZ+YidZZ15qiz2NhYM0QiXhqHgnD3gtzhFkIIkS1UTbhXrlzJsmXLWL58OeXKlePYsWMMHToUHx8f+vTpk6r8qFGjGDZsmPH3qKgofH19adasGc7Ozi8US1JSElu3bqVp06ZYWVm90L7yi2fWmaLD8uep+tev9Kdh09R/0/wmf77PWqEccoEtn1Du1kpK1++IUrxRprfOn3X2YqTOss6cdWbofSVyCUfDXNyScAshhDA/VRPukSNH8sknn9C9e3cAKlSoQGhoKJMnT04z4baxscHGxibVcisrK7N9qTTnvvKLdOvs1O8QfhJsnLFo8DEWUq9G+e59VutduHMKTchSLNcO0A+i5h6QpV3kuzozA6mzrDNHnUmd5zKGhDtGEm4hhBDmp2rCHRsbi1ZrOm6bhYUFOp1OpYiE2ehSYNfX+te1BsnIr/mdRgOtZ0DEWbhxBFa+AW9uAWvzPAoiXoyiKITdj+XglfucuB7JnUcJ3I9J5F5MAtEJybjaWVPA0ZoCjjZ4u9hSM8CdmsUK4GijahMihHnIHW4hhBDZSNVvS23btmXixIkULVqUcuXKERISwowZM3jzzTfVDEuYw5m1cOecfvTXV99VOxqRE1jaQNcl8EN9uH0S1n8Ir83L8iBqwjySU3TsunCHP4/dZP9/94h4lJBu2fCoBAj//+8//vMflloNVYq60qBUQbq/4ksBx9S9j4TIFRwk4RZCCJF9VE24v/vuO7744gvef/99IiIi8PHx4Z133mH06NFqhiVe1JN3t18d+P8pV4RwKQxdFsIvHeDEb1C4GtQcoHZU+crlO9GsPHyNP47e4M4TSbaVhYaKRVyp7udGYTc73B2scXewxsnGisi4JO7FJHA3OpFLEdHsvXyX0HuxHLr6gENXH/Ddjot0re7L2/WK4esuvRZELiNdyoUQQmQjVRNuJycnZs6cycyZM9UMQ5jbk3e3a76jdjQipwmoD03Hw5bPYPMo/VzdfrXUjirPuxTxiG+2XeTvE7eMywo4WNOhSmGali1EZV9XbK0sMr2/a/dj+ffiXX47FMaJ65H8si+UpftD6VC5MJ+2LoOH3PEWuYV0KRdCCJGN5AE8YV46nendbTtXVcMROVStgfpnuU//Aav6wrv//v9LrzCr0HsxzNp2kbXHbqBT9Msaly5Il+q+NCpdEGtLbcY7SIevuz09axalRw1f9v13j3m7/uOfC3f4I+QGwRfuMKFDeVpV8DbjmQiRTQxdymPu6Nsw7fP9nxBCCCHSIgm3MC+5uy0yQ6OBdt9BxBn9+2X1m/DGWrCQjyRzSdbBdzsuM++fKySm6AeibFa2EB82LUkZ7xebRvFJGo2G2sU9qF3cg+PXHvLx7yc4d/sR7y87SpuK3oxvXx53B2uzHU8Is3Pw1P/UJUPcA3AooG48Qggh8hS5jCvMR6eDXVP0r+XutngWG0fo+gtYOcDVfyF4ktoR5RmHrj7g6xMWfLvzMokpOuoFerBuUB1+7F3drMn20yr5urJuUF0GNyqBhVbD+hO3aDXrX87dlnmpRQ5maQ12bvrX8hy3EEIIM5OEW5jP+b/1dytt5O62yCTPUtDuW/3rf6fD+U3qxpPLxSWm8Pnak/T8+RDhcRo8HK2Z3bMKv7xZg4pFXF9KDNaWWoY3K8Wa92tTzNOB21HxdJm7j32X772U4wvxXGSkciGEENlEEm5hHooCu2fqX9foL3e3ReZV6Aw1Ho9UvmYAPLiqaji51cXwR7Sfs5ul+8MAqFVQx6YhdWhT0QeNClOvVSziyh/v1eYVfzceJSTTZ8FBkwHbhMhRZOA0IYQQ2UQSbmEeoXvhxmGwsIGaMu+2yKJmE6FwdYiP1A+ilpz+nNDClKIorDp8jXaz93AhPBoPRxsW9a1G9+I6XOysVI3N1d6aJW/VpEU5LxJTdAz69ShL9l1VNSYh0iRTgwkhhMgmknAL89gzU/+zck8ZbVpknaU1dFmkf47yZghs+ULtiHKFhOQURq4+wcjVJ4hLSqFuCQ82flCPOsVzzqBPtlYWzOlVlTde9UNR4Is/T7Pu+E21wxLClGMh/c/ocHXjEEIIkedIwi1eXMQZuLgFNFqoPVjtaERu5eoLr/2gf33wBzi9VtVwcrp70Qn0+ukAq49cR6uBkc1L8cubNfB0ynnzX1toNYxvX45+dfwBGLHyuDzTrZI5c+bg7++Pra0tNWvW5ODBgxmWX7VqFaVLl8bW1pYKFSqwYcMGk/WKojB69Gi8vb2xs7OjSZMmXLx4MdV+/v77b2rWrImdnR1ubm506NDBnKf14gwjlUffUTcOIYQQeY4k3OKFWeyfrX9Rph0UKK5uMCJ3K9kc6nygf71uMNz/T914cqgL4Y9oP2cPh0Mf4GRryeI3azCwYQm02pf/rHZmaTQaPm9dlpbl9d3LByw5zIXwR2qHla+sWLGCYcOGMWbMGI4ePUqlSpVo3rw5ERFpd6Peu3cvPXr04K233iIkJIQOHTrQoUMHTp06ZSzz9ddf8+233zJv3jwOHDiAg4MDzZs3Jz4+3ljm999/54033qBfv34cP36cPXv20LNnz2w/3yyRO9xCCCGyiSTc4oXYJd5Fc/oP/S+GREmIF9HoC/B9FRKiHj/PHf/MTfKTXRfu0PH7vVx/EIdfAXvWvF+HeoGeaoeVKRZaDd90q6wfSC0+mb4LDnI7Uv6+L8uMGTN4++236devH2XLlmXevHnY29uzYMGCNMvPmjWLFi1aMHLkSMqUKcOXX35J1apVmT1bf5FVURRmzpzJ559/Tvv27alYsSK//PILN2/eZO3atQAkJyfzwQcfMHXqVN59911KlixJ2bJl6dq168s67cyRZ7iFEEJkE0u1AxC5W/GITWh0yRBQHwpXVTsckRdYWEHnBfBDPbh1HO22MUCQ2lHlCOuO32TYimMk6xRqBLjzw+vVcHOwVjusLLG1suCn3tXpNHcvl+/EMGDJYVa/WxtrS7n+m50SExM5cuQIo0aNMi7TarU0adKEffv2pbnNvn37GDZsmMmy5s2bG5PpK1eucPv2bZo0aWJc7+LiQs2aNdm3bx/du3fn6NGj3LhxA61WS5UqVbh9+zaVK1dm6tSplC9fPs3jJiQkkJDw/4ETo6L087gnJSWRlJT0XOf/TDZuWAFKdATJ2XWMLDKca7adcz4l9Wp+UqfZQ+rV/MxZp1nZhyTc4vnFPcTv3i796zpDVQ1F5DEuheG1H2FZJyyO/Ix3gD3QSu2oVLV0fyhf/HkKRYG2lXyY3qVSrk1SXe2tWdSvBm2+282J65FM33qeUS3LqB1Wnnb37l1SUlIoVKiQyfJChQpx7ty5NLe5fft2muVv375tXG9Yll6Z//7TPxYyduxYZsyYgb+/P9OnT6dBgwZcuHABd3f3VMedPHky48aNS7V8y5Yt2NvbZ+Z0s8w26QHNASX6Dhv+Xq8fkySH2Lp1q9oh5ElSr+YndZo9pF7Nzxx1Ghsbm+myknCL56Y9vgwLXQJKwbJoijdSOxyR1wQ20T+msGcWlcN+hof9wDP/jRGgKArfB19m6ubzALzxqh/j2pXL0c9rZ4avuz1TOlXk3aVH+GHXf9Qr4UndQA+1wxJmptPpAPjss8/o1KkTAAsXLqRIkSKsWrWKd955J9U2o0aNMrmzHhUVha+vL82aNcPZ2Tl7Ak1JglMfoEVHqwavgoP678WkpCS2bt1K06ZNsbJSd4q/vETq1fykTrOH1Kv5mbNODb2vMkMSbvF8dCloD/8MQEr1t7HU5O4v/yKHavQFuqt7sL5xGN3aAfDmJn2X83xCURS+3nyeucGXARjcqATDmpZEk0f+v7Uo70XPmkVZfiCMD1ceY9MH9SjgmPNGWc8LPDw8sLCwIDzcdFCw8PBwvLy80tzGy8srw/KGn+Hh4Xh7e5uUqVy5MoBxedmyZY3rbWxsKFasGGFhYWke18bGBhub1O8DKyur7PvSaWUF9gUg9h5WCffB1fvZ27wk2Xre+ZjUq/lJnWYPqVfzM0edZmX7nNNnSuQu5zeiiQwj0cIBpXxntaMReZWFFSkdfiTRwh7tjcOw40u1I3ppnk62P29dhuHNSuWZZNvgi9ZlCSzoyJ1HCYxcfQJFUdQOKU+ytramWrVqbN++3bhMp9Oxfft2atWqleY2tWrVMikP+m54hvIBAQF4eXmZlImKiuLAgQPGMtWqVcPGxobz588byyQlJXH16lX8/PzMdn5m4fB44LRoGThNCCGE+UjCLZ7PgXkAXPVoCFZ2Kgcj8jTXohwr+pb+9Z5ZcHGbuvG8BIqiMPWJZHt8+3L0r1dM5aiyh521Bd/2qIK1pZYd5yJYuj9U7ZDyrGHDhvHTTz+xePFizp49y3vvvUdMTAz9+vUDoHfv3iaDqn3wwQds2rSJ6dOnc+7cOcaOHcvhw4cZNGgQoJ/qbejQoUyYMIF169Zx8uRJevfujY+Pj3GebWdnZ959913GjBnDli1bOH/+PO+99x4AXbp0ebkV8CyOj0f7j5G5uIUQQpiPdCkXWXf7FFz9F0VjwRWPxgSoHY/I8265vkJK9f5YHJ4Pa9+Fd/eAU6Fnb5gLKYrC9C0X+P5xsj22bVl61/JXN6hsVsbbmVEtSzPurzNM2XSe5uW8KOhsq3ZYeU63bt24c+cOo0ePNo4WvmnTJuOgZ2FhYWi1/78OX7t2bZYvX87nn3/Op59+SmBgIGvXrjUZXfyjjz4iJiaGAQMG8PDhQ+rWrcumTZuwtf3/32/q1KlYWlryxhtvEBcXR82aNdmxYwdubm4v7+QzQ+biFkIIkQ0k4RZZd/AHAJTSbYi3LqByMCK/0DUei8W1/RB+Cta8A6//Adq810nn2+2XmL3zEgBj2palb538cUmrTy1/1h67yfFrD5m44SyzuldRO6Q8adCgQcY71E8LDg5OtaxLly4Z3onWaDSMHz+e8ePHp1vGysqKadOmMW3atCzH+1JJl3IhhBDZIO99WxXZK+YenFgJgO6Vt1UORuQrlrb6+bkt7eC/nbD3W7UjMrtFe67wzbYLgP6Z7X75JNkG0Go1TGhfHo0G/jx2k72X76odkshvHCXhFkIIYX6ScIusOboYkuPBuxJKkZpqRyPyG89S0HKK/vWOL+H6YXXjMaM/jl5n7F9nAPiwSck8+8x2RioUceH1mvqBtEb/eZrEZJ3KEYl8xZBwx0jCLYQQwnwk4RaZl5IMh/RTgVHzXchjoyWLXKJqbyj3GuiSYfWbEB+pdkQvbOuZcEauPgFAvzr+DGlcQuWI1DOiWSkKOFhzKSKaBXuuqB2OyE+kS7kQQohsIAm3yLxLWyHqOti5Q7mOakcj8iuNBtrMBNei8DAU/h6hdkQv5MB/9xi4/CgpOoVOVYvwReuyeW7qr6xwsbfik5alAZi17SI3H8apHJHIN6RLuRBCiGwgCbfIvMML9T8r9wQrGUFYqMjOFTrOB40FnFwJx1eoHdFzuRj+iLd/OUxiso6mZQsxpVMFtNr8m2wbdKpahOp+bsQlpTBt8/lnbyCEORgS7ti7oEtRNxYhhBB5hiTcInMeXtPf4Qao1k/dWIQAKFoTGnyif/33MLj/n7rxZFF4VDx9Fx4iKj6Zan5ufNejCpYW8pEM+gHURrctC8CaYze4GP5I5YhEvmDvAWhA0UHsPbWjEUIIkUfItzuROUd/0X8J8a8HHvn3+VKRw9QbDkVrQ2I0/N4fUpLUjihTHsUn0XfhIW48jKOYhwPze1fH1spC7bBylIpFXGlWthCKAjO3XVQ7HJEfWFiC/eOpLqVbuRBCCDORhFs8W0oyhCzRv64ud7dFDqK1gI4/gq0L3DgCOyepHdEzJaXoeH/ZUc7eisLD0ZrFb9bAzcFa7bBypGHNSqLRwN8nb3H6Zu4fHE/kAo6F9D+jw9WNQwghRJ4hCbd4tgub4NEtfXe70m3VjkYIU66+0PbxnNy7v4Er/6obTwYUReGLtaf49+Jd7KwsWND3FXzd7dUOK8cq7eVM24o+AMzYckHlaES+4Oip/xlzR904hBBC5BmScItnO7xA/7PK62Apd+JEDlSug/79iQJr3oHY+2pHlKb5/17ht0PX0Gpgds8qVCziqnZIOd7QJoFYaDVsPxfB0bAHaocj8jqZGkwIIYSZScItMvbgKlzeoX9drY+qoQiRoRZTwL04RN2A9UNBUdSOyMTWM+FM2ngWgM9al6VxmUIqR5Q7FPN0pFPVwoDc5RYvgXFqMOlSLoQQwjwk4RYZO7IYUKBYQ3AvpnY0QqTPxhE6/QRaSzjzJxxbrnZERqdvRvLBbyEoCvSsWZQ36/irHVKuMqRxIFYWGnZfusv+/2T0aJGNDAm3dCkXQghhJpJwi/SlJP8/aZHB0kRuULgaNPxU/3rDSLh3Wd14gIhH8fRffJjYxBTqBXowrl05NBqZazsrirjZ0+0VXwB+2KX+31TkYTJomhBCCDOThFuk7/J2iL6tnyalZEu1oxEic+oMBb+6kBQDf7yt6lRhCckpvLvkCLci4ynu6cDsnlWxkrm2n0v/usXQaGDn+TtcipB5uUU2cXg8aFq03OEWQghhHvLNT6Tv2DL9zwpdZbA0kXtoLaDjD/+fKmzXFFXCUBSFz9ec4mjYQ5xtLZnf5xVc7KxUiSUv8PdwoFlZ/d3Hn3dfUTkakWfJM9xCCCHMTBJukbbY+3B+o/51lV7qxiJEVrkUgTYz9a//nQ6h+156CAv3XGXVkeuPRySvSoCHw0uPIa/pX08/jsTvR29wNzpB5WhEnmToUh57T/9YlRBCCPGCJOEWaTu5ClISwasieFVQOxohsq58R6jUAxQdrBkA8ZEv7dC7L95l4gb9iOSftipD/ZKeL+3YeVl1Pzcq+bqSmKxj6f5QtcMReZF9AdBoAUWfdAshhBAvSBJukbaQpfqfleXutsjFWn4Nrn7wMEw/iNpLEHYvloHLj5KiU+hUtQhv1Q14KcfNDzQaDf0f1+eSfaHEJ6WoHJHIc7QWYO+hfy3dyoUQQpiBJNwitdsn4fYJ0FpBhS5qRyPE87N1ho4/6e9YnVgBJ1dn6+FiE5MZsOQwkXFJVPJ1ZeJr5WVEcjNrWd6Lwq523ItJZE3IDbXDEXmR8TnuCHXjEEIIkSdIwi1SM0wFVqolOBRQNxYhXlTRmlD/I/3r9cPg4bVsOYyiKHzy+0nO3X6Eh6MNP7xeDVsri2w5Vn5maaGl3+N5zOf/+x86naJuQCLvMc7FLQm3EEKIFycJtzCVnKi/EwhQ5XV1YxHCXOqPhMLVISES1rwLOvN3Rf559xXWHb+JpVbD972q4uVia/ZjCL1ur/jiaGPJ5Tsx/HNRpm8SZuYgd7iFEEKYjyTcwtTFzfqBYhwLQfHGakcjhHlYWELHH8HKAUJ3w97vzLr7vZfuMunxIGlftClLjQB3s+5fmHKytaJztSIArDiUPT0WRD7m+HiQwxi5mCOEEOLFScItTB37Vf+zYjd9kiJEXlGgOLR8PCf3jglw67hZdnvjYRyDfg1Bp0CnqkXoXcvPLPsVGev2ii8AW8+Ec+eRTBEmzMgwNZgMmiaEEMIMJOEW/xd7Hy5u0b+u1EPdWITIDlVeh9JtQJcEv78NibEvtLuE5BTeX3qE+zGJlPNxlkHSXqIy3s5U8nUlWafwx9Hraocj8hLpUi6EEMKMVE24/f390Wg0qf4NHDhQzbDyrzN/6hORQuWhUFm1oxHC/DQaaPstOHrB3fOwdfQL7W78X2c4fj0SFzsr5skgaS9dj8d3uVccuoaiyOBpwkxklHIhhBBmpGrCfejQIW7dumX8t3XrVgC6dJGpqFRxcpX+p0wFJvIyhwLQ4Xv960M/wcWtz7Wb1Ueus+xAGBoNzOxeGV93ezMGKTKjTSUf7K0t+O9uDAev3Fc7HJFXyCjlQgghzEjVhNvT0xMvLy/jv/Xr11O8eHGCgoLUDCt/irwOoXv0ryt0VjcWIbJbicZQ81396z8HQszdLG1++mYkn605CcDQxiVpWKqguSMUmeBoY0m7Sj6ADJ4mzMjwDHfsPUhJUjcWIYQQuV6OGRUrMTGRpUuXMmzYsHSfgUxISCAh4f+D40RFRQGQlJREUtKLNYqG7V90P7mV9vgKLABd0Vqk2BeCTNRDfq+z5yF1lnXZVmdBn2F5eSeau+fR/TmYlM6L9V3OnyEqLol3lxwhIVlHUEkP3q3nl+P+nvnpfdapije/HbrG3ydv8WnLkrjYWT3XfsxZZ/mh3vM0O3fQWICSor8Y5+ytdkRCCCFysRyTcK9du5aHDx/St2/fdMtMnjyZcePGpVq+ZcsW7O3N053T0K09v2lwdiEuwAmlNKEbNmRp2/xaZy9C6izrsqPOnAu8TtC9sWgvbOD40k8IK5Bx7xqdAj+f13LtgRZ3G4XmzrfZtGmj2eMyl/zwPlMU8Laz4Facjq9+3UY9rxd7ltscdRYb+2KD8QmVabXg4KEfpTw6XBJuIYQQLyTHJNw///wzLVu2xMfHJ90yo0aNYtiwYcbfo6Ki8PX1pVmzZjg7O7/Q8ZOSkti6dStNmzbFyur57pDkWhFnsQq5hqK1olyXTyln55apzfJ1nT0nqbOsy+46U/Ylw45xVL71G+VbDwD34umW/eGfK5x6cBErCw0/93uV8oVf7HMnu+S399ld91AmbDjPqThXJrV89blGijdnnRl6X4lczLGgPtmWubiFEEK8oByRcIeGhrJt2zb++OOPDMvZ2NhgY2OTarmVlZXZvlSac1+5xll9vWsCm2HlnPVnUfNlnb0gqbOsy7Y6qzsU/tuB5uq/WK17H97cDBapj7Pv8j1mbLsIwLh25aniX8D8sZhZfnmfda5elK+3XOTc7UdcuBNH+cIuz70vc9RZfqjzPE+mBhNCCGEmOWIe7oULF1KwYEFat26tdij5j04HJ1frX1eU0clFPqTVwmvzwNYFbhyBf6alKhIRFc/gX0PQKdCxamF61PBVIVCRHld7a5qU0SdIfx2/qXI0Ik8wDJwWHa5uHEIIIXI91RNunU7HwoUL6dOnD5aWOeKGe/5y7QBEXgNrJyjZQu1ohFCHSxFoPUP/+p+pcO2gcVVyio5Bv4ZwNzqB0l5OTOxQ4bm6LIvs1bai/nGkv47fRKeTObnFC3L01P+ULuVCCCFekOoJ97Zt2wgLC+PNN99UO5T86dTju9tl2oKVnbqxCKGmCp2hQlf9yMR/DICERwBM23KBg1fu42hjyfe9qmJnbaFyoCItDUsXxNHGkpuR8RwJe6B2OCK3kzvcQgghzET1hLtZs2YoikLJkiXVDiX/0aXAmT/1ryt0UjcWIXKCVlPBxRceXIFNn7D9bDjzdl0G4OvOFSnm6ahygCI9tlYWNCunT5LWHZNu5eIFyTPcQgghzET1hFuoKHSPvrucnRsEZDwdkhD5gp2r/nluNBCylPUrfgCgb21/WlWQqYFyunaV9N3KN5y8RXKKTuVoRK7mKAm3EEII85CEOz87vVb/s3SbNEdlFiJf8q9Lcu0PABit/EBDnxQ+bVVG5aBEZtQp4YG7gzX3YhLZe/me2uGI3MyQcMdIwi2EEOLFSMKdX+lS4Ow6/etyHVQNRYicZkJ0B07q/HHTRDPXcT7W8kmZK1hZaGlVwQuAdTJauXgRhi7lcQ8gOVHdWIQQQuRq8jUyv5Lu5EKkad3xmyw6eJOhSQNJsbDFNmwXHPxB7bBEJrWrVBiAzaduE5+UonI0IteycwPt45lTZKRyIYQQL0AS7vzq9Br9T+lOLoTRpYhoPvn9BAAtGwRh0WKifsXWMRB+WsXIRGZV93PD28WWRwnJBJ+XREk8J60WHAxTg0m3ciGEEM9PEu78KCUZzhi6k7+mbixC5BBxiSm8v+wIsYkp1CpWgA+bloTqb0Fgc0hJgN/fhqR4tcMUz6DVamj7ePC0v05It3LxAmTgNCGEEGYgCXd+FLoHYu8+7k5eX+1ohFCdoih8vvYUF8Kj8XSyYVaPylhoNaDRQPvZYO8BEadhx5dqhyoyoW1FfcK9/Ww4sYnJKkcjci2ZGkwIIYQZSMKdHxm6k5dpK93JhQBWHr7G70evo9XAt92rUNDJ9v8rHQtC+zn61/tmw+Ud6gQpMq18YWeKutsTn6Tjnwt31Q5H5FaO+nndiQ5XNw4hhBC5miTc+U1KMpz9S/9aupMLwZmbUYz+U/989vBmpahVvEDqQqVawCv99a/XvAcxMuVUTqbRaGhaVp8sbTlzW+VoRK7laHiGW8YCEEII8fwk4c5vQnc/7k7uDv7SnVzkb1HxSby/7AgJyToalvLkvaDi6Rdu+iV4lITo2/DXEFCUlxeoyDJDwr3jXATJKTqVoxG5krFLudzhFkII8fwk4c5vDIOllWkDFpbqxiKEihRF4aNVJ7h6L5bCrnZ8060yWq0m/Q2s7aHTfNBawbn1cPSXlxesyLLqfm642VvxMDaJw6EP1A5H5EbGQdPkDrcQQojnJwl3fqLTwfkN+tdl2qkbixAqW7DnKptO38bKQsOcXlVxtbd+9kbelaDxaP3rTZ/A3UvZG6R4bpYWWhqVftyt/LTcoRTPwZBwy7RgQgghXoAk3PnJzaPw6BZYO8no5CJfOxL6gMkbzgLweeuyVPZ1zfzGtQbp//8kxcLvb0FyYvYEKV6YoVv51rO3UeQRAADmzJmDv78/tra21KxZk4MHD2ZYftWqVZQuXRpbW1sqVKjAhg0bTNYrisLo0aPx9vbGzs6OJk2acPHiRZMy/v7+aDQak39fffWV2c/N7GTQNCGEEGYgCXd+YhgsLbApWNqoG4sQKrkXncCg5UdJ1im0ruhN71p+WduBVguv/aCfVu/WMdg5MVviFC+ufkkPbCy1XLsfx/nwR2qHo7oVK1YwbNgwxowZw9GjR6lUqRLNmzcnIiLtO7h79+6lR48evPXWW4SEhNChQwc6dOjAqVOnjGW+/vprvv32W+bNm8eBAwdwcHCgefPmxMebzlk/fvx4bt26Zfw3ePDgbD1Xs3B4PGhafCQkJ6gbixBCiFxLEu78QlH0z52C/vltIfKhFJ3C0BXHuBUZTzFPB6Z0qohGk8Fz2+lx9oF23+lf75kFV/4xb6DCLOytLakX6AFIt3KAGTNm8Pbbb9OvXz/Kli3LvHnzsLe3Z8GCBWmWnzVrFi1atGDkyJGUKVOGL7/8kqpVqzJ79mxAf3d75syZfP7557Rv356KFSvyyy+/cPPmTdauXWuyLycnJ7y8vIz/HBwcsvt0X5ydm37MBpC5uIUQQjw3GTUrv7h7Ae5dAgtrKNFU7WiEUMW32y/y78W72FlZMLdXNRxtXuAjsExbqNoHji6GP96B9/aAvbv5ghVm0bRsIbadjWDrmXCGNA5UOxzVJCYmcuTIEUaNGmVcptVqadKkCfv27Utzm3379jFs2DCTZc2bNzcm01euXOH27ds0adLEuN7FxYWaNWuyb98+unfvblz+1Vdf8eWXX1K0aFF69uzJhx9+iKVl2v//EhISSEj4/x3lqKgoAJKSkkhKSsraib8gSwdPNI9ukhx5E8XB66Ue23CuL/uc8zqpV/OTOs0eUq/mZ846zco+JOHOLwzdyYs1AFtnVUMRQg3B5yP4dof+2dKJr5WnlJfTi++0xWQI3Qv3LuqnCuu6BJ7njrnINo1KF0KjOcnJG5HcfBiHj6ud2iGp4u7du6SkpFCoUCGT5YUKFeLcuXNpbnP79u00y9++fdu43rAsvTIAQ4YMoWrVqri7u7N3715GjRrFrVu3mDFjRprHnTx5MuPGjUu1fMuWLdjb2z/jTM2rfooNbsDhXRsJd1FnTvetW7eqcty8TurV/KROs4fUq/mZo05jY2MzXVYS7vzC0J28tHQnF/nPjYdxfLjiGIoCPWsWpWPVIubZsbWDfqqw+U30F7WOLILq/cyzb2EWnk42VC3qxpHQB2w7G07vWv5qh5TvPHmXvGLFilhbW/POO+8wefJkbGxSjycyatQok22ioqLw9fWlWbNmODu/3AvGFo+WwKUrVC9dFKVKq5d67KSkJLZu3UrTpk2xsrJ6qcfOy6RezU/qNHtIvZqfOevU0PsqMyThzg8ir8PNEEADpVqqHY0QL1VCcgrvLzvKg9gkyhd2ZnSbsuY9gE9laPwFbB0Nm0ZB0VpQsLR5jyFeSLOyhTgS+oCtZ/Jvwu3h4YGFhQXh4abPsoeHh+PllXZXaS8vrwzLG36Gh4fj7e1tUqZy5crpxlKzZk2Sk5O5evUqpUqVSrXexsYmzUTcysrq5X/pdNLfvbeMvwcqfeFV5bzzAalX85M6zR5Sr+ZnjjrNyvYyaFp+cO5v/c+ir/5/XlEh8okv15/h+LWHuNhZMbdXNWytLMx/kFqDoVhDSI7TTxWWFP/sbcRLY5gebP9/94hJSFY5GnVYW1tTrVo1tm/fblym0+nYvn07tWrVSnObWrVqmZQHfTc8Q/mAgAC8vLxMykRFRXHgwIF09wlw7NgxtFotBQvmgvbIODWYDJomhBDi+UjCnR9Id3KRT/1x9DpL94eh0cDMbpXxdc+m5z8NU4XZe0D4Kdg2JnuOI55LgIcDvu52JKUo7Lt8T+1wVDNs2DB++uknFi9ezNmzZ3nvvfeIiYmhXz/9YxC9e/c2GVTtgw8+YNOmTUyfPp1z584xduxYDh8+zKBBgwDQaDQMHTqUCRMmsG7dOk6ePEnv3r3x8fGhQ4cOgH7gtZkzZ3L8+HH+++8/li1bxocffsjrr7+Om5vbS6+DLHN4fFFAEm4hhBDPSbqU53Wx9+HqHv1rmQ5M5CNnb0Xx6ZqTAAxuFEjD0tl8N82pEHSYC8u7wIF5ULwRlGyevccUmaLRaKgf6MmyA2H8c/EOTcoWevZGeVC3bt24c+cOo0eP5vbt21SuXJlNmzYZBz0LCwtDq/3/dfjatWuzfPlyPv/8cz799FMCAwNZu3Yt5cuXN5b56KOPiImJYcCAATx8+JC6deuyadMmbG1tAX338N9++42xY8eSkJBAQEAAH374YarRz3MsR0m4hRBCvBhJuPO6C5tBSYFCFcDNX+1ohHgpIuOSeG/pEeKTdNQL9OCDlzUdVMlmUPM9ODAX1r4H7+4BZ+9nbyeyXVDJxwn3hTtqh6KqQYMGGe9QPy04ODjVsi5dutClS5d096fRaBg/fjzjx49Pc33VqlXZv3//c8WaIxgS7hhJuIUQQjwf6VKe113YpP8pg6WJfEKnUxi+8hhX78VS2NWOWd2rYKF9iVN1NR0HXhUg9h6sGQC6lJd3bJGuWsULYKnVcPVeLGH3Mj+Vh8jn5BluIYQQL0gS7rwsOREuPR7MpmQLdWMR4iWZs/MS285GYG2pZe7rVXF3sH65AVjaQOeFYOUAV/6B3d+83OOLNDnZWlHVT//M8K6L+fsut8gCB0/9z4QoSIpTNxYhhBC5kiTceVnYXkh8pB/0xaeK2tEIke12no9gxrYLAEzoUJ6KRVzVCcQjEFpNfRzUJAg7oE4cwkRQSX3ytOu8JNwik2xdwOLxRTu5yy2EEOI5SMKdl13YrP9Zspl+FGUh8rCwe7EM/e0YigI9axala3VfdQOq3BMqdNWPofD7WxD3QN14BPUD9Qn3vst3SUzWqRyNyBU0mv93K4+RCzVCCCGyTrKwvEpR4PxG/euS8vy2yNviElN4Z+kRIuOSqOzrypi2ZdUOSf9Fvc0McAuAyGuwbrD+/6VQTTkfZwo4WBOTmMLRMLkAIjLJ0K1c7nALIYR4DpJw51V3L8KDK/qucMUaqB2NENlGURQ++v0EZ29F4eFozdzXq2JjaaF2WHo2TtB5AWit4OxfcGi+2hHla1qthnqBHgD5frRykQXGgdPC1Y1DCCFEriQJd1514fHdbf96YOOobixCZKOf/v2Pv47fxFKr4fte1fB2sVM7JFOFq0LTx1Mmbf4Ubh1XN558rv7j57j/kYHTRGY5Pr7DLV3KhRBCPAdJuPMqw/PbMh2YyMP+vXiHrzaeA2BM27LUCHBXOaJ0vPoelGoFKYmwqi/ER6kdUb5V7/Fz3KduRHE3OkHlaESuIHe4hRBCvABJuPOi2PsQtl//OrCZurEIkU3C7sUyaHkIOgW6Vi/C66/6qR1S+jQaaD8HnIvA/f9g/VB5nlslnk42lPV2BmD3xbsqRyNyBYeC+p+ScAshhHgOknDnRZe260dGLlgW3HJwEiLEc4pOSGbAksPGQdLGty+PRqNRO6yM2bvrn+fWWMCp3+HoYrUjyreCSj3uVi7PcYvMMHQpj5b3ixBCiKyThDsvMjy/XbKFunEIkQ10OoVhK45x7vYjPJ1smPd6NWytcsggac9StCY0Hq1/vfFjuH1S3XjyKcPAaf9euosiPQ3EsxinBZNRyoUQQmSdJNx5TUoSXNqmfy0Jt8iDvtl2gS1nwrG21PLDG9XwcrFVO6SsqT1E/6hHcjys7CPPc6ugalE3rC213HmUwH93Y9QOR+R0xi7lcodbCCFE1knCnddcOwjxkWBfAIpUVzsaIczqr+M3+W7HJQAmv1aBqkXdVI7oOWi18NoPj5/nvizzc6vA1sqCqkVdAdj/3z11gxE5n+PjhDvxESTGqhuLEEKIXEcS7rzm0lb9z+KNQZtLutkKkQmnb0YxcrV+Sq0B9YvRqVoRlSN6Afbu0GURaC3hzFqZn1sFrxYrAMD+/+6rHInI8WycwPJxTxrpVi6EECKLJOHOawzdyQObqhuHEGYUmQjvLAshPklHg1KefNyitNohvTjfV6Dpl/rXm0bBjSPqxpPP1HqccP+vvfuOj6pK/zj+mUnvoaUSIBTpHUSKYkFAbKyKwrKK6OJaEFisqOCCBVdXRdQVe/uJuDZsiMQoRaV3pDcpIQkQQhpJJpn7+2OSgUiATDKTO0m+79crr5ncueWZQ7nzzDnnOUt3HdU8bjk7i+VkL3eOEm4REXGNEu7aJDu1pAiTBVpcanY0Im5xorCYN7f6kJZVQMuoUGaO6IqP1csrklfUBXdBm6vAboP/3epY0k+qReeESAJ8rRzJKWDXYc3jlnMIUcItIiKV42t2AOJGO5Mdj3FdIaShubGIuIHdbnD/5xvZn2uhXrAf74zqSXign9lhuY/FAkP/C29sdqzP/cUY+Ounjnne4lGOedz1WLr7KMt2H6Vp9zizQzoju93Ozp07SU9Px263l3ntoosuMimqOqa0h1tDykVExEVKuGuT0vnbLQeYG4eIm/xnwTYWbE7Hx2Lw3792oUmDYLNDcr/ACLjxQ3hrgGNKyOLn4OKHzI6qTujdogFLdx9l6e6j3OSlCfeyZcv461//yh9//HHa0HeLxUJxcbFJkdUxGlIuIiKVpG6U2qK4CHb97Hiu+dtSC3y6aj//XbgLgOEt7PRoWgMrkldUTAe46kXH84XTYceP5sZTR5QWTlu+23vncd9555306NGDTZs2kZGRwbFjx5w/GRmaglBtNKRcREQqyfSE++DBg/ztb3+jQYMGBAUF0bFjR1atWmV2WDXPwdWQnwmBkRDf3exoRKrk151HmPTFRgDuuiiR8xt5ZzLkVl1GQI/bAAO++Dtk7jM7olqvc0JEyTzuQq+dx71jxw6efvpp2rZtS2RkJBEREWV+pJo4e7jTzI1DRERqHFMT7mPHjtG3b1/8/Pz4/vvv2bx5M88//zz16tXinixPKa1O3uJSLQcmNdq21Gzu/HA1RXaDqzvHMeGylmaHVH0GPwNx3eDEMfjkb2A7YXZEtVqArw89mjnuN8v3HjM5mvL16tWLnTt3mh2GOOdwHzY3DhERqXFMncP973//m4SEBN59913ntsTERBMjqsE0f1tqgbSsfEa/u4LsgiLOb1af527ohBX7uQ+sLXwD4MYP4I3+cGg9fPtPGPqao7iaeMQFiQ34dedRlu/OYHC42dGc7t577+W+++4jNTWVjh074udXtmhgp06dTIqsjglRD7eIiFSOqQn3119/zaBBgxg2bBiLFi0iPj6eu+++mzFjxpS7f0FBAQUFBc7fs7KyALDZbNhstirFUnp8Vc9jitzD+KWsBcDW9CKopvdQo9vMJGqzM8stKGL0uytJOZ5P84bBvDqiMz7Y616bhcRg+ctb+My+Acv6jymO7oy9599dOkWda7Mq6NHUMSx7+Z4MBnVyT5u5s92vv/56AG677TbnNovFgmEYKppWnZxDytXDLSIirjE14d69ezevvfYaEydO5JFHHmHlypWMGzcOf39/Ro0addr+06dPZ+rUqadtX7BgAcHB7qlenJSU5JbzVKfGGb/SHcgMasKiJWuq/fo1sc3MpjYrq8gOb261svW4lVBfg5EJWfy2sGwb1bU2axF3Ix0OfoxlwaMs25NNRmhrl89R19qsMors4Gf1ISPPRtoJ97RZXl6eGyJz2LNnj9vOJVVQmnDbcqEgBwJCzY1HRERqDFMTbrvdTo8ePXj66acB6Nq1K5s2bWLWrFnlJtyTJk1i4sSJzt+zsrJISEhg4MCBhIdXbSygzWYjKSmJyy+//LQhe97O56uvAQjreh1DLhlSbdetyW1mFrXZ6ex2gwe/2MTW44cI8rPy3m096dz4ZDGoOttmxhXY5+Zj3fwl/VLepOi2ZAiPrdChdbbNKunzI6v4bVcGO7Is3HztgCq3WenoK3do2rSp284lVeAfCn7BYMtzrMWthFtERCrI1IQ7NjaWdu3aldnWtm1bPv/883L3DwgIICAg4LTtfn5+bvtQ6c5zVQu7HXY7lgPzOW8gPibEXuPazAuozU566rvNfLX+EL5WC6/9rTs9EhuWu1+dbLOhr8LRHVjSNuH3+SgYPQ/8gip8eJ1ss0ro06Ihv+3KYOdxi1varKrHf/3111xxxRX4+fnx9ddfn3Xfa665pkrXkgqyWCCkEWT+4RhWXr+52RGJiEgNYWrC3bdvX7Zt21Zm2/bt2/WNvitS10PeUfAPg4TzzY5GxCVvLt7Nm0scQ2afvaETF7eOMjkiL+MfAsM/gjcuhpQ18M14+MvrKqLmZhc0b0B0WADh/t5RFX7o0KGkpqYSFRXF0KFDz7if5nBXs9DokoRbhdNERKTiTF0W7J///CfLli3j6aefZufOncyePZs33niDe+65x8ywapZdjt5tEi8EH/VkSc3x+eoDPDVvCwCTrmjDdd0amxyRl6rXDIa9DxYf2PAJ/Pay2RHVOt2b1mPJAxdxfaJ3VMS32+1ERUU5n5/pR8l2NXMuDZZubhwiIlKjmJpw9+zZky+//JKPP/6YDh068MQTTzBjxgxGjhxpZlg1S8lwcppfYm4cIi6YvymVBz5bD8Dt/RK54yINzzyr5v0da3QD/Pg47FAxNHeyWCxYNGpAziWkkeNRlcpFRMQFpibcAFdddRUbN24kPz+fLVu2nHFJMClHYR7sW+Z43kIJt9QMS3YcZtzHa7EbcEP3xjw6pK2SnYo4fwx0uwUMO3x2OxzebnZEUk2Sk5O56qqraNGiBS1atOCqq67ixx9/NDusuic02vGoIeUiIuIC0xNuqYJ9v0FxIYQ3hgYtzY5G5JxW/3GMOz5YTWGxnSs6xPDMdR2xWpVsV4jFAkOehya9oeA4zL4Rco+aHZV42H//+18GDx5MWFgY48ePZ/z48YSHhzNkyBBeffVVs8OrW0JLerhz1cMtIiIVZ2rRNKmi0vnbLS5WESXxepsOHmf0uys4YSvmwlYNmTG8C74++s7PJb7+cNP/wZuXwrE98Mnf4Ja54Hv66g1SOzz99NO8+OKLjB071rlt3Lhx9O3bl6efflo1T6qTerhFRKQSKvVpd82aNWzcuNH5+1dffcXQoUN55JFHKCwsdFtwcg67FzkeNX9bvNzmlCz+9vZysvKL6NG0Hq/f3J0AXx+zw6qZQhrCXz+BgHDHKJdvJoBhmB2VeEhmZiaDBw8+bfvAgQM5fvy4CRHVYSElRdNyVDRNREQqrlIJ9z/+8Q+2b3fMH9y9ezfDhw8nODiYTz/9lAcffNCtAcoZ5KRDWsmXHon9zY1F5Cy2pWbzt7eXk5lno3NCJO+O7kmwvwbXVElUWxj2LlissH42/DrD7IjEQ6655hq+/PLL07Z/9dVXXHXVVSZEVIeVDinPSdeXXCIiUmGV+tS7fft2unTpAsCnn37KRRddxOzZs/n1118ZPnw4M2bMcGOIUq7S3u2Yjic/BIh4mZ3p2Yx8axkZuYV0jI/gg9vOJyxQy9e5RcsBcMWzMO9++PFfjuXD2v/F7KjEDWbOnOl83q5dO5566ikWLlxI7969AVi2bBm//vor9913n1kh1k2lPdxFJ6AwBwLCzI1HRERqhEol3IZhYLc71iv98ccfnd+yJyQkcOTIEfdFJ2em5cDEy+1Mz2bEm8s5klNIu9hwPrz9fCKClGy71flj4MgOWPE6fPEPCIuFJheYHZVU0Ysvvljm93r16rF582Y2b97s3BYZGck777zDY489Vt3h1V0BoeAXArZcRy+3Em4REamASiXcPXr04Mknn2TAgAEsWrSI1157DYA9e/YQHR3t1gClHIZxSsE0JdzifbamZjHyzeUczS2kTUwYH/29F5HB/maHVTsNng7HD8C27+Dj4XD7jxDR1OyopAr27NljdghyJqFRjoKFOenQoIXZ0YiISA1QqTncM2bMYM2aNYwdO5ZHH32Uli0dS1J99tln9OnTx60BSjmObIfsFPAJcCwRJOJFNh08zvA3lnE0t5AO8eF8POYC6oUo2fYYqw9c/xbEd4cTx+Cj67VskYinhJYMK89V4TQREamYSvVwd+rUqUyV8lLPPfccPj6qPOxxuxc6HptcAH5BpoYicqq1+44x6p0VZOUX0Tkhkg9u0zDyauEfDCM+gbcHwLG9+PxvJD4N7zI7KpHaJ1SVykVExDWVXgQ3MzOTt956i0mTJpGRkQHA5s2bSU/XTcjjNJxcvNBvu47wt7dOLv31f5qzXb1CG8HIzyGoPtaUNfTc8zIU28yOSqR20dJgIiLiokol3Bs2bKBVq1b8+9//5j//+Q+ZmZkAfPHFF0yaNMmd8cmfFRfB3l8cz1UwTbzE/E2HuPWdleQWFtOnRQPeVzVyczRsCX/9H4ZfMNHZG/D59l4oKXApIm6gIeUiIuKiSiXcEydOZPTo0ezYsYPAwEDn9iFDhrB48WK3BSflSFkLhdkQVA9iOpkdjQhzVuzj7o/WUFhsZ3D7GN65tSchAVpn2zQJPSm+/l3s+GDd9Bn88IjWDBZxFw0pFxERF1XqU/HKlSt5/fXXT9seHx9PampqlYOSs9hb8oVG075grfSMAJEqMwyD1xbt4tn52wAY3jOBp/7SER+rxeTIxGhxGWubjqH7H7Ng+WsQ0hAuut/ssKSSMjMzWbFiBenp6c4lOUvdcsstJkVVR2lIuYiIuKhSCXdAQABZWVmnbd++fTuNGjWqclByFntKEu7E/ubGIXVaUbGdx7/+nY+W7wPgrotb8OCg1lgsSra9xYH6fehyXmN8kh6Dn56AwAjHut1So3zzzTeMHDmSnJwcwsPDy/wbs1gsSrirW2jJ0qdKuEVEpIIq1UV6zTXXMG3aNGw2R0Eei8XCvn37eOihh7j++uvdGqCcoqgA9i13PE+80NxYpM7KLShizAer+Gj5PiwWmHxVOx4a3EbJtheyn38nXPSA45d598Paj8wNSFx23333cdttt5GTk0NmZibHjh1z/pQWLJVqFFrSqZCbrqkaIiJSIZVKuJ9//nlycnKIiorixIkT9O/fn5YtWxIWFsZTTz3l7hil1IFVUHQCQhpBozZmRyN1UHpWPje+vpSftx0m0M/KayO7c3u/RLPDkrO55FG44G7H86/HwqbPzY1HXHLw4EHGjRtHcHCw2aEInBxSXpQPBaeP9BMREfmzSg0pj4iIICkpiV9++YUNGzaQk5NDt27dGDBggLvjk1PtXeJ4bHYhqDdRqtmGA5nc8cFqUrPyaRDiz1ujetC1ST2zw5JzsVhg0NNgy4PV78EXd4BfMLS+wuzIpAIGDRrEqlWraN68udmhCDjWvPcPcxQvzTnsmKohIiJyFlUqJdyvXz/69evnrljkXJzzty8yNw6pc75ad5AHP9tAQZGdFo1CePfW82nSQD1uNYbFAle+AIV5sPF/8MnNcNP/QevBZkcm53DllVfywAMPsHnzZjp27IifX9nl9q655hqTIqvDQhtBRrZjWHnDlmZHIyIiXq7CCffMmTO54447CAwMZObMmWfdd9y4cVUOTP7EdgIOrHQ8V8It1aTYbvDsD1t5fdFuAC5rE8WM4V20xnZNZPWBoa9BcSFsnguf/A1u+lA93V5uzBhHobtp06ad9prFYqG4uNil87366qs899xzpKam0rlzZ15++WXOP//8M+7/6aefMnnyZPbu3UurVq3497//zZAhQ5yvG4bB448/zptvvklmZiZ9+/bltddeo1WrVqedq6CggF69erF+/XrWrl1Lly5dXIrda4RGQ8ZuyEkzOxIREakBKpxwv/jii4wcOZLAwEBefPHFM+5nsViUcHvC/uWOD8rh8VBfQwvF8zJyC5nwyToWbz8MwN0Xt+C+ga217FdN5uML17/l6PH+/UtHT/eN70ObK82OTM7gz8uAVcUnn3zCxIkTmTVrFr169WLGjBkMGjSIbdu2ERUVddr+v/32GyNGjGD69OlcddVVzJ49m6FDh7JmzRo6dOgAwLPPPsvMmTN5//33SUxMZPLkyQwaNIjNmzcTGBhY5nwPPvggcXFxrF+/3m3vyRQhJYXTcg6bG4eIiNQIFS6atmfPHho0aOB8fqaf3bt3eyzYOq10OLnmb0s1WP3HMa6cuYTF2x3F0WaO6MqDg9so2a4NfPzgureg/XVgt8H/boEt35gdlVSDF154gTFjxjB69GjatWvHrFmzCA4O5p133il3/5deeonBgwfzwAMP0LZtW5544gm6devGK6+8Ajh6t2fMmMFjjz3GtddeS6dOnfjggw9ISUlh7ty5Zc71/fffs2DBAv7zn/94+m16nnNpMPVwi4jIuVVqDve0adO4//77T6uaeuLECZ577jmmTJniluDkFHtKCqZpOLl4kGEYvP3LHp75fitFdoPmjUL478hutIkJNzs0cScfX7juTceXd5s+h/+NgqH/hc7DzY5M8MwUrsLCQlavXs2kSZOc26xWKwMGDGDp0qXlHrN06VImTpxYZtugQYOcyfSePXtITU0tUzA1IiKCXr16sXTpUoYPd/x9SktLY8yYMcydO7dC1dYLCgooKChw/p6V5agGbrPZnMuRmska1AAfwJ6dSrEH4yl9r97wnmsTtav7qU09Q+3qfu5sU1fOUamEe+rUqdx5552n3Tjz8vKYOnWqEm53K8iGg6sdz7X+tnjI0ZwCHvp8Az9uSQfgqk6xPHN9J0IDqlRbUbyVjy/85Q3wDYR1H8GX/4D8LOh1h9mR1XmemMJ15MgRiouLiY6OLrM9OjqarVu3lntMampqufunpqY6Xy/ddqZ9DMPg1ltv5c4776RHjx7s3bv3nLFOnz6dqVOnnrZ9wYIFXrE8WtMjaXQB0nb/zop58zx+vaSkJI9foy5Su7qf2tQz1K7u5442zcvLq/C+lfokbRgGlnKGNa9fv5769etX5pRyNvuWgVEMkU0hsonZ0Ugt9PO2dB74dANHcgrw97Hy6JVtuaV303L/nUst4uML17wCAWGwfBZ8/wAUHIcL79fUFRPt2bOn3Oc10csvv0x2dnaZnvVzmTRpUpme9aysLBISEhg4cCDh4eaPtrFst8D+d4kOoUwBOXez2WwkJSVx+eWXn1adXipP7ep+alPPULu6nzvbtHT0VUW4lHDXq1cPi8WCxWLhvPPOK/NhvLi4mJycHO68805XTikVoeXAxEPybcU88/1W3vttLwDnRYcy46autIsz/0OtVBOrFQY/41hPeNG/4acnIS8DBj7leE1qhYYNG+Lj40NaWtl5x2lpacTExJR7TExMzFn3L31MS0sjNja2zD6lFch/+uknli5dSkBAQJnz9OjRg5EjR/L++++fdt2AgIDT9gfw8/Pzjg+dEXEAWPOOYK2GeLzmfdcyalf3U5t6htrV/dzRpq4c71LCPWPGDAzD4LbbbmPq1KlEREQ4X/P396dZs2b07t3blVNKRSjhFg9YsSeDhz7fwJ4juQDc2qcZD1/RhkA/H5Mjk2pnscAljziS7h8egWX/heMH4Lo3wC/I7OjEDfz9/enevTvJyckMHToUcFRAT05OZuzYseUe07t3b5KTk5kwYYJzW1JSkvM+n5iYSExMDMnJyc4EOysri+XLl3PXXXcBjvnoTz75pPP4lJQUBg0axCeffEKvXr3c/0arQ2hJRfecNDAMjQYREZGzcinhHjVqFOC4yfbp00fftlSH/OOQusHxvFk/c2ORWiG3oIhn52/l/aV/ABAdHsAz13fiktanLwskdUzvexwVmOfeBVu+hg/SYPjHENLA7MjEDSZOnMioUaPo0aMH559/PjNmzCA3N5fRo0cDcMsttxAfH8/06dMBGD9+PP379+f555/nyiuvZM6cOaxatYo33ngDcMwhnzBhAk8++SStWrVyLgsWFxfnTOqbNCk7DSo0NBSAFi1a0Lhx42p6524WUvJ/ZXEhnDgGwZpKJyIiZ1bhhDsrK8s5d6pr166cOHGCEydOlLuvN8yxqjX2rwDDDvUSITzO7GikhkveksbjX//OgWOOf7s39UjgkSvbEhGkL8+kRMcbICwG5vwV9i+Hty+HkZ9CgxZmRyZVdNNNN3H48GGmTJlCamoqXbp0Yf78+c6iZ/v27cN6yjSCPn36MHv2bB577DEeeeQRWrVqxdy5c51rcINjbe3c3FzuuOMOMjMz6devH/Pnzz9tDe5axS8QAiMhP9PRy62EW0REzqLCCXe9evU4dOgQUVFRREZGlltMqbSYWnFxsVuDrNP++NXx2LSvuXFIjbbvaB7Tvv3dWYE8PjKIZ67vyIWtGpkcmXilZv3gtgXw0Q2QsQvevBSGvQctLjE7MqmisWPHnnEI+cKFC0/bNmzYMIYNG3bG81ksFqZNm8a0adMqdP1mzZphGEaF9vVqYTGOhDs7FaLamh2NiIh4sQon3D/99JOzAvnPP//ssYDkT/aWJNzNlHCL6/IKi3h90W5eW7SLwiI7vlYLf7+wOfde2pIQLfclZxPVBv7+I8wZCQdXwf9d5yikdsFdmrNazZYsWcLrr7/Orl27+Oyzz4iPj+fDDz8kMTGRfv001cgUodFweCvkpJsdiYiIeLkKf+Lu379/uc/FgwrzIGWN43nTPubGIjVKsd3gs9X7eSFpO2lZBQD0bdmAqde0p2VUmMnRSY0RFgO3fgff/hPWz4YfJkHaJrjyBcewWvG4zz//nJtvvpmRI0eydu1aCgoc/56PHz/O008/zbxqWAdayhFWUtk9J9XcOERExOtVOOHesGFDhU/aqVOnSgUjf3JgJdiLIDzesQa3yDkYhsFPW9P59/ytbE/LASChfhAPDW7DlR1jta62uM4vEIb+F2I6woJHYd1HjkKOw97XvO5q8OSTTzJr1ixuueUW5syZ49zet2/fMtW/pZqVVirPTjv7fiIiUudVOOHu0qULFovFOU/7bDSH201Onb+tREnOojTRfil5BxsOHAcgIsiPey9tyc29mxLgq6W+pAosFuh9t2Ou6ud/h9SN8PpFcPVLjiJr4jHbtm3jootOXxIyIiKCzMzM6g9IHELVwy0iIhVT4YR7z549zudr167l/vvv54EHHnCux7l06VKef/55nn32WfdHWVf98ZvjUcPJ5QyK7QY/bknjlZ92svGgI9EO8vPhlj5Nubt/SyKCVX1c3KjFJXDnL/D57Y4vBD+/HfYugUFPg3+I2dHVSjExMezcuZNmzZqV2f7LL7/QvHlzc4KSk0PK1cMtIiLnUOGEu2nTk0Oahw0bxsyZMxkyZIhzW6dOnUhISGDy5MnO9TelCooKHEPKQRXK5TR5hUV8tvoA7/yyh71H8wAI9vfh5t5NGXNhcxqGBpgcodRa4bFwy9ew6BlY/B9Y/R7sWQxDZ0GTXmZHV+uMGTOG8ePH884772CxWEhJSWHp0qXcf//9TJ482ezw6q5Qx1Jq5CjhFhGRs6tUmeKNGzeSmJh42vbExEQ2b95c5aAEOLgGivIhpBE0bGV2NOIldh/OYc7K/Xyycj/HT9gACA/05W8XNOXvFzanfoi/yRFKneDjC5c+5vgycO7dkLEb3h0Mfe6Fix9RQTU3evjhh7Hb7Vx22WXk5eVx0UUXERAQwP3338+9995rdnh1lxJuERGpoEol3G3btmX69Om89dZb+Ps7PuAXFhYyffp02rbVepRu4Zy/3Ufzt+u4fFsxP/yeyscr9rFsd4Zze9MGwdzeL5HruzXWEl9ijhaXwN1LYf4kRxXzX1+CbfPhqhcca3lLlVksFh599FEeeOABdu7cSU5ODu3atSM0NNTs0Oq2sJKEuyDLsaKIf7C58YiIiNeq1Kf0WbNmcfXVV9O4cWNnRfINGzZgsVj45ptv3BpgnXVqwTSpc4qK7fy26yhfrUthwe+pZBcUAWC1wMWtoxhxfhMubROFj1VfxojJgiLhL69B26vhm/FwZBu8dyV0ugkuf+JkYiJV4u/vT7t27cwOQ0oFhINvEBSdcBROq6/59CIiUr5KJdznn38+u3fv5qOPPmLr1q0A3HTTTfz1r38lJESFc6qsuAj2LXc8V8JdZ+Tbivlt1xGSNqeTtDmVIzmFztfiI4MY1qMxN/ZIIC4yyMQoRc6gzRBocgH89ASsehc2fALbvoeLJ0HP28FXdQUq6rrrrqvwvl988YUHI5EzslgcXyYd2+sonKaEW0REzqDS41BDQkK444473BmLlDq0Hmy5EBgBUerRqK0Mw2Dv0Tx+3XmExdsPs2THEU7YTi6pVz/EnyEdY7i2Szzdm9TDqt5s8XbB9eGqF6Hr3+C7+yBlLfwwCZa9Bhc/DJ2Hg1VL1J1LRESE2SFIRYTGOBJuzeMWEZGzqHDC/fXXX3PFFVfg5+fH119/fdZ9r7nmmioHVqeVDidv0gesVnNjEbex2w12Hc5lWbqFxV9uYtnuYxzMPFFmn9iIQAa0jebydtH0btEAPx/9+UsNFN8d/p4Ma/8PFk6H4/vgq7sdc7wvfgjaXusovCblevfdd80OQSoiNMrxqIRbRETOosKfeIYOHUpqaipRUVFnXfbLYrFQXFx8xtelAkrX326m4eQ1lWEY7M84we8px/k9JYtNKcdZtz+TzDwb4AOkAODnY6Fbk3r0bdmQS9tE0T4uHIuK5EltYPWB7qOg042w4g345UXH/O7PboOIJnDBndD1ZggMNztSyE7FuvwNoo/bgSHn3N0M6enpbNu2DYDWrVsTFRVlckRyci3uVHPjEBERr1bhhNtut5f7XNzMbof9yxzPm/Q2NxY5p7zCIvZnnOCPo7nsPpLLzvQcdqTnsDs9x1no7FQBvlYaBxVzWZdE+raKomezegT7q6dPajG/IOg7Hrrf6hhavuINR4/3D4/AwmccCXmn4dC4R/WvyHBwjSOm37/Ex27jvJCWwKPVG8M5ZGVlcc899zBnzhznl9k+Pj7cdNNNvPrqqxp+biYtDSYiIhVg6if9f/3rX0ydOrXMttatWzsLsdVJR7bDiWOO6qexnc2Opk47UVhMenY+6dkFHM4uIPV4PoeOnyDleD4pmSc4cOwEh7MLzni8v4+V82JCaR8bQfv4cDo3jqRlwyB+XDCfIQPPw8/PrxrfjYjJAiMc87j7jncUVFv6quP/u5VvOX7qt3BUNm97laN2haeS78z9sPVb2PQ5HFjp3Gxv3Iudvj3pahieuW4ljRkzhrVr1/Ltt9/Su7fjS9ilS5cyfvx4/vGPfzBnzhyTI6zD1MMtIiIVUOmEOzk5mRdffJEtW7YAjrW5J0yYwIABA1w6T/v27fnxxx9PBuRbx3v7Snu3G/cAHyVk7lRQVExGbiFHcwodj7kFzueO3ws5mlPA0dxCjmQXkFtYsakR4YG+NG0QQtMGwbSKCqNVdCgto0Jp1iAEf9+yc7BtNpsn3ppIzeEX5Ojt7noL7P4Z1s9xJMAZu2Dh046f0Bhocaljne/47lAvsfL1LPKPO3qyD6yCbfMgZc3J16x+0OE66HUnxVEdOTRvHl29bErHt99+yw8//EC/fifXNR80aBBvvvkmgwcPNjEyIbQk4c5JNzcOERHxapXKbv/73/8yfvx4brjhBsaPHw/AsmXLGDJkCC+++CL33HNPxQPw9SUmJqYyYdROpcuBJfQyN44axDAMDmcXsP/YCQ4dP0Hq8XzHT1Y+h7MLOJLj6KHOyj99iPe5BPpZiQoLJCosgOjwQGIjAomNDCIuIpD4ekE0rR9CRLC+GBFxmdUKLS9z/BRkw5aSXue9vzjWNV4/2/ED4BcC0e0cPd9hsRDaCEKiHGuAGwYYdsdPYQ5kpcDxA47HtN8dveic2mttcUzXaXcNtP/LyV5KL/0yrEGDBuUOG4+IiKBevXomRCROzqJp6uEWEZEzq1TC/fTTT/Piiy8yduxY57Zx48bRt29fnn76aZcS7h07dhAXF0dgYCC9e/dm+vTpNGnSpNx9CwoKKCg4OYQ3KysLcPQaVrXnsPR4s3sgffctxQIUxffE8NIPgKWqu82y84vYkZ7DrsO57Dqcw+4jufxx9AQHM09QUFSxugK+Vgv1gv2oH+Jf9qdkW4MQfxqE+tMw1PE8NMD3nEXMXHn/3vL3rCZRm7muxrWZNRDa3+D4KcrHsn8Flj0/Y9n7C5bDW7DYch3Dv08ZAu4KI6IJRnw3jCZ9sbcecnLuLTgTbXe2mTvb/bHHHmPixIl8+OGHzi+nU1NTeeCBB5g8ebLbriOVUPplTe4RKC5S5X0RESlXpe4OmZmZ5Q5lGzhwIA899FCFz9OrVy/ee+89WrduzaFDh5g6dSoXXnghmzZtIiws7LT9p0+fftqcb4AFCxYQHBzs2ps4g6SkJLecpzICbJkMPrYHAws/bD5G0bZ5psXiCk+0mc0O+3Jgb7aF/bmOnyP5Z058LRjUC4B6/hDhbxBxymOYH4T5GYT5QbAvWCw2IK/sCU44foqPQDqOH08y8+9ZTaU2c13NbrOeENMTS3QxIQWpRJzYT2h+CgG24wQWZRFQdBzf4hM4/vVbMCwWiq0BnPCrT75/fU741SM3IJpjwS0o9CuphJ4GpK0+61Xd0WZ5eXnn3qmCXnvtNXbu3EmTJk2cX0bv27ePgIAADh8+zOuvv+7cd82aNWc6jZzit11HaBsTTr0Q/6qdKLghWHzAKIbcdAiPc0+AIiJSq1Qq4b7mmmv48ssveeCBB8ps/+qrr7jqqqsqfJ4rrrjC+bxTp0706tWLpk2b8r///Y/bb7/9tP0nTZrExIkTnb9nZWWRkJDAwIEDCQ+v2tIyNpuNpKQkLr/8ctOKWVm2fgObgKh2DLz6BlNicIU726yo2M7a/cdZsvMIK/ceY/2B49iKTy9eFBMeQItGobRoFELzRiE0axBMQr0gYiMCa8Sa1d7w96ymUZu5ri63WWXvBO5ss9LRV+5wtmU4xXUvJ+/g+aTt3NC9Mf8ZVsXCpFarY1h59iFH4TQl3CIiUo4KJ9wzZ850Pm/Xrh1PPfUUCxcudFZNXbZsGb/++iv33XdfpYOJjIzkvPPOY+fOneW+HhAQQEBAwGnb/fz83Pah0p3nctnBVQBYmlxQoz4kV7bNcguKSN6aTvKWNBZuO8zxE2WHYTYKC6Bns3p0jI+kQ3w47eMiqF/VHgkvYerfsxpKbeY6tZnr3NFm7mzzxx9/3G3nEujTsiEv/Lidz1Yf4Lqu8fRp2bBqJwyNdiTcKpwmIiJnUOGE+8UXXyzze7169di8eTObN292bouMjOSdd97hscceq1QwOTk57Nq1i5tvvrlSx9d4+5Y6Hmvx+ttFxXZ+2XmEL9ceZMHvaZywnawEHhnsR//zGtGnRQPOT2xAswbB55w/LSJSF6xevdq5Kkj79u3p2rWryRHVTN2b1uPmC5rywdI/mPTlRn6YcBGBfj6VP6FzLW4VThMRkfJVOOHes2eP2y9+//33c/XVV9O0aVNSUlJ4/PHH8fHxYcSIEW6/ltcrzIXUDY7nTWpfhfL07Hz+b9k+Pl6xr8za1YkNQxjUPoYBbaPokhCJbw0YFi4iUl3S09MZPnw4CxcuJDIyEnDUUbnkkkuYM2cOjRo1MjfAGuiBQa1Z8HsafxzNY2byDh4c3KbyJwsrSbiz09wTnIiI1DouZzc2m40WLVo4v2mvigMHDjBixAhat27NjTfeSIMGDVi2bFnd/ABxcDXYiyA8HiISzI7GbTYdPM7E/62j7zM/MTN5B4ezC6gf4s+o3k2Ze09ffrqvPw9f0YYezeor2RYR+ZN7772X7Oxsfv/9dzIyMsjIyGDTpk1kZWUxbtw4s8OrkcIC/Zh2bXsA3li8my2HqjDn3rkWt3q4RUSkfC4XTfPz8yM/P98tF58zZ45bzlMrnLr+di0YRr3lUBbPL9jOj1tOfuvfrUkkt/VLZFD7mBpR4ExExGzz58/nxx9/pG3bts5t7dq149VXX2XgwIEmRlazDWwfw+D2Mcz/PZWHv9jIF3f1wcdaiXuverhFROQcKlWl/J577uHf//43b731Fr6+WnfSLfYvczw2ucDcOKpo1+EcXkzazrcbDgFgtcBVneK4rV8iXRIizQ1ORKSGsdvt5RZh8/Pzw263mxBR7TH12vb8uvMI6/dn8uHSvdzaN9H1kzjncCvhFhGR8lUqW165ciXJycksWLCAjh07EhISUub1L774wi3B1Rn2Yti/wvG8hibceYVFvJS8g7eX7KHI7ljO66pOsUwYcB4to0JNjk5EpGa69NJLGT9+PB9//DFxcY5lpw4ePMg///lPLrvsMpOjq9miwwN56Io2PDZ3E8/9sI0hnWKJCgt07STOIeVKuEVEpHyVSrgjIyO5/vrr3R1L3ZW+BQqywD8UotqbHY1LDAOSNqfz5LytpBx3TDW4tE0U9w9sTbu4qq2NLiJS173yyitcc801NGvWjIQER32P/fv306FDB/7v//7P5Ohqvr+e34RPV+1n/YHjvJi0nenXdXLtBGGn9HAbRq2YEiYiIu5VqYT73XffdXccdVvpcmCNe4JPzRmifzSngLe3Wdm4bB0A8ZFBTL2mPQPaRZsbmIhILZGQkMCaNWv48ccf2bp1KwBt27ZlwIABJkdWO1itFiZf1Y4bZi3lk5X7uaV3M9rGuvBlcemQ8uJCOHEMgut7JlAREamxqpTdHT58mG3btgHQunXrulld3B32lxRMq0HDyRdtP8x9/1vHkRwrvlYLYy5qzr2XtiTYv+Z8YSAiUhNYLBYuv/xyLrzwQgICArCoF9WtejSrz5UdY/lu4yGe/G4z/3d7r4q3sW8ABNVzJNvZqUq4RUTkNJUqFZ2bm8ttt91GbGwsF110ERdddBFxcXHcfvvt5OXluTvG2q90/nbC+ebGUQEFRcU8+e1mRr2zgiM5hcQEGXx51wU8NLiNkm0RETez2+088cQTxMfHExoayp49ewCYPHkyb7/9tsnR1R4PX9EGfx8rv+48yk9b0107WIXTRETkLCqVcE+cOJFFixbxzTffkJmZSWZmJl999RWLFi3ivvvuc3eMtVtOOmT+AVggvrvZ0ZxV6vF8bnhtKW/94vjAN/L8BO7rWEybmDCTIxMRqZ2efPJJ3nvvPZ599ln8/f2d2zt06MBbb71lYmS1S0L9YEb3awbAU/O2YCt2oQK8Em4RETmLSiXcn3/+OW+//TZXXHEF4eHhhIeHM2TIEN58800+++wzd8dYux1Y6Xhs1AYCI8yN5SzW78/kmld+YePB49QL9uPNW3rwr6vb4u9jdmQiIrXXBx98wBtvvMHIkSPx8Tn5H27nzp2dc7rFPcZe0pIGIf7sPpzLR8v+qPiBYSWVyrNTPROYiIjUaJVKuPPy8oiOPr0wVlRUlIaUu8o5nLynuXGcxdfrU7jx9aWkZxdwXnQoX4/tx+UqjCYi4nEHDx6kZcuWp2232+3YbDYTIqq9wgL9mDjwPABmJO/geF4F21c93CIichaVSrh79+7N448/Tn5+vnPbiRMnmDp1Kr1793ZbcHVCaQ93Y++bv20YBi8n72Dcx2spKLJzaZsoPr+rDwn1g80OTUSkTmjXrh1Lliw5bftnn31G165dTYiodrupRwKJDUPIzLPxy84jFTtIPdwiInIWlapy9dJLLzFo0CAaN25M586dAVi/fj2BgYH88MMPbg2wViu2wcE1judeVjDNMAye+X4rry/eDcAdFzXnocFt8LGqOq6ISHWZMmUKo0aN4uDBg9jtdr744gu2bdvGBx98wLfffmt2eLWOr4+VdnHh7DmSS1pW/rkPgFN6uF0stiYiInVCpRLuDh06sGPHDj766CPnHLIRI0YwcuRIgoKC3BpgrZb2OxSdcMzdbtDK7Gic7HaDf33zOx8sdcxhm3JVO27rl2hyVCIidc+1117LN998w7Rp0wgJCWHKlCl069aNb775hssvv9zs8GqlqLAAANKzCyp2gDPhVg+3iIicrtLrOAUHBzNmzBh3xlL3lA4nj+8B1kqN7ne7YrvBpC828L9VB7BY4KmhHflrryZmhyUiUmddeOGFJCUlmR1GnREVFghAekV7uJ1DyjWHW0RETlfphDslJYVffvmF9PR07Payy2eMGzeuyoHVCV62/rZhGDz0+QY+W30AqwX+M6wz13VrbHZYIiJ13qpVq9iyZQvgmNfdvbt3LyNZk1W6h7swGwpzwT/EQ5GJiEhNVKmE+7333uMf//gH/v7+NGjQAIvl5Lxei8WihLuiDpQk3I29o0L5M99v5bPVB/CxWpg5vCtXdoo1OyQRkTrtwIEDjBgxgl9//ZXIyEgAMjMz6dOnD3PmzKFxY30p6m5R4aUJdwV7uAPCwC8YbHmQsg4i3PRnUlREUMFhyNwHvqd8XAtpBP4qXioiUlNUKuGePHkyU6ZMYdKkSVi9ZCh0jZNzGI7tBSzQuIfZ0fDm4t3OAmnTr+uoZFtExAv8/e9/x2azsWXLFlq3bg3Atm3bGD16NH//+9+ZP3++yRHWPs4h5RXt4bZYIDTKcU9/b4jb4vADBgJs/tMLQfVh3FoIinTbtURExHMqlXDn5eUxfPhwJdtVUTp/u1EbR9E0E32x5gBPzXMMVXxocBtu7JFgajwiIuKwaNEifvvtN2eyDdC6dWtefvllLrzwQhMjq72iS3q4M/NsFBQVE+Drc+6DuvwNfp0Bhv2cu1aUARQXF+Hj44tzHKEtD05kQMZuiO/mtmuJiIjnVCrhvv322/n00095+OGH3R1P3eEcTm5u7/ai7Yd58LMNANzeL5E7+zc3NR4RETkpISEBm8122vbi4mLi4uJMiKj2iwjyw9/XSmGRncPZBTSuV4Hh2/0fcPy4UZHNxrx58xgyZAh+fn6OjS/3gKM7wHbCrdcSERHPqVTCPX36dK666irmz59Px44dT94ISrzwwgtuCa5W21/Sw21iwbS9R3K5d/YaiuwGQ7vE8eiQtmXm44uIiLmee+457r33Xl599VV69HB8Qbtq1SrGjx/Pf/7zH5Ojq50sFguNQgM4mHmC9Iom3NXFr2TpVSXcIiI1RqUT7h9++ME5xO3PRdPkHIqLIGWN43ljcxLu3IIi7vhwFVn5RXRrEsm/b+iE1ao/OxERb3LrrbeSl5dHr1698C0pnFVUVISvry+33XYbt912m3PfjIwMs8KsdaLCSxLurArO464ufiXJvy3P3DhERKTCKpVwP//887zzzjvceuutbg6njkjb5LhZBkZAw/Oq/fKGYfDAZ+vZnpZDVFgAs/7WvWJz1EREpFrNmDHD7BDqpNKlwQ5XtFJ5dXH2cCvhFhGpKSqVcAcEBNC3b193x1J3lBZMi+8BJhSe++/CXczbmIqfj4XX/taNqPDAao9BRETObdSoUWaHUCeVVipP87Ye7tI1vpVwi4jUGJXK9saPH8/LL7/s7ljqjoOrHY8mFExbsuMw/1mwDYCp13Sge9P61R6DiIiINyvt4a7wWtzVRXO4RURqnEol3CtWrOD999+nefPmXH311Vx33XVlfuQcDqxyPMZXb8KdkVvIxP+txzBgxPkJ/LVXk2q9voiImOvVV1+lWbNmBAYG0qtXL1asWHHW/T/99FPatGlDYGAgHTt2ZN68eWVeNwyDKVOmEBsbS1BQEAMGDGDHjh1l9rnmmmto0qQJgYGBxMbGcvPNN5OSkuL29+ZOUeGlCbeX9XBrSLmISI1TqYQ7MjKS6667jv79+9OwYUMiIiLK/MhZnMh0LOkBEN+92i5rGAYPf76Bw9kFtIwK5fGr21fbtUVExHyffPIJEydO5PHHH2fNmjV07tyZQYMGkZ6eXu7+v/32GyNGjOD2229n7dq1DB06lKFDh7Jp0ybnPs8++ywzZ85k1qxZLF++nJCQEAYNGkR+/sme4UsuuYT//e9/bNu2jc8//5xdu3Zxww03ePz9VkXpkHLvLZqmHm4RkZqiUnO43333XXfHUXeUViev1wxCGlTbZT9ZuZ8Fm9Pw87Hw0vAuBPqpSJqIiDfasGEDHTp0wOrmGh8vvPACY8aMYfTo0QDMmjWL7777jnfeeYeHH374tP1feuklBg8ezAMPONaXfuKJJ0hKSuKVV15h1qxZGIbBjBkzeOyxx7j22msB+OCDD4iOjmbu3LkMHz4cgH/+85/OczZt2pSHH36YoUOHYrPZTltW1Ft4fw+3Em4RkZqiUgl3qcOHD7Ntm2M+cOvWrWnUqJFbgqrVSudvV2Pv9p4juUz9ZjMADwxqTfs4jUIQEfFWXbt25dChQ0RFRdG8eXNWrlxJgwZV+4K2sLCQ1atXM2nSJOc2q9XKgAEDWLp0abnHLF26lIkTJ5bZNmjQIObOnQvAnj17SE1NZcCAAc7XIyIi6NWrF0uXLnUm3KfKyMjgo48+ok+fPmdMtgsKCigoOJnoZmVlAWCz2bDZbBV7w1VUP8jxpfTR3AJO5Bfg61P9BU5L3+up79lqDcAHKC7IwV5NbVHblNeuUjVqU89Qu7qfO9vUlXNUKuHOzc3l3nvv5YMPPsButwPg4+PDLbfcwssvv0xwcHBlTls3HCzp4a6m+du2YjsT5qzlhK2YPi0a8Pd+zavluiIiUjmRkZHs2bOHqKgo9u7d67zPVsWRI0coLi4mOjq6zPbo6Gi2bt1a7jGpqanl7p+amup8vXTbmfYp9dBDD/HKK6+Ql5fHBRdcwLfffnvGWKdPn87UqVNP275gwYJq+3xhN8CKD3bDwqffzCfCv1ouW66kpCTn8xZpf9ABSNm7kzV/mk8vrjm1XcU91KaeoXZ1P3e0aV5exWtpVCrhnjhxIosWLeKbb75xLg/2yy+/MG7cOO677z5ee+21ypy29jOMUwqmVU8P9xuLd7P+wHEigvx4/sbOWK2WarmuiIhUzvXXX0///v2JjY3FYrHQo0cPfHzKnwa0e/fuao6uch544AFuv/12/vjjD6ZOncott9zCt99+i8Vy+j1p0qRJZXrWs7KySEhIYODAgYSHh1dbzM/8voi07AI69uxHh/jqu24pm81GUlISl19+uXM0gHV1GqR8THxUPWKGDKn2mGqD8tpVqkZt6hlqV/dzZ5uWjr6qiEol3J9//jmfffYZF198sXPbkCFDCAoK4sYbb1TCfSbHD0BuOlh9IbaTxy+390guM5MdBdr+dU07YiOCPH5NERGpmjfeeIPrrruOnTt3Mm7cOMaMGUNYWFiVztmwYUN8fHxIS0srsz0tLY2YmJhyj4mJiTnr/qWPaWlpxMbGltmnS5cup12/YcOGnHfeebRt25aEhASWLVtG7969T7tuQEAAAQEBp2338/Or1g+dUeGBpGUXkHGiyNQPu2Xed6Dj74G1KB+rPoBXSXX/faoL1KaeoXZ1P3e0qSvHVyrhzsvLO20IGUBUVJRL3et1zsGS3u3o9icLn3iIYRg8OncjBUV2LmzVkKFd4j16PRERcZ/BgwcDsHr1asaPH1/lhNvf35/u3buTnJzM0KFDAbDb7SQnJzN27Nhyj+nduzfJyclMmDDBuS0pKcmZJCcmJhITE0NycrIzwc7KymL58uXcddddZ4yldIj8qfO0vVHpWtxp3lSpXEXTRERqnEol3L179+bxxx/ngw8+IDDQsXTGiRMnmDp1arnfVkuJaiyY9uXag/y68ygBvlaeHNqh3GF7IiLi3U5dFeTAgQMANG7cuFLnmjhxIqNGjaJHjx6cf/75zJgxg9zcXGfV8ltuuYX4+HimT58OwPjx4+nfvz/PP/88V155JXPmzGHVqlW88cYbAFgsFiZMmMCTTz5Jq1atSExMZPLkycTFxTmT+uXLl7Ny5Ur69etHvXr12LVrF5MnT6ZFixZe/3nhZKXy/HPsWY2cy4Kpc0NEpKaoVMI9Y8YMBg8eTOPGjencuTMA69evJyAggAULFrg1wFrFWTDNswl3Rm4hT363BYBxl7WiaYMQj15PREQ8w2638+STT/L888+Tk5MDQFhYGPfddx+PPvqoS0uH3XTTTRw+fJgpU6aQmppKly5dmD9/vnPE2r59+8qcr0+fPsyePZvHHnuMRx55hFatWjF37lw6dOjg3OfBBx8kNzeXO+64g8zMTPr168f8+fOdX8YHBwfzxRdf8Pjjj5Obm0tsbCyDBw/mscceK3fYuDdpVLoWtzctDaYebhGRGqdSCXfHjh3ZsWMHH330kbO66YgRIxg5ciRBQZonXK7iIkhZ63ju4QrlT8/bQkZuIa2jw7jjIlUlFxGpqR599FHefvttnnnmmTJFSv/1r3+Rn5/PU0895dL5xo4de8Yh5AsXLjxt27Bhwxg2bNgZz2exWJg2bRrTpk0r9/WOHTvy008/uRSjt4gu7eH2qiHlpT3cSrhFRGqKSiXc06dPJzo6mjFjxpTZ/s4773D48GEeeughtwRXqxze6hgC5h8GDVt57DIr92bw2eoDWCzw9HUd8TNh7VAREXGP999/n7feeotrrrnGua1Tp07Ex8dz9913u5xwS8VFlfRwH/aqIeWlPdy55sYhIiIVVqls7PXXX6dNmzanbW/fvj2zZs2qclC1Uun87bguYC1/eZeqstsNnvx2MwA39Uige9N6HrmOiIhUj4yMjHLvt23atCEjI8OEiOqO0qJpXjWk3F893CIiNU2lEu7U1NQyS4CUatSoEYcOHapyULVSaYXyxp4bTv71+hTWHzhOiL8P9w1s7bHriIhI9ejcuTOvvPLKadtfeeUVZw0V8YzSommHswuw2w2ToylxatE0w0tiEhGRs6rUkPKEhAR+/fVXEhMTy2z/9ddfiYuLc0tgtY6HC6bl24p5dr5jPv3dl7SkUZh3F6MREZFze/bZZ7nyyiv58ccfnVW9ly5dyv79+5k3b57J0dVuDUMDsFigyG5wLK+QBqFecF89dUnRonyPLzEqIiJVV6mEe8yYMUyYMAGbzcall14KQHJyMg8++CD33XefWwOsFQpyIN0x1NtTCffbv+wh5Xg+cRGB3N4v8dwHiIiI1+vfvz/bt2/n1VdfdRYpve6667j77rv1BbeH+flYqR/sz9HcQtKzC7wj4fY9JcG2nVDCLSJSA1Qq4X7ggQc4evQod999N4WFhQAEBgby0EMPMWnSJLcGWCscWg+GHcLiINz9H5DSs/P57887AXjoijYE+nlmjriIiFS/uLg4FUczSaOwAGfC3fb0mXTVz8cXfPyhuLBkLe76ZkckIiLnUKmE22Kx8O9//5vJkyezZcsWgoKCaNWqldevqWmalNLh5N08cvoXk7aTW1hM54RIru6kHg8RERF3iAoPZGtqNmlZXlapvLgQCvPMjkRERCqgUgl3qdDQUHr27OmuWGqv0vW347q6/dQ703P4ZOV+ACZf2Rar1eL2a4iIiNRFpZXKD3tTpXK/EMg/XtLDLSIi3s5rFml+5plnsFgsTJgwwexQ3M+DCffM5B3YDbi8XTQ9mmlomYiIiLtEl1QqT/e2Hm7Q0mAiIjWEVyTcK1eu5PXXX6dTp05mh+J+J45Bxm7Hczcn3DvSsvlmQwoAEwa0cuu5RURE6rqosEDAy9biPnVpMBER8XqmJ9w5OTmMHDmSN998k3r16pkdjvsdWu94jGwKwe7tgX4peQeGAYPaR9M+LsKt5xYREe9y5MgRvvvuO77++msOHTpkdjh1QumQcu9KuNXDLSJSk1RpDrc73HPPPVx55ZUMGDCAJ5988qz7FhQUUFBw8qaXlZUFgM1mw2azVSmO0uOrep4/s+5fjQ9gj+1CsRvPvSMth+82Oj5wjb24udvjrghPtVltpjZzndrMdWoz17mzzTzR7p9//jm333475513HjabjW3btvHqq68yevRot19LTooqHVKerSHlIiJSOaYm3HPmzGHNmjWsXLmyQvtPnz6dqVOnnrZ9wYIFBAcHuyWmpKQkt5ynVI893xMPbDkeyM5589x23ne3WzEMK13q29m9Zgm73XZm17m7zeoCtZnr1GauU5u5zh1tlpdX9aG+OTk5hIaGOn+fOnUqK1as4LzzzgPgu+++Y8yYMUq4Pcw5pDyrAMMwsFi8oDCphpSLiNQopiXc+/fvZ/z48SQlJREYGFihYyZNmsTEiROdv2dlZZGQkMDAgQMJDw+vUjw2m42kpCQuv/xy/Pz8qnSuU/m+8hgArS8ZznnNLnTLObelZrNu6VIsFnjqr305LzrMLed1lafarDZTm7lObeY6tZnr3NlmpaOvqqJ79+48++yzXHvttQD4+vqSnp7uTLjT0tLw9/ev8nXk7BqVDCkvKLKTlV9ERJAX/Hty9nAr4RYRqQlMS7hXr15Neno63bqdXJu6uLiYxYsX88orr1BQUICPj0+ZYwICAspd69vPz89tHyrdeS5yj8LxfQD4JnQHN5331UV7ABjSMZb2jc2vTO7WNqsj1GauU5u5Tm3mOne0mTva/IcffuCee+7hvffe49VXX+Wll17ipptuori4mKKiIqxWK++9916VryNnF+jnQ3igL1n5RRzOzveOhNtfPdwiIjWJaQn3ZZddxsaNG8tsGz16NG3atOGhhx46LdmukUqXA2vQEgLdU9RsZ3oO329KxWKB8ZepMrmISG3UrFkzvvvuOz7++GP69+/PuHHj2LlzJzt37qS4uJg2bdpUeHSYVE1UeCBZ+TmkZRXQMsqcEWVlOIeUaw63iEhNYFqV8rCwMDp06FDmJyQkhAYNGtChQwezwnIvD6y//fYvjtnaA9pGmzaUXEREqseIESNYuXIl69ev5+KLL8Zut9OlSxcl29Uo2tsKp6lomohIjWJ6lfJazc0J9+HsAj5fcxCAf1zU3C3nFBER7zRv3jy2bNlC586deeutt1i0aBEjR47kiiuuYNq0aQQFBZkdYp1wauE0r6CiaSIiNYrp63CfauHChcyYMcPsMNzHzQn3B0v3Ulhkp2uTSLo3rYVrlouICAD33Xcfo0ePZuXKlfzjH//giSeeoH///qxZs4bAwEC6du3K999/b3aYdYLXrcWtHm4RkRrFqxLuWiU7FbJTAAvEdKry6fIKi/hw2R8A3HFhc+9YmkRERDzivffeY968ecyZM4eVK1fy4YcfAuDv788TTzzBF198wdNPP21ylHVDTISjh/vAMS/pUVYPt4hIjaKE21NS1jkeG7WGgNCz7loRn646QGaejaYNghnYPqbK5xMREe8VEhLCnj2OFSn2799/2pztdu3asWTJEjNCq3NK66VsS802OZISpT3chUq4RURqAiXcnuLG4eTFdoO3Soql/b1fIj5W9W6LiNRm06dP55ZbbiEuLo7+/fvzxBNPmB1SndUmxpFw/5GRR25BkcnRoCrlIiI1jIqmeYobE+4ffk9lf8YJ6gX7cUP3hCqfT0REvNvIkSMZPHgwu3fvplWrVkRGRpodUp3VIDSAqLAA0rML2J6WTdcmJtdQ0ZByEZEaRT3cnmAYbk2431zi6N2+uXczgvxrwfrkIiJyTg0aNKBnz55Ktr1Am9hwALZ6w7ByFU0TEalRlHB7QlYK5KaDxQeiq7am+KaDx1m7LxM/Hws3X9DUTQGKiIhIRbUtGVa+9VCWyZGgHm4RkRpGCbcnHFrneIxqC/7BVTrV/5VUJr+iQyyNSpYmERERkerTJtaRcG85pB5uERFxjRJuTzi03vEY26VKpzl+wsbcdQcBuLm3erdFRETM0CbGMaR8S2oWhmGYG4wSbhGRGkUJtyc4E+7OVTrNF2sOkG+z0zo6jB5NTS7SIiIiUke1aBSKn4+F7PwiUo7nmxuMf4jj0ZbrqBkjIiJeTQm3JzgT7k6VPoVhGM7h5H+7oAkWi5YCExERMYO/r5UWjUIBL5jHXdrDbdihuNDcWERE5JyUcLtbdhpkHwIsVSqYtnT3UXYdziXE34ehXePdF5+IiIi4rK23VCr3O6U2jAqniYh4PSXc7pa6wfHYsBUEhFb6NB8t2wfA0K7xhAX6uSMyERERqaQ2JZXKN5vdw+3jB1Zfx3PN4xYR8XpKuN2ttEJ5FeZvp2fl88PvqQD8TUuBiYiImM65FrfZCTecsjSYEm4REW+nhNvdDpX0cFch4Z6zcj9FdoPuTes5h7CJiIiIeUrX4t5zJJd8W7G5wTgrlWtIuYiIt1PC7W5VrFButxt8snI/4CiWJiIiIuZrFBZAgxB/7AbsSMsxN5jShLtQCbeIiLdTwu1OJ45BpqOyODGVq1C+bM9RDmaeICzAlys6xLoxOBEREaksi8VCm1hHL/eWVLMrlZcuDaaEW0TE2ynhdqfS4eT1mkFQZKVO8fnqgwBc1TmWQD8f98QlIiIiVdYmpnQet9mVykuHlGsOt4iIt1PC7U6lw8kr2budW1DE95sOAXBD98buikpERETcoLRS+RazC6dpDreISI2hhNudqjh/+/tNqeQVFpPYMIRuTeq5MTARERGpqpNrcWdhGIZ5gahKuYhIjaGE252cCXeXSh3+2WpHsbTru8VjsVjcFJSIiIi4Q8uoUHysFo7l2UjPLjAvEA0pFxGpMZRwu0tBNhzd6Xge6/qQ8v0ZeSzbnYHFAn/ppuHkIiIi3ibQz4fmDR0Fy0wdVu7s4daQchERb6eE211SNwEGhMVBaJTLh3+xxlEsrW+LhsRHBrk5OBEREXGHNs5h5SYWTtMcbhGRGkMJt7ukllQor8T8bcMw+HzNAQCu7x7vzqhERETEjUoLp201s4fbXz3cIiI1hRJud6lCwbSVe4+xLyOP0ABfBrWPcXNgIiIi4i5tS9fiNnNpMBVNExGpMZRwu0sVEu4vSnq3r+wYS7C/rzujEhERETc6L9qRcO8+koPdblKlchVNExGpMZRwu4MtH9K3OJ67mHAXFtn5flMqANd2jXN3ZCIiIuJGMeGB+Fgt2IoNDueYVKlcRdNERGoMJdzukL4ZjGIIbgDhriXNv+48wvETNhqFBdArsYGHAhQRERF38PWxEhMeCMCBYyb1MKuHW0SkxlDC7Q6pGx2PMR3BxfWzv9mQAsCQDjH4WLX2toiIiLcrXU0kJdOshFs93CIiNYUSbnc4NeF2Qb6tmKTf0wC4qrOGk4uIiGe9+uqrNGvWjMDAQHr16sWKFSvOuv+nn35KmzZtCAwMpGPHjsybN6/M64ZhMGXKFGJjYwkKCmLAgAHs2LHD+frevXu5/fbbSUxMJCgoiBYtWvD4449TWFjokfdXXeLrORLug2Yn3IVKuEVEvJ0SbndwJtydXDps8fbDZBcUERsRSPcm9TwQmIiIiMMnn3zCxIkTefzxx1mzZg2dO3dm0KBBpKenl7v/b7/9xogRI7j99ttZu3YtQ4cOZejQoWzatMm5z7PPPsvMmTOZNWsWy5cvJyQkhEGDBpGfnw/A1q1bsdvtvP766/z++++8+OKLzJo1i0ceeaRa3rOnlPZwH9SQchEROQcl3FVlt0NayYcPF3u4v9lwCHBUJ7dqOLmIiHjQCy+8wJgxYxg9ejTt2rVj1qxZBAcH884775S7/0svvcTgwYN54IEHaNu2LU888QTdunXjlVdeARy92zNmzOCxxx7j2muvpVOnTnzwwQekpKQwd+5cAAYPHsy7777LwIEDad68Oddccw33338/X3zxRXW9bY+I05ByERGpIK1BVVXH9kBhDvgEQINWFT7sRGExyVs0nFxERDyvsLCQ1atXM2nSJOc2q9XKgAEDWLp0abnHLF26lIkTJ5bZNmjQIGcyvWfPHlJTUxkwYIDz9YiICHr16sXSpUsZPnx4uec9fvw49evXP2OsBQUFFBScrP6dlZUFgM1mw2aznf2NVpOYMD8ADhzL81hMpect9/wWP/wAw5ZHkZe0SU1x1naVSlGbeoba1f3c2aaunEMJd1WVDiePbgc+FW/On7amk1dYTEL9IDo3jvBQcCIiInDkyBGKi4uJjo4usz06OpqtW7eWe0xqamq5+6empjpfL912pn3+bOfOnbz88sv85z//OWOs06dPZ+rUqadtX7BgAcHBwWc8rjql5gH48seR7NPmtbtbUlLSadtC8g8xACg64fnr11bltatUjdrUM9Su7ueONs3Lq/gIIyXcVVU6nDy6g0uHfVtSnfzKjnFYXKxsLiIiUtMcPHiQwYMHM2zYMMaMGXPG/SZNmlSmZz0rK4uEhAQGDhxIeHh4dYR6TnmFRUxf/xP5xRYuvPRywgL93H4Nm81GUlISl19+OX5+fzp/VgpseQhfw8aQIUPcfu3a7KztKpWiNvUMtav7ubNNS0dfVYQS7qqqRMG0nIIiftrqKFJzVadYT0QlIiLi1LBhQ3x8fEhLSyuzPS0tjZiYmHKPiYmJOev+pY9paWnExsaW2adLly5ljktJSeGSSy6hT58+vPHGG2eNNSAggICAgNO2+/n5ec2Hzgg/P+qH+JORW0haThH1wzzX817u+w4KA8Bit+FnBXy8o11qEm/6+1RbqE09Q+3qfu5oU1eOV9G0qqrEkmDJW9IoKLKT2DCE9nHe8W29iIjUXv7+/nTv3p3k5GTnNrvdTnJyMr179y73mN69e5fZHxzD8Er3T0xMJCYmpsw+WVlZLF++vMw5Dx48yMUXX0z37t159913sVprx0ePuMhAwKRK5f4hJ5+rcJqIiFdTD3dV5B6FrIOO59HtK3zY/E2OuW1XdozVcHIREakWEydOZNSoUfTo0YPzzz+fGTNmkJuby+jRowG45ZZbiI+PZ/r06QCMHz+e/v378/zzz3PllVcyZ84cVq1a5eyhtlgsTJgwgSeffJJWrVqRmJjI5MmTiYuLY+jQocDJZLtp06b85z//4fDhw854ztSzXlPERwax6WAWKcdNSLh9/MFiBcPuWBosULVgRES8lRLuqkgr6d2ulwiBFeupzrcVs2i74wPHoPY1+8OGiIjUHDfddBOHDx9mypQppKam0qVLF+bPn+8serZv374yvc99+vRh9uzZPPbYYzzyyCO0atWKuXPn0qHDyZolDz74ILm5udxxxx1kZmbSr18/5s+fT2Cgo/c3KSmJnTt3snPnTho3blwmHsMwquFde058pGMYuSk93BaLY2mwwhz1cIuIeDkl3FVRieHkv+06Ql5hMbERgXSI13ByERGpPmPHjmXs2LHlvrZw4cLTtg0bNoxhw4ad8XwWi4Vp06Yxbdq0cl+/9dZbufXWWysTqtdzDik3bS3uoJKE26Tri4hIhdSOiVRmqUTBtAW/OwrQXN4uWsPJRUREaqjG9YIAkxNuUMItIuLllHBXhYs93MV2gx+3OBLuge00nFxERKSmiossSbjNGFIOjiHloCHlIiJeTgl3Zdny4ch2x/OYiq3BvW7/MY7kFBIW6Euv5vU9GJyIiIh4UnxJwp2eXUBBUXH1B1Daw12ohFtExJuZmnC/9tprdOrUifDwcMLDw+nduzfff/+9mSFV3OGtYC+CoHoQHl+hQxZsdvRuX9omCj8ffdchIiJSU9UP8SfQz3EvTz2eX/0B+JUsDaYebhERr2Zq1te4cWOeeeYZVq9ezapVq7j00ku59tpr+f33380Mq2JOHU5egbnYhmGUmb8tIiIiNZfFYjF3WLnmcIuI1AimJtxXX301Q4YMoVWrVpx33nk89dRThIaGsmzZMjPDqhgXC6btOpzDniO5+PtY6X9eIw8GJiIiItWhdFi5KYXTnAm3erhFRLyZ1ywLVlxczKeffkpubi69e/cud5+CggIKCgqcv2dlZQFgs9mw2WxVun7p8RU9j8+h9ViBokbtMCpwzPyNhwDo3bw+gT4Vv443c7XNRG1WGWoz16nNXOfONlO71x2mVip3Fk1TD7eIiDczPeHeuHEjvXv3Jj8/n9DQUL788kvatWtX7r7Tp09n6tSpp21fsGABwcHBboknKSnp3DsZBkMOOhLuxduPkb1/3jkP+XSjD2AhujiNefPOvX9NUqE2kzLUZq5Tm7lObeY6d7RZXp56HOuKuAgNKRcRkbMzPeFu3bo169at4/jx43z22WeMGjWKRYsWlZt0T5o0iYkTJzp/z8rKIiEhgYEDBxIeHl6lOGw2G0lJSVx++eX4+fmdfefMffitO4Fh9ePCobeBz9n3T8vK54+liwEYf8OlRIUFVClWb+FSmwmgNqsMtZnr1Gauc2eblY6+ktovvqSHO+W4mT3c+oJHRMSbmZ5w+/v707JlSwC6d+/OypUreemll3j99ddP2zcgIICAgNOTVT8/P7d9qKzQuY5uA8DSqA1+gefuWV+80zGcvEtCJPH1Q6sco7dxZ/vXFWoz16nNXKc2c5072kxtXnd4R9E0JdwiIt7M69amstvtZeZpe6W0kirq0e0rtPvCbekAXNYmylMRiYiISDUrLZqWcjwfu92o3ov7q4dbRKQmMLWHe9KkSVxxxRU0adKE7OxsZs+ezcKFC/nhhx/MDOvc0jY5HiuQcBcW2fl15xEALm6thFtERKS2iIkIxGpx3OuP5BYQFRZYfRdX0TQRkRrB1IQ7PT2dW265hUOHDhEREUGnTp344YcfuPzyy80M69xc6OFe9UcGuYXFNAz1p31c1eaZi4iIiPfw87ESHR7IoeP5HDx2opoTbhVNExGpCUxNuN9++20zL185hXmQscvxPLrDOXdftO0wABed1wir1eLJyERERKSaxUcGceh4PimZ+XRtUo0XVtE0EZEawevmcHu9w1vBsENwQwg99xDxhSUJt4aTi4iI1D7OwmmZ1Zz4qodbRKRGUMLtqlOHk1vO3mOdknmCbWnZWC1wUauG1RCciIiIVKfSpcGqvVJ5acJdqB5uERFvpoTbVc6EuwLDybc7ere7JEQSGezvyahERETEBPHOHu786r2wX4jjUUPKRUS8mhJuV7lQobx0OTANJxcREamdTibcJvVwa0i5iIhXU8LtCsOocIVyx3JgRwG4uHUjT0cmIiIiJjg5pLy653CraJqISE2ghNsV2alwIgMsPtCozVl3Xf3HMXIKimgY6k+HuIhqClBERESqU2nRtKz8IrLzbdV3YfVwi4jUCEq4XVHau92wFfidfa3Nhdsdw8kvaqXlwERERGqr0ABfIoL8AEipznncpT3cxQVgL66+64qIiEuUcLvChfnbpetv99dwchERkVotob6jt3nPkZzqu2hpDzeol1tExIsp4XZFBedvpx7PZ2tq6XJgSrhFRERqs47xjqlj6w8cr76L+p4y0k7zuEVEvJYSbldUcEmwJTscvdudGkdSL0TLgYmIiNRmnRtHArBuX2b1XdRqVeE0EZEaQAl3RRUVwpFtjufn6OH+decRAPq1bOjpqERERMRkXZpEArDx4HGK7Ub1XViF00REvJ4S7oo6sh3sRRAYAeHxZ9zNMAx+3eVYDqxPywbVFZ2IiIiYpFVUGMH+PuQUFLH7cHXO41YPt4iIt1PCXVGnDie3nLnq+M70HA5nFxDga6Vbk3rVFJyIiIiYxcdqoUPJPO61+zOr78Lq4RYR8XpKuCuqghXKS4eT92xWn0A/H09HJSIiIl6gS0IkAOuVcIuIyCmUcFdUBSuUazi5iIhI3eNMuA9kVt9FNaRcRMTrKeGuqNKEO+rMCXdRsZ1lux0Jd98WKpgmIiJSV3QuSbi3Hsom31ZcPRct7eEuVMItIuKtlHBXRF4G5KQ6nke1PeNum1KyyM4vIjzQ1zmXS0RERGq/uIhAGoYGUGQ3+D2lmtbjVg+3iIjX8zU7gBohfbPjMbIpBISecbfS+dsXNG+Aj/XMhdVERESkdrFYLHRJiOTHLWms23+c7k3re/6ipQl3+hb447eKH+fj7+hA8A/xTFwiIuKkhLsi0rc4HqPanXW30oS7r9bfFhERqXO6JETw45a06iuc5l+ScK980/HjCovVMU2ucXdo3BPaXQsBYe6PUUSkjlPCXRGlPdxnGU6ebytm1R/HAOirgmkiIiJ1Tuk87nXVlXB3Gg4H17hepbwg2zFVLm2j42f1e/DbK3D7AggM90ioIiJ1lRLuikgrTbjP3MO9+o9jFBbZiQoLoEWjMw87FxERkdqpU+NIAPZl5JGRW0j9EH/PXrBpb7hzSeWOzUqBA6vgwEpY/zEc3gJfjIHhs8GqZU1FRNxFRdPOxTBODimPPnPCfepwcotF87dFRETqmoggP5o3csyLrtblwSojPA7aXQMDn4ARn4BPAGyfD8nTzI5MRKRWUcJ9LlkpUHAcrL7QoNUZd3Ouv91Cw8lFRETqqi4lvdzr9mWaGodLGneHa191PP91Bqz/xNRwRERqEyXc51Lau92gJfiWPzQsK9/GxpJvslUwTUREpO4qncft9T3cf9ZpGPT7p+P51/c6hpuLiEiVKeE+l/TfHY9nKZi2am8GdgOaNQgmLjKomgITERERb9OlNOHen4lhGOYG46pLp8B5V0BxAcwZ6RjlJyIiVaKE+1wqsCTY8j0ZAPRK1HByERGRuqxNbBj+PlaO5dnYl5FndjiusVrh+jehUVtHFfM5f3W9ArqIiJShhPtc0s9doXxFScLdM7F+dUQkIiIiXirA14e2cY6ltapteTB3CgiDER9DUH1IWQtfjXUUkBURkUpRwn029mI4vM3x/AxDyk8UFrPxwHEAeinhFhERqfO6lgwr/3bDoZo3rBygfiLc+IGjYOymz+CXF82OSESkxlLCfTbH9kJRPvgGQb1m5e6ydt8xiuwGsRGBNK6n+dsiIiJ13Q3dG+NrtZC0OY3ZK/aZHU7lJF4IVzzreJ48DbZ9b248IiI1lBLus0krKZjWqDVYfcrdpXT+9vmJ9bX+toiIiNAhPoKHBrcBYOo3m9lyKMvkiCqp5+3Q43bAgM//Dkd3mR2RiEiNo4T7bEoLpkW3P+MuK05JuEVEREQAbu+XyCWtG1FYZGfs7DXkFhSZHVLlXPFvaHw+FObAxs/MjkZEpMZRwn02zoJp5c/fLiyys3b/MUDzt0VEROQkq9XC8zd2ITo8gF2Hc5ny1e9mh1Q5Pn7Q8QbH8wMrzY1FRKQGUsJ9Ns4lwcpPuDcePE6+zU79EH9aNAqtxsBERETE29UP8Wfm8K5YLfD5mgN8vvqA2SFVTuMejscDK1WxXETERUq4z6SoAI7udDw/w5JgzuXAmtXT/G0RERE5Ta/mDRh/2XkATP5qE7sO55gcUSVEdwSfAMjP1DxuEREXKeE+kyPbwSiGwAgIiy13lxV7jgJwfmKD6oxMRESkUl599VWaNWtGYGAgvXr1YsWKFWfd/9NPP6VNmzYEBgbSsWNH5s2bV+Z1wzCYMmUKsbGxBAUFMWDAAHbs2FFmn6eeeoo+ffoQHBxMZGSku99SjTD20pb0bt6AvMJi7vloDfm2YrNDco2vP8R1cTzXsHIREZco4T4T53Dy9lBO73Wx3WDVH5q/LSIiNcMnn3zCxIkTefzxx1mzZg2dO3dm0KBBpKenl7v/b7/9xogRI7j99ttZu3YtQ4cOZejQoWzatMm5z7PPPsvMmTOZNWsWy5cvJyQkhEGDBpGfn+/cp7CwkGHDhnHXXXd5/D16Kx+rhZeGd6FBiD9bU7N58rvNZofkusY9HY9KuEVEXKKE+0zOUTBta2oW2flFhAb40jY2vBoDExERcd0LL7zAmDFjGD16NO3atWPWrFkEBwfzzjvvlLv/Sy+9xODBg3nggQdo27YtTzzxBN26deOVV14BHL3bM2bM4LHHHuPaa6+lU6dOfPDBB6SkpDB37lzneaZOnco///lPOnbsWB1v02tFhQfywk1dAPi/ZfuYt/GQuQG5Sgm3iEil+JodgNdKO3vCXTp/u0ezevhYNX9bRES8V2FhIatXr2bSpEnObVarlQEDBrB06dJyj1m6dCkTJ04ss23QoEHOZHrPnj2kpqYyYMAA5+sRERH06tWLpUuXMnz48ErFWlBQQEFBgfP3rCzHGtY2mw2bzVapc3qLPomR/OPCRF5fsocHP9tAm+hgEuoFl7tv6Xv1mvcc0xU/wEj7naLcTPAPMTuiSvG6dq0F1KaeoXZ1P3e2qSvnUMJ9JofPXqH8ZME0DScXERHvduTIEYqLi4mOji6zPTo6mq1bt5Z7TGpqarn7p6amOl8v3XamfSpj+vTpTJ069bTtCxYsIDi4/OS0Jmlth2ahPuzNKWL0G0sY36EYn7N8b5+UlFR9wZ3DQL96BNmOsXzu6xwNbWN2OFXiTe1aW6hNPUPt6n7uaNO8vLwK76uEuzwFOZC5z/G80ekJt2EYrNzrSLg1f1tERMR9Jk2aVKZnPSsri4SEBAYOHEh4eO2YwtW93wmuenUpf+QUUa91L/q1PL34qs1mIykpicsvvxw/Pz8Tojydz4lPYes39G7si73PELPDqRRvbNeaTm3qGWpX93Nnm5aOvqoIJdzlObLd8RjSCEJOvwnuPZrHkZxC/H2tdGwcUc3BiYiIuKZhw4b4+PiQlpZWZntaWhoxMTHlHhMTE3PW/Usf09LSiI2NLbNPly5dKh1rQEAAAQEBp2338/OrNR86mzby49I2UXy1LoWNKdlc0rb8PwPwsvedcD5s/QafQ2vw8ZaYKsmr2rWWUJt6htrV/dzRpq4cr6Jp5TlcMryuUfnDpdaUVCfvGB9BgK9PdUUlIiJSKf7+/nTv3p3k5GTnNrvdTnJyMr179y73mN69e5fZHxzD8Er3T0xMJCYmpsw+WVlZLF++/IznlJO6JkQCsHbfMXMDccWphdMMw9xYRERqCFMT7unTp9OzZ0/CwsKIiopi6NChbNu2zcyQHM6VcJfcHLs1iaymgERERKpm4sSJvPnmm7z//vts2bKFu+66i9zcXEaPHg3ALbfcUqao2vjx45k/fz7PP/88W7du5V//+herVq1i7NixAFgsFiZMmMCTTz7J119/zcaNG7nllluIi4tj6NChzvPs27ePdevWsW/fPoqLi1m3bh3r1q0jJyenWt+/t+napB4Aa/dnYtSU5DWuC1h9IScNju83OxoRkRrB1CHlixYt4p577qFnz54UFRXxyCOPMHDgQDZv3kxIiInVL9NLEu6oMyXcmQB0K7lZioiIeLubbrqJw4cPM2XKFFJTU+nSpQvz5893Fj3bt28fVuvJ7+H79OnD7Nmzeeyxx3jkkUdo1aoVc+fOpUOHDs59HnzwQXJzc7njjjvIzMykX79+zJ8/n8DAQOc+U6ZM4f3333f+3rVrVwB+/vlnLr74Yg+/a+/VNjYcf18rmXk29h7NI7FhDaj67RcEMR0hZa2jlzuyidkRiYh4PVMT7vnz55f5/b333iMqKorVq1dz0UUXmRQVJyuUl9PDnVNQxLZUxyT5bk2VcIuISM0xduxYZw/1ny1cuPC0bcOGDWPYsGFnPJ/FYmHatGlMmzbtjPu89957vPfee66GWuv5+1rpEBfOmn2ZrN13rGYk3OAYVp6yFg6sgg7Xmx2NiIjX86o53MePHwegfn0TK38X5p61QvmG/ZnYDYiPDCI6PPC010VEREQqonRY+br9meYG4opT53GLiMg5eU2VcrvdzoQJE+jbt2+Z4WqnKigooKCgwPl7aTl2m81W5QXMS48vTt2MH2AEN6TIPxz+dN6Ve44C0CUhos4vRO/OxePrCrWZ69RmrlObuc6dbaZ2l4rqWlILZm3JVLUaoXEPx+Oh9VBUAL6nV5QXEZGTvCbhvueee9i0aRO//PLLGfeZPn06U6dOPW37ggULCA4Odkscmxd9STfgiLURv82bd9rrP2yxAlYCsg8yb94Bt1yzpnPH4vF1jdrMdWoz16nNXOeONsvLy3NDJFIXlPZwbzmUxYnCYoL8a8DKJ/USIbgB5B2F1I0nE3ARESmXVyTcY8eO5dtvv2Xx4sU0btz4jPtNmjSJiRMnOn/PysoiISGBgQMHEh4eXqUYShdC7xjjB/ugfus+DBk8pMw+hmHw+LqFgI2Rg/rQqY6vwe3OxePrCrWZ69RmrlObuc6dbVY6+krkXOIiAokKCyA9u4BNKcfp2czEKXUVZbFA4/Nh+/ew80cIaeTascENwd89nSQiIjWBqQm3YRjce++9fPnllyxcuJDExMSz7h8QEEBAwOlDl9y5ILxPxnbHY3RbfP50zl2Hc8g8YSPA10rHhPr4+XrVFHjTuLP96wq1mevUZq5Tm7nOHW2mNpeKslgsdG0SyQ+/p7F237GakXCDo1d7+/ewcLrjx1XBDSA8HiISHOfqfisE15D3LiLiIlMT7nvuuYfZs2fz1VdfERYWRmpqKgAREREEBQWZEpPliCPhJur0gmlr/nCsv92pcQT+SrZFRESkiro2qVeScGeaHUrFtf8LrH4f8o64dpy9GIoLHMPR845C6gbY9h0sfg663gy974Z6zTwSsoiIWUxNuF977TWA09bhfPfdd7n11lurPR6f4gIsmX84filnSTCtvy0iIiLu1DUhEqhhhdMatIB/bnT9OMOA/ONw/IDj59geWPsRpG2EFa/Dyjeh3VDocy/Ed3N72CIiZjB9SLk3CS1IcTwJbgghDU97fe0+Rw93VyXcIiIi4gYdG0fgY7WQmpXPoeMniI0wZ4RftbBYICjS8RNTsiJNrzth90L4bSbs+gl+/8Lx0+xCR+Ld8nKwalShiNRc+h/sFOEnDjqelDOcPDvfxra0bAC6NY2sxqhERESktgr296V1dBhQw3q53cVigRaXwM1fwp2/QKfhYPWFvUtg9o3wWm9Y86FjCTIRkRpICfcpwvJLEu5GrU97bf3+4xgGNK4XRFRYYDVHJiIiIrXVyfW4j5kbiNliOsJ1r8P4DY7ebf8wOLwVvh4LMzrCkufhRB1vIxGpcZRwn+Jkwl3e/G3Hf/Cavy0iIiLuVDpVrU72cJcnIh4GPgkTf4fLn4CwOMhJg+Rp8EJ7+P5hyNxndpQiIhWihPsUFUm4uzdVwi0iIiLuU9rDvfHgcWzFdnOD8SaBEdB3HIxfD395HaLagy0Xlr8GL3WBz26DlHVmRykiclZKuEsV5hJcWLK8xZ/mcBuG4fzWufSmKCIiIuIOiQ1CiAjyo6DIztZD2WaH4318/aHzcLjrV/jb59D8YjCKYdPn8EZ/eP9q2PGjowq6iIiXUcJd6ugOLBgY5VQo/+NoHsdP2PD3tdImJtykAEVERKQ2slotdClZHuyJbzfz1pLdrNufSZE6u8uyWKDlALjlK/jHYuh4I1h8YM9i+Oh6eK0PrJsNRYVmRyoi4mTqsmDexHJ4GwBGw/Ow/Om1DQePA9A2Nhx/X31HISIiIu7V/7xGLNp+mBV7M1ixNwMAP4sPsw+toEez+nRrWo9uTerRKCzA5Ei9RGxnuP5NuGwKLJ8Fq9+D9M0w9y7HXO9e/4Duox1LkImImEgJdwnLkZKEu5z52xv2ZwLQuXFEdYYkIiIidcTovs3o2DiClXszWPPHMVb/cYxjeTZW78tk9SnF1Do1juDjMRcQEqCPcABEJsCgp+CiBxxJ9/JZkH0IfvwXLP4PdBsFF9zl2E9ExAT637qE5fBWx5OGpy8JtuGAo4e7U+PIaoxIRERE6gqLxULPZvXp2aw+AIWFhbz/xfeEN+/MugPZrPnjGNvTs9lw4DgLNqfyl66NTY7YywRFQr8JcMHdsPFTWPqKo8d72auOJLzD9Y6lxhq2PdeZRETcSuOjS9g7/5WdUVdgT+hVZnux3WBTiiPhVg+3iIiIVAeLxUJUEFzXNZ7p13Xkh39exL2XtgLguw2HTI7Oi/n6Q9eRcNdvMPIzaHaho8Daxv/B6xfiM/t6GmVtVIE1Eak2SrhLGG2u4vf4ERDdocz2nek55BUWE+LvQ/NGoSZFJyIiInXdVZ1iAVi8/QhZ+TaTo/FyFgu0uhxu/RbuWOjo4bb4YN2ziD67nsPnw6uhMM/sKEWkDlDCfQ7rD2QC0CE+Ah/rn8upiYiIiFSP86LDaBkVSmGxnR83p5kdTs0R1xVueAfGraX4/H9QZPXHun8ZbJ9vdmQiUgco4T6HDSUJd+eS5TpEREREzHJlR0cv97yNGlbusnpNsV/+FH80uNjx+94lpoYjInWDEu5zKC2Y1jFe87dFRETEXFdqWHmVHQlr53iyZ7G5gYhInaCE+ywKiorZcigLgM6qUC4iIiImOy86jFYaVl4lR0JaY1iscHQnZKWYHY6I1HJKuM9iW2o2tmKDesF+JNQPMjscEREREWcvt6qVV06RbwhGTCfHL3s0rFxEPEsJ91msLx1O3jgSi0UF00RERMR8pfO4l+w4wvETGlZeGUazCx1PNKxcRDxMCfdZbNifCWj9bREREfEeraLDOC9aw8qrwmjaz/FkrxJuEfEsJdxnUVowrZPmb4uIiIgXGaJq5VViJPQCqy9k7oNje80OR0RqMSXcZ5BXWMSO9GwAOqmHW0RERLxI6bDyxTsOa1h5ZfiHQnwPx3MNKxcRD1LCfQabDmZhNyA6PIDo8ECzwxERERFxKh1Wbis2uOWdFbycvIO1+45RbDfMDq3mSCydx63CaSLiOb5mB+CtNhzIBDScXERERLzTzRc0ZfJXv7N+fybr92fyfNJ2wgN96duyIf1aNeTClo1o0iDY7DC9V+JFsPg5Rw+3YYAK5IqIByjhPoPSCuUqmCYiIiLe6ObezbjovEYs2XGEX3Yc4dddR8jKL+L7Tal8vykVgCb1g7m6cyz/HHAevj4a2FhG4/PBJwByUh1rcjdsZXZEIlILKeE+g00HTy4JJiIiIuKNmjYIoWmDEP52QVOKiu1sOHicX0oS8DX7jrEvI49Xf97Focx8/jOsM1arenGd/AIh4XzYuwT2LFLCLSIeoa86y5FTUMSeI7kAdIgLNzkaERERkXPz9bHSrUk9xl3Wiv/d2Zt1jw/k39d3xMdq4Yu1B5ny9SYMQ3O8y0js73hU4TQR8RAl3OXYcigLgJjwQBqEBpgcjYiIiIjrQgN8ualnE164sTMWC/zfsn08M3+rku5TJV7keNz7C9jt5sYiIrWSEu5y/F4ynLy9erdFRESkhru2SzxPDe0IwOuLdvPqzzuVdJeK7wZ+IZB3FNI2QnFRxX+UoItIBWgOdzl+T3H0cCvhFhERkdrgr72akFtQxFPztvCfBdt5+5c9tIkJp01sGG1LHs+LDiPQz8fsUKuXjx807Q07f4TXL3LtWN8gGP4RtLzMM7GJSK2ghLscpQl3uzhVKBcREZHaYcxFzbHZ7byYtJ1jeTaW7j7K0t1Hna9bLZDYMIQ2seGcFxVGs4bBJDYMoVnDEMID/UyM3MM6j4BdP4NR7NpxRSdgzQdKuEXkrJRw/0lhkZ0d6dmAerhFRESkdrn74pbc1jeRnek5bDmUxdbUbLYcymLLoSyO5dnYdTiXXYdz+Y5DZY5rUj+YmSO60iUh0pzAPanjDdD6CigurPgxKevgw6GweyHYi8Fax0YGiEiFKeH+kx3pOdiKDSKC/GhcL8jscERERETcKtDPhw7xEXSIPzmSzzAMDmcXsKUkAd+VnsPeo7nsPZrH4ewC9mXkcdf/rea7cRdSP8TfxOg9xD8ECKn4/s0uhMAIyM+Eg2sgoaenIhORGk4J959sPuTo3W4XG47ForUqRUREpPazWCxEhQcSFR5I//MalXktI7eQG177jd1HcpnwyTreu7Wn1vP28YXmF8Pmr2BXshJuETkjVSn/k9IlwTScXERERATqh/jz3791I9DPyuLth3n1551mh+QdWpTM3d6ZbG4cIuLVlHD/SWkPd/t4JdwiIiIiAG1iwnni2g4AvPDjdn7decTkiLxAabG0g6vgxDFzYxERr6WE+xR2A7amlhZMU4VyERERkVLDeiRwY4/GGAaMn7OW1X8c49DxExQUuVjdu7aIaAwNW4Nhh92LzI5GRLyU5nCf4kg+5BYWE+BrpXlDFwpniIiIiNQBU6/pwIYDx9mams31r/3m3B4a4Ev9EP8yPw1OfR7qT/2QAOe2YH+f2lErp+VlcGSbYx53+6FmRyMiXkgJ9ykO5Dr+428TG46vjzr/RURERE4V5O/DrL9155EvN7IjPYdjuYUU2Q1yCorIKShiX0Zehc4T4Gt1JN9/SsTrh/gTFujr+AnwIzTQF183FmgrKipidxas/uMYvr4nPwaHB/lxXnSY6ydscRks+y/s/AkMA2rDlwgi4lZKuE9RmnCrYJqIiIhI+Zo1DGH2mAsAx3JiWflFZOQWkpFbwNGcQjJyCzmaW8jRnEKO5TmeZ+QWkJHjeF5QZKegyE7K8XxSjueb8A58een3ladtHXNhIo9e2c61UzXtAz4BkHUAjmyHRq3dFKOI1BZKuE9xMNfxqIRbRERE5NwsFgsRQX5EBPmRWIHpeIZhkFdY7EzKT03SS3+y84vILrCRnV9ETn4RdsNwW7yGAbl5uYQEhzg7ow3gj6N5vLlkD92b1mdwh5iKn9A/2JF07/7ZUa1cCbeI/IkS7hKGYZzSw62CaSIiUvu8+uqrPPfcc6SmptK5c2defvllzj///DPu/+mnnzJ58mT27t1Lq1at+Pe//82QIUOcrxuGweOPP86bb75JZmYmffv25bXXXqNVq1bOfTIyMrj33nv55ptvsFqtXH/99bz00kuEhoZ69L2Kd7JYLIQE+BIS4EtC/eBqv77NZmPevHkMGdIPPz8/5/an523hjcW7eeCz9bSPC3cttpaXORLuXcnQ+24PRC0iNZkmKpdIyy4gp8iCj9VCm5hKzOERERHxYp988gkTJ07k8ccfZ82aNXTu3JlBgwaRnp5e7v6//fYbI0aM4Pbbb2ft2rUMHTqUoUOHsmnTJuc+zz77LDNnzmTWrFksX76ckJAQBg0aRH7+yWHCI0eO5PfffycpKYlvv/2WxYsXc8cdd3j8/Yq44oFBrenaJJLs/CLGzl5DYZG94geXrse991ewmTFEXkS8mRLuEqXrbzdvGEygn4/J0YiIiLjXCy+8wJgxYxg9ejTt2rVj1qxZBAcH884775S7/0svvcTgwYN54IEHaNu2LU888QTdunXjlVdeARy92zNmzOCxxx7j2muvpVOnTnzwwQekpKQwd+5cALZs2cL8+fN566236NWrF/369ePll19mzpw5pKSkVNdbFzknPx8rL4/oSkSQH+sPHOeZ77dW/OCothAWB0UnYN9v595fROoUDSkvsTklC4B2sZq/LSIitUthYSGrV69m0qRJzm1Wq5UBAwawdOnSco9ZunQpEydOLLNt0KBBzmR6z549pKamMmDAAOfrERER9OrVi6VLlzJ8+HCWLl1KZGQkPXr0cO4zYMAArFYry5cv5y9/+Ysb36VI1TSuF8x/hnVmzAereOfXPfj5WKgX4l+hYy8O6UGb7K/Z/+2/OVjvZw9H6l3shgHp6azMXIlVVdrdRu3qfnbDwDh8GBhyzn3dydSEe/HixTz33HOsXr2aQ4cO8eWXXzJ06FBTYint4W4Xq+HkIiJSuxw5coTi4mKio6PLbI+Ojmbr1vJ78lJTU8vdPzU11fl66baz7RMVFVXmdV9fX+rXr+/c588KCgooKChw/p6V5fhC3GazYbPZzvo+a5PS91qX3nN1OFe7XtyqPrf1aco7v/3B64t3V/i8G6zN+K8/JBxbRsKxZW6JtcbZY3YAtZTa1a1shg8224tVP48L/zebmnDn5ubSuXNnbrvtNq677jozQ+HCVg04nJZKtyaRpsYhIiJSl02fPp2pU6eetn3BggUEB1d/kS2zJSUlmR1CrXS2dm1vh2uaWEg9UfFexQyjO58XXksDe4Y7whMRD7FbrGS74f/VvLy8Cu9rasJ9xRVXcMUVV5gZgtOInglEHN5Il4RIs0MRERFxq4YNG+Lj40NaWlqZ7WlpacTElL8EUkxMzFn3L31MS0sjNja2zD5dunRx7vPnomxFRUVkZGSc8bqTJk0qM5Q9KyuLhIQEBg4cSHh43Zn2ZbPZSEpK4vLLLy9TTVuqpqLtek2lzn5lZcOq0fR31TPUru7nzjYtHX1VETVqDrcnh5lp6Jbr1GauU5u5Tm3mOrWZ69zZZt7Y7v7+/nTv3p3k5GTn1C273U5ycjJjx44t95jevXuTnJzMhAkTnNuSkpLo3bs3AImJicTExJCcnOxMsLOysli+fDl33XWX8xyZmZmsXr2a7t27A/DTTz9ht9vp1atXudcNCAggICDgtO1+fn518kNnXX3fnqZ2dT+1qWeoXd3PHW3qyvE1KuGujmFmGrrlOrWZ69RmrlObuU5t5jp3tJkrw8yq08SJExk1ahQ9evTg/PPPZ8aMGeTm5jJ69GgAbrnlFuLj45k+fToA48ePp3///jz//PNceeWVzJkzh1WrVvHGG28AjvWUJ0yYwJNPPkmrVq1ITExk8uTJxMXFOZP6tm3bMnjwYMaMGcOsWbOw2WyMHTuW4cOHExcXZ0o7iIiIVKcalXB7cpiZhm24Tm3mOrWZ69RmrlObuc6sYWbV6aabbuLw4cNMmTKF1NRUunTpwvz5851Fz/bt24fVenK10D59+jB79mwee+wxHnnkEVq1asXcuXPp0KGDc58HH3yQ3Nxc7rjjDjIzM+nXrx/z588nMDDQuc9HH33E2LFjueyyy7BarVx//fXMnDmz+t64iIiIiWpUwl0dw8w0bMN1ajPXqc1cpzZzndrMddU9zKy6jR079oxDyBcuXHjatmHDhjFs2LAzns9isTBt2jSmTZt2xn3q16/P7NmzXY5VRESkNrCeexcRERERERERcZWpPdw5OTns3LnT+fuePXtYt24d9evXp0mTJiZGJiIiIiIiIlI1pibcq1at4pJLLnH+Xjo/e9SoUbz33nsmRSUiIiIiIiJSdaYm3BdffDGGYZgZgoiIiIiIiIhHaA63iIiIiIiIiAco4RYRERERERHxACXcIiIiIiIiIh6ghFtERERERETEA5Rwi4iIiIiIiHiAEm4RERERERERD1DCLSIiIiIiIuIBSrhFREREREREPEAJt4iIiIiIiIgHKOEWERERERER8QBfswOoCsMwAMjKyqryuWw2G3l5eWRlZeHn51fl89UFajPXqc1cpzZzndrMde5ss9J7Uuk9SqrGnff6mkT/jj1D7ep+alPPULu6n1n3+hqdcGdnZwOQkJBgciQiIiJlZWdnExERYXYYNZ7u9SIi4q0qcq+3GDX4K3i73U5KSgphYWFYLJYqnSsrK4uEhAT2799PeHi4myKs3dRmrlObuU5t5jq1mevc2WaGYZCdnU1cXBxWq2ZuVZU77/U1if4de4ba1f3Upp6hdnU/s+71NbqH22q10rhxY7eeMzw8XH+pXaQ2c53azHVqM9epzVznrjZTz7b7eOJeX5Po37FnqF3dT23qGWpX96vue72+ehcRfzt/+AAAD51JREFUERERERHxACXcIiIiIiIiIh6ghLtEQEAAjz/+OAEBAWaHUmOozVynNnOd2sx1ajPXqc3E2+jvpGeoXd1PbeoZalf3M6tNa3TRNBERERERERFvpR5uEREREREREQ9Qwi0iIiIiIiLiAUq4RURERERERDxACbeIiIiIiIiIByjhLvHqq6/SrFkzAgMD6dWrFytWrDA7JK8wffp0evbsSVhYGFFRUQwdOpRt27aV2Sc/P5977rmHBg0aEBoayvXXX09aWppJEXufZ555BovFwoQJE5zb1GanO3jwIH/7299o0KABQUFBdOzYkVWrVjlfNwyDKVOmEBsbS1BQEAMGDGDHjh0mRmyu4uJiJk+eTGJiIkFBQbRo0YInnniCU+tg1vU2W7x4MVdffTVxcXFYLBbmzp1b5vWKtE9GRgYjR44kPDycyMhIbr/9dnJycqrxXUhtp/us5+k+7B66T7uf7uVVVyPu9YYYc+bMMfz9/Y133nnH+P33340xY8YYkZGRRlpamtmhmW7QoEHGu+++a2zatMlYt26dMWTIEKNJkyZGTk6Oc58777zTSEhIMJKTk41Vq1YZF1xwgdGnTx8To/YeK1asMJo1a2Z06tTJGD9+vHO72qysjIwMo2nTpsatt95qLF++3Ni9e7fxww8/GDt37nTu88wzzxgRERHG3LlzjfXr1xvXXHONkZiYaJw4ccLEyM3z1FNPGQ0aNDC+/fZbY8+ePcann35qhIaGGi+99JJzn7reZvPmzTMeffRR44svvjAA48svvyzzekXaZ/DgwUbnzp2NZcuWGUuWLDFatmxpjBgxoprfidRmus96lu7D7qH7tGfoXl51NeFer4TbMIzzzz/fuOeee5y/FxcXG3Fxccb06dNNjMo7paenG4CxaNEiwzAMIzMz0/Dz8zM+/fRT5z5btmwxAGPp0qVmhekVsrOzjVatWhlJSUlG//79nTd6tdnpHnroIaNfv35nfN1utxsxMTHGc88959yWmZlpBAQEGB9//HF1hOh1rrzySuO2224rs+26664zRo4caRiG2uzP/nwTrkj7bN682QCMlStXOvf5/vvvDYvFYhw8eLDaYpe6RfdZ99F92H10n/YM3cvdy1vv9XV+SHlhYSGrV69mwIABzm1Wq5UBAwawdOlSEyPzTsePHwegfv36AKxevRqbzVam/dq0aUOTJk3qfPvdc889XHnllWXaBtRm5fn666/p0aMHw4YNIyoqiq5du/Lmm286X9+zZw+pqall2iwiIoJevXrV2Tbr06cPycnJbN++HYD169fzyy+/cMUVVwBqs3OpSPssXbqUyMhIevTo4dxnwIABWK1Wli9fXu0xS92g+6z76D7sPrpPe4bu5Z7lLfd6X7ecpQY7cuQIxcXFREdHl9keHR3N1q1bTYrKO9ntdiZMmEDfvn3p0KEDAKmpqfj7+xMZGVlm3+joaFJTU02I0jvMmTOHNWvWsHLlytNeU5udbvfu3bz22mtMnDiRRx55hJUrVzJu3Dj8/f0ZNWqUs13K+3daV9vs4YcfJisrizZt2uDj40NxcTFPPfUUI0eOBFCbnUNF2ic1NZWoqKgyr/v6+lK/fn21oXiE7rPuo/uwe+k+7Rm6l3uWt9zr63zCLRV3zz33sGnTJn755RezQ/Fq+/fvZ/z48SQlJREYGGh2ODWC3W6nR48ePP300wB07dqVTZs2MWvWLEaNGmVydN7pf//7Hx999BGzZ8+mffv2rFu3jgkTJhAXF6c2E6mhdJ91D92H3U/3ac/QvbxuqPNDyhs2bIiPj89plSnT0tKIiYkxKSrvM3bsWL799lt+/vlnGjdu7NweExNDYWEhmZmZZfavy+23evVq0tPT6datG76+vvj6+rJo0SJmzpyJr68v0dHRarM/iY2NpV27dmW2tW3bln379gE420X/Tk964IEHePjhhxk+fDgdO3bk5ptv5p///CfTp08H1GbnUpH2iYmJIT09vczrRUVFZGRkqA3F7XSfdR/dh91P92nP0L3cs7zlXl/nE25/f3+6d+9OcnKyc5vdbic5OZnevXubGJl3MAyDsWPH8uWXX/LTTz+RmJhY5vXu3bvj5+dXpv22bdvGvn376mz7XXbZZWzcuJF169Y5f3r06MHIkSOdz9VmZfXt2/e0ZXC2b99O06ZNAUhMTCQmJqZMm2VlZbF8+fI622Z5eXlYrWX/C/fx8cFutwNqs3OpSPv07t2bzMxMVq9e7dznp59+wm6306tXr2qPWWon3WfdT/dh99N92jN0L/csr7nXu6X0Wg03Z84cIyAgwHjvvfeMzZs3G3fccYcRGRlppKammh2a6e666y4jIiLCWLhwoXHo0CHnT15ennOfO++802jSpInx008/GatWrTJ69+5t9O7d28Sovc+p1VENQ232ZytWrDB8fX2Np556ytixY4fx0UcfGcHBwcb//d//Ofd55plnjMjISOOrr74yNmzYYFx77bV1elmMUaNGGfHx8c6lRL744gujYcOGxoMPPujcp663WXZ2trF27Vpj7dq1BmC88MILxtq1a40//vjDMIyKtc/gwYONrl27GsuXLzd++eUXo1WrVloWTNxK99nqoftw1eg+7Rm6l1ddTbjXK+Eu8fLLLxtNmjQx/P39jfPPP99YtmyZ2SF5BaDcn3fffde5z4kTJ4y7777bqFevnhEcHGz85S9/MQ4dOmRe0F7ozzd6tdnpvvnmG6NDhw5GQECA0aZNG+ONN94o87rdbjcmT55sREdHGwEBAcZll11mbNu2zaRozZeVlWWMHz/eaNKkiREYGGg0b97cePTRR42CggLnPnW9zX7++edy//8aNWqUYRgVa5+jR48aI0aMMEJDQ43w8HBj9OjRRnZ2tgnvRmor3Werh+7DVaf7tPvpXl51NeFebzEMw3BPX7mIiIiIiIiIlKrzc7hFREREREREPEEJt4iIiIiIiIgHKOEWERERERER8QAl3CIiIiIiIiIeoIRbRERERERExAOUcIuIiIiIiIh4gBJuEREREREREQ9Qwi1iMovFwty5c92+r4iIiHgH3etF6i5fswMQqesOHTpEvXr1zA7D69x6661kZmbqQ4eIiNR4uteXT/d6qQuUcIuYqLCwkJiYGLPDcLvCwkL8/f3NDkNERMR0uteL1G0aUi5SjS6++GLGjh3LhAkTaNiwIYMGDSozdKywsJCxY8cSGxtLYGAgTZs2Zfr06Wc83+OPP05sbCwbNmw457X/+9//0qpVKwIDA4mOjuaGG25wvma323n22Wdp2bIlAQEBNGnShKeeesr5+saNG7n00ksJCgqiQYMG3HHHHeTk5Dhfv/XWWxk6dChPPfUUcXFxtG7dGoD9+/dz4403EhkZSf369bn22mvZu3fvOWP917/+xfvvv89XX32FxWLBYrGwcOHCcx4nIiJiNt3rda8XOZV6uEWq2fvvv89dd93Fr7/+CkCbNm2cr82cOZOvv/6a//3vfzRp0oT9+/ezf//+085hGAbjxo3j22+/ZcmSJbRs2fKs11y1ahXjxo3jww8/pE+fPmRkZLBkyRLn65MmTeLNN9/kxRdfpF+/fhw6dIitW7cCkJuby6BBg+jduzcrV64kPT2dv//974wdO5b33nvPeY7k5GTCw8NJSkoCwGazOY9bsmQJvr6+PPnkkwwePJgNGzac9Vvx+++/ny1btpCVlcW7774LQP369c/RsiIiIt5B93rd60WcDBGpNv379ze6du1aZhtgfPnll4ZhGMa9995rXHrppYbdbi/3eMD49NNPjb/+9a9G27ZtjQMHDlToup9//rkRHh5uZGVlnfZaVlaWERAQYLz55pvlHvvGG28Y9erVM3JycpzbvvvuO8NqtRqpqamGYRjGqFGjjOjoaKOgoMC5z4cffmi0bt26zHspKCgwgoKCjB9++OGcMY8aNcq49tprK/T+REREvIXu9brXi5xKQ8pFqln37t3P+Nqtt97KunXraN26NePGjWPBggWn7fPPf/6T5cuXs3jxYuLj4yt0zcsvv5ymTZvSvHlzbr75Zj766CPy8vIA2LJlCwUFBVx22WXlHrtlyxY6d+5MSEiIc1vfvn2x2+1s27bNua1jx45lvslev349O3fuJCwsjNDQUEJDQ6lfvz75+fns2rWrQnGLiIjURLrX614vUkoJt0g1O/Vm9mfdunVjz549PPHEE5w4cYIbb7yxzPwrcNxQDx48yA8//FDha4aFhbFmzRo+/vhjYmNjmTJlCp07dyYzM5OgoKBKv5dT/fl95eTk0L17d9atW1fmZ/v27fz1r391yzVFRES8ke71uteLlFLCLeJlwsPDuemmm3jzzTf55JNP+Pzzz8nIyHC+fs011zB79mz+/ve/M2fOnAqf19fXlwEDBvDss8+yYcMG9u7dy08//USrVq0ICgoiOTm53OPatm3L+vXryc3NdW779ddfsVqtzoIp5enWrRs7duwgKiqKli1blvmJiIg4Z7z+/v4UFxdX+P2JiIjUFLrXO+heL3WBEm4RL/LCCy/w8ccfs3XrVrZv386nn35KTEwMkZGRZfb7y1/+wocffsjo0aP57LPPznneb7/9lpkzZ7Ju3Tr++OMPPvjgA+x2O61btyYwMJCHHnqIBx98kA8++IBdu3axbNky3n77bQBGjhxJYGAgo0aNYtOmTfz888/ce++93HzzzURHR5/xmiNHjqRhw4Zce+21LFmyhD179rBw4ULGjRvHgQMHzhlzs2bN2LBhA9u2bePIkSPYbLZzHiMiIuLtdK8/Sfd6qQtUpVzEi4SFhfHss8+yY8cOfHx86NmzJ/PmzcNqPf27sRtuuAG73c7NN9+M1WrluuuuO+N5IyMj+eKLL/jXv/5Ffn4+rVq14uOPP6Z9+/YATJ48GV9fX6ZMmUJKSgqxsbHceeedAAQHB/PDDz8wfvx4evbsSXBwMNdffz0vvPDCWd9LcHAwixcv5qGHHuK6664jOzub+Ph4LrvsMsLDw8/ZFmPGjGHhwoX06NGDnJwcfv75Zy6++OJzHiciIuLNdK8/Sfd6qQsshmEYZgchIiIiIiIiUttoSLmIiIiIiIiIByjhFqkFlixZ4lyOo7wfb3S2eJcsWWJ2eCIiIl5F93qRmklDykVqgRMnTnDw4MEzvt6yZctqjKZidu7cecbX4uPj3baEiYiISG2ge71IzaSEW0RERERERMQDNKRcRERERERExAOUcIuIiIiIiIh4gBJuEREREREREQ9Qwi0iIiIiIiLiAUq4RURERERERDxACbeIiIiIiIiIByjhFhEREREREfEAJdwiIiIiIiIiHvD/uiRMdVYaVgUAAAAASUVORK5CYII=\n"
          },
          "metadata": {}
        }
      ],
      "source": [
        "#@title Plot the comorbidities curves\n",
        "plot_comorbidities(X_cr, 'Decorrelate model')"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 14,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 0
        },
        "id": "77hCX9orSEcr",
        "outputId": "850a7a91-4213-4f7d-a32a-dc76389c3f99"
      },
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1200x600 with 4 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ],
      "source": [
        "#@title Heatmaps\n",
        "f, (ax1, ax2, ax3) = plt.subplots(1, 3)\n",
        "f.suptitle(\"Correlation in datasets\")\n",
        "#ax1.get_shared_y_axes().join(ax2,ax3)\n",
        "plot_heatmap(X_toProcess, \"Original\", ax1, n = 1)\n",
        "plot_heatmap(X_cr, \"Decorr.\", ax2, n = 2)\n",
        "plot_heatmap(X_cr_alpha, \"Decorr. \\u03B1 \", ax3, 3)\n",
        "plt.tight_layout()\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 15,
      "metadata": {
        "id": "UcKTOTslehF6"
      },
      "outputs": [],
      "source": [
        "del var, var2, X_cr, X_cr_alpha, X_toProcess, cr, cr_alpha, model_r, risk"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "hV3bQvrtFGox"
      },
      "source": [
        "#Augmented Model\n",
        "\n",
        "---\n",
        "The approach consists in creating synthetic observations for the minority class, with the goal of giving the algorithm the opportunity of learning the underlining patterns, as showed in Fu, R., Huang, Y., Singh, P. V., *AI and Algorithmic Bias: Source, Detection, Mitigation and Implications*, http://dx.doi.org/10.2139/ssrn.3681517, 2020.\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 16,
      "metadata": {
        "id": "m1ZVscyNPWCR"
      },
      "outputs": [],
      "source": [
        "#@title Clean the data\n",
        "# Load the dataset\n",
        "url = \"https://gitlab.com/labsysmed/dissecting-bias/-/raw/master/data/data_new.csv\"\n",
        "df = pd.read_csv(StringIO(requests.get(url).text))\n",
        "\n",
        "# Extract race feauture\n",
        "race_mapping = {'white': 0, 'black': 1}\n",
        "df['race'] = df['race'].map(race_mapping)\n",
        "\n",
        "race = df['race']\n",
        "# Substitute NaN values in order to use the data\n",
        "df.fillna(0, inplace=True)\n",
        "\n",
        "\n",
        "X = df.drop(['cost_t', 'gagne_sum_t', 'cost_avoidable_t'], axis = 1)\n",
        "\n",
        "# Extract target variables (y1, y2, y3) from DataFrame\n",
        "y1 = df['cost_t']\n",
        "y2 = df['gagne_sum_t']\n",
        "y3 = df['cost_avoidable_t']\n",
        "y = df[['cost_t', 'gagne_sum_t', 'cost_avoidable_t']]"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 17,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "1g-JqcjKFJE2",
        "outputId": "bdf60211-3ecc-4158-9ef4-ce8a2377cd60"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Before oversampling:\n",
            "0    43202\n",
            "1     5582\n",
            "After oversampling:\n",
            "1    43570\n",
            "0    43202\n",
            "Target Variable          MSE  Race: Black MSE  Race: White MSE\n",
            "       Cost_tot 2.714576e+08     3.465122e+08     1.953666e+08\n"
          ]
        }
      ],
      "source": [
        "#@title Model\n",
        "from imblearn.over_sampling import ADASYN\n",
        "'''\n",
        "ADASYN showed better results in comparison to SMOTE, due to the fact that the\n",
        "algorithm takes into account the local distribution of the samples\n",
        "'''\n",
        "\n",
        "# Identify the minority class in the 'race' column\n",
        "values = df['race'].value_counts().to_string(dtype = False)\n",
        "print(\"Before oversampling:\\n\", values, sep='')\n",
        "# Instantiate OverSampler\n",
        "oversampler = ADASYN(random_state=42, sampling_strategy = 'auto')\n",
        "\n",
        "# Generate synthetic samples\n",
        "df_augmented, race_aug = oversampler.fit_resample(df.drop('race', axis=1), df['race'])\n",
        "\n",
        "values = race_aug.value_counts().to_string(dtype = False)\n",
        "print(\"After oversampling:\\n\", values, sep='')\n",
        "# Extract the data for training\n",
        "y1_augmented =  df_augmented['cost_t']\n",
        "X_aug = df_augmented.drop(['cost_t', 'gagne_sum_t', 'cost_avoidable_t'], axis = 1)\n",
        "\n",
        "# Train\n",
        "mse_df, model_aug = train_lasso(X_aug, y1_augmented, race_aug)\n",
        "X_aug['race'] = race_aug\n",
        "print(mse_df.to_string(index=False))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 18,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "UAI6kOZv3uog",
        "outputId": "2223d231-06e8-4d71-cbba-d41e57338243"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "max: 18\n",
            "*-------------------------------------*\n",
            "Percentage of White People in the high risk section: 0.6666666666666666%\n",
            "Percentage of Black People in the high risk section: 0.3333333333333333"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "<ipython-input-4-df564a953dd6>:41: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  df_plot1.loc[:,'risk_score_t_scaled'] = scaler.transform(df_plot1['risk_score_t'].values.reshape(-1, 1)).astype(int)\n",
            "<ipython-input-4-df564a953dd6>:43: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  df_plot2.loc[:,'risk_score_t_scaled'] = scaler.transform(df_plot2['risk_score_t'].values.reshape(-1, 1)).astype(int)\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "%\n",
            "*-------------------------------------*\n",
            "\n",
            "*-------------------------------------*\n",
            "Num of White People in the high risk section: 6\n",
            "Num of Black People in the high risk section: 3\n",
            "*-------------------------------------*\n",
            "\n",
            "mean: 2.047451506874682\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1200x600 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ],
      "source": [
        "#@title Plot the comorbidities curves\n",
        "\n",
        "y1_r = model_aug.predict(df.drop(['cost_t', 'gagne_sum_t', 'cost_avoidable_t','race'], axis = 1))\n",
        "y1_r = np.reshape(y1_r, (-1,1))\n",
        "risk_reg = LinearRegression().fit(y1_r, X['risk_score_t'])\n",
        "risk = risk_reg.predict(y1_r)\n",
        "\n",
        "X['risk_score_t'] = risk\n",
        "\n",
        "plot_comorbidities(X, 'Augmented model')"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "del risk_reg, X_aug, y1_r, oversampler, race_aug, y1_augmented"
      ],
      "metadata": {
        "id": "h47OJJqET-HQ"
      },
      "execution_count": 19,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Bzq6_3VzXhwl"
      },
      "source": [
        "# Model on a new label\n",
        "\n",
        "---\n",
        "\n",
        "The original work suggests working on a new label (cost_avoidable instead of cost_tot) but they keep producing the risk score as a function of the label used, the next experiment tries to use some basic pre-processing on the data and works with a model that retrievies the risk score directly.\n",
        "\n",
        "**Risk score equation**\n",
        "$$\n",
        "R_i = \\left( \\sum_{k\\in \\text{ comorbidities-i}} mortalityRate(k) \\right) + \\left( \\sum_{j\\in \\text{ biomarkers-i}} OutOfBounds(j) \\right) \n",
        "+ \\left( \\sum_{v\\in \\text{ costs-i}} \\frac{v}{10}\\right) + AgePenalty(i)\n",
        "$$\n",
        "\n",
        "**Biomarkers equation**\n",
        "$$\n",
        "OutOfBounds(j) = \\begin{cases} 0 \\text{ if } lowerBound \\leq j\\leq upperBound \\\\\n",
        "2 \\text{ else}\n",
        "\\end{cases}\n",
        "$$\n",
        "**Age Penalty equation**\n",
        "$$\n",
        "AgePenalty(i) = \\begin{cases} 0 \\text{ if } (i.age)\\leq 65 \\\\\n",
        "5 \\text{ else} \n",
        "\\end{cases}\n",
        "$$\n",
        "\n",
        "Mortality rates taken from:\n",
        "\n",
        "*   https://www.who.int/data/gho/indicator-metadata-registry\n",
        "*   https://pubmed.ncbi.nlm.nih.gov/21828943/\n",
        "*   https://ourworldindata.org/grapher/infectious-disease-death-rates\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 20,
      "metadata": {
        "id": "2mm1V5STYiPm"
      },
      "outputs": [],
      "source": [
        "#@title Clean the data\n",
        "# Load the dataset\n",
        "url = \"https://gitlab.com/labsysmed/dissecting-bias/-/raw/master/data/data_new.csv\"\n",
        "df = pd.read_csv(StringIO(requests.get(url).text))\n",
        "\n",
        "# Extract race feauture\n",
        "race_mapping = {'white': 0, 'black': 1}\n",
        "df['race'] = df['race'].map(race_mapping)\n",
        "\n",
        "race = df['race']\n",
        "# Substitute NaN values in order to use the data\n",
        "df.fillna(0, inplace=True)\n",
        "\n",
        "\n",
        "X = df.drop(['cost_t', 'gagne_sum_t', 'cost_avoidable_t','risk_score_t'], axis = 1)\n",
        "# Extract target variables (y1, y2, y3) from DataFrame\n",
        "y1 = df['cost_t']\n",
        "y2 = df['gagne_sum_t']\n",
        "y3 = df['cost_avoidable_t']\n",
        "y = df[['cost_t', 'gagne_sum_t', 'cost_avoidable_t']]\n",
        "y_r = df['risk_score_t']"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#@title Array of basic data\n",
        "# Multiplicators\n",
        "BIO_MULTIPLICATOR = 2\n",
        "GAGNE_MULTIPLICATOR = 5\n",
        "COST_MULTIPLICATOR = 1/10\n",
        "\n",
        "columns = ['alcohol_elixhauser_tm1','anemia_elixhauser_tm1',\n",
        "             'arrhythmia_elixhauser_tm1','arthritis_elixhauser_tm1','bloodlossanemia_elixhauser_tm1','coagulopathy_elixhauser_tm1',\n",
        "             'compdiabetes_elixhauser_tm1','depression_elixhauser_tm1','drugabuse_elixhauser_tm1','electrolytes_elixhauser_tm1',\n",
        "             'hypertension_elixhauser_tm1','hypothyroid_elixhauser_tm1','liver_elixhauser_tm1','neurodegen_elixhauser_tm1','obesity_elixhauser_tm1',\n",
        "             'paralysis_elixhauser_tm1','psychosis_elixhauser_tm1','pulmcirc_elixhauser_tm1','pvd_elixhauser_tm1','renal_elixhauser_tm1','uncompdiabetes_elixhauser_tm1',\n",
        "             'valvulardz_elixhauser_tm1','wtloss_elixhauser_tm1','cerebrovasculardz_romano_tm1','chf_romano_tm1','dementia_romano_tm1','hemiplegia_romano_tm1',\n",
        "             'hivaids_romano_tm1','metastatic_romano_tm1','myocardialinfarct_romano_tm1','pulmonarydz_romano_tm1','tumor_romano_tm1','ulcer_romano_tm1']\n",
        "hospitalization_percentages = {\n",
        "    'alcohol_elixhauser_tm1': 10,       #\n",
        "    'anemia_elixhauser_tm1': 10,       #\n",
        "    'arrhythmia_elixhauser_tm1': 10,     #\n",
        "    'arthritis_elixhauser_tm1': 10,    #\n",
        "    'bloodlossanemia_elixhauser_tm1': 10,#\n",
        "    'coagulopathy_elixhauser_tm1': 10,#\n",
        "    'compdiabetes_elixhauser_tm1': 10,  #\n",
        "    'depression_elixhauser_tm1': 4,   \n",
        "    'drugabuse_elixhauser_tm1': 10,     #\n",
        "    'electrolytes_elixhauser_tm1': 10, # \n",
        "    'hypertension_elixhauser_tm1': 13,  \n",
        "    'hypothyroid_elixhauser_tm1': 10,   #\n",
        "    'liver_elixhauser_tm1': 10,        #\n",
        "    'neurodegen_elixhauser_tm1': 10,     #\n",
        "    'obesity_elixhauser_tm1': 10, #\n",
        "    'paralysis_elixhauser_tm1': 10,#\n",
        "    'psychosis_elixhauser_tm1': 10,#\n",
        "    'pulmcirc_elixhauser_tm1': 20,\n",
        "    'pvd_elixhauser_tm1': 10,#\n",
        "    'renal_elixhauser_tm1': 15,\n",
        "    'uncompdiabetes_elixhauser_tm1': 8,\n",
        "    'valvulardz_elixhauser_tm1': 10,     # \n",
        "    'wtloss_elixhauser_tm1': 10,#\n",
        "    'cerebrovasculardz_romano_tm1': 10, #\n",
        "    'chf_romano_tm1': 10,#\n",
        "    'dementia_romano_tm1': 10,           #\n",
        "    'hemiplegia_romano_tm1': 24,        \n",
        "    'hivaids_romano_tm1': 10,          #\n",
        "    'metastatic_romano_tm1': 10,       #\n",
        "    'myocardialinfarct_romano_tm1': 34, \n",
        "    'pulmonarydz_romano_tm1': 10,      #\n",
        "    'tumor_romano_tm1': 14,            \n",
        "    'ulcer_romano_tm1': 15             \n",
        "}\n",
        "\n",
        "hospitalization_percentages_array = [hospitalization_percentages[col] for col in columns]\n",
        "\n",
        "bio_columns = ['cre_min-low_tm1',\n",
        "             'cre_min-high_tm1','cre_mean-low_tm1','cre_mean-high_tm1','cre_max-low_tm1','cre_max-high_tm1',\n",
        "             'crp_min-low_tm1','crp_min-high_tm1','crp_mean-low_tm1','crp_mean-high_tm1','crp_max-low_tm1','crp_max-high_tm1',\n",
        "             'esr_min-low_tm1','esr_min-high_tm1','esr_mean-low_tm1','esr_mean-high_tm1','esr_max-low_tm1',\n",
        "             'esr_max-high_tm1','ghba1c_min-low_tm1','ghba1c_min-high_tm1','ghba1c_min-normal_tm1','ghba1c_mean-low_tm1','ghba1c_mean-high_tm1',\n",
        "             'ghba1c_max-low_tm1','ghba1c_max-high_tm1','hct_min-low_tm1','hct_min-high_tm1',\n",
        "             'hct_mean-low_tm1','hct_mean-high_tm1','hct_max-low_tm1','hct_max-high_tm1','ldl_min-low_tm1','ldl_min-high_tm1',\n",
        "             'ldl-mean-low_tm1','ldl-mean-high_tm1','ldl_max-low_tm1','ldl_max-high_tm1',\n",
        "             'nt_bnp_min-low_tm1','nt_bnp_min-high_tm1','nt_bnp_mean-low_tm1','nt_bnp_mean-high_tm1',\n",
        "             'nt_bnp_max-low_tm1','nt_bnp_max-high_tm1','nt_bnp_max-normal_tm1','sodium_min-low_tm1','sodium_min-high_tm1',\n",
        "             'sodium_mean-low_tm1','sodium_mean-high_tm1','sodium_max-low_tm1','sodium_max-high_tm1',\n",
        "             'trig_min-low_tm1','trig_min-high_tm1','trig_min-normal_tm1',\n",
        "             'trig_mean-low_tm1','trig_mean-high_tm1','trig_max-low_tm1','trig_max-high_tm1']\n",
        "\n",
        "cost_columns = ['cost_dialysis_tm1','cost_emergency_tm1','cost_home_health_tm1','cost_ip_medical_tm1','cost_ip_surgical_tm1','cost_laboratory_tm1',\n",
        "             'cost_op_primary_care_tm1','cost_op_specialists_tm1','cost_op_surgery_tm1','cost_other_tm1','cost_pharmacy_tm1','cost_physical_therapy_tm1',\n",
        "             'cost_radiology_tm1']"
      ],
      "metadata": {
        "id": "1EMqyrWs8qxl"
      },
      "execution_count": 21,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "#@title Create new risk score\n",
        "df_y = df[columns]#.astype(float)\n",
        "# Mortality rate proportional term\n",
        "df_y = df_y * hospitalization_percentages_array\n",
        "# Penalization for biomarkers\n",
        "df_bio = df[bio_columns].multiply(BIO_MULTIPLICATOR)\n",
        "# Penalization for multiple comorbidities\n",
        "df_num = df['gagne_sum_tm1'].multiply(GAGNE_MULTIPLICATOR)\n",
        "# Penalization for age \n",
        "df_age = df[['dem_age_band_75+_tm1', 'dem_age_band_65-74_tm1']].multiply(GAGNE_MULTIPLICATOR)\n",
        "# Penalization for high costs\n",
        "df_cost = df[cost_columns].multiply(COST_MULTIPLICATOR)\n",
        "\n",
        "df_y = pd.concat([df_y, df_bio, df_num, df_age, df_cost], axis=1)\n",
        "df_y_sum = df_y.sum(axis=1)"
      ],
      "metadata": {
        "id": "AEKigCWkqiK-"
      },
      "execution_count": 22,
      "outputs": []
    },
    {
      "cell_type": "code",
      "execution_count": 23,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "u0hV2Gx6XhDJ",
        "outputId": "7aaa2b19-f3b4-4a9d-cea9-93e10daca16a"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Target Variable         MSE  Race: Black MSE  Race: White MSE\n",
            "       Cost_tot 1341.219418      2406.917426      1201.210623\n"
          ]
        }
      ],
      "source": [
        "#@title Train the model\n",
        "mse_df, model_r = train_lasso(X, df_y_sum, race)\n",
        "\n",
        "risk = model_r.predict(X)\n",
        "# Print\n",
        "print(mse_df.to_string(index=False))\n",
        "\n",
        "X['risk_score_t'] = risk"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#@title Plot the results\n",
        "plot_comorbidities(X, 'New Label model')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "eWWOtXEe0CSk",
        "outputId": "2d379ccc-c411-4d04-e69d-a294dd9712c7"
      },
      "execution_count": 24,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "max: 18\n",
            "*-------------------------------------*\n",
            "Percentage of White People in the high risk section: 0.6923076923076923%\n",
            "Percentage of Black People in the high risk section: 0.3076923076923077%\n",
            "*-------------------------------------*\n",
            "\n",
            "*-------------------------------------*\n",
            "Num of White People in the high risk section: 9\n",
            "Num of Black People in the high risk section: 4\n",
            "*-------------------------------------*\n",
            "\n",
            "mean: 1.5370584695153002\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "<ipython-input-4-df564a953dd6>:41: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  df_plot1.loc[:,'risk_score_t_scaled'] = scaler.transform(df_plot1['risk_score_t'].values.reshape(-1, 1)).astype(int)\n",
            "<ipython-input-4-df564a953dd6>:43: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  df_plot2.loc[:,'risk_score_t_scaled'] = scaler.transform(df_plot2['risk_score_t'].values.reshape(-1, 1)).astype(int)\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1200x600 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "#ANN\n",
        "\n",
        "---\n",
        "The model, taken from \n",
        "ZHANG, Brian Hu; LEMOINE, Blake; MITCHELL, Margaret. *Mitigating unwanted biases with adversarial learning.* In: Proceedings of the 2018 AAAI/ACM Conference on AI, Ethics, and Society. 2018. p. 335-340.\n",
        "uses a neural network and an adversary to reduce bias w.r.t. the sensible feature.\n"
      ],
      "metadata": {
        "id": "QQjHG3X8yerV"
      }
    },
    {
      "cell_type": "code",
      "execution_count": 25,
      "metadata": {
        "id": "jzp-BH_zymjR"
      },
      "outputs": [],
      "source": [
        "#@title Clean the data\n",
        "# Load the dataset\n",
        "url = \"https://gitlab.com/labsysmed/dissecting-bias/-/raw/master/data/data_new.csv\"\n",
        "df = pd.read_csv(StringIO(requests.get(url).text))\n",
        "\n",
        "# Extract race feauture\n",
        "race_mapping = {'white': 0, 'black': 1}\n",
        "df['race'] = df['race'].map(race_mapping)\n",
        "\n",
        "race = df['race']\n",
        "# Substitute NaN values in order to use the data\n",
        "df.fillna(0, inplace=True)\n",
        "\n",
        "scaler = MinMaxScaler(feature_range=(0, 100))\n",
        "scaler.fit(df['risk_score_t'].values.reshape(-1, 1))\n",
        "df.loc[:,'risk_score_t_scaled'] = scaler.transform(df['risk_score_t'].values.reshape(-1, 1)).astype(int)\n",
        "\n",
        "scaler = MinMaxScaler()\n",
        "scaler.fit(df)\n",
        "df_notscaled = df.copy(deep=True)\n",
        "df = pd.DataFrame(scaler.transform(df), columns=df.columns)\n",
        "\n",
        "X = df.drop(['cost_t', 'gagne_sum_t', 'cost_avoidable_t','risk_score_t'], axis = 1)\n",
        "# Extract target variables (y1, y2, y3) from DataFrame\n",
        "y1 = df['cost_t']\n",
        "y2 = df['gagne_sum_t']\n",
        "y3 = df['cost_avoidable_t']\n",
        "y = df[['cost_t', 'gagne_sum_t', 'cost_avoidable_t']]\n",
        "y_r = df['risk_score_t']\n",
        "\n",
        "indices = np.arange(df.shape[0])\n",
        "\n",
        "X_train, X_val, yr_train, yr_val, race_train, race_val , indices_train, indices_val= train_test_split(X, y_r, race, indices, test_size=0.35, random_state=42)"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#@title Modules for the Adversarial NN\n",
        "import torch\n",
        "import torch.nn as nn\n",
        "import torch.optim as optim\n",
        "\n",
        "# Define the nn architectures for the main task and adversarial task\n",
        "class RegressionNN(nn.Module):\n",
        "    def __init__(self, input_dim, hidden_dim, output_dim):\n",
        "        super(RegressionNN, self).__init__()\n",
        "        self.fc1 = nn.Linear(input_dim, hidden_dim)\n",
        "        self.relu1 = nn.ReLU()\n",
        "        self.fc2 = nn.Linear(hidden_dim, hidden_dim)\n",
        "        self.relu2 = nn.ReLU()\n",
        "        self.fc3 = nn.Linear(hidden_dim, output_dim)\n",
        "\n",
        "    def forward(self, x):\n",
        "        out = self.fc1(x)\n",
        "        out = self.relu1(out)\n",
        "        out = self.fc2(out)\n",
        "        out = self.relu2(out)\n",
        "        out = self.fc3(out)\n",
        "        return out\n",
        "\n",
        "\n",
        "class AdversarialNN(nn.Module):\n",
        "    def __init__(self, input_dim, hidden_dim, output_dim):\n",
        "        super(AdversarialNN, self).__init__()\n",
        "        self.fc1 = nn.Linear(input_dim, hidden_dim)\n",
        "        self.relu1 = nn.ReLU()\n",
        "        self.fc2 = nn.Linear(hidden_dim, hidden_dim)\n",
        "        self.relu2 = nn.ReLU()\n",
        "        self.fc3 = nn.Linear(hidden_dim, output_dim)\n",
        "        self.sigmoid = nn.Sigmoid()\n",
        "\n",
        "    def forward(self, x):\n",
        "        out = self.fc1(x)\n",
        "        out = self.relu1(out)\n",
        "        out = self.fc2(out)\n",
        "        out = self.relu2(out)\n",
        "        out = self.fc3(out)\n",
        "        out = self.sigmoid(out)\n",
        "        return out\n",
        "\n",
        "\n",
        "# Define the training loop\n",
        "def train(model, adv_model, criterion, adv_criterion, optimizer, adv_optimizer, X_train, y_train, sensitive_train, epochs):\n",
        "    device = torch.device(\"cuda\" if torch.cuda.is_available() else \"cpu\")\n",
        "    model.to(device)\n",
        "    adv_model.to(device)\n",
        "    X_train = X_train.to(device)\n",
        "    y_train = y_train.to(device)\n",
        "    sensitive_train = sensitive_train.to(device)\n",
        "    for epoch in range(epochs):\n",
        "        # Forward step for the network\n",
        "        y_pred = model(X_train).squeeze()\n",
        "        loss = criterion(y_pred, y_train)\n",
        "\n",
        "        # Forward step for the adversary\n",
        "        adv_pred = adv_model(X_train).squeeze()\n",
        "        adv_loss = adv_criterion(adv_pred, sensitive_train)\n",
        "\n",
        "        # Total loss is a weighted sum of the main task loss and the adversarial task loss\n",
        "        total_loss = loss - adv_loss\n",
        "\n",
        "        # Backward step\n",
        "        optimizer.zero_grad()\n",
        "        total_loss.backward(retain_graph=True)\n",
        "        optimizer.step()\n",
        "\n",
        "        # Backward step for adversarial task\n",
        "        adv_optimizer.zero_grad()\n",
        "        adv_loss.backward()\n",
        "        adv_optimizer.step()\n",
        "\n",
        "        if epoch % 100 == 0:\n",
        "            print(f'Epoch {epoch}/{epochs}, Main Loss: {loss.item():.4f}, Adv Loss: {adv_loss.item():.4f}')\n",
        "\n",
        "# Create the models and optimizers\n",
        "input_dim = 157\n",
        "hidden_dim = 64\n",
        "output_dim = 1\n",
        "model = RegressionNN(input_dim, hidden_dim, output_dim)\n",
        "adv_model = AdversarialNN(input_dim, hidden_dim, 1)\n",
        "criterion = nn.MSELoss()\n",
        "adv_criterion = nn.BCELoss()\n",
        "optimizer = optim.Adam(model.parameters(), lr=0.001)\n",
        "adv_optimizer = optim.Adam(adv_model.parameters(), lr=0.001)\n",
        "\n"
      ],
      "metadata": {
        "id": "9wyw-rbI00DI"
      },
      "execution_count": 26,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "#@title Training\n",
        "epochs = 1000\n",
        "\n",
        "X_tensor = torch.Tensor(X_train.values)\n",
        "yr_tensor = torch.Tensor(yr_train.values)\n",
        "race_tensor = torch.Tensor(race_train.values)\n",
        "train(model, adv_model, criterion, adv_criterion, optimizer, adv_optimizer, X_tensor, yr_tensor, race_tensor, epochs)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "yc_slbyNNosV",
        "outputId": "4979316b-88ca-45f7-9238-02a0b27ac346"
      },
      "execution_count": 27,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 0/1000, Main Loss: 0.0043, Adv Loss: 0.7360\n",
            "Epoch 100/1000, Main Loss: 0.0006, Adv Loss: 0.1688\n",
            "Epoch 200/1000, Main Loss: 0.0001, Adv Loss: 0.0042\n",
            "Epoch 300/1000, Main Loss: 0.0000, Adv Loss: 0.0009\n",
            "Epoch 400/1000, Main Loss: 0.0000, Adv Loss: 0.0004\n",
            "Epoch 500/1000, Main Loss: 0.0000, Adv Loss: 0.0002\n",
            "Epoch 600/1000, Main Loss: 0.0000, Adv Loss: 0.0002\n",
            "Epoch 700/1000, Main Loss: 0.0000, Adv Loss: 0.0001\n",
            "Epoch 800/1000, Main Loss: 0.0000, Adv Loss: 0.0001\n",
            "Epoch 900/1000, Main Loss: 0.0000, Adv Loss: 0.0001\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#@title Validation loss\n",
        "device = torch.device(\"cuda\" if torch.cuda.is_available() else \"cpu\")\n",
        "X_val = torch.Tensor(X_val.values).float().to(device)\n",
        "yr_val = torch.Tensor(yr_val.values).to(device)\n",
        " \n",
        "y_loss = model(X_val)\n",
        "loss = criterion(y_loss, yr_val)\n",
        "print('Loss on validation data:', loss)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "PGy6YHqg7B7E",
        "outputId": "afe24780-5b39-4f13-d6c9-51f64db83057"
      },
      "execution_count": 28,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Loss on validation data: tensor(0.0060, device='cuda:0', grad_fn=<MseLossBackward0>)\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.10/dist-packages/torch/nn/modules/loss.py:536: UserWarning: Using a target size (torch.Size([17075])) that is different to the input size (torch.Size([17075, 1])). This will likely lead to incorrect results due to broadcasting. Please ensure they have the same size.\n",
            "  return F.mse_loss(input, target, reduction=self.reduction)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#@title Get results\n",
        "\n",
        "# create a tensor\n",
        "X_test_tensor = torch.tensor(X_train.values).float().to(device)\n",
        "\n",
        "# get predictions for the data\n",
        "y_pred = model(X_test_tensor)\n",
        "\n",
        "# convert the predictions to a numpy array\n",
        "y_pred_array = y_pred.cpu().detach().numpy()"
      ],
      "metadata": {
        "id": "z_BmsPVDyiVF"
      },
      "execution_count": 29,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "#@title Plot\n",
        "df_notscaled.loc[indices_train,'risk_score_t'] = y_pred_array\n",
        "X_toplot = df_notscaled.loc[indices_train,:]\n",
        "plot_comorbidities(X_toplot , 'ANN')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "eoVLdi4b9fJd",
        "outputId": "485f6a9c-dd61-43ac-9834-ce9faa6d51af",
        "cellView": "form"
      },
      "execution_count": 30,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "<ipython-input-4-df564a953dd6>:41: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  df_plot1.loc[:,'risk_score_t_scaled'] = scaler.transform(df_plot1['risk_score_t'].values.reshape(-1, 1)).astype(int)\n",
            "<ipython-input-4-df564a953dd6>:43: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  df_plot2.loc[:,'risk_score_t_scaled'] = scaler.transform(df_plot2['risk_score_t'].values.reshape(-1, 1)).astype(int)\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "max: 18\n",
            "*-------------------------------------*\n",
            "Percentage of White People in the high risk section: 0.75%\n",
            "Percentage of Black People in the high risk section: 0.25%\n",
            "*-------------------------------------*\n",
            "\n",
            "*-------------------------------------*\n",
            "Num of White People in the high risk section: 9\n",
            "Num of Black People in the high risk section: 3\n",
            "*-------------------------------------*\n",
            "\n",
            "mean: 3.9396702161757897\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1200x600 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Discussion\n",
        "\n",
        "---\n",
        "Acknowleging bias in tools that are used to manage the health of people results fundamental to offer a fair and optimal service.\n",
        "\n",
        "The results show that debiasing a ML model is more tricky than what it seems and static models applicable to all scenarios do not exist. \n",
        "\n",
        "Basic libraries implement some tools to detect bias and apply *trasformations* on data but they still have the need to be tuned on the specific instance in order to avoid degrading performances in accuracy; even when tuned some of them do not perform well on the specific problem.\n",
        "\n",
        "The best way to get rid of bias is to *reformulate the problem* using labels that are less affected by it, then deriving the target ones; finding a new formulation may not be possible only in some specific instances, contrary to what happened above.\n",
        "\n",
        "Bias evaluations should be carried in any context where data targets humans. \n",
        "Some other examples of models where bias arises are:\n",
        "\n",
        "*   car insurance companies (a 2021 Unipd study that showed how they were affected by bias between people born in diffent areas of Italy)\n",
        "*   job offering\n",
        "\n"
      ],
      "metadata": {
        "id": "oQqUxOtNcRpm"
      }
    }
  ],
  "metadata": {
    "colab": {
      "provenance": [],
      "collapsed_sections": [
        "UAkCD6jjQ4Cg"
      ],
      "gpuType": "T4",
      "authorship_tag": "ABX9TyNVsEV2RgGeKzq9o5yNbmbL",
      "include_colab_link": true
    },
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    },
    "language_info": {
      "name": "python"
    },
    "accelerator": "GPU",
    "gpuClass": "standard"
  },
  "nbformat": 4,
  "nbformat_minor": 0
}