desafio 3, graficos y prueba de modelos.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 74
    },
    "colab_type": "code",
    "id": "YbY9ekh6su1g",
    "outputId": "e9f8119c-4457-4a03-cd5d-fc82f71d6202",
    "tags": []
   },
   "outputs": [],
   "source": [
    "\"\"\"\n",
    "En este paso, levantamos el nuevo dataset y convertimos al tipo de dato category las columnas que consideramos categoricas\n",
    "\"\"\"\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "from sklearn.model_selection import StratifiedKFold, train_test_split, KFold, cross_val_score\n",
    "from sklearn.preprocessing import OneHotEncoder,RobustScaler\n",
    "from sklearn.naive_bayes import GaussianNB, CategoricalNB, BernoulliNB, ComplementNB, MultinomialNB\n",
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "from sklearn.metrics import accuracy_score, confusion_matrix, classification_report\n",
    "from sklearn.pipeline import make_pipeline, make_union, Pipeline\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "from sklearn.impute import SimpleImputer\n",
    "from sklearn import metrics\n",
    "from sklearn.base import BaseEstimator, TransformerMixin\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.model_selection import GridSearchCV, RandomizedSearchCV\n",
    "from sklearn.discriminant_analysis import LinearDiscriminantAnalysis\n",
    "from sklearn.svm import SVC\n",
    "from sklearn.tree import DecisionTreeClassifier\n",
    "import statsmodels.api as sm"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 305
    },
    "colab_type": "code",
    "id": "TubZ_JKR5lwJ",
    "outputId": "e3ebbfa6-8d38-4bd1-c4ea-5a3746abbcf3"
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>sexo</th>\n",
       "      <th>indice_socioeconomico</th>\n",
       "      <th>nivel_desemp_matematica</th>\n",
       "      <th>nivel_desemp_lengua</th>\n",
       "      <th>tiene_notebook</th>\n",
       "      <th>tiene_pc</th>\n",
       "      <th>tiene_tablet</th>\n",
       "      <th>tiene_celular</th>\n",
       "      <th>tiene_smartphone</th>\n",
       "      <th>tiene_consola</th>\n",
       "      <th>...</th>\n",
       "      <th>inicial+primaria</th>\n",
       "      <th>privada</th>\n",
       "      <th>secundaria</th>\n",
       "      <th>sin_discriminar</th>\n",
       "      <th>snu</th>\n",
       "      <th>bienes_y_servicios_no_personales</th>\n",
       "      <th>erogacion_de_capital</th>\n",
       "      <th>personal</th>\n",
       "      <th>transferencias_educacion_privada</th>\n",
       "      <th>transferencias_otras</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>f</td>\n",
       "      <td>medio</td>\n",
       "      <td>0</td>\n",
       "      <td>avanzado</td>\n",
       "      <td>no</td>\n",
       "      <td>si</td>\n",
       "      <td>si</td>\n",
       "      <td>si</td>\n",
       "      <td>no</td>\n",
       "      <td>no</td>\n",
       "      <td>...</td>\n",
       "      <td>14153.324289</td>\n",
       "      <td>6424.569041</td>\n",
       "      <td>9509.173296</td>\n",
       "      <td>5318.844228</td>\n",
       "      <td>2564.942087</td>\n",
       "      <td>3060.587468</td>\n",
       "      <td>1537.060193</td>\n",
       "      <td>25583.003705</td>\n",
       "      <td>6424.569041</td>\n",
       "      <td>1365.632533</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>f</td>\n",
       "      <td>alto</td>\n",
       "      <td>0</td>\n",
       "      <td>avanzado</td>\n",
       "      <td>si</td>\n",
       "      <td>si</td>\n",
       "      <td>no</td>\n",
       "      <td>si</td>\n",
       "      <td>no</td>\n",
       "      <td>si</td>\n",
       "      <td>...</td>\n",
       "      <td>14153.324289</td>\n",
       "      <td>6424.569041</td>\n",
       "      <td>9509.173296</td>\n",
       "      <td>5318.844228</td>\n",
       "      <td>2564.942087</td>\n",
       "      <td>3060.587468</td>\n",
       "      <td>1537.060193</td>\n",
       "      <td>25583.003705</td>\n",
       "      <td>6424.569041</td>\n",
       "      <td>1365.632533</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>m</td>\n",
       "      <td>medio</td>\n",
       "      <td>0</td>\n",
       "      <td>satisfactorio</td>\n",
       "      <td>si</td>\n",
       "      <td>si</td>\n",
       "      <td>si</td>\n",
       "      <td>si</td>\n",
       "      <td>no</td>\n",
       "      <td>no</td>\n",
       "      <td>...</td>\n",
       "      <td>14153.324289</td>\n",
       "      <td>6424.569041</td>\n",
       "      <td>9509.173296</td>\n",
       "      <td>5318.844228</td>\n",
       "      <td>2564.942087</td>\n",
       "      <td>3060.587468</td>\n",
       "      <td>1537.060193</td>\n",
       "      <td>25583.003705</td>\n",
       "      <td>6424.569041</td>\n",
       "      <td>1365.632533</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>m</td>\n",
       "      <td>medio</td>\n",
       "      <td>0</td>\n",
       "      <td>avanzado</td>\n",
       "      <td>no</td>\n",
       "      <td>si</td>\n",
       "      <td>no</td>\n",
       "      <td>si</td>\n",
       "      <td>no</td>\n",
       "      <td>si</td>\n",
       "      <td>...</td>\n",
       "      <td>14153.324289</td>\n",
       "      <td>6424.569041</td>\n",
       "      <td>9509.173296</td>\n",
       "      <td>5318.844228</td>\n",
       "      <td>2564.942087</td>\n",
       "      <td>3060.587468</td>\n",
       "      <td>1537.060193</td>\n",
       "      <td>25583.003705</td>\n",
       "      <td>6424.569041</td>\n",
       "      <td>1365.632533</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>m</td>\n",
       "      <td>medio</td>\n",
       "      <td>0</td>\n",
       "      <td>avanzado</td>\n",
       "      <td>si</td>\n",
       "      <td>si</td>\n",
       "      <td>si</td>\n",
       "      <td>no</td>\n",
       "      <td>no</td>\n",
       "      <td>no</td>\n",
       "      <td>...</td>\n",
       "      <td>14153.324289</td>\n",
       "      <td>6424.569041</td>\n",
       "      <td>9509.173296</td>\n",
       "      <td>5318.844228</td>\n",
       "      <td>2564.942087</td>\n",
       "      <td>3060.587468</td>\n",
       "      <td>1537.060193</td>\n",
       "      <td>25583.003705</td>\n",
       "      <td>6424.569041</td>\n",
       "      <td>1365.632533</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 43 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "  sexo indice_socioeconomico  nivel_desemp_matematica nivel_desemp_lengua  \\\n",
       "0    f                 medio                        0            avanzado   \n",
       "1    f                  alto                        0            avanzado   \n",
       "2    m                 medio                        0       satisfactorio   \n",
       "3    m                 medio                        0            avanzado   \n",
       "4    m                 medio                        0            avanzado   \n",
       "\n",
       "  tiene_notebook tiene_pc tiene_tablet tiene_celular tiene_smartphone  \\\n",
       "0             no       si           si            si               no   \n",
       "1             si       si           no            si               no   \n",
       "2             si       si           si            si               no   \n",
       "3             no       si           no            si               no   \n",
       "4             si       si           si            no               no   \n",
       "\n",
       "  tiene_consola  ... inicial+primaria      privada   secundaria  \\\n",
       "0            no  ...     14153.324289  6424.569041  9509.173296   \n",
       "1            si  ...     14153.324289  6424.569041  9509.173296   \n",
       "2            no  ...     14153.324289  6424.569041  9509.173296   \n",
       "3            si  ...     14153.324289  6424.569041  9509.173296   \n",
       "4            no  ...     14153.324289  6424.569041  9509.173296   \n",
       "\n",
       "  sin_discriminar          snu bienes_y_servicios_no_personales  \\\n",
       "0     5318.844228  2564.942087                      3060.587468   \n",
       "1     5318.844228  2564.942087                      3060.587468   \n",
       "2     5318.844228  2564.942087                      3060.587468   \n",
       "3     5318.844228  2564.942087                      3060.587468   \n",
       "4     5318.844228  2564.942087                      3060.587468   \n",
       "\n",
       "  erogacion_de_capital      personal  transferencias_educacion_privada  \\\n",
       "0          1537.060193  25583.003705                       6424.569041   \n",
       "1          1537.060193  25583.003705                       6424.569041   \n",
       "2          1537.060193  25583.003705                       6424.569041   \n",
       "3          1537.060193  25583.003705                       6424.569041   \n",
       "4          1537.060193  25583.003705                       6424.569041   \n",
       "\n",
       "  transferencias_otras  \n",
       "0          1365.632533  \n",
       "1          1365.632533  \n",
       "2          1365.632533  \n",
       "3          1365.632533  \n",
       "4          1365.632533  \n",
       "\n",
       "[5 rows x 43 columns]"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = pd.read_csv('datasets/df_final.csv', na_values=['nc'], low_memory=False)\n",
    "ponderadores = pd.read_csv('datasets/df_ponderadores.csv', na_values=['nc'], low_memory=False)\n",
    "poderador_matematica = ponderadores.ponderador_matematica\n",
    "df = df.drop('Unnamed: 0', axis=1)\n",
    "for i in df.columns:\n",
    "    if (df[i].dtype == object) or i in ['nivel_id', 'escuela_id', 'provincia_id','departamento_id']:\n",
    "        df[i] = pd.Categorical(df[i])\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 72
    },
    "colab_type": "code",
    "id": "JCfBxV5QrKKC",
    "outputId": "18822cd8-0878-48cf-a60b-9f7263f2de98"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    0.614062\n",
       "1    0.385938\n",
       "Name: nivel_desemp_matematica, dtype: float64"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.nivel_desemp_matematica.value_counts(normalize=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 926
    },
    "colab_type": "code",
    "id": "Pnm-EqnMpkyz",
    "outputId": "a507e06e-02de-4b3c-a225-a69395ce6db8"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 277573 entries, 0 to 277572\n",
      "Data columns (total 43 columns):\n",
      " #   Column                            Non-Null Count   Dtype   \n",
      "---  ------                            --------------   -----   \n",
      " 0   sexo                              277573 non-null  category\n",
      " 1   indice_socioeconomico             277573 non-null  category\n",
      " 2   nivel_desemp_matematica           277573 non-null  int64   \n",
      " 3   nivel_desemp_lengua               277573 non-null  category\n",
      " 4   tiene_notebook                    277573 non-null  category\n",
      " 5   tiene_pc                          277573 non-null  category\n",
      " 6   tiene_tablet                      277573 non-null  category\n",
      " 7   tiene_celular                     277573 non-null  category\n",
      " 8   tiene_smartphone                  277573 non-null  category\n",
      " 9   tiene_consola                     277573 non-null  category\n",
      " 10  tiene_smarttv                     277573 non-null  category\n",
      " 11  tiene_cable                       277573 non-null  category\n",
      " 12  tiene_internet_x                  277573 non-null  category\n",
      " 13  repeticion_primaria               277573 non-null  category\n",
      " 14  escuela_id                        277573 non-null  category\n",
      " 15  ambito                            277573 non-null  category\n",
      " 16  gestion                           277573 non-null  category\n",
      " 17  icse                              277573 non-null  float64 \n",
      " 18  indice_socioeconomico_medio       277573 non-null  float64 \n",
      " 19  indice_socioeconomico_medio_cat   277573 non-null  category\n",
      " 20  tecnica                           277573 non-null  category\n",
      " 21  tiene_internet_y                  277573 non-null  category\n",
      " 22  subvencion                        277573 non-null  category\n",
      " 23  dependencia                       277573 non-null  category\n",
      " 24  departamento_id                   277573 non-null  category\n",
      " 25  provincia_id                      277573 non-null  category\n",
      " 26  matricula                         277573 non-null  int64   \n",
      " 27  director_primaria                 277573 non-null  float64 \n",
      " 28  director_secundaria               277573 non-null  float64 \n",
      " 29  maestro_de_grado                  277573 non-null  float64 \n",
      " 30  maestro_inicial                   277573 non-null  float64 \n",
      " 31  profesor_secundaria_15hs          277573 non-null  float64 \n",
      " 32  profesor_snu                      277573 non-null  float64 \n",
      " 33  inicial+primaria                  277573 non-null  float64 \n",
      " 34  privada                           277573 non-null  float64 \n",
      " 35  secundaria                        277573 non-null  float64 \n",
      " 36  sin_discriminar                   277573 non-null  float64 \n",
      " 37  snu                               277573 non-null  float64 \n",
      " 38  bienes_y_servicios_no_personales  277573 non-null  float64 \n",
      " 39  erogacion_de_capital              277573 non-null  float64 \n",
      " 40  personal                          277573 non-null  float64 \n",
      " 41  transferencias_educacion_privada  277573 non-null  float64 \n",
      " 42  transferencias_otras              277573 non-null  float64 \n",
      "dtypes: category(23), float64(18), int64(2)\n",
      "memory usage: 49.7 MB\n"
     ]
    }
   ],
   "source": [
    "df.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 817
    },
    "colab_type": "code",
    "id": "oldn7ZQ2t9hT",
    "outputId": "dba1eeff-053b-454d-ebc9-5d029c76a4ea"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "sexo                                0.0\n",
       "indice_socioeconomico               0.0\n",
       "nivel_desemp_matematica             0.0\n",
       "nivel_desemp_lengua                 0.0\n",
       "tiene_notebook                      0.0\n",
       "tiene_pc                            0.0\n",
       "tiene_tablet                        0.0\n",
       "tiene_celular                       0.0\n",
       "tiene_smartphone                    0.0\n",
       "tiene_consola                       0.0\n",
       "tiene_smarttv                       0.0\n",
       "tiene_cable                         0.0\n",
       "tiene_internet_x                    0.0\n",
       "repeticion_primaria                 0.0\n",
       "escuela_id                          0.0\n",
       "ambito                              0.0\n",
       "gestion                             0.0\n",
       "icse                                0.0\n",
       "indice_socioeconomico_medio         0.0\n",
       "indice_socioeconomico_medio_cat     0.0\n",
       "tecnica                             0.0\n",
       "tiene_internet_y                    0.0\n",
       "subvencion                          0.0\n",
       "dependencia                         0.0\n",
       "departamento_id                     0.0\n",
       "provincia_id                        0.0\n",
       "matricula                           0.0\n",
       "director_primaria                   0.0\n",
       "director_secundaria                 0.0\n",
       "maestro_de_grado                    0.0\n",
       "maestro_inicial                     0.0\n",
       "profesor_secundaria_15hs            0.0\n",
       "profesor_snu                        0.0\n",
       "inicial+primaria                    0.0\n",
       "privada                             0.0\n",
       "secundaria                          0.0\n",
       "sin_discriminar                     0.0\n",
       "snu                                 0.0\n",
       "bienes_y_servicios_no_personales    0.0\n",
       "erogacion_de_capital                0.0\n",
       "personal                            0.0\n",
       "transferencias_educacion_privada    0.0\n",
       "transferencias_otras                0.0\n",
       "dtype: float64"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# ---------------------\n",
    "# para ver los nulos\n",
    "def pct_nulos(data):\n",
    "  return data.isnull().sum() / data.shape[0]\n",
    "pct_nulos(df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "ND6tltTUsu1n"
   },
   "outputs": [],
   "source": [
    "\n",
    "\"\"\"A continuación ejecutamos modelos de regresión logistica para validar que el dataset final sea adecuado para trabajar .\n",
    "\"\"\"\n",
    "# ---------------------------------------------------\n",
    "# elimino algunos nulos que quedaros \n",
    "df.dropna(inplace=True)\n",
    "# ------------------------------------------\n",
    "X = df.drop(columns=['nivel_desemp_matematica', 'escuela_id', 'ambito', 'icse',\n",
    "       'indice_socioeconomico_medio_cat', 'tecnica',\n",
    "       'subvencion', 'dependencia', 'director_secundaria', 'profesor_secundaria_15hs',\n",
    "       'profesor_snu', 'inicial+primaria', 'privada', 'secundaria',\n",
    "       'sin_discriminar', 'snu', 'bienes_y_servicios_no_personales',\n",
    "       'erogacion_de_capital', 'personal', 'transferencias_educacion_privada',\n",
    "       'transferencias_otras'])\n",
    "y = df['nivel_desemp_matematica']\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, stratify = y, random_state=42)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "HgvN7k6ysu3H"
   },
   "outputs": [],
   "source": [
    "class ColumnSelector(BaseEstimator, TransformerMixin):\n",
    "    def __init__(self, columns = 'n'):\n",
    "        self.columns = columns\n",
    "    \n",
    "    def transform(self, X, *_):\n",
    "        if isinstance(X, pd.DataFrame):\n",
    "            #c para tomar columnas categoricas, cualquier otro valor para tomar continuas.\n",
    "            if self.columns == 'c':\n",
    "              # --------------------------------------------------\n",
    "              # este codigo: pd.api.types.is_categorical_dtype(df[col])\n",
    "              # Cambio en la ultima parte lo que era igual a df[col] por X[col]\n",
    "              # Da el mismo resultado, pero ahora si le paso otro data frame no me va tirar error\n",
    "              #  Tambien, lo cambio en el else\n",
    "              # ----------------------------------------------------\n",
    "                categorical_columns = [col for col in X.columns if pd.api.types.is_categorical_dtype(X[col])]\n",
    "                return pd.DataFrame(X[categorical_columns])\n",
    "            else:\n",
    "                scal_columns = [col for col in X.columns if (pd.api.types.is_categorical_dtype(X[col]) == False)]\n",
    "                return pd.DataFrame(X[scal_columns])\n",
    "        else:\n",
    "            raise TypeError(\"Este Transformador solo funciona en DF de Pandas\")\n",
    "    \n",
    "    def fit(self, X, *_):\n",
    "        return self\n",
    "cat_pipe = make_pipeline(ColumnSelector('c'),OneHotEncoder(handle_unknown='ignore'))\n",
    "num_pipe = make_pipeline(ColumnSelector('n'),StandardScaler())\n",
    "union = make_union(num_pipe,\n",
    "                   cat_pipe)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 395
    },
    "colab_type": "code",
    "id": "QTxOgHoesu3N",
    "outputId": "45755ba1-2958-4ecf-f683-2232f8bf264d",
    "tags": []
   },
   "outputs": [],
   "source": [
    "#Este paso tarda bastante dado que ejecutamos un random grid search.\n",
    "\"\"\"\n",
    "RESULTADO: \n",
    "Best score: 0.739\n",
    "Best parameters set:\n",
    "\t log__C: 0.1\n",
    "\t log__penalty: 'l2'\n",
    "              precision    recall  f1-score   support\n",
    "\n",
    "           0       0.77      0.81      0.79     44028\n",
    "           1       0.69      0.63      0.66     29395\n",
    "\n",
    "    accuracy                           0.74     73423\n",
    "   macro avg       0.73      0.72      0.72     73423\n",
    "weighted avg       0.73      0.74      0.74     73423\n",
    "\n",
    "0.7373302643585797\n",
    "\"\"\"\n",
    "# ---------------------------------\n",
    "# Con los cambios que hice este fue el score\n",
    "# Best score: 0.772\n",
    "# Best parameters set:\n",
    "# \t log__C: 0.1\n",
    "# \t log__penalty: 'l2'\n",
    "# \t log__solver: 'saga'\n",
    "#               precision    recall  f1-score   support\n",
    "\n",
    "#            0       0.84      0.89      0.86       966\n",
    "#            1       0.69      0.58      0.63       400\n",
    "\n",
    "#     accuracy                           0.80      1366\n",
    "#    macro avg       0.76      0.74      0.75      1366\n",
    "# weighted avg       0.79      0.80      0.80      1366\n",
    "\n",
    "# 0.801610541727672\n",
    "\n",
    "\n",
    "#  el solver elegido saga, segun la documentación de sklearn funciona mejor para dasets grandes como el nuestro\n",
    "# ---------------------------------------------------\n",
    "\n",
    "# -------------------------------------------------------\n",
    "# para que la regresion logistica sea representativa de una poblacion se tienen que usar los ponderadores\n",
    "# LogisticRegression(solver='liblinear').fit(X,y,class_weight=ponderador_matematica)\n",
    "#  hay que ver como se lo agregamos al grid search ya que no me deja hacerlo \n",
    "# -----------------------------------------------------------\n",
    "\n",
    "\n",
    "\n",
    "pipeline = Pipeline([('union', union), \n",
    "                     ('log', LogisticRegression(solver='liblinear'))])\n",
    "# ----------------------------------------------\n",
    "#  agrego solver para elegir \n",
    "parameters = {'log__solver':['newton-cg', 'lbfgs', 'liblinear', 'sag', 'saga'],\n",
    "              'log__penalty': ['l1', 'l2','elasticnet','none'],\n",
    "              'log__C': [0.01, 0.1, 1, 10, 100],\n",
    "              'log__fit_intercept':[False,True]\n",
    "              }\n",
    "folds=StratifiedKFold(n_splits=5,shuffle=True, random_state=42)\n",
    "grid_search = RandomizedSearchCV(pipeline, parameters,cv=folds, n_jobs = 5)\n",
    "grid_search.fit(pd.DataFrame(X_train, columns=X.columns), y_train)\n",
    "\n",
    "print(\"Best score: %0.3f\" % grid_search.best_score_) \n",
    "print(\"Best parameters set:\" )\n",
    "best_parameters = grid_search.best_estimator_.get_params()\n",
    "for param_name in sorted (parameters . keys()): \n",
    "                    print(\"\\t %s: %r\" % (param_name, best_parameters[param_name])) \n",
    "# -----------------------------------------------------------------------------\n",
    "# habria que evitar usar los datos de testeo, primero probar con varios modelos, una vez elegido el modelo usar los datos de testeo\n",
    "y_pred = grid_search.best_estimator_.predict(X_test)\n",
    "print(classification_report(y_test, y_pred))\n",
    "print(accuracy_score(y_test, y_pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.77      0.81      0.79    127835\n",
      "           1       0.67      0.60      0.64     80344\n",
      "\n",
      "    accuracy                           0.73    208179\n",
      "   macro avg       0.72      0.71      0.71    208179\n",
      "weighted avg       0.73      0.73      0.73    208179\n",
      "\n",
      "0.7326051138683537\n"
     ]
    }
   ],
   "source": [
    "#   Agrego el sample_Weight para usar los ponderadores\n",
    "pipeline = Pipeline([('union', union), \n",
    "                    ('log', LogisticRegression(solver='saga',penalty='l2',C=0.21,fit_intercept=True))])\n",
    "lr = pipeline.fit(pd.DataFrame(X_train, columns=X.columns), y_train,**{'log__sample_weight': poderador_matematica.loc[X_train.index]})\n",
    "\n",
    "y_pred_train = lr.predict(X_train)\n",
    "print(classification_report(y_train, y_pred_train))\n",
    "print(accuracy_score(y_train, y_pred_train))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 297
    },
    "colab_type": "code",
    "id": "QGi2fbiI75fy",
    "outputId": "6cb0036c-0d0d-4158-9011-6269ab9c450d"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(33,0.5,'verdaderos')"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "matriz = confusion_matrix(y_train,y_pred_train)\n",
    "sns.heatmap(matriz,annot=True,fmt=\"d\")\n",
    "plt.xlabel(\"predichos\")\n",
    "plt.ylabel(\"verdaderos\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 283
    },
    "colab_type": "code",
    "id": "1Un8XzHVtncr",
    "outputId": "faef06b0-2736-4ad9-8c6a-ddea3e682d30"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x2a880334288>"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Histograma de probabilidades para ver si cambiando el umbral de decición puedo mejorar el accuracy \n",
    "sns.distplot(lr.predict_proba(X_train[y_train==0])[:,1])\n",
    "sns.distplot(lr.predict_proba(X_train[y_train==1])[:,1])\n",
    "ylim = plt.ylim()\n",
    "plt.vlines(0.5, ylim[0], ylim[1])\n",
    "plt.ylim(ylim)\n",
    "plt.legend(['Umbral', 'TenYearCHD = 0', 'TenYearCHD = 1'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 35
    },
    "colab_type": "code",
    "id": "Ii0LJMsT13Bu",
    "outputId": "6df174ab-29b8-4514-9cd3-0a10319edecd"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "mejor accuracy es [0.7324706142310223, 0.5053846153846153]\n"
     ]
    }
   ],
   "source": [
    "accuracy = []\n",
    "for umbral in np.linspace(0.01,0.7,40):\n",
    "  probabilidad = lr.predict_proba(X_train)[:,1]\n",
    "  y_pred_proba = probabilidad > umbral\n",
    "  accuracy.append([accuracy_score(y_train,y_pred_proba),umbral])\n",
    "  # print(f\"umbral igual {umbral}:\",accuracy_score(y_train,y_pred_proba))\n",
    "print(\"mejor accuracy es\",max(accuracy))\n",
    "# Hay que tener en cuenta que al hacer esto podemos tener problemas de overfitting, \n",
    "# si lo hicieramos en un esquema de validacion cruzada seria mejor"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 244
    },
    "colab_type": "code",
    "id": "Wz1j8b656OnM",
    "outputId": "a28e0625-e8e5-42e2-89cc-5e56a3df1ff4"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best score: 0.731\n",
      "Best parameters set:\n",
      "\t modelo: LinearDiscriminantAnalysis(n_components=None, priors=None, shrinkage=None,\n",
      "                           solver='svd', store_covariance=False, tol=0.0001)\n",
      "EN ENTRENAMIENTO: \n",
      "               precision    recall  f1-score   support\n",
      "\n",
      "           0       0.76      0.82      0.79    127835\n",
      "           1       0.67      0.59      0.63     80344\n",
      "\n",
      "    accuracy                           0.73    208179\n",
      "   macro avg       0.72      0.71      0.71    208179\n",
      "weighted avg       0.73      0.73      0.73    208179\n",
      "\n",
      "EN ENTRENAMIENTO 0.732134365137694\n"
     ]
    }
   ],
   "source": [
    "# -----------------------------------------------\n",
    "#  Ahora pruebo con un modelo naive bayes, como el union devuelve una matriz esparza primero lo aplico por fuera\n",
    "#  y luego hago el grid search lo paso como array y busco entre dos modelo de tipo naive bayes \n",
    "#  Para este grid search tuve que reducir las columnas del data frame porque sino se hace imposible de correr por el tamaño\n",
    "# ---------------------------------------\n",
    "X_naive_bayes = df.drop(columns=['nivel_desemp_matematica', 'escuela_id', 'ambito', 'icse',\n",
    "       'indice_socioeconomico_medio_cat', 'tecnica',\n",
    "       'subvencion', 'dependencia', 'director_secundaria', 'profesor_secundaria_15hs',\n",
    "       'profesor_snu', 'inicial+primaria', 'privada', 'secundaria',\n",
    "       'sin_discriminar', 'snu', 'bienes_y_servicios_no_personales',\n",
    "       'erogacion_de_capital', 'personal', 'transferencias_educacion_privada',\n",
    "       'transferencias_otras'])\n",
    "\n",
    "X_train_naive, X_test_naive, y_train, y_test = train_test_split(X_naive_bayes, y, stratify = y, random_state=42)\n",
    "\n",
    "\n",
    "X_train_union= union.fit_transform(X_train_naive)\n",
    "X_test_union= union.transform(X_test_naive)\n",
    "\n",
    "pipeline = Pipeline([('modelo', GaussianNB())])\n",
    "\n",
    "parameters = {'modelo':[GaussianNB(),LinearDiscriminantAnalysis()]}\n",
    "\n",
    "folds=StratifiedKFold(n_splits=5,shuffle=True, random_state=42)\n",
    "\n",
    "grid_search_GNB_LDA = GridSearchCV(pipeline, parameters,cv=folds)\n",
    "grid_search_GNB_LDA.fit(X_train_union.toarray(), y_train)\n",
    "\n",
    "print(\"Best score: %0.3f\" % grid_search_GNB_LDA.best_score_) \n",
    "print(\"Best parameters set:\" )\n",
    "best_parameters = grid_search_GNB_LDA.best_estimator_.get_params()\n",
    "for param_name in sorted (parameters . keys()): \n",
    "                    print(\"\\t %s: %r\" % (param_name, best_parameters[param_name]))\n",
    "y_pred = grid_search_GNB_LDA.best_estimator_.predict(X_train_union.toarray())\n",
    "print(\"EN ENTRENAMIENTO: \\n\",classification_report(y_train, y_pred))\n",
    "print(\"EN ENTRENAMIENTO\",accuracy_score(y_train, y_pred))\n",
    "resultados_NB = pd.DataFrame(grid_search_GNB_LDA.cv_results_)\n",
    "#  Se ve el algoritmo linear discriminant analysis anda bien.\n",
    "#  Aqui la documentación https://scikit-learn.org/stable/modules/generated/sklearn.discriminant_analysis.LinearDiscriminantAnalysis.html"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "28NqZORjH5_o"
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>param_modelo</th>\n",
       "      <th>params</th>\n",
       "      <th>split0_test_score</th>\n",
       "      <th>split1_test_score</th>\n",
       "      <th>split2_test_score</th>\n",
       "      <th>split3_test_score</th>\n",
       "      <th>split4_test_score</th>\n",
       "      <th>mean_test_score</th>\n",
       "      <th>std_test_score</th>\n",
       "      <th>rank_test_score</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>GaussianNB(priors=None, var_smoothing=1e-09)</td>\n",
       "      <td>{'modelo': GaussianNB(priors=None, var_smoothi...</td>\n",
       "      <td>0.687098</td>\n",
       "      <td>0.688323</td>\n",
       "      <td>0.688419</td>\n",
       "      <td>0.688611</td>\n",
       "      <td>0.690020</td>\n",
       "      <td>0.688494</td>\n",
       "      <td>0.000930</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>LinearDiscriminantAnalysis(n_components=None, ...</td>\n",
       "      <td>{'modelo': LinearDiscriminantAnalysis(n_compon...</td>\n",
       "      <td>0.729153</td>\n",
       "      <td>0.729081</td>\n",
       "      <td>0.729609</td>\n",
       "      <td>0.729609</td>\n",
       "      <td>0.733397</td>\n",
       "      <td>0.730170</td>\n",
       "      <td>0.001629</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                        param_modelo  \\\n",
       "0       GaussianNB(priors=None, var_smoothing=1e-09)   \n",
       "1  LinearDiscriminantAnalysis(n_components=None, ...   \n",
       "\n",
       "                                              params  split0_test_score  \\\n",
       "0  {'modelo': GaussianNB(priors=None, var_smoothi...           0.687098   \n",
       "1  {'modelo': LinearDiscriminantAnalysis(n_compon...           0.729153   \n",
       "\n",
       "   split1_test_score  split2_test_score  split3_test_score  split4_test_score  \\\n",
       "0           0.688323           0.688419           0.688611           0.690020   \n",
       "1           0.729081           0.729609           0.729609           0.733397   \n",
       "\n",
       "   mean_test_score  std_test_score  rank_test_score  \n",
       "0         0.688494        0.000930                2  \n",
       "1         0.730170        0.001629                1  "
      ]
     },
     "execution_count": 61,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "resultados_NB.iloc[:,4:]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.71      0.67      0.69    127835\n",
      "           1       0.52      0.57      0.55     80344\n",
      "\n",
      "    accuracy                           0.63    208179\n",
      "   macro avg       0.62      0.62      0.62    208179\n",
      "weighted avg       0.64      0.63      0.64    208179\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# Corro el gaussian para ver los scores\n",
    "\n",
    "gaussian = GaussianNB()\n",
    "gaussian.fit(X_train_union.toarray(), y_train,sample_weight = poderador_matematica.loc[X_train_naive.index])\n",
    "# pipeline.fit(X_train_union.toarray(), y_train,**{'log__sample_weight': poderador_matematica.loc[X_train.index]})\n",
    "\n",
    "y_pred_train_gaussian = gaussian.predict(X_train_union.toarray())\n",
    "print(classification_report(y_train, y_pred_train_gaussian))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "zMFsy03FIglp"
   },
   "source": [
    "### Comentarios\n",
    "Era de esperase que el modelo gaussian_NB sean malo, ya que supone que las features no se correlacionan, cosa que es muy poco problable. Ya que un chico que se encuentra en una clase economicamente baja es de esperarse que no tenga celular, computadora y varios elementos mas que aparecen como variables dummy.\n",
    "Por otra parte el algoritmo LinearDiscriminant Analysis funciona muy bien, llegando a resultados parecidos al de la regresión logistica en entrenamiento.\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "QyznDQjD6RJz"
   },
   "source": [
    "## Graficos"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Primero vemos el balance de las clases que tenemos en el data frame"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5,1,'Desempeño en Matemática')"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXUAAADECAYAAABgHLGsAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvhp/UCwAAIABJREFUeJzt3Xd4VGX2wPHvyQQCKZQECIQWSnqkI0U6KKKQFZFV0KUo66o/xYK9gKgrumtB1rbKqhQ7tqiAK0qUIrhEehmKhE4gQBoJafP+/rgTDJAGJDPJ5HyeZx6SueU9cyeceee9955XjDEopZTyDF7uDkAppVTF0aSulFIeRJO6Ukp5EE3qSinlQTSpK6WUB9GkrpRSHkST+llE5EER2SMi0SKy1N3xVEci8qaIPF/GOn1FxO6qmJSqKcqd1EUkSUSyRSRDRFJFZKWI3CYinvbB0BkYBMwCfnBzLIhIqIgYEfntrOcbiUiuiCSVcz8TRGR5pQR5Zjv9gAjg0bOeNyLSvvB3Y8wyY0xEZcejVE3jfZ7rjzDGLBGR+kB/4BWgBzCxwiNzE2PMGOePQ9wayLn8RCTWGLPJ+ftYYDfg48aYitMaGGOMKXB3IErVRBfUyzbGpBlj4oHrgfEiEgsgIj4i8oKI7BWRZOfX8LrOZY1E5BtnL/+4iCwr7OWLSIiIfCYiR0Vkt4hMLmxLRJ4UkU9FZL7zW8JGEQkXkUdE5IiI7BORK4qsnyAiM0TkVxFJE5GvRCSwyPKezm8ZqSKyXkQGnLXt0yKywtnWf0WkUZHlcSKy2bltgohElXSMRCRSRL53vla7iPy5yLL3ROQ1EfnW2c5qEWlXxmGfB4wv8vs4YO5ZbT4sIruc+9wiIiOdz0cBbwK9RCRTRFLL8X4NEJH9zuGoIyJySESuEZGrRGS783U9WqTtS0XkF+BfwFoReVVEajuX/excbb2z/esL919k+5Yi8rnzb+CYiLzqfL6diPzofC5FRN4XkQZlHCulai5jTLkeQBIwpJjn9wK3O3+eCcQDgUAA8DUww7lsBlZiqeV89AUE64MlEZgK1AbaAr8DQ53bPQmcAoZifbOYi9VDfcy5n78Cu4vEkwAcAGIBP+AzYL5zWXPgGHCVs93Lnb83LrLtLiAcqOv8/TnnsnDgpHObWsCDwE6gdjHHxA/Yh/UNxhvoAqQAMc7l7wHHgUudy98HPirhuIcCxvnvPsAGRAF2rG8TSUXWHQ2EOF/b9c54mzmXTQCWn7Xv0t6vAUC+830pPM5HgQ+c68Y435e2zvW7Ab2crycU2ArcU6QtA7Qv8vsAYL/zZxuwHnjZeezqAH2cy9o7j7kP0Bj4GZhZ3r9bfeijpj3Kv2LJSX0VVoIVZxJpV2RZL5wJF3gK+Krof2zn8z2AvWc99wjwrvPnJ4HviywbAWQCNufvAc6E0cD5ewLOROz8PRrIdSaOh4B5Z7X1HTC+yLaPF1l2B7DY+fMTwCdFlnlhfXgMKOaYXA8sO+u5fwPTnD+/B8wusuwqYFsJxz3U+fq8gSVYH27POY/5GUm9mG3XAX9y/jyBIkm9HO/XACC7mOPco8j6icA1JbR9D/BFkd9LS+q9sD4wvMvxd3gNsNbd/3H0oY+q+jjfMfXiNMfqdTYGfIFEESlcJljJFOCfWAn6v87lbxljnsMagw0pHBJwsgHLivyeXOTnbCDF/DFmm+381x8o3Me+IuvvweppNnK2NVpERhRZXgsoepXL4SI/Zzn3C1YPeE/hAmOMQ0T2OV//2VoDPc56Td5YQyhltVOauVjJuTfQDwgrulBExgH3YX0Q4NxnI4pX1vsFcKyY43z2e+HvbLst8A+sbyXezn1vLcdrAmgJ7DHG5J+9QESaYJ207ov1weIFnCjnfpWqcS4qqYtId6ykthxreCEba4jhwNnrGmMygCnAFBGJAZaKyP+wEvBuY0zY2dtchJZFfm4F5Dnj24fVU//rBezzIHBJ4S9iZcKWWL31s+0DfjLGXH4B7ZTmM+BVINEYs0dETh8zEWkNvA0MBn4xxhSIyDqsRA1WT7moUt+vC/BvYA1wozEmR0TuBUaVc9t9QCsR8S4msc/Air2DMeaYiFyDdQyUUsW4oBOlIlJPRIYDH2GNV280xjiwksrLzt4VItJcRIY6fx4uIu2dyTAdKHA+fgXSReQhEakrIjYRiXV+YFyom8S6ztwXa9hngbPHOR8YISJDne3UcZ6wa1GOfX4CXC0ig0WkFtYHVA6wsph1vwHCReQvIlLL+ehe2onV8jDGnMS63HJSMYv9sJLfUQARmYh1XqFQMtCi8ORlWe/XBWiAdTxyRSQSuO2s5clY50uK8ytwCHhORPyc78tlzmUBWMNtqSLSHHjgAuNTqkY436T+tYhkYPWsHgNe4szLGR/COnm4SkTSscaAC69FDnP+ngn8ArxujElwJtsRQCesE6ApwGyg/gW9Iss8rHHrw1gn3SYDGGP2AX/Cuob6qPN1PEA5joMxxg7chHV1R4oz5hHGmNxi1s0ArgBuwOrhHwaepwIuPzTGrDHG7Crm+S3Ai1jHNhnrW8WKIqv8CGwGDotIivO50t6v8zUF6/VmYH1YfHzW8ieBOc4rh/5cdEGRv4H2WCfe92OdlwCYjjWkkwZ8C3x+gfEpVSOIMZ41SYaIJGB9e5jt7liUUsrVPO1uUKWUqtE0qSullAfxuOEXpZSqybSnrpRSHkSTulJKeRBN6kop5UE0qSullAfRpK6UUh5Ek7pSSnkQTepKKeVBNKkrpZQH0aSulFIeRJO6Ukp5EE3qSinlQTSpK6WUB9GkrpRSHkSTulJKeZCLmnhanUmmizfQAmuy61ZAMNYHpznrgfPfbKy5OQ9gTXuXbKYZh4vDVkp5EK2nfgFkugQAlwGXYQjDEAq0QghGLuLbjyEHYRewA9gOrAUSzDRzqALCVkrVAJrUy0GmSxDQl3wGAYOwEXVRyfv87QASCh9mmjnowraVUtWIJvUSyHTpRB7jEYZjox2CuDumInYCPwAfmmnmJ3cHo5SqOjSpFyHTpQWnuAMbY6lFa3fHU047gfeAOWaa2e/mWJRSblbjk7pMFxvZXItwDz70dPGwSsUxOBC+B94BvjLTTI67Q1JKuV6NTepyh9jwYQp+PEAtGrk7ngp2HPgX8KKZZjLcHYxSynVqXFKX60RoxmQCeBwfj0vmZ0sBZgCvm2nmlLuDUUpVvhqV1OUuuRk/nqYuIe6OxcX2A9OBd800U+DuYJRSladGJHW5Q0YSwEv4EuruWNxsO/CEmWY+cXcgSqnK4dFJXa6QZoQzj0YMrlIXJLrfD8AtZprZ4+5AlFIVyyOTukSI0JVxtGImdWng7niqqEzgQeBNM80D/wiUqqE8LqnLAGlIJO/TlCur2A1DVdV/gfFmmjns7kCUUhfPo5K63CAjacVb+Hn8VS0V7QgwwUwzi9wdiFLq4nhEUpcI8aU7L9GOv+JVTW8ecjeDQfgn8LAOxyhVfVX7pC4R0oSefEIb+utgS4X4GBhnpplcdweilDp/1TqpSwdpRy++IoQYd8fiYRKAa8w0k+buQJRS56faJnXpKT24lE8JoqW7Y/FQm4BhWiRMqeql2iV1iRChGdfQjbcJIMjd8Xi4/cCVZprZ7O5AlFLlU61OKkqEeNGMv9KLuZrQXaIFsFymy2XuDkQpVT7VJqlLhHjRkDvpwT+pg7+746lBGgBfy3SJcncgSqmyVYukLhEi+DOBPkzDl3rujqcGaggskunSzN2BKKVKV+WTukSIUItr6MvfCSDQ3fHUYK2Bhc5Jt5VSVVSVT+pAP/rwTxrS1N2BKDoBC2S6eLs7EKVU8ap0UpcIieFSXqYZ7dwdizrtCuBtdwehlCpelb2kUSKkJWG8RXeurLRGsoF4rMonAvwJSMe69eYo8FegeSnbO4C3gADgRudznwHJQDgwxPncT0AwEFmh0bvbE2aaecbdQSilzlQle+oSIf748zCd6F+pDS0G2gN3AbcBjYAmwPVYI8hlWeXcplBhncM7gL3AKSADOICnJXSAJ2W69HB3EEqpM1W5pC4RIsAYejOcWtSttIZOAXuALs7fvYG6QGMoV43HNGBHke0BbEAeVg++AKv3vxQYWDEhVzE2YK5MF193B6KU+kOVS+pAVy5hIo1oVamtnAB8gS+BN4GvgPMpYbUYuBzOKCLWGKgP/BuIAY4DBvDcCwHDgX+4Owil1B+qVFKXCAkkiHuJplulN+YADgHdsYZeagPLy7mtHfCDYqevHgbcDvTmj176z8AnQOLFhVwlGe6Q6XKFu8NQSlmqTFKXCPHCi5vpxWBs1Kr0Bus5Hy2cv0djJfny2IeV2F8GFgC7sU6QFrUNK+nnYZ2I/TOwnvP7NlAdWLNLvSPTRacNVKoKqErXGw+gC2OoR7BLWgvAGipJwRpD/x1r+KQ8hvDHlS27gZXAqCLLC7BOoo7FGoIpZJzLPE9z4DX+uAZIKeUmVaKnLhESgh+30I5YlzY8DKuH/TrWlSt9ga3Ai1j1CT8A5jnXTQfml3O/v2LdplMbTn9EvQ60gko89etuY2W6DCl7NaVUZXL7derOq10m049JtHBxUlcVbS3QVafDU8p9qkJPvR0N6EsIWgWw+usM3OTuIJSqydya1J299OvpSixe2NwZi6owz8h0qe3uIJSqqdzdU4+hCd1pQrib41AVpxVws7uDUKqmcltSlwixATfQhQ7Oy+KU53hEe+tKuYc7e+qdaU4nAmnjxhhU5WgFTHR3EErVRG5J6hIh3sANRBLqjvaVS9zj7gCUqonc1VOPwoemNNaxdA8WKdOlp7uDUKqmcVdSH0QkzfFyQTkA5U4T3B2AUjWNy5O6REhDoCOhhLm6beVyN8h08dx7aJWqgtzRU+9GExrgV+qcQsoz1AdGujsIpWoSlyZ1iRAv4AoiK7lWuqpK9CoYpVzI1VUa22GjCU21JECNYRgk06WlmWb2VXZTiYmJTby9vWcDsbj/xjqlKoMD2JSfnz+pa9euR4pbwdVJvQ+tCMIbnQKtphC8gBuAf1Z2U97e3rObNm0a1bhx4xNeXl5aVEx5HIfDIUePHo0+fPjwbCCuuHVc1puRCKkF9KIlga5qU1UZA1zUTmzjxo3TNaErT+Xl5WUaN26cBiVXtHXlV9TWgI1A2rqwTVU19JHp4oqCbV6a0JWnc/6Nl5i7XZnUI/GjDnVdNLORqkrqYU0bolzIZrN1jYyMjA4LC4sZNGhQ+5SUlDI/WJ955pkmbdu2jYmLi3N5+Y6kpKRaV155pXb6LpIrk3p3WhOgpbtqrP7uDqCm8fHxcWzbtm3Ljh07Njdo0CD/n//8Z5kTNv7nP/9pvHDhwh3x8fG7y9NGXl7exQfqFBoamrd48eLfK2yHNZRLkrpEiB/QkiblngVUeR5N6m7Us2fPkwcOHDhdOfOJJ54Ijo2NjQoPD4++9957QwDGjh3bav/+/T5xcXHtp0+f3iQ9Pd1r9OjRobGxsVFRUVHR8+fPbwAwa9asoGHDhrUdNGhQ+759+4aXtD+73V67bdu2MTfccEPr9u3bx1x22WVhmZmZArBp0yaf3r17h0dERERHR0dHbd682cdut9cOCwuLKdy2a9euEdHR0VHR0dFR33//vZ+rj1l15aqeekvAUF+vT6/B+sp00csM3SA/P5+lS5cGXHPNNakAn3/+eb2dO3fW2bBhw9atW7duWbdune+iRYv8P/jgg71NmjTJ++mnn7ZPmzbtyKOPPtps4MCB6Zs2bdq6bNky++OPP94iPT3dC+C3337z//DDD3evWrVqe0n7A9i7d2+dyZMnH9m5c+fm+vXrF8ydO7chwNixY9vcdtttR+x2+5Y1a9Zsa9Wq1Rld/pCQkPxly5Zt37Jly9aPP/7493vvvVdzRzm56pLGNtjwwpemLmpPVT0NgUuA9e4OpKbIycnxioyMjD5w4EDt2NjYrGuuuSYdYPHixfV+/vnnetHR0dEAWVlZXtu2baszbNiwzKLbJyQk1Pvuu+8azJo1q6lzf7Jz587aAH379k0PDg4uKG1/bdu2zW3evHlO7969swE6d+6clZSU5HPixAmv5OTk2uPGjUsF8PX1NcAZJ7hzc3Pllltuab1ly5a6Xl5e7Nmzx6dSD5YHcVVS70AgXs5rllXN1QFN6i5TOKZ+7Ngx2xVXXNH+ueeea/L4448fMcZwzz33HHrggQdSStveGMOCBQt2duzYMafo88uXL/fz9fV1FF2vuP3Z7fbatWvXPp2sbTabyc7O9irPZPd///vfg5s0aZL32Wef7XY4HNStW7druV94DeeqJNuChuhMOKq1uwOoiYKCggpmzZq197XXXgvOycmRYcOGpc+bN69RWlqaF8Du3btrHThw4JwO3sCBA9NffPHFYIfDyt8rVqwotjhbefdXKDAw0NG0adPcefPmNQDIzs6WjIyMM3JRWlqarVmzZnk2m43XX389qKCg4IJff01T6UldIsQH8KMeAZXdlqryNKm7yWWXXZYdFRWVPXv27IbXXntt+ujRo4937949Mjw8PHrkyJHtUlNTz7nc8bnnnjuYn58vhZdFPv7448UW4Svv/oqaP3/+7tdee61JeHh4dLdu3SL37dt3xofAPffcc+TDDz8M6tixY+T27dvr1K1b11HSvtSZpDxfhS6qgQhpCjxDPyJpwWWV2piq6r4308wVlbXz9evXJ3Xs2LHUIQWlPMH69esbdezYMbS4Za4YfmkAQF0auqAtVbWFujsApTydq5K6F3WcyV3VXIZWMl309jOlKpErknoTwIGP9tRrPMEHtEyEUpXJFUk9BBu5eKPTminQk6VKVSpXJPVgvMh1QTuqevB3dwBKeTJXJPXaCFoOVRWq5e4AlPJkrkjqeh+pKsqjk/pDDz3UtH379jHh4eHRkZGR0T/++GOphahmzZoVlJSUdPqYXH/99a0TExPrlLT+2rVr60RGRkZHRUVFb968+bxunf/mm28CLqQw1vvvv1//0UcfrZQSH4MHD27XqVOnyMrY98WYNWtW0Lhx46plvRlXlAkQLbcLOCggnd1QQ7+1FFAXX/ZTF5ddRy7TpUJvLTfTTGJpy5csWeL33XffNdi4ceOWunXrmkOHDnnn5OSU+tc/f/78Rp06dcoODQ3NA/j444/3lLb+p59+2mDYsGGpL7/88sHzjf/HH38M8Pf3L7j88stPlnebvLw8brzxxjQg7XzbK0tKSopt8+bNfr6+vgXbtm2rHRkZqcO0FcBVSb1mp/V8cljNUvZw2N2huFED4F/Gbn5xdyCV5cCBA7UCAwPz69atawCaNWuWX7js/vvvb7Z48eIGOTk5Xt26dct8//3398yZM6fhpk2bfMeNG9e2Tp06jjVr1mwdNGhQ+AsvvLCvd+/eWddff33ohg0b/ETE3HjjjSmRkZE5b731VrCXl5dZuXKl/+rVq7cPGTKk3aFDh2rn5OR43Xbbbcn3339/CsCCBQvqTZ06tXlBQYEEBgbmv/fee0lz585t7OXlZT755JOgmTNn7m3btm3u+PHjQ48dO+YdFBSUP3fu3KSwsLDcUaNGhTZs2DB/48aNvh06dMi65JJLstesWeM3d+7cvdu3b69d3DaFr7OgoIC2bdvG/vLLL9tCQkLyCwoKaNOmTezq1au3FT0eAPPmzWs4ZMiQ1ODg4Lw5c+YEzpgx4/CxY8dsHTp0iN67d+9Gm81GRkaGV1hYWOyePXs2vvrqq0Hvvvtu47y8PAkNDc1ZsGDB7oCAAMeoUaNCAwICCtavX+939OjRWk8//fT+iRMnnrjnnntCFi9e3ADg+PHj3v369UtfsGBBUknH7JVXXgl6+eWXmzVu3DivXbt2pwrr1pT1mqsa7alXtlNkspJ5HOYzd4dSBaxzdwCV6ZprrkmfMWNGSGhoaGyfPn3Sx4wZc/zqq6/OBHjggQeOvPDCC4ec67X56KOP6k+cOPHEG2+80eSFF17Y169fv6yi+/rll198Dx06VGvHjh2bwerVNmrUqGD16tVH/f39C5566qlkgPfffz8pODi4IDMzUzp37hx90003nXA4HHLnnXeGJiQkbIuMjMxNTk62BQcHF4wbN+6MbQcNGtR+7Nixx+66665jM2fODLr99ttbLlmyZBfArl276qxYsWK7t7c3s2bNCiqM67bbbmtV0jYANpuN66677tjs2bMDp06deuSrr76qFxUVlX12Qgf49NNPA6dOnXowJCQk77rrrms3Y8aMw0FBQQWRkZFZCxcuDBgxYkTGRx99VL9///5pPj4+5sYbbzwxZcqUFIDJkyeHzJo1q9Fjjz12BCA5ObnWmjVrtq1bt67OyJEj20+cOPHEzJkzDwIHjx07ZuvVq1fE3XfffaSkY5aTk+P13HPPhSQmJm4NDAws6N27d0RsbGxWeV5zVeOKpO5VY9N6BsdYxiuk8qKxG+s/rcilQE0tIxqDde/RbozZ7+5gKlr9+vUdmzZt2rJ48eKAH374IWD8+PHtpk6dun/y5MnHFi1aFPDSSy81PXXqlFdqaqp3dHR0NqUMaURGRubs27fPZ/z48S1HjBiRNnLkyPTi1nv++eeDv/322wYAhw8frrV58+Y6ycnJ3pdeemlG4XBGYYncs61du9Zv0aJFuwBuv/3249OnT29RuOzaa6894e19bnoobZtCt99+e0pcXFz7qVOnHnnnnXcaTZgw4Zwht3379nnv2bPH54orrsj08vLC29vb/O9//6vTvXv3U6NHjz7x4YcfNhwxYkTGJ598EnjHHXccBUhMTKw7derU5hkZGbaTJ0/a+vfvf/r4xcXFpdpsNrp27Xrq2LFjp89ROBwOrrvuujb/93//l9y3b9+sko7ZwYMHa/Xs2TMjJCQk3/n6j2/fvr1OeV9zVeKannoeFTfnVXWRwn6WMZ1s5hi7yRsv0vIfMCsYrnF3aFXAQ8A/3B1EZfD29mb48OEZw4cPz+jQoUP2vHnzgiZNmnR8ypQprVevXr2lffv2effdd1/IqVOnSr18oHHjxgWbNm3a8sUXX9R7/fXXm3z88ceBn376aVLRdb755puAn376KWDNmjXbAgICHJdeemlEYWlbucgbd/39/S+4gFb79u3zGjVqlB8fHx+wdu1avy+//PKcKermzJkTmJ6ebmvZsuUlAJmZmbZ58+YFdu/e/eCYMWNSn3rqqebJycm2TZs2+Y4YMSId4NZbb22zYMGCnb169cqeNWtW0E8//XS6SGCdOnVOn6sqWs9qypQpIc2aNcu9++67j0HJxwy46GNWVbjiuhRDDvnkc8oFbVUN+7DzA7eTzX+M3eTdL9LjCVihCf00j6yjun79ep+NGzee/ha2du3aui1atMjNysryAmjatGl+Wlqa19dff3367mp/f/+CtLS0cyoaHjp0yLugoIAJEyakPvPMMwc2btzoe/Y6qamptvr16xcEBAQ41q5dW2f9+vV+AAMHDjy5evXqgG3bttUGSE5OtgEEBAQUZGRknG6rc+fOJ2fPnt0Q4N///ndgt27dMs9u42zl3ebmm28+OmnSpDZxcXHHi+vxL1iwIPCLL77YceDAgY0HDhzYuHr16i1ffvllIFjfeDp27Hjyb3/7W6vBgwenFW6flZXl1apVq7ycnBz56KOPAsuK9cMPP6yfkJBQ75133tlX1jHr16/fyVWrVgUcPnzYlpOTI1988cXp9+hCjpM7uaKnfhxoQh7peFPipVoewQDb+R+J3Gns5tc4EXlaZNT98FZTKPOPsAbxyA/49PR02+TJk1ulp6fbbDabCQ0NzZkzZ86eRo0aFdx4441Ho6OjY1q0aJHbsWPH01efjBs3LuWuu+5q/cADDzjWrFmztfD5pKSkWrfcckuow+EQgKeeeuqc4apRo0alvfXWW43Dw8Oj27Vrd6pwvyEhIfmzZs1KGjlyZHuHw0FQUFDeypUrd4waNSr1uuuua7do0aIGM2fO3PvGG2/sHT9+fOgrr7zStPAEYFmvsbzbjBkzJu3OO++03XrrrcfOXma322sfPHiw9qBBg04fh8jIyFx/f/+CH3/80W/QoEEn//znP5+4+eab237zzTf2wnUefvjhg5deemlU8+bNc6OiorIyMzNLLe87c+bM4CNHjtTq1KlTFMCVV16ZOmPGjEPFHbPWrVvnPfTQQwd79uwZ1bhx47wOHTpkFRQUyPm85qrCFaV3bwa6cRVDaEBYpTbmTg4KWMcPbGOysRt7nIhtCNw9AZ6qBzpp7pnGYMxHFb1TLb1bdfz888++9957b8vExER72Wur81Va6V1X9NSTAR9OVfx1rlVGHqf4lS/Yw/3Gbg7GifhcC8/fAHfU8fCbbS6QJl4P9uijjzZ97733Gr/77ru73R1LTeSKpH4MEE5R7Nn7ai+bdFbwLkd40thNapxIvZvhPyPgWpvrpgusbjSpe7Bnn3328LPPPluT78lwK1ck9TTAwUkP7Kmnc5SfeYl0XjF2kx0nEjwFPuuPzvBUhnPGWZVSFcMVSd3qoZ/0sJ76EfawjGnk8L6xm/xJIhFPw5cdocrVsahKDOSLNSSnlKoEruqpe5HMERe05Rp72MxK7sPwvbEb84hIv8fgwzYQ4u7Qqrp82FPLmCp7i7VS1Z0rkvpJIJcMcskhFZ9qPK2dwbCVVazjTmM3v8WJyLMiN94LrzahGr8uFxLY5O4YlPJklX4iz9iNAXYA9cig+t4aXkA+iSxiHX9xJnTvofDwZHhbE3r52WCDu2NwtzfeeCNw165delWUB6oK762rrs7YBPhxgn1lrlkV5ZLNSj5gO7cYu9kVJ1L3zzDrr/CUPzpN3/kQ2OK6xqRrhT7KwWazdY2MjIyOiIiIjo6Ojjq7fvmPP/7ot2TJknrt2rUrtXRGWXXVq4NvvvkmYODAge1LW+dC6pb379+/fUpKSqk3Hl1MeyWtc99994VMnTq1xDl2q8p764rhFwCrRvRB9lS724+ySGU5b5PCM8Zu0uNEGtwKc6+C4V7U0EJlF2ezuwOoTD4+Po5t27ZtAfjss8/qPfrooy0uv/zy0zfg/P7777Xfe++9vWXtp6y66lWBw+HAGIPNdkH59YL99NNPO13aYDlVlfdN5Y26AAAT30lEQVTWVT31/YBwgCPkk1Xm2lVFKodZwtOk8ISxm/TrREIehu+GwwhN6OfPWDVfaswdhmlpabb69evnO3/26tWrV/hLL73UtEuXLlHz589vAJCenu41YMCA9hEREdFhYWExb7/9dkOASy+9NOLnn3/2Bas2enR0dFRERER0r169wsGq5zJkyJB24eHh0R07doxcvXr1Od8Yu3btGrFy5crTz3fp0uWc9WbNmhU0ePDgdn379g0LDQ2NnTJlSrPCZU8++WRwWFhYTFhYWMxTTz3VBKxb/Nu2bRtz0003tYqJiYnetWtX7aL7W7BgQb02bdrEdO3aNWLBggWnhyXT09O9Ro8eHRobGxsVFRUVXfj6wapDX1z7Q4YMaRcTExPVvn37mBdeeKFR4fPNmze/5NChQ94lxXi2V155JSg0NDS2e/fuEStXrjw9R+7Bgwe9hw4d2i42NjYqNjY26r///W+Zd35v2LDBt2fPnuGtW7eOffHFFxuBe97b0rikp27sJlMiZB/gTxpJBBHtinYvymF2s5zHyOUTYzcFd4jEPAlfxkKpXydVyfJhp6df+ZKTk+MVGRkZnZOTIykpKbUWLly4HcDX19fx7bff7gwMDHQcOHDAu2fPnpFjx45N/fzzz+s1bdo0LyEhYSfAsWPHzuj2Hjx40Pvs2ugADz74YEjHjh2zlixZsis+Pj5g/PjxbQq/IRSaMGFCyuzZsxv17t1734YNG3xyc3OlR48e2WfHvGHDBr+NGzdu9vf3d3Tu3Dn6T3/6U5qI8MEHHwQlJiZuNcbQtWvXqMGDB2c0atSoICkpqc7bb7+dNH/+/DN6pVlZWXLnnXeGfv/99/aYmJic4cOHty1c9uijjzYbOHBg+qeffpqUkpJi69atW1RcXFx6Se3369cvq7i6502bNj1dDG7ZsmW+xcV42WWXnX6Ne/bsqVVSnfS//e1vLe+7777koUOHZu7YsaP20KFDw37//fdSv0lu3bq1bmJi4taMjAxb586do0eNGpXWvHnzPFe/t6Vx5R2PiUB9UjinDGeV8zsbWMot5PKRsZuCx0Uufxh+0IR+cQpgobtjqGyFwy+7d+/e/MUXX+yYOHFiG4fDqmL74IMPhnTp0iVy5MiR7VJSUmrt37/fu0uXLtnLli2rd/vttzdfvHixf1BQ0BkVLBMSEvyKq43+66+/Btxyyy3HAOLi4jJSU1O9z04aEyZMOLFkyZL6OTk58uabbzYaO3ZssXfy9unTJ71p06YF/v7+5uqrrz6RkJDgn5CQ4H/VVVel1qtXz1G/fn3H1VdffWLp0qUBAM2aNcsdPHjwOVPirVu3rk6LFi1yLrnkkhwvLy9uvPHG0zeZJSQk1Hv55ZebRUZGRvfp0yciJydHdu7cWbuk9sGqex4RERHdtWvXqMK652cdmxJjLPTzzz/7FdZJr1Onjrn22muPFy5bsWJFvbvvvrtVZGRk9IgRI9pnZmbaTpw4UWpOHDZsWKq/v79p1qxZfq9evdKXLVvm5473tjSuGlMH2A4I29lKGFfhVQVvoTc42MwKNnCnsZsNcSLyvMjN98LLQVDP3eFVd7XgE3fH4EpDhgw5eeLECe9Dhw55f/755/WPHDlS69dff93m7e1NSEjIJdnZ2V4dOnTI+e2337Z89tln9R977LHmS5YsSS+cIQms2uDF1fkurhCfiJzxZEBAgKNv377pH3zwQYP4+PjAxMTEYnt7Z+9fRIrdfyFfX98Sa62XVJPcGMOCBQt2duzYMafo88uXL/crrv3S6p4X3Wd5lBbTmjVrtvr7+5e7qmFxsb755ptBrn5vS+PKxLobyCODPNLY7sJ2y6eAPP7H12xgnDOhe18F0+6E1zShX7xcOG6D1e6Ow5XWrl1bx+FwEBwcnH/ixAlbw4YN8729vfn6668DDh06VBusErsBAQGOO+644/g999yTvG7dujPqppdUG71nz54Z7777bhBYV5k0bNgwPzAw8Jxke9ttt6U89NBDLTt27HiypBmQli9fXi85OdmWmZkpCxcubNC/f//MQYMGZS5cuLBBRkaGV3p6utfChQsbDhw4MKO019upU6dT+/fvr71582YfgKI1zwcOHJj+4osvBhd+a1mxYsXpceLi2i+p7nlR5YmxtDrpffr0SX/++edPj8MXPf9QkkWLFjXIysqSw4cP21atWhXQp0+fk+56b0visp66sZsciZBlwAD2sJ6GVeh2+hxOsoqPOMAjxm6Oxon4joV/jYJxtVz7bcZj5cDC2pVd5/lsxiS6tD3+GFO3mje88cYbSd7e3kyaNOn4lVde2T42NjYqJiYmq02bNqfAmqLtkUceaVE4pdvrr79+xpURJdVGf/755w+OHTs2NDw8PLpu3bqO9957r9iKiH379s3y8/MrmDhxYolF1Lp165Z5/fXXt0lKSqozatSoY4XzpY4dO/ZYly5dogD+8pe/HL3sssuy7XZ77ZL24+vra/71r3/tGT58ePvAwMD8Hj16ZG7durUuwHPPPXfw1ltvbRUZGRltjJEWLVrkLF26dGdJ7WdnZ2cXV/e8kIiYPn36ZBUXY9H1SquT/tZbb+2bNGlSq/Dw8OiCggLp0aNHRu/evUu9eqVz584nBw8eHHbw4MHa999//6HQ0NA8d723Jan0eupnNBYhbYEnsLGPUUzBm3Nmc3G5TE6wnDc4zgxjN5lxIoH/Bx9cDlfoFS4VxwHDvIxZXJltaD31cyUlJdUaMGBAxK5duzYVd+nhrFmzgtasWeM3d+7cMi/Fqwry8/Np1KhRp+Tk5PU+Pj6u7SRUIaXVU3f1uHYSkEIBdTlaBW4XP85BfmAqx5lu7CZzjEirx2HJUBiqCb3i5EOWF/zo7jhqmldffTWoZ8+eUVOnTj3g6mvJK0tYWFjMmDFjjtbkhF4Wl/bUASRCrgCupyX59OVWlzZe1EF2spyHyOdLYzeOySKd74DPIqGN22LyUBnwbYAxwyu7He2pq5qiKvXUAX4DvNjHYbI56ob2YSdrSeBm8vlixHbMNJGrH4TvNKFXDh/4wEVNOQrn9FTKUzn/xks8cerypG7sJgXYBjRkD6tc2riDAtazlF+5ydjNshHbkf5w+z3wYQto7NJYaggH5NeGb1zU3KajR4/W18SuPJXD4ZCjR4/Wp5Rqp+66smMJcBfrWEcb+uBDwzK3uFj55LKGeH7nPmM3++JEasfBU2PhHl/wqfT2a6g0+KahMS6ZICU/P3/S4cOHZx8+fDgWnUpQeSYHsCk/P39SSSu4fEwdQCLEG/g7UIuOhBLDyEpt8BSZ/MI8DvGEsZtjcSL+f4E3RsIYb/CMM0hVkAPMYegUYkyNL7erlKu4pTdj7CYf+BgIYgMbya7EiYgzOMZS/skhHnAm9MaT4avr4CZN6JUrBX7RhK6Ua7nzK+p6YB+G+uxgaaW0kMJ+lvAIJ5hh7ObkX0TaToMfhsAgHXStfKfgEXfHoFRN47akbuymAKsWSAM2sYUsDldoA/uw8wN3kM1sYzd594lcOhUSusIlFdqOKlYKbGhlzM/ujkOpmsbdJ5M2YdWECcReQb11A9j5H8uYYLaYr0dsh6dFrn0AFoVBywppQ5UpC6a6OwalaiK3JnVjNw7gU6AeW9lOKjsuaocOCljHf0nkL8ZuVsWJeA2BuyfD3GYQWPYOVEU4DjtbGfOVu+NQqiZyd08dYCvWvJVNWUk8+ZxTxL9c8jjFL3zCViYau7HHifhcCy/eCv+oD2XOaKIqhgH2w2R3x6FUTeX2pG7sxgBzABup5LONRee9k2wy+Il/s4c7jN0cjBOpdzPM+wtMrgNVZtb2U8ClQEcgBpjmfP4HoAvQCegDlDQB4wagl3PbS5z7ywGuBGKB14useyuwtmLDL5ckWNbBmPN/D5VSFcLtSR3A2E0yMB8IYQMbOc7Wcm+cxlF+4FmO8Iixm9Q4keD74NtrYLStiry+Qj5YVa3WA+uAxcAq4HbgfedzY4Fnitk2H7gJeBNr5uYErE+r74CuWAn/Lee667HuUOhcOS+jRDmQuwNucXGzSqkiqlLSW4aVj5qxgm/I45zpss5xhL0s4UHSecHYTfZfRcKfhqUDrA5vlSNA4ay3ec6HOB+Ft1ymASHFbPtfoANWLx8gCOsi+1pANlbSL/QE8FRFBl5OO+GdK4y5uPMiSqmLUmWSuvOk6RwAMjBsKaNeyB628CO3ksMcYzf5D4n0eQR+7AhRLgj3ghVgDbM0AS4HegCzgauAFsA84OFitrPmAoShWEM1/3A+fzlw2LmfB4F4rJ57cR8MlekIHFoC97i4WaXUWapMUofTxb7mAE3ZjJ0jrDt3JQxb+IUVjDNbzXcjtsPfRcZMgfi20NzlQZ8nG9Ywy37gV6xrOl/GmpF5PzARuK+Y7fKB5VjDNMuBL7DG4r2xSiCuBUYDM4Epzn1ch5XkK1s+OFbApLuNySl7baVUZapSSd3pFyARaEYC35DJ/tNLCsgnkUWsY5yxm8Q4EdtQeGgy/KcJLigKVoEaAAOARVhjTj2cz18PrCxm/RZAf6AR4IvVs//trHVeB8ZjHcDaWHUYihufr2j/g/kjjVnogqaUUmWockm9yNUw6eTTgAQ+Ipc0cslmJR+ynVuM3eyME6kzGmZNgqcCoMwJY6uCo0Cq8+dsrFKVUVjj6IUzcX9P8eNHQ7FOhmZh9dp/AqKLLD+BVd92nHMdL6zhmlMV+grOlQS734TbKrkZpVQ5VclJlY3dpEmEzASeIJ0cfmYphp0c5WljN+lxIg1uhfeGQZytGk07dwirJ12AdXXKn4HhwNvAKKxE3BB4x7l+PLAG66RnQ6whle5YL/gq4Ooi+34KeJw/xt1fw7rssTKzbRpkfg2j5xhzYfcWKKUqnFtK75aXRMglwANYIwrvGLvJuU6k2b3wxWV/jFgoN8iD/Dlw9yRjXi97baWUq1TppA4gERIMpBi7KfibSPRk+DIGwtwdV033Bcx+F/4Wb0yJ02oppVyvyif1QnEi3WfAJzEQ6u5YarqVsOw5GBqvwy5KVTlV7kRpceJEegN3fQU/5UGuu+OpydbA5udglCZ0paqmapHUsS4WMavgwH/g4/wzb6BULvIb7HgWro035qi7Y1FKFa9aJPV4Y9Zi1YZpuRD2/gc+1MTuWhvg9xlw7efGbC97baWUu1SLpO60BKv2eqtvYe/b8EG+VT5FVbItkDQDrv3UmE3ujkUpVbpqk9TjrTO632DdKNlqEez/N3yQp4m9Utlh33Nw/YfGrHd3LEqpslWbq18KxYkIVgnxMcC+XhB0F4zxh3puDs3j/ArbX4cJ7xnzi7tjUUqVT7VL6nA6sQ/BKjF+pCUwFW4ItkqkqIuUD/kfwapP4MF4TehKVSvVMqkXihPpANwJ5NSF1GkwIvqPkuPqAqRC+kxY+Bs8GW+M3d3xKKXOT7VO6gBxIi2Au4H6wMHJ0GsQXO5VjWrCVBV22PcPeO8ozIw35ri741FKnb9qn9QB4kTqYdWuigb2xkHbm2BkHatKrSqDA8x3sP5N+IeBBfHG6Mlnpaopj0jqAHEitYAbsCYDOhQMTIbBsdBVu+wly4ZTb8HSH6x5sNfEe8ofhFI1lMckdTh9ArUX1gnU2sDBARAyHoYHQbBbg6uCNsKut2FxEsyIN+aAu+NRSl08j0rqheJE6mOVKO8HpHlD6u3QYwAMrGUl+xotGZL/A7+tgq+BOfHGZLk7JqVUxfDIpF4oTiQca9rPZsDBNuAzGYa1q+KTU1eWLDj5Baz7BFYaa47rTTrcopRn8eikDqfH2gdhzcPsAA7HQbvRMKw+BLo3OtfIh/wVsOVtSEy35qn+Od4YrXaplAfy+KReKE6kMdZdqF2B416Qfi1EDIGeIdDazeFVihw4tQq2vA/bD8Ni4Kt4Y1LL3FApVW3VmKQOp0+kdgBGA82x5mU+0hOCr4GeERBrA5tbg6wAJyFjKWz+CHalwzrg03hj9rg7LqVU5atRSb2QM7mHYc3R3AWrjO+RllB7DHTvDF38IMCtQZ4nBziSYNf3kPQdHMi35nVdHG/MXnfHppRynRqZ1IuKEwkG+mPVkvEGUgXSB0GrfhATAdG+4O/eKIvnAHME9v8Kv8fDwSOQASQAP8Qbc8TN4Sml3KDGJ/VCcSJ+QHesk6otAAOkekHmIGjZA8LbQtsgaOrOEgSnICsJdq6BAz/A8WNW6eGdwPfARp1mTqmaTZP6WZxDM42BS4ABWGPvYE2pl9oEbJdBqxho1QyaBkET30oaqsmF3GNw+BAc2gVH1kH2ZjjpsBbvAlYBG3R6OaVUIU3qpSiS4Ntijb3HAnWciwuAk8DJYPDuAE3aQOOG4O8Pfn7g6wt+vuBXB3xrQ10ByYfcPMjJLfLIgZxTzn+z4dReSFsHGdsg11iXYQqQBfwGrAV2xhuT6fIDopSq8jSpn4c4ES+scgMtgFZAO6zLIevyR/IFa0apfCAXa3gkzwY2AzaHdXWNN3/8a4o8CrdNB3YAduAAcAhI0xuFlFJl0aR+kZy9+QAgCKsqpJ/z0bDIox5Wgs/C6t1nFvn3lPOR4/w3GcjQBK6UuhCa1JVSyoNUm4mnlVJKlU2TulJKeRBN6kop5UE0qSullAfRpK6UUh5Ek7pSSnkQTepKKeVBNKkrpZQH0aSulFIeRJO6Ukp5EE3qSinlQTSpK6WUB9GkrpRSHkSTulJKeRBN6kop5UE0qSullAfRpK6UUh5Ek7pSSnkQTepKKeVBNKkrpZQH0aSulFIeRJO6Ukp5EE3qSinlQTSpK6WUB/l/IZjS7A0N/X0AAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 432x216 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(figsize=(6, 3), subplot_kw=dict(aspect=\"equal\"))\n",
    "porcentaje = df.nivel_desemp_matematica.value_counts()\n",
    "ax.pie(porcentaje,autopct='%1.1f%%',explode=(0.05,0),shadow=True,colors=[\"Green\",\"Red\"])\n",
    "ax.legend((\"Satisfactorio y Avanzado\",\"Básico y por debajo del básico\"),\n",
    "          title=\"Referencia\",\n",
    "          loc=\"center left\",\n",
    "          bbox_to_anchor=(1, 0, 0.5, 1))\n",
    "ax.set_title(\"Desempeño en Matemática\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Tenemos un 61.4% de gente que aprobo y un 38.6% de gente que desaprobo"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5,1,'Gestión')"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x216 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df_gestion = df[df.gestion != \"social/cooperativa\"].copy()\n",
    "df_gestion[\"gestion\"] = df_gestion.gestion.astype(object)\n",
    "fig, ax = plt.subplots(figsize=(6, 3), subplot_kw=dict(aspect=\"equal\"))\n",
    "\n",
    "porcentaje_gestion = df_gestion.gestion.value_counts()\n",
    "ax.pie(porcentaje_gestion,autopct='%1.1f%%',colors=[\"C4\",\"c\"])\n",
    "ax.legend((\"Estatal\",\"Privada\",\"Social\"),\n",
    "          title=\"Referencia\",\n",
    "          loc=\"center left\",\n",
    "          bbox_to_anchor=(1, 0, 0.5, 1))\n",
    "ax.set_title(\"Gestión\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Tenemos un total de 68.5% de alumnos que van a escuelas estatales en la muestra y un 31.5% de alumnos que van a una escuela privada.<br>\n",
    "Hay que destacar que se eliminaron 95 alumnos que asisten a escuelas social o cooperativa"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5,1,'Desempeño en matemática y gestion escolar')"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 482.75x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "grafico_gestion = sns.catplot(x=\"gestion\", y=\"nivel_desemp_matematica\",hue=\"indice_socioeconomico\",hue_order=[\"bajo\",\"medio\",\"alto\"], kind=\"bar\", data=df_gestion)\n",
    "plt.ylim((0,1))\n",
    "plt.title(\"Desempeño en matemática y gestion escolar\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "gestion             indice_socioeconomico\n",
       "estatal             alto                     0.268507\n",
       "                    bajo                     0.549543\n",
       "                    medio                    0.455459\n",
       "privada             alto                     0.132829\n",
       "                    bajo                     0.456364\n",
       "                    medio                    0.276874\n",
       "social/cooperativa  alto                     1.000000\n",
       "                    bajo                     0.684211\n",
       "                    medio                    0.763889\n",
       "Name: nivel_desemp_matematica, dtype: float64"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Para ver los valores del grafico\n",
    "df.groupby([\"gestion\",\"indice_socioeconomico\"])[\"nivel_desemp_matematica\"].mean()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "En este grafico se aprecia que al mismo nivel economico, si un alumno pertenece a una escuela publica tiene mas chances de desaprobar"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Para ver si los que se desempeñan bien en lengua lo hacen igual en matematica se realiza el siguente grafico:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "h9vsHOvJpifR"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5,1,'Desempeño en matemática vs desempeño en lengua')"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.catplot(x=\"nivel_desemp_lengua\", y=\"nivel_desemp_matematica\", kind=\"bar\",order=[\"por_debajo_del_basico\",\"basico\",\"satisfactorio\",\"avanzado\"], data=df,height=6);\n",
    "plt.ylim((0,1))\n",
    "plt.title(\"Desempeño en matemática vs desempeño en lengua\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Se observa que el desempeño en lengua esta correlacionado con el desempeño en matematica. Ya que por ejemplo dentro del grupo de alumnos que obtuvieron una calificación igual a basico en lengua, mas del 60% desaprobo matematica. Dentro del grupo que obtuvo un puntaje por debajo del basico en lengua, mas del 70% desaprobo matematica. Por el contrario si su puntaje fue avanzado en lengua, menos del 15% desaprueba matematica. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "nivel_desemp_lengua\n",
       "avanzado                 0.123657\n",
       "basico                   0.651534\n",
       "por_debajo_del_basico    0.723312\n",
       "satisfactorio            0.408851\n",
       "Name: nivel_desemp_matematica, dtype: float64"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Para ver los valores del grafico\n",
    "df.groupby(\"nivel_desemp_lengua\")[\"nivel_desemp_matematica\"].mean()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Indice socioeconomico\n",
    "Por ver si hay alguna correlación entre el indice socioeconomico, que va del 1 al 3, y la nota obtenida en matematica se realiza un box plot."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5,1,'Nivel socioeconomico promedio y desempeño en matematica')"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 405.625x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.catplot(x=\"nivel_desemp_matematica\", y=\"indice_socioeconomico_medio\", hue=\"sexo\", kind=\"box\", data=df)\n",
    "plt.title(\"Nivel socioeconomico promedio y desempeño en matematica\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th>count</th>\n",
       "      <th>mean</th>\n",
       "      <th>std</th>\n",
       "      <th>min</th>\n",
       "      <th>25%</th>\n",
       "      <th>50%</th>\n",
       "      <th>75%</th>\n",
       "      <th>max</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>sexo</th>\n",
       "      <th>nivel_desemp_matematica</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th rowspan=\"2\" valign=\"top\">f</th>\n",
       "      <th>0</th>\n",
       "      <td>88405.0</td>\n",
       "      <td>2.091609</td>\n",
       "      <td>0.331096</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.882353</td>\n",
       "      <td>2.066667</td>\n",
       "      <td>2.304348</td>\n",
       "      <td>3.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>58576.0</td>\n",
       "      <td>1.931898</td>\n",
       "      <td>0.262474</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.770492</td>\n",
       "      <td>1.917808</td>\n",
       "      <td>2.075758</td>\n",
       "      <td>3.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"2\" valign=\"top\">m</th>\n",
       "      <th>0</th>\n",
       "      <td>82042.0</td>\n",
       "      <td>2.076225</td>\n",
       "      <td>0.331201</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.871429</td>\n",
       "      <td>2.047619</td>\n",
       "      <td>2.280488</td>\n",
       "      <td>3.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>48550.0</td>\n",
       "      <td>1.919830</td>\n",
       "      <td>0.258532</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.766667</td>\n",
       "      <td>1.907216</td>\n",
       "      <td>2.060000</td>\n",
       "      <td>3.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                count      mean       std  min       25%  \\\n",
       "sexo nivel_desemp_matematica                                               \n",
       "f    0                        88405.0  2.091609  0.331096  1.0  1.882353   \n",
       "     1                        58576.0  1.931898  0.262474  1.0  1.770492   \n",
       "m    0                        82042.0  2.076225  0.331201  1.0  1.871429   \n",
       "     1                        48550.0  1.919830  0.258532  1.0  1.766667   \n",
       "\n",
       "                                   50%       75%  max  \n",
       "sexo nivel_desemp_matematica                           \n",
       "f    0                        2.066667  2.304348  3.0  \n",
       "     1                        1.917808  2.075758  3.0  \n",
       "m    0                        2.047619  2.280488  3.0  \n",
       "     1                        1.907216  2.060000  3.0  "
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.groupby([\"sexo\",\"nivel_desemp_matematica\"])[\"indice_socioeconomico_medio\"].describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Del siguiente grafico se puede ver lo siguiente. En promedio los alumnos que tienen un nivel socio economico mayor, tienen menos probabilidad de desaprobar."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Diferencia entre provincias\n",
    "Para ver si hay una diferencia significativa en el porcentaje de aprobador y desaprobados entre provincias"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "prov = pd.read_csv(\"datasets/app_provincia.csv\")\n",
    "df_grafico = pd.merge(df,prov,how=\"left\",left_on=\"provincia_id\",right_on=\"id\",suffixes=(\"_df\",\"_prov\"))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5,1,'Desempeño en matemática por provincia')"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.catplot(x=\"name\", y=\"nivel_desemp_matematica\", kind=\"bar\", data=df_grafico);\n",
    "plt.ylim((0,1))\n",
    "plt.xticks(rotation=65, horizontalalignment='right')\n",
    "plt.title(\"Desempeño en matemática por provincia\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Se aprecia que el porcentaje de aprobados y desaprobados varia mucho entre provincias, esto puede ser por diferentes condiciones economicas que afectan a la educacion"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [],
   "source": [
    "indice_socioeconomico_provincia = df_grafico.groupby(\"name\")[\"indice_socioeconomico_medio\"].mean().sort_values(ascending=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Hago el mismo grafico pero ordenando por indice socioeconomico medio de la provincia.<br> Este indice socio economio medio se calcula para cada escuela por lo que se hace un promedio por provincia y ahi se ordena"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5,1,'Desempeño en matemática por provincia.\\n Ordenado por indice socioeconomico medio')"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.catplot(x=\"name\", y=\"nivel_desemp_matematica\", kind=\"bar\",order=indice_socioeconomico_provincia.index, data=df_grafico)\n",
    "plt.xticks(rotation=65, horizontalalignment='right')\n",
    "plt.ylim((0,1))\n",
    "plt.title(\"Desempeño en matemática por provincia.\\n Ordenado por indice socioeconomico medio\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Mirando el desempeño en matematica teniendo en cuenta el indice economio del alumno que se divide en bajo, medio y alto se observa lo siguiente:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5,1,'Desempeño en matemática y índice económico')"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.catplot(x=\"indice_socioeconomico\", y=\"nivel_desemp_matematica\",order=[\"bajo\",\"medio\",\"alto\"], kind=\"bar\", data=df);\n",
    "plt.ylim((0,1))\n",
    "plt.title(\"Desempeño en matemática y índice económico\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Se puede ver que los alumnos que tiene un indice socioeconomico bajo, mas de un 50% de ellos no aprueba las pruebas aprender.<br>\n",
    "Dentro de la clase media un 40% desaprueba y de los de indice alto casi un 20%."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Matricula y nivel de desempeño"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5,1,'Desempeño en matemática y cantidad de matriculados por escuela')"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAqQAAAMyCAYAAAC7DtNDAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvhp/UCwAAIABJREFUeJzs3XmYJWdZN/7vnYSQbZIACY4kYV804AKETRFZFAKYRBAFVBAU+KEi+oqOLAqIG46vIiKKgOwqshN2VAyuIEFADIgvsphM0hIiJJmwZOH5/VE14UwvM32m+/TT0+fzua6+qk9VPafuU+ec6m8/tVVrLQAA0MshvQsAAGC+CaQAAHQlkAIA0JVACgBAVwIpAABdCaQAAHQlkMIaVdWOqvpsVZ1aVX/bu56DVVXdfFyPp+xnvndU1Y9tVF29VNWNq2p3VR26wvRnVtWrDvC5H1VV/zDF/J+pqu85kGVttKr6rqr6xDo8zzlV9ZgDbPuyqvr1tdYwr9ay7jl4CaTMzPhH7MtVdXlVfbGq/qmqHl9VW+1zd/sk907yB0n+pnMt624Dw8gLkjyutXb+xLKXhK7W2v1bay/fgHo21OL13Fr779baMa21a3rWtZlUVauqW+5rntba37fWbrNRNQHr47DeBbDlndFa++uqOi7Jdyd5bpK7JHl037LWT2vt4eOvB0UP0mY09oq+pLX2rt61cPCqqsNaa1f3rmMzs47YrLZaTxWbVGvt0tba2UkemuTHqup2SVJV162q/1tV/11V/1NVL6iqI8dpJ1TVW8fe1f+tqr/f07taVTeqqtdX1cVV9emqeuKeZY29aq+tqleNvbMfrapbV9VTqupzVXV+Vd13Yv5zquq3qupfqurSqnpzVV1/Yvpdx97dL1bVR6rqnova/lpV/eO4rHdX1QkT08+sqvPGtudU1TevtI6q6puq6q/G1/qJqvqhiWkvq6rnV9XbxuW8v6puscLz3HTsSXr0+Fq/MPZM36mq/m2s5Q8n5r9FVb2nqi6pqs9X1Z9V1fHjtFcmuXGSt4y7j3escp38+jh9d1W9papuMD7vZVX1gaq66cT8z03yT0leWFUfrKrvGsefnuSpSR46Ps9HJp7/MRPtH1tVHx/Xy8eq6g7j+CdX1X9NjH/QCutre1V9qapuMDHujuNn6zrLzH9oVT114rk/OAbqVNVzx3V+2eRrGac9s6peU1WvGNudV1WnrbSeJ97Hw8Z5blZV7x3b/lWSExbV9dqqWhg/w39XVbedmHaDqjp7rOtfkiz72ZmY/xE1HD5xSVU9bdG0QybW7SXja7r+Cs9zz6q6YHw9n6uqi6rq+6vqAVX1n+Nn/akT89+5qv55/FxdVFV/WFWHj9P+bpztI+M6eujE8/9SVS0keemecRPPeUpVvWF8Py/Z89mvRb3vi9f3otex4ndknH77qvrX8b35yyRHLGr/2Kr65Ph6z66qG43jq6qeM66bS2v4ft5uhXW5v+3UituaGnrff6mq/i3JFSu8xn1tfx5Qw3fo8qraVVW/MDHtrKr68PjZ+q8avrdLevyXWd8rbkOmWfdsIa01P35m8pPkM0m+Z5nx/53kJ8fffz/J2Umun2Rbkrck+a1x2m9l2I17nfHnu5JUhn+kPpjk6UkOT3LzJJ9Kcr+x3TOTfCXJ/TLsBXhFkk8nedr4PI9N8umJes5JsivJ7ZIcneT1SV41TjspySVJHjAu93vHxydOtP2vJLdOcuT4+NnjtFsnuWJsc50kO5J8Msnhy6yTo5Ocn6Hn+LAkd0jy+SS3Hae/LMn/JrnzOP3Pkrx6hfV+0yRtXHdHJLnvuD7elOSG42v6XJLvHue/5VjjdZOcmOTvkvz+Su/jKtfJJzOEnuOSfCzJf2boQd7zfrx04vkemeQG47RfSLKQ5IiJ9/JVi17fOUkeM/7+g+N7d6cMn41bJrnJxLQbjTU+dHwvvnGFdfb2jJ/J8fFzkjxvhXl/MclHk9xmXOa3JbnBOO1HJ17Lk5Z5LV8Z19uhGT7f79vHet7zPh42Pv7nJL83vk/3SHL55LpJ8uMZvkPXzfC9+vDEtFcneU2Gz9ntxnX2Dyu8vlOT7B6Xcd1xmVfvqS3JzyV5X5KTx+l/kuQvVniue45tn56vf/cuTvLnY623HdfJzcf575jkruP6u2mSjyf5uYnna0luuczz//ZYy5HjuAvG6Ycm+cj4fh6d4ftw9+U+W8us73Py9c/Zit+RDNugzyb5P+NrfEiSq5L8+jj93hm+y3cY2z8vyd+N0+6XYVt2fIbP0jdn5c/oOVl5O7XPbU2Gz9aHk5yS5MgD2P5clOS7xt+vl+QO4+93TnLpuNxDMmwbvmmFz/O16zur24bsd9372Vo/3Qvws3V/Fm+QJsa/L0M4rHEjeouJaXfLGBaTPCvJmzPxB2gcf5ck/71o3FMyhpxxw/dXE9POyPAH9tDx8bYMf3iOHx+fkzFEjo9PTXJlhj9mv5TklYuW9a4kPzbR9pcnpv1UkneOv/9KktdMTDtk/INyz2XWyUOT/P2icX+S5Bnj7y9L8uKJaQ9I8h8rrPebjq/vpIlxlyR56MTj12fiD/2i9t+f5EMrvY+rXCdPm5j2u0nesej9+PByyx6nfzHJt028l/sKpO9K8rOr/Dx+OMlZK0x7aJJ/HH8/NEOQvPMK835ipedZZt4vLHotf73oc/blfaznPe/jYRl6T69OcvTE9D9fvG4mph0/tj1ufD1XZQwK4/TfzMqB9OmZ+GcnQ1i5Ml8PpB9Pcp+J6d84Pv9hyzzXPZN8OUu/e3eZmOeDSb5/hVp+LskbJx4vF0ivzBj6J8btCaR3yxCAl6ttr89W9hFI9/UdyRDcL0xSE9P/KV8PpH+aZOfEtGPG9XXTDGH1PzOE8EP281k6Jytvp/a5rRk/Wz++j+fe3/bnv5P8f0mOXWae56zwnJ/JyoF0NduQ/a57P1vrxy57ejgpQ2/fiUmOSvLBcbfNF5O8cxyfJL+T4b/8d1fVp6rqyeP4myS50Z42Y7unJvmGiWX8z8TvX07y+fb1k0O+PA6PmZjn/InfP5uhl+GEcVk/uGhZd8/wR3iPhYnfvzTxvDcanytJ0lr72rick5ZZJzdJcpdFy/mRJNtXsZyVLF4Hix8fkyRVdcOqevW4K+6yJK/Kot3By9S6v3WyqmWPy39CVX2ohl3dnxmn7Wv5k07J0EO9RFU9ctyVuKfG2+3jed+c5NSqunmG3phLW2v/cgDLfFINhw9cOi7zuEXLXPweHrHc7tNl3CjJF1prV0yMu/azVcNhBM8ed5leliEMZFz2iRlC7eLP+L6Wde284zIvmZh+kyRvnFivH09yTfb+/k26ZJnv3kqfxVvXcJjOwvg6fjP7/yxc3Fr7ygrTTkny2bbGYyb38x25UZJdbUxLo8n1u3g7sDvD+jyptfaeJH+Y5PlJ/qeqXlhVx+6jlJW2U6vZ1ky2XWx/258fyPBP8GdrOGzkbuP4Fb8L+7GabUiSA9o+cZASSNlQVXWnDBvJf8iwS+jLGXYLHT/+HNdaOyZJWmuXt9ae1Fq7eYZetZ+vqvtk2LB+eqLN8a21ba21B6yhtMlLDd04Qw/G58dlvXLRso5urT17Fc95YYYN757XXuNydi0z7/lJ3rtoOce01n7ygF/R6v1Whp6hb22tHZtht3NNTG+L5l/LOtlLVX1nht6dH2qtndJau2mG3uw9y1+87MXOzzLHQ1bVTZK8KMkTMuxOPz7Jvy96XdcaA81rMvwRfkSSVx7AMr8rQ8/PDyW53rjMS1da5nJl7GPaRUmuV1VHT4y78cTvP5zkrAyHRRyXofct47IvztC7uvgzvq9lXTtvVR2V4TCEPc5Pcv9F7/8RrbXlPtfT+uMk/5HkVuNn8anZ//rb13o7P8mNVwj9V2T4h3iP7cvMs8e+viMXJTlp/H7vMbl+F28Hjs6wPnclSWvtD1prd8xw+MKtMxwSspKVtlOr2dbsbz2tuP1prX2gtXZWhkN+3pThu7Kn3UrHI+9r/U6zDdnf9oktQiBlQ1TVsVX1fRmOZXtVa+2j43/xL0rynKq64TjfSVV1v/H376uqW44b18sy9MJck+Rfklw2HqR/5Ng7dLsx7B6oH63hOqJHZThU4HVjr86rkpxRVfcbl3NEDSdNnLyK53xNkgdW1X1qODnmSUm+mmF33mJvTXLrGk4muc74c6fax0lQ62hbhhD4xao6KUv/IP5PhuN091jLOlns+CRfy3CixeFV9fSxnsll37RWvlTYi5P8Qg0nIdX4eblJht3MLUMYS1U9OkMP6b68Ismjkpw5vsaVvDjJr1XVrcZlfmsNJ0RtyxD8Lk5y2Pha9tXbtdji9Xyt1tpnk5yb5FfH9XT3DP+k7bEtw2frkgwh4Dcn2l6T5A1JnllVR1XVqUl+bB91vC7J91XV3Ws4oehZ2ftvxQuS/Ma4nlNVJ1bVWVO8zn3ZluG7vruqvinJ4n/IVlxHK/iXDIHx2VV19PhZ/c5x2oeT3KOG670el+Gwn33VtdJ35J8zvO9PrKrDqurBGY6t3OPPkzy6qr69qq6b4b15f2vtM+N3/C7j9uGKDMfT7usyXyttp6bZ1ixnxe3P+Hn7kao6rrV2Vb6+LU6GwxEePS73kHH7/U3jtA8nedj4XKdlOLZ2j2m2IfvbPrFFCKTM2luq6vIM/xE/LcMJEpOXfPqlDLvl3zfujvnrDCeLJMmtxse7M2z0/6i1ds64AT4jybdnOFnp8xlCwnFrqPOVGY7TXMhw4sMTk6QN18Q8K0NPzcXj6/jFrOK701r7RIb/5p831nhGhstgXbnMvJdnOPnoYRl6Oxby9RM1Zu1XM5zEcGmSt2UIL5N+K8kvj7vWfmEt62QZ70zyjgy9Yp/N8Ad5ctfia8fhJVX1r4sbt9Zem+Q3MvzRvzxD7831W2sfy3Ds6j9nCDHfkuQf91VIa+0fM4Tjf22tfWYfs/5ehgDw7gx/nP80w8k07xpfy3+u8Fr2Z6/1vMz0H85w/PT/JnlGhgC9xyvGZe7KcBLZ+xa1fUKG3eILGT7nL12piNbaeUl+OsM6vSjDcbAXTMzy3AwnIr57/G6/b6xrPfxChtd5eYZ/Vv9y0fRnJnn5uI5+KPsxsa24ZYbjIC/IcLxkWmt/NT7/v2U4jvWt+3iqFb8j4/f5wRn+mfnC+PyT0/8mw16A12dYn7fI8D1Phn9YXjS2+2yGfyj+7z7qWGk7teptzXJWsf15RJLPjNvox4/LynhYy6MznDR2aZL35us9tb8yvtYvZFh/fz6xvGm2IfvbPrFF1N6HvcD8qapzMvTavrh3LfRVVe9J8uc+C2w2tlNsdS6MD5Brj2++Q4aeGwA20Ex32VfVS2q44O+/rzC9quoParhg8L/VeEFrgI1UVS/PcHjIz427LwHYQDPdZV9V98hw/N8rWmtLTiioqgck+ZkMl5O4S5LnttbW61gkAAAOAjPtIW2t/V2GA/BXclaGsNpaa+9LcnxVLbkOGQAAW1fvY0hPyt5noV4wjrto8YxV9bgkj0uSU0899Y7nnXfehhQIAMABW9V1Y3tf9mm5Ipc9hqC19sLW2mmttdOOPPLIGZcFAMBG6R1IL8jed544OcM10AAAmBO9A+nZSR45nm1/1wz3j16yux4AgK1rpseQVtVfJLlnkhOq6oIMdxe5TpK01l6Q5O0ZzrD/ZJIvZe87+AAAMAdmGkhbaw/fz/SW4RZ1AADMqd677AEAmHMCKQAAXQmkAAB0JZACANCVQAoAQFcCKQAAXQmkAAB0JZACANCVQAoAQFcCKQAAXQmkAAB0JZACANCVQAoAQFcCKQAAXQmkAAB0JZACANCVQAoAQFcCKQAAXQmkAAB0JZACANCVQAoAQFcCKQAAXQmkAAB0JZACANCVQAoAQFcCKQAAXQmkAAB0JZACANCVQAoAQFcCKQAAXQmkAAB0JZACANCVQAoAQFcCKQAAXQmkAAB0JZACANCVQAoAQFcCKQAAXQmkAAB0JZACANCVQAoAQFcCKQAAXQmkAAB0JZACANCVQAoAQFcCKQAAXQmkAAB0JZACANCVQAoAQFcCKQAAXQmkAAB0JZACANCVQAoAQFcCKQAAXQmkAAB0JZACANCVQAoAQFcCKQAAXQmkAAB0JZACANCVQAoAQFcCKQAAXQmkAAB0JZACANCVQAoAQFcCKQAAXQmkAAB0JZACANCVQAoAQFcCKQAAXQmkAAB0JZACANCVQAoAQFcCKQAAXQmkAAB0JZACANCVQAoAQFcCKQAAXQmkAAB0JZACANCVQAoAQFcCKQAAXQmkAAB0JZACANCVQAoAQFcCKQAAXQmkAAB0JZACANCVQAoAQFcCKQAAXQmkAAB0JZACANCVQAoAQFcCKQAAXQmkAAB0JZACANCVQAoAQFcCKQAAXQmkAAB0JZACANCVQAoAQFcCKQAAXQmkAAB0JZACANCVQAoAQFcCKQAAXQmkAAB0JZACANCVQAoAQFcCKQAAXQmkAAB0JZACANCVQAoAQFcCKQAAXQmkAAB0JZACANCVQAoAQFcCKQAAXQmkAAB0JZACANCVQAoAQFcCKQAAXQmkAAB0JZACANDVzANpVZ1eVZ+oqk9W1ZOXmX7jqvrbqvpQVf1bVT1g1jUBALB5zDSQVtWhSZ6f5P5JTk3y8Ko6ddFsv5zkNa212yd5WJI/mmVNAABsLrPuIb1zkk+21j7VWrsyyauTnLVonpbk2PH345JcOOOaAADYRGYdSE9Kcv7E4wvGcZOemeRHq+qCJG9P8jPLPVFVPa6qzq2qcy+++OJZ1AoAQAezDqS1zLi26PHDk7ystXZykgckeWVVLamrtfbC1tpprbXTTjzxxBmUCgBAD7MOpBckOWXi8clZukv+J5K8Jklaa/+c5IgkJ8y4LgAANolZB9IPJLlVVd2sqg7PcNLS2Yvm+e8k90mSqvrmDIHUPnkAgDkx00DaWrs6yROSvCvJxzOcTX9eVT2rqs4cZ3tSksdW1UeS/EWSR7XWFu/WBwBgi6qDMfuddtpp7dxzz+1dBgAA+7bc+URLuFMTAABdCaQAAHQlkAIA0JVACgBAVwIpAABdCaQAAHQlkAIA0JVACgBAVwIpAABdCaQAAHQlkAIA0JVACgBAVwIpAABdCaQAAHQlkAIA0JVACgBAVwIpAABdCaQAAHQlkAIA0JVACgBAVwIpAABdCaQAAHQlkAIA0JVACgBAVwIpAABdCaQAAHQlkAIA0JVACgBAVwIpAABdCaQAAHQlkAIA0JVACgBAVwIpAABdCaQAAHQlkAIA0JVACgBAVwIpAABdCaQAAHQlkAIA0JVACgBAVwIpAABdCaQAAHQlkAIA0JVACgBAVwIpAABdCaQAAHQlkAIA0JVACgBAVwIpAABdCaQAAHQlkAIA0JVACgBAVwIpAABdCaQAAHQlkAIA0JVACgBAVwIpAABdCaQAAHQlkAIA0JVACgBAVwIpAABdCaQAAHQlkAIA0JVACgBAVwIpAABdCaQAAHQlkAIA0JVACgBAVwIpAABdCaQAAHQlkAIA0JVACgBAVwIpAABdCaQAAHQlkAIA0JVACgBAVwIpAABdCaQAAHQlkAIA0JVACgBAVwIpAABdCaQAAHQlkAIA0JVACgBAVwIpAABdCaQAAHQlkAIA0JVACgBAVwIpAABdCaQAAHQlkAIA0JVACgBAVwIpAABdCaQAAHQlkAIA0JVACgBAVwIpAABdCaQAAHQlkAIA0JVACgBAVwIpAABdCaQAAHQlkAIA0JVACgBAVwIpAABdCaQAAHQlkAIA0JVACgBAVwIpAABdCaQAAHQlkAIA0JVACgBAVwIpAABdCaQAAHQlkAIA0JVACgBAVwIpAABdCaQAAHQlkAIA0JVACgBAVwIpAABdCaQAAHQlkAIA0NXMA2lVnV5Vn6iqT1bVk1eY54eq6mNVdV5V/fmsawIAYPM4bJZPXlWHJnl+ku9NckGSD1TV2a21j03Mc6skT0nyna21L1TVDWdZEwAAm8use0jvnOSTrbVPtdauTPLqJGctmuexSZ7fWvtCkrTWPjfjmgAA2ERmHUhPSnL+xOMLxnGTbp3k1lX1j1X1vqo6fcY1AQCwicx0l32SWmZcW6aGWyW5Z5KTk/x9Vd2utfbFvZ6o6nFJHpckN77xjde/UgAAuph1D+kFSU6ZeHxykguXmefNrbWrWmufTvKJDAF1L621F7bWTmutnXbiiSfOrGAAADbWrAPpB5LcqqpuVlWHJ3lYkrMXzfOmJPdKkqo6IcMu/E/NuC4AADaJVe+yH8+G/60kpyY5Ys/41trNV2rTWru6qp6Q5F1JDk3yktbaeVX1rCTnttbOHqfdt6o+luSaJL/YWrvkgF4NAAAHnWpt8SGdK8xY9Q9JnpHkOUnOSPLosf0zZlfe8k477bR27rnnbvRiAQCYznLnEy0xzS77I1trf5MhhH62tfbMJPc+kMoAAGCPac6y/0pVHZLk/4274XclcRF7AADWZJoe0p9LclSSJya5Y5IfTfJjsygKAID5seoe0tbaB8Zfd2c4fhQAANZs1T2kVfVXVXX8xOPrVdW7ZlMWAADzYppd9idM3j1pvPe8Y0gBAFiTaQLp16rq2nt2VtVNsvQ2oAAAMJVpzrJ/WpJ/qKr3jo/vkfHe8gAAcKCmOanpnVV1hyR3zXCR0//TWvv8zCoDAGAu7HeXfVV90zi8Q5IbJ7kwwzVIbzyOAwCAA7aaHtKfz7Br/neXmdbibk0AAKzBfgNpa23PcaL3b619ZXJaVR0xk6oAAJgb05xl/0+rHAcAAKu23x7Sqtqe5KQkR1bV7TOc0JQkx2a4lSgAAByw1RxDer8kj0pycpLfmxh/eZKnzqAmAADmyGqOIX15kpdX1Q+01l6/ATUBADBHprkO6eur6oFJbpvkiInxz5pFYQAAzIdVn9RUVS9I8tAkP5PhONIfTHKTGdUFAMCcmOYs++9orT0yyRdaa7+a5G5JTplNWQAAzItpAumXx+GXqupGSa5KcrP1LwkAgHmy6mNIk7y1qo5P8jtJ/jXDXZpePJOqAACYG9Oc1PRr46+vr6q3JjmitXbpbMoCAGBerDqQVtWhSR6Y5KZ72lVVWmu/t692AACwL9Pssn9Lkq8k+WiSr82mHIB+duzYkYWFhWzfvj07d+7sXQ7A3JgmkJ7cWvvWmVUC0NnCwkJ27drVuwyAuTPNWfbvqKr7zqwSAADm0jQ9pO9L8saqOiTDJZ8qSWutHTuTygAAmAvTBNLfzXAx/I+21tqM6gEAYM5Ms8v+/yX5d2EUAID1NE0P6UVJzqmqdyT56p6RLvsEwMHClRRgc5omkH56/Dl8/AGAg4orKcDmNM2dmn41Sarq6NbaFbMrCQCAebLqY0ir6m5V9bEkHx8ff1tV/dHMKgMAYC5Mc1LT7ye5X5JLkqS19pEk95hFUQAAzI9pAmlaa+cvGnXNOtYCAMAcmuakpvOr6juStKo6PMkTM+6+BwCAAzVND+njk/x0kpOSXJDk25P81CyKAgBgfkzTQ3qb1tqPTI6oqu9M8o/rWxIAAPNkmh7S561yHAAArNp+e0ir6m5JviPJiVX18xOTjk1y6KwKAwBgPqxml/3hSY4Z5902Mf6yJA+ZRVEAAMyP/QbS1tp7k7y3ql7WWvvsBtQEAMAcmeakpi9V1e8kuW2SI/aMbK3de92rAgBgbkxzUtOfJfmPJDdL8qtJPpPkAzOoCQCAOTJNIL1Ba+1Pk1zVWntva+3Hk9x1RnUBADAnptllf9U4vKiqHpjkwiQnr39JAADMk2kC6a9X1XFJnpTh+qPHJvk/M6kKAGCD7dixIwsLC9m+fXt27tzZu5y5supA2lp76/jrpUnuNZtyAAD6WFhYyK5du3qXMZdWHUir6mZJfibJTSfbtdbOXP+yAACYF9Pssn9Tkj9N8pYkX5tNOQAAzJtpAulXWmt/MLNKAACYS9ME0udW1TOSvDvJV/eMbK3967pXBQDA3JgmkH5LkkckuXe+vsu+jY8BAOCATBNIH5Tk5q21K2dVDAAA82eaOzV9JMnxsyoEAID5NE0P6Tck+Y+q+kD2PobUZZ8AADhg0wTSZ8ysCgAA5tY0d2p6776mV9U/t9butvaSAIDNwK002SjT9JDuzxHr+FwAQGdupclGmeakpv1p6/hcAADMifUMpAAAMLX1DKS1js8FAMCcmOoY0qranuTOGXbPf6C1tjAx+RHrWRgAAPNh1T2kVfWYJP+S5MFJHpLkfVX143umt9b+ff3LAwBgq5umh/QXk9y+tXZJklTVDZL8U5KXzKIwAADmwzTHkF6Q5PKJx5cnOX99ywEAYN5M00O6K8n7q+rNGY4hPSvJv1TVzydJa+33ZlAfAABb3DSB9L/Gnz3ePA63rV85AADMm2luHfqrsywEAID5tOpAWlWnJXlakptMtmutfesM6gIAYE5Ms8v+zzKcaf/RJF+bTTkAAMybaQLpxa21s2dWCQAAc2maQPqMqnpxkr9J8tU9I1trb1j3qgAAmBvTBNJHJ/mmJNfJ13fZtyQCKQAAB2yaQPptrbVvmVklAADMpWnu1PS+qjp1ZpUAADCXpukhvXuSH6uqT2c4hrSSNJd9AgBgLaYJpKfPrAoAAObWqnfZt9Y+m+SUJPcef//SNO0BAGA5qw6UVfWMJL+U5CnjqOskedUsigIAYH5M08P5oCRnJrkiSVprFybZNouiAACYH9ME0itbay3DtUdTVUfPpiQAAObJNIH0NVX1J0mOr6rHJvnrJC+aTVkAAMyLVZ9l31r7v1X1vUkuS3KbJE9vrf3VzCoDAGAurDqQjrvo39Na+6uquk2S21TVdVprV82uPAAAtrppdtn/XZLrVtVJGXbXPzrJy2ZRFAAA82OaQFqttS8leXCS57XWHpTErUQBAFiTqQJpVd0tyY8keds4bpo7PQEAwBLTBNKfy3BR/De21s6rqpsn+dvZlAUAwLyY5iz79yZ5757rj7bWPpXkibMqDACA+TDNrUPvVlUfS/Lx8fG3VdUfzawyAADmwjS77H8/yf2SXJIkrbWPJLnHLIoCAGB+THVSUmvt/KqaHHWPghY0AAAgAElEQVTN+pYDAOvjB1//0SXjvrj7yiTJRbuvXDL9tT/wLRtSF7DUNIH0/Kr6jiStqg7PcPzox2dTFgAA82KaXfaPT/LTSU5KckGSbx8fAwDAAZvmLPvPZ7gGKQAArJv9BtKqel6SttL01ppLPwHAQe7df/H5JeO+dPnXrh0unn7fh5+wIXUxH1azy/7cJB9MckSSOyT5f+PPt8dJTQAArNF+e0hbay9Pkqp6VJJ7tdauGh+/IMm7Z1odAABb3jQnNd0oybaJx8eM4wAA4IBNc9mnZyf5UFXtuX/9dyd55rpXBADAXJnmLPuXVtU7ktxlHPXk1trCnulVddvW2nnrXSAAAFvbtHdqWkjy5hUmvzLDSU8AALBq0xxDuj+1/1kAAGBv6xlIV7xWKQAArGQ9AykAAExtPQPplev4XAAAzInV3Dp0nycqtdb+dRzedb2KAgBgfqzmLPvf3ce0luTe61QLAABzaDW3Dr3XRhQCAMB8WvUxpFV1VFX9clW9cHx8q6r6vtmVBgDAPJjmpKaXZjhx6TvGxxck+fV1rwgAgLkyTSC9RWttZ5KrkqS19uW4GD4AAGs0za1Dr6yqIzNeAL+qbpHkqzOpqpMdO3ZkYWEh27dvz86dO3uXAwAwF6YJpM9M8s4kp1TVnyX5ziSPmkFN3SwsLGTXrl29ywAAmCurDqSttXdX1QeT3DXDrvqfba19fmaVAQAwF1YdSKvq7CR/keTs1toVsysJAGC2Ltp5/pJx13zh6muHi6d/445TNqSueTXNSU2/m+S7knysql5bVQ+pqiNmVBcAAHNiml32703y3qo6NMPdmR6b5CVJjp1RbQAAzIFpTmrKeJb9GUkemuQOSV4+i6I2wsUveMGScddceum1w8XTT3z84zekLgCAeTPNMaR/meQuGc60f36Sc1prX5tVYQAAzIdpekhfmuSHW2vXzKoYAADmz34DaVXdu7X2niRHJTmrau+bM7XW3jCj2lgFF/MHAA52q+kh/e4k78lw7Ggy3qkpw7VIWxKBtCMX8984wj8AzMZ+A2lr7Rnjrz+Z5AeS3HSiXVuuzaSqOj3Jc5McmuTFrbVnrzDfQ5K8NsmdWmvn7rfyGTjx6KP3Gk5DWNn6hH/YHGxvYeuZ5hjSNyX5YpJ/TfKVcdw+A+l4iajnJ/neJBck+UBVnd1a+9ii+bYleWKS909Rz7p72j3uccBthRWAjWF7C1vPNIH05Nba6VM+/52TfLK19qkkqapXJzkryccWzfdrSXYm+YUpn3+ufOL5Zy0Zd9WlV4zDC5dMv81Pv3lD6tpoekcAYGuZ5k5N/1RV3zLl85+UZPLeWxeM465VVbdPckpr7a37eqKqelxVnVtV51588cVTlsF627FjRx75yEdmx44dG77sPb0jCwsLG75sAGD9TdNDevckj6qqTyf5asaTmlpr37qPNrXMuGt381fVIUmek+RR+1t4a+2FSV6YJKeddtp+j11ltg7WXWZ6VwFg85kmkN7/AJ7/giSnTDw+OcmFE4+3JbldknPGy0ltT3J2VZ3Z68Sm1Vj4499YMu6aS//32uHi6dt/8mkbUhf7d7AGaViJf7KArWCae9l/9gCe/wNJblVVN0uyK8nDkvzwxHNemuSEPY+r6pwkv7CZwyjAZuKfLGArmOpe9tNqrV1dVU9I8q4Ml316SWvtvKp6VpJzW2tnz3L58+D6Rx2S5GvjkPXyolfcb8m4yy6/ehzuWjL9sY9814bUBfSjNxpmZ6aBNElaa29P8vZF456+wrz3nHU9W80T737kzJfxdy964JJxX77sq+PwwiXT7/HYt63bst/w0qUXdth92VXjcNeS6Q9+9DvXbdnryR8yOPjpjYbZmXkgnRcnHHXEXkOY5A8ZcDA6btuJew1hVgTSdfKUe9y+dwlAR3rB2Yoeen8n5bIxBNJNwB8yOPjpBQc4cALpJnAw/iE7/qjaa8jW4x8lADaKQMoBecS9Du9dwn696mVLz5S//LKrx+GuJdN/9FHOlJ90MP6jBMDBybWCAADoSg/pBtv1h49fMu7qL37u2uHi6Sc94QUbUtdGW8vu4GOPqSRtHAIABzuBlC7Wsjv4zPv0+dgedfQQhIchs+T4VYD5IpDCKt3rvof2LmFuOH4VYL5suUCqZwUA4OCy5QLpwdizcuLR19lruFGEd8B2gM3I53L+bLlAejDa8Z0ndVnuwRjeDwbP/bOll5v64uVXj8NdS6b/7I+43BT92A6wGc3j53LeQ7hAyky9808fsOz4L1125Ti8cMk8p//E22dWzzHjGfrHOEMfgE1kHkP4JIGUuXK/73FiEgBsNgLpnPjgC85YMu6rl355HF64ZPodH/+WDamL/Zv33TgAbH0CKWxy874bB4Ct76ANpBf/8auWHX/NpZdfO1w8z4k/+aMzrwsAgOkctIGUg9uxRydJjUMAYJ4JpHPseuMtMK/X4VaYP3Tvwzd8mfPI8acAHAwE0jn22Hsc0bsEZszxpwAcDA7pXQAAAPNNIAUAoKstt8v+xKOO2WsIm8FR4x2ijnKHKABYYssF0qfdY+l9xKG3u52+ue8Q9QNvPn3JuMuuuCpJctEVu5ad/vqz3jnzumC9HbLt+nsNgc1hywVSAFjJsWf+ZO8SgGUIpACduTwXMO8EUoDOXJ4LmHfOsgcAoCs9pEBXD3jj05eMu3L3JUmSC3dfsmT62x/0rA2pC4CNI5DCJvIbf7n0KhH/u/vqcbhryfSnPfRdG1IXAMySXfYAAHSlhxTgIPF9r3v1knFf2X15kuTC3Zcvmf7WhzxsQ+oCWCs9pAAAdKWHFLaIJ71u6d2UPr/7qnG4a8n0332IOy0BsDnoIQUAoCs9pABTeuDrX7Rk3Fd3X5YkuXD3ZUumv+0HHrshdQEcrPSQAgDQlUAKAEBXdtkDc+mBb/jDJeO+uvuLSZILd39xyfS3PfgJG1IXwDwSSIGZ2LFjRxYWFrJ9+/bs3LmzdzkAbGICKTATCwsL2bVrV+8yADgIOIYUAICu9JACyzpkWyVp4xAAZkcgBZZ1zINsHujvwa//5yXjLt/9lSTJRbu/smT6G37gbhtSF7C+/MWBTe6IY4aeymEIAFuPQAqb3O0feGjvEgBgpgRSAIAkJxx5wl5DNo5ACgCQ5Mmn/WLvEuaWQApb2OHjmfKHH2xnyh97RGocArD1CaSwhd3ijIPzK374WbftXQIAG8iF8QEA6EogBQCgK4EUAICuBFIAALoSSAEA6OrgPAUXgKmc8bo3Ljv+y7t3J0ku3L17yTxveciDZl4XQCKQAsDM7dixIwsLC9m+fXt27tzZuxzYdARSAJixhYWF7Nq1q3cZsGkJpAAb6Pte94ol476y+/IkyYW7L18y/a0PeeSG1AXQk5OaAADoSiAFAKAru+yBNbv/m39qybgrr/hckmTXFZ9bMv0dZ/3RhtQFwMFBDykAAF0JpAAAdCWQAgDQlWNIAQ5itW3bXkOAg5FACnAQu+4ZD+hdwpb0yDd8dsm4hd1XXztcPP0VD77JhtQFW5Vd9gAAdKWHFGAd1Laj9hoCsHoCKcA6OPzMe/UuAeCgZZc9AABdCaQAAHQlkAIA0JVACgBAVwIpAABdCaQAAHQlkAIA0JVACgBAVwIpAABdCaQAAHQlkAIA0JVACgBAVwIpAABdHda7AACAefI/v3/uknHXfPGr1w4XT/+GnzttQ+rqSSAFZqK2XSdtHALAvgikwExc50Gn9C4BgIOEY0gBAOhKDykAG27Hjh1ZWFjI9u3bs3Pnzt7lAJ0JpABsuIWFhezatat3GcAmYZc9AABd6SEF6Ky2Hb3XEGDeCKQAnV33jPv2LgGgK7vsAQDoSiAFAKAru+wBgG4+8qLPLRl35WXXXDtcPP3bHnvDDamLjSWQAluK61tufYdsO36vIXDwE0iBLcX1Lbe+o898dO8SgHUmkALMsdp27F5DgB4EUoA5dsQZZ/UuAcBZ9gAA9CWQAgDQlUAKAEBXAikAAF0JpAAAdCWQAgDQlUAKAEBXAikAAF0JpAAAdCWQAgDQlVuHAgAwtR07dmRhYSHbt2/Pzp071/RcAikAAFNbWFjIrl271uW57LIHAKArgRQAgK4EUgAAuhJIAQDoSiAFAKArgRQAgK4EUgAAuhJIAQDoyoXxgYPWA9/4O0vGfXX3F5IkF+7+wpLpb3vQL25IXQBMRw8pAABdCaQAAHQlkAIA0JVACgBAVzMPpFV1elV9oqo+WVVPXmb6z1fVx6rq36rqb6rqJrOuCQCAzWOmgbSqDk3y/CT3T3JqkodX1amLZvtQktNaa9+a5HVJds6yJgAANpdZ95DeOcknW2ufaq1dmeTVSc6anKG19rettS+ND9+X5OQZ1wQAUzvs2BNy6HHfkMOOPaF3KbDlzPo6pCclOX/i8QVJ7rKP+X8iyTuWm1BVj0vyuCS58Y1vvF71AcCqnHDmk3qXAFvWrHtIa5lxbdkZq340yWlJll7pOklr7YWttdNaa6edeOKJ61giAAA9zbqH9IIkp0w8PjnJhYtnqqrvSfK0JN/dWvvqjGsCAGATmXUP6QeS3KqqblZVhyd5WJKzJ2eoqtsn+ZMkZ7bWPjfjegAA2GRmGkhba1cneUKSdyX5eJLXtNbOq6pnVdWZ42y/k+SYJK+tqg9X1dkrPB0AAFvQrHfZp7X29iRvXzTu6RO/f8+sawAAYPNypyYAALoSSAEA6EogBQCgq5kfQwrAfPv+171nybjdu7+cJLlw95eXTH/TQ+69IXUBm4ceUgAAuhJIAQDoSiAFAKArgRQAgK4EUgAAuhJIAQDoymWfgC2ljj1yryEAm59ACmwph591p94lADAlu+wBAOhKDykAQGcnHnm9vYbzRiAFAOjsKXd7TO8SurLLHgCArvSQAsA6+vU3XrRk3P/uvuba4eLpv/ygb9yQumAz00MKAEBXAikAAF3ZZQ/AAdmxY0cWFhayffv27Ny5s3c5wEFMIAXggCwsLGTXrl29y9jyBH/mgUAKAJuY4M88cAwpAABdCaQAAHQlkAIA0JVACgBAVwIpAABdCaQAAHQlkAIA0JVACgBAVwIpAABdCaQAAHQlkAIA0JV72QMAm8r1jjpxryFbn0AKAGwqj7/XU3uXwAazyx4AgK4EUgAAuhJIAQDoSiAFAKArgRQAgK4EUgAAunLZJwD266zXvX3JuCt2fylJcuHuLy2Z/uaHPGBD6gK2Bj2kAAB0JZACANCVXfYAAOzT557/piXjrrn0imuHi6ff8Ke/f6rn10MKAEBXAikAAF3ZZQ/Ahjtk23H52jgEEEgB2HBHnfHDvUsANhG77AEA6EogBQCgK4EUAICuBFIAALoSSAEA6MpZ9gCjOvbovYYAbAyBFGB0+Jnf2bsEgLlklz0AAF0JpAAAdCWQAgDQlUAKAEBXAikAAF0JpAAAdOWyTwAckEO2HZuvjUOAtRBIATggR57xkN4lAFuEXfYAAHQlkAIA0JVd9gCwSbzoDZ9bMu6y3ddcO1xu+mMffMOZ1wWzpocUAICuBFIAALoSSAEA6EogBQCgK4EUAICuBFIAALoSSAEA6EogBQCgK4EUAICuBFIAALoSSAEA6EogBQCgK4EUAICuBFIAALoSSAEA6EogBQCgK4EUAICuBFIAALoSSAEA6EogBQCgK4EUAICuBFIAALoSSAEA6EogBQCgq8N6FwAAW90Rx56w1xDYm0AKADP27Wc9uXcJsKnZZQ8AQFd6SAEAmNqJR23ba7gWAikAAFN76t2/f92eyy57AAC6EkgBAOhKIAUAoCuBFACArgRSAAC6EkgBAOhKIAUAoCuBFACArgRSAAC6cqcmANjEjjr2xL2GsBUJpACwid3rzKf2LgFmzi57AAC6EkgBAOhKIAUAoCuBFACArgRSAAC6EkgBAOhKIAUAoCuBFACArgRSAAC6EkgBAOhKIAUAoCuBFACArgRSAAC6EkgBAOhKIAUAoCuBFACArmYeSKvq9Kr6RFV9sqqevMz061bVX47T319VN511TQAAbB4zDaRVdWiS5ye5f5JTkzy8qk5dNNtPJPlCa+2WSZ6T5LdnWRMAAJvLrHtI75zkk621T7XWrkzy6iRnLZrnrCQvH39/XZL7VFXNuC4AADaJaq3N7smrHpLk9NbaY8bHj0hyl9baEybm+fdxngvGx/81zvP5Rc/1uCSPGx/eJskn9rHoE5J8fh/T90XbjWnbc9nabkzbnsvW9uBYtrYb07bnsrU9OJY9y7afb62dvt9naa3N7CfJDyZ58cTjRyR53qJ5zkty8sTj/0pygzUu91xtN3fbg7VubQ+OZWt7cCxbW++xtptj2T1f856fWe+yvyDJKROPT05y4UrzVNVhSY5L8r8zrgsAgE1i1oH0A0luVVU3q6rDkzwsydmL5jk7yY+Nvz8kyXvaGLkBANj6Dpvlk7fWrq6qJyR5V5JDk7yktXZeVT0rQxfv2Un+NMkrq+qTGXpGH7YOi36htpu+bc9la7sxbXsuW9uDY9nabkzbnsvW9uBYds/XnGTGJzUBAMD+uFMTAABdCaQAAHQlkAIA0NVBFUjHW5EeSLvjq+q0qjriANqeVFWPqqrrdVj21G3Wqe3RB9p2bH/sGtp2ec1rUVXfWFV3OtDPyBqWe7Oq+sGquskBtD3g93iNyz3gtuvRvoden49x2Qe6zez2PnVse8Dv0xrbrqXm61fVDavqgP6Wj1e/OSBr3IZcZw1tD/hk7DW+3jWt6zUs96iq+uEDXWdrqXuj/6YeNCc1VdVZSe6e5BUZbkf65SnavjDJNyb52yQfznCG/2WrbPuyJN+e5NNJ3p/kVW28q9QGLPu5SS5K8tbW2r+vdpnr0PY1Ga548IrW2j9N03Zs/w9JrkjynNbaO6dsu+GvedywPirJi9pwi9upVNXbkhyZ5E1JPpTkP1prF6+y7bcleUyS17TW/n7K5b45ydFJ/jHDDSXe31rb1x3MJtv+TZIjkvza5HtUVbW/y66tcbkH3HYt7avqmCQ/m+SlrbXF10JezXIP+DNSVW/PsK4P5PNxwHVX1bYkv5rktUn+vbV2+RRtu7xPnduu5Xu8lrZrqfndSa6b4Szn9ya5qLV2zWraju3fkWRbkqccwPZnLduQlyU5PMnvttY+OOVyfztJS/Ly1trHp2y7ltd7wOt6jd/j30jyxAx3p3xLkue21r64QXUf8N/jA3Ew9ZD+WpLvSPI7SZ5eVXcZ3+R9GjfKN0jyF0mOTXJWkp+pqgdW1Yn7aXvdJB9P8uAkz0+yPcmfVNVvVtU3r2LZxya5fpJXH8CyvzvJ6RkuzfWzVfUrVXWX/S1zHdreNclNk3wsyY6qeklV7f+WX19vf4ckX8nwmn+lqv62qh6+AXUfcNskz0jyvCQXVdUfVdUJq2yXqrpBhlum/XaSm2UIl4+pqu+pqhut4imekuQWSX6qql5WVQ9c5XKPy/DZ3DEu/w5JfqKqHjuG3H21vVGGz+NfJtlZVX9fVQ+f/ENSVbVC22PXsNwDrnlsv20N7Z+ZYRvyN+N7fOv9LW+RA/qMjD1mFyd5dobv1bSfj7XUvSPJo5P8SpLfqqr71Cp68NbyHo/t97zPv5TkkmnaL2o77edrLZ/N649tnp0pv8drbLuW7/ENk1SSP8lwV8Q/SvK4qrpFraI3rapOznDLx7cleU1Vfaiqvn9/7ca2e7Yhr07y21NuQ26S5I4ZAtYLq+pNVXWfVS735hn+Hl+TYdv13Kq68yrbruX1nphhXb8gB7Cus7bv8a4kt03y2CS3TPL3VfWcWkWP+lrqXuPf1ANyMPWQ3qm19oGqOjXJT2T4QH80yRuSfGhf/zFU1TGttd3jxvhuSe6UISh+YXyO9y/u9dzzxaqq67TWrqqhq/+4DB+M+yb51gz/zf55a+39yyxzT/ttrbXLx43WXVez7LH9t2T4IJ2f5F5JviXJSRn+uP11a+1v9/F675jkqgwf5Hsmud0UbW+ZpLXW/msMl9+VoWf6qiRvTPK6ff33O/5ROKq1tjCu7zOSPDLJMRl6t/94pfZrrHstbR/cWnvDuIF/XpL7J3ln8v9zd97hWlTX/v8MvXcERBEFFFCxYkPEHhuKYgWxt4i9xFjAgqjYFQsidsXeK1aw90SvJTGJJqbce43JTXJvYnKTOL8/vmuc/e6z9545h2Lu7zzP+5z3nZk1u629+lqb0/M8/0kCrlhjlwiPQYrHAOBT4E10PO4v/HFnUnh2RgRyMLAVmu+vkaXlnjzPv0602ybP83/YtZHAZsAwe+wzdDDFO8UzHmzHwsuQZdm+SANvA8xDWnzUA5FlWauiXy1ot9l9ducrz/O/Ob9rw9u6vI728AXA1sBrwOV5nr8VG6sDvyvwMGJotXHEYN35GoPWfHkq8MOe38T62RM4H9imbr+zLFsb0ZhOwAnAOOAXaB+/BvwuhF+B96yO5nmoXUrNc7HOrQsrjNGyTavgI7DNwpGW9tued3FzLKJdlfs4AFtrjSPjrdXnCOxOiNZ2A54HHgM+zfP8r5H5aQv0zPP8C/t9BBKM/wlckOf5DRG4xaIhmZT45fI8/zjLsoHAXsCewH8Dc/I8vz8EZ7BDgL55nr9u87wlsD7wBXBXnufPJWBbGex/OuM9BfhHarwOfNs8z/9u33dCR6F3p95cjwVepmX7+Jt27fdaSDjdDHgbWUzfW9L9dnjqbxDtqM1TW/r3f0Yg9f9M2ykW5VNEZF/N8/z3zjMdgHUQUf6LwxQ6IWvrhkg4/AewIM/z5x3YDBXzHwH82ntvT7u+CRIins3z/BKvfxkSKFdGJvK/2PWOwJiKtjsBf/eQcBBCiLWQ8PInFAbwVBVTMU1q9RbCdgIGIYayJTL9PwbcFmBGnYB/ukKDM+ZtkSKxArLMXeqObyn0eyBSGtauA+sJlb2Rq3NfxEzOzPP8tQBMR7S+A5FS9KVzbz0kmK5gbd+fV7iI7H2DEH5sgdx4C9Bc/8V5rjswMM/zjwLvGIqE2lWAfsDTeZ7f59xvjTwGfwys03gkuCyHPApX5AE3r72jlYefqyBFYOVIuy3usz3TD1kZjveJZ1XboT9TEE9GVoOPbazPpmA8+N7AOcBkIjhiCs5heZ6fG4BfG9gVHZucxA9P2OmJhIZm99uUxYOBHZDl8mFgUe64EK3P30En6P1P7rj2Mll1xqL9FFun1kAP4G/AAFdYN2FiHHEcaYVo4v8i3vRH514Vfi1uv7vlef6nLMv65o6rPcuydYAJJNapgMX7q1pjw8G+eZ7/2BUs7d5QxNti422PPHbtgM88YXVzFF4yAHgHCYdBpcnmvLW3l3cHpiPl6/I8zy/0YDoBnfM8/62Lm3ZvR+AkEjTEF7DsWg9gIhKW2qJDdJICosF1Qx6mLRDd/AoZTR6qGqdzbw/kRYiNty+wY57nNwdgx6I9VTnXHlwvtI93J7GPsywr5IeQwDgM8ahdgOeAkzyFZ4n2O8uywcBIWsCPa//lzTj4/tv4IIY+FWkVKwTu90ab72PgKO/ehcB1zu8MEcnid1ukib4KbOrBHowI9q1IOOnq3S+E+ZuAbQL9OghtyLuAnexah5ptX4I0qHaB9w5AQvDhSOPy+7UeIvr7BN67XAXsOMRgjwJW9u61t/7uhFwtPQJ9uwBYL7GWbZCC8Hug9xLs91rIYrUD0vh92K0TsHsU6wK08e61A85AGuH8Ys2d++cATyML5x52baj3TA8U/rB2ZE4yJNz57RZz/QBwrnf/XBQWcQuwlY+TznotBNbxYI9FQu7HwMaRPq2CrLRrBvCjn/O7FdDWe6ZDpN0W99nuXWF9WgCMsmutAzgaarsvssCsYDjYxrnX1ubkY7Tfs8h8HEoY56M4Alxlfb4P2094e7oKP7y9U7vfwI5IgG/yXnvXJOAJm89Wzr0bEHMEKeQ9EKOvu05H2XhvAa5Hgs1mNdfpSOBJu3dUZB5i+NXifqOwhtuA+1Hozf7AGnXWyZ69H1mqJvs4XQE7B7jIu9a95lydBjyL8hq2tWs+7RuAFI9RgXns7uFTRtO9PBLh70jv+hnApAp8jdGQyYYbbwLDAnBdkXD5BTAk8u5WgTnuhLyIeyED1CDv/jTgRGRN7R55b2y819r12UCfyD5OzfVkJL+sHrgX3ceIZv0N7al1kecRmvLO05Gi7r+7xf1GBqR9UKjSbt69/iT48eJ8lshLlubHFukCZJW7BQku6wBrec/d6yKhbbB3gVXs92HAZYh4XAr0d56dQSNRzpCLZYx9HgDOce63d77vGtgcmW2KzVA80R1Ia5wOHOcuYKDtNYCfO7+HI4H8YK+NDYEbAhvqBRQnMg25KG8EtqsJ+waKR5oFnBZZj9WA2YHrI4AfOr+LWKoDvOf6AbOWYL9HoEDtJ63vhxdr4D23QQB2P+TmKX63wjau99xhKAjevdYd+ClywWxo7d+K3FWzaFR8Tgq8c2+aMqhWgedmA+sHYO9D1syX0R65HzjFe27/QJ9/hojQgUjxuQIJLlt5z14XgP0z8CNE6Fb17vfCBES/3cXps13rhhI+uiFmOAeHKNO4f/wx72ftXIfcny9gDDywxtMC16cgq9sj1ocudXDE+voyctueB0yPrXUEPzZB+8dnjq1xmEqo37a2TyIh7QPkKRiJvAXtvbGd7PzujKwl/e33LES77kFejREV69QdhRj1sjavQULvZcBGNXDzp8jbsDUSWE5w7ncs5iwA2+J+W7ufoz0xDvGZpxEjHpQ7tMRfJ4P9CaLRewHvA3s599skYLshl20hKOxlfb4UGTPaVfT5Z8gYMx4J03MN9jBMcCnmIrKf7kQCx3APJ3y6eXMA9l2n30MQzT4oAOvTkG7AJ4gWT0N4ehxS9tz90BGFK7mwXRH/dOlqSJlfHnjcu5Yho8FLyEh1LAoBmUxTQSw03peQ3HE5cIi/j501Ds31wbkThEUAACAASURBVIZPNyClZXXEL9an0UAV2sdDbb6eQjxuhs3btd5zW7hruLj9tj6+h+Su45Hi8xpNeXkTfry4nyX2oqXxsYl52/n9CXA7EhDvBlZzF8CD7Y82ai9nE01AhP5qZMr+ZnE82IOAB5zfgxBDKjbh6XgatAc/FRN0bIN8YZtgX+v3Bom2j8E2MiJSzwMzEaG8g9Ka14em2vStwNH2fSUkHF0MXIncpSRgbwOOtO9rIea7p3N/PQfROwfGfDOK/wFpT08iYapQJNrYvbY01dIWp9+3AlPt+xaIMW3s3B+WgH0R2MS+T0IE/TMsptOut0FE2+/z5sg1BLAqil3d2MZ+M7B3Aj+6I831N0iRGuvdX6FoDxgdeccM4Lu2HscCf0GC/IEErHwGsz9wr31fDblcpiBL77t4lmkP9nAkUK6IBNn3DB83tPsz8azDS6LPzrtvtO8r2/y+hGd9icA+76xx4ZF4F+3D5RBhbmNr0jYAv4jSAnUPsnr+G8Lv1WI4YmOdZ983QoLwAszqQ0D58Nr9ORLaZwJHIAtJOxcWCadN+m193sy+32Hr9gKinTs6sG0p92WGBIHLELPugfbHEGTtOheFH6T6vFmBX86eu99w5x08r4sHewoKTflm31qf+9nv7wHLB+AWq98o1tPt80AUv30eit3sl4A9A7k6i99jkeu0q/0OWtUd2Pfs+3BkDNgDCTALgS0S7U6gpD1DUTLUjohvPQVsX2Mf/h7h8mxEF0YBXez+NgWe0ZTuzQQW2vfBaH/Nsjn/BFgx0e6JxRqjELD/Qrx0IQq7K4xH7WnKzy9ENPMjpOj4CvE69r87ik/1297I1uZ4w5Vbkcdmd8yybLjkj/ccSn68i43xJkrlp2ofP0vJO+9AsstbiCcWntMU/dnZPn0MZ/5o7xzj4FnIq9PifiP56AL73tY+E5ASc5DzXG8i1uaWfv7Vs+zHIGJKlmU7oLjKfZHV45fAXlmWtbL4v4YYnjzP/wNZOecgJjg/z/OHcpUxKphgETvoxz/8B7Agy7K2mRIpPgf+E9jDYpV2zdMlEH6PhBsQMlyV5/kVeZ7fjgjzpETbj6ONARIMZ+V5fjraOH9F1mHyPP8yb4yxypArvbXd/wVCpHet7+dbEHsIthsixC/Y/fcQUu5p9/dH2jR5nv8pz/M/ux222MfWwP9acPtVyIp6NBJ4WiGCR57nf8+dsjmL2e+eiJgvsPsvIHw52O5PQUJ9CHY0shoNtHit45CQswGymOxhcP/I8/xHedNSPx8BXbIs+yWyMt6Z5/lruYLqF6J1L2KX/L8xSOBeGQk387Ise9piOEEEc31rvyHg3XnfbGC44f2fkYA0Awm3uc2r//c68I9MpciuRu7C2/I8n46I5TaJPi9Ee+lXeZ6fiCwzPwEuzbLsfeTC+2mo3aysG9iSPmNzdIbNx2d5nh+AGODJFpdYtNMAb3v1v4E2WZa1y/P8p2ht90QKxDjbf//M8/yPedO4tg2Ar/I8f9rGMA4R5fWRpXiq9SmEI79FTIE8z19HLrCPUAZ1x7yMZ28y11mWrYYsm5ciwXQosuZMtTiu+VmWfTfP8yb9zpQE9Mc8z1+0OMOtkBV0C7SGB2VZ1tNg/55b/F+uv69sXPci/FyQ5/nP8jz/FJUy2ik0z87fh0C3LMuuMjpwJvCzPM+vQwr9hgn4921uMZr7E7RWa2RZNgopeL/xYZ1+f9LCfr8LtM+y7Pgsy7ZBnqx/y/P8NGTZWttgQ3viU7Rvijjfl4A/ACsZfTkqz/M/RGBfAP6YZdk7SGm/Oc/ze3PFTd6P4mFjff4B0DrLsoVIOJuT5/njeZ7fiOZ5fKLPIJzaEVnbfojweiowPsuy45A3gTzPvw7Qvd8Df81UDuhR4ME8z0/J83wza3vdRNvvA19mqtpyDhJ6ZuZ5vjlS9jaydv/m83PEH9ZB+NwKeCrLsvlZlq2SZdnGKKwO2w9f+A3bHrwe+FOe58ej3I2fIdyYZM/kgfFmSDkhz/OHkSD9FYp3pWIf90J0YKzFIo9HAt1otP5HZlk2wOiHv4+LdX8XWU9XQcLsLxC/m4GSz8hzSYeBvxb1GynCBc8tclkeQ3ixo8Vkk+f571yeukT+YpLqv8oH6JiX0vhg5/qeeGb9AGxbpCG/gIjHd+z6rcCJeUJToKn7YTMk7DyCuV9isIF3uZaTOQ58SLPpgBjRz5ClYxvn3qvAmATsaKRpF9bJl5x7z+BYGwOwI4CVnN/tUaLYYIPdMjIvhdl/RST8XYOC9/H6vWFsjq3frzen3w7shjiWE2R9exxZxp/HXNGBfg9BispMtOnvcO6thZh3B79du++6xLYz3FiAFBUQ4Tg0hiPIer+ed+0Y5Kb8BaphWAevjkOE/EvM/Y9p2SH8cPp7gn1uorTy3Us83KFTRT9+DRxYZ084ff4vSqtGVZ8Li2DmfO+JrCZvIutBDPZAZFGdihjSI3Z9C8PvJnHaDuxymLsXWcH3du71Q26sjhXjbUsZyjDc2lyELCIpK8XKzvfByIo1AzGF3xMIG3CeLyw+nXD2HWLk72GWsAT89kjQ+goJDv3R3jykao2RR+gUZKk7Huhl1x934GNr5Vun9rHxPgbsG2rbnX8UQ3679fvsuv1GQk4RFjaL0ir7GOamDPUZs1B7OHq0zdk9wJQa87UxonkrONfuqTFXI5AgtSey6I616/dhFmEflpJm9nbxB4U87IM8H3/AwsNC/TacGoHifZ+gMfzsYWC/xHx1Qsr7tbauRzn3HnTaDcGuTGNIXk9bsw+QYHZ4rM/eProdeYd+iDyw62AhUW67zlwVVsh2lN6EzZFwfT8K4Ul5dwol9jZk2XRDBT+kRvwl8hpeghSR/Yu5jOyHot8FXrZtbr8NPx5AMsjh3r13iHjtlsRnqbx0iXRMiUprJe6/QlxI2ohGV+86SMh71xZjDhFGZEi7NZ7LwO7dhUzmTUzrzjM9kVbdDc9dg+I5Pg61bYjjxpRsbQj8NQqIPgtl81fNW38Ur3cYZQzUROD1Zsx9ISAcDvyKhOCPiHJvJBi2RgxpVef+zsBzMVjneyHQLk6/i3CG01G8zt01YPoh69PazrUzKBMkQkT5J/ZMW+faeETsfogOT2jWXDu/f0WEISBCejSOSwoJlpfE+urBj/Bw7DRkzb4feD4BV4z3G/euc28o8FYC9vvI2uO2eyxwXp0+e+/y93lbvNhe/3n77IfqF++NhRUgYfaqxBr3pGmSiDuG6TEcQcwzlgTVCRUSj/W5yLxuMl679hKlK85vty1e4qf7DFK+Yn0+Bsf9Zv3YC4U23I2F8yT63R8JNU0YLGKEHyRgV0cKw/qG452LPqJ9/J8VuDmNMpxhIPIm3VnVb6RIDrTvPWikR1vixPJH1nFzPOEe0f3Pgc8TsMshYXJQ4N7GJBRSPOHVrh2IBLxFqIB5bdpDo7FkfeC/asK2pzHBcRTwfuL57lioGgqx6I346Vm2l95OwLrx4q29dRpdgR/tkXWwEOC2QKFSC2uMsbPhQUgW6AOcn4AdRhkqNMTm+hDEmyahHIlrCjz3YE9GfLQQLjOk3L3SjLUdRcCQUNVv79kdkFL2CxQ+N7s5fWjJZ6m9eLE6Jen9K6SBFVaUYnHaGpIEhQ1kAfrSEH0KZVzK8vZ/sPOsz+C2QxbQH6GEhD3tuqthfM++tw60vT3KtH2RMqlgO2T1bGX9DlrOkLY3J/DO4YjgjsPicwKw2yLL5B00zS5sjWIyd4rA9vN+uwxsRaR97h2CtWvHI9fNncBEu+YSkOso49ZCTPAQ/512r731e3wE9jvI7Xw7jmXX7q2HBPk9IrDbIqJwGmY1d+4tj6y1RVa0jyN7Iq18HrCPc70rpsVjgogPa9duxgSOwL1+KLQkdG8KsjI9jkIZTrHrvSgtYk1w0q7vavM0GytGjRh3RySMb0YktjEx3kLY64jFcgZgt0JhJtcjgjbZ1vVQSstZsM92L5bYk9G0IoK/TpMQ8b8U7QFXiO6LLHhDQrB27WpUgDuE80ORQtwER6iZBFXR7q6hdu3+Y5Rx0f6Yz0NCyZfAut58DTLcCfV5XxrjKDvSKGy0q+jzPtavx5A7eJLXdncsrjuAI4cj1/hdNmdXG34UisO2lHTYr6oQxE3nfttYv21t/4EOptgNGSIyJBC3Qntx80ifd6dMylvoP4cExCmRPvtJdi/hJNmhpNZtI+0eiAmczpqOQKFHY5AQsnwI1tkTvyMSU4vo5u6RfjeJ33XudUDepv0j/d4L8Yj7sCQ6g1nZ1vtISmHVhz0EjyZ66zqFkqf6fZ6IrM13YTGRdn0IpRU8aFxCe+IJg/0tott+pYNCcQrtiQexJCbn2iAUjnUtUpD7+fC2nj92fneza0Mola42oT7bvZ1szPNp9IB9QzMpY4VjFlJ/HtdGVv+tKI1FtQ0Jzfks8RcukU6JSNyCyh08SdMSHH0wocdbzIL4PYdM3Bci181MnID5RLtPAbvY9wm2iYqQgcoFMPgiLGB7ZC27k0jGutfv9yjdg1sigfp7mIu+AvZjFLN4hyHOJGSxKwSFJhqeA/saElbGR55JuQaLagIbofjDx22j3YcEge4Vbb+JYgpBQsJBKDGgS41+F2O+3ca8t23G1e2ZJiVp7LqbQXgCcuu/SulqWpFSsA4R9UW2MbdEcWvTSBAID3Yf4FH73hrFBW2MNnxr5BYqGIrPQF/CEh1QeZO7KYW6YGiBA/uyzVWGmMOn6Di4E2v0OTpev48R+IuQG/RgFN/3tI2lMhieGok9EbgRaO9NQcR/J+RxmOI8MzA2BiQUfOr8Hm34UoTa9KIUGvx1SiVBDa8Yb7LdAm8i7Y5EsY8g79I0RHvOo9xjMSXrZUqB4FCkQHyBGPLIVJ8d3Nzevm+LmHfhrYh6k+z+s5TCalfDz3kohCTZdgA3pzu4GVV07H4ntO/vRu7U85E7OagQBsa7GxLcD8UsXTXHG0uyu4uIourAPkIZgnSyre9nNt/RcnsO/BRkvX0PhcxMQ4JXYWRJCTr/bng8JfVcBPZFJByuioTxs+3aLAIJsoG53s6+b0oZEnGwzX93H58d2IUoF2A44kmn2v/LSQjYBvsqJS/vgWjQL9CeroIdYXM8HSmlftmzIF+ze9ejwzYwvL4byRXzMB5V0fbLiP+PRHv/QudeHRmmoXJBbG6X1meZNdRMBH4Xc0sY0r6EEhCSE1TcQ27fiw0xJiM36DNGPIKZnvbsmzRqX49hcS6GzFMSbfdFcYSbOdfuotSmD03ADqfMmuyHLHTnIDfpAwTqlzmwx1Fm9A9GjOQyZJ19kjJbPMR4j0FEZl/KEimTsBISiIhEY1wQgXjAvvdBxG53pK3PASbENgJKHnjSvveyTXc/ivd9HHNLR/qdGvMCHCt4ADaVQXhgBV6OLfpsv0fb+hzpXEvh5+uUSsIpNuc/QDHN0cxYFNP6lH1vY587KC0DV8f6jqwQz3nXLkJhLDcQqJtXY7zfrRov5V5cH1mDBtjnp0jRu4mEsoWY12PIM3Ao2s+XIIv8YLS3vhuBvZFSeDzV2jzL8OqOED568PtSZvXvgxjbcfb/VeLhPqO9dfpPtBfaG35euZjt9krAno2Vg0FC3ecoceUKxEiDwjBShv6M9uOaiAZuhITDS1Gh8FSfR2OhRMiy2AXRu53t2lVY/GcAtr21cTWNsXVtbb2uJ271r8TNqg/yshyOrFCTER1ZYOsQzBZHVvsXnN+9Ee0qYvtnADtEYJezudmMUrG6Cwmml2IenUR/j7O56o1i+Xoanp2C6FcyNtjBsWFofz2CrMQLSYehbUwp6LyKhK3jKD1BJ+JVCnFgh+OEA6Fk5P2Q8nU9cESi3YHWx4KWvIEUh6mIdqXyKTYHnnF+/xYptWsi2hOl9Ta/DyGlw/UOtEVepigvt+fmUlaN+T6SPbZ27se8Hz2QYnU6kideRIaEFZHB5X6csleRMS9yfi9vuFkopMdj1vzYfgxcK+Z+ndA8L+nPUn15izokBnShd+04W4xoqaXAe2Yjq90oxJAORppRkKgjoWgbGmuyrYeYWE8UDFxV1uYARDCOQNbZ5+z66siE3mTBHdhHUALWdBzLFWI05yXgTqG0bpyBTkBy5+CgBOxEm5fOiEAdgCyOdyIB4FcV4x2AGNirmJDh3DsIZWDGYGchReFc2/zT7HqGNP6UgHbaYoz5O+7zdq2NbeYHSFixUAb+Wva9YCg7III+LbVhEdP5GjGkcZQ1Klsjt9R9eDGLDmxfpC27CQSbIEVtBaTA9U60PcfWdD0koBV4eazhaYxAbkBZhL5tc8brvaeowbsLslwOQYwxmhCEhJVmJ/YggWgeZVD/PZRhHz2Q52X9iv4WJeMG23xt79y7Gq8+rnNvIIuRBFXR7lV44SUe7HqIwc9ESs5k5955eMkJzr3WaE+cjeIxXdd9Eb6S6vNgxLi7ULqsd0FMuEcN3Oxn+Hms4bQbH/1jIhY0JDQXuNncvVgw2R5oX+yOFKf3kdB1A5FEPsRPDjD8LKzVRUm0djbelOJwoL3/SJqfZNcT0bej0B5whfh/i81Vsc72f3NKQ8BgFPJwC6pGERRobcwTnN/b2Pp+iPjEr2NjRpboexGNvwV42bk3Flm5ox4eJKhfhPb+RQ7OHolTTjAAN4bSiLUlcIZzb2uk7Kbmehuk2G7iXR+JyjbFEl57Ig+tm/h6lM1XUGj3cHIzw4u3gRe9Z14lUerOYI8x3Cz2xFmIZmWILsTKkH3XcPhhPB6I+M+9sXaX5GepN9CiTpVul8JS1w4RzI8xzTsBWxDFwYgwPYOdYkBZgytm1XEzNgsGfJUtUjQBwoHpjFwKFyJtpIh/PRlLdEm03ds23k1ICC5czzeQyMovrlOGK7gW3hucsddxr7ZDwsKWSKOMxst6Yz4JEdT5lBmL11NdyWAockn+ksYMSrcuaZUFrltzxowIxt20IIMwMY51EePeMwZb4DUSqP6OjgJ0r79PDQuHux6Ikb+FKXCJ/vVEjOMdJOisa9dnOnhdW/utM14fZ5Di9DFlaETStVnRfjSxx64V1vW2eBYBRNQ3qnh/a0TY30cu1iOce69hjM5f28A117oSTYLyYI5qTrsO/mSISU9F9GYGYiQZUn62DsANohSg+yIm7iYknlnVZ3/clPT3Vix0qsaYVzX8vB4pm9chxfiGKtiW4Kb3fAdr+30HN4swrRjt8RMROyHB9sEa89UF8YiLUAxskeOQTLJz4As3bhF2dLDN15wqWOcd+6K9/xpl3GfXin63o2nx9YEoJjXZNjJcXIgE0NmUvG1ajfnqiwSq+21PrGrXT8Ws9z6sj5PFvDvfvx9r1/q6so33RKRA342UnfHI+nl9RZ8L766bfHWSzfceITgf15BXbJzze2fqJTX7uDnYcPNW7ECbwJgnI0VuNFJeb0v0a6laSZfai1vUGQkXvQxxQ5ltJ1DGk9QRsL6PiFMyxq7iHUcgy9basXaRW2In5GZrT6OA1AEJAhuE4BHDGGnfRyBr8KPIsrMAEbmgZdU2TtCljgTLD4m7F8dQCvyt3L4Z7L/H2rVnDvbXCDHDl5Em9kSi33vgaGq25kUfBlX0ey7OwQJ1x2zz3J4ydnAnm+fPKQtERzMIkYIxD4VBhAp0jyMRF+nPBY5L0IhAjEAegoSLF5AW61pJ90YJLA0nyjj310dJdTtjjM+519/WKprY41+3OSxwZPPYeCkZ3oOU8ZRbEDjJJAJfJJxdQ6Awua1bMLGn4r27EKn4EJn7scjl9SVi/pcDD0dgz8Q5qCOAly8TieEMPL8JsuD81nAj1W4nJDx39q5fg5SzuxKwdyJl282uL4wByyHrVbTPiFbdgBTogd69qaiE0IoheGRVPQQJoUXC5ShkQTwGuXW7R2C3RolfZxBwjyNLUQw3d0dC/y2U1SzG47j/a+BRRlOF61QU+rBCqM8V+6oXjYcm+OMdhfZxUbC+HfJunYd44rEEcioc+JFovxbCXAcU/vJajbHujJcITKOg9Xqi3+NpSneOQh7Lt1D8cCyueXVK/tQf0ZRnDCcfRvxllQjsuajqS6jQ/DCkfMTa/Sb73X53RMLk3UhxuYAKw1ZinY8CZkae2xdZKO8lQEdsroJJunZtCAq/WQ8vxA8pA38hEjdva7i5gxuPAcfY734kDmlY0p9l0kjtzohA3m9I8RaKf3IZd4fIhuuLhM/VaFrLrkj8aFYgtveO3WIIaJvlDaxAMRaQ7CIOlgUeQIQtkHb7Is5xnIjJ9EXafq8I7I4oNuxQJFwWp0j1RvFfB2BH7/kIjBKubkmMdzUiGZN2bTvgiaJflAkMHWw+xhLIIHT6/SiBjEEUg3YopWXW7/c2KGt7Dk3PjO+KmNjxEdgjECG7FzuRCBH2VQx3ohmESFH4ALmwz0RM7XsEjnyMzOcB1sYipHW7lvg+KFwjlP28Bkpa2BJZU55DGvupztzHqiesjjTyexCBfgAnec3WKVY2bQTag352ex3ry0jsCEUkQN9ERa3OQNtFQtLVhqvb0OiCrp1U5cB0R/GjO8fGghj7k4ZHruC/InLLru7gdSuvzx86v5dDe6RYmwFEkqBS82vvPSzWrv2eRWktzhADGYKUr71Q/N+ASJ8/QeExr9O0Lu4AEnH7NN0TE9GeKLKoO2KhM4E+j0RK62421/+FFMnk+egO7BtISDjaxvAsiZPRvP30Y0RjDkQ85jMaT6SLxawOQ0Kce3RlK0pP2mAi9amJ7ydXsAvGzNN4NPL1VJTfiszX8/ZZhIwYGVJWi0oGMbf3Kug0t2eR67w4caiISVyeMsnKH7MLewXmeTK86oKUwxiPGEGkdB7aW5sTqWSC9spnzu+BSKgu6N5qlMJWCC9/afP0DM6RswVON2fuA31vTUAotPX4BCUEForvicgIUiRZp07AGoF46qOGo48hj0zhrdwEuDgy5v5I0HYtyBti4XYoKfSSxRl3s+ZoWTVUY7F2RISmoyHdmqhUzEvIpJwi5PNQcPZDSGseRWNM06oJ2CZIRiPhXqWi37OBs+37uobMbuxYk1pzzr1HEdMejAjWgciqdGyN+RqMXKC3ISvFVKSBFXE2HYgT10cpGfMYxIhfpUbWtcE8Qqmt74IsjJ8gxlRVdPs5ygLyO9gG+imlkNWTOIF8xOaryF70z3jvHhqzbfiPkOXkCCSYnYk2fp3MxWuAc+z7VJT9fZyt9eukKxEUAtpaKHnq94iRF+EMbYhYVZAQO9e7thFiLndU9Plm7Px0tJ9WRRapuSQC2+35u5Ar7kMk0PpxVFslYG+hVIS6IIH0Kuf+hIq2UwlJyThsZ61j1t4q3Jxj+2C27ZHDkLUuOl6nz7OKuTE8nY8U7DPr7CnvfQ21FiuefY3SCnkOooW/QoLA4ATcdZTFxGfaOg1rRh+r9kQqSebqAjft9xFI4HmQCqEU0bpi73RG3oMLrf9VoRgX4CWWIeHmaRK1Ye25+1AN6t9ZP30rVDBZ1u5V7adxCdjQ0chjnPvJNUNCbKGkX2Y4+YjNRdTyZfuoDRImT0DhIA8ifhOtGVoBewfwbg3cmuus8arIAHA3NcIwEG8oXPkTkeV5ru3HqjWeQ3nAx+XIAt3ZHVeN9oP0h3QYxq6UycG9kHeyOLf+PspE2Bhdu53y2O91kUB6OTJE+C78kId2EI3Hp3ax9RqD6GE0BnxJf5ZJI7U6IqQL1eHcARHZWO3GNigeZ21bjOsoz9cdgoTZoDvGJv4y4q7lEZiAF7nfHrn0hjrX9sCsj0gzmxuBXQ+noDgidichi9DbWAxlxZwdh4TQdVGM7ReIcW9KPEFmbRTAXoQQLLR2d0XMLZmliqwMX1NaUF9EFqwNkUIwJ7b5kJA8m7Ks0gdIK9zC+pGa6w1wXOpIoHwAp95iAnZbSo2vB9KCj0FukvkkznB31nQmEgZ/hNXqs3vXEsmqtfsX0VgDbw+bo6dwrOkR2FURQ9o4cO8mzMIZgZ2ER4ARsTscWYpjWcQdUCZrP6Qcno2UjQUoxuggIsKwwV6OYwlClpLn7ftBeKd4BfZjixKSkBWzu3fNVSy3pPpUpS6Iee2O9u61tleuJnKmucE87+DHOygcpLPNV9AF58B3xyt07t2PZrciT8pVKHSjN7IsF2XDziFiwUfelw8p6yj2sXV7AeeAiMXYE3Mw63AEdnecRE0kJO2I6Nh8AmeRO88eTqMwezuytO5jsKmqIKNtTZfzrg+wPgRLJ9la3mT7pyPiGb9Hwveq1n4T3tWM/RSrPdwTJY26/OVUYJ59n4KTrBNZ508orYm/QN6a4SiT+34ifMJ5x+aG3+uhBMq3bb1vp7okYQz2Djyh3IHpj3hhURrrXqSsnGiwSa8U2i+X2fdrsURlZOCaH8NvW5/PKEsOjkRhAU9RQ1FDez6VdLh5DDcN7581vHqIRm/puXhJ3h5sG7R3D3Su3WR4NQfJQzHDVGskgA52rhUhWUciPl/M5VKpO9qkT8uikVodkcv1YaSBrejdi564YRO6Co3ZmSOQ5e0pxFCKwsq+ufp0yjIr3RBxOYjSldmBSByTs6Bb4TAcRNyfR8TkPiLufuTW3sG+b4wTX2cIei/VDLRnscmQq+JTpFndTVwgXQsJ6xeguLbHnXsbWLuprMe22LGtNrcLnXsdkQU0VWdtnG2UrQpkt+vtkJsktmmnYJZVpx8zEGFP1uBDisnrhl+PYS5Ou3ciTnWACPwKBjsHEbktnXuvkLZwbIUsBT0Mx+9GxKuL4XrKstIKKRwvIEa0PGUIx/ukS3gMQgxzIY6wYPfeIJ2t2Y/GwugZEuB/gIhU8DhWZ64GOXCZs95PUFG+jRYmJNn63oKswCsV+Gj/+1JdvqhIFNsBi/NClr+30X6J1qdE+286YiZPevdeS+EnYh4/QHuxcGEWIQl9qD4eeSxSWr6Lc2oVcsk2iYmmZDhF6uousgAAIABJREFU/J3rNp6BLCN1lLzF2RMr2rz8Clk8H/HWOFp/FLlcf0BZEP9Z594iLJ4xAtsJCTcf43mD7J2pMmSDaDyiOENxzr+yPVF4jEJ7YnH202hacDSy83yRn9AZz/OGjAlRZcl5bhKivz1QMuimyKtWp0Zts2CRwWMu2nP3Ae949+4grXR0Qxbpd/Hc7oaXwfwDu18kWxWe1VZIjriaasH91+gwn2MQL3Pj7XtQTX+2tTF/H1m1C6/ZdVQnNW9uY52D+Pobzr3HiBvzzkR8fB5eKSsk6D5NJAZ8aX2WegPN6ow22Rykvd+BBLWNkDsoWXKpmDR34pDW/MvE8x9RFmU+H8umR0R5txb0v0Dkqbb5oscxFotu/zu6CI+YS2FlrcoyXxNl1X7j6idRe9SuL0dZfH9T5/qRVe26Y0XJBe5Z2UdTXU2gPbKsvI2Ixm6I2H+PshZrc2IDT0MCVhMrovfcUBvvZKTdF3FQlRveeUc7pIG/jcJJZuLUJIzA9EHWl7eRG/0OZ+1eJ1LA33vHaCTIvoGY/8NUCCoObGEFXoRcUGcCP2jG/LoxvjuROB4wsR+2Qoz3ibqwgXcFE5KwGGYkDF5OmaV9uK3z8v7zNdq6Fe3hj5GFpTWROHAPbgiOBdf6HM2MRaVsHrfvFyBm9AiqSNDkdKrIOzKktHyKYvYOQYz7PiqqXPjrZN+vwCuLtqT3hAO7MY3HOk6si5u2LrtRuhonkjjC1oMdh4SxXyJ6fxuwoIV7YsdluJ+KuPvTqHk0cuJdO1PBnwrcQBbky6zNIqM+WdB+cWCRwrKDwe3lXN+h7joZbryEEgNPQ0Lwy3VgDb5QUPuj8J2rE8/ubTi0OlKKU56rOvRnmu3DexCPSIYpGcxmiGadSClYT4yN2WjEDxBvmWD7YWfvmZ51+7ykPsukkZoI4BLFlZCA+C6KDzrWf6bOu1DA/OQQrBHDtxABn4KYTyekSe6HtLQuLVkMZJH5A3BAc/ptz3a3fhWleZL19BCzvAYR1yqLqm+l7UxpRepo7a5X1W7onUgoeIMykLpKuFsfaYKPIy30TsoSNJWlJmhUPE4jEfQdgD0RWYJfQBbEqrhEP6N2FFaYnjLbu4rpj0KW+0IJ2R34YTPXaxWULDOI0lIaC48o9kAbxBS2svnelzKZoVluGNsrW1e0GxOcniKSgFVzT1QlJG1pbeyKsnsfRczoBGoI/d67eiPLZa1zwSP9aYcUh90Sz9xMGft1nT2/N2JGL5Bw5Uf6cSjyGMy0PdHs0lqGL5XVABZnT1gbPm53Qgx18wrYUIx4e+TaDB65GVsn5JY9DHmN+qdgY3iOchaCRyPXeEflforArY8UvKDnL9Vf+94WuYh3bwb8aKTgRU84W9KwyEDVzfn9NIlT9NzriC/2QkaTa5FXL3g8co1+tKM8gjO0j1+ntJCfgKzmp1EqWrX2oYNTqyNP7STKZLDm9rk14q3BNUZW39Oc33vSWEd8OglL9NL6LNPGaixEqwChqjqKLRUsHCxu7NzfElmNXqQxLmkY8OpijmmzFsBkyCJ8mjsvqTlzfhdlP1pUTQAxlBOr2k30ux81SnIFNsZAEqdP1MWbZj6/HNIod6QshZKs8VoH3+qM1651Qtpr8Hxvv13bF7WSXBzYZgmMdcZHopB0VVvUsKhUvCuakETJhHanVALfQNmjMzEm1pz1QgprYUVPJei0Cn2338Ei1M798cizMRfF+q3k3LuXSAH+QLu+cOgW5V7i1g3Dx+i4m4ETfr9r064A/aui9e5+qp00VtV+Ci9Dc+Xda9Z+8mArjwoNrY2NvRvVcfPBajahdavZj9qwEXo4EssUXxofwqW86gjqKwJnedc2RKEo0cNdlnC//VrABW4Gk5EMJ49DltFWSEHpjpTgEcizE431X6rj+TYarTHJbdz/kWdGEbCKOYsxNMZIAsRsMI0xPjcQKT0UgW8QjEiEF9RhEM1hIoZMRbupZIBvSr/E3t+cdv3ncU4NqQnb2pm3aBmfqj6RiFf1caqZ81oVpL4FzdQga679cJysfTzh0ohF9Czl0J5x5jkaX2f3C40+dIRcNK4QWW43RlbYUPu7kTinG7lQBybwsk5CUitkOb4JxUL+0K6vFKMDLVmfZsxXpacBeWAmo9CimchVtyISGH5ConpD4F11GH1vfx69Pbx2jf0WzTpO7QlkZeph+NwEFww2dipTVeJX1JtUYx9vnuhzPx93cOgnCYubPecn2blCcWWcbgw/qagdHIKjkd6m6MdowrV/C9jUOnTz58LabVUFW3Mcbd2+NGO+kt4zAgKdAxuN8Xfn1r63t3Vvgyyc/45V4AnA9aexpGXRXjHGfahWGvwSl26/izj+1KE2fv3xo1DI5EJKy/0yc9fn+bcskCIGsiciyn3xBBMUjxFECBSj+duC2NBIWLuTPgf5ABSfcQ5yE61N6bpeDQXcp4rCH4sSNGbjaebW9jwfyUPI4C+29SNYT8829rrIXbMtXkkXZOaPHQ/YGbkzg65L5LLaK9HXLig7NCjUWdunRmBboTjXWNvrEjhFpuZ8rUWkFAhigF1R0keDu8r+H0naAtXiIHWUVdo9tpmRZhoU4FGYykso1rOw0BWnqPTCFKUI7GQURzkbL+YRue1Thf/PNLx/AMUwf1NcHBHPaMwqcqXdj1zOqxu+FJbntqk+2zM/wdyt9rs9ZXJT19hc2zxfRONxvxuj5IXJqTadvvWgaRkeN9N02wjs4szXILwTwQxfrkTK8PPA+QXee89tjNyfsYMjjsZLYnPuPYfiqLentFi1q9Nne6Y4aOR5PIEKJVim9sRlKGbtLuvDfpShAb0rYG+ihYlfLN4+nktEeEMKVDQmEYUQvIMs9X28e/0q+tybOO1I7mN7Zrzh7mV4WeUoMfKBBOyPDb+ahAQZjqRgLyRulRtYATsc8bYCL12avQLpU+FSXp0BpI+w3hd5Se+haRLl8hVr3A7t22AyHCqbVJSR8nnXwzSWh2xD48lu40kr8YehcnPX4iUh23wF5xrJEZug2O3tcI4JRfzlM2rEFi+tz7fSqA2+Czr+7BhU9PcPhhjuEZLBs+tRQPtjKMZiLk6tUIfYxArKT0AMaz0US/hze8d+zjNBWLu2IyKMo5FQOgUJqLtRWiGDp2YYku1D40lOrsY9mvKsdB/2e8iNNx8J2w+ieJXBdn91IqfXoGoCv0Nxsvtam267G8fatWtnYMeJIXfzQGS1KiwTy1OWmwn1+wFU/PoqSqtSL2c+YyVAFme+rkaC2c22xufQ9Cz4oEWZxQxSR0xolHetYKBdCQho8E3tvmsQIzkR4fh5SJAYnGoX7YWXUDH0CzGBJvKsL+QcinC6OxJqr0BK1dSq8aLauY/a9yNR7Ob9OPUPK+ZqH+CpAieQpfAJlOSzY0XbD1MeMdvR8HhlKiwLDvylKHb5LmvPP+1oU8LWohbPl117AJji/O6BmNsAw/mRlPvEH/MDlDH1vdE59OMoaxWuQsCyanhXJMXNQDFi30FhDb7AFOpzcVLOACSMP0/8vHe/z4di53Jbm8ehuPFpBKyI3u/9aWHiF4uxj9F+Ks6Z72hrMgOLFazYT5MRT2uD6NDBWI3LAI757e6JHWBha9km0U7IMjsJuV53RnTgK2SMcJNXu0Ta3gdVObgS8ZkQHsVgvymviBTKFZHQNKYG7B5oz79n89QB7YdCKR1D5GAJFPp0KTJ6BK3/iXb3RbR1DZSz8jWiX5VrbNeC9APzahEJDUHCt1vy8UC0H+6xOWtNOkxoEtp/WyA6uyEyrG3tPBMb82kotvRZG/NcVMGlqMZwKokToZb2Z5k25k3MCcC1zu/Tkfb9E6zGZQLWPUbrOlSfbCfnfspNNhs7QcTpx1xDxLOqFsEQeJJ9vxCrn4k28E1ESiYZkv4aMdHCquIe2Zd0PaMYs+JEpOK0o1l4TD/S7quIeW6FCHSt+CPnHT9wEPZMRDxuRUw4VUYjQxa/ImHgEcQY7kMMJVVeqsXzhZjHu5ilAQl1n6Ei63VOdGlxkDpigkWlgR6IoS5Agvmq/hgD8CNsfre3cdyPlImrSJfSeRMj/oiRvQjsY79741RDCMBOB07xrm2FGNkjpOM3v4/VU0XM9zZE4HeydU4mFCFhqjjM4Qwk3G2JLH1PEAkzQNaat53f8xGhfQftw6hb0p5fHWX+roo8E3dj8ad2Pxrft5jzNYLGUjbHI/rzluFaKnylD05sO6os8IC1eRulRT0WszgJ0Y19kFC4APgPFAZSFc7xCo7r2+b4bPs+gEScLtq7Z3vX1kXM9DnSB0vcTAsTv1i8ffwecIQDey/iHZ/b/5Qr9A0sfhDRuwVIKH0CCRKpI4Y/QIL/dUhZ2JOS9vak+pCGlyhLQbVGQsd1SLmtKo/3PGVc+5025qI8WBVffAtTstDefcDW9y2qa4cuoqRdD9ia34SU0yqX+Y9R/d1Cid2Q0lPRE88T4cEuwE5Cst9nWrtvU32Ax+LQj0spK8pMQvLMIchYNj+F0wbzGmVpyvuwHBjgNzjHnkZgf0mZPT8cKS5nItlnmbrng/371hqWZfRiTLBAAspkVELkqtjkIM3rZu/aIUirHFWj3X0Neb+DwgQWoESi7rYRUsykK3Cu8/sxSithRxI1/JAQ+QoqDbMnpRB8KiIc1xGx7CCL5EM0lpRZhNzRL2OCRwT2YMqi8F0QQf4CWbPa4sQWReC3RYR8AhIe3kOC1iDkxr8Mc4UFYPfDztFGcbpfG9wW1v+TE+u8OPO1F421Rje0tsYgDTyaYIPcHed612oHqSOGdYF9P9n6eSBiYi+QZr6FMLItpeVvEdobJxM5VQrFSd5Mo+t6AqVgPBsnaS8AvzFi3vvRtEzSPaRrlo5GSmRRxmuUc28+cFjFfPU1HJ6FGOAQDz4WktERWSb2Q+VgiqNsexmuVClqF+MUFUfZuA87v68hXr9vQ8QUWjJfM5ES28/gn7G210fMJegVcuDnGT7sSLm3uiLB43yqk0APxZgtYuS3YEe0JmB6ImWhF04cozPnV5EoFYUUq/cQ0/SPY7yXdP3P8Yj2NSvxi8VINkHKzp1I6J2O9nQhmPVGynjMld+dUoBuZes12Ll/J/Hi7D2QMNcD7enjrK3ZhuNP4Zx6FoDPkGDhlkp6lvLo2/OI0HvEA92a1D2R1e5mKk7qQV6J5xBt3h+VICtCdoo57xWB3YpGa+E/EZ0eZ2M/j0jeg63TmciiWtQeX2j4OBIpacGz4w3+WPu0pkzq6YPiii9Flt4Yf7qEFtAPytrSd9jY3qcxXOk6EicmIj56sfP7m4N5bM1uxaoZBGA7ICVhgnd9DcRbZ7AYiXZL4vPtNSwh625ERK9ABH4lu/cMEWsQIr5FnEmh6Xa2jfihP9kB+NaIoN9oCOgu7kckjvoMvKeXd+0D0oXOhxdIbuMfjxWyBX5X0d5RyFp2J9Lirrfr47CTOyJwG+Edf4oEwnnUrASAtKg7keBxjjeehQm4fpSEfBTOsYBI8bivot0RLZkvFFz/KrJcj0Hus+I4y1upYSW1Z9tQutKrgtQLRn00im/+GbJsDnSeuRsYG4Hv4HzvjQjTbVidQsPxlFVmCA4BRJUEHkEx0W9SzVS2Q4yvEHZGI2HxdyRimZz53hQpew+gkJp+yCJdmZiDGMs0RFyfdeA/T8EbDl1pfb6QMg58N9IxehlSWopyPa2RcvYEss6OJxHv5s3XqXXny9rdxXD4AsOR7Zz7F+Hsr8g7VkRxwjMRDetj18dTsZ/suTUQUzoEszCj8KWot8JZo8Ky0sbmbD5ymb5SA79GYKfOIFpWlC/7fQq/KBO/JtHMxC8HH9pSJpscTMU+ts8IVELsKpzjRq3dfyNRKJ0yTr0NjXuyG+IR0blC+9ZNHOuPFJe7bK6S+8nw9zdoz1+H1SpFNGURgTALShq3vNf/fojuRw8foXRPb4CEu7txanYiQe+HRKzCtr5r2/chOOFMNvZniCQ0Wtv9vPkahPbkq6gubwo/1rXnnkU89Qa73hPxmFQB/oksHv1YHrnZL6eR9i+kLH0WEsJbUYYEdsfxUBp+vV+Bm7si48MsD3Y4zajFu7Q+327jEjKmIuawjl1bA/ikAi6WEHIMaUtQF8oTEAYY4hVu8JlYqQPSR1+u7F0rMuxOxk5zicEH3lckFLyGuQBpWnqiLaWwszVluYaCOdxG6TKtypBtbQjdARGPr0ifS97R2ywb02ihmEekGkFkM7kJBdc7Y07GODZzvopapmsi9979OCEgiDhuHnn/hog4rRF5dzRIPfCuvXCOfDNi8SHhuKzNEPF0j19cFVmAdk+1h1xHLhH/JisfWcO/pnSJ+2u0IXLTFS7BcbYPrkDC7HOUgnwsucYPqD8EEeZFlAcOhOKv1kCu0CLuahgi8hciofYu4OAEfHEc4gHIU/EB2v8nIhdWMObMrrUv9pZ3fSdk4XyNwAk4Tp8LPNwEWQ5nWx+i8+W10wspeSfhMFukIG8R2RMdnXXdCQlKxSlSs20dm4wZMej1vXftjYSWQ1I4bM+uRFwYmWz4dWUER4ZhR78WexMJVzNsrh7HLOgB2OVpGjfdC1mm5pJO/Bpo446FT6WSTbp5vwfRmNdwFhZqFunzSsQTzs7FPDcB2K7e73Y0xo9eDdyV2A+tvP/7I4Wt8EBegB35G2i74GErRO5fDewbGVPRXtHOmjQedToNcyOH+l2Be2dQFtMPxs/iJLl5eH9NbLyBdrZDVt6CLlyYgqWscuDHrO5Imn5kaE+4Ryt/Y4W1NYsmFDnPxU5hPDM21zR6z9ZElv+b7HMC4pXntWSdluRn2TcoreQgFDP6fRpLH7RF2nPwhBFEwM9FDGsBXpC6beJYMsDpiJAV2oGb4JKhuKrVQ7B27WxD1DeRhuy6RfojQrVmpO21URzV1AASd0euzhjxPAMR7wtoKiANRUwpWBEAOAIJWDMi99cnXVrrSmS1ut6Z18KtvAoSAmJtH25tN7H4IMb0UQXsc6F+p+YLWRFfwTlb3tuIO+C4pQLw8w0/prm4R2kFaFKrznlmLDoruklAvN0/H5gbweubkVvyGpyQDzzLduS996G4oEMD95ZDQvA3R3lGxnumd30w8kT0jOEH2oPH2fc+KOZ1HLK+L4dTNSMC/yjOccCIIXR38Kuz31+73omSmB7rPLsjitM9knRGblcUttE/MB/tkcISPF0p0OeB9r7CYtErNebA+75hokhIfDryXBcUGuPicgekUExEQt64xDrt670ro2YJHpomYHWlUYh2jzn05/M+GuPqunhjiB48YvvhQxRiUNDlTsjFuztyyXaOtFvAHuzAFvj8TcJJpN1HUE5Ck6ojyIhxA/H9FGu3BwoF+j6R4vBOu01wF/HFXYkkjtq1M1D4xU145ajQvjo/sU5nIBo0l0ABeBt3MGbfafcGvKRaZJV9gsjJgYheXEngWHBk+fwRpYetuX2emhjvUciSOxtTAJ17PZHRJ7ZOnZHA2SRUBO3Jd4nTj+ORgeQ/UHjfat79CZSxoSFBuOj3ZXiVaRA/vovIiXIoBOFQJIxmiNaNRXRnvuFXi2qYL8nPsm9QJvgZiIE8hOJNbqaMOekR2XADUBzSBojZPQz8DwrijgYuO7A/RFad1QyJd2hGn/tb2wVBuAH4EgmDRb2umNayPVaGBlluvkLCq3sCRcHQfEFlZ4PdAFkT9jLYvSgz+vtEYMcjd8S2NudjkSYUzGgP9HsXm9thyApbuEVPxs4TJ65RR9tGTHgVyiD2Zvc7MV9noIz6T5C22M19DrklgqcU2b03DD+etj6sH3o2Ml+3IiF6Fp4FH1lH96BM7nK15pEo8L8TcjX9HCcmmAjDdvr8KrKcF6eaFSWiCsFu5ZrjfY3GY2BT7aaSa26kZBAxYWMFnFOqKE/s+pXNY8rldJq1NQFZbjawa1sEng3RkXMorR/dEIE+gnIf70/g5J2KPt9IdRH8KciC6wpkrjV7s2L+A2t1OnCjfe+BFLrJVNO9UALVPKRUn4pjdW0GfJGANQMJSgX98RngSOBdb97vQvHrc6io4Wvr+6nh1gvW9pM0Kpupsmou7HEGmwzVQQaNi5BQ+T5y7R+IkwhKaZEL4Vas3b3sfsz17Lf7G2T5v42SBqXCdUYi/rYhElYuRorZdMrTzWL8yYe9EO2H6YgHNKfdiwLtxoT3Iq74CESvt7LPUOTJ60fkFKuqPtszsQoQa1i72xpOvY+Xi+C3591zeczplIaawsp8IGH6kSFDVnE2/Lk0hoJUlYn0+/0eXuw1pdfIn+sVkLxyNdqH30VGh0P9Z7/tz7JtTK6S14vFst8XGyJFXe32/KE0Hm21EbKYHo7jGo3Ang1c4vzeG6e+mG2kVFH5yTgxaYiJn4di5m4iXbP0URqtqXORIPBjfyMEYG+ldC2dh4jc1Xb9ptims+fvxVzVNscvIO3qRVT6pSr54WrM6mbz9wxi/OcgjTgVX5Nq+6qKDV/V75hVNUNCVQckbN2MVzaoYrxzgJOd36cjQlfnHOERiMkPsPVtqKtZ4HsEdqbX7p7I7RO0tHqwdxSwiGDdChy1tMdrz4eSa7ohS34yuQZZUO9HyQPbAc/Y9b4oZi1aQ5RGov4iUkynI8H85Br9fpLSCjHH+nEJEnKD9Tvt2T6RPi+HBK1UnzMkiC1CisM5OHGENu5Uot1HlNnP59uYr8E5mjSBWwsJJ1Ddg1nxFgM+moCFGPcHKLb2YKT0DENu7fuJxFJ77/gucjv3QeEh/4ti7jai4iCAAOzfbJ2TsEgxPBZZwkYjxv81UjTrKKahdi+1dpOnCibarQoPmofVgUaevg/R3pyKhNqUgpeCvZWEorWY7d7swE5F+/dhRO/PT425Rp9T7c7FqZlt6+XmkRxKopwZLeQx1s9Hnd/d0b7ewH5fQSTZrRn9jh4egow6jyKl+Hxkof0cGemSh44sy08rlu3fz4EvsixbJ8/zrxEhXgktxpgsy7ZNwD4F9MuybLssy7oi83XHPM+vA4ZUwH6K3McA5Hl+F/CnLMu2yrJsa+Qm/SIB/zjwzyzL9smybF3EUP6a5/ntQI6SKJr8ZVnWBrkd/uFc7oY06bOQQE6WZVmk3beArbMsOwYJ0YfneT4Vxen1QYwh1G4rxLiHZVk2EZnjj8jzfC+k9Q5EzCH19yZwQJZls5DV6Jg8z8/L83w6stJsE2k7Q1pnrO2VkAAXg32vot9DIv09APgiz/O/ohJPi4AbsyybkWVZl8z+Iu22Rus027l8J9LEX82ybKNIm8XfYcC9eZ7/e57nhyFBemKWZUOLBwzf/XZbIaHqCuf3/cgyPSvLsimxBrMsawf8ESkO5Hn+ASLGh2VZNi/Lsj4J2NZIaG7peEFKygBk9f5dlmV98jz/E7LoD8vz/O8xQNtr85A1fBgKvyHP898iF992iTG/AGyWZdmW6GCEiXmen4MsJKOzLOsXgS3W/k5glP1eESkAZ6G5G59lWc8QnuR5/iXyjEzw+vwFog/BPtvfWLTO2yCGsCKwMMuy2VmWdUJCx6aRfm8M/BlY3fBhAmJuJyNc2S7Lss6R8b6NlN8TkcfgkjzPn83z/G0UXrJXrMMG/w4K6Tg+Ar9nYswLKM8QPxdlh/8kz/PfoDJyExOwxd8niOl+iayVtwB/RQr+P5sJeyvw9xSs7YuvkCB4aZ7nb1GWuNsACUzBP9u7oES1S7x2/9fabUIDvHZz4LJAu48k2m2N1uIGu7Ql8P08z2eh+QfR0Rjs5wnYjDRvS8Gm2u2EPGVX2aUdUWztLmgfj0KW0li7v6zoc6zdNmhN3rXfbZGQNjrLsl5Zlu2GDAl/CcFTzWNaOXjg/70NzC74UJ7nf0Q0cLssy1ZDBogfJvr964p+fyfP868ibYMUuWdROMGpiA9/iOSKLRJwy/ZvWUvAyMr2ESJK92AuGCS1fz8CU7i1DkNC7QJExIvirwsp677F3DgNxe5RPOFliJlMdNuJwG+DAunvQNakzk7b42Nto832DmL+12CZbIiZLyJSDsOe6Ytca2chjfNAp91v4msisKuhLMtZNsZtKd0KP6Kivps9VzC+i60fhav8YxJ1C5FL/npkeW5W20hYnYusGbVh0QYb5F0biTZ9tCxW4D2+u+Mk60/QUoAs/VMoA+JboVjCK5FSkNSgiR9vu7vhR511cg8OWAER+hOW0nhXo6zzVyTXFOEC0eQab33bI8XsXJujjxDj2xWVgIoeW2f76efIEnuTc30YTl3SxHhHISHtVeRabe3Av1YB2wZZGj5AAs/EOn125s3NBl4b0bxfAl9WtLuljXchjaVmhuGETkRgO9mcnUTjGfcvEEmg8uD72hhrJ2B5ba+HmJ479oUxWEQX3bCG7ZE36DeUe6x9BHYDHOuYwc5DzDwJ619DCtelKCSjwJFgfkLgPbX7HIA9B/GlXza3XXvGd28vIpLE6cO0BLYl7UJ5fCyyNhZWwoI3v4aFhS2l8fqnK16CFK5nSNOelfCOIaWFPAbRkgHIyvokTqmwBFxDsXvEl+v0u1jfnZG3cUvEw3sjGhaVQZb159tpVIRqe8qYhy6IKQ2vCe8mQm2BE6sUWvzI9Z7IbV7FhDZArvmi6HQv59444Ac1+rsmEio3p3Q3fo/yhBufsHbHy+RDzOdKRLAeoowp8zdkX6ycE2Um/oZIcDgIuZFuCcHateGGtG65oj5IuLwFud6uj7TdGll/VnQ2Te227bqb6LERdpBBHdjAu1ohQedPOAcnJJ53z3sumMFQKo6+TLzvYOQaaVIVAsXvnUwZ1xMiJE3iIu16O2SRc+OQ3XilHbCM3ABsF8O9LNLmUCKub5Sg8RCNJ6O1RkLeRGRJH5eYj85IeHNLHfVBys5cZMlrctpRAk/ORRa805HL/bS6+IG8DJ8hunM0Uo5P9+FRnOT2KAZ4f8Pt9ihU6DokqNWUAtm5AAAgAElEQVTqc2TPLKB0P1a5oUfRSPtuIFDlwvBjVxvj6GK+nPv7EEmgsvtFPHN/55pbbWOvGLzh1wziitau2Ik+Efy6DWW2t3bGcj4wq1iPBG69h1VmcN53AxbSlYBtj2j8VMqYzeWRNb2oKRzLnG+Lwhh2pLTuZUhYqGq3NaKPYwx+MOJL853x1qoLSSBRi+qs7f40raNbmfFtba1NvOb2AQnYJnSHRkVgZywcJgDbEdGZzIer0ecuKFm1QwBuZaQ8RPdEZA7aUIPHGF48Q6DUGAr3+QvpsL++KGzGTwpcGSlbzen3YagcX2V407fxWfYNRhgvAWZgvzehadmSb4QGu79JCLZGX46lPHUpJJwdghJVrkNCVTukERXa3UjKzHq/39fiZKB693ogghU79vJaGjNji7pwW6FYmT0ok5p82Hk4RcEptap9UMzc4UQy8Zy2P7PNW5S3yZAwvReyFMcSGS5Cbqb3kWWzVd22kfBWlNC5hjJmrjh277AE7EQkUJyNCPtwGq2Gg4lni4eSTVpF8MFvt63Nx142Xl+Y7U68XMppyOW6h41vfZQQsGsNvD0NWVCutHkp8HFojT6fQZnU0xZpyasQsLYHYM/FDqVA2v0EJEBET+kJtP1zZF2cTqNA3aai7WB9YFv7+5EVLpXVPxYJNqcga2UPG/sEw9tgbCHaB08iGnAWstpfZjgVZaxeuxcTKHZt+PEo8bjo/ZDAeiFNj/cchixJTWANPx5FzO56FOqyBmVpva0pExNDeL4fclu/h5TQIkmt2JM7xOBpPGa4M7LYb0F5gMgGxGnmDOBW+97b5m46jUcixhTZ01Giy1tIuenjwyRgT7M1vh3RnxWt7d0paWAM9lSkDJ2D9uMbhmcrU9KDVLvFMY4XINp9ElI2O6Vg7d5mtpauslG02QHRhu0icz0Vefv+gpRmV9HpgISoGGzsOOieSLj/bgL2LGt3QxwvDKK5fW3+J0RgT8U5gIGyZGMrJKym2j2TsqxjOyTgDadMpr6cssReaE9MRqETwfhWIjyGiiRdpPjsZN9D9OdYpFT9wOb8dBRj3Brx5WS/A3jR09Y2eiLct/n51jvgTFhMi/wcxYA+ildqgYos0SXQpzcpi/Y+Yxv3GsQkNk/AtTPk+QvwB4PthywxKyPNKlZLdQBygbqhBWch5hI90cSeXQ74mfP7YETkXsdLskm0/TYKdH4QxUzVLQ9TVDIYhCzH1yNG/DrOkYMR2OUNdgIiVM8goeVG5NpNJSFMRIxzPBIsbrZ2k+egF9dIJ5v0x3PRePDfR7Fe9xnRaFJ7Lta+9bmwlr9m7/keIl7Rkzrs+TUMv45ABLI4U/niCrjM1qMoaXISYsQPotCO5KEQiFnsb9/vQtbdY5AL+8oabScTAogz7u1QuMbV1EiGCcDvZPO6o83bX1HCzuAasD+jrCbRB3k7TkSKW9URf9shAeVIJHS4p4cVSkvhxfDxo+jztkgo/ACvgHwI1ub5XcpzwF8w3Lra5rzSFYoEg/Os7VMRTXifGoWzSR8zXFUVYD6mkBnMrfaOl0gcP+ngVhck5N9IxQlhHvyPKIWqZ5GQOA8JtlWJpz+mDKNaDzsdyD5V9S9DxzhOQ0J90jJqY/6FtXcJsjy7YREpi1sRq786Emwuo6bFzMGv2HHQ0URbe/4EdHhEUU5wbWSoKELuYkcFF0dgF8r3JESDnscSfyvafZfSyzkd8fB7EN2r2sc7oNj4bQ0Xt0a8dUO/jwHYVJLu3BCMB/8RpQHodLQPb6CG4eL/4ufb70A6fmlzFG/UBwlzrxgCTrb7pxJxa9Zot8o9tjXlebPtbPOvgwSvE7AamTGigzSmI5CVYBrSJr+mwpqEhKJF9n0sEpaOtnc9TUJoQEkejyCBd3PbhKORdvc21QxhJmWx6R7IcnclJXNLaWCHU1rPiuNGNzLC8S5mXYnAHgPMdn7vjKwHp+CdGR6AvRglexW/l0cJby8jq0OVheE+W99NUdzXvyHBq5Nt/KAgjwjkR5RhJ0UmcRFn2ze2VkjoLooQL2dr3Ma5dyvxo/YKReUs6+cqyAL9R2Qp/CbWNwA7kfJovpVRbHN/w+vbkIW5ar6uQ/vxOhpPpLmX9LniB1Jm47e33/+JrGLR43rt+YU21iNQyMDNOO4xG3+q3/fQWErrQiR4/IzqUkDn49TStGsDkWWioTZoAPZZSutFD6RUFyVhOpA+RWaB2ze0D6fa9y5EDrRAgklxOs8KNsdd7PuZ1u8mbsvAe/a0/nZFyvQXiBm+RESQp9z3uxA/ZrjJcYwOTu+ECXJIwSr2xEgkrMZOwNofeMzZl9+xvs6jIj7O5uUhpIQPQVnxHRD9PsXwvFNovpDwezuNFtxXkdLyEErIjLUbO8ZxTZv3s0lb/Ccgmr4ZKi91lX32s/tXET+idD/gIef3cOTGLUrEzcJqikZgU8dBn5TCLWT5vgzFMRa1qv+JwpdSe+lAa2switF/E/G2vZBCkRRK0T7eyfDxDbQn1rX1vQjnAJrIPt7Hvl+GhMp5ttbXxNbJcHEa8ixNRPSm8Db0tHWOVrpA3oTnaDSSPITw/S08z3GdT2pt/hU+334H0sjbAwlWhQbaFcWnPIkEhz/TzJIF1I8/7Elpzh9OY4Hn4UjgSZW2aYWIfxGDdJ4h8H8TOeLPEHiqIfmpttHc4zYvw4qRJ9o9HZWA2J1Gt//RRV8S/Z1Go6trRcQI7sFCBBLwg5BQcgqykkx37h2PnRYUgV0LKRtF7OtcJJD2QJp0MFbJnp2ABN7v0Jh0McD6E7Vw2nNDPLhaySbI2vakd+0bJoSE2S0T8EWN1O40FsMfBrxRE0evtTU6CAkKO2ChLwmYgxBB/srFB8QIg7F9zjNFUfmPsGx3B/b9CtjBPg5RIyEACbtHIIGhi7V1tK3tXGRpuC0BnyGL08mUloYHUILCpsC5xXMR+G2QcHMfTiku5PZ/OdHuACRUtXOuTcHOIkfK0uwIbHdk9RpIKaxtBDxo38/COZ4x1Lb9749zbrv1eVENvCraPAGFCe1JmYyZPN0J0Z17aOYxw/bMSsg78mP7X1jhGuqhBuDaB3BrALKU1ikFdgSy2s0H7nSuD0mtsT1zIOIFLyJFqThZaDSROG4Hdje0F5t9jKPthbUMvzsbfhxpePUQ8F8J2I3wEnERzTvGcORHCdjFOg7antuVMqzjaMTLn8FJUgzALIcsm4tQ1Rr3FLzNsZyKBPwOSLC8DDugxK4PJR1P3Ru43Pn9NGW94KJ8XSq5eBgtTPC1Z2Yh+nUIkguKw1UOAKZVzfX/tc/Sb6CpNlwsRvLs4xrvXUT8SMSeyOrVJPgZMZhmaxbe+8+jDDqvcsucZxvwR5SCdbBYuQMzyjbqVTQKS49TxtfErAxrI+HuCyS490EM7knsFBAfNrVmSAi5zjZetFarPTsJWbwm4FRMsH5XHYF5ghGmd5CgVcS+LsI7lSIAe4Bt+P0RIyislP9RteET70wmmyBBaR1vfbazuVqNagEttvY3EjmO1W3bweUZSPPe1K41VJNItL+lOzc0HuVahdM7I9f/j5HV7WFUFiwJ6+BoRhkDtSs1k86c93RGluFDUFjMEPf9gefXoTwJaz6l56MHsrT0q2ivA1KqHjZ8nI5c0efExmwwG7l7Blnv30ZWuZdi/bZ56eOsZSv7PI32VhS2YhzRo34jz6+N4v3+h8QRwxHYTTz8qt02snj/CBXz/h4KKTm5JmwbSh4z2vbGLjX625Yy6/lSw8s7MMW6Aq9XRoLpWErX/20ObMrosjoyBNzEYh7jSGnN/oDy2N6kJ9DBt+LAmLuBo5uBI24yXfQ46ADcEciS/grloQNrVMGiEK5DArhV2S4SXGehOPYrkfJ0G5ZsFBqvzU1RVaQ9TU9X+oC4x6A9kkUKfGxWgq/dWxEpC5fbXijCh26jPNHSpx+pMLMxVNQg/zY/y66hcPLNoyTKByXe1RExxOUi736GeMbbDCyouiV9R8LiT4i4j5znCiQch4jrnaG+VsxRrwJZUQhBssSLB7sr0r6fRcLa7c1ZH7tWtL0czcg0N6LxAHIX3gQ81wzY1ZzvWwFv1YDpiRjCHMRMHjcciJ6hTNMzq901TiabRGDaICXoBsOPw0Nt47nCaExIWB0pDrEkl3Z++0ggvLnOmiTuDUFlQGLtpkIPdkWMPEVUu1fg+WASbq/Ee0+lrFQRE94LHF4ZKS6jKBWW0yjdj6ETw+5BITTFXl4eCeOXIutqJbMP9OcMFFd+U2x8iL41YSrI7fd3ytAYfx6LPk/3+2a49X4Kpw3+bhceCdWVe99pe0ag7WGo5mEMv4KwyA18DLIYxta3SZ+9+31i60Qg4QzRutORZyZ6wIPT52mB8a6MhMrYeCch+nI+2sPdbJ6nUOMYRyTsnwUcFNprwG+JH0U9xubqAO96W+T1+H2i3zsg+naKdz1D3o7UcdArIS9WkQw3HAlzH9TArTGh8Vq7QxF/jbW7CQp/ONB+r4IE4ROQHHBwaq4r+nU8kTPvEW25Ahk2xjnXp1AvuXgTW6fdivVxxrwqEqpja/wYjZVMCkPAIKSUNzlm+l/ls3RfrlIa69FIZIrJaWeIcm4z31nAF2cF+4iwAY3H3e2CkhCmA6Ps2v449fxaMK7RobZTz1MeOVdLY/V/28bZrapd//1GKHpTMpgWxZBQIysvsBZbIytez9D9ijEX5TS+E3n3injhGpQhHiMND7pE3l3EQu5ZEAXnXrHxg8ew2bXl/L47eHkMinUKZT+PRwk1jxGpT0r8fO6NEZFryE61e33cvgfemRpvO6TJj4jMc62Eotja1pnrluKl4Vf0mFIiBNvudUXCRJNKF4jxvobi055FTGWJJE+i2OKPiFiDkDB0vbV/ORIw3L49QEnHYn1+BllTC1wuLKyDY2sVgS/2fMcYXAK2ndtHykQ6H79CsNF1qwHbNjA3ofH6CWeuG7cIpwnSzAB+PI233ymtaiHYV5Bw9jJSuiuPgfX6/RqiMw8hJWkcjQlN/ULjroJFvHrfxDq9amN+HoXO7EZpsVuFSLUbtE8XISPUVYhetkWejiJTPqaQun1+GIWxuH0eSfyIURf2EaQ4b1AXvyrWYUVkZQ3RjwEobG0zVOngUsS/r6deUmHR72OR4jIAKWcFP2tDSff8MX8XeRbewCmBZvc6Ic/Y7UW//9U+S+/FCm7+NQpSPxZpOU0QgYrYxBa0ewilK39/xExmoozoYMzW/48fWmC9SbyrWcKCv0mWwthWQFam85FloXtz2kRhAc8ZXpyPAvVHOPdTRf+3QtaT72GCQeCZQrjzlYMHkYVsX8SMFgFTnPsnJwjze8jCNRtp92OxGoKIwKcOWKgabzSsgeqEor1Tc1+j7Wj81RLAkd+jhIIBzvVKawgSMPZ0ft9Byai7UrNecuC9hXC2sfvbuV9UqtiY8mCL7b1ngkeMBvp8O2XyZ8eqPkfGPMW+d0rB12h7RDPbLWC7LGa7KdhQwllRfqcdiZCyinY7YVUGEu26CTILUZjOq8hlXnWm+dPYUdQou/1JFKv838jwktqLKdjCxR9TOp5x+n2tjWM+SmKbWdHnp5BC2hqFft1uuP0YFaEgFX2uyqfwYZ9C1vT/ZjHrcCIFLyi7IJpTyCDbo/C5A5FV9HkqqntE+j3f+p088ARZuYci/vAEgXAVpDzUOiJ6WX+W5tGhayPz8LtIq9kDODbLstFZlm2aZdnjAHme/3vdF6aOf3T+HgHWybJsO+RWOzHP88LqsEKWZdHj8v5V/xLHkUX/8uqj9ZrzrryZz38NLet38VexzpORoPM3FBN0HrBFlmW9DPayLMuWi7y3KyKEx1CGFawDHJxl2a5Zlj2ILPexvxmoYgLA0VmWnZll2SbO+8fmef4xNK6BHdN3N0pWuB1pwXOB/bIseyHLsreQgPsPf+xZlvVGQt1E6/tKSPE6KMuyNZEb7YIWjvchG1MItg0qrTIfMZFpaN73z7JsbpZlLyJLb+xIxDpzfU4Idgn8HYxikvsDr2RZdm2WZavlef4P69vJWZb1CPS5D3YanIO/dyCLFkixPaYlHSr2UZ7nr7m/nb8jUHjLa3me/xgxl9Odvs1C1uw6fZ5PeZzptFSfE2MujmM+PQZfo+0zUDx8c9otYE9dzHZjsAOQW/sRgDzP/4Dc7wXsOchF3JJ2T0dVPkKwvZGL+k67tDpyfx+EYoN7I7dq8C/LssEoWeluuzQcHTW6D1KUhyNBvEWwdgRtE1pv8/W50+8RKO5yErLcDYkdVWz0qX2e5/cYPVwbhVRdivB7nD3XhN7X6PMaoWNzE7CX5jqGeitg1Rhsnb88z78OyS42jt9RHou6AfDdPM9vynXM+btEjmOt6Pck6/eIxJgnIYXmp3mev4Qd9pFl2UFu//I8/xQdZ/uv97c0pFykPQyhDMrvi5jpeYj5/ZSyvFBMI0vVnqxTsuQelPnoHrX3NjVM1Tgn9oTG1pJ+tRSu5ny3qN0UfEvhlkW/kQuocFsOQMznBUQELqLiCElkUXWzn9dEAt5lSNgcGuofSlQ5CZULWQnFE05Hys7JyBrfrFhdZN3cGcUGxpJcWiNG5Z6WMxZZHW+3Pg9LrGWLxhvpf3MTilrcdgV+VNGASZSF4FdCyvFnSAk4lUg8NnKHjaSx2kRX5EJbH9GUZicU1RzT/jS1iD6BGNHWxE+hWaw+Lw78/1HYxUk4W5x2W5wgY/e7UO7zflgIl/3u+P/aO+9wO4ojb7+tiCREFCAECAmERJBNzhiRc44mi5yjEcFkAwIBBpGTQSIjGWxsMCaLnGwyxtnY2F7v2rvrXaddf/Yy3x+/Gqbv3O6eOedc6Uow/Tzz3HtmpqZreqqrqysiK8pSswF2IIU1ZkE81x17p3eJBNKgDeGG9v9YvMhwpKh6mUj6s956307ncem+cu7gaGByD7zzWEryDXITm0WNXK1zwzFnOxPxHoJ2EMF0TXSv2JIHJeyGFv2Yn1w/ulZ+WBMt2C9SmEPy/HyxxbPsB5Sb2HZBC+jQOu9Y+hs1V+VEFhijHHZdWsxGULffCGwfD34LEtkIyt+p1Pdsx5tS7krEKCci/83KyhUhOkBCy2NVsHT1iV4aCQxnolygMWE2JYgcRlFFpFWcbwO+MzvetwLnyoCidvtGQTCxObo99X0My8FUixjeHwN71sXd7jsa+f9Oj40NRdLu0FiuQj3/8S5ZEpDv3lVoIdu9p3GeXfDzKGxlwNlsHOtogEwL/d/dDnyHsMe3AuuPC7Ji3NsBbc6W96WDKPXUe9hcTpYq74l3xnx0vd/bIEE2WeRhbjh6/oHSWG2Edvt5vi6/PvnRFLm0ygvRRGSC/R6lVD9GCPejPJ3dqligaMnbkM/GSDs3EC1uq6HAotw5P7SYHIyclH+BTFyfJMpFUcBfRz6A3fxG0M74K5TqeKMdzbPIB2RcGc7ueYxSBSrv2p3IkXv1xFjvj3b783n4LljVr923PjL17lw6PxAJ8/cSccI2uLGB8wNRSopvEa913BHepWc5iuCDkSRy6KWeYX8PoajM1SqDvJIawmwEdj2KgJO6O+98ozaJQhNYF7Yn3jcZUNRJ3ygV07EBmCVQ1obJpAOWQgFyeYDOSOCHbdBIPpc/F+rDzk0jXL51jNFHtGJZbAxRBokf08ZCVgfn2QU/j8ImA85mY7/RAJma8MsiTXpLfKCHYO+uy0P86/bOP6IoKTsncU7CMhui1JEMdDwtboTr4o1M+2digWUB+B0pBcPOjUfPP1A+QN9CKQ+uIpAeiXAk5GCk+t+Soq753kjQzAXMFYhUv7D73zWCuQyZNE8iIuyVYAcjc8lmSHCdQUm4tPu6pYpCQt2rhufvULLiz+PlNkRBKN1qmiNN4ve83xshoXx/ikjJYwg4b1u/zyNfp3eRG8TBdN0ZnRzq14N/DQmWryANzJDSN9mPQPJdG6OXvd8Low2DHzxyFHByT+Nd8R3HUDD12RpY5fWZM6tdCUQ/z4H+kyUG57UDOf7/2OjxtNK13BXoMiJVaGo8fz3M5FiXRrxvnArOOQZpXt+ne/WdRRBPe5xIIFxF/ydiCch7EufZBT+PwwYDzmbzWEcDZGrC98Vyec5h2D4ElBI1YQcAO7U61rP7fZlLo9RTeBtPewXJPr9FQXJ5nmc/T+wcW5fafs8eHrR1Mf89ZMq8E7jRuz6GQOk4u3YmVm0BOTr/DaVMuBVpLfOqSTFT/wikQT3Q4O5BJclOQoJsyif1NLwqEUgg/D6FVuVLSFgL5bmbiZyWQULWwyitx+MUuSj7EajxiwTV8+3/g9AifB8Sbi/y7guN1x1Y1RQjvK8jzfKrFJrp/qF+PbyPKD1vcw/fUYnxegjLY4eC1e5Ctecfw6vkMjvwbo5P94EyD6yEfE2DpQiR79lcJYgjP61RyL3nbgKCJ/Lx6jjlTHM0R3PMnoPZFKXObFSQ2FqeVwgcgiybfvW/YGaOufHo6Sj7XZAgRpZlv0HC4HLOueUsIvEy5NeVBWD/jgJSQJFpX8qy7JYsyw5HprDN7bn/E+o4y7J/QXnO1sqy7BYUCfk2MtcfkKWjzv+APmr+rLuBj1C0/nooIe9fy8+wqMkhyD8VpBafnmXZhijAZhfn3OJZlv0zy7K/Bfp9BNjaObcu8i+ZlCmC8GZgLefcFoZPl/Fyzg1A/kofO+cG2fU/YvXkUS5Osiz7R6hfiy7uh1Jo5O05ZEYFaXu+FBoog10d2M/wOwpVnBhr77ybc26xBN5/bxfvVKuRfWGubJ1kIqjx7NkyJrPxuXsC/8yy7IdZlr2Hkv5vjDTmn7Qsy36bZVmPR4mm3iv1nZxz+6AF55corcvPgAedc7vlz7Xo1p8g+m8Fp76zA+eafXeSJWNehG2brnur307hewu2kza7cO4kSr0KpyySiaRTvE0GGYqsumRZ9lfkL3qwXT+ZSBaVubL1sKS+HKow0BfbQaDcgxNRUNLX7VxZ45EnUc794sbQNXBkJqY+L8OWnpPXrd4WJaYdZ7isFOk3N7UMpkgunuOwP0rhMR2rEEF3n9f5KBI+L0gppxpKkZP0hUSEcz3SrB7unX8T80eKwG2McrpNMvjn7fziyJcpagJCwuhKWML6/BsYvmsa/Dqx8UaVRY5AAv/7pWtvkjDjIO33DKT5ahXvHt9lUj/YpEf7Ru4QlVrgUL8ezbUT5V3rfSOwrQQUhawJyb5RRG6XhM9GL68jphqtNuU9o6XMDVVjSI1Se8jVZs3Suf0R39qkzbFuOytHTZyjWpMUfIQf9GkXthW828W5anxTdNkTYz07YA2+Ez7QKzwkNP49jXO7PBMvSp1CNqgdpU7XakotB/i2gzeKyt8QrZ95n8NQ5b1lkKIpupbPbUfPP7C70LY2MmF/QMSBGQkmsQCYVYF3Kvr0CWBlZD5+tia+QwxmdIkQh6Fa6D9t4d0HePD7Aw9F3nd+ivq9g1Fy8WlIu/IgclP4RhURIbPCZJRSK3emPwQLrqn7rby/hyDtzv0JmMW8/wfhCZ+oLNq3Qngjf6N847Ez8l1tGW/71n42gMoJjwkFAZxWQH7O0WCTiue2taAgn9v7kXvGqMg9q5d+52M3HgWcpfwZN0Ga6w1K58eh6j8tvy/1A4omRPAeC3y1qm+PFn1f5nEoEXer9dTrZrpYG7n6bFo6PwqV+asVxEDXALt+KB/lL0hnq+hDWMirynLREc4o1+6ydOfXo5GFJgpvfayE/OkGeeeXRUnTU7AjkdUqDxzKv/cyNWA7wXkJlAB/ZbqmIVuB6oCzTnDeDW99auM7rYasZ8uWzo9Hyp5Kv9XAeK1MDR6S03TpfCXPREGrC5f7p2YZ2g5wbptnVuCSjFJHluCbMNe90rVkYHJP4+3BnmY4f72dd+6to2ceIo3Zmcg/8Gz7QH41lm9hQmWAwPdE5ucHKVW4QALPSRTa0TKRroJXZ5tiARqMpW4g4WuGNKmPIuHgSkoVa1Bi5VhN8l2QqfkO5Ds61Lu2BPJh3TLyzhcSDxbaFCWkXjgCu6eN9d2UaowjQfcWinKboYVuhM8s8vvQwjgCCfNbRPo+0cbqP/Cq9di1BW3yxd751PxaAKchaEEJ4o0iCM/A8tmV6OMulOA6tqE5AUtP5J0bajQyBAWbPEa86tIuiPHvUTqfXATtO52AFrNY9aUTQngj4fxf0cZkD+/8AojZX4pSmYX8FI9AQWPXoYohfdDGZXH7zgfH3hflQ92tdC6fU/2QX/jlRAKKUO7Pj+35vn/yIkgIP5JIYA+l8sIBGtiEhIaUNjM3oJRbT6MFZSrycV+ZIoXTpiSCGFCGjQsIBEHa9QMozbfIO/ubrGSWix7A+VSKVGED0Lzfn6IM5IQYvL3Pt5ES4XbE249Cwt5AJEjHYPdHVWieRH7yS5d4wMaIb4booxOc90ZWmW8iAfJqFIQ4xGjzsARddoLzJ7yHonzrChRliZMBMog3PYesZ2fauTFoLg1F8zvGB1al+3qaC4YjUFaYV6rw9s7V4pnAPigW4l/oGk8wAG0sn0GuaiN7GOe2eabHXy5Dm7zdsVyg3vVglDpSSnzfaOhFo7V9kAVsuN0TDEzuFG/SMshKqIrWJvZ7rteOZlnPCaS3Yrs9G+CL7PfBdn0wRVBSWbD7LuY/aR90o/J9scG0D/lXxBxXKl0LBj+V7nnFCG1lxNhnUNTIjdZ+t/d524huYxRZ+5/2jIWQoBRL1TQY+afmAue2iLHfVH6HRL8H2+Q8wfo/1SP+xRPwQ2zyTDK4bvcSKZtpff8ACWKbowVzMnJpyDUHwVJ9BvuBh+OySOg+lCLvYhBvEhGE3j1HE4jot2uzKLTRu6HJO83GIHcrGUu49vxRBj8JbViGoqILdRbBfze6+IZ9pxUo0o4NQ8xmUcL5XLdEi/4xhuudSMC82rtnxQjO71FonZ9CAsNdaJ5tVPG+/0wPbEsAACAASURBVG10/AIKsuvnXVsICXfRgCIknN1j4zbTvtsMPI17qG/kk/wE0ih20ayUvnOMD7SduQG5A+Rpa56073UDElDyaPxgEAM1Mmwk5mL0nb3/9yG8cW0bZ7v2EJaAH2227kPC73sUJQtj7/wjo735gT2QsHQz8vmnAvYdpPUZjoTCh+nOb2OwneD8E8N5OAoknYV4tV/+Mxhw1iHOL2EbCuS2dh/iF9dTWORS3+kHaG0aQhHwegdaK5at4APvI2H2WuKpgGKwnfDM9ynmwXS0gb0JKztq92xFOHVjJzh3wjMnoHX1XKOLrxqN7UUF/7F3O8L+/5LhPwOtF5NTsJ3gTVoGWdTu2TbU59x8dP4ACQdPUmgk+6HULHvaBMrrA4c+5nbAa97vUxFzXaVGv+PRYruifcBfWH95VYgzsfrCEfgNKeX2Ax6gYOgnEzFLoFyo93m/tzECvgn4cgXeK1GY8j+HGN4OyAzyB2DjBOwZwK1en/+OtDM3ISbSLcVWCf5k4EMkTN6JfGS3pBCwjicelX8o8E37f1WUBWFHm7z/SYSB2P2ne7ArUDDXG9FmJqXFjkYQGq3lJooQfe2CEl0PRJH7byM/yL3QLv0Y794Q/FteX08ZzN1oocjrOndbUBD9T0Gauq2RkPaEjfdiKLPAybF+7fxXUG7Cccil449I6DoBKwpQhkWa+QcQs+pv32U8EiKPRoJTcLzsvpvs/73RrvwdrHYyWkQPTuFs1w6n2Jhtajj82MZjwUjfjyILyRU2xl+mq+VjA9I+lW1lbkDavLwgwQDgV0hQXBIJLhcRcCEo0WYsw8ahMXxrvvM6EbiOcLbzR3r09yRFWr3NjeaC+ZqRlv3reJpqNJc3RZqv3WPzCW28ny+d+y6wnf1/AbBZYqzbxXm0jW/fEs77ovm91ezAGfHVX6P5OwoJzhujTcz1VJiuEV/5jvf7L8g9aT0k5J1PoVEv9709mnNrGl09hDbyOd4nUwicZdi2eSaap896vz9GmuiT0SYqtzZ2c3PqBGfvGS3zTDt3BZ4GEwnTxyJePzHxjfqiNfkBtIZ/RNeqVE/69JV4Tju8vm0ZZG4+euYhEjiepbtj/65IwIyVBtsO1cLOfw+wQX3bPx+BXQIveT7S3lyMdvBPoYUwqqlA5v7j8EobGtHni/J7RAQ85NczzSOWy40wh9pESvmLOCTEHYF2zSd61ybilToNwF6GCetIO+nDXo23C43Ab4X5pSHGeIlNutORq8W/JGBHIy3MMfatL/GuHQZ8JQF7BxKAJ6DF+kQ7vzQSTGOC/zC0APgapNOBa73vdVWi3xFIGPwAbViu8a5tgASKWOnaBdEOfzl7zu8ptBpbIWE6xizmQ4K3X+ozz7zwOtp4xNKX5YtMX/veg6y/N9B8uZ64hrIvMvF8bO/mL2jjrO8YbH+ja98ncBu0CH2EmGTU6kBXzd4lSAu+DhIKP49cSULaiUH2TlsgLdS2aHP2KEXA3l2JfgcgDd0hFELJLai861GoVngMdhGKMn0r4i0+9vtF4gE6w5Bgk2vJ3qCopLSj0W1M6z8IzaXUO9+T6Hd0Ozh7961i3+VbwG3e+ZFoExLT2PVBi/f37O9FFJvrLRAPigkLyyBT6GAKfnu40fh8iG+n6Gt8mzj3R/PuAaTwOJdic7wLlnYvAru0h3O/VnBGG88pyLT9DjDVu7a6nQ+68tg9A40m/oI2Lg971z6HBJYYbS6G5u4QO1ZFrm/3GM3+D3HFw5KIV38A/JyumroqnjkU8cWZhvNM71ru2hCDXbwDnNvmmQa3G+KNZZ/7TZFrzOgE7PyI392OLFE7e9feJJF3mGLj2A6vr5JBgi5sc/vRcw8q/BrPpUguvBLwQRvP2h8thMvVuLdPmcgRw5zmf/Sa/Y5EjOsh4LoYvE2Y29FO8kE7cg3xI1jC3xKMr/ofY/c9jQSy5RADmgkcH7g/n3ALUzDG1ehqUn0AOLAMW8KhP93LKa6GNKN/JuIva+cGIAZxKdqdXeDhdR9WWSfUN1p0jzL6eAEvWhb5dX0xBGvjvD5diwy0HEGIGPQZdM3NdglweYpGkHDwN/vWvnA3msgiWD5H91K4j2KCcRVtok3L9UiYzit8BH2LS3Aj0MJyn73nGKS1vDLUr71PrnFyAZwfwwT/CG2Un7cT0qT9B4UmuUprN9j+9kWL08ZIc/B/Vd+YNjJOUCrRGZhrk4EpqXemsCzMTzjDRjJpOEVUbl+0wa71zpSC+FrA2ecXw5FZ8W9IuJmAFtRLq2gTCdDTkFtEHgV8TV269uhsWWTenIW5JgRoaQlsY2c0neP8XBXO3rjMhzbN96BgjzEezlf69yZwzX1Al0Z+9FGcS7ALW9+reucuAS5LfCc/6Go1JMA+hPjv4kh4ic7HCB4L2vF6ikZK434enpKJejxzHLLG7IPWs1wWmIwJt4HvFBOsW8LZrh+OeMCHmEsG9XjmKWhDdRRd14mfUvJrtfPDMG2xR2erI4H8EpSC8uFYvxQb2ZwvHkYLvB5tkqYhGeQBasgg88LR+QOKj7EA2uGcYwP0QyRs5CatMhEuj3YmqyNTxPzetcFoB57amXRbSLyP+xDmzxgjwsT505CGadlIP77pZ0nkmJ7v+FcFPkzgPB8SDvoZQZ9lk+1rSGPyUAI2qGW2a6sBb1V8p7KPrS/wLgX8ZwJ2Re//fkg4vR35vkwFXk/A+gwtz2HqPLzfaIHW8klcGUGINCrnoIj+Fei6GI9ADCumAR9FIaAtgRaVa1Fp2jWRWSTImI0ZrB95bj8kyOcpxsq0NQJpyX2GeAKepiH1jRBT3c7mVj8Kv7dfIGa3SAR2FrIIjC2dd2ijdDNFwEyIOZ5heI/wzu2ELfYVeAcDfuzaicBzsX5L97aUcQKZ03aMzIfxyA81Rh+hjd0nfs2kM2wsSKksct13RhrJryGhdb7StSTOds+llPy8jeYuNPrYFS8wIkBfXybgmoMEyxcT47UJXQNGxnjXLkb8NpgpA7mOvICXjcJwvrgGzhORxqlbMBzahP2EuBY7WorRvkMK5xWRP+G2iJf4tDUK5chO+frfj7kCeOdWRmviu0iIj/W9jdFCqBDLIOs7rwIY3IxFnjvCYGPfeFVkvVrFO3cA4pVft+8Ys5ReHRrnmjh3K9uN5tB9sXfx7htGIZ8MRX7zl6I1LVcWBYVK+xZ/wxRIdq4vcgU5G/HisSFYO3cK3YOSTwQeqMK7BLMk2pjVkkHm9qPnH6iPPAyZ65aJ3LMSYmCPIq3ZlTb5J9Tso1sUrndtIBUlHCsm3kjiu/X1kTnwSbwoYrvWHzHHYK1aYxTfwHa63vm+SMgZQRF9WSb+nZALwuWU/GuR4PGJ+b7cr507APhu6dxCpXeekOMTg6VYcAdYv4ejTcVykXc+CPMp8mB9/7MrMJNqAHYiWsRep7SAI5Pjb4mnJtkYuX1cYeN9B2KWuT/ocMwlJDJes5CTuM9chxqd/gwJlaGIy7Xw/JKxogxIWzDKzsX8KNdGC+8tSFi+yOs3F45j5psNUQDR2WiT8AskPOTm6IEUWqYQQ3/FaOgBvFrJ3jcLLgZ2blngT2geX4cY7WL2/XJNVCzLwDH2vn/ECzDxrp8Q+8Z2rq2ME4g3veH93hkJEOdSaCbzYJQQfZyB5y9ZujYc+TnHsk1cAlzhz3+/n9g7I3Pp68gd4H7krnIM2nStWwPncrnfRdE8qaxvTffAwumUtFoUrkTleTyB7gEjl1IEIi1A4epQhl0TWbumIqGlVj5Hg10eCY3XoY3/rmgubU4RlLR2pN9YIGWfGjiX5+KHiPf4czFauhatMd+jmHsrG+4bo3VzMeIZWFa0d74Uzb/c53JVCneWdSJ4j0UWhg3pvtkZgASfHSKw6xtt3mC4L4Z49PzI7L0R8TUip+t8Hgw3mvE3tzGcy0GFq1u/Yyh4Vsot4qt0FypXsP43Q5v7bpYUpDB7HfH1bwJnlJ4b7TN/H+Al7/d89p3no+DTyTyvRo/ltTopg8wLR070HTWrUpFlNR/mnLsO+H2WZV9xzg1Hi+KaiJBvyrLs/QTsYlmW/aEDXPtm6apNKdjXkV9PH2S2fiHLstNrwr6ABJo/oEX7B/acvwPXZ1n2HwnY3ZH/05uIWb6PNFpLZ1l2kt3TJ4tUg3DOPY7MJY865zZHTHlBFBR1U5Zlv0v0XYbdAmkMfwV8NUtUzHHOPYNMSw8753ZEmpJByJR7bZZlvw99D+fcYMRovoSE5eFIEFzb4H7lnNsmy7LHIv1eD/wgy7Ib7PcEimwKt2ZZ9s0EzmshbegsxKQnZ1n2ZumePlmWfWxVOzLv/APArCzLrnfOHYCikPsY7n/Ksuy8RL8zkWbseufcaLSonJOpuk+yOeduQwzudvt9M9rgzIfMmU8nYL8OPJpl2TTn3GXI5DapTr/eM45Gc/hptCCsjLQ0u2dZ9m4EZjCqF70XMoNugxaV0cANWZb9oKLPwUjrcrXB/DsSfNYB7s6y7F+tStrvA7CHoRRQk5xzByGtxveR28rwfE4l+n0Vmej/4Jwbhyw8yyDt5e+RefbtAOwgZDlaI8uy/3TO7WV9/xlVO3sq0e/VwI+yLLvROXc8Cn54EZnsRmRZdnzFeD2EtLbTrSLWToiXfIg08F9LwM4Ensmy7Cbn3BAUiHhFlmVvWgWZgVmW/W8E9grgN1mWTbXfn0fa7PXRXLmtot/HkU/jDLQIH59l2UcpfufBT0HuUO8gATx3Adohy7Jftfm+fZEwHnvf2FwchHjJMxU4nwv8I8uyS5xzB6IN/zAU8PNOlmXXVMDfgyxR/4qqCf0X4rvbZln2Xwm4WYbjM2iD+CbatP2XXe/C60qwdwMv2nhdgTapa6DUTxdlWfZ4ot8pSBb4qnNuP8SnV0P+qw/m4xiBnYlo6Ebn3PlIhlgUrS8PVdDWYOTOcxXiQd/Lsmxy7P4S7OlIwXC0rS2XIN53WZZlf66SMZxzDyIXsNudczsjwXc7FAR3WZZlD9XB41PZ2pVk0W5zG7rmvsp3OcPQohaDPRT5/vjBKguineTzxM0oSyL/qnMIRFWiSLdUUNAiaLcc8zXaATg3ArsZ8Kr3e1HELEfZ792J+IwhAfAF7/cfkN/mHkjb0C0dTeAZx9q4LYOE0z+hRTSvPx+rNLIRmtwLIobzfeSLtA/a+e9bHseasHeQzmIwBC3QuY/Z95BGdQ9kapsYwxtpvW63/yegXewRaGH4BRWuHMgP6HG657TbHWktUmbir1NEk1+GgiiSvoB272AUYf0mMk3OotjZj0N+mOtGYJdAzNH3HbsUOM/+PwXzkwvA9jM8L8J8JSnKsO6FF0wRgF0MCaP57/mQ1uB5LNsD9fwBhxtcroF+wd7nNiKBdkjLeIf9vy7wD7QYXYEWxA0r+mw74wTyw3vK4G6lSFu2KNJ47F3R7yz7fzl7zlR7zpMkNI5IMPk62qSsiATodZFm6iNglwRNH4IE4Ylooc+1bIuh6P5oQCMSUD4y/LZAi2eea3gHZKmKmVOHGe22G1i4K+mAkWCcAOL1T3m/ByH/vHvw/J0raGQkikYfac97E2kvLyLiNkF1IOUpRFxRSM/FPUnMRe8ZeSGZsfZdc/eT9ez7RYNk7L4xhuPCaHP2E5RN4LgYLOLv96HMALsid4ir0Tq7FpqTt0ZgByGN/Y5IJvgB5gqDrEOPkJ4T26IMJIvZ3zzodhvEw5eOwA0zug0FFe5k3zDFA04HbrT/J6DN7UWYPEO6YtiNdHU7GY/8ZSuT/iMe80tkvVkT8apd7dpByEoVC9CbDwn7A71zvvyzGYlKh/PC0R6QmMm37SM8jBisL5iOpHAoDgk6CyBh6AojHr/6z2tESmYiJ+tHbLLsFrg+kFLFpdL1KRTO5H2RP8/40gcPwqP0Pcfl/djfmzBhEu1gY0x9XYocepvlE8F77jcSRJgLyyNtzPII7h+gVCivkJ7wa9iYzbC/foTqthUToG1Yu+cUxERPpWuE+6aIYcQiVE+hKPc6Ca8CBmKSx8T69O67yOhlG7r6n/2USDlXJKh81xvzlgU0FLn7Ap5Jxs6/kaDrYYihL0FhFhyLongHowCO3LwYoutxyMF9ChJ4HstxNfqIuc7MT+EI7+fs/BJaZIJ5ZSPP2hGZqjdEm5hxaG4HGSTa1OQm/cPpWjb3KKrT4nSacWIvxL+ex9vEok1XMFG5R1fTUUDPq8AJ3rXrSAizHvxkZJo/3Tu/G+aXXDGfbkGuOxO9898jUqzAu2cBJNS+RQvlfpHAsR6WQ9ij19qBhWj+X4uCXSoDRuza8PwahRlzmD3nAQKJ1SPPORZtjFYwutzAxjEm6AxGG/FF23lfZFa/FWnNas/F/B77e7LRyN14wSloAxNMiUjBs/ohgXIK2oj8HM3NG4lXheuLrBR5wOyCKA3Tuciq918U7iChd94drQ9fs3Ea5l17h1KS+QD8uSgu4CY8/3uDjfIgCmVHy0GFdCZULlAeC6OrV43WUgqPgUgQPRlZOL4fwDu2SZuK1tPf4aWHpAi4u5gaLjhz89EekITRw+3/6cgp/jm0kx5T8xmj0MJ3JdIGXYsEiGCZUMQo3rGJszOqQnAhhaa0SlAYjITG3H9oik343xphjK+Bczm6dTukKbySRFR+4DnlhN231nyHQ23SvkshHM9fE3ZTYxgTWu27Q9ijkLD0cwrh/TRMuC3DUjhn58x5DF0DyWZSVO4KMcfF7e8QxOhuR07v0+xbPZqAnY9Cs+HvQisFNLrnuFzE+/8IIiVVU/SBFrUXsbyTkXvzhWhZpJHYDCtugDYs363br/fMgYhht+pgPxFpwyZX9Yn8nXLtxuI+HdhYH5F6Bl03wLUyTqDF2vedXhMx8ReRhuQWYEZFvwujBf4ixAeHVvUbgL8eRcU+g7RneX7PbrlpkVVn19IzxhldP4g0ht9O0PTCeEKyzYvlvd/Rcr92rhy8lQstLZUmRBuAy1BkezJgJAKf9zvI3vmaOv0azElGl5e0ANMlK0Qr74tcEg5DfDNXcNSei0jTeBpS9jyENsU3EgmUizzjArRRmdLCO4eKdNxEjcBCpKBYCmn5jkeb69Pq4IzSwt1s+OaFNa6M0TUSQFcrnfPX1GhQoXdPW0JlxXssT/Wm0l9XVsFLEYk26bHS22thwitKhn+djdWFdA+Aq83r57ajdQDtAH2H3F/bZFsUOTVPbuFZg5Gz9r5I8zWJwvk7FCBznfd7vH2QQ2r2tQ3y75pkH/RVCoflybSZTBYtCh9TmO5DTL1bCif7fzgyqawQgw0860TgyRbwKzs+54x2GPJtjPbdIWx+72C06E5B5tDvI+F2+RAsWqxjZUBXJbJh8cbmPmS+zU2aSyJ/u/3Q4jsi0m+e0mVA+f2pENCQhm8K0j5NQ2bRwd77X068HGueOPoeJBz5QV/roepJX4jAHoCE7ZfQZm6Ud20Q8o/aKgIbKyHr0+eQEKyd29v6fgotIH2RoHQ+kcCtmvT6eUoavBbhoxknkEXmNrR45mbfgUh7thoylwYTrAeeNdjet6+HdzLThQe7ENpczkRaj7uBryVwvsWj0VxbuCXaABxKoeWOWYU+omu+53xuDkEKhVgA1knIOhIqFbwKch3YJAK7HxJIXrR3WAatEWug+REMGLHfeUnEOymVRLTrCxDJ+oDmxAVIG3k1EhLyevXRTBHeeJR5Xi6QjkOaqdj7DiayMUdBQVOJzEU7t5F9z9MoNMOr2PvsiuZbjHftiVxJ7qOwpKxp3254xTufjTTXvqDkbw4vowgALfd7AiVLBPIff8i++22Yq0YA9kRKxR/QHDoNuRccQWFBKcNeiOdaR1eelae7C9J0jblZKVRG4PI5lcpGciRaR17DXBO8awuiTWlsLs4ETrX/j0PKtQ2Qe9sHJKw689LROoAm5gSP+M7wro1AJrBYepl8cn8SsdhCv+tSpMvJTeY7IxX3FBKmY7t3IZv0ZyNHfp+gN0FO7JUEF3iXrWkxJ5zd2x/5kZ2QgvX68YXBcfZ/NNmvN86fjLn/LmiBOK4O3u3ABiZUvpCOIFCiz67tCfwvEvLL/p+D0AJ5aKhfupY23QJpFycjzXvQRF+CPw35JH2MFtPQuHUT0Cg078vZ8STaZFxHdXT8ELv3aLRAz0DZFG6hiBI9OAI7GAWWbW9z4y7EmO6kWIRifsWVJWQrxiq3VhyIzPJ3lr9XO4fhtT9FXtpWecRApLk8KUIjZyPrwg1oof2C0dQ2LfThAs/tgwSCqjnRJWcyElRWtPEM0dYgtPHL+d6RyAo1nYT/duk7vY385V7Hc33J5ySRcsWl+bQ5havBdIosJsEqNAb7PuK3K9mY/xvS9CUj5T2cQyURo6bQwJxYx3D9wPrNI9Nj32YIsvJNRPwt5Nv+hRqwI6ngyxG8X0c84ytGn8FUVonxOhgJ4sfauO1HkbUlxgf6IFP7PxE/ONvOb0+FSwSyNnwP8ctfYnIARbrFlRP95jwzF7AXR8qGtaku6lAuvb01UgJMpxDkk362iWcnhUoK/9IQbVRFxedzYoJ9q0koS8aNFBvKoN884jlbUig5TqCryX4KVqluXj86f4BnqrSBvif0MSOw/ShMMevXhSs9Y0WkpdmoBZjxdPWJmkEiKbxPGJHz0br3bY5pSHvX6rgEhb6asD2SLgLi6bki938XCWaT0A57ozI+iW8QK216FVrQNkn0Oxj5Pg1FVVBuQELh3VSUfkOmSL+KynCDm2rvs2AC9kBKeTLRDv9mrLBDgiZ2AZ4onVvCYJ/BS5sS6LeqhOzJNhaxsT6Twl0jL5/3hHf9IAJ5EO1aqGxgy/SWoqvY89Bm6H4b9yOQVvr/kFC6AhWLSrmPwDepRes2L5ILr9031L7Pwmiz8yYS8A5CAsRBFfB+MM4OKEikrkWpqlTwegnYQ4BveL8/jxbe26guFdxJWebYnLitxpyYhCLT70VC2rHIXShXfuwTw7sG7ETEY1Jayuvt/yWRpsxXmGyc6DsW3HcbCphJBrkgTerlyL1gJvLr/ZiI8O3TOVLknIwyFzxrfX9MxFe+RJfT7P+xaP192mjkrArYUOnt7Y0u/0Ck5G7FXOymfAjctx4Blw8PdhsSbn9Iq3m//b8B8ss9wr7Tz6nwsy33Vzr3CpGA2Xnt6LkHaZfzIkVOvDKzzheBqXRX8y9MIom2TegjQ7A5sVTgdpDBX0Up3yFito9H4Aaind7xdBVg853UjphWpI3x2ptIuU2kOVnaxmxkoN+9gQsSzx5uBH8oEvq7LJ4Gf2EEdiG0segmFNeA7Y8W/T3o6kOZC+ybEdHsIEH0Ne/3qSjYK+jEH4CvKm16fgJ2H6wSE2KQf0dMdhLyVQ5GP9v9w9AiuzXKg3dN/l0Rk0zBLo4W2rLpqo89syqLwQ32Pcoau0soacNK17ckXUL2NxVjfQZWY9qbJ48hoW4DvFysJbj10FwcErm+G4mNpdH1sqVzvgZ7c7x5GnnGFhQVhS5GFpbppEvf5nTdl0BQSqpfNI83QWbXUIL2bUkvZGcgofQcugop2wI3V7zrQ3jWAWRRej5FGy3Mp9R4jUYL7TjET66i0L5NJ5FvmuqSiNHgrYo5MRk4KgG7H4W5ezu0cXkCCRIHk6g4WAM26YJi9+/i/V7dxnwAChKMFsagOrjvizW+9WQsKw5aY3+G3BOS7nc2H+6kyJhyDdJQ/x+RrCB239PIorMMEkZPQn6hGyArVVSoRALkFchNaiLdS28Hs+TY9X5og+KwgKjS9ahQaTh/jPzEc3ezvh5OxxCY3x78GkhZsAXSpn/Fu3Y6Zo4PwPVH8QEHEq44dxDmwsc87DuaH31ooVm+0fz/vvZ3XTu1INr1vAZKSurdu5cN3F+RJupi59xLzrlD7Jl/QX4hWF47SrAH2D0hWLIsezWB8172Mf+CzARTnHMvGrxDfqUnh/pGfnlnokX7DOdcf+fcslmR/+49ZKIN9buzc25759zSgWt9kcP6vZF+z7fxWA34qXPuMefcOl6/307Agnz63kM7yDWzLPvY/x4Gf3cE/iy0iGV2fRHn3Biv74cTsF9GQtjuSBjGckHmOdn+gjQ7oZahnX3erkHf6x7n3PYRGL/9Gu3uRyINw989/DZHu+fYeD0BfOicuwoJiNdkWfZKlmWXI9qYkOj3P5Dp6ni7F8R4QFqSoYl+/4C+0/3OuduccyMAbKyXQCbbIGyWZX9FGtijgBudc2s45xZ1zvVHvmd9Ev3OQmOFveeZFJVHTqeYizH+cAOai3l+wr8bLlsi7djtEfgrkEXkr865oc65sc65HSwnIIgGuuXv9NptwE3OudOdc2vnJ51z/ZxzA9B3DuaH9NrTwO+dc9si4WFfRKv35e8TgJls158EjnbOHW/j7Wy8N0v0ewUysR0JbOKcG+ScW905N9CuL4dMkMGWZdmliEZGA7s75w50zo01nH+UwBmkDf1xfj3Lsm/Zu+ztnJtkeVFjrWo+/Vui798g/9IHkNZrERRQCPIjXSLRb57/9jdOeX2XQ1a3PyOBYmQMsGJOjEcCVAznB5HpmSzLHs2y7ItoHiyP6O5agw3NiSrY6xOwIPp4P8cty7K3EO/bBCsuksD7wqzI2Tsz78vaUkioTdEIhusCzrkxSCjcEilivpuCNb5+IrCic24JtOHcGG0cn0zAHooE1/vQd7kly7K/ZFn2Mkr/tlQMUVuXpqHN1QHAes655Ww+bYdSxsX6/Qra6D8AnOecO8opLy5GI8sTmItOuan/kWVZH5R2cB/v/cnUbsiy7E8xvBG//CvaoDyFgqnztgbKRRxbj69C3+ROT97CObcwslZ8KdHvvNXakWIp1NTzo51UtKyl3fckFgmGtCkXU6R5WH12wdaAT+22cz+55ZCv1Uto9/g+isiNZhMw2I8RozofaVJHU5hw8gi/UKWOQdZv7ge4LjLV/RYxt6gJgexQ/gAAIABJREFU2IN/CwlDJ6CF43iKyO2US8Jg5HOVB6WcjYSL36BdXdU7v41MKiuhhf9SZCacgUxRdesu+9rc/ZEZLJgKIwDbx77z7SiI6ioSpU09uN2QRuOLSOAahRbA+7BKHj5eAfiF8FLJIC3Hr2vivJTh+3sbq1uIaBkDsIsjH6Jc0zeTUlWuqvHy/h9BooRs6ltRJKd/L3LfOJR8Ov89w8b2LqQhSmo2DWaKveOJNkaTkNl94xBOieeshASfZ2v0uR7wS/t/TbSgTDbaLvs5l60K62KBTkjTNQtp+W41+KTfu/ec+ZHJ/TKkzX4IuLjuHApc2wYvhVsVfbQznwx2MTzzLdpgf1QTtu2SiD0wJ3yt+3Dg32cXbIBmche2zZDw80LdvkvPqSwnnX9f+7srsqrV7o9CQ3goclmoM5983r4oXTO3fJ5SGqTEc4YhBcjraC2vKr29DsabUMT6I8gCdyHVbg1+Boz1kSX4GYpYjkr3G+9Z/Y0+ZyFt8DmkMwu9h6x2/ZAm+WZv3Nt2y5tbj7qDOAAtILnTcD5pDsbK4KU+ChIAL/J+v2BEcRiKuA6a8GrCBn3VasLfhpmGAnD7UiTOXQHtnPKo8qmUSoeWYDdDi87qRnB3ocVkL6QteA2LpgvArmQTJY9AXTDHE2kb1qx436Ppmud0U7Sj3LTGd94N+RbuTpHQehVkkpiKBRRFYA+jiAYeg4TYjZHJ8mYCeWMrcMkFncFIyxRNhm/3lQNGBqN8g7tSmFjqCsTnIGHpDizlSQrP8nPRAj6WitylZVj7xnujxXfJFmEHoQV8KSKlBWuM3zIUUcQpf6qYL9yVwCllvA3XwciEOBwFL9xr55ZGZtlK30Yk9D5itPYFtAj9FQlp27ZIX4MpfGZTPo2bo8pP/rllkebiRySCuWw+TfPmx2v2vhsafeUl/ur6nvbzvu3AVmADz1oqBU93N5/a84lwWcOBKMhpuzZga5VE7Ok5YfevQZHovbaPcQ/ADkJBZZem3jkCmwzuS8CtQxHZXlvIsvv3oMgdXRWY1I1n2vkbMJN7CufAN94EbWCi3xhZKG72fm+I1uWv2LwM5u+0sTy0jA/iORe0MLbl+TQ/2pQeS7zs7r7AdO/3MOA7FHLIDD4lvqOfvGONgcwFg4Mp+UsijUEeAZpawFZBZodf2yDe4V17jbTvxYpIO/cRJiTUhfX6fr5VeLRbW9b+3wRvwUORxd9J9PlJJLz9Ho18Im9G2o0fV+B8BdKuHom0RzfY+ZNRGc8U7FYUwkx/O05Cu+1TSW8clkNmkAuRicHPoLBrxTtvTLFQ7ULX3IcHYQ7dLRFne2mDHBW1hKv6QUxoP7QbzhfuOvllXeh5de5PfZeefl8P9hMhvp2xjoxdaiE5BQlxD+HxEqTpvKGin1wrsAaFj+4MpAE7hRrFErxn5RvqOoFFi6JF4GYkBPv5Dr9KJAuCXR+OLAS/QZaaQ71r56DSuyG4cp34PhQWjlYCr6Lp2Fr8xi0JUz5cO/3luLcwj7pEP3cyJ+a2o5P3qBp7umpzOwpkbQU+9k5o01OL1lr9xmgT+W3jFQci17O97dq1lKwseR9lvD3esQrScL6KfJdb2fzXwtt43S5oPcrn/1lIOTQWlWvtdRrtyaMVglsSaYxeoZSTrgIuZxL9kNluZQozzHkUUflVk2dV+0C1YemaLHsNJEC33Lfd4y9Ed1MjKj/ynH8Fjo3B2qTMnaSvR+b+XFv6HQ82ptmITfb1DT6aEoNC67Ik8tEZ4V27t5V3Lo39nVhAQaeMr9wHaSf17UgERRHQQtZ8t9VIRJIi4T2WIqYqMGdTItHAaIEPObbngm00QAbtyMcgH7xuri4GmxqrquCcHYinEPJzu25s9PAPpDFdB6UFCgZD2rlPhDHkPnM1cif5ldHA0AhOg5Ff2FIEcqMiZp+KJs77XQj5kl9lfw9Em7dfJL6zT/+rImH2aSwXLnKr6Vb9hurAryht1aSvTTqgr+h86oS+0AZ+deKlLaMlEamOfk7NieUopTfC02ShuRjrt6qMYxQ2hmvev/39AlImxNa2nJ67afdJB9mti7kgld+7Bn0shaWjK9HMJ5lySJiR6V5koQ+FfLBmmVZrjJX/naIZFLz7N0Ka4zvpuiF+AwtKK90/gHRA0UASlp0a9LV5HbxL8Hkp3zdoUwaZm4/WAaQpfQWZc49EvnbbElB524e8B0vnYASRV9JZGPlijSoTmv1OVcVZIAVr5/ZHdcgfoHsUcxIepZHYE/leblO6NoFSWcgA7F7IH3Hr0rVBpP1cLkam/isC18YDj1R8m0uQJuc6umtZ+mEmrwjsZOSX100Di5jY8wnY0ylFeHrXtgLebIs4q7MJTEa73gcN96MociT2R+aQmAZ8A7TBijH8lFD5DJZyh8LHbjgFY96VuEDxHRQscDpFZG5f+z4D7J1i5qOLkKXhXYwZedcccteIve+NyFLwDaQR+Cqat/0MNjpWBn+/wT6GLASDkBCRb2KC8GjD8IyNT86Ih6AFd6rhsUNoHtq5xZBpzS/lOBxpOKrSAF2NrCuzCESx2nvEXHbyfnN+tTjyvTwNucA8Q5FDuMw/FkH8rl/p/M42ji9jmSoCsC9S5NodijQhO1DkIIzS1rxKX2gePoF80Dc1mCUpNgQpnNuOfka+pdtHrvUnUYqRDss4ItesGF59keY/ls/7LGQdeBK5gm1Lkbd4QAXez9l4XU2h6Ojj4X5RAnY6kfR5iAdOS+B8uo33TyiV9ETC/ePES29vSClXMgUvqcI5j1Lfj0AeW6PplyJz8Xyj57tsrNf1rlVqOTukr5QccSnwk6r+58WjPSAR/a5IZX2jDexxdN29jEPpI5ZHWoG3EdM4iUJLGUucm9cdnoH8KEIlzVKmwRWQaXBN5CPyId0rI8T63tsmx/k28d9EQnW+aK5IID9mTdj+RKrYIB+c55A24BqkGc1r/M5nx8gQbAB+KtJmnIfMocGa8RHYq9ECeK7BzocWh/Ui77wn8ss9BjGk9VH6jQONTlamqPRRht3ZJl63utKIKQ+mSLRcHq+2ndQN5gngT0iwWNPHDy1kuxBY+K2vp73fp9ozriFhvvXubyswx973LcNtGxTZv1lVfwa7LkVgzgr2nU80OqlMCk8HwTnIr/Rq+38hFC26G6X5mIC/Gphq/+d11dfG22SW6cofL/t/feSec6B3vWpOlPv9nPWbBzL0C/XrfePLPDpezuBzQWkoBQ8s88xeCfzqLfoy2Lft/x3tfR9G8+nwCpzXoqgVfxmWWqomzmvhFUOxvr9DqeJZot+2yzjaO38fBeXsQKIwReQ7vWv/b4rWmEftWYNq4P040tDfRDoNVgj2Be/3hshSOImSljYC+yba1J2JhMALkQtaGba8RqyHsj3cYHCj6uJs585H68I9KLp9bf9+xJO6VQ2kOqAouJHtIfqKyRG5P/LnsJiM8njN60fnD4hL+MdjFYzs95bI7DsN7RyiC4JNlm8hoeBuFN15FEVuumNIBDMZvJ/n6wjgJu/3nkT8x5CGIDcfLomY+jmIuS4R67MHYJ/EmDda6F+0MZxJRb7BOvApwu2kbyTI5gL6dKQluRSZRaIaLDrIRGDn23JSt3v9hSz/PkkG48HeS1HbeW+j03VRKpCXsQIPCfi2AnNsPE8v0fAT9r9DG4hYBaw1MPcU79wwtFF8k+o8vm0F5yBh7GEsQ4J9n3sQY34QY7CJfvth2lX7fZuN/ztIM57KWfo14Bzv90Zofg5GWpUriVTDifR7H9pUP0rA36xE1z+mqLw0BfGw3yJBPma27tXAr96iL8TX82DIQ5BQPB7Np2dIlICkg+hn+455bfkdjU73QwqWn5EIoqTDMo42B55GQs61iP/tSaHlPIlIMRQkeF7t/d7Uvt00e6fU5nAmRcaQHZHSZjqF61IquO9hisTuO6E5fKL9/YC0UP11CmvCacjVZV+kyf8xiY2Wvdc3kCzwVbSmnkoR2HNBgi6rhMrc9zhmYfXjTcoBRfdX0HUn9PUsbcoR8/ox+x6sxeoZI7xxRliH2bWvUVLbl2B3QYnG50MuAXuhReF++9DR1BBoETwLzycF7YKeQBq73YGnIrD90AIyqUQcGyEt8LXEBdkc9tQI7DUk1Pw2yS5Gws2fKRbxhWxCJjVKNeCj0flolxuD/SYB/xq73h9pJs5BZpH/QT5kfZF2+lvEF+DNjA5azkRg8MvTopO6B/sg8CX7fwWkifo+RbL4VIDeWtbvj1Cao029a5Pz58ZoxP62FJiDBIJ9KNwRcnPVd23cDwW+neh3fgpz+2509XubjEXGJ+CXpAjOeYKawTkeXd6DBIbnvPP72vlUJSuHzHFXIT/Ml71rxyPNVKyCzZaYLyPFAjTT6HJ/imovocUo1e8J1m9w4Ufaxd+jObUfCnrITaOXUO1m0EngV04XrdJXHzTvxndAXw+0Q18UwSavIoFyP+/aRcB5EbgBSOgub4JqRT/b+34HbSL/TNd5fDWJrCDIDWmQ/X8iLZRxtLHeFqvkhHw2z0Q86zTEM1MuYavYN90FJVn/FrCzXbuBSCpE5CbzZbr6Ey+B1qVjK8aqP9JmzkB8+g94pnukANoiApuXfM5LR19Zgr0+BmvX10EblLzU9hftG3/Vvt3ziXm8H+ko9ahQad9nZ7Su5/yjdkAR2mA82ip90YEM8mk4Zu/DJShMRZqBqRTar1eoaQ6y+/tTBFX8GfOpCBGhB7O4fw/S/B2JtAdbJYh4HbTovo4WzJl2fgDS/qX8t9ZF2sZ2YFc04r3GJrhfoekdKuqN2yR52Yj9ZrzAhhQ8Eh7HIQGw5b6R+eADe9978LROSKO0WARuEbxNCS1kIqBYDDZDGtHp1HBSz785YjRl09CX0U40VV5wce//lQxfv3Tuy8Qrlc3n/49Vr0HuJNHAHLs/nze51i1nkCejxXcWsEak38W8/w+mKIs6CQlnH2K+smVYv2/7f1UUlPQ0CryJBueUnnExMtE9iGmP0Hx+I0XTdt9CRtNnoDmc130eTWRj6o1X7neZbwQORJugl2Lj5T1jYev3zLr92vUF0OJxJtIE+dWVNsEz45Xnoff/F5Bf5T8Q31yb6sCv3N0kD/ya2gJ9OYPL0+bk41WHvny8JyIN0C1IuKqkLzs/Cs3l3dHGaSv77m9Qo6Q0XQNrPkfN6GcK4fAqvFREiBe2WoIyh60s44jWg8ElPNZGm6z/ptgYx8Yrp+M78XzsbbzWCH3f0m9fyNoE8e9vELBa+rBoY3oYRTrE/J1fJ6AwoVh7+3j9ji5dey0Emxg7Z/SyHUoKHx0r5C7WslCJMuwsSSnTA1qzkgFFgbHe3qMv/51T1ajalkHm9aPnH6hBm2kfvC/aaSxA4aB+AJGEvR5svlMvBwVsj/nPJODvxwSdEjGtjHwGX47A9qFYyBZFEXBbUJiOzwfujBDhEAqfsEUMdquasMPouuj3Rzu7d5GgdQu2yyvD2rll8KKbbQz2Mvg7UvBIy7hvpO/bK2AHohyp/bwxOA756UxBO7m7EnjP7/1fnsTRTAQUZVEPst/lyM2ok7qd8wU0P8pzcaT5eo9wcM6SqCTeBQTM+0jYeybyPl1Kudq5FdDmqKp28yfvi+aTz9hGInPsrAhsjvP5SEiZH22a9qVIdH5gYqzygKRyQMF2RpsvEQ/O2YhCUF0XCaWzkEbkdqTZOSZGH/lcsr8bI8b8vj3nyzZ23eA9nBcNPG8ISv8W5AGB+9e09/xBVb8B2AVRNoZh3rkZBBYyiiCo/t63HYCEtJuoDvzqVu4XRUR/l0TJXB82cm25CvrK8c5531C0Udsf8Z4ofVm/u1kfvlB7GLKmPQdcWwG7AuEo82j0s+G4OnF3jf0xC1qi3zEEtFRUlHE0+lvY+92HrtkYDqDwiw+lolsZlcwciviV/70Px4ppJGCPI2Aet2cFy8lSlM3dxWi6L+IjOZ3uh+faEYHdjUCwk8E+XoHz8RGc94uNlZ1bIvKNKqPUEV8LljKnIqAICb8LofW/W0rMCvpqWwb5tBw9+7CCSA+mlFqCIvpwbYrKSX0isOV8p3k99rHAhhEiyuEPCcDnDPNyCvNGGf5otEg+Rve0KUPRbrSb87P9vg7tWHenFMmHNC0p2LvwhELv/HgkEH4SHRiZeI8CpwXGayzSeEbhkSbyY2RW8zWbq1f1jTRej6NFbwXv/BYUJsLFIrCTiWcDmJ90JoLzURDVNQTMLYhpjo70mwto51BoWcv3fDFy/jwkjHQzt1CYk7pthEo4T6VrpOZSFKasmFAWg83pfSKFe0oK5929831D9wf6LgckbYMC4LZEUehDiSRoR0Exd+JpXZC7w65oYxl1qbB7L0bCmO8vtypaVCcCe9TEeSs0L3M/582wCm2hMUf842Qb75xXjEem+oP8caxzeN9pVUxYCdxTDoIaSSIFVwD+crza4WjRHUFXgSVGX2XYRTBeVYO+yniPoKvVI6WdnISsXWehDdJYpLjYCmmoFqewCJT7nYTKEJ9psGPQvE8WDTHYc1CJ3NOQJno+NH+XRD6He1O4eoT6/XOg3zXQXNgBSwcXenej6X/ac5b1vw1FYYqVIn1PRmvFDIpgu9He9XEUvrMx2HuNrvvTfY2LuaKdh/yn78fWGbpu6k/EtH2BfnPY+zATNIXlsr99+6B1pQpn+27BNdXOPYA2lKFrlxMRKpGp/T20rs3yvkfOM1chEVBE4XZxF5IJfJznRy4HMWtD2zLIp+WYPQ8tcpa+SkXgQgL25VZhq+CJBEIhxvID+3sDYjhfRsw4r6LQLSrWfvc34r0PmY9vRAvguR7hxWCXR35TudZsAyRoHYWZCCvedTSexhgxumlImIz6MXl9P2sTbBo1AiY82HHIz20EEpi+anifW/XNqJENgHgmgnJZ1N/b38qyqHY9KFTimesicHkZ2aWRCeg3SCBPRmpHcP43+5sLcil/1WgZWMM5mlA/gvPFFGbsqmoqoYCku5AP1r1YxZ3EN34NbTrexzaRLdBXOevDTmjxOiyFdwLnR9AClcyfjPx084C+g9Bm7zW0iFQJ71+0+9ctnc/NlUMJW25CQVD3oCCol0nkDfbgy+V+p6Fa8k9SXe43Vir4BcQjUvRVxvte6/dFqgN7FkYmyWloob0MLcRTa9BHCPa7eBXqErDjEe+aicyfV9j/F9eYj233a/DboLRH3wF+iTZPtxNISRaZy6Ey1i/hbSBahK1TAtsvB/0MXctBL0aEBwZg/VLS9yMNYGw9rsJ5hYrx+iJmjqfw052ChMVBiJ/u7s9PD/Zxig3YFLSm101kP9hw/Dxam6/FNmx2PcW72pZBPk3H7H14kbP0NYqcpduTiIDuCdgE/FaEc5HdRhEEsDfy/TrNCGMWRYnBmE/PqojJTEQL2e1oB38iEW2dnZtBYdbeCPmNXGL9PkdF0lyklcsjVDdEC9D2SAt1PwnfHMPxfPt/LyT4XE4hsKQY84nANfb/dojBHwl8CS3+qRyynUT0d1IWNSlUpia5fdPrvN/jkdBQJ+q5E5w7gZ3YLs4eTCwgaT87H8uw8QSF79/ZdPdLrto4hGjkBOSDWpUxopMgqocpXEFy94jNkVC7dwXOr6GF810kWPmV3RbG/NYDcKkgqMlUB0F1Uu63CjYVnLNtu3hTCOlbeuN9EPKfnImyqMR8z/smYO832GA+S+8ZuyK+lQfA/hW5F+xMPGLbVfR7LAE3kcBz9kc+gkOR8uL/oY1MMBjJo91UGesjO4BNlcA+DOPJhMtB50JdaG1Lwd5SAds2znbPxRQlSC9CWsfj7TtFNw/IeuunXVsF8aJnKTTfKd5zGGZSt99LI7kgz55wHZGCKnQog3xajtnfQY2cpbMDti48MpV8CWMmaNfq+1V+MvEifeTauV0osgjciRalS/FSqQRgzwL+iPwuPwB28q7dCuxa8X59UUqN6Ua4u3rXLsIzxQXG5XK6+nGuhJeeo6LfjZDAfCdaxA7yrl1T8c5tZxNAi/eS3ju0Uha1E6FyXQp/oHyXujPyQ5pCOtXK1rRfynVbbFOC3DBagV2TwmWifys4l55zES0EJCG/SV8QHI7m3veBDWr22XbGCLvvUoNvKYgKaWKnls4NQG4o3yaSqgWZyM+hEJZOQGa/l5AAcz2BakLeO22EBPfaQVDePWNov9zvWBQgcz7iP63Azk8bwVuBcfsm2mgdamO8TWysSrCLdQA7DgkKo5Hw87A94+4ETC5EL4FcnVrq16ONddCmvS+yBDxitPNcAnYVijLWG9NaGWu/BPaEFmEnUNS2b6kctNFAu7Cfa/d97Z6V0Lo9CM29PCBxfrTGpTLOhKr3XUgiTZx336bIyjKAYp2YigIylyMdv9K2DPJpOuZsZzW1mz0N2wo8hcYsZ0DPUqqUEIEbgXZi+yNzzDKI4eUEntLAnYIEPN9Z/Lma/Q5Ai8JDSDuzhZ1/OZ/IFX1/Um/aJtQsCu1najc4AS28JyPheayHd7RfYxYvIXPGrdTMBlAxBnXKoq5NDwhopWeuiDYT0XyYneDs3VsO7kvCks7xW4kzEnLyHfmaSLh7AWk6KwOSKPyOfWf+01A+zjrvm2eMaDfjxDIo+OwFJAzXDaJaEpn2P0KLSL6ozIesAFHNGxKuyv7je6E8nP+kolgD0kquSo0gqAj8UkbPS3rnKsv9ovm/EhIalqgDS9cI8Tx4a/E6eBtt+TgujLRDf6B7mq5QoJyf2HwRtFj/axVsjqv3/0pGj5X9Bp4z1Gjq963C2j2b2/j+zoPP3bbqlC72FSotlbHuENafz7XKQVOsZ31ahY30W4kzWkevR1rZNyhiVgahNbIb/8ALKMrxplgnRiJeck4VroHnroI2Hc/VfWfalEE+DUevI9CrLy9hbgawsv3uW7q+E/HUEF1g7dyqyFfm2hr9fh0Lrghc35lEJLDXd56XcgTSpl2JhI2HiWtkBiB/19g7r0HErEGRBWHF0vmz0KLyFPDNBN7lvnYzBjGN6mwChyHB/VpKWk1M29AmDSQFNGSOnIRK3430+suZRVTrh8x4Z+MJN961qlKudwATItdc6n3RYhkVgKhOhP8wpQT/SDO0jdFmEK/Qd/bGKS8tmPSXC3zXfamX9eFAVADjSorKYEujQJMdqA6i8qOdJyDB9N/Q4jkDuC1Bm/1LOPu/TwNmRHDez66fBizljx2JICgP/hikXZsUuLYO6XK/04mXgVyzAvY24vXOV8Mip1ugra2AKeXvEIDNA+WGeOcOpdhAp2DzILnrKfy398OsJqStDTltTQUOsHPbY/l3U/3a9Z2QVvAkzAqGTMi3VPXtz6MSjTqj01cSMNvY+OxB16DVOrB7G33tSveNwZYkykEjl41Y2r3NK2CvJ+L6gLSuqRytDo/P2hx4FAWyXo82Lzfk41mCPYYioGhU5PnB2AY7dzpyt+tmYbS59svIM9uWQT6NR68j0Gsvno7q74ucro+g0PZFMwLQdde5JYUmLlTytAzbp9TvYMQou/VbA++88sQwCkHABWAPoZQMGZmD+/h9RmBDGRRWNcb1BQqhLSQw3GNM4Swbp37IpHQ76Yh+P9gkT4L/hjGRWg7noe9AhVCJNFzPIsf2S2lhh4yiNZ9DWuQ7bWy2QgJWtCKKwa4C/C+KBv4RsL+d75aNIQC7B0Vakbx062HIrJvs12C+SJFsejDaNNyHBOuqcpvljUq3ajMkzF5leO+8n/Uh91P0aXMvtBHcD/nd/goFrEQF/tLz90eL0YN0LUm6DOIB44hHfOewMwmUJ0YBit1K/gZo69wS7IJEoqYD9HUXEjA+oS/DPffjLc9Fn75+TJGaaV/7uxLxUsE+fQ20Zx1q49AfmdJXiYyVT1uDkFb2XiSo5VqhWElnP1DuPcIZNmIas3Ig5faIBx1KKbdvANanrSORz+0nVqAUbD6mdv91yJqUR6zvQZHOJ9b3bnavn9bPUWhTU984x/seZInKKzTlfHzZBOzeRpvXIbeIzyMXp80RTX+OeDno9ZBf7A9RgFBe7jJPW7Z2AnYP4Fn7vx/STG6J6DpPIRks223n8ij1p+iaKWIEmsdjCMxjugYUXY8sjmegTcR4f+wj3ylWPnsimovrYEoUum7W25ZBPq1HryPQ2wfpqPwq1boPu1NP9dsi/CtYepo50Xep31qwaEH/GYra3R9FXz5ARYUgDz4UbLIZ2nlWBZsko58rYF+kSEa+ClpY8mCboaSrQT1NYSq6y8brYmRyOqOiX2d474A03+8B/wH8qQbOz1P4Mp+IzEXXIqG4UjtpMJfb/5cgTf6eyFfuu8SjalcAfo52+/vS3c0gGTxWBV+B85PeWA+0cT4dCbHRYBGv3x8hreAFSOCoTB+UgF2jdE/sfatoK6hhmpfpK0FbjxhtpXyxQ4Fyy7ZAH9EguTZp62tVtGUwsyiEsoUp/FZvpLrM8FvIR//39pyNvGsLELGw2fUXKDYVG9rvPDBncGrskEC3if2flzn9mh371HjnU5BAvCeKzH8f+FkNuGewOAhkXXqYovT1XhWwIaHyy2jTMbICtiqgqMrdpq3y2R582zLIp+3odQTmloNwVP42BKLyE7Cve7Db0n42gVr9JuC366DvVmFfRzvTUUSyICDN5lXe7y2RZmQaWhSqNG87+fB2rjLYxO6rin7eJQK3HBJkfb+64yjMr1OICNTGFPMExg6ZGpe13+sjYTyKs923LGLQeVDOuyjw5GNMwxCAWQZpNs5D2oafIi1DP5RS7EEi5ijvGSMRUx2NApp8TcO9RFI4IQHhW0jrdTfSrByFBc4ZfXUrKNAC/NFEKpSgxeMi7/cLKOfpYWgRDaaX8fr9ivf7COAm7/fuxMuEVsHuQTg5d05bw71zZdpKlaCtQ1/RVEB236gEfQXTY9Wkr9RcTNHWPQna6ihQjs7KKrdNW3b/4ZSCS5Hv67GolXVKAAAMNklEQVTWd7A2OdLqXe/9/jLylX3bxvpS4OwI7IpISFrEO3cFhWvDJZg/YwB2PF6ie+RPPQZtkvZDAmKVgLYWEspWtN8/Qr7UPyGSIQRt7j608V7L6HE04tH7W7+pmvcpofJ54i4mHQc1o/RfZxMun/1tauYUpkNZ4NNw9DoCc9NBL2UE6AR2XsAb7dCfQZqBcUgQzbUsX8NLph3ow6Ek2S0Hm1BEP+em+XL083VY/sEI/DJ4/kxoAZuJFsi3SPtpLuiN0Yala+9QL0XMBsh/9QuY3xUyeUUZFAoImoS0CzeWrr2d6tfGui9awH9u73iEXVsMaSCCiwISJLe2bzIKaUhuQybHR0mU2+wUnmIB/giZRO/wrr1GvGRmnvZnTe/3QkgbNwAJo0/1NKzdP4quAUy1acvuz10XWqYvCj/VDdugrxUN5pE26Ku/0dYvWqGtfC77uNv/tQLlkAvCK7RXGnkVJNC0RFvePaOR6fp55Crl+zi+VzHWS5THEwktH6GNQ4r/rISnHEC87JtIuHwnBEthQs7Hej5KwTT23eooaXZBG8oNgXfs3A4klB0ofuFaG5e7StfejPVLD0ap00ZAEZqzPzZarF0+O/KsjmSBef3odQTm1iM1ceZW2LkZb+T7ORWZQKZSBBa8gpnEavSxEVq0/w2ZOJLBJnZ+4TJutBD97MHkflsnoRyzkyv6jZlqz6XIz5eKfO6HTGtTkE/W8VXPLj1jCUwTZL/PoQiiqOOusDJFovOpaBNxRQvw/VFQ0SZIK7V1XdzbhUe+zGtQbFjOA+6pwpmieky+IN+KNBSPYPlDE9+zZVi6Bj05D9+Ta9JWKgAnSV90DYwZitK/1aKvEt5L4SVVr6KvUr+fQ8JdLdoq9ftJcA8SmKaRcBUo9TsQadsqg+QCsGsgjVct2rLv6kdt74FMuN9DG63pwH0heLpHfPehqyB7OnBvHdj8ve3vV1AGmCB9GWwwD2vVN/bp2H4vimq4/5GSf2TguV36RZu1lUv9Jnlm6Xm1hEo6DCgKfKfByC3lKaTBTpbPrvEeHckC8+LR6wg0x6f3oIjKH0vhlL4ARTqNA4C3EvCxIJfKYJMInL/ApKKfo8E5SFD7h8fEysJGGdbvc7gxx1iJv27vixZdf7cfExZSxQgWQu4KwdKmFWM9AtgRaZvnK8N733h8+X3t9/Z41cQSNNIyfMU7L4CifUdFxnoA0r6GqietDPyJeN7ATmBTQVBJ2irBzwjAV9FXN1h7lzr0lcM+gBf45dHX4zH6KsH65W+XTNFW1ft696Tq0n8Lmcb9YLU6pZFT75ukLTuXB9g86d23ADKJn4c21wtE+s5hH6fkYoP46IUUbgdl2Gi0OErr9zGFKT1VurIMuxjycYzRlg+bF4NZCjg5NkcDsF2CkbwxeyWBc1tCJQU/bzugqIS3v/nfGmlLo+WzmyNCC72NQHN8Oo/ShC9H5ee1m9emCBgoC4W1glxiE520UJmKfq7sl0j9+ZqwsdrvUViqS5vW6bd/qN9OxrqCqfezMR+LmZQD49U2fM13TmqvkQm13G+uAbucopZ9KMNGO7B1gqAGxXCvCZ+irzLsWqF7a/a7ZuTd6/S7TqCPGF1G35dCU1oXNhkoFoH9ij9W3j2pyPpQgM1pKGp7VEXflRHfIVqvAZvXZB8VmYsx2Ilev8Mj3zgGuw+FS0tsLsaCkfz3XSjSb08IlW0FFEXwPssf6+Zo/eh1BJrj0314E/7V8oSvgKsKcvnk/wBsVcS3KzO3Fvo9jnikeRXsMcTrN7cdFFQD9thYv52Odekbd5oxojZ8J2Ndp9/ZAUt1ENTeZVptEX4v4rXn6wRvBdOCzYF+2w0a260D2D06eN89Y+9r10MBNqdTBNgs1SJstxKSbcA+10G/z5AoY5143xupiFLv5H1rzsVapnJaDCiqwPvZ1Fg3R+I79DYCzfHZOCITPhiVb/d3EuTSScR2J/3Oc7A9Ad/uN+4EvqdwTvS7Fa1l2EjC0nkQVK8EYDWwLX+ntgNs5gBssJjGHIDt8fdtYS62krGmbpBuU+pzNh29jkBzfHaOuhM+Als7yIWeFVbaDs6ZF2F7oO9eyRjRA+88RzJV0EEAVafwDeycgS09p+0ykA1s6yUz6aEodVoIKOoJvJvDxrK3EWiOz+YRm/B0GCQTeWalsNJJv/Mi7Owa6zrfuBP42Y1zp3gncG4rCKpT+AZ2zsB68D1SirqB7bxkJrMhSn1O4P1ZPvrQtKb1Qsuy7L/K55xzLsuy/4eqtmxi9/3TrvVzzvVBybiPtnN9SvADnHMznXPjcxh7xj+yLPsNMAT4MMuyx+181mm/8yJsT8DXaaFv3An8nMA51G8P4fwEMuX7dDcgy7IPkIlvSgjnTuAb2DkDW4J/DFWRI8uy/7NrfZ1zi6IsCBcn+m5gK2BbaZ3yn3KbU3h/pltvS8TN0Rzlg/aCXDqK+G6333kZtifg5xX66O0jhTMVAVSdwjewcwa2Bnwrpagb2Iqx7q1jXsV7XjjyRbxpTZvrmnPuYBTh2gclsn4cVU55KYvsfp1zS6KkxCsAl2RZ9vCc6Hdehu0J+N5onyKcxwLfy7Lsj7MTvoGdM7AJ+BWB19rsu4Gdy9q8ivfc3BqBtGlzdXPODUB+gYcDvwL+E/gdqvMcJd4eENDa6ndehe0J+N5on0Wc50X6+qzBzqt4z4uwvdnmVbzn1tYIpE2bZ5pzbqFWNF89xSxa7Xdeh+0J+N5on0Wc50X6+qzB9mbfnzXY3mzzKt5zU2sE0qZ9JlrDLJrWtKY1rWlNm3tbI5A2rWlNa1rTmta0pjWtV1uTlqBpTWta05rWtKY1rWm92hqBtGlNa1rTmta0pjWtab3aGoG0aU1rWtOa1rSmNa1pvdoagbRpTWta05rWtKY1rWm92hqBtGlNa9pntjnnRjnn3g+c/5pzbuXewKncnHO7pHBxzh3lnDuwzWc/65xbq33smta0pjWtZ1q/3kagaU1rWtPmtpZl2WG9jYPXdgEeAT4oX3DO9cuy7KY5j1LTmta0pvVsazSkTWta0z7rrZ9z7g7n3LvOuQecc4N9zaFz7i/OuYudc+845151zi1h5/d0zr1v55+PPdw5N9E595Bz7mHn3IfOueOcc6c4596y5y1i9x3unPuePe9Bw2MDYCfgcufc28655Q23yc6554ATnXPnO+dOtWeMcc49Zc940+7fxDn3iIfPdc65iQE8b3TOfd859wPn3AU9OcBNa1rTmlbVGoG0aU1r2me9jQNuybLs88CfgGNK14cAr2ZZtirwPKr8BXAusLWd36mij/HAvsA6wMXA37IsWx14BcjN7d/Ismxte94PgUOzLHsZ+DYwKcuy1bIs+7ndu1CWZROyLPtqqZ97UCWyVYENUGWyuu2sLMvWAj4PTHDOfb4F2KY1rWlN66g1AmnTmta0z3r7dZZlL9n/dwMbla7/P2QyB3gDGGX/vwRMd84dDvSt6GNWlmV/zrLsD8B/Aw/b+fe85413zr3gnHsP2A9YJfG8GeUTzrmhwFJZln0TIMuy/82y7G8VePltL+fcm8Bb1vdc4UPbtKY17bPRGoG0aU1r2me9lcvVlX//IytK2v0f5nufZdlRwNnAMsDbzrlFE3383fv/Y+/3xxS+/NOB47Is+xxwATBf4nl/DZxzkXv/SVde3+25zrnRwKnA5qYp/k5F/01rWtOa1qOtEUib1rSmfdbbSOfc+vb/PsCLdYCcc8tnWfZalmXnAv+OBNNO2lDgd865/khDmrc/27Vky7LsT8BvnHO7GH4DnXODgV8BK9vvBYHNA+ALICH3v81HdtvOXqVpTWta01prjUDatKY17bPefggc5Jx7F1gEuLEm3OXOufcsbdTzwDsd4nEO8BrwJPAj7/z9wCQLglq+4hkHACfYu7wMDM+y7NfATOBd5GP6Vhkoy7J37PwPgNuRO0LTmta0ps2x5gpLVNOa1rSmNa1pTWta05o251ujIW1a05rWtKY1rWlNa1qvtiYxftOa1rSm9UBzzm0NTCmd/jDLsl17A5+mNa1pTZuXWmOyb1rTmta0pjWtaU1rWq+2xmTftKY1rWlNa1rTmta0Xm2NQNq0pjWtaU1rWtOa1rRebY1A2rSmNa1pTWta05rWtF5tjUDatKY1rWlNa1rTmta0Xm3/HxmVmvRj7eKuAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "df_grafico[\"bins_matricula\"] = pd.cut(df_grafico.matricula,bins=41)\n",
    "sns.catplot(x=\"bins_matricula\", y=\"nivel_desemp_matematica\", kind=\"bar\", height=10,data=df_grafico)\n",
    "plt.xticks(rotation=65, horizontalalignment='right')\n",
    "plt.ylim((0,1))\n",
    "plt.title(\"Desempeño en matemática y cantidad de matriculados por escuela\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [],
   "source": [
    "indices_bins = df_grafico.groupby(\"bins_matricula\")[\"indice_socioeconomico_medio\"].mean().sort_values(ascending=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5,1,'Desempeño en matemática y cantidad de matriculados por escuela')"
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.catplot(x=\"bins_matricula\", y=\"nivel_desemp_matematica\",order=indices_bins.index, kind=\"bar\", height=10,data=df_grafico)\n",
    "plt.xticks(rotation=65, horizontalalignment='right')\n",
    "plt.ylim((0,1))\n",
    "plt.title(\"Desempeño en matemática y cantidad de matriculados por escuela\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "bienes = pd.get_dummies(df[['tiene_notebook', 'tiene_pc', 'tiene_tablet', \\\n",
    "       'tiene_celular', 'tiene_smartphone', 'tiene_consola', 'tiene_smarttv', \\\n",
    "       'tiene_cable', 'tiene_internet_x']],drop_first=True)\n",
    "cantidad_bienes = bienes.sum(axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5,1,'Desempeño en matemática y cantidad de bienes')"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.catplot(x=cantidad_bienes, y=\"nivel_desemp_matematica\", kind=\"bar\", data=df)\n",
    "plt.ylim((0,1))\n",
    "plt.xlabel(\"cantidad de bienes\")\n",
    "plt.title(\"Desempeño en matemática y cantidad de bienes\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "bienes\n",
       "0    0.521058\n",
       "1    0.608582\n",
       "2    0.496834\n",
       "3    0.455388\n",
       "4    0.402621\n",
       "5    0.355341\n",
       "6    0.311035\n",
       "7    0.270363\n",
       "8    0.246271\n",
       "9    0.232992\n",
       "Name: nivel_desemp_matematica, dtype: float64"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Para ver los valores del grafico\n",
    "df[\"bienes\"] = cantidad_bienes\n",
    "df.groupby(\"bienes\")[\"nivel_desemp_matematica\"].mean()"
   ]
  }
 ],
 "metadata": {
  "colab": {
   "name": "limpiar_dataset.ipynb",
   "provenance": []
  },
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}