{
 "cells": [
  {
   "attachments": {},
   "cell_type": "markdown",
   "id": "a48e9306-f459-4d8a-8608-9bd71a7600ae",
   "metadata": {},
   "source": [
    "# 6. Exploring Advanced Features\n",
    "\n",
    "This notebook is target at advanced users that want, among other things, access MOA objects directly using the Python API from capymoa. \n",
    "\n",
    "* Examples on how to use any MOA Classifier or Regressor from capymoa\n",
    "* An example of how preprocessing (from MOA) can be used.\n",
    "* Comparing a SKLearn model against a MOA model\n",
    "* A variation of **Tutorial 5**: `Creating a new classifier in CapyMOA` which uses MOA learners, thus accessing MOA (Java) objects directly\n",
    "* How to log experiments using TensorBoard alongside the PyTorch API. This extends **Tutorial 3**: `Using Pytorch with CapyMOA`\n",
    "* Creating a synthetic stream with concept drifts using the MOA CLI directly\n",
    "* An example utilising a multi-threaded ensemble\n",
    "\n",
    "---\n",
    "\n",
    "*More information about CapyMOA can be found in* https://www.capymoa.org\n",
    "\n",
    "**last update on 28/07/2024**"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d2bb536e-4716-48fe-bf9b-05455b9e5a85",
   "metadata": {},
   "source": [
    "## 1. Using any MOA learner\n",
    "\n",
    "* **CapyMOA gives you access to any MOA classifier or regressor**\n",
    "\n",
    "* For some of the MOA learners there are corresponding Python objects (such as the HoeffdingTree or Adaptive Random Forest Classifier). However, MOA has over a hundred learners, and more are added constantly.\n",
    "\n",
    "* To allow advanced users to access **any** MOA learner from CapyMOA, we included the ```MOAClassifier``` and ```MOARegressor``` generic wrappers."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "3d1a9e23-a272-4c01-ab9b-e7f3ec5f7395",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Cumulative accuracy = 59.57599999999999, wall-clock time: 1.4879562854766846\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>instances</th>\n",
       "      <th>accuracy</th>\n",
       "      <th>kappa</th>\n",
       "      <th>kappa_t</th>\n",
       "      <th>kappa_m</th>\n",
       "      <th>f1_score</th>\n",
       "      <th>f1_score_0</th>\n",
       "      <th>f1_score_1</th>\n",
       "      <th>f1_score_2</th>\n",
       "      <th>f1_score_3</th>\n",
       "      <th>...</th>\n",
       "      <th>precision_1</th>\n",
       "      <th>precision_2</th>\n",
       "      <th>precision_3</th>\n",
       "      <th>precision_4</th>\n",
       "      <th>recall</th>\n",
       "      <th>recall_0</th>\n",
       "      <th>recall_1</th>\n",
       "      <th>recall_2</th>\n",
       "      <th>recall_3</th>\n",
       "      <th>recall_4</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>4500.0</td>\n",
       "      <td>50.000000</td>\n",
       "      <td>33.449507</td>\n",
       "      <td>34.858135</td>\n",
       "      <td>27.768860</td>\n",
       "      <td>45.089567</td>\n",
       "      <td>53.487412</td>\n",
       "      <td>33.012821</td>\n",
       "      <td>62.374245</td>\n",
       "      <td>13.945578</td>\n",
       "      <td>...</td>\n",
       "      <td>39.845261</td>\n",
       "      <td>67.685590</td>\n",
       "      <td>29.496403</td>\n",
       "      <td>47.644540</td>\n",
       "      <td>43.338015</td>\n",
       "      <td>57.142857</td>\n",
       "      <td>28.180575</td>\n",
       "      <td>57.835821</td>\n",
       "      <td>9.131403</td>\n",
       "      <td>64.399421</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>9000.0</td>\n",
       "      <td>49.466667</td>\n",
       "      <td>32.401117</td>\n",
       "      <td>35.287422</td>\n",
       "      <td>28.332808</td>\n",
       "      <td>41.328952</td>\n",
       "      <td>56.127745</td>\n",
       "      <td>25.129983</td>\n",
       "      <td>61.844725</td>\n",
       "      <td>2.307692</td>\n",
       "      <td>...</td>\n",
       "      <td>34.855769</td>\n",
       "      <td>68.228404</td>\n",
       "      <td>6.185567</td>\n",
       "      <td>45.596376</td>\n",
       "      <td>41.013784</td>\n",
       "      <td>59.175084</td>\n",
       "      <td>19.647696</td>\n",
       "      <td>56.553398</td>\n",
       "      <td>1.418440</td>\n",
       "      <td>68.274303</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>13500.0</td>\n",
       "      <td>52.422222</td>\n",
       "      <td>36.245962</td>\n",
       "      <td>38.846044</td>\n",
       "      <td>32.031746</td>\n",
       "      <td>45.928811</td>\n",
       "      <td>59.177456</td>\n",
       "      <td>25.333333</td>\n",
       "      <td>53.851590</td>\n",
       "      <td>18.657938</td>\n",
       "      <td>...</td>\n",
       "      <td>39.820359</td>\n",
       "      <td>65.017065</td>\n",
       "      <td>31.318681</td>\n",
       "      <td>51.899384</td>\n",
       "      <td>43.756287</td>\n",
       "      <td>66.071429</td>\n",
       "      <td>18.575419</td>\n",
       "      <td>45.958987</td>\n",
       "      <td>13.286713</td>\n",
       "      <td>74.888889</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>18000.0</td>\n",
       "      <td>59.311111</td>\n",
       "      <td>46.626403</td>\n",
       "      <td>47.309353</td>\n",
       "      <td>41.632133</td>\n",
       "      <td>57.974404</td>\n",
       "      <td>59.917012</td>\n",
       "      <td>34.180139</td>\n",
       "      <td>58.620690</td>\n",
       "      <td>64.179104</td>\n",
       "      <td>...</td>\n",
       "      <td>44.223108</td>\n",
       "      <td>56.342857</td>\n",
       "      <td>75.704225</td>\n",
       "      <td>64.530457</td>\n",
       "      <td>56.441207</td>\n",
       "      <td>62.946818</td>\n",
       "      <td>27.854454</td>\n",
       "      <td>61.090458</td>\n",
       "      <td>55.699482</td>\n",
       "      <td>74.614820</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>22500.0</td>\n",
       "      <td>64.555556</td>\n",
       "      <td>54.112305</td>\n",
       "      <td>54.258675</td>\n",
       "      <td>48.943662</td>\n",
       "      <td>63.436739</td>\n",
       "      <td>67.889126</td>\n",
       "      <td>52.967359</td>\n",
       "      <td>60.546875</td>\n",
       "      <td>63.816475</td>\n",
       "      <td>...</td>\n",
       "      <td>62.962963</td>\n",
       "      <td>61.917443</td>\n",
       "      <td>57.954545</td>\n",
       "      <td>68.314763</td>\n",
       "      <td>63.573514</td>\n",
       "      <td>70.629991</td>\n",
       "      <td>45.710627</td>\n",
       "      <td>59.235669</td>\n",
       "      <td>70.997680</td>\n",
       "      <td>71.293605</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>27000.0</td>\n",
       "      <td>60.111111</td>\n",
       "      <td>48.455842</td>\n",
       "      <td>48.670289</td>\n",
       "      <td>44.375581</td>\n",
       "      <td>58.755137</td>\n",
       "      <td>68.893204</td>\n",
       "      <td>49.307075</td>\n",
       "      <td>58.830549</td>\n",
       "      <td>55.581395</td>\n",
       "      <td>...</td>\n",
       "      <td>58.885017</td>\n",
       "      <td>56.536697</td>\n",
       "      <td>54.566210</td>\n",
       "      <td>60.080321</td>\n",
       "      <td>58.558489</td>\n",
       "      <td>73.671096</td>\n",
       "      <td>42.409034</td>\n",
       "      <td>61.318408</td>\n",
       "      <td>56.635071</td>\n",
       "      <td>58.758837</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>31500.0</td>\n",
       "      <td>60.666667</td>\n",
       "      <td>48.797681</td>\n",
       "      <td>49.225473</td>\n",
       "      <td>44.058154</td>\n",
       "      <td>59.622479</td>\n",
       "      <td>69.438669</td>\n",
       "      <td>47.246608</td>\n",
       "      <td>56.351039</td>\n",
       "      <td>61.557478</td>\n",
       "      <td>...</td>\n",
       "      <td>56.813820</td>\n",
       "      <td>52.928416</td>\n",
       "      <td>66.223404</td>\n",
       "      <td>58.839590</td>\n",
       "      <td>58.587577</td>\n",
       "      <td>70.227082</td>\n",
       "      <td>40.437158</td>\n",
       "      <td>60.246914</td>\n",
       "      <td>57.505774</td>\n",
       "      <td>64.520958</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>36000.0</td>\n",
       "      <td>62.488889</td>\n",
       "      <td>51.365841</td>\n",
       "      <td>51.977240</td>\n",
       "      <td>46.666667</td>\n",
       "      <td>62.079191</td>\n",
       "      <td>67.826087</td>\n",
       "      <td>55.932203</td>\n",
       "      <td>53.044496</td>\n",
       "      <td>67.179487</td>\n",
       "      <td>...</td>\n",
       "      <td>61.682243</td>\n",
       "      <td>50.727884</td>\n",
       "      <td>67.007673</td>\n",
       "      <td>65.527489</td>\n",
       "      <td>61.832537</td>\n",
       "      <td>68.997473</td>\n",
       "      <td>51.162791</td>\n",
       "      <td>55.582822</td>\n",
       "      <td>67.352185</td>\n",
       "      <td>66.067416</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>40500.0</td>\n",
       "      <td>57.933333</td>\n",
       "      <td>45.320224</td>\n",
       "      <td>45.852403</td>\n",
       "      <td>39.559387</td>\n",
       "      <td>56.433686</td>\n",
       "      <td>65.095729</td>\n",
       "      <td>35.967742</td>\n",
       "      <td>56.277603</td>\n",
       "      <td>61.290323</td>\n",
       "      <td>...</td>\n",
       "      <td>43.984221</td>\n",
       "      <td>57.106274</td>\n",
       "      <td>59.507830</td>\n",
       "      <td>64.431725</td>\n",
       "      <td>56.395784</td>\n",
       "      <td>75.298126</td>\n",
       "      <td>30.422920</td>\n",
       "      <td>55.472637</td>\n",
       "      <td>63.182898</td>\n",
       "      <td>57.602339</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>45000.0</td>\n",
       "      <td>72.000000</td>\n",
       "      <td>63.510428</td>\n",
       "      <td>63.636364</td>\n",
       "      <td>59.420290</td>\n",
       "      <td>71.308445</td>\n",
       "      <td>71.458075</td>\n",
       "      <td>66.176471</td>\n",
       "      <td>72.941176</td>\n",
       "      <td>69.961977</td>\n",
       "      <td>...</td>\n",
       "      <td>70.977918</td>\n",
       "      <td>71.480583</td>\n",
       "      <td>75.000000</td>\n",
       "      <td>74.647887</td>\n",
       "      <td>70.422343</td>\n",
       "      <td>74.121680</td>\n",
       "      <td>61.983471</td>\n",
       "      <td>74.462705</td>\n",
       "      <td>65.558195</td>\n",
       "      <td>75.985663</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>49500.0</td>\n",
       "      <td>66.177778</td>\n",
       "      <td>56.005059</td>\n",
       "      <td>55.820029</td>\n",
       "      <td>51.926721</td>\n",
       "      <td>66.012305</td>\n",
       "      <td>62.703583</td>\n",
       "      <td>61.891516</td>\n",
       "      <td>62.087186</td>\n",
       "      <td>70.573871</td>\n",
       "      <td>...</td>\n",
       "      <td>65.058480</td>\n",
       "      <td>65.096953</td>\n",
       "      <td>71.534653</td>\n",
       "      <td>69.770674</td>\n",
       "      <td>65.432668</td>\n",
       "      <td>63.900415</td>\n",
       "      <td>59.018568</td>\n",
       "      <td>59.343434</td>\n",
       "      <td>69.638554</td>\n",
       "      <td>75.262369</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>50000.0</td>\n",
       "      <td>65.400000</td>\n",
       "      <td>54.995471</td>\n",
       "      <td>54.869565</td>\n",
       "      <td>50.712251</td>\n",
       "      <td>65.018349</td>\n",
       "      <td>61.386139</td>\n",
       "      <td>60.445682</td>\n",
       "      <td>61.479100</td>\n",
       "      <td>68.827930</td>\n",
       "      <td>...</td>\n",
       "      <td>63.265306</td>\n",
       "      <td>63.479416</td>\n",
       "      <td>70.408163</td>\n",
       "      <td>70.110957</td>\n",
       "      <td>64.466325</td>\n",
       "      <td>62.155388</td>\n",
       "      <td>57.866667</td>\n",
       "      <td>59.600998</td>\n",
       "      <td>67.317073</td>\n",
       "      <td>75.391499</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>12 rows × 23 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "    instances   accuracy      kappa    kappa_t    kappa_m   f1_score  \\\n",
       "0      4500.0  50.000000  33.449507  34.858135  27.768860  45.089567   \n",
       "1      9000.0  49.466667  32.401117  35.287422  28.332808  41.328952   \n",
       "2     13500.0  52.422222  36.245962  38.846044  32.031746  45.928811   \n",
       "3     18000.0  59.311111  46.626403  47.309353  41.632133  57.974404   \n",
       "4     22500.0  64.555556  54.112305  54.258675  48.943662  63.436739   \n",
       "5     27000.0  60.111111  48.455842  48.670289  44.375581  58.755137   \n",
       "6     31500.0  60.666667  48.797681  49.225473  44.058154  59.622479   \n",
       "7     36000.0  62.488889  51.365841  51.977240  46.666667  62.079191   \n",
       "8     40500.0  57.933333  45.320224  45.852403  39.559387  56.433686   \n",
       "9     45000.0  72.000000  63.510428  63.636364  59.420290  71.308445   \n",
       "10    49500.0  66.177778  56.005059  55.820029  51.926721  66.012305   \n",
       "11    50000.0  65.400000  54.995471  54.869565  50.712251  65.018349   \n",
       "\n",
       "    f1_score_0  f1_score_1  f1_score_2  f1_score_3  ...  precision_1  \\\n",
       "0    53.487412   33.012821   62.374245   13.945578  ...    39.845261   \n",
       "1    56.127745   25.129983   61.844725    2.307692  ...    34.855769   \n",
       "2    59.177456   25.333333   53.851590   18.657938  ...    39.820359   \n",
       "3    59.917012   34.180139   58.620690   64.179104  ...    44.223108   \n",
       "4    67.889126   52.967359   60.546875   63.816475  ...    62.962963   \n",
       "5    68.893204   49.307075   58.830549   55.581395  ...    58.885017   \n",
       "6    69.438669   47.246608   56.351039   61.557478  ...    56.813820   \n",
       "7    67.826087   55.932203   53.044496   67.179487  ...    61.682243   \n",
       "8    65.095729   35.967742   56.277603   61.290323  ...    43.984221   \n",
       "9    71.458075   66.176471   72.941176   69.961977  ...    70.977918   \n",
       "10   62.703583   61.891516   62.087186   70.573871  ...    65.058480   \n",
       "11   61.386139   60.445682   61.479100   68.827930  ...    63.265306   \n",
       "\n",
       "    precision_2  precision_3  precision_4     recall   recall_0   recall_1  \\\n",
       "0     67.685590    29.496403    47.644540  43.338015  57.142857  28.180575   \n",
       "1     68.228404     6.185567    45.596376  41.013784  59.175084  19.647696   \n",
       "2     65.017065    31.318681    51.899384  43.756287  66.071429  18.575419   \n",
       "3     56.342857    75.704225    64.530457  56.441207  62.946818  27.854454   \n",
       "4     61.917443    57.954545    68.314763  63.573514  70.629991  45.710627   \n",
       "5     56.536697    54.566210    60.080321  58.558489  73.671096  42.409034   \n",
       "6     52.928416    66.223404    58.839590  58.587577  70.227082  40.437158   \n",
       "7     50.727884    67.007673    65.527489  61.832537  68.997473  51.162791   \n",
       "8     57.106274    59.507830    64.431725  56.395784  75.298126  30.422920   \n",
       "9     71.480583    75.000000    74.647887  70.422343  74.121680  61.983471   \n",
       "10    65.096953    71.534653    69.770674  65.432668  63.900415  59.018568   \n",
       "11    63.479416    70.408163    70.110957  64.466325  62.155388  57.866667   \n",
       "\n",
       "     recall_2   recall_3   recall_4  \n",
       "0   57.835821   9.131403  64.399421  \n",
       "1   56.553398   1.418440  68.274303  \n",
       "2   45.958987  13.286713  74.888889  \n",
       "3   61.090458  55.699482  74.614820  \n",
       "4   59.235669  70.997680  71.293605  \n",
       "5   61.318408  56.635071  58.758837  \n",
       "6   60.246914  57.505774  64.520958  \n",
       "7   55.582822  67.352185  66.067416  \n",
       "8   55.472637  63.182898  57.602339  \n",
       "9   74.462705  65.558195  75.985663  \n",
       "10  59.343434  69.638554  75.262369  \n",
       "11  59.600998  67.317073  75.391499  \n",
       "\n",
       "[12 rows x 23 columns]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from capymoa.evaluation import prequential_evaluation\n",
    "from capymoa.base import MOAClassifier\n",
    "from capymoa.datasets import RBFm_100k\n",
    "# This is an import from MOA\n",
    "from moa.classifiers.trees import HoeffdingAdaptiveTree\n",
    "\n",
    "rbf_100k = RBFm_100k()\n",
    "\n",
    "# Creates a wrapper around the HoeffdingAdaptiveTree, which then can be used as any other capymoa classifier\n",
    "HAT = MOAClassifier(schema=rbf_100k.get_schema(), moa_learner=HoeffdingAdaptiveTree)\n",
    "\n",
    "results_HAT = prequential_evaluation(stream=rbf_100k, learner=HAT, window_size=4500, max_instances=50000)\n",
    "\n",
    "print(f\"Cumulative accuracy = {results_HAT['cumulative'].accuracy()}, wall-clock time: {results_HAT['wallclock']}\")\n",
    "display(results_HAT['windowed'].metrics_per_window())"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3c102052-1a19-4f30-b3d1-f0163cab6af0",
   "metadata": {},
   "source": [
    "### 1.1 Checking the hyperparameters for the MOA CLI\n",
    "\n",
    "* MOA objects can be parametrized using the MOA CLI (Command Line Interface)\n",
    "* Sometimes you may not know the relevent parameters for ```moa_learner```,  ```moa_learner.CLI_help()``` presents all the hyperparameters available for the ```moa_learner``` object."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "3fbca563-e87f-41f2-98f2-dcad2ab65fb6",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "-l treeLearner (default: ARFHoeffdingTree -e 2000000 -g 50 -c 0.01)\n",
      "Random Forest Tree.\n",
      "-s ensembleSize (default: 100)\n",
      "The number of trees.\n",
      "-o mFeaturesMode (default: Percentage (M * (m / 100)))\n",
      "Defines how m, defined by mFeaturesPerTreeSize, is interpreted. M represents the total number of features.\n",
      "-m mFeaturesPerTreeSize (default: 60)\n",
      "Number of features allowed considered for each split. Negative values corresponds to M - m\n",
      "-a lambda (default: 6.0)\n",
      "The lambda parameter for bagging.\n",
      "-j numberOfJobs (default: 1)\n",
      "Total number of concurrent jobs used for processing (-1 = as much as possible, 0 = do not use multithreading)\n",
      "-x driftDetectionMethod (default: ADWINChangeDetector -a 1.0E-3)\n",
      "Change detector for drifts and its parameters\n",
      "-p warningDetectionMethod (default: ADWINChangeDetector -a 1.0E-2)\n",
      "Change detector for warnings (start training bkg learner)\n",
      "-w disableWeightedVote\n",
      "Should use weighted voting?\n",
      "-u disableDriftDetection\n",
      "Should use drift detection? If disabled then bkg learner is also disabled\n",
      "-q disableBackgroundLearner\n",
      "Should use bkg learner? If disabled then reset tree immediately.\n",
      "\n"
     ]
    }
   ],
   "source": [
    "from moa.classifiers.meta import AdaptiveRandomForest\n",
    "\n",
    "arf = MOAClassifier(schema=rbf_100k.get_schema(), moa_learner=AdaptiveRandomForest)\n",
    "\n",
    "print(arf.CLI_help())"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "id": "55d070de-8697-4f98-a11b-eab4e3d5c281",
   "metadata": {},
   "source": [
    "## 2. Using preprocessing from MOA (filters)\n",
    "\n",
    "We are working on a more user friendly API for preprocessing, this example just show how one can do that using MOA filters from here\n",
    "\n",
    "* Here we use ```NormalisationFilter``` filter from MOA to normalize instances in an online fashion.\n",
    "* MOA filters syntax wraps the whole stream, so we are always composing commands like `Filter(Stream, \n",
    "* We obtain the MOA CLI from the rbf_100k stream, since it can be mapped to a MOA stream, it is possible to obtain that. Comment out the print statements if you would like to inspect the actual creation strings (perhaps to copy and paste that into MOA?)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "ae9bb646-e0d1-4de6-b5a1-cff0f0a1b172",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-04-29T11:52:48.998749Z",
     "start_time": "2024-04-29T11:52:45.889095Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy with online normalization: 61.815\n",
      "Accuracy without normalization: 60.357000000000006\n"
     ]
    }
   ],
   "source": [
    "from capymoa.stream import Stream\n",
    "from capymoa.classifier import OnlineBagging\n",
    "from capymoa.evaluation import prequential_evaluation\n",
    "from moa.streams.filters import StandardisationFilter, NormalisationFilter\n",
    "from moa.streams import FilteredStream\n",
    "\n",
    "rbf_100k = RBFm_100k()\n",
    "\n",
    "# print(f'MOA creation string for data: {rbf_100k.moa_stream.getCLICreationString(rbf_100k.moa_stream.__class__)}')\n",
    "\n",
    "# Create a FilterStream and use the NormalisationFilter\n",
    "rbf_stream_normalised = Stream(CLI=f\"-s ({rbf_100k.moa_stream.getCLICreationString(rbf_100k.moa_stream.__class__)}) \\\n",
    "-f NormalisationFilter \", moa_stream=FilteredStream())\n",
    "\n",
    "# print(f'MOA creation string for filtered version: {rbf_stream_normalised.moa_stream.getCLICreationString(rbf_stream_normalised.moa_stream.__class__)}')\n",
    "\n",
    "ob_learner_norm = OnlineBagging(schema=rbf_stream_normalised.get_schema(), ensemble_size=5)\n",
    "ob_learner = OnlineBagging(schema=rbf_100k.get_schema(), ensemble_size=5)\n",
    "\n",
    "ob_results_norm = prequential_evaluation(stream=rbf_stream_normalised, learner=ob_learner_norm)\n",
    "ob_results = prequential_evaluation(stream=rbf_100k, learner=ob_learner)\n",
    "\n",
    "\n",
    "print(f\"Accuracy with online normalization: {ob_results_norm['cumulative'].accuracy()}\")\n",
    "print(f\"Accuracy without normalization: {ob_results['cumulative'].accuracy()}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f74c58fb-dd90-49f4-8b4f-81a9e36e47ff",
   "metadata": {},
   "source": [
    "## 3. Comparing a MOA and SKLearn models\n",
    "\n",
    "* This simple example shows how it is simple to compare a MOA and a SKLearn regressors. \n",
    "* For the sake of this example, we are using the wrappers\n",
    "* SKClassifier (and SKRegressor) are parametrized directly as part of the object initialization\n",
    "* MOAClassifier (and MOARegressor) are parametrized through a CLI (a separate parameter)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "afe7193c-5bab-4b46-8627-c74b28a3b7c5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from capymoa.base import SKClassifier, MOAClassifier\n",
    "from capymoa.datasets import CovtypeTiny\n",
    "from capymoa.evaluation import prequential_evaluation_multiple_learners\n",
    "from capymoa.evaluation.visualization import plot_windowed_results\n",
    "\n",
    "from sklearn.linear_model import SGDClassifier\n",
    "from moa.classifiers.trees import HoeffdingTree\n",
    "\n",
    "covt_tiny = CovtypeTiny()\n",
    "\n",
    "sk_sgd = SKClassifier(schema=covt_tiny.schema, sklearner=SGDClassifier(loss='log_loss', penalty='l1', alpha=0.001))\n",
    "moa_ht = MOAClassifier(schema=covt_tiny.schema, moa_learner=HoeffdingTree, CLI=\"-g 50\")\n",
    "\n",
    "results = prequential_evaluation_multiple_learners(stream=covt_tiny, learners={'sk_sgd':sk_sgd, 'moa_ht':moa_ht}, window_size=100)\n",
    "plot_windowed_results(results['sk_sgd'], results['moa_ht'], metric='accuracy')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "df198282-7a87-4e03-ba0b-b3de3ccf9163",
   "metadata": {},
   "source": [
    "## 4. Creating Python learners with MOA Objects\n",
    "\n",
    "* This example follow the example from `06_new_learner` which shows how to create a custom online bagging implementation.\n",
    "* Here we also create an online bagging implementation, but the base_learner is a MOA class"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "a0a0906a-c953-4d50-8be8-c1c94e3eac4d",
   "metadata": {},
   "outputs": [],
   "source": [
    "from capymoa.base import Classifier, MOAClassifier\n",
    "from moa.classifiers.trees import HoeffdingTree\n",
    "from collections import Counter\n",
    "import numpy as np\n",
    "import random\n",
    "import math\n",
    "\n",
    "def poisson(lambd, random_generator):\n",
    "    if lambd < 100.0:\n",
    "        product = 1.0\n",
    "        _sum = 1.0\n",
    "        threshold = random_generator.random() * math.exp(lambd)\n",
    "        i = 1\n",
    "        max_val = max(100, 10 * math.ceil(lambd))\n",
    "        while i < max_val and _sum <= threshold:\n",
    "            product *= (lambd / i)\n",
    "            _sum += product\n",
    "            i += 1\n",
    "        return i - 1\n",
    "    x = lambd + math.sqrt(lambd) * random_generator.gauss(0, 1)\n",
    "    if x < 0.0:\n",
    "        return 0\n",
    "    return int(math.floor(x))\n",
    "\n",
    "class CustomOnlineBagging(Classifier):\n",
    "    def __init__(self, schema=None, random_seed=1, ensemble_size=5, moa_base_learner_class=None, CLI_base_learner=None):\n",
    "        super().__init__(schema=schema, random_seed=random_seed)\n",
    "\n",
    "        self.random_generator = random.Random()\n",
    "        self.CLI_base_learner = CLI_base_learner\n",
    "        \n",
    "        self.ensemble_size = ensemble_size\n",
    "        self.moa_base_learner_class = moa_base_learner_class\n",
    "        \n",
    "        # Default base learner if None is specified\n",
    "        if self.moa_base_learner_class is None:\n",
    "            self.moa_base_learner_class = HoeffdingTree\n",
    "        \n",
    "        self.ensemble = []\n",
    "        # Create several instances for the base_learners\n",
    "        for i in range(self.ensemble_size): \n",
    "            self.ensemble.append(MOAClassifier(schema=self.schema, moa_learner=self.moa_base_learner_class(), CLI=self.CLI_base_learner))\n",
    "        \n",
    "    def __str__(self):\n",
    "        return 'CustomOnlineBagging'\n",
    "\n",
    "    def train(self, instance):\n",
    "        for i in range(self.ensemble_size):\n",
    "            k = poisson(1.0, self.random_generator)\n",
    "            for _ in range(k):\n",
    "                self.ensemble[i].train(instance)\n",
    "\n",
    "    def predict(self, instance):\n",
    "        predictions = []\n",
    "        for i in range(self.ensemble_size):\n",
    "            predictions.append(self.ensemble[i].predict(instance))\n",
    "        majority_vote = Counter(predictions)\n",
    "        prediction = majority_vote.most_common(1)[0][0]\n",
    "        return prediction\n",
    "\n",
    "    def predict_proba(self, instance):\n",
    "        probabilities = []\n",
    "        for i in range(self.ensemble_size):\n",
    "            classifier_proba = self.ensemble[i].predict_proba(instance)\n",
    "            classifier_proba = classifier_proba / np.sum(classifier_proba)\n",
    "            probabilities.append(classifier_proba)\n",
    "        avg_proba = np.mean(probabilities, axis=0)\n",
    "        return avg_proba\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c971ac60-0aa5-4295-be56-bcb6ee1ccb40",
   "metadata": {},
   "source": [
    "### 4.1 Testing the custom online bagging\n",
    "\n",
    "* We choose to use an HoeffdingAdaptiveTree from MOA as the base learner\n",
    "* We also specify the CLI commands to configure the base learner"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "0740765c-35c0-416a-99ca-f8e55f921032",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/gomeshe/Dropbox/ciencia_computacao/dev/main-projects/CapyMOA/src/capymoa/stream/_stream.py:38: UserWarning: target variable includes 2 (< 20) unique values, inferred as categorical, set target_type = 'numeric' if you intend numeric targets\n",
      "  warnings.warn(f'target variable includes {num_unique} (< 20) unique values, inferred as categorical, '\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "82.58077330508475"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from capymoa.evaluation import prequential_evaluation\n",
    "from capymoa.stream import stream_from_file\n",
    "from moa.classifiers.trees import HoeffdingAdaptiveTree\n",
    "\n",
    "elec_stream = stream_from_file(path_to_csv_or_arff=\"../data/electricity.csv\")\n",
    "\n",
    "# Creating a learner: using a hoeffding adaptive tree as the base learner with grace period of 50 (-g 50)\n",
    "NEW_OB = CustomOnlineBagging(schema=elec_stream.get_schema(), \n",
    "                             ensemble_size=5, \n",
    "                             moa_base_learner_class=HoeffdingAdaptiveTree, \n",
    "                             CLI_base_learner=\"-g 50\")\n",
    "\n",
    "results_NEW_OB = prequential_evaluation(stream=elec_stream, learner=NEW_OB, window_size=4500)\n",
    "\n",
    "print(f\"Accuracy: {results_NEW_OB.cumulative.accuracy()}\""
   ]
  },
  {
   "cell_type": "markdown",
   "id": "62e3e70a-2422-4b3a-b2bf-b8f96a3efdeb",
   "metadata": {},
   "source": [
    "## 5. Using TensorBoard with PyTorch in CapyMOA\n",
    "\n",
    "* One can use TensorBoard to visualize logged data in an online fashion\n",
    "* We go through all the steps below, including installing TensorBoard"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8fda8006-e0e9-4547-a2c9-8fc43d16ca57",
   "metadata": {},
   "source": [
    "### 5.1 Install TensorBoard\n",
    "Clear any logs from previous runs\n",
    "\n",
    "```sh\n",
    "rm -rf ./runs\n",
    "```"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "f11baceb-2c77-4636-8e91-d19aadf7b3b3",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Requirement already satisfied: tensorboard in /Users/gomeshe/miniconda3/envs/capymoa/lib/python3.9/site-packages (2.17.0)\n",
      "Requirement already satisfied: absl-py>=0.4 in /Users/gomeshe/miniconda3/envs/capymoa/lib/python3.9/site-packages (from tensorboard) (2.1.0)\n",
      "Requirement already satisfied: grpcio>=1.48.2 in /Users/gomeshe/miniconda3/envs/capymoa/lib/python3.9/site-packages (from tensorboard) (1.65.1)\n",
      "Requirement already satisfied: markdown>=2.6.8 in /Users/gomeshe/miniconda3/envs/capymoa/lib/python3.9/site-packages (from tensorboard) (3.6)\n",
      "Requirement already satisfied: numpy>=1.12.0 in /Users/gomeshe/miniconda3/envs/capymoa/lib/python3.9/site-packages (from tensorboard) (1.26.3)\n",
      "Requirement already satisfied: protobuf!=4.24.0,<5.0.0,>=3.19.6 in /Users/gomeshe/miniconda3/envs/capymoa/lib/python3.9/site-packages (from tensorboard) (4.25.4)\n",
      "Requirement already satisfied: setuptools>=41.0.0 in /Users/gomeshe/miniconda3/envs/capymoa/lib/python3.9/site-packages (from tensorboard) (69.5.1)\n",
      "Requirement already satisfied: six>1.9 in /Users/gomeshe/miniconda3/envs/capymoa/lib/python3.9/site-packages (from tensorboard) (1.16.0)\n",
      "Requirement already satisfied: tensorboard-data-server<0.8.0,>=0.7.0 in /Users/gomeshe/miniconda3/envs/capymoa/lib/python3.9/site-packages (from tensorboard) (0.7.2)\n",
      "Requirement already satisfied: werkzeug>=1.0.1 in /Users/gomeshe/miniconda3/envs/capymoa/lib/python3.9/site-packages (from tensorboard) (3.0.3)\n",
      "Requirement already satisfied: importlib-metadata>=4.4 in /Users/gomeshe/miniconda3/envs/capymoa/lib/python3.9/site-packages (from markdown>=2.6.8->tensorboard) (7.1.0)\n",
      "Requirement already satisfied: MarkupSafe>=2.1.1 in /Users/gomeshe/miniconda3/envs/capymoa/lib/python3.9/site-packages (from werkzeug>=1.0.1->tensorboard) (2.1.5)\n",
      "Requirement already satisfied: zipp>=0.5 in /Users/gomeshe/miniconda3/envs/capymoa/lib/python3.9/site-packages (from importlib-metadata>=4.4->markdown>=2.6.8->tensorboard) (3.18.2)\n"
     ]
    }
   ],
   "source": [
    "!pip install tensorboard"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b1e64bd6-b0c7-4296-aee7-a29985a9da21",
   "metadata": {},
   "source": [
    "### 5.2  PyTorchClassifier\n",
    "* We define `PyTorchClassifier` and `NeuralNetwork` classes similarly to those from **Tutorial 3**: `Using Pytorch with CapyMOA`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "ea9a4d94-7515-424c-a9fd-76c78ddf52d1",
   "metadata": {},
   "outputs": [],
   "source": [
    "from capymoa.base import Classifier\n",
    "import numpy as np\n",
    "import torch\n",
    "from torch import nn\n",
    "\n",
    "torch.manual_seed(1)\n",
    "torch.use_deterministic_algorithms(True)\n",
    "\n",
    "# Get cpu device for training.\n",
    "device = (\"cpu\")\n",
    "\n",
    "# Define model\n",
    "class NeuralNetwork(nn.Module):\n",
    "    def __init__(self, input_size=0, number_of_classes=0):\n",
    "        super().__init__()\n",
    "        self.flatten = nn.Flatten()\n",
    "        self.linear_relu_stack = nn.Sequential(\n",
    "            nn.Linear(input_size, 512),\n",
    "            nn.ReLU(),\n",
    "            nn.Linear(512, 512),\n",
    "            nn.ReLU(),\n",
    "            nn.Linear(512, number_of_classes)\n",
    "        )\n",
    "\n",
    "    def forward(self, x):\n",
    "        x = self.flatten(x)\n",
    "        logits = self.linear_relu_stack(x)\n",
    "        return logits\n",
    "\n",
    "\n",
    "class PyTorchClassifier(Classifier):\n",
    "    def __init__(self, schema=None, random_seed=1, nn_model: nn.Module = None, optimizer=None, loss_fn=nn.CrossEntropyLoss(), device=(\"cpu\"), lr=1e-3):\n",
    "        super().__init__(schema, random_seed)\n",
    "        self.model = None\n",
    "        self.optimizer = None\n",
    "        self.loss_fn = loss_fn\n",
    "        self.lr = lr\n",
    "        self.device = device\n",
    "        \n",
    "        torch.manual_seed(random_seed)\n",
    "        \n",
    "        if nn_model is None:\n",
    "            self.set_model(None)\n",
    "        else:\n",
    "            self.model = nn_model.to(device)\n",
    "        if optimizer is None:\n",
    "            if self.model is not None:\n",
    "                self.optimizer = torch.optim.SGD(self.model.parameters(), lr=lr)\n",
    "        else:\n",
    "            self.optimizer = optimizer\n",
    "        \n",
    "    def __str__(self):\n",
    "        return str(self.model)\n",
    "\n",
    "    def CLI_help(self):\n",
    "        return str('schema=None, random_seed=1, nn_model: nn.Module = None, optimizer=None, loss_fn=nn.CrossEntropyLoss(), device=(\"cpu\"), lr=1e-3')\n",
    "\n",
    "    def set_model(self, instance):\n",
    "        if self.schema is None:\n",
    "            moa_instance = instance.java_instance.getData()\n",
    "            self.model = NeuralNetwork(input_size=moa_instance.get_num_attributes(), number_of_classes=moa_instance.get_num_classes()).to(self.device)\n",
    "        elif instance is not None:\n",
    "            self.model = NeuralNetwork(input_size=self.schema.get_num_attributes(), number_of_classes=self.schema.get_num_classes()).to(self.device)\n",
    "            \n",
    "    def train(self, instance):\n",
    "        if self.model is None:\n",
    "            self.set_model(instance)\n",
    "    \n",
    "        X = torch.tensor(instance.x, dtype=torch.float32)\n",
    "        y = torch.tensor(instance.y_index, dtype=torch.long)\n",
    "        # set the device and add a dimension to the tensor\n",
    "        X, y = torch.unsqueeze(X.to(self.device), 0), torch.unsqueeze(y.to(self.device),0)\n",
    "\n",
    "        # Compute prediction error\n",
    "        pred = self.model(X)\n",
    "        loss = self.loss_fn(pred, y)\n",
    "    \n",
    "        # Backpropagation\n",
    "        loss.backward()\n",
    "        self.optimizer.step()\n",
    "        self.optimizer.zero_grad()\n",
    "\n",
    "    def predict(self, instance):\n",
    "        return np.argmax(self.predict_proba(instance))\n",
    "\n",
    "    def predict_proba(self, instance):\n",
    "        if self.model is None:\n",
    "            self.set_model(instance)\n",
    "        X = torch.unsqueeze(torch.tensor(instance.x, dtype=torch.float32).to(self.device), 0)\n",
    "        # turn off gradient collection\n",
    "        with torch.no_grad():\n",
    "            pred = np.asarray(self.model(X).numpy(), dtype=np.double)\n",
    "        return pred\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2b166ade-23b3-445c-a382-6f0cb6231d66",
   "metadata": {},
   "source": [
    "### 5.3  PyTorchClassifier + the test-then-train loop + TensorBoard\n",
    "* Here we use instance loop to log relevant log information to TensorBoard\n",
    "* These information can be viewed while the processing is happening using TensorBoard"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "9e93527d-26cb-4a0b-a4e4-2f3399724502",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Processed 10000 instances\n",
      "Processed 20000 instances\n",
      "Processed 30000 instances\n",
      "Processed 40000 instances\n",
      "Processed 50000 instances\n",
      "Processed 60000 instances\n",
      "Processed 70000 instances\n",
      "Processed 80000 instances\n",
      "Processed 90000 instances\n",
      "Processed 100000 instances\n"
     ]
    }
   ],
   "source": [
    "from capymoa.evaluation import ClassificationEvaluator\n",
    "from capymoa.datasets import RBFm_100k\n",
    "from torch.utils.tensorboard import SummaryWriter\n",
    "\n",
    "# Create a SummaryWriter instance.\n",
    "writer = SummaryWriter()\n",
    "## Opening a file again to start from the beginning\n",
    "rbf_stream = RBFm_100k()\n",
    "\n",
    "# Creating the evaluator\n",
    "evaluator = ClassificationEvaluator(schema=rbf_stream.get_schema())\n",
    "\n",
    "# Creating a learner\n",
    "simple_pyTorch_classifier = PyTorchClassifier(\n",
    "    schema=rbf_stream.get_schema(), \n",
    "    nn_model=NeuralNetwork(input_size=rbf_stream.get_schema().get_num_attributes(), \n",
    "                           number_of_classes=rbf_stream.get_schema().get_num_classes()).to(device)\n",
    ")\n",
    "\n",
    "i = 0\n",
    "while rbf_stream.has_more_instances():\n",
    "    i += 1\n",
    "    instance = rbf_stream.next_instance()\n",
    "\n",
    "    prediction = simple_pyTorch_classifier.predict(instance)\n",
    "    evaluator.update(instance.y_index, prediction)\n",
    "    simple_pyTorch_classifier.train(instance)\n",
    "    \n",
    "    if i % 1000 == 0:\n",
    "        writer.add_scalar(\"accuracy\", evaluator.accuracy(), i)\n",
    "\n",
    "    if i % 10000 == 0:\n",
    "        print(f\"Processed {i} instances\")\n",
    "\n",
    "writer.add_scalar(\"accuracy\", evaluator.accuracy(), i)\n",
    "# Call flush() method to make sure that all pending events have been written to disk.\n",
    "writer.flush()\n",
    "\n",
    "# If you do not need the summary writer anymore, call close() method.\n",
    "writer.close()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9da96643-1900-41e8-96aa-6af460194ac6",
   "metadata": {},
   "source": [
    "#### 5.4 Run TensorBoard\n",
    "Now, start TensorBoard, specifying the root log directory you used above. \n",
    "Argument ``logdir`` points to directory where TensorBoard will look to find \n",
    "event files that it can display. TensorBoard will recursively walk \n",
    "the directory structure rooted at ``logdir``, looking for ``.*tfevents.*`` files.\n",
    "\n",
    "```sh\n",
    "tensorboard --logdir=runs\n",
    "```\n",
    "Go to the URL it provides\n",
    "\n",
    "This dashboard shows how the accuracy change with time. \n",
    "You can use it to also track training speed, learning rate, and other \n",
    "scalar values."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "38b1f9ce-c3a1-4944-8cb1-f2a84cd4ff25",
   "metadata": {},
   "source": [
    "## 6. Creating a synthetic stream with concept drifts from MOA\n",
    "\n",
    "* Demonstrates the flexibility of the API, these level of manipulation of the API is expected from experienced MOA users.\n",
    "* To use the API like this the user must be familiar with how concept drifts are simulatd in MOA\n",
    "\n",
    "EvaluatePrequential -l trees.HoeffdingAdaptiveTree **-s (ConceptDriftStream -s generators.AgrawalGenerator -d (generators.AgrawalGenerator -f 2) -p 5000)** -e (WindowClassificationPerformanceEvaluator **-w 100**) **-i 10000 -f 100**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "7b79ac8e-d7ac-48fb-b983-22301d272364",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from capymoa.stream import Stream\n",
    "from capymoa.classifier import OnlineBagging\n",
    "from capymoa.evaluation import prequential_evaluation\n",
    "from capymoa.evaluation.visualization import plot_windowed_results\n",
    "from moa.streams import ConceptDriftStream\n",
    "\n",
    "# Using the API to generate the data using the ConceptDriftStream and AgrawalGenerator.\n",
    "# The drift location is based on the number of instances (5000) as well as the drift width (1000, the default value) \n",
    "stream_sea1drift = Stream(moa_stream=ConceptDriftStream(), \n",
    "                          CLI=\"-s generators.SEAGenerator -d (generators.SEAGenerator -f 2) -p 5000 -w 1000\")\n",
    "\n",
    "OB = OnlineBagging(schema=stream_sea1drift.get_schema(), ensemble_size=10)\n",
    "\n",
    "results_sea1drift_OB = prequential_evaluation(stream=stream_sea1drift, learner=OB, window_size=100, max_instances=10000)\n",
    "\n",
    "plot_windowed_results(results_sea1drift_OB, metric='accuracy')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0f2f9fb6-0994-4f3f-aaf3-c73b09847019",
   "metadata": {},
   "source": [
    "## 7. Drift, Multi-threated Ensemble and Results\n",
    "\n",
    "* Generate a stream with 3 drifts, 2 abrupt and one gradual. \n",
    "* Evaluate utilising test-then-train (cumulative) and windowed evaluation.\n",
    "* Execute a multi-threated version of AdaptiveRandomForest.\n",
    "* For more on multi-threaded ensembles, see **parallel_ensembles.ipynb** notebook"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "3142e7e7-7175-40da-a89c-b528d71eb00c",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "None\n",
      "Cumulative accuracy = 89.37\n",
      "wallclock = 14.674108982086182 seconds\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>instances</th>\n",
       "      <th>accuracy</th>\n",
       "      <th>kappa</th>\n",
       "      <th>kappa_t</th>\n",
       "      <th>kappa_m</th>\n",
       "      <th>f1_score</th>\n",
       "      <th>f1_score_0</th>\n",
       "      <th>f1_score_1</th>\n",
       "      <th>precision</th>\n",
       "      <th>precision_0</th>\n",
       "      <th>precision_1</th>\n",
       "      <th>recall</th>\n",
       "      <th>recall_0</th>\n",
       "      <th>recall_1</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>5000.0</td>\n",
       "      <td>88.26</td>\n",
       "      <td>73.743687</td>\n",
       "      <td>74.333188</td>\n",
       "      <td>67.096413</td>\n",
       "      <td>87.033480</td>\n",
       "      <td>82.534960</td>\n",
       "      <td>91.158307</td>\n",
       "      <td>88.176897</td>\n",
       "      <td>87.951807</td>\n",
       "      <td>88.401987</td>\n",
       "      <td>85.919338</td>\n",
       "      <td>77.746637</td>\n",
       "      <td>94.092040</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>10000.0</td>\n",
       "      <td>88.88</td>\n",
       "      <td>75.482652</td>\n",
       "      <td>76.520270</td>\n",
       "      <td>69.700272</td>\n",
       "      <td>87.903177</td>\n",
       "      <td>83.939919</td>\n",
       "      <td>91.495870</td>\n",
       "      <td>88.990120</td>\n",
       "      <td>89.305470</td>\n",
       "      <td>88.674770</td>\n",
       "      <td>86.842465</td>\n",
       "      <td>79.182561</td>\n",
       "      <td>94.502370</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>15000.0</td>\n",
       "      <td>89.26</td>\n",
       "      <td>76.181571</td>\n",
       "      <td>77.187766</td>\n",
       "      <td>70.315091</td>\n",
       "      <td>88.238991</td>\n",
       "      <td>84.302835</td>\n",
       "      <td>91.837665</td>\n",
       "      <td>89.310872</td>\n",
       "      <td>89.454094</td>\n",
       "      <td>89.167651</td>\n",
       "      <td>87.192532</td>\n",
       "      <td>79.712548</td>\n",
       "      <td>94.672516</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>20000.0</td>\n",
       "      <td>88.98</td>\n",
       "      <td>75.124468</td>\n",
       "      <td>75.938865</td>\n",
       "      <td>68.586089</td>\n",
       "      <td>87.733308</td>\n",
       "      <td>83.297969</td>\n",
       "      <td>91.777347</td>\n",
       "      <td>88.966743</td>\n",
       "      <td>88.932039</td>\n",
       "      <td>89.001447</td>\n",
       "      <td>86.533606</td>\n",
       "      <td>78.335234</td>\n",
       "      <td>94.731978</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>25000.0</td>\n",
       "      <td>89.90</td>\n",
       "      <td>77.706053</td>\n",
       "      <td>78.157439</td>\n",
       "      <td>72.114854</td>\n",
       "      <td>88.955206</td>\n",
       "      <td>85.400405</td>\n",
       "      <td>92.279468</td>\n",
       "      <td>89.829793</td>\n",
       "      <td>89.623786</td>\n",
       "      <td>90.035800</td>\n",
       "      <td>88.097484</td>\n",
       "      <td>81.557151</td>\n",
       "      <td>94.637817</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>30000.0</td>\n",
       "      <td>89.50</td>\n",
       "      <td>76.471976</td>\n",
       "      <td>76.933216</td>\n",
       "      <td>70.522179</td>\n",
       "      <td>88.417299</td>\n",
       "      <td>84.314311</td>\n",
       "      <td>92.108823</td>\n",
       "      <td>89.663782</td>\n",
       "      <td>90.102171</td>\n",
       "      <td>89.225393</td>\n",
       "      <td>87.204997</td>\n",
       "      <td>79.225154</td>\n",
       "      <td>95.184840</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>35000.0</td>\n",
       "      <td>90.10</td>\n",
       "      <td>77.634753</td>\n",
       "      <td>78.232190</td>\n",
       "      <td>71.535365</td>\n",
       "      <td>88.949725</td>\n",
       "      <td>84.986351</td>\n",
       "      <td>92.615247</td>\n",
       "      <td>90.051553</td>\n",
       "      <td>89.922978</td>\n",
       "      <td>90.180128</td>\n",
       "      <td>87.874534</td>\n",
       "      <td>80.563542</td>\n",
       "      <td>95.185526</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>40000.0</td>\n",
       "      <td>89.70</td>\n",
       "      <td>77.159091</td>\n",
       "      <td>77.676636</td>\n",
       "      <td>71.325167</td>\n",
       "      <td>88.687154</td>\n",
       "      <td>84.963504</td>\n",
       "      <td>92.167300</td>\n",
       "      <td>89.601454</td>\n",
       "      <td>89.318600</td>\n",
       "      <td>89.884307</td>\n",
       "      <td>87.791326</td>\n",
       "      <td>81.013363</td>\n",
       "      <td>94.569288</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>45000.0</td>\n",
       "      <td>89.58</td>\n",
       "      <td>76.658626</td>\n",
       "      <td>77.406765</td>\n",
       "      <td>70.531674</td>\n",
       "      <td>88.460627</td>\n",
       "      <td>84.461676</td>\n",
       "      <td>92.161878</td>\n",
       "      <td>89.515036</td>\n",
       "      <td>89.337539</td>\n",
       "      <td>89.692533</td>\n",
       "      <td>87.430769</td>\n",
       "      <td>80.090498</td>\n",
       "      <td>94.771040</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>50000.0</td>\n",
       "      <td>89.54</td>\n",
       "      <td>77.044005</td>\n",
       "      <td>77.697228</td>\n",
       "      <td>71.637744</td>\n",
       "      <td>88.656788</td>\n",
       "      <td>85.052872</td>\n",
       "      <td>91.955084</td>\n",
       "      <td>89.633308</td>\n",
       "      <td>89.909366</td>\n",
       "      <td>89.357250</td>\n",
       "      <td>87.701317</td>\n",
       "      <td>80.694143</td>\n",
       "      <td>94.708492</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   instances  accuracy      kappa    kappa_t    kappa_m   f1_score  \\\n",
       "0     5000.0     88.26  73.743687  74.333188  67.096413  87.033480   \n",
       "1    10000.0     88.88  75.482652  76.520270  69.700272  87.903177   \n",
       "2    15000.0     89.26  76.181571  77.187766  70.315091  88.238991   \n",
       "3    20000.0     88.98  75.124468  75.938865  68.586089  87.733308   \n",
       "4    25000.0     89.90  77.706053  78.157439  72.114854  88.955206   \n",
       "5    30000.0     89.50  76.471976  76.933216  70.522179  88.417299   \n",
       "6    35000.0     90.10  77.634753  78.232190  71.535365  88.949725   \n",
       "7    40000.0     89.70  77.159091  77.676636  71.325167  88.687154   \n",
       "8    45000.0     89.58  76.658626  77.406765  70.531674  88.460627   \n",
       "9    50000.0     89.54  77.044005  77.697228  71.637744  88.656788   \n",
       "\n",
       "   f1_score_0  f1_score_1  precision  precision_0  precision_1     recall  \\\n",
       "0   82.534960   91.158307  88.176897    87.951807    88.401987  85.919338   \n",
       "1   83.939919   91.495870  88.990120    89.305470    88.674770  86.842465   \n",
       "2   84.302835   91.837665  89.310872    89.454094    89.167651  87.192532   \n",
       "3   83.297969   91.777347  88.966743    88.932039    89.001447  86.533606   \n",
       "4   85.400405   92.279468  89.829793    89.623786    90.035800  88.097484   \n",
       "5   84.314311   92.108823  89.663782    90.102171    89.225393  87.204997   \n",
       "6   84.986351   92.615247  90.051553    89.922978    90.180128  87.874534   \n",
       "7   84.963504   92.167300  89.601454    89.318600    89.884307  87.791326   \n",
       "8   84.461676   92.161878  89.515036    89.337539    89.692533  87.430769   \n",
       "9   85.052872   91.955084  89.633308    89.909366    89.357250  87.701317   \n",
       "\n",
       "    recall_0   recall_1  \n",
       "0  77.746637  94.092040  \n",
       "1  79.182561  94.502370  \n",
       "2  79.712548  94.672516  \n",
       "3  78.335234  94.731978  \n",
       "4  81.557151  94.637817  \n",
       "5  79.225154  95.184840  \n",
       "6  80.563542  95.185526  \n",
       "7  81.013363  94.569288  \n",
       "8  80.090498  94.771040  \n",
       "9  80.694143  94.708492  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA/oAAAHWCAYAAADHF/LFAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjkuMCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy80BEi2AAAACXBIWXMAAA9hAAAPYQGoP6dpAACvzUlEQVR4nOzdd3hT5fvH8XeSpntAB7SFUsreyJbpYgkiSxQFAUVxoKAoAv7Er4qIoIIKynCjDEWGqDgQRfbee+9SVvdMk/P7o1KprFLapuPzuq5eJCcnz7kTniTnPs8yGYZhICIiIiIiIiJFgtnZAYiIiIiIiIhI7lGiLyIiIiIiIlKEKNEXERERERERKUKU6IuIiIiIiIgUIUr0RURERERERIoQJfoiIiIiIiIiRYgSfREREREREZEiRIm+iIiIiIiISBGiRF9ERERERESkCFGiLyIiIiIiIlKEKNEXERERERERKUKU6IuIiIiIiIgUIUr0RUREpEBJTEx0dggiIiKFmhJ9ERGRIu7o0aM8/fTTVK1aFQ8PDwICAujRowdHjhy5bN+YmBief/55ypcvj5ubG2XLlqVPnz6cO3cuc5+UlBRee+01qlSpgru7OyEhIXTr1o2DBw8CsHTpUkwmE0uXLs1S9pEjRzCZTHz55ZeZ2/r164e3tzcHDx6kQ4cO+Pj40KtXLwCWL19Ojx49KFeuHG5uboSFhfH888+TnJx8Wdx79uzh/vvvJygoCA8PD6pWrcr//d//AfDXX39hMpmYP3/+Zc+bOXMmJpOJ1atX3+jbKiIiUmC5ODsAERERyVvr169n1apV9OzZk7Jly3LkyBEmT57M7bffzq5du/D09AQgISGBli1bsnv3bh599FHq16/PuXPnWLhwISdOnCAwMBC73c4999zDkiVL6NmzJ4MHDyY+Pp7FixezY8cOKlaseMPxpaen065dO1q0aMG7776bGc+cOXNISkriqaeeIiAggHXr1jFx4kROnDjBnDlzMp+/bds2WrZsidVqZcCAAZQvX56DBw/y448/Mnr0aG6//XbCwsKYMWMGXbt2zXLsGTNmULFiRZo2bXoT77CIiEjBokRfRESkiOvYsSP33Xdflm2dOnWiadOmzJ07l4cffhiAd955hx07djBv3rwsCfErr7yCYRgATJ8+nSVLljB+/Hief/75zH2GDx+euc+NSk1NpUePHowZMybL9rFjx+Lh4ZF5f8CAAVSqVImXX36ZY8eOUa5cOQCeffZZDMNg06ZNmdsA3n77bQBMJhO9e/dm/PjxxMbG4ufnB8DZs2f5/fffM1v+RUREigp13RcRESniLk2WbTYb58+fp1KlSpQoUYJNmzZlPjZ37lzq1q17Was3ZCTLF/cJDAzk2Wefveo+OfHUU09dM+7ExETOnTtHs2bNMAyDzZs3AxnJ+rJly3j00UezJPn/jadPnz6kpqby/fffZ2779ttvSU9Pp3fv3jmOW0REpCBSoi8iIlLEJScn8+qrrxIWFoabmxuBgYEEBQURExNDbGxs5n4HDx6kVq1a1yzr4MGDVK1aFReX3OsU6OLiQtmyZS/bfuzYMfr164e/vz/e3t4EBQVx2223AWTGfejQIYDrxl2tWjUaNWrEjBkzMrfNmDGDW2+9lUqVKuXWSxERESkQ1HVfRESkiHv22Wf54osveO6552jatCl+fn6YTCZ69uyJw+HI9eNdrWXfbrdfcbubmxtms/myfdu0acOFCxcYNmwY1apVw8vLi5MnT9KvX78cxd2nTx8GDx7MiRMnSE1NZc2aNUyaNOmGyxERESnolOiLiIgUcd9//z19+/blvffey9yWkpJCTExMlv0qVqzIjh07rllWxYoVWbt2LTabDavVesV9SpYsCXBZ+UePHs12zNu3b2ffvn189dVX9OnTJ3P74sWLs+xXoUIFgOvGDdCzZ0+GDBnCrFmzSE5Oxmq18sADD2Q7JhERkcJCXfdFRESKOIvFctlEeRMnTryshb179+5s3br1isvQXXx+9+7dOXfu3BVbwi/uEx4ejsViYdmyZVke//jjj28o5kvLvHj7gw8+yLJfUFAQrVq14vPPP+fYsWNXjOeiwMBA7r77br755htmzJhB+/btCQwMzHZMIiIihYVa9EVERIq4e+65h6+//ho/Pz9q1KjB6tWr+eOPPwgICMiy39ChQ/n+++/p0aMHjz76KA0aNODChQssXLiQKVOmULduXfr06cP06dMZMmQI69ato2XLliQmJvLHH3/w9NNP07lzZ/z8/OjRowcTJ07EZDJRsWJFfvrpJ86cOZPtmKtVq0bFihV58cUXOXnyJL6+vsydO5fo6OjL9v3www9p0aIF9evXZ8CAAURERHDkyBF+/vlntmzZkmXfPn36ZK5AMGrUqBt/M0VERAoBJfoiIiJF3AcffIDFYmHGjBmkpKTQvHlz/vjjD9q1a5dlP29vb5YvX87//vc/5s+fz1dffUWpUqW46667MifLs1gsLFq0iNGjRzNz5kzmzp1LQEAALVq0oHbt2pllTZw4EZvNxpQpU3Bzc+P+++/nnXfeue6keRdZrVZ+/PFHBg0axJgxY3B3d6dr164888wz1K1bN8u+devWZc2aNYwcOZLJkyeTkpJCeHg4999//2XldurUiZIlS+JwOLj33ntv9K0UEREpFExGThe9FRERESlk0tPTCQ0NpVOnTnz22WfODkdERCRPaIy+iIiIFBsLFizg7NmzWSb4ExERKWrUoi8iIiJF3tq1a9m2bRujRo0iMDCQTZs2OTskERGRPKMWfRERESnyJk+ezFNPPUWpUqWYPn26s8MRERHJU2rRFxERERERESlC1KIvIiIiIiIiUoQo0RcREREREREpQlycHUBh5XA4OHXqFD4+PphMJmeHIyIiIiIiIkWcYRjEx8cTGhqK2Xz1dnsl+jl06tQpwsLCnB2GiIiIiIiIFDPHjx+nbNmyV31ciX4O+fj4ABlvsK+vr5OjkQIlMRFCQwGw7d2LNTjYyQGJ5C1bXBy/L19OWy8vrO7uzg4n+5KToUuXjNsLFoCHhzOjuXGpqZCWBs2bg6ens6MpNmw2G7///jtt27bFarU6OxyRPKX67gSXnEdy8CAEBjo3nmKksNT3uLg4wsLCMvPRq1Gin0MXu+v7+voq0ZesLJbMmzYfH6yqH1LE2QBPT098S5bEWpgSzuTkf28HBBS+RD8lBeLjwddXiX4+stlsGfXd17dAnwiK5AbVdye45DwSH5+M73jJF4Wtvl9v+Lgm4xMREREREREpQpToi4iIiIiIiBQhSvRFREREREREihCN0RcRERERKWYMwyA9PR273X7VfWw2Gy4uLqSkpFxzP8lFqakQHp5x22bLmI9F8kVBqe8WiwUXF5ebXsJdib6IiIiISDGSlpZGZGQkSUlJ19zPMAyCg4M5fvz4TScdkk0OB0yZknH7wgWIjXVuPMVIQarvnp6ehISE4OrqmuMylOiLiIiIiBQTDoeDw4cPY7FYCA0NxdXV9apJjcPhICEhAW9vb8xmjfjNF3b7v6vChIdDIZj9vagoCPXdMAzS0tI4e/Yshw8fpnLlyjmORYm+iIiIiEgxkZaWhsPhICwsDM/rLM3pcDhIS0vD3d1diX5+ubTLuLu7Ev18VFDqu4eHB1arlaNHj2bGkxP6xIqIiIiIFDNK3EUKrtz4fOoTLiIiIiIiIlKEKNEXERERERERKUKU6IuIiIhIgXX4XCJjf93Ds7M2M/bXPRw+l+jskKQQee2117jlllvy5Vj9+vWjS5cu+XKs/LB06VJMJhMxMTHODkVyQIm+iIiIiBRI3204zl3vLWXaskP8vO0U05Yd4q73ljJnw3FnhyY47yLM6tWrsVgsdOzYMV+O919HjhzBZDKxZcuWLNs/+OADvvzyy1w9Vvny5TGZTJhMJjw9Palduzaffvpprh7D2S5eUPjv3yuvvOL0mP57kaNfv36YTCaefPLJy54zcOBATCYT/fr1u2z/i38BAQG0b9+ebdu25fErUKIvIiIiIgXQ4XOJDJ+7DYcBdoeR5d9hc7dxRC37TuXMizCfffYZzz77LMuWLePUqVN5frzs8vPzo0SJErle7htvvEFkZCQ7duygd+/ePP744/zyyy+5fhxn27t3L5GRkZl/w4cPz1E5drsdh8ORy9H9KywsjNmzZ5N8cRlEICUlhZkzZ1KuXLnL9m/fvn3ma1qyZAkuLi7cc889eRbfRUr0RURERKTA+W7D8auu724ymfhWrfq5xjAMktLSr/iXnGa/bNuuyNhrXoTZHRl71fL++2cYxg3FmpCQwLfffstTTz1Fx44dL2tBf/vttyldujQ+Pj7079+flJSULI+vX7+eNm3aEBgYiJ+fH7fddhubNm3Kso/JZGLy5MncfffdeHh4UKFCBb7//vvMxyMiIgCoV68eJpOJ22+/HcjadX/atGmEhoZelnB27tyZRx99NPP+Dz/8QP369XF3d6dChQq8/sYbpKenZ3mOj48PwcHBVKhQgWHDhuHv78/ixYtv+DV9+umndO3aFU9PTypXrszChQuz7LNo0SKqVKmCh4cHd9xxB0eOHLns/Z87dy41a9bEzc2N8uXL895772V5vHz58rz55pv06dMHb29vwsPDWbhwIWfPnqVz5854e3tTp04dNmzYcFnZpUqVIjg4OPPP29sbgOjoaPr06UPJkiXx9PTk7rvvZv/+/ZnP+/LLLylRogQLFy6kRo0auLm5cezYMVJTU3nxxRcpU6YMXl5eNGnShKVLl2Y+7+jRo3Tq1ImSJUvi5eVF7dq1+f333zly5Ah33HEHACVLlryslb5+/fqEhYUxb968zG3z5s2jXLly1KtX77LX5ebmlvmabrnlFoYPH87x48c5e/bsZfvmJpc8LV1EREREJAdORCdfNQk0DIMT0clXfExuXLLNTo1Xf8uVshwG3P3Bimzvv+uNdni6Zj8l+e6776hWrRpVq1ald+/ePPfcc4wYMQKTycR3333Ha6+9xkcffUSLFi34+uuv+fDDD6lQoULm8+Pj4+nbty8TJ07EMAzee+89OnTowP79+/Hx8cncb+TIkbz99tt88MEHfP311/Ts2ZPt27dTvXp11q1bR+PGjfnjjz+oWbMmrq6ul8XZo0cPnn32Wf766y/uuusuAC5cuMCvv/7KokWLAFi+fDl9+vThww8/pGXLlhw8eJABAwbA6dP87/HHL39vHQ7mz59PdHR0lmNm9zW9/vrrjBs3jnfeeYeJEyfSq1cvjh49ir+/P8ePH6dbt24MHDiQAQMGsGHDBl544YUsx9+4cSP3338/r732Gg888ACrVq3i6aefJiAgIEsiPGHCBN566y1GjhzJhAkTePjhh2nWrBmPPvoo77zzDsOGDaNPnz7s3LnzqhfzLtWvXz/279/PwoUL8fX1ZdiwYXTo0IFdu3ZhtVoBSEpKYuzYsXz66acEBARQqlQpnnnmGXbt2sXs2bMJDQ1l/vz5tG/fnu3bt1O5cmUGDhxIWloay5Ytw8vLix07dmCxWAgLC2Pu3Ll0796dvXv34uvri4eHR5aYHn30Ub744gt69eoFwOeff84jjzyS5ULClSQkJPDNN99QqVIlAgICrvvab4Za9EVERESkwClbwp1rNfaWLelx9QelyPrss8/o3bs3kNElOjY2lr///huA999/n/79+9O/f3+qVq3Km2++SY0aNbI8/84776R3795Uq1aN6tWrM23aNJKSkjLLuKhHjx489thjVKlShVGjRtGwYUMmTpwIQFBQEAABAQEEBwfj7+9/WZwlS5bk7rvvZubMmZnbvv/+ewIDAzNbi19//XWGDx9O3759qVChAm3atGHU668z9ZKWYoBhw4bh7e2Nm5sb9913HyVLluSxxx674dfUr18/HnzwQSpVqsRbb71FQkIC69atA2Dy5MlUrFiR9957j6pVq9KrV68syTvA+PHjueuuuxg5ciRVqlShX79+PPPMM7zzzjtZ9uvQoQNPPPEElStX5tVXXyUuLo5GjRrRo0cPqlSpwrBhw9i9ezdRUVFZnle2bFm8vb0z/86fP5+Z4H/66ae0bNmSunXrMmPGDE6ePMmCBQsyn2uz2fj4449p1qwZVatW5dy5c3zxxRfMmTOHli1bUrFiRV588UVatGjBF198AcCxY8do3rw5tWvXpkKFCtxzzz00b94ci8WS+X96sZeBn59fllh79+7NihUrOHr0KEePHmXlypWZ9fK/fvrpp8zX5OPjw8KFC/n2228xm/M2FVeLvoiIiIgUOK4uZq6W5zsMKO3rlq/xFGUeVgu73mh32XaHw0F8XDw+vj5ZkpLxi/fxxYoj2K9wJcZiMvFIi/IMaVMl28fOrr1797Ju3Trmz58PgIuLCw888ACfffYZt99+O7t3775skrSmTZvy119/Zd6PiorilVdeYenSpZw5cwa73U5SUhLHjh277Hn/vf/fyfeup1evXjz++ON8/PHHuLm5MWPGDHr27Jn5Xm7dupWVK1cyevTozOfY7XZSUlJISknB859tQ4cOpV+/fkRGRjJ06FCefvppKlWqdMOvqU6dOpm3vby88PX15cyZMwDs3r2bJk2aXPM92L17N507d86yrXnz5rz//vvY7XYsFstlxyldujQAtWvXvmzbmTNnCA4Ozty+fPnyLD0QSpYsycqVK3FxcckSW0BAAFWrVmX37t2Z21xdXbMcd/v27djtdqpUyVoPU1NTM1vSBw0axFNPPcXvv/9O69at6dq1K+XLlyc7goKCMoeOGIZBx44dCQwMvOK+d9xxB5MnTwYyhiF8/PHH3H333axbt47w8PBsHS8nlOiLiIiISIGy6sA5Jv11EAATYDabMAwDE2D/J7d8beEuziek8XzrKpjN1+/+K1dnMpmu2H3e4XCQ7mrB09UlS6Lfq0k4n684fMWyDAx6Nwm/oe742fXZZ5+Rnp5OaGjov8czDNzc3Jg0aVK2yujbty/nz5/ngw8+IDw8HDc3N5o2bUpaWlqux9upUycMw+Dnn3+mUaNGLF++nAkTJmQ+npCQwOuvv063bt3+fZLdDrt24X5J1/zAwEAqVapEpUqVmDNnDrVr16Zhw4aZvRWy+5oudnO/yGQy5cmkdZce52LX/Ctt+++xIyIicjyZoYeHR5ZhAAkJCVgsFjZu3Jh5AeKii2P/H3vsMdq1a8fPP//M77//zpgxY3jzzTd58cUXs3XMRx99lGeeeQaAjz766Kr7eXl5Zbkw8+mnn+Ln58cnn3zCm2++me3XeKPUdV9ERERECoyj5xN5euYm7A6DLreE8ucLtzGgVQU61gllwG0V+WNIKx5rkTEZ2sQ/DzDg6w3Ep9icHHXxEhHoxdjudTCbwGI2Zfl3bPc6lA/0yvVjpqenM336dN577z22bNmS+bd161ZCQ0OZNWsW1atXZ+3atVmet2bNmiz3V65cyaBBg+jQoUPmpHLnzp277Hj/fd6aNWuoXr06QOb4eLvdfs2Y3d3d6datGzNmzGDWrFlUrVqV+vXrZz5ev3599u7dm5nEZ/6FhV21W3dYWBgPPPAAI0aMuOHXdC0X5x7472v+7z4rV67Msm3lypVUqVLlsmQ6t1SvXp309PQs/6/nz59n7969lw3LuFS9evWw2+2cOXPmsvf30l4EYWFhPPnkk8ybN48hQ4bw1VdfAdn7P27fvj1paWnYbDbatbu8R8zVmEwmzGZzlln784Ja9EVERESkQIhPsdH/qw3EJNmoW9aPt7vXwd1qYVj7aln2e+WeGtQI9WX4vO38sfsMXT9exbSHG1AhyNtJkRc/PRqG0ai8P99uOM6J6GTKlvTggYZheZLkQ8Y45+joaPr373/ZeOnu3bvz2Wef8eKLL9KvXz8aNmxI8+bNmTFjBjt37swyGV/lypX5+uuvadiwIXFxcQwdOvSyidYA5syZQ8OGDWnRogUzZsxg3bp1fPbZZ0DGuG0PDw9+/fVXypYti7u7+2UxXdSrVy/uuecedu7cedkY7ldffZV77rmHcuXKcd9992E2m9m6eTM7lizhzaeeuup7MXjwYGrVqsWGDRto2LBhtl/TtTz55JO89957DB06lMcee4yNGzdetqLBCy+8QKNGjRg1ahQPPPAAq1evZtKkSXz88cc3dKwbUblyZTp37szjjz/O1KlT8fHxYfjw4ZQpU+ayYQSXqlKlCr169aJPnz6899571KtXj7Nnz7JkyRLq1KlDx44dee6557j77rupUqUK0dHRLF26lKpVqwIQHh6OyWTip59+okOHDnh4eGT2BLjIYrFkDh+41oWO1NRUTp8+DWR03Z80aRIJCQl06tTpZt+ea1KLvoiIiIg4nd1hMHj2Fg6cSaC0rxvT+jTE/Rrjt7vVL8ucJ5oS7OvOgTMJdP5oJUv3nsnHiKV8oBfD2ldj4oP1GNa+Wp4l+ZDRbb9169ZXTKi7d+/Ohg0bqF69OiNHjuSll16iQYMGHD16lKf+kzB/9tlnREdHU79+fR5++GEGDRpEqVKlLivz9ddfZ/bs2dSpU4fp06cza9aszBZkFxcXPvzwQ6ZOnUpoaOg1E84777wTf39/9u7dy0MPPZTlsXbt2vHTTz/x+++/06hRI2699VYmvP8+4Ze0OF9JjRo1aNu2La+++uoNvaZrKVeuHHPnzmXBggXUrVuXKVOm8NZbb2XZp379+nz33XfMnj2bWrVq8eqrr/LGG29cNmlfbvviiy9o0KAB99xzD02bNsUwDBYtWnTZUIQrPa9Pnz688MILVK1alS5durB+/frMte7tdjsDBw6kevXqtG/fnsqVK/Puu+8CUKZMmczJEkuXLp3ZRf+/fH198fX1vWYcv/76KyEhIYSEhNCkSRPWr1/PnDlzMpdlzCsm40YXrxQA4uLi8PPzIzY29rr/uVLMJCbCP1f8bKdOYQ0JcXJAInnLFhfHor/+ooO/P1ZPz+s/oaBIToaWLTNuL18ON9j64XQpKRAfD61aQWF63ws5m83GokWL6NChw3VPMuXGvP3LHqb8fRBXFzNznmhK3bAS2XremfgUnvpmExuPRmMywbD21XiiVYVsLdtVHKWkpHD48GEiIiJwd3e/5r4Oh4O4uDh8fX3zfIbwgsRkMjF//ny6dOmS/we322Hz5ozbdeuCvmfyTUGq79f6nGY3Dy0+n1gRERERKZDmbz7BlL8zJt9757462U7yAUr5uDPz8SY82DgMw8i4YDB49haS0649flpEpChToi8iIiIiTrP5WDTD5m4H4KnbK9L5ljI3XIabi4W3utZmVJdauJhNLNx6ivumrOJkTN5OdiUiUlAp0RcRERERpzgdm8ITX28kLd1B6+qlGNq2ao7LMplMPHxrODMea0KAlys7T8Vx78QVrD10PhcjluLAMAzndNsXyUVK9EVEREQk36XY7Az4egNn4lOpUtqbCQ/cgtl88+Pqm1QIYOGzLagZ6sv5xDR6fbqWr1cfQdNSiUhxokRfRERERPKVYRi89P02tp2IpYSnlU/7NMLHPfcmHStTwoPvn2xGp7qhpDsMRv6wkxHztpOarnH7F+nCh0jBlRufTyX6IiIiIpKvJv99kIVbT+FiNvFxr/qUC8j9lSM8XC182PMWht9dDZMJZq8/zkOfrOVMfEquH6swubhaRFJSkpMjEZGrufj5vJnVXVxyKxgRERERkev5Y1cU7/y2F4D/3VuTZhUD8+xYJpOJJ2+rSNVgHwbN2szGo9HcO3ElUx9ucEMz+xclFouFEiVKcObMGQA8PT2vuhShw+EgLS2NlJQUpy83VmzYL+l1kpKS9b7kqYJQ3w3DICkpiTNnzlCiRAksFkuOy3J6oh8fH8/IkSOZP38+Z86coV69enzwwQc0atQIyHix//vf//jkk0+IiYmhefPmTJ48mcqVK1+z3I8++oh33nmH06dPU7duXSZOnEjjxo0zH09JSeGFF15g9uzZpKam0q5dOz7++GNKly6dp69XREREpLjaezqewbM3YxjQq0k5Hr41PF+Oe0fVUvwwsDmPT9/AwbOJ9Ji6mjFda9O9Qdl8OX5BExwcDJCZ7F+NYRgkJyfj4eFx1YsBksscDjh3LuO2uzvcRKInN6Yg1fcSJUpkfk5zyumJ/mOPPcaOHTv4+uuvCQ0N5ZtvvqF169bs2rWLMmXKMG7cOD788EO++uorIiIiGDlyJO3atWPXrl24u7tfscxvv/2WIUOGMGXKFJo0acL7779Pu3bt2Lt3L6VKlQLg+eef5+eff2bOnDn4+fnxzDPP0K1bN1auXJmfL19ERESkWIhOTOOx6etJTLNzawV/Xru3Zr4ev0KQNwsGNuf5b7fwx+4zvDBnK7si4xhxdzVcLMWrtdpkMhESEkKpUqWw2WxX3c9ms7Fs2TJatWp1U12I5QYkJUHHjhm3V60Cf3/nxlOMFJT6brVab6ol/yKnJvrJycnMnTuXH374gVatWgHw2muv8eOPPzJ58mRGjRrF+++/zyuvvELnzp0BmD59OqVLl2bBggX07NnziuWOHz+exx9/nEceeQSAKVOm8PPPP/P5558zfPhwYmNj+eyzz5g5cyZ33nknAF988QXVq1dnzZo13Hrrrfnw6kVERESKB5vdwVMzNnL8QjJh/h583KsBVick1z7uVqY93JAJf+xj4p8H+GzFYfaejmfSQ/Uo4ema7/E4m8ViuWZCYbFYSE9Px93dXYl+frHb4ejRjNtWa0arvuSLolbfnZrop6enY7fbL2uZ9/DwYMWKFRw+fJjTp0/TunXrzMf8/Pxo0qQJq1evvmKin5aWxsaNGxkxYkTmNrPZTOvWrVm9ejUAGzduxGazZSm3WrVqlCtXjtWrV18x0U9NTSU1NTXzflxcHJBx5edaV0KlGLLZuPjVYEtPB9UPKeJs6ekZ/xpGRpfDwsLh+Pez6nAUrtghI17DAH3P5KuLv/n67b8x//txF2sOXcDL1cKUh27Bx9Xk1Pdw0B0VqFLKi2HzdrDiwDk6TVzBlF63UKW0j9NiKohU351A55FOU1jqe3bjc2qi7+PjQ9OmTRk1ahTVq1endOnSzJo1i9WrV1OpUiVOnz4NcNm4+dKlS2c+9l/nzp3Dbrdf8Tl79uwB4PTp07i6ulKiRIlslztmzBhef/31y7b//vvveHrm/kyxUnhZUlK455/bf65ahV1XYqWYWBwdDdHRzg4j2y79rP4WGVl4P6t//eXsCIqlxYsXOzuEQmPFaRNzDlswYfBgRBoHNi7ngLOD+sez1eDTvRaORyfT9eNV9K7koG6Alp37L9X3/JPlt2nZssL721SIFfT6nt0VM5w+Rv/rr7/m0UcfpUyZMlgsFurXr8+DDz7Ixo0bnR1aFiNGjGDIkCGZ9+Pi4ggLC6Nt27b4+vo6MTIpcBITM2/e2awZ1pucSEOkoLPFx7N4+XLalCyJ1cPD2eFkX3Jy5s12ISFQmGKHjNmYExKgefPCF3shZrPZWLx4MW3atCkSXTvz2ppDF5i/diNgMKR1ZZ68rYKzQ7pM96Q0Bn+7jdWHLvD5PgvP3F6BZ++oiNmsyedU353gkvPIdq1aQWDerUohWRWW+n6xZ/n1OD3Rr1ixIn///TeJiYnExcUREhLCAw88QIUKFTJnGoyKiiIkJCTzOVFRUdxyyy1XLC8wMBCLxUJUVFSW7VFRUZnlBQcHk5aWRkxMTJZW/Uv3+S83Nzfc3Nwu2261Wgt0RRAnuKQ+WF1cVD+k6HPJ+CmxmkxYC9PyS5fEajWbs9wvFMxmMJky3n99z+Q7/f5f37HzSTz77VbSHQadbwnlmbuqOH0m6ysp5Wfl6/5NeGvRHj5feZhJSw+x90wiEx64BW83p58qFwiq7/noP+eR+n7PfwW9vmc3tgJzVuPl5UVISAjR0dH89ttvdO7cmYiICIKDg1myZEnmfnFxcaxdu5amTZtesRxXV1caNGiQ5TkOh4MlS5ZkPqdBgwZYrdYs++zdu5djx45dtVwRERERyZ74FBuPTV9PTJKNOmX9GNu9ToFM8i9ysZh5tVMN3u1RF1cXM4t3RdH1o5UcOZd4/SeLiBRATk/0f/vtN3799VcOHz7M4sWLueOOO6hWrRqPPPIIJpOJ5557jjfffJOFCxeyfft2+vTpQ2hoKF26dMks46677mLSpEmZ94cMGcInn3zCV199xe7du3nqqadITEzMnIXfz8+P/v37M2TIEP766y82btzII488QtOmTTXjvoiIiMhNcDgMnv92C/uiEijl48a0hxvibi0ca4Hf16As3z3RlNK+buw/k8C9k1bw976zzg5LROSGOb0/UmxsLCNGjODEiRP4+/vTvXt3Ro8endkl4aWXXiIxMZEBAwYQExNDixYt+PXXX7PM1H/w4EHOnTuXef+BBx7g7NmzvPrqq5w+fZpbbrmFX3/9NcsEfRMmTMBsNtO9e3dSU1Np164dH3/8cf69cBEREZEi6N3f9/LH7jO4upiZ1qchwX6FazKxW8JK8OMzLXjim41sPhbDI1+sY1j7agxoVaFA90oQEbmUyTAMTS2aA3Fxcfj5+REbG6vJ+CSrxETw9gbAduoU1kvmlxApimxxcSz66y86+PtjLUyrkCQnQ8uWGbeXLy98E9qlpEB8PLRqBYXpfS/kbDYbixYtokOHDgV6DKez/LDlJINnbwFgwgN16VqvrHMDugmp6XZeXbCTbzccB6DzLaGM7V6n0PROyA2q705wyXkkZ85AUJBz4ylGCkt9z24e6vSu+yIiIiJS+G09HsNL328D4InbKhTqJB/AzcXC291r8/q9NbGYTfyw5RT3TVnFyZjk6z9ZRMTJlOiLiIiIyE2JikthwNcbSE13cGe1UrzUrpqzQ8oVJpOJvs3K803/Jvh7ubLjZBydJ61g3eELzg5NROSalOiLiIiISI6l2OwM+HojUXGpVC7lzQc9b8FSxNagb1oxgB8GNqd6iC/nEtJ46JM1fLPmqLPDEhG5KiX6IiIiIpIjhmEwfO42th6Pwc/Dyqd9G+LjXnDHtt6MMH9P5j7VlI51Qkh3GLyyYAcvz99OWrrD2aGJiFxGib6IiIiI5MjUZYdYsOUUFrOJyb3qEx7g5eyQ8pSnqwuTHqzH0HZVMZlg5tpj9Pp0DWfjU50dmohIFkr0RUREROSGLdkdxdhf9wDwv041aFYp0MkR5Q+TycTAOyrxWd+G+Li5sP5INPdOWsG2EzHODk1EJJMSfRERERG5Ifui4hk8ewuGAQ81KcfDt4Y7O6R8d2e10ix4pjkVgryIjE2hx5TVLNh80tlhiYgASvRFRERE5AZEJ6bx2FcbSEhNp3GEP691qonJVLQm38uuikHeLBjYnDurlSI13cFz327hrUW7sTsMZ4cmIsWcEn0REZFC5nB8OmN3J/PsLgdjlxzk8LlEZ4ckxYTN7mDgzE0cu5BE2ZIeTOndAFeX4n066etu5ZM+DRl4R0UApi07RL8v1hGbZHNyZCJSnBXvb2YREZFC5rvDSdz16zmmHUzl5zMG01Yd4673ljJnw3FnhybFwKifdrHq4Hm8XC182rch/l6uzg6pQLCYTQxtV42PHqqPh9XC8v3nuPejFeyLind2aCJSTCnRFxERKSQOx6czfEMcDsBu8O+/Bgybu40jatmXPDRj7VGmr85YO37CA7dQLdjXyREVPB3rhDD3qWaUKeHB0fNJdP1oJb/tPO3ssESkGFKiLyIiUkh8dySZqw2FNplMfKtWfckjaw6d538/7ATgxbZVaFsz2MkRFVw1Qn358dkWNK0QQGKanSe+3sj7f+zDoXH7IpKPlOiLiIgUEicS7RhXyRUMw+BEdHL+BiTFwvELSTz1zUbSHQad6oYy8I5Kzg6pwPP3cmV6/8b0a1YegPf/2M+T32wkITXduYGJSLGhRF9ERKSQKOtluWqLPkDZkh75F4wUCwmp6Tz21Qaik2zULuPHuO51iu0M+zfKajHz2r01GXdfHVwtZn7fFUW3j1dy9LyG2IhI3lOiLyIiUkjcX96Dq/X+dRjg4+6SvwFJkeZwGDz/7Rb2RsUT5OPGtD4N8HC1ODusQuf+hmHMfuJWSvm4sS8qgXsnrWT5/rPODktEijgl+iIiIoWEv5sZ93/yLPM/f5ZLGlfH/bqX1xbuxGZ3OCM8KWLGL97H4l1RuLqYmfpwA0L81GMkp+qXK8mPz7bglrASxCbb6Pv5Oj5dfgjjamNxRERukhJ9ERGRQmLq3kSS7VDe28KAim50LGViQLNy/PnCbTzXujIAX646Qp/P1nEhMc3J0UphtnDrKSb9dQCAt7vVpn65kk6OqPAr7evO7AG3cl+DsjgMePPn3Qz5bispNruzQxORIkiJvoiISCEQlWzn8/0ZY3v/r64Pw2t4MLGGmWF3VaRCkDfPta7C1Icb4OVqYfWh89w7aQW7I+OcHLUURttPxDJ0zlYAnmhVgW71yzo5oqLD3Wrhnfvq8L9ONbCYTczffJL7p64mMlYTaYpI7lKiLyIiUgi8vyuBFDs0DLDSOsTtivu0qxnM/IHNCQ/w5ER0Mt0+XsXP2yLzOVIpzM7EpfD49A2kpju4o2oQL7Wv5uyQihyTycQjzSP4+tHGlPC0su1ELJ0mrmTDkQvODk1EihAl+iIiIgXcwfh0vjuc0eI3vI7PNWc9r1Lahx8GNqdl5UCSbXYGztzEO7/t0Rrecl0pNjsDvt7I6bgUKpXy5oMH62Exa4b9vNKsUiA/PtOCasE+nEtI5cFP1jBr3TFnhyUiRYQSfRERkQLu3e3x2A1oHepGw0DX6+5fwtOVL/o14vGWEQB89NdBHp++gbgUW16HKoWUYRi8PG87W47H4Odh5dM+DfF1tzo7rCIvzN+TuU81o0PtYGx2gxHztvPKgu2kpWtCTRG5OUr0RURECrDN59P45WQqZuClWt7Zfp6Lxcz/dazBhAfq4upiZsmeM3T5aCUHzybkXbBSaH2y/BDzNp/EYjbx0UP1KR/o5eyQig0vNxc+eqg+L7atgskE36w5Ru/P1nIuIdXZoYlIIaZEX0REpIAyDIOx2zMS8+7lPajid+MtrF3rleX7J5sS4ufOobOJdJm0kr/2nMntUKUQ+2vPGcb8sgeAkR2r06JyoJMjKn5MJhPP3FmZT/s0xNvNhXWHL3DvxBXsOBnr7NBEpJBSoi8iIlJA/R2Vxpqzabia4fma2W/N/686ZUvwwzPNaRhekvjUdB79aj0fLz2gNbyFA2fiGTRrM4YBDzYOo2+z8s4OqVi7q3ppFgxsToVAL07FpnDflFX8sOWks8MSkUJIib6IiEgB5DAMxm6LB6BvJU9CPS03VV4pH3dmPn4rDzYuh2HAuF/38uyszSSnaQ3v4iomKY3HvtpAfGo6jcv78/q9ta450aPkj0qlvJk/sDm3Vw0ixeZg8OwtjPllN3ZNqCkiN0CJvoiISAG08FgKu2PT8bGaeLpazlvzL+XqYmZMt9q82aUWLmYTP22LpPvkVZyITsqV8qXwSLc7eGbmZo6cT6JMCQ8m966Pq4tOCwsKPw8rn/VtxFO3VwRg6t+HePTL9cQmaUJNEckefaOLiIgUMKl2g3d3ZIzNf7KqFyXdcvfnuvet4cx4rAkBXq7siozj3kkrWXPofK4eQwq2N3/ezYoD5/B0tfBp34YEeLs5OyT5D4vZxLD21fjwwXq4W838ve8snT9awf6oeGeHJiKFgBJ9ERGRAmbmoSROJNkp5W7m0cp5M/t5kwoBLHy2BbXK+HIhMY3en65l+uojGrdfDMxad4wvVx0BYPz9t1A9xNe5Ack13Vs3lO+fbEaZEh4cOZ9E149XsXhXlLPDEpECTom+iIhIARJvczBxV0Zr/nM1vfFwybsx02VKeDDniWbcWzeUdIfBqz/sZPjc7aSma9x+UbXu8AVe/WEHAEPaVKF9rWAnRyTZUauMHwufaU7jCH8SUtN5fPoGPlyyH4fG7YvIVSjRFxERKUA+2ZfIhTSDCt4W7i/vkefH83C18EHPW3i5QzXMJvh2w3EenLaGM3EpeX5syV/HLyTx5DcbsdkNOtYJ4dk7Kzk7JLkBAd5uzHisCX2ahgMwfvE+Bs7cRGJqupMjE5GCSIm+iIhIAXE2xc6nezMmxhta2wcXc/7MgG4ymRjQqiJfPNIYX3cXNh2LodOkFWw5HpMvx5e8l/hPK/CFxDRqlfHl3fvqaob9QshqMfNG51q83a02VouJX3acpvvkVRw7rwk1RSQrJfoiIiIFxMRdiSTZDer6W2lfJv8nR7utShA/PNOCSqW8iYpL5f6pq5m78US+xyG5y+EwGPLdFvacjifQ241pDzfEw/XmlmsU5+rZuByzB9xKoLcbe07Hc+9HK1h54JyzwxKRAkSJvoiISAFwNCGdmYcyWuWG1/Z2WmtrRKAX859uRuvqpUlLd/DCnK288eMu0u0Op8QjN+/9P/bx284oXC1mpj7cgNASeT8kRPJeg3B/fny2OXXL+hGTZKPP5+v4bMVhTagpIoASfRERkQLhvR0JpBtwW7ArTUs5d6kzH3cr0x5uwKC7KgPw+crD9P1iHdGJaU6NS27cT9tO8eGfBwB4q1ttGoSXdHJEkptC/Dz49ommdKtfBrvDYNRPu3hxzjZSbJpQU6S4U6IvIiLiZDuibSw8noIJGFbbx9nhAGA2mxjSpgpTetfH09XCygPnufejFew5Hefs0CSbdpyM5cU5WwF4vGUE9zUo6+SIJC+4Wy2816MuI++pgdkEczed4IFpazgdqwk1RYozJfoiIiJONnZ7PACdy7lTo4TVydFk1b5WCPOebkaYvwfHLyTT7eNV/LI90tlhyXWciU/h8ekbSLE5uK1KEMPvru7skCQPmUwm+reIYPqjTfDzsLL1eMaEmhuPRjs7NBFxEiX6IiIiTrQyKpXlUWlYTfBCLW9nh3NF1YJ9WTiwBc0rBZCUZuepGZsY//tereFdQKWm23ny641ExqZQIciLDx+shyWfVnAQ52pROZAfn2lB1dI+nI1P5cFpa/h2/TFnhyUiTuDURN9utzNy5EgiIiLw8PCgYsWKjBo1KsskIlFRUfTr14/Q0FA8PT1p3749+/fvv2a5t99+OyaT6bK/jh07Zu7Tr1+/yx5v3759nr1WERGR/3IYBm//05rfq6InYV4uTo7o6kp6ufLVI43p3yICgA//PMCArzcSn2JzcmRyKcMw+L/5O9h0LAZfdxc+69sIP4+C1UtE8la5AE/mPd2Mu2sFk2Z3MGzudl79YQc2TagpUqw4NdEfO3YskydPZtKkSezevZuxY8cybtw4Jk6cCGT8WHXp0oVDhw7xww8/sHnzZsLDw2ndujWJiYlXLXfevHlERkZm/u3YsQOLxUKPHj2y7Ne+ffss+82aNStPX6+IiMilFp1IYXt0Ol4uJp6p7uXscK7LxWJm5D01eK9HXVxdzPyxO4quH6/i8Lmr/yZL/vpsxWG+33gCswk+6lWfiMCCX68k93m5ufDRQ/UZ0qYKANNXH6X3p2s5n5Dq5MhEJL84NdFftWoVnTt3pmPHjpQvX5777ruPtm3bsm7dOgD279/PmjVrmDx5Mo0aNaJq1apMnjyZ5OTkaybl/v7+BAcHZ/4tXrwYT0/PyxJ9Nze3LPuVLKmZaEVEJH/YHAbv7kgAYEBVLwLdC8+65t0blOW7J5pS2teNA2cS6DxpBUv3nnF2WMXeX3vP8Nai3QC80rEGLSsHOTkicSaz2cSguyrzSZ+GeLu5sPbwBe6dtJKdp2KdHZqI5AOn9hFs1qwZ06ZNY9++fVSpUoWtW7eyYsUKxo8fD0BqasZVR3d398znmM1m3NzcWLFiBY899li2jvPZZ5/Rs2dPvLyyXtVeunQppUqVomTJktx55528+eabBAQEXLGM1NTUzHgA4uIyZh222WzYbOq2KJew2bjYSdKWng6qH1LE2dLTM/41DHAUoq6hDse/n1WHI99jn3komSMJdgLcTPSt5J4Rw41wOMAwwEnfMzWDvZj35K08M2sLm4/H8uiX63mxbWUea14ek6nojge/+Jtf0H77D55N5NmZm3EY0KNBGXo3LlPgYhTnuL2yP98NaMxTM7Zw9EIS3Sev4u2utehYO/i6zy2o9b1I03mk0xSW+p7d+EzGpQPi85nD4eDll19m3LhxWCwW7HY7o0ePZsSIEUDGi6hUqRJNmjRh6tSpeHl5MWHCBIYPH07btm357bffrnuMdevW0aRJE9auXUvjxo0zt8+ePRtPT08iIiI4ePAgL7/8Mt7e3qxevRqL5fJWlddee43XX3/9su0zZ87E09PzJt4FKWosKSnc07MnAD/Nno39kgtVIlJwOPOzmmqHNzdbiLOZuC/CTsvgwjupXboD5hw2s+ZMRifB+gEOHqzowLXwdFAo9JLSYfx2C2dTTFTwMRhYw46LpluW/0hKh6/2mdkTm1E5Woc66FjOgeZpLFh0HinXk5SUxEMPPURsbCy+vr5X3c+pif7s2bMZOnQo77zzDjVr1mTLli0899xzjB8/nr59+wKwceNG+vfvz9atW7FYLLRu3Rqz2YxhGPzyyy/XPcYTTzzB6tWr2bZt2zX3O3ToEBUrVuSPP/7grrvuuuzxK7Xoh4WFce7cuWu+wVIMJSZi/WcYSNKxY1iDr3/FXKQws8XHs3j5ctqULInVw8PZ4WRfcjLW224DwPb335CPsX+8J5EJOxMJ8zLza9sAXHNypp2SAgkJ0Lx5vsZ+JYZhMGPdcUYv2ku6w6BGiA+TH7qF0BKFqD5kk81mY/HixbRp0war1fmT3KXbHTz29WZWHjxPqJ87855sQoC3m7PDkgLK7jB4d/F+Pl1xBIDbKgcyvkdtfK8yYWNBq+/FwiXnkbZTpyAw0MkBFR+Fpb7HxcURGBh43UTfqV33hw4dyvDhw+n5z1Wr2rVrc/ToUcaMGZOZ6Ddo0IAtW7YQGxtLWloaQUFBNGnShIYNG163/MTERGbPns0bb7xx3X0rVKhAYGAgBw4cuGKi7+bmhpvb5T+cVqu1QFcEcYJL6oPVxUX1Q4o+l4yfEqvJhNVciJoRL4nVajZnuZ+XLqQ6+GRvEgAv1vLByyWHTd9mM5hMGe9/AfieeaRFRaqHluDpGZvYFRlPtylrmdy7AY0j/J0dWp4oKL//b/26k5UHz+NhtfBJ34YElyyYSzRKwWAFXrmnJrXLluCl77fx9/5z9Ji2jml9GlKp1NXrTkGp78XCf84jC8L3e3FT0Ot7dmNz6hlZUlIS5v+cWFksFhxXGKfo5+dHUFAQ+/fvZ8OGDXTu3Pm65c+ZM4fU1FR69+593X1PnDjB+fPnCQkJyf4LEBERuUEf7U4gId2gZgkXOoUVrS6Zt1YIYOEzzakR4sv5xDQe+mQN36w56uywiqxv1x/ji5VHAJjwQF1qhvo5NyApNDrfUobvn2xGqJ87h84l0vWjlSzZHeXssEQkFzk10e/UqROjR4/m559/5siRI8yfP5/x48fTtWvXzH3mzJnD0qVLM5fYa9OmDV26dKFt27aZ+/Tp0ydzXP+lPvvsM7p06XLZBHsJCQkMHTqUNWvWcOTIEZYsWULnzp2pVKkS7dq1y7sXLCIixdqJRDtfH8xozR9W2wdzEZy0rmxJT+Y+1Yx76oSQ7jB4ZcEORszbTlp6IZqosRBYf+QCryzYAcDzravQvpYaKuTG1C7rx8JnW9C4vD/xqek8Nn0DH/11ACeO6hWRXOTUrvsTJ05k5MiRPP3005w5c4bQ0FCeeOIJXn311cx9IiMjGTJkCFFRUYSEhNCnTx9GjhyZpZxjx45d1jNg7969rFixgt9///2y41osFrZt28ZXX31FTEwMoaGhtG3bllGjRl2xe76IiEhuGL8znjQHNC/lSsvSrs4OJ894uFqY+GA9aob6Me63Pcxad4z9UfFM7t2AIB/9zt6sE9FJPPn1Rmx2gw61g3n2zkrODkkKqUBvN755rAlv/LSTb9Yc453f9rLrVBzv9KhDVFwqs9ceZf0+M7tc9tOzSTgRgV7XL1RECgSnTsZXmMXFxeHn53fdSRCkGEpMBO+McW62U6ewajiIFHG2uDgW/fUXHfz9sRamVUiSk6Fly4zby5fn+YR2e2Jt3P37eQxg4V0B1PG/yfF/KSkQHw+tWkEBft//2nOGQbM3E5+SToifO1MfbkCdsiWcHVaO2Ww2Fi1aRIcOHZwyhjMpLZ3uk1ezOzKOGiG+fP9UUzxdndpuI0XEzLXH+N/CHdjsBsG+7pyJT8GECYfhwGwyY2AwtnsdejQMc3aoRdsl55GcOQNBQc6Npxhx9vd7dmU3Dy1EsyaJiIgUXuO2J2AAHcu633ySX4jcUa0UPwxsTsUgLyJjU+gxZTXzN59wdliFksNh8MJ3W9kdGUegtyuf9G2oJF9yzUNNyjHz8Vsp6WnldFwKDgPshoGBCbth4DBg2NxtHDmX6OxQRSQblOiLiIjksbVn0/gzMhWLCV6sVfxmRa8Q5M38gc25q1opUtMdPP/tVt78aRfpdo3bvxEfLNnPLztOY7WYmPpwA8oUweULxbkalffnnjqhV33cZDLx7Ybj+RiRiOSUEn0REZE8ZBgGb2+LB6BnhAcRPsWzBdbX3confRryzB0Z48k/XXGYR75cT0xSmpMjKxwWbY/kgyX7ARjdtTYNwovmsoXifDHJNsxXmSfU7jD4adsp5m48wfELSZq4T6QAK55nGyIiIvnk91OpbL5gw8NiYnCN4teafymz2cSL7apSPcSXF+dsZfn+c9w7aSWf9GlI1WAfZ4dXYO08FcsL320FoH+LCO7XGGnJQ2VLemAymeAqSfzxC8m8MCejPob6udM4wp/GEQE0qeBPhUCvjOeKiNMp0RcREckj6Q6Dd7ZntOb3r+JJKQ+LkyMqGDrWCSEi0IsBX2/g2IUkun68kvH330L7WsHODq3AORufyuNfbSDZZqdVlSBG3F3N2SFJEXd/wzCm/n3wio+ZTNCzUTn2nI5j+4lYTsWmsGDLKRZsOQVAoLdrRuJf3p8mFQKoWtoH89W6B4hInlKiLyIikkfmHk3mQLydkq4mBlTVslSXqhHqy8JnWjBwxiZWHzrPk99sZPBdlRl8V2UlBv9ITbfz5DcbORWbQoVALyY+WA8Xi0ZdSt6KCPRibPc6DJu7DZPJhMPhwGwyYUCWWfeT0tLZfCyGtYcvsO7weTYfi+FcQhqLtp9m0fbTAPi6u/zT4p/R6l8r1Fd1WCSfKNEXERHJAyl2gwk7EwAYWN0bX6tObv/L38uV6f0bM/rn3Xy56ggfLNnP7sg4xj9wC95uxfsUxTAMXpm/g41Ho/Fxd+GTvg3x8yg+qzWIc/VoGEaj8v7MWnuU9bsO0qhGBR5sEk75wH8vWHq6utC8UiDNKwUCGRemtp2IZd3hC6w9fIGNRy4Ql5LOH7vP8MfuM/88x0KD8JI0+Sfxr1PWD3erejqJ5IXi/SsqIiKSR77cn8jpZAdlPM30rlhw17l3NqvFzGv31qRGqC+vzN/B77ui6PbxSqY93DBLUlHcfL7yCHM2nsBsgo8eqk/FoOI9v4Pkv/KBXrzYtjKL0vfToW3l664r7uZioVF5fxqV92fgHZBud7DzVFxm4r/+yAVik20s33+O5fvPAeDqYqZeWInMxL9+eAktGSmSS/RJEhERyWWxaQ4+3pOx1vSQmj64W9QV/XrubxhGpVLePPn1RvZFJXDvpBVMeqg+raoEOTu0fPf3vrOM/nkXAP/XsUaxfA+k8HOxmKkbVoK6YSV4vFUFHA6DvVHxrDt8ITP5P5eQytp/bsMBXMwmapf1o3GEP00i/GkQ7q+eLCI5pERfREQkl03ek0iczaCqrwtdwt2dHU6hUb9cSX58tgVPfrORzcdi6PfFOkbcXZ3HWkYUm5m8D51N4JmZm3AY0KNBWR5tXt7ZIYnkCrPZRPUQX6qH+NK3WXkMw+DQucR/E/9D5zkVm8LmYzFsPhbD1L8PYTJB9WBfmlTISPwblfcnwNvN2S9FpFBQoi8iIpKLIpPsfLE/ozV/WB1vLMUkQc0tpX3dmT3gVkYu2MF3G04wetFudkXGMaZb7SI/ljc22cZjX20gPiWdBuElebNrrWJzgUOKH5PJRMUgbyoGefNg43IAnIhOYu2hjMR/3ZELHD6XyK7IOHZFxvHFyiMAVCrlndni3yQigGA/XUwVuRIl+iIiIrnog10JpDqgcaCVO4LV8pQTbi4WxnavQ81QP974aRfzN5/kwJkEpj7cgNASHs4OL0+k2x08O2szh84lEurnzpTeDXBzKdoXNkT+q2xJT8o28KR7g7IAnIlLYd2RC5nJ/96oeA6cSeDAmQRmrj0GQDl/z8yZ/W+NCCDM30MXyERQoi8iIpJrDsSl893hZACG1fHRyeZNMJlM9G1WnsqlvRk4YxPbT8Zy76QVTO7dgEbl/Z0dXq57+5c9LNt3Fg+rhWl9GhLko4tEIqV83bmnTij31AkFIDoxjfVH/h3jv/NULMcuJHHsQhLfbzwBQLCv+7+JfwV/KgZ567tYiiUl+iIiIrnknR3xOIC2oW40CHB1djhFQrOKgSx8pgWPT9/AntPxPPTJGl6/txYPNSnn7NByzZwNx/l0xWEA3ru/LrXK+Dk5IpGCqaSXK21rBtO2ZjAA8Sk2Nh6Nzkz8t52I4XRcCgu3nmLh1lNAxjKejcv7Zyb/1UN8sZiV+EvRp0RfREQkF2w8n8ZvJ1MxAy/V9nF2OEVKmL8n855uxtA52/h5eyQvz9/OzlOx/K9TTVxdzM4O76ZsPHqB/5u/A4DBd1WmQ+0QJ0ckUnj4uFu5vWopbq9aCoDkNDubj0dnTvC36Vg0FxLT+HXnaX7defqf57jQ6JLEv3YZP6yWwv09InIlSvRFRERukmEYjN0WD0CPCA8q+ernNbd5urow6aF61Fjqy7u/72XG2mPsj0rg4971CSyks3CfjEnmia83kmZ3cHetYAbfVdnZIYkUah6uFppVDKRZxUAA0tIdbD8Zw9p/Ev8NR6KJT0nnzz1n+HPPmYznWC00CC+ZmfjfElaiyE/8KcWDzkRERERu0tLTaaw7Z8PNDM/V8HZ2OEWWyWRi4B2VqB7iw+BZW1h35AL3TlzBtD4NC11396S0dB7/agPnEtKoHuLLe/fXxazuxCK5ytXFTINwfxqE+/P07RmTXu6OjGft4fOZM/vHJNlYceAcKw6cy3iOxcwtYSUyE//64SXxdlPKJIWPaq2IiMhNsBsGY7dntOb3q+xFiKdagvLandVKM39gcwZM38Chc4l0n7yKcffVofMtZZwdWrYYhsHQOdvYFRlHgJcrn/RpgKerTslE8pqLxUztsn7ULuvHYy0r4HAYHDibwNpD51n7zzj/s/GprDuScRGAv8BiNlEr1JcmFQJoXN6fRuX98fO0OvuliFyXflVERERuwg/HUtgTm46v1cTT1bycHU6xUamUN/MHNue52Zv5a+9ZBs/ews5TcQxrX63AT7T14ZID/Lw9EqvFxJSHG1C2pKezQxIplsxmE1VK+1CltA8PNy2PYRgcPZ/EusMXWPNPq/+J6GS2nohl64lYpi07hMkEVUv7cGuFABpHZCT+WiVDCiIl+iIiIjmUajd4b0dGa/5T1bzwc9WETvnJz8PKp30b8d7ve/l46UGmLTvEntPxTOxZr8C2uP26I5IJf+wD4M0utYrkUoEihZXJZKJ8oBflA724v1EYkDGXxvrDF1h7OKPV/9DZRPacjmfP6Xi+XHUEgApBXjSJ8KdJREbyH1rCw4mvQiSDEn0REZEc+uZgEieTHAR7mHmkslrzncFiNvFS+2pUD/Fl6PdbWbbvLJ0/WsEnfRpSuXTBWv1g16k4nv92KwCPNC/PA42KzhKBIkVVmRIelKlXhi71MoYGnY1PZf2RC5lL+u05Hcehs4kcOpvIrHXHAShb0oPGEf6ZyX94gCcmU8HuaSRFjxJ9ERGRHIizOZi0OwHImIDP3aKTOGfqVDeUCkFeDJi+kSPnk+jy0UomPHBL5nrbznYuIZXHp28g2WanZeVA/q9DdWeHJCI5EOTjRofaIZlLYcYkpbHhSDTrjmQk/jtOxnIiOpkT0SeZt+kkAKV83DIT/8YRAVQu5X3VyTePnEuk/D+3xy89RNc7PIkI1IVkuXFK9EVERHLgk72JRKcZVPSxcF95ddMsCGqG+rHwmeY8PWMTaw9fYMDXG3m+dRWevbOSU2e0T0t38NQ3GzkZk0xEoBeTHqyPi9btFikSSni60rpGaVrXKA1AQmo6m45GZ8zqf/gCW47HcCY+lZ+2RfLTtkgASnpaaVTe/5/kP4Aaob5YzCa+23Cc12etY+c/ZX++5gSTNp1jbPc69GgY5qRXKIWVEn0REZEbdCbFzqf7kgAYWtsHlwI++VtxEuDtxjePNeHNn3bx1eqjTPhjH7sj43jv/rp4OWGJLMMwGLlgB+uPROPj7sInfRoW2PkDROTmebu50KpKEK2qBAGQYrOz5XhMZuK/8Wg00Uk2ft8Vxe+7ojKfUyPUh/WHo3E3/i3LboDDgGFzt9GovD/l1bIvN0CJvoiIyA36cFcCyXaDev5W2oVqtuWCxmox83rnWtQI9WXkgp38uvM0hz9O5JM+DSkXkL8z3H+56gjfbjiO2QQTH6xHpVLe+Xp8EXEud6uFWysEcGuFAABsdgc7Tsay9p/Ef/2RC8SnpLPucPRVyzAMeP67LXSsHUKAtyv+Xm74e7ri7+1KgJcr7lYt6yqXU6IvIiJyAw7HpzP7UDIAw+v4aIKlAuyBRuWoVMqHJ7/ZyN6oeO79aAWTHqxPi8qB+XL85fvPMuqnXQC83KE6t1ctlS/HFZGCy2oxU69cSeqVK8mTt1XE7jDYczqOYd9vY8epuCs+xwA2H4th87GYKz7u6WqhpKfrPxcBXDMuAnj9eyHg38fc8PdyxdfdRb9dxYASfRERkRvw7o4E0g24I9iNJkGuzg5HrqNBeEl+fKYFT3yzka3HY+jz+Vpe7lCd/i0i8vRE9/C5RAbO2ITDgPsalKV/i4g8O5aIFF4Ws4maoX60rBLE7tPxV9zHBNQq40dEoBcXEtM4n5hGdGIaFxLTSLM7SEqzk5SWzMmY5Gwd08VsoqTXvxcBLl4Q8P/PX4CXGyW9rPh7umpekUJIib6IiEg2bY+28fOJFEzAS7XVBbuwCPZz59sBt/J/83cwd9MJ3vx5N7si43ira+086fIal2Ljsa/WE5eSTv1yJRjdtZZaz0Tkmu5vGMbUvw9e8THTP0N//jtG3zAMElLTufBP0n/xIsCFfy4EnL9k28ULAwmp6aQ7DM7Gp3I2PjXb8fl5WP9zEcA182LBlS4QeLhqOIGzKdEXERHJprHbMlpbuoa7U72EJlQrTNytFt7tUYeaob6MXrSbeZtOcvBMAlMfbkiwn3uuHcfuMHh25mYOnk0kxM+dKQ83wM1FJ7wicm0RgV6M7V6H12aty9xmMYHZBGO717niRHwmkwkfdys+7lbCA7I3UV+KzU5Mko3zialZLhBkXiRISONC0r/bopPSMAyITbYRm2zj8LnEbB3H3Wom4J+hApdfCLj8IoGvu9Wpq6MURUr0RUREsmF5VCorzqThaobna6o1vzAymUw82iKCqsE+DJy5ia0nYuk0aQVTetenQbh/rhxj7K97+HvfWdytZj7p05BSPrl3EUFEirYeDcNoHOQG4zPuP3prWbrdUStXZ9t3t1oI9rNk+wKn3WEQm2zjQmIq5xP+uQCQlHFB4PwlFwMyH/tnOEGKzcHJmOwPJ7CYTRlzCXi5UtLLesWLBJdeICjp5Yo1F4cTHD6XyOy1R1m/z8wul/30bBJORCFf5UCJvoiIyHU4DCOzNb93RU/CvPTzWZg1rxTIwoEtGPD1BvacjqfntDWM6lyLno3L3VS53288wbRlhwB4t0ddapXxy41wRaQYCb8kuRxyewVwcrJpMZsyE+1K2ZhP1DAMEtPs/1wIuLzXwJWGGMSnpmN3GJxLSOVcQvaHE/i6u1xyIcDt2sMJvF3xsFquOIzquw3HGT53GyZMOAwTW1cc4ZMVhxnbvQ49GobdyNtVoOhMRURE5Dp+Op7Cjph0vF1MPFNdrflFQbkAT+Y+1YwX52zllx2nGT5vO7si4xh5T40ctRJtPBrNy/O2AzDozkrcUyc0t0MWESnwTCYT3m4ueLu5ZHs509R0O9GJGcMJLv57rQsD0UlpOAyIS0knLiWdI+eTsnUcNxdzxkWAzCUKrVjMZuZtOoEBZKxvYMJuZNwbNncbjcr752qPivykRF9EROQa0hwG7+1IAOCJql74u2nm4aLCy82Fj3vVZ9KfB3hv8T6mrz7K3tPxfNyrPgHebtku51RMMk98vZE0u4N2NUvzXOsqeRi1iEjR4uaS8+EEFxL/GVZwhfkFzif8M6wgMY20dAep6Q5OxaZwKjYlW8cxmUx8u+E4w9pXu5mX5zRK9EVERK5h9qEkjibaCXQz079K9lonpPAwmUw8e1dlqoX48vy3W1h7+AL3TlrJ1IcbZKvrfXKanQFfb+BcQirVgn0Yf/8tmlBKRCQPXTqcIDsuDif4dyWC1MyLAN+uP87Bs1eeYNAwDE5EZ2+OgYJIib6IiMhVJKY7+HBXxgnA4JreeLqoNb+oalOjNAsGNuPx6Rs5fC6R+6asYtx9dbm37tW74BuGwdDvt7LjZBz+Xq580qchXm46tRIRKUguHU4Q5p/1gn10ko1pyw5hdxhXfF7Zkh75FWau0xmLiIjIVXy2L4lzqQ7Ke1voGVF4f+wleyqV8mHBwObcViWIFJuDQbM28/Yve654Aggw6c8D/LQtEqvFxJTeDS47gRQRkYLt/oZhGMaVv+MNw+CBQjwZn1MTfbvdzsiRI4mIiMDDw4OKFSsyatSoLG92VFQU/fr1IzQ0FE9PT9q3b8/+/fuvWe6XX36JyWTK8ufunnXMh2EYvPrqq4SEhODh4UHr1q2vW66IiBQf51MdTN2b0Zr/Yi1vrOqOXSz4eVj5vF8jnritAgBT/j5I/6/WZ64f/e7v+/lqn5mBM7fw3uJ9AIzqXIvGEbmzPJ+IiOSfiEAvxnavg9mUMSTAhIHFBGYTjO1ep9BOxAdO7ro/duxYJk+ezFdffUXNmjXZsGEDjzzyCH5+fgwaNAjDMOjSpQtWq5UffvgBX19fxo8fT+vWrdm1axdeXld/4319fdm7d2/m/f8upTBu3Dg+/PBDvvrqKyIiIhg5ciTt2rVj165dl10UEBGR4mfS7gQS0w1ql3ShQ1n9LhQnFrOJEXdXp0aILy99v42le89y13tLuZCYlrn8knH+DADNKwXc9LJ8IiLiPD0ahtGovD+z1h5l/a6DNKpRgQebhBfqJB+cnOivWrWKzp0707FjRwDKly/PrFmzWLduHQD79+9nzZo17Nixg5o1awIwefJkgoODmTVrFo899thVyzaZTAQHB1/xMcMweP/993nllVfo3LkzANOnT6d06dIsWLCAnj175ubLFBGRQuZ4YjrfHMhYrmdYbR/MV1h3V4q+zreUoWKQN/2/XE9U/MW1nTOWX7po9cHzHDmXWOhPCEVEirPygV682LYyi9L306FtZaxWq7NDumlOTfSbNWvGtGnT2LdvH1WqVGHr1q2sWLGC8ePHA5CamvGjemkLu9lsxs3NjRUrVlwz0U9ISCA8PByHw0H9+vV56623Mi8WHD58mNOnT9O6devM/f38/GjSpAmrV6++YqKfmpqaGQ9AXFwcADabDZvNdhPvghQ5NhsXvxps6emg+iFFnC09PeNfwwCHw8nR3ACH49/PqsORJfZ3d8RjM6B5KStNgqwZjxc0DgcYBuh7Jk9VLeVJu5qlmL7m+BUfNwGz1h7lxbaV8zcwkXxw8RxX57r5SOeRTlNY6nt243Nqoj98+HDi4uKoVq0aFosFu93O6NGj6dWrFwDVqlWjXLlyjBgxgqlTp+Ll5cWECRM4ceIEkZGRVy23atWqfP7559SpU4fY2FjeffddmjVrxs6dOylbtiynT58GoHTp0lmeV7p06czH/mvMmDG8/vrrl23//fff8fTU5DvyL0tKCvf8c/vPVauwayiIFBOLo6MhOtrZYWTbpZ/V3yIjMz+rJxNh4TELYKJp6WQWnSrgS+v89ZezIyjyduw3kzFy8/KeHQ7DYP2ugyxK1zw/UnQtXrzY2SEUG1l+m5Yt03mkExT0+p6UlJSt/Zya6H/33XfMmDGDmTNnUrNmTbZs2cJzzz1HaGgoffv2xWq1Mm/ePPr374+/vz8Wi4XWrVtz9913X3V2RICmTZvStGnTzPvNmjWjevXqTJ06lVGjRuUo1hEjRjBkyJDM+3FxcYSFhdG2bVt8fX1zVKYUUYn/rsV5Z7NmWK8yhESkqLDFx7N4+XLalCyJ1aMQzUyf/G8C3y4kBP6J/bEVMRik0bGsG09UKeWs6K4vJQUSEqB588zYJW/sctnP1hVHsF/h3MNsMtGoRgU6qEVfiiCbzcbixYtp06ZNkejKXChcch7ZrlUrCAx0YjDFS2Gp7xd7ll+PUxP9oUOHMnz48Myu8rVr1+bo0aOMGTOGvn37AtCgQQO2bNlCbGwsaWlpBAUF0aRJExo2bJjt41itVurVq8eBAwcAMsfuR0VFERISkrlfVFQUt9xyyxXLcHNzw83N7YplF+SKIE5wSX2wuriofkjR55LxU2I1mbCaC9GqrZfEajWbwWxm9ZlU/o5Kw8UEQ2v7FOzXYzaDyZTx/ut7Jk/1bBLOJysOX/ExA3iwSbi+66VI0/luPvrPeaS+3/NfQa/v2Y3NqWcwSUlJmP9zEmWxWHBcYSykn58fQUFB7N+/nw0bNmROopcddrud7du3Zyb1ERERBAcHs2TJksx94uLiWLt2bZaeACIiUnwYhsHb2xMAeLCCJ+W9nXotXAqQorz8koiIFE1OPYvp1KkTo0ePply5ctSsWZPNmzczfvx4Hn300cx95syZQ1BQEOXKlWP79u0MHjyYLl260LZt28x9+vTpQ5kyZRgzZgwAb7zxBrfeeiuVKlUiJiaGd955h6NHj2ZO3mcymXjuued48803qVy5cubyeqGhoXTp0iVf3wMRESkYfjuZytYLNjwtJp6tocRNsiqqyy+JiEjR5NREf+LEiYwcOZKnn36aM2fOEBoayhNPPMGrr76auU9kZCRDhgzJ7Gbfp08fRo4cmaWcY8eOZekZEB0dzeOPP87p06cpWbIkDRo0YNWqVdSoUSNzn5deeonExEQGDBhATEwMLVq04Ndff80yw7+IiBQP6Q6DcTsyWvMfq+JJKXeLkyOSgqgoLr8kIiJFk1MTfR8fH95//33ef//9q+4zaNAgBg0adM1yli5dmuX+hAkTmDBhwjWfYzKZeOONN3jjjTeyG66IiBRR844mcyjejr+ricerqoVWRERECrcCPMuQiIhI/vhod8Ysx8/U8MbHqp9GERERKdx0NiMiIsXemRQHZT0t9Krg6exQRERERG6aEn0RERHghVreuFlMzg5DRERE5KYp0RcRkWKvip8LnctpMlYREREpGpToi4hIsRSZZM+8PaSmN2aTWvNFRESkaFCiLyIixdLE3QmZt1uVdnViJCIiIiK5S4m+iIgUO/vj0vnhaErmfZNa80VERKQIUaIvIiLFzrjt8TicHYSIiIhIHlGiLyIixcqGc2ksPpWKJtgXERGRokqJvoiIFBuGYTB2ezwA3cI1y76IiIgUTUr0RUSk2FgSmcr6czbczDCwurezwxERERHJE0r0RUSkWLAbBuO2Z8y0/2gVL0p7WJwckYiIiEjeUKIvIiLFwvyjKeyLS8fPauLJql7ODkdEREQkzyjRFxGRIi/FbjB+R8bY/IHVvfFz1c+fiIiIFF060xERkSLvm4NJnEp2EOJhpk8lT2eHIyIiIpKnlOiLiEiRFmdzMGl3xtj852t646519URERKSIU6IvIiJF2tQ9icSkGVT2daFbuIezwxERERHJc0r0RUSkyIpKtvPZ/kQAhtbyxsWs1nwREREp+pToi4hIkfXBrgRS7NAgwEqbUDdnhyMiIiKSL5Toi4hIkXQoPp1vDycDMLyODyaTWvNFRESkeFCiLyIiRdK7O+KxG9A6xI1Gga7ODkdEREQk3yjRFxGRImfLhTQWnUjFBAyt7e3scERERETylRJ9EREpUgzDYOy2jOX0uoW7U9XP6uSIRERERPKXEn0RESlSlkWlsfpsGq5mGFLLx9nhiIiIiOQ7JfoiIlJkOAyDsdvjAehTyZMynhYnRyQiIiKS/3KU6P/111+5HYeIiMhN+/F4Crti0vFxMTGwmsbmi4iISPGUo0S/ffv2VKxYkTfffJPjx4/ndkwiIiI3LM1h8O6OjLH5T1bzoqSbOq2JiIhI8ZSjs6CTJ0/yzDPP8P3331OhQgXatWvHd999R1paWm7HJyIiki0zDyZxPNFOkLuZRyp7OjscEREREafJUaIfGBjI888/z5YtW1i7di1VqlTh6aefJjQ0lEGDBrF169bcjlNEROSqEmwOJu5OBOC5Gt54uqg1X0RERIqvmz4Tql+/PiNGjOCZZ54hISGBzz//nAYNGtCyZUt27tyZGzGKiIhc0yf7Ejmf6iDC28L9ER7ODkdERETEqXKc6NtsNr7//ns6dOhAeHg4v/32G5MmTSIqKooDBw4QHh5Ojx49cjNWERGRy5xLsfPpviQAhtb2wWo2OTkiEREREedyycmTnn32WWbNmoVhGDz88MOMGzeOWrVqZT7u5eXFu+++S2hoaK4FKiIiciWTdieSmG5Qt6SVu8u4OTscEREREafLUaK/a9cuJk6cSLdu3XBzu/JJVWBgoJbhExGRPHUsIZ0ZBzNa84fV8cZkUmu+iIiISI4S/SVLlly/YBcXbrvttpwULyIiki3v7UzAZkCr0q40K6XWfBERERHI4Rj9MWPG8Pnnn1+2/fPPP2fs2LE3HZSIiMj17Ii28cOxFABequ3j5GhERERECo4cJfpTp06lWrVql22vWbMmU6ZMuemgRERu1OFziYz9dQ/PztrM2F/3cPhcorNDkjw2bns8AJ3LuVOrpNXJ0YiIiIgUHDnqun/69GlCQkIu2x4UFERkZORNByUiciO+23Cc4XO3YTKZMAwDk8nE1L8PMrZ7HXo0DHN2eJIHVp1JZVlUGlYTvFDT29nhiIiIiBQoOWrRDwsLY+XKlZdtX7ly5Q3NtG+32xk5ciQRERF4eHhQsWJFRo0ahWEYmftERUXRr18/QkND8fT0pH379uzfv/+a5X7yySe0bNmSkiVLUrJkSVq3bs26deuy7NOvXz9MJlOWv/bt22c7dhEpGA6fS2T43G04DLA7jCz/Dpu7jSNq2S9yDMNg7LYEAHpV9KScd46uWYuIiIgUWTk6O3r88cd57rnnsNls3HnnnUDGBH0vvfQSL7zwQrbLGTt2LJMnT+arr76iZs2abNiwgUceeQQ/Pz8GDRqEYRh06dIFq9XKDz/8gK+vL+PHj6d169bs2rULLy+vK5a7dOlSHnzwQZo1a4a7uztjx46lbdu27Ny5kzJlymTu1759e7744ovM+1dbQUBECq7vNhzPmGn9kguEF5lMJr7dcJxh7S8faiSF1y8nU9kabcPLxcQz1a/8OyAiIiJSnOUo0R86dCjnz5/n6aefJi0tDQB3d3eGDRvGiBEjsl3OqlWr6Ny5Mx07dgSgfPnyzJo1K7P1ff/+/axZs4YdO3ZQs2ZNACZPnkxwcDCzZs3iscceu2K5M2bMyHL/008/Ze7cuSxZsoQ+ffpkbndzcyM4ODj7L1xECpwT0ck4HJcn+QAOw+BEdHI+RyR5yeYweOefsfmPV/Ei0N3i5IhERERECp4cJfomk4mxY8cycuRIdu/ejYeHB5UrV77hFvFmzZoxbdo09u3bR5UqVdi6dSsrVqxg/PjxAKSmpgIZFxEuMpvNuLm5sWLFiqsm+v+VlJSEzWbD398/y/alS5dSqlQpSpYsyZ133smbb75JQEDAFctITU3NjAcgLi4OAJvNhs1my/6LlqLPZuPitGC29HRQ/cgzqekODp6J58ppfkYj/44TMRyIiiXc3zNfYytObOnpGf8aBjgceXqsWYeSOZxgx9/NRN9K7thu5ngOx7+fVYcjz2PPdQ5HRiXX90y+uvibr99+KQ5U351A55FOU1jqe3bju6mBjd7e3jRq1CjHzx8+fDhxcXFUq1YNi8WC3W5n9OjR9OrVC4Bq1apRrlw5RowYwdSpU/Hy8mLChAmcOHHihib9GzZsGKGhobRu3TpzW/v27enWrRsREREcPHiQl19+mbvvvpvVq1djsVzeQjRmzBhef/31y7b//vvveHoqgZB/WVJSuOef23+uWoX9kgtVknsupMKX+ywcTTD9s8UATJfskZH+Hz6fRLv3l3NbsEHbsg48NJw7zyyOjobo6DwrP9UO7+6wACbuCEln2dnTN1XepZ/V3yIjC+9n9a+/nB1BsbR48WJnhyCSb1Tf80+W36Zlywrvb1MhVtDre1JSUrb2MxnGFQa2ZsOGDRv47rvvOHbsWGb3/YvmzZuXrTJmz57N0KFDeeedd6hZsyZbtmzhueeeY/z48fTt2xeAjRs30r9/f7Zu3YrFYqF169aYzWYMw+CXX3657jHefvttxo0bx9KlS6lTp85V9zt06BAVK1bkjz/+4K677rrs8Su16IeFhXHu3Dl8fX2z9XqlmEhMxFqyJABJx45h1fCQXLds/zlemLOdmGQbfh4udK0XyvTVx/6ddZ+MNH/QnRXZcDSGFQfOA+DvZWXwnZW4v0EZXCw5motUrsAWH8/i5ctpU7IkVg+PPDvO5D2JjN+ZSFlPM7+1C8DVbLr+k64lORnrbbcBYPv7b8jD2PNESgokJEDz5oUv9kLMZrOxePFi2rRpg9WqZR2laFN9d4JLziNtp05BYKCTAyo+Ckt9j4uLIzAwkNjY2GvmoTlq25o9ezZ9+vShXbt2/P7777Rt25Z9+/YRFRVF165ds13O0KFDGT58OD179gSgdu3aHD16lDFjxmQm+g0aNGDLli3ExsaSlpZGUFAQTZo0oWHDhtct/9133+Xtt9/mjz/+uGaSD1ChQgUCAwM5cODAFRN9Nze3Kw5NsFqtBboiiBNcUh+sLi6qH7nI7jD4YMl+Jv65H8OAOmX9+Oih+oT5e9KvWQW+3XCcE9HJlC3pwQMNwygf6IVhGCzde5Y3f97FwbOJ/O/H3cxYd5xXOtagVZUgZ7+kosEl46fEajJhNefNBZToVAef7M24gj20tg9eLrkwNv+SWK1mc5b7hYLZDCZTxvuv75l8p99/KU5U3/PRf84j9f2e/wp6fc9ubDlK9N966y0mTJjAwIED8fHx4YMPPiAiIoInnniCkJCQbJeTlJSE+T8nVhaLBccVxkn6+fkBGRP0bdiwgVGjRl2z7HHjxjF69Gh+++23bF0UOHHiBOfPn7+h+EUk/1xITGPw7M0s338OgF5NyvFqpxq4/ZPwlQ/0uuLs+iaTiTuqlaJF5UBmrj3GhD/2sS8qgT6fr+OOqkH8X8caVCqlddgLuo/2JBCfblCjhAudwtSNUURERORactR8cfDgwcyZ8l1dXUlMTMRkMvH8888zbdq0bJfTqVMnRo8ezc8//8yRI0eYP38+48ePz9IrYM6cOSxdupRDhw7xww8/0KZNG7p06ULbtm0z9+nTp0+W2f4vThT4+eefU758eU6fPs3p06dJSMhYdzkhIYGhQ4eyZs0ajhw5wpIlS+jcuTOVKlWiXbt2OXlLRCQPbToWTccPl7N8/zncrWYmPFCX0V1rZyb52WG1mOnbrDx/v3gHjzaPwMVs4q+9Z2n3/jL+98MOohPTrl+IOMWJRDvTD2S05g+r7YPZdJNd9kVERESKuBwl+iVLliQ+PmN5ozJlyrBjxw4AYmJisj05AMDEiRO57777ePrpp6levTovvvgiTzzxRJbW+sjISB5++GGqVavGoEGDePjhh5k1a1aWco4dO5Zlcr7JkyeTlpbGfffdR0hISObfu+++C2T0Gti2bRv33nsvVapUoX///jRo0IDly5ff8MoBIpJ3DMPgy5WHeWDqaiJjU6gQ6MUPA1vQtV7ZHJfp52nl1U41+P35VrSuXhq7w+Cr1Ue57Z2/+GzFYdLSC9nM68XAhJ3xpDmgaZArrUq7OjscERERkQIvR133W7VqxeLFi6lduzY9evRg8ODB/PnnnyxevPiK49uvxsfHh/fff5/333//qvsMGjSIQYMGXbOcpUuXZrl/5MiRa+7v4eHBb7/9ls0oRcQZElPTGT5vOz9uPQVAh9rBjO1eBx/33BkzVSHIm0/7NmTlgXOM+mkXe07HM+qnXXyz5igvd6hO6+qlMKnl2On2xNqYdzQFgOF1fPR/IiIiIpINOUr0J02aREpKxonX//3f/2G1Wlm1ahXdu3fnlVdeydUARaT42R8Vz5PfbOTg2URczCZe7lCdR5qXz5Mkr3mlQH4e1JI5G47z7u97OXwukcenb6BZxQBG3lOD6iFaVcOZ3tmegAF0KOtGXf+COzGOiIiISEFyw4l+eno6P/30U+ZYdrPZzPDhw3M9MBEpnn7YcpIR87aTlGantK8bH/eqT4Nw/zw9psVsomfjcnSsE8LHSw/y2YrDrDp4no4fLueBRmEMaVOVIB8N68lv68+lsSQyFYsJXqzl4+xwRERERAqNGx6j7+LiwpNPPpnZoi8ikhtS0+28+sMOBs/eQlKaneaVAvh5UMs8T/Iv5eNuZVj7aiwZchsd64TgMGDWuuPc8e5SPl56gBSbPd9iKe4Mw+DtbRlzwTwQ4UEFnxx1QBMREREplnI0GV/jxo3ZsmVLLociIsXVyZhk7p+6humrjwLw7J2VmP5oEwK9ndOKHubvyUcP1ef7J5tSt6wfCanpjPt1L63H/81P205hGIZT4ipOFp9KZeN5Gx4WE8/V0PKHIiIiIjciR00kTz/9NEOGDOH48eM0aNAALy+vLI/XqVMnV4ITkaLv731neW72ZqKTbPh5WJnwQF3urFba2WEB0LC8P/Ofbs4PW08y9pe9nIhO5pmZm/ky/Agj76lB3bASzg6xSEp3GLyzI2M51Ecre1LKI/vLKIqIiIhIDhP9nj17AmSZDd9kMmEYBiaTCbtd3VtF5NrsDoMPl+znwz/3YxhQu4wfH/eqT5i/p7NDy8JsNtG1Xlna1Qxm2rJDTP37EBuORtP5o5V0rVeGl9pXJcTPw9lhFinzjiazPy6dEq4mnqjmdf0niIiIiEgWOUr0Dx8+nNtxiEgxciExjcGzN7N8/zkAejUpx8h7auBuLbgtt56uLjzXugo9G5Vj3G97mLfpJPM3n+SXHZE80aoiT9xWAU9XjSO/WSl2gwk7M1rzn6nuja81RyPMRERERIq1HJ2VhoeH53YcIlJMbD4WzcAZmzgVm4K71cxbXWvTrX5ZZ4eVbcF+7oy//xb6NSvPqJ92sf5INB8s2c/s9cd4qV01utYrg9mstd5zavqBJCKTHYR6mOldsWD17hAREREpLHKU6E+fPv2aj/fp0ydHwYhI0WUYBtNXH+XNn3dhsxtEBHoxuXd9qgUXznXq65QtwXdPNOWXHad5a9FuTkQn88KcrXy1OmP8fqPy+bdaQFERm+bgo90ZrflDavngbtEFExEREZGcyFGiP3jw4Cz3bTYbSUlJuLq64unpqURfRLJITE1n+Lzt/Lj1FAAdagcztnsdfNytTo7s5phMJjrUDuHOaqX4YuURPvrrANtOxNJjymo61A5mxN3VC9ycAwXZlL2JxNoMqvq60DXc3dnhiIiIiBRaORr8GB0dneUvISGBvXv30qJFC2bNmpXbMYpIIXbgTDydP1rJj1tP4WI2MfKeGnz0UP1Cn+Rfyt1q4anbK/LXi7fzYONymE2waPtp7nrvb97+ZQ/xKTZnh1jgnU628/m+RABequ2NxaTWfBEREZGcyrVZjipXrszbb799WWu/iBRfC7ee4t5JKzlwJoHSvm7MHnAr/VtEYCqiSVyQjxtjutXm50EtaVEpkDS7gyl/H+SOd5cyc+0x7A7D2SEWWB/sTCDVAY0CrdwZ4ubscEREREQKtVydztjFxYVTp07lZpEiUgilpTv43w87GDRrM0lpdppVDODnQS1pWEzGrVcP8eXr/o35rG9DKgR6cS4hjZfnb6fjh8tZ8c9KA/KvA3HpfHs4GYDhtX2K7IUgERERkfySozH6CxcuzHLfMAwiIyOZNGkSzZs3z5XARKRwOhmTzMAZm9hyPAaAZ+6oxPNtqmApZjPRm0wm7qpempaVg/hmzVE+WLKfPafj6f3ZWu6qVoqXO1anYpC3s8MsEN7dEY8DaBPqRoNAV2eHIyIiIlLo5SjR79KlS5b7JpOJoKAg7rzzTt57773ciEtECqFl+84yePZmopNs+Lq7MOGBW7iremlnh+VUri5mHm0RQbf6ZXj/j/18s+YoS/ac4e99Z3m4aTiD76pMCc/im9xuOp/GrydTMQMv1fZxdjgiIiIiRUKOEn2Hw5HbcYhIIeZwGHz4534+WLIfw4BaZXyZ3KuBZpy/RAlPV167tya9bw3nrUW7+XPPGb5YeYR5m07yXOvK9L41HKslV0dTFXiGYTB2ezwA95X3oLJvjn6SREREROQ/itdZpYjkuguJafT7cj3v/5GR5D/UpBzfP9lMSf5VVCrlzef9GvF1/8ZULe1DbLKN13/cRbv3l7FkdxSGUXwm7Ft6Oo21Z224muG5mhrGICIiIpJbcpTod+/enbFjx162fdy4cfTo0eOmgxKRwmHL8Rju+XA5y/adxd1q5r0edXmra23crRZnh1bgtawcxM+DWjC6ay0CvFw5dDaR/l9t4OHP1rHndJyzw8tzjkta8/tV8iTUU3VGREREJLfkKNFftmwZHTp0uGz73XffzbJly246KBEp2AzDYPrqI/SYsopTsSlEBHqxYGBzujco6+zQChUXi5leTcL5a+jtPHFbBVwtZlYcOEeHD5bz8vztnEtIdXaIeeaHYynsiU3Hx2ri6epqzRcRERHJTTlK9BMSEnB1vXzyKKvVSlxc0W+JEinOElPTGTx7C6/+sBOb3eDuWsEsfKY51YJ9nR1aoeXrbmXE3dX5Y8htdKgdjMOAmWuPccc7S5ny90FS0+3ODjFXpdoN3t2R0Zr/VDUvSrhqFJmIiIhIbsrR2VXt2rX59ttvL9s+e/ZsatSocdNBiUjBdOBMPJ0/WsnCraewmE280rE6H/eqj4+71dmhFQnlAjz5uFcDvh1wK7XL+BGfms7bv+yh9fi/WbQ9ssiM359xKImTSQ5Ku5t5pJKXs8MRERERKXJyNMXxyJEj6datGwcPHuTOO+8EYMmSJcyaNYs5c+bkaoAiUjD8uPUUw+ZuIynNTmlfNyY9VJ9G5f2dHVaR1KRCAD8MbM68zSd557c9HL+QzNMzNtG4vD8j76lB7bJ+zg4xx+JtDibtSgAyJuDzcDE5OSIRERGRoidHiX6nTp1YsGABb731Ft9//z0eHh7UqVOHP/74g9tuuy23YxQRJ0pLd/DWot18ueoIAM0qBvBBz3oE+bg5N7Aizmw2cV+DsnSoHcyUvw8xbdlB1h25QKdJK+hevywvta9KaV93Z4d5wz7Zm8iFNIMKPhZ6lPdwdjgiIiIiRVKOFy3u2LEjHTt2zM1YRKSAORWT0ZK85XgMAAPvqMiQNlWxmNUKm188XV0Y0qYKPRuF8c5ve5m/+SRzN51g0fZInrytIgNaVcDDtXDMWH8mxc6n+5IAeKmWDy6qRyIiIiJ5Ikdj9NevX8/atWsv27527Vo2bNhw00GJiPMt23eWjh8uZ8vxGHzdXfisb0OGtqumJN9JQkt4MOGBW1gwsDkNwkuSbLMz4Y993PneUuZvPoHDUfDH70/clUiS3eAWfyvtyqhHiIiIiEheyVGiP3DgQI4fP37Z9pMnTzJw4MCbDkpEnMfhMPjgj/30/WId0Uk2apXx5edBLbmremlnhybALWEl+P7Jpkx8sB5lSngQGZvC899upevHK9l49IKzw7uqIwnpzDqU0Zo/vI4PJpMuGImIiIjklRwl+rt27aJ+/fqXba9Xrx67du266aBExDkuJKbR78v1TPhjH4YBDzYux/dPNiPM39PZocklTCYTneqGsuSF2xjariperha2noil++TVDJy5ieMXkpwd4mXe25FAugG3B7tya9Dly7OKiIiISO7JUaLv5uZGVFTUZdsjIyNxccnxsH8RcaItx2PoNHEFy/adxd1q5t0edRnTrTbu1sIx/rs4crdaGHhHJf4aejs9G4VhMsHP2yK5a/zfjPt1Dwmp6c4OEYDt0TZ+PJ6CCXipto+zwxEREREp8nKU6Ldt25YRI0YQGxubuS0mJoaXX36ZNm3a5FpwIpL3DMPg69VH6DFlFSdjkokI9GL+0825r0FZZ4cm2VTKx523u9fhp2db0LRCAGnpDj5eepDb31nK7HXHsDt5/P647fEAdCnnTo0SVqfGIiIiIlIc5Kj5/d1336VVq1aEh4dTr149ALZs2ULp0qX5+uuvczVAEck7ianpvDx/Oz9sOQVA+5rBjOtRB193JWOFUc1QP2Y+3oTFu6J4a9FujpxPYvi87Xy1+igjO1anWaXAfI9pRVQqy6PSsJpgSC3vfD++iIiISHGUo0S/TJkybNu2jRkzZrB161Y8PDx45JFHePDBB7FalSCIFAYHziTw1Dcb2X8mAYvZxIi7q9G/RYQmSSvkTCYTbWsGc3vVUkxffYQPl+xnd2QcD326ltbVS/N/HasTEeiVL7E4DIOx/7Tm967kSZiXhnaJiIiI5Iccn3V5eXnRokULypUrR1paGgC//PILAPfee2/uRCcieeKnbacY9v02EtPslPJx46Ne9WlU3t/ZYUkucnUx81jLCnSrX5YP/tjHN2uP8cfuKP7ed4Y+Tcsz6M7K+Hnm7YXZn0+ksD06HW8XE89UV2u+iIiISH7JUaJ/6NAhunbtyvbt2zGZTBiGkaUV0G6351qAIpJ70tIdvLVoN1+uOgJA0woBfPhgPYJ8tKZ5UeXv5crrnWvR+9ZwRi/azdK9Z/lsxWHmbTrBc62r8FCTclgtOZqu5ZpsDoP3diQAMKCqFwFuuX8MEREREbmyHJ15DR48mIiICM6cOYOnpyc7duzg77//pmHDhixdujSXQxSR3HAqJpkHpq3OTPKfvr0iX/dvrCS/mKhc2ocvH2nMV482pnIpb6KTbPxv4U7av7+Mv/aeyfXjzT6czJEEO4FuZvpX0fKMIiIiIvkpR4n+6tWreeONNwgMDMRsNmOxWGjRogVjxoxh0KBBuR2jiNyk5fvPcs/EFWw+FoOvuwuf9mnIS+2r4ZIHLblSsN1WJYhfBrdkVJda+Hu5cvBsIo98sZ4+n69jX1R8rhwjMd3BBzszWvMH1/DGy0X1TERERCQ/5ejsy2634+OTsRZyYGAgp05lzNgdHh7O3r17cy86EbkpDofBh0v20+fzdVxITKNWGV9+erYlrWuUdnZo4kQuFjMP3xrOXy/ezoBWFbBaTCzbd5b27y/j/+Zv53xC6k2V//m+JM6lOgj3stCzgkcuRS0iIiIi2ZWjMfq1atVi69atRERE0KRJE8aNG4erqyvTpk2jQoUKuR2jiORAdGIaz327hb/3nQXgwcZh/K9TTdytFidHJgWFn4eVlztU56HG5Rjzy25+2xnFjLXHWLjlFM/eVYm+zcrj5nJj9eV8qoOpexMBeKGWN1azVnEQERERyW85atF/5ZVXcDgcALzxxhscPnyYli1bsmjRIj788MNsl2O32xk5ciQRERF4eHhQsWJFRo0ahWEYmftERUXRr18/QkND8fT0pH379uzfv/+6Zc+ZM4dq1arh7u5O7dq1WbRoUZbHDcPg1VdfJSQkBA8PD1q3bp2tckUKg63HY7hn4gr+3ncWNxcz7/aoy5hudZTkyxWVD/Ri6sMNmfX4rdQM9SU+NZ23Fu2h7YRl/LojMst38vV8tDuBhHSDWiVcuCfMPQ+jFhEREZGryVGi365dO7p16wZApUqV2LNnD+fOnePMmTPceeed2S5n7NixTJ48mUmTJrF7927Gjh3LuHHjmDhxIpCRjHfp0oVDhw7xww8/sHnzZsLDw2ndujWJiYlXLXfVqlU8+OCD9O/fn82bN9OlSxe6dOnCjh07MvcZN24cH374IVOmTGHt2rV4eXnRrl07UlJScvKWiBQIhmHw9Zqj9JiympMxyZQP8GTBwObc16Css0OTQqBpxQAWPtOCcffVIcjHjaPnk3jym030nLaGHSdjr/v8E0kOvjmYBMCwOj6YTWrNFxEREXGGHHXdvxJ//xtfg3vVqlV07tyZjh07AlC+fHlmzZrFunXrANi/fz9r1qxhx44d1KxZE4DJkycTHBzMrFmzeOyxx65Y7gcffED79u0ZOnQoAKNGjWLx4sVMmjSJKVOmYBgG77//Pq+88gqdO3cGYPr06ZQuXZoFCxbQs2fPy8pMTU0lNfXfcatxcXEA2Gw2bDbbDb92KcJsNi6uTm5LT4d8qh9JaemM/GE3C7dFAtCmeinGdquJj7tVdVRuSNe6wbStFsjU5Yf5fOVR1h6+QKdJK+hWL5QhrStT6j8rNdjS0wGYsC+FNAc0K2Xl1iArtn96fhVYDse/n1WHAwp6vP/lcIBhQD5+zwiZ36f6XpXiQPXdCZx0HimFp75nN75cS/RzolmzZkybNo19+/ZRpUoVtm7dyooVKxg/fjxAZmLt7v5v90+z2YybmxsrVqy4aqK/evVqhgwZkmVbu3btWLBgAQCHDx/m9OnTtG7dOvNxPz8/mjRpwurVq6+Y6I8ZM4bXX3/9su2///47np5aOkr+ZUlJ4Z5/bv+5ahV297zvvhyVDJ/vtXA62YQZg07hDu7wO8XyP0/l+bGl6KoGDK8DPx41s+m8mbmbTvHjlpO0LuPgjhADVwucSYa1Z8ycSDSzJ9YGmGhWOplFp5KdHf51XfpZ/S0yMl8+q3nir7+cHUGxtHjxYmeHIJJvVN/zT5bfpmXLCu9vUyFW0Ot7UlJStvZzaqI/fPhw4uLiqFatGhaLBbvdzujRo+nVqxcA1apVo1y5cowYMYKpU6fi5eXFhAkTOHHiBJGRkVct9/Tp05QunXVW8dKlS3P69OnMxy9uu9o+/zVixIgsFw/i4uIICwujbdu2+Pr63viLl6LrkmEldzZrhjU4OE8Pt2j7aT5YsJPENDulfNx4//46NCpfMk+PKcVLb2Dz8RhGL9rL1hOxLDpuYXOcO60qBzBn60lMgN0wgIyu+oGuJekQWghm20/+92JEu5AQ8CgEMV8qJQUSEqB588IXeyFms9lYvHgxbdq0wWq1Xv8JIoWY6rsTXHIe2a5VKwgMdGIwxUthqe8Xe5Zfj1MT/e+++44ZM2Ywc+ZMatasyZYtW3juuecIDQ2lb9++WK1W5s2bR//+/fH398disdC6dWvuvvvuG5ocKje4ubnh5uZ22Xar1VqgK4I4wSX1werikmf1Iy3dwVuLdvPlqiMA3FrBnw8frEcpH135ldzXuEIQCwYGsnDrKcb+sodTsSl8u+HkJXv8Ox7/5Y3x3FrKjfLeTv2JuT7zv9PUWM3mLPcLBbMZTCZwccnyvSP5Q7//Upyovuej/5xH6vs9/xX0+p7d2Jx6VjN06FCGDx9Oz549qV27Ng8//DDPP/88Y8aMydynQYMGbNmyhZiYGCIjI/n11185f/78NZfxCw4OJioqKsu2qKgogv9pWb3477X2ESnIImOT6TltdWaS/9TtFfmmfxMl+ZKnTCYTnW8pw58v3k7ja/QaMZng28MFv+u+iIiISFHl1EQ/KSkJ839aUCwWS+bSfZfy8/MjKCiI/fv3s2HDhsxJ9K6kadOmLFmyJMu2xYsX07RpUwAiIiIIDg7Osk9cXBxr167N3EekoFqx/xwdP1zBpmMx+Li78EmfhgxrXw0XSyFrjZRCy91qobSfB+arTKpvGHAi0Z6/QYmIiIhIJqf2q+zUqROjR4+mXLly1KxZk82bNzN+/HgeffTRzH3mzJlDUFAQ5cqVY/v27QwePJguXbrQtm3bzH369OlDmTJlMnsCDB48mNtuu4333nuPjh07Mnv2bDZs2MC0adOAjFap5557jjfffJPKlSsTERHByJEjCQ0NpUuXLvn6Hohkl8NhMOmvA0z4Yx+GATVDfZncqwHlAjQZpOS/siU9MJlMGVn9f5hMUNbL4oSoRERERAScnOhPnDiRkSNH8vTTT3PmzBlCQ0N54oknePXVVzP3iYyMZMiQIURFRRESEkKfPn0YOXJklnKOHTuWpWdAs2bNmDlzJq+88govv/wylStXZsGCBdSqVStzn5deeonExEQGDBhATEwMLVq04Ndff80yw79IQRGdmMbz321h6d6zAPRsFMZr99bE3apkSpzj/oZhTP374BUfMwx4IEKTw4mIiIg4i8nI71ntioi4uDj8/PyIjY3VrPuSVWIieHsDYDt1CmtIyE0Vt/V4DE/P2MTJmGTcXMy82aUWPRqG5UakIjdlzobjDJu7DRPgMAzMJhOGAWMb+dKjfCHoaZKcDC1bZtxevrzwzVyfkgLx8dCqFWiZ13xjs9lYtGgRHTp0KNCTNYnkBtV3J7jkPJIzZyAoyLnxFCOFpb5nNw8t4FMiixRfhmHwzdpjjPpxF2l2B+EBnkzu1YAaobqwJAVDj4ZhNCrvz6yVB1i/7xiNSrjxYBXfgj/bvoiIiEgRp7MxkQIoKS2dl+dtZ8GWUwC0rVGad++vi697wb26KMVT+UAvXrwjgkUcoYO/O1ZP/ayIiIiIOJvOyEQKmINnE3jqm43si0rAYjYxrH1VHm9ZIWPiMxERERERketQoi9SgPy8LZKXvt9KYpqdIB83Jj1YjyYVApwdloiIiIiIFCJK9EUKgLR0B2N+2c0XK48A0CTCn4kP1aOUj1aBEBERERGRG6NEX8TJImOTGThjE5uOxQDw5G0VebFtFVws5ms/UURERERE5AqU6Is40Yr95xg0ezMXEtPwcXdh/P230KZGaWeHJSIiIiIihZgSfREncDgMPvrrAOP/2IdhQI0QXyb3rk94gJezQxMRERERkUJOib5IPotJSuP5b7fw196zADzQMIzXO9fE3WpxcmQiIiIiIlIUKNEXyUfbTsTw1DebOBmTjJuLmVFdanF/wzBnhyUiIiIiIkWIEn2RfGAYBjPWHuONH3eRZncQHuDJx73qUzPUz9mhiYiIiIhIEaNEXySXHTmXSPl/bn+w7CgdWnnyybLDzN98EoC2NUrzTo+6+HlYnRajiIiIiIgUXUr0RXLRdxuO8/qsdez85/4X604yafN5ACxmEy+1q8qAVhUwmUzOC1JERERERIo0JfoiueTwuUSGz92Gm/HvNsclt8f3qEvnemXyPzARERERESlWzM4OQKSo+G7D8au21FtMsCcqPp8jEhERERGR4kiJvkguORGdjGEYV3zM+OdxERERERGRvKZEXySXlC3pcdXHTCbTNR8XERERERHJLUr0RXLJXdVLZRmTfynDMHigYVj+BiQiIiIiIsWSEn2RXJCW7mDsL3su224xgdkEY7vXoXyglxMiExERERGR4kaz7ovkgtE/72L9kWh83FyY3LMBTMjY3q9xGe67q7aSfBERERERyTdK9EVu0pwNx/lq9VEAJjxwCy3CvTMfG9wqHKuSfBERERERyUfqui9yE7adiOH/FuwA4LnWlWldo7STIxIREZH/b+/e42M88/+PvyfnISIhSCKRxLlCqGiVbvGt1GGt6rbdqqoqu8Xi2/rZ1WJLq2rTqlWtbVHb1YNWVrvla7dasohDKQ3i2DoVoRXqEDmKSeb6/ZGaNXUKTdzJ5PV8POZh7rmv3PfnHp/MzDsz9zUAUNUR9IEbdDK3UMPe36zzRU4l3lJXT97dxOqSAAAAAICgD9wIR7FTIz7You/PnlPD0Oqa3reNvLxsVpcFAAAAAAR94EYkLf1GGw+eVnU/b731WIKCAnytLgkAAAAAJBH0geu2aOtR/f2Lg5KkvzzURo3r1rC4IgAAAAD4L4I+cB12fndWY/+5Q5I08n8aq0fLMIsrAgAAAAB3BH2glE7nndfQ9zersMipLs3q6P/d09TqkgAAAADgEgR9oBSKip363wVb9F1WgaJrV9NrfW+VN5PvAQAAAKiACPpAKbyybI++2H9K1fy89daAdqpZjcn3AAAAAFRMBH3gGv617XvNWfOtJOmVB1urWRiT7wEAAACouAj6wFV8fSxbT3+8XZI0rHMj9YoPt7giAAAAALg6gj5wBVn5JZPvFTiKdVeTUI3p3szqkgAAAADgmgj6wGUUO42eTE5Xxul8RdWya2Y/Jt8DAAAAUDkQ9IHL+MvyPVqz9wcF+HppzqPtFFzNz+qSAAAAAKBUCPrATyzdcUxvph6QJL38QLxaRARZXBEAAAAAlJ6lQb+4uFgTJkxQbGys7Ha7GjVqpMmTJ8sY4xqTm5urkSNHKjIyUna7XS1atNDs2bOvut0uXbrIZrNdcunVq5drzOOPP37J+h49epTbsaJy2Hs8R3/8aJsk6Ym7YtWnTX2LKwIAAACA6+Nj5c5ffvllzZo1S++++67i4uKUlpamQYMGqWbNmnryySclSaNHj9bKlSs1f/58xcTEaPny5Ro+fLgiIiJ07733Xna7n3zyic6fP+9aPnXqlFq3bq3f/OY3buN69OihefPmuZb9/f3L4ShRWZwtcGjIe2nKP1+sjo1q65keza0uCQAAAACum6VBf/369erTp4/rnfaYmBgtWLBAmzZtchszcOBAdenSRZI0ZMgQzZkzR5s2bbpi0K9Vq5bbcnJysqpVq3ZJ0Pf391dYWFgZHhEqK6fTaFTyVh06la/6wSWT7/l4c2YLAAAAgMrH0qDfsWNHvfXWW9q7d6+aNm2qbdu2ad26dZo+fbrbmCVLlmjw4MGKiIhQamqq9u7dq1dffbXU+3n77bf18MMPq3r16m63p6amqm7dugoJCdHdd9+tF198UbVr177sNgoLC1VYWOhazs7OliQ5HA45HI7rOWxUQDNW7NeqPT/I38dLb/RrrSB/rxv/f3U45HvhalGRRH/AwzmKikr+NUZyOi2u5jo4nf/9XXU6K1ftUkm9xkg8ztxUF54beO5HVUC/W4DXkZapLP1e2vosDfpjx45Vdna2mjdvLm9vbxUXF2vKlCnq37+/a8zMmTM1ZMgQRUZGysfHR15eXpo7d646depUqn1s2rRJO3fu1Ntvv+12e48ePXT//fcrNjZWBw4c0Pjx49WzZ09t2LBB3t7el2wnKSlJkyZNuuT25cuXq1q1atd55KhIdpy26W97Sv7PfxPj0OH0dTqcfuPb8z53Tr/68frK9etVHBDws2sEKoOUM2ekM2esLqPULv5dXXbsWOX9XV21yuoKqqSUlBSrSwBuGvr95nF7blqzpvI+N1ViFb3f8/PzSzXOZi6e+e4mS05O1pgxY/TKK68oLi5O6enpGjVqlKZPn66BAwdKkqZNm6a5c+dq2rRpio6O1po1azRu3DgtWrRIiYmJ19zH0KFDtWHDBm3fvv2q47799ls1atRI//nPf9S1a9dL1l/uHf2oqCidPHlSQUHMyl5ZHfghTw/M+VJ5hcV67I4GmtCrDM7Lz8uTb0iIJCk/I0O+nB4CD+fIyVHK2rW6JyREvna71eWUXkGBfDt3liQ5Vq+WKlPtknTunJSbK915Z+WrvRJzOBxKSUnRPffcI19f32v/AFCJ0e8WuOh1pOP776XQUIsLqjoqS79nZ2crNDRUZ8+evWoOtfQd/TFjxmjs2LF6+OGHJUmtWrXS4cOHlZSUpIEDB6qgoEDjx4/XokWLXOfxx8fHKz09XdOmTbtm0M/Ly1NycrJeeOGFa9bSsGFDhYaGav/+/ZcN+v7+/pedrM/X17dCNwKuLOecQ8MXpCuvsFi3x9bShN5x8i2L8/Iv6gdfHx/6A57Pp+SpxNdmk69XJZrb4qJafb283JYrBS8vyWYruf95nLnpeP5HVUK/30Q/eR3J4/vNV9H7vbS1WfqqJj8/X14/eWHl7e0t54/nSV44//1qY67mo48+UmFhoR599NFrjj169KhOnTql8PDw6zgCVFZOp9Hohdv07Q95Cq8ZoDceaVs2IR8AAAAALGbpO/q9e/fWlClT1KBBA8XFxWnr1q2aPn26Bg8eLEkKCgpS586dNWbMGNntdkVHR2v16tV677333Cbse+yxx1S/fn0lJSW5bf/tt9/Wfffdd8kEe7m5uZo0aZIeeOABhYWF6cCBA3r66afVuHFjde/evfwPHJb766r9Stl9XH4+Xpr9aILq1OCrFQEAAAB4BkuD/syZMzVhwgQNHz5cJ06cUEREhIYOHaqJEye6xiQnJ2vcuHHq37+/Tp8+rejoaE2ZMkXDhg1zjcnIyLjkXf89e/Zo3bp1Wr58+SX79fb21vbt2/Xuu+8qKytLERER6tatmyZPnnzZj+fDs6z4+rhe/c9eSdKL97VU66hgawsCAAAAgDJkadCvUaOGZsyYoRkzZlxxTFhYmObNm3fV7aSmpl5yW7NmzXSleQbtdruWLVt2PaXCQ3z7Q65GJafLGGnAHdF6qF2U1SUBAAAAQJnipGRUGbmFRRr6/mblFBapXXSIJvyqhdUlAQAAAECZI+ijSjDG6I8Lt2nfiVzVC/LXm4+2lZ8P7Q8AAADA85B0UCW8mXpAn+/KlK+3TW/2T1DdGgFWlwQAAAAA5YKgD4+XuueEpi3fI0madG9LJUSHWFwRAAAAAJQfgj482uFTeXpywVYZI/W7PUqPtG9gdUkAAAAAUK4I+vBY+edLJt/LPlekNlHBev7eOKtLAgAAAIByR9CHRzLG6OmPt+ubzByFBvpr9qMJ8vfxtrosAAAAACh3BH14pLlrv9W/tx+Tj5dNsx5tq7CaTL4HAAAAoGog6MPjrNt3Ui999o0k6bneLXRbTC2LKwIAAACAm4egD49y5HS+Ri7YIqeRfpMQqUfviLa6JAAAAAC4qQj68BgF54s19P3Nysp3KD6ypibf11I2m83qsgAAAADgpiLowyMYYzTuk+3afSxbtav7afajCQrwZfI9AAAAAFUPQR8e4e9fHNLi9O/l7WXTG/3bKiLYbnVJAAAAAGAJgj4qvfUHTurPS7+WJD3b6xbd0bC2xRUBAAAAgHUI+qjUvssq0MgPt6rYafTrW+vr8Y4xVpcEAAAAAJYi6KPSOuco1rD3N+t03nnFRQTpz79uxeR7AAAAAKo8gj4qJWOM/rRop3Z8d1Yh1Xw1+9EE2f2YfA8AAAAACPqolN7/8rD+ueWovGzSXx9pq6ha1awuCQAAAAAqBII+Kp1NB0/rhX/tliSN63mL7mwcanFFAAAAAFBxEPRRqRw7W6DhH2xWkdOod+sI/e6uWKtLAgAAAIAKhaCPSqOwqFjD5m/Rydzzah5WQy8/wOR7AAAAAPBTBH1UCsYYTVy8S9uOZKmm3VdvDWinan4+VpcFAAAAABUOQR+VwoebMvSPtCPyskkz+92qBrWZfA8AAAAALoegjwpv8+HTen7JLknSmO7N1alpHYsrAgAAAICKi6CPCu149jkNm79FjmKjX7YK07DODa0uCQAAAAAqNII+KqzzRU79fv5m/ZBTqKb1AvXKg62ZfA8AAAAAroGgjwpr0r92aUtGlmoE+GjOgHaq7s/kewAAAABwLQR9VEjJmzL0wcYM2WzSaw+3UWxodatLAgAAAIBKgaCPCmdrxhlN/L+SyfdGJzbV3c3rWVwRAAAAAFQeBH1UKD/kFOr387fofLFT3VrU04j/aWx1SQAAAABQqRD0UWE4ip0a8cEWZWafU6M61fWXh1rLy4vJ9wAAAADgehD0UWFM+fRrbTp0WoH+PnrrsXaqEeBrdUkAAAAAUOkQ9FEhfLz5qN5Zf0iS9GrfNmpUJ9DaggAAAACgkiLow3I7jp7V+EU7JElPdW2ie1ow+R4AAAAA3CiCPix1KrdQQ99P0/kip7o2r6unujaxuiQAAAAAqNQI+rBMUbFTIz7cou/PnlPD0Op69eE2TL4HAAAAAD8TQR+WSfrsG3357WlV9/PWnAEJCmLyPQAAAAD42SwN+sXFxZowYYJiY2Nlt9vVqFEjTZ48WcYY15jc3FyNHDlSkZGRstvtatGihWbPnn3V7b7zzjuy2Wxul4CAALcxxhhNnDhR4eHhstvtSkxM1L59+8rlOHGpxVu/09vrDkqS/vJQazWpV8PiigAAAADAM/hYufOXX35Zs2bN0rvvvqu4uDilpaVp0KBBqlmzpp588klJ0ujRo7Vy5UrNnz9fMTExWr58uYYPH66IiAjde++9V9x2UFCQ9uzZ41q22dw/Ej516lS9/vrrevfddxUbG6sJEyaoe/fu2r179yV/FEDZ2vndWY39ZLskacT/NFKPluEWVwQAAAAAnsPSd/TXr1+vPn36qFevXoqJidGDDz6obt26adOmTW5jBg4cqC5duigmJkZDhgxR69at3cZcjs1mU1hYmOtSr95/Z3I3xmjGjBl69tln1adPH8XHx+u9997T999/r8WLF5fX4ULS6bzzGvr+Zp1zONW5aR2NvqeZ1SUBAAAAgEex9B39jh076q233tLevXvVtGlTbdu2TevWrdP06dPdxixZskSDBw9WRESEUlNTtXfvXr366qtX3XZubq6io6PldDrVtm1b/fnPf1ZcXJwk6eDBg8rMzFRiYqJrfM2aNdW+fXtt2LBBDz/88CXbKywsVGFhoWs5OztbkuRwOORwOH7W/VBVFBU7NfLDLfouq0BRIXb95cGWchYXyVlsdWVlzOHQhdkGHEVFEv0BD+coKir51xjJ6bS4muvgdP73d9XprFy1SyX1GiPxOHNTXXjO57kfVQH9bgFeR1qmsvR7aeuzNOiPHTtW2dnZat68uby9vVVcXKwpU6aof//+rjEzZ87UkCFDFBkZKR8fH3l5eWnu3Lnq1KnTFbfbrFkz/f3vf1d8fLzOnj2radOmqWPHjtq1a5ciIyOVmZkpSW7v8l9YvrDup5KSkjRp0qRLbl++fLmqVat2I4df5Sw57KX133vJz8vokagcfbEqxeqSyoX3uXP61Y/XV65fr2JOBUEVkXLmjHTmjNVllNrFv6vLjh2rvL+rq1ZZXUGVlJLimc9hwOXQ7zeP23PTmjWV97mpEqvo/Z6fn1+qcZYG/YULF+qDDz7Qhx9+qLi4OKWnp2vUqFGKiIjQwIEDJZUE/S+//FJLlixRdHS01qxZoxEjRigiIsLtHfmLdejQQR06dHAtd+zYUbfccovmzJmjyZMn31Ct48aN0+jRo13L2dnZioqKUrdu3RQUFHRD26xKlu7I1IoNJeflv/Jga/2yVZjFFZWjvDzX1bs7dpRvmAcfKyDJkZOjlLVrdU9IiHztdqvLKb2CAtfV7uHhUmWqXZLOnZNyc6U776x8tVdiDodDKSkpuueee+Try7fFwLPR7xa46HVk906dpNBQC4upWipLv1/4ZPm1WBr0x4wZo7Fjx7o+Kt+qVSsdPnxYSUlJGjhwoAoKCjR+/HgtWrRIvXr1kiTFx8crPT1d06ZNu2LQ/ylfX1/deuut2r9/vyQp7Mfgdfz4cYWH/3ciuOPHj6tNmzaX3Ya/v7/8/f0vu+2K3AgVwTeZ2Rq7aJckaWjnhurTNsriisrZRf3g6+NDf8Dz+ZQ8lfjabPL1qkTf2npRrb5eXm7LlYKXl2Szldz/PM7cdDz/oyqh32+in7yO5PH95qvo/V7a2ix9VZOfny+vn7yw8vb2lvPH8yQvnP9+tTGlUVxcrB07drhCfWxsrMLCwrRixQrXmOzsbG3cuNHtkwD4+bLyz2vIe5tV4CjWXU1C9XT35laXBAAAAAAezdJ39Hv37q0pU6aoQYMGiouL09atWzV9+nQNHjxYUslX5HXu3FljxoyR3W5XdHS0Vq9erffee89twr7HHntM9evXV1JSkiTphRde0B133KHGjRsrKytLr7zyig4fPqzf/e53kkpm5B81apRefPFFNWnSxPX1ehEREbrvvvtu+v3gqYqdRk8lpyvjdL4iQ+x6/eFb5e1lu/YPAgAAAABumKVBf+bMmZowYYKGDx+uEydOKCIiQkOHDtXEiRNdY5KTkzVu3Dj1799fp0+fVnR0tKZMmaJhw4a5xmRkZLi963/mzBk98cQTyszMVEhIiBISErR+/Xq1aNHCNebpp59WXl6ehgwZoqysLP3iF7/Q559/rgAmvCgz01P2aPXeHxTg66U5AxIUUt3P6pIAAAAAwONZGvRr1KihGTNmaMaMGVccExYWpnnz5l11O6mpqW7Lr7766jW/fs9ms+mFF17QCy+8UNpycR0+23FMb6w6IEl6+YF4xUXUtLgiAAAAAKgaKtnMQ6gM9h7P0R8+2iZJ+t0vYtWnTX2LKwIAAACAqoOgjzJ1tsChoe9vVv75YnVoWFtjezL5HgAAAADcTAR9lBmn0+j//SNdB0/mKaJmgP76yK3y8abFAAAAAOBmIoWhzMxYsU8rvzkhPx8vzRnQTrUD/a0uCQAAAACqHII+ysTyXZl6fcU+SVLSr1upVSST7wEAAACAFQj6+Nn2n8jV6IUlk+893jFGDyREWlwRAAAAAFRdBH38LDnnHBr6fppyC4t0e0wt/anXLVaXBAAAAABVGkEfN8zpNPrDwm068EOewoIC9Eb/tvJl8j0AAAAAsBSpDDfsjVX7tXz3cfl5e2n2gATVqcHkewAAAABgNYI+bsjKb45r+n/2SpJevK+l2kQFW1sQAAAAAEASQR834ODJPD2VnC5jpEfvaKCHbouyuiQAAAAAwI8I+rguuYVFGvJemnLOFSkhOkQTfxVndUkAAAAAgIsQ9FFqxhiN+Wib9p3IVd0a/prVv638fGghAAAAAKhISGkotVmrD+iznZny9bZp1qMJqhsUYHVJAAAAAICfIOijVFL3nNAry/ZIkp6/N04J0SEWVwQAAAAAuByCPq4p41S+a/K9h2+L0iO3N7C6JAAAAADAFRD0cVX554s05P00nS1wqE1UsCb1iZPNZrO6LAAAAADAFRD0cUXGGD3zzx36JjNHoYH+mv1ogvx9vK0uCwAAAABwFQR9XNHf1h7Uv7Z9Lx8vm97s31ZhNZl8DwAAAAAqOoI+LmvdvpNK+uxrSdLE3i10e2wtiysCAAAAAJQGQR+XOHI6X/+7YIucRnowIVID7oi2uiQAAAAAQCkR9OGm4Hyxhr6/WWfyHYqPrKkX72vJ5HsAAAAAUIkQ9OFijNG4T7Zr97Fs1a7up9mPJijAl8n3AAAAAKAyIejDZd4Xh7Q4/Xt5e9n010faKiLYbnVJAAAAAIDrRNCHJGnDgVOasrRk8r3xv7xFHRrVtrgiAAAAAMCNIOhD32UVaOSHW1TsNLqvTYQG3xljdUkAAAAAgBtE0K/izjmKNez9zTqVd14twoOUdH88k+8BAAAAQCVG0K/CjDF6dvFO7fjurIKr+WrOgATZ/Zh8DwAAAAAqM4J+FTb/y8P6ePNRedmkv/Zrq6ha1awuCQAAAADwMxH0q6ivDp3WpH/tliSN7dlcv2gSanFFAAAAAICyQNCvgjLPntPv529RkdPoV/HheuKuhlaXBAAAAAAoIwT9KqawqFjD5m/WydxCNQ+roakPMvkeAAAAAHgSgn4V8/ySXUo/kqWa9pLJ96r5+VhdEgAAAACgDBH0q5APN2ZowaYjstmk1/vdquja1a0uCQAAAABQxng714MdPJmnhWlHdPRMgXy8bPrXtu8kSWO6N1PnpnUsrg4AAAAAUB4I+h5qYdoRjf3ndtlsNhlj5DQlt8dHBun3nRtZWxwAAAAAoNxY+tH94uJiTZgwQbGxsbLb7WrUqJEmT54sY4xrTG5urkaOHKnIyEjZ7Xa1aNFCs2fPvup2586dq7vuukshISEKCQlRYmKiNm3a5Dbm8ccfl81mc7v06NGjXI7zZjt4Mk9j/7ldTiMVO/8b8iVp53fZOnwq37riAAAAAADlytJ39F9++WXNmjVL7777ruLi4pSWlqZBgwapZs2aevLJJyVJo0eP1sqVKzV//nzFxMRo+fLlGj58uCIiInTvvfdedrupqanq16+fOnbsqICAAL388svq1q2bdu3apfr167vG9ejRQ/PmzXMt+/v7l+8B3yQL046UzKR/0R9MLrDZbPpH2hE906O5BZUBAAAAAMqbpUF//fr16tOnj3r16iVJiomJ0YIFC9zefV+/fr0GDhyoLl26SJKGDBmiOXPmaNOmTVcM+h988IHb8t/+9jf985//1IoVK/TYY4+5bvf391dYWFgZH5X1jp4pcPtUxMWMMTp6puAmVwQAAAAAuFksDfodO3bUW2+9pb1796pp06batm2b1q1bp+nTp7uNWbJkiQYPHqyIiAilpqZq7969evXVV0u9n/z8fDkcDtWqVcvt9tTUVNWtW1chISG6++679eKLL6p27dqX3UZhYaEKCwtdy9nZ2ZIkh8Mhh8NxPYdd7iKC/GWTTdJl3tH/cX1Fq9mjOBzyvXC1qEjivoaHcxQVlfxrjOR0WlzNdXA6//u76nRWrtqlknqNkXicuakuPH/yPIqqgH63AK8jLVNZ+r209Vka9MeOHavs7Gw1b95c3t7eKi4u1pQpU9S/f3/XmJkzZ2rIkCGKjIyUj4+PvLy8NHfuXHXq1KnU+3nmmWcUERGhxMRE1209evTQ/fffr9jYWB04cEDjx49Xz549tWHDBnl7e1+yjaSkJE2aNOmS25cvX65q1apd55GXr9ACyWkuHIPtojUl5+vXydmnpUv3WVFaleB97px+9eP1levXqzggwNJ6gJsl5cwZ6cwZq8sotYt/V5cdO1Z5f1dXrbK6giopJSXF6hKAm4Z+v3ncnpvWrKm8z02VWEXv9/z80s23ZmnQX7hwoT744AN9+OGHiouLU3p6ukaNGqWIiAgNHDhQUknQ//LLL7VkyRJFR0drzZo1GjFixCXB/UpeeuklJScnKzU1VQEX/aI8/PDDruutWrVSfHy8GjVqpNTUVHXt2vWS7YwbN06jR492LWdnZysqKkrdunVTUFDQz7kbykWN2O80fvEu16z7Je/v2/Tn++L0QNv61/x5/Ax5ea6rd3fsKF8PPD0EuJgjJ0cpa9fqnpAQ+drtVpdTegX/PY2pe3i4VJlql6Rz56TcXOnOOytf7ZWYw+FQSkqK7rnnHvn6+l77B4BKjH63wEWvI7t36iSFhlpYTNVSWfr9wifLr8XSoD9mzBiNHTvWFbpbtWqlw4cPKykpSQMHDlRBQYHGjx+vRYsWuc7jj4+PV3p6uqZNm3bNoD9t2jS99NJL+s9//qP4+Pirjm3YsKFCQ0O1f//+ywZ9f3//y07W5+vrWyEb4eH2MbqjUR39I+2Ijp4pUGSIXX3bRSkmtLrVpXm+i/rB18enQvYHUKZ8Sp5KfG02+XpZ+mUu1+eiWn29vNyWKwUvL8lmK7n/eZy56Srq8z9QHuj3m+gnryN5fL/5Knq/l7Y2S4N+fn6+vH7ywsrb21vOH8+TvHD++9XGXMnUqVM1ZcoULVu2TO3atbtmLUePHtWpU6cUHh5+nUdRccWEVmd2fQAAAACoYiwN+r1799aUKVPUoEEDxcXFaevWrZo+fboGDx4sSQoKClLnzp01ZswY2e12RUdHa/Xq1XrvvffcJux77LHHVL9+fSUlJUkq+dq+iRMn6sMPP1RMTIwyMzMlSYGBgQoMDFRubq4mTZqkBx54QGFhYTpw4ICefvppNW7cWN27d7/5dwQAAAAAAGXE0qA/c+ZMTZgwQcOHD9eJEycUERGhoUOHauLEia4xycnJGjdunPr376/Tp08rOjpaU6ZM0bBhw1xjMjIy3N71nzVrls6fP68HH3zQbX/PPfecnn/+eXl7e2v79u169913lZWVpYiICHXr1k2TJ0++7MfzAQAAAACoLCwN+jVq1NCMGTM0Y8aMK44JCwvTvHnzrrqd1NRUt+VDhw5ddbzdbteyZctKWSUAAAAAAJVHJZt5CAAAAAAAXA1BHwAAAAAAD0LQBwAAAADAgxD0AQAAAADwIAR9AAAAAAA8CEEfAAAAAAAPQtAHAAAAAMCDEPQBAAAAAPAgPlYXUFkZYyRJ2dnZFleCCicvz3XVkZMj3+rVLSwGKH+O7Gzl5+cr22aTb0GB1eWU3sW1njol2e3W1XIjCgul8+el7GypqMjqaqoMh8NR0u/Z2fL19bW6HKBc0e8WuOh1pHJyJH9/62qpYipLv1/Inxfy6JXYzLVG4LKOHj2qqKgoq8sAAAAAAFQxR44cUWRk5BXXE/RvkNPp1Pfff68aNWrIZrNZXQ4qmOzsbEVFRenIkSMKCgqyuhygXNHvqErod1Ql9DuqksrS78YY5eTkKCIiQl5eVz4Tn4/u3yAvL6+r/gUFkKSgoKAK/UABlCX6HVUJ/Y6qhH5HVVIZ+r1mzZrXHMNkfAAAAAAAeBCCPgAAAAAAHoSgD5QDf39/Pffcc/JnplRUAfQ7qhL6HVUJ/Y6qxNP6ncn4AAAAAADwILyjDwAAAACAByHoAwAAAADgQQj6AAAAAAB4EII+AAAAAAAehKAP/GjNmjXq3bu3IiIiZLPZtHjxYrf1xhhNnDhR4eHhstvtSkxM1L59+9zGnD59Wv3791dQUJCCg4P129/+Vrm5uW5jtm/frrvuuksBAQGKiorS1KlTL6nlo48+UvPmzRUQEKBWrVpp6dKlZX68qLqSkpJ02223qUaNGqpbt67uu+8+7dmzx23MuXPnNGLECNWuXVuBgYF64IEHdPz4cbcxGRkZ6tWrl6pVq6a6detqzJgxKioqchuTmpqqtm3byt/fX40bN9Y777xzST1vvPGGYmJiFBAQoPbt22vTpk1lfsyoumbNmqX4+HgFBQUpKChIHTp00GeffeZaT6/Dk7300kuy2WwaNWqU6zZ6Hp7i+eefl81mc7s0b97ctb7K97oBYIwxZunSpeZPf/qT+eSTT4wks2jRIrf1L730kqlZs6ZZvHix2bZtm7n33ntNbGysKSgocI3p0aOHad26tfnyyy/N2rVrTePGjU2/fv1c68+ePWvq1atn+vfvb3bu3GkWLFhg7Ha7mTNnjmvMF198Yby9vc3UqVPN7t27zbPPPmt8fX3Njh07yv0+QNXQvXt3M2/ePLNz506Tnp5ufvnLX5oGDRqY3Nxc15hhw4aZqKgos2LFCpOWlmbuuOMO07FjR9f6oqIi07JlS5OYmGi2bt1qli5dakJDQ824ceNcY7799ltTrVo1M3r0aLN7924zc+ZM4+3tbT7//HPXmOTkZOPn52f+/ve/m127dpknnnjCBAcHm+PHj9+cOwMeb8mSJebTTz81e/fuNXv27DHjx483vr6+ZufOncYYeh2ea9OmTSYmJsbEx8ebp556ynU7PQ9P8dxzz5m4uDhz7Ngx1+WHH35wra/qvU7QBy7jp0Hf6XSasLAw88orr7huy8rKMv7+/mbBggXGGGN2795tJJmvvvrKNeazzz4zNpvNfPfdd8YYY958800TEhJiCgsLXWOeeeYZ06xZM9fyQw89ZHr16uVWT/v27c3QoUPL9BiBC06cOGEkmdWrVxtjSnrb19fXfPTRR64xX3/9tZFkNmzYYIwp+cOYl5eXyczMdI2ZNWuWCQoKcvX3008/beLi4tz21bdvX9O9e3fX8u23325GjBjhWi4uLjYREREmKSmp7A8U+FFISIj529/+Rq/DY+Xk5JgmTZqYlJQU07lzZ1fQp+fhSZ577jnTunXry66j143ho/tAKRw8eFCZmZlKTEx03VazZk21b99eGzZskCRt2LBBwcHBateunWtMYmKivLy8tHHjRteYTp06yc/PzzWme/fu2rNnj86cOeMac/F+Loy5sB+grJ09e1aSVKtWLUnS5s2b5XA43PqwefPmatCggVu/t2rVSvXq1XON6d69u7Kzs7Vr1y7XmKv18vnz57V582a3MV5eXkpMTKTfUS6Ki4uVnJysvLw8dejQgV6HxxoxYoR69ep1SV/S8/A0+/btU0REhBo2bKj+/fsrIyNDEr0ucY4+UCqZmZmS5PZAcGH5wrrMzEzVrVvXbb2Pj49q1arlNuZy27h4H1cac2E9UJacTqdGjRqlO++8Uy1btpRU0oN+fn4KDg52G/vTfr/RXs7OzlZBQYFOnjyp4uJi+h3lbseOHQoMDJS/v7+GDRumRYsWqUWLFvQ6PFJycrK2bNmipKSkS9bR8/Ak7du31zvvvKPPP/9cs2bN0sGDB3XXXXcpJyeHXpfkY+neAQCWGjFihHbu3Kl169ZZXQpQbpo1a6b09HSdPXtWH3/8sQYOHKjVq1dbXRZQ5o4cOaKnnnpKKSkpCggIsLocoFz17NnTdT0+Pl7t27dXdHS0Fi5cKLvdbmFlFQPv6AOlEBYWJkmXzNR5/Phx17qwsDCdOHHCbX1RUZFOnz7tNuZy27h4H1cac2E9UFZGjhypf//731q1apUiIyNdt4eFhen8+fPKyspyG//Tfr/RXg4KCpLdbldoaKi8vb3pd5Q7Pz8/NW7cWAkJCUpKSlLr1q312muv0evwOJs3b9aJEyfUtm1b+fj4yMfHR6tXr9brr78uHx8f1atXj56HxwoODlbTpk21f/9+Ht9F0AdKJTY2VmFhYVqxYoXrtuzsbG3cuFEdOnSQJHXo0EFZWVnavHmza8zKlSvldDrVvn1715g1a9bI4XC4xqSkpKhZs2YKCQlxjbl4PxfGXNgP8HMZYzRy5EgtWrRIK1euVGxsrNv6hIQE+fr6uvXhnj17lJGR4dbvO3bscPvjVkpKioKCgtSiRQvXmKv1sp+fnxISEtzGOJ1OrVixgn5HuXI6nSosLKTX4XG6du2qHTt2KD093XVp166d+vfv77pOz8NT5ebm6sCBAwoPD+fxXeLr9YALcnJyzNatW83WrVuNJDN9+nSzdetWc/jwYWNMydfrBQcHm//7v/8z27dvN3369Lns1+vdeuutZuPGjWbdunWmSZMmbl+vl5WVZerVq2cGDBhgdu7caZKTk021atUu+Xo9Hx8fM23aNPP111+b5557jq/XQ5n6/e9/b2rWrGlSU1PdvpImPz/fNWbYsGGmQYMGZuXKlSYtLc106NDBdOjQwbX+wlfSdOvWzaSnp5vPP//c1KlT57JfSTNmzBjz9ddfmzfeeOOyX0nj7+9v3nnnHbN7924zZMgQExwc7DYDLvBzjB071qxevdocPHjQbN++3YwdO9bYbDazfPlyYwy9Ds938az7xtDz8Bx/+MMfTGpqqjl48KD54osvTGJiogkNDTUnTpwwxtDrBH3gR6tWrTKSLrkMHDjQGFPyFXsTJkww9erVM/7+/qZr165mz549bts4deqU6devnwkMDDRBQUFm0KBBJicnx23Mtm3bzC9+8Qvj7+9v6tevb1566aVLalm4cKFp2rSp8fPzM3FxcebTTz8tt+NG1XO5Ppdk5s2b5xpTUFBghg8fbkJCQky1atXMr3/9a3Ps2DG37Rw6dMj07NnT2O12Exoaav7whz8Yh8PhNmbVqlWmTZs2xs/PzzRs2NBtHxfMnDnTNGjQwPj5+Znbb7/dfPnll+Vx2KiiBg8ebKKjo42fn5+pU6eO6dq1qyvkG0Ovw/P9NOjT8/AUffv2NeHh4cbPz8/Ur1/f9O3b1+zfv9+1vqr3us0YY6z5LAEAAAAAAChrnKMPAAAAAIAHIegDAAAAAOBBCPoAAAAAAHgQgj4AAAAAAB6EoA8AAAAAgAch6AMAAAAA4EEI+gAAAAAAeBCCPgAAAAAAHoSgDwAAAACAByHoAwDg4X744Qf5+fkpLy9PDodD1atXV0ZGxlV/5vnnn1ebNm3KrIYuXbpo1KhRZbY9AABwZQR9AAA83IYNG9S6dWtVr15dW7ZsUa1atdSgQQOrywIAAOWEoA8AgIdbv3697rzzTknSunXrXNevx+OPP6777rtP06ZNU3h4uGrXrq0RI0bI4XC4xrz55ptq0qSJAgICVK9ePT344IOun129erVee+012Ww22Ww2HTp0SMXFxfrtb3+r2NhY2e12NWvWTK+99tp177ewsFDPPPOMoqKi5O/vr8aNG+vtt992rd+5c6d69uypwMBA1atXTwMGDNDJkydd6z/++GO1atVKdrtdtWvXVmJiovLy8q77PgIAoKLwsboAAABQ9jIyMhQfHy9Jys/Pl7e3t9555x0VFBTIZrMpODhYjzzyiN58881Sb3PVqlUKDw/XqlWrtH//fvXt21dt2rTRE088obS0ND355JN6//331bFjR50+fVpr166VJL322mvau3evWrZsqRdeeEGSVKdOHTmdTkVGRuqjjz5S7dq1tX79eg0ZMkTh4eF66KGHSrVfSXrssce0YcMGvf7662rdurUOHjzoCvJZWVm6++679bvf/U6vvvqqCgoK9Mwzz+ihhx7SypUrdezYMfXr109Tp07Vr3/9a+Xk5Gjt2rUyxpTJ/wMAAFawGZ7JAADwOEVFRTp69Kiys7PVrl07paWlqXr16mrTpo0+/fRTNWjQQIGBgQoNDb3szz///PNavHix0tPTJZW8s56amqoDBw7I29tbkvTQQw/Jy8tLycnJ+uSTTzRo0CAdPXpUNWrUuGR7Xbp0UZs2bTRjxoyr1j1y5EhlZmbq448/LtV+9+7dq2bNmiklJUWJiYmXbO/FF1/U2rVrtWzZMtdtR48eVVRUlPbs2aPc3FwlJCTo0KFDio6Ovub9CgBAZcBH9wEA8EA+Pj6KiYnRN998o9tuu03x8fHKzMxUvXr11KlTJ8XExFwx5F9JXFycK2xLUnh4uE6cOCFJuueeexQdHa2GDRtqwIAB+uCDD5Sfn3/Nbb7xxhtKSEhQnTp1FBgYqLfeeuuSiQKvtt/09HR5e3urc+fOl93+tm3btGrVKgUGBrouzZs3lyQdOHBArVu3VteuXdWqVSv95je/0dy5c3XmzJnrul8AAKhoCPoAAHiguLg4BQYGasCAAdq0aZMCAwPVtWtXHTp0SIGBgYqLi7vubfr6+rot22w2OZ1OSVKNGjW0ZcsWLViwQOHh4Zo4caJat26trKysK24vOTlZf/zjH/Xb3/5Wy5cvV3p6ugYNGqTz58+Xer92u/2qNefm5qp3795KT093u+zbt0+dOnWSt7e3UlJS9Nlnn6lFixaaOXOmmjVrpoMHD5b2bgEAoMIh6AMA4IGWLl2q9PR0hYWFaf78+UpPT1fLli01Y8YMpaena+nSpWW+Tx8fHyUmJmrq1Knavn27Dh06pJUrV0qS/Pz8VFxc7Db+iy++UMeOHTV8+HDdeuutaty4sQ4cOHBd+2zVqpWcTqdWr1592fVt27bVrl27FBMTo8aNG7tdqlevLqnkDwd33nmnJk2apK1bt8rPz0+LFi26gXsAAICKgaAPAIAHio6OVmBgoI4fP64+ffooKipKu3bt0gMPPKDGjRuX+fno//73v/X6668rPT1dhw8f1nvvvSen06lmzZpJkmJiYrRx40YdOnRIJ0+elNPpVJMmTZSWlqZly5Zp7969mjBhgr766qvr2m9MTIwGDhyowYMHa/HixTp48KBSU1O1cOFCSdKIESN0+vRp9evXT1999ZUOHDigZcuWadCgQSouLtbGjRv15z//WWlpacrIyNAnn3yiH374QbfcckuZ3j8AANxMBH0AADxUamqqbrvtNgUEBGjTpk2KjIxUeHh4uewrODhYn3zyie6++27dcsstmj17thYsWOA6ReCPf/yjvL291aJFC9WpU0cZGRkaOnSo7r//fvXt21ft27fXqVOnNHz48Ove96xZs/Tggw9q+PDhat68uZ544gnX1+NFREToiy++UHFxsbp166ZWrVpp1KhRCg4OlpeXl4KCgrRmzRr98pe/VNOmTfXss8/qL3/5i3r27Fmm9w8AADcTs+4DAAAAAOBBeEcfAAAAAAAPQtAHAAAAAMCDEPQBAAAAAPAgBH0AAAAAADwIQR8AAAAAAA9C0AcAAAAAwIMQ9AEAAAAA8CAEfQAAAAAAPAhBHwAAAAAAD0LQBwAAAADAgxD0AQAAAADwIP8fsfV7OXXVnnIAAAAASUVORK5CYII=",
      "text/plain": [
       "<Figure size 1200x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from capymoa.stream.generator import SEA\n",
    "from capymoa.stream.drift import DriftStream, AbruptDrift, GradualDrift\n",
    "from capymoa.classifier import AdaptiveRandomForestClassifier\n",
    "from capymoa.evaluation import prequential_evaluation\n",
    "from capymoa.evaluation.visualization import plot_windowed_results\n",
    "\n",
    "SEA3drifts = DriftStream(stream=[SEA(1), \n",
    "                                  AbruptDrift(10000),\n",
    "                                  SEA(2), \n",
    "                                  GradualDrift(start=20000, end=25000), \n",
    "                                  SEA(3), \n",
    "                                  AbruptDrift(45000),\n",
    "                                  SEA(1)])\n",
    "\n",
    "arf = AdaptiveRandomForestClassifier(schema=SEA3drifts.get_schema(), \n",
    "                                     ensemble_size=100, \n",
    "                                     number_of_jobs=4)\n",
    "\n",
    "results = prequential_evaluation(stream=SEA3drifts, \n",
    "                                 learner=arf, \n",
    "                                 window_size=5000, \n",
    "                                 max_instances=50000)\n",
    "\n",
    "print(f\"Cumulative accuracy = {results.cumulative.accuracy()}\")\n",
    "print(f\"wallclock = {results.wallclock()} seconds\")\n",
    "display(results.windowed.metrics_per_window())\n",
    "plot_windowed_results(results, metric='accuracy')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "c5431a62-64c5-4634-a86e-29a86f2397a7",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.19"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}