Notebook.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "# Twitter classificator\n",
    "\n",
    "The aim of this project is to test several Machine Learning classification methods in the context of textual data from Twitter. Being given a tweet, we want to be able to assign it to one of a few pre-defined categories. We will achieve this by using supervised Machine Learning techniques, so we will use a training set of a couple of hundreds of manually labeled tweets. \n",
    "\n",
    "## Acquiring data\n",
    "\n",
    "First step is to acquire labeled tweets for training the model. At first, I tried searching on the Internet (I gathered three sample data sets, 'dataset1', '2' and '3'), but the data sets did not satisfy me, so I decided to gather and manually label Tweets myself.<br>\n",
    "<br>\n",
    "I defined six categories for tweets:<br>\n",
    "Music, Sport, Technology, Weather, Movies, Politics<br>\n",
    "I also created a seventh, \"Other\" category for tweets which don't fit in any of the former six.<br>\n",
    "<br>\n",
    "I used Twitter streaming API for getting all tweets in English from around the world. Then I filtered them using a set of keywords for each category (I tried to include as many keywords as possible, ~200 per category). In the end, I manually labeled those filtered tweets. This way I gathered dataset with circa 1500 Tweets (~200 per each category and ~300 labeled \"Other\")<br>\n",
    "I decided not to put the original script for acquiring data in the repository, as I don't think it is useful after getting the data set. However, there is a script for expanding the training set, of which I will tell more later.\n",
    "<br>\n",
    "The acquired data set can be found in data/gathered_dataset/external/training_set.txt file.<br>\n",
    "\n",
    "## Word Embeddings\n",
    "\n",
    "Now having a set of labeled sentences (I use the term 'sentence' interchangeably with 'tweet', although it is not strictly correct) we would want to train one of popular classification models, e. g. Neural Network or an SVM classifier. Alas, all these methods require our data to be numerical, not textual. To employ them in our problem we need to firstly get a Word Embedding to represent words from our dataset as vectors of numbers.<br>\n",
    "<br>\n",
    "We can find a few pre-trained Word Embeddings on the Internet. I have found a good list in this repository:<br>\n",
    "https://github.com/3Top/word2vec-api#where-to-get-a-pretrained-models<br>\n",
    "<br>\n",
    "By default I decided to use two embeddings for comparison:<br>\n",
    "- Google News embedding (Trained by Google, using word2vec architecture)<br>\n",
    "- Twitter embedding (Trained by GloVe, using GloVe achitecture)<br>\n",
    "\n",
    "Google News embedding contains 3 million word vectors of 300 dimensions.<br>\n",
    "Twitter embedding was trained on actual tweets, so potentially it could be a really good fit. It has several versions in terms of dimensionality, I used the largest one with 200 dimensions. I do not have information about its vocabulary size, however.<br>\n",
    "<br>\n",
    "Having at least one word embedding is required for further work. They have sizes of a couple of GB each, so they can take some time to download.<br>\n",
    "You can also try other embeddings, but it is required that GloVe embeddings are in textual form (.txt) and word2vec in binary form (.bin).<br>\n",
    "<br>\n",
    "In code, the default locations for word embeddings are:<br>\n",
    "'models/word_embeddings/google/GoogleNews-vectors-negative300.bin'<br>\n",
    "'models/word_embeddings/glove_twitter/glove.twitter.27B.200d.txt'<br>\n",
    "<br>\n",
    "You can change them by editing src/common.py file (It contains some constants which are used in scripts for testing model parameters). To do this you need to edit the WORD_EMBEDDINGS list.<br>\n",
    "<br>\n",
    "Meaning of each entry in the list is:<br>\n",
    "(word embedding class (Word2VecEmbedding/GloveEmbedding), ['relative_path_to_file', number of dimensions])<br>\n",
    "For scripts to work there needs to be at least one item in this list.<br>\n",
    "<br>\n",
    "Caution: These embeddings are loaded directly to memory. This means that they can crash on 32-bit version of Python. It is also required to have at least 8 GB RAM.<br>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "# Word2Vec allows us to see the most similar words for a given word\n",
    "from src.features.word_embeddings.word2vec_embedding import Word2VecEmbedding, run_interactive_word_embedding_test\n",
    "\n",
    "# change this line for embedding location\n",
    "model = Word2VecEmbedding('google/GoogleNews-vectors-negative300.bin', 300)\n",
    "\n",
    "# loading model to memory can take up to a few minutes\n",
    "model.build()\n",
    "\n",
    "run_interactive_word_embedding_test(model)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "# Unfortunately, with a library for GloVe model that I used it is not so easy to find the nearest words.\n",
    "# In this script we can only see raw word vectors\n",
    "from src.features.word_embeddings.glove_embedding import GloveEmbedding, run_interactive_word_embedding_test\n",
    "\n",
    "model = GloveEmbedding('glove_twitter/glove.twitter.27B.200d.txt', 200)\n",
    "model.build()\n",
    "run_interactive_word_embedding_test(model)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "# This scripts tries to visualize embedded words\n",
    "# Unfortunately, we see only 3 dimensions, so it uses PCA to reduce dimensionality of word embedding\n",
    "# PCA is trained on all words which occur in our dataset\n",
    "# Reducing number of dimensions from 300 to 3 may make our word embedding worthless\n",
    "# Maybe even in this small dimensionality we can see some patterns\n",
    "# Color of word point is based on its occurences counts in tweets from each category\n",
    "# If you run this outside the notebook in standalone python you can click on dots to see words they represent\n",
    "\n",
    "from src.visualization.visualize_word_embedding import visualize_word_embedding\n",
    "\n",
    "# You can also use GloVe embedding here\n",
    "word_emb = Word2VecEmbedding('google/GoogleNews-vectors-negative300.bin', 300)\n",
    "word_emb.build()\n",
    "\n",
    "visualize_word_embedding(word_emb)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "## Preprocessing tweets\n",
    "\n",
    "Next step is to look at tweets and see if we can preprocess them so they work better for our model. We want to convert each to tweet a list of words. As an input we have tweets which are unicode strings. I simply convert them to list of words by splitting on whitespaces. Then, I do these conversions:\n",
    " - convert all text to lowercase (I assume case does not affect meaning, also SoMe PeOPle WrItE LiKe tHiS)\n",
    " - remove hyperlinks\n",
    " - remove @UserNames (marks answer to another user's tweet)\n",
    " - remove stopwords (words like 'a', 'the' etc) - I use NLTK stopwords corpus\n",
    " - convert numbers from 0-20 to textual representation - I use 'inflect' library\n",
    " - remove words that contain non-alphanumeric characters\n",
    "\n",
    "We will also ignore words which don't exist in our Word Embedding. This filtering will be done later, because I wanted the preprocessed dataset to be a single file, fit for all embeddings and models.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "from src.data.dataset import run_interactive_processed_data_generation\n",
    "\n",
    "# When prompt for folder name, insert 'gathered_dataset'\n",
    "run_interactive_processed_data_generation()\n",
    "\n",
    "# data will be saved in data/gathered_dataset/processed/training_set.txt (script shows a relative path)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "## Embedding tweets\n",
    "\n",
    "We have aquired a Word Embedding which allows us to get a vector of numbers for each word. However, our training set consists of lists of words, so we need to specify how to represent a sentence (tweet) based on representations of its words. I test two different approaches:<br>\n",
    "<br>\n",
    "First is to simply concatanate word vectors, so for each sentence we get vector of size (number of words in sentence) * (size of a single word vector). To use machine learning algorithms each tweet vector must have the same size, so we need to specify a constant length of a sentence. For instance, we can decide that sentences longer than 30 words will be cut at the end and sentences shorter than 30 words will be filled with zeros.<br>\n",
    "<br>\n",
    "Second approach is not to concatanate, but sum word vectors (element-wise). This means that the size of sentence vector will be the same as the size of word vector, so we don't have to artificially cut or lengthen tweets.<br>\n",
    "<br>\n",
    "Concatanation yields longer vectors, so models will obviously need more time to compute. On the other hand, we don't lose information about order of words. However, if we look at this example we can see that in training for finding categories order of words may not be so important:<br>\n",
    "<br>\n",
    "Tweet 1: \"President Obama was great!\"<br>\n",
    "Tweet 2: \"What a great president Obama was!\"<br>\n",
    "<br>\n",
    "Both are obviously about politics and we know this whatever order of words is.<br>\n",
    "Furthermore, if we have a small training set, we may encounter only one of these tweets. In sum embedding it is not a problem, because both will have very similar representation. In concatanation embedding we may end up with a model which labels correctly only one of them. So it is not obvious which embedding is going to work better.<br>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "# 3D visualization of sentence embeddings, similar to visualization of word embeddings.\n",
    "# It transforms sentence vectors to 3 dimensions using PCA trained on them.\n",
    "# Colors of sentences align with their category\n",
    "\n",
    "from src.visualization.visualize_sentence_embeddings import visualize_sentence_embeddings\n",
    "from src.features.sentence_embeddings.sentence_embeddings import ConcatenationEmbedding, SumEmbedding\n",
    "from src.features.word_embeddings.word2vec_embedding import Word2VecEmbedding\n",
    "\n",
    "# You can also use GloVe embedding here\n",
    "# To use GloVe embedding: from src.features.word_embeddings.glove_embedding import GloveEmbedding\n",
    "word_emb = Word2VecEmbedding('google/GoogleNews-vectors-negative300.bin', 300)\n",
    "word_emb.build()\n",
    "\n",
    "sentence_embeddings = [\n",
    "    ConcatenationEmbedding,\n",
    "    SumEmbedding\n",
    "]\n",
    "\n",
    "visualize_sentence_embeddings(word_emb, sentence_embeddings)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "## Finding the best model hyperparameters\n",
    "\n",
    "I chose four standard classifiers from sklearn package to test: \n",
    " \n",
    "- SVC (SVM Classifier) \n",
    "- RandomForestClassifier \n",
    "- MLPClassifier (Multi-Layer Perceptron classifier - neural network) \n",
    "- KNeighborsClassifier (classification by k-nearest neighbors). \n",
    " \n",
    "For all these methods, I defined a set of hyperparameters to test: \n",
    " \n",
    "CLASSIFIERS_PARAMS = [<br>\n",
    "&nbsp;(SVC,<br>\n",
    "&nbsp;&nbsp;{\"C\": list(log_range(-1, 8)),<br>\n",
    "&nbsp;&nbsp;\"gamma\": list(log_range(-7, 1))}),<br>\n",
    "<br>\n",
    "&nbsp;(RandomForestClassifier,<br>\n",
    "&nbsp;&nbsp;{\"criterion\": [\"gini\", \"entropy\"],<br>\n",
    "&nbsp;&nbsp;\"min_samples_split\": [2, 5, 10, 15],<br>\n",
    "&nbsp;&nbsp;\"min_samples_leaf\": [1, 5, 10, 15],<br>\n",
    "&nbsp;&nbsp;\"max_features\": [None, \"sqrt\"]}),<br>\n",
    "<br>\n",
    "&nbsp;(MLPClassifier,<br>\n",
    "&nbsp;&nbsp;{\"alpha\": list(log_range(-5, -2)),<br>\n",
    "&nbsp;&nbsp;\"learning_rate\": [\"constant\", \"adaptive\"],<br>\n",
    "&nbsp;&nbsp;\"activation\": [\"identity\", \"logistic\", \"tanh\", \"relu\"],<br>\n",
    "&nbsp;&nbsp;\"hidden_layer_sizes\": [(100,), (100, 50)]}),<br>\n",
    "<br>\n",
    "&nbsp;(KNeighborsClassifier,<br>\n",
    "&nbsp;&nbsp;{\"n_neighbors\": [1, 2, 3, 4, 7, 10, 12, 15, 30, 50, 75, 100, 150],<br>\n",
    "&nbsp;&nbsp;\"weights\": ['uniform', 'distance']})<br>\n",
    "]<br>\n",
    " \n",
    "I test all possible combinations of these parameters, for all word embeddings' and sentence embeddings' combinations. This yields a lot of test cases. Fortunately, classifiers from sklearn package have similar interfaces, so I wrote a generic method for grid-search. This method looks more or less like this: \n",
    "\n",
    "for word_embedding in word_embbedings: \n",
    "   - build word embedding \n",
    "   for sentence_embedding in sentence_embeddings: \n",
    "       - build features using sentence embedding and word embedding \n",
    "       for classifier_class, tested_parameters in CLASSIFIERS_PARAMS: \n",
    "           - run multithreaded GridSearchCV for classifier_class and tested_parameters \n",
    "             on traing set labels and built features \n",
    "           - safe GridSearchCV results in a summary file in /summaries folder \n",
    " \n",
    "With models from sklearn we can also use GridSearchCV method from this package, which allows to perform grid search for best parameters in parallel. For each combination of parameters it saves average score in folds. I use folds count equal to 5.<br>\n",
    "<br>\n",
    "Testing all models can take up to a couple of hours. Because of this, I save grid search result in summaries folder, which is included in repository. We can use these results to get the best possible model or do some interpretation of the results.<br>\n",
    "<br>\n",
    "By viewing grid search results files we can see that the best scores are somewhere around 85%.<br>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "# run grid search on chosen models\n",
    "# consider doing a backup of text summary files from /summary folder before running this script\n",
    "\n",
    "from src.common import DATA_FOLDER, FOLDS_COUNT, CLASSIFIERS_PARAMS, \\\n",
    "                       SENTENCE_EMBEDDINGS, CLASSIFIERS_WRAPPERS, WORD_EMBEDDINGS\n",
    "from src.models.model_testing.grid_search import grid_search\n",
    "\n",
    "classifiers_to_check = []\n",
    "for classifier_class, params in CLASSIFIERS_PARAMS:\n",
    "    to_run = raw_input(\"Do you wish to test {0} with parameters {1} ? [y/n] \"\n",
    "                       .format(classifier_class.__name__, str(params)))\n",
    "    if to_run.lower() == 'y' or to_run.lower() == 'yes':\n",
    "        classifiers_to_check.append((classifier_class, params))\n",
    "\n",
    "print(\"*\" * 20 + \"\\n\")\n",
    "grid_search(DATA_FOLDER, FOLDS_COUNT,\n",
    "            word_embeddings=WORD_EMBEDDINGS,\n",
    "            sentence_embeddings=SENTENCE_EMBEDDINGS,\n",
    "            classifiers=classifiers_to_check,\n",
    "            n_jobs=-1) # uses as many threads as CPU cores"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "# now that we have performed grid search, we can train model using the best parameters and test it interactively\n",
    "\n",
    "from src.common import choose_classifier, LABELS, SENTENCES\n",
    "from src.models.model_testing.grid_search import get_best_from_grid_search_results\n",
    "from src.features.build_features import FeatureBuilder\n",
    "from src.visualization.interactive import interactive_test\n",
    "\n",
    "classifier = choose_classifier() # choose classifier model\n",
    "best_parameters = get_best_from_grid_search_results(classifier) # get best parameters\n",
    "\n",
    "if best_parameters is None: exit(-1) # no grid search summary file for this model\n",
    "\n",
    "word_emb_class, word_emb_params, sen_emb_class, hyperparams = best_parameters\n",
    "\n",
    "print (\"\\nEvaluating model for word embedding: {:s}({:s}), sentence embedding: {:s} \\nHyperparameters {:s}\\n\"\n",
    "           .format(word_emb_class.__name__, ', '.join(map(str, word_emb_params)), sen_emb_class.__name__, str(hyperparams)))\n",
    "hyperparams[\"n_jobs\"] = -1 # uses as many threads as CPU cores\n",
    "\n",
    "print (\"Building word embedding...\")\n",
    "word_emb = word_emb_class(*word_emb_params)\n",
    "word_emb.build()\n",
    "\n",
    "print (\"Building sentence embedding...\")\n",
    "sen_emb = sen_emb_class()\n",
    "sen_emb.build(word_emb)\n",
    "\n",
    "print (\"Building features...\")\n",
    "\n",
    "# this helper class builds a matrix of features with provided sentence embedding\n",
    "fb = FeatureBuilder()\n",
    "fb.build(sen_emb, LABELS, SENTENCES)\n",
    "\n",
    "print (\"Building model...\")\n",
    "clf = classifier(sen_emb, probability=True, **hyperparams)\n",
    "clf.fit(fb.features, fb.labels)\n",
    "\n",
    "print (\"Model evaluated!...\\n\")\n",
    "interactive_test(clf)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "# This scripts visualizes how models work in 2D\n",
    "# It scatters training samples on a 2D space and colors backgrounds according to model prediction\n",
    "# It uses best possible parameters from grid search result file\n",
    "# I use PCA to reduce number of dimension to 2D\n",
    "# Models perform poorly in 2D, but we can see how boundaries between categories may look for different methods\n",
    "\n",
    "from src.visualization.visualize_2d import visualize_2d\n",
    "from src.features.word_embeddings.word2vec_embedding import Word2VecEmbedding\n",
    "from src.common import choose_multiple_classifiers\n",
    "\n",
    "classifier_classes = choose_multiple_classifiers()\n",
    "\n",
    "# use same word embedding for all classifiers - we don't want to load more than one embedding to RAM\n",
    "word_emb = Word2VecEmbedding('google/GoogleNews-vectors-negative300.bin', 300)\n",
    "word_emb.build()\n",
    "\n",
    "visualize_2d(word_emb, classifier_classes)\n",
    "\n",
    "# Attention: There is probably a bug in matplotlib, which causes the color of the fifth category\n",
    "# not to be drawn in the background (cyan, 'Movies' category)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "## Applying PCA\n",
    "\n",
    "Until now, I applied PCA several times to make plotting 2D or 3D figures possible. Maybe there is a chance that applying a well fit Principal Component Analysis to our data set will increase model accuracy. The possible option is to fit PCA to our sentece vectors. Then we can train model on sentence vectors transformed with PCA. With reduced number of dimensions, it is also possible that our model could work faster.<br>\n",
    "<br>\n",
    "The script below allows user to see how applying PCA to a chosen model works. It cross-validates model built on PCA with dimensions reduced to numbers distributed linearly between 1 and the initial number of dimensions.\n",
    "There is however a limit - we can't train PCA to number of dimensions which is higher than the number of training samples. So, if we have a dataset of 1500 tweets and we use 5 folds for cross-validation, that means that maximum dimensions after PCA are 1200 (4/5 * 1500). It can be an issue when we use Concatenation Embedding.<br>\n",
    "<br>\n",
    "The script measures: cross-validation scores, model training time (including fitting PCA) and average sentence prediction time for a trained model.<br>\n",
    "<br>\n",
    "My conclusion after seeing the results is that in our problem applying PCA probably wouldn't be advantegous, as there is a noticable decreasement in training speed and no apparent improvement in accuracy, as compared to model without PCA."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "# training for PCA visualization may take a while, so I also use a summary file for storing the results\n",
    "from src.visualization.visualize_pca import visualize_pca\n",
    "from src.common import choose_classifier\n",
    "\n",
    "classifier_class = choose_classifier()\n",
    "visualize_pca(classifier_class, n_jobs=-1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "## Expanding training set\n",
    "\n",
    "After having a way to choose the best parameters for a model I modified the script for mining tweets. In the beginning I used keywords to search for tweets possibly fit for the training set. This gave me quite clear tweets, but may seem my models not general, because they fit mostly to those keywords.<br>\n",
    "<br>\n",
    "The approach that I use to expand my current training set is to use a trained classification model to test tweets incoming from Twitter. I declare a threshold, for instance 0.7 and I take into consideration only these tweets, which are assigned to one category with probability more than the threshold.<br>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "from src.data.data_gathering.tweet_miner import mine_tweets\n",
    "\n",
    "# At first, we train the best possible model we have so far (searching in grid search results files)\n",
    "# Then, we run as many threads as CPU cores, or two if there is only one core.\n",
    "# First thread reads stream from Twitter API and puts all tweets in a synchronous queue.\n",
    "# Other threads read from the queue, test tweets on model\n",
    "# and if threshold requirement is met, put them in mined_tweets.txt file\n",
    "\n",
    "# by setting this flag to False we ignore tweets that are classified as \"Other\"\n",
    "include_unclassified = False\n",
    "threshold = 0.7\n",
    "\n",
    "mine_tweets(threshold, include_unclassified)\n",
    "# this method can be stopped only by killing the process"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "# when we have some tweets in mined_tweets.txt, we must label them manually.\n",
    "from src.data.data_gathering.tweet_selector import select_tweets\n",
    "\n",
    "select_tweets()\n",
    "\n",
    "# Results will be stored in file selected_tweets.txt.\n",
    "# We can manually merge this file with already existing training set"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Some analysis\n",
    "\n",
    "Having performed grid search on several models, we can analyse how various hyperparameters perform. Use script below to see some plots showing cross-validation performance for different parameters."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using TensorFlow backend.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Found Grid Search results in /summaries/gathered_dataset_SvmAlgorithm_grid_search_results.txt\n",
      "Found Grid Search results in /summaries/gathered_dataset_RandomForestAlgorithm_grid_search_results.txt\n",
      "Found Grid Search results in /summaries/gathered_dataset_MLPAlgorithm_grid_search_results.txt\n",
      "Found Grid Search results in /summaries/gathered_dataset_NearestNeighborsAlgorithm_grid_search_results.txt\n",
      "Found Grid Search results in /summaries/gathered_dataset_KerasNeuralNetworkAlgorithm_grid_search_results.txt\n",
      "Plot saved in /reports/figures/sentence_embeddings_comparison.svg\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAaQAAAEICAYAAAAQkoCgAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\noAFJKaVUofD/89+0rIxSQJYAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f0d9034d1d0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# compare how different sentence embedding perform for all tested models\n",
    "from src.common import SENTENCE_EMBEDDINGS\n",
    "from src.visualization.compare_sentence_embeddings import get_grid_search_results_by_sentence_embeddings\n",
    "from src.visualization.compare_sentence_embeddings import compare_sentence_embeddings_bar_chart\n",
    "\n",
    "sen_embeddings = [sen_emb.__name__ for sen_emb in SENTENCE_EMBEDDINGS]\n",
    "\n",
    "grid_search_results = get_grid_search_results_by_sentence_embeddings(sen_embeddings)\n",
    "compare_sentence_embeddings_bar_chart(grid_search_results)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Found Grid Search results in /summaries/gathered_dataset_SvmAlgorithm_grid_search_results.txt\n",
      "Found Grid Search results in /summaries/gathered_dataset_RandomForestAlgorithm_grid_search_results.txt\n",
      "Found Grid Search results in /summaries/gathered_dataset_MLPAlgorithm_grid_search_results.txt\n",
      "Found Grid Search results in /summaries/gathered_dataset_NearestNeighborsAlgorithm_grid_search_results.txt\n",
      "Found Grid Search results in /summaries/gathered_dataset_KerasNeuralNetworkAlgorithm_grid_search_results.txt\n",
      "\n",
      "SvmAlgorithm: Best result: 90.55%, Average result: 53.48%\n",
      "For embeddings: GloveEmbedding(glove_twitter/glove.twitter.27B.200d.txt, 200), SumEmbedding\n",
      "And for parameters: {'C': 1000, 'gamma': 1e-05}\n",
      "\n",
      "RandomForestAlgorithm: Best result: 71.78%, Average result: 59.39%\n",
      "For embeddings: Word2VecEmbedding(google/GoogleNews-vectors-negative300.bin, 300), SumEmbedding\n",
      "And for parameters: {'max_features': 'sqrt', 'min_samples_split': 2, 'criterion': 'gini', 'min_samples_leaf': 10}\n",
      "\n",
      "MLPAlgorithm: Best result: 91.01%, Average result: 87.21%\n",
      "For embeddings: GloveEmbedding(glove_twitter/glove.twitter.27B.200d.txt, 200), SumEmbedding\n",
      "And for parameters: {'alpha': 1e-05, 'activation': 'logistic', 'learning_rate': 'constant', 'hidden_layer_sizes': (100,)}\n",
      "\n",
      "NearestNeighborsAlgorithm: Best result: 80.97%, Average result: 55.3%\n",
      "For embeddings: GloveEmbedding(glove_twitter/glove.twitter.27B.200d.txt, 200), SumEmbedding\n",
      "And for parameters: {'n_neighbors': 12, 'weights': 'distance'}\n",
      "\n",
      "KerasNeuralNetworkAlgorithm: Best result: 45.01%, Average result: 23.65%\n",
      "For embeddings: Word2VecEmbedding(google/GoogleNews-vectors-negative300.bin, 300), SumEmbedding\n",
      "And for parameters: {'nb_epoch': 64, 'batch_size': 4}\n",
      "Plot saved in /reports/figures/models_comparison.svg\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAZQAAAEICAYAAAB4YQKYAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nrkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f0d90346790>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# compare overall performance of all tested models\n",
    "from src.visualization.compare_models import get_available_grid_search_results\n",
    "from src.visualization.compare_models import compare_models_bar_chart\n",
    "\n",
    "best_results_for_models =  get_available_grid_search_results()\n",
    "\n",
    "for classifier_class, parameters in best_results_for_models:\n",
    "    word_emb_class, word_emb_params, sen_emb_class, params, best_result, avg_result = parameters\n",
    "\n",
    "    print (\"\\n{0}: Best result: {1}%, Average result: {2}%\".\n",
    "           format(classifier_class.__name__, best_result, avg_result))\n",
    "    print (\"For embeddings: {0}({1}), {2}\".format(word_emb_class.__name__,\n",
    "                                                  ', '.join(map(str, word_emb_params)),\n",
    "                                                  sen_emb_class.__name__))\n",
    "    print (\"And for parameters: {0}\".format(str(params)))\n",
    "\n",
    "compare_models_bar_chart(best_results_for_models)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Choose a classifier by typing a number:\n",
      "0 - SvmAlgorithm\n",
      "1 - RandomForestAlgorithm\n",
      "2 - MLPAlgorithm\n",
      "3 - NearestNeighborsAlgorithm\n",
      "4 - KerasNeuralNetworkAlgorithm\n",
      "0\n",
      "Found Grid Search results in /summaries/gathered_dataset_SvmAlgorithm_grid_search_results.txt\n",
      "Choose one or two of the following parameters by typing a number or two numbers, e.g. '1' or '3,4': \n",
      "0 - Word Embedding\n",
      "1 - C\n",
      "2 - Sentence Embedding\n",
      "3 - gamma\n",
      "1,3\n",
      "Use logarithmic scale for C? [y/n] y\n",
      "Use logarithmic scale for gamma? [y/n] y\n",
      "Plot saved in /reports/figures/parameters_SvmAlgorithm_C_and_gamma.svg\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAZQAAADuCAYAAAAeC/GzAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAIABJREFUeJzsvXecHNWVt//cCp0na6QZSaORRgQJJBBBSKwDxmZZbAub\nxSCEbBDBpNe8a/iBFxYcwGnx4n2xCQsLAoEJL7YxLC/RGEwwBhGNwSQhzSjMaDTS5Olc4fz+qO5W\n92hykATu5/MZUHWlW1W37rfuueecq0SEIkWKFClSZLxoe7oARYoUKVLkk0FRUIoUKVKkyIRQFJQi\nRYoUKTIhFAWlSJEiRYpMCEVBKVKkSJEiE0JRUIoUKVKkyIRQFJQiRYoUKTIhFAWlSJEiRYpMCEVB\nKVKkSJEiE4KxpwvQj2LYfpEiRYqMHrWnCwDFHkqRIkWKFJkgPjaC8txzzzFz5sxJO34kEqGxsXHQ\n9bNnz+bpp5+etPNPJIlEguOPP56ysjJOPvnkPV2c3YKIcOaZZ1JRUcERRxwxKefIrwM//elP+eY3\nv5lb99BDD1FXV0ckEuEvf/kLH374IYsWLaKkpITrr79+Usqzp1BKsX79+j1djBz97/3exJ133smn\nP/3pPV2M3cawgqKU+rRS6iWlVI9SqlMp9Wel1OLJKtCdd96JUopf//rXk3WKAYlGozQ0NABwxhln\n8N3vfne3nn8ieeCBB2hra6Ojo4Pf/va3e7o4u4UXX3yRP/zhDzQ3N/Pqq69O+vmuuOIKVq9enVu+\n9NJLufHGG4lGoxxyyCH8x3/8B0cffTR9fX38y7/8y6SXJ5/J/vja2+h/7/sjIlx//fUsWLCAcDjM\nzJkzOfnkk3nnnXf2QGk/2QwpKEqpUuBR4AagEpgBXA2kJqtAd911F5WVlfzqV7+arFMUYNv2bjnP\n7sJxHDZt2sR+++2HYYx+iOzjej82bdrE7NmzCYfDo953Iq5506ZNHHjggYMu7+7y/D0x3L3+9re/\nzS9/+Uuuv/56Ojs7WbduHSeccAKPPfbYbizl3wkiMugfcDjQPcg6P9ANLMj7rRpIAFOBzwHNwL8C\n24FW4ATgS8A6oBO4Iv+YGzduFKWUPPDAA6LrurS2tkqWZ599VmbMmJFbfuONN2TRokUSiUTkpJNO\nkuXLl8uVV16ZW3/rrbfK3LlzpaKiQo4//nhpaWnJrQPkxhtvlH322Udmz56d++2jjz6S//7v/xbD\nMMQ0TQmHw7Js2TIREamvr5drr71WFi5cKKWlpbJ8+XJJJBIFZfvZz34m1dXVUlNTIw899JA89thj\nsu+++0pFRYX85Cc/kcFYtWqVnHfeeXLMMcdIJBKRz372s7Jx48bc+vfff1+OOeYYqaiokP32209+\n/etfF+x7/vnnyxe/+EUJhULyD//wD2KaphiGIeFwWFavXi2O48iPfvQjmTVrllRXV8tpp50m3d3d\nIiLS1NQkgKxevVrq6urkM5/5TO63O+64Q2bOnCnl5eVy8803y6uvvioLFy6UsrIy+da3vpUrw/r1\n6+Xoo4+WyspKqaqqkpUrV0pXV1du/VD3TkTkf/7nf+Tggw+WkpISaWhokCeeeEJERLq7u+Wss86S\nmpoamT59ulx55ZVi2/Yu92/16tXi9/tF0zQJh8Py/e9/f0x1oD+/+tWvZNasWVJZWSk//vGPpb6+\nXv7whz+IiMgPfvAD+frXvy7JZFLC4bAAEgqFpKGhQY4++mjRNE38fr+Ew2H58MMPJZlMyiWXXCJ1\ndXUydepUOe+88yQejxfUn2uuuUamTZsm3/jGN0RE5JFHHpGDDz5YysrK5Mgjj5S//vWvw97TaDQq\ngUBAlFISDoclHA4XXLeIyNq1a2XatGkF9/LBBx+UhQsXiojIK6+8IkuXLpWysjKpqamRb33rW5JK\npQru3UcffSQiIkcddZTcdtttuXVr1qyRT33qU7nloeruY489JvPnz5dIJCLTp0+Xa6+9dsDnMFj9\nHeje92fdunWiaZq88sorAx57IO644w6ZN2+eRCIRmTNnjtxyyy25ddln9fOf/zz3rt9xxx259e3t\n7XL88cdLSUmJLF68WL773e8W3I/+3HXXXbk69sMf/rCgjo3kOdx0002yzz77CNAH/AiYC7wE9AK/\nAXzepqNrj4EjgJczbXwrcGP2WEP9DScopUAHcBfwRaCi3/o7gJ/kLX8LeDLvAmzg+4AJnAPsAO4D\nSoADM+IzJ7v/D3/4Q1m8eLGIiCxYsEB+/vOf7/IgRURSqZTMmjVLfvGLX0g6nZbf/e53YppmTlCe\neeYZqaqqkjfeeEOSyaRceOGF8pnPfKbgQRxzzDHS0dGRe6nzX5JVq1YViJOI9wIvXrxYWlpapKOj\nQ+bNmyc333xzrmy6rsvVV18t6XRabr31VpkyZYqceuqp0tvbK3/7298kEAhIY2PjgJVq1apVEolE\n5Pnnn5dkMin/8i//kquE0WhUZs6cKXfccYdYliVvvvmmVFVVybvvvpvbt7S0VF588UVxHEcSiUSu\nscty++23y9y5c2XDhg3S19cn//zP/5xrtLLicdppp0k0GpV4PJ777bzzzpNEIiG///3vxe/3y1e/\n+lVpa2uT5uZmqa6ulueee05ERD766CN56qmnJJlMyvbt2+Uzn/mMfPvb3x7RvXvllVektLRUnnrq\nKXEcR5qbm+X9998XEZETTjhBzj33XIlGo9LW1iaLFy8ueLnz6d+QjaUO5PPuu+9KOBzOPZOLL75Y\ndF3fRVDyj5etPyK7NrQXXXSRHH/88dLR0SG9vb2ybNkyufzyywvqz7/+679KMpmUeDwub775plRX\nV8vatWvFtm258847pb6+XpLJ5LD3tP/H10A0NDTIU089lVs+6aST5N///d9FROT111+Xl19+WSzL\nkqamJpk3b55cd911A17rUIIyXN2tqamRF154QUREOjs75Y033hiwrEPV34HufT4333yzzJo1a8h7\n0Z9HH31U1q9fL67rynPPPSfBYDBXtuyz+t73vifpdFoee+wxCQaD0tnZKSIip5xyipx88skSjUbl\nnXfekenTpw8qKNk69qc//UlSqZRccsklYhhGro6N5Dl85StfkZ6eHsm0pyngGaABKAPeA1bJGNpj\n4DBgKZ4n8GzgfeAiGY+gZA48H7gzo2428P+AaZl1xwAb8rb9M3B63gUkAD2zXILnFrwkb/s3gBOy\ny/vss0/uhv30pz+Vgw46KHfz8l+S559/XqZPny6u6+bWf+pTn8qJwFlnnSXf+c53cuv6+vrEMAxp\namrKPYhnnnmm4OGORFDuvvvu3PJ3vvMdOe+883JlCwQCuS++3t5eAWTt2rW57Q899FB56KGHZCBW\nrVolp5xySkF5NU2TzZs3y/333y+f/vSnC7Y/99xz5aqrrsrte9pppxWs79/Yff7zn5ebbropt/zB\nBx+IYRi5igrIhg0bcuuzvzU3N+d+q6yslPvvvz+3fOKJJxZU7nweeughWbRoUW55qHt37rnnykUX\nXbTLMbZt2yY+n6+gsb/vvvvkc5/73IDn7C8oY6kD+Vx99dUFzyQajYppmmMSFNd1JRQKyfr163Pr\nX3rppVzP6NlnnxXTNAt6beeff75897vfLSjTfvvtlxPx4erjcIJy5ZVXyplnnikiXn0NhUIFveJ8\nrrvuOjnhhBMGvNahBGW4ultXVye33HKL9PT0DFnWoepv//L058c//rEsWbJkyOMPx1e/+lX5xS9+\nISI73/XsuUVEqqur5eWXXxbbtsUwjNwHkYjIv/3bvw0qKFdffbWsWLEitxyLxQrqWH8Geg4vvvhi\n3iJvAJfJzvb1P4FfyBjbYynUgYuAhwZal/837KC8iLwvImeIyExgATAd+EVm9bNASCm1RCk1G1gE\nPJS3e4eIOJl/JzL/b8tbnwAiAEqpTzU1NbFixQoAVq5cyTvvvMNbb721S5m2bt3KjBkzUGqn63Vd\nXV3B+vr6+txyJBKhqqqKlpaWAbcfKTU1Nbl/h0IhotFobrmqqgpd1wEIBoMATJs2Lbc+GAwWbN+f\n/PJEIhEqKyvZunUrmzZt4pVXXqG8vDz3d++997Jt27YRX0v/+1FfX49t27S17XwUAx2jf/kHu562\ntjZWrFjBjBkzKC0t5Rvf+Abt7e0Fxxrs3m3ZsoW5c+fucu5NmzZhWRa1tbW56z7vvPPYvn37kNc6\n2DWPtg5s3bq1YH04HKaqqmpE5+7Pjh07iMfjHHbYYblrOe6449ixY0dum+rqagKBQG5506ZN/Od/\n/mfBc9+yZQtbt27NbTNUfRyOlStX8uCDD5JKpXjwwQc59NBDc/dr3bp1LFu2jJqaGkpLS7niiit2\neZ4jYbi6+7vf/Y7HH3+c+vp6jjrqKF5++eUBjzOS+jsYVVVVtLa2jqrcTzzxBEuXLqWyspLy8nIe\nf/zxguuvqqoqGJ/M3vsdO3Zg23ZBvckv90DXlb9tKBQqqGMjeQ757yReezpg+5phNO3xfkqpR5VS\n25RSvcBPgSmDXkyGUbkNi8gHeL2VBZllB89Od2rm71ER6RvNMfNYJSIsWrSImpoalixZAniD9P2p\nra2lpaUlq5yA1zBlmT59Ops2bcotx2IxOjo6mDFjRu63fDHqz1DrJov88kejUTo7O5k+fTp1dXUc\nddRRdHd35/6i0Sg333zziMvb/35s3rwZwzAKKuN4rvmKK65AKcU777xDb28v99xzT8GzGYq6ujo2\nbNgw4O9+v5/29vbcdff29vLuu++O6LjjrQO1tbUFzyQej9PR0TGic/dnypQpBINB3n333dy19PT0\nFAhA/7LU1dVx5ZVXFjz3eDzOqaeeOuz5RvIsDzjgAOrr63niiSe47777WLlyZW7dBRdcwLx58/jo\no4/o7e3lpz/96aDPMxwOE4/Hc8v9P3SGqruLFy/m4YcfZvv27ZxwwgksX758wHOMpP4Oxhe+8AWa\nm5t5/fXXh90WIJVK8bWvfY1LL72UtrY2uru7+dKXvjSi+lxdXY1hGAX1ZvPmzYNuX1tbS3Nzc245\nkUgU1LHRPIdJ4GbgA2BfESkFrmAEwZPDeXnNU0pdopSamVmuwxOOtXmb3QecAnw98+9Ro5QKAMtv\nvfVW3nrrrdzfDTfcwH333beL18uRRx6JruvceOON2LbNww8/XOAqeuqpp7JmzRreeustUqkUV1xx\nBUuWLGH27NkjKs+0adOGjEmZDB5//HFefPFF0uk03/ve91i6dCl1dXUsW7aMdevWcffdd2NZFpZl\n8dprr/H++++P+Ninnnoq1113HU1NTUSjUa644gpOOeWUMXmBDURfXx+RSISysjJaWlq49tprR7zv\n2WefzZo1a3jmmWdwXZeWlhY++OADamtrOfbYY7nkkkvo7e3FdV02bNjA888/P6LjjrcOnHTSSTz6\n6KO5Z/L9738f13VHfF35aJrGOeecw8UXX5zrYbW0tPD73/9+0H3OOeccbrnlFl555RVEhFgsxmOP\nPUZf3/Dfa9OmTaOjo4Oenp4ht1u5ciW//OUveeGFFwrilfr6+igtLSUSifDBBx8UfLz0Z9GiRTz4\n4IPE43HWr1/P7bffnls3VN1Np9Pce++99PT0YJompaWlaNrAzdF46u++++7L//pf/4tTTz2V5557\njnQ6TTKZ5P777+eaa67ZZft0Ok0qlcqJwxNPPMFTTz017HkAdF3nxBNP5KqrriIej/Pee+8N+EGc\n5aSTTuKRRx7hpZdeIp1Oc9VVVxUIxmiewyRQgjewH1VKzQMuGMlOw/VQ+oAlwCtKqRiekPwNuCS7\ngYi8AsTwTGFPjL7cgOdtkDj99NOpqanJ/Z111lnYts2TTz5ZsLHP5+PBBx/k9ttvp7y8nHvuuYdl\ny5bh9/sBOOaYY/jRj37E1772NWpra9mwYQP333//iAtz9tln895771FeXs4JJ5wwxksaHStXruTq\nq6+msrKSN954g3vuuQeAkpISnnrqKe6//36mT59OTU0Nl112GanUyD23zzrrLE477TQ++9nPMmfO\nHAKBADfccMOElf0HP/gBb775JmVlZXz5y1/mxBNPHPG+RxxxBGvWrOHiiy+mrKyMo446Kvc1+qtf\n/Yp0Os0BBxxARUUFJ5100ojNF+OtAwceeCA33XQTK1eupLa2loqKinHFdvzsZz9jn332YenSpZSW\nlnLMMcfw4YcfDrr94Ycfzm233caFF15IRUUF++yzD3feeeeIzjVv3jxOPfVUGhoaKC8vLzCT5XPq\nqafy/PPP8/nPf54pU3ZaM37+859z3333UVJSwjnnnMMpp5wy6LkuvvhifD4f06ZNY9WqVXz961/P\nrRuu7t59993Mnj2b0tJSbrnlFu69994BzzHe+nv99ddz4YUX8q1vfYvy8nLmzp3LQw89xPHHH7/L\nttlA1OXLl1NRUcF9993HV77ylRGfKxsPU1NTwxlnnMGZZ5456LYHHnggN9xwAytWrKC2tpZIJMLU\nqVNz7dhonsMkcCmwEk8DbgNGFBiodmMXaiSMuTBLlizh/PPPH/IB7q2cccYZzJw5kx//+Md7uihF\nihTZQ0SjUcrLy/noo4+YM2fOaHcv5vIaD88//zzbtm3Dtm3uuusu3n77bY477rg9Xawik4yIYFkW\ntm2P2QRVpMjewiOPPEI8HicWi3HppZeycOHCEZtl90b2tmzDI+bDDz9k+fLlxGIxGhoaeOCBB6it\nrd3TxSoySYgItm1j23aBuU8pha7ruT9N09A0DaXUHnGuKFJkNDz88MOcdtppiAiHH344999//8e6\n3n5iTF5FPpm4rovjODnHDKUU6XQapVRuADPPVz6HUgpN09B1HcMwikJT5JPOXlGpi4JSZK9DRHBd\nl0QikYvtyQqBiOQEZbhjDCQ0QEGPJturKQpNkY85e0Xl/diavIp88sgKQHaM5PXXX2fp0qVjaugH\nE4isWDmOs8u6/iKj63pRaIoUGQVFQSmyx8k28vkD7ZqmISIDNubjaeBHIjT5521tbWXGjBk5s1n/\ncZoiRYrspCgoRfYYIpIbH8k24nuqRzDYebdu3cr06dNJp9O7CFxWYAZyCChS5O+RoqAU2e0MJCSD\nRUnvaQYrW755rv+YTlFoivy9UhSUIruNrOtvvllpNELiOA5btmyhpaUFv99POBzO/YVCodwA/u5g\nKNMZeJNkWZZVsK4oNEU+6RQFpcikkx9DAoN/9Q+GZVls3ryZbdu2UVtby6JFi7Btm1gsRiwWo7Oz\nk3g8juu6BAKBPS40+f/P0l9o8s1nWVfoUCiEaZo5odlbe21FigxGUVCKTBoDCclovsZd12XdunXs\n2LGDuro6li5diqZppNNpTNMkGAwW5KASEZLJZC7yuKuri1gslhOaUChUIDZ7k9CsX7+e+vp6QqFQ\nwT7FoM0iHyeKglJkwsl6bGVdc0fbACaTSZqamojH44TDYfbZZ5+Cr/VsPEr/YyqlCAaDBIPBgnkl\nRIRUKpXr0bS0tBCPx3EcZ0DT2URlYR4J+deQFY5smYGCoM78fYpBm0X2RoqCUmRCyA5Sp9NpGhsb\nmTNnzqgbuHg8TmNjI319fcyZM4fOzs6C+UvGilKKQCBAIBAYtdBYlkVvb++kC01/gRysR5PdNuvi\nnE6nC9ZlRakoNEX2BEVBKTIu8r2dsjEk27ZtG3AWxsHo6+ujqamJRCLBnDlzOPDAA1FKTfqcNEMJ\nTTqdJhaL0dbWxtatW4nFYjmh6W862509mmy5xyo0xaDNIpNJUVCKjImBghFHG0fS09NDY2Mjtm3T\n0NBAZWXlXtHAKaXw+/34/X58Ph/z5s0DCoUmFovR2tqaExqfz1cgMqMVmsGCOEdb7uGExrZt1q9f\nn7smTdOKQZtFJoyioBQZFRMRjNjZ2UljYyOapuUmgfo4kC80lZWVud/7C822bduIxWLYtr2L0GQ9\nufozEYIyVLmzx3Zdl3Q6ja7rBWbKYtBmkYmgKChFRsR4gxFFhPb2dhobGwkEAuy///6UlJSMaL+9\nvREbSmgsyxpQaEzTLBAa13V3y3X2d1cerEdTDNosMhaKglJkSMYbjCgitLW10dTURElJCQsWLCAc\nDo9o38G8uT4uKKXw+Xz4fD4qKioK1uX3aNra2ujr6+Mvf/nLgKazgXo0Y8V13WGfXzFos8hYKQpK\nkQEZbzCi67q0trayadMmKioqWLRoEcFgcFRlyJ/z5JNGf6GJx+MsWLAAEckJzfbt24lGo7v0aEKh\nEJFIZExCMx6BHm3Qpuu6dHZ2UlNTU+B1lnUIKPLJoygoRQoYaEKr0bz8WQ+jl19+mSlTpnDYYYfh\n9/vHVJbBBOXvSWiy5JvOduzYwcaNG7EsC8MwdunR+Hy+QY8/Gaa1wYTGcRxaWlqorq4eMJamGLT5\nyaMoKEWA8Qcj2rbNli1b2Lp1KyLC4sWLh2zYRoKmabvMGy8idHZ2opQiEonsdpfdyWK4noNpmpSX\nl+/iwDAaoTFNExHZbSldRCQnFv1/h2LQ5ieRT8bbWGRM5A++bty4kUgkQlVV1aheXMuy2LRpE21t\nbcyYMYMlS5bw+uuvT0hak/yeiIiwbds2mpqaiEQiAAWxIf0bz92ZVmWiGEuDOZTQxONxotEo7e3t\nOaFRSuE4Ds3Nzbl4Gp/PNymNteM4A4rXWII2i0IzNpRSZUBMROxhN54AioLyd8hAwYjZr8WRvpyp\nVIqNGzfS3t5OXV0dRx55ZK7xyPYsxtuoZxu/rVu3snHjRiorKzn00EPRdT1nusm6vUaj0Vy0e37+\nrj2ZKHI0TLQJzzRNysrKKCsrK/i9s7OT5uZmNE2jo6ODzZs359yIBzKdjaexHksdGG/QZnFKZw+l\nlBKvUv0YWA38VSl1C/B14ErgvyZDZIqC8nfEUMGIuq4POC1ufxKJBE1NTXR3d1NfX8++++67y1fo\nQKaq0eK6LqlUijfffJPq6moOP/zwnAkt38Mo32W3f7R7MpkcMCNxMBjcRWj2dGbf3eXNls0OMH36\n9ILf87M3T5TQjMSjbDTlHk5oPvzwQ2bNmpVz/hhojObv0PPsM8CVSqmjgAhwAHAP8FugdaJPVhSU\nvwNGEow4nKDEYjGamppyebbmz58/6Is5HkFxXZfm5ma2bNmC67osXLhwTIGP+Yki+2ckTiQSucaz\nvb2deDwOsIvQjNYr7ePAYGMohmEM2KOxbTtnOuvs7GTLli2kUil0Xd8lBY3f7y+oE4OZvCaS/Hqc\n9YbLBm0ONKUz/H24OMvOLq8FlAPfAB4WkS1KqQgQn4zzFgXlE8xoghE1TdsltgC8PFuNjY0kk0ka\nGhpyebaGYiyCkrXrNzc3M23aNI444gjee++9CR90V0oRCoUIhUJUV1fnfnddt0Botm/fnlt+9913\ndxGaiW58dlcPZbTnMQyD0tJSSktLC37PCk12moDm5mZSqRSapuXuU7Yx313XZtt2zsRWDNrMcSvw\nKJAE/rdSqgIIAn2TcbKioHwCGUswoq7rJJPJ3HJ3dzeNjY24rpvLszVSRiMo+d5htbW1LFmyJCci\nE2E6Gyn5DWE+r776KvX19cRiMfr6+ti2bRvJZDInTPlCEwgExtX47I6Ga6LMUIMJjeM4OVHu6ekh\nGo3y2muvFdzfrJlxvPerP47jDDtm83cYtPm4iNyWXVBKmcA/icikvFhFQfkEMZ5gRE3TcByHjo4O\nGhsbMQyDuXPn7mICGemxhhMC27bZvHkzra2tTJ8+vUBIsgwVh7K7yLonZz3LsjiOk/tC7+npYevW\nrSSTyV0azoFMQXuSye4t6LqeExqlFKWlpdTX1xfcr+7ublpaWnL3ayKFeXcFbYJXz2+++WYuvPDC\nEWd/2J0opWqBPwL7K6V0EXHw2vwbga9OxjmLgvIJYLzBiCJCX19f7iWfN2/eiPJsDcZQgpI/ne/M\nmTNZunTpoF+VmqYhIuz/g6cLfn/3e58bc9kmCl3XKSkp2eU+5X+h55uCJsOLaixM5ED5SM6VfbZD\n3a/BhHmie4BjYTih+e1vf8tFF120W8s0HEopH7AEb0DeVkotBXxKqU68QfmGyTp3UVA+xow3GDEb\n27Fx48ZcYsODDjpo3OXKpt3IJ51Os2nTJrZv376Lm/FojrO3k/+Fns9gXlTZAMR0Ok1XV9ewke7j\nZXfmRhuJeI1VaLLpZ0Kh0KSMaQ1HviPAZD6vMaIDM4AjAQV8BygFTDzPrssn68RFQfmYkT+o2NTU\nlAtqG80L5bouW7duZfPmzVRUVHDIIYfkGvyJIL+Hkk6naWpqor29nfr6+hEJSZaJTrGy8Ccv5P59\n7WFw3HGfnbBjD8dgXlTZSPf29vaCSPf+2YgnKknk7u6hjNWpYiihyTpL9Pb20traSiKRQClFIpFg\n48aNk+o8kc9enP7HBR4GngX2FZEXsyvyTF+TQlFQPiYMFENiWdaoghGzuZW2bNmyS2yH4zgjikMZ\nCZqmkU6n+eCDD+js7Bw0XmU4soLy4dXH5Mxel/A+75/9AHo4jH9GLf666QQaZhPcdy5mv/xXQ/Gd\nN+C440a27WR+1Wcj3X0+H/vtt1/u9/7ZiAebX2UsE3ntLkGZDLdhXdcHHNOyLIu//OUvBIPBAucJ\nYBfT2UQJTVZQ9pbxsTzmAxcC2/BMXsfgeXkl8GIe3xaRP07GiYuCspfTX0jyY0gMw9glF9JAZD2p\nWlpaqK2t5Ygjjtjla3eiPKqSySQdHR1s3bqV/fbbj/3333/ML1y2TH1vvc3/6/o9PS+/iWtZxJRC\nHEH6ldcsj6CFI+glEfzTawjUzyQwdw6hefsSmFM/7mvbnQyUJDLbM81mBcifMXKk6Wd2t8lrd2Um\nEBF8Ph/Tpk3bpQxZ09lAQtPfdDYaAUwmk3trrFIP8CqQBvx4bsJBoAyoBrZM1omLgrKXMhHBiOl0\nms2bN+fybB155JGDvuAjjZQfjEQiQWNjI729vYTDYerq6qitrR1w2/6D7B9efcwu2zjpNMlHnqTn\n2Zdwd7SjaQrNb4Ch4Tg2rqEh8SRK19B8JpKySXdEoSOK8plE311fcDzNb8IRZ8HUqaDbPL9Yx45G\nMfp96Y4sdD3wAAAgAElEQVSUf/zZC/zhst1nMoOd86tUVlYOOGPkSNLP7E6b/+40rw3mMqxp2oA9\nmvy4o2g0SltbG4lEAhg4wHWg6+ju7h6TF+RkIyJNwK1KKR0IAwkR2TXIbBIoCspexmiCEQcTgfw8\nW7NmzRrRuMVYBSUej9PY2Eg0GqWhoYEDDjiAzZs3j9m+HN/QyLZ77qf9D3/G6urxBFSBuIUBcmbY\nj5SVYXX24ooNmgJ/JkJaHDA0JG1jhnygaTjxNLf/6b9RugaOy/rfgFJgTqkgOKeO8IHzKFl6GCWH\nHow2RIP75JMv8J03vH8v/MkLu30sZiBGk36mq6sLpRTbt2+f9PQze4OgDMZgcUeDBbhCodBYlkVn\nZ+deKSh54yQrgRVAe+a9SeD1VO4Qkecn49xFQdlLGGswYn6ivPw8W7Nnzx7VuMVoTV7RaJTGxkYS\nicQuEfRjMZ8lP/qA9t/eS/crb6H5TCLTQ6TCDq4juJYLtuBaDppu4iQsrN4UZ514KRgGiHDs/T/i\nFDLebpqGXhbGduLYloPSXDA1xHYRy0b5fGBZCJDe0YkbjxN79wO23fMAStfxVVcQbJhF+MD5UBrE\nWbQII2MizIpJltGMxexuBko/09jYSElJCeFweMTpZ8YqCqNt5MeD4zgTklVhOKHJpqG55557eOih\nh+jt7WX58uUceOCBnH766cyZM2dU5/vlL3/JbbfdhohwzjnncNFFF9HZ2ckpp5zCxo0bmT17Nr/5\nzW92mR9nKPIG3dcC2zP/1oF/AuqB2KgKOQqKgrKHGU8wYnYMJRqN0tTURCwWGzbP1mCMdPtsKpZU\nKsXcuXOprKzcZd/B0rhkyR9k/8vyalqu/lcSH63DcTREwE0kcEVwHRvlupim5jk8RgJY0SRmmY6v\nwvTExCs8T518BSsevQZd17DjaZyuPjSfD0nlZaYVQRkaKIGSIGLbKHGw40m8S1AoXSe1dTuprdvp\nfvF1xHV5+4Y7Ccyagb9uBieVz+EB5uYOee1hI7ptew3ZQfnRpJ+BsQ1s784eSn7alckgX2iqq6v5\n3ve+xxFHHMHatWs599xzeffdd0d9/r/97W/cdtttvPrqq/h8Po477jiWLVvGrbfeyhe+8AUuv/xy\nrrnmGq655hp+9rOfjerYmWzDHwEf5f38uFLqdmDkaS9GSVFQ9hDjDUYEz7S1bds2uru7aWhoGPVc\nJqOht7eXDRs2YNt2TkgGYyQ9lL+cVEXnb+6m9drNiGWBZpCOJhHXyTkfGH4DN+kJkzJN0r3eYKpr\nOyA2iED2etNpnFgafAZ6wAcaiAviGpB1XFAqs07hJu3M74LSNZSpe7+7LmIakEqj+U00zcBNWcTX\nNZFobObL9os0ciBvfvHLLH79MRo2Kra8/SbhgxcQWXQQvmlTR3Q//8+TL7Amr7dzx7G7Z3B3qEZ+\nqK/z/gPbiURi2ODDvdnkNRH09PRQWVnJfvvtV+ChN1Lef/99lixZQigUAuCoo47iwQcf5OGHH+a5\n554DYNWqVXzuc58btaCIiCilDgQcoBPoBmxgAXDfqAs7QoqCspsZbzAiQFdXF42NjViWRTgc5rDD\nJu8zuaenhw0bNiAiNDQ0jKjrPWSkfFc7bf/5E5Ibm/BirjxScQvHsdEATddRgCuC8vuQdBo33W/m\nRoCtW6GmBmybO3//SyQUQlzBtQRxHcRxQfMG7ZWpozSFOC6SshBX0EwT17I46/AzYPp06Ovj9j/e\ngFkaxs02iuKiTB2xHdx0CkyTC+2/weN/A6WIJ/zEPlzH9t896vUuAz7MqVUEZs4guG8DoQPmEzl4\nIUZ5oa19TT/T2ZNPJVi8eNhbO27G4uU12MD2cOlnsrEiWWeCyfQu21OCMpZM2FkWLFjAlVdeSUdH\nB8FgkMcff5zDDz+ctra2nENLTU0NbW1tozquUkrL5Or6D+Af8ETFwPP6uhHPA2xSKArKbiAbjJhN\nAT5t2rQxRbV3dnbm8mxlx0fWr18//M5jICtaSqlR5/QaSlDab7uBZFOTt2DoiJXGFg07ZaFrGvl3\nRAGaEghHSLX3ej+KgGiI4+fedQ9ibjFQusKtKcWJpbBiqYza6PgqSlC6TrIzgTgOYkum16J5pi9N\n8XRJGGbN8o5dUcHZS8/m9pdWe55jaavgK1vpOrppQCiAOBYiLul4ApUptTJN3FgCZ2MLyY0tdL/4\nKprfQOk6WjiMWV5KoL6O4L4N+JhN/jRRx+0FPZTRMlz6me7ubrq7u2ltbR0w/UwkEsE0zQkRmj0l\nKIN5Mo6E+fPnc9lll3HssccSDodZtGjRLtcwlg/ObOJHEflyv2NpgCoGNn5M6R9Dkk6nC74+RnqM\nHTt20NjYSCgUYv78+bkvxWQyOWHBiFk6OzvZsGFDTrT6pxAZCYMJSvzNtcTf/isAomlIOgVKw7UU\nOgN7hSlN4aS9a3TTDiJ+xBY0bJy0jZNJwi2aBrYFmiJUW47ruIBCLBvdb+Im7IxCkf0PoNhQPrvw\nhFVVGEEfroDym5BIecdHMAI+lAZuykIyzhC6pqN8BkrXEAQJhHFiMTTTe7XspIXCgngKq6Ob+Ecb\n4ek/cStwxhHnQlUV09veQd25mbaPGoksXEBw/v5ooxhgzs8A8M6VQ3uc7Y44lGz6mWwdyk9TMlT6\nmf5CMxocx5mQTAKjobe3d9xeXmeffTZnn302AFdccQUzZ85k2rRptLa2UltbS2trK1OnjsyM2h+l\n1OnAkyKyXSm1AlgM/IJiHMrHi8GCEX0+34gCEbPHaG1tZdOmTZSWlnLQQQflbK1Zxhs7kn+uzs5O\nYrEYmzdvnpTkkK5l0XHPalxXcBGU4w0OixFAer3ex9fmXwB+P0Sj/G79agCUz4/T0wdaAH9VGOW6\nOMm01xNxdoqQ0jUClaVY0RTxrd2FBdINjKAPI+z3BMp2MPwG6Z44/7tjA6e5nmkMEVb+4ToQhRKv\nM4TfRPcZKJV5rmkbsR2UaSKWZzqTzDgPuoYZMdEMX2YbHYIBRFzEcXAEVHamTNPgrnfWePsKuJpG\n82tvZ65Fx6wswzd9GoE5swkdsD+RgxYQmDN7FzHIF5Ps8lCisjvHNaDQ2WO49DOxWGzM6Wc+jiYv\ngO3btzN16lQ2b97Mgw8+yNq1a2lqauKuu+7i8ssv56677uKrXx1zYuAfiMivlFKLgH8F7sKLT1k2\nWb2UoqBMIMMFI44ksj2bZ2vTpk1UVVVxyCGHEAgEBtx2vIIiIrS3t9PY2JhzF12wYMG43S/7C4rr\nurTc9V8kWttQSqEbJjg2RkkINB+aBl/yfwWyUcelpTwJHKvrWB0xnKQX1Gh3e8IjgGZq6CUmRsBE\nD/pxXYXTG2XAaR4cBztho3w6us/A6oqTFsEI+vCVBvnNa/+NWBZWPOX1k5RglgbQAybx9j6UIwgK\ncRXiaKA0lG6gDB3ERUTQTZ3TPn8J6DqIsOZ/fgRW4bPRlMI3pRInkQAE13ERTcO1bHBdXE1DiSC2\nTaqjCyceI/beOjoeeQpxXTTTwJxSiX9GDcG5cwjN35/PTwnwx/ad55g7TBb13RkpP1Ky6Wf6N86j\nST9j2/aET8Y2HBMhKF/72tfo6OjANE1uuukmysvLufzyy1m+fDm333479fX1/OY3vxn1cZVSfqBX\nKRUGTgf+E/gNsKpo8trLGWkw4lBjC/kzFk6dOpXFixcPG9GcTe8+lvJmzWiRSISFCxcSCoV4/fXX\nJ8SfP3udWXFsffsNKl960ZuUyDTxhfz4S/zYmFidXZ73b78B3z+VHM4X2l7D6rXRw2EvuJH8YXzx\nYkpCfmJb2r2V4nUrjIDJ02EfXzKNzE+CoJHa3k2+M7OdSGMn0qAb4HjPLlRbjlLemVzbAdNEpS0U\n4lnKNK8E4riYU8oR1yHd2YeTlgI35jOX/RtrHr8Gz3YGesgPtovV2euljMl7brqm4WoKX3kpTirh\nnTudxupL5LZRPhMnlcbd2kZqaxu9r/0VPejjTOCPh58HpaUQi/Ff5Vvpfg7CBx+EWbFrY5ftMQ9E\nfm/nkGr41bl7NmBzsPQz+UKTTT+T9UArKysbNv3MRNHT0zOq+JCB+NOf/rTLb1VVVTzzzDPjOi6e\no/064AfAZ/EyDJdlfp80ioIyDkYbjDjQi5ydaGrr1q2DTjQ1keVta2ujqamJ0tJSDj744IJcRBNl\nQgMv8PHll1+mesoU6v66lmjKwldZSiDsB9eGQBCrdeen9VM993Js8JueG7Dr8uO+t0n0pRGfP+ca\nrDQdXK98ZkkIsWwS23u8AyhA6dzn2DzxT5eCUqwR4dd/vpFkex8YJmYkgB70BoGdlI0VTSKOizJ0\nAtUhnHia1I7egutwUZhBH0bIhzI0XMvN/N/G6ouC46AbAzTQto3yBTHCJk7CcxQQTQNNULqZG4MB\nLwsAmsJNJBBLvLElPG83zTTAUIgCMQ3seBLdb4DjYsW8mSPXvHijJ7QitGYPqsAsL8U/owZzajXB\nfRoILzgAV40sOeRfdozkKe9+8rMC5Luuv/POO8ycORPXdYdNPxMOhyfE7Nfb2zvuHspkISJRpdQ5\nwEnAahFJZybc+uVknrcoKGNgPMGIWfLnB5k5c+aQebbGS3Y8ZuPGjbl09QOZ0cabIDLbI9m4cSOu\n67J06VLstc/Q0bKF0plVKF15brxGEFsZKJ8PSXmNJwoeb16NOC6az0fK5xCZXYpeEsYIBL3B+7Rg\n9/ZiWy7Jtj7chAUKz8vKcwnjiS/8fztjU5Ti39yD+QEvgghWNIkV3TnNsR4w8U8txVU6ON6Avigv\n7jGLEhcnkca1HPyVYZxEGieRRg/68JcGUbpCbAc7aUEs5pntHIcH3ruTlF/DjafQAEF5YzNKIQh6\nJAKOhWgKTYN0NIUTS6F0nTM/fQFUV4MIp/zPjzgO79rMyjJU0kLS4glNyPTGdRzXc3POBHEqn4Fm\n6KS7erG6elG+jXT98c/eMxLhg4pS/DVTCcyeRfCA/QnPn0fogHljfu6DsTvTu7uuSygUGlH6mXg8\njuu6u2QFGG36mXQ6Pag5ei/BB2wEZuL1VpqADyfzhEVBGQUTEYzoui7vv/8+XV1dI86zNVZc180N\n7FdWVnLYYYfh9/sH3X6sPZT8cZ/q6moWLVrEBx98gJFO0vvCkwTCJph6Jr5EEH+IdOsOlKbA0DGC\nPpK9KRCH0roqOt5vAaXQfCZuMo2d9gxVjuOiNAE7RbBKB93nNayajriCbYO/uZnU3J2R7D/ozEwF\n4e68LiMSwAiapDqixFuzA/iKbJCjURLwegEipOJJAmUR0h1Rktt39l6ywoKmQUaEf/3mavSwL2t5\nQy8JId19iCMoxNO5zDdDYGo5TjRGui/pFU1pXgJLTXliohQoxa9P+B7Lnr4WJ5HGau/2nAFsC8cu\nfE5KKcyKMi9mB68AStdwLBuVEe1sL8bq7kXSaRIbNtLx1PPetpqCw77pJc/UhD/u30Ni3UcE9t1n\nTGMuOfPZH4Z2EpgoBhuUHyj9DHhCk58VYLTpZ/biuVAAUErNwRuIPw4v7uSPwD8DhwIXT9Z5i4Iy\nAiYiGDEej9PU1EQikaC0tJR58+ZN2OBof8+d/g18/rwnQzHaHkr/82THfSzLi9+IPXIPEu1D+QMo\nDZQI6Dp2xhVXaRqh+ukkWtsR26WyoZqeFs+EpfzZLz/JtNeCpikMvx8nlgIUOC6uk/kq1xSGFuS2\nDx7gdPNUqKlhztq1uIDu80xHRlkYUpZnVbOcjBhkGmZNB9dGHJd0dxwz4ufpiJ87jjiPxa8+xaUR\nC830AiftRBo7lvIaaKVhlvrRgyapzijpvB6QMgxwBTPsRw+YoHuxL5KySG7vAsfxLHUKDFP3hmcG\naKhEFHrA73WddB03pXBTVm5bI+xHHBe7L+FlHchDVxrKNNBLwjjJBI7rgEA6Gi+InxHb5s5Xb83t\ntyXTy0EzMEpC+GfUeh5n8/cjcshB+GfMGLRe9Pc82x2M1stLKTXm9DPJZDL3nuxtDg55QY2fxksG\nuQz4Xma1BdRM5vmLgjIE2fknxiMk2SSK8XichoYGEokEU6ZMmbCKaBhGbiIj13Vpbm5my5YtIx7Y\nz2ekPZTBhCSLpmkEtjZhbW5E8/tRug6OFwcipp90eztmaRgzYOImU1iOTuXcKTi2S6K9B+XzobJp\nUMQTEqU0tECAVGfhGAcKNCNEfGsfTqITUYq73rynYL2vNISTsEhs6fYG3wHw0raYkQBGyET5A7jR\nGKKBGfDE4Y6M+ey1Y07itA0buP29nd42mqHhn1KC8geRZBw7kc58tea5DbgCAlYsiR4wdwZe4pnb\njLIgmqHhWg6p3hg4eAGX0WjOSWHuiy+ie10Lz9PMEZRuoPkVRiSIk5nbQ+kayvWcCPJF5YwvXeY5\nC7S0cNc79yKJtJc8U9PRfCboXk9IfCZuKuX1kJTCTaY9M5pSOPEEqdYd9L7+9s7rD/oJ1E3HN20q\ngfo6QgvmEznk4FFNcjbRTMQ7NZL0M2+99RarV69m8+bNLFmyhHnz5rFq1So+//nPj+pc1113HatX\nr0YpxcKFC1mzZg2tra2sWLGCjo4ODjvsMO6+++6xTjeg4c2LcmDm/+ANyvcOuscEUBSUIcjOGDgW\nIenp6aGxsRHbtmloaMglUWxtbcW27SFNT6Mhm3G4paWFlpYWampqBpxAayQM10MZTkiyiLjUbF8P\niVgmXXzGVVo3sGJJgjVVkEoilkU6paiaUwlWmp6WDlC6F/TouqBU5gveu/d6MACqL3MOwbVBiQ8n\nnsYs8WOWBMAwIZ1EFGi6gdguruUiluuZdST3XQ6AE01hRZP4pnojHXaf1+CbFeGdYzFAetYseM/7\nt1kWRNN1ktt6QOvLmbw0n4ER8aP5DMR2sOJpzHAYcVySHdGCe+QkLdy0i2R6SK7rEJlRibguv1t/\nD6KbOL19WMl0pqyZe4Hgqy5D0ims3iSeEREEL/pfAfhNEIfTj7rIi+sBmDULeWvnsxVXcJJe704A\nX0UZWA5i2Wg+HRUKIgiu4+CK2pkPTfPyobnpTH6zpma6//RK7rh6JMSZ06awZv7pnlihYcdiGOFh\n/JkHIX+6gN1hOhuM/PQzJ5xwAkceeSQXXHABjzzyCB9++OEuaWmGo6Wlheuvv5733nuPYDDI8uXL\nuf/++3n88ce5+OKLWbFiBeeffz633347F1xwwYiPKzv95l/F642cDTQqpS4AvgLcMKqCjpKioAzD\naM1AXV1dbNiwIZeypL8XyEhnWRwJtm2TTCZ54403mDlz5rg9xAbroeQLyUh6PvLa05ixXq8nkDXh\n6Dr4/PgMFzdjQnDNECWlCoVg6z7i27sRnx9dZQevPeH4St0ZoBQPbbgDlMLqTWL12SA6aKnCkysd\nX4kfO5rGivYVXp/fxCiLoJk6ruNgJyz8pV4GY7urj2yjbfcksTLjEFlR8W/ejFkdQdM0byxFMpLk\nuqC8TJRu2ibd6T1bszyIGfbjWILh1wlWl+BaDum+BJINyNQUuBCoCmOn0gVjNOg6OA6armOE/eg+\nHXSF0jRSXXHP5JW9ZPBcnTP6p4f8GCET+n1U6AETN+THcG3EFhzLQg/4PfHr6vFMX47nhFBwSwG9\nNIIS1wvSzNwXV1xIpbyyuq5376JxPpfeytFb/8PzSPP5+ev/szFKI/hrp+Kvn0lo330ILTyA8IID\n0Yb58MmfLmC4oM3dSTYGJRAIcPDBB4/pGLZtk0gkME2TeDxObW0tf/zjH7nvPi9346pVq7jqqqtG\nJSh5bAR+jzdj4yJgX+ByEfnrmAo7QoqCMgGICB0dHTQ2NubmBh8sZclECEq+q7Gu6xxwwAHj9ocH\nTzzzBWW0QgLgdLfjbHwXu6sL3fCBY3Mky8AxIB7jz/IH71wlJfiU1wiJCD3b+iAQwtAKe4InzPmm\n12AB/7zvN7nnmZ+iG36CtZWZ2BPXGwR3BaU07KRLoq2XgTK5OCkLx3IwQiZm2OfN8tidwIj4McIB\nXNFId/WAKygUpz787/zfz1/Isrfv5Bua4c0IKZ6Lr17iRzP1zJe+gyRSGROaHz3kI9UZA1Iow8CO\n2tyHlzP8iwovaNJnovx+lNgk23txLRtNyxsDyHzEiOPipCx0n05ye1/uurSgD8NveGUQcNM2TsrC\nVxoiHU2Rau9j8TMP8do/fs0TRcdBA3TbRmk6ZlUEO57ETdtopkIzNTBM3ITCTe8cn1Gm7qWuSaY5\n/VMXQCrFmpf+21unNDRTRwsFENvyknkqsOOpnOHPtb2gTaunD7unj9gHG+j4/fMYIT+g0MMhfDXV\nBOpnEZq3L6GDFxDcZx+0UYyH7IkB8p6ennGlXZkxYwaXXnops2bNIhgMcuyxx3LYYYdRXl6e+yic\nOXMmLS0tozpu3hjKT4BnReSHYy7kGCgKyjAMZeoSEbZv305TUxPhcJgDDzxwF9trf8YjKJZlsWnT\nJtra2nKuxuvWrZuwF0rX9YKAxNEISRb3lSexdnR6brIKjuRoPO9FQEp4EvjK1Gp0U8Pp6/NSsbiK\n1I5elC8Adr8eR753jVIEqks905W78ytaAXogQGJbFHGFQG0pmiZeoyTgWJmvbl1D8wVItnZgR3em\nwrezX+S6iRLBLA1iBE1ONDSOf3M1dl8KSzcyg/BehKXbl8Ilo2majq8yjObPeJul0ri4XvoWx+X0\nf7wkF/R4r2Xxf1+6CRWA1PZOL/BRAz3sxxcK5DIiOwkL23IIVIRIdcZItheazFzLJZ3wvJIEIVBV\ngpu2sWJJzJIwKuLjMnc7zsvXY8dTkOlk+0r8GD4f6a4+777lxme8Ccg0w8h4nBkgjlc+EVZ96fLc\nszjz+CtY88hPIZOKRulptGAQp7MHIeMI4NNxlUI3NMT04Wai3I2gD2wXO+bdf8dySO/oJPrOh/Co\n97GhTAP/9KlMW/4VoHCO+AHr3G5OJwPe9L/jiUHp6uri4YcfpqmpifLyck4++WSefPLJCSwhCpih\nlDIAXURSw+0wERQFZQy4rsu2bdvYuHEj5eXluwQIDoWu66MWlPyYlbq6ugJX44kMRtQ0jR07dox5\nUN9Z/1eksw2rpxejegqRsjBsyO7vueUG1SzC+04j+t77XuiIppGIuxCKFE6GldmDVAqyvv6W5WUB\ndiX3BQ+gGSapjmTOlOT1RAq9nZ6tjLD6sAtyX+u/fum/sKIp7Fjee5ZJx2IETVJdMVzbm5vFDPox\nK0pQOF6jHU3k8oj5qyIo0yTV3gN5FjZd80xV+I1cDwsAw8BJp3A6Ml/xGbdjJ5YilcjO2wKBKSVI\nysVJ2fgqwjnPNDuewk3vfN6+8hDiuqTyxmicxM4eBngWOV9lGGXq2GkbZZroQR92IpUbmcmOzyi/\nj1CZn1RXDHEyQpMtZ941GJEAIIiAE0vhpuzcmKOIICkv84CjFOgWRiiARhrlKM8UZ+KlnrEzThJ5\n41C+aWX4pwTpe+FZXr/0Mvz19QzFRM3WOBrG20N5+umnmTNnTs7D7MQTT+TPf/4z3d3duTQyzc3N\nzBjCm24YNgHfxksI+aFSysIzf90lIq1D7jkOioIyClzXpaWlhc2bNzNlypRh4zoGYjQ9lPy54evr\n6weMWZkIQcn2SDZs2EAwGBy1kAC4Vgp5by2ubROZW4fYFkpX/G3eRyz44CBAA2Xx1SOneuYdN2OH\ntx0klSY8JYgySzEyrrMiYFsWT7TdQyDiI540SbV7MSNKUzstWprCSXqD3FmUSIHFy1caZPXB5+wc\nZNd10t1eA2yW+jDCAZTm5ehK7ujB7hct76QsnG1dmSUvviNU632duo6D3RfL9Fx2nlVcwepLeiLj\nujtFxXXxlUcyiSYt0kkHLWveEsFfFcZN2SR39HniM0Bd0f0GZlUZyrUR28VJu15XIzs2kzHL4Xoz\nVPorwqQ6o4gjuK6D5vODbaMZmfEZ0/B8yHSFnXK9c3t3cmf8jOPsvAbLwldeQrorBq6LHgqiNLxe\nSjzuCWTKAheMiB8naeOkbCQ9cL03QiGUCf4pJRhVYVwBy7ZJpdN8uPpWtOUriGSmLR5otsjJnq1x\nIMabx2vWrFmsXbuWeDxOMBjkmWee4fDDD+foo4/mgQceYMWKFWNKDJk3KL8RL7OwHwjjmQmmDLLb\nhFEUlGFQSmHbNs3NzbS0tDBt2rQxNbhZDMMglRq695lMJmlqaqKrq2vYueHH0uPJ0t+0td9++9HX\n1zema5M3nwFsr80RF0dpaOEyrL4YfzYfJzxvH3wlYZTScPPMWlbSxs2MHYhl4boOOF4cC6kUwYoQ\n4i8h1dy86zlF0FSQVE8/T0hxsx0i9KCPdE/SaxD7DQALYIYD2AmbdLdnPlJmJkVLwEAcl1RPzDP7\nIGg+H4HKIFY0QaK1MKOxF9GvoTQNJ5X2piqOeAkm73n2l3zjH84H4O7nbyD/6euGiR4KoId9nsCm\nHRzb8eJFbNBkp7cbKL5+zKW5Xs29z/x851iRAiPsQw+YKE3D1UyUa2H1xvMEInfjvP/ZLlZPApXp\n5Vi9SUDQA95xNDNjMkxb3Pf8/2Hl/isp7+3lpuZHsDOalbGZecdLJtCUl5fMrI5g98W9AM+A6ZlA\nzSBOIsmZx1+ZE6eGF17gmvL3Ce9Xj9vbQ3+ku4uydBo7FNpltsj+aVR2Z+LL3t5eGhoaxrz/kiVL\nOOmkkzj00EMxDINDDjmEc889ly9/+cusWLGC7373uxxyyCG51PajRUT+p/9vSilDRCbGI2gQ1F4W\n8blXFQZgx44dvP3228yYMYO6urpxd63b29vp6Ohg//3332VdMpmksbGRnp4eZs+eTU1NzbAvSHNz\nM47jUD+MWSCf/kJSX1+Pz+ejo6OD7du3M3/+/FFdkzc/fPbR2fxtwQfELBfDH2RH8w5mNEzD8Pty\nA7VWZxeplq24tk20PYb4d5oLNREklUDTlBejIUJfWxKrN1ZwTrFtdH+IeHMXAyEolCYo9Nx4yWlf\n/Gy+UD0AACAASURBVI7XkCUS/ObtNdi9KdK9icIdddNLw5K7V0KwOoIR9oHpx+rswYkP8EFg7Jxq\n2CwPo+kK13G9iHufHzcWx06kcK18j0EhMK0CJ54k3RP3ekkZlK4wwgE0n9fwOrbNyhn/jGQnAwOI\nRrn3pZvIj30B8FeEcNGxuvoyx/FcmZVSWMk0YgtuJn2MEfSR6iwcn8l6mO1cVgQqI9jxNHokjHIt\nxHVz+dBU3vl9U0q92KJo0ovAV4pMngBv7MlxWfWly3Ye23F4pOMBjKnTsLcPPDOhWTuD6Zf9W8H9\nyc4WGY1G6ejooLe3F8MwCuZWiUQiQ6a8Hw9XXXUVRx99NMuWLZvwY4+Rgkqg8hoO2Y2NfLGHMgxl\nZWUTmmdrIJNXNoq+t7eXOXPmMH/+/BF/aWXjUEbCcIPt2UH50eBt77Bz0iodW/PhuHFSepC6Qxeg\nuxak4jv3SXoNcro7ipiFuZBcw8SnSy52Je3ohWKiIFBdgfh1pDtKyT6VuJaLE7exepM4aQfxBRDD\nxKyswOroxj9zBnrA4LdbHgTRSG/rJd5loQR81ZXofg3JZPd10jvfPbM0iPJrpHsSpHviuazEN/j3\n4/V/OhFcl/tfugkrmvCcAaoimUj7neW1oCA9ix400cN+b257EdK9MW/Mox/iCFZvAmUYBCr/f/be\nPErSqzzz/N1vjTX3pVZVqRaVVCAhJLkloM0itmpoY2NsgxE21rHbjOkeaLtnenSmp3368IeRZ6bH\nx8MY7PawGQMyeIb2htnE1hiDJIPFIlGqrEW1ZWVWbrF+6713/rhfrBmZlVsJnzn1nFMnKzMivvgi\n4ov73Pd9n/d586hYcujcOU73EYpV8LKiu8Yt5bBch3C5YZpJ28fpdO8rJbFdl/yuEWQYI2xBbqKM\njFPSZoROVU//jT9eIm1G7SgnDeIsYjOwbAun4Jt6EQKpTOSTzQGguz6DbeFkBNle/pyUodtvI6k3\nsfbtI754if59ZTJ7ifpj36Z834vaf+ueFul5Hvl8nsOHD7eHeNXrdebn56nX6z2W9y2S2a4T8U5Y\n119PPJck0o0bhHINOI6zo7LEbkJpNBqcOXOGRqPBoUOHOH78+KZD9o3UUDaq2tpSPeaZx+klFE0q\nU2S+xOTUiFk3bB/SuE0SKgxRUpM0Iyh3hoa5u3ahohBWjAuxEoL6hUUA/LEh/PEiOokQSUqqLZKo\naSzttcbOC8RQEQplRLFMGjnIqwsIR5GuLKOHhtCJILw4n8nCHDQQBxICiWm5z+FM5PHHy4g0IV5c\nIlyoYdvZa8vemydOZFJcy+ItL/k3fOqxP0Rr41qMEDilXLZzb38AICy0VrjlPGkYt1NsAHYhh3DB\n9VxUpoCScUpucgiV6nbB/T18iweie8HzIEn4+A8/jpLSiAbKeWO3L0zjZZoZWrbTZdl15Y2X0M2E\n4GqNB172bhONrKzw8cf/2JxL3sUZKmMJiRCCuBaYekjfa2nLim2BnXON67OiHakJNyMazzVOyYkk\nNzXE9EuOM3T5NFVuBlK+e+RH5rhDecTIBPKmMikuaaxImglJtU68vMLK3/wVxbvuxhpw3Xbbrgwa\n4tVyvKjX6wOdiFskUyqVBvp2DcI/dUIBEEKMYD6VQGu9etdyHXCDUK6Bnc7JOo5DGIZ873vfIwgC\nDh06tC0rlvVqKJuV/260ifMX3/8lvjMHv3cHvLZ0ge+9LM8dX/MBl+/fdRFL5KgM7wUdd3a7fgGC\nGlqZxsZopYp2zLlYxSL5vdM4QlM/PY8WFgKFEjkKe8fxbJBBE5oNY8NSKJJeWcwWSwG+j3Y8LCHQ\ntk24GJFcuQRa4kyPgXSRsUTYAm/3OGmlgQqiXt8sIcjt34UKYxoXFrPbbKzhMbyxsskC1RvEK7We\nHTyW1Zl335UqshzLRAyeY2aQ5Iy1S9ifXsIU/XU9Ic0WMn+8iIPZ1TulPLZjIgMZJHz8ax33ceE6\n+GNlIyueX+2o4eRNLUg4NlhG6BAt1SFV/NIr/4fO/Ja+eSNCK0MQ2dsjLHDK5rUIS6CFTVyp4Q8V\niCt9NZpWfSaRJNWAlJD81BDTr7ydwq4RRHGIrxWqwPdXna9OEyyh8YjxfMAXMFqGA2WUttAXz8Ch\n1a7I1xqu1ZqWOjY21mN53+1EXK/XewwiW75dLbLJ5XI939F/yoSSzY5/BaYzfphOBmxZa/1vr+dz\n3yCU5xC1Wo2ZmRlqtRpHjhxp27FsBy0vr25stY9kIxHK6//3LzGTrR+/+T34xr/YRWFljn/82X24\nz/4QEOh8CWtoBN2c7+zSLRvcHLpRRSpIKnUYncQ/cIBCTiGICVKT69bCwvI8VK2OFSfIvvEAMlF8\nEM1nDr4DlOIvFo3Hli4OE1yuInKC3JF9iEgTnLqMbEZojK9W6ye+hzdSxMr7xuU3SggvzJvbuj4T\nFSdE8yu0cjTO6AQkSafAv7jYRSRdBepUEa00yE0OmeL31SqWDf5Y0cyfTxVJLUC100NGAmxZ9EQv\nWM12isnybJyCj3BtU3iPE8LlBsjBmwAlNZY0PRrRYgNTxFC4xVyvDBiQQuGPFEkrAcHcEggrM3UR\naIVRrGXwJkdxPJd4pYFT8LGGTbSnY0kaxm2/TSfvMXXfLYzctrdznbvrXIdBc82bLKEQ9aWBt21V\nNryWE3HLILJer1OtVpmdnW0LAaSUfOlLX6JerxPH8ZaEACdPnuTNb35z+/czZ87wnve8h1/+5V/m\nzW9+M+fOnePgwYN86lOf2mrD8n7gg8B7gEsYhVcB2FhufBu4QSjPAarVKqdPn0ZK2TaI7J7ZsB10\nk8B2GhJhYxHKTI9gSPP1b7v87AvyxE6n8Cn3HMJqhj3WJQB4PrImiJdWsMtFCsduxumaoZhWzcEV\nwuzol2pZr0rnGCKXI7qyzGf2/ao5tm3z0xNv5r+mXyYNFVbeQWtBdLFmopiWK2zWaNmenZJKbN8j\nrdSJVurmVF3byHFzPiKKSZaqEGUDvUZLuKNlwtkFPvq1jisvgDUxiTOcKdjqdeKlFdNoGAREy43W\nm4uMFXKpE00KAb/0wrcbu/pKhU+d/FPiel/BvzXvXmlknOKW88hGRNrM1gbHwSm6mbrLNEWmgVHL\n+eMjhLMLpN09bUIgg4TXfeX/4LOvNIPISFP8Qg4ShTdipmOmWqDqTVSYtNNmdsHDzfvGoDO75uKK\nETU8cOtbYHSUj//9+3FKecpHpxm/8yBeoW9eyHqLr0zAz0EUDrxZVxYHP0zKHfPGg7UNIqWUXLly\nhZGRERYWFviN3/gN5ubmOHHiBA8//PCGj3/s2DH+8R//sX3MvXv38sY3vpGHH36YV77ylTz00EM8\n/PDDPPzww/zu7/7uVl5CHviG1vpD3X/MIpfrihuEcg1sJ4JYWVnh9OnTAAN9vXYCrZTXxYsXt0wk\n3cdaL0JZXl7mHgeeaK+Jgn95YhS91GgX3bWXIxkax6qdGyDZE6TKwfUditMT0EUmEhu5vIRWCjE+\nSTo7i2UJlOo9ipLGAbdnYbJtorPz5PZNkbt5F83Tl8lN5RB7DkAqUWGKrIfEyw1UI8KbNpFhdNGM\nJRTCynotNOn8cpa1Mbtz/+AuU+BOU6LFFVScdD21+U9abZJWm2AJ8vuncB0X7VjYI6PYk5JkpYqs\nN0BJhMwMIbXmkwUfdmVu4mNj/Pf+P+c/V76IU8xlBGEinTRWOL6FVrqngbGFtNFp0Gz1ncTVABlG\npkGSLHrIrFYAHlCKB5/4AFq2UlMtZOTjOJAqHN/FKeUQnpFRp1GCEsrY/mdE87aX/dt28+kDr/73\nfPvgD9G6iT0gatDJ+ptk4efRaxAK6xDKc9GHYts2e/fu5Z3vfCePPPIIn/vc50ydaYOimEF49NFH\nOXz4MAcOHOAv/uIv+OpXvwoYH6+Xv/zlmyIUIYTIivEWkBdC/BLwt0BTa93UWquu+1wX3CCUDaDV\nAbxRtAwiLcviyJEj2+qoXQ9KKebn51laWmJoaGhb/TGwdoSysrLCzMwMjuPwR795H9/4xrd479/D\nl9+SJ41CZHkCa/GKOac9h0FY2LK3W7uN5Tm8oQJ0SYW1hjDRaGWsP+yhIaILF+mncuH5hFeXSe1c\nb6OdVWPs5jJaN6ldaaKa5h+ApbOahoDcGLi33URSDbFcD2dkL7IR8Zbjb6Y1evgjX3g/CE1+/250\nkhCeu2LkqhnJWMU81kjJSJqjmDST5ub2TCIrNaJL8wNft10u444Ogy1QaYJl2/zN8Ct67jN3223o\nuS/2TJZ0h3LYxQJptYaT9/AnyiZaiRIzV6b1XBb442WSatCuaai4Zii7fToa23fJjRSz6ZQ2st40\ny0//x641Go03UiSpBsjlTjrKAuxiDtt3wRFGJNC+URDmHZyaGalgBA1dn+Q6aS3zRq29JOlGFZ3E\niL602XNFKP1obTa385175JFH+MVf/EUA5ubm2L17NwC7du1ibm6wjHoDGANeAtwD/F52rmPAI1rr\nt275ZDeAG4SyQ9Bas7S0xJkzZ3Bdl2PHjlEul6/Lc/XbyBcKBY4ePbrt4/ZHY5VKhZmZGYQQPa/n\nxIlX8ZrbT5LGMU71PPHkTTjnnkY7HvGIsZKwZdzfHmFQryBsG23bbVNHMH0RlmXh3XwzzdPnTSW4\n/3ULB2U5qPFpPuM+hptU8ZVpxgNI3CGSq+f6X1RHkeQ4xPMVZKPZvu2tx9/WiXYsiwef9wb+LP4u\n8fwyaaXeexwEqhmhghi0xi765PdNI8PQRFLFvCn+B0GPukoAstbEyuew8x7JFVMLeKT+Wd4y+kD7\nHH/5B5+AQgmhjBrMciC8sgSNBDJ35J6XZgnsooM/PGSaSaO09z3XmKFesmNrYhV9goVqjyILC9yS\nqSUJy/TPYDvoIFzdFJlBxhIZJKbPJoqgkG8/ad73SWqmWJ4kSTbTxjL/7AQrV8ROB+/q9bVEIdUl\nGO+dEfVcE0qSJDti9RLHMX/5l3/Je9/73lW3bWVkhtZaZ+aQ36BvkJYQwsZ0zV9X3CCUDWC9CKXb\nadj3fW677bZrzkZoRQKbNbRbq0aysLCwqeNcC9VqlZmZGbTWa0dYUYPYGaIQByilEUoi9x4xxXfA\nkTHa6f2SS5lCmqDzBbMAa51Zntio5auIYgmzaDcQPe1ypoM9qjfx73sxzsJVqLQKtKYlXjsujTOX\nB7yarGUecMYmCM513UcDYQiFjnT5hT98jCZXjPy3nEeU8jiei2xGxIvGidgZLuKOlYkuXyU8f6X3\n2YTALhbNdEjbQjZDtC2wXZtobilbKMx5JSs1PvaNPwQ0SplF19+/q+0gjFQ4do50uQpaIZRCS4lQ\nEi0l3mgBpCKYq/Bxpfjsa/599ro0n378D0ynu+sjGw1s3yFaahAv1bFaZN26phUk9QjqEW7ZDESL\nlypYno03nEe4xhhTJSlJM8pmy9iMHN/L5E8c4bv5Z3jTDJzhJr575DzoEqkQ+Dm//Vxaa6SSKKlo\npAqnFUG2iMaysG0bEa/vIqGrS4gfM6FUKpU13cQ3g7/927/lrrvuYnraGGBOT08zOzvL7t27mZ2d\nZWpqatPHzNJaHvAy4NWYaOUc8Kda63M3Ul7/RKG1ZmFhgTNnzlAoFDbkNNxCqxdlww6+fUSy1QFa\n10KtVqPZbPLMM89w5MiRNWs+Wil01ESJMro8hh3U0bZN0vqia8UXv6p4iFFaC/qPXtsgXVkGNInt\n4ghhXIOBSDsgJflDN1H9oak5aUEWpSgsz0WUhijesh/PjVipdrrjdUY8QeigwtW7Xi2EmZxbLhP0\nLf4An1r4Ir9Q/lmzW08S3kV2H63NSN1KncQyaRtvehSnmM/6KlKEY6P7ZrujNbLeJK01ye+fwnYd\nksUKYrRMfu80WoCMYtKVqjlGJhJwJ0ewPdekzPogXAd3pIyV99u7Vq1S4stXTaTheXz2RV0zM4Tg\nL1eavBZNbveEMbSsBtg5F7vk47iO6XSPFUnNjAK2cg5uKWdqNC2ekZ2iezeGju5i8v578J1OxPT/\nHAE4b34J6uZ979oSCCFwbAds8EdH0VnVpuVuLaU0o6ObQdbiYwimRTbt3fqAOsq1ZMM7jZ2SDH/y\nk59sp7sA3vCGN/DRj36Uhx56aEs+Xl3W9W8CHgS+DnwBMwb4D4QQv621/of1jrFd3CCUTaLbsr5U\nKnH77bdT6NrhbgQbJZRuM8rrSST1ep3Tp08TxzGe53HPPfes/4C4CZaDlYQkpTGs+YvoXQfRlrmc\nbCV5iDKd/IvgvZ8PeddNV7BsB79YQqVd6q5miLN7NypOUdlQLGHb+Lv2YBGDEljDYwhX0tS5nhqF\nQoBfJjj17MBT1bpFaU6PQ7E3NYZKLMILC3yi8WnsvG/IYHqUZKmKTlpEIfCmx7Acm/DSAjGGFFuv\nzR0t4wwVwQJZb/LWOx8wqqxajY/9w8faz5dkxpat9JYQAndkCG9qFKU1SaVu0lvtWeWtuwt0mhIv\nVgyp5DyiWROROkMlnKGiGVLVaEDXIveGvdNGSLBQhTQBbXpZdDMi7UonWr5HbrSISg25ecMFZBBn\n6bM+S5fxEtMvOUZp/wQUctBcu1cucXzW8t/uTmu1CKOHEHJFZKOKUoo0TZFSttNmyfkzxLuO9vh4\nPdcRSrVa3XZdtNFo8MUvfpE/+qM/av/toYce4hd+4Rf44Ac/yIEDB/jUpz61zhEGovWBnQA+orX+\nRPb7l4QQ7wdeCNwglB83Wimvubk5zp49y/Dw8KYs6/txLcfh54pIGo0Gp0+fJgxDDh8+zPj4ON/8\n5jevra0P6yT+EG5lkWh8D/naMtGeDgnZSiKoo7tStv/uZaB+FCL8PEa9mKWqEMiVCkPHD7P0g7N4\nu6ZxS3mssIGVxsRKUxgeIs0sO9Kor+cGi2B2talgN5zpacKzxrHb3zuJsH3qT18yEY0QxHN9/Q1C\n4I6W8XZNkMYRuh4QXem+j2gTQ7JcM55Zjs3bX9YZBsbQEO6uCdKlCjpOeh+Lxh0fwsr5NE9fNO+D\n1tiOizs2jF3MmcW02iCt1LAKOcRwmXh2gRaZCYG5vWqcjj+mP8kv3f12o7aan8caGUHFCbZwkPXA\nGGZKiW6NY1YSf6RAGiuCK71GlwAP3P9b5rWkKZ/50YcZPraHkdv2taNKgtUOy92Qlo0Qa9RDotVR\nTw/8HHbYWEUSWmucJKAWRSwtLdFoGEl2EAScP3+ecuZI3N+EuNPY7iwUgGKxyOJib7Q1Pj7Oo48+\nuq3jZrgA3CKEmARCrXUN4zQ8uCC2g7hBKBvAwsICTz/9NKOjo7zwhS8kl8td+0HrYK3u9u0QyWYa\nrJrNZtvypUUkrce2yHPdwWJhg9AtU+QqQqYwOoXu6kMhiXjiJ1Pu/m9LwBAjhNhaomSKKI/0PE+C\ni7d/LzpVFIY8SGJomNRVOjJFPlpGuR6tzZesN9ozNwCioV2IYAFnJDVF9L71TXgu8UKV/IFdaMsm\nnmsSXhhUa+nAmxpBWBaNH50zNQvLjN71pkaxCz5ISbzcQNabWJ6Dv2+SZLHSO/cEiC4tmHTW6BDO\nUMH0FaYSO+cQXpqDld7vt1aKeHEFsnVGODbFI/tRUYzlusg9k8acMox73gMB4Dh8euHzJFcW0GlK\nl/ELVs7DGSoifA8lpXEUUJpo9ipoC/J5ExlJCVLywN2/2umid13+0x1385EjfUuFVohCGd3cwhoV\nNk2tTQ2WqAvLGugSK4TATWNu2j1thrFhvjOPPfYYhUKBarXK5cuXCcMQ27Z7Ot1LpdKObcoqlcqO\nTEjdaWitW2/o+4D/AnwcmBFCvAb4DCb9dV1xg1A2gHw+v6XZJ2uhP0LZbkTS6h+5Vh45CALOnDlD\nrVbj8OHDAy1fWsdaVzCQEUrBz2M3q8ST+wCzlidxTP3KZcoIvvcK8Dwjf03nVsyONpN8iszbSgmH\n4kiO4PxlQyYZxMQuciokdXMgzEKttCZdnEdo83jluFiVRSxVxy5odKmMyOV5vfsaEBF/Jb+NlBbJ\nxVnCSytEc1VAYHku9lABO59DuKZ+o+LEzEV3XeP31UarW1wTz3dqN3YhR+HIPnSagjZ1jhNf/gCf\ne+U7aQ3xaj08WaqipcQdLZPMLyJzHv70JMKxkM2AeKlq7ti128/tn0ZW64RtEUGnSdSfGMIuFkyT\nYiNA+B5pvUl0YTZTN7c8UzKFWRijwsiozEaNmADALpuUmbAttJTIetMYcbau8+zS+B67QCyZKKcb\nng9rEIqlZOZvPxgiX0Q3VtvFAJ0oai1UlmByj3merNbSKmy30DKJbDQaXL16lbNnz7bTzN0k021/\nv1Fsd7jW9UKrhqK1ngN+WgjxQmA38LvAxS7CuW64QSgbQKlU2vYc+G60CGWnUlut461FKN22+Ncy\nobxWt7zWGiklLgFxYRg7iUldnyRJiKMIx3WZHC6RdvUtaKXQ1RWE10XIQqA1+CNFdHUZ3agb63jA\n2rMPJ6gihEUqXGxtelRCfGyt2guVHp3CunAOy3KwsiLvv3Bfny28Hj+lXsZH//79JDWNZVvtAriK\nE9RChQSTKrN8D3/3GMGzc9g5D3/3BFbORaeSaLmKqgXtcoJTzuOOjxBdmqc50zuj5YFCjl956tMI\nx0YGIbEAt1zAGxsivDRPVDcpGhVERBc7PQaW52KPjeJkdRy0JL7cX5w3smW0Jl2pk67UcceGsTyH\n5OoSzugw1kjZyH7jhLRaR2ZuBQrw90ySzC+RXpwzn70QyFrdNFzqTk0ot2+ab7wS/vm3O9fHd95+\nF2gJUQxBExE0oF5dd9aEk4Tg5ldFjG2sZ8ESrp8S05VFREYoaz7/GiaRcRy3TSIvXLhAs9lEa00+\nn2+TTKlUWjdtVqlUuKnb9fmfDh4UQvwJpijvAVeBBkZCPCaE+P71nodyg1A2gJ3Ox7ZG7Z49e3ZH\naiRrdbhHUcSZM2dYXl7m0KFDG7LFv6afVxKSuEW8uElSyBMmkqTRwLHNHAohBNZK1NsOkcboMEAU\nOlJLYdlw8BBefYFwfg6dkYm9/wBuqYgOa4SpbSYMWqYgK9PO5ar8PHoxcyVGYLfkwd2vz3FwJveB\nVUE1mtlirTru6wJy+3eRLFUJzs0CAhlEyKAjXZXZCGB/7wSW6xj58JWFtpNvd1Qhm6FZxAF3bIjC\ngV0mgrEE7uSoKcy3Pb+6hAVxjIgTtOsQX1kAIfjX4y+jcms2l0YpPv6tjt2LlfdwJ0aJLnaaKJPF\nlVX1DKeQx9sziZYKnaRQLhoPNaXaRINSpl0l5zP+up9k/FX3Yrsu373V7NqF66OTyESJuTzk8ujR\nzDbIcSE8BlGEaDZMtFKvoGsVLK0hVzS1lkFY7zqMAnPstSKVaqf2sBkFrBAC3/fxfb/H+khrTbPZ\npNForBri1R3NtNJm/1QjFAyJKOB24BZM34mP6UUdAV4FrF9w3CZuEMpziFZEcvbsWQqFwo4V2/tJ\nII5jzp49y+LiIgcPHuTWW2/d1HyVdQklbBDYJXR6lWZd4QtNoVDA6jq+iMNsM505zzZraKURbhah\n2A7ceheWlsgrZ5HNAHJl7JsOUpgYIlqpkUpNLAUCK/PyEuZ+GfTIFJyb6fwuBELrjvdVJjs+9H8+\nZJ5yZYX4wmXiuUWiK0tE80uoIKL+1FlkI8gW417frwdf+qumR0VrPvqlD2RNmIAAb2wIe7iMlinJ\nUtUUvgF3chh3KE80u0B0oS8d5NhGMVYooMKQZLGCXSpgl3zCS1dR1Xq7Y79yS9cANstCeB7OxAhO\nwUfHifEZ61lMNd3qM3uoiFMqEMxcoEW0OvNWc0eGsEt5YzCZpORu3s/kT78Uq1RAKkUchJmqSiCj\nGN8W2Jkaq6dzUiZm4XdcdLEETLfPJbl6FXI5RKOKrtegvgKNzvuxrgWLECYlVh+89nV7eu2EwksI\n0VaNdfd+SCnbTsRXr17l3LlzvOtd7yKKIs6fP8/KygoveMELuPPOOzf1fCsrK/zar/0aP/jBDxBC\n8KEPfYhjx45t2xhSa/2B7PV8SGt9uu815rXW11BDbB83COU5QH9q69ixY9RqtR0rErZIIEkSzp07\nx/z8PAcPrj86eC1cK+VVW5xnNhlmLEnIj42T072+S5Y2PlUC0S4c96S7/Dz6trvMYrO8QHplDiwH\n56abyY+X0Uoj6zVi5WXnAq3dvFoxSitdKKMWelNCGoGOE/7r0v/L/zL587zpAJw4caRzXqPDFAsO\nxWOrUxVpMyK+NE90ZZH4yiLh7ALR3BK0+oqE4O2v/O/48Bf+oN1bES9VYbnWXtTzN+8G20IoZZoZ\ntV5tFJBKoygTy7hTY0YyHMYgLLzdE+hGQJoRE2naY2niTY6SVmsEXakwy3dxRspYvmdm1Icxab2B\nPzlGdPkqUXswWSddhlakKzXSlRq5/dPs+aWfonj8KHpVQ6H57BINOk2IsxQtdDUj2ja24yK0ot8W\nwS0WEcNj6HJXA6CSEIaIoIkOGuDloLbcUztrw1knJVZZbAtHrqdk2LZthoaGepoYv/71r/Pggw/y\nile8gsXFRR555JFNE8q73/1uTpw4wZ//+Z8TxzHNZpPf+Z3f2bYxZNYNbwNfA/YJIWyttcxMIb8C\n3LepA24BNwhlA9hqymutGsny8jLLy4NH124FlmVx/vx56vU6Bw4c4EUvetGmiaSFtSKUlZUVTp06\nxZSrKI1NUShNkFg+pH2E0vVYjdmJ6iiCXBFRHkHdckfb+l0szpHWatiHbsEfK5kFSwhEokzfSBuC\nBLedPtEjk3D6ZM/zJlGKs/cAkw88wB97q4k60h44XraPNwtsazStLoNz627yt0qKpFhaYukUPnSR\ndmrKEZSO30xwbhbZCNEosC38fVPoICQ4c7knjSMcG3dyFCvvg1KmphGE5PdOIYOQZG6pfWy9YGaV\nWEJgFfO4o0N8+vLf8PN7Xg+Ow/DJk0TLWc2lK82moqQtedYCcvt3Y4URaaVuDDBt03iZ1hq8Hi8x\nOQAAIABJREFU5ZY3G4KKIh45+QiTP30/oy+92yiqBm4gjPWHJ7RxAG5HQ9o0IiqFTFPqQYwrzOJu\n252udxiQjrJsKBTRhSIwaX6XqSHPMEAEdXS9hqhX0OupDKPQpMVyhee8B0UIQbPZ5HWve92W6iiV\nSoWvf/3rfOQjHwGMD5jneTthDGlhuuNfDEghxBuznxVMumv7rf0bwA1CuQ64VrH9Wn0oG0Wapjz7\n7LNcvnyZqampbRFJC/0RSrVa5dSpUwghuPXWW6G+QsOyQAmkWv1ctsx2m9nCp+MArS3E9D7Uzbfy\ngj/ptokRfHf/FPb0BGTd0akUBGnvZakVxGQjbYdGe6MTv4AcmUD/5OsYPzREfxVY47CQjHKpUWak\nIEnl4IWq6EnCpP+2i7R23m/aazP9a/8zPiHy8iWi088Snr9C9alzhHNLWTTWet8MKYazSwg0dt7H\n2zOOcF2UVFj5PO6kRby4guhbzFUjIIpTcnsneOTkp0irdZyRMs7eaRNkxCnJ0kpPh743PY5OU6KL\nV9qzU7p9yB448FMd5VY+x3tf+CAfeXmXrYdM6U6X9UIgbAfdbkQVWJaNZdnguORKpsaitUYpiVTK\ndLwrRW1lCce2O0Rj21lqNJOo27ZxXnZdcF0TzUxln6CwzAaiWwRQWzGSYzAd8z8GQoHtyYbPnj3L\n5OQkDz74IE8++SR33303v//7v78TxpAWZubJIUyd5GcwJJIHloHf2tIJbxI3CGUD2GiEslHV1nYJ\nRUrJ+fPnuXz5Mvv27ePgwYP4vr9tMoFOhFKv15mZmSFN07YNi04TrtZ8HCvrkxuwsRXSLDwCzK62\nUYObbkXdfIRew0cNjOLv24PqWhAawepLUmpBGmbHHRlHn3waMTyCGJ8iPXw74gUvNmVuaxmhWqkb\nixU5wvnaiOmmh8zPZTASZdFPRp94+/0EQUCp4CK1w8XWmjp5E2LyXkovCZmyA7y0iTz/LPHJGZrP\nPEvj1EWiy1exRss4Qzni2QV+96zmO696K2jNR77yAWOZb1s4E2NYOReVphAlOKMlkoUVwi4VWLJU\nNTUTaEuBnZES7vgw2Baq3swUXWu8uNaUwsxD7Mn++2lt6lpr1c7WGaPRIlEToTjYNmhHGeVULofC\nQsmki2h0Zq1ioxOJZ+mMEPo/G22aNLtFAGC6/qMQbVtmrM1zbLsCpo9rs+4YLaRpyne+8x3e9773\nce+99/Lud7971SyVLRpDpsBfA38thDjcX0N5rnCDUDaI9QwiNyv/3SqhSCm5cOECly5dYu/evdx3\n333Yts2FCxc2Pwt+DSilOHfuHABHjx7tGZmqoiYNkWdIB2jh4rD6Oa0kW9CFAC1Jp2+GvTcDg5qq\nFaLHrtyi3pCrLDukAlmtwPgUKoxwb7oJPbWP6CU/1WNlnggPj5iGLnOuNkaieneutdAi7w2uD0WJ\nIOdq+q25AOLUwnN7CVRjUZMFmrqAVOOwbz/sfQn+q1PGrAAvWEaePk3y9I9onjzDd6ZPtFNiv3L/\nO/nQF/8vhAQ5u4AzPoRTzJHWTErPHR/FHZEklZoRDPTBLhdwygXTo9L1plqFnLFi8Vzj1dUIsYs5\n/ttLJT/5RGeB+u7bB5gOWusQynoYEN202y2Fhe24WKJ/odFIqYg1JElKFEUm5WcJU5dp1Wgct8cu\nB2iLALANyT3XEUqrLrjVzdu+ffvYt28f9957LwA/93M/x8MPP7wjxpAAmTFkWQjx28AoEGI+nKe0\n1n+ypYNuAjcIZRvYah/JZglFKcWFCxe4ePEiu3fv5t577+3ZlTmOQxBsT8DR6lW5evUqU1NTA5Vh\nkRRoYYGShCpHzgpWzdEQsampmAhFoPYcQWhJmiZEUcTbRuFPlwESnr73AtgtQ01BKF3scPWY11Tb\nUF3G3bsXNT9L8s9eQ/KCl67a09ZSn4X6KIF0+Y0/+3z77x9482sBE+l4jiZOV+/+hDBrVDehvPWj\nX27//74JeNfr71/1OKnAcyBOzUEi5RIpF9whuPUA1m33m3Duo1/NHmGUVvm9U1i2QKOJZheIV0wE\nEi2uEC2umF0q4AyXcLL+EhkEuOUi0ew80aXVc0VUMyTOZqQ4I2Um3/ByRl58J8Ky+O7zoBkneGus\ng71C5l7oNTrazY3akFH3fbTulJMGRjcC27bJWza0Hak1SmuUVCglSZKUsBHg26IjAGgbRVptldiP\nexbKZrFr1y7279/PyZMnOXbsGI8++ijHjx/n+PHj2zWGbLkIvwiT3oqBvcCjwM8DZ7d0wpvEDULZ\nArbbkLiVFNquXbtWEUkLG5kFvxbiOObMmTMsLS1x6NAhyuUySqmB5xhKG7TAIqWWDFNyV3c6twhF\n2R7KUYTaRjer2LZFoVDgf/yZEr+VJOTPPgkp4OZApShsImnhxqs7rxvSpzBcRs1fJn7VA8hDzx/o\nIfX0hTwjI6vfn9/4s8+3SSUZQCYtROnay+q3FuBdazzOFmvVH8BzNGHSvagKM9zqf/sv+PU59Mkf\nYv3w+8RP/wg9t9DuCwGzG04qNRQaZ7hEvFwjrTVxRodxfA+NRjdDkkqtM3c+5zP+6vsYf/WLsfze\na7Ilih6EdeeQyCSLONd6A3ptVHrFzOv0iajUEE7WWGkJgeVYtJalguujkxCpFErKzChSmTRbs8nV\nakySmO73rYyD2Ap2gsDe97738cADDxDHMYcOHeLDH/4wSqmdMIbUwG0YA8gvAv9Ka/0fhRDngOPb\nOukN4gahbBAtieJzYdq4Wbv6rRBKv8T42LFjCCG4fPnymtFTMxXZBluxFPjssXufU2htFh+g6Q9T\nE2W8oEYpn88m92X3ixqAAstG2DZCKYLEJkxsBr1K1agjaksEb/pNk+qKIpLEPI9lWdiWRRBbXK14\nlIty3al/tchiKCdRA+opUgnyrhoYwYDZTA9KiUVrFPqhw3ufePvq6EaPTRPcs4vkzpehlMKLIvTJ\np3BO/QD9zI8gbGK7EF+eJ2mGCDQqUkberDtLteXYuGPDDP+z5zHxupfh5AdLbi3d1/jZ8+LXdg42\nhXm3qzDfd6sQvbTR7QW3XnSzquDfi1bqzLZsbMvuuUVrzVhumIuzV2g2m/zDPxgT3UKh0G5CLJVK\neJ63o43J1Wp127NQ7rzzTp544olVf9+mMWTrRWpgBVOQb33pdmOaHq87bhDKBnH58mVOnz59XYlE\na83s7Cznzp1jYmJiwyN9HcfZMKGkacr58+eZnZ1l//79q5Rha5GT1ppEaoTSaNvK6gm9u09Lp0il\nWNGu+ZnfxZGipG+5wQ6bnUIwkGgHjSBN1CpC0TJlpH6Js694F8WRKeJ6Hc/zKBaMl5XK5Ks/POcQ\nxiEzzybsnu4173z/L7y26zeBba23znUWn9dNw2ez2vgn3n4/jqUHqsRUd9qrD2/64Ffa/x8GPtBF\nLHHXsQQgymXEPfei77kXz9ZEtYBo5hnUyacRp55CP3sWIVM+j+YTL/lXhhyShM8EX2PiZ+7HmRqh\nkSS4zSST73YUVkJYWCgEa+/ihe2s7aO1bmF+nShEdkchmzvu2h+SKVqPlgtUG0MUCgWmpqZQStFs\nNqnX6ywvL3Px4kWiKMJxnB6SKRaLW44ydsJp+Hqgy6fr74Ai8H3gDUKIBDMX5T8+F+dxg1A2iGKx\nuKNEIrLF0LIstNZcuXKFs2fPMj4+zj333LOpOdVruRd3o1WHuXDhAvv27WsX9PvRmi/Rj0RmO2Kl\nSB2fste7q1RaEywtkgtC4iN3MxZe4XwyjGD1QCQrzgglcyhuW6JHvc11Wiv85hLVQ3dxzj3AbTJo\n27u0FjHLslgOCli2055CWy5q3vemV7YHNzWbDYzc1cw4r6LxPbunu7+FMLGwLYXW8LYT9/OmMMTP\nPot0nayQtU7aq4X+vu9W/SUZ8NFFqcApFRB33Al3mMY5HYWomRk+8cRsR7HmeYy889/hEOLqkFxq\nXAqUlEiputJEZlefSo3v6cGd762+kM2i7zEmg9Uuoqwbhax/3IT13lOdRD0qr5ZVSv/E1CRJqNfr\n1Ot1Ll26RKPRQClFPp/vIZqN2N7v1HCtnYYQoiURnsHUTyyt9b8WQrwr+/8WPoDN4wahbBAjIyM7\npqSCTmF+aWmJs2fPMjIysmVH4/VSXt3ps127dnHfffetK7O0bXtgp3ySKgQCW6Q00hzFbFqf1poo\nikilpGRb2DffjpsvIHUJZwDJSaWNcaDS7W5oJ9uk2lFv/cTXIfb0FD/Y9VoKhQK5vA39yjItOHWp\nlxjjBBy3T1eks4Y8qWiEChVXwHIRwpBMq+grLIu8A9GAr18qBY49OEqJ10l7rQfLalmm9EIIcG2N\nVF1RjJ/Dft7z4Yn5nnV2NpnofqUM2xGeHZIXIZ4O8VSIUClhGKBsF63kgM53GzS4qGxeTS/WLcxL\nmdVRWtdNn0vAOrUN3e9g3AOxbtSkk2hDNQ3XdRkdHe3pHdFaEwQB9XqdWq3G7Oxs2/a+37+r+/uy\nU+N/rwN+GjNAaxaT7qoKIVaACFBCiEe11rPX+yRuEMqPAS3H3scff3xHZqwMIpT+9NlGo6u1yClO\nFVLZFJ2ES40SE16FMIpIkwTP9ynlcngOLBT3UVZ1ms4orqqvOo5MFSJNwBJtybCdZUTcjFB0ZhHi\nlfNc9m4mLe/GERZhYpNze8/t9GWH/l3slUXBvl29y7Tpk7CzxcdlOO+SKFOMbhV9kzhBa0VUj7C9\nPHYWPXYfx7EG11GUAt/RWWG/g/60WT+SrC41aCOuup64W3H2ibff3/N7LwSR8KkkvdeTK1KUvsq4\nB0UnxVMhlopBqzbRxlFEqlrTEbMGRtvUqCzTyj+4MC8EwrLbhf3eCIW1i/lwjaZK0OtFTWm85SK5\nEIJCodBOl7UPmdne1+t15ubmOH36NFJKcrkc3/72t3n22WeRmUBgK/0vBw8epFwuY9s2juPwxBNP\nsLS0tG0fL2ABOA28BHgB8CMMubwG8+b+Y/b7dcUNQtkgdqKw15pD3xq3e/z4cSYmJq79wGugW4bc\nGlF85syZLUU9a3l5JammGduM+QnnFwX53ByFskWxVGovBZXSXjQWbtrkir4Jl0X6FwqRBAi1unD+\nom/kMUIUyVf3PoVVHka7CU9NvKK9HlUDm1wXJ0axxdXK6sUkTcERikSvvTNu9aSIVuqna3FQqoAt\nYlJpopowCECYlFkSg7BdE830XRODLpG3nbiftwE5VxMOiHqkAltI0gHnGqWm3vOWD68mj0/+yv3Z\n7JPVxxwULSXaIRCjpNJF6I5Ut2hHFN2QHIGJCFWEUInZ9GREGyUpWisSBb7TLeG1O+9BX9Nqj9v0\nZmXHXVhXzpxESJnuqGx4Ldv7MAy5ePEi3/zmN5mZmWlv0D772c/2OBdvBF/5yld6vvcPP/zwtn28\ntNZ/CyCEeCXwRq31M63bhBCf5lr52B3CDUJ5DqC1ZmlpiZmZGQqFAnfccQfnzp3bsS9Cqx6zsLDA\nzMwMpVJpy1HPehHK1cWYYqFCIvcwPTGEozumfqmVIxGuWSCkpC4KjBhdcM9xnDAzVOwy/7v185qO\nIMXmVZdGeOwuxSXnFhK3TGt+X6VpMT0s2saLTz27jpqrqcjl1yaUNJZcqSoakW08E7WJCJQWoAXj\nQx4CgYxDxkcL5HwbraUhmSQgViZ6EZnKzLJttDI79UFQ6+zSBQpY/TgBePbgB2oNvquJVtnFrCMS\n0Gbkst++TdCQORoyh7F7MlGWQlKyAgoixCfEVyGWisD2kHHQJeGN29GMihJ8m3ZNsCdtdo26TNuC\nZeDrXOeN0xqhrn8fihCCfD7P6173Ok6fPs3rX/96HnzwQYIg2Pb0VmDbPl7ZOTpZt/wocJsQ4gKm\ndtIADmDqKtcdNwhlg9hqhNIiklwux/Of/3yKmYPtTvl5gVGeNJtNLl++zB133LFlWwhYHaForZmb\nm2exqkgSH788jd/0sek994vBGKOlGFfFxE4JJHiil1CU1jhx0CaU1sAj+pRHOfaj5FlO7X6F6VVp\nrynCuBijWao6hIm15rZrbklw8z69Sh7sCMWFyynLNU0pDziDj1BpCLQWNIM8S4GJSMBBCM30iIfr\nQt7X2Jbxr5JKESYSlTbBzrdJxvy02tHGoHYPqddeELVeO8W13hVpajO993j3p/6eVhJyUPoNMuGY\nclhWZZYpd92iGBYBOT8mR4CnI3wVgkqNAlBplExJkhQp00xmb6IH0/XuZa7EA5917Rei1v+OOKjn\ntLGxWq0aTzvMJNfNQgjBa17zGoQQvOMd7+DXf/3Xd8LHCzrfkj8E/g0m7TUvhPgp4Gng0lYOulnc\nIJTrhOXlZWZmZnBdl+PHj69SnuwEoVQqFWZmZrAsi1wuxx133LGt40FvhLK4uGginvIwxeH97BGC\negp5J+1ZHKTlsxLnGSXG1xEr2W7XFb0LSJoqPJkAZgGKwwau6/L0axxu+4LEEIvF1158lfPidpTt\ng+xVfkWpjWdrTl66xiKiyc7R3M9C02yknLnUOad6AHsmJdVw9bESKRgraZp9BgRaC+Yr2WhdRfv4\nOV8zWtKU8zlsyxhFpkqTJAkyMgOtHJGAU2iTTUtlprCwkfRHc+b1CiwxmADWEwKkA3ppuitab/3o\nlwcecy3zTLCopUUquojZBANa49spRSsgLwJyGJJJglpb4NCKZppBjCd0n72KjWWJ9ZsfpTRF/TUi\nFVds3QZlK9juPPlvfOMb7N27l/n5eV796le3yamFrfh4QUc2rLX+MyHEtzBF+ingD4DP6fXVDzuG\nG4SyQWz0Q65UKpw6dQrbtrn11lspl8sD77cdQqnX65w6dQopJUePHmV4eJhvfvObWzpWP2zbJooi\nHn/8cTzP4/nPfz7C8Tl/VVHymlxuDDGU642eLwXjlH1DQk4asSTLeFaadZB3oKWENDapJaU6EmDg\nyVt/xOLk8/DGS6jzFzm958XZo3q7WGqhTaXWcaxdD0sVGBoCm5SZZyXhgKC/1lBGoTQAcbKWdxuM\nlDQr9c45hJFgNhJcEVbbwsWyNKNlGMorfE9j6YRUmQbZOEnQXY4EmhhsMwOm/5W5ziZTWxiJs2tr\nkk2qz1K1dgOn0uB2y5y7rWYYMik4KYh1kyErYcSX5ESIq0JGhEYkzcxeRaKUIkmM0kzqJp7T6Znp\nMYsU68uOfWtn6psbxXanNe7duxeAqakp3vjGN/LYY4/tiI+XEOJXtdYfFEL8PHAB+AugCgTPFZnA\nDULZMVSrVWZmzATBW2655ZrSQsdxiKL+oUbro9lsMjMzQxiGHD16dNVOqTV0aKtoEVUURbzgBS9o\nv4aVRkK1abO3KFkOPQ6PdCxXlOUxszTErbuNsWGKgxYWeat3TkqaphCHiDQG1+vJPWthY+3aT3n3\nOPrKBc4O3W0KtbSWlc7CvlKRLNU2tiNdqQNpyKXFtd+TWhN2T0pqA6KUemTh2zF6QJNxFA/WJmkN\npbwhG6UEixVYzIQDGoecq/FcGC5qCr7GthRhFKK0TRzH7XRjK1VmWTaevTaBrtf/YluQrFEPf8O+\nwX8H1mzgBGMz8+HPfYXPzsEvT8OJE50op/UIpS0aukSYdN5T11bYVkrRCsl7IT6Bqc3IBLRCZkTT\naxZp3gOUNpGIWP0++M+xjdd2IpRW/0u5XKbRaPCFL3yB3/7t3+YNb3jDtny8wjAEOJQN2PpPmLB5\nCDP+NyeEqGqtd2/ppDeJG4SyTdRqNWZmZlBKceTIkQ3vXhzHodFYY952H8Iw5PTp09RqNY4cOcL4\n+Pgq4milqrYiZQyCgNOnT9NoNDh69ChBEPQQYpIqrlY99hQt0IKC09nqz4ZjptBrSyyVEghTv8lZ\nHfO+MDTEWUiaxp7F61Wdab+InhzBSRVhEHF+8oW9J5jxiSU0T35vmcOHhgmlMM32ttWeotgPn5D5\nuQic4sDbW2g0jQ3MIAihBiZkglhQzmvqwernDtYgGwHkPVhpCGrN1q02cWKK4fumbDzX+H9pjAAg\nTVOWoxS07pLxmp+WsNaNQPrHC/zfP38f+ULBFPsdPTCygfUFBP/Tx7/CqSwN+CdzcKLrtp4UXN/1\nGUsLW3iEqUf3rCcLRckJKNohvjZ9M7aMQMt2NBOmKUr2Wt/bLbK1xLY3UpvBdghlbm6ON77xjYDZ\nYL31rW/lxIkT/MRP/MS2fLxyuRxa6/+Q/fq8LZ3cDuEGoWwQ/Rdsa15IkiQcOXJk0xfZRlJeURRx\n5swZVlZWOHToEMePH1/zi7MVQmkZQy4vL3P48GEmJycHHj9OzMLZsoPP2QloUMLl5JKplzi2xE9C\nZqWZK+6oppHchmEmW7awGtlwrS6Fl3Y84qkDOFEVp7HI7PDtvYtRl240qDao1RKeeWaWsd0TgCCK\nFKDbkwJNEdjGdzUnn6qQpnDoljxBvHZUU23A7klFLVx9n2bsUy4q1ABZr73GGhatQzaNWKzSwmqt\niaVFtSmoNFrP4yAszWgRygVNMS/bUl6pZHuIlRACx1II229HNK1njVOBY3eIpZsn1hMJrHfbqb6a\n0uc+9+V2lKKyNFswYPyxYHDqTmFRlUViu9AZcKY1OTuhZAemOZMEVzayWTumh8ukzExE9+Q/PI7t\nb77rfSuo1+ur6qEbxaFDh3jyySdX/X18fHxbPl5Xrlxh9+7d/yvGUbiB8fJazn5WgWWt9Won1+uA\nG4SyCQghqNfrnD59mjAMOXLkSM+8kM1gPUJJkoSzZ8+ysLDAwYMHB1rJ92MzBpFpmnLu3Dnm5ua4\n+eab28aQayFIBGOFhOXIpKk8YQhlPh5DZyoqgURri0jZRHFILZpj2DLNY0II4igx6a7uGoEQLOx6\nIWWySK1ZozFxCHqc2U3R1tYJ33psDqUUmhyHyz5B0lJ9mZqMUpI0lSgVc/XCVao1hW3ZNJZXID+y\nbvG2GcjBvlIC8p6kEa3+qlSaorem0AVrjbczSWCooKk2V9+hFlhYXQu5VoKlGizVBOW8oBYYJirl\nspRZTuM4CqFiYqlJ4gSZDRhrmWbaaDQdg8SOIYqRJIdq9Xmsd9sRH2a6MrXdKS/oKkcNuJ7Wu4J7\nLl0hCJVHqDyMA5o5nCUkJSuk4AbkCHFUQFpb5o7jhwlxV3W976SHF3QkzD8Ou/z1kKXO92Ps6ieA\nEsaGxct+/ybws8/FudwglE3gqaeeolqttolkOzugQYTSGul75coVbrrpJu67774NK1g2EvF0D+ga\nZAw5CKlUNCObkULC1bpJHdkkaMvhR4smOsl7xitqqZbQkE18z2PX+BDhStfKkyYmX251Lrnq1HGq\n7ijD8TKETWRxlGI530comjRJ+P5TsziO2+72l3EEwhCcwNQcbMvCdSDvJlw6myOXUyipuHApZO+B\nOkFqGvHaKqvM2wsElTrsHk8IU9ukkbo+23poryk0KuYUK/XV72F1HbJZq7VCKRgpalYaq6+raiDw\nXYgTQT2EethJmfmugwJGS1DKKXxXQ5YyC7u631Vmk9NSWel1JliulfV6z1vu5yuPfpk/vjhYeSbV\n2n3v3em53/urL/N41+ibj7/9/jWjIjCRT5zaVGSRijTXodKaSAQcSDymh1nV9d7y8Go0Gj0eXv2O\nxL7vb+q7/FyKADaCAwcOoLX+xR/3ecANQtkUDh48uGOhdDcB9E9i3Mps+I36eQ0a0DUIrbx0kmqW\n6zYjfkrctBBCY6mUq3ICqS20Bo8azVqVQBTayi1bxe01WWuNLSNjuZJ1yEdjN3G1sB+hNULG0Kgg\np48yVtKcyWT4cRybAm3cpNbM47qd9/3ChQbTN61uKhNCc/5c1bj3Cqs9X6PgODh+ocvTy6SNwlDi\niISwVmfxgiaUOTzPZnTEo1R0UHEdyxuilNM0E6fHWwugGVlrtnOvSTZtclj9mHitKbxAwVfEyerj\nRYmglNfMLwvm2w2SDgjNcNFntKjI+VCrLmHnfKRSpElCEBgnYFOT6fTMCMS6c2Puf9X9vILBrzlO\nhWlSHPBw2aUge7xvjprAkEY0ICqCNezAtAZhUYtspgdMD13Lw6vlSFypVLh06dKGHYnXmhP040Y2\nQdLGrOfHMDNRXCDANDSe7O6cv564QSibQLFYHGhLshU4jkOSJFy4cIHz589veKFfC2v5ec3NzXHm\nzBnGx8c37OfVcvMVQhCnivkVl+Fd5hs95MdoYfH0whhJYqYwTudDivkc9WTSRAuodkOaWcDBUilC\npeDlSYujXBoy+ntPpJCmSNvDKpYZK5rZ41Ec4Tgu+ZzHkz9cRPQprRpNRcFJCdLe1+PqkEpldVjw\n7MWAQ0fzBHHH00sIKNghz5ysEsUCKRXDwxG1GlSrjXaTp+umTIx61JqCcslmbNSjWHRxPYdU2wzl\nNdXm6hWvEQ4mG0MOmniAFLgRCAo5TRCtvq0ZWWvu/u2BC66JvIS2WGkIGs0ixYpP3jey52JO4zkS\npXS7Z0ZFqj3i1rUVWH5PzwyYdXzNyZcAKsJkXVZjPQXZhlNirfPILF6CRBi58wb2YEIIisUixWKR\n6enp9t8HORJrrSkUzCYpiiLiON5y/eR6InOSlkKInwR+FXg5pnZSB+7AzIa7QSj/f0XLrr5erxOG\n4Y7Y4vfPRFlcXOTUqVOUy2XuuuuuTVlEtMjJsiwaoSBRgjgryJe9hCtBgZVagGWZKYxT4xqpO+mt\nvB0jJLSWCKUUTmrSZMLLc3niTjNKGPBJ0M0qsjhMrASnvvMYtr6dQqGMJSxqC4s0ArPD7ke9GmIX\nOu+b5yhO/nCN2qMGGYZgme7moie5cmmFc8tx9ppN+mts2G9PNVZaEwYBSimuXA0o+Alz8y4Li3aX\nAMBiYtTC9z3KQw6e76Itmyi1SNK16yWNcHVxHkymzXcgGKAojxMYyuusltKLWiB66i89j+vhV0EQ\n0Sas4aJFpWFclEdLUMorcp5GIEEnxGlvz0wrknEEaNzBJDCgKN9CS0HW3f3f0kWup1hPAnKtAAAg\nAElEQVSLU9FraEzHhPKtH+m4CHz/P7x0zWOsh0HRjFKKIAio1Wr83d/9HR/84Ac5c+YMr3rVq7jj\njjt4xzvewbFjxzb1PFJK7rnnHvbu3ctf//Vfc/bsWd7ylrewuLjI3Xffzcc+9rFNja6AHnuafwl8\nDvg8MK21/s9CiP/E+vacO4obhLIJbDfc7Y4YxsbGKBQKHD16dEfOrTUTpVKp8Mwzz+B53pZtWFqE\n4rou1WzxasbG1j6ozDMTjJPLdaYwerakGXVW/Lwwi7QQ2Zc+TRAyQXsF5nfdRWJ17uvoCPn/sffm\ncZLc9Xn/+1tVfR9zz86xu1rtMXtJ2l0dgEAmEraJlBArsjAoPwzGQBwwCAlIkDhisLEtZA6JX16x\nZWKwMZhDEbHNIUsBCQwCBAQdBKTdndmd+z777q7rmz+qq6a6p3t2pmdmR8lrHr1Gs9NHdXV11/ep\nz/E8HxRMKRgam+D48eOE5xOMzCkEFJtfPJ+rezqMjOY5eDSOZTuF+exituaVrIvh0TyH+oIUcwWe\nO7ty1DDA5IxOPBGipDu6B0VRCASDqIrCrrY40/PWsmmiXsK2bHI5QTwKpVHNSx9pCrS2BlFaAoSC\nGkLT0C3VWxBXI5tMUTgzqWqQQ71iv207hfpUjfpLruhEPbmVY+g9IjItwWwKn9mmRjAQJByEpmhZ\nM6M6HWaWJcnmS1jSNzWz3GGmKgoo9RdE3RTe96K6BmOtIqp07fwrUmJS8qYv/6Dua20UiqJ40czr\nXvc6LrvsMh544AHuu+8+fv7zn3s2SuvBpz71KY4ePUo67Vz43HXXXbzrXe/itttu461vfSuf+cxn\neNvb3tboLgdwzIq6cYrx4KjlFxvd4HqxQygXAX6XYX/EsFnqdnAK+uPj44TDYQ4fPryhmQ1+P690\nXhBULWaWDGxbx2jqJCArw35hm+StsJezCCmuTkUgpY1i6RgEyPccI6ct75e0bYrzE2iBEDIU5cqj\nTgt9W0kyMgdjw6m6BWwoD6gSOh/6wfe92+LAjS2X13x8awKGz45jyvoLnm1LmmMwXUNVP7doEQoq\nTqrKl1+XQEvSYnreLqeNSkgpyeYKjI2rxKMKerkhoDmp0dIcJBrTkJZKQNUoysp0j2XVJ4dUQdRd\ndOuJGN2op/b7rf9auiEIqJLR2WXjTggQDkqa4yESEYmmVmpmdMsZr1zIp1G0UIVmxkWwjvIf1pcS\nW9WyZQvgDtfq6OjgV3/1V9f9/LGxMb75zW/ygQ98gE9+8pNIKXn88cf54he/CDjGkB/+8IfXTSi+\nWs9jOJ5dzwMfEkI8jrPGP7TunW0QO4SyDjQSobjmkJFIpGbEsFFRVqFQYGBggKWlJVpbWzl+fOO6\nJjdCKZV0hidNSqUi0aY2ApqGqVReNgsh0aUKYnmBDcrlCMWUCpptU2g9QCpcFutKSUnXMQyD1qCC\nIgJoieVUQ2vcRpMWw6MFEKsvHDPTlcKIlRNYnIJuc9Tk3EAKISARt4jFI+QKtbc7M6+jBoJYVuX9\nlg2dScHUfOXtAkjnVcJhFcvymWGWi/9BTSeVcYc6KUzPKuUajiAR1ygUSnS2h2ltixEIakhFLRNG\nje+FhHi4dieYU3+pnS7LFt1wcSWqBZB+BNSVVZuiLphaEBS9jjTNsZmJQyJiUyzO09mexLBYoZlR\nVBVpWthKqEIz4729VTiiOiUmJdzWCl9eqPOETcZGpzXeeeed/Nmf/RmZjBMdz8/P09zc7NVNd+/e\nzfj4hjwcfwzkpJRZIcRbgGuAZ6WUDblNNoIdQtkiuJ5emqbVNIeE5bpHI4V4Xdc5d+4cS0tLHDx4\nkM7OTi+M3iiEEIyOjjK/VCBvnKKnI0rR1kCs9I1qiliYVuX+q2VCkQjSVpxYQJJJ9IIEQ9cp6TrB\nQIB4JIKWV0BaaLFlh4FIEBZn3JSUqJnykuX/JqbKFi+m81B0mNRLaAFBQBPEIoKmsM7k5HLXWSZr\nYFmSZEuMdM5ZTb+WOwM4jfs3cpjdbYLZxZUvnMrWXn1NE3a1qUzPL4cJrtjSlAHicScacIerWbaF\nbdnohQyZrEa+YDE2udzSq6mCnu4g4XCAcFgDVaVkqUgbikYdJb5wLO1rFfRNC0JqEWfceCUyeUEo\nWLvrLFNUllOXVfA7Ati2YD4N82mVXD5OzgyRL0Ei4tPMlFNmpmVjmSs1M4qqIKXqyIFqvN6KlJiU\n3HTTr/Ab5VHSu1ut2k/cJKTT6YYJ5Rvf+AadnZ1cddVVnlX9ZsF3UfpWQAghvgWcllL+z019oTVg\nh1DWgbVEEplMhv7+fqSUF/T0cluH10MopmkyODjIzMwMl156qSd6nJ+f3/CIYikl4+PjzMzM0NXV\nxd6DJxk9HcAud2wFFXuFw20ibKJble2VarlAP2V30CTz5BNd6IZBqei0Z7qtxcI2UaUJQkULL1uB\nCwG/9coEX3lE8txAlTS7vJ/SBr2oMzy05DRHdgACXsRBDNPGMMEQBscvCfCGV3cT0ATZnMX0nMHM\nnM7MvMFiyiQeDfDA0C+9bbuZrkzGANRy4Xx5kcoXJR2tKnOLK4mlUKq9mJV06GxVmVmwHGW7pqGV\nTz01GCEQKCKEM0fE0HWK5c8xICSpfHDZAl4RtDQ7KbN4XAOhYUi1IkWULdQu9q+2zApRvyXZsurX\nekqGIB6WToNB1evkioJICDIFfE0ETsosGJB0NDmz7UPasmbGNC10S0daRYRWNQKg3DfsT4k5U3T8\neiGFlujmjemuxtLSUsMD8X7wgx/wta99jYcffphisUg6neaOO+5gaWnJWwPGxsY888j1wLcuPQS8\nHrgfeL48WOvbUsqVJ9EWYYdQNgm5XI6BgQF0XV+zFct6HIcvJEpcj1K+GlJKZmdnGRgYoK2tjZ6e\nHtrb25nIagTUcmeOAGwLy67sRgsHTIqG5mW8HF2JQV5rZdLqpFV7nom0QThqOKp53z47A7okMhRd\nobsJaIL/718l+e+PGPzkFyVvPyXQ1Sq49kQYhSD95zWuHo0xOVNiMW0hlRJoQY5cInj9v+2gtTmA\nlCaWpRCLKuzfG+Lgvso5Fg/8SeXx2NcbZHrOpK0lwPjMyuNl1zG7SmdtmhMKS5mVZGNa9cmmOW6T\nKQbQqnQPum2hKDaGaWBbdllDoTA5rbCrVWEho6EIQTym0NoSJB4PoGkakbBCurAynVSywoQ1J5qq\nRkEXqwzirY9AHdG4EBAOSop67dpMrigZn1tuABCKU7tKRiWxWAQpnLHErmbGtm0QAl2YoC2PZ/YL\nUHMlx99tq6Qi6XSaAwcONPTce+65h3vuuQeA7373u3z84x/n7/7u7/it3/otHnroIW677baGjCH9\nkFL+ErgbQAjxGuBO4G+EEEeklLMNb3gd2CGUDcJvrOgaN64VayGUalHiS17ykprWD40SyuLiImfP\nniUWi3nNAq7Z5XxWoGGiqCrYzjz46hXHMm1nSl/577BiYClRfpntQTOXyIXiRCOthMIrr361ciSj\nRGpb/CuK4KbrgmTSMwxOh+np1HjpyRCXH1oWl544VhkBSilJZ02aEoGyHYvt9ul7x8f97c6eOBCF\nc74OqH//Wy1IKcnkbH7y9CyWaCWVDzA1Z7GUsZlbtEjGVbL5SpJ4cPZZErNwU8eJFe9lfskiGVPI\n5FcSiy1rL+SWrdLTGWRmYflzdd/T7KKJbRuYpiSfF8wvLLcyd7ZqpHOSttYgyWSAUEjDFgoFAfGQ\nZKmGfqSkr4w2XKTzAq0OEWWr2599NcFCqb5qvlRFNNIWLGZgMSMIBQVFXUMgiYUd94BoWBLQJLZt\nYlqObsQ0TUzT9CIZU1HI5EokYsEN1SXrYaOzUGrh3nvv5bbbbuODH/wgp06d4s1vfnPD2xJCHMPp\n6uoBTuKM4fwJjiblomCHUNYB/5fUb9y4mrHianBbfWvB32Lc3t5+Qa3Kegklm81y9qyjdTp+/HhF\njcfd1lJWUMiVCEed+3TDotrJvVCCUGR5PRGYPLXYScEs0hHUkZEmnCdVvU8pUS0nwaTFVqYFD3/o\n296///61PfSOjxIPF5GFKIODCRKJBMlksmw8uQwhBE0J5zgpvlSJC5dg/L//+zuuLe+SI+izLCc1\nlYyr7Ok06epSaGqKld+vZHreYm7JYmzaYmreIZivzDimfxkcYnlNFakIIYhGRE1CWcoqNCegVKOG\nUR3ZLL8nzavZVMx/10sMjxcIaJJ0RvPciYWiUCoWsAtNRGMBwpEAiqZhWIqn/tdqFOBdxMOV81+W\n988xr3TdkysMKA1BLCzJ1yCpgi6IhGrXe0q6IBaS5EuCXJEqktNojjkjACx9idbmCJqmYJYbIAaG\npzFzUwQCAU/1nkgkiEZXRsHrxUaL8i6uv/56rr/+esAxjPzJT36y4W2W8XbgcmAIJ/31ASll7UEy\nW4QdQlkndF1ncHCQ+fn5ihpGI6gXobiz4ZPJ5JpFiWtNnxWLRQYGBsjlcvT19dW84lJVlXzJKf7O\nz5dItiVQFYlVw/upWBIEIsLr3FrIZFAj7cTCAYJqDkvWm19lo0oDhEIgslwollJy7MOVzquDgxPc\neOMN5S6pAul0mqWlJUZHRymVSoRCIZLJJImEQzSRSKTuZ+IuKv4orxbJgFOvKpVK5fnpDslEQgr7\nejT29Whcfcx5vmFKvlI1AlzTBKZZSQbzKRtVXan6FgLikdqEMr9kEY8qNTvSdMMuP1846TLfe9rV\npjAxY5RdeQ3PoffcUJpkXFAoaV6nWTKh0doSIC81IoEAJqqjF/G/1ipfrQptTFXXYlCtsmbzIRSo\n3ZEGzuCw/CrjgmaWBPlCmNlcyFHxCydl1t1yKVce34uu657yfXh4mHze2YtYLEYikfDIZj2C4nQ6\nvekRymZBCKEAD0op317+O3ixyQR2CGVdsG2bp556it27d3Po0KENX/FUk8DS0hL9/f0NiRIvFKEY\nhsH58+eZn5/n4MGDq0ZUiqKQyqoENYkorxaqtAgEVfzf0KBqk9MVwuXOrUAgQLKpE106XytNcRTZ\n8cTKmoKQNqptIIMx7zj601K1IITjXhyNRunq6vKeUyqVyGQypNNpJicnKRQKBAIBL4q50BVqLZJx\n273dbfg9wNzXdRfkWrYnd/1ugvklm6k5J4qZmrOYnrdIxpWKTjBnY84wsFrdVEIIElGFXGHlcVlI\n2STjCpncSrLJFaRnMRMIOOk/l3yjYZNs3kLXda+VeXbOTZepLKQF4bBCW0uQZFIjEAxgGiqh4MpU\nFVSq9Kv3JFeq3SQA9efGABTqdLK521yOiMuQgqUsZAoax3ZbBINBWltbK9zALcsin8+TyWSYnZ1l\ncHAQ0zQJh8MV0Uw9v77NilC2AuWpjP8MIIRoB94PvPti78cOoawDqqrykpe8ZNO25/p5uZMSbdte\ndWzwavCLEf2wLIuRkREmJia45JJL1kSEqqqSKagotklTwvmKFIsGgVAA/C9h5VlcSNGcSBCLRlEV\nQaFslqiW54cv5TUSSVmj7dRCtQ2UeJu3WLv7/7//8/Uc/8h3vUfeeOMNdfdVCEE4HCYcDtPR0eHd\nrus6mUzGWzzy+TyKonhRTDKZJB6PrzgWrq7HNE2OHz9eoYZ2989fm3Fve+q91/KfvvQjfq0Vbrrp\nZSAlHS0qHS0ql5fNEKSUpLKSqXmL6TmLqXmLyTmbYtFp2f3qwvKsjCRwYzlttpix6pJNLCxqEko2\nL2lJKiymK78TiqJQ0IOEw8vbcwdZWbZFJlMgl1PI5wVLqUrjyO4OFQuV5mSAUERDqMvq/3riSMN0\nJljWmg1T0uunxPxpr5rbDEvy+ZUVeMuCyUWFPe0rzwVVVb3P34WUkmKx6FnfT01NUSwWvce6s1WC\nwSDpdHpDguGthhBCk1KaOD2P15VvU3ZGAL+A4RonbgZs2/badGuN9F3vfvnhtgAPDw/T09NTt5hf\nC4qikC4GyWV0YjHnK7K4aNDR7dQrLMuiWCqR1wv0dDcTjjiFFd2wnYhG4kxwVGAuE2DPruoeVulF\nKEok5kVpbpEc4Lk/qE8ia0EwGKStra2iScI0TbLZLOl0mtHRUbJZRwbpusvmcjkymQyHDh2q2Vzh\nkk+9uswnf/u6CpKpLv4LIWhOKDQnFI7sW061PPPzCdRQF1/16Zn9VdSSDh0tCrM1WpUX03ZdD6+g\nVrUYl//UDWhvUbzWZ0UIlHIrsyGDtLYoFEv2sitzuZV5RLcx7QBjPh8zIQTNTRq0BQiGg1jCwlYq\nowtVqV+bCWr1U2LBQP20V0CreEsVGJtXaxJKLQghiEQiRCKRigsSv1nk008/zfvf/35mZ2d54xvf\nyMmTJ/n1X/91Tp48uabXACfV/PKXv9xLob761a/mD//wDzfFy8tFmUzA6c9eOcnrImCHULYBbkF/\nbm6OWCzGqVOnNq0rpboFuBHjSVVVyZUCTE4XOHa0CQRMzeg0dUCx6JglhsNhTKkQiy7v9+yCTUs5\nwxBULaQUKML1hPXDRilrW5RQtIJIthKaptHc3FyRtrAsi+HhYUZGRohEIiiKQn9/P+Pj4xV1mdVO\n8gvVZfzRl2VZXvHffd+hgM3unkqSSgJvvDnG9JzN1LxFvmizkLZX1F9KRn2ymU9ZXs3GWdKXj3H9\noy1oSShM6cspMz9aozYLKauilblQUJieUYiGVYq6855bWvK0twSJxQMYJZWgpqFbyorUV75Uv115\ntS6xWpGLi4lFBcuu48C8RvjNIvfs2cOTTz7Jddddx/ve9z6effZZ5ubm1rW9UCjE448/TjwexzAM\nrrvuOm666SY++clPboqXlxAiCbRIKYellL8QQrxXCJGQUtY2rdsi7BDKOrGRCMUVJc7OzrJv3z66\nu7sZGxvbtMXUNE1++tOfEolEOHXqFJFI5MJPqgEbFcOUZHMWmqYAJlKWyOXyhEJhAuXLw4WcTkBz\nLdUl0/OSppZyzUWxKBkKsaCJ41lXhpQIaaFiIoPRhu36NwNuzSqZTHLttdd6xCul9KKVubk5BgcH\nMQyDSCRSUZdZbTDTWov/bvHYtm1+dQ88Ngoh4Adl19w9yw7rWLZkbtEhGKcm4/zbn9B4cHb5wvQ1\nHSfobFWYWbBX5MtmFy3CQYWivvK7nK8j0ASIRgJk8goBKi1mbNsmHNBJZ2XZu85icdHXytymsZSR\ndLQFSCSCy+p/UyUekeRqpcRqCCdd6AYElRKwss5omjC9pNDTunmZHleNftlll3HZZZet+/lCCK+T\n0jAMDMNACLFhLy/btt3v2suBXwHuEkL8JvAZ4DkhxO9LKS9atLJDKBcB/jrGnj17vEmMuVxuzcLG\n1eDWYHRd59SpUw3VYCq2Vwpg6gXn6rNUIr2UoimpEYvFvZS1gkm+KJ3cPo4zbjxWTgkJiYrNYkEj\nHDTwCKUsTBTSRpMGIrw9syWKxSL9/f2YpsmxY8dWuMa6J388Hqe72/Efc3Pt6XSaVCpV0WHmRjFu\n8X8tJGPbNmNjY4yPj7Nv3z7i8TifeN3LPM2MYRje67rFf4Fj77KrTeVEn7NNW0qWMjZTsxaf+/KT\nFa/34OyzvDV5yvfGKt4lzUmFqbmVxf501in2p2vYzMynrBUpNreVORAKEAw6hf5AIOB1lpVKJRYX\n0mQKQbLZ5XSZM2cFLukNoAYCRKMBRFUrcz3hJDgjgethbH5zCSWbzW74vLIsi6uuuoqBgQHe/va3\nc+DAgQ17efkubo/ieEYAvAK4GdgLvAW4fUM7vg7sEMo6sZ5o4kKixPUo5WuhugVY1/WGoxI/MsUA\niwuL5HIWNi0EQ01Eo0pF/bOYNVDEco1+KW0TL6e/VCRBzWY+E6IjXvSIBMpKepz6iRpduwh0M2BZ\nFkNDQ8zNzXHgwIF12Wj4c+3+wUylUol0Ok0mk2F6epp8Po+maRXpslgsVlF3WVhYoL+/30tJ1poM\n6P6uLv7Xqsu0JlVakyr3fuDlPPwn36vY1hV9AWYXbYYnLOaXKr+79cwxAaIhSNdw2jTN1es5kbAk\nX8SzmHEhlQiJBOi6tWwYWXZlHhgU2FLziv+OaaZKa3OQnK0SDgaw0Sj5WpklYNgBnAHOKzG+oGJL\ns67d/3qRSqVoamq68ANXgaqqPPPMMywtLXHLLbdw+vTpDe+Xbz0qAj1CiPcAhpTye0KIu4GLZJ3p\nYIdQtgBrFSU2SiiGYTA4OOgtjMePH3faV8utwxtJIy0uLnJ2MMfMnEl7e4JgKMLYyAKXXhLzyENV\nJLm8RXPCPZ0lQ5M2l/Y6r6sbNkJIlgoB2iIlikWlYkoiOG3IWvTiRCjuQLOhoSF6e3u55pprNtzy\n7SIUCtHR0bGioOu2MQ8NDZHLOdMfI5EIuVwOVVU5duxY3SveCxX/VyMZP/zDpubnC8zMZog2tZZb\nmZ2UmaoqLFV1gj04+yyUjTqqBZpQ3xFYCEE0JCkUVx5b26acfiunAQP++2yaoiYLKcf+3q3LzM07\n35uWpEomLwgGFdpbgyQTGlpQRZc2sZDtjGGugm7ATEqhq3lzopTNIBQXzc3N3HDDDfzoRz/asJeX\n7zvyj8B/BG4D/nP5tg4u0qRGFzuEsk6sFqFIKZmfn/dEiVddddUKJbcf9Vp966G6BdhNnbnYiJ9X\nLpfj7NmzSCmJxI5hmtO0tYTQFEkqY6EFFNx0uzANLClojznHQgEyeQgGheMBbFkYljOcqqO92XHX\ntSwM00QVNiENdNNiamqaZDK5KSrmenAHjiUSCa6++uoNT8ZcCwKBQIUGwrIsBgcHmZ6epq2tDdu2\nee6555BSetoH92e1/VtL8f+p917r3efm6YUQWJZFKCi4pFvjku7l0960JDMLNtPluszUvOWRCdRW\n/c+nLIIBBd2opX+pf1zq2eQrikI8GiFX9FnM+FqZjdJyK3M6U1b+C4FhGFy6xy63MmuEI4GKQWZj\n85tHKEtLSxsilNnZWQKBAM3NzRQKBb71rW9x1113ccMNN2yWl9eklPKd7h9CiG7gHmpPdNgy7BDK\nJsEt8IZCoTWLEteaPpNSMjExwdDQ0Kb7eZVKJc6dO0c6naavr4/m5lYGv+/kwZMJDcXtRFQUKG96\nZKyAJQLEy4SSy0uQ5YdIp76SLQYQYjn14UZNwi4hsNGCYfJl1wH3qt2vEalOE60XbjpQ13WOHj26\nLbPA3Y678+fP09XVtcLQ07Ztcrkc6XSamZkZzp07h2manprbPRYb6TCTUqLrOjMzM16Hkfs8IQSK\ngJ4OlZ6O5ed/8heVr7G/Vyt3msny+xK0JAXT87UK+tCSqN3uO79Un4gWqrQ2/lZmtDCxmKwQl7ru\nBaMTS5h2gPGqVuampIad17hyf9BR0m8QGxU1Tk5O8ju/8zteXek1r3kNr3rVqzh27NiGvLx8Rfnb\nhRA/lVJ+XwjxVuA3cexX/rrhnW4AO4SyQbh29UDDosR6cCc9DgwM0NrayjXXXLPq4rJe9+KhoSGm\npqbYv38/R48eRQjBYhaWUo7HViSiUcg727NkeQa86oyAtVWIhJwTdXzGmUMhEajCJhqVzGYCJCNV\nV4fl+omQFrHmZloTyyeoaZqeEHF4eJhsNutcucbjJJPJukLEeu9rdnaWgwcPNmw3vlHkcjnOnDlD\nMBjk1KlTNSNVv9DSheMo7Ki5FxYWGB4eRtd1wuFwRYdZPTW3u133t5vq27NnD93d3RVmmbVamauP\n78eughtvjDmmmznHx2xyziKVtSmWasyHkU6kmi8td5wtRziClmQNpwDAMKC9WWFuaWVEUSwtizTd\n1KllOXWYQDBMS8RmMb2ylXlmTuEXexUOX+q0fW+km3KjhHLFFVfw9NNPr7h9o15evqL8rcAjQog9\nOLPl7wfeBTyDYxB5UbBDKOuE56RaKNDf30+xWNywKLHW1MalpSXOnj1LJBLh5MmTayq2ryVC8Qse\ne3t7V1w1L2VhYtIZWhUMaUxOFQkGhNd1k14qEYupZIoQCIDAZmJOoilgS4G0JJGIZCat0ttaTW7O\nl1+VJoFo5cmpaZrX9+/CsiyPZKqFiP6it6qqK+okL3rRi7YsjbYaTNP0TEOdiG99i5AQwptj7reX\nKRaL3rEYHx+nWCwSDAYrIhl/h1k2m+XMmTNEIpGKVF91JOP+donGsix+9p9e4m3HTZcJIWiKKzTF\nFfouWU7L5Ys2U/O2p/wfGMqTzWkV7cv+tFmt6MSFskpHV6gqEyhZPhfj0QDZQu1W5l/2F5ClcUql\nkmfH4zeMXCvJvFBtV3z7X8KZJf/7wNellA8LId4F5C7m/uwQyjqh6zqnT5/2JiW2t7dv6MrHbSF1\nT3S/DcvRo0fXFfGsRijV0U69RoG5JQvDdLcnmJrRaWtxvbkkzw8X2LUrXDYBlJRKjmyxOekcg0zW\nZldCgBREQ1XjgstlfUV1OnrW8n5qCRFdm4yJiQmy2azX1x+LxbxZNBebTKSUTE5OMjw8zJ49ezh0\n6NCm6Yv8HWadnZ3e7bquex1mfnsZ27YxTZP9+/fT1dV1QQ+zRor/7k80rLC/V2F/uSHj9OkROjq7\n+eJfVL6WpgpMS7KQsohGFArFGmmvJRtVqV1rSWVXui247V2Z/MonuNHWbLaJK67oRQjh2fFks1nm\n5ua84+USjNsqXiudnE6n2bdv38od22b4PruHgFcDlwEfLd8WA5Yu5v7sEMo6IYSgubl5Qy7Dfrhp\nKsMwvBbgQ4cOVZjarRX1CMUtTIdCoQtGO9PzlQalli1JxMtfE0PHtCS5soIZIZhbcE70ZFmDspS1\nKZnOvyPB6nRImVACF3ZPrgdVVWlqaqKpqcmrk5RKJQ4ePIhpmp4Q0bZtYrFYRSSzVQX5dDrNmTNn\nLmrhHxx7mfb2dtrb25FSenWYtrY2wuEwCwsLjIyMeLqa6qiuHtar/IdlknGK/yu3fdfvJphbcoSY\niymb4Uknoin5hJWmBR2tCrMLtadh+r3JpJSe8j9XkDTFlZrjmVMZm8k5m54OtfCuGyQAACAASURB\nVK4djytinZycJJvNIqUkGo16RCOlfMFGKC6klH8uhLgCZ4JCqmwQeTcwdTH3Y4dQ1olgMOiJ3TYD\nQggGBgZIpVIVLcCNoLqGks/n6e/vxzAMDh8+fEFju3f+7bd59By8tvMkAjDL9uiRiDPke2isQCKm\nYFsQCwMIhifLjwkLNEXSFLOZzzhfq4Dqu6qUElGu6sdjjRMKLNulzMzMeHoS95i5bZduwTuTyXgL\nrWVZRKNRb2FNJpMbWvx1XWdgYIBCobDp9bP1IJ/Pe/Waq6++ekWdrVZU5xKuvy6zkQ4zKSVLS0uk\n02mEEDz13mt59X/5EQj4+zuuAySdrY6bsQvXLHNyzvLcmEtVKTF/6uz2tlMV9/nPk2hYkKrTz3R6\n0KhoOvBD0zTvAsX/vvL5PNlslvn5ee644w7Onz/P008/zfe//31e+tKXctNNN9U9VtUYHR3lDW94\nA9PT0wgh+L3f+z3uuOMOFhYWeO1rX8vQ0BD79u3jwQcfbDh1LoS4BLgW6MUZPvRD4DG5WcaDa92P\ni/x6F8ILamdqwe2a2Shs22ZkZIRz586xZ88eDh48uOE0zcTEBKVSid7eXs6fP8/i4iKHDh1aU2Ha\nP9AK4LbOgxw53E3/YIFTlydQNJWnn03R3RkkW1Lp6RDs363yxNMOgV15NIBQBIpdYnQhgmULXnQo\nhyd9lDaKLFEwwhzZ2xihuPqewcFBenp62LNnz7qOmVvwTqfTXqrIMIwVJHMhcz7X1HNsbIz9+/fT\n2dl5UbzIquEXah4+fHhdV9B+wnU1M6ZpEo1GK+oyq7W9u3Cj60KhQF9fH9Fo1Cv0uz8uqov/tT6/\nbMGpyfzbv/1Rxe2dwPWdJ0A6ZC6E8EgwHhUrJmi6aG1S+P3XxDf0Gf32b/8273rXuzynhNe//vVr\nfu7k5CSTk5NceeWVZDIZrrrqKv7hH/6Bv/mbv6G1tZW7776bj370oywuLnLvvfdeeINVMAyDYDD4\nt0Ac+CegCfh3wBeklPeve4MbwE6Esk5sdOGobgHu7u6mra1tU3L+Qgjm5uaYmJhg3759HD58uOH9\nTcQVJqcdJbOiqqQWnUK9WnbcS8ZEhSBO0wSprE1rbLmAj/APyLAp6oKejsacVFOpFP39/cRiMa66\n6qqGHFn9BW+/pYo7tGtxcdHrqqr27XKHnF1I5X6xMDs7y7lz5+ju7m5IqFmvw2w9A8z8BL9v376K\nNHB1e7T7u3quTK26TDyiEN+jEMKpNLv46ze/tKyXsekfKpApLEdU2bysaxezkLKZW7LpaGn8s0qn\n0xw7dqyihrVWuOc5QCKR4OjRo4yPj/OP//iPfPe73wUcH6/rr7++IUIpG1W+SEp51L1NCPHfgO/h\ndHttKoQQCeDlUspvVt+3QygNoBGDSH9RvKWlxWsBdmdvbARuQbi/v59wOLwuq/p6iEYCDI2nURSF\nfCHEwGABoagUDWfBiIShf9hv6CSYXZQ0R8v1k1DltKVsXiBEkGR0fQtfsVjk3LlzlEoljhw5sul6\nknpDu6p9u4rFIoZhoGkae/fupb29fVu6yAqFAmfOnEFVVU6ePLmmaZ5rxXoGmCmK4hHv0aNHaWpq\nWlMbsx+rFf+llDx510s5de8Pvcfv7dLY2+UsWf09Y7S0tGESL8+VsSgUJf0lWbOT7BV//gPAbYN+\n+Yr7L4TNqqEMDQ3x9NNP8+IXv5jp6WmPaLq6upienm5om2VSnhRCvAwYxJkIcBJIbXiHa6MZR/dy\nCPihlPIn7tyVHUK5CFitBXijfl7z8/OcPXuW5uZmjh07xuzs7LrJ5Oywzut7T/IP48+QAX6r4wSJ\neIRYzHSuJg0b3TDR1BLzixJVVchkbJYyobKYrNyqWoCiUY5gosvNAUs5DWEb7Otau8/YanWSrYa/\nq6q9vZ2RkRGmpqY4dOgQmqaRyWQ4ffo0xWLRM4d0r+BXGz+8Edi2zdDQEDMzM/T19TXUtNEIqgeY\n2bbN8PAwU1NT7N27FyklIyMjax5g5sda6jLVyn/3eaZpEgxqtCdVutpVTh4uP19KFtM2U3O2p/z/\n6NM/87bxn34GN964/uNgWdaGmy2y2Sy33nor999//4p65kZGOOzevRvgvwB/ihOV7Ab6gD/Z0A5X\n7l8LjkdYVko5KoR4O07N5i+FEK+QUi7CToTSENYaobh2Jqu1ADdKKJlMhrNnz6KqKidOnCAajZLJ\nZNa9re/9LM+3n8whpTMh0P1KG4aNAJJxldkFZ1Hp7gyQLWmOg6yul2eu60RCJosLGgFFZXIx6liK\nhJzjo5sq/RMaR3tUmuMXJrrqOsl26UnciPLcuXN0dXXx4he/2NsPf9rDbw7pXr27+hD/+OGNkMz8\n/Dz9/f3s2rVr244HlH3ezp6ls7Oz4ni4WG2Amd9eZjWvubWQjBs57tmzB8MwKhyZAdqaVNqaVI4f\ncAjgoyv1hOtCLZ3YemEYBrfeeiuve93r+M3f/E0Adu3axeTkJN3d3UxOTjaUTgMYGxsDGANeB/wq\n8GPgB+4iv0n4XeCwEOKXUsr/X0p5DjgnhLgW+DyOmHKHULYCbpomm81esAVY0zRKpTpj6epsu7+/\n3yuA+sPw9Viv6IbNNx5fYGisRFOo7AYswbYl4YjKfCpAQJM0xWBqwSGpQECBknPS50ph3GzL7k6B\nokI8XCRfsrGsPLn0Iqqt8OxwM+1Jnc7mCy+C6XSas2fPbqhOshlwLwQCgUBdlbuLWuaQrt4hnU6v\ncCB2SSYWi11wkSoWi56/2okTJzbFSboRGIbhiXgvv/zyurZCtQaY2bbtkczU1JSnsfI3Qqx1gBng\n2QRdccUVxGKxuo7M9ZT/jcBtLGiUVKSUvPnNb+bo0aO8+93LY95/4zd+g8997nPcfffdDfl4ubYr\njz32GMB/lFK+FvhcQztZB0KIkJSyJKX8ZDnF9UdCiEullO8CkFK+XQjxnBDid6WUf71DKA2g3hfL\nP0DrwIEDHDt27IJfwrVGKK4Ce25ujoMHD9LR0bFi25qmrYlQFlM6n/67CUYniuU35P2P3b0xSjJM\ngBLZdJFFrRmQNAV1sgVnQWuO2Sykl0/UWFRhKQuxeAy76FwVdnYq/GJYRSIJKSlGzvVz/ozpaUP8\nraqlUomBgQGKxSKHDx/etvZb9/NbXFxsSOXuopbewe9A7IoQ6/mXuR2AU1NT22of43cfuPTSS9m1\na9e6F1VFUbzP27/d9Q4wc6O03t7eCtForbpMdfHfnzbzK//XSjaFQmFN3nz18IMf/IDPf/7zXH75\n5d7Y4D/90z/l7rvv5jWveQ2f+cxnuOSSS3jwwQcb2n655TkihLgRGMIxhMwBKdngPHkhxL8BnsA3\njVpK2S+E+H3gcSHEbcD/kFLqwDuBjwkhvrJDKJsAdwEYHx9n7969K1yAV8OFCMW2bUZHRxkbG7vg\ntlVVXXVbUkrODef4qy9Pks1aFUQCcOhgE0IRTEymKBZt+g63sZSDCAW0YBjdcgYeWboOLBeDgwHB\nYkYSDS9v69xkgJKhEAjAyb4mOpqu8RYSd1F1W01t26ajo4N9+/ZtapF5rfAvnG4L92bXQaodiGHZ\nvyydTnv+Ze4UR1c8eyHt0FbB9SKrtm7ZDKxngFkgEPBck93W6NU+m1pRyYWK//5xzLXOrY06DV93\n3XV1U+Tl6KIhuMehfOHTA/wxMIrTDrML+EvgbxvY7vuA9wKPAFEhxATwHqAkpVwUQnwIh0QeB2aA\nfwb+J3B8h1A2ALe7anBwkK6uroa6q+oRin+mipuzvtCck3p2+O5J9MOfLfHQw3NYlsQ/LUvTJPv3\nJZieLZLLOxHO3r1x0nlQjBwT8wbHd7WgF0EYOUQ0DD5BvS0hGpTOzHAgqNlkCs6/4xHoaFr2hYrH\n48RiMVRVZXFxkb1799LW1kY2m/Wcef0CRPdnq0YFb5fKHSr9y0qlkjd1c//+/ei6zvj4OGfOnAFY\nUezeqnZlt/g/Ozu7bm3LRlA9wMxV/g8MDNDR0YGmaYyMjHD69OkLDjCrxlrHMQMVEb5LNC9UlbwQ\ngoWFBfr6+gD+FU4JI4KjR4nhdHytZ3tKOaLZBbwW+D5wKU4xPu8+Tkr5NSHE63CU+O8GgjizV8wd\nQmkQ7hW2vwW4EdQiFLf4GY/HLzhTxY/qKzdXrHjPCcjLI5weyBMNgWWBadmYpiQe0whFwpwfLnrP\nC4dV1GCQvz7zv7zbvvHTfu5+0cs4/VyGPQcqR+bqhkRYOrrl7KcqDSzbWZirx7C6dZJoNMqVV17p\nHbdkMklPTw9ARSTjV7n7ld0bJZkXisrdPwr4wIEDNVOZrtI9nU4zPj5OJpMBlovd7uK6UZJZXFzk\nzJkzdHV1beoQsvWiWCxy+vRpAoFAzXOr3gAzv73MhUh3LSTj/vuf/umfmJiY2Jo3uwF87WtfY3Fx\nkZtvvhkp5Yx7uxDiIJDEl65aC3zpsS5gSkpZAJ7zP8ZHOu8B/psQoldKOS6E+CrQvkMoDWB6eprJ\nyck1uwCvBj+huMaQUkqOHz++Ic2FX/n+vmfhV/WVLel798SYW7LJpyoJbd++BJOTK30shs8v0NMT\nXWHel81LikWTeJlnSkUDRJlQWpxQfz11En9KZC0k41e5X4hk/Cr3RusCmwW3ndw166y3APr9y1z4\ni92Tk5NeN2E1yayFdHVd96Kj7Sz+u07YY2NjHDp0qKIG5Uet9KHfmXp8fNxLH25kgNn8/Dzvfe97\nKRQK/Nf/+l838Z1uDh566CFe+cpXetGTEEKUrVZGgI8Anwa+s9o2hBD/FqdDbFRKOS2E2Isz+Wio\n1uN9pGMCz1LOVUgpvymEUHYIpQF0d3dvWqFUVVUMw+C5554jk8k0bAzpR0VhXrLC0CYaVmhrjzI1\nZ5b3ATRVQVWhrSXIyGiOUmllzjebM7lkXzNpX1NaKACGbhGNlf27NBu9CGjOwK2OpMng4AjT09Ps\n37+/5hX4WlCLZFzPpXokU72oLi4u0t/ff8EFfKvhRkfFYpHjx48Ti8Uu/KQq1Cp2++1UpqenGRgY\nWNW/zO+QvJ0WMuDUbJ5//nkSiQTXXHPNuj+bWs7UjQ4wk1LyjW98gz/+4z/m/e9/P7fddtu2HZfV\nMDExweWXXw6AEEKTUppCiICUUhdCNOGNxFsJIUQv8Fc4xdDR8m13SylHcNqPV4WUckoIUcBJsbm3\n7QgbtxOmaTI0NEQ+n+fgwYPekKtG4IbnsViMrz/85PIdwvlpa9GQtqS5OcRSxmYxpaMgsSwwTIkB\ndO8KMzNvYpa/hq8KHeIbJWd42L+JHAJAl5VfmaaIydycTk9vnJIETJ1ASMO0IB7I8/RTT9HV1bUl\n+gk3zVFNMv5Ftb+/35vup6oql156KR0dHdtCJu4V+Ojo6JYs4H5hYXVk53ZUnT9/HtM0CYVC5PN5\nYrGY14K7HXCFkjMzMxw5cmTT5rbD+geYPfnkk6iqyre//W3C4TDf/va32bVr16btz2ZDURRGRkY4\nceIEUjqjVaWUbnWzlTopLyFEDPgg8D0p5T1CiEtxiuwvBv6+/BhF1ukQ8933qerX2DGHbAC2bXuq\n3UafPz4+zsjICL29vYyPj/Oyl72soW35LcWXUgZf/9YcP30mhWU5LZOWbWFbNq3NAqGFKJTU5VGp\nvu3s7okwuySprulfdqKD4fOLZHMmba0hor7oKaoUicYC/PL5FFdf00XRVFmcXqC1PcJizmJ/6zwv\nP9W65hrQZsOv6t69ezeKoni5d7+9/XrSQ43CLf43NTWxf//+LX2t1eDOt5+dnWXXrl1ep5lro+KP\nZLb6c0ulUpw+fZrOzk4uueSSbavZuB1m9913H48++qjXbtzV1cXDDz+8bft1ITz66KP85V/+JTff\nfDNvfOMbL8exXDGBW3DI4Z1SyrlazxVCvBg464ofhRAPAl+WUv6PjezTToTSADYSRbjFfNdcMBAI\nNFTw8xcNC0WTf/7REk/+LIVtS9pagygKqIpAVQXhsMb4tI6uWyAtdF33evFVRWHP7giTc2b5xFl+\nb81NAYq5Etmckxrr6IiSK0cvyZDB4Pks+w8kURQoWQqqYjI4lEKLKITDca6+rJtt4hLPPLFa5e7C\nnw7xC+4aqUGsBv+cm+2abe9iYWGBs2fP0t3dveKYuMaQmUymwhgyHA6vIJmNRlWWZXkCxcsuu2zb\noiMXc3NzvOc970HTNL75zW96ivWFhYUXLJkA/Mt/+S8ZGBjgm9/8JsCdOGRyAFCBN9QjEwAp5Y8B\nyikyA1goP8+DryazZuxEKA2gEQv7VCrl9fUfPHiwovD55JNPrislJMujWi3L5kc/S/HwY3NksivT\npR3tQTJZi5LuhB2JuEosqhEMKNhSks1bJBIqEzMmluVEOUIIJ4JRVQ4fjDM4slwwOXbZLjJFhUTY\n5vTzs/QdTDIxVSKR0GjelaSUnmVyCvYf6SESgluv3ZjpZSPwq9wPHTq0rqtsP8m4diouyfgX1bWk\ny/z1iX379tHV1bVteXhd1zl79iymaXL48OE1F93dK3c3qstkMp5/mf94rDbfvhquQHH37t309vZu\na21CSsnf//3fc++99/KhD32IW2+99QVZK7kQMpkMyWTyCiCK0501vJ7nCyEUHC3Jv5dSnhZCHJNS\nPneh59XCToTSANbzpfMPuaonVHM7vVZrPf53f/5tniqbkf7ig/+CX57J8o+PzjI1U5vYertDTM3q\nFSmsTNbyiEcISU9XhLFxHSEgqAmC4QCKIkHahEIGA+cNiiXHciIS1ljMWsTCksFzC0gbQkGFTK5E\nPFYEkth2jK5O55qgul14q7EZKnd/zr16UFetSMafLvOTTCaT4cyZM8Tj8YuubfHDHZUwMjJStyV5\nNfi1IbX8y/zuw8FgsEIbUu1fZhgGZ8+exTCMTXdJbgQzMzO85z3vIRQK8fjjj1dY5/zfhvJUyf9d\n7/41RBptOJ1h00KIzwJhIcTbpJTrdiveiVAahK7rqxpE6rq+5iFXzzzzjDeYqBaklBz5cKWi9nVN\neymWbDJZC9P074dkb2+E0cn6/mCqCp3tYaZma5NRR0cIw9a88axSSnp7QshglNmxGVJpk1hMJR5X\nWEwLXvKibrRwnJ8/PcWhg3FkMMZ1xyz2dW79x1mtcr8YV731IplYLEapVPIuHhqdvrcZyGaznD59\nmkQiwYEDB7a8ZuOfb59Op8nn8wQCARKJBLZts7CwwIEDB7a1TRuc78tXv/pVPvaxj/GHf/iH3HLL\nLf9XRiU1UPEmahXVhRB7y11cVN3eDZzF6fb6Rynl+xrdiZ0IZZNhWRYjIyPrGnK1mlreP78bcAYg\n2jAxtSxEjEVUYjGVSFghEFBJZUw0DWq5sISCgmQiWJdMeroiZIoKprVMBkIIWtuSpBaLmFYIVZW0\ntwpmlwKoqiRfKmGkUuTyJQxTQ6gh2mM6joB26+CKJC92JFAdybjprfPnz9Pc3Ew0Gl1TJLMVcIvu\nCwsLF9W6xT/f3kUmk+H555/3yHZoaIjR0dGKGtWFVO6bienpad797ncTi8X4zne+s20eaVsJl0ik\nlLYQIoKjcjeFELcCJ4A/qPG0dqAf+F0p5bPl7ahSyrU5zfqwQyibBL8NS3d397psWKo9uNzIx7Is\niiWLx76/4JTb3POuqsEsV7BIJjTmFgxyeYdohIDmZIBo1JlXUijaGKaNpmnMLdbuUNuzO8pCRmDb\nlZGFEIJirsS5wSUeNQeda6Fx575bWy4jnmxjsbBEOCQIhsJoao7Tzz+PYRg1zSA3Cl3XOXfuHPl8\nflvNJMGp2Zw+fZpIJMKLXvSiirSlX3zon+W+VSTj1id6enq4+uqrt7VrylX/VwsUXZV7JpOpULn7\nSeZCc1TWC9u2eeihh/jEJz7BRz7yEW6++eb/V6KSFXCjkrKJ428APxdCPCil/Crw1erHl9Nh/xu4\n0v27vJ11kwnspLwahmEYXuTgH3J14MCBdduw9Pf309TURGdnZ0XB/cdPpXjkO3OoiiAcVrBtyGRN\n0tll8gmHFNpagoz7IpZa6OkK83epESfCKZ+rr4nuQVUFqgrRWADdFNg22JbEtiW2lCgKBIOC0Skd\nTdN41Bqq2O6bLz3JpQc7+fnTU0gJx090cmS35Ip9dsUM91QqRSaT8YSHTU1N615Q3YXqhaBytyzL\nS2kePnx4zfoJP8mk02mPZPxCu/WSjK7rnDlzBtu2OXz48LbWJ1yBYjKZ5MCBA2t6H27rspsuy2az\nCCFWkEwjxDs1NcWdd95JU1MT999/f131/f/NeOCBB3jb2972GinlfxdCBIDXAzcCbwS+hLOu3iml\nHIL60UejUYkfOxHKBuAWXzVN84ZcNQJN0zAMA8uysG2b5/uz/MM/zTAxXbsOEgoqNCc14jEN3bRZ\nWKiviYmEVVqbg862gjjRhQRsmFswaGkOoAmNwdGVKbDOdo3FjI5hqkQjUYQinE53H9pbQyi2jpSO\n67BuCa8gL2rMcPfXH/yWIX6PrkQiseIK9YWicndbv8+dO8fu3bu55ppr1kVq9RTu/kjG79XlL3RX\nv2e/UPLAgQMND2jaDLimknNzc+siWKg0yXTh9y8bGxtbMazrQsRr2zZf+cpX+NSnPsWf/Mmf8KpX\nver/2aik3IDy34QQ/yClNMrCxYeBt+H4ct0ppRwSQgSB/cB7hRD/wSeCBBqPSvzYiVAaxNmzZ5me\nnl73yVMN96p7cnISobXx1C9sfnG6iGHWPxQdrUGkhLnFZRKIRlSaEhqaJjAMyWLKoK01yGLKpFh0\nFvjHA1Oech4T3tizn6lZY4U3V0CDRNxmLuVMavSftF/Ln6l47Ef+xa+wsKgzNl5gV2eIjp5mXv1S\nk/Wcu+6CmkqlvCtUwOsWWlhY8OzLNzKXYqPI5/OcOXOGQCBAX1/flg4Aq45kqklG0zRGR0dpamq6\nKEX31XCxBIr1ojuXZFRVJRaLUSwWueOOO2hvb+e+++7b1uaIrYI7XEuWB38JIb4CaFLKVwsh7gDe\nBXxDSvkOACHEFcCclHJCCHFIStm/Ffu1QygNolAobGgOtGveeKoD/stt1/I/vjnOT55OY5oWtrSJ\nRwWJRIBgIIBlq6QzFtGIRiKu1o1cXLQ0aYRCGgtLBsGAgqYJNE3wrfwoS+XHvKlzP5btWM9LWzr/\ntm1sq0g2XyKRiBOPR1AVBYnAsqBkQiCokC0sLxjXXN3OU08vICUcOhBn394o1x3b8IWONyVwbm6O\nWCyGYRgVV/fuQKqLcdVpWZZ39d3X17dtC5Rt26RSKQYHB0mn0wSDQVRVrajJbKW1fTUsy2JgYIBM\nJsPRo0e3RaDoj3ifeOIJ7r33Xqanpzlx4gT/+l//a2655RYOHjx40fdrK+GSCTjpzmAwSDn6mAB+\nG6db6wM4hPIlIcTrgfcD75BSPgaNiRbXgh1CaRCWZTU0C75WC/CNVpcnPvQ/zqmlWChYJBM2hqkQ\njwYJBAOUdEEma1ccsEhYpb01yPhUiVofazym0twcZHJGX3G/aZqEg0UsGUYoQWot05GISmtHglR2\n+cmnTrbw7M+ddvXLjzdz5RGN/V0b+xj9Kve9e/d6J4+ba3cjGXfqoZ9kNjq/vd6+dHd3s2fPnm1V\nTs/NzTEwMEBPTw979uzxpg+6V+1uHQLYcpJ5IQkUwTFKvOOOO9i1axf33nsvMzMzPPXUUxVTEv9f\nwtjYGLfffjvHjh3j6quv5pZbbhFCiNcCnwT2AL8G3I6Tj2jBIZOnt3q/dgilQTRCKC5JHP/Idytu\nv/2SAwyPFchkK7cnBOzuCjO3oFMo2eUIwsa0LGzLQlEkTQmFaDRAJBpEEMAsmz2apsQwbAzTCYl7\nusJMzRlVmhXnaqdYLNLZplA0IkhZe2GIxwPEm2Nk88vPj0YVOtsjjIwVADh5so0br4FIg5kgv8r9\n4MGDayouG4bhpUBcknFFdu7PepTcLgqFAmfOnEFRFPr6+ra10F0qlThz5gxSyjUV3f0k408hbgbJ\n+FX3R44c2XaBom3bfOELX+DP//zP+ehHP8pNN910UcntkUce4Y477sCyLN7ylrdw9913b8nr+KOS\nRx55hPe+97188IMfJBAI8KY3vYmlpaVjUsrny3NJslLK3ym3De9201tlRbzcisjExQ6hNIj1EEq1\nnuTm+/+Zc75xI8/9wQ0ALCwZDI8VGBotMD1bYnqmxMz86hYvnW0qSymddNYp6iNBUVVUVUFVVdqa\nw04xXTgRjKYJJIJiyWJuPkuxZHNgX5KZBVn36Dc1BwlGoxSKlQ/Y3RNictrwop2XXdvOK69c/0e4\nWbPcXei67kUx6XS6wi4kmUzS1NRU15LFNZScnp5edSbHxYC/q+3gwYMbUnPXI5nq7rJ6EZgsTxAd\nHBzcdqt7F+Pj49x+++3s2bOHj33sYxd9qqJlWfT19fGtb33La9D40pe+xLFjxzbtNfxEkkqlaGpq\n8mYmzc/Pc9ddd5HL5Xjqqae+LaX89XLqqwj8upvegs3p4FoLdgilQazFcbiaSNZbc7FtyfRsieGx\nIsNjBYZHC4xPFTEtSVtzgIdK457Q8Qaj3OEjwbKdVFlzUrKUtjBNx5/L8ehSME0TwzAIhYLsvyRJ\noWgTCiloqtOaXDIkuYKNrtu0t4eQWsRTzbsvoqqC3l1BxqbcmSqC33hlGyf3r/0jvFgqdyllhV1I\nOp2mVCp57rruTyaTob+/n127dm2r+y04HYSnT5+mubmZ/fv3b0ldZDWS8XeX6brO888/TzAYpK+v\nb9usZFzYts3nP/95HnjgAf7sz/6MV77yldtCbj/60Y/48Ic/zKOPPgrAPffcA8D73tew0LwuHnjg\nAT796U/zhS98gSNHjjA+Ps4b3vAGPv7xj3PVVVchhFgE7pdS/pEQ4qiU8vlN34k1YKdteAuwUSJx\noSiC7l1huneFeclVztWXYdpMTBW56bM/Lj8IEPAdY4Yb6AThCCUv3Rtn3PUCNwAAIABJREFUbLJE\nKATBkJNqMwwDs2g6j1EU2poFgyPZsp292/61jN7uEEXDwixksCyJZTmpNMuGrs4wM/PLi1xLs0b3\nOuaCXUy/KyGcbrVwOOy11rrGh+l0mtnZWX75y19i27bXsbe0tLRpQsz1wHXiTaVSWz6auNYkSD/J\njI6Osri4iK7rtLS00NTURKFQ8MYfbAdGR0e5/fbb2b9/P9/73vc2dX7KejE+Ps6ePXu8v3fv3s2P\nf/zjDW/XH5UAfOxjH+PRRx/ly1/+sjs/Hl3XaW1tpbu7m8XFRYD/BewvRyhnYPWZJluFHUJpELUI\nwq9w97XzberrBjSFS3ZXtc4K+A+9cPTyXQyPFiiWbM6cWxaMSNtGL+mOLiQeIxhQaG0JMDld8CIt\n90vsRjK93RFmFiysGkFyb0+Ekh3GNJajkeZkgI6mC0cnLxSVuxCCUChEsVgklUpx/Phx2traPCGm\nW4xfbQLkZsMdbbB7924OHTq0LVfdLslomsb09DS7du1i37593lCq0dHRFZGMW5PZSpKxbZvPfe5z\nfPrTn+bjH/84v/Zrv7btKbetgJ9M5ubmaG9vZ3x8nA996EP09fVRKpW8dO2BAwf4vd/7PQYGBgD+\nRkr5Rf+2LjaZwA6hbBr8UYkQYsuv4J77gxs49kfL46LveMcNFfenMyYDgxmeenaMkfES2WIUXReE\nggrJpMbsvIGmBdA0eIxJAG6wd2FZFq3NkpHxLLYtUVQFVVE9otm7N05WD1a4GD8y/3My83DP/1re\nt2pUq9yPHDmyrQvC4uIiZ8+epb29vWLkbLUQ0514mE6nvQmQrvbBVftvtIuqWCxy5swZhBBceeWV\n2zaQDCoFin4vsGAwWFGjcGe4u5FMJpPxJmhuNsmMjIzwjne8g0OHDvHEE09sq9WOH729vYyOjnp/\nj42NeS7VjUJRFM6dO8f73/9+brjhBt761rcyPDzM17/+dX7lV37F0z51d3fzwQ9+kO985zscOnSI\nY8eOfREuXq2kHnZqKA3CnYmyWemtzYRt256Xkt+iZGZeZ3KqyMSMzuy8zsyczmdnBpePugFv7N3P\nxLRRvkliW7ZXk+nsUClYcYdkVBVNVQkGVb40/fOK168mFL/K/dJLL902lTs4HVPuOIFGhZL1RIfr\nvWKXUjI6OsrExAQHDx7cdrPCjQoU/STj1mQ2QjK2bfPZz36Wz372s3ziE5/gFa94xbafW36Ypklf\nXx+PPfYYvb29XHPNNXzxi1/k+PHjDW/ziSee4J3vfCfvfOc7ef3rX4+qqpw/f55XvOIVfPrTn+aV\nr3wlf/EXf8ETTzzBX/3VX3mzbYQQKlvcwbUW7BBKg3BnUre2tnok8kL4srupGndRuNDi7Y9yAD53\n81VomkJAEwQCgkBZFBnQFMJhFVWRFPJZ8vkM+ZyzmL7pfxYqtvHL/3y9YyhZLNLf3+91w2ynyt0f\nITUyG+RC8NceUqmUt5gmEgkvkvELMS9G0X09+z4wMEA2m+XIkSObKlCsRzL+7rJaJDM8PMw73vEO\njhw5wr333rutky5Xw8MPP8ydd96JZVm86U1v4gMf+MCGtvf1r3+dBx98kPvuu49nn32W4eFhrrzy\nSsbHx7nvvvu8+tXHP/7xauLa/sWHHUJpGM888wx33HGH1+p61VVXcfXVV3Pq1CkikchFJ5dsNsvZ\ns2cJBoNr1nDASkKpla66EEzT5Io//b7392d+PeyJMnt7e9m9e/emjI5tFO60zJaWlou6eLtCTHcx\ndZ113ZnlfX19tLe3b+uFiCuWvJgCxXokMz4+ztDQEOl0mkceeYT777+f66+//gVxobaZcMd314rU\nRkdHectb3kKxWOTqq68mnU7zwx/+kJ/+9Keoqsrp06c5ceIEsKJ4/4I4SDuEskGYpslzzz3Hk08+\nyU9/+lOeeeYZAE6cOOGRzOHDh7eskOsO8spkMvT19TXU9fKKe77DvzoF//HG9ZNJNdwIqb39/7R3\n5mFRVf0D/xxAxSUVNZDcUERAEBEBl8osRctcMEvTyjIl38o1Ne1nKuZOuLwpZpqWlUuWr0tqbhkE\nGqDihuziLoqioOwynN8fM/c26FiiwIw2n+eZR2buzLlnrnfO93z3emoorpILYm1tre7Wa9asWa61\nsEB7bVJSUsjLyyvznfeDkJ6eTkpKCnXr1qVy5colGlHpX5cHScQsLUqCokajMXqFYtAKmX379rFw\n4UKuXbtGpUqVqFq1KrNmzeL55x/+vjQV9IXA6dOnycjIoE2bNiU2OTk5OWrFh7S0NCZOnMiiRYuo\nU+evMEqNRnPnxsgsUAxgUpN5EBQn7uHDh4mKiuLgwYMkJCRQp04dvL29adu2Lb6+vtjb2z/UoqHv\nJzF2z3IoWTjRkIakH6arJB0q/VL0F9Oy0B70q/Aau8w9/OV0V7Lu73S6K90OlUdeXl6JRExFyJQF\n+rk/zZo1w87OrkzGfRg0Gg0rVqzg+++/Z9GiRXTq1AkhhFr8saKahAUGBrJixQo1gXT27Nn06NGj\nTMZWIrYUQkJCmD9/PrNnz6ZPnz6qLwRQ88QWL17M+vXrGThwIBMmTPinU5gFigFMajJlhZJlHB0d\nrWoyaWlpNGvWDG9vb7y9vfHy8uKJJ564r4VPMVM8+eSTODg4GNX+/jBZ7voRVEq/FP1S9rVq1Sp1\npNDNmzdJTEykVq1aNGvWzKhVePWd7qXNus/Pzy9Rt6ygoABra+sS2f6l1fDy8/OJj4+nSpUqODk5\nGT1BEeDUqVOMHDkST09PZs+ebVQ/W2BgIDVq1GD8+PFlOm5MTAy7du1izJgxVK1alRMnTjB27FhW\nrlxJkyZNDH7m/PnzhISEMGTIEJydne/nNGaBYgCTmkx5UlxcTFJSElFRUURFRRETE0NhYSHu7u6q\nqczNza3EoqHUurKyssLJycmoZoryynIvLi5WzWT6zm39hdRQAcjbt29z6tQp1bFsbCfuzZs3Vc20\nLCLb9DU85VFYWEi1atVKaDKGhIR+B8UWLVqUMJ0YC41Gw1dffcWaNWv473//S6dOnYw9pXITKNev\nX8fGxobjx4/j5ubG77//zoYNG1ixYgU5OTlqiwgld+1OlFSEf/h9mQWKAUxqMhVNfn4+R48eJSoq\niujoaE6ePEm1atVwc3PjwoULODk5MXHiRKP3d9DPcnd0dCz3nW5RUZG6iGZlZd1VADI/P181/T2s\nKbEs5nrq1Clu3bpV7oJNSkleXp6qxdy6dYuioiJVyNSqVQsLCwu1m6ixo8kUkpOTGTVqFG3btmXm\nzJlG1Ur0CQwM5Ntvv6VmzZp4e3szf/78B/6t3enjOH78OLNmzeK1117D29ub9u3bk5qaqn73U6dO\nUalSJRo3blxinHsJGQOYBYoBTGoyxkZKyaJFi1i8eDE+Pj7cunWLc+fOqYXo2rZtS9u2bdXQ5fLm\n9u3bpKSkkJOTY/Re7oWFhVy5coUzZ84A2gzvqlWrVqjTXx/9To6NGzfmqaeeMopgU8yIWVlZXLhw\ngZycHKpUqVLiupRlH/vSoNFo+PLLL1m/fj2LFy/m6aefrvA5dO3alcuXL9/1+qxZs2jfvr0adTdl\nyhTS0tJYtWrVQ50vPDwcT09PatSowerVq4mIiGDBggUEBQWxf/9+Jk6cSHJyMkuWLOGHH37Ax8fn\nQU9lFigGMKnJmAL79u2jXbt2aoSSksmsmMoOHTqk7oYVf0zr1q3L1BxmSr3cQasFpKamkpWVhbOz\nMzVr1qxQp/+dKKXuraysyr2T4/2QmZlJYmIidnZ26o5XETJKmK6U8q5s//Ks7pCYmMioUaNo3749\nn332WQkntCly5swZevbsSWxs7H1/Rl8ryc7O5pVXXuHatWs0b96coUOH8uyzzzJjxgysra2ZNm0a\nX3zxBSkpKVy9epUZM2Y8bCMws0AxgElN5lHh9u3bnDhxQhUyx48fx8rKijZt2qj+GCcnpwdaTPWz\n3B0cHIzu5FZKqN+P38aQ078sF9Li4mLOnz9PWlqaSfgmFHPb/SQoGqo0LIQoke1fvXr1hxYyRUVF\nhISE8PPPP7N48WI6duz4UOOVJ2lpaWrJnYULFxIVFcX69evv67P64cBbtmzBwsKCjIwM3nnnHX78\n8UeCg4MJDQ0lISGBZcuW0alTJ956660SJi0DocClwSxQDGBSk3lUkVJy69YtDh06pPpjUlJSsLW1\nxcvLCx8fH7y9vf9W01Cy3IuKiozeyx20O+zExESsra1p3rz5A2sBf+f0V4TM/XR9VJIl69ata/RI\nOyibBEVDCYcP0xEzISGBUaNG8fTTTzN9+nSj57r8E2+99RZHjx5FCIGDgwNfffWVKmDuh7NnzzJ5\n8mROnz5Neno6r732GtOnT6dSpUoMHz4cKysrQkJCWLJkCQUFBYwaNUrNfL+zwvADYBYoBjCpyTxO\nSCm5dOmSqsVER0dz7do1nJycSmT5W1hYEBYWho2NjVqixJhoNBpOnz7N9evXcXZ2Lpdy5X/n9FeE\njJLpr0ST5eTkmESypH6CoouLS5kXlrx9+3YJIZObm4uVlVUJIXNnZYiioiIWL17Mpk2bWLp0Kb6+\nvmU6J1NAWTeV7x0fH89LL73E2LFjGT16NAsWLODcuXO88847eHp6kpaWxtNPP82SJUvo3r17eWxA\nzALFACY1mccdjUZDfHy8moC5b98+bty4QYcOHejevTve3t64uroaxcylOLlTU1PV8i0V6bfR7/qY\nlZVFQUEBFhYW5OfnY29vj4ODg1F9JcZMUCwsLCyh5eXl5VG5cmX+97//8eSTT7Jx40a6devGtGnT\nTF4reRAuXbrEU089BWgTehWB+vzzz1OnTh02btxIVlYWEydOpEWLFgwaNIj69esTFhaGi4uL+n9V\nBlqJPmaBYgCTmsy/icWLFxMVFUVgYCBpaWlER0cTHR1NQkICNWvWpG3btqqprEGDBuXqwNXPundy\ncjJqOXfQOt3j4+O1LZX1eqYoIbrl7fQ3NJ+EhASTSlC8desWkydP5vDhw1SvXp2bN2/SqlUr1qxZ\nY+yplRm5ubksWrQICwsLJkyYwGeffUZMTAyurq4EBQWRkpLCq6++yueff46fnx/h4eEEBQUREBBA\nr1691A1RKUKBS4NZoBjApCbzb+JeDkEpJdeuXSthKlNyPhQh4+XlRa1atR76R6LRaNReHC1atDB6\nvk1xcTHnzp3j8uXLBp3u5e30vxP9zHtTCAJQiIuLY+TIkXTp0oUpU6ZQpUoV9b4pb5PpTz/9RGBg\nIPHx8URHR+Pt7a0emzNnDitXrsTS0pIvvviC7t27P/B5lN/HunXrOHz4MPn5+eTn5zN+/Hh69erF\n6NGjGTFiBCtXruSbb77hjz/+wMLCgo0bN9K1a9eK6CxpFigGMKnJmDFMcXExp06dIjIykujoaA4f\nPkxubi5ubm6qkHF3dy+VZqE4le3t7WnUqJFR+7nDX6G3Snmb0vTwuNPpb2lpWaKMfWkc2wrZ2dnE\nx8ebVILi7du3WbhwITt27ODLL7+kbdu2FT6H+Ph4LCwsGD58OMHBwapAiYuLY+DAgURHR3Pp0iW6\ndu1KUlJSqa+bRqMhPDycatWq4evry8GDB9mzZw87d+5k0qRJ9OjRg/379/Phhx/yww8/4O7uTo8e\nPXBxcWHBggXqOOWklehjEgLF3LHRTKmxsLDAyckJJycn3nrrLUBrVz927BiRkZEsX76c2NhYrK2t\n1dBlHx8fmjVrdtfCnJeXR1JSEkIIPD09jW5zV5I38/LycHd3L7XT3cLCQu3TrvQb13f6p6SkGHT6\n3+t7FxcXq1Vp9TsoGpvY2FhGjRpFt27diIiIMJo/ydXV1eDrW7Zs4fXXX6dKlSo0bdqU5s2bEx0d\nTYcOHe577B9//JEBAwaQn59PcHAwFy5c4MMPP6RPnz6cOXOG2NhYnn32WZ5++ml69uzJJ598wpYt\nW1ixYsVd1+NxK8F/L8wCxUyZULlyZXx8fNRMXyklmZmZHDp0iMjISD799FPVwe7l5UXr1q2JiIig\nVatW9OjRo1SFE8sD/RwXBweHMm1RbGVlRZ06dUqYqPSd/hcuXKCgoICqVauWEDJKqLSdnR3e3t5G\n19qUeS9YsIBdu3axbNky2rRpY+wpGeTixYu0b99efd6wYUMuXrx4359PT08nKysL0N4bv/76K717\n9yYgIACALl26EBkZyb59++jTpw8zZ87khRdeICEhgZYtWwJl7nR/JDALFDPlghACGxsb/Pz88PPz\nA/5KBFy5ciVjxoyhSZMmhIeHs3fvXlWLad26dYU3KMvNzSUhIQFra2u8vb0rxMlduXJlnnzySdXH\noGT6Z2VlcfXqVU6ePElRURE2NjYIIbh586bRSqYoHD9+nFGjRtGjRw/Cw8MrTCv5u3Ipffr0KdNz\nKb4SW1tbhgwZwtChQ1m8eDG7d+9m+/btrF27lkGDBtGnTx/Onj1LREQEjRo1wsvLi127dpW4d/5t\nwgTMAsVMBWJhYUHjxo0pLCwkMjISBwcHioqKOHnyJJGRkaxbt46JEyeq5i/9BmXlsZAWFxdz9uxZ\n0tPTjR4EIISgatWq5OTkkJmZiaOjI/b29uTm5pKVlUVaWhpJSUlIKUvdu/5hKSwsJDg4mL1797J8\n+XI8PT3L9Xx3snfv3lJ/pkGDBpw/f159fuHCBRo0aPCPn1Pus+nTp9O7d2/y8vKYPHkyCxcuJDk5\nmdDQUF544QXq16+Pm5sbkZGRqrnSysqqInwlJo3ZKW/GpJBSkp2dXaJBmZKRrt+g7GEbiilOd1tb\nW5o0aWL03WRhYSGJiYkUFxf/bYKiUjJFvy5XWTj974XS6rp3795MmjTJJEKUDdG5c+cSTvmTJ08y\naNAg1SnfpUsXkpOT/3Fjcv78eQIDA7G0tGTBggXcuHEDf39/goKC8PHxYc6cOaSmpnLjxg0++OAD\nunfvbip1yUxCipkFigmwc+dORo8ejUajYdiwYUyaNMnYUzIplCQ+/QZlly9fxtHRsUSDsho1avzj\nQnr79m2Sk5PJz8/HxcXF6CVl9BMUHR0dsbW1LfUY95PpX9pgh4KCAoKCgggNDeWrr77Cw8Oj1POq\nCDZt2sTIkSO5evUqtWvXxtPTk127dgFak9iqVauwsrJi0aJFvPTSSyU+e2e2e05ODitWrGDevHns\n2rVL/c5r1qwhJCSE0NBQCgsL+fzzz2nSpAnvvvuuOo4JaCVGnwCYBYrR0Wg0tGjRgj179qhl6det\nW6c69swYRqPRlGhQduTIEbVBmSJk3Nzc1B11cXExly9f5uzZsyZRMRnKN0GxoKCghJAx5PS/1/mO\nHDnC6NGj6du3Lx9//LHJaiUPg77D/NixY9SoUYNGjRpx8+ZNxo4di4ODAzNmzFDf+8Ybb2BpackP\nP/xQYhwTESZgFigGManJVAR//vkngYGB6q5qzpw5AHzyySfGnNYjSX5+PkeOHCnRoKxGjRo0b96c\nY8eO8d577zFw4ECjZ94bI0FR3+mvCJqioiKqV69OzZo1OXv2LC1btiQkJISIiAiWLVtGq1atyn1e\nxubjjz9m9+7d+Pv7k5iYyLp169izZw8bNmzA39+fl19+GYBz584RGxtLjx49TEmI6GMSE3rsnfKr\nVq3Cw8OjRAatKXHx4kU1XwG04Y1RUVFGnNGji7W1NR06dFBzDYqLi5kyZQo///wznTt3Ztu2bSxd\nupRGjRqVaFCmRFJVBPoJij4+PhUWtaU4/atWrUr9+vWBvzL9MzMzCQkJ4ciRIwgh6NatG4cOHXrs\nBMqdYbxr166lqKiImJgY5s+fz+7du5kxYwZTpkwhMTGRjRs30qpVKxo3bqw+4N+TU/IgPPYCJT4+\nnoyMDLy8vNSbKSoqioyMDLp162bU/h5myhcLCwueeeYZAgMDS5i+Tp8+TVRUFPv27SMoKIjs7Gxc\nXV1VU5mHh0eZJ1jqJyi6uroatdulghACKysrVqxYwY0bN9i1axeOjo4cO3aMU6dOlfv571U25cyZ\nM7i6uuLs7AxA+/btWbZs2QOfR7HCKL//77//nr59+9K3b1969uxJQEAAWVlZbN68mSFDhtCnTx9e\neeUVLl68SHZ29kN+y38Xj/1q+vzzz7Nw4UImTJgAaHeIs2bNwsvLix49egDaH7uU0igx/g8a3vgg\nODg4qLkMVlZWHDp0qFzOY0rc6Yi1sLDA0dERR0dHBg0aBGgd9cePHycqKopvvvmGEydOYGVlhZeX\nV4kGZQ8aCabfQdFUEhQBDh48yNixY3n99dcJDQ1VN1ft2rWjXbt25X5+d3d3/ve//zF8+PC7jjk6\nOnL06NEyOY+iUYSHh7N79262bt3KpUuXmDhxIleuXEGj0fDTTz8hhKB27dq88cYbnDhxghkzZpg3\nnKXksb9amZmZJZyKmzZtIiMjg8DAQAoKCqhSpcpdP/D09HQGDx7Mzp07y31+Pj4+JCcnc/r0aRo0\naMD69etZu3ZtuZ3v999/p169euU2/qNIpUqVVPPXBx98gJSSmzdvqg3Kpk+fTkpKCvXr1y/RoMzW\n1vZvzR9FRUWkpKSQk5NDq1atjB5RppCXl8fs2bOJjo5mzZo19yxfUt6U53nv9HNs27aNkSNHMnfu\nXG7dukVoaCi+vr7Y2tpy/vx5Dhw4wPbt2/H391erPZjzSkrPYy9Q6tSpQ8OGDcnMzFQTk4YMGQLA\n5s2bmTVrFrVq1eKtt95iyJAhVKpUibCwMM6cOWNwvLK+waysrNSmOxqNhnfffRc3N7cyG99M6RFC\nUKtWLbp06UKXLl0A7f/7xYsX1aiykJAQMjIyaNGiRYkGZUoOyKlTp0hPT6dx48Y4OzubzKIUFRXF\nuHHjGDRoEL///rvJ7sBPnz5NmzZtqFmzJjNnzuTZZ5+9788aqpwdExPDpEmTGDBgAC+88AJbt25l\n2bJl/Pjjj/Tt25cvvviCnJwc1q1bV8IcaSr/b48Kj32UV25uLl26dGHbtm2MHTsWd3d3AgICsLGx\nITc3F9C2c502bRoBAQH4+PgwfPhwnnjiCYKDg0lKSlKruxq6uTQaDUIIkzFj/B1NmzZVHdDDhw/n\nvffeM/aUHmk0Gg1xcXFqAmZMTAyFhYUANG7cmP/7v//Dzc3NJBbtvLw8Zs6cSUxMDF999RUuLi4V\nct77KZtyZ1JiQUEB2dnZ1K1bl8OHD+Pv78/JkyfvqzCm/obvm2++wdnZmQ4dOrB27VpWrlzJvn37\nAG32/bhx4wgICGDEiBHk5uaqGuQjWoPLJCSf8e/0cqZatWpcvXqVQ4cOER0dzfz587GxsSEjI4NZ\ns2YRGRlJy5YtWbdunWpT379/P8HBwWRnZzNixAjGjRtH9+7dCQsLo3LlyiXqPRnyuxQXFwOmV8sn\nIiKCBg0akJ6ejp+fHy4uLnTq1MnY03pksbS0pFWrVrRq1Yphw4axb98+xowZw4ABA6hSpQoLFy4k\nISGB2rVrq1qMj4/PA/d8f1D+/PNPJkyYwJtvvklwcHCF+gofpGxKlSpV1NDutm3b4ujoSFJS0j0j\nNQsLC0lJSaFly5YIIYiLi2PAgAG4ubmxbds26tWrxzvvvIOHhwfTp09n2rRpgNbcHBkZyUsvvYSj\noyNw775AZu6Px16gZGdn07p1az744ANGjBjBk08+SV5eHiEhIWRlZbF161ZiYmLYvn07zZo1A7Tl\nF9q3b8/58+extrZWM2a3b9+OlZUVnp6e3Lp1i7lz55Kbm8vzzz9Pr1691GJ5piZIFBRnv62tLX37\n9iU6OtosUMoQNzc3wsPDSzRTUloZK6ay1atXc+nSJZo2bVqiQVnNmjXLXMjk5uby2Wefcfz4cdav\nX0+LFi3KdPzy4urVq9SpUwdLS0tSU1NJTk5Wf5uGWLx4MSkpKXz00Uc4OTlx8OBBBg8ezIQJE0hJ\nSWHjxo18//33jB49mh49ehAbG8uJEyeYPn06O3bsKKH5mIXJw2GaK18ZUr16dTp06ICHhwejR48G\n/lKLW7RoQb169cjPz6dq1ao0btyYhIQEqlWrRu3atdUsZnt7e4qKisjPz8fV1RVLS0teffVVXnvt\nNbp168bmzZuJiYkBtLbaefPmsWPHDrX8tSE0Go2qyVQEOTk53Lp1S/179+7duLu7l9n47777Lra2\ntiXGvH79On5+fjg5OeHn58eNGzfK7HymiJ2d3V2d+YQQ2Nra0qtXL2bOnMnu3bs5fvw48+fPp1mz\nZvz666+88sorPP3007z33nt89dVXJUxnD4KUkv379+Pn54ejoyO//fabSQqTTZs20bBhQ/78809e\nfvlltaPiH3/8gYeHB56enrz66qssW7bMYLfMzZs3k5qayoABAygsLCQsLAwpJUlJSURGRgLQqFEj\nOnToQHZ2No6OjoSFhfHpp58SHx9PcXExFy5cQKPRVPh3f1x57H0o92Lr1q2MGTMGX19fTpw4Qf36\n9fntt99YunQpW7ZsYdeuXQQHB5OSksKyZctISUlhwYIFam/onj17MmDAANq1a8fly5e5ePEiq1at\nwtfXl0GDBvHnn38ihGDt2rX/WLpCsdlGRERw8eJFOnXqhL29fZl+39TUVPr27Qtoo48GDRrE5MmT\ny2z8P/74gxo1ajB48GBiY2MBbRZynTp1mDRpEnPnzuXGjRvMmzevzM75OFFQUKA2KDt48CAnTpyg\natWqauiyj48PTZs2/UftNycnh+nTp3Py5EmWL1+Ok5NTBX2DiiU3N5chQ4YwZcoU3N3d2bhxI1u3\nbiUgIAB7e3vGjRvH8OHDeemllzhx4gSBgYFs2LABS0tLrl27xuzZs4mLi2PlypXlFqZfwZiED+Vf\nIVCKioru6RhNS0vj2LFjVKtWjU6dOtG/f3+aNGnC559/zpQpU7hy5QrLly9n2bJlbNmyhS+++ILt\n27eTmJjI+++/z08//URiYiK+vr688cYb+Pj48Pvvv+Pk5KSGJevRs3oqAAAT40lEQVRz69Yt5syZ\nQ2ZmJp06deK1115T1eyOHTvi6upK8+bNGTNmjKlUMb1vzpw5Q8+ePVWB4uzsTGhoKPb29qSlpdG5\nc2cSExONPMtHA6VBWXR0tOr0V0LLFX+Mt7c3devWRQiBlJKIiAgmTpzI0KFD+eCDDx5b841iYQgO\nDubSpUtqq91PP/0UKSUDBw4kPj6eMWPGMG3aNFauXEn37t0JDAzEwsKCoqIiYmNjK7wMfzljFigG\nqLDJGIrkkFIipSQvL4/q1atz8OBBJkyYQHFxMUVFRTRt2pTvvvuOKVOmUFBQwPz58+/6/Pr16/nt\nt9/o1KkTb775JkII1TZ+7tw5Fi5cSH5+Pr6+vvzyyy989NFHODs70717dy5evEhERMRdu8q8vDzW\nrFnDt99+S6NGjfD19aV///4mt7O6U6DUrl2bzMxMQHttbGxs1OdmSk9xcTHnzp0jMjKS6OhoDh48\nSFZWFi1atCAjIwMLCwuWL1+uOpgfd8LCwtixYwdvv/02LVu25OzZs0ydOpXnnnuOd999l927d5OU\nlIS9vT39+vUDTKqYY1ljGl9KWURN5GEUNBrN3x6/efOmjIuLk9HR0TIvL09mZWXJvn37yn79+smZ\nM2fKX375RWZnZ8vCwkIppZRXr16VzZs3l6tXr5ZSSvX1r7/+Wo4YMUKeOXNGSillUFCQ7Nu3r5RS\nyuHDh0t3d3c5YcIE9biUUl67dk1OnjxZdu3aVcbHx8uffvpJjho1SgYFBZX5dXhYTp8+Ld3c3NTn\ntWrVKnG8du3aFT2lx57CwkIZExMjp06d+o/3cVkzfvx46ezsLFu1aiX9/f3ljRs31GOzZ8+Wjo6O\nskWLFnLnzp0PdZ7i4mKDr6ekpMjhw4fLdevWyezsbCmllBs3bpSvvvqq3Lt3713vr+jrU8EYe+1G\nSvn4R3ndD/eySxcXFyOE4IknniiR1Wttbc3SpUs5cOAA4eHhZGRkUL16dQYMGECNGjXo168fNjY2\najE5ZXyl54XSPz0nJwcvLy9AG10yevRohg0bRnFxsapBbdiwgYSEBNatW0e9evVwcXHB39+f0NBQ\noOSO68aNG3zzzTd4eHjw/PPPc/z4cdzd3Y1WftzOzo60tDTV5PUgvT7M/D2VKlWiTZs2Runt7ufn\nx5w5c7CysmLixInMmTOHefPmERcXx/r16zl58iSXLl2ia9euJCUlldoEp/wG9DUK/fvd0dERHx8f\n9u/fT7Vq1ejduzevvPIKCQkJBrV3U42+fJwwX+G/Qf9mlneYBuvXr88rr7zCwoULefvttwFt29D2\n7duzceNGRo8eTefOnYG/QhHd3d3V3gugjdF3d3ensLCQEydOqJE4Qgg1AmzHjh34+vpSr1491fRm\nZWVF165d1fcq/Pzzz8TExNC1a1eOHj3KtGnTOHLkCKD9cWo0mru+R3nSu3dvVq9eDcDq1avLrP+3\noYiywMBAGjRogKenJ56enuzYsaNMzmXm3ugXV23fvj0XLlwAYMuWLbz++utUqVKFpk2b0rx5c6Kj\no0s9viIAdu3axdSpU4mLiyMvLw/Q+kUB3nnnHVxdXdm+fTvffvstAP/3f/9XYYmbZkpiFij3iSG7\nq6JJKLi4uBAQEMDKlSt544037nr/yy+/zPXr1+nXrx+jRo2iSpUq9O7dm/T0dLKzs/H19VXPpQih\n1NRUOnbsCGh/YFZWViXOqQiImJgYYmNj6d+/PwBxcXE4OTlhZ2enftZQtn9ZCZiBAwfSoUMHEhMT\nadiwIStXrmTSpEns2bMHJycn9u7dW2adKN955x2DddbGjh3L0aNHOXr0qFr400zFsGrVKrUQp6GW\nDBcvXiz1mBqNhvHjxzNr1izq1q3L5MmT2bx5M6AtWaQkIQ4dOpRhw4bx9ddfs2fPHjIyMsrmS5kp\nNWaT10NgyKmvmMkMqdc1atTg22+/JTQ0lGvXrjFr1iwsLCyIiYkhLy8Pa2trVaUXQqDRaOjYsaNa\njdhQVWRFQOzatQs7Ozu1F0hKSgp169bF1taW7Oxsli9fjpSSl19+ucTuTV/AKHN/EKflunXrDL7+\n22+/lXqsf6JTp073rLVmpmy5n9Ips2bNwsrKyuAmqjTcmaUeGxtL06ZNCQ4OZunSpRw6dIi6deti\nZ2dHly5d1PdWrlwZHx8fQkJCsLCwMOeVGBGzhlKGKJrF39lqbWxs6Nu3LwEBAWoRumeffVZV1/U1\nBktLS3r37s2qVauAv7SMtLS0EmPevn2b1NRUnJycePLJJwG4dOkSjRo1omrVqrz44otkZGRw7tw5\nPvzwQ86ePQtoY/l3795NcnKyOv6dwqSizWQPw5IlS/Dw8ODdd9997JMoK4q9e/cSGxt710MRJt9+\n+y3btm1jzZo16r3zoC0ZLC0tKSgoYMOGDdy+fVutcBEYGEh4eDj79u2jsLCQrVu3lhhfOW/r1q1p\n1aqV2VdnRMwCxQgomoaCjY0N7dq1Q0p5lzDy8/OjTZs2uLi48PLLLzNu3DgiIiLUcQCuXLmCtbW1\nKqDS0tIoKCjA3t6e8PBwNBoNs2bN4r///S+urq788ssvAAQEBPDFF18wePBgJk6cyPjx48nNzSU/\nP19dkO80k0kpS2T5m4qwef/99zl16hRHjx5VE9vMlC87d+4kKCiIrVu3lijN37t3b9avX09BQQGn\nT58mOTlZNefeib42ERsbS5cuXQgODmbkyJGEh4cjhGD//v1MmTIFJycnKleuzIULF9RNkBnTwixQ\njIAhTQAM+2msra0JCgriwIEDjBgxgjZt2qg+FYWqVatSo0YN1WGZlJSEtbU1Tz31FElJSWokWU5O\nDh06dCA+Pp6IiAhiY2PZtm0bf/75J8ePHycpKYlq1aoRGxuLv78/7dq1IyAgQB1XMccpWphGo+Ht\nt9/m9OnT5ObmcuXKlQotJ6OPnZ2dOq+AgIAHcgKbKR0jRozg1q1b+Pn54enpyX/+8x9AW9Osf//+\ntGzZkhdffJGQkJC7IryUjYilpSW5ubn8+uuvRERE8OmnnxIdHY2dnR07d+4kKysLHx8fhgwZwvDh\nw0lNTWXGjBm88MILFf59zfwzZh/KI0KdOnUMdh8sLi6mbt26nD9/Xt0FHjx4ECEETk5OfPfdd6qg\nysrK4tixYzRq1IiEhAR69eoFaJuQeXt7q/WjXF1dCQsLIzMzk+XLl7N371569OjBl19+qdY68/Ly\nYsCAAXz33XcAHDhwgJMnTxIQEADAjz/+SN26dXnuuecqJGxZCU8GbY2osqpTdv78eQYPHsyVK1cQ\nQvDee+8xevRorl+/zoABAzhz5gwODg5s2LABGxubMjnno0JKSso9j02ePPlvS/so9+TBgwd5//33\nadiwIWFhYWppnp49e7J27VrWr1/PtGnTaNasGZcuXWLJkiXq/aQfQmzGNDBrKI8Qd5rK4K/AADs7\nO9V/8swzz9CnTx80Gg1Dhgzh/PnzTJ06lalTp5Kfn89//vMfQkND1fdnZmaSnp5O69atycvLY/Hi\nxXh5edG3b19+/vlnDhw4gKWlJYcOHeLIkSN4eHiQkZHB7t276dOnD9euXWP06NFs376dnJwcbt++\nTVRUFEePHjUoTB7WTGYoouzjjz+mVatWeHh48Pvvv7Nw4cKHOoeClZUV8+fPJy4ujsjISEJCQoiL\ni2Pu3Ll06dKF5ORkunTpwty5c8vkfI8zdzrLN2/ezGeffUZAQACbN28mKCiImTNnAtrS8m3btiUi\nIoLDhw8zbNgwpk6dSqVKldRxzMLE9DBrKI8Qf+fsHz9+PKdOnQK0OQEKzs7OfPLJJ+zZswcHBwdG\njBhB7dq1eeaZZ4iKiuLIkSNs3bqVQ4cOMW3aNObOnUtmZia//PILlpaW+Pv707p1a9LT07GwsGDE\niBEMHDgQgClTpuDi4oKlpaVaFuTDDz+kV69eVKlShaKiIqKjo2nZsqWaewN/LQQP2jfGUETZ0KFD\nSzXG/WJvb69qPkqC68WLF9myZYuaXPr222/TuXNnc+HLf0Axe127do169erh6OhI5cqVSUlJ4fbt\n2wQEBLBp0yaGDh3KypUr6d27N/Xq1VMjF4G7ohzNmBZmgfKYYG9vT/369YG765S1b9++hJABGDJk\nCHl5ecybN486derQpEkT6tWrR+3atbGwsKBBgwaEhYWRkZGBp6cnZ8+eRQih+mOklBw7doxhw4ZR\nWFhI586defPNN2nTpg0REREcPXoUW1tb/vjjD/Lz81m2bBktWrQgMTERS0tLmjdvfk9BUlRUxJkz\nZ7CzsyvRjtXYnDlzhiNHjtCuXTuuXLmiCpr69etz5coVI8/ONNG/FwsLCxk4cCBXr16lY8eOTJs2\njbfeeouwsDBCQ0Px8/Nj7dq11KlTh7fffptOnTrx4osvApQIpzdjuphNXo8Ryo/tzoVayZJXNAKl\nRIuzszMTJ04kLy+P5557jsqVK+Pv78+2bdto374969ev59KlS7i4uJCYmIi1tTVNmjQB4ObNm1y4\ncAEfHx/i4uLIzMxUtZDU1FRq1arFJ598omasHzhwAICPPvqI/v378+KLL9K7d29OnjxJYmIiqamp\n6nyLioqIj49Xc1iUBFJjRpRlZ2fTr18/Fi1adFcrWvNCZxh9YZKcnMzhw4fx9fXl+++/5/r163z8\n8cf4+/tTq1YtDhw4QEpKCrVr12b//v139ZA3X99HBGMXEzM/Kv4BVAZeAr4DdgMfABYG3uMA9Nc9\n/xAI0jvuBRzX/f028F+9Y58AE4EauudRwLO6v1OBN3R/bwK+BaYCV4CRutfrAvWNfZ30vk8lYBfw\nkd5riYC97m97INHY8zTFB1AVWAkcBQ4BA/Wu2UGgm+4++0HvmFIF3cIYczY/Hvxh1lD+hUgpC6WU\nv0opB0spu0kpl0opi4UWCyGEhe49Z6SUG3SfCQGm6A1ji3ZRBbgJtBVCNBBC1AFqAtellNlCiCeA\nasApIYQDkAP8qPtcNnBTSvkZ8BzQWQhRA2gHLBVCNAMQQvQXQvxHCNFe3GOrqpu35b2OPyi68VYC\n8VLKBXqHtqIVpOj+3VKW530cEEK0BnagFSYdgRSgoRCivpQyDZgHLEG7mQiWUq4DkDppIqU0Tgy6\nmQfGLFDMqEgtxcoP+c7FWUpZoPf3Tinla7qnMcBJYALQFaiNVlgA+AIZUspLQEvgqpSySAhhC2iA\nzbr3WQFNpJTZQB3ggpQyVQgxB+gBuAHBQG9DQkM3b42yGCnzLwMh8zTwFvCCEOKo7tEDmAv4CSGS\ndd/5Xx3mJYQwtJbkAe2BfCllLrAM8AB8hRCVpJQ/AxuBp6SUR3XjmG1bjzBmp7yZe6K/ON+JTosp\n1r3vLDBc71g4oOwuXYFTur+fB5SaGc7AbUCp5NcaUCoINgHOCSHqAwMBbynlNd3YNe8QGpWAz3Xj\nnQC+k1LG6s2/RKyqbsESusNSee1e31VKGcG9mxd1ucfr/yp010/ZhAxCayKMlVIeFkKMB4YLIb6W\nUoYKITwBf7T/14ellJ/oj/V395wZ08esoZh5IPTNEYq5Se9YmpTyik7oLAH+ozt0EFir+9sD7UKt\ntHD0BaKFEJWBJ4FYtGaSZCnlNSFETd3CdfOOqbwMNADGAteAVro5NRBCLBBCjBNCqHGnelqYIkws\npJRSCNFXCGFdBpfmX4OileiuXw0hxBpgMNrNxK9CiGY6U2kioJgLV6IVJndXnDTzyGMWKGYeGsXc\nZOh1nRC4rXu+QUq5S3f4e7Q2dGVhcQYOA02Bp9AuOtbAJd1nb95j91oZaAvUklIGSSnXCSHaAd8A\nV9E6fz/QLXgNhRBLhBABQghX/TmiNb2opr7y8Mc8TuhrJToc0G4YeqC95lUBJbt0GuAvhHhJSnlL\nSjlFSln6evZmTB5T6ylv5jFGCGFpSPDoHbdAuxi9BGyVUqYLIVYDt4A9aBesK8oYQggrnT/mebR+\njuNSykVCiM/RRpi9rzd2JeAJwA+tP6YVME1KeVwI4QuslVI21zflmbkbffOgEKIqsAKYJ6U8IYSo\nCSwCcqSUI4UQKcDXUsq5QojXgAQp5Yk7xzHz+GDWUMxUGH8nTHTHi6WUF6WUX0sp03UvzwPOoY2k\ncrljjDpCiIZSyt+BOcBQIUQjwAXYCaCLGkOnJTmidfj/D6328qZunO5o/S8AjYQQE4QQs4QQHfWd\nzYYczzpz31OluAyPNHrC5F3gM7SmymW6wzloI/qW654fAoYIIWyllD8pwkR/HDOPF2anvBmTQjEz\n6YWOxgFx93h7XeA7IUQu2vyW42hNZVboHOm6qDF0WsxwtCY2D2AIfwUSdAbW63bcw9GGt14BxqAN\na/1DN5YhzaUJME/nbD4FLJNSPtYhxEKI9sA4tFrhWrT/B4FSykAhxDXgPV1AxWWgn97mwMxjjllD\nMWNS6JzmhkJ/77pXpZTxUkof4D20eSEjdIv+FGCsEGKhEGKAzhTTGq25bIzOFJbKX4LKDa1JbRDw\nGtrordOABDro5jFMCOFhYMq5wBgpZQvdHLoLIWopc9f920AI8aIQosqj5JcRQjgJIeYJIVrccegp\nYLuUMkZKeQToB4zTaYcz0EbyJUopR0opYx+l72zm4TALFDMmjU6+aAxpB3raTKKUcpOU8obu+SG0\nUV8X0WojuUAk0FUIMUloc1uaAMd1wsZKSnkGbWBACFpz2UC0+TQFOmGWDhQamN8V/gos2AvYoBVe\nSvTTa2jNdj+grQQghRBVdPO30f8epoROw1gN/CalTLpDoF8GBimRfVLKJLTCeZWU8oouOOJT3TgW\nZvPWv4f/B1w4tV9gDWn9AAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f0d9c2bba90>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# For a given model, visualize how it performs for a chosen parameter or pair of parameters\n",
    "from src.common import choose_classifier\n",
    "from src.visualization.visualize_parameters import get_all_grid_searched_parameters\n",
    "from src.visualization.visualize_parameters import choose_parameters_to_analyze\n",
    "from src.visualization.visualize_parameters import analyze_single_parameter\n",
    "from src.visualization.visualize_parameters import analyze_two_parameters\n",
    "\n",
    "classifier_class = choose_classifier()\n",
    "parameters_list = get_all_grid_searched_parameters(classifier_class)\n",
    "if not parameters_list: # grid search results not found\n",
    "    exit(-1)\n",
    "tested_parameters = list(parameters_list[0][0].iterkeys())\n",
    "parameters_to_analyze = choose_parameters_to_analyze(tested_parameters)\n",
    "\n",
    "# if we choose a single parameter, draw 1D plot\n",
    "if len(parameters_to_analyze) == 1:\n",
    "    analyze_single_parameter(parameters_to_analyze[0], classifier_class, parameters_list)\n",
    "\n",
    "# if we choose two parameters, draw 2D plot\n",
    "elif len(parameters_to_analyze) == 2:\n",
    "    analyze_two_parameters(parameters_to_analyze[0], parameters_to_analyze[1], classifier_class, parameters_list)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "anaconda-cloud": {},
  "kernelspec": {
   "display_name": "Python [conda root]",
   "language": "python",
   "name": "conda-root-py"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}