{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### LIME for income prediction"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "LIME is a model-agnostic explainer for tabular data. If using this explainer, please cite the original work: https://github.com/marcotcr/lime."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "# This default renderer is used for sphinx docs only. Please delete this cell in IPython.\n",
    "import plotly.io as pio\n",
    "pio.renderers.default = \"png\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "import sklearn\n",
    "import xgboost\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "from omnixai.data.tabular import Tabular\n",
    "from omnixai.preprocessing.tabular import TabularTransform\n",
    "from omnixai.explainers.tabular import LimeTabular"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The dataset used in this example is for income prediction (https://archive.ics.uci.edu/ml/datasets/adult). We recommend using `Tabular` to represent a tabular dataset, which can be constructed from a pandas dataframe or a numpy array. To create a `Tabular` instance given a numpy array, one needs to specify the data, the feature names, the categorical feature names (if exists) and the target/label column name (if exists)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "      Age         Workclass  fnlwgt   Education Education-Num  \\\n",
      "0      39         State-gov   77516   Bachelors            13   \n",
      "1      50  Self-emp-not-inc   83311   Bachelors            13   \n",
      "2      38           Private  215646     HS-grad             9   \n",
      "3      53           Private  234721        11th             7   \n",
      "4      28           Private  338409   Bachelors            13   \n",
      "...    ..               ...     ...         ...           ...   \n",
      "32556  27           Private  257302  Assoc-acdm            12   \n",
      "32557  40           Private  154374     HS-grad             9   \n",
      "32558  58           Private  151910     HS-grad             9   \n",
      "32559  22           Private  201490     HS-grad             9   \n",
      "32560  52      Self-emp-inc  287927     HS-grad             9   \n",
      "\n",
      "           Marital Status         Occupation   Relationship   Race     Sex  \\\n",
      "0           Never-married       Adm-clerical  Not-in-family  White    Male   \n",
      "1      Married-civ-spouse    Exec-managerial        Husband  White    Male   \n",
      "2                Divorced  Handlers-cleaners  Not-in-family  White    Male   \n",
      "3      Married-civ-spouse  Handlers-cleaners        Husband  Black    Male   \n",
      "4      Married-civ-spouse     Prof-specialty           Wife  Black  Female   \n",
      "...                   ...                ...            ...    ...     ...   \n",
      "32556  Married-civ-spouse       Tech-support           Wife  White  Female   \n",
      "32557  Married-civ-spouse  Machine-op-inspct        Husband  White    Male   \n",
      "32558             Widowed       Adm-clerical      Unmarried  White  Female   \n",
      "32559       Never-married       Adm-clerical      Own-child  White    Male   \n",
      "32560  Married-civ-spouse    Exec-managerial           Wife  White  Female   \n",
      "\n",
      "      Capital Gain Capital Loss Hours per week        Country  label  \n",
      "0             2174            0             40  United-States  <=50K  \n",
      "1                0            0             13  United-States  <=50K  \n",
      "2                0            0             40  United-States  <=50K  \n",
      "3                0            0             40  United-States  <=50K  \n",
      "4                0            0             40           Cuba  <=50K  \n",
      "...            ...          ...            ...            ...    ...  \n",
      "32556            0            0             38  United-States  <=50K  \n",
      "32557            0            0             40  United-States   >50K  \n",
      "32558            0            0             40  United-States  <=50K  \n",
      "32559            0            0             20  United-States  <=50K  \n",
      "32560        15024            0             40  United-States   >50K  \n",
      "\n",
      "[32561 rows x 15 columns]\n"
     ]
    }
   ],
   "source": [
    "feature_names = [\n",
    "    \"Age\", \"Workclass\", \"fnlwgt\", \"Education\",\n",
    "    \"Education-Num\", \"Marital Status\", \"Occupation\",\n",
    "    \"Relationship\", \"Race\", \"Sex\", \"Capital Gain\",\n",
    "    \"Capital Loss\", \"Hours per week\", \"Country\", \"label\"\n",
    "]\n",
    "data = np.genfromtxt(os.path.join('../data', 'adult.data'), delimiter=', ', dtype=str)\n",
    "tabular_data = Tabular(\n",
    "    data,\n",
    "    feature_columns=feature_names,\n",
    "    categorical_columns=[feature_names[i] for i in [1, 3, 5, 6, 7, 8, 9, 13]],\n",
    "    target_column='label'\n",
    ")\n",
    "print(tabular_data)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "`TabularTransform` is a special transform designed for tabular data. By default, it converts categorical features into one-hot encoding, and keeps continuous-valued features (if one wants to normalize continuous-valued features, set the parameter `cont_transform` in `TabularTransform` to `Standard` or `MinMax`). The `transform` method of `TabularTransform` will transform a `Tabular` instance into a numpy array. If the `Tabular` instance has a target/label column, the last column of the transformed numpy array will be the target/label. \n",
    "\n",
    "If some other transformations that are not supported in the library are necessary, one can simply convert the `Tabular` instance into a pandas dataframe by calling `Tabular.to_pd()` and try different transformations with it.\n",
    "\n",
    "After data preprocessing, we can train a XGBoost classifier for this task (one may try other classifiers). "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training data shape: (26048, 108)\n",
      "Test data shape:     (6513, 108)\n",
      "Test accuracy: 0.8668816213726394\n"
     ]
    }
   ],
   "source": [
    "np.random.seed(1)\n",
    "transformer = TabularTransform().fit(tabular_data)\n",
    "class_names = transformer.class_names\n",
    "x = transformer.transform(tabular_data)\n",
    "train, test, labels_train, labels_test = \\\n",
    "    sklearn.model_selection.train_test_split(x[:, :-1], x[:, -1], train_size=0.80)\n",
    "print('Training data shape: {}'.format(train.shape))\n",
    "print('Test data shape:     {}'.format(test.shape))\n",
    "\n",
    "gbtree = xgboost.XGBClassifier(n_estimators=300, max_depth=5)\n",
    "gbtree.fit(train, labels_train)\n",
    "print('Test accuracy: {}'.format(\n",
    "    sklearn.metrics.accuracy_score(labels_test, gbtree.predict(test))))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The prediction function takes a `Tabular` instance as its inputs, and outputs the class probabilities for classification tasks or the estimated values for regression tasks. In this example, we simply call `transformer.transform` to do data preprocessing followed by the prediction function of `gbtree`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "predict_function=lambda z: gbtree.predict_proba(transformer.transform(z))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "To initialize a LIME explainer, we need to set:\n",
    "  \n",
    "  - `training_data`: The data used to train local explainers in LIME. ``training_data`` can be the training dataset for training the machine learning model. If the training dataset is too large, ``training_data`` can be a subset of it by applying `omnixai.sampler.tabular.Sampler.subsample`.\n",
    "  - `predict_function`: The prediction function corresponding to the model.\n",
    "  - `mode`: The task type, e.g., \"classification\" or \"regression\"."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "explainer = LimeTabular(\n",
    "    training_data=tabular_data,\n",
    "    predict_function=predict_function\n",
    ")\n",
    "# Apply an inverse transform, i.e., converting the numpy array back to `Tabular`\n",
    "test_instances = transformer.invert(test)\n",
    "test_x = test_instances[1653:1655]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "LIME generates local explanations, e.g. `explainer.explain` is called given the test instances. `ipython_plot` plots the generated explanations in IPython. Parameter `index` indicates which instance in `test_x` to plot, e.g., `index = 0` means plotting the first instance in `test_x`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAArwAAAH0CAYAAADfWf7fAAAgAElEQVR4XuydBXQVV9eG3yTEQ4J7KV6g0GIFSrEWK+7uTnDX4BqCFwvu7u5epEgpVkqRFneXKPnWPjT3S4hMEu6Euck7a33/+ntzZs+ZZ5+QZ87sc65VUFBQEHiQAAmQAAmQAAmQAAmQQBwlYEXhjaOZ5W2RAAmQAAmQAAmQAAkoAhReDgQSIAESIAESIAESIIE4TYDCG6fTy5sjARIgARIgARIgARKg8HIMkAAJkAAJkAAJkAAJxGkCFN44nV7eHAmQAAmQAAmQAAmQAIWXY4AESIAESIAESIAESCBOE6Dwxun08uZIgARIgARIgARIgAQovBwDJEACJEACJEACJEACcZoAhTdOp5c3RwIkQAIkQAIkQAIkQOHlGCABEiABEiABEiABEojTBCi8cTq9vDkSIAESIAESIAESIAEKL8cACZAACZAACZAACZBAnCZA4Y3T6eXNkQAJkAAJkAAJkAAJUHg5BkiABEiABEiABEiABOI0AQpvnE4vb44ESIAESIAESIAESIDCyzFAAiRAAiRAAiRAAiQQpwlQeON0enlzJEACJEACJEACJEACFF6OARIgARIgARIgARIggThNgMIbp9PLmyMBEiABEiABEiABEqDwcgyQAAmQAAmQAAmQAAnEaQIU3jidXt4cCZAACZAACZAACZAAhZdjgARIgARIgARIgARIIE4ToPDG6fTy5kiABEiABEiABEiABCi8HAMkQAIkQAIkQAIkQAJxmgCFN06nlzdHAiRAAiRAAiRAAiRA4eUYIAESIAESIAESIAESiNMEKLxxOr28ORIgARIgARIgARIgAQovxwAJkAAJkAAJkAAJkECcJkDhjdPp5c2RAAmQAAmQAAmQAAlQeDkGSIAESIAESIAESIAE4jQBCm+cTi9vjgRIgARIgARIgARIgMLLMUACJEACJEACJEACJBCnCVB443R6eXMkQAIkQAIkQAIkQAIUXo4BEiABEiABEiABEiCBOE2Awhun08ubIwESIAESIAESIAESoPByDJAACZAACZAACZAACcRpAhTeOJ1e3hwJkAAJkAAJkAAJkIDhhHfngZPoPmQajmz8BYndEjJDHxG49/ApvKYvx9FTFxEQEIA8ubKib4cGyJIxbYxY3XvwBEvW7cbBY3/g/sMnsLa2RqrkSVDku1zo1KIGEiSwQb6yreE10B0VShWK0TVi+6QFq3bAa/qKMJetVPp7eHq0NX3u6+ePKXPXYuue43j+8jUyf5kG7ZtWQ6li+Uxtnr14haJVO2HCkA4oV/I70+dBQUHoP3oOjp66gCVTB+CLNCli+zZ5PRIgARIgARIggSgSiHfC27DDCAzs1gTZs6SPIiLjNBNBq9bcA0kSJUTnVjXh5GCPRWt2KvndMG8EkidNFK3OHj99CV0HT0XyJG6oX700smRIi3c+vrhy/Rau/nMHngPaQq5pacI7afYaHD5xDgO6NArFI0kiV2T4IpXpsz4jvSEM+nduhC/TpVTSP23Bekwf3Q1FC+ZW7SISXs9py7FxxxEsmtI/xg8b0UoWG5MACZAACZAACcSYQLwSXpkdLVO3B9bMHmqRwrt6ywGMnrIUe1ZNUNIrR2Dge1RrPgBFC32DPh3qR3kgPHryHFWbDUCu7Bnxy8gusLezDfdcowjvrbsPMXbacrRtUgW5vsoY6X0OGbcAT1+8xJThnSNs98+t+6jYuK+S2xLff2tqJ+eev3wda+cMi1B4Zy7ahLnLt2LehD7InSNTlJmzIQmQAAmQAAmQwOchYBHCW899GCqWKozXb99hw/YjePD4GVImS4yqPxdFu8ZVYG1tpei9ev0WE2evwYGjv+Pp81dwS+iM3NkzoVf7ejj/53UMHjcfPr5+JtKZM6TFpgUj1X+v23YI67cfwfWbd+Hn569eb3dpVQvfF/ja1D6q/ZATRCgnz1mrZhpfvHyNZEncULvyj2jbuLKK9/jpC4yfuQoHj5/FOx8/fJ0tg+rntzkzRzgSug6aCh9fX8z07BGqjVxn297j2LncS30+buZKrNt6SIl9mlTJwo33y7x1mLt8G/asHK/6FtERnvC+feeL+Su2Ye+RMxARtbVNgDxfZ0Hfjg2RPu2HV/uR5UJmU6UkYP7K7Vi37TDu3n8MB3s7ZM6QBq0bVkLxwv8X0OB+ST/mLN2CZRv2ImfWDGjXpAryf5Mt3G4LJzdXZwzt2TzC+1q6bg8mzV6N41tmwMbG2tTu2KmLaNXTSz1UpE6RJMwM78qN+zBm2nLMGNMNhfPl/Dy/tbwqCZAACZAACZBAtAhYjPD+ff02KpQqjPZNqypBO37mEjp5TEHv9vXRoHopddO9hs/AjZv31UyntHnw6Jlq16ZRZdgmsMHFK/9AShqWTvNAtkzpVL2qiFawJKZImggFvv1KfS6vxU+fu4LdK8YhkZuLaiPCG5V+vH7zDjVbDVIiJTWh6dIkx607D5E8WSIlSSKMtVoPUrOqIrlSq7xs/V4lrVsWjUbqlEnDTWLV5gPwff6v0bdjg1A/37zrKPqOmoXTO2ep+5H/f8vuY9i4YKQS9/COWq0Hq1niWV49Ix0w4QlvQGAgeg6dgdLF8yNH1i/x5OlLDPKah6SJXRVbrVw4Odpj1eYDmOi9CkN7tVC5ePHqDU6f+0vdn8SM6JCHA3k4WbhqB1KlSKLEt0iBXKGaN+0yGnlzZUXX1rUijDNy8mKcOf+3aSY3uOH9R09RqnZ3eI/tocoaQpY0iKT3Hemt6nl/Kvr/Ot9o/caxMQmQAAmQAAmQQKwTsBjhffzkOXYs90ICGxsTpO5DpuPtOx/M9OyuPqvSbAB++C5XhK/2L1+9qURUXldr1fAGv/KeO6G3aSZPhDcq/Zi/Yjsmz12LHcvGqgVgHx+LVu/EBO9V2LbE0zQD+/59ECo27qOuNbhHs3AHwk+1u6FmheLo0Lx6qJ8fOHoWHfpPwoG1k1Qdr8R6+vxlpDO3P1TtiPI/FoJH18bRFt7wThCBHT5xIc7snK1mfLVyMWLSYpy9eBWrZw2BldWHGfroHFLKsePAb5i3fJt6mGnfrJppZljqnEVU5fPXb31UDooWyo2W9SuYFkJK/e79h0+xcHK/UJeV8fRd+XYYO7CdeqsQLLy1KpVQNbtpUyfH+rnDYRdBCUh07oFtSYAESIAESIAEYoeAxQivlDBMHt4pFJXRvyzFqT/+Ms3SySzpqClLULNicdSt8iNyZssQqn10hNffPwB5yrQKtTpfhDcq/XDvO1Et/lowqW+4WWzfbyKevXiN5dMHhvp5nxHeuHHrHlZ5D4lQeGtVLKHkLuSx/+jv6Nh/skl4ozJ0ClV0R5WyRTCgi3mE99eTF9Cm1zjT7hpauZASk+bdxuDbnFnQtE45/PBd7lClBVG5h+cvXmPDziOYvmADCubJgamjuqjTrt64g7sPHiNFssSqPOWv67cwe8kWNXO/bu4wODk6QIRX3gB8nKM3b31QsEJY4ZVZ6Q7NqmPSnDWQHGg9KESl/2xDAiRAAiRAAiQQOwQsRnhzZEkfZuZThPfk2ctYN3e4iZa8pl68Zif2//o7Mn2ZBh2bVze9fo5IeKWuV2ZdZab0zv3HaiGY1AU/efYyjPBGpR/12w9H2lTJMG6Qe7hZlJ9fuHw9zEIxf/9AVT5xcN3kcM+r3sIDhfLljLCk4dSOWXB0+FCioXVUbtIP6dKkULWokR3hlTTIa3+ZWRX2D588h421NeQB4eXrt6G2k4ssF3JNYb1w1U5s2X1UzQqL+Dat/bOm+J45fwUrN+6HiP6PRfKiYY3S+CaS2me51rV/7qhZ51H9WqNquR8waspSVUIRvDgtmEFwSYO8NShW6BvTDO+IPi1RvXwxSA3vsImLMHFoR5QtUUALM39OAiRAAiRAAiRgAAJxTniDmYqszl22FQtX71SLvIoVyo2IhFf2/f3j4jWM7NcK+XJlVa+rZdFV4UrtYyS8MtMp0izlEOEdbXuPV9t9DesVdlGVjY2NkuXwjh5Dp+PN23fhLlrbtPNX7F09IcpDauj4Bdi8+yj2rZkEVxenCM/7WHjlvys07KO29+rbqQEypU+jBHXfkTOqpjq8/ZPDy0XIC0rMDdsPq8V29auVQve2dcL0R2ZqZZeKlZsOqEWAMoNfp8qPkZZtfBxEZuy7tKyJ5vXKK3GV60W0aE3KUWRv3fC2JZOZ+APHzqpFgdx/N8pDjg1JgARIgARI4LMRiLPCG0xUZkVLfJ9HLWC6cfMeKjXpp74oQBY1BR8itvWq/hRqkZMsdmvZfWyMhNd78WZ4L96EXSvGhStk81ZsU6/Yty31jNaXa8jiNNlpYt/qiaaFdFKvK9uS5cudDUN6fqj9FYGUXSpkl4GIjivXb6uFc2VLfKf22w25U8HHMhpyH97ghwZZ7Cb10sGH7Pog23VF9oUhIXMRXr9k1lVmvpd9VOohbW/eeagWjDWsWUb1Wepzo3Ocu3QNMrMe/PAjO0OUrd9LzXDLTG7wIQ8Cx8/8ie1LPdVH4QmvLDqs124oHBzssHSqh5qd5kECJEACJEACJGBcAnFKeGV7rtzZMyJj+tTqG8JOnPkTIyYvxpThndSCJpklLFGjCwrnz6lmEWUWVmYqZVX/s+evMLJvK7i5uuDiXzcwb8V2/H39Fjw92pm+YUtqeKNS0iC1pTVaDURCFydVUiG7LsgCKRFRWQgldaJ12g5RfZQdJDKkS6UW313/965adBbRDgD+AYGo02YwXJyd0K1NLbUjw6I1u7D38Gn1aj592pRqpIkUr916KNJdGqSdzHIOn7QYObN9idqVSyLjF6nh5++Pu/efqO3bGtUsE+aLJ6TvJWt2UZIo38T2/v17HD5xHss37MXte49MwquVizVbDirJlh0ZZPu4q//cxdAJC9RCuh7tws7wRvVXSPZalu3LZOeGNKmSqgV8ItHTF25UXxAxZ1wv0yI5KU3Y/+sZeHRp8uGLJ47/obYq8xzQzvStchF98cS1f++ibtshqFmxBPp1ahjV7rEdCZAACZAACZDAZyAQp4R32vz12LLnGB4+fq4WKInMNq5VBlXK/mBCK/viyrdk3b77EDmyZVALx2T2cMSkRfj9wlUEBb1H3lzZ0K9zQ/QbOQst6leItvDKxeQreyfOWo1fT13A69fvkDypG1rUr2jaQk1E6pd565VwyZ68Iq+yyM69SVUl5BEd0la+gOHQiXNK4OWLI3q51wv1BQgim6s3H8Dq2UMjneWVa/xx6Zra4ku2YJM+ST9kYV7lskWUjIdXwysL1GSXies376ma4VJF8yuxL1WnOw5vmKJmrbVyIV8hPWvJZiXJfv4BSJU8Mcr/VAjuTatFe/Y2JCvZEs5rxgrIfrpSSiGzr1J2UO3noqhTuWSo2Vh5gJAFbxt2HFYPPOnTpVI7OUiNb/ARkfDKz2WsSXmDfMFFyK8j/gy/x7wkCZAACZAACZBAJAQMJ7zMFgmQAAmQAAmQAAmQAAmYkwCF15w0GYsESIAESIAESIAESMBwBCi8hksJO0QCJEACJEACJEACJGBOAhRec9JkLBIgARIgARIgARIgAcMRoPAaLiXsEAmQAAmQAAmQAAmQgDkJUHjNSZOxSIAESIAESIAESIAEDEeAwmu4lLBDJEACJEACJEACJEAC5iRA4TUnTcYiARIgARIgARIgARIwHAEKr+FSwg6RAAmQAAmQAAmQAAmYkwCF15w0GYsESIAESIAESIAESMBwBCi8hksJO0QCJEACJEACJEACJGBOAhRec9JkLBIgARIgARIgARIgAcMRoPAaLiXsEAmQAAmQAAmQAAmQgDkJUHjNSZOxSIAESIAESIAESIAEDEeAwmu4lLBDJEACJEACJEACJEAC5iRA4TUnTcYiARIgARIgARIgARIwHAEKr+FSwg6RAAmQAAmQAAmQAAmYkwCF15w0GYsESIAESIAESIAESMBwBCi8hksJO0QCJEACJEACJEACJGBOAhRec9JkLBIgARIgARIgARIgAcMRoPAaLiXsEAmQAAmQAAmQAAmQgDkJUHjNSZOxSIAESIAESIAESIAEDEeAwmu4lLBDJEACJEACJEACJEAC5iRA4TUnTcYiARIgARIgARIgARIwHAEKr+FSwg6RAAmQAAmQAAmQAAmYkwCF15w0GYsESIAESIAESIAESMBwBCi8hksJO0QCJEACJEACJEACJGBOAhRec9JkLBIgARIgARIgARIgAcMRoPAaLiXsEAmQAAmQAAmQAAmQgDkJUHjNSZOxSIAESIAESIAESIAEDEeAwmu4lLBDJEACJEACJEACJEAC5iRA4TUnTcYiARIgARIgARIgARIwHAEKr+FSwg6RAAmQAAmQAAmQAAmYkwCF15w0GYsESIAESIAESIAESMBwBCi8hksJO0QCJEACJEACJEACJGBOAhRec9JkLBIgARIgARIgARIgAcMRoPAaLiXsEAmQAAmQAAmQAAmQgDkJUHjNSZOxSIAESIAESIAESIAEDEeAwmu4lLBDJEACJEACJEACJEAC5iRA4TUnTcYiARIgARIgARIgARIwHAEKr+FSwg6RAAmQAAmQAAmQAAmYkwCF15w0GYsESIAESIAESIAESMBwBCi8hksJO0QCJEACJEACJEACJGBOAhRec9JkLBIgARIgARIgARIgAcMRoPAaLiXsEAmQAAmQAAmQAAmQgDkJUHjNSZOxSIAESIAESIAESIAEDEeAwmu4lLBDJEACJEACJEACJEAC5iRA4TUnTcYiARIgARIgARIgARIwHAEKr+FSwg6RAAmQAAmQAAmQAAmYkwCF15w0GYsESIAESIAESIAESMBwBCi8hksJO0QC4RO4++SdRaNxc7aFs0MCvHjjjzc+ARZ9L0bqvKO9DRxsbfDstZ+RumXxfUnsYgcf/0C88w20+Hsx0g2kSuKIh8/e4X2QkXoV+31Jk9Qx9i8az69I4Y3nA4C3bzkEKLyWk6vY7CmFVx/aFF59uFJ4P3Cl8OozviKLSuGNfea8IgnEiIB5hdcqRn34lJMSOiaAs6MNXr0N4Azvp4D86FwKrxlhhghF4dWHK4WXwqvPyNKOSuHVZsQWJGAIAuYS3qAr/yBw5yEgIHZf1VonSACnKsVh81VmCq8ZRxSF14wwKbz6wAwRlcJL4dV9kEVwAQrv5yLP65JANAmYS3gDz/4JvxZ9gbexXBPs5AC3RWNhWyAXhTeauY+sOYXXjDApvPrApPCG4cqSBt2HWpgLUHhjnzmvSAIxIkDhjRG2OH8ShVefFLOkQR+unOHlDK8+I0s7KoVXmxFbkIAhCFB4DZEGw3WCwqtPSii8+nCl8FJ49RlZ2lEpvNqM2IIEDEGAwmuINBiuExRefVJC4dWHK4WXwqvPyNKOSuHVZsQWJGAIAhReQ6TBcJ2g8OqTEgqvPlwpvNET3jv3H6NsvZ7qJBsbayRPkggF8+ZApxbVkSZVsk9OUsvuY9GwZhn89ENeFatu26FY6T3YFPf46UuYNGcNVswY9MnX+twBKLyfOwO8PglEkQCFN4qg4lkzCq8+Cafw6sOVwhsz4f1141RYWVnh1r2H+GXuOtx7+BQb549Qn33KsX77YeT/5iukT5sC12/eQ6OOI3B00zRTyPuPnuLg0bOoW/WnT7mMIc6l8BoiDexEdAnkKdMKZ3fPie5pFt2ewmvR6dOt8xRefdBSePXhSuGNmfCe2TUb9na26uQnz16iePXO2L50rBLVB4+eYdSUJTh66oIS4BLffwuPLk3g5uqs2p85fwVjpy3Hleu34ehoj/zfZMP4wR1gm8AGtdsMQfO65ZE5Qxq06zMeDx8/N523eeFonP/zOrxmrMDWxWNQtfkANKlVDjUrFjcNDhHmecu3YfOi0fD3D4DXjJXYsvsoYAWU/D4P+nVqiIQuTvoMpmhGpfBGE1h8bn7t37uYPGcNTv3xFwICApE+bUoM6dkMub7KGCMsQUFB+LlBb2xZNBq2tglw5Lfz6knT0cFOM15Ewlul2QCMH+yOrBnTacaIzQbnL9+Ah+ccPHryHNkzp4enR1skT5ooTBeevXiFfqNm4dyf15EkkSuG9WqBfLmzqnYU3tjMmOVci8KrT64ovPpwpfB+uvDK34miVTspCc3wRSrUcx+GbJnSoXPLmnj/PgheM5bj5au38B7bQ12sRI0u6NW+HsoWL4CnL17h8t83UbJIHvWzYOGtUKoQ/rh0De59J4Sa4T1w9KxJeGcv3YITv/+JOeN6mQZH297jlUC3aVQZE2etxtV/7mBQt6aws0uAYRMWIkECG3gNdNdnMEUzKoU3msDia/N/bz9A/fbD0LZxFVT7uSicnRzUk1+WDGk/6elN6oMK588Jkd/67YdjxphuSOyWUBOzJQlvYOB7VGjUBx5dm6BYodxYsnY3jp2+iGmjuoa5zz4jvZE2VTJ0bF4DFy5fR/eh09UDgYO9HYVXc1TEzwYUXn3yTuHVhyuF99OEV2ZgRWhv3LyPNbOH4vS5K+jYfxIOrJtsmgF+89YHJWt2wZKpHvgq8xcoVq0TBnVvijLFC4RJanSE996DJ2qSav/aSUiSKCFevHyjZpq3LhmDlMmToHBFd2xZPAapUyRR17l64w5qtBqI0ztnq9nkz31QeD93Bizk+j2HzUDqFEnRo12dcHsc/FrDx88fSRO7Ytwgd6RLnRzyRPj0+Sv8e/s+Xr95p4ruZdbyizQpVJxgcfWavkKJoLxWcU3ojAWT+uLcpWvqNc2zF69hbW2N/p0bKWEMed7HnYlohleuLbFkdlr6ULtySbSoV0GdfvLsZXhOW4436osYrNC/c0MUK/QNlq3fi2Xr98DPz19J/bTRXZEq+YdfZDl++/0yhk5YEIaHPP2mTpnU9Lk8GIz+ZSmWTR+oPpMn8BI1OmPbEs9QDwvyeZEqHbB/zSTTLHcnjymoWaG4ehrnDK+F/LLEcjcpvPoAp/Dqw5XCGzPhlfIEX19/WFtboVzJgujauhaSJXHDqs0HsHHHESyd5hEqYTVbDUKL+hVQsVRhHD5xHv1Hz0L2LF+iUc0yKF74G1Ptb3SEVy7QuNMoVCpdWNX0rtlyEBt2HMGSqQMgk2IysSNva0MegYGB2LVivEmC9RlVUYtK4Y0ap3jfSl6fzBnfC9mzpA+XxV/XbiFViiRwS+iM8TNX4Z2PLzy6NlbCK7+Q6+YMU3K3Y/9vWLpuDxb/0j+MuP5YqyvWzR1umuGVJ1kfX19VOnH4xDlVG7RpwcgYCe8gr3lKpHu0rYO373zRptc4NKldVv3DUav1YIzu31qVQbx49Ub9QyDLAKo198C2pZ7qqVlWysrMa0yOLbuPqRndkX1bmU6XV1ADujRG7uz/LweROqxGnUZi94pxpnYTvFchkZuLknMKb0zox/1zKLz65JjCqw9XCm/MhFfqdeXva3BdbnB2lq7bjW17T0QqvNLWx9cP8rdo7vKt+DJdKvWGUU3+/FfDG5WSBomzatN+bNt3Qk1KterphdLF8qNe1Z/UgrfKTfrh5HZvODna6zN4PjEqhfcTAcaX078u2QxHNv4SpXIDKZwXqZVfKBHeh4+fKbmTQ0oXilTugB3LvdQvb8jShI+FNyRb/4BAdd7J7TNjJLwFK7TDnlUT4Ppf8fzuQ6ewaeev+GVkF/QaPgOJXF3Qvlk10/0FlyG0bFAR1X8uGuapNTp5X73lAP78+yYGdWtiOq1Z1zFwb1oVhfLmMH0mT8jyakqK/4OP6Qs2qBnhji2qU3ijAz0etaXw6pNsCq8+XCm8MRPekIvWQmZG3lp2GjA53JKGRVP6I0fWL0Ml8p2PH8rV74kpIzojz9dZQgmvrDVp3dMLx7dMN50TsoZXPpQyBvlbvXbOMMgs8t7VE9TfTVmwVrCiO6aO7IIfvsulz+D5xKgU3k8EGF9OL1TRXb22iGgxmEjdnkOnldDKL0SSxK6qHleEV2ZUu7SqaUJVvmEfVR6QKX3qSIVXiuPl6dXHx0+dKytNT+2YFW3hlXomqWGSfzCCDymXGDF5MVZ5D1H9W7ByO9ZtO4QS3+dRZRtOjg64e/8xZizahGOnLqBJ7XLqfyEP6d9gr/lhhsD8iX1ClTRs2XMMh4+fUwvVgg/5h2JIj2bInSOT6TOZ0a7bbogqaQg+pNRCXlu1rM8Z3vjyuxbd+6TwRpdY1NpTeKPGKbqtKLzmFV6ZEKkvi9Yyf6H+zsp/j5uxQu3kMHdCb3Ux+fv8Y5G8Skwv/PVBajcuGKXKDELO8Mo5Uvu7YFI/fJMjE6ysrXDkxHnTorXgXHfsPxm+fv6wtbXB9NHdTENg0uw12HngJEb0aYmsmdLhwaOnaoFc5bJFojtMdGlP4dUFa9wL6t53onr9LrOgHx8Hj/2BqfPXY96E3qpsQZ4IV27abxJembmUXwA5PszUtlezrR/P8P5Uu5t6apRfSpHm8g17q9c0GdOnxtt3Pqo4PibCKxIePMMr15Qj5Axv8P1Ire7wSYvhmtAJvdzrmW7z8dMXkPvv2Ly62u4luselK/9g8LgFWD1riDo1IDBQrbDducwr1Osp6ecPVTqq2e/gmeh2fSagdqWSKFUsH2d4ows+nrSn8OqTaAqvPlwpvOYVXokm62RGTl6sSv/kkDUo8lZVFpbJITPAsrjtna+fWj/j3qQKyv9USP0spPDKf89ZtlVtMyZ/j9bMGYa/r98OI7xSmthj6HSMHdhO1QgHH/L3fdr89aqu99nzV0iaxBU1yhdXbyiNcFB4jZAFC+jD2YtXVd1rz3Z11S+Ki7Mjbt19hJTJE2Pb3uPqqU5mdOWpr88Ib/j5B5iEd/7K7Vg5c7D6RVuwcgf2HjkTbg1vjZYDVd1vvtzZcPveI9RrNwy7VoxT9UBzl2/DzEUbVX2QHNHdpWHg2HlI6OyotmYJruGV4n25l8tXb5pqk2Xh3M07D9CtTR31dCpbvsgTczdstDMAACAASURBVPch01CxdOFwV7lqpU/Or9y0n9qPsGjBD7s07DtyBvMm9lGxZaa5fdOqaiZXao2TJnZDpxYfdmnoOGCy2mtRdsVgDa8W6fj5cwqvPnmn8OrDlcIbPeHVJwvxMyqFN37mPUZ3La/w5RteZMZS9tYTgZ09rhcc7G3RZeBU/HP7vpK2xjXLYvPuoybhldIAmeUVkUydMhlG9WsVZpcG6dCug6cwZupSNfO7ft4IyM4N2/YdV/vRyq4KUiwvi9q0hPfm7fvAf98+I7Is3xojZQ3DJi7EiTN/qh0QalYsocoEZIFa/9GzcfzMJdjb2eGLNMkxql9r2NnZok1PLzx+9lJtp1KkQC61S4QU+cfkkEV9ch3Z1iXTl2kwZkAbtYuFPCBUaNhH7ZeYJWNavHz9Fv1HzcaZC1fg6uKMgd2amOqhKLwxIR/3z6Hw6pNjCq8+XCm8FF59RpZ2VAqvNiO2+AQCUsMrxezhlUJ8Qth4eSqFN16mXfOmKbyaiGLUgMIbI2yaJ1F4Kbyag0SnBhRencAy7AcCIrxSG9uhuTFqeCw5LxReS86efn2n8OrDlsKrD1cKL4VXn5GlHZXCq82ILT6BAIX3E+B9dCqF13ws41IkCq8+2aTw6sOVwkvh1WdkaUel8GozYgsSMAQBCq8h0mC4TlB49UkJhVcfrhReCq8+I0s7KoVXmxFbkIAhCFB4DZEGw3WCwqtPSii8+nCl8FJ49RlZ2lEpvNqM2IIEDEGAwmuINBiuExRefVJC4dWHK4WXwqvPyNKOSuHVZsQWJGAIAhReQ6TBcJ2g8OqTEgqvPlwpvPoJ741nr/DcxyfKiUvu5Ih0bi5Rbm/pDSm8lp5B9j/eEKDwxptUR+tGKbzRwhXlxhTeKKOKVkMKr37Cu/rS3+iwf3eU8mFnbY155SqgbKb0UWofFxpReONCFnkP8YKAuYQ36Mo/CNx5GAgIiFVu1rYJ4FSxOGxyZMYbn9i9dqzeaCxfjMKrD3AKrz5cKbz6Ce/yi3+hwfbNUUqcvY0NNlSpgZ8zfxml9iEbbdr1K0ZOXqI+eufjC9sECdSXUclxeP0U9cVNRjwovEbMCvtEAuEQMJfwfggdFOuMEzrawtkxAV69DaDwmpE+hdeMMEOEovDqw5XCa/nCG3JkdBowGdV+LoZSxfLpM2DMGJXCa0aYDEUCehIwr/Dq2dPwY7s528LZIQFevPGn8JoRP4XXjDApvPrADBGVwhv3hbdSk37o1ro2Rk5ZjIJ5c2Boz+b4oUoHnNoxyzQS+ozwxk9F86Fcye/w9p0Phk1chD8uXoWDvR36d26E7/JkN/tYpPCaHSkDkoA+BCi8+nC19KgUXn0yyBlefbhSeOO+8Bas0A7FCn2jRNfF2RG+fv6RCq/IrouTA7q3rYNr/95Fqx5jsXXxGDg5Oph1EFJ4zYqTwUhAPwIUXv3YWnJkCq8+2aPw6sOVwhv3hTdPmVbYOH8kvkyXUt2slvAWrtQeO5d7wS2hs2rftMtouDetisL5cpp1EFJ4zYqTwUhAPwIUXv3YWnJkCq8+2aPw6sOVwhs/hPf3XbNhZWUVofD2Hj4TpYrlVzPB35VviwxfpDINuDdvfVRZQ9kSBcw6CCm8ZsXJYCSgH4G4ILwOdjZ4+TYA73wD9QMVzyI72FnDPoENXrz1j2d3ru/tujnZwjcgED5+7/W9UDyLnszNHk9f+uJ97K+bNQNp83U6TVJHM/QndIjY2qUh5FXDW7QmM7xnd88xNQsIDEShCu44vfP/NbytenqhdqWSqoZXZnj3rZ4IJ0d7szMJGZDCqyteBicB8xGwdOF1dbLF0bt3sffmTQv9Y2e+XJozkrUVIBMpgfQyc2KFjTUQFASOVbNSBRLYWCEg0HziaObuRRgugbU1yn6RAV8lSmyWS8Yn4RVgpev2gLdnd2TOkBa37z1Clab9Mbp/GyW8UsNrb2eranhtE9jg5p2HSJ0iCWxtE5iFdXAQCq9ZcTIYCehHwOKF19kW665cQdNd2/SDxMgkQAIkoAMBlwS2WFquCgqk+P+r90+5THwT3oPH/sCUuWtVnW7SJK6ws7VF8cLfmnZp8Jq+AoeOn4Ofv78qb/Ae29PsM74U3k8ZsTyXBGKRAIU3FmHzUiRAAiQQggCF1/KHA4XX8nPIO4gnBCi88STRvE0SIAHDETC68EqRyO5rN3Hi3t0osZMFZT9+8QV+SJ8mSu3jQiMKb1zIIu8hXhCg8MaLNPMmSYAEDEjA6MIryD7UnEe9Ptra2gof9lGIHweFN37kmXcZBwhQeONAEnkLJEACFknAEoTXIsHGYqcpvLEIm5cigU8hQOH9FHo8lwRIgARiToDCG3N2RjmTwmuUTFhwP2Yv3QI/P390aF7dgu/C+F2n8Bo/R+whCZBA3CRA4bX8vFJ4LTiHss3HrCWb8feN20hgY4OcX2XAxCEdkNDFKUZ3denKP/hl3nrMGNNNnb/n8GmULpZfM5aW8G7c+SuWr9+DG7fuw9rKCpm+TIO+HRsgd45MkcbuNngqGtcqh3y5s2r2ISYNgoKCMHPxJixatRPHtkw3hThz/gqadR0DuxB7AO5fM0lxffbiFfqNmoVzf15HkkSuGNarhal/2/aegPeSTXj+4jWSJEqovinmuzzZQ3Xt8dMXav9Bj65NUKFUoTDdjiw+hTcmWeY5JEACJPDpBCi8n87wc0eg8H7uDMTw+lv3HsfYacsxtGdzfF/ga7x/H4Q/Ll5F4fwx/+7pdz5+uHL9Fr7NmRkvX79Fi26eWDN7qGYPIxPeyXPWYv+vv2NgtybInT2j+qrB85dvIPOXaeDm+uF7syM6zv95Xcmxs5ODZh+i28A/IBC9hs1AimSJsGXPMRzdNM0UYvehUzh++pLq88dHn5HeSJsqGTo2r4ELl6+j+9Dp2LJoNBzs7TBz0SZUKVsEaVIlw8mzl9Fj6HQcXDfZ9PWKEku+lUa+NrFWpZLhCm9k8Sm80c0y25MACZCAeQhQeM3D8XNGofB+TvoxvLbMTP5YqxuG926hvof640NmGPuOmoUbN+8hICAQ9ar9hNYNK6lmbXqNQ6F8OZXQvX77Dl+mS4nB3ZvB0cEOf1y6hgneqzDLqyfa9RmPsxevIUuGtCiYJzt6ta+H9dsPY97ybfDx80fSxK4YN8gd6VInR0TCe+f+Y1Rv4YEti8YosQzv0OprqwaVUDBvdtXvkkXy4uCxs3j7zgeODvbwGugeRprnLt+GddsOhbpUsiRuWDi5X5jLHz5xHj98lwvfV26PE1tnmH6+cuM+PH3xCu5NqoY6Rx4qilTpAJntFV5KYD2moGaF4ihZJE+Y+IUqumPXinFqo205Nu36FWfO/Q17e1t8mzNLGOHVik/hjeEvDE8jARIggU8kYAnC++rWIwS+fB3lO7VN6gbnVEmi3N7SG1J4LTCD1/65g4YdR+J4iNfwIW9DxOnsxb+RL3c2yCv0Sk36YdsST/WaXcTRydEBE4Z0gGxJIl/pl9jNBZ1a1DAJr8jhzTsP0HHAFGxaMNIU+q9rt5AqRRIlcONnrsI7H194dG0cofCu2rQfv568gMnDO0VIWauvIYVXvmZwyvDOsLGxVrPbLi5OaN80tJRGN53yHd8/VOkYSnjnr9iOhat3qq86FLGuU+VHNKheCg8ePUOjTiOxe8U402XkASGRmwta1KsQ6tLX/r2Ljv0nYfvSserzR0+eo12fCVg0pT8mzlqtcvNxSYNWfApvdLPL9iRAAiRgHgKWILwPthzB614f/uZoHrYJ4DZ1EJIVDztZo3luDBuIfwT/TY9hiE86jcL7Sfg+z8nyunz4pMWhZDSynjTsMAIDujRCzmwZlPDK6/SyJQqoUy5fvalqUtfPG6EpvCGvcfTUBSxdtwfTRnWNUHinL9yIFy9fo1+nhurUI7+dV6/5379/j0plimBw96Zhuv1xX0MKb82KxVGuZEF1zvZ9J3Dg2Fl4Dmj7SUkIT3gDA9/D189ffa2hiL/M4nZsXh3ZMn2hJHbzotGma05fsEGVk3Rs8f8FexKzdU8vNKpRFqWK5VNtO/SfhCa1y6FQ3hwYMWlxuML77+0Hkcan8H5SqnkyCZAACcSYgEUI78aDeN3KI2r3aGcLt6XjkKzkh79R0T12HjiJ3sNnwuG/t51yfueWNdGwRukIQ1F4o0uZ7ZWktujuGaruNCSWW3cfwnvxZjUjaWUFSC3s7PG9kOurjEp4m9ctr+p+5Xj4+DlqtR6EQ+unaArv6i0HsOfQaUhJxYuXb5Aksata4BZRScOStbshC8BkNjnksWbLQXUtKcnQ6mtI4W3ZoKISRjnkl23XwZMYP7h9qNjSl7VbQ5c0JE/qhsW/DAh35IQnvB83XLX5gKrXlbrduu2GqJKG4MNz2nJIyUTL+h9meEV++470Rvq0KU0SLKUgsiBwQJfGqk1Ewiu5iCw+hZe//CRAAiTweQhQeENzl7/BMvE0aVjHKCeEwhtlVGwYTEAWXBWv1glTRnQOswuAtKnTdgia1CqHSmW+V6fUbjMEg3s0NQmvfF6l7A/qZ2cvXsXQ8QvCmeF9iI4DJptmkWVHiKnz12PehN5qt4IDR89i5ab9kQqvEvNunti+bKypjlWuGVJ4tfoaUnhDvgqJSHijO0qiIrzLN+zFjZv30a9TA1X+sGO5F1z/2wlDyhRqVyqpZnLlQWDwuPmqZER2oQg+Wvcchz8uXTUtXvP19YONjQ2q/lwUg0IsjJPzI4tP4Y1udtmeBEiABMxDgMIbdeEdNWUJDh0/h4CAAHyTMzM8PdrBNoGNmnAL/ju+bP1eLFu/R21pKk4xbXRXpEqeBPKmc+DYuWoyThaIj+jbCqlTmKfOmCUN5vldiPUoUme6dP0eDO/VAvm/yQZrG2vcvPMQmdKnRuk63TFxaEe17dfpc1fQssdYLJk6wCS8UnvrPbYn7OwSoPuQaciaMV2YGt4Xr97g5/q9sHf1BCVwMkspkikzuvK6v88Ib/j5B0QqvAJF4j968gL9OzdE1kzpEPQ+CIvW7MI/t+6rGV6tvn4O4ZUFfcJOdoe4e/8x2vQer8ovZIuxQV7zkDSxm+KlZn0HTFZ1utJ21JSl8uWOajuyyI6QM7wyIzxi8mJViywzxZHFp/DG+q8ZL0gCJEACigCFN+rC+/uFv03bjjbrMgaNa5VFuZLfmYQ3R9b0qNbcA9uWeqq1MrLAXeRW/h5Wb+mBnu3qqgX5sgBdvMN7bA+zjEIKr1kwfp4gKzbug5QNyGBxsLNVZQpSPiBblsnCKJGwvF9nhZ+/PxrUKG0S3jy5smLfkTN4+vwlCubNEWaXhuAdDWRh2pY9R1G88Lfo06E+ugycin9u31di1rhmWWzefVRTeP1FihdtxObdx/Dk6Qs1uFOnTKrqWav9XFSzr59DeKUsQuqT5XBN6IxWDSqYZsRlu7b+o2bjzIUrcHVxVluXyU4PIvAVG/dVCwFDHl1b1zaVOwR/HlJ45eGhQsM+6hc6S8a0aju48OLLuRTez/N7xquSAAmQAIU3rPB+XMM7d0Jv5Rkhj0mz16iF3c3q/GwSXpmkq9CoD6RMsfrPRSEL0uWQt8IDxszB2jnD1H/LG9gC5drg5I5Zaob4Uw8K76cStLDz5ZWCbFH28RciWNhtxMvuUnjjZdp50yRAAgYgQOENK7zh1fDKfv7eizfhwl831BdNyZvnulV+RPN65UOVNMjb0xmLNuHYqQtqAkz+J6WTPYZOQ8rk/y9hkPVCG+aPUBNtn3pQeD+VoIWd/7mLxi0Ml6G6S+E1VDrYGRIggXhEgMIbNeH1mrECb9/5qp2h5BtgvaavULL6sfAGR5OtU937TlQ7Icl+/bKmaMXMwbqMLAqvLliNG5TCa9zcaPWMwqtFiD8nARIgAX0IUHijJrz9Rs1W5Xmyc5EsPGvaZRTqVA49wytrZB48eooMX6RSdbuy1qdi6cL46Yd8ateodk2qqppfWcgt5YIZ06c2S1IpvGbByCAkoD8BCq/+jHkFEiABEgiPAIU3asJ75fpt9Bo+Q33Lq3wTa6F8OWAFq1AzvNmzpkebnl54/Oylqs0tUiCXWuwtXyolJRAjJi2CfNGVCG/p4gVC7Wb0KaOTwvsp9HguCcQiAQpvLMLmpUiABEggBAEKr+UPBwqv5eeQdxBPCFB440mieZskQAKGI2B04Q0C8PjQWfifvhgldkHWVnD4IS+SFvjwZU7x4aDwxocs8x7jBAEKb5xII2+CBEjAAgkYXXgFaVAQ8F7+TxQP2UYz9EaaUTzRQptReC00cex2/CNA4Y1/OecdkwAJGIOAJQivMUgZtxcUXuPmhj0jgVAELF54nWxx9O5d7Lt5U63M5WEeAtbWUItCAsnUPED/i2JjbYUgBOH9e7OGjffBEiSwRmDAe1javwAJrK1RNn0GZEtknq+5TZPUMd6PhdgGQOGNbeK8HgnEkIClC6+bsy0c7RPgxRs/vPMNiCEFnvYxAQc7G9glsMHLt36EY0YCrk528AsIhI9foBmjMlQyNwc8fekDy3w+s1ZfH2+Og8JrDorRi0HhjR4vtiaBz0YgLgivs4MIrz/e+FB4zTWQHO1t4GBrg2evKbzmYipxErvYwcc/EO98Kbzm5JoqiSMePntnocJrPhIUXvOxjGokCm9USbEdCXxmAhTez5wAg16ewqtPYii8+nCl8H7gSuHVZ3xFFpXCG/vMeUUSiBEBCm+MsMX5kyi8+qSYwqsPVwovhVefkaUdlcKrzYgtSMAQBCi8hkiD4TpB4dUnJRRefbhSeCm8+ows7agUXm1GbEEChiAQF4RXFli9fBsQx+oizbOIJaaDjMIbU3KRn0fh1YcrhZfCq8/I0o5K4dVmxBYkYAgCli68rv9tS7ZXtiX7vI5otnwmtrdHjUzZ4GpnZ7aY0Q1E4Y0usai1p/BGjVN0W1F4KbzRHTPmak/hNRdJxiEBnQlYvPA622LdlStoumubzqRiL3yupMmxrFwVJLV3iL2LfnQlCq8+6Cm8+nCl8FJ49RlZ2lEpvNqM2IIEDEGAwmuINITqBIXXeDkxV48ovOYiGToOhZfCq8/I0o5K4dVmxBYkYAgCFF5DpIHCa7w06NIjCq8uWEHhpfDqM7K0o1J4tRmxBQkYggCF1xBpoPAaLw269IjCqwtWCu9/WLkPrz7jK7KoFN7YZ84rkkCMCFB4Y4RN15NY0qAr3s8anMKrD37O8HKGV5+RpR2VwqvNiC1IwBAEKLyGSANneI2XBl16ROHVBStneDnDq8/AikJUCm8UIP1x6RomeK/Cwsn9otCaTcxNoMfQ6Shb4juUK/mduUNbVDwKr/HSxRle4+XEXD2i8JqLZOg4nOHlDK8+I0s7qiGFd+eBk+g+ZBo8PdqiUunvQ93F5DlrMWvJZmxeNBqZ0qfWvsMQLYKCgvBzg97Ysmg0bG0T4Mhv55H/m6/g6BD5HpqRCe/2fScwdf56PHz8DG6uLmhQvRRa1KsAPz9/nPj9TxQr9I1mH6PTVjNYNBtUaTYAAQEB2LbEM9SZb976oHj1zihZJA/GD24fzajhN+82eCoa1yqHfLmzRitedIX34zxH62IGbkzhNV5yKLzGy4m5ekThNRdJCm94JFnDq8/4iiyqYYVXpDZpYlfM8upp6r+ITOWm/WFtbY1JwzpGW3gl0PHTl1A4f05IrPrth2PGmG5I7JYwUvIRCe/9R09Rq9VgzJ/UB1kzpsODR8/w7MUrZM+SHoeO/4FDx8/Bo2tjzaxGp61msGg2EOG1TWCDQd2b4tucmU1nr9t2CMs37EP6tCnMJrzn/7yOTF+mgbNT9PYsja7whsxzNHEYujmF13jpofAaLyfm6hGF11wkKbwUXn3GUnSjGlZ4ZXZURHD59IFInjSRuq/ffr+MFRv34d/b9+E1yF0J77lL1zBqyhI8e/FaiXD/zo1QrFBu1b5Sk37o1ro2Rk5ZjIJ5c2BM/zbIU6YVzu6eA6/pK7Bk7W5kzpAGrgmdsWBS3whjRSS8cu3hkxZj9awhobhfuX4bMpv54uUbpE6ZFM3q/oyKpQqrfooEy4zqNzkzw9OjHW7cvBemrauLM5at36NkPPj4sVZXrJs7HAldnDB84iLFwt/fH9mzfImpo7qEur78bOiEBWHGwpxxvVR/Qh4ivJVKF8b9R88wqFsT04+adR2DogVz48+//zUJb3j9F1mevXQLXJwdcfTkBTWrvXLmYIz+Zam673EzVkJmi3cu90KbXuPQqkElFMybHW/f+WDYxEX44+JVONjbqbx9lye7uv75yzcwZNx8vPPxRc5sGRAQEIjyPxUKt6Th9Zt3iuux0xfx/n0Q+nRogAqlCpnyXLX5AIzo3RK5c2RSsS/+9Q/6jPRWs/whD3lY6T96Nu49fKJm52tX/hFtG1dW9/b0+Ss15uRaNjbWGNarBb5Ik0KdLvkbNnEhbt19BBcnR/R0r6u4yVHPfZh64Mn1VUb132u2HMTlqzfVZyfPXobntOV48/YdACv079xQvQ2IjAuFN7r/vOnfnsKrP+PPdQUKrz7kWdLwgStnePUZX5FFNazwHj9zSQlEkkQJ0bxeeXUPHp5zUbZEAYz3XoWJQz/M8D58/Bw+vr5InzYlDp84B68ZK7FpwUjVvmCFdkoihvZsroRMjmDhlf8/WCKDZ3gjihWR8Pr7B6BGy4GoUu4HNKpZNlRpREi5CU7A7xf+NolXsy5j0LhWWSVxH7c9fOJ8hMJ79uJVbNr5q7p/OW7fe4R0qZPHeOSI8MpseYtunti9Ypwq9bh7/zG6D52uSjN2HvjNJLwR9V+kUB4eRvRpZXrYELl9/vI1xg1yV7mRI6Twiuy6ODmge9s6uPbvXbTqMRZbF4+Bra0tyjfsjWG9mqNIgVy48NcN1HcfhnGD2ocrvIPHzYe9nS16d6iP94Hv4ecfoHIdnGd5U/Di1Rv0cq+n+jBx1mo4ONjBvUnVUMzGTluOtKmTo2GN0pC8iuSmTJ5YCe+qzQewbs4w9bCxY/9vWLpuDxb/0h+Bge8hQt2zXV1V+vHPrfto3m0Mlk0fhNQpkkQqvLVaD8bo/q3VmwHpn5WVFVxdnNRDQHhcnBwdQOGN8TDX7UQKr25oP3tgCq8+KaDwUnj1GVnaUY0rvKcvomGNMug1fAbWzxuBdz5+Si43LxqFGi0HhVvS4B8QiCKVO+Dk9pkmud04fyS+TPdBuLSENySukLEiq+GVEoYZCzdhz+FTqFjqe7RvVk2Jb3jCGzL+pNlrkMjNBc3q/Bwt4X389AU69p+sJF5KMz71EOGdN6E3Rk5ejIqlv0fpYvkxY9FGyCxzsiRu2HXwZLglDSH7L1L49/XbGDuwnak7IrdlShRA7UolQ30WPMNbuFJ7NevrltBZ/bxpl9Fwb1oVzk6OGOw1T81mBx+NO41UDxThLVqTh5rdK8eb4nyc51t3H6JF97FK5uUQmZ7p2SPUmJDPRWq37T2uSjtC1obLvUl99oAuH0pTpBRGxtiO5V64c++ReggL2VeZbZYHkCa1y0UqvDKuE7m6qPESsqQmIi6F8+Wk8H7qYNfhfAqvDlANEpLCq08iKLwUXn1GlnZUQwvv4B7NULPVIIzs2wpXb9zBxSv/oE+H+gielRQxkVfoS9ftho+Pn7rbM+ev4NSOWSa5/X3XbDV79rEIyX9/PMMbUayo7NIgr7vlFfWjJ88x07N7GIkVYfdevEnNWFpbWeHmnYeoW+VHNXsdlRnekjW7Yv284UqOzpz/GzMXbVSz2yFfoQffo9zHYK/5YbI/f2KfcEsaRHjP/Xkd67cfxi8jOiu+iyb3U2yDhTey/osU+vr6o2OL6qHktmWDiiiUN0cY4c2VPSO+K98WGb5IZfqZlD1IWYOdbQKs2LhXSWnw0XXQVFXSIALetvcHca1cpgh6ta+PolU74syu2WHuNeRMvpQWiLCqWmWveVjlPUQtfJT/yTGkR3NUKvM9Nuw4gvkrtyNV8iQY0KWRmpmWe3v7zhddWtU0XaN8wz6YNrqrKnNYuXFfqL7OWbZV1XHLjPLHJQ2rtxzAX1dvqZIGiblg5XZIrXSJ7/OgR7s6qrQhIi7yZoMzvNr/oMV2CwpvbBOPvetRePVhTeGl8OozsrSjGl54F67eiQcPn+LvG3eUFMiCsGDhTZrIVc3YLZ3mgYzpU6v6R9lZIKTwSr1uyCOkCP1UuxvWzhmmJFLqbSOKFRXhlWu8fP0WJWp0gUj22q2HVP1r8KI1rxkrlOSISCWwsVE1xCJwIrwftz126iIWrt5hEimZVcxbtjX2r5kYajZQ6lHb9RmPLYvGwM31w0xpdI/gGV7ZYaJM3R4Y0qMZ1m47pMRXdssIFt7I+i9SKHWvHZqHFt7g2dzgPoUsaZCZzH2rJ8LJ0T5Ul6V+d9iEhaHqoqVMoF7VUuHO8Iog7l09UZUDRJTnxWt2qQeRBAlsVL22zKpHdkg+Vm7ap8RY7u3f2w8wok9LdcqHmf/22LNqgnpo+Xg2OuQMb6OOI9UDWnD9sAi2PKSEXMgo3KQO3DWhk5LkiLjItSm80R3d+ren8OrP+HNdgcKrD3kKL4VXn5GlHdXwwiuv8Ku38ECKZImVnMoRLLwyG1iv3TDsWjFOidPc5dvUzOfJ7d6qXUi5DUYR8jMpkRD5yJc7m6qFjShWRMIrr8uljlNKJmTBlCyok9fiIuB7D5/Bsg17MHd8b3XpfqNmI0vGtGhZv4KSnqZdRqFO5Q8zvB+3lb406zIaW5d4qvrUbXtPqNKOIxt/gY+vPxzt7VQ5hAh+1eYeWOU9WHOniYiGQrDwinyPmLRY1agO6dlMlTaEFN7I+h8T4ZVaVbk3qeGVmVeRR6l7lUMePEb3b6MWsclDg9TwygK/8EoapF+J3VzQo11dvA8KwstXb9TuHiHzLGOoZY+x6kFj+uhuqjb34+PaP3fUQjQ7O1v8dBj0EwAAIABJREFUde0W+ozwxob5I5TwyqyvLMSTny9YuQN7j5xRNbyS80pN+qJH27ooVSzf/2t4pw1UM+myCE5kt361UvDx9UPDDiOQN1dWNeZk8Zo8vMkh9c837zxQM9wRcVG11U9kgZvlHq7Otlh35Qqa7tpmuTfxUc8pvHEmlWFuhMKrT24pvBRefUaWdlTDC6/cQtve4/F9/q/Vqv+QwislDTJTum3fcSRJ5IralUti1ab9pppKLeHddfAUxkxdquo/pU44olgRCe/pc1cgi6akxtPO1hbffp1ZSUvaVMnUjKd7v4lqdrBd4ypqVwaRVtlxQGo8C+XLAStYKeH9uG2tSiUwc9EmNbuaNLGbEqOjpy5gzvheSqqkblTiiAQ1rVMuVJ2sdspDtwgpvLLrRLs+E3Bw3WQVO6Twys4TEfU/JsIrsi68ZdcKP39/Vd7gPbanenAR3kPHL1C7O2TLlE7tbyyLD8MTXpmZl10STv3xl7qxvh0bqPKHj3PfsvtYBAQGRvjlIcs37MXcZVvVfbs4O6Ffp4Zqv2C5N1nEJ3kUKU2dMhlG9WsVapcGKZOQhxS3hCLedUx7L0t7EXLZ2UHu69uvs+DJ05dKeEWGZWGmvZ0dvkiTHKP6tVYz/pFxofBGd3Tr357Cqz/jz3UFCq8+5Cm8FF59RpZ2VEMKr3a32YIEYoeACK/s2iCLyz73QeH93BkIe30Kr/FyYq4eUXjNRTJ0HAovhVefkaUdlcKrzYgt4jGB8GavPxcOCu/nIh/xdSm8xsuJuXpE4TUXSQpveCS5D68+4yuyqBTe2GfOK1oQAQqv+ZLFGl7zsQwZydHeBg62Nnj2+sNONTzMQ4DCax6OH0fhDC9nePUZWdpRKbzajNiCBAxBgDO8hkhDqE5whtd4OTFXjyi85iLJGV7O8OozlqIblcIbXWJsTwKfiQCF9zOBj+SyFF7j5cRcPaLwmoskhZfCq89Yim5UCm90ibE9CXwmAhTezwSewms88LHQIwqvPpBZ0vCBK2t49RlfkUWl8MY+c16RBGJEgMIbI2y6nsQZXl3xftbgFF598FN4Kbz6jCztqBRebUZsQQKGIEDhNUQaQnWCwmu8nJirRxRec5EMHYfCS+HVZ2RpR6XwajNiCxIwBAGLF14nWxy9exf7bt5U31IXF47EDvaolukruNnZfbbb4S4N+qCn8OrDlcJL4dVnZGlHpfBqM2ILEjAEAUsXXjdnWzjaJ8CLN3545xtgCKbm6YSVecLEMAqFN4bgNE6j8OrDlcJL4dVnZGlHpfBqM2ILEjAEgbggvM4OIrz+eOMTl4T38w4PCq8+/Cm8+nCl8FJ49RlZ2lEpvNqM2IIEDEGAwmuINBiuExRefVJC4dWHK4WXwqvPyNKOSuHVZsQWJGAIAhReQ6TBcJ2g8OqTEgqvPlwpvBRefUaWdlQKrzYjtiABQxCg8BoiDYbrBIVXn5RQePXhSuGl8OozsrSjUni1GbEFCRiCQFwQXgc7G7x8G4B3voGGYBp+JyxrBwkKrz5DicKrD1cKL4VXn5GlHZXCq82ILUjAEAQsXXhd/9uWbK9sS2ZQp7SGFcp9mQHZEyUxRM6j0gkKb1QoRb8NhTf6zKJyBoWXwhuVcaJHGwqvHlQZkwR0IGDxwutsi3VXrqDprm060DFPSHsbG6z4uSoKpkhtnoCxEIXCqw9kCq8+XCm8FF59RpZ2VAqvNiO2IAFDEKDw6p8GCq/+jC3lChRefTJF4aXw6jOytKNSeLUZsQUJGIIAhVf/NFB49WdsKVeg8OqTKQovhVefkaUdlcKrzYgtSMAQBCi8+qeBwqs/Y0u5AoVXn0xReCm8+ows7agUXm1GbEEChiBA4dU/DRRe/RlbyhUovPpkisJL4dVnZGlHpfBqM2ILEjAEAQqv/mmg8OrP2FKuQOHVJ1MUXgqvPiNLOyqFV5tRnGhRvHpnbFwwEondEsaJ+4mPN0Hh1T/rFF79GVvKFSi8+mSKwkvh1WdkaUel8GozMmyLKs0G4N6DJ7C2tlJ9dHK0x/41k8Ltr57Cu+fwaZQull9dNygoCD836I0ti0bD1jaB2djtPHASfUfNwqYFI/FFmhSmuKN/WYrsWdKjevliZrvWpwT69/YDdOg/CXUql0ST2uVMoVp2H4uzF6/C6kOqUL9aafRoVyfMpZ69eIV+o2bh3J/XkSSRK4b1aoF8ubOqdhTeT8lM1M6l8EaNU3xoReHVJ8sUXgqvPiNLOyqFV5uRYVuI8I4f7I6sGdNp9lEv4X35+i1adPPEmtlDTX04fvoSCufPqdmn6DQQ4fWctgzZMqXDTM8ehhTeE7//iZGTFiNLxnTI83XmUMJbrbkH5k3sgySJIp9h7zPSG2lTJUPH5jVw4fJ1dB86XT08ONjbUXijM2Bi2JbCG0NwcfA0Cq8+SaXwUnj1GVnaUSm82owM2yIy4V237RBmL90CW1tblCtRAOt3HMHqWUPg5OiAH6p0wKkds0z31WeEN34qmg/lSn6nZmjnLNuKNVsO4u07H1QoVRj9OjXEuUvXMGrKEjx78RrW1tbo37kRCubNjnZ9xuPsxWvIkiEtCubJjl7t6yFPmVY4u3uOin/j5j0Mm7gQt+4+gouTI3q610XRgrnVz9r0GoeSRfLi4LGz6lqODvbwGugON1fnMMxFeKUPJ/+4jNYNK6FM8QKqTcgZ3gneq1TJRvN65dXPbt55gI4DpqhZYTlK1+mO+tVL47ff/8Tte49Qr+pP8PMPwKHjf+DFyzeoUaFYKEmVc975+KFW60Fh+tO5ZU3FK+Rx9cYduLg4YvXmA3BL6Bwq1k+1u2HvqgmwCp7iDWdUvX8fhCJVOqhZekcHO9Wik8cU1KxQHCWL5KHwxsJvIoU3FiBbyCUovPokisJL4dVnZGlHpfBqMzJsi4iE9/rNe2jedQxWeg9GquRJsHD1ToydthxHNv6iKbxb9hzD8vV7McOzOxI6O+L5y9dKIh8+fg4fX1+kT5sSh0+cg9eMlUokP5ZKgRUsvIGB71G1+QD0bFdXCds/t+6jebcxWDZ9EFKnSKKEV8oepgzvDBsba9VHFxcntG9aNVzhPfXHZVW60GXQVGycP1KVcERHeKVffTrUR/1qpfDoyXOUrtMDbRtXRvtm1fDmrQ/K1e+FHcvGwsXZ8ZNyPnnOWiR2cwklvIUquiNl8iRK7HNm+xK929dHutTJQ13nwaNnaNRpJHavGGf6XCQ+kZsLWtSrQOH9pKxE7WQKb9Q4xYdWFF59skzhpfDqM7K0o1J4tRkZtsXHNbzf5/8ak4Z1xLwV2yDyJDOzcvj7ByBfudY4tH6KpvB27D8Z1coXNdXkhnfz/gGBKFK5A05unxmp8F668g88POdi3dzhpjAySyyiJ/WtIrw1KxZHuZIF1c+37zuBA8fOwnNA23CF9/jpixjco5mSXBFlEenoCu+BNZOUQMpRqnZ3zBnfCxnTf/ga2bpth2JE35ZRKhGJbFCEJ7yv37yDs5MDAgLfY+m63di44wjWzxsRKozU/3bsPwmbF402fT59wQbIzG/HFtUpvLHwm0jhjQXIFnIJCq8+iaLwUnj1GVnaUSm82owM2yKiGV6v6SuQJLErWtavYOq7vCrfunhMuMLbe/hMlCqWX72ir9N2CAZ2bYLcOTKFum+pTxVR8/HxU5+fOX9FlUVENsO7/+jvWLlxX6iaWymXkIVZvdzrKeFt2aAiCuXNoWJK2cKugycxfnB7tO09HqfP/aU+37xwtFrEFSy8MhtbrYUHZozuhjVbD+KrzF+omd+PSxpEIKUkILikQWZ4f98121RWULpuD6yYMQjJkrip69RzH4YhPZqpRXDBh5Q0VG/hEWYMdGtTyyTqH/8wPOH9uE2JGl2wynsIUiZPbPqRzKLXbTck1MJDz2nLVf8kl1y0pv+vIoVXf8aWcgUKrz6ZovBSePUZWdpRKbzajAzbIiLhlRneJ09fqnpaOaS0IF/Z1jiwbhISujihUAV3nN75/xreVj29ULtSSSW87n0nolbFEihVLJ/pvqW+tXzD3lg6zUPNhspreVkE90F4H6LjgMkmqZSTgksazl++gcFe8yKd4W3VoJKqBf5YeD+GLjIcLLzys92HTmHR6l3I83UWZPoytRJeEU0pRwgW/d8v/I3B4xaEEt7g2mKJERXhjUnyoyK8Rat2wtYlY1Stb/Ah9dM/VOmIHcu94OripD5u12eCyo3kg8Ibk2xE7xwKb/R4xeXWFF59skvhpfDqM7K0o1J4tRkZtkVEwiuLp2SGdPmMQUiRLBGWrd+LkZMXqxpeqccV0fP27I7MGdKqxVtVmvbH6P5tlPCu334Y67Ydxowx3ZQ8SmmEf0AA6rUbhl0rxqm62bnLt2Hmoo04ud0bL169wc/1e2Hv6glq9jik8Mqr+EpN+qJH27pK2Ew1vNMGInXKpGqGN6bCK9dp328iZBa3VYOKSnil71JfPGFIB9WPoRMW4vS5K59deGXm9uGTZ8j1VUa1KHD+yu2qn/Mn9lXlCiMmL1Z1yzKTO8hrHpImdkOnFh92aZCHie1Lx6pyCAqv/r+KFF79GVvKFSi8+mSKwkvh1WdkaUel8GozMmwLEd7bdx+G2IfXQdXpyiG7LMguDbIY7McieSElCbPH9VTCe/DYH5gyd62aXUyaxBV2trYoXvhbJbwiYFPnr8OmXUfh6+uHSmWKqIVeUiaxbd9xtTds7colsWrTftPM7fiZq7Blz1EVY2jP5mF2aRCJE7F2S+ii9p4tVugb1cdPFd679x+jSrP+GNClsRJePz9/yLZesjdxQhdnJdnLN+xVC9xCinhwQmNrhlf6KduL3bn3CPZ2tvj26yzo27EBkidNBF8/f1Ro2AfeY3sgS8a0kG3e+o+ajTMXrsDVxRkDuzXBD9/lUl2m8Or/q0jh1Z+xpVyBwqtPpii8FF59RpZ2VAqvNiO2IAFDEKDw6p8GCq/+jC3lChRefTJF4aXw6jOytKNSeLUZsQUJGIIAhVf/NFB49WdsKVeg8OqTKQovhVefkaUdlcKrzYgtSMAQBCi8+qeBwqs/Y0u5AoVXn0xReCm8+ows7agUXm1GbEEChiBA4dU/DRRe/RlbyhUovPpkisJL4dVnZGlHpfBqM2ILEjAEAQqv/mmg8OrP2FKuQOHVJ1MUXgqvPiNLOyqFV5sRW5CAIQhQePVPA4VXf8aWcgUKrz6ZovBSePUZWdpRKbzajNiCBAxBgMKrfxoovPoztpQrUHj1yRSFl8Krz8jSjkrh1WbEFiRgCAIUXv3TQOHVn7GlXIHCq0+mKLwUXn1GlnZUCq82I7YgAUMQsHjhdbLF0bt3se/mTfUFJ0Y8rK2sUCZ9BuRInNSI3Qu3T472NnCwtcGz134W02dL6CiFV58sUXgpvPqMLO2oFF5tRmxBAoYgYOnC6+ZsC0f7BHjxxg/vfAMMwTS8TljBGkEwppCH118Krz5DicKrD1cKL4VXn5GlHZXCq82ILUjAEATigvA6O4jw+uONj3GF1xDJjkYnKLzRgBWNphTeaMCKRlMKL4U3GsPFrE0pvGbFyWAkoB8BCq9+bC05MoVXn+xRePXhSuGl8OozsrSjUni1GbEFCRiCAIXXEGkwXCcovPqkhMKrD1cKL4VXn5GlHZXCq82ILUjAEAQovIZIg+E6QeHVJyUUXn24UngpvPqMLO2oFF5tRmxBAoYgEBeE18HOBi/fBuCdb6ABmFrOwrTIYFF49RlKFF59uFJ4Kbz6jCztqBRebUZsQQKGIGDpwuv637Zke2Vbss/smkkcHFA7UzY42doaIref0gkK76fQi/hcCq8+XCm8FF59RpZ21EiFd8/h08iWKR3Sp02pIh357TwWrtqJtKmSoVvb2nBL6Kx9BbYgARIwCwGLF15nW6y7cgVNd20zC49PCZInWQqsKFcVCe3sPiWMIc6l8OqTBgqvPlwpvBRefUaWdtRIhbdK0/4Y0rM58uXOikdPnqNi475o36waLl+9iaD3QfD0aKt9BbYgARIwCwEKr1kwqiAUXvOxjKuRKLz6ZJbCS+HVZ2RpR41UePOWbY2jm6bB0cEOk+esxeOnLzC8dwu8ePUGFRr1wa8bp2pfgS1IgATMQoDCaxaMFF7zYYzTkSi8+qSXwkvh1WdkaUeNVHiLV++MdXOHw8nRAeXq98S8iX2QNWM6+Pj6oUjlDjiza7b2FdiCBEjALAQovGbBSOE1H8Y4HYnCq096KbwUXn1GlnbUSIV31JSluPjXDQS+f480KZNiwpAOKuLZi1fh4TkXWxaN1r4CW5AACZiFAIXXLBgpvObDGKcjUXj1SS+Fl8Krz8jSjhqp8AYEBmLd1kNKeGtUKA57uw8rmk+evYw3b31Qskge7SuwBQmQgFkIUHjNgpHCaz6McToShVef9FJ4Kbz6jCztqNyWTJsRW8RRAn9cuoYJ3quwcHI/i7hDCq/50sRFa+ZjGVcjUXj1ySyFl8Krz8jSjhqp8L5/H4Q1Ww5g697jePj4ObYv9VQRj566oGZ4yxQvoH0FtjA7gWcvXqkdM2RBYchj9C9L1ZZxTWqXM/s142JAPYT32KmLmDx3LR49fo40qZJh8S/9Fbrzl2/Aw3OO2u0ke+b0aoeT5EkThcEque03ahbO/XkdSRK5YlivFmqXFDkovOYbhRRe87GMq5EovPpklsJL4dVnZGlHjVR4ZyzaiG17jqN+9dIYO20Zzu6ZqyKeOX8FQycsxMb5I7WvwBZmJ0DhNQ9ScwvvuUvX4DF2HsYNclf7VwcfgYHv1a4mHl2boFih3FiydjeOnb6IaaO6hrmRPiO91UNLx+Y1cOHydXQfOl3VyjvY21F4zZN2FYXCa0aYcTQUhVefxFJ4Kbz6jCztqJEKb6na3TFnfC9kTJ8aecq0wtndc1TE5y9e46fa3bhLgzZfXVpERXiDgoIwZ9lWrNt2GFKLXShvDvTv3FDtuHH4xHksW78HM8Z0M/Xvx1pd1Y4cid0Sok2vcWhW92eMm7FSzeRvXTIGwycuwm+/X4a/vz+yZ/kSU0d1CXNv5er3QuUyRXDuz2t4+vwVypX8Dq0bVlLt3r7zwbCJi/DHxatK3vp3boTv8mRXP6vUpB+6ta6NkVMWo2DeHBjTv40pdotunmjVsCKKFMgFXz9/tTvI8N4tUaFUIdWmWLVO2LRwlKovjyj+v7cfYODYueothcjkiL6tkDpFEoQU3rfvfNGs62i0a1wFPxXNZ7r+Ox8/1Go9KMy9dm5ZU91fyKPzwCmoXakkihX6JtTn5/+8Dpl9XzZ9oPpc3pyUqNEZ25Z4IqGLk6mtfF6kSgfsXzNJbQUoRyePKahZobiql+cMr/l+nSi85mMZVyNRePXJLIWXwqvPyNKOGqnw5ivbGofWT4GLs2Mo4b3271007DACx7dM174CW5idQFSEd8vuY1i95QBmjOmu5Gn0L8sQGBiIgd2aREl4n798rWYq5Vv29h/9HZt2/oqJQzuqe7l97xHSpU4e5r7koahnu7poVLOM2rquUceR6OleF4Xz5VQy6uLkgO5t60DGT6seY7F18Rgl4AUrtFOSOLRnczXWQh7yliEgIBCdWtRQ/Z40ezWyZkqnpPj6zXvoPngaNswfEWF8B3t7VG/pofol11i37RB2HjgJ77E9TMI7f2JfdPKYjO/zf636HtNDZLVLy5pYs/UQ3r9/j7pVfkSdKj9CciEzuiP7tjKFruc+DAO6NEbu7BlNnz149AyNOo3E7hXjTJ9JjXEiNxe0qFeBwhvTxIRzHoXXjDDjaCgKrz6JpfBSePUZWdpRIxVe974T1R9k+Xa14BneFy/foPeImXBN6ASvge7aV2ALsxMQ4S1atVMYOfT19VNCKTW8HfpPQo3yxVGq2IfZSslb2fo9cWLrjCgJb5kSBdRspRx/37iNjv0nKyEtnD9nhPcjY+TAmklK0ORYvGYXbt19pGaWC1dqj53LvUxfR920y2i4N62qZFjOk/KYL9N9+ArrkMfpc1cwdf46iJSOnLwY3+TIjEmz12DPqvFYveUgrt64rWaLI4qfyNUFA8bMwdo5w1RYme0uUK4NTu6YhUtX/lECnSPLl3gfFIS+HRvEOFcy+5y/XBu0alBR3ZfMaDfpNEpJ7l/Xb+HPv29iULcmpvjNuo5R7WTmPfiQmeiO/Sdhc4jt/qYv2KBmhDu2qE7hjXF2wp5I4TUjzDgaisKrT2IpvBRefUaWdtRIhffew6foNniqKmGQWb2vs2XAtX/v4KvM6TF5eCckS+KmfQW2MDuBqMzw1mo9GIO7N0XuHJlM15dvzpNvxxOJ/LikoWTNrlg/7/8lDS0bVAwlY2fO/42ZizaqsgCZtS1aMHeY+xJxPbXDGwlsbNTPtuw5hv2//q5KEL4r3xYZvkhlOkdKJURUy5YooIT3912zYWVlFSamv38AStbqioPrJkO+6nrlzMHoOngquraujSVrd6mFk1LuEFF8KXXoMXQaUiZPYoot8i+zwnfuP4Z73wmwsbZG/Wql1IPdx4eUNFRv4RHm825taqFcyYKmz6Wf0offts2E3X/b98nstLBInTIpDh8/F+qruGu2GoQhPZqFyo+wrdtuiCppCD48py1Xv2ct63OG15y/SBRec9KMm7EovPrklcJL4dVnZGlHjdK2ZCI7IrrvA98jc4a0KPDtV9qR2UI3AlER3nZ9JqgZ2pAzvGXq9VBCJjsJLFy9AzM9e6g+Sr2vyPD+NRNNNbytGlRCwbwfamxDHhf/+gft+ozHlkVj4ObqHOpnIq7bl45V9bFyzF66BY+evDDN8O5bPRFOjvZhYoasDw8PmtTx1qhYHGu2HMSCSX2xcPVONYMq9clrZg1V/ZAZ3vDi//n3vxg6fgFWzBwcJrTU8MrP5ozvjTptBmPS8E7I9dX/Swyim0CpaxchD959Qb6OW2a7v/v2KwwetwCrZw1RIWWWWWbody7zCsVQ8vBDlY7YsdwLrv/V9obMI2t4o5uRiNtTeM3HMq5GovDqk1kKL4VXn5GlHTVS4V26bjeqlP0h1MIa7ZBsoTeBqAjv5l1HsWrzAXiPlRpee4yZugx+/gFq1ldm65t1GY2tSzzVYq9te0+g1/AZOLLxl3CFV2b6He3tlLyJaFZt7oFV3oNV25CHSHO9qj+hd/t6atFag/bDMbRXc1MNr1xLSi5sE9jg5p2HSoxtbROEqg8Pj53MlO45dBqVSn+P5vXK48bNe+g2eBqsra3+1955QFV1bWv4RzooKJbYE7sxMbGbYu8ae++9C9hFsYuiYu9i7733bqKJxmiiMZrEGGvsvdLLG2t5OQ8U2ecczzps4N9jvDHu5cw999rfnL77ncXcG/mgnTjEjHB8+VOlSiUfOuvetp58yExI5fX/7skHMWM/tHb81HmI3dRNC0fLh+rMOWYu3oy7959grE8niBqJGeaZY72R95NsqNNuCIZ4tZI74+ItDUd+/E3+qW4xrjB2xkr0bFdP7uSOmLQE6dO5y5ll8ZYGz6Ez5JcIVxcnjjSYU5T3nEPhtSDMZJqKwqumsBReCq+aztLOmqDwCoHZvWK8fJ8oD/0QMEZ4xWoDV+7Ehh1Hkco2FUoVKShHCIQ4iWP+ih048MNpKVcF8+aU71YWb+SIeUtD7B3esxcuyz8lLR4eE4Larml1w3xvbCpil1W8lWHb3uMIDglFs3qV4rylYdLcdTj283mEhYfL8YbAgAFyx1drh1e8Bq+Nl798G0Oej7PKS9ZoOQjlv/5SSqQ4hIi/L7+Q67HTV+DSlf+k8FYpV0LO0779WjKx22tnZysfJjPnEA/qjZqyTO6giy8Z3drUQYOaZWUqcW3f8Qtx9/5j5P44KyYM7Sof/BOzv7Va+ciH6PLmyoYXr4Lg678Qv134B26pXeVDht+W/Fzm4A6vOVWJ/xwKr+VYJtdMFF41laXwUnjVdJZ21gSFd8CYefJXvOIVVTxIQIuA+IIkZnF5qCFA4bUcVwqv5Vgm10wUXjWVpfBSeNV0lnbWBIX33MV/MWryMhTImwOlinwqd61iHzHvQtW+DCNSAgGtndqUwEDlPVJ4LUeXwms5lsk1E4VXTWUpvBReNZ2lnTVB4W3S9c1DNu87Yh7C0b4MI1ICAQqv2ipTeC3Hl8JrOZbJNROFV01lKbwUXjWdpZ3VqLc0aKdhBAmQgGoCFF7LEabwWo5lcs1E4VVTWQovhVdNZ2lnpfBqM2IECeiCAIXXcmWg8FqOZXLNROFVU1kKL4VXTWdpZ01QeCs27pNghtgvyNe+FCNIgAQ+hACF90PoxT2Xwms5lsk1E4VXTWUpvBReNZ2lnTVB4f3l7N9xMohXKP1354H8K1292tdHzUqlta/ACBIgAYsQoPBaBKNMQuG1HMvkmonCq6ayFF4Kr5rO0s5q1kjD/YdP5Z93XTt3uPYVGEECJGARAhRei2Ck8FoOY7LOROFVU14KL4VXTWdpZzVLeEXaqs0H4OC6ydpXYAQJkIBFCCR54XWxx4k7d3Dk5k35F+YS8/BwckaTvPnhbGefmMuwyLWdHW3hZG+Lp6/CLJKPSd4QoPCq6QQKL4VXTWdpZzVLeP/+9yb6jJiNfWsCtK/ACBIgAYsQSOrC6+5qD2dHOzx/HYbg0AiLMPmwJDYfdrpOzqbwqikEhVcNVwovhVdNZ2lnTVB4W3uOeyeD+POpV2/cwRDvVvH+eVntSzKCBEjAHALJQXhdnYTwhuN1iB6E15wq6O8cCq+amlB41XCl8FJ41XSWdtYEhXfr3uPvZHB1cUbBvDmRM1sm7eyMIAESsBgBCq/FUCarRBReNeWk8KrhSuGl8KrpLO2sCQrv46cvkD6d2ztZIiIjIR5cy5Y5g/YVGEECJGARAhRei2BMdkkovGpKSuFVw5XCS+FV01naWRMU3m/q9sKJHXPeyRIeHoFyDbxxctdc7SswggRIwCJ7YVKqAAAgAElEQVQEKLwWwZjsklB41ZSUwquGK4WXwqums7SzmiW8v57/B15DZ+DEzndlWPuSjCABEjCHQHIQXicHW7wIikBwaKQ5CEw/JzoaSB7Ppr333im8preFMWdQeI2hZHoMhZfCa3rXWOaMeIW3WbfR+O/uAzx/8Rrubq5xriReJ/Q6KBh9ujRBpxa1LLMKZiEBEtAkkNSF183FHsdv38K+69cRHaV5ux8ckNbBEU3yF0B6R+cPzqXnBBReNdWh8KrhSuGl8KrpLO2s793hvXrzLjr2nYgZfl5xsqRKlQpZP0of72yv9uUYQQIkYC6BJC+8rvbYevky2u7fbS4Ck84rkNYDa6rXRWaXuF/aTUqSBIIpvGqKROFVw5XCS+FV01naWRMcabh05T8UyJNDOwsjSIAElBOg8JqGmMJrGi9GxyVA4VXTERReCq+aztLOqvmHJy5cuobLV28hLCz8nWzN6lXSvgIjSIAELEKAwmsaRgqvabwYTeG1Rg9QeCm81uiz+K6RoPDOX7EDi9fuwReFcuP8n1fxecFPcOPWfbx6HYzm9SqhX7emibVuXpcEUhwBCq9pJafwmsaL0RRea/QAhZfCa40+M1l4yzfsjQWTBsixhoadhmPLYj+Ih9YmzVuHdO6p0bV1ncRaN69LAimOAIXXtJJTeE3jxWgKrzV6gMJL4bVGn5ksvMWrd8XPu+bC3t4ODToOw9YlY2WO5y9fo07bITi2dWZirZvXJYEUR4DCa1rJKbym8WI0hdcaPUDhpfBao89MFt5GnUfAb1BHFMr/CTr1D0D/bk3lf3705DlqtvLB6b3zE2vdvC4JpDgCFF7TSk7hNY0Xoym81ugBCi+F1xp9ZrLw7jp0ElkypUfxL/Jj39FfMG7GSpQt/QXO/3UVBfPmxOQRPRJr3bwuCbxDIGDOWjx4/CzR+lLMvM9fuQN2tqnk2vLnzoE1c4cb1vnDyd8x0G8eVsz0lf9+4juePn+JIf4L5L8xj7RuGDOwI4oVzidDKbymNT2F1zRejKbwWqMHKLwUXmv0mcnC+/YJR346i1/O/oXsWTKiSZ0KcHSwT6x187okEIdARGQkGnceKcdvFk8dBLfULlYn5D9zlfxyWL1CqXeuvWz9Phw9cRbBIaFSYt8nvD7jApEtcwZ4dmiIC39fRb/Rc7FrxXg4OTpQeE2sKIXXRGAMj0OAryVT0xAUXgqvms7Szqr5WjKRIjwiUo4xZMnkoZ2RESSQCASETB796Sw+ypAOGdOnRdO6FeUqoqOjMXb6Svz9703cuf8IDx8/l1/Ydi73l3K8cPUubN//EyIjo9CqYRW0blQ1zuqDQ8LQuMuId+7Iu1MjVK9QMs7P+4+ei5YNqkjpffs4dfYvFP0sLzr1n4ShvVvHK7zigdBv6vbC0U3T4ezkIFN4DZuJRrXKocI3RSi8JvYVhddEYAyn8FqhByi8FF4rtFm8l0hQeMXDaX7TVuDAD6elOPxxZKlMsvPACdy8fR+9OjRIrHXzuiQQh4D38Jlo16QGMmVIi8HjFmD1nGHy84PHzmDHgROYNdYboWHhqNvOV36WwcNdjums234EgQH9ERERiRY9/eTM+peF8phFt8fgabj+3135BTFzRg94dWqI0kU/jZOrtec4DOvTJl7hvf/wKVp7jcPBdZMN50wN3IC07qnRsXktCq+JVaHwmgiM4RReK/QAhZfCa4U2M114B/svQGhoGHq1b4DGXUfi3MFFMsk/V2+h55BpOLR+SmKtm9clAQMBMffa1ssfO1eMlz9r6+2P0QM6IFfOLAhcuROpUtmgS6va8rOuAyejZ/v6KPJZXngNnYEGNcuiUpli8rM5S7ciKjoaXh0bmkU3JDQMtra2cob39LlL6DdqDjYuHB3nNyMJCa94x7Wn73TDfYhFzF22Tb4K0LNjAwqviVWh8JoIjOEUXiv0AIWXwmuFNjNdeL+p0wu7V01AOvc0KFK1s0F4xc5v+QbeOHdocWKtm9clAQOBlZsOyHdDp0r15mGxqKgodGhWE327NsGh479iw46jmDuhL14HhaB59zFYN28E3N1c0bTbKDx59tIwiy52gKuWKwGfXi0MucVIg3gl39tH366N453VjR0nHlAr/1UR1K76teHHCQnvg0fP0Kz7KDnSEHNMnLNW7kZ3asEdXlNbnsJrKjHGxybAGV41/UDhpfCq6SztrAmONHxduye2L/OXvyaOLbynz/2NQWPnx/kfZu1LMYIE1BAQr8+bPsYTObJmkhcQEitk9sDaybCxAboNmoIXL18jrXsatGtSHV+X+EzGefrOQLN6FeWbR1QcYoe3VuWvUKVscaOEV4wNfVvXE/vWTjI8dNfdZyqa1K6AymWLcYfXxCJReE0ExvA4BCi8ahqCwkvhVdNZ2lkTFF7xGrI79x9jWJ+2qNlqEE7smIPf/vhHPgRUo2Ip9OnSWPsKjCABhQTEw2ijpiyTu7axDzG60LZJdfne6C4DJmHzojHvrELM8K7ddhgz/bzlju+z56/kSINH2jRmrfj7E+dQpnRh2NnaQnwpFG9c2Lp4rMwdc7y9wyvGFcbOWIme7erJndwRk5YgfTp3OVYh3tLgOXQG9q4OgKuLE4XXxKpQeE0ExnAKrxV6gMJL4bVCm8V7iQSFNzw8AjMXb8GarYcg5hPF4eBgL3fJxEyh+B92HiSQmATGz1otX+Ml5Db2sePATzj283kM690G9ToMxeugYNjZ2SFT+rTw7tzIsOu6fON+rN16GEHBIUjrlhr+vl3weYFcZt2SeHDu7B+X5dsfxJoG9WyBwp/mjpPrbeEVYxS1WvnIB+fy5sqGF6+C4Ou/EL9d+AduqV0xvG9bfFvyc5mD7+E1rSwUXtN4MTouAe7wqukICi+FV01naWc1+rVkd+49QmRUFHJkySj/B50HCSQFAs27j8aAHs3lq8Kio4FLV27KUYbDG6cmheXHWSOF17SSUXhN48VoCq81eoDCS+G1Rp/Fd414hVe8XmnS8O5I7eosz9m463vUqvSV/LUqDxJISgTK1PPCjuX+hjGFYz//Lt/cEPPasqR0LxRe06pF4TWNF6MpvNboAQovhdcafWa08H5WoT1+3D5Lvp1BHKW/64Eti/3kr2l5kEBSIrD/+1+wcPVu2KZKJedzs36UHgN7Npd/fCKpHRRe0ypG4TWNF6MpvNboAQovhdcafUbhTSzKvC4JWIAAhdc0iBRe03gxmsJrjR6g8FJ4rdFnFN7EoszrkoAFCFB4TYNI4TWNF6MpvNboAQovhdcafWaS8E4Z2ROuLm9mePuOnCVfTSZelxT7KFu6cGKtm9clgRRHgMJrWskpvKbxYjSF1xo9QOGl8Fqjz4wW3irN+hu1Hv5pYaMwMYgELEKAwmsaRgqvabwYTeG1Rg9QeCm81ugzo4U3sRbD65IACbyfAIXXtO6g8JrGi9EUXmv0AIWXwmuNPqPwJhZlXpcELEAgyQuviz2O376FA9euQ/wZZdWHu6MTGuYpgIzOb0azkuvh7GgLJ3tbPH315o8D8bAMAf7hCctwfDsLhZfCq6aztLMa9YcntNMwggRIQDWBpC687q72cHa0w/PXYQgOjVCNS+a3gQ3Uq7VVbuW9F6HwquFP4VXDlcJL4VXTWdpZKbzajBhBArogkByE19VJCG84XodYR3h1UTjFi6DwqgFM4VXDlcJL4VXTWdpZKbzajBhBArogQOHVRRl0twgKr5qSUHjVcKXwUnjVdJZ2VgqvNiNGkIAuCFB4dVEG3S2CwqumJBReNVwpvBReNZ2lnZXCq82IESSgCwIUXl2UQXeLoPCqKQmFVw1XCi+FV01naWel8GozYgQJ6IJAchBeJwdbvAiKQHBopHqmNtFI9k+sAaDwqmklCq8arhReCq+aztLOSuHVZsQIEtAFgaQuvG7/ey3ZvuvXER2lFmlqO3vUy5MPOVKnUXshHWSn8KopAoVXDVcKL4VXTWdpZ6XwajNiBAnogkCSF15Xe2y9fBlt9+9WzjN76jRYXbUO8qZNp/xaiX0BCq+aClB41XCl8FJ41XSWdlYKrzYjRpCALghQeI0vA4XXeFaMjJ8AhVdNZ1B4KbxqOks7K4VXmxEjSEAXBCi8xpeBwms8K0ZSeK3ZAxReCq81+y32tSi8iUWe1yUBEwlQeI0HRuE1nhUjKbzW7AEKL4XXmv1G4U0s2rwuCXwAAQqv8fAovMazYiSF15o9QOGl8Fqz3yi8iUWb1yWBDyBA4TUeHoXXeFaMpPBaswcovBRea/YbhTexaPO6Sgms2nwQKzcdgKuLEzYtHINUqWyUXs/c5L//eQVTAzdg+YwhJqWg8BqPi8JrPCtGUnit2QMUXgqvNfuNwquI9vrtRyCk6/a9R0jt6ozKZYujf7em8j/r5bh28y6ioqKQ55NsckmrtxxEUHAourSqbfUljp2+Etv2HYeDgz3s7exQuuin8PFsifTp3Exey+ugENRoORD71kyCs5NjoshudHQ0arQchF0rxsPe3u699/C28B46/iuqlC2uec8UXk1EhgAKr/GsGEnhtWYPUHgpvNbsNwqvAtpL1+3F5j3HMG5wZ3xW4BM8efoSMxdvxo1b97B8hm+iCFh8tzl5/noULpgb1SuUlB/fufcIEZFRyJktkwIqCacUwpsrZxa0algFQcEhmDR3HV4FBWPS8B4mr+Xfa7fhMy4QmxeNMflcS57w869/4qvihRJMGVt4X7wKQse+E7Fp4WjNZVB4NRFReI1HxEgNAnwtmZoWofBSeNV0lnZWvqVBm5FmRHhEJMrV98KiKYOk7MYcUVHRqNveFz69WqJs6cJ49ToY/jNX4eSvFyE+Ez+vVbk0xK7rmGnLcfXGXbnbuWjyQNy8/QBrth7CvAl9DfkqNu6DLYv9kM49DboOnIzSxQpBCJaQxI+zf4SR/drD2ckBz56/wmD/BTJvREQkmtevJHdwdx/+GeNmrEQaVxe4pXHFpOHdcfDYGYSFhaNXhwYQO5SL1uzGlj3HEREZKXdcfb1bwcXZCULSFqzaicwZPXDj9n08fvIc7ZrWQP0aZeLwCQ4JQ+MuI95h5t2pkUGyYz6MLbziZ+cu/gu/aSuktC5cvUvujJ84fQGnzv6F9fNHImP6tJLfmd8vwdY2FZrUqYCOzWtByK7gd/HSdeT+OCs6taiFGhVLxVnD6XN/Y+KctXgdFAzARt5X2dJfJMhRJBDr2L7/J0RGRkkxb92oqsz7vloWqdoZ5w4ukjFb9x7HkrV7EBIWLnetJ4/ogexZMkqWYqRhwaQB6O4zBecuXkHeT7KhVJGC8l6HeLVC8S/yyxzi3ryGzcTe1RNB4dX8p0jhNR4RIym8idIDFF4Kb6I0nvhf/mhhOTw+iMDf/95Et0FT8MOWGe/kmTB7DZwcHdCnS2OMnLwUjg72GNSrBaIioxAWHiF//V6/w1B4dmyA6hVKSVkVMvrT6QuawitEdOqoXnL3eMy0FUjnnhpeHRtKmT538TKKFc6PR0+eo3bbIdizaiI80qaBz9hAVCpTzCCfQuhihHfXwZPYuOt7zJvQT4rz+FlrEBkZieF920pJa9nTD2vnDscXhfLg8dMXqNN2CL7fPF1KujlHbOEV65w4Zw0ypU+HgT2bS9EU4yFjfTrLLwviGDFpiWQjxkTEGIaQ/rZNqkluogbDA5Zg44JR8S6lcZeRGO/bBflyZcfzl69hY2MDt9QuMsf7OO47+gvWbT+CwID+8otDi55+8BvUEV8WyhNvLYWgxxbeS1f+Q+ZMHnBP44op8zcgOCQUw/q0MQivmOG9efs+PIfOxI5l4+S6xYiJkNyR/dvL/z5ryRbYwEb2B4XX+C7jSIPxrBgZPwHu8KrpDAovhVdNZ2lnpfBqM9KM+OWs2D1cE++v04W43bn/GCP7tUOpWt1xcP0UKUAxx1+Xb8B3/EJsXTI2znWOn/pDU3gb166AauVLyPOE8A3xX/BOHvFZq15jMbR3axTK/0mCwtvLdzoa1iyHymWLyZzPX7xGtRYDcGr3PClpwycuxo7l/oZ1CpGeO76v2eMQQnjF7qmQ67TuaVClbDF0b1NXCrTgdvnqLQQM7264nuB3aMNUKariELvTO/b/hFnjemsK70C/eUjrlho929eXO+QxhxDe93H0GjoDDWqWlV8QxDFn6VZERUfLLxXx1VLExBbe2AU9ceYCVm85hDn+fRIUXvGFp16HoTi0foqcA/6uzWDMHtdbjn5QeDX/KRoCKLzGs2IkhdeaPUDhpfBas99iX4vCawHyxuzwdm1dB2XqeeK3AwvjXPGHk79j7bZDmD+xv6bwVmjUB1uX/P9IQ4dmNfF1ic/keQ8ePZOjBMe2zsR/dx4gcOVO3H/4FDY2wB9/XcXCKQPxeYFcCQqv2AUVYl7409yGtRSt1gU/bZ+Ny9duYcr89Vgx09fwWd32QzF9jCdy58xi+JkYaWjQcdg7VPt2bSx3YmMfb480xP5MCG9oaLjc2RSHeCitbH2vOPzO/3kFY2esxIbAUe8Ib4kaXeV5WT/KICVd7AgvW78XW/YcQ/mvi6B/96ZyZ1cI7/s4Nu02Ck+evZS78uIIDQtH1XIlpPDGV8u3hVfslh869qscFRFfHjzSuckRldgzvG/v8Ioc3sNnon71MsiWJaP8MhQzl0zhNf4fK4XXeFaMpPBaswcovBRea/YbhdfCtMUMb/kG3lg8dRA+zfexIbsYLRDjCmKEoUypwihZsxsOb5xm2KEUgRcuXcOIgCVyNjf2cfLMRSzfuM8gwkKahHwe3TTNMMNbu+rXqFvtW3mamH8dPWWZ3OEVota2cXWIz8XRpOsojOzf7o3wjgtEpW/jH2no7jMVTWpXiLPDW7V5f/yyZ34cSYtZZ3zCawpaLeGNGbUQOcX9x+zwxuyQm7LDG7MukdNv+kq4pXHBwB7NpfC+j6On7ww0q1dRzvq+fcRXy9jCK77IzF66FUumDkKa1C74/sQ5rN9xNB7hfQDPoTMMIw0ix5Eff4MYp8ieNaOct+7QvKa8PIXX+O6i8BrPipEUXmv2AIWXwmvNfqPwKqC9bP0+bNl7HP5DOqNQvk/w7MUrOX959cYdLJs+WM6MDvFfKOds+3dvJn81/uLla7i7uaJ2myFSvsQogXgYSszkip3F9r3HY/eqiXKHcc/hUxC/lv9x+yyD8IqZ0MCAAXBwsEO/UXPkfKrYfazStB+mjfaUO7W/nv8HnfoHYNXsoVJ4x89aLc/v3raupBB7hnfngRPYsPN7BAaIGV5HiPljMWcsdn3je3esNYVXrFXM6KZxdZYzvjEzvOIhspqVSmuONIhd+IJ5c8p7FrPBYmfV17u1FN73cRTSuXbbYcz085Z1EuMGom5iFjq+WooH02JGGsQDa/u/Py0FV+wMi9lpwfLtHV4xT1yjxUAc3jhV7jiLQzwwWL/DMDjY22HO+L7IksmDwmviv1kKr4nAGP4OAc7wqmkKCi+FV01naWflSIM2I6MjxANOQqbu3n8MF2dHVC1fEv26NjG8h1f8Wlu8TUC8ZUAcgz1bSlkTDzeJtxMICROiKR6S+iRHZsxfsQMHfjiN9OncpayJOdBFUwYahLfI5/nkbuCTZy9Qquinhrc0iLcxTFuwUf4BhqKf5UNYeDhaNqwihVe8ucF72ExERkXJB96OnzpveGhNrEmMQmzYcRSpbFPJtwYIKRR59CC8YqxB8Dv1219y7rfRd+XlGxnElwmth9bEaMDPv/0JRwcH5MiaEf5DuiCDh7sU3vdxFDyWb9yPtVsPy9emiRlgf98ukuP7ahkjvCK+9/DZuH7rnrxOm0bVsPPgiXeEV1xDPNC269AJlPvqS4we0EH2hviy8ec/1+OMkHCH1+h/iqDwGs+KkfEToPCq6QwKL4VXTWdpZ6XwajPSZYQQNfGqsZJFCupyfUllUXrlOHrqcnyaNyea1q1oQEnhNb6rKLzGs2IkhdeaPUDhpfBas99iX4vCm1jkP/C6QtQ6t6yNUkUpvB+CUo8cxTuYO/SZIB+2E7vrMQeF1/hKU3iNZ8VICq81e4DCS+G1Zr9ReBOLtgWvq0dRs+DtWS2V3ji28fLHg0dP5ShJ+a+/jMOBwmt8W1B4jWfFSAqvNXuAwkvhtWa/UXgTizavSwIfQIDCazw8Cq/xrBhJ4bVmD1B4KbzW7DcKb2LR5nVJ4AMIUHiNh0fhNZ4VIym81uwBCi+F15r9RuFNLNq8Lgl8AAEKr/HwKLzGs2IkhdeaPUDhpfBas98ovIlFm9clgQ8gQOE1Hh6F13hWjKTwWrMHKLwUXmv2G4U3sWjzuiTwAQQovMbDo/Aaz4qRFF5r9gCFl8JrzX6j8CYWbV6XBD6AQJIXXhd7HL99CweuXZd/Klrl4WrvgLq58iJHGjeVl9FFbmdHWzjZ2+LpqzBdrCe5LIJ/eEJNJSm8FF41naWdle/h1WbECBLQBYGkLrzurvZwdrTD89dhCA6NUM/UxgZQ69Xq78GIK1B4jYBkRgiF1wxoRpxC4aXwGtEmSkIovEqwMikJWJ5AchBeVychvOF4HWIF4bV8CXSZkcKrpiwUXjVcKbwUXjWdpZ2VwqvNiBEkoAsCFF5dlEF3i6DwqikJhVcNVwovhVdNZ2lnpfBqM2IECeiCAIVXF2XQ3SIovGpKQuFVw5XCS+FV01naWSm82owYQQK6IEDh1UUZdLcICq+aklB41XCl8FJ41XSWdlYKrzYjRpCALggkB+F1crDFi6AIBIdGGsXUxiYail/oYNQ69BxE4VVTHQqvGq4UXgqvms7Szkrh1WbECBLQBYGkLrxu/3st2b7r1xEdZRzSPO5p0SRvfgA2xp2QAqMovGqKTuFVw5XCS+FV01naWSm82owYQQK6IJDkhdfVHlsvX0bb/buN5tkyfyFM+qYCIF4xxiNeAhReNY1B4VXDlcJL4VXTWdpZKbzajBhBArogQOHVRRl0twgKr5qSUHjVcKXwUnjVdJZ2VgqvNiNGkIAuCFB4dVEG3S2CwqumJBReNVwpvBReNZ2lnZXCq82IESSgCwIUXl2UQXeLoPCqKQmFVw1XCi+FV01naWel8GozYgQJ6IIAhVcXZdDdIii8akpC4VXDlcJL4VXTWdpZKbzajBhBArogQOHVRRl0twgKr5qSUHjVcKXwUnjVdJZ2VgqvNiNGkIBSAl0HTkbnlrVRqmjBBK9D4VVahiSbnMKrpnQUXjVcKbwUXjWdpZ2VwqvNiBE6JbD3yCnMXroVDx49hbtbarRsUBkdm9ey2mqfPn+JMvW80Lh2eYwe0CHOdU+euYjOAyZhRN+2aFavUoJrovC+Hw9fS6bdzhRebUbmRFB4zaGmfQ6Fl8Kr3SVqIii8argyq2IC9x4+QePOI7F0ug/y5cqO+w+fQghowbw5FV/5/9OL67X2HIfXQSHYv3YSHB3sDR8O9l+AK9fvoPF35Si8/6PixvfwKulNCq8SrKDwquFK4aXwquks7awUXm1GjNAhgfN/XoHf9JXYuGBUvKtbuHoXtu//CZGRUWjVsApaN6qK2/ceoZ23PzYsGA2PtGnw6/l/MHrqcmxeOBr29naGPIvX7sGWPcfi5M3g4Y7lM4bE+ZkQ3na9J6BAnuyoXKY4alQsJT8PCg5Fo84jULb0F8jzcRYpvM+ev4KQ4Gs37yIiIhLN61dCl1a1ZXzsHd6g4BCMmbYCv1/8F06ODvD1bo2SRd6MOnCkQYeNqIMlUXjVFIHCq4YrhZfCq6aztLNSeLUZMUKHBMLDI9Cw03DUrf4tWjeqBmcnB8Mq9x39Beu2H0FgQH8ply16+sFvUEd8WSgPVm85iPN/XsVYn05o3HUk/AZ2xBeF8ph1h0J423j5w6dXC6zZehjzJvSVeYRoX752C6GhYcj7STYpvFFR0Th38TKKFc6PR0+eo3bbIdizaqIU79jCK2Q3tYsT+nVriis37qBz/wDsXjkBLs5OFF6zqpT8T6LwqqkxhVcNVwovhVdNZ2lnpfBqM2KETgkI4Zy3fAcOHT+D7yp/jZ7t60vx9Ro6Aw1qlkWlMsXkyucs3Yqo6Gh4dWyI6OhotPUej9SuzsiVIzMG9Wph9t3FjDTsWOaPqs37Y+OC0Uifzg2d+gVgsFdLrN9+FPlyvRHet49WvcZiaO/WKJT/kzjC+1XtnnI8wj2NqzylXe/x6NGuHr4qVojCa3alkveJFF419aXwquFK4aXwquks7awUXm1GjNA5gVevgzFxzlo8fPwM8yf2Q9Nuo/Dk2UvDTG1oWDiqlishd2LFIR52GzBmHg6um4ysmTO8c3diHGLz7rgjDRnTu2PlrKFxYmOEV+zAiutn/Sg9qpYvAe9hM7EhcBTGTl9pEN7/7jxA4MqdctbYxgb446+rWDhlID4vkMsgvJ8XzIWSNbvhkxyZDdcR88FirKFa+RIUXp33YWItj8KrhjyFVw1XCi+FV01naWel8GozYkQSIPDiVRDKN+yNswcWwtN3BprVqyhnaN8+hPw27jwCXxUvJKV4ysieZt9dbOH96/INDA9YguoVSsrZ2zaNq8URXiHhbRtXR+2qX8vrNek6CiP7t4sjvOK1ZGKH98jGaXBxdnxnXZzhNbtUyfpECq+a8lJ41XCl8FJ41XSWdlYKrzYjRuiQgNgxFQ+kfZz9IzkfK2Z29xz+GavnDIOY4V277TBm+nnD3c1VPjAmRhrEvOzk+evl3fTv1lSOC7RrUgOVy74ZfTD1iC284ty67YfixcvX2LxojBxtiL3DW6VpP0wb7YnCn+aWD8t16h+AVbOHviO8YoZXvO1BzPDa29ni5u0HyJLJQz5UR+E1tUIpI57Cq6bOFF41XCm8FF41naWdlcKrzYgROiQgpHHk5KXyHbwO9vb48rM88lf/2f43orB8436s3XoY4q0Had1Sw9+3i7wLn7GBUkjFLuz1/+7JcYKNC0cbZmZNuawGmUgAACAASURBVNW3hXfRmt347Y9/MHf8m4fXYgvv7sM/Y9qCjXB1cULRz/IhLDwcLRtWeUd4xXonzV2HYz+flzFivCEwYIDc8aXwmlKdlBNL4VVTawqvGq4UXgqvms7Szkrh1WbECBLQBQEKry7KoLtFUHjVlITCq4YrhZfCq6aztLNSeLUZMYIEdEGAwquLMuhuERReNSWh8KrhSuGl8KrpLO2sFF5tRowgAV0QoPDqogy6WwSFV01JKLxquFJ4KbxqOks7K4VXmxEjSEAXBCi8uiiD7hZB4VVTEgqvGq4UXgqvms7Szkrh1WbECBLQBQEKry7KoLtFUHjVlITCq4YrhZfCq6aztLNSeLUZMYIEdEGAwquLMuhuERReNSWh8KrhSuGl8KrpLO2sFF5tRowgAV0QoPDqogy6WwSFV01JKLxquFJ4KbxqOks7K4VXmxEjSEAXBCi8uiiD7hZB4VVTEgqvGq4UXgqvms7Szkrh1WbECBLQBYEkL7wu9jh++xYOXLuO6Ohoo5jmck+HJnnyAzY2RsWnxCAKr5qqU3jVcKXwUnjVdJZ2VgqvNiNGkIAuCCR14XV3tYezox2evw5DcGiEUUyjYQOqbsKoKLxGtZLJQRRek5EZdQKFl8JrVKMoCKLwKoDKlCSggkByEF5XJyG84XgdYpzwquCY3HJSeNVUlMKrhiuFl8KrprO0s1J4tRkxggR0QYDCq4sy6G4RFF41JaHwquFK4aXwquks7awUXm1GjCABXRCg8OqiDLpbBIVXTUkovGq4UngpvGo6SzsrhVebESNIQBcEKLy6KIPuFkHhVVMSCq8arhReCq+aztLOSuHVZsQIEtAFgeQgvE4OtngRFIHg0Mj3MxVvcOCTakb3HIXXaFQmBVJ4TcJldDCFl8JrdLNYOJDCa2GgTEcCqggkdeF1+99ryfZdv47oqPgppXVwROP8BZDB0VkVxmSXl8KrpqQUXjVcKbwUXjWdpZ2VwqvNiBEkoAsCSV54Xe2x9fJltN2/+708C6T1wKqqdZA1dWpdME8Ki6DwqqkShVcNVwovhVdNZ2lnpfBqM2IECeiCAIVXF2XQ3SIovGpKQuFVw5XCS+FV01naWSm82owYQQK6IEDh1UUZdLcICq+aklB41XCl8FJ41XSWdlYKrzYjRpCALghQeHVRBt0tgsKrpiQUXjVcKbwUXjWdpZ2VwqvNiBEkoAsCFF5dlEF3i6DwqikJhVcNVwovhVdNZ2lnpfBqM2IECeiCAIVXF2XQ3SIovGpKQuFVw5XCS+FV01naWSm82owYQQImEfj9zyuYGrgBy2cMMek8rWAKrxahlPk5hVdN3Sm8arhSeCm8ajpLOyuFV5sRI6xEoG77obh7/zFSpbKBi7MjqpUviYE9m8PO1tZKK4j/MkJgh01YhJ0rxhsC+o+ei8jIKEwf42n4Wad+AWhevxIyZUj3XuGNjo5GjZaDsGvFeNjb2+HHX/5A8S8KwNnJQfMeKbyaiFJkAIVXTdkpvGq4UngpvGo6SzsrhVebESOsREAI75SRPZAvV3Y8fvoCfUbMRq3KpdGifmUrrSD+y0RERuLr2r2wf+0keKRNg6ioaNRsNQiRkZHYu2YS7O1sERNzcN1k3Lh9P8Ed3p9//RNfFS8EIb8tevph3oS+SOeeRvMeKbyaiFJkAIVXTdkpvGq4UngpvGo6SzsrhVebESOsRCC28IpLLlqzGw8ePYOvdyuc//MK/GeuwtPnr5AqVSr4erdG2dKF5cqu3byLMdOW4+qNu3BwsMeiyQPxcfaPcOzn3zE1cCNCQsNQqmhBDOvdRn4e+xBSffnarTg/q/htUQzo3izOz7r7TEHDWuVRrXwJiB3fecu3wc7WDq0bV8VXxQrJn42esgxbFvvJ/zx94Ubk+Tgbvj9xVq7Xq2ND1Kn2jcxZpGpnnDu4CJPmrsOqzQeR55OscEvjimXTByMoOARjpq3A7xf/hZOjg7zPkkUKyvMovFZqxCR2GQqvmoJReNVwpfBSeNV0lnZWCq82I0ZYiUBs4b374Am8h81Ej7Z1UalMMSm+IaGhyJntIxw/dR6T5q3HjmXj5FhB/Q5D4dmxAapXKIVnz19Jebz/6Cla9xqLVbOHInMmD/iOXyTFsnPL78y6myXr9uDeg6dSvucu2wYnJwe4ODvh5q37GNSrBRav3YNHT57Dp1cLKbwte/ph1rjeqPRtUVz/7x6a9xiDY1tmSOGOEV6xkIqN+0hJjtnhFbKb2sUJ/bo1xZUbd9C5fwB2r5wgr0XhNat0yf4kCq+aElN41XCl8FJ41XSWdlYKrzYjRliJgBDeR0+ewcbGBi9evsak4T1Qo2Kpd64eHhGJb+r0wum98/HX5RvwHb8QW5eMjRO3YuN+3Ln/GIM9W8qf/3L2b8xasgUrZ/madTcXLl3DyElLsXnRGDmGMHpABymmXQdNkfO4PQZPQ9O6FVDxm6JSeIf4L8CeVRMN1/quzWDMm9APObNlSlB4v6rdU45OuKdxlee26z0ePdrVk7vIFF6zSpfsT6LwqikxhVcNVwovhVdNZ2lnpfBqM2KElQjE3uFt6+0vZ3drViotr37q7F9YveUgQkLC5H//7Y9/cGbfAvxw8nes3XYI8yf2j7PKSfPWYfu+n+Du9kYco6KipESumz8yTpzXsJm4fDXuSEPlMsXkw3KxD7GTXLa+FzYv9pM7x4c3TpUf123niznj+8gd3H2rA5AmtYsU3inz12PFzP+Xa3Fv4gG33DmzvFd4g4JDUbJmN3ySI7Ph0q+DQuRYgxiloPBaqRGT2GUovGoKRuFVw5XCS+FV01naWSm82owYYSUCsYX3zO+XMDxgMXYsH4+goBD5kNjqOcOQK2cWOedaroG3FF6x8zoiYIkcC4h9LN+4H48eP0f/7k0ttvpevtORP3cOPHz8DGN9Osm8Yg5XvG3h5JmLWB/4Rqbjey3Z+4S3UpO+ctc4ZqRB7PAe2ThNvqXi7YPCa7FSJqtEFF415aTwquFK4aXwquks7awUXm1GjLASgbcfWhOv+apavgTKlCqM5t3H4MC6yVIExbzs/BXbcXpvoHw7Qu02QzCwR3NULlsMr14Hy9eaPX/xGm28/TF/Qj/kzZUNoWHhUlSzZ8lo9t0IiV68ZjeGeLUy7DyLNy4M9JuH+jXKGuTaFOFt2Gk4hvVpg2KF88t1iRleRwd7OcMr3v5w8/YDZMnkIaWawmt26ZL1iRReNeWl8KrhSuGl8KrpLO2sFF5tRoywEoG3hffcxX/Re/gs7F0dgDlLt2LPkZ/hkdYNTepUwIYdRw27upeu/Ae/aStw8/Z9ODs5IjCgvxwLELuuk+evl6LrYG+Hnu3ro2Gtcmbfzd//3kTjLiPx0/bZhlGJ8PAIfFPXE1NH9TK8NcIU4T3wwxlMmL1ajluIOWSxey12jY/9fB5h4eHyPgIDBkjRp/CaXbpkfSKFV015KbxquFJ4KbxqOks7K4VXmxEjSEAXBCi8uiiD7hZB4VVTEgqvGq4UXgqvms7Szkrh1WbECBLQBQEKry7KoLtFUHjVlITCq4YrhZfCq6aztLNSeLUZMYIEdEGAwquLMuhuERReNSWh8KrhSuGl8KrpLO2sFF5tRowgAV0QoPDqogy6WwSFV01JKLxquFJ4KbxqOks7K4VXmxEjSEAXBCi8uiiD7hZB4VVTEgqvGq4UXgqvms7Szkrh1WbECBLQBQEKry7KoLtFUHjVlITCq4YrhZfCq6aztLNSeLUZMYIEdEGAwquLMuhuERReNSWh8KrhSuGl8KrpLO2sFF5tRowgAV0QoPDqogy6WwSFV01JKLxquFJ4KbxqOks7K4VXmxEjSEAXBJK88LrY4/jtWzhw7Tqio6PjZeru6ISGeQogo7OzLpgnhUVQeNVUicKrhiuFl8KrprO0s1J4tRkxggR0QSCpC6+7qz2cHe3w/HUYgkMj3svUBjaIX4d1UQbdLYLCq6YkFF41XCm8FF41naWdlcKrzYgRJKALAslBeF2dhPCG43XI+4VXF7CT0CIovGqKReFVw5XCS+FV01naWSm82owYQQK6IEDh1UUZdLcICq+aklB41XCl8FJ41XSWdlYKrzYjRpCALghQeHVRBt0tgsKrpiQUXjVcKbwUXjWdpZ2VwqvNiBEkoAsCFF5dlEF3i6DwqikJhVcNVwovhVdNZ2lnpfBqM2IECeiCQHIQXicHW7wIikBwaOR7mPJxNVObjcJrKjHj4im8xnEyNYrCS+E1tWcsFU/htRRJ5iEBxQSSuvC6/e+1ZPuuX0d01Luw0jk6oVm+gkjr6KiYZPJKT+FVU08KrxquFF4Kr5rO0s5K4dVmxAgS0AWBJC+8rvbYevky2u7fHS/PQh4ZsKZ6XWR04jt4TWk4Cq8ptIyPpfAaz8qUSAovhdeUfrFkLIXXkjSZiwQUEqDwKoSbhFNTeNUUj8KrhiuFl8KrprO0s1J4tRkxggR0QYDCq4sy6G4RFF41JaHwquFK4aXwquks7awUXm1GjCABXRCg8OqiDLpbBIVXTUkovGq4UngpvGo6SzsrhVebESNIQBcEKLy6KIPuFkHhVVMSCq8arhReCq+aztLOSuHVZsQIEtAFAQqvLsqgu0VQeNWUhMKrhiuFl8KrprO0s1J4tRnpPmL99iO4fO02hvVpo/u16mmBZ36/hFFTliE8PAKz/XsjX67sZi+vSNXOOHdwkdnnv+/E2HkpvBbHmywSUnjVlJHCq4YrhZfCq6aztLNSeLUZJUrE/u9PY5DffDg5ORiu792pEVo1rPLOelQKb1hYOE6d/QtlS38hr7t6y0EEBYeiS6vaVucydvpKbNt3HA4O9rC3s0Ppop/Cx7Ml0qdzM2stLXv6oV+3pij+RX55vo2NjVl5xEkUXm10bnwtmTYkMyIovGZAM+IUCq8RkMwIofBSeM1oG4ucQuG1CEbLJxHCu/fIKUwf46mZXKXwHvv5dxz7+bxh9/jOvUeIiIxCzmyZNNdl6QAhvLlyZpHSHxQcgklz1+FVUDAmDe9h1qW+reeJ/WsmIbXrh7/3lcKrXQIKrzYjcyIovOZQ0z6HwqvNyJwICi+F15y+scQ5FF5LUFSQIyHhffz0BYZOWISrN+4gcyYPFCucH69eB0spPX7qD6zZegjzJvQ1rKpi4z7YstgP6dzT4NrNuxgzbTmu3rgrd0oXTR6Ij7N/BP+Zq6TYRkRE4ItCeTBxWHcZ23fkbDx/8RpZPkqP9s1qQAiv2PXt1aGBzP/Dyd8xbcFGKZ4fZ/sII/u3N8hwlab90KZJdZw8cwGPnrxA0c/zYmjvd8cu+oyYjcvXbsWhWPHbohjQvVmcn8UWXvHBuYv/wm/aCmxeNAYLV++S4nri9AW5I71+/kh8kiMzFq3ZjS17jiMiMlLuCPt6t0KqVKkwPGAx9h39Bflz58A3JT5H/+5N41zr/sOn8B2/EHcfPJb326RORXRrUwfR0dEy56ZdP0jprlX5KwzxaoWSNbtjsGdLuY6w8HBULVdC/lwcMee8vQ4XZydNhhxpUPCPK5mlpPCqKSiFVw1XCi+FV01naWel8GozSpSIhIR3wJh5yJE1E3p3boQnz15C/Gq+TKnCmsLrltoV9TsMhWfHBqheoRSePX8FtzSuSJXKBmcvXEbhT3PLe23fewLaNK6G6hVKSrH7+9+bhh1eKXT/E95bdx+irbc/lk0fIiX3wA9nMHfZNinXIqeQNa+ODdGpRS1ERkbJdQ7s2RwlvixgFtPYwvvoyXNMnLMGmdKnkznFulZtPoixPp1RtnRhmX/XwZPYuOt7zJvQD85ODhg/aw0iIyMxvG9b+Xnp73rg6KZpiBHP2IsKmLMW2bJklLvJYsZXcP4oYzrsOnQSa7cexryJ/ZDG1RnPXrySXySKVuuClg0qS0kPDglD8+6jpfyLcYmE1mEMw5jZYM7wmtU2yf4kCq+aElN41XCl8FJ41XSWdlYKrzajRImIb4Z38dRB+LxALpSo0RVHNk2HW2oXubaZizfjxcsgTeG99+CJ3LXcumRsgvc0feEmpHVPjfZNayQovMs27MPd+48NO5kiaf0OwzBhaFcUzJtTCu+RjdPgkTaNvN7oqcvl+ht9V84spkJ4t+//ScprWvc0qFK2GLq3qSt3qoXwXr56CwHDuxty9/KdjoY1y6Fy2WLyZ2KnulqLATi1e56m8G7Y+T32HP4ZI/q1Q+6cWQw5PX1noH7NMqhStnicexD3enDdZGRMn/bNvU5Zhs8L5pb3mtA6jGFI4TWrXVLMSRReNaWm8KrhSuGl8KrpLO2sFF5tRokS8b4dXvFr9DL1vPDbgYWGda3eckiOH7xvpKFCoz7YusQP5/+8irXbDmH+xP5x7knsSAau3IELl64hlY0Nbt5+gGZ1K6JD85oJCq/YBU3v4S53cGOOboOmoGWDKij/9ZdSeM8eWGh4GEwIa75c2dCsXqU41/caNlPKauyjcplicuc29vH2SEPsz4TwhoaGy93rmKNxl5EY2a+dYeda/FzsxP60fTZcnB3j7PCKdf96/pI8defy8XKEY9u+H7F0/V5kzuiBob1bI2e2j9C02ygM79M2Tk5xTkL3mtA6Zi/ZosmQwpso/wSTzEUpvGpKReFVw5XCS+FV01naWSm82owSJSKhkYbi1bvi2NaZcHV5MwM6d/l2PHn6QgrvyTMXsXzjPoPUivlRIXniV/e37z3CiIAlcuQg9jFp3jr55gUhdXa2tvJhsAwe7lJ4N+8+hr8u34h3pGHJuj0Qs64xs6oi59s7vLFf1fU+4TUWsJbwxp4tFjm7+0xFk9oV4uzwVm3eH7/smS8vmdBIQ+w1CQbrdxzBhsBR6DF4Ghp/V96QMybu7YfWYt9rQuswhSFHGoztlJQVR+FVU28KrxquFF4Kr5rO0s5K4dVmlCgRCQlvv1FzkS93NvRoW0/O4bb2GoevihWSUipmQtv3Ho/dqybC0cEeew6fwkC/efhx+yykSe2C2m2GYGCP5lLYxINuYtbWb9pK5M2VTe7UPnj0DO16+6NpnTc7vIeP/4Y12w5h8ZRBkkPsGV6xE9y+z3gsnTZYPvh28NgZzFqyFduWjDXM8Cam8O48cAJiNCEwQMzwOmLC7DUIC4+Qu75awnvl+m05Jy3GJS5d+Q8+YwOxbelYbN17XD4EJx4KFA/JCeEXs70JCW9C6zCFIYU3Uf4p6v6iFF41JaLwquFK4aXwquks7awUXm1GiRIhhHfAmLlSWmOOPl2aoHWjqhAPbIm3NAgpy5A+LSp+UxQPHz8z7MLOX7EDB344jfTp3OUs7YkzF7BoykD5cJWQN/Fmg5u370sJDAzoLyVQSHFERCSyZ8mI0sU+hQ1spPCKXdMeQ6bhxq37cl726fOXcd7ScOSns5gauAEhIaHIkS0TRvXvIOVXHAlJoDlQTd3hFdcIXLkTG3YcRSrbVChVpCB8vVsbdsYT2uFdu+0wFq/ZDXt7O6R2dZG72MUK50NUVDRmL92CHQdOIDQ0DLWrfgOfXi007zWhdRjLkMJrTtck/3MovGpqTOFVw5XCS+FV01naWSm82owYQQK6IEDh1UUZdLcICq+aklB41XCl8FJ41XSWdlYKrzYjRpCALghQeHVRBt0tgsKrpiQUXjVcKbwUXjWdpZ2VwqvNiBEkoAsCFF5dlEF3i6DwqikJhVcNVwovhVdNZ2lnpfBqM2IECeiCAIVXF2XQ3SIovGpKQuFVw5XCS+FV01naWSm82owYQQK6IEDh1UUZdLcICq+aklB41XCl8FJ41XSWdlYKrzYjRpCALghQeHVRBt0tgsKrpiQUXjVcKbwUXjWdpZ2VwqvNiBEkoAsCFF5dlEF3i6DwqikJhVcNVwovhVdNZ2lnpfBqM2IECeiCAIVXF2XQ3SIovGpKQuFVw5XCS+FV01naWSm82owYQQK6IJDkhdfFHsdv38KBa9ch/uT120c6Jyc0yfsp0jo66oJ3UlkEhVdNpSi8arhSeCm8ajpLOyuFV5sRI0hAFwSSuvC6u9rD2dEOz1+HITg04j1MbXTBOiktgsKrploUXjVcKbwUXjWdpZ2VwqvNiBEkoAsCyUF4XZ2E8Ibjdcj7hFcXqJPUIii8aspF4VXDlcJL4VXTWdpZKbzajBhBArogQOHVRRl0twgKr5qSUHjVcKXwUnjVdJZ2VgqvNiNGkAAJkAAJkAAJkAAJJGECFN4kXDwunQRIgARIgARIgARIQJsAhVebESNIgARIgARIgARIgASSMAEKbxIuHpdOAiRAAiRAAiRAAiSgTYDCq82IESRAAh9I4I+/r2HYxEV4+PgZCubJiYnDuiFj+rQfmDVlnb5173HMWrIFYWERqFK2OIb3bQtb21RxIDx68hxT5m/AT6f/gJ2dLcqW/gLD+rSFvZ1tyoJl5N1GRUVjwuw12HP4Z9jb26Jbm7poXq9SgmdPX7gJB4+dwe6VE4y8SsoLe/r8JYb4L8D5v67CI60bxgzsiGKF88UL4uSZi5ixeDMePnqGrJkzYOUs35QHjHdsFQIUXqtg5kVIIOUSiIyMQq3WPlK8ypYujFWbD+Lkrxcxx79PyoVi4p1f/+8eOvabiFWzhiJjhnTwGRuILwrlRvumNeJkOnfxX4jY2lW/RlRkFLyGzUSFb4qgRf3KJl4xZYRv3n1Myu6c8X0QFByKNl7jEDCsOz4r8Em8AMQXt9lLtuDW3YcU3gRaxGdcILJlzgDPDg1x4e+r6Dd6LnatGA8nR4c4Z53/8wqGBSzB5BE9kD939pTRdLzLRCNA4U009LwwCaQMAn/8dRXjZ63GmrnD5Q2LXbXyDb2xZ9VEpEntkjIgfOBdLl67By9fBaFPl8Yy09//3sTwgCXYuGBUgplXbNyPW3cfwde71QeuIHme3t1nCto0ro5vS34ub3DlpgO49+AJBvZs/s4Nh4WFo42XP4b3ayu/cHCHN/6eEP++v6nbC0c3TYez0xvBFV+8GtUqJ798xT68h89Ek9oV5G8ieJCAagIUXtWEmZ8EUjiBXQdPyh3dcYM7G0g07zEGQ3u3QeGCuVI4HeNuf8SkJShWOD/q1ygjTwgNC8c3dXrh1/0LEkzQb9RcVPq2qNzx5fEugZqtBmHxlEHyV+niOH7qD6zechDzJ/Z7J1iMimTO5IFalUujtec4Cu97Gur+w6do7TUOB9dNNkRMDdyAtO6p0bF5rThnCTHu3akRNu0+hqioKDSrWxFN61Zkq5KAEgIUXiVYmZQESCCGwMZd3+Ovyzcxom9bA5T2fSagR7t6KF30U4IygsAgv/moVKYYalQsZYj+rEJ7XDi6FDY28f855hNnLmDmos1YNWcY7Gw5wxsf5gqN+mDzojFIn85Nfnz63N9yTnrFzLhzpOJX72LOdNHkgXj24hWFN4GevXHrPjx9p2PnivGGqLnLtsnf7Hh2bGD4mfjSVrx6V3Ru+Z38/wVBwSFo6+Uvvxh/USiPEf8qGEICphGg8JrGi9EkQAImEth16CSO/3xePqgWczTqPAKj+rdH4U9zm5gtZYaPnLwUhQvmRuPa5SWAV6+DUa6BN347sDBeIGLW1Nd/ARZMHogsmTxSJjQj7rpmKx8EBvRDzmwfyegjP53Fhh1HMH9i/zhiJkRs6uheci5VPJDFHd73w33w6BmadR8lRxpijolz1iKDhzs6tfj/Hd7w8AiUrNkNv+yZDwcHexk6b8V2+eWsS6vaRlSPISRgGgEKr2m8GE0CJGAigT//uY6Rk5cZ5k0jIiNRpp4X9q+ZBHc3VxOzpczwZRv2yafYY2ZLhdCOnLQEWxb7vQPk0pX/MHDMPEz380LunFlSJjAj77qX73Q5QxozW7pk3R6IX8kP8fr/medfz/+DrgMny7deiCM6Olo+4Obq4oRtS8Yiy0fpjbxayggTfL6t64l9ayfB7X8z+t19pkrOlcsWiwOhUpO+WD9/pOGNLTMWbZajD+2aVE8ZsHiXViVA4bUqbl6MBFIeAfGrzDrthkiJKFPqzVsajvz4G5ZM80l5MMy849v3HqGdt7/8VfubtzTMR77c2dGjbT0cP3Ued+8/lrOPV2/eRb+RczBlVE/k+TirmVdLOaeJ+XLxureYtzS06jUWY306ofgX+bF03V58+VkeOTsd++AOr3Z/iJnz9Onc4dXxzVsaPIfOwN7VAXB2csTYGSvRs109ueM7c/Fm3L3/RDKP4TpzrDff2KCNmBFmEKDwmgGNp5AACZhGQOw6+o5fKMUs98dZMWFoV2TPktG0JCk8es/hU5gyfz2CQ0PlU+1+AzvKXwWLXcl/rt7CBN+uGOg3DyIuVar/n+t1dLDHmX0JP9yWktGKh9G27TsuZ6HbN6theLCq55BpqPRtMcMYSQwjCq92t7x4FQRf/4X47cI/cEvtKt8ZLd6EIeZ2a8kxkv7ImysbQkLDMGrKMoh38QoZ7tamDhrULKt9AUaQgBkEKLxmQOMpJEACJEACJEACJEACSYcAhTfp1IorJQESIAESIAESIAESMIMAhdcMaDyFBEiABEiABEiABEgg6RCg8CadWnGlJEACJEACJEACJEACZhCg8JoBjaeQAAmQAAmQAAmQAAkkHQIU3qRTK66UBEiABEiABEiABEjADAIUXjOg8RQSIAESIAESIAESIIGkQ4DCm3RqxZWSAAmQAAmQAAmQAAmYQYDCawY0nkICJEACJEACJEACJJB0CFB4k06tuFISIAESIAESIAESIAEzCFB4zYDGU0iABEiABEiABEiABJIOAQpv0qkVV0oCJEACJEACJEACJGAGAQqvGdB4CgmQAAmQAAmQAAmQQNIhQOFNOrXiSkmABEiABEiABEiABMwgQOE1AxpPIQESIAESIAESIAESSDoEKLxJp1ZcKQmQAAmQAAmQAAmQgBkEKLxmQOMpJEACJEACJEACJEACSYcAhTfp1IorJQESDIAnUgAAC2lJREFUIAESIAESIAESMIMAhdcMaDyFBEiABEiABEiABEgg6RCg8CadWnGlJEACKZDA7XuPUK35APx2YCEcHex1R2D7/p/w6nUwWjWsoru1cUEkQAIkEEOAwsteIAESIAEdE9C78PYfPRfFCuen8Oq4h7g0EiABgMLLLiABEiABHRN4W3gnzV0HVxcn3Lh9H2f/uIzQsHA0qV0B5b76Av6zVuPx0xdwcrDH0N5t8FXxQvLOxDnOTo64fuse/rnyH54+f4lalb/CoJ4tYGubSsaEh0dg2oKN2HXoJF68CkLhgrkxtHdrFMyb05Aja+b0CAoOxZK1e2Ru9zSpsXXvcdjb28LBwR4NapTFwJ7NcfHSdUxfuAkX/7mGqKholClVGH6DOsHZyQFBwSEo18AbE4d2x6wlW/A6OAQuzo4Y2KO5jIs5fj3/DybNW4e//72JNK7O6Nq6Dto0ribXOWneeuw6eAKwASp8XQRDvFohTWqXd6r44NEzjJm6HKd//1uuo0CeHPAb1BG5cmaRsRt2fi/v5d6Dx/goowf8h3RB8S/ym8Vi2mhPzbVt3PXmencfPEH6tG5oWKssenVooOPu49JIIPkQoPAmn1ryTkiABJIhgfiEd9Xmg1g63UfurP577TYadR6Bj7N/hAWTByBzRg/sOPATpgZuxNFN02BjYyOFd+XmA1gQMECK6rPnr9Cx30TUq1EG7ZpUl9SmzN+A0+f+gr9vV2TwcJciu2j1LuxbM0kKtshx/Jc/UDBPDvRoV08KpojrOnAyyn9dJM4O783bD3Dz9n0UK5wPwSFh6DF4KmpWLI0OzWtK4S1Zszu+KlYIM/y8kNrVGYeO/4oRAUtwdPN0Obbx6Mlz1Gw1CL3aN0DNSqXx8lWQFHMhqkLK/71+GyP6toODg50UWjs7W0wa3uOd6g8YMw/p3FOjb9cmCI+IxJlzl1CmdGF5jR9O/o4h/gsw1qcTvvwsr1xv7pxZ4e7majaLhNZ29eZdtOrph2UzhiBXjsz4784DPH8ZJBnxIAESUE+AwqueMa9AAiRAAmYTiE94/7p8A0um+RhyNug4DNUrlEL3tnXlz0JCw1C8elf8sGWGlFIhq39evo6l0wYbztn//S+YuXgLdq+cIHcmS33XAytmDEHhT3MbYtp6+6Na+ZJo3aiqzHHg2BnsWTUR9na2hpj4hPftmw1cuRM3bt2TO6gxwrt46iApvTGHWO+GwJHI80k2zF2+Xe5eL5w8IE4qIa1ffdcDu1ZOQJZMHvIzIfwNOw/Hr/sXxlmX+KzfqDnI4JEWvt6t3uHfoe8EfFPic3RpVTvuNcxkobW2m7fuoUVPP2xdMhbZMmcwux94IgmQgHkEKLzmceNZJEACJGAVAvEJ77MXrzBucGfD9Zv3GIPm9Sqhfo0yhp99VqE99q6eiJzZPpKy+vzla7mbGXPE7AyfPbhI7m5+12Ywzh1cBHt7O0PMxDlrERwcilED2sscj54+x8Sh3eLcd3zCe+HSNSxdtxfX/7uH6OhoOWZR4ssCmDKyp0F4D66bjKyxxO+bur0wb0I/fFkoDwb6zUO2zBnRp0vjONe6ces+arX2ibNGERAZGYkD66YYJDjmpFt3H8Jr6AxERUejZf3KckfbydFBfly+YW8p4N+W/DzONcSazWFhzNoWrt6FRWt2o2q5EvJLRMy4iFUaiRchgRROgMKbwhuAt08CJKBvAvEJr9glHdm/fRzhbdWgCupU++a9wvvw8TMEDO9u+FzMxjbvPhq/HVgkRwTELvHZAwvlLG7MMWH2GoSEhBmE9+3riri3hVfMzdZuOxjD+rTBd5W/lqMIYsf2yvXbcYQ3Zvc55lqxhVc8CJcja6Z3hFeMBdRpOwSn9wbKuV9jDiHcP/5yAUvW7cbd+0+wbMZgOfYh5ognDO0qd3ljH/9cvWUWC2PXJsY11m8/gpWbD6JTi1rv7DAbc0+MIQESMJ0Ahdd0ZjyDBEiABKxGwFLCe+rsX9i0cLRh3Zt2/YBlG/Zh14rxcgTi27qeWDrNB18UymOIESMNlcsWl3O+Yoc3PuHt7jNV7pKKB8rEceTH3+TDc4fWTzHkGeQ3H5FRUUYLr3iY7cLf1xAY0D8O55jRi9njer+zM2tMQXoOmYain+eTktnGyx/lv/4SnVt+F+dUc1mYujbxYF8rz7HyS4aYs+ZBAiSglgCFVy1fZicBEiCBDyJgKeFdveUgBvZsgbrVvoF4qKz38Jlo26S6/D9xCKH99fwl+dBaRg93bNl7HIErd8iH1txSu7xXeMdOXwmxxuljPOWbEMSsrtg5XjlrKPLlzi4fSPOfsQpfl/jMaOG99/AJvms9GP26NUWtyqXle37FSMbnBXLJtz/s//60HM8Q+e8/fIK/L9+Ms7sdA1zMKRfM+zGyZcmAh4+eQQhvp5bfoXaVr2WOMdOWyxENIfl37j2SD9Blz5LRbBYJrU08pHb1xl35kJoYq9i0+wes2HhAjp3wIAESUE+AwqueMa9AAiRAAmYTsJTwiteXvQ4KwaHjZ2BrayvnfcWrwGJeSxYRGYnZS7bKtzMIuSxcMBcGe7bCZwU+MQhxfDu8QhTFw2GXrt5Ck9rl4evdGvNX7JBvhQgNDUOFb4rKB9+EfMae4U1opEFc8OSZi5gSuAGXr96Sb4To1aE+WtSvLN+2MGfpVmzb9yOePnuJ9B5uaFizHDw7vvt6L/Gw3Lrth2WcRzo31KteBt6dGhp2VFduOoAVmw7gwcOnyJzJAxOHdUORz/LCXBYJrU18yRg8LhCXr92WPD/Nl1O+Tu3TfB+b3Rs8kQRIwHgCFF7jWTGSBEiABJIkATmOEBKKkf3aJcn1c9EkQAIk8KEEKLwfSpDnkwAJkIDOCbxv/lbny+bySIAESMBiBCi8FkPJRCRAAiSgTwIUXn3WhasiARKwHgEKr/VY80okQAIkkCgEKLyJgp0XJQES0BEBCq+OisGlkAAJkAAJkAAJkAAJWJ4AhdfyTJmRBEiABEiABEiABEhARwQovDoqBpdCAiRAAiRAAiRAAiRgeQIUXsszZUYSIAESIAESIAESIAEdEaDw6qgYXAoJkAAJkAAJkAAJkIDlCVB4Lc+UGUmABEiABEiABEiABHREgMKro2JwKSRAAiRAAiRAAiRAApYnQOG1PFNmJAESIAESIAESIAES0BEBCq+OisGlkAAJkAAJkAAJkAAJWJ4AhdfyTJmRBEiABEiABEiABEhARwQovDoqBpdCAiRAAiRAAiRAAiRgeQIUXsszZUYSIAESIAESIAESIAEdEaDw6qgYXAoJkAAJkAAJkAAJkIDlCVB4Lc+UGUmABEiABEiABEiABHREgMKro2JwKSRAAiRAAiRAAiRAApYnQOG1PFNmJAESIAESIAESIAES0BEBCq+OisGlkAAJkAAJkAAJkAAJWJ4AhdfyTJmRBEiABEiABEiABEhARwQovDoqBpdCAiRAAiRAAiRAAiRgeQIUXsszZUYSIAESIAESIAESIAEdEaDw6qgYXAoJkAAJkAAJkAAJkIDlCVB4Lc+UGUmABEiABEiABEiABHREgMKro2JwKSRAAiRAAiRAAiRAApYnQOG1PFNmJAESIAESIAESIAES0BEBCq+OisGlkAAJkAAJkAAJkAAJWJ4AhdfyTJmRBEiABEiABEiABEhARwQovDoqBpdCAiRAAiRAAiRAAiRgeQIUXsszZUYSIAESIAESIAESIAEdEaDw6qgYXAoJkAAJkAAJkAAJkIDlCVB4Lc+UGUmABEiABEiABEiABHREgMKro2JwKSRAAiRAAiRAAiRAApYnQOG1PFNmJAESIAESIAESIAES0BEBCq+OisGlkAAJkAAJkAAJkAAJWJ4AhdfyTJmRBEiABEiABEiABEhARwT+D2tdQpgskFepAAAAAElFTkSuQmCC"
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "explanations = explainer.explain(test_x)\n",
    "explanations.ipython_plot(index=0, class_names=class_names)"
   ]
  }
 ],
 "metadata": {
  "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": "3.7.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}