Text Classification using scikit & nltk.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "from __future__ import print_function"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 264,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "TensorFlow version: 2.0.0\n",
      "Keras version: 2.3.1\n",
      "NLTK version: 3.4.5\n"
     ]
    }
   ],
   "source": [
    "import sys, csv, re\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "\n",
    "from collections import defaultdict\n",
    "import tensorflow as tf \n",
    "print(\"TensorFlow version: \" + tf.__version__)\n",
    "\n",
    "import keras \n",
    "import nltk\n",
    "print(\"Keras version: \" + keras.__version__)\n",
    "import keras.backend as K\n",
    "from keras.layers import Dense, GlobalAveragePooling1D, Embedding, AveragePooling1D, Flatten, concatenate\n",
    "from keras.callbacks import EarlyStopping\n",
    "from keras.models import Sequential, Model\n",
    "from keras.preprocessing.sequence import pad_sequences\n",
    "from keras.preprocessing.text import Tokenizer\n",
    "from keras.optimizers import Adam\n",
    "from sklearn.feature_extraction.text import CountVectorizer\n",
    "from keras.utils.np_utils import to_categorical\n",
    "from sklearn.metrics import accuracy_score\n",
    "from sklearn.metrics import confusion_matrix\n",
    "from nltk.tokenize import word_tokenize, sent_tokenize\n",
    "from keras.utils import to_categorical\n",
    "\n",
    "print(\"NLTK version: \" + nltk.__version__)\n",
    "from collections import defaultdict\n",
    "from sklearn import preprocessing\n",
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "np.random.seed(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "def mapped_text(x):\n",
    "    x_new = []\n",
    "    for i in range(len(x)):\n",
    "        x_clean = clean_str(x[i])\n",
    "        x_strip = strip_char(x_clean, character_list)\n",
    "        x_new.append(x_strip)\n",
    "    return x_new"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "def strip_char(text, list_char):\n",
    "    newString = \"\"\n",
    "    for i in range(len(text)):\n",
    "        if text[i] in list_char:\n",
    "            newString += text[i]\n",
    "        else:\n",
    "            newString += \"a\"  # replace character that not in the char list\n",
    "    return newString"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "strip_char(\"hello\", character_list)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "import re\n",
    "character_list = ['a', 'b', 'c', 'd', 'e', 'f', 'g', 'h', 'i', 'j', 'k', 'l', 'm', 'n', 'o', 'p', 'q', 'r',\n",
    "                  's', 't', 'u', 'v', 'w', 'x', 'y', 'z',  '1', '2', '3', '4', '5', '6', '7', '8', '9', '0',\n",
    "                  '-', ',', ';', '.', '!', '?', ':', '\"', '`', '/', '\\\\', '|', '_', '@', '#', '$', '%', '^',\n",
    "                  '&', '*', '~', \"'\", '+', '=', '<', '>', '(', ')', '[', ']', '{', '}', '\\n', \" \"]\n",
    "\n",
    "\n",
    "def clean_str(string):\n",
    "    string = re.sub(r\"\\'s\", \" \\'s\", string)\n",
    "    string = re.sub(r\"\\'ve\", \" \\'ve\", string)\n",
    "    string = re.sub(r\"n\\'t\", \" n\\'t\", string)\n",
    "    string = re.sub(r\"\\'re\", \" \\'re\", string)\n",
    "    string = re.sub(r\"\\'d\", \" \\'d\", string)\n",
    "    string = re.sub(r\"\\'ll\", \" \\'ll\", string)\n",
    "    string = re.sub(r\",\", \" , \", string)\n",
    "    string = re.sub(r\"!\", \" ! \", string)\n",
    "    string = re.sub(r\"\\(\", \" \\( \", string)\n",
    "    string = re.sub(r\"\\)\", \" \\) \", string)\n",
    "    string = re.sub(r\"\\?\", \" \\? \", string)\n",
    "    string = re.sub(r\"\\s{2,}\", \" \", string)\n",
    "    return string.strip().lower()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "def load_data(data_path):\n",
    "    \"\"\"\n",
    "    data_path: location of the data file\n",
    "    return: value(text) of data x to train and labels which were authors' names. \n",
    "            Labels are converted into numbers by Label Encoder. \n",
    "    \"\"\"\n",
    "   # load data \n",
    "    data = pd.read_csv(data_path, header=0)\n",
    "    x = np.array(data['article'])\n",
    "    y = np.array(data['class'])\n",
    "    x_mapped = mapped_text(x)  # mapping with listed character\n",
    "    \n",
    "    # transform y (label of author name) into integer label (one hot vector)\n",
    "    auth_class = list(set(y))\n",
    "    le = preprocessing.LabelEncoder()\n",
    "    le.fit(auth_class)\n",
    "    y_numeric = le.transform(y)\n",
    "    return np.array(x_mapped), np.array(y_numeric)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 412,
   "metadata": {},
   "outputs": [],
   "source": [
    "# C50/C50_train.csv or C10/C10_train.csv\n",
    "x_train, y_train = load_data('C10/C10_train.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [],
   "source": [
    "def turn_to_integer(sequences, indices, mode):\n",
    "    new_sequences = []\n",
    "    for input_list in sequences:\n",
    "        new_list = []\n",
    "        if mode == 'word':\n",
    "            for word in word_tokenize(input_list):\n",
    "                if word in indices:\n",
    "                    new_list.append(indices[word])\n",
    "                else:\n",
    "                    new_list.append(indices['unknown'])\n",
    "            new_sequences.append(new_list)\n",
    "        else:\n",
    "            for character in input_list:\n",
    "                if character in indices:\n",
    "                    new_list.append(indices[character])\n",
    "            new_sequences.append(new_list)\n",
    "    return new_sequences\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "def create_ngram(input_list, n):\n",
    "    return set(zip(*[input_list[i:] for i in range(n)]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "def add_ngram(sequences, token_indice, ngram_range, mode):\n",
    "    new_sequences = []\n",
    "    for input_list in sequences:\n",
    "        new_list = input_list[:]\n",
    "        ngram_list = []\n",
    "        if mode == 'word':\n",
    "            ngram_list = new_list\n",
    "        for i in range(len(new_list)-ngram_range+1):\n",
    "            for ngram_value in range(2, ngram_range+1):\n",
    "                ngram = tuple(new_list[i:i+ngram_value])\n",
    "                if ngram in token_indice:\n",
    "                    ngram_list.append(token_indice[ngram])\n",
    "        new_sequences.append(ngram_list)\n",
    "    return new_sequences"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 286,
   "metadata": {},
   "outputs": [],
   "source": [
    "# average length of sentence\n",
    "def average_sentence_length(doc):\n",
    "    sent_tokenized = sent_tokenize(doc)\n",
    "    average_sent_len = sum(len(sent) for sent in sent_tokenized)\n",
    "    return average_sent_len\n",
    "def average_nwords(doc):  # per sentence\n",
    "    total_nwords = len(word_tokenize(doc))\n",
    "    total_nsents = len(sent_tokenize(doc))\n",
    "    return total_nwords/total_nsents # total number of words / total number of sentences\n",
    "def average_word_len(doc):  # per sentence\n",
    "    word_tokenized = word_tokenize(doc)\n",
    "    total_word_len = sum(len(word) for word in word_tokenized)\n",
    "    total_nwords = len(word_tokenized)\n",
    "    return total_word_len/total_nwords # total length of words / total_nwords\n",
    "def average_ncharacters(doc):  # average number of characters in a sentence\n",
    "    total_ncharacters = sum(len(word) for word in word_tokenize(doc))\n",
    "    total_sentence_len = len(sent_tokenize(doc))\n",
    "    return total_ncharacters/total_sentence_len  # total number of characters / total number of sentences \n",
    "def freq_ndigits(doc):  # proportion of the use of digits in a document\n",
    "    return sum(d.isdigit() for d in doc)/len(doc)\n",
    "def ratio_hapax_legomenon(doc): # per document \n",
    "    word_indoc = word_tokenize(doc)\n",
    "    fdist = nltk.FreqDist(word_indoc)\n",
    "    fdist_hapx = nltk.FreqDist.hapaxes(fdist)\n",
    "    return len(fdist_hapx)/len(word_indoc)\n",
    "def letter_freq(doc):\n",
    "    letter_indoc = ''.join([c for c in doc if c.isalpha()])\n",
    "    letter = ['a', 'b', 'c', 'd', 'e', 'f', 'g', 'h', 'i', 'j', 'k', 'l', 'm', 'n', 'o', 'p', 'q', 'r', 's'\n",
    "              , 't', 'u', 'v', 'w', 'x', 'y', 'z']\n",
    "    count = dict()\n",
    "    letter_freq = dict()\n",
    "    for li in letter_indoc:\n",
    "        if li in count.keys():\n",
    "            count[li] += 1\n",
    "        else:\n",
    "            count[li] = 1\n",
    "    for l in letter:\n",
    "        if l in count.keys():\n",
    "            letter_freq[l] = count[l]\n",
    "        else: \n",
    "            letter_freq[l] = 0\n",
    "    return np.array([*letter_freq.values()])\n",
    "    \n",
    "\n",
    "def create_feature(x_train):\n",
    "    lexical_feature_all = []\n",
    "\n",
    "    for doc in x_train:\n",
    "        lexical_feature = []\n",
    "        lexical_feature.append(average_sentence_length(doc))\n",
    "        lexical_feature.append(average_nwords(doc))\n",
    "        lexical_feature.append(average_word_len(doc))\n",
    "        lexical_feature.append(average_ncharacters(doc))\n",
    "        lexical_feature.append(freq_ndigits(doc))\n",
    "        lexical_feature.append(ratio_hapax_legomenon(doc))\n",
    "        lexical_feature.extend(letter_freq(doc))\n",
    "\n",
    "        lexical_feature_all.append(lexical_feature)\n",
    "\n",
    "    return lexical_feature_all\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 97,
   "metadata": {},
   "outputs": [],
   "source": [
    "def input_rep(ngram_range, indices, maxlen, tr, vl, ts, mode):\n",
    "    tr = turn_to_integer(tr, indices, mode)\n",
    "    vl = turn_to_integer(vl, indices, mode)\n",
    "    ts = turn_to_integer(ts, indices, mode)\n",
    "    max_features = len(indices)\n",
    "    if ngram_range > 1:\n",
    "        print('Adding {}-gram features'.format(ngram_range))\n",
    "        ngram_set = set()\n",
    "        for input_list in tr:\n",
    "            for i in range(2, ngram_range+1):\n",
    "                ngrams = create_ngram(input_list, ngram_value=i)\n",
    "                ngram_set.update(ngrams)\n",
    "        start_index = max_features + 1\n",
    "        token_indice = {v: k+start_index for k, v in enumerate(ngram_set)}\n",
    "        indice_token = {token_indice[k]: k for k in token_indice}\n",
    "        max_features = np.max(list(indice_token.keys())) + 1\n",
    "        tr = add_ngram(tr, token_indice, ngram_range, mode)\n",
    "        vl = add_ngram(vl, token_indice, ngram_range, mode)\n",
    "        ts = add_ngram(ts, token_indice, ngram_range, mode)\n",
    "    print('Pad sequences')\n",
    "    tr = pad_sequences(tr, maxlen=maxlen)\n",
    "    vl = pad_sequences(vl, maxlen=maxlen)\n",
    "    ts = pad_sequences(ts, maxlen=maxlen)\n",
    "    return tr, vl, ts, max_features\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "def eval(y_pred, y_ts):\n",
    "    acc = accuracy_score(y_pred, y_ts)\n",
    "    return acc"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 227,
   "metadata": {},
   "outputs": [],
   "source": [
    "def single_model(train, \n",
    "              valid, \n",
    "              test, \n",
    "              y_tr, \n",
    "              y_val, \n",
    "              y_test, \n",
    "              max_features, \n",
    "              nb_class, \n",
    "              hidden_layer, \n",
    "              seq_length, \n",
    "              nb_epoch, \n",
    "              batch_size, \n",
    "              lr):\n",
    "\n",
    "    model = Sequential()\n",
    "    model.add(Embedding(max_features, output_dim=hidden_layer, input_length=seq_length, dropout=0.75, init='glorot_uniform'))\n",
    "    model.add(AveragePooling1D(pool_length=model.output_shape[1]))\n",
    "    model.add(Flatten())\n",
    "    model.add(Dense(nb_class, activation='softmax'))\n",
    "    adam = Adam(lr=lr)\n",
    "    model.compile(loss='categorical_crossentropy', optimizer=adam, metrics=['accuracy'])\n",
    "\n",
    "    earlystop_cb = keras.callbacks.EarlyStopping(monitor='val_loss', patience=5, verbose=2, mode='auto')\n",
    "    print (\"Training the model...\")\n",
    "    hist = model.fit(train, y_tr,\n",
    "                  nb_epoch=nb_epoch,\n",
    "                  batch_size=batch_size,\n",
    "                  validation_data=[valid, y_val],\n",
    "                  callbacks=[earlystop_cb])\n",
    "    fitted_model = model\n",
    "    print (\"Evaluate on test data..\")\n",
    "    loss, acc = model.evaluate(test, y_test, verbose=2)\n",
    "    # predicted classes\n",
    "    pred = model.predict_classes(test)\n",
    "    return hist, acc, pred, fitted_model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 396,
   "metadata": {},
   "outputs": [],
   "source": [
    "def merged_model(train_char, \n",
    "              valid_char, \n",
    "              test_char, \n",
    "              train_wd, \n",
    "              valid_wd, \n",
    "              test_wd, \n",
    "              y_tr, \n",
    "              y_val, \n",
    "              y_test, \n",
    "              max_features_char, \n",
    "              max_features_wd,\n",
    "              nb_class, \n",
    "              hidden_layer,\n",
    "              seq_length_word, \n",
    "              seq_length_char, \n",
    "              nb_epoch, \n",
    "              batch_size, \n",
    "              lr):\n",
    "    \n",
    "    # 모형 클래스 객체 생성 \n",
    "    # Create model\n",
    "    modelw = Sequential()\n",
    "    # Create Layers\n",
    "    modelw.add(Embedding(max_features_wd, hidden_layer, input_length=seq_length_word, dropout=0.75, init='glorot_uniform'))\n",
    "    modelw.add(AveragePooling1D(pool_length=modelw.output_shape[1]))\n",
    "    modelw.add(Flatten())\n",
    "\n",
    "    modelc = Sequential()\n",
    "    modelc.add(Embedding(max_features_char, hidden_layer, input_length=seq_length_char, dropout=0.75, init='glorot_uniform'))\n",
    "    modelc.add(AveragePooling1D(pool_length=modelc.output_shape[1]))\n",
    "    modelc.add(Flatten()) \n",
    "    \n",
    "    #Merge two models\n",
    "    model = Sequential()\n",
    "    merged_layer = concatenate([modelw.output, modelc.output], axis=1)\n",
    "    merged_layer = Dense(nb_class, activation='softmax')(merged_layer)\n",
    "    model= Model(inputs=[modelw.input, modelc.input], outputs=merged_layer)\n",
    "    \n",
    "    adam = Adam(lr=lr)\n",
    "    model.compile(loss='categorical_crossentropy', optimizer=adam, metrics=['accuracy'])\n",
    "    earlystop_cb = keras.callbacks.EarlyStopping(monitor='val_loss', patience=5, verbose=2, mode='auto') \n",
    "    print (\"Training the model...\")\n",
    "    # 트레이닝 시작\n",
    "    # Start to train - save the fitting history into hist\n",
    "    hist = model.fit([train_wd, train_char], y_tr,\n",
    "                      nb_epoch=nb_epoch,\n",
    "                      batch_size=batch_size,\n",
    "                      validation_data=([valid_wd,valid_char], y_val),\n",
    "                      callbacks=[earlystop_cb])\n",
    "    \n",
    "    fitted_model = model\n",
    "    print (\"Evaluate on test data...\")\n",
    "    loss, acc = model.evaluate([test_wd, test_char], y_test, verbose=2)\n",
    "    # predicted values \n",
    "    pred = model.predict([test_wd, test_char])\n",
    "\n",
    "    return hist, acc, pred, fitted_model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 447,
   "metadata": {},
   "outputs": [],
   "source": [
    "def train_test(x_tr, x_val, x_ts, y_tr, y_val, y_ts,\n",
    "               mode, mini_batch_size, nb_class, nb_epoch,\n",
    "               seq_length_word, seq_length_char, lr):\n",
    "    # turn the class label into categorical\n",
    "    y_tr = to_categorical(y_tr)\n",
    "    y_val = to_categorical(y_val)\n",
    "    y_ts = to_categorical(y_ts)\n",
    "\n",
    "    # create vocabulary for word\n",
    "    word_vectorizer = CountVectorizer(analyzer='word', max_features=700, max_df=0.95, min_df=2, ngram_range=(1,1))\n",
    "    wrd = word_vectorizer.fit(x_tr)\n",
    "    wrd.vocabulary_['unknown'] = 0\n",
    "    vocab = wrd.vocabulary_.keys()\n",
    "    word_indices = dict((c, i) for i, c in enumerate(vocab))\n",
    "\n",
    "    # create vocabulary for character\n",
    "    chars = character_list\n",
    "    char_indices = dict((c, i) for i, c in enumerate(chars))\n",
    "\n",
    "    acc = 0\n",
    "    # can change ngram range\n",
    "    ngram_range_word = 2\n",
    "    ngram_range_char = 4\n",
    "\n",
    "    # represent the input and train test the model\n",
    "    if mode == 'word':\n",
    "        train, valid, test, max_features = input_rep(ngram_range_word, word_indices, seq_length_word, x_tr, x_val, x_ts, mode)\n",
    "        acc = single_model(train, valid, test, y_tr, y_val, y_ts, max_features, nb_class, hidden_layer, seq_length_word, nb_epoch, batch_size, lr)\n",
    "\n",
    "    elif mode == 'char':\n",
    "        train, valid, test, max_features = input_rep(ngram_range_char, char_indices, seq_length_char, x_tr, x_val, x_ts, mode)\n",
    "        acc = single_model(train, valid, test, y_tr, y_val, y_ts, max_features, nb_class, hidden_layer, seq_length_char, nb_epoch, batch_size, lr)\n",
    "\n",
    "    elif mode == 'wordnchar':\n",
    "        train_word, valid_word, test_word, max_features_word = input_rep(ngram_range_word, word_indices, seq_length_word, x_tr, x_val, x_ts, \"word\")\n",
    "        train_char, valid_char, test_char, max_features_char = input_rep(ngram_range_char, char_indices, seq_length_char, x_tr, x_val, x_ts, \"char\")\n",
    "        acc = merged_model(train_char, valid_char, test_char, train_word, valid_word, test_word,\n",
    "                              y_tr, y_val, y_ts, max_features_char, max_features_word,\n",
    "                              nb_class, hidden_layer, seq_length_word, seq_length_char, nb_epoch, batch_size, lr)\n",
    "    return acc\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 439,
   "metadata": {},
   "outputs": [],
   "source": [
    "x_tr, x_val, y_tr, y_val = train_test_split(x_train, y_train, test_size=0.1, random_state=50)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 414,
   "metadata": {},
   "outputs": [],
   "source": [
    "# C50/C50_test.csv \n",
    "# C10/C10_test.csv\n",
    "x_ts, y_ts = load_data('C50/C50_test.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 440,
   "metadata": {},
   "outputs": [],
   "source": [
    "nb_class = 50\n",
    "\n",
    "mini_batch_size = 5\n",
    "hidden_layer = 100\n",
    "nb_epoch = 150\n",
    "length_word = 1000\n",
    "length_char = 9000\n",
    "lr = 0.001"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 464,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Pad sequences (samples x time)\n",
      "Adding 4-gram features\n",
      "Pad sequences (samples x time)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.7/site-packages/ipykernel_launcher.py:23: UserWarning: The `dropout` argument is no longer support in `Embedding`. You can apply a `keras.layers.SpatialDropout1D` layer right after the `Embedding` layer to get the same behavior.\n",
      "/usr/local/lib/python3.7/site-packages/ipykernel_launcher.py:23: UserWarning: Update your `Embedding` call to the Keras 2 API: `Embedding(701, 100, input_length=1000, embeddings_initializer=\"glorot_uniform\")`\n",
      "/usr/local/lib/python3.7/site-packages/ipykernel_launcher.py:24: UserWarning: Update your `AveragePooling1D` call to the Keras 2 API: `AveragePooling1D(pool_size=1000)`\n",
      "/usr/local/lib/python3.7/site-packages/ipykernel_launcher.py:28: UserWarning: The `dropout` argument is no longer support in `Embedding`. You can apply a `keras.layers.SpatialDropout1D` layer right after the `Embedding` layer to get the same behavior.\n",
      "/usr/local/lib/python3.7/site-packages/ipykernel_launcher.py:28: UserWarning: Update your `Embedding` call to the Keras 2 API: `Embedding(91584, 100, input_length=9000, embeddings_initializer=\"glorot_uniform\")`\n",
      "/usr/local/lib/python3.7/site-packages/ipykernel_launcher.py:29: UserWarning: Update your `AveragePooling1D` call to the Keras 2 API: `AveragePooling1D(pool_size=9000)`\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training the model...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.7/site-packages/ipykernel_launcher.py:47: UserWarning: The `nb_epoch` argument in `fit` has been renamed `epochs`.\n",
      "/usr/local/lib/python3.7/site-packages/tensorflow_core/python/framework/indexed_slices.py:424: UserWarning: Converting sparse IndexedSlices to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"Converting sparse IndexedSlices to a dense Tensor of unknown shape. \"\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train on 2250 samples, validate on 250 samples\n",
      "Epoch 1/150\n",
      "2250/2250 [==============================] - 66s 29ms/step - loss: 3.8779 - accuracy: 0.0618 - val_loss: 3.8061 - val_accuracy: 0.1080\n",
      "Epoch 2/150\n",
      "2250/2250 [==============================] - 62s 28ms/step - loss: 3.6971 - accuracy: 0.1836 - val_loss: 3.6088 - val_accuracy: 0.1560\n",
      "Epoch 3/150\n",
      "2250/2250 [==============================] - 63s 28ms/step - loss: 3.4088 - accuracy: 0.3044 - val_loss: 3.3110 - val_accuracy: 0.2600\n",
      "Epoch 4/150\n",
      "2250/2250 [==============================] - 62s 28ms/step - loss: 3.0436 - accuracy: 0.4107 - val_loss: 2.9771 - val_accuracy: 0.3840\n",
      "Epoch 5/150\n",
      "2250/2250 [==============================] - 72s 32ms/step - loss: 2.6654 - accuracy: 0.4871 - val_loss: 2.6511 - val_accuracy: 0.4720\n",
      "Epoch 6/150\n",
      "2250/2250 [==============================] - 67s 30ms/step - loss: 2.3201 - accuracy: 0.5676 - val_loss: 2.3468 - val_accuracy: 0.5080\n",
      "Epoch 7/150\n",
      "2250/2250 [==============================] - 68s 30ms/step - loss: 2.0191 - accuracy: 0.6276 - val_loss: 2.0957 - val_accuracy: 0.5520\n",
      "Epoch 8/150\n",
      "2250/2250 [==============================] - 68s 30ms/step - loss: 1.7707 - accuracy: 0.6622 - val_loss: 1.8747 - val_accuracy: 0.6240\n",
      "Epoch 9/150\n",
      "2250/2250 [==============================] - 64s 28ms/step - loss: 1.5604 - accuracy: 0.7209 - val_loss: 1.6880 - val_accuracy: 0.6720\n",
      "Epoch 10/150\n",
      "2250/2250 [==============================] - 66s 29ms/step - loss: 1.3831 - accuracy: 0.7600 - val_loss: 1.5439 - val_accuracy: 0.7120\n",
      "Epoch 11/150\n",
      "2250/2250 [==============================] - 64s 29ms/step - loss: 1.2327 - accuracy: 0.7858 - val_loss: 1.4156 - val_accuracy: 0.7040\n",
      "Epoch 12/150\n",
      "2250/2250 [==============================] - 66s 29ms/step - loss: 1.1000 - accuracy: 0.8156 - val_loss: 1.2908 - val_accuracy: 0.7640\n",
      "Epoch 13/150\n",
      "2250/2250 [==============================] - 61s 27ms/step - loss: 0.9849 - accuracy: 0.8422 - val_loss: 1.2072 - val_accuracy: 0.7600\n",
      "Epoch 14/150\n",
      "2250/2250 [==============================] - 66s 29ms/step - loss: 0.8822 - accuracy: 0.8578 - val_loss: 1.1203 - val_accuracy: 0.7960\n",
      "Epoch 15/150\n",
      "2250/2250 [==============================] - 64s 28ms/step - loss: 0.7898 - accuracy: 0.8791 - val_loss: 1.0437 - val_accuracy: 0.8000\n",
      "Epoch 16/150\n",
      "2250/2250 [==============================] - 67s 30ms/step - loss: 0.7078 - accuracy: 0.8911 - val_loss: 0.9758 - val_accuracy: 0.8200\n",
      "Epoch 17/150\n",
      "2250/2250 [==============================] - 63s 28ms/step - loss: 0.6341 - accuracy: 0.9067 - val_loss: 0.9161 - val_accuracy: 0.8200\n",
      "Epoch 18/150\n",
      "2250/2250 [==============================] - 66s 29ms/step - loss: 0.5665 - accuracy: 0.9178 - val_loss: 0.8687 - val_accuracy: 0.8160\n",
      "Epoch 19/150\n",
      "2250/2250 [==============================] - 60s 27ms/step - loss: 0.5061 - accuracy: 0.9302 - val_loss: 0.8146 - val_accuracy: 0.8320\n",
      "Epoch 20/150\n",
      "2250/2250 [==============================] - 61s 27ms/step - loss: 0.4521 - accuracy: 0.9382 - val_loss: 0.7773 - val_accuracy: 0.8200\n",
      "Epoch 21/150\n",
      "2250/2250 [==============================] - 62s 27ms/step - loss: 0.4027 - accuracy: 0.9493 - val_loss: 0.7340 - val_accuracy: 0.8360\n",
      "Epoch 22/150\n",
      "2250/2250 [==============================] - 57s 25ms/step - loss: 0.3584 - accuracy: 0.9573 - val_loss: 0.6950 - val_accuracy: 0.8440\n",
      "Epoch 23/150\n",
      "2250/2250 [==============================] - 61s 27ms/step - loss: 0.3188 - accuracy: 0.9649 - val_loss: 0.6627 - val_accuracy: 0.8600\n",
      "Epoch 24/150\n",
      "2250/2250 [==============================] - 64s 28ms/step - loss: 0.2828 - accuracy: 0.9738 - val_loss: 0.6407 - val_accuracy: 0.8600\n",
      "Epoch 25/150\n",
      "2250/2250 [==============================] - 65s 29ms/step - loss: 0.2513 - accuracy: 0.9800 - val_loss: 0.6111 - val_accuracy: 0.8600\n",
      "Epoch 26/150\n",
      "2250/2250 [==============================] - 64s 28ms/step - loss: 0.2222 - accuracy: 0.9827 - val_loss: 0.5902 - val_accuracy: 0.8560\n",
      "Epoch 27/150\n",
      "2250/2250 [==============================] - 63s 28ms/step - loss: 0.1961 - accuracy: 0.9853 - val_loss: 0.5703 - val_accuracy: 0.8760\n",
      "Epoch 28/150\n",
      "2250/2250 [==============================] - 64s 29ms/step - loss: 0.1728 - accuracy: 0.9911 - val_loss: 0.5461 - val_accuracy: 0.8760\n",
      "Epoch 29/150\n",
      "2250/2250 [==============================] - 66s 29ms/step - loss: 0.1519 - accuracy: 0.9933 - val_loss: 0.5330 - val_accuracy: 0.8680\n",
      "Epoch 30/150\n",
      "2250/2250 [==============================] - 64s 28ms/step - loss: 0.1336 - accuracy: 0.9947 - val_loss: 0.5191 - val_accuracy: 0.8680\n",
      "Epoch 31/150\n",
      "2250/2250 [==============================] - 66s 29ms/step - loss: 0.1175 - accuracy: 0.9956 - val_loss: 0.4984 - val_accuracy: 0.8880\n",
      "Epoch 32/150\n",
      "2250/2250 [==============================] - 64s 29ms/step - loss: 0.1026 - accuracy: 0.9956 - val_loss: 0.4908 - val_accuracy: 0.8720\n",
      "Epoch 33/150\n",
      "2250/2250 [==============================] - 60s 27ms/step - loss: 0.0901 - accuracy: 0.9973 - val_loss: 0.4683 - val_accuracy: 0.8880\n",
      "Epoch 34/150\n",
      "2250/2250 [==============================] - 57s 25ms/step - loss: 0.0785 - accuracy: 0.9973 - val_loss: 0.4748 - val_accuracy: 0.8760\n",
      "Epoch 35/150\n",
      "2250/2250 [==============================] - 57s 25ms/step - loss: 0.0691 - accuracy: 0.9982 - val_loss: 0.4631 - val_accuracy: 0.8560\n",
      "Epoch 36/150\n",
      "2250/2250 [==============================] - 64s 28ms/step - loss: 0.0600 - accuracy: 0.9996 - val_loss: 0.4579 - val_accuracy: 0.8840\n",
      "Epoch 37/150\n",
      "2250/2250 [==============================] - 62s 27ms/step - loss: 0.0526 - accuracy: 0.9987 - val_loss: 0.4547 - val_accuracy: 0.8720\n",
      "Epoch 38/150\n",
      "2250/2250 [==============================] - 63s 28ms/step - loss: 0.0461 - accuracy: 1.0000 - val_loss: 0.4428 - val_accuracy: 0.8720\n",
      "Epoch 39/150\n",
      "2250/2250 [==============================] - 65s 29ms/step - loss: 0.0397 - accuracy: 0.9996 - val_loss: 0.4338 - val_accuracy: 0.8840\n",
      "Epoch 40/150\n",
      "2250/2250 [==============================] - 59s 26ms/step - loss: 0.0349 - accuracy: 1.0000 - val_loss: 0.4248 - val_accuracy: 0.8760\n",
      "Epoch 41/150\n",
      "2250/2250 [==============================] - 53s 24ms/step - loss: 0.0305 - accuracy: 1.0000 - val_loss: 0.4302 - val_accuracy: 0.8720\n",
      "Epoch 42/150\n",
      "2250/2250 [==============================] - 57s 25ms/step - loss: 0.0267 - accuracy: 1.0000 - val_loss: 0.4298 - val_accuracy: 0.8800\n",
      "Epoch 43/150\n",
      "2250/2250 [==============================] - 57s 25ms/step - loss: 0.0235 - accuracy: 1.0000 - val_loss: 0.4200 - val_accuracy: 0.8840\n",
      "Epoch 44/150\n",
      "2250/2250 [==============================] - 59s 26ms/step - loss: 0.0202 - accuracy: 1.0000 - val_loss: 0.4180 - val_accuracy: 0.8720\n",
      "Epoch 45/150\n",
      "2250/2250 [==============================] - 60s 27ms/step - loss: 0.0177 - accuracy: 1.0000 - val_loss: 0.4089 - val_accuracy: 0.8840\n",
      "Epoch 46/150\n",
      "2250/2250 [==============================] - 58s 26ms/step - loss: 0.0155 - accuracy: 1.0000 - val_loss: 0.4095 - val_accuracy: 0.8840\n",
      "Epoch 47/150\n",
      "2250/2250 [==============================] - 60s 27ms/step - loss: 0.0133 - accuracy: 1.0000 - val_loss: 0.4046 - val_accuracy: 0.8760\n",
      "Epoch 48/150\n",
      "2250/2250 [==============================] - 57s 25ms/step - loss: 0.0118 - accuracy: 1.0000 - val_loss: 0.4117 - val_accuracy: 0.8760\n",
      "Epoch 49/150\n",
      "2250/2250 [==============================] - 60s 26ms/step - loss: 0.0102 - accuracy: 1.0000 - val_loss: 0.4171 - val_accuracy: 0.8760\n",
      "Epoch 50/150\n",
      "2250/2250 [==============================] - 60s 27ms/step - loss: 0.0091 - accuracy: 1.0000 - val_loss: 0.3998 - val_accuracy: 0.8680\n",
      "Epoch 51/150\n",
      "2250/2250 [==============================] - 58s 26ms/step - loss: 0.0078 - accuracy: 1.0000 - val_loss: 0.3935 - val_accuracy: 0.8960\n",
      "Epoch 52/150\n",
      "2250/2250 [==============================] - 58s 26ms/step - loss: 0.0067 - accuracy: 1.0000 - val_loss: 0.4204 - val_accuracy: 0.8680\n",
      "Epoch 53/150\n",
      "2250/2250 [==============================] - 60s 27ms/step - loss: 0.0061 - accuracy: 1.0000 - val_loss: 0.3941 - val_accuracy: 0.8760\n",
      "Epoch 54/150\n",
      "2250/2250 [==============================] - 58s 26ms/step - loss: 0.0051 - accuracy: 1.0000 - val_loss: 0.4008 - val_accuracy: 0.8880\n",
      "Epoch 55/150\n",
      "2250/2250 [==============================] - 59s 26ms/step - loss: 0.0046 - accuracy: 1.0000 - val_loss: 0.4097 - val_accuracy: 0.8760\n",
      "Epoch 56/150\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2250/2250 [==============================] - 58s 26ms/step - loss: 0.0039 - accuracy: 1.0000 - val_loss: 0.4027 - val_accuracy: 0.8840\n",
      "Epoch 00056: early stopping\n",
      "Evaluate on test data...\n"
     ]
    }
   ],
   "source": [
    "mode = 'word'\n",
    "# mode = 'char'\n",
    "# mode = 'wordnchar'\n",
    "hist, acc, pred, model = train_test(x_tr, x_val, x_ts, y_tr, y_val, y_ts, mode, mini_batch_size, nb_class, nb_epoch,\n",
    "                             length_word, length_char, lr)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 458,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt \n",
    "%matplotlib inline\n",
    "\n",
    "plt.plot(hist.history['accuracy'])\n",
    "plt.plot(hist.history['val_accuracy'])\n",
    "plt.title('model accuracy')\n",
    "plt.ylabel('accuracy')\n",
    "plt.xlabel('epoch')\n",
    "plt.legend(['train', 'validation'], loc='upper left')\n",
    "plt.show()\n",
    "# summarize history for loss\n",
    "plt.plot(hist.history['loss'])\n",
    "plt.plot(hist.history['val_loss'])\n",
    "plt.title('model loss')\n",
    "plt.ylabel('loss')\n",
    "plt.xlabel('epoch')\n",
    "plt.legend(['train', 'validation'], loc='upper right')\n",
    "plt.show()\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 459,
   "metadata": {},
   "outputs": [],
   "source": [
    "# convert the history.history dict to a pandas DataFrame:     \n",
    "hist_df = pd.DataFrame(hist.history) \n",
    "\n",
    "# or save to csv: \n",
    "hist_csv_file = 'history.csv'\n",
    "\n",
    "with open(hist_csv_file, mode='w') as f:\n",
    "    hist_df.to_csv(f)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 460,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential_107\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "embedding_81 (Embedding)     (None, 9000, 100)         9158400   \n",
      "_________________________________________________________________\n",
      "average_pooling1d_80 (Averag (None, 1, 100)            0         \n",
      "_________________________________________________________________\n",
      "flatten_79 (Flatten)         (None, 100)               0         \n",
      "_________________________________________________________________\n",
      "dense_44 (Dense)             (None, 50)                5050      \n",
      "=================================================================\n",
      "Total params: 9,163,450\n",
      "Trainable params: 9,163,450\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 346,
   "metadata": {},
   "outputs": [],
   "source": [
    "target_names_10 = ['AlanCrosby',\n",
    " 'AlexanderSmith',\n",
    " 'BenjaminKangLim',\n",
    " 'DavidLawder',\n",
    " 'JaneMacartney',\n",
    " 'JimGilchrist',\n",
    " 'MarcelMichelson',\n",
    " 'MureDickie',\n",
    " 'RobinSidel',\n",
    " 'ToddNissen']\n",
    "\n",
    "target_names_50 = ['AaronPressman',\n",
    " 'AlanCrosby',\n",
    " 'AlexanderSmith',\n",
    " 'BenjaminKangLim',\n",
    " 'BernardHickey',\n",
    " 'BradDorfman',\n",
    " 'DarrenSchuettler',\n",
    " 'DavidLawder',\n",
    " 'EdnaFernandes',\n",
    " 'EricAuchard',\n",
    " 'FumikoFujisaki',\n",
    " 'GrahamEarnshaw',\n",
    " 'HeatherScoffield',\n",
    " 'JanLopatka',\n",
    " 'JaneMacartney',\n",
    " 'JimGilchrist',\n",
    " 'JoWinterbottom',\n",
    " 'JoeOrtiz',\n",
    " 'JohnMastrini',\n",
    " 'JonathanBirt',\n",
    " 'KarlPenhaul',\n",
    " 'KeithWeir',\n",
    " 'KevinDrawbaugh',\n",
    " 'KevinMorrison',\n",
    " 'KirstinRidley',\n",
    " 'KouroshKarimkhany',\n",
    " 'LydiaZajc',\n",
    " \"LynneO'Donnell\",\n",
    " 'LynnleyBrowning',\n",
    " 'MarcelMichelson',\n",
    " 'MarkBendeich',\n",
    " 'MartinWolk',\n",
    " 'MatthewBunce',\n",
    " 'MichaelConnor',\n",
    " 'MureDickie',\n",
    " 'NickLouth',\n",
    " 'PatriciaCommins',\n",
    " 'PeterHumphrey',\n",
    " 'PierreTran',\n",
    " 'RobinSidel',\n",
    " 'RogerFillion',\n",
    " 'SamuelPerry',\n",
    " 'SarahDavison',\n",
    " 'ScottHillis',\n",
    " 'SimonCowell',\n",
    " 'TanEeLyn',\n",
    " 'TheresePoletti',\n",
    " 'TimFarrand',\n",
    " 'ToddNissen',\n",
    " 'WilliamKazer']\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 461,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x17283a910>"
      ]
     },
     "execution_count": 461,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAACk0AAAsRCAYAAAB47S1CAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy8li6FKAAAgAElEQVR4nOzdebjkdXnn7/fpbqDZRMVBRBCN4se4oOjELS4IInHXGBNjXBIXxIhiJKiTEG0FHIwYQRQRASEgoklm4hYdtyhRYPxFBATjB0VEZJHIAEoD3Zzu/v1x6mAHm16Q+lY/1H1fl9fp2l/9XOXpOnUevjWzatWqAAAAAAAAAAAAANzZLZh0AAAAAAAAAAAAAMAQLE0CAAAAAAAAAAAAU8HSJAAAAAAAAAAAADAVLE0CAAAAAAAAAAAAU8HSJAAAAAAAAAAAADAVLE0CAAAAAAAAAAAAU2HRpAMAAAAAAAAAAAAA1kdr7ewkvxidvDjJh5McmWQ2yRd77+9Y2+0tTQIAAAAAAAAAAAAbvdba4iQzvffdVzvvnCQvSPKjJJ9rre3We//Obd2HpUkAAAAAAAAAAACggocn2aK19sXM7T8uSbJZ7/2iJGmt/Z8kT01iabKyzZ937KpJN2yIa/5xn0knwG/spptXTDphgy3eZOGkEwAAYKMzu6LUj9RJkkULZyadAAAATKGKvxtZtGDBpBM2mJ/5gDuzxYvimxzrZfPd9qv3xi2DufE7H1jX95Ibkhye5LgkuyT5fJJrV7v8l0l+a213YGkSAAAAAAAAAAAAqODCJD/sva9KcmFr7bokd1/t8q3zX5cof029//QGAAAAAAAAAAAAmEavSPLeJGmt7ZBkiyRLW2v3b63NJNk7yb+t7Q4caRIAAAAAAAAAAACo4PgkJ7bWvpFkVeaWKFcm+ViShUm+2Hv/v2u7A0uTAAAAAAAAAAAAwEav9748yYvXcNFj1/c+fDw3AAAAAAAAAAAAMBUsTQIAAAAAAAAAAABTwdIkAAAAAAAAAAAAMBUsTQIAAAAAAAAAAABTwdIkAAAAAAAAAAAAMBUsTQIAAAAAAAAAAABTYdGkAwAAAAAAAAAAAJgSM47zx2R5BgIAAAAAAAAAAABTwdIkAAAAAAAAAAAAMBUsTQIAAAAAAAAAAABTwdIkAAAAAAAAAAAAMBUsTQIAAAAAAAAAAABTwdIkAAAAAAAAAAAAMBUsTQIAAAAAAAAAAABTwdIkAAAAAAAAAAAAMBUsTQIAAAAAAAAAAABTwdIkAAAAAAAAAAAAMBUWTToAAAAAAAAAAACAKTEzM+kCppylySnz37ZZnDPe+/t55ts/lwsvuy5J8kdPun9e+4yHZve3fmrCdWu2cuXKHHrwklzYezbddNO8/R2H5D477zzprLXSPH7VepNk9uabc/CSg3LF5Zfl5uU3589e/Zo8afc9Jp11myrOWPP4VetNNA+hWm+ieQjVehPNQ6jWm9Rrrtab1GxOku+ed26OOuLwHHvCyZNOWaeKM9Y8ftV6E81DqNab1Guu1ptoHkK13kTzEKr1JpqHUK03qfe7kXl+5huvas3VehPNQ6jWm9RsBhiXwT+eu7X25tbaFa21xWO6/z9trf2ktfa11tq/ttbObK394Tgeq5pFC2fygdc+MTcum73lvIffb9u8/KkP2qgXuL/6lS9n+bLlOfnUT2T/vzgg733PYZNOWifN41etN0k+/7nPZJtt7ppjP3pKjjj6wzn8sEMmnbRWFWesefyq9Saah1CtN9E8hGq9ieYhVOtN6jVX601qNp90wnE5eMlBWbZs+aRT1kvFGWsev2q9ieYhVOtN6jVX6000D6Fab6J5CNV6E81DqNab1PvdSOJnviFUa67Wm2geQrXepGYzwLgMvjSZ5CVJTkvyojE+xqm99917709J8swkf9da24jXAodx2J8+Nh/5wn/kimtuSJLcfevN8o6X/E4OPP6MCZet3XfO/nYe/4QnJkl2ffgjcsEF50+4aN00j1+13iTZ82l75zWve8PciVXJwoUb98F+K85Y8/hV6000D6Fab6J5CNV6E81DqNab1Guu1pvUbN5xp51y+PuOmnTGeqs4Y83jV6030TyEar1JveZqvYnmIVTrTTQPoVpvonkI1XqTer8bSfzMN4RqzdV6E81DqNab1GwGGJdBX5W21nZPclGSY5KckuTE1tqTk7w9cwucWyV5cZLlST6T5Ook/5LkS0mOSrIiyU1JXj26/seTXJrk/km+1Xt/7Roe9q5Jbuy9r2qtnZ/kwtH9vybJ8Um2HV3vDb3377bWPprkAUk2T3Jk7/3k1tqhSZ6SuXn9U+/93a21ryU5N8lDk1yf5N+S7D16vKeNWo8bnd4hyQd77x8a3e6c0e3ukuSFvfdLbudI19tL9nhg/vMXN+XL5/w0B/7BI7JwwYIcs9+T85aPnpUbl8+u+w4maOnS67P11lvdcnrhgoWZnZ3NokUb7w9VmsevWm+SbLHFlkmSpUuX5q1/+cbsO/8mwUaq4ow1j1+13kTzEKr1JpqHUK030TyEar1JveZqvUnN5j332juXX/bTSWest4oz1jx+1XoTzUOo1pvUa67Wm2geQrXeRPMQqvUmmodQrTep97uRxM98Q6jWXK030TyEar1JzWaAcRn6SJOvSnJc770nWdZae0yShyR5Se999yT/K8kLR9fdPsnTeu9/m+QjSfbrvT85ydFJ/m50nQcmeWWSRyd5Rmtt+9H5Lx59PPdXk7w/yUtH52+V5ODe+4uS/FWSr4yORrlPkg+11rZO8qQkv5/k9zK3+Jgkf5K5Zc4nJrl2tb/Pt3rveybZLMkNvfe9knwvyZMzt3h5Wu/9aZlbonzTrW731Mwtg/7xhg7x9nj5ni17Pvze+T+HPCu73m/b/PuRf5CH7Hz3vH/fJ+TkA/bMg3a6W97zyscNkbLBttxyqyxduvSW0ytXrdzo/9HWPH7Veuf97Mor8uev/tM8/VnPzt7PeNakc9aq4ow1j1+13kTzEKr1JpqHUK030TyEar1JveZqvUnN5moqzljz+FXrTTQPoVpvUq+5Wm+ieQjVehPNQ6jWm2geQrXeeZV+N1JRxedFteZqvYnmIVTrTWo2A4zLYEuTrbW7JXlGkv1ba19Isk2S/ZJcluT9rbUTM3c0x01GN7m497589Ocdeu/njP58euYWLZPkh733X/beVyS5Isni0fnzH8+9R+/9Wb33s1ZL6aOvD0vyitGRHz+S5O69918meWOSY5N8InPLkMnc0uRhSf5P5o4cOe/s0ddrM7csmSTXjDp+luR5rbVTkhy02t8rSb4z+nrpas1jtddffyZPO+iz2fugz+a8i6/Obq//ZB6y72nZ+6DP5qXv/Uq+f+k1OfD4M4dI2WC77fbIfOP005Mk5517TnbZ5YETLlo3zeNXrTdJrr7653nDa1+d/fZ/U57zvBdMOmedKs5Y8/hV6000D6Fab6J5CNV6E81DqNab1Guu1pvUbK6m4ow1j1+13kTzEKr1JvWaq/UmmodQrTfRPIRqvYnmIVTrTer9bqSiis+Las3VehPNQ6jWm9RsBhiXIVfGX5Lk+N77gUnSWtsiycWj8+/Se/9la+2kJDOj669c7baXt9Z27b2fl7mjOF44On/V7eiYv9/vJzml935qa227JK9qrd0ryaN6789vrS1Ocmlr7dTMHf1y/oiQ32utnbYej39AkjNHH8n9lCTPXO2y29M9tfZ46l4588xv5mV/8qKsWrUq7zzkXZNOWifN41etN0lOPO7Y/OIX1+WEY4/JCccekyR53wc/nMWLB9md3mAVZ6x5/Kr1JpqHUK030TyEar2J5iFU603qNVfrTWo2V1NxxprHr1pvonkI1XqTes3VehPNQ6jWm2geQrXeRPMQqvUm9X43UlHF50W15mq9ieYhVOtNajYDjMvMqlXD7O+11s5N8tLR4uP8eUcn2TfJeUmWZu7ojFcnOTRzH2392NH1dktyZOYWKmcz95HcK291nbOSvCjJ7kke1Ht/6xoafjy67KbW2rZJjs/ckSPvkmRJks8k+VCSXTP30dyf7b2/u7X2tswtPd6Y5NzMHY3yX5Ps23v//miJ8pje+9daa0ckOWv0dzlq9Pe5NslDkzw4c0ernL/dvkm2770vWdvsNn/esaWWLK/5x30mnQC/sZtuXjHphA22eJOFk04AAICNzuyKUj9SJ0kWLZxZ95UAAADuYBV/N7JowWAfrHiH8TMfcGe2eFF8k2O9bP6o/eu9cctgbvz2kWP/XjLY0iS3n6VJGF7FNwYsTQIAwK+zNAkAALB+Kv5uxNIkwMbF0iTry9IkazPE0mS9V5EAAAAAAAAAAAAAt8OiSQcAAAAAAAAAAAAwJWYc54/J8gwEAAAAAAAAAAAApoKlSQAAAAAAAAAAAGAqWJoEAAAAAAAAAAAApoKlSQAAAAAAAAAAAGAqWJoEAAAAAAAAAAAApoKlSQAAAAAAAAAAAGAqWJoEAAAAAAAAAAAApoKlSQAAAAAAAAAAAGAqWJoEAAAAAAAAAAAApsKiSQcAAAAAAAAAAAAwJWZmJl3AlHOkSQAAAAAAAAAAAGAqWJoEAAAAAAAAAAAApoKlSQAAAAAAAAAAAGAqWJoEAAAAAAAAAAAApoKlSQAAAAAAAAAAAGAqWJoEAAAAAAAAAAAApoKlSQAAAAAAAAAAAGAqWJoEAAAAAAAAAAAApoKlSQAAAAAAAAAAAGAqWJoEAAAAAAAAAAAApsKiSQcAAAAAAAAAAAAwJWYc54/JsjRZwDX/uM+kEzbIPp88b9IJG+TYP9x10glshBZvsnDSCQCs5pKf3zDphA2y8z22mHQC3CHO+OHVk07YII9/wLaTTpgKsytWTTphgyxaODPpBAAAgBL8bgQAgGlhbRcAAAAAAAAAAACYCpYmAQAAAAAAAAAAgKlgaRIAAAAAAAAAAACYCpYmAQAAAAAAAAAAgKlgaRIAAAAAAAAAAACYCpYmAQAAAAAAAAAAgKlgaRIAAAAAAAAAAACYCpYmAQAAAAAAAAAAgKlgaRIAAAAAAAAAAACYCosmHQAAAAAAAAAAAMCUmJmZdAFTzpEmAQAAAAAAAAAAgKlgaRIAAAAAAAAAAACYCpYmAQAAAAAAAAAAgKlgaRIAAAAAAAAAAACYCpYmAQAAAAAAAAAAgKlgaRIAAAAAAAAAAACYCpYmAQAAAAAAAAAAgKlgaRIAAAAAAAAAAACYCpYmAQAAAAAAAAAAgKmwaNIBAAAAAAAAAAAATIkZx/ljsjwDAQAAAAAAAAAAgKlgaRIAAAAAAAAAAACYCj6eewqtXLkyhx68JBf2nk033TRvf8chuc/OO086679YOJO86rE75R5bbppFC2fy6fOvyuPue9dss3juKXuPLTfNRVffkKO/+ZMJl65ZhRnfWrXmar1JveZqvYnmIVTrTTQPoVpvkszO3pwj3vW2XHXl5VmwYGH2O/BvsuPO95t01lpVm3O13kTzuF3cL8j//vuj86ZDP3jLed/6+hfztc/9Q978tx+ZYNnaVZpxUq93dd8979wcdcThOfaEkyedsk7V5lytN9E8hGq9ieYhVOtN6jVX6000D6Fab6J5CNV6E81DqNabaB5Ctd6kXnO13kTzEKr1JjWbAcblTnGkydbam1trV7TWFo9Of6219qDbcT87tdY+Obr9/22tHd1a2/Q3bLvyN7n9OHz1K1/O8mXLc/Kpn8j+f3FA3vuewyad9Gsef7+75fplszn0yxfl8H+9OC/77zvk6G/+JP/zKz/KkadfkhtuXpGPffvySWfepgozvrVqzdV6k3rN1XoTzUOo1ptoHkK13iT597O+kRUrVuRvjz4pf/TyfXLycR+YdNI6VZtztd5E8zh98X+dklM++D9z8/Llt5x36Y96zvjyZ7Jq1QTD1kOVGc+r1jvvpBOOy8FLDsqyZcvXfeWNQLU5V+tNNA+hWm+ieQjVepN6zdV6E81DqNabaB5Ctd5E8xCq9Saah1CtN6nXXK030TyEar1JzWaAcblTLE0meUmS05K86PbeQWttYZJPJXlv73333vtjktyc5J13TOLG4ztnfzuPf8ITkyS7PvwRueCC8ydc9Ou+9ZPr8k/n/SxJMpNkxWq/Sf39Xe+ZL/Wf57qbZidUt24VZnxr1Zqr9Sb1mqv1JpqHUK030TyEar1Jcu8dd86KFSuycuXK3HjD9Vm0aOM/AHu1OVfrTTSP0z22v3de89b/ecvp639xXf755A/nha984wSr1k+VGc+r1jtvx512yuHvO2rSGeut2pyr9Saah1CtN9E8hGq9Sb3mar2J5iFU6000D6Fab6J5CNV6E81DqNab1Guu1ptoHkK13qRmM8C4bPy/HV6H1truSS5KckySU5KcuNplOyb5UJLFSe6V5KDe+z+31s5L8vUkuyZZleS5SR6R5NLe+/9d7e7fkmRBa+2+ST6T5Ook/5LkS0mOSrIiyU1JXp3kqiSfTLJNki2S/HXv/YtJNmutnZZkpyTnJfnzJN9Isk/v/YLW2tOTPLv3/ud36GDWYunS67P11lvdcnrhgoWZnZ3dqJYFls2uTJIsXrQg+z1x5/zTuXMLlFtvtjAPvudW+djZG+9RJpMaM761as3VepN6zdV6E81DqNabaB5Ctd4kWbzFFrnqysvz5y99fn5x3bX5m8OOnHTSOlWbc7XeRPM4PfLxT8nVP7siSbJyxYqc8oF35Q9e8YZsuulmEy5btyoznletd96ee+2dyy/76aQz1lu1OVfrTTQPoVpvonkI1XqTes3VehPNQ6jWm2geQrXeRPMQqvUmmodQrTep11ytN9E8hGq9Sc1mgHG5Mxxp8lVJjuu99yTLWmuPWe2yB2XuyJF7JdknyetG598lycd7709OclmSpyfZIcmPVr/j3vtNvfcbRie3T/K03vvfJvlIkv1Gtz86yd8luX+SeyR5dpI/zq8WUjdP8pbe++8m2XZ0+XFJXj66/BWj04PZcsutsnTp0ltOr1y1cqP8R/DuW2yS/7Hnb+WMi6/JmZdcmyR59H3umjMvuXaj/wi/KjNeXbXmar1JveZqvYnmIVTrTTQPoVpvknz6k6dkt0c/Lsd87FM58oRP5Ih3vS3Lly2bdNZaVZtztd5E81B+clHPVZf/NB8/5j057vC35cpLL84njzti0lm3qdqMq/VWVW3O1XoTzUOo1ptoHkK13qRec7XeRPMQqvUmmodQrTfRPIRqvYnmIVTrTeo1V+tNNA+hWm9SsxlgXEovTbbW7pbkGUn2b619IXNHedxvtatckeQ1rbWTk+ybZJPVLvvO6OulmTsS5SWZOxrk6ve/bWvt2aOTF/fel4/+vEPv/ZzRn09P8pDe+wVJPpzk45lbpJyf7U9675eM/nxGkpa5I1I+p7W2XZIde+9n364B3E677fbIfOP005Mk5517TnbZ5YFDPvx6ucviRXnzU+6XT5xzZU7/0TW3nP+Q7bfKeZf/coJl66fCjG+tWnO13qRec7XeRPMQqvUmmodQrTdJttz6Ltlyy7n/mnLrrbfJihWzWbly5YSr1q7anKv1JpqHct8HPjhv+8DH8qZDP5hX/eU7s/1O98sfvmrj/ZjuajOu1ltVtTlX6000D6Fab6J5CNV6k3rN1XoTzUOo1ptoHkK13kTzEKr1JpqHUK03qddcrTfRPIRqvUnNZoBxqb4y/pIkx/feD0yS1toWSS5O8vPR5Qcn+Ujv/fOttT9L8qer3fbWxyo8K8n9WmuP7r1/q7U2k2RJkhuTfDfJ6r9Fv7y1tmvv/bwkT05yYWvtYUm27r0/s7V2r8wtSH42yY6ttXv13q9I8oRR79LW2r8mOTJzHyk+qD2eulfOPPObedmfvCirVq3KOw9519AJ6/Tsh2yXLTZdmOc+dLs896HbJUkO/9rF2X7rzfKf12/cR4FKasz41qo1V+tN6jVX6000D6Fab6J5CNV6k+S5L3xJ3v/uJXnrfq/I7OzNeemrXp/Fm28+6ay1qjbnar2JZtas2oyr9VZVbc7VehPNQ6jWm2geQrXepF5ztd5E8xCq9Saah1CtN9E8hGq9ieYhVOtN6jVX6000D6Fab1KzGWBcZlZt7J9zvBattXOTvHS0vDh/3tGZ+8juXZPsluSgJFcn+WmSh/feH9Ja+3GSB/Xeb2qtHZbk+733E1trv5XkA0m2HP3vrCRvytxHd5/We3/s6DF2y9zC40yS2SSvTHJ55hYgt8vcUSY/3Hs/ubV26eh+dkxyRu/9gNF9PDLJNzJ31Mpr1/b3vGn21xY8N2r7fPK8dV9pI3LsH+466QQAYB0u+fkNk07YIDvfY4tJJ8Ad4owfXj3phA3y+AdsO+mEqTC7otSPqFm0cGbSCQAAAAAwFRYvijfjWC+bP/Yttd5oZlA3nvXusX8vKb00WVlr7XeSvL73/rJ1XdfS5HhZmgSAjZ+lSZgMS5OsiaVJAAAAAGBNLE2yvjZ/3FtrvdHMoG4887Cxfy+p/vHcJbXW9svc0Sn/cNItAAAAAAAAAAAAMC0sTU5A7/0DmfsYcAAAAAAAAAAAAGAgCyYdAAAAAAAAAAAAADAES5MAAAAAAAAAAADAVLA0CQAAAAAAAAAAAEwFS5MAAAAAAAAAAADAVLA0CQAAAAAAAAAAAEwFS5MAAAAAAAAAAADAVLA0CQAAAAAAAAAAAEyFRZMOAAAAAAAAAAAAYErMOM4fk+UZCAAAAAAAAAAAAEwFS5MAAAAAAAAAAADAVLA0CQAAAAAAAAAAAEwFS5MAAAAAAAAAAADAVLA0CQAAAAAAAAAAAEwFS5MAAAAAAAAAAADAVLA0CQAAAAAAAAAAAEwFS5MAAAAAAAAAAADAVLA0CQAAAAAAAAAAAEwFS5MAAAAAAAAAAADAVFg06QAAAAAAAAAAAACmxMzMpAuYco40CQAAAAAAAAAAAEwFS5MAAAAAAAAAAADAVPDx3Nzhjv3DXSedsEFefNK3J52wwU59+aMmnQBTaXbFqkknbLBFCx3WnDuHne+xxaQT4Df26fMvn3TCBnvOQ3eYdAIbIa8vWJNqr5U9jwEA4NdVe12feG0/BM8LAODOyJEmAQAAAAAAAAAAgKlgaRIAAAAAAAAAAACYCpYmAQAAAAAAAAAAgKlgaRIAAAAAAAAAAACYCpYmAQAAAAAAAAAAgKlgaRIAAAAAAAAAAACYCpYmAQAAAAAAAAAAgKmwaNIBAAAAAAAAAAAATIkZx/ljsjwDAQAAAAAAAAAAgKlgaRIAAAAAAAAAAACYCpYmAQAAAAAAAAAAgKlgaRIAAAAAAAAAAACYCpYmAQAAAAAAAAAAgKlgaRIAAAAAAAAAAACYCpYmAQAAAAAAAAAAgKlgaRIAAAAAAAAAAACYCpYmAQAAAAAAAAAAgKmwaNIBAAAAAAAAAAAATImZmUkXMOUcaRIAAAAAAAAAAACYCpYmAQAAAAAAAAAAgKlgaRIAAAAAAAAAAACYCosmHcDwVq5cmUMPXpILe8+mm26at7/jkNxn550nnXWbqvQunEle96T7ZrutNs0mCxfkH8+5IlcvXZ7X/O7OuXnFyvz46htz/FmXZtWkQ29DlTnPq9ab1Guu1pvUbE6S7553bo464vAce8LJk05Zp4oz1jx+1XoTzUOo1pvUar70B9/LF089Nq98+xH5xBHvzPXX/b8kybX/eWV2fMCD80dvfNuEC9es0oznVWuu1ptoHkK13tV5rTw+1XoTzUOo1pvUa67Wm2geQrXeRPMQqvUmmodQrXd1XtuPT7XeeZ4T46V5/Kr1JjWbAcZlYkeabK29ubV2RWtt8ej011prDxrosfdtrS3ZgOu/vLX21dbav7bWvtlae9p63u4RrbW3jf78/NbaDq21+7bWzrqd6XeIr37ly1m+bHlOPvUT2f8vDsh733PYJHPWqUrvkx+wba6/aTYHfe7CHPyFH+RVj7tP9v3dnXPCWZfmoM9dmKU3r8gT73/3SWfepipznletN6nXXK03qdl80gnH5eAlB2XZsuWTTlkvFWesefyq9Saah1CtN6nT/G+f+nj++cOHZ3b53L8df/TGt+WVbz8iLz7g4CzeYqs84+Wvm3Dhbasy49VVa67Wm2geQrXeeV4rj1e13kTzEKr1JvWaq/UmmodQrTfRPIRqvYnmIVTrnee1/XhV6008J4agefyq9SY1mwHGZZIfz/2SJKcledEEG9aptbZNkr9J8nu996ckeWGSE1pr65xd7/2c3vs7Ryf3T3KX8ZWuv++c/e08/glPTJLs+vBH5IILzp9w0dpV6T3j4mty6tmXz52YSVasXJVtt9w0/aqlSZLv/+z6/PY9t5pg4dpVmfO8ar1JveZqvUnN5h132imHv++oSWest4oz1jx+1XoTzUOo1pvUab779jvkxQe889fO/+o/fDSP/b3nZ+u7bTuBqvVTZcarq9ZcrTfRPIRqvfO8Vh6var2J5iFU603qNVfrTTQPoVpvonkI1XoTzUOo1jvPa/vxqtabeE4MQfP4VetNajYDjMtEPp67tbZ7kouSHJPklCQnrnbZNkmOTzL/m743JLk2yVeTPCnJbyd5R5KnJNk3ye8n2TLJz5M8P8mLkzwjyRZJ7p/k3b33E1trT0hyZJJrkswmOWv0eK8f3WZVktN67+9vrZ04evxtk7wgyaZJXtta+2zv/aLW2v177ytH17s5yc5JNsvcEuizk9wnyXOT7DRqPDnJI5L8feaWRf9ba+2fk9wryXm991f/hiPdIEuXXp+tt/7V8t7CBQszOzubRYs2zk9rr9J70+zKJMniTRbkwD3un49/+7I8/cHb5cHbb5XvXXl9fuc+22SzTSa5p7x2VeY8r1pvUq+5Wm9Ss3nPvfbO5Zf9dNIZ663ijDWPX7XeRPMQqvUmdZof8pgn55qrrvwv511/3TW56Pyz8/SN+CiTSZ0Zr65ac7XeRPMQqvXO81p5vKr1JpqHUK03qddcrTfRPIRqvYnmIVTrTTQPoVrvPK/tx6tab+I5MQTN41etN6nZDDAuk9rgelWS43rvPcmy1tpjVrvsr5J8ZXRUx32SfKj3fmmSNyc5Kcn7kvxxkpWZW81L3SwAACAASURBVGp8au/9MZlbAP2d0X1s03t/VpLnJHnr6LwPJfnj3vtTk1ycJK21Byf5oyRPSPLEJM9rrbXR9b/ae3987/2KJHsk2SXJF1prlyR5xWq9P+69Py3JfyS5X+/9GUn+KXPLk0mS3vvnkpyT5GVJlmfuiJN/luRxSfZsrW13e4Z4e2255VZZunTpLadXrlq5Uf8jWKl32y03yTuf/sB8/YdX599+dE0+8G8/zgsevn2WPH2XXHfjbH550+ykE29TpTkn9XqTes3VepOazdVUnLHm8avWm2geQrXepGbzvAvO+np2/d2nZsGChZNOWauKM67WXK030TyEar1VVZtztd5E8xCq9Sb1mqv1JpqHUK030TyEar2J5iFU662q2pyr9VZUccaax69ab1KzGWBcBl+abK3dLXNHgty/tfaFJNsk2W+1qzwsyStaa19L8pEkdx+d/89Jdkzy9d77T3vvKzO3gPjx1trxo8s2GV33nNHXS5MsHv35nr33C0d//ubo60Mzd5TIr4z+t23mliOTpI96d0iyee99v977Lkn2SnJga+1ho+udPfp6bZLvjf58zWqPuyY/6r1fM/o7XJW5o2IOZrfdHplvnH56kuS8c8/JLrs8cMiH32BVerdZvChv/71dcvL/d1m++oOrkySP2mmbvO9rF2fJ53+QrTdblHMv+8WEK29blTnPq9ab1Guu1pvUbK6m4ow1j1+13kTzEKr1JjWb51303bPzwEc8etIZ61RxxtWaq/UmmodQrbeqanOu1ptoHkK13qRec7XeRPMQqvUmmodQrTfRPIRqvVVVm3O13ooqzljz+FXrTWo2A4zLJFbGX5Lk+N77gUnSWtsic0d+/Pno8u8nOaX3furoCIyvGp1/QJIvJnl0a+2xSW5I8rze+2NG9/HtJDOj665aw+Ne1lr77d77f2TuiJTXZG4x8oIkT++9r2qt/UWS85L8QeaOZJkk2yf5aGvtCb33Xya5ZNS6fC2PtSYr86sl1fW9zVjs8dS9cuaZ38zL/uRFWbVqVd55yLsmmbNOVXpf8Ijts+Wmi/LC3e6VF+52ryTJp7/7s7zj6Q/MstmVOf+KX+bsn268S5NV5jyvWm9Sr7lab1KzuZqKM9Y8ftV6E81DqNab1Gye9/MrfpK73XOHSWesU8UZV2uu1ptoHkK13qqqzblab6J5CNV6k3rN1XoTzUOo1ptoHkK13kTzEKr1VlVtztV6K6o4Y83jV603qdkMMC4zq1YNu7/XWjs3yUt77+etdt7RmVuO3DXJfyY5PsldM/cx1kuSXJ7k2Mx9nPVvZe7jrx+X5H8n2Wx0N8tGt9skyYN6729trS1O8v3e+31ba49O8sEkv0jyyyTn9N6XtNYOTPK80f18K8nrR/dzWu/9C6O+VyV5XZIbkyzM3NLnsa21E+ev11o7bPRYJ7bW3pi5I02elWTf3vuLWmuHJPm9zH3k+NG998eO7vusJC/qvf/4tmZ20+xklyzv7F580rcnnbDBTn35oyadAFNpdkW9b8eLFs6s+0oADOLT518+6YQN9pyHbvyLmMDGodprZa+TAQDg11V7XZ94bT8EzwtgQyxeFP8HZL1s/qQl9f6BYTA3nr5k7N9LBl+aZMNZmhwvS5PA+vLGAAC/CUuTwJ1ZtdfKXicDAMCvq/a6PvHafgieF8CGsDTJ+rI0ydoMsTS5YN1XAQAAAAAAAAAAAKjP0iQAAAAAAAAAAAAwFSxNAgAAAAAAAAAAAFPB0iQAAAAAAAAAAAAwFSxNAgAAAAAAAAAAAFPB0iQAAAAAAAAAAAAwFSxNAgAAAAAAAAAAAFPB0iQAAAAAAAAAAAAwFRZNOgAAAAAAAAAAAIApMeM4f0yWZyAAAAAAAAAAAAAwFSxNAgAAAAAAAAAAAFPB0iQAAAAAAAAAAAAwFSxNAgAAAAAAAAAAAFPB0iQAAAAAAAAAAAAwFSxNAgAAAAAAAAAAAFPB0iQAAAAAAAAAAAAwFSxNAgAAAAAAAAAAAFPB0iQAAAAAAAAAAAAwFSxNAgAAAAAAAAAAAFNh0aQDAAAAAAAAAAAAmBILZiZdwJRzpEkAAAAAAAAAAABgKliaBAAAAAAAAAAAAKaCpUkAAAAAAAAAAABgKiyadABM2qkvf9SkEzbYJT+/YdIJG2zne2wx6QT4jS1aODPpBDZCN928YtIJG2zxJgsnnXCn1y//5aQTNljbYetJJ9zpPeehO0w6AWBsvFYGYJp4f5Y18R4RdwZe1w/j3Euum3TCBnn4zttMOmGDza5YNemEDeb/f6zJZf/vxkknbLD7b7f5pBMA1osjTQIAAAAAAAAAAABTwdIkAAAAAAAAAAAAMBUsTQIAAAAAAAAAAABTwdIkAAAAAAAAAAAAMBUsTQIAAAAAAAAAAABTwdIkAAAAAAAAAAAAMBUWTToAAAAAAAAAAACAKTHjOH9MlmcgAAAAAAAAAAAAMBUsTQIAAAAAAAAAAABTwdIkAAAAAAAAAAAAMBUsTQIAAAAAAAAAAABTwdIkAAAAAAAAAAAAMBUsTQIAAAAAAAAAAABTwdIkAAAAAAAAAAAAMBUsTQIAAAAAAAAAAABTwdIkAAAAAAAAAAAAMBUWTToAAAAAAAAAAACAKTEzM+kCppwjTQIAAAAAAAAAAABTwdIkAAAAAAAAAAAAMBUsTQIAAAAAAAAAAABTwdIkAAAAAAAAAAAAMBUWTTqA4a1cuTKHHrwkF/aeTTfdNG9/xyG5z847TzrrNlXrTWo2z87enCPe9bZcdeXlWbBgYfY78G+y4873m3TWbao442rN1XoTzUOo1pvUbJ69+eYcvOSgXHH5Zbl5+c35s1e/Jk/afY9JZ92mijOu1vyD/zg/Hzvu/Vny3mNz5WWX5oPvWZKZmZnsdN/755Wvf0sWLNj4/luoajNONA+hWm9Sr7lab6J5CNV6E81DqNabaB5Ctd6kXnO13kTzEKq9N5vUm3FSr7na+0NJvRkn9Zqr9Saax+2i75+fT3z0A/mrdx+TSy66MCcf854sWLAwizbZJK85YEm2udu2k05co0ozXt13zzs3Rx1xeI494eRJp6xTxRlXa67WmyRf+pdP5cuf/3SSZPny5fnRD3s+9s9fzlZb32XCZQDDu0N+u9pa2721dlVr7Wutta+31s5qre22gffx1tbao3+Dht9rre2zlsv/tLV22Gqn92+tndFau9vtfcxb3f+PW2uL1/CYz7kj7v+O9NWvfDnLly3Pyad+Ivv/xQF573sOW/eNJqhab1Kz+d/P+kZWrFiRvz36pPzRy/fJycd9YNJJa1VxxtWaq/UmmodQrTep2fz5z30m22xz1xz70VNyxNEfzuGHHTLppLWqOONKzZ/6xEk55u8Ozs3LlydJTjrm7/KiP3tt3vm+47Jq1ar8+xlfn3DhmlWa8TzN41etN6nXXK030TyEar2J5iFU6000D6Fab1KvuVpvonkI1d6bTerNOKnXXO39oaTejJN6zdV6E83j9Ll/+Pscf+Sht7xveMqH35uX7ntg/urdx+S/P/4p+ew//P2EC29blRmv7qQTjsvBSw7KsmXLJ52yXirOuFpztd4k2esZz827jzo+7z7q+Dyg/Xb23f/NFiaBqXVHHpLmq7333XvvT07ytiQHb8iNe++H9d6/dXsfvPf+hd77setz3dbagUmenWSv3vs1t/cx16PpxN77p8d1/7fXd87+dh7/hCcmSXZ9+CNywQXnT7ho7ar1JjWb773jzlmxYkVWrlyZG2+4PosWbdwHoq0442rN1XoTzUOo1pvUbN7zaXvnNa97w9yJVcnChb4n39EqNd9zhx3zl29/zy2nf/SD7+fBuz4qSbLbox+f886+3S+hx6rSjOdpHr9qvUm95mq9ieYhVOtNNA+hWm+ieQjVepN6zdV6E81DqPbebFJvxkm95mrvDyX1ZpzUa67Wm2gep+3utWPecNC7bzn9urcemp3v/8AkycoVK7LppptNKm2dqsx4dTvutFMOf99Rk85YbxVnXK25Wu/qLvz+BfnJxRfl6c/5g0mnAEzMuH7CuluSq1prD0vy/iQzSa5O8ookuyV5S5LlSX4ryWm990NbaycmOS3JGUmOS3LXJDsk+WDv/UOtta8lOTfJQ5Ncn+Tfkuw9ut7Tkjw3yYOSHJPk40kuTXL/JN/qvb92Pqy19tdJnpDkmb33ZaPz/iDJ65JskmRVkuePHmdNnQ9IcmKSm5NckuS+vffd1zSE1tqSJFcm+X6S/5FkWZKdRo17JHl4kiN77x/aoOn+hpYuvT5bb73VLacXLliY2dnZjfaNmGq9Sc3mxVtskauuvDx//tLn5xfXXZu/OezISSetVcUZV2uu1ptoHkK13qRm8xZbbJkkWbp0ad76l2/MvvNvkG+kKs64UvNjn7hnrrry8l+dsWpVZmZmkiSbb75lbrjh+gmVrV2lGc/TPH7VepN6zdV6E81DqNabaB5Ctd5E8xCq9Sb1mqv1JpqHUO292aTejJN6zdXeH0rqzTip11ytN9E8Tr/zhD3ynz/71fuGd737PZIkP/jeefnSZ/8hf/23H55U2jpVmfHq9txr71x+2U8nnbHeKs64WnO13tV98uTj8+I/23fSGQATdUceaXKP0cdzn5nko5lbgPxIkteNlgr/JcmbR9fdOckLkjx2tfPmPSBzC4pPy9wy5JtWu+xbvfc9k2yW5Ibe+15Jvpfkybe6jwcmeWWSRyd5Rmtt+9H5f5JkzyTbZ26Rc/XrP7P3/oTR/e29ls73JHlX7/0pSb65HnOZt+Povl6b5KAkL03y9CSv2YD7uENsueVWWbp06S2nV65auVH/w12tN6nZ/OlPnpLdHv24HPOxT+XIEz6RI971tixftmzSWbep4oyrNVfrTTQPoVpvUrM5SX525RX581f/aZ7+rGdn72c8a9I5a1VxxhWb580s+NVL+BtvXJott9xqLdeenIoz1jx+1XqTes3VehPNQ6jWm2geQrXeRPMQqvUm9Zqr9Saah1Dtvdmk3oyTms2V3h9Kas64WnO13kTz0M76+pdy4gcOywFL3pe7bHO3SefcpsozrqLijKs1V+udd/0vf5Gf/uTHefgjf2fSKQATNY6P535c5o4medro69Gjo0S+Ism9R9f9bu99tve+NMmNt7qfnyV5XmvtlMwtF26y2mVnj75em7nlxiS5JsniW93HD3vvv+y9r0hyxWqXfyfJU5N8JckHVrv+VUlOaq19NMmuqz3mmjp/O3NHw0zmjna5vs7vvd88ar+o9778NtrHbrfdHplvnH56kuS8c8/JLrs8cOiEDVKtN6nZvOXWd7ll6WLrrbfJihWzWbly5YSrblvFGVdrrtabaB5Ctd6kZvPVV/88b3jtq7Pf/m/Kc573gknnrFPFGVdsnnffB7RccO6/J0m+860z8tsP223CRWtWccaax69ab1KvuVpvonkI1XoTzUOo1ptoHkK13qRec7XeRPMQqr03m9SbcVKvudr7Q0m9GSf1mqv1JpqH9M2vfj5f/uwn8z/e/aFsd697r/sGE1R1xpVUnHG15mq9884/9+w84lGPmXQGwMSNa839Z6Ov5yV5We/9J621301yr9H5q9Zy2wOSnDn6SO6nJHnmapet7Xaru63rfa/3vrK19ldJzmytvTTJp5O8I8l9Rtf5Un51FMo13c/5SR6X5POZOwLl+lrf9rHb46l75cwzv5mX/cmLsmrVqrzzkHdNOmmtqvUmNZuf+8KX5P3vXpK37veKzM7enJe+6vVZvPnmk866TRVnXK25Wm+ieQjVepOazSced2x+8YvrcsKxx+SEY49Jkrzvgx/O4sWD/7cW66XijCs2z3vZa96YD//doZmd/WDufZ/75rFP3HPSSWtUccaax69ab1KvuVpvonkI1XoTzUOo1ptoHkK13qRec7XeRPMQqr03m9SbcVKvudr7Q0m9GSf1mqv1JpqHsnLFipxyzHuz7Xb3zPsPeUuS5EEPe2R+/yX7TLhszSrOuJqKM67WXK133k9/8uNsv8PGvVjNlJj5/9m78yhLq/re/5+iC2xGQUGZ2saJTZwQogQVY8uoRE2MiXI18RKNilfjhOSiMTKpYEBBcQIUURTRxJ9eiZh4kYtEBU0cQFC2giMioIAMLfRYvz/qNGmxhyrlPE9tntdrrVpd59QZ3r3XqTqnTn3Xfu7Off5g9iampn7/Wb5SyqIkn8j07o8rkmye5D1Jvp3kbZkezpzK9CGzt09ySK31oNF1r621bltKOSPTu1MuSXJykhsyvSvjI5I8LMm/j653RSnl7CTvq7VeUEo5KcnFmd6xcZck78v04b33HN3+xUkOSrIoyS611sNH55ckFybZO8kRSXZKsjzTuz9+JdOH3l5T54OTnD667M1JNq+17ldK+dGoedWCnpVkiyTXJrli1W2VUnYZtS8qpWyZ5OJa6y7rWt87ls+dgUvmhh//8td9J8zawq036TsBYCzuWLai74RZm7/hvL4T7vHqNbf2nTBrZfvN+04AAABogvdnWRPvEQEzdcmPb+47YVZ2XXjvvhNmbfmK9v68PjlvYv0XYnB+duNdD9w69z34fht7MDMjG+97XHs/rOnM7ecdPvafJXfL0OTdoZTy0SQfqLWe33fLupRSnpfkq7XWK0spf5vk8bXWF4zzPg1NclfelAOYO7whzpoYmgQAALjn8v4sa+I9ImCmDE2On6FJ7ikMTXJPZmiSdeliaHJch+eelVLK6zK9m+Q3+26ZgZ8mObuU8utM76r5wp57AAAAAAAAAAAAgBmYE0OTtdZjkxzbd8dM1FovTPKYvjsAAAAAAAAAAACA2dmg7wAAAAAAAAAAAACALhiaBAAAAAAAAAAAAAbB0CQAAAAAAAAAAAAwCIYmAQAAAAAAAAAAgEEwNAkAAAAAAAAAAAAMgqFJAAAAAAAAAAAAYBAm+w4AAAAAAAAAAABgICYm+i5g4Ow0CQAAAAAAAAAAAAyCoUkAAAAAAAAAAABgEAxNAgAAAAAAAAAAAINgaBIAAAAAAAAAAAAYBEOTAAAAAAAAAAAAwCAYmgQAAAAAAAAAAAAGwdAkAAAAAAAAAAAAMAiGJgEAAAAAAAAAAIBBMDQJAAAAAAAAAAAADMJk3wEAAAAAAAAAAAAMxIR9/uiXRyAAAAAAAAAAAAAwCIYmAQAAAAAAAAAAgEEwNAkAAAAAAAAAAAAMgqFJAAAAAAAAAAAAYBAm+w4AZm/h1pv0nTBrT3/fxX0nzMo5h+zZdwIM1vIVU30nzMr8Def1ncAcVLbfvO8EABr2Xz+8qe+EWXvMA7fqOwFowK9+vazvhFnbcpMN+064x2vtfYCkzfdnGT/vEQEztevCe/edcI83OW+i7wS4W+xwn437TgC4x7LTJAAAAAAAAAAAADAIhiYBAAAAAAAAAACAQTA0CQAAAAAAAAAAAAyCoUkAAAAAAAAAAABgEAxNAgAAAAAAAAAAAIMw2XcAAAAAAAAAAAAAAzEx0XcBA2enSQAAAAAAAAAAAGAQDE0CAAAAAAAAAAAAg2BoEgAAAAAAAAAAABgEQ5MAAAAAAAAAAADAIBiaBAAAAAAAAAAAAAbB0CQAAAAAAAAAAAAwCIYmAQAAAAAAAAAAgEEwNAkAAAAAAAAAAAAMgqFJAAAAAAAAAAAAYBAm+w4AAAAAAAAAAABgICbs80e/PAIBAAAAAAAAAACAQTA0CQAAAAAAAAAAAAyCoUkAAAAAAAAAAABgEAxNAgAAAAAAAAAAAINgaBIAAAAAAAAAAAAYBEOTAAAAAAAAAAAAwCBM9h1A91auXJk3H3NkvldrNtpooxxx1JvygIUL+85aq9Z6E83jssFE8uonPygLtto4U1NTeccFP8yPbrw9SXLIXgtz9U23518vv77nyrVrYY1X11pvorkLrfWu7tuXXpKTTzohp55+Zt8p69XaOrfWm2juQmu9ieYutNabtNfcWm+iedx+UC/Lv3zw3fn7496ba37yw3zoXccmU1O53/YLcvArXp958+bmW0MtrfEqrTW31pto7kJrvau76cYb8qK/fnbe9u7TsnCnB/Wds1YtrnGLzYn3AcZN8/i11pto7kJrvYnmLrTWm7TX3FpvorkLrfUmbTYDjMuc2GmylLKolHJ9KeWCUsoXSykXl1J267jh2tG/R5ZSDrnL1y4upexUSjm4lPKMtVx/USnl7C5af1/nf+G8LF2yNGee9fG88tWH5m3HH9d30jq11ptoHpc9d9oqSfKqT16eD3716vzNngty7/mTefPTd8njHrhVz3Xr18Iar6613kRzF1rrXeVDp78/xxz5hixZsrTvlBlpbZ1b6000d6G13kRzF1rrTdprbq030TxOn/uXM3PGO4/NsmXTr4E++eH35s+f/9K87vjTkiSXfPVLfeatUytrvLrWmlvrTTR3obXeVZYvX5YTjj0q95o/v++U9WpxjVts9j7A+Gkev9Z6E81daK030dyF1nqT9ppb6000d6G13qTNZoBxmRNDkyPn11oX1VqflOSNSY7pO+iuaq1n1Fo/03fH7+ub3/h6Hr/XE5Mkj9r10bn88st6Llq31noTzePylR/elBP/3w+SJPfffKPctnRFNt5wXs782tU5r/6y57r1a2GNV9dab6K5C631rrLjggU54cST+86YsdbWubXeRHMXWutNNHehtd6kvebWehPN47TNdjvkZa8/9s7TL3vdsSmP2C3Lly3LLTfdmI033azHunVrZY1X11pza72J5i601rvKe046IX/658/J1ltv03fKerW4xi02ex9g/DSPX2u9ieYutNabaO5Ca71Je82t9Saau9Bab9JmM8C4zM1jMCVbJbm+lPLIJO9MMpHkhiQvSLJbkrcmWZrk1CSHJflikkclmUryp0luS3JKkgVJtkvymVrrG0opZyS57+jj6Un+KcnDk1yV5F7riyqlHJnk2tFtn5xkjyQbJTkiyc2jy2yS5JNJPlJr/Wgp5dgkT0wyL8nbk3w+yTeS7FxrXVFKeWuSr9daP/G7LdXsLV58Wzbf/L//MDJvg3lZvnx5Jifn5sOhtd5E8zitnEoO2/fBecKDtsoxn/t+rr11Sa69dUkeu3DLvtPWq5U1XqW13kRzF1rrXWWf/Q7INT+7uu+MGWttnVvrTTR3obXeRHMXWutN2mturTfRPE6PecLe+eV119x5eoN58/LL63+et/3D32XjTTfLggc+tMe6dWtljVfXWnNrvYnmLrTWmySfO+fT2XKr+2SPxz0hHz3jtL5z1qvFNW6x2fsA46d5/FrrTTR3obXeRHMXWutN2mturTfR3IXWepM2mwHGZS7tNLn36PDcFyX5YJKzk5yW5GW11kVJzk3y96PLzq+1PrHWemaSLZJ8bLRD5c+SPDXTw5IX11oPyPRg4+qH2z6/1vr4JE8e3c6eSV6XZJPVLvOaUcsFpZQLkjzsLq1/lmTrWuseo9t5zOj8zZKck+S9o4HJpyZ5YK11r9Hl/iHTA6BfSnJAKWXeqPfTv+Oa/U423XSzLF68+M7TK6dWzuknwdZ6E83jdvx5V+VvPnJJXv3kB2X+5Fz6MbZuLa1x0l5vorkLrfW2qrV1bq030dyF1noTzV1orTdpr7m13kRz17a+33Y59rR/yaKnPjMff/87+s5ZqxbXuLXm1noTzV1orTdJzj3nU/mvr16UV77k4Fz5vZq3HPH63PDLuXtEkhbXuMXm1rS4xprHr7XeRHMXWutNNHehtd6kvebWehPNXWitN2mzmXuwiQkfPtb+0YG5NG206vDcj8v0bpJnj/59z2hw8QVJdhhdtt7lut8c/fvTJPOT3JjksaWUjyY5Mb+5i+Sq6+6c5GtJUmv9yei6q7x91LJoNLD5nbvcX0ly0ei6N9Va/3F0/pOSbLza/T0yyR+O+v8tyYZJdsr0MOjBmR6YPK/WunQd63K322233fOlCy9Mklx6ybfy0Ifu3OXdz1prvYnmcdm3bJ2D/nD7JMmSZSuzMlNZOTXVc9XMtbDGq2utN9HchdZ6W9XaOrfWm2juQmu9ieYutNabtNfcWm+iuUvvPPq1ue5nP0mSzN94k0xs0M2bT7+LFte4tebWehPNXWitN0lOPvVDeeepZ+Qdp5yRh+xc8vqj3pL7br1131lr1eIat9jcmhbXWPP4tdabaO5Ca72J5i601pu019xab6K5C631Jm02A4zLXB0Zv27076VJnl9r/Ukp5QmZPtR2kqy8y+XvOjV1cJJf1VpfUkp5SJIXl1JW/SVg1XW/k+SgJO8opWyf/x7InInvJvnLJCml3DvJJ5Icm+SzSV6Z5D9KKV9OckWS/1drfXEpZYMk/5jkqlrrraWUdyR5YZI3zOJ+7xZ777tfLrroy3n+8w7K1NRUjn7TW7pOmJXWehPN4/Klq27Ma/d5cN72zIdlct5E3vsfP87SFe0MTbawxqtrrTfR3IXWelvV2jq31pto7kJrvYnmLrTWm7TX3FpvorlLB/7l83P6Scdk3uSG2ehe83PwK17fd9JatbjGrTW31pto7kJrvS1qcY1bbG5Ni2usefxa6000d6G13kRzF1rrTdprbq030dyF1nqTNpsBxmViag7s0lZKWZTpwcPvJFmRZPMk70ny7SRvy/Rw51Smhwy3T3JIrfWg0XV/lGSXWusdpZTjMj2o+J9Jzkpyc5IlmT5c9z5J3pzk7Frrv42GKN+V6UNr/zjJ42qtC0opRya5ttb6vtX6Ls70gOXBSa5NckqSd2Z6J8zJJEcluX1VVynlf4wu+5RR/2MzfejuT9Vajx7d5muS/OVoZ811umP5bw2FQnOe/r6L+06YlXMO2bPvBBis5Q0NQyfJ5Ly5u0MTANCm//rhTX0nzNpjHrhV3wlAA37162V9J8zalpts2HfCPV5r7wMk3gsAAIC1mT8ZL5aZkY2femJ7uaganAAAIABJREFUvwzSmds/9+qx/yyZE0OTQ1RKOSzJDbXW09d3WUOT3BMYmgRmqrU/lvhDCQBwdzM0CdxTGZpkTVp7HyDxXgAAAKyNoUlmytAk69LF0ORcPTz3PVop5YxM75j59J5TAAAAAAAAAAAAYDAMTfag1npw3w0AAAAAAAAAAAAwNBv0HQAAAAAAAAAAAADQBUOTAAAAAAAAAAAAwCAYmgQAAAAAAAAAAAAGwdAkAAAAAAAAAAAAMAiTfQcAAAAAAAAAAAAwEBP2+aNfHoEAAAAAAAAAAADAIBiaBAAAAAAAAAAAAAbB0CQAAAAAAAAAAAAwCIYmAQAAAAAAAAAAgEEwNAkAAAAAAAAAAAAMgqFJAAAAAAAAAAAAYBAMTQIAAAAAAAAAAACDYGgSAAAAAAAAAAAAGARDkwAAAAAAAAAAAMAgTPYdAAAAAAAAAAAAwEBMTPRdwMDZaRIAAAAAAAAAAAAYBEOTAAAAAAAAAAAAwCAYmgQAAAAAAAAAAAAGwdAkAAAAAAAAAAAAMAiGJgEAAAAAAAAAAIBBMDQJAAAAAAAAAAAADMJk3wHAMJxzyJ59J8zKVgce33fCrN107mF9JzAH3bFsRd8JszZ/w3l9JwAA9OoxD9yq7wSAsdhykw37Tpi15Sum+k6Ytcl5E30nzEprvXTD9x4AAADjZKdJAAAAAAAAAAAAYBAMTQIAAAAAAAAAAACDYGgSAAAAAAAAAAAAGITJvgMAAAAAAAAAAAAYiAn7/NEvj0AAAAAAAAAAAABgEAxNAgAAAAAAAAAAAINgaBIAAAAAAAAAAAAYBEOTAAAAAAAAAAAAwCAYmgQAAAAAAAAAAAAGwdAkAAAAAAAAAAAAMAiGJgEAAAAAAAAAAIBBMDQJAAAAAAAAAAAADIKhSQAAAAAAAAAAAGAQJvsOAAAAAAAAAAAAYCAm7PNHvzwCAQAAAAAAAAAAgEEwNAkAAAAAAAAAAAAMgqFJAAAAAAAAAAAAYBAMTQIAAAAAAAAAAACDYGgSAAAAAAAAAAAAGARDkwAAAAAAAAAAAMAgGJoEAAAAAAAAAAAABmGy7wC6t3Llyrz5mCPzvVqz0UYb5Yij3pQHLFzYd9ZatdabaO5Ca73bbLlJvvLuv86fHP7PuenWO/LuV+2frTafn3kbbJAX/tO5+eHPf9V34m9pbY0TzV1YvmxZjjnyDfn5NT/LsqXL8jcvekn+eNHefWetU2trnLTX3FpvorkLrfUmmrvQWm/SXnNrvYnmLrTWm2juQmu9ieYutNabtNmcJN++9JKcfNIJOfX0M/tOWa8W17i15tZ6kzabE99746Z5/FrrTTR3obXepL3m1noTzV1orTdpsxlgXOb0TpOllEWllOtLKReUUr5YSrm4lLLb73A7h5RSjhx9vnR0exeUUr5aSjm6lDIxi9t6eSnlu6WU58y2Y644/wvnZemSpTnzrI/nla8+NG87/ri+k9aptd5Ecxda6p2ct0He9cr9c/uS5UmSN//tk/Lx87+b/Q49O0ee8R8pC+7Tc+GatbTGq2gev8999pzc+95b5tQPfiQnveeUnHDcm/pOWq/W1jhpr7m13kRzF1rrTTR3obXepL3m1noTzV1orTfR3IXWehPNXWitN2mz+UOnvz/HHPmGLFmytO+UGWlxjVtrbq03abPZ9974aR6/1noTzV1orTdpr7m13kRzF1rrTdpsBhiXOT00OXJ+rXVRrfVJSd6Y5Jjf8/ZuHN3eoiR7Jrl/kpfP4vp/nuTZtdaP/54dvfnmN76ex+/1xCTJo3Z9dC6//LKei9attd5Ecxda6j3uxYty2r9+Kz+/4bYkyeMevkN22GbzfPa4Z+egvR+WCy/9ac+Fa9bSGq+iefz22f+AvORlr5g+MZXMmzf3N61ubY2T9ppb6000d6G13kRzF1rrTdprbq030dyF1noTzV1orTfR3IXWepM2m3dcsCAnnHhy3xkz1uIat9bcWm/SZrPvvfHTPH6t9Saau9Bab9Jec2u9ieYutNabtNkMMC5zf9LhN22V5PpSygVJrk9ynyTPSnJaki2TbJ/k3bXW95ZS9kryjiQ3JVme5OK73litdaqU8rYkpyc5uZTyvCSvSrIkyfeTvDjJ85K8INMDpmcm2T3JB0Y7TZ6d5KdJdhp9/ogkuyX5bK319aWUJyU5YnTdzZI8N8nSJB8bXe/BSb5Wa33p3bdE67d48W3ZfPPN7jw9b4N5Wb58eSYn5+bDobXeRHMXWun9q/0enl/c/Ouc9/Uf5bCD/ihJsvD+W+SmW+/Inxz+ibzueY/Loc/eI8d8+Ms9l/62VtZ4dZrHb5NNNk2SLF68OIe/9lU5ZNUA5RzW2hon7TW31pto7kJrvYnmLrTWm7TX3FpvorkLrfUmmrvQWm+iuQut9SZtNu+z3wG55mdX950xYy2ucWvNrfUmbTb73hs/zePXWm+iuQut9SbtNbfWm2juQmu9SZvN3INNzPigwDAWLew0uffoUNoXJflgpocTk+RjtdZ9Mz14eHatdf8k+yd5zejr703yP0aX+eE6bv+6JFuXUu6b5Kgke9da90ryqyQvGV3mplrrXrXWU5J8K8nzk0wleVCSFyZ5WqZ3wHxNkj8anZckD0/yV6NdLf+/JH85On/n0WX2SHJgKWXb2S/L727TTTfL4sWL7zy9cmrlnH4SbK030dyFVnr/51MemX123yn/fvxz8qgH3y8fOOzArFg5lc9edGWS5NyLr8ruO3f6I2DGWlnj1WnuxnXX/jz/60UH56lPe3oOOPBpfeesV4tr3Fpza72J5i601pto7kJrvUl7za31Jpq70FpvorkLrfUmmrvQWm/SZnNrWlzj1ppb603abG5Ni2usefxa6000d6G13qS95tZ6E81daK03abMZYFxaGJpcdXjux2V6F8ezk2ycpI6+fl2SPyulfCTJG5JsODr//rXW740+X9cWbguTXJ3pAcjLa623js6/MNNDj1ntvu7qB7XWmzM9YHldrfXGWusdmR6oTJKfJXlnKeWMJE9ere3KWuuttdYVSX6eZP46V+Butttuu+dLF16YJLn0km/loQ/ducu7n7XWehPNXWild79Dz87+rz07Bxz28Vx61fV54fHn5nNfvSoH7PGgJMlej9wx3/3xL3uuXLNW1nh1msfvhht+mVe89EV5+Stfk2f82bP6zpmR1tY4aa+5td5Ecxda6000d6G13qS95tZ6E81daK030dyF1noTzV1orTdps7k1La5xa82t9SZtNremxTXWPH6t9Saau9Bab9Jec2u9ieYutNabtNkMMC6tjYxft9rnK0f/HprkotEhuZ+c5E9G5/+slPIHtdbvJnlspg/T/RtKKRskeW2mBzF/mORhpZRNa62Lkzwpyaqhy5V3ve7I1FrOX+W0JA+utd5aSvlQklV7y67vemO197775aKLvpznP++gTE1N5eg3vaXPnPVqrTfR3IXWeld3+CkX5D2vOSAvftqjc/Ovl+TgY/+176Q1anGNNY/fGe8/NbfccnNOP/V9Of3U9yVJTnz3KZk/v9P5/1lpbY2T9ppb6000d6G13kRzF1rrTdprbq030dyF1noTzV1orTfR3IXWepM2m1vT4hq31txab9Jmc2taXGPN49dab6K5C631Ju01t9abaO5Ca71Jm80A4zIxNdXr/N46lVIWJflEku8kWZFk8yTvSXJwkkNqrVeMBiVPTnJDpnd8fESShyXZNcm7k9yS5NYk36q1HllKWZrkK5kehNwwyf9NckytdaqU8twkrxp97cokf5vkoCS71FoPHzVdkOSQJHdk+rDge5ZS5ie5ota60+gy19Zaty2lvD3J3kkWZ3rg84Ykb151vdFlL05yUK31R2tbhzuW9ztkCUO01YHH950wazede1jfCcxBdyxb0XfCrM3fcF7fCQAAAJAkWb6ivbdmJ+dNrP9CMMf53gMAaNP8yXhRxIxs/Iz3tvein87c/pmXjv1nyZwemmSaoUnonqFJ7ikMTQIAAMDvzuAW9MP3HgBAmwxNMlOGJlmXLoYmNxj3HQAAAAAAAAAAAADMBYYmAQAAAAAAAAAAgEEwNAkAAAAAAAAAAAAMgqFJAAAAAAAAAAAAYBAMTQIAAAAAAAAAAACDMNl3AAAAAAAAAAAAAAMxYZ8/+uURCAAAAAAAAAAAAAyCoUkAAAAAAAAAAABgEAxNAgAAAAAAAAAAAINgaBIAAAAAAAAAAAAYBEOTAAAAAAAAAAAAwCAYmgQAAAAAAAAAAAAGwdAkAAAAAAAAAAAAMAiGJgEAAAAAAAAAAIBBMDQJAAAAAAAAAAAADMJk3wEAAAAAAAAAAAAMxMRE3wUMnJ0mAQAAAAAAAAAAgEEwNAkAAAAAAAAAAAAMgqFJAAAAAAAAAAAAYBAMTQIAAAAAAAAAAACDYGgSAAAAAAAAAAAAGARDkwAAAAAAAAAAAMAgGJoEAAAAAAAAAAAABmGy7wCAueimcw/rO2HWLr/6lr4TZu3hO27Rd8I93vwN5/WdwBy0+xs/33fCrH3j6P37TgAYi+UrpvpOmLXJeRN9JwBAZzzvjZ/XQ6yJNQZmyvMIAAC/CztNAgAAAAAAAAAAAINgaBIAAAAAAAAAAAAYBIfnBgAAAAAAAAAAoBsT9vmjXx6BAAAAAAAAAAAAwCAYmgQAAAAAAAAAAAAGwdAkAAAAAAAAAAAAMAiGJgEAAAAAAAAAAIBBMDQJAAAAAAAAAAAADIKhSQAAAAAAAAAAAGAQDE0CAAAAAAAAAAAAg2BoEgAAAAAAAAAAABgEQ5MAAAAAAAAAAADAIBiaBAAAAAAAAAAAAAZhsu8AAAAAAAAAAAAABmJiou8CBs5OkwAAAAAAAAAAAMAgGJoEAAAAAAAAAAAABsHQJAAAAAAAAAAAADAIhiYBAAAAAAAAAACAQTA0CQAAAAAAAAAAAAyCoUkAAAAAAAAAAABgEAxNAgAAAAAAAAAAAINgaBIAAAAAAAAAAAAYhMm+A+jeypUr8+Zjjsz3as1GG22UI456Ux6wcGHfWWvVWm+iuQut9SZtNV95xWX52AdOzj8ef8qd5535vrdnux0XZt+nPavHsnVraY1Xaa25td5E8zhtMJEc/cyHZ6dtNsnUVHLU//lubrxtaY5+5sOyxcYbZoMNktf982X56Y239536W1pZ49W11txab6K5C631Jm02J8m3L70kJ590Qk49/cy+U9arxTVurbm13kRzF1rrTTR3obXepL3m1nqTNpsTr4fGTfP4tdabaO5Ca71Jm82J55Fxa625td5Ecxda603abAYYl7HuNFlKWVRKub6UckEp5YullK+UUp49zvsc3e82pZRPllI+P7rP95dSNl7LZXcqpVz8e97ffUopzx19/oBSytNHn19QStnl97ntcTj/C+dl6ZKlOfOsj+eVrz40bzv+uL6T1qm13kRzF1rrTdppPucTH85pJ74py5YuTZLc8qub8tZ/eEW+fvGFPZetXytrvLrWmlvrTTSP05N32SZJ8len/Gfe+X+vzKv2e0gOfcpD86+X/DzPP236vAdus2nPlWvWyhqvrrXm1noTzV1orTdps/lDp78/xxz5hixZsrTvlBlpcY1ba26tN9HchdZ6E81daK03aa+5td6kzWavh8ZP8/i11pto7kJrvUmbzZ5Hxq+15tZ6E81daK03abOZe66JiQkfPtb60YUuDs99fq11Ua31SUn2T/K/SymPHvN9Hpbk/9Za96+1Pj7JbUkOGeP9PSrJM0af753kCWO8r9/bN7/x9Tx+rycmSR6166Nz+eWX9Vy0bq31Jpq70Fpv0k7z/bfbMa964z/defqOO36dZ/31i7PXPgf2WDUzrazx6lprbq030TxOX/juL3LEp7+TJNl+y41zyx3Ls/vCLXP/LebnAy/4wzxt1+3ynz+4qefKNWtljVfXWnNrvYnmLrTWm7TZvOOCBTnhxJP7zpixFte4tebWehPNXWitN9HchdZ6k/aaW+tN2mz2emj8NI9fa72J5i601pu02ex5ZPxaa26tN9HchdZ6kzabAcal08Nz11pvK6WckuQ5pZSXJ1mQZLskn6m1vqGUckaS+44+jk9yeJKlSU5N8pMkb06yIslVSV6S5HlJDkyySZIHJ3lrrfWMJNcl+YtSypVJvpzktUmmkqSU8oYkf5bp//t7k/x7km1KKZ8etVxaa33RqOXsWuu/lVKekuSgWuvBpZS/TPKaUceXaq2HJ/mHJLuWUl6S5NVJNimlfGXV/7uUcu8kHxj9v5LkFbXWb5dSfpzkiiTfqbW++m5Z5BlYvPi2bL75ZneenrfBvCxfvjyTk3PzaO2t9Saau9Bab9JO8x5P3Du/uPaaO0/fb9sdcr9td8i3/vMr67jW3NDKGq+utebWehPN47Zi5VTe8hePyL4Pu19eddYl+ZNdt80tty/LC0//el6694PywiftlHedd1Xfmb+lpTVepbXm1noTzV1orTdps3mf/Q7INT+7uu+MGWtxjVtrbq030dyF1noTzV1orTdpr7m13qTNZq+Hxk/z+LXWm2juQmu9SZvNnkfGr7Xm1noTzV1orTdpsxlgXLrYafKurkuyW5KLa60HJNkjv7kL5Pmj3SFvSjK/1vrEJB9JclqSPx/tWPmzJAePLn/vWuvTMr3T4+Gj805Mclamd5y8JsmnkmxfStktyVOT/NHofndOMpFkiyR/k+RxSfYppdxvTeGllPskOSrJPrXWvZLsUErZL9PDnOfXWk9JclySs2qtn1ntqq9P8oVa65OTvDjTw5rJ9NDoc7scmEySTTfdLIsXL77z9MqplXP6SbC13kRzF1rrTdpsbk2La9xac2u9ieYuvP5fLsuBb/9Sjn7mw3LrHctz/nd/kSS54Lu/yCN22KLnujVrbY2T9ppb6000d6G13qTN5ta0uMatNbfWm2juQmu9ieYutNabtNfcWm/SZnNrWlxjzePXWm+iuQut9SZtNremxTVurbm13kRzF1rrTdpsBhiXPoYmFyb5SpLHllI+mukBx3ut9vW6hs+3yfQukJ8opVyQ6cN8Lxx97Vujf3+aZP7o872TfHg0lLltkq8lOSlJSfK1WuuKWuvSWuuhmd6B8ge11ptqrSuTXJ/pnStXt+pg6Q8ZtZw76nhYpne4XJ9HJnnB6DqnJbnP6Pxf1lpvmMH171a77bZ7vnThhUmSSy/5Vh760J27TpiV1noTzV1orTdps7k1La5xa82t9Saax+npj94uL3rSA5Mkty9bkZVTyX/98Kb8cdk6SfKHD9wqV163eF030ZtW1nh1rTW31pto7kJrvUmbza1pcY1ba26tN9HchdZ6E81daK03aa+5td6kzebWtLjGmsevtd5Ecxda603abG5Ni2vcWnNrvYnmLrTWm7TZDDAunY6Ml1K2SPKiJO9Psmmt9SWllIckeXEpZdVg4srVrrLq818muTrJn9Zaby6lPCPJbUkekNFht+/iFUm2z/Tg5JJSyuVJ/iDTh8J+aSllgyTzkpyb5OVruY07Mj2omSS7j/79YaaHM/ertS4rpRyc6aHNLfLfA6gr89vDqFck+Uit9azRLpZ/u4b/a2f23ne/XHTRl/P85x2UqampHP2mt/SRMWOt9Saau9Bab9Jmc2taXOPWmlvrTTSP03mXX583/8XD8+EXPTaT8yZy3GevyBU/vzVHP/PhOeiPFuS2O5bnsI9f2nfmGrWyxqtrrbm13kRzF1rrTdpsbk2La9xac2u9ieYutNabaO5Ca71Je82t9SZtNremxTXWPH6t9Saau9Bab9Jmc2taXOPWmlvrTTR3obXepM1mgHGZmJpa07zg3aOUsijJJ5J8J8mKTA9pviPTO0ieleTmJEsyfZjqfTJ9mOuza63/NrruIbXWg0a3tX+SN2Z6IPGWJM9PcmCSXWqth5dS5ie5ota6Uyll+yTvyfRulLcn+UWSl9ZarymlvC7Th/LeINOHyb5gdJ97ju7n4iQHJdk6yemZ3nnye0k2qbUeXEr5qyT/K9NDlz/K9GG9t0pyXpJTknwxydlJjsj0YccPGd3/B5JsmekByyNrrZ8ppVxba912fet4x/I1DnUC/IbLr76l74RZe/iOc/OwvXBPt/sbP993wqx94+j9+04AGIvlK9r7dW9y3sT6LwQAMENeDwHw+/A8AjC3zJ+MH3LMyCbPOr29J3E68+tPvmDsP0vGOjTJ3cPQJDAThiaBmTI0CTB3+OMOADB0Xg8B8PvwPAIwtxiaZKYMTbIuXQxN3vUw0gAAAAAAAAAAAAD3SIYmAQAAAAAAAAAAgEEwNAkAAAAAAAAAAAAMwmTfAQAAAAAAAAAAAAzDxMRE3wkMnJ0mAQAAAAAAAAAAgEEwNAkAAAAAAAAAAAAMgqFJAAAAAAAAAAAAYBAMTQIAAAAAAAAAAACDYGgSAAAAAAAAAAAAGARDkwAAAAAAAAAAAMAgGJoEAAAAAAAAAAAABsHQJAAAAAAAAAAAADAIhiYBAAAAAAAAAACAQZjsOwAAAAAAAAAAAICBmOg7gKGz0yQAAAAAAAAAAAAwCIYmAQAAAAAAAAAAgEEwNAkAAAAAAAAAAAAMgqFJAAAAAAAAAAAAYBAMTQIAAAAAAAAAAACDYGgSAAAAAAAAAAAAGARDkwAAAAAAAAAAAMAgGJoEAAAAAAAAAAAABmGy7wDo2/IVU30nzNrkvIm+E+7xLr/6lr4TZu3hO27Rd8KsXVB/0XfCrCwq2/SdAHeLbxy9f98JAIx4bT9+fucDgLnN8143WntNdNuS5X0nzNqWm2zYdwIMkucRYKY+c9k1fSfM2jMesX3fCQD3WHaaBAAAAAAAAAAAAAbBTpMAAAAAAAAAAAB0YmLCbtH0y06TAAAAAAAAAAAAwCAYmgQAAAAAAAAAAAAGwdAkAAAAAAAAAAAAMAiGJgEAAAAAAAAAAIBBMDQJAAAAAAAAAAAADIKhSQAAAAAAAAAAAGAQDE0CAAAAAAAAAAAAg2BoEgAAAAAAAAAAABgEQ5MAAAAAAAAAAADAIBiaBAAAAAAAAAAAAAZhsu8AAAAAAAAAAAAAhmFiYqLvBAbOTpMAAAAAAAAAAADAIBiaBAAAAAAAAAAAAAbB0CQAAAAAAAAAAAAwCIYmAQAAAAAAAAAAgEEwNAkAAAAAAAAAAAAMgqFJAAAAAAAAAAAAYBAMTQIAAAAAAAAAAACDYGgSAAAAAAAAAAAAGARDkwAAAAAAAAAAAMAgTPYdQPdWrlyZNx9zZL5XazbaaKMccdSb8oCFC/vOWqvWelf37UsvycknnZBTTz+z75T1am2dW+u98orL8rEPnJx/PP6UO887831vz3Y7Lsy+T3tWj2Vr19oa/+h7l+ecM9+bvzvmXbn6B9/LqW/5+2yz3Y5Jkicc8Mzsvtc+PReuWWvr3FpvorkLrfUmmrvQWm+iuQut9SbtNbfWm7TZnPh9b9w0j19rvYnmLrTWm7TX3FpvorkLrfWurqXXREly04035EV//ey87d2nZeFOD+o7Z51afFxoHr/WehPNXWitN2mvubXeRPO4/fT738nnzzo1LzzipHz8pKNz2803Jkl+9Ytrs+NDHpbnvOqNPReuWUtrzD3fxMRE3wkM3JzfabKUsqiUcn0p5YJSyhdLKV8ppTx7lrdxeCllj7ucN7+U8qPR52eUUp5y91X/5u3PNed/4bwsXbI0Z5718bzy1Yfmbccf13fSOrXWu8qHTn9/jjnyDVmyZGnfKTPS2jq31HvOJz6c0058U5YtnX4s3PKrm/LWf3hFvn7xhT2XrVtLa/yFT300Z7/nrXeu8U+vqln09Ofk7455V/7umHfN2YHJpK11TtrrTTR3obXeRHMXWutNNHehtd6kvebWepM2m/2+N36ax6+13kRzF1rrTdprbq030dyF1npXae010fLly3LCsUflXvPn950yIy0+LjSPX2u9ieYutNabtNfcWm+ieZz+4/98LJ8+5YQsH/1d8jmvemNeeMRJee6hx2T+JpvlwP/5sp4L166VNQbowpwfmhw5v9a6qNb6pCT7J/nfpZRHz/TKtdbjaq1fG19eW775ja/n8Xs9MUnyqF0fncsvv6znonVrrXeVHRcsyAknntx3xoy1ts4t9d5/ux3zqjf+052n77jj13nWX784e+1zYI9V69fSGt932x3ygr9/852nf/qDmu98/aK88w0vy1nvPjZ33P7rHuvWraV1TtrrTTR3obXeRHMXWutNNHehtd6kvebWepM2m/2+N36ax6+13kRzF1rrTdprbq030dyF1npXae010XtOOiF/+ufPydZbb9N3yoy0+LjQPH6t9Saau9Bab9Jec2u9ieZxus+22+e5hx79W+ef/88fzJ5PeWY23+q+PVTNTCtrDNCF5g7PXWu9rZRySpLnlFJenmRBku2SfCbJUUm+m2TXWuviUsprk6xIsmuSs5N8KclHk2yV5Mp13U8pZV6SU+5y+ycm+UKt9dGllD2TfC7JfZNsn+QDSZ61ptsvpTwyyTuTTCS5IckLkuyW5K1JliY5tdba2bErFi++LZtvvtmdp+dtMC/Lly/P5OTcfDi01rvKPvsdkGt+dnXfGTPW2jq31LvHE/fOL6695s7T99t2h9xv2x3yrf/8So9V69fSGj/6cYtyw/U/v/P0Ax7yB3ncvk/Lggfvks//y4fybx8/PX928Mt7LFy7ltY5aa830dyF1noTzV1orTfR3IXWepP2mlvrTdps9vve+Gkev9Z6E81daK03aa+5td5Ecxda612lpddEnzvn09lyq/tkj8c9IR8947S+c2akxceF5vFrrTfR3IXWepP2mlvrTTSP08P/6Em56fprf+O8226+KVdd9o08dQ7vMpm0s8YAXWhlp8m7ui7TQ4cX11oPSLJHkkNqrcuSfDLTw4tJ8twkH17teockuazW+scF+ygAAAAgAElEQVSZHohclwVruP0bktxQSlmQ5KlJfpLkMUmekeRT67j905K8rNa6KMm5Sf5+dP78WusTuxyYTJJNN90sixcvvvP0yqmVc/pJsLXeVrW2zq31tqjlNX7Unn+cBQ/eZfrzP/rjXP3D7/dctHatrXNrvYnmLrTWm2juQmu9ieYutNabtNfcWm/SZnNrWlxjzePXWm+iuQut9SbtNbfWm2juQmu9LTr3nE/lv756UV75koNz5fdq3nLE63PDL3/Zd9Y6tfi40Dx+rfUmmrvQWm/SXnNrvYnmrl1+8RfzqCfsmw02mNd3yjq1vMYAd7dWhyYXJvlKkseWUj6a6R0g7zX62vuTPL+UskeSOhp0XGXnJF/L9Be+mmTZOu7jxrXc/qeSHJjk8UmOS7Jfkj8Znb+22/+DJO8ppVyQ6V0mdxidX2f7H7877Lbb7vnShRcmSS695Ft56EN37iNjxlrrbVVr69xab4taXuP3Hf2a/Pj730mSfO/Sr2fBg0vPRWvX2jq31pto7kJrvYnmLrTWm2juQmu9SXvNrfUmbTa3psU11jx+rfUmmrvQWm/SXnNrvYnmLrTW26KTT/1Q3nnqGXnHKWfkITuXvP6ot+S+W2/dd9Y6tfi40Dx+rfUmmrvQWm/SXnNrvYnmrl317W9k50fv0XfGerW8xgB3t+ZGxkspWyR5UaaHIzettb6klPKQJC8upUzUWr9fSplIclj+f/buPMyysr739rfoAloGQcUZxAkfTkxU4lFxBBnEWaOJIZooxgmHOMRoHFBQQDFiUInKoIjinBiNHiOeo76EREFPFCeMj8PrxKSCoNKhu6nuev+oat5Ohx4K2Wv1j33f19VXd+3ae61PP6zatXfx67WSd2zw8G8nuW+Sf2qt7ZNk203s6rAkV2y4/SQfz8IluC9NcmaS/53kV733n7XWNrb9nuTJvfeftNbun4XLfSfJ2t9iKa6zAw46OOec84U8+UmHZn5+Pq895nVjZGyxar1VVVvnar0VVV7jP3rmX+Wj73xzls0uy8673iyHPvulm3/QSKqtc7XeRPMQqvUmmodQrTfRPIRqvUm95mq9Sc3maiqusebJq9abaB5Ctd6kXnO13kTzEKr1MoyKx4XmyavWm2geQrXepF5ztd5E89AuvfgnucktbzN2xmZVXmOA69vM/Pz82A2b1FrbP8lHsjDwuCYLg55vycIg4geS/CrJqixcTvvA3vuFrbUnJnltkr167/OttdOTfCjJWVm4XPdtknwnyQN7723x8/dL8uvF3fYkr9vE9r+U5D2997e31s5J8pHe+wmtteUb2f49k7xpsX0+ydMW73N47/3Qza3Byrls3f+RiptbU295Z5fNjJ1wg3f+Bb/e/J22Mnfd/cZjJyzZWf0XYycsyf7t5mMnAACwRN7zAQDUe0105aq5sROWbNcdNnWeDgBgbJ/41kVjJyzZo3936x/E3NDy2fjBFlvkxoe+t9abFAb16w89eeLPJVv90CSGJiet2g+LEv8DbQiGJodhaBIAgEnzng8AoN5rIkOTAMD1zdDkMAxNsqUMTbIpQwxNbjPpHQAAAAAAAAAAAABsDQxNAgAAAAAAAAAAAFNhduwAAAAAAAAAAAAApsPMjCu5My5nmgQAAAAAAAAAAACmgqFJAAAAAAAAAAAAYCoYmgQAAAAAAAAAAACmgqFJAAAAAAAAAAAAYCoYmgQAAAAAAAAAAACmgqFJAAAAAAAAAAAAYCoYmgQAAAAAAAAAAACmgqFJAAAAAAAAAAAAYCoYmgQAAAAAAAAAAACmwuzYAQAAAAAAAAAAAEyJmbEDmHbONAkAAAAAAAAAAABMBUOTAAAAAAAAAAAAwFQwNAkAAAAAAAAAAABMBUOTAAAAAAAAAAAAwFQwNAkAAAAAAAAAAABMBUOTAAAAAAAAAAAAwFQwNAkAAAAAAAAAAABMBUOTAAAAAAAAAAAAwFQwNAkAAAAAAAAAAABMBUOTAAAAAAAAAAAAwFSYHTsAxja3du3YCUs2u2zZ2Ak3eHfd/cZjJ0yF/dvNx05YkkeddO7YCUv2ycP3HTsBYCJWXr1m7IQlW76t13CT5rjg2swumxk7YSpU+/rztQfAtKn2mmjXHbYdO2EqzK2ZHzthSaodx3BDceXKubETlmyn5cYgJq3azwGS5H577jZ2ArCemRmv7RiXM00CAAAAAAAAAAAAU8HQJAAAAAAAAAAAADAVDE0CAAAAAAAAAAAAU8HQJAAAAAAAAAAAADAVDE0CAAAAAAAAAAAAU8HQJAAAAAAAAAAAADAVDE0CAAAAAAAAAAAAU8HQJAAAAAAAAAAAADAVDE0CAAAAAAAAAAAAU8HQJAAAAAAAAAAAADAVZscOAAAAAAAAAAAAYDrMzMyMncCUc6ZJAAAAAAAAAAAAYCoYmgQAAAAAAAAAAACmgqFJAAAAAAAAAAAAYCoYmgQAAAAAAAAAAACmgqFJAAAAAAAAAAAAYCoYmgQAAAAAAAAAAACmgqFJAAAAAAAAAAAAYCoYmgQAAAAAAAAAAACmgqFJAAAAAAAAAAAAYCrMjh0AAAAAAAAAAADAdJiZmRk7gSlnaHIKrV27NscefVS+23u22267HPmaY3K7PfccO2ujqvUmydzVV+foo47IxRddmKtXX52nPuNZedD+B4ydtUnV1rlab1KvuVpvUqd5m5nkRQ++Y/a4yY0yPz+ft5z1w/zol1clSQ5/wJ654PKr8r/O//nIldeuyhqvT/PkVetNNA+hWm/iNdwQqvUm9Y6LimusefKq9Sb1vvaSeutcrTfRPIRqvUm95mq9ieYhVOtNNA+hWu/6vvmNr+fENx+fU047Y+yUzaq2ztV6E81DqNabJGvWrMkbjjkyP/nxDzOTmbzkFUfmjnfea+ysjaq4xhWbK/4s4FlPfkJ23HHHJMmtbnPbvPRVx4xctGkVjwuASRn08tyttf1baz9vrZ213q+/3+A+h7fWjroO2z69tfaNDbZ9u+stfmktH2qt7T/GvrfE5z/32axetTpnfODDecGLXpw3vfG4sZM2qVpvknz6U5/MLrvsmlPe/b68+e0n5/jjtu4XR0m9da7Wm9Rrrtab1Gne9/Y3SZK88KPn591fuiBP3XeP7LJ8Nsc+au/c9w43Gblu06qs8fo0T1613kTzEKr1Jl7DDaFab1LvuKi4xponr1pvUu9rL6m3ztV6E81DqNab1Guu1ptoHkK13kTzEKr1rvOe096Zo486IqtWrR47ZYtUW+dqvYnmIVTrTZIvnH1WkuSk096fZzzn+Tn57W8ZN2gzKq5xxeZqPwtYvWpVkvn87Tvenb99x7u3+oHJpOZxATApY5xp8vO990MntO2X9t7PnNC2bzDO++pXcr8HPDBJcre73yPnn/+tkYs2rVpvkhz4kENywMGHLHwwnyxbtvWf1LXaOlfrTeo1V+tN6jR/8YeX59wfXZ4kueXO2+XK1Wtyo22X5YwvX5B77bnryHWbVmWN16d58qr1JpqHUK038RpuCNV6k3rHRcU11jx51XqTel97Sb11rtabaB5Ctd6kXnO13kTzEKr1JpqHUK13nd332CPHn3BiXvWKvx47ZYtUW+dqvYnmIVTrTZIHPfjA3O+B+yVJLrn4ouy0084jF21axTWu2FztZwE/+F7PypUr89LnPzNr1qzJ0579/PzO79597KxNqnhcAEzKVvFdprX2gCRvSXJ5krkk57bWbp/kg0l+muROSb7ce392a233JO9IsjzJrZMc0Xv/+Ca2/XtJ3ppkJsllSf48yT5J3pBkdZJTkrwkyb8kuVuS+SSPSXJlkpOT7LG4n0/03o9orZ2eZFWS2y/efljv/auttecmeXqSi5PcYnHf2yY5KcleWTir5xG997Naa8cmeXAW1v+jvfc3/BbLt2QrVlyZnXfe6ZqPl22zLHNzc5md3SoOh/+mWm+S7LDDwim4V6xYkZf91Qtz+HOfP3LR5lVb52q9Sb3mar1Jrea188lLDrpT7n/Hm+ToT38vl/xmVS75zaqtfmiy0hqvo3nyqvUmmodQrTfxGm4I1XqTesdFxTXWPHnVepN6X3tJvXWu1ptoHkK13qRec7XeRPMQqvUmmodQrXedAw8+JBddeMHYGVus2jpX6000D6Fa7zqzs7M5+tUvz9lnfS7HvOGEsXM2qeIaV2yu9rOA7ZcvzxOe+JQ8/DGPzwU//XFe/qJn5z0f/mSWbcVrXPG4AJiUQS/PveiADS6h/ZIsDEH+Se/9oCQ/XO++d0nytCT3TvLw1tqtkuyd5E2994OTPDPJc9e7/9+st91XLt52apLn9t73T/LPSV66ePvy3vsDe+9nJLlxkg/23vdLcmGSh2VhWPLc3vshi/s/fL39/Hjx9hOTPLO1dsskL0iybxYGLrdbvN/Tk1zae3/Q4u1vW7z9SUmemOSBSa5Y8gr+lnbccaesWLHimo/Xzq/dqr8JVutd52eXXJznPOOwPOyRj8ohD3/k2DmbVW2dq/Um9Zqr9Sb1mt/42R/kqe/7el704Dtm+ewY35KXrtoaJ5qHUK030TyEar3reA03WdV616l0XFRcY82TV613nUpfe0m9da7Wm2geQrXepF5ztd5E8xCq9Saah1Ctt6pq61ytN9E8hGq963vVa1+fD/3jp/KGY47MVVf959g5G1VxjSs2J7V+FrD77W6fgx76yMzMzGSP290+N95l11x22aVjZ21S1eMCYBLGmND4fO99//V+vTHJLXvv3138/BfWu+/3e++/6b2vycIZHJcv/v6s1toZWRhk3Ha9+790ve0eu3jb/0jy9tbaWVk4y+RtF2/vG3Sdt/j7Txf388sk92qtvT/JCUm238R975Tk/N77qt771Um+vPj538vCsOdZST6aZLa1tlsWhiaPS/KZJIOfUmyffX4//3b22UmSb3z9a9lrr7sMnbAk1XqT5LLLLs3zn/2MPO8Ff5lHP/bxY+dskWrrXK03qddcrTep03xQ2y2H3vM2SZJVV6/N2sxn7fz8yFVbpsoar0/z5FXrTTQPoVpv4jXcEKr1JvWOi4prrHnyqvUm9b72knrrXK030TyEar1JveZqvYnmIVTrTTQPoVpvVdXWuVpvonkI1XqT5MxPfSLvPe3UJMny5TfKNttsk21mtt6TS1Rc44rN1X4WcOYnP5aT3np8kuTSX/w8/7niytzsZruNXLVpFY8LgEnZWkbGL2yt/Y/e+38kuVcWLtOdLFwqe0NHJzm19/7p1tpTkxy2mW33JE/uvf+ktXb/LFxSO0nWbnC/Dfd1WJIreu/Paq3dOQtnlJzZyH2/l+SurbUbZeGS3/skeV+S7yS5oPf+usXPvTLJb5L8UZI/WXzst1trH+q9/3gzf4/rzQEHHZxzzvlCnvykQzM/P5/XHvO6oXZ9nVTrTZLT33lKfv3rX+W0U07KaaeclCQ54W0nZ/ny5SOXbVy1da7Wm9Rrrtab1Gn+tx/8Mn914J3ypj/4ncwum8k7/vXHWb2mxtBklTVen+bJq9abaB5Ctd7Ea7ghVOtN6h0XFddY8+RV603qfe0l9da5Wm+ieQjVepN6zdV6E81DqNabaB5Ctd6qqq1ztd5E8xCq9SbJfgcclNcddUSe8/QnZ25uLi948cuyvfd716uKzdV+FvCwRz8uf3P0K/OCZz45mZnJS1559FZ9ae6k5nEBMCkz8wOe2aq1tn+SjyT59gafOjLJ8Ul+nYWhwq8lOT3Jh3rv+y4+9twkhya5b5IjklyW5IIkd++937W1tu7+Z26wz3smeVMWBkTns3C579skObz3fujifX6UZO/e+8rW2nFZGHb8v0k+kORXSVZl4XLdByY5dt1+WmsPTXJo7/2w1tqfJ/mLJL/IwtknX53knCxcHnzPLFwC/O2991Nba69O8ogkVyX5epIX9t43+h9i5dy1Do9yPVl59ZqxE5Zs+bbLxk6AqfSok84dO2HJPnn4vmMnAEyE13BcG8cFjKfa15+vPQCAZK7IP+ReZ3bZzObvBFzvrlw5N3bCku20fOseXLshqPZzgCS5cmW95t123m7shCVbPhvfsNkiN/2zD9R6McqgfnnGEyf+XDLo0CTXjaHJyar4gs7/3IFxGJoE2Hp4Dce1cVzAeKp9/fnaAwAwNAlsGUOTXJtqPwdIDE0OxdAkW8rQJJsyxNDkNpPeAQAAAAAAAAAAAMDWwD+xAAAAAAAAAAAAYBAzM05KyricaRIAAAAAAAAAAACYCoYmAQAAAAAAAAAAgKlgaBIAAAAAAAAAAACYCoYmAQAAAAAAAAAAgKlgaBIAAAAAAAAAAACYCoYmAQAAAAAAAAAAgKlgaBIAAAAAAAAAAACYCoYmAQAAAAAAAAAAgKlgaBIAAAAAAAAAAACYCoYmAQAAAAAAAAAAgKkwO3YAAAAAAAAAAAAAU2Jm7ACmnTNNAgAAAAAAAAAAAFPB0CQAAAAAAAAAAAAwFQxNAgAAAAAAAAAAAFPB0CQAAAAAAAAAAAAwFQxNAgAAAAAAAAAAAFPB0CQAAAAAAAAAAAAwFQxNAgAAAAAAAAAAAFPB0CQAAAAAAAAAAAAwFQxNAgAAAAAAAAAAAFNhduwAAAAAAAAAAAAApsPMzMzYCUw5Q5NMveXbLhs7ga3Q3Jr5sROWbHaZFxWT9snD9x07YckOe/95Yycs2elP2mfsBKAAr+G4No4LGM/sNrUuZuI9H7ClPF9wQ3HJr1aOnbAkt9pl+dgJU8HzxeRd8Z9Xj52wJCuvXjN2wpJ5vpi8nZYbKRhCta+/ij+Hq9gMwOTU+ok2AAAAAAAAAAAAwHVkaBIAAAAAAAAAAACYCoYmAQAAAAAAAAAAgKlgaBIAAAAAAAAAAACYCoYmAQAAAAAAAAAAgKlgaBIAAAAAAAAAAACYCoYmAQAAAAAAAAAAgKlgaBIAAAAAAAAAAACYCoYmAQAAAAAAAAAAgKkwO3YAAAAAAAAAAAAA02FmZmbsBKacM00CAAAAAAAAAAAAU8HQJAAAAAAAAAAAADAVDE0CAAAAAAAAAAAAU8HQJAAAAAAAAAAAADAVDE0CAAAAAAAAAAAAU8HQJAAAAAAAAAAAADAVDE0CAAAAAAAAAAAAU8HQJAAAAAAAAAAAADAVDE0CAAAAAAAAAAAAU2F27AAAAAAAAAAAAACmw8zMzNgJTDlnmgQAAAAAAAAAAACmgqFJAAAAAAAAAAAAYCq4PPcUWrt2bY49+qh8t/dst912OfI1x+R2e+45dtZGVetNNA+hWu/6vvmNr+fENx+fU047Y+yUTaq4xponY9lMcvj998zNd9ous8tm8rFvXJJLr7w6T7/vHlmzdj4X/3pVTvniTzI/duhGVFjjDVVrrtabaB5Ctd5E8xCq9Sb1mqv1JpqHUK13fVXeP61TqbficaF58qr1JjWbE88Xk1atuVrvOh9677ty7r+elavnrs6jHveEPPRRjxs7aZOqrXO13kTzkC7/5WV5xp89IW9626nZ8/Z3HDtnszxfTFa13qRec7XeJJm7+uocfdQRufiiC3P16qvz1Gc8Kw/a/4Cxszap2jpX601qNgNMSrkzTbbW9m+t/by1dtZ6v/5+g/s8tLX2zM1s596ttdWttXtdx44ftdaWX5fHrreN5a21H/0227guPv+5z2b1qtU54wMfzgte9OK86Y3HDZ2wJNV6E81DqNa7zntOe2eOPuqIrFq1euyUzaq4xpon4wF3uml+s2ouR535vbz+//wgT733Hnn8PW6Vj379khx15vey7bKZ7LP7jcfO3KgKa7yhas3VehPNQ6jWm2geQrXepF5ztd5E8xCq9a5T6f1TUq+34nGhefKq9SY1mz1fTF615mq9SfL1r/7ffPubX8vfnvyeHP+20/KLn/1s7KTNqrbO1XoTzUOZm7s6x7/+Ndl++W/1vykH4/li8qr1JvWaq/Umyac/9cnsssuuOeXd78ub335yjj/umLGTNqvaOlfrTWo2A0xKuaHJRZ/vve+/3q8/Wv+Tvfcze++nbGYbz0jypiTPnVjlVuq8r34l93vAA5Mkd7v7PXL++d8auWjTqvUmmodQrXed3ffYI8efcOLYGVuk4hprnoxzf3RFPnLexUmSmZlkzfx8fnTZVdlp+2VJkuXbLsuatVvreSZrrPGGqjVX6000D6Fab6J5CNV6k3rN1XoTzUOo1rtOpfdPSb3eiseF5smr1pvUbPZ8MXnVmqv1JslXvvTF3P5Oe+U1L3tRXv3Sv8h97v+gsZM2q9o6V+tNNA/l7W8+Po953B9nt91uPnbKFvF8MXnVepN6zdV6k+TAhxySZz33+QsfzCfLlm39FyGtts7VepOazQCTsvV/Z9xCrbWzkvw8yU2TfDDJXr33l7XWjkjy2Cz8Xd/Rez+5tbZTkgOS3DXJN1tru/XeL22tHZXkkt77Sa21vZOc1Hvfv7X2yCRHJplJ8tUkhy/u9h2ttTss/vkPkqxJ8s4kuya5TZK39d7fsUHb45O8N8lNknx/ciuycStWXJmdd97pmo+XbbMsc3NzmZ3dOg+Har2J5iFU613nwIMPyUUXXjB2xhapuMaaJ2PV3NokyfLZbfKi/e6Qj5x3ceaT/Pl9ds8f3O1WuWr1mnz7kivHjdyECmu8oWrN1XoTzUOo1ptoHkK13qRec7XeRPMQqvWuU+n9U1Kvt+JxoXnyqvUmNZs9X0xeteZqvUnyqyuuyM8vuSivPf7vcslFF+aov35+3vnBf8rMzMzYaRtVbZ2r9Saah/DpT348u97kprn3fe+f959+6tg5W8TzxeRV603qNVfrTZIddtgxSbJixYq87K9emMPXDVBuxaqtc7XepGYzwKRUPdPkARtcnvsli7d/sPd+UBaGF9Na2yfJw5LcJ8m9k9yltTaT5NAk/9h7X5nkw0metrEdtdZmk/xdkkf03v9nFgYdd1/89Lt67/sn+VGSg5PcOcmHeu8PSfKQJH+53qbWtT0jybd67w9KcvJvuQ7XyY477pQVK1Zc8/Ha+bVb9TfBar2J5iFU662o4hprnpyb7bBtXnXIXvnX//eX+cIPL89T7nXbHHXm9/Lij/9Hzv7BL/On97rt2IkbVWWN11etuVpvonkI1XoTzUOo1pvUa67Wm2geQrVehlHxuNA8edV6k5rN1VRc42rN1XqT5Ma77JJ73ud+2XbbbbPHnrfPttttn19d/suxszap2jpX6000D+GfP/mx/PuXzskLnnVYvv/dntcd+YpcdumlY2dtkueLyavWm9Rrrta7zs8uuTjPecZhedgjH5VDHv7IsXM2q9o6V+tNajYDTErVockNL8/9xsXb+wb3a0m+3Htf03tf3Xt/ce99PsnTk9y3tXZmkgcleVZrbcO1WPfPm3ZLcnnv/edJ0nv/m977TxY/95XF3y9JskOSnyV5bGvtfUmOSLLtettb13aXJF9e3NaXklx9nVbgt7DPPr+ffzv77CTJN77+tey1112GTliSar2J5iFU662o4hprnoxdls/mFQffOR/8yoU56/sLP8y6cvWaXLV6TZLk8quuzk7bLRszcZMqrPGGqjVX6000D6Fab6J5CNV6k3rN1XoTzUOo1sswKh4XmievWm9Ss7maimtcrblab5Lc9e775N+/9MXMz8/nsl/8PCuvuio777Lr2FmbVG2dq/Ummodw4invyVtPOT1vOfn03PkuLa94zetys912GztrkzxfTF613qRec7XeJLnsskvz/Gc/I897wV/m0Y99/Ng5W6TaOlfrTWo2A0zKDW1kfO0GH38nybMXByKXJfnnJH+VZFnvfd91d2qt/Z8kj0yyMsmtF2/+/cXff55k19baTXvvv2ytvTXJ+xY/N7/B/l6c5JzFS3I/OMkjrqXt20num+SfFs+EuW0GdsBBB+ecc76QJz/p0MzPz+e1x7xu6IQlqdabaB5Ctd6KKq6x5sl47O/dMjtuvyyPu/ut8ri7L9x26hd/kufvd/usWTufubXzOfWLPx03chMqrPGGqjVX6000D6Fab6J5CNV6k3rN1XoTzUOo1sswKh4XmievWm9Ss7maimtcrblab5Lse//98q2vfTXPf9qTsnZ+bZ734pdn2bKt9x/VJvXWuVpvoplr5/li8qr1JvWaq/UmyenvPCW//vWvctopJ+W0U05KkpzwtpOzfPnykcs2rto6V+tNajYDTMrM/PyGc39bt9ba/kk+koXhw/XdKMlTeu/faa0dlmTv3vvLWmsvT/LoLJxV8x1ZGIb8fu/9rett8wlZOPvksxa3vSILZ5G8Z+99/9baw5K8OguX/T4vyfOT/HBxHytba8dlYUDzx0lOTHJZkiuS/G6S30nymSSHL7YtT/LeJLdZfMwDe+9tU3/nlXP/bTgTmLC5NfW+7GaXzWz+Tkydw95/3tgJS3b6k/YZOwEAgCWq+B6qGu/5YBwVn988X3BtLvnVyrETluRWu2y9Ax2wFFf85+AXnPutrLx6zdgJS+b5ghuKal9/y7fduoefGc/y2XhDwha5xZ9/pN4bbgbz89OeMPHnknJDk9PI0CQMzw/EuaEwNAkAwBAqvoeqxns+GEfF5zfPF1wbQ5MwDkOTk+f5ghuKal9/hibZGEOTbKlbPM3QJBv383dNfmhym0nvAAAAAAAAAAAAAGBrYGgSAAAAAAAAAAAAmAqGJgEAAAAAAAAAAICpYGgSAAAAAAAAAAAAmAqGJgEAAAAAAAAAAICpYGgSAAAAAAAAAAAAmAqGJgEAAAAAAAAAAICpYGgSAAAAAAAAAAAAmAqGJgEAAAAAAAAAAICpYGgSAAAAAAAAAAAAmAqzYwcAAAAAAAAAAAAwHWZmZsZOYMo50yQAAAAAAAAAAAAwFQxNAgAAAAAAAAAAAFPB0CQAAAAAAAAAAAAwFQxNAgAAAAAAAAAAAFPB0CQAAAAAAAAAAAAwFQxNAgAAAAAAACIewdoAACAASURBVAAAAFPB0CQAAAAAAAAAAAAwFQxNAgAAAAAAAAAAAFNhduwAAAAAAAAAAAAAgC3RWrtFkq8kOTjJXJLTk8wn+VaS5/be127q8YYmAQAAAAAAAAAAGMTMzMzYCRTWWts2yclJrlq86W+THNF7P6u1dlKSxyT52Ka24fLcAAAAAAAAAAAAQAXHJzkpyUWLH98zyb8s/vnTSQ7a3AacaRLgWswu868auGE4/Un7jJ2wZDfZ75VjJyzJ5f9y7NgJAACj8x4KuKHy/MYNxa12WT52AvzW5tbMj52wZLvusO3YCUtUrbfmceH1Bddm+bbLxk4AgBJaa4cl+UXv/TOttZcv3jzTe1/3wvA3SXbZ3HYMTQIAAAAAAAAAAABbuz9PMt9aOyjJPZK8N8kt1vv8zkmu2NxGXJ4bAAAAAAAAAAAA2Kr13h/Ue9+v975/kq8leXKST7fW9l+8y8OS/OvmtuNMkwAAAAAAAAAAAEBFL05yamttuyT/keQfNvcAQ5MAAAAAAAAAAABAGYtnm1xnv6U81uW5AQAAAAAAAAAAgKlgaBIAAAAAAAAAAACYCoYmAQAAAAAAAAAAgKlgaBIAAAAAAAAAAACYCrNjBwAAAAAAAAAAADAdZmZmxk5gyjnTJAAAAAAAAAAAADAVDE0CAAAAAAAAAAAAU8HQJAAAAAAAAAAAADAVDE0CAAAAAAAAAAAAU8HQJAAAAAAAAAAAADAVDE0CAAAAAAAAAAAAU8HQJAAAAAAAAAAAADAVDE0CAAAAAAAAAAAAU8HQJAAAAAAAAAAAADAVDE0CAAAAAAAAAAAAU2F27AAAAAAAAAAAAACmw8zMzNgJTDlnmgQAAAAAAAAAAACmgqFJAAAAAAAAAAAAYCq4PPcUWrt2bY49+qh8t/dst912OfI1x+R2e+45dtZGVetNNA+hWm9Sr7lab6J5CNV6b77rjvniac/JI1747izfftuc+JJHZ25ubb7308vy7OM+lvn5+bETr1W1da7Wm2geQrXeRPMQqvUm9Zqr9Saah1CtN9E8hGq9ieYhVOtN6jVX6000D6Fab6J5CNV6k5rNSfLNb3w9J775+Jxy2hljp2xWxTWu2Jw4LiatWnO13kTzEKr1JjWbASZl1DNNttZu31r7dWvtrPV+vfo6buse6x7bWrtkEvtvrb2stXbv1tpDW2vPXGLfYa214za4bcnbuT58/nOfzepVq3PGBz6cF7zoxXnTG4/b/INGVK030TyEar1JveZqvYnmIVTqnV22Tf7upY/NVavmkiSvfOoBed27/58c+JxTs/12y/Kw+7WRCzeu0jon9XoTzUOo1ptoHkK13qRec7XeRPMQqvUmmodQrTfRPIRqvUm95mq9ieYhVOtNNA+hWm9Ss/k9p70zRx91RFatWj12yhapuMYVmx0Xk1etuVpvonkI1XqTms0Ak7I1nGny2733/X/bjfTev5bka5Pcf+/9ev2O0Xs/8/rc3pY676tfyf0e8MAkyd3ufo+cf/63xsjYYtV6E81DqNab1Guu1ptoHkKl3uOe97Cc+vEv5SV/tl+S5Gvfuyg32flGSZKddtg+V8+tGTNvkyqtc1KvN9E8hGq9ieYhVOtN6jVX6000D6Fab6J5CNV6E81DqNab1Guu1ptoHkK13kTzEKr1JjWbd99jjxx/wol51Sv+euyULVJxjSs2Oy4mr1pztd5E8xCq9SY1mwEmZWsYmvwvWmv7Jzm8937o4seX9N5v1Vo7PcnVSfZMsn2SDyV5VJLbJXlMkj3Wf9ziY1+XZJckz0vyxCQvTLIqyfeSbPQMj5tp+FCSWyXZO8lRST6yuI8dkryy9/6/W2vPS/K4JDsmuTTJH6y37Zsn+XiSVy827917f9l1WqzraMWKK7Pzzjtd8/GybZZlbm4us7Nb3eGQpF5vonkI1XqTes3VehPNQ6jS+6cP3ye/uGJFPvvl718zNPmDn16WE178qLzssAfn11euzNnn/XDkyo2rss7rVOtNNA+hWm+ieQjVepN6zdV6E81DqNabaB5Ctd5E8xCq9Sb1mqv1JpqHUK030TyEar1JzeYDDz4kF114wdgZW6ziGldsdlxMXrXmar2J5iFU601qNgNMyqiX5170O+tfHjvJbTdx3x/13h+S5D+S3KH3/vAkH83C8OR/0Vo7Psls7/25SW6a5DVJDui9PyDJFUmedW37b61tav8bulOS3Rb3/ydJZltr2yS5WZKDeu/3ycJg6r0W73/LJJ9I8pe9988tYT/Xqx133CkrVqy45uO182u36m+C1XoTzUOo1pvUa67Wm2geQpXepzzinjnwXnfOZ058Wu62163zrlf9UU494vE56Dmn5h5PfHPef+Z5Oe55Dxs7c6OqrPM61XoTzUOo1ptoHkK13qRec7XeRPMQqvUmmodQrTfRPIRqvUm95mq9ieYhVOtNNA+hWm9Ss7maimtcsbmaimtcrblab6J5CNV6k5rNAJOyNQxNfrv3vv+6X0ku3ODzM+v9+auLv1+R5NuLf748yfINHnPLJHdLsm5E/o5Jzu+9/2bx47OT3PXa9t9733D/GzZco/d+fpKTk3wwyduTbNN7X5tkdZIPttbelWT3JNsuPuShWThL5qjrvs8+v59/O/vsJMk3vv617LXXXcbM2axqvYnmIVTrTeo1V+tNNA+hSu/Bz31nHvK8d+aQv3hXvvG9i/O0o/8+P7zo8vxmxaokycWX/uaaS3Vvjaqs8zrVehPNQ6jWm2geQrXepF5ztd5E8xCq9Saah1CtN9E8hGq9Sb3mar2J5iFU6000D6Fab1KzuZqKa1yxuZqKa1ytuVpvonkI1XqTms0Ak7I1joyvTHLrJGmt7ZmFs0SuM7+F2/hZkkOSnNVae2iSf8/CGSV37L2vSLJfku9ex4ZrtNZ+L8nOvfdHtNZuneSLrbWfJHls7/0+rbUdknwl///Q5XuSnJHkI621e2/h3+V6d8BBB+ecc76QJz/p0MzPz+e1x7xurJQtUq030TyEar1JveZqvYnmIVTrXd9zjvtY3vuaP87cmrVZPbcmzznuY2MnbVS1da7Wm2geQrXeRPMQqvUm9Zqr9Saah1CtN9E8hGq9ieYhVOtN6jVX6000D6Fab6J5CNV6k5rN1VRc44rN1VRc42rN1XoTzUOo1pvUbAaYlJn5+S2dQ7z+tdZun+RDvfd917ttNguX3L5VFi7Dfb/e+11aa6cv3vfM1tpxSb7Tez+9tfbCLJxp8twkh/feD22tXdJ7v1Vr7c5JzkxynywMUb4wydok30/y9MV9/Jf9b0nD4u17JzkqyfuS3CILZ488efFx/ysLZ5RMklVJ3pWFs03u3Xt/WWvt5UnukOSL627b1DqtnNviYVEAKO8m+71y7IQlufxfjh07AQAAAABu8ObW1PvfZbPLrvVidlyPHBcAW5fls9d+JVfY0G0O/8d638QZzEUnPW7izyWjDk1W01p7f5J39d4/P+R+DU0CME0MTQIAAAAAGzIcx7VxXABsXQxNsqUMTbIpQwxNbjPpHdxQLJ4d8neSnDd2CwAAAAAAAAAAALB0s2MHVNF7f32S14/dAQAAAAAAAAAAAFw3zjQJAAAAAAAAAAAATAVDkwAAAAAAAAAAAMBUMDQJAAAAAAAAAAAATAVDkwAAAAAAAAAAAMBUMDQJAAAAAAAAAAAATAVDkwAAAAAAAAAAAMBUMDQJAAAAAAAAAAAATIXZsQMAAAAAAAAAAACYDjMzM2MnMOWcaRIAAAAAAAAAAACYCoYmAQAAAAAAAAAAgKlgaBIAAAAAAAAAAACYCoYmAQAAAAAAAAAAgKlgaBIAAAAAAAAAAACYCoYmAQAAAAAAAAAAgKlgaBIAAAAAAAAAAACYCoYmAQAAAAAAAAAAgKlgaBIAAAAAAAAAAACYCrNjBwAAAAAAAAAAADAdZmZmxk5gyjnTJAAAAAAAAAAAADAVDE0CAAAAAAAAAAAAU8HQJAAAAAAAAAAAADAVZscOAABY3+X/cuzYCUtykz88ZeyEJbv8H545dgJbobk182MnLMnsspmxEwAAgClV7f1T4j0UNwyOYwCmzZUr58ZOWLLlOxlDAmpwpkkAAAAAAAAAAABgKhiaBAAAAAAAAAAAAKaCoUkAAAAAAAAAAABgKhiaBAAAAAAAAAAAAKaCoUkAAAAAAAAAAABgKhiaBAAAAAAAAAAAAKbC7NgBAAAAAAAAAAAATIeZmZmxE5hyzjQJAAAAAAAAAAAATAVDkwAAAAAAAAAAAMBUMDQJAAAAAAAAAAAATAVDkwAAAAAAAAAAAMBUMDQJAAAAAAAAAAAATAVDkwAAAAAAAAAAAMBUMDQJAAAAAAAAAAAATAVDkwAAAAAAAAAAAMBUMDQJAAAAAAAAAAAATAVDkwAAAAAAAAAAAMBUmB07AAAAAAAAAAAAgCkxM3YA086ZJgEAAAAAAAAAAICpYGgSAAAAAAAAAAAAmAqGJgEAAAAAAAAAAICpMDt2AMNbu3Ztjj36qHy392y33XY58jXH5HZ77jl21kZV6000D6Fab1KvuVpvonkI1XqTes0332V5vvimx+URR34q373wV0mSP37QnfLsh/9u9n/ZP41cd+2qrXGieUjf/MbXc+Kbj88pp50xdspmVVzjas3VepN6zdV6E81DqNabaB5Ctd5E8xCq9Sb1mqv1JpqH5P3TZFVrrtabaB5Ctd6kZnPiOXnSqjVX6000D6Fab5KsWbMmbzjmyPzkxz/MTGbyklccmTveea+xswBGcb2dabK1dsfW2j+01s5trX2+tfap1tpdt/Cxp7fWHno9tqxurZ21wa/bXl/bX28/Z7XW9r6+tztpn//cZ7N61eqc8YEP5wUvenHe9Mbjxk7apGq9ieYhVOtN6jVX6000D6Fab1KreXbZTP7u2Q/MVavmrrnt7ne4WZ5y0N6ZmRkxbDMqrfE6mofxntPemaOPOiKrVq0eO2WLVFzjas3VepN6zdV6E81DqNabaB5Ctd5E8xCq9Sb1mqv1JpqH4v3T5FVrrtabaB5Ctd6kZrPn5Mmr1lytN9E8hGq9SfKFs89Kkpx02vvzjOc8Pye//S3jBgGM6HoZmmyt7ZDkE0ne1Hvft/d+QJLXJHnb9bH96+CXvff9N/h14UgtW53zvvqV3O8BD0yS3O3u98j5539r5KJNq9abaB5Ctd6kXnO13kTzEKr1JrWajzts35x65n/k4sv/M0ly0523z2v+9F55ybu+OHLZplVa43U0D2P3PfbI8SecOHbGFqu4xtWaq/Um9Zqr9Saah1CtN9E8hGq9ieYhVOtN6jVX6000D8X7p8mr1lytN9E8hGq9Sc1mz8mTV625Wm+ieQjVepPkQQ8+MC995VFJkksuvig77bTzuEEAI7q+Ls/9qCSf772fs+6G3vuXW2sPbq2dnuRmi78eleQNSfZIcuskn+i9H7H4kGe11l6aZJckz158/OuT/M/Fx3699/7U1tpRSe6cZLfF29+W5PFJ7pLkKb33czcW2VrbPck7kixf3P8RvfePt9a+leS7SVYn+U6SOyS5RZI9k7yo9/6Z1tqxSR6chTX7aO/9DYubPbK1dsskOyb5kyQ/TnLy+n/HJCck+Vzv/R6ttX2TfHqx/TZJ3tV7P2QJa/1bW7Hiyuy8807XfLxsm2WZm5vL7OzWebX2ar2J5iFU603qNVfrTTQPoVpvUqf5Tw+4S37x65X57NcuyEv+8B5Zts02Oel5++Wv331urlo9t/kNjKjKGq9P8zAOPPiQXHThBWNnbLGKa1ytuVpvUq+5Wm+ieQjVehPNQ6jWm2geQrXepF5ztd5E81C8f5q8as3VehPNQ6jWm9Rs9pw8edWaq/UmmodQrXed2dnZHP3ql+fssz6XY95wwtg5AKO5vi7PfYck31/3QWvtn1prZ2VhAHH3LAxU3i/JzknOXRwSvHeSw9fbxlcWz1B5YpLDWms3TnJ57/3gLAxO7rveJbav6r0/NMlHkzy89/6oJMclOXTx8zfd4NLc71+8fe8snA3z4CTPTPLcxdt3SnJ0733d41f13h+W5AVJXrR425OSPDHJA5NcsV73pxa7P53kD7MwLPlf/o6998uSXNZa2yPJw5L8ZPHv9OgkH9uiFb4e7bjjTlmxYsU1H6+dX7tVf+Ou1ptoHkK13qRec7XeRPMQqvUmdZqfcmDLgXe/bT5zzCNztzvcLP/+lj/MXfe8ad56+ANyxosPzN573CRvfNp9x868VlXWeH2auTYV17hac7XepF5ztd5E8xCq9Saah1CtN9E8hGq9Sb3mar2JZq5dxTWu1lytN9E8hGq9Sc3maiqucbXmar2J5iFU613fq177+nzoHz+VNxxzZK666j/HzgEYxfU1NPnTLAxOJkl674/pve+f5PIkFyTpi5/6ZZJ7LQ4xnpBk+/W28ZXF3y9JskOSq5LcorX2wSycuXGnJNsu3ueri79fkeTbi3++PAtnkEz+++W5n7R4+8VZOKPlGfn/2LvzMDvLwu7jv0kmIUDYKiIKCrXAjSIg1K2IVZFFlKqtS3lrS3FhUayoVIsCguwIyCIgiyKIImhbbRGxrkgVXF4XEJTbfQERJaJIIGSb94+Z8KaWLKPM8+TO8/lc11xyzpw58537spMT+fU544PNJc+XpRqT5BtL/VxLnvOlGR9m/leS9ZfTvayf8SNJnpNkp4nn2S3Jc9PDaHKHHXbMF665Jklyw/XfzJZbbtV1wqS01pto7kJrvUl7za31Jpq70Fpv0k7zboddkd0P/1j2OPxjueFHc7LDP30o2xx4WfY4/GP5h1M/k5t/dmfe+J7rVvxEPWjljJemmQfS4hm31txab9Jec2u9ieYutNabaO5Ca72J5i601pu019xab6KZB9biGbfW3FpvorkLrfUmbTa3psUzbq25td5Ecxda602ST1z5n3nfhRckSWbNWjPTpk3LtJEHazYE0JYHa+b+H0kOLaU8ZcnbY5dStsj4VSbvTbJ44nH7JvlNrfWAic/vX0oZmfjc2O89555JHllr/dtSykOT/HWSZT12ZR2T5IJa61WllJdN9CyxeKl//h/PX0pZI8mLM/7220ny7VLKZcto2TcP/DN+NMkHktyR5BNJPpnkt7XW2//An+UPtsuuu+W6676YfV66d8bGxnL0scd3nTAprfUmmrvQWm/SXnNrvYnmLrTWm7TZ3JoWz1gzD6TFM26tubXepL3m1noTzV1orTfR3IXWehPNXWitN2mvubXeRDMPrMUzbq25td5Ecxda603abG5Ni2fcWnNrvYnmLrTWmyRP32XXHH/U4Xn1K/fJwoULc/Ahh2aNWbNW/IUwBUZGRlb8IJhCI2Njf+j+8H8qpWye8SsoPjzjY8xFGX+r7ecmuazW+olSyjZJLk3y2yT3ZfytrJ+V5LilHvPsjL/N9qFJrsj46HIsyZoZf6vs3ZL8otZ6binlwCQb11qPKqW8IMmza60HllLmJ7n29xLfnGTzJIcnmZPxK2BuX2vdppTy4yRb11rnlVKOWur5t05ybq31GaWUt078LPcmuT7J65J8LuNvv33zkpYkH36gn7HWemsp5ctJLq61nlNKuS7Jh2qtp63obOct/INHogDAFNvgRef3nTBpd/7r/n0nsApauKitl5yj0/1lGgAA6Edrf39K/B0KWH35nQyszu6et7DvhEnbcPaoX3KslEf903+294c4nfnpO5835b9LHrTRJFPHaBIAVl1Gk6wuWvsfmP2PywAAQF9a+/tT4u9QwOrL72RgdWY0yerMaJLl6WI0OW2qvwEAAAAAAAAAAADAqsBoEgAAAAAAAAAAABgEo0kAAAAAAAAAAABgEIwmAQAAAAAAAAAAgEEwmgQAAAAAAAAAAAAGwWgSAAAAAAAAAAAAGASjSQAAAAAAAAAAAGAQjCYBAAAAAAAAAACAQRjtOwAAAAAAAAAAAIBhGBkZ6TuBgXOlSQAAAAAAAAAAAGAQjCYBAAAAAAAAAACAQTCaBAAAAAAAAAAAAAbBaBIAAAAAAAAAAAAYBKNJAAAAAAAAAAAAYBCMJgEAAAAAAAAAAIBBMJoEAAAAAAAAAAAABsFoEgAAAAAAAAAAABgEo0kAAAAAAAAAAABgEEb7DgAAAAAAAAAAAGAYRkZG+k5g4FxpEgAAAAAAAAAAABgEo0kAAAAAAAAAAABgEIwmAQAAAAAAAAAAgEEwmgQAAAAAAAAAAAAGYbTvAAAeHN+//e6+EyZti4fN7jsB/mh3/uv+fSdM2nGf/l7fCZN22K5b9p0wKTfdclffCZO2zabr9p0Ag/Sbexb0nTBp6681o+8EAIBejU4f6TsBgAl+JwMra96CRX0nTNqsGdP7TgBYbbnSJAAAAAAAAAAAADAIRpMAAAAAAAAAAADAIBhNAgAAAAAAAAAAAINgNAkAAAAAAAAAAAAMgtEkAAAAAAAAAAAAMAijfQcAAAAAAAAAAAAwDCMjI30nMHCuNAkAAAAAAAAAAAAMgtEkAAAAAAAAAAAAMAhGkwAAAAAAAAAAAMAgGE0CAAAAAAAAAAAAg2A0CQAAAAAAAAAAAAyC0SQAAAAAAAAAAAAwCEaTAAAAAAAAAAAAwCAYTQIAAAAAAAAAAACDYDQJAAAAAAAAAAAADILRJAAAAAAAAAAAADAIo30HAAAAAAAAAAAAMBAjfQcwdK40CQAAAAAAAAAAAAyC0SQAAAAAAAAAAAAwCEaTAAAAAAAAAAAAwCAYTQIAAAAAAAAAAACDYDQJAAAAAAAAAAAADMJo3wF0b/HixTnumKPy3Vozc+bMHPm2Y/OozTbrO2uZWutNNHehtd6kveYF8+fnrLcfldtvuzVrrbV2XnnwoXnEpo/qO2u5WjvjpL3m1noTzVNl8aKF+eqlZ2Tur2/P4oUL8pjd987shz48X7vsrIxlLOs89BF5wt6vzbTp0/tOfUAtnPES37/5xnzwPe/MESefd/99l5z7jjx8082y614v7LFs+Vo64yU0T73WepM2m5Pkzl/PyX7/8JKcevYF2WzzR/eds1wtnnFrza31Jpq70FpvorkLrfUm7TW31pto7kJrvYnmLrTWm2juQmu9ieYutNabtNfcWm+iuQsLFyzIMUcdntt+fmsWzF+Ql+13QP7yGbv0nbVSvnXD9Xnn6afk/Asv6TsFoDdTfqXJUsozSimX/d59J5ZS9p3k8+xfSpnxQM83iefYopRyZSnlk6WUL5VSTiqlTOoMSimPLqXcXEq5uJTy5FLK90spJ5RSLiulzFzG12xeSvnSA9w/6XN4MHz2M5/O/Pvm55JLL8/Brz8kp558YtcJk9Jab6K5C631Ju01f+rKj2TWmmvlxLMvziv+6U1595kn9Z20Qq2dcdJec2u9ieap8pOvfi4z114nuxz89jztwKPzjX89N9/62Puy7V775FmvOzlJ8vMbv9xz5bK1cMZJcsWH3pcLTjs2C+bPT5Lc9Zs7c9Jhr83XvnRNz2Ur1soZL03z1GutN2mzeeHCBTnlhLdljVmz+k5ZKS2ecWvNrfUmmrvQWm+iuQut9SbtNbfWm2juQmu9ieYutNabaO5Ca72J5i601pu019xab6K5C1ddeUXWW2/9nP/e9+f0c87LKSce23fSSrn4wnfnmKMOz333ze87BaBXLb0991uS/LGXKjo+yTtrrbsn+YskWyV5/iSfY+ckV9Za/zHJHknOqLW+uda6d621iT9VvvH1r2WnnZ+WJNlu+8fnpptu7Llo+VrrTTR3obXepL3mW37yw+z4pJ2SJJs8avPc+tMf9Vy0Yq2dcdJec2u9ieapsukOO+dxz/n7iVtjGZk+LTu9/M156BaPy6KFCzLvrjszY821e21cnhbOOEke9vBN87q3vv3+2/Pm3ZMX/sP+2flZz+mxauW0csZL0zz1WutN2mw+5/RT8vy/+dtsuOFD+05ZKS2ecWvNrfUmmrvQWm+iuQut9SbtNbfWm2juQmu9ieYutNabaO5Ca72J5i601pu019xab6K5C8/afY8ccNBrx2+MJdOnt/FGr5s+8pE55bR39p0B0Ltef2uXUk5I8rSMjyHfUWv9cCnl6UmOzPigc3aSv5t4zMZJLktyepItSylXJdkoyRW11qNKKdsmOTPJSJI5SV6eZIckJyWZn+T8JLcn2beU8rskX0nykiQLSykjSd6Z5ElJZiY5stb6H6WUUzM+kkySS5N8JOPjzbVKKXcleVmS+aWUW5KclmTrJA+d+F5rJrk3yf6/9zO/MMnhSX418b1ufhCOclLmzr0766wz+/7b06dNz8KFCzM6umr+Id5ab6K5C631Ju01b75Fyf/90hfypJ2fme9958b8+o5fZdGiRZm+ir7VbtLeGSftNbfWm2ieKjPWWDNJsmDePbn2whPyuOf8Q6ZNm565v/5lPn/2YZmx5tpZf5M/7bly2Vo44yR50tN2ya9+8fP7b2+08SbZaONN8s2vXttj1cpp5YyXpnnqtdabtNd81RUfzfob/Eme9BdPzQcuuqDvnJXS2hkn7TW31pto7kJrvYnmLrTWm7TX3FpvorkLrfUmmrvQWm+iuQut9Saau9Bab9Jec2u9ieYurLXW+IUj5s6dm0P/+XU5cMmAchX3rN32yM9vvaXvDIDedXWlyV1KKVcv+cj4EHLNJH9aa905yTOTHFZKWT/JNkn+vtb6jCT/nuTFtdb3JPlFkr0nnm9WkhdkfEz5mon7Lkhy0MTXfTzJm5Y8ttb6tFrrJUn+OcmXkpyQ5JdJ3ptkvYnn2rDW+qSJlieUUvZK8qdJnpLx4eTfTTz2xCSX1lrfluSijI89P7LUz3pKkjMnOk6ZeHySpJQyI8k7kuya8atU3vOHHOYfa+21Z2fu3Ln33148tniVfaGRtNebaO5Ca71Je83P2vN5WWuttXP4wa/Il7/wuTx6y8es0oPJojJsrAAAIABJREFUpL0zTtprbq030TyV7rnzV7n6rLdksyc+M5s94RlJkrX/ZKM854gL8mdP3TPf/Mi7+w1cjlbOuGUtnrHmqddab9Je88ev+Ej+75evy8EH7Jvvf7fm+CPfkjl33NF31nK1dsZJe82t9Saau9Bab6K5C631Ju01t9abaO5Ca72J5i601pto7kJrvYnmLrTWm7TX3Fpvorkrt//itrx6v32z515/lT2es1ffOdCUkZERHz6W+dGFrkaTn621PmPJR8av2rhOkj+fGFF+IsmMJJsnuTXJmaWUizI+YJzxAM93Y631vlrrPUkWTtz3mCTnTDzfy5NsMnF/XerrnllrPb3W+pdJHpnk7iRHJClJrkuSWuudtdYjJp7vv2utY7XWBRkfWz52JX7WbZO8ZaLjrUkettTnHprk17XWObXWsSS9XCJohx12zBeuuSZJcsP138yWW27VR8ZKa6030dyF1nqT9pq/f/O3s+2OT8xxZ16Yv3j6rnnYIzZZ8Rf1rLUzTtprbq030TxV5t11Zz7/riOy3V/tm0c/ZfckyRcuODq/++WtSZLRNdbMyEhXLzUnr4Uzbl2LZ6x56rXWm7TX/M7zL86Z51+UM867KFtsVfKWtx2fh2y4Yd9Zy9XaGSftNbfWm2juQmu9ieYutNabtNfcWm+iuQut9Saau9Bab6K5C631Jpq70Fpv0l5za72J5i7MmXNHXvuq/fKag9+Q573ghX3nADBJfc7y5yX5XK11/1LKtIyPF3+Q5JNJ/qzW+rtSysUZf7vtJFmc/z/yHHuA56tJ9qm1/rSU8tQkD1/q65Z4eynl3lrr52utd5dSvptkwyTfSfLiJCmlrJfkQxl/u+6XJTlt4gqROyW5OMl2K/i5bk5ySq312lLK1kmevtTnfplk/VLKQ2utv0ryxCSdX/d4l113y3XXfTH7vHTvjI2N5ehjj+86YVJa6000d6G13qS95odv+sh88Jh35d8+cGHWnr1OXv3Pb+07aYVaO+OkvebWehPNU+U7n/pQFtxzd779ycvy7U9eliTZ9rn75CuXnp5p00czOnONPGHvVfetKFo449a1eMaap15rvUmbza1p8Yxba26tN9HchdZ6E81daK03aa+5td5Ecxda6000d6G13kRzF1rrTTR3obXepL3m1noTzV246N3n5667fpsLzz83F55/bpLktLPPy6xZs3ouA2BljIyNPdD+8MFTSnlGkgNrrXsvdd+JGR85bpvx4eDsJB+ptR5dSnlHkl2SzE1ye5I5tdb9JgaUmyV5W5IDljxfKeUXtdaNSyl/nuTUjA9Bx5K8Iskjlv7epZTHJDkzyQZJ5if5YZJXZfyKk2cm2WHi699Wa72qlHJKkqcmmZnkQ7XWk0op+ybZutZ6aCnlqCS/qLWeW0r5cZKtJ77nuzL+FuJrJjk4yW1JLqu1PqWU8twkxyT5dZIFSS6vtV60vDOct/ABR6IA/8P3b7+774RJ2+Jhs/tOgEE67tPf6zth0g7bdcu+Eyblplvu6jth0rbZdN2+E2CQfnPPgr4TJm39tR7oDSEAAAAAYNU1b8GivhMmbXTaqvuOXssye42O3leX5j36DR+3hWKZfviO50z575IpH03yxzOaBFaG0SSwsowmp57RJLCyjCYBAAAAYOoZTXbDaJKVZTTJ8nQxmmzvNywAAAAAAAAAAADAH8BoEgAAAAAAAAAAABgEo0kAAAAAAAAAAABgEIwmAQAAAAAAAAAAgEEwmgQAAAAAAAAAAAAGwWgSAAAAAAAAAAAAGITRvgMAAAAAAAAAAAAYhpGRkb4TGDhXmgQAAAAAAAAAAAAGwWgSAAAAAAAAAAAAGASjSQAAAAAAAAAAAGAQjCYBAAAAAAAAAACAQTCaBAAAAAAAAAAAAAbBaBIAAAAAAAAAAAAYBKNJAAAAAAAAAAAAYBCMJgEAAAAAAAAAAIBBMJoEAAAAAAAAAAAABsFoEgAAAAAAAAAAABiE0b4DAAAAAAAAAAAAGIaRkb4LGDpXmgQAAAAAAAAAAAAGwWgSAAAAAAAAAAAAGASjSQAAAAAAAAAAAGAQjCYBAAAAAAAAAACAQRjtOwCAB8cWD5vddwLQiMN23bLvhEnb+pCP9Z0wKTefulffCfCguP4nv+07YVK232y9vhMmbf21ZvSdAAAAAACrvVkzpvedAMAqxJUmAQAAAAAAAAAAgEEwmgQAAAAAAAAAAAAGwWgSAAAAAAAAAAAAGASjSQAAAAAAAAAAAGAQRvsOAAAAAAAAAAAAYBhGRkb6TmDgXGkSAAAAAAAAAAAAGASjSQAAAAAAAAAAAGAQjCYBAAAAAAAAAACAQTCaBAAAAAAAAAAAAAbBaBIAAAAAAAAAAAAYBKNJAAAAAAAAAAAAYBCMJgEAAAAAAAAAAIBBMJoEAAAAAAAAAAAABsFoEgAAAAAAAAAAABgEo0kAAAAAAAAAAABgEEb7DgAAAAAAAAAAAGAYRkb6LmDoXGkSAAAAAAAAAAAAGASjSQAAAAAAAAAAAGAQjCYBAAAAAAAAAACAQTCaBAAAAAAAAAAAAAbBaBIAAAAAAAAAAAAYBKNJAAAAAAAAAAAAYBBG+w6ge4sXL85xxxyV79aamTNn5si3HZtHbbZZ31nL1FpvorkLrfUm7TW31pto7kJrvYnmLrTSO20kOXHv7fLojWZnLMlhH/pWRqeN5LiXbJuFi8fyo1/enX+57IaMjfVd+sBaOeclWutNNE+1H9x8Yy5/71l5y0nn5ic/+G4uOffkTJs2PaMzZuSAQ47Kehs8pO/EB9TSGSft9Saau9Bab6K5C631Jpq70Fpv0l5za72J5i601pto7kJrvYnmLrTWm2juQmu9SXvNrfUmmrvQWm/SZjPAVFktrjRZSrm6lLL1JB6/eSnlS1PU8tellEdM5ff4Y332M5/O/Pvm55JLL8/Brz8kp558Yt9Jy9Vab6K5C631Ju01t9abaO5Ca72J5i600rvr4x6WJHnRGdfm1Ctr3vjckoOfvVXO+K/v5cVnXJuZo9Ozy2M36rly2Vo55yVa6000T6UrP/y+vOeM47Jg/vwkyfvPOzX/cOAb85aTzs0TdnpmPvbh9/VcuGytnPESrfUmmrvQWm+iuQut9Saau9Bab9Jec2u9ieYutNabaO5Ca72J5i601pto7kJrvUl7za31Jpq70Fpv0mYzwFRZLUaTq5iDk6zbd8TyfOPrX8tOOz8tSbLd9o/PTTfd2HPR8rXWm2juQmu9SXvNrfUmmrvQWm+iuQut9H7yW7fnzZd/K0myyQZr5q57F+SmW36b9deakSRZe43pWbhoFb3MZNo55yVa6000T6WNHr5pXnv4SfffPujQ47LZn22VJFm8aFFmzlyjr7QVauWMl2itN9HchdZ6E81daK030dyF1nqT9ppb6000d6G13kRzF1rrTTR3obXeRHMXWutN2mturTfR3IXWepM2mwGmyur09twbllKuSDIrycOTHF5r/Wgp5YYkn0+yXZKxJM9f1hOUUnZI8s4ki5LMS7JfxoelH05yW5JNk1xVaz2slPK4JO9IMj3JhklelWSDJI9P8r4kfz/xnNOTXJTkplrriaWUE5I8IclDklxfa33Zg3kIK2Pu3Luzzjqz7789fdr0LFy4MKOjq+Z/HVrrTTR3obXepL3m1noTzV1orTfR3IWWehctHsupL90+u2+3cV594deywdozc/SLHpd/2n3L/O7eBfnS9+f0nbhMLZ1z0l5vonkqPXHnXfKr239+/+31/2TDJMn3vn1DPvWxD+ewt5/XV9oKtXLGS7TWm2juQmu9ieYutNabaO5Ca71Je82t9Saau9Bab6K5C631Jpq70FpvorkLrfUm7TW31pto7kJrvUmbzay+RkZG+k5g4FanK00+Psmptdbdkuyf5KCJ+9dN8sFa69OT3Jpkz+U8xwVJXjPx2HMyPopMks2T7JvkiUl2KaXsmGSbJIfUWp+V5KQkL6u1Xpnkm0n2STI/46PUDyS5bmIwuW6SOycan5DkKaWUTR6MH34y1l57dubOnXv/7cVji1fpPwRb6000d6G13qS95tZ6E81daK030dyF1noP+cD12eXYq3PC3tvlyBdukxefeW2edfzV+bev3pLDXvDYvvOWqbVzbq030dy1L33+U7norBNzyFGnZd31Nug7Z5laO+PWehPNXWitN9HchdZ6E81daK03aa+5td5Ecxda6000d6G13kRzF1rrTTR3obXepL3m1noTzV1orTdpsxlgqjQ7miylzC6lzFjqrv9OckAp5ZIkByZZ+nPfmPjPn2X8SpTL8oha6zcn/vmajA8jk/ErQv661rooyZeTlIwPMI8opVyc5EW/9/2W2D7JRkmWTPXvTbJRKeWDSc6buP+Bvm5K7bDDjvnCNdckSW64/pvZcsutuk6YlNZ6E81daK03aa+5td5Ecxda6000d6GV3r9+wiZ59a5/liS5d/6ijI0lv5m7IHfPW5gk+eVd92W9tTp/abbSWjnnJVrrTTR36YufvSqf/tiH8uaT3pWNHt75/x/ZpLR2xq31Jpq70FpvorkLrfUmmrvQWm/SXnNrvYnmLrTWm2juQmu9ieYutNabaO5Ca71Je82t9Saau9Bab9JmM8BUaXkyfnGSs0opn8/4MPG0JCfXWq8qpbws41eGXGJsJZ/z56WU7WqtNyR5epLvTtz/mFLKWknuS/LkJO+d+HhprfU7pZS3ZfxqlEmyOP9/jPq1JM9N8pVSyicmHvPIWuvfllIemuSvk3R+vdlddt0t1133xezz0r0zNjaWo489vuuESWmtN9HchdZ6k/aaW+tNNHehtd5Ecxda6f3EDb/IyX+3fS7/p7/IjOnTcvS/35Q775mfd/7jjlm0eCzzFy3Omy+7oe/MZWrlnJdorTfR3JXFixbl/eeemods9LCceey/JEm23nbH/M3f799z2QNr7Yxb6000d6G13kRzF1rrTTR3obXepL3m1noTzV1orTfR3IXWehPNXWitN9HchdZ6k/aaW+tNNHehtd6kzWaAqTIyNraye8JVSyllpyRnTtz8eJLvJDk8yZwktyTZvta6TSnlx0m2rrXOK6WcmOTmJFcnuXHin5c4JMldSc7I+JBxYZJXZHwE+bkk307ysCT/OvFW229I8vIkd058vw1rrbuVUo5N8uyMv0X4ObXWp5RSdk5yVpLnJ/nXjF9xcizJmkleX2v94vJ+1nkLV3r0CQCwWtr6kI/1nTApN5+6V98J8KC4/ie/7TthUrbfbL2+EwAAAAAABmvWaPcXDqNN5V/+yxaKZaon7THlv0uaHU12pZSyeZLLaq1P6avBaBIAGDqjSeiH0SQAAAAAACvLaJKVZTTJ8nQxmpy24ocAAAAAAAAAAAAAtG+074BVXa31x0l6u8okAAAAAAAAAAAA8OBwpUkAAAAAAAAAAABgEIwmAQAAAAAAAAAAgEEwmgQAAAAAAAAAAAAGYbTvAAAAAAAAAAAAAIZhZKTvAobOlSYBAAAAAAAAAACAQTCaBAAAAAAAAAAAAAbBaBIAAAAAAAAAAAAYBKNJAAAAAAAAAAAAYBCMJgEAAAAAAAAAAIBBMJoEAAAAAAAAAAAABsFoEgAAAAAAAAAAABgEo0kAAAAAAAAAAABgEIwmAQAAAAAAAAAAgEEwmgQAAAAAAAAAAAAGYbTvAAAAAAAAAAAAAIZh2rSRvhMYOFeaBAAAAAAAAAAAAAbBaBIAAAAAAAAAAAAYBKNJAAAAAAAAAAAAYBCMJgEAAAAAAAAAAIBBMJoEAAAAAAAAAAAABsFoEgAAAAAAAAAAABiE0b4DgMm7e97CvhMmbfYsv26A1dO8BYv6Tpi0WTOm950waTefulffCZNy7ffn9J0waTtt8ZC+E1gFbb/Zen0nAADAKueO383vO2HSNlxnZt8JAEw4+4s/7DthUg566qP7ToDBavF156YbeN0JtMGVJgEAAAAAAAAAAIBBMJoEAAAAAAAAAAAABsH75QIAAAAAAAAAANCJkZG+Cxg6V5oEAAAAAAAAAAAABsFoEgAAAAAAAAAAABgEo0kAAAAAAAAAAABgEIwmAQAAAAAAAAAAgEEwmgQAAAAAAAAAAAAGwWgSAAAAAAAAAAAAGASjSQAAAAAAAAAAAGAQjCYBAAAAAAAAAACAQTCaBAAAAAAAAAAAAAbBaBIAAAAAAAAAAAAYhNG+AwAAAAAAAAAAABiGkZGRvhMYOFeaBAAAAAAAAAAAAAbBaBIAAAAAAAAAAAAYBKNJAAAAAAAAAAAAYBCMJgEAAAAAAAAAAIBBMJoEAAAAAAAAAAAABsFoEgAAAAAAAAAAABgEo0kAAAAAAAAAAABgEEb7DqB7ixcvznHHHJXv1pqZM2fmyLcdm0dttlnfWcvUWm/SZvOiRYty0rFH5qc/+VFGMpI3vuXIPHqLLfvOWqYWz7i15tZ6E81daK03abN54YIFOeaow3Pbz2/NgvkL8rL9DshfPmOXvrOWqcUzbqn5R/WmfOR95+QNx519/31f+fwnc/WVH86b3n5Bj2XL19IZL6F56rXWm7TX3FpvorkLrfUmmrvQWm+iuQut9SbtNbfWm2juygH7vCRrr712kmTjR2ySNx1xbM9Fy9fiGbfW3FpvorkLrfUmmqfK4oULc/XFp+XuObdn0cIF2eE5/yebP/4pSZJrLz8v62+8aR779Of2XLlsLZzx0lrrTTR3obXeJVp73QkwVXq/0mQp5epSytYPwvNcVEr5TSlljaXu27GUMlZKecYf+/wP8P22LaX85YP9vF347Gc+nfn3zc8ll16eg19/SE49+cS+k5artd6kzeYvXnN1kuTcCz+Q/V792px3zhn9Bq1Ai2fcWnNrvYnmLrTWm7TZfNWVV2S99dbP+e99f04/57yccuKq/RfWFs+4leZP/vv78/6zT8iC+fPvv+9nP6y59tNXZGysx7CV0MoZL03z1GutN2mvubXeRHMXWutNNHehtd5Ecxda603aa26tN9Hchfn33ZdkLO9413vzjne9t4l/cd3aGSftNbfWm2juQmu9ieap8r0vfzazZq+T573plOx58LH54gfPyb2/+02uOuOI/OT6L/Wdt0ItnPHSWutNNHehtd6kzdedAFOl99Hkg+y2JHsudfulSX44Rd/rhUkeO0XPPaW+8fWvZaedn5Yk2W77x+emm27suWj5WutN2mz+y2c+K2867KgkyS9u+3lmz16n36AVaPGMW2turTfR3IXWepM2m5+1+x454KDXjt8YS6ZPX7UvDt7iGbfSvOHGm+SAQ0+4//bdd/02H73kvLz4Fa/rsWrltHLGS9M89VrrTdprbq030dyF1noTzV1orTfR3IXWepP2mlvrTTR34Qffq5k3b17e9Nr9c8hBr8i3b7y+76QVau2Mk/aaW+tNNHehtd5E81R59J8/LU94/j7jN8bGMm369Cy4b17+/K9emi2f8qx+41ZCC2e8tNZ6E81daK03afN1J6uvkREfPpb90YVV5d/Ab1hKuSLJrCQPT3J4rfWjpZQbknw+yXZJxpI8v9b621LKCUmelmR6knfUWj888TwfTPJ/kny0lDItyY5JvpokpZR1k7w7yfpJHpHk7Frru0opT05yesYHpLdmfGj5pCRHTtw3O8nfJZmf5Iokc5J8Lsm+SeaXUr4+8bwr7EzyySRfT7JVrXVRKeWkJF+rtX7oQT3NFZg79+6ss87s+29PnzY9CxcuzOjoqvJfh/+ptd6kzeYkGR0dzTFvfXOuufozOfak0/rOWa4Wz7i15tZ6E81daK03abN5rbXG3xZh7ty5OfSfX5cDlwwoV1EtnnErzTvu9MzMuf22JMniRYvy/rOOz4te/trMnLnGCr6yf62c8dI0T73WepP2mlvrTTR3obXeRHMXWutNNHehtd6kvebWehPNXVhj1qy85O/+Mc95/gtzy89+kje//lW5+PIrMn0V7U3aO+OkvebWehPNXWitN9E8VWbMWjNJMn/ePfnUucflic/fJ+tuuHHW3XDj/PTG/9tz3Yq1cMZLa6030dyF1nqTNl93AkyVVeVKk49Pcmqtdbck+yc5aOL+dZN8sNb69IwPGvcspeyZ5E9rrTsneWaSw0op6088/itJti6lrJ1kl4yPG5fYIslltdbdk+ye5A0T95+X5OW11icnuTLJY5Jsk+Tva63PSPLvSV488diNk+xea31bkosyPtj8ysp2JhlJ8oUke5RSpmf8qpgf/eOObvLWXnt25s6de//txWOLV+k/uFvrTdpsXuKIo0/IZf9+ZU469sjce+89fecsU4tn3Fpza72J5i601pu02Zwkt//itrx6v32z515/lT2es1ffOcvV4hm32PzTH9T88ue35IPnnpx3n/LW/OJnP8qH3n1631nL1OIZa556rfUm7TW31pto7kJrvYnmLrTWm2juQmu9SXvNrfUmmruw6aM2z67P3isjIyN55KM2z7rrrZ85c+7oO2u5WjvjpL3m1noTzV1orTfRPJXu/vWv8rFTD82WT9klWzz5mX3nTEorZ7xEa72J5i601pu0+boTYKr0MpospcwupcxY6q7/TnJAKeWSJAcmWfpz35j4z59l/EqU2yb581LK1Uk+MfHYzZd6/H8keX7Grw75/qXuvz3JC0op709y+FLfY+Na63eSpNb6nlrr1zM+fDyzlHJRxgePSx77o1rr/GX8WCvbeUHGr1K5Z5JPL+f5pswOO+yYL1xzTZLkhuu/mS233KrrhElprTdps/kTV/5n3nfhBUmSWbPWzLRp0zJtZFXZVf9vLZ5xa82t9Saau9Bab9Jm85w5d+S1r9ovrzn4DXneC17Yd84KtXjGLTZvvtVj89azPpA3HHd2XvnPR2fjR/5pXvLKVfdtuls8Y81Tr7XepL3m1noTzV1orTfR3IXWehPNXWitN2mvubXeRHMXPnHFR3LumackSe741S9zz9y785CHbNhz1fK1dsZJe82t9Saau9Bab6J5qtxz1535+OmH5cl/87JsvfMefedMWgtnvLTWehPNXWitN2nzdSfAVOlr5n5xkrNKKZ9PslGS05KcXGu9qpTysoyPCpcY+72vvTnJ52qt+0+8BfcRSX6w1OcvzfjbbY/VWn9YSlly/yFJrpt4S+5nJnnuxP0/L6VsWWv9XinlX5J8N+NXn/yzWuvvSikXZ/wKkUmyeKnvszj/c3S6Up0Tz3lGkldkfLzZuV123S3XXffF7PPSvTM2Npajjz2+j4yV1lpv0mbz03fZNccfdXhe/cp9snDhwhx8yKFZY9asvrOWqcUzbq25td5Ecxda603abL7o3efnrrt+mwvPPzcXnn9ukuS0s8/LrFX093KLZ9xic2taPGPNU6+13qS95tZ6E81daK030dyF1noTzV1orTdpr7m13kRzF/Z83t/k7cccloP33ycZGckbDztmlX+LxNbOOGmvubXeRHMXWutNNE+Vb3788tx3z935+pUfzNev/GCSZM/XHpPRmWv0XLZyWjjjpbXWm2juQmu9SZuvOwGmysjY2O9v/aZeKWWnJGdO3Px4ku9kfEA4J8ktSbavtW5TSvlxkq1rrfNKKSdmfIh4cZJTkzwxyewkH6m1Hj1xVcjLaq2fKKV8Lcl7aq3nlFIuS3JuxoeP75z4Hr9J8rgkj02yXZJ3ZHwEeVuSf0xyQsbf3ntuxq9QOSfJcRPP/5SJn+G5SU7O+FuJv3dlOye+9g1JXlxr/YuVOa95C//XIJOBu3vewr4TJm32LC+2gNXTvAWL+k6YtFkzpvedsNq79vtz+k6YtJ22eEjfCQAAAE2443edv4HUH23DdWb2nQDAhLO/+MO+EybloKc+uu8EGKwWX3duusHMkRU/CpJtj/iULRTL9K1jdpvy3yW9jCaHrpTyxiRzaq0XrszjjSb5fUaTAKsOo0keiNEkAADA6qvFf3ltNAmw6jCaBFZWi687jSZZWUaTLE8Xo0krpo5NXBHzEUn+qucUAAAAAAAAAAAAGBSjyY7VWvftuwEAAAAAAAAAAACGaFrfAQAAAAAAAAAAAABdMJoEAAAAAAAAAAAABsHbcwMAAAAAAAAAANCJkZGRvhMYOFeaBAAAAAAAAAAAAAbBaBIAAAAAAAAAAAAYBKNJAAAAAAAAAAAAYBCMJgEAAAAAAAAAAIBBMJoEAAAAAAAAAAAABsFoEgAAAAAAAAAAABgEo0kAAAAAAAAAAABgEIwmAQAAAAAAAAAAgEEwmgQAAAAAAAAAAAAGwWgSAAAAAAAAAAAAGITRvgMAAAAAAAAAAAAYhpGRkb4TGDhXmgQAAAAAAAAAAAAGwWgSAAAAAAAAAAAAGASjSQAAAAAAAAAAAGAQjCYBAAAAAAAAAACAQTCaBAAAAAAAAAAAAAbBaBIAAAAAAAAAAAAYBKNJAAAAAAAAAAAAYBBG+w4AJm/2LP+nC7CqmDVjet8JrIJ22uIhfSdM2gZPfE3fCZNy51fP6jsBAAAYqA3Xmdl3AgANO+ipj+47AWiE150AU8eVJgEAAAAAAAAAAIBBcLk6AAAAAAAAAAAAOjEy0ncBQ+dKkwAAAAAAAAAAAMAgGE0CAAAAAAAAAAAAg2A0CQAAAAAAAAAAAAyC0SQAAAAAAAAAAAAwCEaTAAAAAAAAAAAAwCAYTQIAAAAAAAAAAACDYDQJAAAAAAAAAAAADILRJAAAAAAAAAAAADAIRpMAAAAAAAAAAADAIBhNAgAAAAAAAAAAAIMw2ncAAAAAAAAAAAAAwzAyMtJ3AgPnSpMAAAAAAAAAAADAIBhNAgAAAAAAAAAAAINgNAkAAAAAAAAAAAAMgtEkAAAAAAAAAAAAMAhGkwAAAAAAAAAAAMAgGE0CAAAAAAAAAAAAg2A0CQAAAAAAAAAAAAyC0SQAAAAAAAAAAAAwCEaTAAAAAAAAAACs3KgvAAAgAElEQVQAwCCM9h1A9xYvXpzjjjkq3601M2fOzJFvOzaP2myzvrOWqbXeRHMXWutN2mturTfR3IXWehPNXWitN9E81a699F/yu7nzkiQ/vnVO3vNvX8gpb3xRFi5alE9fd3OOP/+qngsfWEtnvERrza31Ju01t9abaO5Ca72J5i601pto7kJrvUl7za31Jpq70FpvorkLrfUmmrvQWm+iuQut9SbtNbfWm2juQmu9SZvNAFNltbjSZCnl6lLKgaWU503ia15TSrmulHLNxMcRE/dvXEo5Z+Kff1xKmbWMr7+olPLsFXyP00spj1rG52aVUl65sr0Pps9+5tOZf9/8XHLp5Tn49Yfk1JNP7CNjpbXWm2juQmu9SXvNrfUmmrvQWm+iuQut9Saap9IaM0czMpLssd8Z2WO/M3LAUe/PmYftnX98y0XZ5WWn5Ynbbp7ty6Z9Zz6gVs54aa01t9abtNfcWm+iuQut9Saau9Bab6K5C631Ju01t9abaO5Ca72J5i601pto7kJrvYnmLrTWm7TX3FpvorkLrfUmbTaz+hoZ8eFj2R9dWJ2uNHl1rfXmlXlgKeVVSXZK8sxa67xSyowkHyil7F5r/WSSVz8YQbXW1y3n0xsneWWSdz8Y32syvvH1r2WnnZ+WJNlu+8fnpptu7DphUlrrTTR3obXepL3m1noTzV1orTfR3IXWehPNU2m7rTbJWrNm5opzDsro9Gk59ryPZ40Zo/nRLXckST597Xeyy5NLrq+39Fz6v7Vyxktrrbm13qS95tZ6E81daK030dyF1noTzV1orTdpr7m13kRzF1rrTTR3obXeRHMXWutNNHehtd6kvebWehPNXWitN2mzGWCqrE6jyaNK+X/s3XmcZHV97/93zzQwMCCIqLiwuOAXjaKYG2NAlKCCGDUuMRKNBm+uAnGPxmCCggKKOiqKCxJFiAlilhu3KP5EAkQFNYIgRI9LolEQF5RtcPa+f3QP6Z+ZVahz5sN5Ph8PH3RVV1W/+mPZfaj+eKqdn+QbSV6VZHmS3ZKcmuSgJA9O8vau696T5AVJDuy6blmSdF23srX2jK7rZlpreyY5u+u6h6994NbaXpldbtw6yc1JDpv71BGttVcm2THJUUl+nOTjSa5N8skkj09yZJI7JXlLkpVz9/+9JH+Z5AGttdd0Xfe6iUxkPZYuvSk77LD9LZcXLliYVatWZXp6y3w6VOtNNPehWm9Sr7lab6K5D9V6E819qNabaJ6km5etzMl//dl84J++kPvufpd89J1H5bobf3HL52+8eVnudY9dBixcvyoznq9ac7XepF5ztd5Ecx+q9Saa+1CtN9Hch2q9Sb3mar2J5j5U600096Fab6K5D9V6E819qNab1Guu1pto7kO13qRmM8Ck3C7ennsd7pnkaZldZDwmybOTHJrkiLnP79x13U+TpLX2lLlly4tba0vW83hLkryh67rfSvL2JPvOXf+VrusOSnJKksPnrts1ycFd171p3v2fnOTvkjwqyXuS3DHJiUn+ve+FySRZvHj7LF269JbLa2bWbNG/BKv1Jpr7UK03qddcrTfR3IdqvYnmPlTrTTRP0re+9+N86JNfTpJ8+79+nOtvWpadd9zuls/vsN2iXD9viXJLUmXG81Vrrtab1Guu1pto7kO13kRzH6r1Jpr7UK03qddcrTfR3IdqvYnmPlTrTTT3oVpvorkP1XqTes3VehPNfajWm9RsBpiUskuTrbXt595We62ZeR9f0XXdyiTXJflO13Urkvw8yaK5z9/YWts5Sbqu+6eu6w5McmyS9Z3ypiW5aO72H5t7C+8k+crcP69JsvYvwf859/Xme32Suyf5bGbPMrlyk7/RCdh334fmcxdemCS5/LKvZq+97jdkzkZV600096Fab1KvuVpvorkP1XoTzX2o1ptonqQ/evLDc9KfPiVJcrc775jtFm2Vpb9YkXvdc/ZQ+zH73T+fv/TbQyauV5UZz1etuVpvUq+5Wm+iuQ/VehPNfajWm2juQ7XepF5ztd5Ecx+q9Saa+1CtN9Hch2q9ieY+VOtN6jVX600096Fab1KzGWBSKq+Mn5nkna21C5LcJcnl8z43s+673OJdSU5urT2v67rlrbWFSQ7YwP2+nuQ3kpzbWntWkp038HXWrOO6P0xyRtd1r2itvSrJ85N8IAMtrR70mMfmoos+n+c867DMzMzkdSe8foiMTVatN9Hch2q9Sb3mar2J5j5U600096Fab6J5ks74p4vyV697dj57+ssyMzOTI1/7t1mzZiZnnPhHWbhwQc696Bv58hXfGzpznarMeL5qzdV6k3rN1XoTzX2o1pto7kO13kRzH6r1JvWaq/UmmvtQrTfR3IdqvYnmPlTrTTT3oVpvUq+5Wm+iuQ/VepOazQCTMjUzs7H9wi1Ta22/JO+Yu/jJzC4rXpPkG0mO7LrusNba3klO7bruwNbaTkku7rpu77n7H5nkmUlWJ9kxs2eSPD6zZ6M8u+u6h7fWvptk78y+3fd7kyxMcnNmlyDfOne7c1prj0tyWJLj1t537mucn+TIucd/e5Klc53PT/LDJBcn+XTXdX++oe912aqNLoECAMCtcsffeOHQCZvl519+59AJAAAAAADAPIumMzV0AzU89HXn2YVivS55zUET/1lSdmlyTCxNAgAwaZYmAQAAAACAW8PSJJvK0iQb0sfS5CBvDw0AAAAAAAAAAADQN0uTAAAAAAAAAAAAwChMDx0AAAAAAAAAAADAOExNeSd3huVMkwAAAAAAAAAAAMAoWJoEAAAAAAAAAAAARsHSJAAAAAAAAAAAADAKliYBAAAAAAAAAACAUbA0CQAAAAAAAAAAAIyCpUkAAAAAAAAAAABgFCxNAgAAAAAAAAAAAKNgaRIAAAAAAAAAAAAYBUuTAAAAAAAAAAAAwChYmgQAAAAAAAAAAABGYXroAAAAAAAAAAAAAMZhamroAsbOmSYBAAAAAAAAAACAUbA0CQAAAAAAAAAAAIyCpUkAAAAAAAAAAABgFCxNAgAAAAAAAAAAAKNgaRIAAAAAAAAAAAAYBUuTAAAAAAAAAAAAwChYmgQAAAAAAAAAAABGwdIkAAAAAAAAAAAAMAqWJgEAAAAAAAAAAIBRmB46AIa2avXM0AmbbXrh1NAJt3ueF6yL5wVwe/bzL79z6ITNsudR/zB0wmb79jufNnTCZqv2e8TvahjOT29cMXTCZtllh62HTgAAAACA0Zqa8to4w3KmSQAAAAAAAAAAAGAULE0CAAAAAAAAAAAAo2BpEgAAAAAAAAAAABgFS5MAAAAAAAAAAADAKFiaBAAAAAAAAAAAAEbB0iQAAAAAAAAAAAAwCpYmAQAAAAAAAAAAgFGwNAkAAAAAAAAAAACMgqVJAAAAAAAAAAAAYBQsTQIAAAAAAAAAAACjMD10AAAAAAAAAAAAAOMwNTV0AWPnTJMAAAAAAAAAAADAKFiaBAAAAAAAAAAAAEbB0iQAAAAAAAAAAAAwCpYmAQAAAAAAAAAAgFGwNAkAAAAAAAAAAACMgqVJAAAAAAAAAAAAYBQsTQIAAAAAAAAAAACjYGkSAAAAAAAAAAAAGAVLkwAAAAAAAAAAAMAoWJoEAAAAAAAAAAAARmF66AD6t2bNmpx4/HH5Ztdl6623zrGvPSG777HH0FnrVa13vq9dfllOOXlJTjv9g0OnbFS1OVfrna/K86LijCs2J3WeE0nNGWuevGq9ieY+VOvdZYdt8uljHp1nvO1fs2r1mrz9ub+RmSTdVdfn6LMuzczM0IXr5/fI5JnxZGmevGq9ax3xnN/P4sWLkyS73v0eeeWrTxi4aMOqzblab6K5D9V6k3rN1XoTzX2o1pto7kO13kRzH6r1Jpr7UK03qddcrTfR3IdqvUnNZm6/pqamhk5g5AY702Rr7fzW2t7ruP6A1ton511+VWvtZ6216bnLB7bWPtJae1xr7fkbePydW2vPvLU967nt7q21J859/KDW2iM39etsCc777LlZsXxFPnjWh/OSl708b3nzSUMnbVC13rXOPP19Of64Y7J8+YqhUzZJtTlX612r0vOi4owrNld6TiQ1Z6x58qr1Jpr7UKl3euFU3vTsh2bZytVJktf+/oPzxo9cmSe/6fxkKnncQ+4+bOAG+D0yeWY8eZonr1pvkqxYvjzJTN76ng/kre/5wBa/MJnUm3O13kRzH6r1JvWaq/UmmvtQrTfR3IdqvYnmPlTrTTT3oVpvUq+5Wm+iuQ/VepOazQCTsiW+PffFSfZpra1tOyTJeUn2n7v820nO6brunK7rTtvA4+yT5EkTajxoXs/TkjxgQl9nIi695CvZ7xEHJEn2efBDcuWVVwxctGHVete65267ZcnbThk6Y5NVm3O13rUqPS8qzrhic6XnRFJzxponr1pvorkPlXqP/b198tcX/Ed+dN2yJMk+e9wxX/jmT5Ik533tmjzy/ncZMm+D/B6ZPDOePM2TV603Sb7zrS7Lli3LK1/8/Lz8BX+cf7/isqGTNqranKv1Jpr7UK03qddcrTfR3IdqvYnmPlTrTTT3oVpvorkP1XqTes3VehPNfajWm9RsBpiUod+ee6fW2ieS3GGu5Ziu685rrV2a2cXJ72Z2sfPsJL+T5IIkj0pyeGvt8CR7Jzk1yYeSfD/JfZJ8qeu6o5L8ZZIHz52N8lNJTkuybZJfJHl+koVJPp7k2iRrz2z5utbaLkmWJ3lO13U/aa29Jckj5j5/VpJ3Jjk6yXatta8nOTzJitbaJUl2THJCkmVzj/u/kzwkyavmHnO3ud6Dkjw4ydu7rnvPbTHIzbF06U3ZYYftb7m8cMHCrFq1KtPTQz8d1q1a71qPfuwhufqqHwydscmqzbla71qVnhcVZ1yxudJzIqk5Y82TV6030dyHKr3P2G+PXHvT8px/5Y/y4kNnTzw//x0hblq+Kjtsu9VAdRvn98jkmfHkaZ68ar1Jss2iRfn9Z/5RHv+7T8sPvv+9vOplR+XMD388C7fg5mpzrtabaO5Dtd6kXnO13kRzH6r1Jpr7UK030dyHar2J5j5U603qNVfrTTT3oVpvUrMZYFKGPtPkK5J8puu6RyZ5epL3t9amknwmyQFJDp77+DNJHttaW5Rkp67rvvtLj3O/JH+c5GFJHt9a2zXJiUnOmzsb5ZIk7+i67sC5j9eeY3jXJAd3Xfemucv/t+u6gzK7TPmq1toTktwrycMzuzj5zMyeVfKkJGd1XXdmkjOSvDXJlzO7mPnUruseldkFz2PmHveemT0j5VFz1z07yaFJjviVJ3crLF68fZYuXXrL5TUza7boX4LVequqNudqvRVVnHHF5moqzljz5FXrTTT3oUrvYfvvmUfd/675v694VH5ttx1zyv/+jeyywza3fH77baZzw80rByy8fanyvKis4ow1T1613iS55+575jGPe0Kmpqay2+575g477pRrr/3p0FkbVG3O1XoTzX2o1pvUa67Wm2juQ7XeRHMfqvUmmvtQrTfR3IdqvUm95mq9ieY+VOtNajYDTEqvS5Otte1ba/NP07I4yYVJ0nXdVUluSHKX/PfS5KFJPtl13fVJrk/yuCTnr+Ohv9113Y1d161O8sMki37p8w9K8hettfOTvCbJXeeu/8+u61bMu92Fc//8QpKW5P5J/rXrupmu61Zm9q3D1/dW3LskuWHu+1j7WL829/EVc/e/Lsl35r7mz9fR2Yt9931oPnfh7Ld6+WVfzV573W+IjE1WrbeqanOu1ltRxRlXbK6m4ow1T1613kRzH6r0PuXNF+QpSy7IU5dckCu/f31edPqXc94V12S/+905SXLQg3bNF7+1ZS/pVFLleVFZxRlrnrxqvUlyzsf/Kae+Y0mS5Kc/+XFuXnpT7nSnXQau2rBqc67Wm2juQ7XepF5ztd5Ecx+q9Saa+1CtN9Hch2q9ieY+VOtN6jVX600096Fab1KzGWBS+l4ZPzPJO1trF2R2OfLizC5HXtpau0eSOya5tuu6H7XW7p5kmySXzN3305k9M+UJ63jcmXVctyb/vRT6jSRLuq77Qmtt78y+xffa28z3sCQfmWu6IsnXkzw3ydvmlj33m/seHjjvsdd+nZ8muUNr7W5d1/1w7mt8cwN9gznoMY/NRRd9Ps951mGZmZnJ6054/dBJG1Stt6pqc67WW1HFGVdsrqbijDVPXrXeRHMfqvXOd9zfX563PPvXs9X0gnzrhzfk41+p89bMW7rKz4sqKs5Y8+RV602SQ5/01Lzp+L/MS57/nGRqKn/2l8dv0W/NndSbc7XeRHMfqvUm9Zqr9Saa+1CtN9Hch2q9ieY+VOtNNPehWm9Sr7lab6K5D9V6k5rNAJMyNTPT3z5fa22/JO+Yu/jJJCcnOT3Jzkm2TfLqruvOmbvt6UkWdF13+Nzlh2T2DJB36rruF621w5PsneTUJGd3XffwudtdnOSwJCuTnJvkvUk+luQ9mT2z47ZJXpLZM1LOv9/5Sb6bZM/MnvHyj7qu+3lrbUmS/ZNsneTvuq57Y2tt3yRnJzk2yY1J3pzkBUkWJjk+s4uUP09yeGYXLI/suu6wuYXNU7uuO7C1tlOSi7uu23tjc1u2astaury9WbW63ninF04NnXC753nBunheAGw59jzqH4ZO2GzffufThk7YbNV+j/hdDcP56Y0rNn6jLcguO2w9dAIAAADA7c6i6XjBk03ym2+4oN4L+vTmi6961MR/lvS6NMmvxtLkZPnDKuviecG6eF4AbDksTfaj2u8Rv6thOJYmAQAAALA0yaayNMmG9LE0uWDjNwEAAAAAAAAAAACob3roAAAAAAAAAAAAAMZhyjlJGZgzTQIAAAAAAAAAAACjYGkSAAAAAAAAAAAAGAVLkwAAAAAAAAAAAMAoWJoEAAAAAAAAAAAARsHSJAAAAAAAAAAAADAKliYBAAAAAAAAAACAUbA0CQAAAAAAAAAAAIyCpUkAAAAAAAAAAABgFCxNAgAAAAAAAAAAAKNgaRIAAAAAAAAAAAAYhemhAwAAAAAAAAAAABiHqampoRMYOWeaBAAAAAAAAAAAAEbB0iQAAAAAAAAAAAAwCpYmAQAAAAAAAAAAgFGwNAkAAAAAAAAAAACMgqVJAAAAAAAAAAAAYBQsTQIAAAAAAAAAAACjYGkSAAAAAAAAAAAAGAVLkwAAAAAAAAAAAMAoWJoEAAAAAAAAAAAARmF66AAAAAAAAAAAAADGYWpq6ALGztIko7dqzZqhEzbb9MKFQydstutuXjl0wmbZabuthk5gC+TnBcCW47vv+b2hEzbbIad8fuiEzfbpF+0/dMJmmV7oVZY+LFu5euiEzbJoK8dDffDvUAAAALdP11y/bOiEzbbrjouGTrjdW7V6ZuiEzea1QwDm8/bcAAAAAAAAAAAAwChYmgQAAAAAAAAAAABGwdIkAAAAAAAAAAAAMAqWJgEAAAAAAAAAAIBRsDQJAAAAAAAAAAAAjIKlSQAAAAAAAAAAAGAULE0CAAAAAAAAAAAAo2BpEgAAAAAAAAAAABgFS5MAAAAAAAAAAADAKEwPHQAAAAAAAAAAAMA4TE1NDZ3AyDnTJAAAAAAAAAAAADAKliYBAAAAAAAAAACAUbA0CQAAAAAAAAAAAIyCpUkAAAAAAAAAAABgFCxNAgAAAAAAAAAAAKNgaRIAAAAAAAAAAAAYBUuTAAAAAAAAAAAAwChYmgQAAAAAAAAAAABGwdIkAAAAAAAAAAAAMAqWJgEAAAAAAAAAAIBRmB46AAAAAAAAAAAAgHGYmhq6gLGzNDlCa9asyYnHH5dvdl223nrrHPvaE7L7HnsMnbVe1XqTZNXKlTn+uGPyw6uvysoVK/Pc5x2RRx540NBZG1Rxzkny859dm+c9+/fzlnf9VfbY895D52xQtRlX601qNlf7eVFxxponr1pvorkP1XqTOs0LppI/e8x9s9vO22ZmJnnrZ7+ThQum8pLfvldWr0lWrl6T13/6W/n5zSuHTv0fqsx4vmrN1XqTesdDSb05V+ud72uXX5ZTTl6S007/4NApG1VtztV6E819qNab1Guu1pto7kO13kRzH6r1Jpr7UK030dyHar1rnf3X78/F/3p+Vq5amSc+9ffzuCc+deik9ao444rNidcBJq1iM8CkjP7tuVtr57fW9t7M+9yrtfaPc/f9fGvt3a21HdZz2xfO/fNxrbXn3xbNt9Z5nz03K5avyAfP+nBe8rKX5y1vPmnopA2q1pskn/rnj2fHHXfKaR/4m5z87vdmyUknDJ20URXnvGrVyix5w2uzzaJFQ6dskmozrtab1Gyu9vOi4ow1T1613kRzH6r1JnWa97v3zkmSF374a3n/F76X/7P/7nnRgffK2//lP/PSf7giF3772jzzf91j4Mp1qzLj+ao1V+tN6h0PJfXmXK13rTNPf1+OP+6YLF++YuiUTVJtztV6E819qNab1Guu1pto7kO13kRzH6r1Jpr7UK030dyHar1JctklX86/f+2reet7z8ySd52en/zoR0MnbVDFGVds9jrA5FVsBpiU0S9Nbq7W2rZJPpbkTV3XHdh13f5JvpjkQ+u5yzFJ0nXdOV3XndZT5gZdeslXst8jDkiS7PPgh+TKK68YuGjDqvUmyaMPPiRHvODFsxdmkoULt/yTulac87tPXpLffeozsssudx46ZZNUm3G13qRmc7WfFxVnrHnyqvUmmvtQrTep0/y57/wsS879dpLkrnfYJjctX53XfrLLt3+yNEmycMFUVqxeM2TielWZ8XzVmqv1JvWOh5J6c67Wu9Y9d9stS952ytAZm6zanKv1Jpr7UK03qddcrTfR3IdqvYnmPlTrTTT3oVpvorkP1XqT5Ctf/EL2vM9eee3RL8trXvmi/Ob+jxw6aYMqzrhis9cBJq9iM8CkbPl/iejHTq21TyS5Q2ZnckzXdee11h6V5MQkq5N8J8kRSX4nyQVd131x7Z27rjuztXZUa+1eSY5Ncqe5//xzkp1ba+9O8qUkeyf5aJI3zN11lySLu667Vx/f5FpLl96UHXbY/pbLCxcszKpVqzI9vWU+Har1Jsl22y1OkixdujRHv+KlOXLtHwC3YNXm/KmPfyQ73XHnPOy39s/fnvFXQ+dskmozrtab1Gyu9vOi4ow1T1613kRzH6r1JrWaV88krzpkrxxwn53zmk90+dnS2bfi/rW77ZCnPvhuedHff23gwnWrNOO1qjVX603qHQ8l9eZcrXetRz/2kFx91Q+Gzthk1eZcrTfR3IdqvUm95mq9ieY+VOtNNPehWm+iuQ/VehPNfajWmyTXX3ddfnzN1XndknfmmquvynF//uK870MfzdTU1NBp61RxxhWbvQ4weRWbASbFmSZnvSLJZ7que2SSpyd5f2ttKslfJXlq13WPSnJVksOT3DuzC5S/7D+T7DH38Xld1+3Xdd2JSX7Wdd2frL1R13UXdV13YJKnJrkhydMm8y2t3+LF22fp0qW3XF4zs2aL/iVYrXetH13zw/zJ8w7PoU94Yg55/BOGztmoanP+5Mf/Kf/2xYvykiMOz7e/2eX1x/5Frv3pT4fO2qBqM67Wm9RsTmr9vKg4Y82TV6030dyHar1JveY3fPpb+cMzLsmfPfY+WTS9IL99v13y8kffJ3/+0X/P9b9YNXTeOlWbcVKvuVrvWpWOh5J6c67WW1W1OVfrTTT3oVpvUq+5Wm+iuQ/VehPNfajWm2juQ7XeRHMfqvUmyR123DG//pv7Zauttspue+yZrbbeJtf//GdDZ61XxRlXbK6m4owrNgNMyiiXJltr27fWtpp31eIkFyZJ13VXZXaZ8S5J7pbk71pr5yc5OLNLkVcl2XMdD3vfJP8193G3sa+f5CNJXtN13SW/8jfyK9p334fmcxdemCS5/LKvZq+97td3wmap1psk117707z4qOflhS/50zzpyb3vxf5Kqs35lNPOzDtOOyNvf+8Zue/9Wv7ita/PnXbZZeisDao242q9Sc3maj8vKs5Y8+RV600096Fab1Kn+eD73znP+o17JEmWrVqTmZnkkXvdKU99yK55yd9fkR9ev3zgwvWrMuP5qjVX603qHQ8l9eZcrbeqanOu1pto7kO13qRec7XeRHMfqvUmmvtQrTfR3IdqvYnmPlTrTZJfe/C++bcvfiEzMzO59ic/zrJf/CI77LjT0FnrVXHGFZurqTjjis0AkzLWlfEzk7yztXZBZpcjL05yQJJLW2v3SHLHJD9P8oMkv9t13fWttScluSmzb7P9l621h3Vd96Ukaa39nyQ/7bruP1prSbJm3tf6/51DvLW2TZJ/TPKuruvOneQ3uT4HPeaxueiiz+c5zzosMzMzed0Jrx8iY5NV602SM953Wm644fqcftqpOf20U5Mkb3vXe7No0aKBy9av4pyrqTbjar1JzeZqPy8qzljz5FXrTTT3oVpvUqf5wm9dm6MP2SvvePoDM71gKqec/585+pD75kc3LM/xT9w7SXLZVdfnAxd9f+DS/6nKjOer1lytN6l3PJTUm3O13qqqzblab6K5D9V6k3rN1XoTzX2o1pto7kO13kRzH6r1Jpr7UK03SR6+/6NyxVcvyYv/+FlZM7MmL3z5q7Jw4cKhs9ar4owrNldTccYVmwEmZWpmZmboht611vZL8o65i59McnKS05PsnGTbJK/uuu6c1trBSV6T2TNy3pDkOV3X/bi1dp8kb0typ8wunl6e5M+6rruutXZGkrO7rjtn7mv9S2bPTnlukr2TXJvklXP3WXvk99Su69Z7vvFlqzK+/5J6tGzl6qETNtuirbbcf2lYn+tuXjl0wmbZabutNn4jRsfPCwBujUNO+fzQCZvt0y/af+gEtkDVjokcD/Vj1epaL11ML5za+I0AAADINdcvGzphs+2645b7f8S8vaj2OkDitYC+LHAGzHkAACAASURBVJqOQbNJHrHkX+v9IKE3n3vFARP/WTLKpclqLE1OVrU/+CU1/+hnaZLbAz8vALg1LE1ye1HtmMjxUD+q/bHEH0oAAAA2jaVJ1qXa6wCJ1wL6YmmSTWVpkg3pY2lyrG/PDQAAAAAAAAAAQM+mpuzXMqwFQwcAAAAAAAAAAAAA9MHSJAAAAAAAAAAAADAKliYBAAAAAAAAAACAUbA0CQAAAAAAAAAAAIyCpUkAAAAAAAAAAABgFCxNAgAAAAAAAAAAAKNgaRIAAAAAAAAAAAAYBUuTAAAAAAAAAAAAwChYmgQAAAAAAAAAAABGwdIkAAAAAAAAAAAAMArTQwcAAAAAAAAAAAAwDlNTU0MnMHLONAkAAAAAAAAAAACMgqVJAAAAAAAAAAAAYBQsTQIAAAAAAAAAAACjYGkSAAAAAAAAAAAAGAVLkwAAAAAAAAAAAMAoWJoEAAAAAAAAAAAARsHSJAAAAAAAAAAAADAKliYBAAAAAAAAAACAUbA0CQAAAAAAAAAAAIyCpUkAAAAAAAAAAABgFKaHDgAAAAAAAAAAAGAcpqaGLmDsLE1ym1u1embohM2yaKuFQyeMwk7bbTV0Atxqfl7AcLqrbxw6YbO0u+8wdAJboE+/aP+hEzbbA4/+1NAJm+WKkw4dOmEUHBOxLtMLvcpJfdVe00r8bw8AwDHc5O2646KhE9gCVXseA8Av8/bcAAAAAAAAAAAAwChYmgQAAAAAAAAAAABGwdIkAAAAAAAAAAAAMAqWJgEAAAAAAAAAAIBRsDQJAAAAAAAAAAAAjIKlSQAAAAAAAAAAAGAULE0CAAAAAAAAAAAAo2BpEgAAAAAAAAAAABiF6aEDAAAAAAAAAAAAGIepqamhExg5Z5oEAAAAAAAAAAAARsHSJAAAAAAAAAAAADAKliYBAAAAAAAAAACAUbA0CQAAAAAAAAAAAIyCpUkAAAAAAAAAAABgFCxNAgAAAAAAAAAAAKNgaRIAAAAAAAAAAAAYBUuTAAAAAAAAAAAAwChYmgQAAAAAAAAAAABGwdIkAAAAAAAAAAAAMArTQwcAAAAAAAAAAAAwDlNTQxcwds40CQAAAAAAAAAAAIyCpUkAAAAAAAAAAABgFLw99witWbMmJx5/XL7Zddl6661z7GtPyO577DF01kZ97fLLcsrJS3La6R8cOmWjKs64WnO13qRec7XeRHMfqvUmmvtQrfdbX78if/u+d+S4t5yWa676ft715uMyNTWV3fa8T/74RX+eBQu2zP9fUbU5V+tNNE/KgqnkxKc/KPe+8+LMZCav/scr84JH3yd3vsM2SZJ73HHbfPV71+Wlf3vZwKXrVmHG81XrTTT3oVpvorkP1XqTms2J17QmrVpztd5Ecx+q9Saa+1CtN9Hch2q9Sc3mxDHcpFVrrtabaO5Dtd6kZjPApGyZfxHegNba+a21vdfzue+21hZt4uOc0Vq7rrW2zbzrHtpam2mtHbiZTbu31p64jusf11p7/gbud3hr7Umb87VuC+d99tysWL4iHzzrw3nJy16et7z5pL4TNtuZp78vxx93TJYvXzF0yiapOONqzdV6k3rN1XoTzX2o1pto7kOl3o9++Myc+tbjs3LF7DHFmae+NYc996i87m3vy8zMTP7tCxcMXLh+leac1OtNNE/KQQ+4S5LkGe+6OG8751t5+ePul5f+7WV51nu+lKPOuCQ3/GJVTvzYNwauXL8KM56vWm+iuQ/VehPNfajWm9Rs9prW5FVrrtabaO5Dtd5Ecx+q9Saa+1CtN6nZ7Bhu8qo1V+tNNPehWm9SsxlgUsotTd7Gfpjk0HmXn5XkP36Fxzkoyf6/fGXXded0XXfa+u7Udd0ZXdd97Ff4erfKpZd8Jfs94oAkyT4PfkiuvPKKvhM22z132y1L3nbK0BmbrOKMqzVX603qNVfrTTT3oVpvorkPlXrvevd75hXHvvmWy//xrW/kAfv8epJk34ftl8sv+dJQaRtVac5Jvd5E86Sce+WPc8w/zHbd/Y7b5oZlK2/53EsO3isf/Pz38pMblw+Vt1EVZjxftd5Ecx+q9Saa+1CtN6nZ7DWtyavWXK030dyHar2J5j5U600096Fab1Kz2THc5FVrrtabaO5Dtd6kZjPApFR9e+6dWmufSHKHzH4Px3Rdd97c597TWrvX3MdPSfK7SR6fZLsk90nyxq7rzpj7/IeS/EGSj7TWFiR5aJIvJ0lr7Q5J3pdkpyR3T/Kuruve01r7kyR/lGTN3G1fluToJNu11r6Q5E+T/DjJznOPv1eSU+c+/v5cw5e6rjuqtXZckmu6rjv1th3Phi1delN22GH7Wy4vXLAwq1atyvT0lvt0ePRjD8nVV/1g6IxNVnHG1Zqr9Sb1mqv1Jpr7UK030dyHSr0PP+DR+fE1V//3FTMzmZqaSpJsu+3i3HzzTQOVbVylOSf1ehPNk7R6zUzedNiDcvADd80L//rSJMnO22+d/fa6U0782NcHrtuwKjNeq1pvorkP1XoTzX2o1pvUbPaa1uRVa67Wm2juQ7XeRHMfqvUmmvtQrTep2ewYbvKqNVfrTTT3oVpvUrMZYFKqnmnyFUk+03XdI5M8Pcn7W2tTc597f9d1Byb5bpLHzl23Y9d1T0jypMwuOK71pSR7t9YWZ/Zskf8y73P3TXJ213UHJzk4s8uQSfLcJC/suu63knw9yVSSk5KcNe+skR/quu4xSVbPe7z7JfnjJA9L8vjW2q634vu/VRYv3j5Lly695fKamTV+Cd7GKs64WnO13qRec7XeRHMfqvUmmvtQrXe+qQX/fTj8i18szeLF22/g1sOqNudqvYnmSXvl2V/LY954QU58+gOz7dYLc+g+u+Zjl/4wa2aGLtuwSjNO6vUmmvtQrTfR3IdqvUnN5moqzrhac7XeRHMfqvUmmvtQrTfR3IdqvUnN5moqzrhac7XeRHMfqvUmNZsBJqXE0mRrbfvW2lbzrlqc5MIk6bruqiQ3JLnL3Oe+MvfPazJ7dskk+ercP7+fZNEvPfxHM3s2ymcm+Zt51/8oyZNba3+T5Jgka7/+c5O8oLV2QZI9Mrs0+cu6dVz37a7rbuy6bnVm3xb8lzt6s+++D83nLrwwSXL5ZV/NXnvdb6iU262KM67WXK03qddcrTfR3IdqvYnmPlTrnW/P+7Zcedm/JUku/dIXcv8H7Ttw0fpVm3O13kTzpDz5oXfPkQfdO0mybMWarJmZyZo1M9lvrzvlwm/8ZOC6jasw4/mq9Saa+1CtN9Hch2q9Sc3mairOuFpztd5Ecx+q9Saa+1CtN9Hch2q9Sc3mairOuFpztd5Ecx+q9SY1mwEmpcrK+JlJ3jm3qHiXJBcnOSDJpa21eyS5Y5Jr5267rvOUbOjcJWclOTnJTNd1/9FaW3v9y5NcNPeW3L+d5Hfmrn9ekiO7rlvWWvt0kv0y+1bd8xdQ12xmQ68Oesxjc9FFn89znnVYZmZm8roTXj900u1OxRlXa67Wm9RrrtabaO5Dtd5Ecx+q9c73nCNemve+9cSsWvWu3GP3PfPwAx49dNJ6VZtztd5E86R8+oof5Y3PeFDO+pPfzFYLpnLiR7+e5avW5N53Xpz/uvbmofM2qsKM56vWm2juQ7XeRHMfqvUmNZurqTjjas3VehPNfajWm2juQ7XeRHMfqvUmNZurqTjjas3VehPNfajWm9RsBpiUqZmZLWaXb71aa/slecfcxU9mdsnx9CQ7J9k2yau7rjuntfbdJHvPLTSelOQbc/fZu+u6o1tri5J8o+u6PVtrZ2T27bfPaa19JbNv6/3u1trZSU7N7BkkT8nsMuZ1SR6Y5AFJnp3kiCQ3Jrkqs0uU909ydpJjkxyZ2aXKb7TWDk+y99zjnd113cPnvp+LkxyW5PAk13Rdd+qGvv9lq7achctNsWp1qdxML1zXyUIBgC1Jd/WNQydslnb3HYZOgNvEA4/+1NAJm+WKkw4dOgGAwqq9ppV4XQsAwDEcwJZl0fQ6360V/oeD3nFRvV/i9Oa8F//WxH+WlFiaHDtLk5PlX0wAYMtnaRKGYWkSgDGp9ppW4nUtAADHcABbFkuTbCpLk2xIH0uTCzZ+EwAAAAAAAAAAAID6LE0CAAAAAAAAAAAAo2BpEgAAAAAAAAAAABgFS5MAAAAAAAAAAADAKFiaBAAAAAAAAAAAAEbB0iQAAAAAAAAAAAAwCpYmAQAAAAAAAAAAgFGwNAkAAAAAAAAAAACMgqVJAAAAAAAAAAAAYBSmhw4AAAAAAAAAAABgHKamhi5g7JxpEgAAAAAAAAAAABgFS5MAAAAAAAAAAADAKFiaBAAAAAAAAAAAAEbB0iQAAAAAAAAAAAAwCpYmAQAAAAAAAAAAgFGwNAkAAAAAAAAAAACMgqVJAAAAAAAAAAAAYBQsTQIAAAAAAAAAAACjYGkSAAAAAAAAAAAAGAVLkwAAAAAAAAAAAMAoTA8dAAAAAAAAAAAAwDgsmJoaOoGRc6ZJAAAAAAAAAAAAYBQsTQIAAAAAAAAAAACj4O25uc1NL6x1Ct1Vq2eGTths1WYMtxfLVq4eOmGzLdpq4dAJcJtod99h6AS2MI7h+nHFSYcOnbBZ7vQHHxg6YbNd+6HnDp0AAAAAZVV8vaUafxsBAG6PnGkSAAAAAAAAAAAAGAVLkwAAAAAAAAAAAMAoWJoEAAAAAAAAAAAARsHSJAAAAAAAAAAAADAKliYBAAAAAAAAAACAUbA0CQAAAAAAAAAAAIzC9NABAAAAAAAAAAAAjMPU1NAFjJ0zTQIAAAAAAAAAAACjYGkSAAAAAAAAAAAAGAVLkwAAAAAAAAAAAMAoWJoEAAAAAAAAAAAARsHSJAAAAAAAAAAAADAKliYBAAAAAAAAAACAUbA0CQAAAAAAAAAAAIyCpUkAAAAAAAAAAABgFCxNAgAAAAAAAAAAAKNgaRIAAAAAAAAAAAAYhemhAwAAAAAAAAAAABiHqampoRMYOWeaBAAAAAAAAAAAAEbB0iQAAAAAAAAAAAAwCpYmAQAAAAAAAAAAgFGYHjqA/q1ZsyYnHn9cvtl12XrrrXPsa0/I7nvsMXTWelXrne9rl1+WU05ektNO/+DQKRtVbc7VepN6zdV6k5rNq1auzPHHHZMfXn1VVq5Ymec+74g88sCDhs5ar4oz1jx51XoTzX2o1jufY7jJqdZ75zssyufe+KQ88fhPZ8GC5JQj9s9Uku9cc0P+5D2fz+o1M0Mn/g/VZpxo7kO13kRzH6r1JjWbE8cWk1atuVpvorkP1XoTzX2o1pto7kO13kRzH6r1Jv420gfNk1etN6nZDDAp5c402Vo7v7W2923wOLu31p54Wz1ma+27rbULW2sXtNb+rbX2yrnrH9dae/46bv+g1tojb83X/FWd99lzs2L5inzwrA/nJS97ed7y5pOGyNhk1XrXOvP09+X4447J8uUrhk7ZJNXmXK03qddcrTep2fypf/54dtxxp5z2gb/Jye9+b5acdMLQSRtUccaaJ69ab6K5D9V613IMN1mVeqcXTuUdR+yXX6xYlSQ57g9+Pced9ZU85tWfTJI8/n/tNmTeelWa8VqaJ69ab6K5D9V6k5rNji0mr1pztd5Ecx+q9Saa+1CtN9Hch2q9ieY+VOtN/G2kD5onr1pvUrMZYFLKLU3ehg5Ksv9t/JgHd133qCT7JTmitXaXruvO6brutHXc9mlJHnAbf/1NcuklX8l+jzggSbLPgx+SK6+8YoiMTVatd6177rZblrztlKEzNlm1OVfrTeo1V+tNajY/+uBDcsQLXjx7YSZZuHDLPgl0xRlrnrxqvYnmPlTrXcsx3GRV6n3Dcx6W9/9/38gPf35zkuSZb/mXfP7rP8pW0wty1522zQ03rxy4cN0qzXgtzZNXrTfR3IdqvUnNZscWk1etuVpvorkP1XoTzX2o1pto7kO13kRzH6r1Jv420gfNk1etN6nZDDApW/bRx/rt1Fr7RJI7ZPZ7OKbruvNaa5cnuSDJPklmkvxukpuSvDfJbknuluRjSY5NcnSS7VprX5h7zGNba3dNsjjJHyT53i/fr+u6Y1prZyRZnmTPuesP77rukl/q2y7JyiQ3t9YOT7J3klOTfDzJtUn+JcnhSVa01i7puu5Lt9lkNsHSpTdlhx22v+XywgULs2rVqkxPb5lPh2q9az36sYfk6qt+MHTGJqs252q9Sb3mar1JzebttlucJFm6dGmOfsVLc+TaFwm2UBVnrHnyqvUmmvtQrXctx3CTVaX3Dw+8b35y/bKce9nVeflT9kmSrFkzk912WZxPvOZxueHmFfnad382cOW6VZnxfJonr1pvorkP1XqTms2OLSavWnO13kRzH6r1Jpr7UK030dyHar2J5j5U6038baQPmievWm9SsxlgUqqeafIVST7Tdd0jkzw9yftba1OZXaL80NzZHq9Kcmhmlx4v7rrukCQPS3Jk13Wrk5yU5Kyu6z4295j//P/Yu/MwScv63sPfnmlgZBE3DC6AUfHBnCjqiZqoRAQUNG6JxoALQiKCoqIhMRpRkMVgXEAxLmAQ44bJiebEoxiPGEJETIwLCMqjJlFQwQVRZGQYeqbPH92jc3BWpd6aH+99X1dfM9Vd9dann2u6u6r6N8/be98nyTlJnryu2611/99YfP9pSdY+9fbHWmv/kqQnuTDJ8pt075yF3ShfmeSsJK8femAySbbbbvssX/6ztNXzq7foH4LVequqts7VepN6zdV6k5rNSfKdq67Mcw87JI9+7OOy/2MeO+2cDaq4xponr1pvonkI1XqrqrbOVXqf8Yjds8+ed845xx2Q+97tdjnj+XvlV25zq1zx/eXZ8wV/n7d/7LKc/MwHTTtznaqs8do0T1613kTzEKr1JjWbq6m4xtWaq/UmmodQrTfRPIRqvYnmIVTrTTQPoVrvGn43MlmaJ69ab1KzGWBSSgxNtta2b61ttda7tktyfpL03r+V5Nokd1z82OcX/7wiybIkP0jywNbae5KckmSb9dzNZxf/vCoLO0Vu6HY3vY811pyee5fFt6fd5D7+u/e+csOf7eTd//4PyCfPPz9JcvFFX8juu99rykUbVq23qmrrXK03qddcrTep2Xz11d/PC55zWJ531B/n8U980rRzNqriGmuevGq9ieYhVOutqto6V+nd/9hzcsCx5+TRx300F3/9BznstH/NaYc/JPfY+dZJkutWzGX1/PyUK9etyhqvTfPkVetNNA+hWm9Ss7maimtcrblab6J5CNV6E81DqNabaB5Ctd5E8xCq9SZ+NzIEzZNXrTep2QwwKVVGxt+Z5E2LuzjeMcmnk+yV5POttbskuW0WTnudLJyWe22HJPlh7/3w1to9kzx7cVfK1fn/h0Y39Xbruu7/p/e+srX2nSRbL97PGjf9+1SGVvfZ75G58MILcvDTDsz8/HyOP/FV08jYZNV6q6q2ztV6k3rN1XqTms1nvf30XHvtj3Lm6W/Nmae/NUlyyl+9LcuWLdvILaej4hprnrxqvYnmIVTrraraOlfrXdvrPvjFvO3Ih2Xl3Opcv3IuR77lgmknrVPFNdY8edV6E81DqNab1GyupuIaV2uu1ptoHkK13kTzEKr1JpqHUK030TyEar2J340MQfPkVetNajYDTMrM/Ba6O8baWmsPSfLGxYsfSXJqkjOT3C7JrZK8vPf+0dba15Ps0Xtf0Vo7OcllST6T5L1JfpTkhizsALlvFoYvz05ybBZOvX1E7/2y1toRWTiN9t+t53YnJTl78f4OSHJg7/2Qxfu+PMmqLAyjXpHk0CQHJdkjyVsXb/ebi5/T7yR5TZIje+//vKHPf8Xchoc0+eXMraq3vLNLZzZ+JeBmt+LGVdNO2GzLtlo67QSAifAYjnW5/UHvmHbCZrv6fYdOOwGARR5fAADAz/O7EWBzLJuNJ6pskke/5d/qvRDDYM55zoMn/r2kxNDk2BmanCwviAObygsDAFsOj+FYF0OTAPwyPL4AAICf53cjwOYwNMmmMjTJhgwxNDmV00MDAAAAAAAAAAAADM3QJAAAAAAAAAAAADAKhiYBAAAAAAAAAACAUTA0CQAAAAAAAAAAAIyCoUkAAAAAAAAAAABgFAxNAgAAAAAAAAAAAKNgaBIAAAAAAAAAAAAYBUOTAAAAAAAAAAAAwCjMTjsAAAAAAAAAAACAcZiZmZl2AiNnp0kAAAAAAAAAAABgFAxNAgAAAAAAAAAAAKNgaBIAAAAAAAAAAAAYBUOTAAAAAAAAAAAAwCgYmgQAAAAAAAAAAABGwdAkAAAAAAAAAAAAMAqGJgEAAAAAAAAAAIBRMDQJAAAAAAAAAAAAjIKhSQAAAAAAAAAAAGAUDE0CAAAAAAAAAAAAozA77QAAAAAAAAAAAADGYWZm2gVU1lpbmuSMJC3JfJIjkqxIctbi5UuSHNl7X72+Y9hpEgAAAAAAAAAAAKjgcUnSe39okmOSnJTk9UmO6b3vlWQmyRM2dABDkwAAAAAAAAAAAMAWr/f+D0mevXhxtyQ/TPI/k/zL4vvOSbLfho5haBIAAAAAAAAAAAAoofc+11p7Z5LTkrwnyUzvfX7xwz9OsuOGbj874T7Y4s0unZl2AlDEsq2WTjsBgEUew7EuV7/v0GknbLbbPvB5007YbNd85k3TTgCYCI8vAADg5/ndCACwpeq9P7O19mdJ/i3Jrdb60A5Z2H1yvew0CQAAAAAAAAAAAGzxWmvPaK29dPHiT5KsTvIfrbW9F9/36CT/uqFj2GkSAAAAAAAAAAAAqOADSd7RWjs/yVZJXpjky0nOaK1tvfj3/7WhAxiaBAAAAAAAAAAAALZ4vfflSZ6yjg89fFOP4fTcAAAAAAAAAAAAwCgYmgQAAAAAAAAAAABGwem5AQAAAAAAAAAAGMRMZqadwMjZaRIAAAAAAAAAAAAYBUOTAAAAAAAAAAAAwCgYmgQAAAAAAAAAAABGwdAkAAAAAAAAAAAAMAqGJgEAAAAAAAAAAIBRMDQJAAAAAAAAAAAAjIKhSQAAAAAAAAAAAGAUDE0CAAAAAAAAAAAAo2BoEgAAAAAAAAAAABgFQ5MAAAAAAAAAAADAKMxOOwAAAAAAAAAAAIBxWDIz7QLGzk6TAAAAAAAAAAAAwCgYmgQAAAAAAAAAAABGwdAkAAAAAAAAAAAAMAqGJgEAAAAAAAAAAIBRMDQJAAAAAAAAAAAAjMLstAMY3urVq3PSCcflK71n6623zrGvPDG77rbbtLPWq1pvonkI1XqTes3VehPNQ6jWm2geQrXeRPMQqvUmmodQrTep1fyp9/5Zfrx8RZLk69+6On/995/Ma//0yZlbtSofv/CyvOr0c6ZcuG6V1niNas3VehPNQ6jWm2geQrXepF5ztd5E8xCq9Saah1CtN9E8hGq9ieYhVOtN6jVX6000D6Fab1KzGWBSyu002Vrbu7V29lqXn9xau6S1tusm3Pa81toerbVDWmuXL17+59baha21p/wCLee11vbY3Nvd5BhX/TK3/0V84tyPZ+UNK/Ou974/R73o6LzuNScPnbBZqvUmmodQrTep11ytN9E8hGq9ieYhVOtNNA+hWm+ieQjVepM6zdtsPZuZmWT/w96Q/Q97Qw4/7t1548sOzDP//Kzsc+gpeeB97pY9212nnblOVdZ4bdWaq/UmmodQrTfRPIRqvUm95mq9ieYhVOtNNA+hWm+ieQjVehPNQ6jWm9RrrtabaB5Ctd6kZjPApJQbmlxba+2gJC9Nsm/v/fLNvPl7e+97994fkeR3kry+tTZzs0dugT7/uc/mIQ/bK0ly3z3vl0svvWTKRRtWrTfRPIRqvUm95mq9ieYhVOtNNA+hWm+ieQjVehPNQ6jWm9Rpvu+97pJtl22dD735yJzztufnoQ+4R7bZajb//c3vJ0k+/qkvZ58HtylXrluVNV5bteZqvYnmIVTrTTQPoVpvUq+5Wm+ieQjVehPNQ6jWm2geQrXeRPMQqvUm9Zqr9Saah1CtN6nZDDApZU/P3Vp7RpLnJ9mv935Na+3hSY7NwiDo9kmemmRlkg8luTrJRzZwuNskub73Pr94nJOSrEryn0kOT/K0JI9Jsm2SeyR5de/9rMXbHtta+5Uk2yU5KMk3krwtyS5J7pTkH3vvx7TWzkpydu/9o621A5Ic2Hs/5OZYi821fPl12WGH7X96eemSpZmbm8vs7Jb5z6Fab6J5CNV6k3rN1XoTzUOo1ptoHkK13kTzEKr1JpqHUK03qdP8kxU35tS/OTfv+OCncs9d75j//abn5Ic/vv6nH//xT1bkV+9yhykWrl+VNV5bteZqvYnmIVTrTTQPoVpvUq+5Wm+ieQjVehPNQ6jWm2geQrXeRPMQqvUm9Zqr9Saah1CtN6nZDDApVXea3CvJs5PcLj8b/PwfSZ7ee987yQeS/P7i+3dO8qje+1/e5BhPXTy99ieSvDHJMxZ3mjwjye/13h+e5FtJDlm8/o6998cmeXySl6x1nA/33vdJck6SJ2dhWPLTvff9kzwoyRE3z6d889luu+2zfPnyn15ePb96i/4hWK030TyEar1JveZqvYnmIVTrTTQPoVpvonkI1XoTzUOo1pvUaf7qN76b933kM0mSr13+3fzouhW53Y7b/vTjO2y7LD9aa4hyS1JljddWrblab6J5CNV6E81DqNab1Guu1ptoHkK13kTzEKr1JpqHUK030TyEar1JveZqvYnmIVTrTWo2A0xK1aHJK5M8MsmpSd7dWluShQHHNy7u6PiIJFstXve/e+8r13GMNafn3qf3/tje+6eT7JSF3SH/trV2XpJHJdlt8fpfWPzziiTL1jrOZxf/vCoLO1H+IMkDW2vvSXJKkm3Wcd9TPQ34/e//gHzy/POTJBdf9IXsvvu9ppmzUdV6E81DqNab1Guu1ptoHkK13kTzEKr1JpqHH9wVnwAAIABJREFUUK030TyEar1JneZnPvE3c/If/26S5E477Zhtl22V5devzK/edWF3yf0ecu9c8PmvTTNxvaqs8dqqNVfrTTQPoVpvonkI1XqTes3VehPNQ6jWm2geQrXeRPMQqvUmmodQrTep11ytN9E8hGq9Sc1mbrlmZma8eVvv2xCqjox/rfe+IsmbWmv7J3lZFk7VfY/e+49ba+/MzwYTV2/Gcb+f5JtJntB7/1Fr7fFJrkuya5L59dzmpu8/JMkPe++Ht9bumeTZiztYrsjCQGaSPGAzmm52++z3yFx44QU5+GkHZn5+Psef+Kpp5mxUtd5E8xCq9Sb1mqv1JpqHUK030TyEar2J5iFU6000D6Fab1Kn+awPXpgzjn9Gzj3zRZmfn88Rr3xPVq+ez1knPTNLly7Jxy+8LJ+55BvTzlynKmu8tmrN1XoTzUOo1ptoHkK13qRec7XeRPMQqvUmmodQrTfRPIRqvYnmIVTrTeo1V+tNNA+hWm9SsxlgUmbm59c3C7hlaq3tneSI3vuBi5d3SvL5JN/LwqDk8iTfSXJ1kpOSnN17/83F656XhdNl/2aSPXrvL1nH8R+V5BVZ2IXz2iQHJ3nMmuu31pYluaz3frc1x+u9X9ZaOyILpwL/uyTvTfKjJDdk4XTd+2ZhYPLMJN9N8pUk2/beD2mtXdV733lDn/OKufUObAIAAFDYbR/4vGknbLZrPvOmaScAAAAAAFugZbPTPfMqdTzhjP8wC8V6/e/DfmPi30vKDU2OkaFJAACAWyZDkwAAAADALYWhSTaVoUk2ZIihySWTvgMAAAAAAAAAAACALYGhSQAAAAAAAAAAAGAUDE0CAAAAAAAAAAAAo2BoEgAAAAAAAAAAABgFQ5MAAAAAAAAAAADAKBiaBAAAAAAAAAAAAEZhdtoBAAAAAAAAAAAAjMPMzLQLGDs7TQIAAAAAAAAAAACjYGgSAAAAAAAAAAAAGAVDkwAAAAAAAAAAAMAoGJoEAAAAAAAAAAAARsHQJAAAAAAAAAAAADAKhiYBAAAAAAAAAACAUTA0CQAAAAAAAAAAAIyCoUkAAAAAAAAAAABgFAxNAgAAAAAAAAAAAKNgaBIAAAAAAAAAAAAYhdlpBwAAAAAAAAAAADAOS2Zmpp3AyNlpEgAAAAAAAAAAABgFQ5MAAAAAAAAAAADAKBiaBAAAAAAAAAAAAEbB0CQAAAAAAAAAAAAwCrPTDgAAAICxuuYzb5p2wma7/UHvmHbCZrn6fYdOO2EU5lbNTzths8wunZl2AgAAcDOo9lwkqfd85LoVc9NO2GzbLzMGMWkrblw17YTNtmyrpdNOAGALYqdJAAAAAAAAAAAAYBQMTQIAAAAAAAAAAACjYGgSAAAAAAAAAAAAGAVDkwAAAAAAAAAAAMAozE47AAAAAAAAAAAAgHGYmZl2AWNnp0kAAAAAAAAAAABgFAxNAgAAAAAAAAAAAKNgaBIAAAAAAAAAAAAYBUOTAAAAAAAAAAAAwCgYmgQAAAAAAAAAAABGwdAkAAAAAAAAAAAAMAqGJgEAAAAAAAAAAIBRMDQJAAAAAAAAAAAAjIKhSQAAAAAAAAAAAGAUDE0CAAAAAAAAAAAAozA77QAAAAAAAAAAAADGYWZmZtoJjJydJgEAAAAAAAAAAIBRMDQJAAAAAAAAAAAAjIKhSQAAAAAAAAAAAGAUDE0CAAAAAAAAAAAAo2BoEgAAAAAAAAAAABgFQ5MAAAAAAAAAAADAKMxOO4DhrV69OiedcFy+0nu23nrrHPvKE7PrbrtNO2u9qvUmmodQrTep11ytN9E8hGq9ieYhVOtNNA+hWm+ieQjVepN6zdV6d7r1snzy1Y/P4074pyxZkpx2+EMzk+Q/r7o2z33LBVm1en7aietUbZ2r9a7tixdflNNOfW1OP/Nd007ZqGrrXK030TyEar1JveZqvYnmIVTrTTQPoVpvonkI1XqTms2J5yKTtmrVqrz6xGNz+Tf+OzOZyZ/++bG5+z13n3bWelVc44rNczfemBOOOyZXfvtbuXHljTn0sMPz23vvM+2s9aq4xhWbASblFr3TZGtt79ba2WtdfnJr7ZLW2q7ruO79WmuvWPz777bW7txau1tr7dM3ud42rbWrWmtLFy//VmttdWvtgYuXl7XWvtVaW+fattZe0lp70M35eW6uT5z78ay8YWXe9d7356gXHZ3XvebkaeZsVLXeRPMQqvUm9Zqr9Saah1CtN9E8hGq9ieYhVOtNNA+hWm9Sr7lS7+zSmbzx8Ifk+pVzSZLjDvqfOe69n81+L/9IkuQxv7HLNPM2qNI6J/V613jnmW/PCccdkxtuWDntlE1SbZ2r9Saah1CtN6nXXK030TyEar2J5iFU6000D6Fab1Kz2XORybvg/POSJG898z057LkvyNve/IbpBm1ExTWu2HzOhz+UHXe8TU5/x7tz6pvflteefOK0kzao4hpXbAaYlFv00OTaWmsHJXlpkn1775ff9OO99y/03o9fvHhUkluv6zi99xuSfCHJ/Rbf9Zgkf7f4Z5L8VpJ/7b2vXs/tT+69//sv/IncDD7/uc/mIQ/bK0ly3z3vl0svvWSaORtVrTfRPIRqvUm95mq9ieYhVOtNNA+hWm+ieQjVehPNQ6jWm9RrrtT7Fwc/KH/9scty5TU/SZI89XX/nAu+/J1sNbskv3KbW+Xan9w45cL1q7TOSb3eNe66yy557SmnTTtjk1Vb52q9ieYhVOtN6jVX6000D6Fab6J5CNV6E81DqNab1Gz2XGTyfvsR++bFLzsuSXLVld/O9tvvMN2gjai4xhWb933U/jn8yBcsXJhPli7dsk+cWnGNKzYDTMqW/VPmZtJae0aS5yfZr/d+TWvtPknemGQmydVJ/jDJ/ZMckeRdWRiI/JskT0+yU2vtH5LcKcnFvffDkvzfJHsl+WySfRev994kr0yyd5KPLt7v85M8Ncl8krN7729srZ2V5OwkOy/e75Ikx/bez53sKvzM8uXXZYcdtv/p5aVLlmZubi6zs1vmP4dqvYnmIVTrTeo1V+tNNA+hWm+ieQjVehPNQ6jWm2geQrXepF5zld6n733PfO9HK/Lxi76do3/3vkmS1avns8sdtsv/ecUBufYnK/PFr/9gypXrV2Wd16jWu8a+j9w/3/7WN6edscmqrXO13kTzEKr1JvWaq/UmmodQrTfRPIRqvYnmIVTrTWo2ey4yjNnZ2Zzwipfm/PPOzYmvPmXaORtUcY0rNm+77XZJkuXLl+clf/LCHLFmgHILVXGNKzYDTMoYdprcK8mzk9wuPxsSPSPJkb33vZN8JMmL11y59/7hLOwkeXCSlVnYcfLQLOwguW9r7Y5ZHJpsrf1KkuW99/9KMtNa2ynJw5P8U2vt15L8QZKHLTY8sbXWbtJ2Te/9YUMOTCbJdtttn+XLl//08ur51Vv0D8FqvYnmIVTrTeo1V+tNNA+hWm+ieQjVehPNQ6jWm2geQrXepF5zld5nPGL37LPnnXPOcQfkvne7Xc54/l75ldvcKld8f3n2fMHf5+0fuywnP/NB085cryrrvEa13qqqrXO13kTzEKr1JvWaq/UmmodQrTfRPIRqvYnmIVTrTWo2V1N5jV9+/F/k7A98OK8+8dhcf/1Ppp2zXhXXuGJzknznqivz3MMOyaMf+7js/5jHTjtngyquccVmbrlmZrx5W//bEMYwNHllkkcmOTXJu1trS5LcO8mbW2vnZWG3x7ts4Pb/1Xu/ZvF0299Nsm2SLybZPckBWdxVMsk/ZWGXyWW99yuT/HqS3ZKcu/h2+8XbrK3/sp/cL+L+939APnn++UmSiy/6Qnbf/V7TyNhk1XoTzUOo1pvUa67Wm2geQrXeRPMQqvUmmodQrTfRPIRqvUm95iq9+x97Tg449pw8+riP5uKv/yCHnfavOe3wh+QeO986SXLdirmsnp+fcuX6VVnnNar1VlVtnav1JpqHUK03qddcrTfRPIRqvYnmIVTrTTQPoVpvUrO5mopr/NEP/2P+5swzkiTLlt0qS5YsyZKZLXd0oeIaV2y++urv5wXPOSzPO+qP8/gnPmnaORtVcY0rNgNMyhhGxr/We1+R5E2ttf2TvCwLw4oH994vb609NAun3l7b6vxsoPTnfkvTe59vrV2c5FlJDl989zlJXpPkvDVXS3JpkkcvXv9FSS5O8uSb3M/g9tnvkbnwwgty8NMOzPz8fI4/8VXTyNhk1XoTzUOo1pvUa67Wm2geQrXeRPMQqvUmmodQrTfRPIRqvUm95mq9a3vdB7+Ytx35sKycW53rV87lyLdcMO2k9aq2ztV6q6q2ztV6E81DqNab1Guu1ptoHkK13kTzEKr1JpqHUK03qdlcTcU1fvg+++VVxx2T5z7r4MzNzeWoo1+SbZYtm3bWelVc44rNZ7399Fx77Y9y5ulvzZmnvzVJcspfvS3LttB/GxXXuGIzwKTMzG/BOzf8slpreyc5ovd+4OLlnZJ8PskpSR6XhaHR+SR/lOTOa67bWjsxC7tIPjvJm3vvv7l4+08nObD3/vXW2jOTvLL3frfFjy3Nwk6UT+q9n7f4vj9N8sQk2yT59yTPT/LXSc5OsnOSPXrvL9nY57Fi7ucHNwEAAGAabn/QO6adsFmuft+h004YhblVtV66mF060DleAACAiar2XCSp93zkuhVz007YbNsvG8PeUdO14sZV007YbMu2WjrthFFYNpta3+SYmt8/63P1fogzmL875AET/15yix6avKUwNAkAAMCWwtAk61LtF5XVfkkJAACsW7XnIkm95yOGJlkXQ5Osj6FJNpWhSTZkiKHJJRu/CgAAAAAAAAAAAEB9hiYBAAAAAAAAAACAUTA0CQAAAAAAAAAAAIyCoUkAAAAAAAAAAABgFAxNAgAAAAAAAAAAAKMwO+0AAAAAAAAAAAAAxmHJzMy0Exg5O00CAAAAAAAAAAAAo2BoEgAAAAAAAAAAABgFQ5MAAAAAAAAAAADAKBiaBAAAAAAAAAAAAEbB0CQAAAAAAAAAAAAwCoYmAQAAAAAAAAAAgFEwNAkAAAAAAAAAAACMgqFJAAAAAAAAAAAAYBQMTQIAAAAAAAAAAACjYGgSAAAAAAAAAAAAGIXZaQcAAAAAAAAAAAAwDjPTDmD07DQJAAAAAAAAAAAAjIKhSQAAAAAAAAAAAGAUDE0CAAAAAAAAAAAAo2BoEgAAAAAAAAAAABgFQ5MAAAAAAAAAAADAKBiaBAAAAAAAAAAAAEZhdtoBAFui7/945bQTNtsddth62glsga5bMTfthM22/TIPTwD4xcytmp92wmabXToz7YTNdvX7Dp12wmZ58AnnTjths/3by/eddgIAwFR5TQumw/Nq1sX3N9Zldon9uQCozU8yAAAAAAAAAAAAYBQMTQIAAAAAAAAAAACjYGgSAAAAAAAAAAAAGIXZaQcAAAAAAAAAAAAwDjMzM9NOYOTsNAkAAAAAAAAAAACMgqFJAAAAAAAAAAAAYBQMTQIAAAAAAAAAAACjYGgSAAAAAAAAAAAAGAVDkwAAAAAAAAAAAMAoGJoEAAAAAAAAAAAARsHQJAAAAAAAAAAAADAKhiYBAAAAAAAAAACAUTA0CQAAAAAAAAAAAIzC7LQDAAAAAAAAAAAAGIclM9MuYOzsNAkAAAAAAAAAAACMgqFJAAAAAAAAAAAAYBQMTQIAAAAAAAAAAACjYGgSAAAAAAAAAAAAGAVDkwAAAAAAAAAAAMAoGJoEAAAAAAAAAAAARsHQJAAAAAAAAAAAADAKs9MOYHirV6/OSSccl6/0nq233jrHvvLE7LrbbtPOWq9qvYnmIVTrXePwg5+S7bbbLkmy853vkhe//MQpF61fxTXWPHmrVq3Kq088Npd/478zk5n86Z8fm7vfc/dpZ21QtTVO6jVX6000D6Fab6J5CNV61/jixRfltFNfm9PPfNe0Uzaq4hpXaV4ykxz7+Htntztsm8wnJ/6fyzK7ZCanPW3PfOPq65Mkf/eZb+afLv3ulEt/XpU1Xhdff5NTrTfRPIRqvUm95mq9ieYhVOtNajZXe12r4hprnrxqvWvz2H5yqvUm9Zqr9SY1mxPfKyatYjPApExsp8nW2t6ttbPXuvzk1tolrbVdN/H292utvWIjx/9ua+281tq/tNY+1Vp7ys3Rvnj8q26uY93kuGe11g6YxLE31SfO/XhW3rAy73rv+3PUi47O615z8jRzNqpab6J5CNV6k2TlDTckmc/r3/KOvP4t79iiByaTmmusefIuOP+8JMlbz3xPDnvuC/K2N79hukGboNoaJ/Waq/UmmodQrTfRPIRqvUnyzjPfnhOOOyY33LBy2imbpOIaV2l+eLtDkuSQv/5s3vSJ/8zz9r1H7n3nW+ddF16RZ531uTzrrM9tkQOTSZ01vilff5NVrTfRPIRqvUm95mq9ieYhVOtNajZXe12r4hprnrxqvWt4bD9Z1XqTes3VepOazb5XTF7FZoBJGeT03K21g5K8NMm+vffLN+U2vfcv9N6P38jVPtF737v3/vAkj0ryZ621+/2Subd4n//cZ/OQh+2VJLnvnvfLpZdeMuWiDavWm2geQrXeJPnPr/asWLEiL37Bs3P0kX+UL11y0bSTNqjiGmuevN9+xL558cuOS5JcdeW3s/32O0w3aBNUW+OkXnO13kTzEKr1JpqHUK03Se66yy557SmnTTtjk1Vc4yrN/3zZ93P8hy5Lktxpx2X58Yq5/Nqdd8heu98+Zx76gBz3hHtn262XTrly3aqs8U35+pusar2J5iFU603qNVfrTTQPoVpvUrO52utaFddY8+RV613DY/vJqtab1Guu1pvUbPa9YvIqNgNMysRPz91ae0aS5yfZr/d+TWvtPknemGQmydVJ/jDJsUku6r2/s7W2c5IPJzk6yRG99wNba19NckGSluQ7SZ500/vpvV/XWntbkie31m6T5NVJViY5Pcn1SY5MslWS+SS/m+SMJCf13v+jtXZZkj/vvX+gtfaxJIcm2WZxp8xdklyc5LlJ7pLkLUmWJblTkmN67//QWvt6kj167ytaaycnuSzJO5P8VZLfSHJVkl9N8rjF3MNbay9OsmOS5/Te//2XW+XNs3z5ddlhh+1/ennpkqWZm5vL7OyWebb2ar2J5iFU602SbZYty1Oe+sw85glPyjev+EZe+qLn5J3v/1CWbqHNFddY8zBmZ2dzwitemvPPOzcnvvqUaedsVMU1rtZcrTfRPIRqvYnmIVTrTZJ9H7l/vv2tb047Y5NVXONKzatWz+eE3/217LPHTvmTv/1i7rjDNvnAZ7+dL1/54zzrt++WI/b+1bz+Y1+bdubPqbTGa/P1N1nVehPNQ6jWm9RrrtabaB5Ctd6kZnNS63WtimusefKq9a7hsf1kVetN6jVX601qNvteMXkVm7nlmpmZmXYCIzfpnSb3SvLsJLfLzwY0z0hyZO997yQfSfLiJG9P8szFjz8jyTtucpy7J3l57/23kuyU5IHrub/vJLnD4t+X9d736r2/K8m9kvxO7/1hSb6UZP8kH0zy6Nbarya5Icl+rbUdF2/3rSS3SvJnvfeHJrl9FgYe90jyut77Ixc/ryM38Lk/Psnte+8PSvJHWRi+XOOzvfd9kpyW5JANHGMitttu+yxfvvynl1fPr96ifwhW6000D6Fab5Lcdde7Zb8DHpuZmZnssuvdcusdb5Orr/7+tLPWq+Iaax7Oy4//i5z9gQ/n1Scem+uv/8m0czao4hpXa67Wm2geQrXeRPMQqvVWVHGNqzW//INfyuNPuzCvePweufA/f5AvX/njJMknvvy97HGnLXO3omprXFW1da7Wm2geQrXepF5ztd5E8xCq9SY1m9eo8rpWxTXWPHnVequqts7VepN6zdV6k5rN1VRc44rNAJMy6aHJK5M8MsmpSd7dWluS5N5J3txaOy8Lu0zepff+pSSzrbXdkvxBknff5Djf771fsfj3K7Kw0+O67JZkzX896Gu9/7tJ3tlae0eS+2Zhx8kPLbYdkIVdKR+U5NGL70+Sy3vv31j8+6eysMvllVnYJfJdSY5YPM5NrRmFvneSC5Ok9/69LOw+ucZnF/+8Ksm26/lcJub+939APnn++UmSiy/6Qnbf/V5DJ2yWar2J5iFU602Sj37og3nrG1+bJPn+976bnyy/Lre//R02cqvpqbjGmifvox/+x/zNmWckSZYtu1WWLFmSJTOTfjjxy6m2xkm95mq9ieYhVOtNNA+hWm9FFde4SvNj77tz/nCv3ZIkK25clfn55PUH3ie/fpdbJ0kefPfb5kvf/vE0E9eryhpXV22dq/UmmodQrTep11ytN9E8hGq9Sc3maq9rVVxjzZNXrbeqautcrTep11ytN6nZXE3FNa7YDDApkx4Z/1rvfUWSN7XW9k/ysiwMMx7ce7+8tfbQLJzmOkn+OslfJvlS7/2HrbW1jzO/sTtqrd06yWFJnrx4zNWL798xySuT7Lp41f+bZGbxVOE/ycKQ5u8t3u6oJE9fvN5dW2t36r1fmeRhi30nJDmj935Oa+3Q/GyXyBVJ7rR4mu77JflykkuysGvmqa2122Zht8tN/nwmaZ/9HpkLL7wgBz/twMzPz+f4E181zZyNqtabaB5Ctd4kefTjfy9/ecLLctSzD05mZvKnLzthiz01d1JzjTVP3sP32S+vOu6YPPdZB2dubi5HHf2SbLNsff+XYctQbY2Tes3VehPNQ6jWm2geQrXeiiqucZXmc7/83bzyib+WMw99QGaXLslfnvOVXHXtDXnJY+6VuVXzufq6lTn+Q1+eduY6VVnj6qqtc7XeRPMQqvUm9Zqr9Saah1CtN6nZXO11rYprrHnyqvVWVW2dq/Um9Zqr9SY1m6upuMYVmwEmZWZ+fjLze621vZMc0Xs/cPHyTkk+n+SULJzqejYLw4N/1Hv/Smtt2yTfTvL43vv5a9++tXZV733nxeOcneSti3fzt1k43faqxeO9off+gZvcdibJ+5PcLclckmuSfKr3flJr7TlJDu29P6i1dniS5/be91y8nyuSfDrJXRevf3Rr7aAkxyS5Ogs7Wu7Ze/8frbU/TPKnSb6e5EdJPprknUnelOT+WdhR8kFJHpzkpCRn994/2lo7IMmBvfdDNrSWK+amO2QJY/T9H6+cdsJmu8MOW087gS3QdSvmpp2w2bZftuUOEwOwZZtbVe+p0+zSmY1fiV/Kg084d9oJm+3fXr7vtBM2W7WvP197ALBl85oWTEe1x/WJx/YwLb5fsD7LZmOh2SRPf/dF9b6RMJh3P33PiX8vmdjQ5Ni11vZIcr/e+9mttdsnuTTJbr33Gzb3WIYmYXiGJrml8AIzAGPixVrWxdDkMKp9/fnaA4Atm9e0YDqqPa5PPLaHafH9gvUxNMmmMjTJhgwxNLlk0ncwYlckOai19uks7Dz5Z7/IwCQAAAAAAAAAAABw8/Df3iak9748yROm3QEAAAAAAAAAAAAssNMkAAAAAAAAAAAAMAqGJgEAAAAAAAAAAIBRcHpuAAAAAAAAAAAABjEzM+0Cxs5OkwAAAAAAAAAAAMAoGJoEAAAAAAAAAAAARsHQJAAAAAAAAAAAADAKhiYBAAAAAAAAAACAUTA0CQAAAAAAAAAAAIyCoUkAAAAAAAAAAABgFAxNAgAAAAAAAAAAAKNgaBIAAAAAAAAAAAAYBUOTAAAAAAAAAAAAwCgYmgQAAAAAAAAAAABGYXbaAQAAAAAAAAAAAIzDzMzMtBMYOTtNAgAAAAAAAAAAAKNgaBIAAAAAAAAAAAAYBUOTAAAAAAAAAAAAwCgYmgQAAAAAAAAAAABGwdAkAAAAAAAAAAAAMAqGJgEAAAAAAAAAAIBRMDQJAAAAAAAAAAAAjMLstAOAcfjhT26cdsJmucMOW087YbOtuHHVtBM227Ktlk474RZv+2V+1PPz5lbNTzths80unZl2wmaxxjAd/h2zLv/28n2nnbDZfv0l50w7YbNdcvKjp50AAGxAtddnb7PtVtNOgFGq+Lza63AwHf4dA1CdnSYBAAAAAAAAAACAUTA0CQAAAAAAAAAAAIyCc3YCAAAAAAAAAAAwiCUz0y5g7Ow0CQAAAAAAAAAAAIyCoUkAAAAAAAAAAABgFAxNAgAAAAAAAAAAAKNgaBIAAAAAAAAAAAAYBUOTAAAAAAAAAAAAwCgYmgQAAAAAAAAAAABGwdAkAAAAAAAAAAAAMAqGJgEAAAAAAAAAAIBRMDQJAAAAAAAAAAAAjMLstAMAAAAAAAAAAAAYh5mZmWknMHJ2mgQAAAAAAAAAAABGwdAkAAAAAAAAAAAAMAqGJgEAAAAAAAAAAIBRMDQJAAAAAAAAAAAAjIKhSQAAAAAAAAAAAGAUDE0CAAAAAAAAAAAAo2BoEgAAAAAAAAAAABgFQ5MAAAAAAAAAAADAKBiaBAAAAAAAAAAAAEZhdtoBDG/16tU56YTj8pXes/XWW+fYV56YXXfbbdpZ61WtN9E8pGt+cHUOe8ZT8rq/OiO73e3u087ZoGprPHfjjTnhuGNy5be/lRtX3phDDzs8v733PtPO2qBqa5zUa67Wm2ge0hcvviinnfranH7mu6adslHWePIqrrHmyavWm9RrrtabaJ6kJTPJSb9/n9x9p+0yn/m8/O8vzZH73iM73XqbJMldbnurfOEbP8wL33PRlEt/XpU1Xlu15mq9ieYhVOtN6jVX6000D6Fa79q8Pjs51XoTzUOo1pvUbE68Djdp1Zqr9Saah1CtN6nZDDApW8xOk621vVtrZ691+cmttUtaa7tu4u3v11p7xUaOP99aO/Am77+4tXbWL9j8ktbag36R207TJ879eFbesDLveu/7c9SLjs7rXnPytJM2qFpvonkoc3M35rV/8cpss2zZtFM2SbU1PufDH8qOO94mp7/9xboyAAAgAElEQVTj3Tn1zW/La08+cdpJG1VtjZN6zdV6E81DeeeZb88Jxx2TG25YOe2UTWKNJ6/iGmuevGq9Sb3mar2J5kna59fumCT5g7/6dE756Fdz9AH3ygvfc1Ge9pZ/z3PO+lyuvX4uJ/3jZVOuXLcqa7y2as3VehPNQ6jWm9RrrtabaB5Ctd41vD47WdV6E81DqNab1Gz2OtzkVWuu1ptoHkK13qRmM7dcM968beBtCFvM0OTaWmsHJXlpkn1775dvym1671/ovR+/katdluSnQ5Ottfsk2e4X7ey9n9x7//df9PbT8vnPfTYPedheSZL77nm/XHrpJVMu2rBqvYnmobz51NfmCb/3B7nDHXaadsomqbbG+z5q/xx+5AsWLswnS5du+ZsTV1vjpF5ztd5E81Duussuee0pp007Y5NZ48mruMaaJ69ab1KvuVpvonmSPn7pd3PM/1pou/Ntb5VrV9z4048d9ajd864LvpHv/fiGaeVtUJU1Xlu15mq9ieYhVOtN6jVX6000D6Fa7xpen52sar2J5iFU601qNnsdbvKqNVfrTTQPoVpvUrMZYFK2uAmY1tozkjw/yX6992sWBxvfmIVB0quT/GGSY5Nc1Ht/Z2tt5yQfTnJ0kiN67we21r6a5IIkLcl3kjxp8fAXLdxF27H3/qMkT0/yniS7Lt7305K8MMkNSb6a5NlJnrZ4n0sW7/fMLAxffinJbZOcneS/krwjydzi9Z7ae7+itfa6JA9bvO/39t7fsLir5Q1J7pbkTkkO6b1/7uZbwY1bvvy67LDD9j+9vHTJ0szNzWV2dov755CkXm+ieQjnfOgfcpvb3i4P+q2H5j1nnTHtnE1SbY233XZhpnz58uV5yZ+8MEesGaDcglVb46Rec7XeRPNQ9n3k/vn2t7457YxNZo0nr+Iaa568ar1JveZqvYnmSVu1ej5/eeB98qhf3znP+5vPJ0lut/3Wecjut/9/7N15mKVVee/vb3UX2DIIKvhzIqhHXEoSHHIcwlEhDCLORo04BgccCM4mwYjKpMGIgsEYQGxR0aCZNSYmRwkSFU0iIIqwCMYpAg44oC1DV3edP6pa62d6jOx398N739fVV/eu2sOnFm/X3vX2w9p5/YcunXLdhlVa43WqNVfrTTQPoVpvUq+5Wm+ieQjVehPnZ4dQrTfRPIRqvUnNZufhJq9ac7XeRPMQqvUmNZsBJmVr22nyIVkYVLxNfjbQ+Y4kv9N73zfJ3yf5vSRnJPntxc8/IwsDi0vdLclreu+/nmTXJPdf8rm/TPKbrbWZJA9I8ukkaa3dNskxSfbrvT84yQ+SPH/xNt/vvT+49/7xJLtlYSjyZUvu88Ak/5rkgCwMVu7UWntUkrsmeVAWBiefujgAmiRf670flOSUxa93UNtvv0NWrVr108tr59du1U+C1XoTzUP4+w//df79s+fnJc8/NFdc3vOG1/1Brvnud6edtVHV1jhJvnX1VTn8sENz8KMenYMe8ahp52xSxTWu1lytN9HM+lnjyau4xponr1pvUq+5Wm+ieQi/d/YXcsAbP5HXP+lXcsttl+fgvW6fD114VdbOT7tsw6qtcVKvuVpvonkI1XqTes3VehPNQ6jWmzg/O4RqvYnmIVTrTWo2V1Nxjas1V+tNNA+hWm9SsxlgUra2ocmrsjCAeHKSs1pry5LcK8nbW2vnZmHHxzv13r+UZLa1tnuSJyc56+fu57u9928s/vkbSVYs+dz7s/AW3Q9N8i9LPn63JJf03n+0ePm8JL+8+Of+c/d9zc893juzMGT50SRHZGHHyXsl+Zfe+3zvfXWSzyTZc/H6F26gbRD3ve/98snzzkuSXPz5i7LHHvcYOmGLVOtNNA/hlNPfnT8+/cy89bQzc/d7tPzBMW/IbXfZZdpZG1Vtja+55rt58QsPyxEveXke87gnbPoGW4Fqa5zUa67Wm2hm/azx5FVcY82TV603qddcrTfRPEmPu98d84L97pYkuf7GtVk7P5+1a+ez9x63zXmXfWfKdRtXZY2XqtZcrTfRPIRqvUm95mq9ieYhVOtNnJ8dQrXeRPMQqvUmNZurqbjG1Zqr9Saah1CtN6nZDDApW9vI+BW99+uTvK21dlCSV2dhYPGZvfevt9b+Txbe0jpZGFT8oyRf6r3/oLW29H42uFdC7/0/W2vbJ3lxkldlYVgySb6SZM/W2va991VJ9kly+eLn1i65i6V/XuexWRiQPKa19pQkv5+FHS2fleSk1to2SfZO8u4kB2+sbwj7HXBgzj//U3nm0w7J/Px8jj3+DdPM2aRqvYlm1q/aGp95xum59tofZuXpp2bl6acmSU76k9OyYsXgs96brdoaJ/Waq/Ummlk/azx5FddY8+RV603qNVfrTTRP0j9+8Vt545N/Ne8//IHZZtlMXv+3l+aGubW5267b5+vX/GTaeRtVZY2XqtZcrTfRPIRqvUm95mq9ieYhVOutqto6V+tNNA+hWm9Ss7maimtcrblab6J5CNV6k5rNAJMyMz+/dbwXU2tt3yQv6L0fsnh51yzsyHhSkkdnYcBzPslzeu+Xt9a2S3Jlksf03s9bevvW2tW999sv3s/ZSU5dfJh1n39Rkmf03h/QWnt4kkN674e21p6a5KVZGIy8Islzs7Ar5T1770cu3t/S+z4zydlJ/iMLA5E3Jlme5GW99wtaaycm+T9Jtk3ywd77G9fdpvf+0aWPvbG1uX5uukOWcFP4wU9WTzthi+y83TbTTthi169eM+2ELbZim+XTToBRmltT76XF7PKZaSdsEWsMwC/iV478h2knbLEvnnDwtBMAgI1wfha4uXIeDmDrsmI2vsmxWZ599hfqPYkzmJWH/OrEv5dsNUOTbJihSW4OnJSbPEOTwOZyInHyrDEAvwhDkwDATc35WeDmynk4gK2LoUk2l6FJNmaIocllk34AAAAAAAAAAAAAgK2BoUkAAAAAAAAAAABgFGanHQAAAAAAAAAAAMA4LJvxTu5Ml50mAQAAAAAAAAAAgFEwNAkAAAAAAAAAAACMgqFJAAAAAAAAAAAAYBQMTQIAAAAAAAAAAACjYGgSAAAAAAAAAAAAGAVDkwAAAAAAAAAAAMAoGJoEAAAAAAAAAAAARsHQJAAAAAAAAAAAADAKhiYBAAAAAAAAAACAUTA0CQAAAAAAAAAAAIzC7LQDAAAAAAAAAAAAGIeZmWkXMHZ2mgQAAAAAAAAAAABGwdAkAAAAAAAAAAAAMAqGJgEAAAAAAAAAAIBRMDQJAAAAAAAAAAAAjIKhSQAAAAAAAAAAAGAUDE0CAAAAAAAAAAAAo2BoEgAAAAAAAAAAABgFQ5MAAAAAAAAAAADAKMxOO4BNu371mmknbJEV2yyfdgJboZ2322baCTd7/u5N3tya+WknbLHZ5TPTTmAr5LiYPGsM3Jx97bs/mXbCFtl9l+2mnbDFvnjCwdNO2GIPPO7j007YIp99zf7TToCbRLWfU71OhulxfhbYHNVeWzCMiseF150AwKbYaRIAAAAAAAAAAAAYBTtNAgAAAAAAAAAAMIiZGbsCM112mgQAAAAAAAAAAABGwdAkAAAAAAAAAAAAMAqGJgEAAAAAAAAAAIBRMDQJAAAAAAAAAAAAjIKhSQAAAAAAAAAAAGAUDE0CAAAAAAAAAAAAo2BoEgAAAAAAAAAAABgFQ5MAAAAAAAAAAADAKBiaBAAAAAAAAAAAAEZhdtoBAAAAAAAAAAAAjMPMzLQLGDs7TQIAAAAAAAAAAACjYGgSAAAAAAAAAAAAGAVDkwAAAAAAAAAAAMAoGJoEAAAAAAAAAAAARsHQJAAAAAAAAAAAADAKhiYBAAAAAAAAAACAUTA0CQAAAAAAAAAAAIyCoUkAAAAAAAAAAABgFAxNAgAAAAAAAAAAAKNgaBIAAAAAAAAAAAAYhdlpBzC8udWrc9zRR+WqK7+Z1TeuzrMOe34euu9+087aoLVr1+b1xx2dy3vPtttum9cdc3x+affdp521UZonr1pvUq+5Wm9SszlJvnDx53PKySfm9JXvnXbKJlVcY82TV6030TyEar2J5iFU603qNVfrTZK5udU5+Q2vzbevvjLLli3PEb/7mtx597tOO2ujqq1zld5lM8nrHnOv7L7Ldsl8cvzfXZbZZTM55Wn3zteuuS5J8uf/9l/5x0u+PeXS9auyzutU6000D8nPqZNTrTfRPIRqvYnmIVTrTTQPoVrvUpVeXyS1eqseF9Z4sjRPXrXepGYzN1/LZmamncDIldppsrW2b2vt7CWXn9ha+2JrbWVr7Zc28z4e31q7Y2vt9q21t2/kendprV3bWjt38ddnWmsfa63dekO3ba2d0Fo79Oc7tzb/8JEPZ6edds7p7zorJ7/9tJx4wvHTTtqocz7+sdx4w4157/s/kJe87BV585tOmHbSJmmevGq9Sb3mar1JzeZ3rzwjxx19VG644cZpp2yWimusefKq9Saah1CtN9E8hGq9Sb3mar1J8u+f+WTWrFmTP3r7u/Pk335e3nvG26adtEnV1rlK7z5tlyTJoe/8XN52zpdzxP7/K/e6463y3vO/keeeeUGee+YFW+3AZFJnndep1ptoHoqfUyerWm+ieQjVehPNQ6jWm2geQrXedaq9vqjWW/G4sMaTp3nyqvUmNZsBJqXU0ORSrbWnJHlVkv1778/uvX99M2/6kiS36r1f3Xs/fBPX/VLvfd/FXw9K8m9JnrOZt91q7f+wg/L833nxwoX5ZPnyrXvD0Qsv+Fz2fvBDkiR73fs+ueSSL065aNM0T1613qRec7XepGbznXfbLSeedMq0MzZbxTXWPHnVehPNQ6jWm2geQrXepF5ztd4kudOdd8+aNWuydu3aXPeTH2d2duv+GTWpt85Vev/5su/m2A9fliS5w04r8qPr57LnHXfMQ/a4bVY+6345+rH3ynbbLp9y5YZVWed1qvUmmofi59TJqtabaB5Ctd5E8xCq9Saah1Ctd51qry+q9VY8Lqzx5GmevGq9Sc1mgEnZ+v8lYj1aa89I8qIkB/Tev99aOzfJC5IckmTvJDskeU6SNybZKcl2SV6dZJsk90nyntba05O8p/f+oNbaxUk+kWSvJPNJHruex5xJsluSK1prd0ly9uJtn5DkqCTfSbJtkst+7nZPSvLyJGuSfLL3fmRr7VNJntd7v6S1dnCSRw85hLnddtsnSVatWpUjX/nSvGDdAOVWatWqH2fHHXf46eXly5Znbm5uq/6HNM2TV603qddcrTep2bz/gQflym/+17QzNlvFNdY8edV6E81DqNabaB5Ctd6kXnO13iRZsd12+fbVV+bwZzw+1/7wB3nNCW+ddtImVVvnSr1r1s7nuMfvmf3uuWte+cEv5HY73iJ/9bkrc+lVP8pzH3qXvGDfu+Yt/3TFtDPXq9I6J/V6E81D8XPqZFXrTTQPoVpvonkI1XoTzUOo1rtOtdcX1XorHhfWePI0T1613qRmM8CkVNxp8iFJnpfkNln/0Oelvfe9s/C17ZLk0UmekmS29/6RJBcleWaSpXt93yrJn/Xe90nyzSQHL358z8W35r44yeVJrkjy7nU3aq1tk+QtSQ5IclCSnywNaa3dJskxWdgN88FJ7tRaOzDJGUl+e/Fqz168PKhvXX1VDj/s0Bz8qEfnoEc8auiH3yLbb79DVq1a9dPLa+fXbvVP2ponr1pvUq+5Wm9Ss7maimusefKq9Saah1CtN9E8hGq9Sb3mar1J8qEPnpX7PuDXc+r7/jZvXfmBnPyG1+bGG26YdtZGVVvnar2v+esv5TGnnJ/XPuaeOf/L38ulV/0oSXLOpd/JPe+w45TrNqzaOlfrTTSzftXWuFpvonkI1XoTzUOo1ptoHkK1XobhuJi8imusefKq9SY1mwEmpeLQ5FVJDkxycpKzWms//zX0JOm9X5LktCR/luTt2fTXeuHi799IsmLxz1/qve+b5IFJvpbkW733uSW32TXJ93rv1/Te55N8+ufu8+6L1/n7xd0w90zyv5J8MMljWmu3S3Ln3vsFm/qib0rXXPPdvPiFh+WIl7w8j3ncE4Z86P+R+973fvnkeeclSS7+/EXZY497TLlo0zRPXrXepF5ztd6kZnM1FddY8+RV6000D6Fab6J5CNV6k3rN1XqTZPsdb5Xtt1/4P9x33HGnrFkzl7Vr1065auOqrXOV3kftdfs8+yG7J0muX70m8/PJWw751fzKnW6VJHng3W6dL135o2kmblSVdV6nWm+imfWrtsbVehPNQ6jWm2geQrXeRPMQqvUyDMfF5FVcY82TV603qdkMMCkVR8av6L1fn+RtrbWDsvC220utTZLW2q8m2bH3/sjW2h2yMND4d4ufX98A5fyGHrD3fl1r7WlJLmqtfTrJDxc/9e0kO7fWdu29fyfJ/ZMs3Uf8K1kYwjyw9766tXZokot676taa/+c5K1JztqSL/6mcOYZp+faa3+YlaefmpWnn5okOelPTsuKFSs2ccvp2O+AA3P++Z/KM592SObn53Ps8W+YdtImaZ68ar1JveZqvUnN5moqrrHmyavWm2geQrXeRPMQqvUm9Zqr9SbJY5/09PzxG4/OkUc8O3Nzq/OM574oK255y2lnbVS1da7S+/FLv51jHrdnVj7rfpldvix/9A+X5+prb8iRj7hH5tbM55of35hjP3zptDM3qMo6r1OtN9HM+lVb42q9ieYhVOtNNA+hWm+ieQjVehmG42LyKq6x5smr1pvUbAaYlJn5+Q3OCm51Wmv7JnlB7/2Qxcu7ZmGHyC8neX6SQ5Jc3Xs/tbW2IgsDibfLwpDkab3397bWjk/y8Cy8xffbe+8Paq19Nck9e+/Xt9ZOSHJZknOTnN17f9CSx39akhcleWqS9y/e9pFJjkvyvSSrk3wgyVfXdbbWnp7k8CTLFz/+rN77T1pr90vyySR37L3/YGNf9w+uW1PnP1KSFdssn3YCwETM1fp2nCSZXT4z7QQA4Gbma9/9ybQTtsjuu2w37YRReOBxH592whb57Gv2n3YC3CSq/ZzqZ1QA2LpVe21RVbXXRBWPi2prDDcnK2bjLyCb5YV/+aV6TzAM5k+fsOfEv5eUGpq8OWmt3T/Ji3rvz9zUdQ1NAmwdnBgAADA0yfoZmoTpqPZzqp9RAWDrVu21RVXVXhNVPC6qrTHcnBiaZHMZmmRjhhiarPj23OW11o5I8pwkvzXtFgAAAAAAAAAAABgLQ5NT0Ht/W5K3TbsDAAAAAAAAAAAAxsTQJAAAAAAAAAAAAIOY8UbuTNmyaQcAAAAAAAAAAAAADMHQJAAAAAAAAAAAADAKhiYBAAAAAAAAAACAUTA0CQAAAAAAAAAAAIyCoUkAAAAAAAAAAABgFAxNAgAAAAAAAAAAAKNgaBIAAAAAAAAAAAAYBUOTAAAAAAAAAAAAwCgYmgQAAAAAAAAAAABGYXbaAQAAAAAAAAAAAIzDzMzMtBMYOTtNAgAAAAAAAAAAAKNgaBIAAAAAAAAAAAAYBUOTAAAAAAAAAAAAwCgYmgQAAAAAAAAAAABGwdAkAAAAAAAAAAAAMAqGJgEAAAAAAAAAAIBRMDQJAAAAAAAAAAAAjIKhSQAAAAAAAAAAAGAUDE0CAAAAAAAAAAAAo2BoEgAAAAAAAAAAABiF2WkHsGkrtlk+7QQAkswun5l2whabWzM/7YQtVnGdAWBMdt9lu2knsBX67Gv2n3bCFrn1Pq+edsIW+/4nXj/tBLZCfn4CNle1c0S+v8F0+LvH+jguAJgEu/wxbY5BAAAAAAAAAAAAYBQMTQIAAAAAAAAAAACjYGgSAAAAAAAAAAAAGAVDkwAAAAAAAAAAAMAoGJoEAAAAAAAAAAAARsHQJAAAAAAAAAAAADAKhiYBAAAAAAAAAACAUTA0CQAAAAAAAAAAAIyCoUkAAAAAAAAAAABgFGanHQAAAAAAAAAAAMA4zMzMTDuBkbPTJAAAAAAAAAAAADAKhiYBAAAAAAAAAACAUTA0CQAAAAAAAAAAAIyCoUkAAAAAAAAAAABgFAxNAgAAAAAAAAAAAKNgaBIAAAAAAAAAAAAYBUOTAAAAAAAAAAAAwCgYmgQAAAAAAAAAAABGwdAkAAAAAAAAAAAAMAqGJgEAAAAAAAAAAIBRmJ12AAAAAAAAAAAAAOOwbGbaBYydockRWrt2bV5/3NG5vPdsu+22ed0xx+eXdt992lkbVK030TyEar1JveZqvYnmIX3h4s/nlJNPzOkr3zvtlE2quMbVmqv1JpqHUK030TyEar1JveZqvYnmIVTrTeo177rz9vn0ysPzyJe+KytusU1O+d3HZG5ubf7jG9fkhSf8debn56ed+N9UW+NE8xCq9Sb1mqv1JpqHUK13KeeIJqdab6J5CNV6E81DqNab1Guu1ptoHkK13qRmM8CkbPLtuVtr+7bWzl5y+YmttS+21n5psmlJa+3Q1toJ6/n4ua21ey7+eYfW2idaa0du5n3ep7X22s287pmttYdv6mPVnPPxj+XGG27Me9//gbzkZa/Im9/035Z4q1KtN9E8hGq9Sb3mar2J5qG8e+UZOe7oo3LDDTdOO2WzVFzjas3VehPNQ6jWm2geQrXepF5ztd5E8xCq9Sa1mmeXL8vbfu9xue6GuSTJq5+1X97wrn/O/oe/I7fYdnkO3rtNuXD9Kq3xOponr1pvUq+5Wm+ieQjVetdxjmiyqvUmmodQrTfRPIRqvUm95mq9ieYhVOtNajYDTMomhyaXaq09Jcmrkuzfe//6ZJK2qOdWST6a5AO99836bt57v6j3fuxky7ZuF17wuez94IckSfa6931yySVfnHLRxlXrTTQPoVpvUq+5Wm+ieSh33m23nHjSKdPO2GwV17hac7XeRPMQqvUmmodQrTep11ytN9E8hGq9Sa3mE444OO/4m8/mqu9emyS56D+uzK13vGWSZIftbpHVc2ummbdBldZ4Hc2TV603qddcrTfRPIRqves4RzRZ1XoTzUOo1ptoHkK13qRec7XeRPMQqvUmNZsBJmWz3567tfaMJC9KckDv/futtfsmOSXJmiTXJzksC0OYZ/feH7R4m88kOSTJoUn2TrJDkuckecTix+eSnNd7//3W2v9J8uYkq5P8JMkTFx/6Qa21f0qya5I/7b2fvvjxnZO8O8nbe+/vXny85UlOS7Jbkjsk+VDv/ajW2plJbrv4601Jntx7P6S1dkWSTye5R5KPJ9kpyQOS9N77M5Z87Q9M8sdJnrT4oee31n5v8fov7L3/a2vtD5P878XH+Hzv/VmttaOT3DXJ7ZLsnuRlSb6S5Kze+wMW7/sDSd7ce//Xzf1v8YtaterH2XHHHX56efmy5Zmbm8vs7Nb5bu3VehPNQ6jWm9RrrtabaB7K/gcelCu/+V/TzthsFde4WnO13kTzEKr1JpqHUK03qddcrTfRPIRqvUmd5qc/4r75zg9W5WP/ekV+9xn7JEm+/I1rctIrHp0jD/2NXPvj63PehV+ZcuX6VVnjpTRPXrXepF5ztd5E8xCq9a7jHNFkVetNNA+hWm+ieQjVepN6zdV6E81DqNab1GwGmJTN3WnyIUmel+Q2+dmg5TuSHNF73yfJ25O8ZRP3cWnvfe/F2/9WFoYo906yR2vtUUkel+SDSfZJ8qdJbr14u9VJDkry+CQvXXJ/Zy1+7k5LPrZbks/03g/KwvDjC5Z87pzFx//+ko/dJclRi1/fixe/jgcmeXBrbefF6+y9+LU9esnump/rve+XhaHRQxd3vPx+7/3ALAxOPqi1tq7rht77wUlekuRlvffLk1zXWtuztXabJHcdcmAySbbffoesWrXqp5fXzq/dqp8Eq/UmmodQrTep11ytN9HM+lVc42rN1XoTzUOo1ptoHkK13qRec7XeRPMQqvUmdZp/+5G/lv3vf/f84ynPyV573CHvfM2T8o6jnpADDn9H7vPUk/O+j16YE444eNqZ61VljZfSPHnVepN6zdV6E81DqNZbVbV1rtabaB5Ctd5E8xCq9Sb1mqv1JpqHUK03qdkMMCmbOzR5VZIDk5yc5KzW2rIkd+y9X7T4+fOS/PJ6bjez5M998fd7ZmGwcXXvfT7Jvyze9g1J7piFHR+fmIWByCS5YPF6VyfZbsn9HZmFHSsPba3ts/ix7yW5f2vtfUlOSnKL9Tz+Utf03r/ee1+dZFXv/UuLj/XDJCsWr/OwLOxquXrJ7T63+Pu6puuS3K619mdZ2OlyhyTbLF7nwsXfv7HkPt+Rhd03n5qF4c9B3fe+98snzzsvSXLx5y/KHnvcY+iELVKtN9E8hGq9Sb3mar2JZtav4hpXa67Wm2geQrXeRPMQqvUm9Zqr9Saah1CtN6nTfODvnJGHHXFGDnrRO3Pxf1yV5xz35/nKld/Pj1bdkCS56rs/+ulbdW9tqqzxUponr1pvUq+5Wm+ieQjVequqts7VehPNQ6jWm2geQrXepF5ztd5E8xCq9SY1mwEmZXNHxq/ovV+f5G2ttYOSvDrJla21vXrvF2dhd8jLs/A23bdbfJvsHbPw1tTrrF38/bIkr2itzWbhrb0fmuQ9SZ6e5Mze+ytba6/Kws6WX0syv4GmL/ber22tPTPJB1tr/zsLb/n9g97781trd0/yvNbausHNteu5jw3d91JHJ7lzFnahfMoGbndwkt16709ure2ahV0xZzZw3ST5iySvTHJNfvaW34PZ74ADc/75n8ozn3ZI5ufnc+zxbxg6YYtU6000D6Fab1KvuVpvopn1q7jG1Zqr9Saah1CtN9E8hGq9Sb3mar2J5iFU601qNq9z+Al/nfcc8+TMrVmbG+fW5PAT/nraSetVcY01T1613qRec7XeRPMQqvVWVW2dq/UmmodQrTfRPIRqvUm95mq9ieYhVOtNajYDTMrM/PzG5wZba/smeUHv/ZDFy7tmYffEN/DbPUwAACAASURBVOdnw4FzSZ7Te//P1tppSe6f5MtZeOvsp2ZhV8Wre++nLt7Hy5M8OQs7XX4yycuz8Hbab02yKgsDjs/LwjDmPXvvR7bWViS5rPd+l9bauYtNly3e31FJ9s3CW2C/Pws7Rd6Qhbfr3j/J65Oc3Xv/6NKvp7V2de/99ov3sfTPFyV5eJITltzuH5O8Ows7T6772MOzMKh5ZJIPZ2HHyfkkt0zysizsznl17/3U1to9k5zae9938TH+OMmuvfd1g5gbdP3cZg13AsB/M7em3lPI7PKZTV8JAAB+Abfe59XTTthi3//E66edAEBh1c4ROT8EAFDTitl4IcdmeenfXlbrhxQGdfJj7znx7yWbHJpkMlprf5LkL3vv52zquoYmAfifqnZCPHFSHACAyTM0CcDYVDtH5PwQAEBNhibZXIYm2ZghhiaXTfoB+O9aa/+U5NabMzAJAAAAAAAAAAAA3DRmpx0wRr33h027AQAAAAAAAAAAYGjL7EnKlNlpEgAAAAAAAAAAABgFQ5MAAAAAAAAAAADAKBiaBAAAAAAAAAAAAEbB0CQAAAAAAAAAAAAwCoYmAQAAAAAAAAAAgFEwNAkAAAAAAAAAAACMgqFJAAAAAAAAAAAAYBQMTQIAAAAAAAAAAACjYGgSAAAAAAAAAAAAGIXZaQcAAAAAAAAAAAAwDjMzM9NOYOTsNAkAAAAAAAAAAACMgqFJAAAAAAAAAAAAYBQMTQIAAAAAAAAAAACjYGgSAAAAAAAAAAAAGAVDkwAAAAAAAAAAAMAoGJoEAAAAAAAAAAAARsHQJAAAAAAAAAAAADAKhiYBAAAAAAAAAACAUTA0CQAAAAAAAAAAAIyCoUkAAAAAAAAAAABgFGanHQAAAAAAAAAAAMA4LJuZdgFjZ2gSYD2uX71m2glbbHZZvc2DZ5d7JTRpc2vXTjthi80uXz7thC0yt2Z+2glbrOL3uB1WeNkKANx0vnPO8dNO2GL3e+0/TTthi1xw7MOmnQDAEtXOw1U831JtjSuqeE5rxTa1znUCbC7P1QBUV2/CBgAAAAAAAAAAAOB/wNAkAAAAAAAAAAAAMAqGJgEAAAAAAAAAAIBRMDQJAAAAAAAAAAAAjIKhSQAAAAAAAAAAAGAUDE0CAAAAAAAAAAAAo2BoEgAAAAAAAAAAABgFQ5MAAAAAAAAAAADAKBiaBAAAAAAAAAAAAEZhdtoBAAAAAAAAAAAAjMPMzLQLGDs7TQIAAAAAAAAAAACjYGgSAAAAAAAAAAAAGAVDkwAAAAAAAAAAAMAoGJoEAAAAAAAAAAAARsHQJAAAAAAAAAAAADAKhiYBAAAAAAAAAACAUTA0CQAAAAAAAAAAAIyCoUkAAAAAAAAAAABgFAxNAgAAAAAAAAAAAKMwO+0AAAAAAAAAAAAAxmHZzMy0Exg5O00CAAAAAAAAAAAAo2BoEgAAAAAAAAAAABgFb889QmvXrs3rjzs6l/eebbfdNq875vj80u67Tztrg6r1JpqHUK03SeZWr85xRx+Vq678ZlbfuDrPOuz5eei++007a5O+cPHnc8rJJ+b0le+ddsomVTwuqjVXPI6rrfE6lf7urVmzJm88/nX5+te+kpnM5Hf/4HW52933mHbWRlU8Lqo1V+tNNA+hWm9Sr7lab6J5CNV6k5rNydb/Gm7ZTHLs4385d9l1u8zPJ8f87aX53o9vzLGP3zO3uuU2WbYsedWffzHf+N51005dr4rHRbXmar1JveZqvYnmIVTrTWo2J1v/c/VSFde4WrNzncPQPHnVepN6zdV6k5rNiefqSavYDDApN+udJltr+7bWzt7C25zQWju0tXaf1tprN3K9fVpr5y75dUVr7bMbuf6RrbUHbEnLpJzz8Y/lxhtuzHvf/4G85GWvyJvfdMK0kzaqWm+ieQjVepPkHz7y4ey00845/V1n5eS3n5YTTzh+2kmb9O6VZ+S4o4/KDTfcOO2UzVLxuKjWXPE4rrbGSb2/e58679wkyakr35fDDn9xTnv7W6cbtBkqHhfVmqv1JpqHUK03qddcrTfRPIRqvUnN5gqv4X7jnrsmSZ5+2r/lj//vFXnpgXfPKx6+R/7u81flme9Y+Nhdd91+ypUbVvG4qNZcrTep11ytN9E8hGq9Sc3mCs/VS1Vc42rNznUOQ/PkVetN6jVX601qNnuunryKzQCTcrMemvxF9N4v6r0fu5HPf6L3vm/vfd8kT04yn+TlG7n+Cb33f73pS7fchRd8Lns/+CFJkr3ufZ9ccskXp1y0cdV6E81DqNabJPs/7KA8/3devHBhPlm+fOvf7PfOu+2WE086ZdoZm63icVGtueJxXG2Nk3p/9x76G/vn9159dJLk6quuzA477DjdoM1Q8bio1lytN9E8hGq9Sb3mar2J5iFU601qNld4DffxS7+T1/3Nl5Ikd9z5lrn2+rncb/ed8//dakXe+exfy6PufYf8239+f8qVG1bxuKjWXK03qddcrTfRPIRqvUnN5grP1UtVXONqzc51DkPz5FXrTeo1V+tNajZ7rp68is0Ak7L1v/q/CbXW7pHkrN77AxYvfyDJm5PsluSoJN9Jsm2Sy1pr+yZ5Qe/9kNbaEUl+M8n2Sb6b5PG99xsX72ObJH+R5E2990+11pYnOW3xPu+Q5EO996Naa2cmOTvJJ5K8K8nui491RO/9/CG+/nVWrfpxdtxxh59eXr5seebm5jI7u3UeDtV6E81DqNabJNttt7Bbx6pVq3LkK1+aF6w7GbMV2//Ag3LlN/9r2hmbreJxUa254nFcbY2Ten/3kmR2djbHvfZVOe/cj+f4N5407ZxNqnhcVGuu1ptoHkK13qRec7XeRPMQqvUmNZurvIZbs3Y+b3jir+SAPW+Xl77/83nkvW+fa69bnees/FxeuN/d8px97pK3fezL085cr4rHRbXmar1JveZqvYnmIVTrTWo2V3muXqfiGldrdq5zGJonr1pvUq+5Wm9Ss9lz9eRVbAaYlFHtNNl7vzzJda21PVtrt0ly1yQXJnlLkgOSHJTkJ0tv01pbluS2SQ7ovT8wC4Om919ylbcmuaT3fvri5d2SfKb3flCSByR5wc9lvCDJV3vvv57kkCQPvAm/xM2y/fY7ZNWqVT+9vHZ+7Vb9JFitN9E8hGq963zr6qty+GGH5uBHPToHPeJR08652al4XFRsrnYcV1zjql5z7B/m7L/6SN54/Oty3XU/2fQNpqjicVGtuVpvonkI1XqTes3VehPNQ6jWm9RsruQP/uKLecRbPpljH79nfnT9XM659DtJknMv/U5+5U63mnLdhlU8Lqo1V+tN6jVX6000D6Fab1KzuZqKa1yx2bnOydM8edV6k3rN1XqTms3VVFzjis0AkzKqoclF70hyaJKnJjkrya5Jvtd7v6b3Pp/k00uv3Htfm+TGJH/WWntnkjsn2SZJWmvPSvKrSV605CbfS3L/1tr7kpyU5BY/9/gtyfmL9/0fvfeTb9KvbjPc9773yyfPOy9JcvHnL8oee9xj6IQtUq030TyEar1Jcs01382LX3hYjnjJy/OYxz1h2jk3SxWPi2rNFY/jamtc0Uc/8qG8Z+U7kiQrVtwyy5Yty7KZrftlZsXjolpztd5E8xCq9Sb1mqv1JpqHUK03qdlcwaPvc4ccts9dkyTXrV6TtfPJv3/l+3lo2yVJ8mt3vXWu+Naqjd3FVFU8Lqo1V+tN6jVX6000D6Fab1KzuZqKa1yt2bnOYWievGq9Sb3mar1JzeZqKq5xxWaASRnjyPhfJHllkmuSPCnJtUl2bq3t2nv/ThZ2kfzpns+ttb2SPK73/sDW2nZJPpdkprV2/yR/kOTBvffVS+7/0CQ/6L0/v7V29yTPa63NLPn8pYuP8bettbslOb73/tRJfbHrs98BB+b88z+VZz7tkMzPz+fY498w5MNvsWq9ieYhVOtNkjPPOD3XXvvDrDz91Kw8/dQkyUl/clpWrFgx5bKbj4rHRbXmisdxtTWuaJ/9Dsgbjj4qhz/3mZmbm8tLXnFkbrEVHxNJzeOiWnO13kTzEKr1JvWaq/UmmodQrTep2VzBxy75dl7/xF/Oew67f2aXz+SEj1yWy676UY59/C/nkAfulh9fP5ff/cDF087coIrHRbXmar1JveZqvYnmIVTrTWo2V1Nxjas1O9c5DM2TV603qddcrTep2VxNxTWu2AwwKTPz8/PTbpiY1tq+Sf4qyX8u+fBTkxyRZNfe+1MWr/fIJMdlYZfI1Uk+kOSrWXgr7Wcn+bv8bMfIG5K8MwvDkXdL8o0l9/3jJL+f5P1Jfrh43d2S7J/k9UnOTnJukpVJ7pRkeZKX9t7/fWNfx/Vzufn+R4Kt1PWr10w7YYvNLtu6d3Vbn9nlM5u+Er+Qisfyim2WTzthi8ytqfc0XfG42GHFGP9fHwBgUiq+hnvAMf932glb5IJjHzbtBAAKq/hc7Vzn5FU8p1XtXCfA5vJczYasmI2FZrMc+feX1/tGwmBOeMQ9Jv695Gb9r8+993OT3ObnP95aW56Ft+led72PJPnIeu7i3MXf91vP5963kYe+93oec5skN/ber8/C4CYAAAAAAAAAAMCo1NsSipub0R2DrbV/SnLr3vs5Az7mq5LsmeTCoR4TAAAAAAAAAAAA+P+7We80uT6998HfG6j3/odJ/nDoxwUAAAAAAAAAAAB+ZnQ7TQIAAAAAAAAAAADjZGgSAAAAAAAAAAAAGAVDkwAAAAAAAAAAAMAoGJoEAAAAAAAAAAAARsHQJAAAAAAAAAAAADAKhiYBAAAAAAAAAACAUTA0CQAAAAAAAAAAAIzC7LQDAAAAAAAAAAAAGIeZmWkXMHZ2mgQAAAAAAAAAAABGwdAkAAAAAAAAAAAAMAqGJgEAAAAAAAAAAIBRMDQJAAAAAAAAAAAAjIKhSQAAAAAAAAAAAGAUDE0CAAAAAAAAAAAAo2BoEgAAAAAAAAAAABgFQ5MAAAAAAAAAAADAKBiaBAAAAAAAAAAAAEbB0CQAAAAAAAAAAAAwCrPTDgAAAAAAAAAAAGAcls3MTDuBkbPTJAAAAAAAAAAAADAKhiYBAAAAAAAAAACAUfD23ADrsWKb5dNOgJuEY3nyfnzD3LQTttjO220z7YSbvetXr5l2whbz/QKAMZldXu/tfy449mHTTtgid3nhX0w7YYt99U+fOO2Em725NfPTTthiFb9fMHmO5cmr1sswnLvg5qLauUN/91gfz9UAVGenSQAAAAAAAAAAAGAUDE0CAAAAAAAAAAAAo2BoEgAAAAAAAAAAABgFQ5MAAAAAAAAAAADAKBiaBAAAAAAAAAAAAEbB0CQAAAAAAAAAAAAwCoYmAQAAAAAAAAAAgFGYnXYAAAAAAAAAAAAA4zAzM+0Cxs5OkwAAAAAAAAAAAMAoGJoEAAAAAAAAAAAARsHQJAAAAAAAAAAAADAKhiYBAAAAAAAAAACAUTA0CQAAAAAAAAAAAIyCoUkAAAAAAAAAAABgFAxNAgAAAAAAAAAAAKNgaBIAAAAAAAAAAAAYBUOTAAAAAAAAAAAAwCjMTjsAAAAAAAAAAACAcVg2M+0Cxs5OkwAAAAAAAAAAAMAoGJoEAAAAAAAAAAAARsHQJAAAAAAAAAAAADAKs9MOYHhr167N6487Opf3nm233TavO+b4/NLuu087a4Oq9Saah1CtN6nXXK030TyEar1JzeYk+f73rslhz/itvPlP3pHd73K3aedsVMU1rtY8t3p1jjv6qFx15Tez+sbVedZhz89D991v2lkbVW2NE81DqNab1Guu1ptoHkK13kTzEKr17rLjLfKPR+2fJ5/0L5lbszZvfdb9M5+kf/OHOfL9F2Z+ftqF61dtnav1LvWFiz+fU04+MaevfO+0Uzaq4hprHk6V4zipucbVmqv1JpqHUK030TwE5w4nr1pvonkI1XqTms0AkzLoTpOttX1ba2cP9Fg7tNbe2lo7r7X2idbah1tr91j83NGLv/ZtrX27tXbu4nU+3Vr7rQHavtpaW9FaO7O19vBJP97PO+fjH8uNN9yY977/A3nJy16RN7/phKETtki13kTzEKr1JvWaq/UmmodQrTep2Tw3tzon/uExucWKFdNO2SwV17ha8z985MPZaaedc/q7zsrJbz8tJ55w/LSTNqnaGieah1CtN6nXXK030TyEar2J5iFU6p1dPpM/esb9cv3qNUmSY37r3nnj31ySx/3RuclM8vD73HG6gRtRaZ2Ter3rvHvlGTnu6KNyww03TjtlkyquseZhVDqOk5prXK25Wm+ieQjVehPNQ3DucPKq9Saah1CtN6nZDDApN+e3535Hkit67w/tve+T5Kgkf9Na2ynJ1UmuXLzeOb33fRev87Akv99au890kodx4QWfy94PfkiSZK973yeXXPLFKRdtXLXeRPMQqvUm9Zqr9Saah1CtN6nZ/PaTT8xjf/PJ2WWXXaedslkqrnG15v0fdlCe/zsvXrgwnyxfvvVvGF9tjRPNQ6jWm9RrrtabaB5Ctd5E8xAq9b7uiXvlPZ/4z3zrB9cnSfba/db59OXfSZKc84Wr89B73W6aeRtVaZ2Ter3r3Hm33XLiSadMO2OzVFxjzcOodBwnNde4WnO13kTzEKr1JpqH4Nzh5FXrTTQPoVpvUrMZYFKm+oppcefHs3r/f+zdeZRkdX3//1fPNMMAg4jLV2PEARQ+qF81aEQkbLKIGI1rfhI1BowKGIUY1ygKChKNg6gEZVGCccNETSKaEH9qABWMcQMl8sENZTWCuDDAzPRMf/+oGpyQnk3oe/sz9/E4h9Pd1VW3nnxOnarumnd/bt1t/PVHk5yU5L1JLkjy8CTTSZ6SZNckr06yPMmOSc6ptb65lLJdkjOSbJHk1iQvGn98WK31j1bfV631klLKuUmenmT1uS0evWZPrfXmUsrpSZ6Z5JullJOS7Dn+9odrre8spZydZFmS7ZP8VpJDa61fL6V8N8mXkpQkP0nyjIyGUk9LstP482NqreffyWW705YuvTlbb73o9q/nz5ufqampTE7OzR+gW+tNNHehtd6kvebWehPNXWitN2mv+V/P/afcfdt7ZLfH/l4+dPaZfedskNbWOGmvecstt0qSLF26NK95xZ/niNVvgs5hra1xorkLrfUm7TW31pto7kJrvYnmLrTS+6w9FufGm5fl/Mt+kqMO3iVJMjHx6+/fvGwqW2+xWU9169fKOq/WWu9q+x94UK695uq+MzZIi2usuRstPY6TNte4tebWehPNXWitN9HcBe8dzr7WehPNXWitN2mzGWC29LrTZK31iiS3llIeUkq5R5Idaq1fSXK3JB8Z7/54TZKDxzdZnNEw4u5JXjW+bEmSd9Va9x1//paMhiq/P8Nd/iDJ4lrr0lrr0rVk/STJvUopT0qyw/i+9kzy7FLKw8bX+VGt9aAkp2Q0pJnxfb6+1vrYJPfOaCDzBUluqLXundHg56kbvjqzZ6utFmXp0l//76+aXjWnXwRb6000d6G13qS95tZ6E81daK03aa/5X879x3z1Py7O0Ycfmu9dUXPisa/NjTfc0HfWOrW2xkmbzT+5/rq8+IWH5uAnPTkHPfFJfeesV4trrHn2tdabtNfcWm+iuQut9Saau9BK7yG/t332efB98olX7JOHbrdNTnn+o3OvrTe//fuLNp/ML29Z0WPhurWyzqu11tuiFtdYMzNpcY1ba26tN9HchdZ6E81d8d7h7GqtN9HchdZ6kzabAWbLXDg995lJDk3y7CQfXOPyb4w/XpVk4fjzb9Vap8YDj7eOL3tYkteWUs5P8oYk98no1NuLZ7ivnZL8eD09i5NcneTBSb5Qa52uta5I8uUkD1lH2w211qvucPnDkjxx3PbxJJOllHut5/5n3a67PjJfvPDCJMmll3wzO+20c89F69Zab6K5C631Ju01t9abaO5Ca71Je82nnPH+vOuMs/PO08/Og3Yuee0bT8w979X7jw/r1NoaJ+0133jjDTnqyBfmJUf/Rf7gqc/oO2eDtLbGieYutNabtNfcWm+iuQut9Saau9BK79PedkGetuSCPH3JBbnsql/kpWf9Zz7/7euzx873TpLs97D75j++O3f/yKiVdV6ttd4WtbjGmplJi2vcWnNrvYnmLrTWm2jugvcOZ19rvYnmLrTWm7TZDDBb5sLI+MeSvCLJjUn+cI3Lp2e47kyXXZ5kSa31olLKLkn2qbVeXUr5finlz2qtpyZJKeWRSZ6c5IS1hZRS7pbkhRmdnnunJIclObmUslmSPZK8P6NdLzem7epa64mllC2SvC7Jz9Z2/13Z74ADc/HFX8rznnNIpqen86YTTuw7aZ1a6000d6G13qS95tZ6E81daK03abO5NS2ucWvNZ7/3jPzyl7/IWWeclrPOOC1JcvKpp2fhwoXruWV/WlvjRHMXWutN2mturTfR3IXWehPNXWitd03H/cOlOemPH5XNJuflu9f9Mud+be6ezra1dW6tt0UtrrFmZtLiGrfW3FpvorkLrfUmmrvgvcPZ11pvorkLrfUmbTYDzJaJ6emZZv1mRyll3ySfyOg02as9O8lLkty71vpH4+tdmWSXWuttpZS3ZDR8eGWSI2qth4yvc32t9b6llB2TvCejnR23SHJ0rfXiUspWSd6W5FFJVia5Kckraq3fuUPP3yf5r/F1JpO8s9b6ifH3lyT5vSQLkvx9rfWtpZSzk5xTaz2vlPKEJIfUWg9d3TO+3TlJTktycUY7aS7O6JTj7661nrn6/298nXNqreeta91um5pxIBMAmAN+PodPBbg2d99ys74TNnm3rVjZd8JGW7jZ/L4TAIBNyPZHfqzvhI125Xue2XfCJm9qZXtvc07On+g7gTnIYxmAO6O19w69bwhsjIWT8YMnG+RN///32vvFis684cAHzfpzSadDk2tTSjk1ycdrrZ/vu2UuMjQJAHOXoUlm0tobn4k3PwGAu5ahSWZi0IxNhccyAHdGa+8det8Q2BiGJtlQhiZZly6GJufN9h2sTynlM0m2NTAJAAAAAAAAAAAAzKbJvgNqrY/vuwEAAAAAAAAAAADY9PW+0yQAAAAAAAAAAABAFwxNAgAAAAAAAAAAAINgaBIAAAAAAAAAAAAYBEOTAAAAAAAAAAAAwCAYmgQAAAAAAAAAAAAGwdAkAAAAAAAAAAAAMAiTfQcAAAAAAAAAAAAwDBMTfRcwdHaaBAAAAAAAAAAAAAbB0CQAAAAAAAAAAAAwCIYmAQAAAAAAAAAAgEEwNAkAAAAAAAAAAAAMgqFJAAAAAAAAAAAAYBAMTQIAAAAAAAAAAACDYGgSAAAAAAAAAAAAGARDkwAAAAAAAAAAAMAgGJoEAAAAAAAAAAAABsHQJAAAAAAAAAAAADAIk30HAAAAAAAAAAAAMAzzJvouYOjsNAkAAAAAAAAAAAAMgqFJAAAAAAAAAAAAYBAMTQIAAAAAAAAAAACDMNl3AABAy+6+5WZ9J2y0qZXTfSdstMn5E30nbJSFm83vOwEAoFdXvueZfSdstG33eEXfCRvtpouW9J2wUVr7uR7WxmMZgDvDe4cAAP2z0yQAAAAAAAAAAAAwCIYmAQAAAAAAAAAAgEEwNAkAAAAAAAAAAAAMgqFJAAAAAAAAAAAAYBAMTQIAAAAAAAAAAACDYGgSAAAAAAAAAAAAGITJvgMAAAAAAAAAAAAYholM9J3AwNlpEgAAAAAAAAAAABgEQ5MAAAAAAAAAAADAIBiaBAAAAAAAAAAAAAbB0CQAAAAAAAAAAAAwCIYmAQAAAAAAAAAAgEEwNAkAAAAAAAAAAAAMgqFJAAAAAAAAAAAAYBAMTQIAAAAAAAAAAACDYGgSAAAAAAAAAAAAGITJvgMAAAAAAAAAAAAYhnkTfRcwdHaaBAAAAAAAAAAAAAbB0CQAAAAAAAAAAAAwCIYmAQAAAAAAAAAAgEEwNAkAAAAAAAAAAAAMwmTfAXRv1apVefPxx+WKWrNgwYIc+8YT8oDFi/vOWqvWehPNXWitN2mvubXeRHMXWutNNHfpW5deklPesSRnnPWBvlPWq8U1bq25td5Ecxda603aa26tN9HchdZ6E81daK03aa/53tsuykV/9+f5/Zecni0XLsgpr3lGlq2YyqVXXJuXn/TPmZ6e7jvxf2ltjZP2mlvrTTR3obXeRHMXWutNNHehtd5Ecxda603aa26tN9HchdZ6kzabAWbLrO00WUrZt5Ryzl18zLNLKU/YyNucX0r5yhof33FXNs1wf08opbxoNu/jzvr85z6b5cuW5wMf/miOftnLc9Lb3tJ30jq11pto7kJrvUl7za31Jpq70Fpvorkr7z/rvTn+uGOybNnyvlM2SItr3Fpza72J5i601pu019xab6K5C631Jpq70Fpv0lbz5Px5+Zu/fEZuXbYiSfI3r31mXvn2f84BL3p3fnHzbXnWQbv2XDizltZ4tdaaW+tNNHehtd5Ecxda6000d6G13kRzF1rrTdprbq030dyF1nqTNpsBZstQTs/9vFrrvkkek+R3Sym/O1t3VGs9r9Z6xmwd/67wja9/LXvsuVeS5OGP+J1cdtm3ey5at9Z6E81daK03aa+5td5Ecxda6000d+X+222XJSef0nfGBmtxjVtrbq030dyF1nqT9ppb6000d6G13kRzF1rrTdpqfsvRT86Zn7g41/30l0mS3/4/2+TL3/pRkuTiS67MHr+zQ595a9XSGq/WWnNrvYnmLrTWm2juQmu9ieYutNabaO5Ca71Je82t9Saau9Bab9JmM8Bs6ez03KWUnZN8sNa62/jrjyY5Kcl7k1yQ5OFJppM8JcmuSV6dZHmSHZOcU2t98xrH2izJaUl2ymjw85gk167l+GvaPMmCJD8rpeyb5K3j+zgjyfVJTkhyW5Ibkzw/yd8meXOt9aullMuTvLbWBu7DVwAAIABJREFU+olSymeSHJbk/CRfSlKS/CTJM5L8cZJdxn0fSXJVkgcm+Uqt9chSyr2SfHjcUpPsV2t90G+6rr+JpUtvztZbL7r96/nz5mdqaiqTk3PzbO2t9Saau9Bab9Jec2u9ieYutNabaO7K/gcelGuvubrvjA3W4hq31txab6K5C631Ju01t9abaO5Ca72J5i601pu00/zc3//d/PTnN+ezX74ir/yT/ZMkV17zs+y564754jd+kCfu9ZBstXBBz5Uza2WN19Rac2u9ieYutNabaO5Ca72J5i601pto7kJrvUl7za31Jpq70Fpv0mYzwGzpbKfJWusVSW4tpTyklHKPJDvUWr+S5G5JPlJr3SfJNUkOHt9kcUZDiLsnedUdDveCJDfUWvfOaMjy1HUcP0n+rpRyfpIrkvw8yep/0V9Ya90ryQczGpx8+rjjgowGMf8xycGllB2SLEtyQCllm/HtrslooPP1tdbHJrl3kkffoXPnJH+aZLckTyyl3DfJ65L80/h+/iEdDq6uttVWi7J06dLbv141vWpOvwi21pto7kJrvUl7za31Jpq70FpvopmZtbjGrTW31pto7kJrvUl7za31Jpq70FpvorkLrfUm7TT/yR/slv132zn/9p4j8/Cd75f3HfdHef2p/5JXHrpf/uXUw/PTm27Ojb9Yuv4D9aCVNV5Ta82t9Saau9Bab6K5C631Jpq70FpvorkLrfUm7TW31pto7kJrvUmbzQCzpevTc5+Z5NAkz85oUHG1b4w/XpVk4fjzb9Vap2qtS5PceofjPCyjIcTzk3w8yeR4B8e1HX/16bm3z2hHytVDmHX88V5JfjkehEySC5M8NMm5SQ5M8oSMdqXcLaOhznPH17uh1nrVDO2rfa/W+qta68ok142//+AkF42//4X0YNddH5kvXnhhkuTSS76ZnXbauY+MDdZab6K5C631Ju01t9abaO5Ca72JZmbW4hq31txab6K5C631Ju01t9abaO5Ca72J5i601pu003zg4e/O4494Tw468j259Ipr86fHfSSPfPD9c9gbPpwn/tnpuec2W+Zz/3FF35kzamWN19Rac2u9ieYutNabaO5Ca72J5i601pto7kJrvUl7za31Jpq70Fpv0mYzwGzpemT8Y0lekdHpr/9wjcunZ7juTJetdnmSq2utJ5ZStsho98afreP4SZJa66pSyjUZnaI7SVaNP96Q5G6llN+qtV6XZJ8kV9Rabyql3JLkWUmenuSZSY5O8twNaFzb97+d5LFJvpnRLpqd2++AA3PxxV/K855zSKanp/OmE07sI2ODtdabaO5Ca71Je82t9Saau9Bab6KZmbW4xq01t9abaO5Ca71Je82t9Saau9Bab6K5C631Jm02r/a9H9+Qfzn18Nx624pc8LXv5d8uurzvpBm1uMatNbfWm2juQmu9ieYutNabaO5Ca72J5i601pu019xab6K5C631Jm02s+maN9F3AUM3MT29vrm/30wpZd8kn0jygzUufnaSlyS5d631j8bXuzLJLrXW20opb8loIPLKJEfUWg8ZX+f6Wut9SylnJzknyb9ntKvk4oxO7/3uWuuZ4+u+6w7HPz/JlkluGTfcktHQ48PvcB8HJDk+o0HKm5IcWmu9oZRyZJLDaq27lVIOT/LiWusj1uwaf35OktMy2s1yl/Hn59Radx9//8tJDklyc5IPZLTr5LVJdqu17rSutbxtar3DmQAAG2xqZXs/WkzO95sTAACza9s9XtF3wka76aIlfScAAADA7RZOxj/osEH++t+/394/WNKZVz3ugbP+XDJrQ5NrU0o5NcnHa62fb/H4d1Yp5YlJflpr/c/xoOZra637res2hiYBgLuSoUkAAPjfDE0CAADAnWNokg1laJJ16WJostPTc5dSPpPkhlkcmJzV499FfpjkrFLKVJL5SY7quQcAAAAAAAAAAAAGodOhyVrr41s+/l2h1vqdJI/tuwMAAAAAAAAAAACGZl7fAQAAAAAAAAAAAABdMDQJAAAAAAAAAAAADIKhSQAAAAAAAAAAAGAQDE0CAAAAAAAAAAAAg2BoEgAAAAAAAAAAABgEQ5MAAAAAAAAAAADAIEz2HQAAAAAAAAAAAMAwTExM9J3AwNlpEgAAAAAAAAAAABgEQ5MAAAAAAAAAAADAIBiaBAAAAAAAAAAAAAbB0CQAAAAAAAAAAAAwCIYmAQAAAAAAAAAAgEEwNAkAAAAAAAAAAAAMgqFJAAAAAAAAAAAAYBAMTQIAAAAAAAAAAACDYGgSAAAAAAAAAAAAGITJvgMAAAAAAAAAAAAYhnkTfRcwdHaaBAAAAAAAAAAAAAbB0CQAAAAAAAAAAAAwCIYmAQAAAAAAAAAAgEEwNAkAAAAAAAAAAAAMwmTfAQAAdGty/kTfCQCz4rYVK/tO2GiT89r7W0avI8Cm6roL3tp3wkZ78mlf7jtho5x7xO59JwCwhnrtr/pO2Cjlflv3nbDJu/m2qb4TNtqihf65G/owtXK674SN5j0tANbU3r/OAAAAAAAAAAAAAPwGDE0CAAAAAAAAAAAAg2BoEgAAAAAAAAAAABgEQ5MAAAAAAAAAAADAIBiaBAAAAAAAAAAAAAZhsu8AAAAAAAAAAAAAhmFiou8Chs5OkwAAAAAAAAAAAMAgGJoEAAAAAAAAAAAABsHQJAAAAAAAAAAAADAIhiYBAAAAAAAAAACAQTA0CQAAAAAAAAAAAAyCoUkAAAAAAAAAAABgEAxNAgAAAAAAAAAAAINgaBIAAAAAAAAAAAAYBEOTAAAAAAAAAAAAwCBM9h0AAAAAAAAAAADAMMybmOg7gYGz0yQAAAAAAAAAAAAwCIYmAQAAAAAAAAAAgEEwNAkAAAAAAAAAAAAMgqFJAAAAAAAAAAAAYBAMTQIAAAAAAAAAAACDYGgSAAAAAAAAAAAAGITJvgPo3qpVq/Lm44/LFbVmwYIFOfaNJ+QBixf3nbVWrfUmmrvQWm/SXnNrvYnmLrTWm2juQmu9ieYutNabaO5Ca71JMrViRY4/7phcd+01WbF8RQ574eHZe9/9+s5ar29deklOeceSnHHWB/pOWa8WHxetNbfWm2juQmu9SXvNrbyGzJtIXva4HbPdtltkeno67zz/h7nyZ7cmSY7Yc3GuvunWfOqy/+65cu1ae1y01pto7kJrvYnmLrTW+93vfDsfeu+7ctxJZ+T6a67KqW87LhMTE9lu+wfmT1/66sybNzf3cWltnVvrTZKVK1fmrSccmx//6IeZyERe+dpjs+ODduo7a51aW+fWepP2mlvrTdpsTrynNdtabAaYLbP6G0opZd9SynQp5ZA7XH5pKeXsu/i+jiulHFFK2X58n6+5w/c/WUo5f/z5OaWUBWs5ztmllCds4H1u8HXnks9/7rNZvmx5PvDhj+bol708J73tLX0nrVNrvYnmLrTWm7TX3FpvorkLrfUmmrvQWm+iuQut9Saau9Bab5L866fPzTbb3D1n/O0H8453n54lbzmh76T1ev9Z783xxx2TZcuW952yQVp8XLTW3FpvorkLrfUm7TW38hqy+/bbJkn+/OOX5W//4+octvt22WbhZN785F3y2B227blu/Vp7XLTWm2juQmu9ieYutNT7zx99f057+/FZsXz0O8j7T3t7DjnsyLzp5Pdmeno6X73ogp4L166ldU7a602SL114fpLktLM+lBe++Kic/u539hu0AVpb59Z6k/aaW+tN2mz2ntbsa7EZYLZ08Wddlye5fWiylPKwJFvN8n1+P8kz1rjPeya5/U+Waq2H1FrbeKWdBd/4+teyx557JUke/ojfyWWXfbvnonVrrTfR3IXWepP2mlvrTTR3obXeRHMXWutNNHehtd5Ecxda602S/R9/UA7/s6NGX0wn8+fP/RM23H+77bLk5FP6zthgLT4uWmturTfR3IXWepP2mlt5Dbnohzfl5H//QZLkPlsvyM3LV2aLzebnA1+5Op+tN/Rct36tPS5a6000d6G13kRzF1rqvc/97p9XHPu227/+wXcvz0Me/qgkya677ZFLv/6VvtLWq6V1TtrrTZK9H7d/XvW645Ik1193bRYt2rrfoA3Q2jq31pu019xab9Jms/e0Zl+LzQCzpYt36i5JUkop29Raf5HkuUk+lOQBpZSXJHl6RkOUNyR5WpJnJ3l+RgOdxybZPsmRSeYn+WSt9dhSyh8m+YskK5N8sdb6mv95l7khyY2llAfXWr+T5P9L8g9J9s4o5sokuyTZLsl7kyxIckt+Pdx5eCnlVUm2SXJkrfUrpZSXjtumk5xTa33X6jsrpeyc5G+TTI27n11rvaqUclKSPcdX+3Ct9Z3jHTaXjf+/fivJobXWr/8G6/obW7r05my99aLbv54/b36mpqYyOTk337htrTfR3IXWepP2mlvrTTR3obXeRHMXWutNNHehtd5Ecxda602SLbcc/c3f0qVL85pX/HmOWD38Moftf+BBufaaq/vO2GAtPi5aa26tN9HchdZ6k/aaW3oNWTWdvPKAB+b3dtw2x//rd3P9r5bl+l8ty6MX373vtPVq7XHRWm+iuQut9Saau9BS7+577Z//vv7aX18wPZ2JiYkkyRZbbJVbbrm5p7L1a2mdk/Z6V5ucnMzxb/jLXHj+53LCW0/uO2e9Wlvn1nqT9ppb603abPae1uxrsRlgtnSx02SSfDzJ00spE0l2S3LR+L7vmeSAWutjMhrgfPT4+jfVWvdM8q0kr0myV5JHJtm8lPKAJG9Msv/4Or9dSjlwhvv8SH49BPmUJP80w3WWJPmrWutjk7wzya7jy79Wa90vySlJDi2lPCTJszIagNwryVNLKWWN4xyY5CtJDsho0HObUsqTkuyQZPfx7Z493mUzSX5Uaz1ofPwXrXvp7npbbbUoS5cuvf3rVdOr5vSLYGu9ieYutNabtNfcWm+iuQut9Saau9Bab6K5C631Jpq70Frvaj+5/rq8+IWH5uAnPTkHPfFJfedsclp8XLTW3FpvorkLrfUmbTa39Bryts9+P4d98JK87HE7ZuFkV28b33mtPS5a6000d6G13kRzF1rrXdPEvF+/jtx669JstdWidVy7X62tc2u9a3r9m/4q53zi03nrCcfm1ltv6TtnnVpb59Z6k/aaW+tN2mxuTYtr3GIzm655E/7z39r/6+Qx2M3d5MMZDTDuneQL48tWJVme5COllPcluX+Szcbfq+OPOyb5dq311lrr9HhHyfsmuXeSfymlnJ/kIUkeOMN9/lOSPyilbJ/k+ox2kryjkuTiJKm1frLW+pnx5V8bf7w+yZZJ/m+SxUk+N/7vf5zuO8n7kvw8yXlJXpLRjpMPTvKFcfeKJF8etybJN8Yfr0qycIauWbXrro/MFy+8MEly6SXfzE477dx1wkZprTfR3IXWepP2mlvrTTR3obXeRHMXWutNNHehtd5Ecxda602SG2+8IUcd+cK85Oi/yB889Rl952ySWnxctNbcWm+iuQut9SbtNbfyGnJAuVcOedT9kiTLVqzKqkxn1fR0z1UbrrXHRWu9ieYutNabaO5Ca71r2v5BJZdd8tUkyTe+clEe/LBd13OL/rS2zq31Jsl5n/5k/u6sM5MkCxdukXnz5mXexNz+A43W1rm13qS95tZ6kzabW9PiGrfYDDBbOhkZr7X+oJSyVZKjkvxlRsOQd0vy1FrrY0opW2Y0qLh6VnTV+OP3k+xSStm81rqslPKxJC/PaNjwwFrrilLKoUm+meSpd7jPm0spNclfZ3QK7pl8J6PdLT9bSnlOknuML7/ju4I1yWVJDq61TpdSXpbk0iTPHH//KRkNSL6xlPJHSV6d0e6ahyU5uZSyWZI9krw/ycEzHL9T+x1wYC6++Et53nMOyfT0dN50wol95qxXa72J5i601pu019xab6K5C631Jpq70FpvorkLrfUmmrvQWm+SnP3eM/LLX/4iZ51xWs4647Qkycmnnp6FCzv/+7dNVouPi9aaW+tNNHehtd6kveZWXkO++P2f5RX7PzAnPe0hmZw/kfd84UdZvrKdocnWHhet9Saau9Bab6K5C631rul5h/95Tn/7mzM1dWp++wHbZ/e99u87aa1aW+fWepNkn/0OyInHHZMXv+B5mZqaytEvf002n2M/D91Ra+vcWm/SXnNrvUmbza1pcY1bbAaYLRPTs/hXw6WUfZMcUWs9pJTy0iR/XGvdrZTyhCTPT3KvJJuPr74sox0bN0uyy3hXyYyHIo/IaNDw3FrriaWU5yZ5cZL5Sa7MaDjxVRntDHleknNqrbuXUp6c5PQk22W0M+RptdZ9SylXJtklo90tTx8f55Ykz03y9vHtzxt3HlJrPbSU8sqMBjM3z+hU3C8d956T5LsZDUQuHx/rZbXWr5dSliT5vSQLkvx9rfWtpZSzZzr+utbxtql+hywBAABacNuKlX0nbLTJeXN7d42ZTM7v6NwYAB1r8XXkD9/3n30nbJRzj9i97wQA1lCv/VXfCRul3G/rvhM2eTffNtV3wkZbtNBpZaEPUw39odVq3tPqxsLJWGg2yLu++MP2nkjozFF77jDrzyWzOjTJXcPQJAAAwPq1OOxiaBJg7mjxdcTQJAB3hqFJ7sjQJLChDE2yNoYm2VCGJlmXLoYm2/vXGQAAAAAAAAAAAIDfgKFJAAAAAAAAAAAAYBAMTQIAAAAAAAAAAACDYGgSAAAAAAAAAAAAGARDkwAAAAAAAAAAAMAgGJoEAAAAAAAAAAAABmGy7wAAAAAAAAAAAACGYWKi7wKGzk6TAAAAAAAAAAAAwCAYmgQAAAAAAAAAAAAGwdAkAAAAAAAAAAAAMAiGJgEAAAAAAAAAAIBBMDQJAAAAAAAAAAAADIKhSQAAAAAAAAAAAGAQDE0CAAAAAAAAAAAAg2BoEgAAAAAAAAAAABgEQ5MAAAAAAAAAAADAIEz2HQAAAAAAAAAAAMAwzMtE3wkMnJ0mAQAAAAAAAAAAgEEwNAkAAAAAAAAAAAAMgtNzAwAAAAAAAAAAAHNaKWWzJGcl2T7J5klOSPJfSc5OMp3k20n+rNa6al3HsdMkAAAAAAAAAAAAMNc9N8mNtda9kjwhyd8keXuSY8aXTSR5yvoOYmgSAAAAAAAAAAAAmOv+Icnrx59PJJlK8qgkF4wv+9ckB6zvIE7PDQAAAAAAAAAAAMxptdabk6SUsnWSjyU5JsmSWuv0+Cq/SrLN+o5jaBLoxNTK6fVfaQ6ZnD/RdwLArGntOTnxvAxsmIWbze87Ae4SN9821XfCRlu00FtMtK/F15Fzj9i974SNsu0zz+g7YaPd9LEX9Z3AHOT3ajYV5X5b953AHOPnemBD+dkCgD6VUrZL8o9J3l1r/XAp5a/X+PbWSX6+vmM4PTcAAAAAAAAAAAAwp5VS7pPkM0leXWs9a3zxN0op+44/PzjJF9Z3HH8uBAAAAAAAAAAAAMx1r02ybZLXl1JeP77s6CTvKqUsSPKdjE7bvU6GJgEAAAAAAAAAAIA5rdZ6dEZDkne0z8Ycx9AkAAAAAAAAAAAAnZiY6LuAoZvXdwAAAAAAAAAAAABAFwxNAgAAAAAAAAAAAINgaBIAAAAAAAAAAAAYBEOTAAAAAAAAAAAAwCAYmgQAAAAAAAAAAAAGwdAkAAAAAAAAAAAAMAiGJgEAAAAAAAAAAIBBMDQJAAAAAAAAAAAADIKhSQAAAAAAAAAAAGAQJvsOAAAAAAAAAAAAYBjmTfRdwNDZaRIAAAAAAAAAAAAYBEOTAAAAAAAAAAAAwCAYmgQAAAAAAAAAAAAGwdAkAAAAAAAAAAAAMAiGJgEAAAAAAAAAAIBBMDQJAAAAAAAAAAAADIKhSQAAAAAAAAAAAGAQJvsOoHurVq3Km48/LlfUmgULFuTYN56QByxe3HfWWrXWm2ju0rcuvSSnvGNJzjjrA32nrFdra9xab6K5C631Jpq75Dl5drXW3FpvorkLrfUm7TW31pto7sLKlSvz1hOOzY9/9MNMZCKvfO2x2fFBO/WdtU6trXHSXnNrvYnmLrTWe+9tFuaik56e3z/207niml8kSZ619wNz5BP/b/Z9zT/3XDez1tY40dwlv1fPrtaaW+tNNHehtd5Ecxda603aa26tN9HchdZ6kzabAWbLnNlpspSybyllupRyyB0uv7SUcvYGHuP6GS67spRyYSnlglLKV0spr7oLWv/X/azxvfuWUt69ju9fWUpZeGcb7ozPf+6zWb5seT7w4Y/m6Je9PCe97S195qxXa72J5q68/6z35vjjjsmyZcv7Ttkgra1xa72J5i601pto7orn5NnXWnNrvYnmLrTWm7TX3FpvorkLX7rw/CTJaWd9KC988VE5/d3v7DdoA7S2xkl7za31Jpq70FLv5PyJ/M2Re+XWZVO3X/aIHe6ZPzlgl0xM9Bi2Hi2t8Wqau+H36tnXWnNrvYnmLrTWm2juQmu9SXvNrfUmmrvQWm/SZjPAbJkzQ5Njlye5fWiylPKwJFvdBcd9fK11nyR7JDm8lPJ/7oJjzqjWen2t9cWzdfy7wje+/rXssedeSZKHP+J3ctll3+65aN1a6000d+X+222XJSef0nfGBmttjVvrTTR3obXeRHNXPCfPvtaaW+tNNHehtd6kvebWehPNXdj7cfvnVa87Lkly/XXXZtGirfsN2gCtrXHSXnNrvYnmLrTU+5ZDd8+Z530n1910S5LkHltvnjc+99F55fsu6rls3Vpa49U0d8Pv1bOvtebWehPNXWitN9HchdZ6k/aaW+tNNHehtd6kzWaA2TLXTs99SZJSStmm1vqLJM9N8qEkDyilvCTJ0zMaorwhydOSPDvJ8zMa/jx29UFKKScm2SbJS+5w/C2TrEhySyllsySnJdlpfPtjaq3nl1IuTXJBkocnmU7ylCQ3JzkjyUOTfD/J5uP72W58+RZJbk3yoiTzk5xTa929lPKkcddEkq8nOWLc8Z5Syg7jz59Wa73pzizaxlq69OZsvfWi27+eP29+pqamMjk51x4OI631Jpq7sv+BB+Xaa67uO2ODtbbGrfUmmrvQWm+iuSuek2dfa82t9Saau9Bab9Jec2u9ieauTE5O5vg3/GUuPP9zOeGtJ/eds14trnFrza31Jpq70Ervc/fbOT/95W357Devziuf+TuZP29eTnvJPnn13345ty6fWv8BetTKGq9Jczf8Xj37WmturTfR3IXWehPNXWitN2mvubXeRHMXWutN2mxm0zVvLp8mgkGYaztNJsnHkzy9lDKRZLckF2XUec8kB9RaH5PRsOejx9e/qda6Z631c0lSSlmSZLLW+me11unxdT5TSrkgSU1ycZKlSV6Q5IZa694ZDUaeOr7u3ZJ8ZLwz5TVJDs5oQHNhrXX3JH+Z0fBlkixJ8q5a677jz2/fu7iUMpnkb5L8fq31d5N8L8n9x99+3/g2VyY58M4t18bbaqtFWbp06e1fr5peNadfBFvrTTQzs9bWuLXeRHMXWutNNDOzFte4tebWehPNXWitN2mvubXeRHOXXv+mv8o5n/h03nrCsbn11lv6zlmnFte4tebWehPNXWil90/2L9n/Eb+dfzvhSXn4DvfMV9/5zDx08T3yriP2zAdevn922W7bvO1PH9t35oxaWeM1aWYmLa5xa82t9Saau9Bab6K5C631Ju01t9abaO5Ca71Jm80As2UuDk1+OKNTdO+d5Avjy1YlWZ7kI6WU92U0fLjZ+Ht1jdveJ6MdIhflf1p9eu7txv89J8nDkjyxlHJ+RoOak6WUe42v/43xx6uSLEyyc5KvJEmt9cfjyzM+xmvHx3jD+P5Xu1dGA53/Pb7dX49vmyRfG3+8Pr8ewOzMrrs+Ml+88MIkyaWXfDM77bRz1wkbpbXeRDMza22NW+tNNHehtd5EMzNrcY1ba26tN9HchdZ6k/aaW+tNNHfhvE9/Mn931plJkoULt8i8efMyb2IuviX0a62tcdJec2u9ieYutNJ74OvOzeOP+VQOOuZTufSHN2bXl/59HnrEOTnomE/lj0/6XC6/6qa88n0X9505o1bWeE2amUmLa9xac2u9ieYutNabaO5Ca71Je82t9Saau9Bab9JmM8BsmXMj47XWH5RStkpyVEa7Ou6Y0e6PT621PqaUsmVGQ4er92ldtcbNf5LkoCTnl1KeUGs97w7HXl5K+UmSBUkuT3J1rfXEUsoWSV6X5Gfjq07nf/qvjAY531lKuV+S3x5ffnmSJbXWi0opuyTZZ43b/HeSu5dS7lFr/Vkp5V1JPriW43dqvwMOzMUXfynPe84hmZ6ezptOOLHPnPVqrTfRzMxaW+PWehPNXWitN9HMzFpc49aaW+tNNHehtd6kvebWehPNXdhnvwNy4nHH5MUveF6mpqZy9Mtfk80XLuw7a51aW+OkvebWehPNXWitt0UtrrFmZtLiGrfW3FpvorkLrfUmmrvQWm/SXnNrvYnmLrTWm7TZDDBbJqane53fu10pZd8kR9RaDymlvDTJH9dadyulPCHJ8zPauXHz8dWXJXlfRrtN7lJrfc34GNfXWu9bSnlQkvOSPCajAcsfJ1mZ0ZDoVUkOGx/nzCSLMxrKfHet9cxSypXjY95WSnlLRoOR78/oVNu/m+RHSR5ba92ulLJjkvdktBvlFkmOTnJdknNqrbuXUg7OaAfKlRntXnlUkh/e8fi11rPXtTa3TfU7ZAl3hamVbT2MJ+dPrP9KAI1q7Tk58bwMwLDcfNtU3wkbbdHCOfd3ucActO0zz+g7YaPd9LEX9Z3AHOT3agAAmNnCyfjBkw1yxpd/1N4vVnTmRbsvnvXnkjkzNMnaGZpkU9DaG4neRAQ2Za09JyeelwEYFkOTwKbK0CSbCr9XAwDAzAxNsqEMTbIuXQxNzpvtOwAAAAAAAAAAAACYCwxNAgAAAAAAAAAAAINgaBIoYXr4AAAgAElEQVQAAAAAAAAAAAAYBEOTAAAAAAAAAAAAwCAYmgQAAAAAAAAAAAAGYbLvAAAAAAAAAAAAAIZhYqLvAobOTpMAAAAAAAAAAADAIBiaBAAAAAAAAAAAAAbB0CQAAAAAAAAAAAAwCIYmAQAAAAAAAAAAgEEwNAkAAAAAAAAAAAAMgqFJAAAAAAAAAAAAYBAMTQIAAAAAAAAAAACDYGgSAAAAAAAAAAAAGARDkwAAAAAAAAAAAMAgTPYdAAAAAAAAAAAAwDDMm5joO4GBs9MkAAAAAAAAAAAAMAiGJgEAAAAAAAAAAIBBMDQJAAAAAAAAAAAADIKhSQAAAAAAAAAAAGAQDE0CAAAAAAAAAAAAg2BoEgAAAAAAAAAAABgEQ5MAAAAAAAAAAADAIEz2HQAMw+T8ib4TNsptK1b2nbDRFm42v+8E5iCPZWbS2nMywIaaWjndd8JG85zMTBYt9HbNbPN8Af246WMv6jtho+3y8k/1nbDRLj/pSX0nbPJafE5u7bWvxTWGmbT4/mxrvJ/MpsJrNQB0y06TAAAAAAAAAAAAwCAYmgQAAAAAAAAAAAAGwfmeAAAAAAAAAAAA6MTERN8FDJ2dJgEAAAAAAAAAAIBBMDQJAAAAAAAAAAAADIKhSQAAAAAAAAAAAGAQDE0CAAAAAAAAAAAAg2BoEgAAAAAAAAAAABgEQ5MAAAAAAAAAAADAIBiaBAAAAAAAAAAAAAbB0CQAAAAAAAAAAAAwCIYmAQAAAAAAAAAAgEEwNAkAAAAAAAAAAAAMwmTfAQAAAAAAAAAAAAyDXf7om8cgAAAAAAAAAAAAMAiGJgEAAAAAAAAAAIBBMDQJAAAAAAAAAAAADIKhSQAAAAAAAAAAAGAQDE0CAAAAAAAAAAAAg2BoEgAAAAAAAAAAABgEQ5MAAAAAAAAAAADAIBiaBAAAAAAAAAAAAAZhsu8Aurdq1aq8+fjjckWtWbBgQY594wl5wOLFfWetVWu9ieYutNabJFMrVuT4447JdddekxXLV+SwFx6evffdr++stWpxjTXPvtYex0l7a5y019xab6K5C631Jpq70Frvat+69JKc8o4l/4+9Ow+XrCzv/f3ddIMtqOCMUcSo8BoHphOHOMEBEVHjlESJU3BA0SioqAcToyionCMGDYqCSnBA0VyakxjHqD+CCmqCAkrkNU5RmVREkFagu/f+/bGrOS3pMVKr+ul139fV195Vu2rVZ7+srloUD6ty8invm3XKBlVcY83TV603qdmceL6YNs3TV603qdO81Vxy7EG75a63u1kWkvzlh7+RpVvN5XVPvE9Wzi/k+z+5Ov/r9POzsDDr0v+qyhqvqVpztd41ee2bnmq9ieYhVHx/tmJztf2iWm9Sr7la75q8Vk9Ptd6kZjPAtJQ/02RrbZ/W2kJr7aAbXH9+a+3UjdzG41trv9Na27G1duJ6bvfu1toT17j8rdba29a4fGpr7XHruO9dWmtfnnz/g9baso1pm4bPf+6zue7a6/K+D3woh7/4iLzpjcfOKmWjVOtNNA+hWm+SfPLjH8v22++Qk//2/XnziSfluGOPmXXSelVcY83TV20/TuqtcVKvuVpvonkI1XoTzUOo1psk7znlXTn6qFfm2muvm3XKRqm4xpqnr1pvUrPZ88X0aZ6+ar1JneaH3fv2SZI/fstZedPHe172qJbDH7Fr3vLp/8ifvOWsbLN0Sfa95+1mXLl2VdZ4TdWaq/Wu5rVvuqr1JpqHUPH92YrN1faLar1JveZqvat5rZ6uar1JzWa2XHNzc/74s84/Qyg/NDlxYZLrhyZba/dJst0m3P/wJLfovV/ae3/+em73z0keMnmMuyX5bpK91/j5g5J8bhMedya+/rVz8sAHPyRJstvue+SCC74546L1q9abaB5Ctd4k2e/hB+S5f37Y4oWFZMmSzftkvxXXWPP0VduPk3prnNRrrtabaB5Ctd5E8xCq9SbJnXbaKccdf8KsMzZaxTXWPH3VepOazZ4vpk/z9FXrTeo0f+Ybl+UVH/pGkuSOt7xprvr1ilzw4yuzw7ZbJ0m2u8mSrFy1GZ5mMnXWeE3Vmqv1rua1b7qq9Saah1Dx/dmKzdX2i2q9Sb3mar2rea2ermq9Sc1mgGnZ/I9KN855SVprbfve+5VJnprktCR3bq29IMkTsjhE+bMkj0/y5CTPzOLQ6BuS7JHkva21pyZ5b+/9Aa2185P8S5LdkiwkeWwWByJfPnnMRyb5xySPaa3dM8mvk1zUe/9la23PJCckWZXkmiSHrC26tXZokocn+dPe+7U35oKsz/LlV+fmN7/Z9ZeXbLUkK1euzNKlm+fuUK030TyEar1Jsu22i7Pcy5cvz5EvfVEOXf0mwWaq4hprnr5q+3FSb42Tes3VehPNQ6jWm2geQrXeJNlv/wNy8UU/nnXGRqu4xpqnr1pvUrPZ88X0aZ6+ar1JreZV8wt501N2z8N32zHPP+Wc3HK7bfLaP753XvjwXfLLX6/Il79z+awT16rSGq9Wrbla72pe+6arWm+ieQgV35+t2Fxtv6jWm9Rrrta7mtfq6arWm9RsBpiWLeVMk0nykSRPaK3NJblfkrOy+PvdOsnDeu/3z+KQ6H0nt7+i9/7g3vvHk5yb5OlJ1jwv9S2SfLD3vneSi5Ic2Hv/aZL51tr2SQ5M8snJnwOT7JPkU5P7vjPJCyb3PTHJX6+l94VZPGvlnww5MJkk2213syxfvvz6y/ML85v1i2C13kTzEKr1rnbZpZfk+YccnAMf/Yc54JGPnnXOelVcY83DqLQfJzXXuFpztd5E8xCq9Saah1Ctt6KKa6x5+qr1JjWbq6m4xpqnr1pvUq/5iNPOy77HnJE3HLRbXv1H98qf/M1Z2e/1Z+Qj//rj/OXj7jnrvLWqtsZJveZqvVVVW+dqvYnmoVR7fzap11xtv6jWm9RrrtZbVbV1rtab1GwGmJYtaWjyA1n8iO6HJvnC5Lr5LA5CfrC19u4kd0qy9eRnfSO2+fXJ1x8lWTb5/nNJHpbkNr33H2VxaPKBWfyY7tVDk7/Tez938v2ZSe61lm0/LMkOvfdVG9Fxo9pzz73yxTPPTJKcf9652WWXXYdO2CTVehPNQ6jWmySXX/6zHPa8Q/KCw1+Sxzzuj2ads0EV11jz9FXbj5N6a5zUa67Wm2geQrXeRPMQqvVWVHGNNU9ftd6kZnM1FddY8/RV603qND/+9++Y5z/sbkmSX1+3KgsLyS+Wr8jV16xMkvzkqmuz/bZbr28TM1NljddUrblab1XV1rlab6J5CBXfn63YXG2/qNab1Guu1ltVtXWu1pvUbAaYli1mZLz3/r3W2nZJDkvyiiR3zeLZIh/Xe79/a23bJOckmZvcZX6Nu89n7QOkC2u57p+TvDrJGWs87q2S3D6LHxOeJBe31nbrvZ+fxWHKb69lO49N8q7W2qG993ds/G/629v3Yfvn7LO/lKc/5aAsLCzktce8fsiH32TVehPNQ6jWmySnvuvkXHXVlTnl5HfklJMX/9of/7aTsmzZsg3cczYqrrHm6au2Hyf11jip11ytN9E8hGq9ieYhVOutqOIaa56+ar1JzeZqKq6x5umr1pvUaf7U+ZfmjU/ePR964R9k6yVb5bUfvSBX/Oq6nPBne2XV/EKuWzWfV5x+/qwz16rKGq+pWnO13qqqrXO13kTzECq+P1uxudp+Ua03qddcrbeqautcrTep2QwwLXMLC2ubC6yjtbZPkkN77we11l6Y5Gm99/u11h6R5JlJbpPkJpObX5vk3Vk82+Q9eu9HTrZxTJJHJHlOkhN77w9orf1gcptrWmvHJrmw935qa21Zkp8neWTv/YzJ/d+cZPve+zMml/dM8pYsDmiuTPKsLA5mnr7mtpNsm+SrWfzo7/9Y1+94zcq1Dm8CU3TNisFPAvtbW7b1klknsBmyLwMwJitX1ftXp6VL5jZ8I+BG5/kC2Fj3OOKfZp2wyS580+b/sacMr9prn9c9thQV35+txvvJbCm8VrOlWLY0dg42ynv+7Ue1nvgY1J/9/k5Tfy4pPzQ5BoYmYXgV38jwxgBrY18GYEyqvbmceIMZZsXzBbCxDE2ypaj22ud1jy1Fxfdnq/F+MlsKr9VsKQxNsrEMTbI+QwxNru0jqQEAAAAAAAAAAAC2OIYmAQAAAAAAAAAAgFEwNAkAAAAAAAAAAACMwtJZBwAAAAAAAAAAADAOc7MOYPScaRIAAAAAAAAAAAAYBUOTAAAAAAAAAAAAwCgYmgQAAAAAAAAAAABGwdAkAAAAAAAAAAAAMAqGJgEAAAAAAAAAAIBRMDQJAAAAAAAAAAAAjIKhSQAAAAAAAAAAAGAUDE0CAAAAAAAAAAAAo2BoEgAAAAAAAAAAABiFpbMOAAAAAAAAAAAAYBy2mpubdQIj50yTAAAAAAAAAAAAwCgYmgQAAAAAAAAAAABGwdAkAAAAAAAAAAAAMAqGJgEAAAAAAAAAAIBRMDQJAAAAAAAAAAAAjIKhSQAAAAAAAAAAAGAUDE0CAAAAAAAAAAAAo2BoEgAAAAAAAAAAABiFpbMOANgcLdt6yawT4EZhX2ZLsXLVwqwTNsnK+flZJ2wyzxdsCZYumZt1AlCE5wtgY134pkfPOmGT3XL/o2edsEmu+Oe/mnXCKHjtg9nwfguwsbxWA8CwnGkSAAAAAAAAAAAAGAVnmgQAAAAAAAAAAGAQzq/LrDnTJAAAAAAAAAAAADAKhiYBAAAAAAAAAACAUTA0CQAAAAAAAAAAAIyCoUkAAAAAAAAAAABgFAxNAgAAAAAAAAAAAKNgaBIAAAAAAAAAAAAYBUOTAAAAAAAAAAAAwCgYmgQAAAAAAAAAAABGwdAkAAAAAAAAAAAAMAqGJgEAAAAAAAAAAIBRWDrrAAAAAAAAAAAAAMZhbm7WBYydM00CAAAAAAAAAAAAo2BoEgAAAAAAAAAAABgFQ5MAAAAAAAAAAADAKBiaBAAAAAAAAAAAAEbB0CQAAAAAAAAAAAAwCoYmAQAAAAAAAAAAgFEwNAkAAAAAAAAAAACMgqFJAAAAAAAAAAAAYBQMTQIAAAAAAAAAAACjsHTWAQxvfn4+rzv6qHy792yzzTZ59WuOyZ133nnWWetUrTfRPIRqvUm95mq9ieYhVOtNNA+hWu+avnH+eTnhzcfl5FPeN+uU9Vq5YkWOPuqVueTii7LiuhV5xiHPzUP32XfWWetVcb/QPH3VepN6zdV6E81DqNabaB5Ctd5E8xCq9Sb1mqv13naHbXPWSc/Oo156Wl558N65/a22S5LsvOMO+eq/X5SnH/3RGReuXbV1rtabaB5Ctd5E8xCq9Saah1CtN6nXXK030TyEar1JzWa2XHNzc7NOYORmfqbJ1to+rbWF1tpBN7j+/Nbaqeu4z7LW2rMn39+qtfbkyfenttYecSM1/aS1dkZr7V9aa19ure352253c/H5z3021117Xd73gQ/l8BcfkTe98dhZJ61Xtd5E8xCq9Sb1mqv1JpqHUK030TyEar2rveeUd+Xoo16Za6+9btYpG/TJj38s22+/Q07+2/fnzSeelOOOPWbWSRtUcb/QPH3VepN6zdV6E81DqNabaB5Ctd5E8xCq9Sb1miv1Ll2yVd76kkfl19euTJI8/eiP5oAXvy9P+qu/yy+uviYvf9tnZly4bpXWOanXm2geQrXeRPMQqvUmmodQrTep11ytN9E8hGq9Sc1mgGmZ+dDkxIVJrh+abK3dJ8l267n9jkmePfl+tySPmULT53vv+/Te907yqiRHT+ExZuLrXzsnD3zwQ5Iku+2+Ry644JszLlq/ar2J5iFU603qNVfrTTQPoVpvonkI1XpXu9NOO+W440+YdcZG2e/hB+S5f37Y4oWFZMmSzf+E8RX3C83TV603qddcrTfRPIRqvYnmIVTrTTQPoVpvUq+5Uu+xz3tY3vmxc3LJ5b/8jev/6hl75+1//6+59OdXz6hswyqtc1KvN9E8hGq9ieYhVOtNNA+hWm9Sr7lab6J5CNV6k5rNANOyufzX1vOStNba9r33K5M8NclpSe7cWntBkidkcYjyZ0ken+Qvk9yztfaqJA9Jsntr7TmTbT23tfbyJNsneV7v/auttRcmeXKShSSnT7b9ud77Hq21ByT5ZJJbJ/mdJO9O8oYb9N0yyU+yGHlGkkN77xe21g7N4gDnqUk+mORHSe6W5Ku99+e11m6b5D1Jdkgyl+Tpk+28e/J4SXJY7/0bv+0Cborly6/OzW9+s+svL9lqSVauXJmlSzeX3eE3VetNNA+hWm9Sr7lab6J5CNV6E81DqNa72n77H5CLL/rxrDM2yrbbLv7/RMuXL8+RL31RDl09QLkZq7hfaJ6+ar1JveZqvYnmIVTrTTQPoVpvonkI1XqTes1Vep96wG756S9+lc/+6/fysic/6Prrb7vDttlnr9/Nyzbjs0wmddZ5tWq9ieYhVOtNNA+hWm+ieQjVepN6zdV6E81DqNab1GwGmJbN5UyTSfKRJE9orc0luV+Ss7LYd+skD+u93z+LQ573TfK6JP/ee3/t5PvP995PnmznnN77vklOSHJwa+2eSZ6U5MFZHLB8XJLbJLm8tbZTkgOT/DDJ72fxjJV/P9nOvpOP5z47yd9mcdhyfXZN8qxJ+yNbazsmeWWSf+y9PzDJEZOf/UUWBzb/Z5LnJHn7f2u1fgvbbXezLF++/PrL8wvzm/WLYLXeRPMQqvUm9Zqr9Saah1CtN9E8hGq9VV126SV5/iEH58BH/2EOeOSjZ52zQRX3C83TV603qddcrTfRPIRqvYnmIVTrTTQPoVpvUq+5Su+fHbhH9vv9u+bTxz8tu919x7z7FY/N7W+5XR6/9+/lQ5/9ZubnF2aduF5V1nm1ar2J5iFU6000D6Fab6J5CNV6k3rN1XoTzUOo1pvUbAaYls1paPIDWfyI7ocm+cLkuvkk1yX5YGvt3UnulGTrDWznnMnXS5Nsm+TeSXZO8rnJn1sn2SWLw5GPTPLAJMcm2T/Jo/L/hiZXfzz3HyTZM8nprbWb3uCx5tb4/ju991/23lcluSTJsiQtydlJ0ns/q/d+WpL7JHnm5IyV70xyqw38Pje6PffcK18888wkyfnnnZtddtl16IRNUq030TyEar1JveZqvYnmIVTrTTQPoVpvRZdf/rMc9rxD8oLDX5LHPO6PZp2zUSruF5qnr1pvUq+5Wm+ieQjVehPNQ6jWm2geQrXepF5zld79X/TePPxF780BL35fzv/OpXnWG/4hl12xPPvuddd85qvfmXXeBlVZ59Wq9Saah1CtN9E8hGq9ieYhVOtN6jVX6000D6Fab1KzGWBaNpuR8d7791pr2yU5LMkrktw1yS2SPK73fv/W2rZZHIicy+Iw5eqBzzW/TxY/gvs3Np3kgiQH9t4XWmsvTnJ+knOz+DHdP0vyqSSfSXJl7/2y1trv3WAbl63x/TVJ7pDkwiR7JbloHY+bJN/K4pkxz2utPTSLQ5kXJnl/7/0DrbXbJXn2+lfmxrfvw/bP2Wd/KU9/ykFZWFjIa495/dAJm6Rab6J5CNV6k3rN1XoTzUOo1ptoHkK13opOfdfJueqqK3PKye/IKSe/I0ly/NtOyrJly2Zctm4V9wvN01etN6nXXK030TyEar2J5iFU6000D6Fab1KvuVrvDe1y51vn+xdfMeuMDaq2ztV6E81DqNabaB5Ctd5E8xCq9Sb1mqv1JpqHUK03qdkMMC1zCwuz/RiN1to+SQ7tvR/UWnthkqf13u/XWntEkmdm8aO0bzK5+bVJ3p3Fj/L+cpJPJ/mbJJ9NclKSPZKc3nv/1OT+B/XeD26tvSyLH8t9kyRfTfLC3vuq1tpXkryn937i5GO4P9x7P37S9OEk/55kVZKbJzmx935qa+2RSd6UxY/0vmjy9dTJ4z5g8jt9OYtnzVye5JTJ/Rey+PHdV05+hx2yOBR6VO/9H9e3RtesXOtAJgDAaKxcVetwaOX8/KwTNtmyrZfMOgEAALgR3HL/o2edsEmu+Oe/mnUCAABwI1m29Dc+sRXW6YNfv6jWf/xjUH+65x2n/lwy86FJNszQJAAwdoYmp8/QJAAAbBkMTQIAALNiaJKNZWiS9RliaHKrDd8EAAAAAAAAAAAAoD5DkwAAAAAAAAAAAMAoLJ11AAAAAAAAAAAAAOPgLH/Mmn0QAAAAAAAAAAAAGAVDkwAAAAAAAAAAAMAoGJoEAAAAAAAAAAAARsHQJAAAAAAAAAAAADAKhiYBAAAAAAAAAACAUTA0CQAAAAAAAAAAAIyCoUkAAAAAAAAAAABgFAxNAgAAAAAAAAAAAKNgaBIAAAAAAAAAAAAYhaWzDgAAAAAAAAAAAGAc5ubmZp3AyDnTJAAAAAAAAAAAADAKhiYBAAAAAAAAAACAUTA0CQAAAAAAAAAAAIyCoUkAAAAAAAAAAABgFAxNAgAAAAAAAAAAAKNgaBIAAAAAAAAAAAAYBUOTAAAAAAAAAAAAwCgYmgQAAAAAAAAAAABGwdAkAAAAAAAAAAAAMAqGJgEAAAAAAAAAAIBRWDrrAABgelauWph1wiZbumRu1glshqrtF0uXLJl1AgAAMFJX/PNfzTphk9zykW+cdcImu+ITL5t1AoyS9zqn75oVq2adsMmWbe19OACoqNZRElsiZ5oEAAAAAAAAAAAARsHQJAAAAAAAAAAAADAKhiYBAAAAAAAAAACAUTA0CQAAAAAAAAAAAIyCoUkAAAAAAAAAAABgFAxNAgAAAAAAAAAAAKNgaBIAAAAAAAAAAAAYBUOTAAAAAAAAAAAAwCgYmgQAAAAAAAAAAABGwdAkAAAAAAAAAAAAMApLZx0AAAAAAAAAAADAOMzNzc06gZFzpkkAAAAAAAAAAABgFAxNAgAAAAAAAAAAAKNgaBIAAAAAAAAAAAAYBUOTAAAAAAAAAAAAwCgYmgQAAAAAAAAAAABGwdAkAAAAAAAAAAAAMAqGJgEAAAAAAAAAAIBRMDQJAAAAAAAAAAAAjIKhSQAAAAAAAAAAAGAUls46AAAAAAAAAAAAgHFwlj9mzdDkCM3Pz+d1Rx+Vb/eebbbZJq9+zTG58847zzprnar1JpqHUK03qddcrTfRPKRvnH9eTnjzcTn5lPfNOmWDKq5xteZqvYnmIVTrTTQPoVpvUq+5Wm+ieQjVehPNQ6jWm2geQrXepF5ztd6kXvNtd9g2Z73taXnUkX+XK355Td72oofnljdfliVbbZVn/Z9P5PuX/GLWif9FtTVONA+hWm9SsznxXue0rVyxIkcf9cpccvFFWXHdijzjkOfmofvsO+us9aq2ztV6k3rN1XoTzUOo1pvUbAaYlpkO7rbW9mmtLbTWDrrB9ee31k5trX10Pfc9o7V2j9/ise/SWvvy5PutWmt/0Vr7wmS7/19r7T7/3W1v7j7/uc/mumuvy/s+8KEc/uIj8qY3HjvrpPWq1ptoHkK13qRec7XeRPNQ3nPKu3L0Ua/MtddeN+uUjVJxjas1V+tNNA+hWm+ieQjVepN6zdV6E81DqNabaB5Ctd5E8xCq9Sb1mqv1JrWaly7ZKm89/OH59bUrkySve/be+dDnv5X9jzg9R536hbSdbjXjwrWrtMaraZ6+ar1JzWbvdU7fJz/+sWy//Q45+W/fnzefeFKOO/aYWSdtULV1rtab1Guu1ptoHkK13qRmM8C0bA5nO70wyfVDk5Nhxe2SpPf+hIEaXp7kNkn27r3vM7n8D621rQd6/EF9/Wvn5IEPfkiSZLfd98gFF3xzxkXrV6030TyEar1JveZqvYnmodxpp51y3PEnzDpjo1Vc42rN1XoTzUOo1ptoHkK13qRec7XeRPMQqvUmmodQrTfRPIRqvUm95mq9Sa3mY5+zT975T+fmksuvTpL8wb3umDve9ub5+LFPzEH73jNnnv+jGReuXaU1Xk3z9FXrTWo2e69z+vZ7+AF57p8ftnhhIVmyZPP/IMRq61ytN6nXXK030TyEar1JzWaAadkcjkrPS9Jaa9v33q9M8tQkpyW5c2vt0t77jq21+yd5cxaHPC9K8pTJfV/dWrt9Focs/zTJfyY5KclOSe6Q5B97769sre2U5OQkN03y6yTPuUHDc5L8j977fJL03v+1tXbf3vuK1tqeSU5IsirJNUkOmXR8MMmPktwtyVd7789rrR2V5HeT3C7Jzkle3Hv/dGtt/yTHTO5/eZJnJtkjyf9Ocl2Sk3vvg53zf/nyq3Pzm9/s+stLtlqSlStXZunSzWF3+K+q9Saah1CtN6nXXK030TyU/fY/IBdf9ONZZ2y0imtcrblab6J5CNV6E81DqNab1Guu1ptoHkK13kTzEKr1JpqHUK03qddcrTep0/zU/e+Vn175q3z2nB/kZQfdP0my8+1vkSt+eU0edeSH84qn/EGOeOL9cvR7vzTj0v+qyhqvSfP0VetNajZ7r3P6tt12uyTJ8uXLc+RLX5RDVw9QbsaqrXO13qRec7XeRPMQqvUmNZsBpmVzONNkknwkyRNaa3NJ7pfkrBv8/KQkz+y93z/Jx5P83uT6j/fe903yySR/nMVhyS/33g+YbOfQye2OS/I3k7NIHpfkhucY3rb3fsWaV/TeL598+84kL+i9753kxCR/Pbl+1yTPmjzOI1trO06uv7b3fmCSw5O8ePI7nZzkCZNt/EuSV05uu6z3/pAhByaTZLvtbpbly5dff3l+YX6zfhGs1ptoHkK13qRec7XeRDNrV3GNqzVX6000D6Fab6J5CNV6k3rN1XoTzUOo1ptoHkK13kTzEKr1JvWaq/UmdZr/7BH3yX573SWffuOTstvdbpd3v+yRWTW/kI+f/Z0kySe+/N3steuOG9jKbFRZ4zVpnr5qvUnN5mqqrvFll16S5x9ycA589B/mgEc+etY5G1Rtnav1JvWaq/UmmogCbGQAACAASURBVIdQrTep2QwwLZvL0OQHsvgR3Q9N8oW1/HzH3vu3kqT3/u7e+9cm158z+Xppkm2T/DzJfVtrpyU5PslNJj+/T5K/aK2dkeRVSW5/g+1f0Vq7xZpXtNYeP7nud3rv506uPjPJvSbff6f3/sve+6oklyRZNrn+65OvP5pcd5skV/XeL1rLNvo61mOq9txzr3zxzDOTJOefd2522WXXWWRstGq9ieYhVOtN6jVX6000s3YV17hac7XeRPMQqvUmmodQrTep11ytN9E8hGq9ieYhVOtNNA+hWm9Sr7lab1Knef8jTs/DX3p6DnjZh3L+d3+SZ73xE/nkV76bA+531yTJg+9zp3zrP38248q1q7LGa9I8fdV6k5rN1VRc48sv/1kOe94hecHhL8ljHvdHs87ZKNXWuVpvUq+5Wm+ieQjVepOazQDTslmMjPfev9da2y7JYUlekeSuN7jJxa21XXrv/9Fa+19Jvj25fuEGtzs4yS96789trd09yXMmZ3q8MMlxvfezWmv3SLL3De73nix+1PdLe+8LrbUHZvGMkm3y2Lv13s+f3G9dj511XP+zJLdord2h937JDbYxv45tTNW+D9s/Z5/9pTz9KQdlYWEhrz3m9bPI2GjVehPNQ6jWm9RrrtabaGbtKq5xteZqvYnmIVTrTTQPoVpvUq+5Wm+ieQjVehPNQ6jWm2geQrXepF5ztd6kZvNqR550Rk58yQF5zqP3yJW/ujYHv+GfZp20VhXXWPP0VetNajZXU3GNT33XybnqqitzysnvyCknvyNJcvzbTsqyZcs2cM/ZqbbO1XqTes3VehPNQ6jWm9RsBpiWuYWFdc3+TV9rbZ8kh/beD2qtvTDJ03rv92utPSKLZ558RO99x9bafbM4xDifxbM6/lmST0/ue2Fr7dAkOyb5uyyetfLKJNdm8eO698viGSffnsUzP940ix+dfUmS03vvD2itLUlydJJ9k6yY/Dm89/6N1tqeSd6SZC7Jyix+JPf86vtOfo8vT3oPTnJp7/0dk+HMd/Te92mtPWyy/fkkV0xud+/Vv/uG1umalesc0ASA9Vq5qt5LyNIlc7NOAAAAAAZyy0e+cdYJm+yKT7xs1gkwSt7rnL5rVqyadcImW7b1klknALCGZUtT68WPmfnoeZfUO7hjME/Y/Q5Tfy6Z6dAkG8fQJAD/Xd5IBAAAADZnhiaBjeW9zukzNAnAb8vQJBvL0CTrM8TQ5FbTfgAAAAAAAAAAAACAzcHSWQcAAAAAAAAAAAAwDnNzTkrKbDnTJAAAAAAAAAAAADAKhiYBAAAAAAAAAACAUTA0CQAAAAAAAAAAAIyCoUkAAAAAAAAAAABgFAxNAgAAAAAAAAAAAKNgaBIAAAAAAAAAAAAYBUOTAAAAAAAAAAAAwCgYmgQAAAAAAAAAAABGwdAkAAAAAAAAAAAAMAqGJgEAAAAAAAAAAIBRWDrrAAAAAAAAAAAAAMZhbtYBjJ4zTQIAAAAAAAAAAACjYGgSAAAAAAAAAAAAGAVDkwAAAAAAAAAAAMAoGJoEAAAAAAAAAAAARsHQJAAAAAAAAAAAADAKhiYBAAAAAAAAAACAUTA0CQAAAAAAAAAAAIyCoUkAAAAAAAAAAABgFAxNAgAAAAAAAAAAAKOwdNYBAAAAAAAAAAAAjMPc3KwLGDtDkwCwBVu6pN7R5spVC7NO2CQV15jpu/TKa2adsMl23H7ZrBMAgC1IteP6xLE9zMoVn3jZrBM22Wlf++GsEzbZU/a686wTtni/+NWKWSdssh223XrWCZvEa/X0Ldt6yawTNlnF486vfv/ns07YJA+8+61nnQCjVfH4Ysdb1Dq+AMbLx3MDAAAAAAAAAAAAo2BoEgAAAAAAAAAAABgFQ5MAAAAAAAAAAADAKBiaBAAAAAAAAAAAAEbB0CQAAAAAAAAAAAAwCoYmAQAAAAAAAAAAgFEwNAkAAAAAAAAAAACMgqFJAAAAAAAAAAAAYBQMTQIAAAAAAAAAAACjsHTWAQAAAAAAAAAAAIzDVpmbdQIj50yTAAAAAAAAAAAAwCgYmgQAAAAAAAAAAABGwdAkAAAAAAAAAAAAMAqGJgEAAAAAAAAAAIBRMDQJAAAAAAAAAAAAjIKhSQAAAAAAAAAAAGAUDE0CAAAAAAAAAAAAo2BoEgAAAAAAAAAAABgFQ5MAAAAAAAAAAADAKCyddQAAAAAAAAAAAADjMDc36wLGzpkmAQAAAAAAAAAAgFEwNAkAAAAAAAAAAACMgqHJEZqfn8/Rr3lVnvbkJ+VZBz8tP/zP/5x10npV6000D6Fab1KvuVpvonkI1XrX9I3zz8tznvm0WWdslGrrXK03qdmcJKe/99150SFPy58/46B86mMfnXXOelVcY83TV603qddcrTfRPIRqvYnmIVTrXZNj++mp1pvUa67Wm2ietou/86188HVHJEku/cF/5L2vfkE+cPSL89n3vjUL8/Mzrlu3Smu8WsXmJLni55fnjx+1X/7zB9+bdcoGVVzjas3VepOazUmdY87v9wvy13/5579x3Vf/5TP5Py8/ZEZFG6faflGtN9E8hGq9a6p0fAEwLeWHJltr+7TWFlprB93g+vNba6f+Fts9uLX2w9baGa21M1tr/9Ja23fys0e01p6zjvvdpbX25Rtct2Nr7cT/bsuN7fOf+2yuu/a6vO8DH8rhLz4ib3rjsbNOWq9qvYnmIVTrTeo1V+tNNA+hWu9q7znlXTn6qFfm2muvm3XKRqm2ztV6k5rN533tX/Pv3zg3f33Se3Lc207JTy+7bNZJ61VxjTVPX7XepF5ztd5E8xCq9Saah1CtdzXH9tNVrTep11ytN9E8TV/5pw/lU+/+66xcsfic9ulTjs9+T3lenvxXx+cmN90u/37252dcuG5V1nhNFZtXrlyR497wmtxk2bJZp2yUimtcrblab1Kzucox52c++v68/21vyIrr/l/nj77Xc9ZnP5aFhRmGbYRq+0W13kTzEKr1rlbt+AJgWsoPTU5cmOT6ocnW2n2SbHcjbPcDvfd9eu8PTfLEJG9vre3Ye/9U7/3kjd1I7/3S3vvzb4SeG8XXv3ZOHvjghyRJdtt9j1xwwTdnXLR+1XoTzUOo1pvUa67Wm2geQrXe1e6000457vgTZp2x0aqtc7XepGbzOV85K3e52y55zZEvzqte/sLc/0EPnXXSelVcY83TV603qddcrTfRPIRqvYnmIVTrXc2x/XRV603qNVfrTTRP0w63/5087vBXX3/56p//LHfc9V5Jkjvueq/8+NubZ3dSZ43XVLH5xDcfl8c+4Um5zW1uO+uUjVJxjas1V+tNajZXOea8zY53zHOPfMP1l6++6sr83/edlD951otmWLVxqu0X1XoTzUOo1rtateMLgGlZOuuAG8l5SVprbfve+5VJnprktCR3bq1d2nvfMYs3OD3JO5LcJckzszg0+uokt0rykiSrknyx937kDR+g935Za+0jSR7dWluZ5B699yNba69M8rgsruXbk3x68lhLkpya5IIkpyc5vff+gNba3kleN3ms7yZ5bu99xRTWZJ2WL786N7/5za6/vGSrJVm5cmWWLt08d4dqvYnmIVTrTeo1V+tNNA+hWu9q++1/QC6+6Mezztho1da5Wm9Ss/nKX/wiP7n04rz2uLfm0osvylH/67C864P/kLm5uVmnrVXFNdY8fdV6k3rN1XoTzUOo1ptoHkK13tUc209Xtd6kXnO13kTzNLX7PiRX/vTS6y9vf9s75IffOi93/r3d852vfzkrrr1mhnXrV2WN11St+ZMf+7/Z4Za3yv3+4EE57dR3zjpno1Rb46Rec7XepGZzlWPOvR74P3P5ZZckSeZXrcr73/r6/PEzD8s229xkxmUbVm2/qNabaB5Ctd6k5vEFwLRsKWeaTJKPJHlCa20uyf2SnLWB21/Re39wkq8neU2S/SaX79ha238d97ksyW1WX2it7ZnkwCT3nzzmrknmsjhAeVqSs3vvx65x+7kk70zyhN773kkuSnLwJv6ev7XttrtZli9ffv3l+YX5zfqFu1pvonkI1XqTes3VehPNQ6jWW1W1da7Wm9RsvsX22+d/3P+B2XrrrbPTznfJ1tvcJFde8fNZZ61TxTXWPH3VepN6zdV6E81DqNabaB5Ctd6qqq1ztd6kXnO13kTzkA58zkvzlY+dntPf8LJse4sdctObbT/rpHWquMbVmj/xsb/Pv33l7Bz+3IPznW/3vP7Vf5HLf/azWWetV7U1Tuo1V+tNajZX9MPv9vzk4h/ng+94Y9513Kty6Y++nw+/682zzlqnavtFtd5E8xCq9SY1jy8ApmVLGpr8QBY/ovuhSb6wjtuseRqePvl69yS3TfKJ1toZSe6Z5G7ruP/OSdb834pakq/23lf13q/rvR+RZCHJ7klul+RmN7j/bZPcIcmHJ4/18Mk2B7Xnnnvli2eemSQ5/7xzs8suuw6dsEmq9Saah1CtN6nXXK030TyEar1VVVvnar1JzeZ77b5n/u0rZ2VhYSGX//QnuebXv87Nt99h1lnrVHGNNU9ftd6kXnO13kTzEKr1JpqHUK23qmrrXK03qddcrTfRPKTvnfuVPPp5r8hBr3hjrrn6qtzl3nvNOmmdKq5xteYTTn5P/ubkU/OWk07N3Xdt+YvXvD63vs1tNnzHGaq2xkm95mq9Sc3miu6y6z3zqreelpe87m159ktfmx13+t088dmb78d0V9svqvUmmodQrTepeXwBMC2b95j7Jui9f6+1tl2Sw5K8IsldJz/aurV2syTXJbnXGneZn3z9fpIfJdm/976itXZwknOT7LHm9ltrd0jy2CTHJHnM5OoLkzyvtbZVkiVJPpHkBUnOSfKoJF9trX0qyVWT2/8si0OXj+29X9lae0ySq2+EX3+T7Puw/XP22V/K059yUBYWFvLaY14/dMImqdabaB5Ctd6kXnO13kTzEKr1VlVtnav1JjWbH/CgvfPNc7+Ww571lMwvzOcFR7wiS5YsmXXWOlVcY83TV603qddcrTfRPIRqvYnmIVTrraraOlfrTeo1V+tNNA/plre/Yz507MuydJtlufM9d8/d9rj/rJPWqeIaV2yupuIaV2uu1pvUbGb6qu0X1XoTzUOo1gvAb5pbWFiYdcNvpbW2T5JDe+8HtdZemORpvff7tdYekcUzT343yZOSfC+Lg41vTHKXJPfovR852cZTkzx/8vMfJHlGkicmee3kfquyeJbKo3rvZ04GK+/Rez+ytfaKLA5RbpXk7UnOSHJ67/0BrbUHJ3nr5PHfM7nu4UleNbn9VUme3nv/yfp+x2tWpvY/JADYBCtX1XrZW7pkbsM3YnQuvfKaWSdssh23XzbrBABgC1LtuD5xbA9svNO+9sNZJ2yyp+x151knbPF+8asVs07YZDtsu/WsE+C3VvG486vf//msEzbJA+9+61knwGhVPL7Y8RZb+5drNso/ffOyei/iDObR97791J9Lyg9NjoGhSQDGpNqbXP7DKmtjaBIAGLtqx/WJY3tg4xmaZG0qDjUYmmRLUPG409AksLEqHl8YmmRjffybP6n3Is5gHnXv2039uWSraT8AAAAAAAAAAAAAwObA0CQAAAAAAAAAAAAwCoYmAQAAAAAAAAAAgFEwNAkAAAAAAAAAAACMgqFJAAAAAAAAAAAAYBQMTQIAAAAAAAAAAACjYGgSAAAAAAAAAAAAGAVDkwAAAAAAAAAAAMAoGJoEAAAAAAAAAAAARsHQJAAAAAAAAAAAADAKS2cdAAAAAAAAAAAAwDjMzc26gLFzpkkAAAAAAAAAAABgFAxNAgAAAAAAAAAAAKNgaBIAAAAAAAAAAAAYBUOTAAAAAAAAAAAAwCgYmgQAAAAAAAAAAABGwdAkAAAAAAAAAAAAMAqGJgEAAAAAAAAAAIBRMDQJAAAAAAAAAAAAjIKhSQAAAAAAAAAAAGAUls46AAAAAAAAAAAAgHHYKnOzTmDknGkSAAAAAAAAAAAAGIW5hYWFWTewAVdfW+sf0tIlpsH5r1auKrUbZ+X8/KwTNtmyrZfMOoHNULW/e4nXEQAAAIAb263/9G9nnbBJLv/gM2adwGbIe53TZ40B+G0tW+r0gWycT13w03oHHgzmEfe67dSfS5xpEgAAAAAAAAAAABgFQ5MAAAAAAAAAAADAKBiaBAAAAAAAAAAAAEbB0CQAAAAAAAAAAAAwCoYmAQAAAAAAAAAAgFEwNAkAAAAAAAAAAACMgqFJAAAAAAAAAAAAYBQMTQIAAAAAAAAAAACjsHTWAQAAAAAAAAAAAIzD3NysCxg7Z5oEAAAAAAAAAAAARsHQJAAAAAAAAAAAADAKhiYBAAAAAAAAAACAUTA0CQAAAAAAAAAAAIyCoUkAAAAAAAAAAABgFAxNAgAAAAAAAAAAAKNgaBIAAAAAAAAAAAAYBUOTAAAAAAAAAAAAwCgYmgQAAAAAAAAAAABGwdAkAAAAAAAAAAAAMApLZx0AAAAAAAAAAADAOMzNzbqAsXOmSQAAAAAAAAAAAGAUDE0CAAAAAAAAAAAAo+DjuUfqG+eflxPefFxOPuV9s07ZoPn5+bzu6KPy7d6zzTbb5NWvOSZ33nnnWWetl+bhVNqXV65YkaOPemUuufiirLhuRZ5xyHPz0H32nXXWOlXcJzQPp9LfvYprXK25Wm+ieQjVehPNQ6jWm9RrrtabaB5Ctd5E8xCq9Saah1CtN6nXXK030TyEar23vcWyfPF/PyZ/ePSns9VWyQnPfVDmknz30qvy/Ld/KavmF2aduFbV1rlab1KzOfFe51Cs8/RU603qNVfrTTQPoVpvUrMZYFq22DNNttb2aa1d2VrbaY3rjm2tHdlaO3E99/tBa23ZDa47o7X2/7N37/GW1nXd/997ZoMDA4KoRSqCIn7VVETTulEMOYiaph1MUjOtFFQOGh2oUEYBwwRFSVMElBs1rKyfloe6FYlEzRQ5KpdaHkFEQUEGmJk9e//+2GvTjpiTzLqu+cx6Ph+Pecysta51rdf+zrX3rL3ms6/1kLvY85LW2jajrvPuyr7uqnPOPjMnrDguq1atHjJjo53/iY9n9arVOfd978/Rrzwmp77h5KGTNkhzP6odyx/98D9mp512zhnvek9Oe9s7csrJJw6dtF4VjwnN/aj2uVdxjas1V+tNNPehWm+iuQ/VepN6zdV6E819qNabaO5Dtd5Ecx+q9Sb1mqv1Jpr7UKl3eulU3nLYvrl19UySZMVvPiYr3veFHPSqjyRJnvZzu63v7oOqtM5Jvd6kZrPXOvthncerWm9Sr7lab6K5D9V6k5rNAOOy1Q5NjqxK8q7W2tSi667tuu5lA7T8aZKlAzzu/3K/3XbLKW86feiMjfbFi7+QfZ+wX5LkkXs/KldeecXARRumuR/VjuUDn3xIDnv5UfMX5pKlS7fsk/1WPCY096Pa517FNa7WXK030dyHar2J5j5U603qNVfrTTT3oVpvorkP1XoTzX2o1pvUa67Wm2juQ6XeP3/B43LWv1yV7/7wliTJc0/9ZC768veyzfSS/PTO2+WmW9YMXLhuldY5qdeb1Gz2Wmc/rPN4VetN6jVX600096Fab1KzGWBctvahyfOT3JDk5YuvbK19dvT701tr/9Fa+3xr7YzW2pJF2xzeWvv71trd7mzHo7NGvqe19unW2r+31p4zuv72s1KO9rGitfa7SXZNsnCGyb1aax9trX2htbZic3/QG3LgwYdkenrLHtZabOXKm7PjjjvcfnnpkqWZmZkZsGjDNPej2rG8/fbLs3z58qxcuTLH/sErcvjCAOUWquIxobkf1T73Kq5xteZqvYnmPlTrTTT3oVpvUq+5Wm+iuQ/VehPNfajWm2juQ7XepF5ztd5Ecx+q9D5//wfl+zfelo9fes3t183OzmW3ey3P59/4K7nnjsty+TduGLBw/aqs84JqvUnNZq919sM6j1e13qRec7XeRHMfqvUmNZsBxmVrH5pMkpcmeWVr7UGLr2ytTSf5yyS/1HXdzyX5WpL7jW4+Msl+SZ7ddd2qdez3sCTf77pu3yQHJTmxtXavO9uw67qzklyb5NDRVcuSPGv0GEf8pB/YpFi+fIesXLny9suzc7Nb/DdWmlmX71373bzsxS/MU5/+jBzytKcPnbNeFY8JzdyZimtcrblab6K5D9V6E819qNab1Guu1pto7kO13kRzH6r1Jpr7UK03qddcrTfR3Icqvb/1pL1ywN73yUdXPCWP3GOXvPPI/fLTO2+Xb/9gZfY+6gM581+uysm//bihM9epyjovqNab1Gyuxhr3o9o6V+tN6jVX600096Fab1KzGWBctvqhya7rrk/yiiTn5H9+vPdK8sOu664bbfcXXdd9a3TbQUl27rpu7Xp2/dAkF47u++MkX0qy5x22mbrjnUau6LpuVdd1tyQxtr8B++zz6HzqwguTJJddekn22uvBAxdtmGbuzPXX/yBHvfTFOeLo388vP+vXhs7ZoIrHhGbuTMU1rtZcrTfR3IdqvYnmPlTrTeo1V+tNNPehWm+iuQ/VehPNfajWm9RrrtabaO5Dld5Djv9onnL8R/PUFR/LZd+4IS8+/d9y+mH7Zs9d754kufm2mczOzQ1cuW5V1nlBtd6kZnM11rgf1da5Wm9Sr7lab6K5D9V6k5rNAOMyESPjXdf9Y2vtV5K8MMkfja6+LsnOrbVduq67obX2liTvGd32zCRnttYO77ru7evY7Zczf6bIf2it7ZjkEUm+nuS2JD+T5Kokj05y9Wj72fz30OaW+6rBFuiAgw7OZz5zUV7wvEMzNzeX1574uqGTNkgzd+bdZ56Rm266MWef8facfcb8l5Y3vfUdWbZs2cBld67iMaGZO1Nxjas1V+tNNPehWm+iuQ/VepN6zdV6E819qNabaO5Dtd5Ecx+q9Sb1mqv1Jpr7UK13sVP/4fK84+VPyOqZ2dy6eiYv/6uLhk5ap2rrXK03qdlcjTXuR7V1rtab1Guu1pto7kO13qRmM8C4TM1twT/1d1e01vZPcnjXdYeOLt89yeVJjh9d/wuttacmeXWStUm+mOSozA8+PiTJ9kk+l+SpSd6Z+TNT3jba/QVJ/nR0/Z5Jtkvylq7rzmmtPS3JqUm+lfmByW91XbeitXZOkt2TvCbJYYu6ru26btf1fSw3r6r1lzS9dF0n2GSSzawtdRhnZnZ26IRNtmybpUMnsAWq9rmX+HcEAAAAYHO752++a+iETXL9X79o6AS2QF7rHD9rDMBdtWx6ne/ICv/D//vyD+o98aA3Bz/0XmP/WrLVDk1uTQxNsjWo9o22oUm2FtU+9xL/jgAAAABsboYm2Rp4rXP8rDEAd5WhSTaWoUnWp4+hySUb3gQAAAAAAAAAAACgPkOTAAAAAAAAAAAAwEQwNAkAAAAAAAAAAABMBEOTAAAAAAAAAAAAwEQwNAkAAAAAAAAAAABMBEOTAAAAAAAAAAAAwEQwNAkAAAAAAAAAAABMBEOTAAAAAAAAAAAAwEQwNAkAAAAAAAAAAABMhOmhAwAAAAAAAAAAAJgMS6aGLmDSOdMkAAAAAAAAAAAAMBEMTQIAAAAAAAAAAAATwdAkAAAAAAAAAAAAMBEMTQIAAAAAAAAAAAATwdAkAAAAAAAAAAAAMBEMTQIAAAAAAAAAAAATwdAkAAAAAAAAAAAAMBEMTQIAAAAAAAAAAAATwdAkAAAAAAAAAAAAMBGmhw4AAAAAAAAAAABgMkxlaugEJpwzTQIAAAAAAAAAAAATwdAkAAAAAAAAAAAAMBEMTQIAAAAAAAAAAAATYXroADZseunU0Alwl1U7jq/+4aqhEzbZ7vfafugEtkDVPveSZGbt3NAJm6TiGgPAJLn2xtuGTthku+60bOgEgLGo9v1ekszMzg6dsMmWbbN06AS2QNf/9YuGTtgkh5x+0dAJm+yfj3z80AlbPa/DjZ81BgBgUjjTJAAAAAAAAAAAADARDE0CAAAAAAAAAAAAE8HQJAAAAAAAAAAAADARDE0CAAAAAAAAAAAAE8HQJAAAAAAAAAAAADARDE0CAAAAAAAAAAAAE2F66AAAAAAAAAAAAAAmw9TU0AVMOmeaBAAAAAAAAAAAACaCoUkAAAAAAAAAAABgIhiaBAAAAAAAAAAAACaCoUkAAAAAAAAAAABgIhiaBAAAAAAAAAAAACaCoUkAAAAAAAAAAABgIhiaBAAAAAAAAAAAACaCoUkAAAAAAAAAAABgIhiaBAAAAAAAAAAAACaCoUkAAAAAAAAAAABgIkwPHQAAAAAAAAAAAMBkmMrU0AlMOGeaBAAAAAAAAAAAACaCoUkAAAAAAAAAAABgIhiaBAAAAAAAAAAAACbC9NAB9G92djYnnbAiX+m6bLvttjn+NSfm/rvvPnTWOlXrTTT3oVpvkszMrMlpr3t1rrv2mixZsjRH/OGrcr/dHzB01jpVXGPN41etd7HLL7s0p592Ss44+9yhUzao2jpX600096Fab6K5D9V6k3rN1XqTms1Jct7/PSuf/bcLsmZmTZ7xq7+RpzzjV4dOWqeKa6x5/Kr1Jpr7UK13sSrf882sWZMTVhyX715zddasXpMXvfiwPHH/A4bOWq+Kx0W15mq9SZ3mJVPJHx70oOy2y3aZm0ve+In/zNIlUzn6SQ/I2tlkzdrZvO6fv5of3rJm6NT/pcoaL6Z5/Kr1Jpr7UK03qddcrTfR3IdqvUnNZoBxGduZJltr+7fWrmutXdBa+2Rr7bOttSPXs/1LWmvb3Mn157XWtl3HfY5trT1uAx0vaa1dOOq4qLW2/yZ/MJugtfao1tqrx/kYd9X5n/h4Vq9anXPf9/4c/cpjcuobTh46ab2q9Saa+1CtN0k+/9lPZe3atfmLt52T5/z2S3LumX85dNJ6VVxjzeNXrXfBOWefmRNWkR8AkAAAIABJREFUHJdVq1YPnbJRqq1ztd5Ecx+q9Saa+1CtN6nXXK03qdl86cX/kS9dfkne+I5zcspbz873v/e9oZPWq+Iaax6/ar2J5j5U611Q6Xu+j374H7PTTjvnjHe9J6e97R055eQTh07aoIrHRbXmar1JneZ9H7hLkuSI91+esz79zfze4++fI/d/QN78ya/nFX93RS782vV57s/dd+DKO1dljRfTPH7VehPNfajWm9RrrtabaO5Dtd6kZjPAuIz77bnP77pu/67rnpTkF5Mc01rbeR3b/mmSpXe8suu6Q7uuu9NXu7quO7nrus+t68Fba4cmOTjJgV3X7Z/k+UnOba3daxM/jo3Wdd0lXde9dlz73xy+ePEXsu8T9kuSPHLvR+XKK68YuGj9qvUmmvtQrTdJ7nu/3bN27drMzs7m1ltuzvT0ln2y34prrHn8qvUuuN9uu+WUN50+dMZGq7bO1XoTzX2o1pto7kO13qRec7XepGbzF/7909ljz73ymmNfmVf/0ZH5+cc/ceik9aq4xprHr1pvorkP1XoXVPqe78AnH5LDXn7U/IW5ZOnSLfv1oaTmcVGtuVpvUqf5U/95Q075+NeSJD9997vl5lVr85qPdPna91cmSZYumcrqtbNDJq5TlTVeTPP4VetNNPehWm9Sr7lab6K5D9V6k5rNAOPS5ysyOyZZm2Tv1trxmR/Y3CHJc5Psl2TXJOe11k5L8vokq5OckeSEJA9JsluSM5Nsm+SWJIcmeUOS85J8enTbzknuk+StXdf9VZLDkvx+13VrkqTruq+31h7Vdd31rbU9kpyd+TWYS3JU13WXtta+Ntrfg5N8IslOSR43f/fut1pr706yJsnuSe42evxnJLl/kmeOOg/vuu7Q1tpXk1yUpCX5XpJfS7JnknclmRmtwXO7rvv2ZlrjjbJy5c3Zcccdbr+8dMnSzMzMbLEDXNV6E819qNabJMu23z7XXXtNXvZbv5KbbvxRXnXym4dOWq+Ka6x5/Kr1Ljjw4ENyzdXfGTpjo1Vb52q9ieY+VOtNNPehWm9Sr7lab1Kz+cYf/SjXXXtNXnvKX+baa67Oij8+Kmf+9QczNTU1dNqdqrjGmsevWm+iuQ/VehdU+p5v++2XJ0lWrlyZY//gFTl8YYByC1bxuKjWXK03qdW8di75k0P2yn577pJX/1OXG1bOvxX3z/7MjvnVvX8mR/7t5QMX3rlKa7xA8/hV600096Fab1KvuVpvorkP1XqTms0A4zLuM00eMHpb7POTvDfJkUl+NsnzR2d+/Pskz+667qwk12Z+EDJJlnVdt1/Xdecu2tcpSf6867r/k+TNSfZZdNuDkpzXdd2Tkzw5ye+Prr9Pkv9aHNR13fWL9vfmruuemOToJGeNrt8jyXGZH+Q8Ksnbkvx8kicsOkvmN0aP9eUkD+i67mlJPpD54cnFHpjkVaPmeyd5bObPfPm5JAclOT7zQ5m9Wr58h6xcufL2y7Nzs1v0P4LVehPNfajWmyQf+pv3ZJ/H/Z+8/b0fzJvPfn9Oe92rs3rVqqGz1qniGmsev2q9VVVb52q9ieY+VOtNNPehWm9Sr7lab1Kz+e477ZTH/Py+2WabbbLb7ntkm23vlht/eMPQWetUcY01j1+13kRzH6r1VvW9a7+bl734hXnq05+RQ5729KFzNqjicVGtuVpvUq/5z//5q3n+uy/OHx68Z5ZNL8mTHnyvHHPgnvnjD34pN946M3Tenaq2xonmPlTrTTT3oVpvUq+5Wm+iuQ/VepOazQDj0tfbcx/Qdd0hXdd9JMnVSd4yOmPjk5Jscyf36+7kupbkM0nSdd2Huq77l0W3fS/Js1pr78n8wOPCPr+Z+TM//vdOWjuktfYzSR6a5MLR/i5ZtN31Xdd9a3R2ypVd132p67q5JDcmWTba5uLR7z9K8qXRn3+46PYFP1h0Fslvj24/a3S/jyU5IvNnnOzVPvs8Op+68MIkyWWXXpK99npw3wmbpFpvorkP1XqTZPmOd8/y5fM/ubPjjjtl7dqZzM5umW/9ktRcY83jV623qmrrXK030dyHar2J5j5U603qNVfrTWo2/+ze++Tz//7pzM3N5frvX5fbbr01O+6084bvOJCKa6x5/Kr1Jpr7UK23ouuv/0GOeumLc8TRv59fftavDZ2zUSoeF9Waq/UmdZqf/NB753mPvW+S5LaZ2czNJU/c65751UftmqP/9op898Yt94faq6zxYprHr1pvorkP1XqTes3VehPNfajWm9RsBhiXIUbG35lkz67rftxaOyfJwntXzea/hzjvbIroy5k/U+PHW2vPS7LLotuOSfKZruv+qrX2pCS/NLr+7CSvaq09r+u6mdbagzP/Nt6PGe1vvyQfaq09KvNnukzm36p7QzZmm3Vt98wk/9Z13Wtaa7+Z5I+TvGgj97dZHHDQwfnMZy7KC553aObm5vLaE1/X58Nvsmq9ieY+VOtNkmc++/l5y+tX5NgjficzM2vyW793ZJZtt93QWetUcY01j1+13qqqrXO13kRzH6r1Jpr7UK03qddcrTep2fwLj//FXHHJxTnqd5+X2bnZHHHMn2Tp0qVDZ61TxTXWPH7VehPNfajWW9G7zzwjN910Y84+4+05+4y3J0ne9NZ3ZNmyO/48/paj4nFRrblab1Kn+cKvXp9jD9krb3n2wzO9ZCqnX/D1HHvIg/K9m1blhGc8JEly6dU35l2f+fYG9tS/Kmu8mObxq9abaO5Dtd6kXnO13kRzH6r1JjWb2XotmdrwNjBOU3NzGzv/t2laa/snObzrukPvcP0bkxyQZGXmzxB5fdd1Lx4NUO6e5DVJDlu4X2vtG0kekuR+Sd6RZGmSW5I8P8kbk5yXZFWS05Ncn/mzOD48ycO6rlvVWntlkuckWT267592XfevrbU9Mj/AebfMn5nyyK7rPt9au7brul1Hj734z5ckeUqSkzP/VuAfa62dnOSqruve3Vp7RebPJPnZhY/7Dvc/L8nbM3/GyXMW9byy67qFM1feqdtmNnpIE9hMvvmDW4ZO2GS732v7oRNgs5hZW+ufvemlntEDwJbs2htvGzphk+2605Y7OANwV1T7fi9JZrbgdwlZl2XbbLmD/LCxDjn9oqETNtk/H/n4oRMAAAa3bDr+44yNcuFXbqj3IgG9eeKDdxn715KxDU2y+RiahP4ZmoThVPtPNEOTALBlMzQJsOWo9v1eYmgShmJoEgCgJkOTbCxDk6xPH0OTSza8CQAAAAAAAAAAAEB9hiYBAAAAAAAAAACAiWBoEgAAAAAAAAAAAJgIhiYBAAAAAAAAAACAiWBoEgAAAAAAAAAAAJgIhiYBAAAAAAAAAACAiWBoEgAAAAAAAAAAAJgIhiYBAAAAAAAAAACAiTA9dAAAAAAAAAAAAACTYSpTQycw4ZxpEgAAAAAAAAAAAJgIhiYBAAAAAAAAAACAiWBoEgAAAAAAAAAAAJgIhiYBAAAAAAAAAACAiWBoEgAAAAAAAAAAAJgIhiYBAAAAAAAAAACAiWBoEgAAAAAAAAAAAJgIhiYBAAAAAAAAAACAiWBoEgAAAAAAAAAAAJgI00MHAAAAAAAAAAAAMBmmpoYuYNI50yQAAAAAAAAAAAAwEQxNAgAAAAAAAAAAABPB0CQAAAAAAAAAAAAwEQxNAgAAAAAAAAAAABNham5ubugGNuC2mfhLAgAo5LY1a4dO2GTTS2r9PNX00qmhEwCA9ZhZW+/lLM8vALgrHn7sR4dO2GRXnPzUoRPYwlR8DleR553jd/UNtw6dsMnuu8t2QyfAZrFsOr7IsVE+9dUfeuLBOj1hr3uM/WtJrf8ZBQAAAAAAAAAAAPgJGZoEAAAAAAAAAAAAJoKhSQAAAAAAAAAAAGAiGJoEAAAAAAAAAAAAJoKhSQAAAAAAAAAAAGAiTA8dAAAAAAAAAAAAwGSYGjqAiedMkwAAAAAAAAAAAMBEMDQJAAAAAAAAAAAATARDkwAAAAAAAAAAAMBEMDQJAAAAAAAAAAAATARDkwAAAAAAAAAAAMBEMDQJAAAAAAAAAAAATARDkwAAAAAAAAAAAMBEMDQJAAAAAAAAAAAATARDkwAAAAAAAAAAAMBEMDQJAAAAAAAAAAAATITpoQMAAAAAAAAAAACYDEumpoZOYMI50yQAAAAAAAAAAAAwEQxNAgAAAAAAAAAAABPB0CQAAAAAAAAAAAAwEQxNAgAAAAAAAAAAABPB0CQAAAAAAAAAAAAwEaaHDqB/s7OzOemEFflK12XbbbfN8a85Mfffffehs9apWm+iuQ/VepN6zdV6E819qNabaO5Dtd6kXvPMmjU5YcVx+e41V2fN6jV50YsPyxP3P2DorI1y+WWX5vTTTskZZ587dMoGVTsuknrN1XqTes3VehPNfajWm2juQ7XexTy/GJ9qvUm95mq9ieY+VOtNNI/TkqnkpGc/Ig+89/LMZS6v+sCVefmBe+bed79bkuS+99gul3zzR3nFey8duPR/q7LGi1Vrrta7WKXncEmt3orHRcXm//eRD+bjH/1QkmT16tX5r691ee//9/HssOPdBy67cxXXuFpztd6kZjPAuAxypsnW2v6ttetaaxe01j7ZWvtsa+3I9Wx/xCbs+4WttZPvcN15rbX970Lyuh7rG621ZZt7v+N2/ic+ntWrVufc970/R7/ymJz6hpM3fKcBVetNNPehWm9Sr7lab6K5D9V6E819qNab1Gv+6If/MTvttHPOeNd7ctrb3pFTTj5x6KSNcs7ZZ+aEFcdl1arVQ6dslGrHRVKvuVpvUq+5Wm+iuQ/VehPNfajWu8Dzi/Gq1pvUa67Wm2juQ7XeRPM4HfCwn0qSPOetn82bPvbVHPOUB+cV7700z/urz+Wl7744N906k5M+dNXAlXeuyhovVq25Wu+Cas/hqvVWPC4qNh/8tGfm9aefldefflYe1B6aw4/+oy12YDKpucbVmqv1JjWbAcZlyLfnPr/ruv27rntSkl9Mckxrbed1bHtcj11bvS9e/IXs+4T9kiSP3PtRufLKKwYuWr9qvYnmPlTrTeo1V+tNNPehWm+iuQ/VepN6zQc++ZAc9vKj5i/MJUuX1jhh/P122y2nvOn0oTM2WrXjIqnXXK03qddcrTfR3IdqvYnmPlTrXeD5xXhV603qNVfrTTT3oVpvonmcPn7ldTnu7+bb7nOP7XLTbWtuv+3oJ++Vcy/6Zr7/41VD5a1XlTVerFpztd4F1Z7DVeuteFxUbF7wlauuzLe+/p956i//+tAp61Vxjas1V+tNajYDjMuW8r+tOyZZm2S31to/JJlKcn2S30lyRJJdWmtvS3J0krcn2SvzA5/HdV13QWvtiiRfSbI6ycfW9SCttRcmeUjXdceOzhB5Vdd1e7TWLkhyaZKHJ7k5yb8lOSTJzkmenOSZSZ416rxXktd2XfeB0W7/qrX2gNGff2W07e+M+o5PskuS3x99fJ8aPfZFSV7Sdd2VrbWnJnlG13Uv+4lXbxOtXHlzdtxxh9svL12yNDMzM5me3lIOh/+pWm+iuQ/VepN6zdV6E819qNabaO5Dtd6kXvP22y9PkqxcuTLH/sErcvjCAOUW7sCDD8k1V39n6IyNVu24SOo1V+tN6jVX600096Fab6K5D9V6F3h+MV7VepN6zdV6E819qNabaB63tbNz+YtDH5EnP3zXHPF/v5gk2WWHbbPvXvfMSR/68sB161ZpjRdUa67Wu6Dac7hqvRWPi4rNC/7m3LPy3BcdPnTGBlVc42rN1XqTms0A4zLkmSYPGL099/lJ3pvkyCTvTPLyruv2T/KRJH/Udd1JSW4YDRX+XpIfdF33xMwPJ751tK8dkpzQdd2ho8vPHe37gtFA5AEb0fO5rusOTHK3JLd0XXdwki9l/iyYSbI8ycGZH6J8Y2tt4V+Ns0a93xjdniQ/7LruCUm+mOQ1SQ4cXb5va+3gJGcm+e3Rtr8zutyb5ct3yMqVK2+/PDs3u0X/I1itN9Hch2q9Sb3mar2J5j5U600096Fab1Kz+XvXfjcve/EL89SnPyOHPO3pQ+dslSoeF9Waq/Um9Zqr9Saa+1CtN9Hch2q9VVVb52q9Sb3mar2J5j5U60009+GPzrs8B73+X3PSsx+e7bZdmqc+ctd86Ivfzezc0GXrVm2Nk3rN1XrpR8XjomJzktz845vynW99I3s/+rFDp2xQxTWu1lytN6nZzNZryi+/1vOrD1vC23Mf0HXdIV3XfSTJQ5O8bTTo+DtJ7nuH+zwiydNGt38gyXRr7V6j27pF271vtO/9RwON59/J499xjS8e/f6jzA9LJskPkywb/flfu66b7brue6Pr7z26/guj369Nsv0dWh402u4jo+aHJdkzyd8k+eXW2k8luV/XdQuP3Yt99nl0PnXhhUmSyy69JHvt9eA+H36TVetNNPehWm9Sr7lab6K5D9V6E819qNab1Gu+/vof5KiXvjhHHP37+eVn/drQOVutasdFUq+5Wm9Sr7lab6K5D9V6E819qNZbVbV1rtab1Guu1pto7kO13kTzOD3r0ffJ4Qc8MEly2+rZzM7NZXZ2Lvvudc9ceNX3B65bvyprvFi15mq99KPicVGxOUmuuPTiPOoxPz90xkapuMbVmqv1JjWbAcZlSxsZ75K8oOu6b7XWHp/kZ0bXLww4XpXkO13Xva61tl2SP0tyw+i22Y3Y/22L9vnoO9y2oZ/Ne0yStNZ+Osndk1y3nvsttHw9ybeTHNx13ZrR24Nf0nXdytbaJ5O8Ocl7NqJ7szrgoIPzmc9clBc879DMzc3ltSe+ru+ETVKtN9Hch2q9Sb3mar2J5j5U600096Fab1Kv+d1nnpGbbroxZ5/x9px9xtuTJG966zuybNmyDdyTTVHtuEjqNVfrTeo1V+tNNPehWm+iuQ/Vequqts7VepN6zdV6E819qNabaB6nf77ie3n9cx6R973s57PNkqmc9MEvZ9XMbB547+X51vW3DJ23XlXWeLFqzdV66UfF46Jic5J851vfyK73ueN5n7ZMFde4WnO13qRmM8C4TM3N9X8e/9ba/kkOX/R22gvXPybJqZkf5pxL8rtd131lNGB4dZLfzfxbeO+e+cHFt3Vd987W2jeSPKTruttGg4kP6bru2EX7PS/J25NckuSDSZZm/gyRT+q67pGjs0Ae3nXdVQvbdl13QWvttCSfzfzZJg9LckuSnZIc13Xdx+7wuCdnfqgzix+/tfb8JC8bPeY3kryo67pbWmuPTvKpJPfpuu5H61uv22Y2ONAJAMAW5LY1a4dO2GTTS4Y8Cf2mm17a18n5AYCfxMzaei9neX4BwF3x8GM/OnTCJrvi5KcOncAWpuJzuIo87xy/q2+4deiETXbfXbYbOgE2i2XTvb2zLsV99ms/8sSDdfqFB+089q8lgwxNVnNng5ibYZ+PTXJk13Uv2NC2hiYBAGoxNDl+XlwGgC1bxf9w9/wCgLvC0CRbg4rP4SryvHP8DE3CcAxNsrEMTbI+fQxNbmlvzz0RWmtHZP6smb8xdAsAAAAAAAAAAABMCkOTG6Hrundv5v39ZZK/3Jz7BAAAAAAAAAAAANav1nvwAQAAAAAAAAAAAPyEDE0CAAAAAAAAAAAAE8HQJAAAAAAAAAAAADARDE0CAAAAAAAAAAAAE2F66AAAAAAAAAAAAAAmxNTQAUw6Z5oEAAAAAAAAAAAAJoKhSQAAAAAAAAAAAGAiGJoEAAAAAAAAAAAAJoKhSQAAAAAAAAAAAGAiGJoEAAAAAAAAAAAAJoKhSQAAAAAAAAAAAGAiGJoEAAAAAAAAAAAAJoKhSQAAAAAAAAAAAGAiGJoEAAAAAAAAAAAAJoKhSQAAAAAAAAAAAGAiTA8dAAAAAAAAAAAAwGSYytTQCUw4Z5oEAAAAAAAAAAAAJoKhSQAAAAAAAAAAAGAiGJoEAAAAAAAAAAAAJoKhSQAAAAAAAAAAAGAiTM3NzQ3dwAbcNhN/SWM0s7be8k4vnRo6AQAAAICBddf8eOiETdLus+PQCZvMa4fA1uwhx/zT0Amb5KpTnz50AgCwAcum4xsSNsq//+eN9b7hpjc/v+dOY/9a4kyTAAAAAAAAAAAAwEQwNAkAAAAAAAAAAABMBEOTAAAAAAAAAAAAwEQwNAkAAAAAAAAAAABMhOmhAwAAAAAAAAAAAJgMU1NDFzDpnGkSAAAAAAAAAAAAmAiGJgEAAAAAAAAAAICJYGgSAAAAAAAAAAAAmAiGJgEAAAAAAAAAAICJYGgSAAAAAAAAAAAAmAiGJgEAAAAAAAAAAICJYGgSAAAAAAAAAAAAmAiGJgEAAAAAAAAAAICJYGgSAAAAAAAAAAAAmAiGJgEAAAAAAAAAAICJMD10AAAAAAAAAAAAAJNhaugAJp4zTQIAAAAAAAAAAAATwdAkAAAAAAAAAAAAMBEMTQIAAAAAAAAAAAATwdAkAAAAAAAAAAAAMBEMTQIAAAAAAAAAAAATwdAkAAAAAAAAAAAAMBGmhw6gf7OzsznphBX5Stdl2223zfGvOTH33333obPWqVrvYpdfdmlOP+2UnHH2uUOnbFC1da7Wm9RrrtabaO5Dtd5Ecx+q9Saa+1CtN9Hch2q9Sb3mar2J5j5U600096Fab6J53L765Svy3jPfkhWnnpFrr/523vqGFZmamspue+yZ3z3yj7NkyZb58/eV1ngxrxuOV7Xmar2J5j5U6V0ylZx86CPzwJ/aIXNJ/uxvLs/0kqmc9BuPyMzsXL5+3c354/Muy9zc0KV3rso6L6jWm2juQ7XepF5ztd5Ecx+q9SY1mwHGZct8pesn1Frbv7V2XWvtgtbaJ1trn22tHdlaO621dv/N9Bj3Hu3/gtbaj1prnxv9+Xc3x/77cP4nPp7Vq1bn3Pe9P0e/8pic+oaTh05ar2q9C845+8ycsOK4rFq1euiUjVJtnav1JvWaq/UmmvtQrTfR3IdqvYnmPlTrTTT3oVpvUq+5Wm+iuQ/VehPNfajWm2gepw++/5y8/Y0nZM3q+dexznn7G3Poi16a177pzMzNzeXzn/7XgQvXrcoaL+Z1w/Gr1lytN9Hchyq9Bz38p5Mkv/7mT+fUD3f5w19qOfopD86b//mrefabP51tp5fmgIf91MCV61ZlnRdU600096Fab1KvuVpvorkP1XqTms0A47JVDU2OnN913f5d1z0pyS8mOSbJiq7rvrU5dt513fdH+98/ySVJXjC6fNbm2H8fvnjxF7LvE/ZLkjxy70flyiuvGLho/ar1LrjfbrvllDedPnTGRqu2ztV6k3rN1XoTzX2o1pto7kO13kRzH6r1Jpr7UK03qddcrTfR3IdqvYnmPlTrTTSP00/f5375g+PfcPvl//rqVXnYIx+TJNnncfvmsos/N1TaBlVZ48W8bjh+1Zqr9Saa+1Cl918u/17+5P2XJ0nue4/tctOta3Lld27MzttvkyRZfrelmVm7hZ5mMnXWeUG13kRzH6r1JvWaq/UmmvtQrTep2QwwLlv723PvmGRtko+31p6f5LtJzkpyz9HtR3Vdd3lr7ZtJrkrypST3GN1+zyS/lOSPkuyXZGmSN3Zd97frerDW2rsX3fcZSV6fZLckP5PkQ13XHTfaZlWSPUbXv7Druos334e8YStX3pwdd9zh9stLlyzNzMxMpqe3zMOhWu+CAw8+JNdc/Z2hMzZatXWu1pvUa67Wm2juQ7XeRHMfqvUmmvtQrTfR3IdqvUm95mq9ieY+VOtNNPehWm+ieZx+Yb8Dc9211/z3FXNzmZqaSpJst93y3HLLzQOVbViVNV7M64bjV625Wm+iuQ+VetfOzuXU5+2dJz9y17zs7C/kHsu3zWt//eE58sl75ce3rslnv3b90InrVGmdk3q9ieY+VOtN6jVX600096Fab1Kzma3Y1NABTLqt8UyTB4zeLvv8JO9NcmSShVf1/jTJJ0ZnoXxJkr8aXb9bkud2XffK0eXzu67bN8kvJHlA13VPSPKkJH/WWtt5A4+/cN8dk3y267pDkjwuyeGLtvnm6PrTRx29Wr58h6xcufL2y7Nzs1v0P4LVequqts7VepN6zdV6E819qNabaO5Dtd5Ecx+q9Saa+1CtN6nXXK030dyHar2J5j5U600092lqyX+/bHzrrSuzfPkO69l6WFXXuJKKa1ytuVpvorkP1XqPee+lOeDEC/Lnhz4yx//az+bZb/l0DnzdBfnAf3wnf/ashw2dt07V1rlab6K5D9V6k3rN1XoTzX2o1pvUbAYYl61xaHLh7bkP6LrukK7rPrLotkck+Z3W2gVJ3plkl9H1P+i6bvGPuXWLtn/MaPuPJdkm82eIXJ+F+96Q5LGttfcmeVOSuy3a5ouj37+dZNnGfmCbyz77PDqfuvDCJMlll16SvfZ6cN8Jm6Rab1XV1rlab1KvuVpvorkP1XoTzX2o1pto7kO13kRzH6r1JvWaq/UmmvtQrTfR3IdqvYnmPu3xoJYrL/18kuSLn/t0HvqIfQYuWreqa1xJxTWu1lytN9Hchyq9v/Jz983LDtozSXLr6rWZm0t+tHJNbr5tJkly3U2rstPorbq3RFXWeUG13kRzH6r1JvWaq/UmmvtQrTep2QwwLpM2Mn5Vkvd0Xfe+1tpPJfm90fWzd9hudtH2n+y67iWttSVJXpXkPzfwGAv3fWGSH3Vdd1hr7UFJXtJaWzi57Nxd+SDuqgMOOjif+cxFecHzDs3c3Fxee+LrhszZoGq9VVVb52q9Sb3mar2J5j5U600096Fab6K5D9V6E819qNab1Guu1pto7kO13kRzH6r1Jpr79ILDXpF3vPGkzMy8Nfe9/x75hf0OHDppnaqucSUV17hac7XeRHMfqvR+7LJr84bn7p33H/l/ss3SJXnt31+ZH96yOqf/9qOzdnYuq9fO5k/Ou2zozHWqss5i1qSuAAAgAElEQVQLqvUmmvtQrTep11ytN9Hch2q9Sc1mgHGZmpsbdH5vs2qt7Z/k8K7rDr3D9Rdk/u2xv5/krCQ7J7l7khVd132otXZt13W7jrZ9d5Lzuq772GjI8dQkj02yQ5J/6LrutXfcb9d1V93JfX82yfuS3JhkVebfAvzAJCct2uYpSQ7tuu6F6/u4bpsZdshyazeztt7yTi+d2vBGAAAAAGzVumt+PHTCJmn32XHohE3mtUNga/aQY/5p6IRNctWpTx86AQDYgGXT8Q0JG+U/vn5jvW+46c1jH7DT2L+WbFVDk1srQ5Pj5YVPAAAAACoyNDl+XjsEtmaGJgGAzc3QJBvL0CTr08fQ5JJxPwAAAAAAAAAAAADAlsDQJAAAAAAAAAAAADARDE0CAAAAAAAAAAAAE8HQJAAAAAAAAAAAADARDE0CAAAAAAAAAAAAE2F66AAAAAAAAAAAAAAmw1Smhk5gwjnTJAAAAAAAAAAAADARDE0CAAAAAAAAAAAAE8HQJAAAAAAAAAAAADARDE0CAAAAAAAAAAAAE8HQJAAAAAAAAAAAADARDE0CAAAAAAAAAAAAE8HQJAAAAAAAAAAAADARDE0CAAAAAAAAAAAAE8HQJAAAAAAAAAAAADARDE0CAAAAAAAAAAAAE2F66AAAAAAAAAAAAAAmw9TU0AVMOmeaBAAAAAAAAAAAACaCoUkAAAAAAAAAAABgIhiaBAAAAAAAAAAAACaCoUkAAAAAAAAAAABgIhiaBAAAAAAAAAAAACaCoUkAAAAAAAAAAABgIkzNzc0N3cAG3DYTf0nQs5m19T7tppdODZ2wyaqtc8U1BgDY3G5bs3bohE0yvaTez4t63gkATLpqrxsmnsP14R6/fsbQCZvsh3/3kqETAKBXy6bjSREb5QvfuKnek35685g97j72ryX1/ucAAAAAAAAAAAAA4CdgaBIAAAAAAAAAAACYCNNDBwAAAAAAAAAAADAZvI87Q3OmSQAAAAAAAAAAAGAiGJoEAAAAAAAAAAAAJoKhSQAAAAAAAAAAAGAiGJoEAAAAAAAAAAAAJoKhSQAAAAAAAAAAAGAiGJoEAAAAAAAAAAAAJoKhSQAAAAAAAAAAAGAiGJoEAAAAAAAAAAAAJoKhSQAAAAAAAAAAAGAiGJoEAAAAAAAAAAAAJsL00AEAAAAAAAAAAABMiKmhA5h0zjQJAAAAAAAAAAAATARDkwAAAAAAAAAAAMBEMDQJAAAAAAAAAAAATARDkwAAAAAAAAAAAMBEMDQJAAAAAAAAAAAATARDkwAAAAAAAAAAAMBEMDQJAAAAAAAAAAAATITpoQPo3+zsbE46YUW+0nXZdtttc/xrTsz9d9996Kx1qtabaO5Dtd7FLr/s0px+2ik54+xzh05ZL2vcj2rrXK030dyHar2J5j5U600096Fab1KvuVpvksysWZMTVhyX715zddasXpMXvfiwPHH/A4bO2iDPOcdL8/hV600096Fab1KvuVpvorkP1XqTms2J53DjVqn53jsty6dP/dX80vEfzleuvjFJ8pwn7pmXPu3h2f/YDw5ct26V1niB5vGr1pvUa67Wm2juQ7XepGYzwLhsNWeabK3t31q7rrV2QWvtk621L7TW/ra1tu06tn93a+0pd7hu19ba2zbwOL/dWjt/9BgXtdaePLr+2Nba4+6w7bLW2jc20Hzexn6Mm8v5n/h4Vq9anXPf9/4c/cpjcuobTu47YZNU600096Fa74Jzzj4zJ6w4LqtWrR46ZYOscT+qrXO13kRzH6r1Jpr7UK030dyHar1JveZqvUny0Q//Y3baaeec8a735LS3vSOnnHzi0Ekb5Dnn+Gkev2q9ieY+VOtN6jVX600096Fab1Kz2XO48avSPL10Kn/50v1y66qZ26/b+wH3zG8f9JBMTQ0YthH+f/buPM7uur77/nuSASIE91oUBKviV8WiaBcXEAwg1cuqrRtVSy1uFOterRUsKKhQQUVbi6CItVr07mXby6W2tyLigm0vN4TKF7VuoGihKBAM2eb+YybcKQ3JBD2/k09+zyePeSRncs6ZVz4ZJmd+55PfqTLjjWmevGq9Sb3mar2J5iFU601qNgNMynazNLng3N77Qb33R/TeH5RkTZLHLvbGvfcreu9H39yvt9Zuk+RVSX6j9/6IJE9KclZrbUnv/aTe+7/+rL+BIXzpi1/IQ/c/IEmy7/0fkIsvvmjKRZtXrTfRPIRqvRvscde75pQ3vXXaGYtixsOoNudqvYnmIVTrTTQPoVpvonkI1XqTes3VepPk4Eceluc+7wXzF+aSpUu3/RfF8Jhz8jRPXrXeRPMQqvUm9Zqr9Saah1CtN6nZ7DHc5FVpPukZD86ZH/tafnD19UmS2++6U1799F/Ny975uSmXbVmVGW9M8+RV603qNVfrTTQPoVpvUrOZ7deM//y3mf+GsO0/E3ELLZxh8s5Jrm6tnZpk/4Vfel/v/bSFnx/dWntZ5ufwzCRrk5zTe39wa+3CJJ9Ksm+SuSSPS3JDkh2T/EFr7cO992+21u7Re1/fWjs7yTlJPpPkvUlul+QbG/X8cpK3JJlJclWSIyf3u9+8lSuvy667Lr/x8tIlS7N27drMzm6bnw7VehPNQ6jWu8HBhx6W719+2bQzFsWMh1FtztV6E81DqNabaB5Ctd5E8xCq9Sb1mqv1JsnOO++SJFm5cmVe8UcvylEbFii3YR5zTp7myavWm2geQrXepF5ztd5E8xCq9SY1mz2Gm7wKzU9fca/85zWr8vEvX5aXPfEBWbpkSU7/wwPzx+/6fH66eu2W72DKKsz4pjRPXrXepF5ztd5E8xCq9SY1mwEmZXs70+SKhZfn/vckX0zyd0luleSXkjw484uTT11YYEySz/XeD05ycpI/u8l93TrJ3/TeD0xyeZJH9d5XJVmRZO8kH2utfSf/c/nxqCQX9d4fnuTtG73/zCTP670flOSjSV7+8/gN3xK77LI8K1euvPHy+rn12/RfgtV6E81DqNZbkRkPo9qcq/UmmodQrTfRPIRqvYnmIVTrTeo1V+vd4IdX/CBHP/sZedRjfjOHPfox087Z7lT8vNA8edV6E81DqNab1Guu1ptoHkK13qRmczUVZ1yh+fcObjn4/rvnn058TPb9pTvk/572xOyz1+3zlqP2z3teenDufdfb5Q3PfMi0M29WhRnflObJq9ab1Guu1ptoHkK13qRmM8CkbG9Lk+cuLCUekGR1km8luU+ST/fe53rva5J8Psl9F65//sKPn0vSNnF/X1r48XtJlrXW7pLkVr33P+y9753k0CQv22gJM0nuleRfk6T3/i+Zf4nwLHS8rbV2XuYXLXf/GX+vt9h++z0wnzl//rd+4Ve+nL33vte0UhalWm+ieQjVeisy42FUm3O13kTzEKr1JpqHUK030TyEar1JveZqvUly1VVX5gV/8Oz84Qtfksc+/gnTztkuVfy80Dx51XoTzUOo1pvUa67Wm2geQrXepGZzNRVnXKH50GM+lEce++EcduyHc+G3rsp+z/9A9jnqnBx27Ifzu6d+Ipd87+q87J0XTDvzZlWY8U1pnrxqvUm95mq9ieYhVOtNajYDTMp2uTLee7+qtfb0JJ9M8kdJHp/kTa21HZI8NMm7kzwqya9lfmHygCQXbeKu5m5yebck72qt7d97vzbJd5JcmfkFzQ3+PclDkvxDa22/JDtsyEpyRO/9u621h2X+pcOnYsUhh+aCCz6bI552eObm5vKaE183rZRFqdabaB5Ctd6KzHgY1eZcrTfRPIRqvYnmIVTrTTQPoVpvUq+5Wm+SnP2OM3LNNT/JWWecnrPOOD1J8qa/eHuWLVs25bLtR8XPC82TV6030TyEar1JveZqvYnmIVTrTWo2V1NxxhWbq6k4Y82TV603qddcrTfRPIRqvUnNZoBJmZmbu+leYE2ttYOSHNV7P3yj9x2T5AGZX258WJIdk3yg935ya+3sJMuS3Cnzy5FHJplJck7v/cGttW8nuXfvfVVr7aQkl/Tez26tPSvJ85L8NMnSJO/svZ+xcH/nJDkvyV8luUuSS5Ic0HtvrbUHJTk184uqc0meuXCd/9a8KavW/o/lTWDC1q6r97/d7NKZaSdstWpzrjhjAICft1Vr1k07YavMLqn3IhsedwIAY1ftuGHiMdwQbvfEM6adsNWu/tvnTDsBAAa1bDYeFLEoX/rOtfUe9DOY/fbadeJfS7abpcntmaVJGJ6DcsOoNueKMwYA+HmzNDl5HncCAGNX7bhh4jHcECxNAsC2z9Iki2Vpks0ZYmmy3jMHAAAAAAAAAAAAALeApUkAAAAAAAAAAABgFCxNAgAAAAAAAAAAAKNgaRIAAAAAAAAAAAAYhdlpBwAAAAAAAAAAADAOMzPTLmDsnGkSAAAAAAAAAAAAGAVLkwAAAAAAAAAAAMAoWJoEAAAAAAAAAAAARsHSJAAAAAAAAAAAADAKliYBAAAAAAAAAACAUbA0CQAAAAAAAAAAAIyCpUkAAAAAAAAAAABgFCxNAgAAAAAAAAAAAKNgaRIAAAAAAAAAAAAYBUuTAAAAAAAAAAAAwCjMTjsAAAAAAAAAAACAcZiZdgCj50yTAAAAAAAAAAAAwChYmgQAAAAAAAAAAABGwdIkAAAAAAAAAAAAMAqWJgEAAAAAAAAAAIBRmJ12AAAAAAAAAAAAAMBitNZ+PcnJvfeDWmv3THJ2krkkFyV5Xu99/eZu70yTAAAAAAAAAAAAwDavtfbyJO9IsmzhXW9Mcmzv/YAkM0ket6X7sDQJAAAAAAAAAAAAVPDNJL+90eUHJfnUws//MckhW7oDL89dwJXXrp52wla54647TjsBfmazS2emnTAK5gxsr65btXbaCVtt+TLfGgCLs2yHpdNOAAC2I2vXzU07Yas5pjV5ZsymXP23z5l2wlbb41nnTDthq1z2jsOnnQAAAFvUe//frbW7bfSumd77hgMM1ya5zZbuw5kmAQAAAAAAAAAAgIrWb/TzXZP8eEs3sDQJAAAAAAAAAADAMGa8edvM29b7UmvtoIWfPyrJp7d0A6/BBwAAAAAAAAAAAFT00iRnttZ2TPK1JH+7pRtYmgQAAAAAAAAAAABK6L1/O8mDF35+aZIDt+b2Xp4bAAAAAAAAAAAAGAVLkwAAAAAAAAAAAMAoWJoEAAAAAAAAAAAARsHSJAAAAAAAAAAAADAKliYBAAAAAAAAAACAUbA0CQAAAAAAAAAAAIyCpUkAAAAAAAAAAABgFCxNAgAAAAAAAAAAAKMwO+0AAAAAAAAAAAAAxmEmM9NOYOScaRIAAAAAAAAAAAAYBUuTAAAAAAAAAAAAwChYmgQAAAAAAAAAAABGwdIkAAAAAAAAAAAAMAqWJgEAAAAAAAAAAIBRsDQJAAAAAAAAAAAAjIKlSQAAAAAAAAAAAGAULE0CAAAAAAAAAAAAo2BpEgAAAAAAAAAAABiF2WkHMB3PPeLJ2WWXXZIku91l97z8VSdOuejmrV+/Pq894fhc2nt23HHHHPfqE7PnXntNO2uzNE9etd6kXnO13kTzEKr1JpqHUK03qdm8bt26nHzicfnud76VmczkZa88Lne/597TzrpZFWesefKq9Sb1mqv1JpqHUK030TyEar2J5iFU603qNVfrTWo2J8lXL/xK3vrmU3LGWe+ZdsoWVZyx5smr1ptonrQ77rpTPnH8YXnCKZ/MN35wbZLkxN/ZL9+44pqc/clvTrnu5lWa8QbVmqv1JvWaq/UmmodQrTep2QwwKSXPNNlaO6i19qPW2nmttU+21r7QWvt/Wms7TuBjfbu1dv7CxzqvtfbBhfdv+PG81tq9W2vHt9aOaq09oLX2pz/vjp+n1TfckGQub/zLd+WNf/mubXphMknO/cTHs/qG1XnP+96fF774pTn1DSdNO2mLNE9etd6kXnO13kTzEKr1JpqHUK03qdn82fPPS5KcftZ78+yjX5C3v+206QZtQcUZa568ar1JveZqvYnmIVTrTTQPoVpvonkI1XqTes3VepOaze8+6x054fhjc8MNq6edsigVZ6x58qr1JponaXbpTE59xq9m1Zq1SZI77LpTznnJw3PYfrtPuWzLqsx4Y9Waq/Um9Zqr9Saah1CtN6nZzPZrZsabt5t/G0LlM02e23s/fMOF1tr7kjw2yd9O4GM9sve+auN39N5/e1NX7L1/OcmXJ9Dwc/PNr/esWrUqL3/Bc7Ju3bo88w9ekPve7/7TzrpZX/riF/LQ/Q9Ikux7/wfk4osvmnLRlmmevGq9Sb3mar2J5iFU6000D6Fab1Kz+eGPODgPPeDAJMkVP/h+li/fdcpFm1dxxponr1pvUq+5Wm+ieQjVehPNQ6jWm2geQrXepF5ztd6kZvMed71rTnnTW/OqV/7xtFMWpeKMNU9etd5E8yS9+ikPyNmf/EZe9Jj7Jkl22Wk2f/b3F+WQfe8y5bItqzLjjVVrrtab1Guu1ptoHkK13qRmM8CkVF6avNHCGSbvnOTq1tqpSfZf+KX39d5Pa63dM8nZSdYk+U6Su/XeD2qtPSnJS5KsS/KZ3vsrWmvHJ3lokuVJnrmZj3lF7323Tbz/oCRH9d4Pb609LcmLktyQ5OtJnpPkaUkenWTnJPdIcnLv/eyfbQJbZ6dly/Lkp/5eHv24J+Sy730nf/LiP8i73/+hLJ3dNj8dVq68LrvuuvzGy0uXLM3atWszu432JpqHUK03qddcrTfRPIRqvYnmIVTrTWo2J8ns7GxO+NM/yfnnfSInnvymaedsVsUZa568ar1JveZqvYnmIVTrTTQPoVpvonkI1XqTes3VepOazQcfeli+f/ll085YtIoz1jx51XoTzZNy+P6/lKuuvSGfvOiKG5cmv3vlynz3ypUlliYrzPimqjVX603qNVfrTTQPoVpvUrMZYFJKvjz3ghULL43970m+mOTvktwqyS8leXDmFyef2lr75SRvSPK63vsjknw2SVprt0/y6iQH9973T7J7a+3Qhfv+Wu/9ob33ry1c/ueNXp77fy0mrrV2h4X7X7Fw/z9O8tyFX75N7/0xmT8z5it+liHcEnvsebcc8huPyczMTO66591y69vcNldddeXQGYu2yy7Ls3Llyhsvr59bv83/pa158qr1JvWaq/UmmodQrTfRPIRqvUnN5g1e9ZrX55wPfiQnn3hcfvrT66edc7Mqzljz5FXrTeo1V+tNNA+hWm+ieQjVehPNQ6jWm9Rrrtab1GyupuKMNU9etd5E86Q87YBfykH77JZ/eMWK3G/P2+Ztz35w7nSbZdPOWrQKM76pas3VepN6zdV6E81DqNab1GwGmJTKS5Pn9t4PSnJAktVJvpXkPkk+3Xuf672vSfL5JPddeP/nFm736YUf75nkF5J8tLV23sL17rHwa/0mH+uRvfeDFt4+ssi+uye5uPd+7cLl85Pss/DzDS/f/b0kg39X87EP/V1Of8spSZIr//NHuX7ldbnDHe44dMai7bffA/OZ889Pklz4lS9n773vNeWiLdM8edV6k3rN1XoTzUOo1ptoHkK13qRm88c+8n/yV2edmSRZtuxWWbJkSZbMbLsP5yvOWPPkVetN6jVX6000D6Fab6J5CNV6E81DqNab1Guu1pvUbK6m4ow1T1613kTzpPzm68/NY086N4876dxc9N0f5+gzP58f/WTVtLMWrcKMb6pac7XepF5ztd5E8xCq9SY1mwEmpfzKeO/9qtba05N8MskfJXl8kje11nbI/MtsvzvJRUkekuQfM38WymR+yfJ7SQ7tva9prT0j88uMj0+y/ueQ9q0k922t7dJ7X5nkwCSXLvza3M/h/m+xRz32t/NnJxyTFz7niGRmJi875oRt9qW5k2TFIYfmggs+myOednjm5ubymhNfN+2kLdI8edV6k3rN1XoTzUOo1ptoHkK13qRm84ErDsnrjj82Rz/riKxduzYvfOkrstOybfesAhVnrHnyqvUm9Zqr9Saah1CtN9E8hGq9ieYhVOtN6jVX601qNldTccaaJ69ab6KZTas442rN1XqTes3VehPNQ6jWm9RsBpiUmbm5qe7v3SKttYOSHNV7P3yj9x2T5AFJvpPkYUl2TPKB3vvJrbV7JDkrydokP0mya+/90IVly6OTLE3y7SS/n+TlSa7ovZ++cL/fTnLv3vt/+ydbrbUreu+7LZyl8qgkhye5IsklG9paa09N8qLML2F+I8mzFq537977K1pry5Jc0nu/2+Z+v5ddvbrUH9Idd91x2gkAAFN13aq1007YasuXbbv/iAYAANh+rV1X6vB3kmR26cy0E4Ai9njWOdNO2CqXvePwLV8JADZj2Ww8WGZRvnrZdfW+GWQwv7zH8ol/LSm5NLm1WmtPS/IvvfdvtNaeleShvfcjp921WJYmAQBqsTQJAACwOJYmge2ZpUkAxsbSJItlaZLNGWJpcizPjH4vyTmtteuTrEvyzCn3AAAAAAAAAAAAAAMbxdJk7/38JL8y7Q4AAAAAAAAAAABgekaxNAkAAAAAAAAAAMD0eR13pm3JtAMAAAAAAAAAAAAAhmBpEgAAAAAAAAAAABgFS5MAAAAAAAAAAADAKFiaBAAAAAAAAAAAAEbB0iQAAAAAAAAAAAAwCpYmAQAAAAAAAAAAgFGwNAkAAAAAAAAAAACMgqVJAAAAAAAAAAAAYBQsTQIAAAAAAAAAAACjYGkSAAAAAAAAAAAAGIXZaQcAAAAAAAAAAAAwEjPTDmDsnGkSAAAAAAAAAAAAGAVLkwAAAAAAAAAAAMAoWJoEAAAAAAAAAAAARsHSJAAAAAAAAAAAADAKliYBAAAAAAAAAACAUbA0CQAAAAAAAAAAAIyCpUkAAAAAAAAAAABgFCxNAgAAAAAAAAAAAKNgaRIAAAAAAAAAAAAYhZm5ublpN7AF191Q6w9pdunMtBMAWHDdqrXTTthqy5fNTjsBAIDt3Ko166adsNWW7bB02gkwSmvXlTo0m8TxWQBuuXu/9MPTTthql5z6mGknwM9FtedzKj6X41jAMJbNxjckLMrFl6+s9w03g9ln910m/rXEmSYBAAAAAAAAAACAUbA0CQAAAAAAAAAAAIyCpUkAAAAAAAAAAABgFCxNAgAAAAAAAAAAAKNgaRIAAAAAAAAAAAAYBUuTAAAAAAAAAAAAwChYmgQAAAAAAAAAAABGwdIkAAAAAAAAAAAAMAqWJgEAAAAAAAAAAIBRsDQJAAAAAAAAAAAAjMLstAMAAAAAAAAAAAAYh5mZaRcwds40CQAAAAAAAAAAAIyCpUkAAAAAAAAAAABgFCxNAgAAAAAAAAAAAKNgaRIAAAAAAAAAAAAYBUuTAAAAAAAAAAAAwChYmgQAAAAAAAAAAABGwdIkAAAAAAAAAAAAMAqWJgEAAAAAAAAAAIBRsDQJAAAAAAAAAAAAjIKlSQAAAAAAAAAAAGAUZqcdwHR89cKv5K1vPiVnnPWeaads0fr16/PaE47Ppb1nxx13zHGvPjF77rXXtLM2S/PkVetN6jVX6000D2HdunU5+cTj8t3vfCszmcnLXnlc7n7PvaedtVnVZpzUa67Wm2geQrXeRPMQqvUm9Zqr9Saah1CtN6nZvHbNmpxw/LH5wfcvz5rVa/L7z35uHn7Qimln3ayKM9Y8edV6k5rNieOzk1atuVpvonkI1XoTzUOo0rtkJjnp8H1z9zstz1ySYz7w1cwumclrn/zLWbt+Lt/60XX543MuzNzctEs3rcqcN6jWm9RrrtabeD5nCNWOAyT1Zsz2bWbaAYxeiTNNttZe0Vr7eGvtU621T7bWHjTQxz27tfYbrbWDWms/aq2dt/DxP99ae/4QDZPw7rPekROOPzY33LB62imLcu4nPp7VN6zOe973/rzwxS/NqW84adpJW6R58qr1JvWaq/Ummofw2fPPS5KcftZ78+yjX5C3v+206QYtQrUZJ/Waq/UmmodQrTfRPIRqvUm95mq9ieYhVOtNajb/40c+lNvc5rY5411/nTe/7e055aQTp520WRVnrHnyqvUmNZsdn528as3VehPNQ6jWm2geQpXeQ+73i0mSJ572uZz6kZ6X/a+WF/7GvXLaP309Tzrtc9lxdmlW3PdOU668eVXmvEG13qRec7XexPM5Q6h2HCCpN2OASdrmlyZba/dN8tgkh/beD0zy4iRnTSHl3N77Qb33RyQ5MMlLW2u3nULHz2yPu941p7zprdPOWLQvffELeej+ByRJ9r3/A3LxxRdNuWjLNE9etd6kXnO13kTzEB7+iIPz8mOOT5Jc8YPvZ/nyXacbtAjVZpzUa67Wm2geQrXeRPMQqvUm9Zqr9Saah1CtN6nZfPAjD8tzn/eC+QtzydKl2/YLvFScsebJq9ab1Gx2fHbyqjVX6000D6Fab6J5CFV6//mrP8yfvP+rSZLdb3erXPPTNbn4sp/ktjvvkCTZZaelWbtuGz3NZOrMeYNqvUm95mq9iedzhlDtOEBSb8YAk7Ttf9VOfpJkzyRHttY+1nv/cmvt11prByY5LvOLn8uTPDXJ6iTvT/K9JHdLck6S+yXZL8lHeu+vbK2dl+So3vslrbWjkuzWez9+4cyRT00yl+Sc3vtbNtO0a5J1Sda21n45yVsyf+bYq5IcufDxTl7oOSPJHye5dOHyXZM8p/d+cWvtUUl+s/d+9M9jUIt18KGH5fuXXzbkh/yZrFx5XXbddfmNl5cuWZq1a9dmdnbb/fTVPHnVepN6zdV6E81DmZ2dzQl/+ic5/7xP5MST3zTtnC2qOONqzdV6E81DqNabaB5Ctd6kXnO13kTzEKr1JjWbd955lyTJypUr84o/elGO2vDEyTaq4ow1T1613qRms+Ozk1etuVpvonkI1XoTzUOo1Ltu/VxOfdr988h9d8vRZ30ht9tlx7zmiffL8x+5d6796YXFJRcAACAASURBVJp8/htXTTvxZlWac1KvN6nXXK13A8/nTFa14wBJvRkDTNI2f6bJ3vvlmT/T5MOSXNBauyTJY5Lsk+TpvfeDknwwyZMWbnL3JM9cuM4JSV6S5NcX3rdJC2ezfEqS/ZMckOTxrbV2k6utWHh57nOTvDfJ83vv1yU5M8nzFjo+muTlC9df1ns/oPf+nswvdZ7Qez88yTuS/N7CdY5cuMxm7LLL8qxcufLGy+vn1m/zf2lrnrxqvUm95mq9ieYhveo1r885H/xITj7xuPz0p9dPO2ezKs64WnO13kTzEKr1JpqHUK03qddcrTfRPIRqvUnN5iT54RU/yNHPfkYe9ZjfzGGPfsy0czar4ow1T1613qRmczUVZ1ytuVpvonkI1XoTzUOo1vvS934lK048L68/fN8c94R98qS3fC4Hv+68/O9/uyzHPP6+0867WdXmXK03qddcrXdjns+ZrErHAZKaMwaYlG1+abK1ds8k1/Tej+y975nk6UlOT7IyyVtaa2cneUSSHRZu8h+9958k+XGSH/be/6v3virzZ5C8qZmFH++XZK8kn1h4u0OSvW9y3Q0vz72i935Y7/2jC++/T5K3LZzB8sgkuy+8v9/k9hsufyDJY1trd0qyR+/9i4udxVjtt98D85nzz0+SXPiVL2fvve815aIt0zx51XqTes3VehPNQ/jYR/5P/uqsM5Mky5bdKkuWLMmSmW374US1GSf1mqv1JpqHUK030TyEar1JveZqvYnmIVTrTWo2X3XVlXnBHzw7f/jCl+Sxj3/CtHO2qOKMNU9etd6kZnM1FWdcrblab6J5CNV6E81DqNL7W7+ye44+5B5Jkp+uXpe5ueTHK9fkulVrkyQ/uuaG3GbnHTZ3F1NVZc4bVOtN6jVX6008nzOEascBknozBpikCivj+yZ5Tmvtsb331Zl/mesfJ3lzkj1779e21t6d/38BclPLkRtbleTOSS5J8sAkl2d+ofHiJI/qvc+11l6c5MIkT1xEX09yRO/9u621hy3cd5Ksv8n11idJ731la+2TSU5L8teLuP/RW3HIobnggs/miKcdnrm5ubzmxNdNO2mLNE9etd6kXnO13kTzEA5ccUhed/yxOfpZR2Tt2rV54UtfkZ2WLZt21mZVm3FSr7lab6J5CNV6E81DqNab1Guu1ptoHkK13qRm89nvOCPXXPOTnHXG6TnrjNOTJG/6i7dn2Tb6eLnijDVPXrXepGZzNRVnXK25Wm+ieQjVehPNQ6jS+7ELr8gbnnr/vP/5D8kOS5fkNR+8OFdfvzpv/b0HZt36uaxetz5/cs6F0868WVXmvEG13qRec7XexPM5Q6h2HCCpN2OASZqZm9vSjuH0tdaOSfLkJNdl/uyYJyd5eJIVmT/j5A+TXJXktUnO6b0/uLW2LMklvfe7LdzHFb333Vprj05yapLvZn5h8ru99+Nbay9L8vgkOyX51yTPT/LOJOdkftHyqIWX175p24MW7m828wubz0xyl42v31r7dpJ7L5zxMq21Byb5TJK79N5/vKXf/3U3FPhD2sjs0pktXwmAQWz4l7uVLF9W4d90AABQ2ao166adsNWW7bB02gkwSmvXlTo0m8TxWQBuuXu/9MPTTthql5y67b8cLixGtedzKj6X41jAMJbNxjckLMrXvr+y3jfcDOY+d9ll4l9LSixNbm9aa7+a5Pm99yMWc31LkwDcUtW+yU5qfqMNAEAtnigBFsvSJABjYmkSpqfa8zkVn8txLGAYliZZLEuTbM4QS5P1/iYrrrX2h5k/G+WTp90CAAAAAAAAAAAAY2JpcmC99z9P8ufT7gAAAAAAAAAAABicc5IyZUumHQAAAAAAAAAAAAAwBEuTAAAAAAAAAAAAwChYmgQAAAAAAAAAAABGwdIkAAAAAAAAAAAAMAqWJgEAAAAAAAAAAIBRsDQJAAAAAAAAAAAAjIKlSQAAAAAAAAAAAGAULE0CAAAAAAAAAAAAo2BpEgAAAAAAAAAAABgFS5MAAAAAAAAAAADAKMxOOwAAAAAAAAAAAIBxmMnMtBMYOWeaBAAAAAAAAAAAAEbB0iQAAAAAAAAAAAAwCpYmAQAAAAAAAAAAgFGwNAkAAAAAAAAAAACMgqVJAAAAAAAAAAAAYBQsTQIAAAAAAAAAAACjYGkSAAAAAAAAAAAAGAVLkwAAAAAAAAAAAMAoWJoEAAAAAAAAAAAARmF22gEAAAAAAAAAAACMw8zMtAsYu5m5ublpN7AFq9bGHxIAABO1dl2th5yzS303DQAAwLbj4suumXbCVttnj1tPOwF+ZiveeP60E7bauS95+LQTtns/vn7NtBO22m133mHaCWyDqh23T5LlO1mFY3H6FdfX+wRnMG23nSf+tcTLcwMAAAAAAAAAAACjYGkSAAAAAAAAAAAAGAVLkwAAAAAAAAAAAMAoWJoEAAAAAAAAAAAARsHSJAAAAAAAAAAAADAKliYBAAAAAAAAAACAUbA0CQAAAAAAAAAAAIyCpUkAAAAAAAAAAABgFCxNAgAAAAAAAAAAAKMwO+0AAAAAAAAAAAAAxmFm2gGMnjNNAgAAAAAAAAAAAKNgaRIAAAAAAAAAAAAYBUuTAAAAAAAAAAAAwChYmgQAAAAAAAAAAABGwdIkAAAAAAAAAAAAMAqWJgEAAAAAAAAAAIBRsDQJAAAAAAAAAAAAjIKlSQAAAAAAAAAAAGAULE0CAAAAAAAAAAAAo2BpEgAAAAAAAAAAABiF2WkHAAAAAAAAAAAAMBIz0w5g7CxNjtD69evz2hOOz6W9Z8cdd8xxrz4xe+6117Szbla13kTzEKr1JvWaq/UmmodQrTfRPIRqvYnmIX31wq/krW8+JWec9Z5pp2xRxRlXa67Wm9RrrtabaB5Ctd5E8xCq9Saah1CtN6nXXK030TyEar2J5kn7xiUX5W/e+da86g1vv/F97zn9jbnzHnvlkMc8YYplm1dpxhtUa67Wm9RpXjKTvOKwe2XP298qc0ne8M9fz/q55I8P2zszSb539U9z0scuzbq5aZf+T1VmvLGKzUly9X9dlWf/7pNz6l+cmb3udvdp52xWxRlXa67Wu7FKx+0BJmWbe3nu1torWmsfb619qrX2ydbag36G+7pba+3zm3j/8a21S1tr57XWPt1a++fW2n634P7Paa3teEv7puXcT3w8q29Ynfe87/154YtfmlPfcNK0kzarWm+ieQjVepN6zdV6E81DqNabaB5Ctd5E81DefdY7csLxx+aGG1ZPO2VRKs64WnO13qRec7XeRPMQqvUmmodQrTfRPIRqvUm95mq9ieYhVOtNNE/Shz7wVznzTSdmzer576Wv+fHVOfmYF+QLnz9/ymVbVmXGG6vWXK03qdO8/z3ukCQ56n1fyRmf/naee8DdctTD75a3n//tHPW+ryRJHnbPO0wz8WZVmfHGKjavXbsmp7z+1dlp2bJppyxKxRlXa67Wu0G14/YAk7JNLU221u6b5LFJDu29H5jkxUnOmtCHe2Pv/aDe+wFJXpDkb1prW/UIp/d+eO+93N8kX/riF/LQ/Q9Ikux7/wfk4osvmnLR5lXrTTQPoVpvUq+5Wm+ieQjVehPNQ6jWm2geyh53vWtOedNbp52xaBVnXK25Wm9Sr7lab6J5CNV6E81DqNabaB5Ctd6kXnO13kTzEKr1Jpon6RfvvEde9Kd/duPlVauuzxN+9znZ/+BHT7FqcarMeGPVmqv1JnWaz//GVTn5ny5Nkux262W59oZ1eeXf/3u+fNlPMrtkJnfYZcesvGHdlCs3rcqMN1ax+W1vPiWP++2n5I53/IVppyxKxRlXa67Wu0G14/YAk7KtvTz3T5LsmeTI1trHeu9fbq39WmvtwCTHZX7Jc3mSpyZZneRDSa5K8tEk/3Iz1/mF1trfJ7lzkgt778++6QftvV/SWvtikv1ba5ck+cskyxZuc2yS/0hyWu/9EUnSWvtwklcl+bsk907y6CR/nGRNku8nOTzJrZP89cKPs0mO7b2f21q7MMmnkuybZC7J43rvP/k5zW9RVq68LrvuuvzGy0uXLM3atWszO7utfTrMq9abaB5Ctd6kXnO13kTzEKr1JpqHUK030TyUgw89LN+//LJpZyxaxRlXa67Wm9RrrtabaB5Ctd5E8xCq9Saah1CtN6nXXK030TyEar2J5kn6tQNW5D+v+P6Nl++02+65026758v/9rkpVi1OlRlvrFpztd6kVvO6ueTYR7ccuPcdcsw/fC3r55Ldbr1TTnvKvrnuhrX5+o+um3biJlWa8QbVmv/xQ3+f297u9vm1hzws7z37zGnnLEq1GSf1mqv1blDtuD3ApGxTZ5rsvV+e+TNNPizJBQsLjI9Jsk+Sp/feD0rywSRPWrjJbkke2Xv/s81c59ZJfj/JQ5Ic3Fq70818+B8muWPmlyBP7b0fmuQ5SZ7Xe78wybLW2l6ttTsnuWPv/Usb3fZ3kryh975/kg8vfMxjk/y/vfeHL7S8s7U2s/Brf7NwJs3Lkzzqlk3rlttll+VZuXLljZfXz63fpv/irtabaB5Ctd6kXnO13kTzEKr1JpqHUK030cymVZxxteZqvUm95mq9ieYhVOtNNA+hWm+ieQjVepN6zdV6E81DqNabaGbTKs64WnO13qRe84kf7XnKO/4trzhs7yzbYUmuuOaGPOXMf8vff/kHecGKe0w7b5OqzTip1/zRD/1d/u+/XJAXPvcZ+calPa877pW56sorp521WdVmnNRrrtYLwH+3TS1NttbumeSa3vuRvfc9kzw9yelJViZ5S2vt7CSPSLLDwk2+tdHLY19+M9f5j9771b339Ul+lGTnm/nweyW5LMkPkjy3tfaeJEdtdD/vTHJEkt9N8q6b3PYlSVa01j6V5KFJ1ie5T5LzkxuXQa9JsmFhc8PC5fcyf0bLQe233wPzmfPPT5Jc+JUvZ++97zV0wlap1ptoHkK13qRec7XeRPMQqvUmmodQrTfRzKZVnHG15mq9Sb3mar2J5iFU6000D6Fab6J5CNV6k3rN1XoTzUOo1ptoZtMqzrhac7XepE7zb9z3TvndX79rkmTVmvVZP5ec9Fv7ZI/bzT+Ve/3qdZmbm5tm4s2qMuONVWt+6xnvzlvOODunvf3s3PNeLa989etyhzvecdpZm1Vtxkm95mq9APx329qa+75JntNae+zCMuSlSX6c5M1J9uy9X9tae3eSmYXrr9/otmcmuccmrrPFR6+ttX2S3DfJ55N8IMmZvfd/bK39fpJnLFztnCSfWPiYj7zJXTwnyfG99x+11t6e5LeSfC3JAUm+1FrbPcntMv9S4otqmqQVhxyaCy74bI542uGZm5vLa0583TRztqhab6J5CNV6k3rN1XoTzUOo1ptoHkK13kQzm1ZxxtWaq/Um9Zqr9Saah1CtN9E8hGq9ieYhVOtN6jVX6000D6Fab6KZTas442rN1XqTOs3nff3KHPOolrf9zv0zu2Qmp537zVx9/Zoc+6iWNevncsOadXn9x74+7cxNqjLjjVVsrqbijKs1V+sF4L+b2db+RUxr7ZgkT05yXebPhHlykocnWZH5M07+MPPLh69Nck7v/cELt3vjIq7z+SSHZ34R8qlJvp9kXZI1Sf6o935Ra+13Mv/S2ldl/syT9++977Nw+zOSzPbej1y4/O3Mv5z3oUleleTahe4jM78YeVaS2ye5VZJX9d4/tuE2vfdVrbWTklzSez97czNZtXa6S5YAAGz/1q6r9ZBzdunMlq8EAAAAA7n4smumnbDV9tnj1tNOgJ/ZijeeP+2ErXbuSx4+7YTt3o+vXzPthK1225132PKVGJ1qx+2TZPlOMw7esyiX/vD6ep/gDOZev7jzxL+WbHNLk/xPliYBAJi0agdfLE0CAACwLbE0CdNhaZJNsTTJ9qLacfvE0iSLZ2mSzRliaXJbe3luAAAAAAAAAAAAtlMzsV/LdC2ZdgAAAAAAAAAAAADAECxNAgAAAAAAAAAAAKNgaRIAAAAAAAAAAAAYBUuTAAAAAAAAAAAAwChYmgQAAAAAAAAAAABGwdIkAAAAAAAAAAAAMAqWJgEAAAAAAAAAAIBRsDQJAAAAAAAAAAAAjIKlSQAAAAAAAAAAAGAULE0CAAAAAAAAAAAAozA77QAAAAAAAAAAAADGYWZm2gWMnTNNAgAAAAAAAAAAAKNgaRIAAAAAAAAAAAAYBUuTAAAAAAAAAAAAwChYmgQAAAAAAAAAAABGwdIkAAAAAAAAAAAAMAqWJgEAAAAAAAAAAIBRsDQJAAAAAAAAAAAAjIKlSQAAAAAAAAAAAGAULE0CAAAAAAAAAAAAo2BpEgAAAAAAAAAAABiF2WkHAAAAAAAAAAAAMA4z0w5g9Gbm5uam3cAWrFobf0gAAIVcee3qaSdstTvuuuO0EwCA7cjadfUOZ80udbgeWJx/+toV007YaofdZ7dpJ2z3Vq1ZN+2ErbZsh6XTToBReu3Hvz7thK1yzCF7Tzthq/l+hO1FxccXt72VT2YW55s/+mm9L9YM5h53utXEv5Z4eW4AAAAAAAAAAABgFCxNAgAAAAAAAAAAAKNgaRIAAAAAAAAAAAAYBUuTAAAAAAAAAAAAwChYmgQAAAAAAAAAAABGwdIkAAAAAAAAAAAAMAqWJgEAAAAAAAAAAIBRsDQJAAAAAAAAAAAAjMLstAMAAAAAAAAAAAAYiZlpBzB2zjQJAAAAAAAAAAAAjIKlSQAAAAAAAAAAAGAULE0CAAAAAAAAAAAAo2BpEgAAAAAAAAAAABgFS5MAAAAAAAAAAADAKFiaBAAAAAAAAAAAAEbB0iQAAAAAAAAAAAAwCpYmAQAAAAAAAAAAgFGwNAkAAAAAAAAAAACMgqVJAAAAAAAAAAAAYBRmpx0AAAAAAAAAAADAOMxkZtoJjJwzTQIAAAAAAAAAAACjYGkSAAAAAAAAAAAAGAUvzz1C69evz2tPOD6X9p4dd9wxx736xOy5117TzrpZ1XoTzUOo1pvUa67Wm2geQrXeRPMQqvUmNZuT5LlHPDm77LJLkmS3u+yel7/qxCkX3byKM9Y8edV6k3rN1XoTzUOo1ptoHkK13o199cKv5K1vPiVnnPWeaadsUbU5V+tN6jVX6000T9p3L/33fPSv356jXnNarvvJ1fnbv3xDfrry2qxfvz6HP/+VucNuu087cZMqzXiDas1r16zJCccfmx98//KsWb0mv//s5+bhB62YdtZmVZtxUq+5Wm+ieVLWr1ubf3vfaVn5Xz/M+rVrcp9HHp7lv3DnfOGcP89c5rLrL9wlv3L4/8fevcf7Ws/5/3+uvVfb7iQUk0OFKW+DUvkyM4kSFaZpDL+hGaSR0uBWqkEOo3QgJQyiA4kwkRlmchaSQ5hBcuoyYpBqKKlps9un9f1jrd13/5r2idZ1rde+7vfbbd32OnwOj/Xan9a+PrvXvj6HZ978+UOn3q4KM749novMrmrN1XqTmscXALNlgz7TZGvtmNbaRa21z7fWPtdae9h6Xv/Q1tpGM++/YObXPVtr59/mcie31g5qre3cWnvlzOeunfn14tbaA2e+vv8d8539fj77mYuy5JYlOe99788RRx6d0049eeikNarWm2juQ7XepF5ztd5Ecx+q9Saa+1CtN6nZvOSWW5JM5fVve2de/7Z3zumFyaTmjDXPvmq9Sb3mar2J5j5U600096Fa70rvOuftOeG4V+SWW5YMnbJOqs25Wm9Sr7lab6J5Nl384fflg2eckmVLp3+mffS8M7LLox+Xvzvhzdn3rw/OL37+04ELV6/KjFdVrfnjH70wW2xxl5z1zvfkjW89M687eW7/PUBSb8ZJveZqvYnm2fKTf/9cFmy6efY64pQ86rDj880PnpFvf+Td2XG/A/PYF56aJLn6O18duHL1Ksz4tjwXmX3Vmqv1JjWPLwBmywa7NNlae1CS/ZPs3XXdHkmOTHLOet7My5Ks/Oc3r1jbhbuuu6zruuNX87Vzu677t/W8/1nxzW98Pbvt/qgkyU4P3Tnf/e53Bi5as2q9ieY+VOtN6jVX600096Fab6K5D9V6k5rNV/5nl8WLF+fFhx+ao59/cL73nW8NnbRGFWesefZV603qNVfrTTT3oVpvorkP1XpXus822+R1b3jz0BnrrNqcq/Um9Zqr9SaaZ9OWW987z3zR//sf1f91xbdz4/W/zFmvOirf/MJF+cMH7zxg3ZpVmfGqqjU/dp9989znHz79wVQyf/7cf/G4ajNO6jVX6000z5b77LJ7HvLEZ8x8NJWJ+fOy27Nfmrtv/5AsX7Y0i2+6IRttvOmgjWtSYca35bnI7KvWXK03qXl8ATBbNuSfgDcm2TbJs1trn+i67rLW2iNaa3+c5I2ZXhj9eZKnJ3lgkjcnWZ5kcZJDkuydZOsk57fW/j3J3Vprb03ygdXdYWttzySHdV13wO187bgk1yb55yTvn7n/hTOXv+wO+Y7X0aJFN2fzzTe79eP58+Zn2bJlmZycmw+Har2J5j5U603qNVfrTTT3oVpvorkP1XqTms13WrgwT/2bZ+WJf/GUXPWzn+SlR/5d3vX+CzN/jjZXnLHm2VetN6nXXK030dyHar2J5j5U613psXvvm6t/ftXQGeus2pyr9Sb1mqv1Jppn045/skd+9Ytrbv34hl9em4033TyHHvv6fPqCc/O5D78v+x5w8ICFq1dlxquq1rzJJtPLTosWLcoxf//CHLZywWEOqzbjpF5ztd5E82zZ6E4bJ0mWLv5NvnzOa/KQJz4z8+bNz6Jf/SKfP/3l2WjjTXOXe99v4MrVqzDj2/JcZPZVa67Wm9Q8vgCYLRvsmSa7rvt5ps80+cgkl7bWrkiyX5Izkzy767o/TvLRJH+U5OwkL5g5I+Vbk7y+67p3ZHrJ8YCu605K8quu6543c/N7zbzs9sWttYuT/M16pD0iyfVJnpDk+Ul6/yc+m266WRYtWnTrxyumVszpP7ir9Saa+1CtN6nXXK030dyHar2J5j5U601qNt9n2/vmcY/fLxMTE9lm2/vmzlvcJddff93QWatVccaaZ1+13qRec7XeRHMfqvUmmvtQrbeqanOu1pvUa67Wm2ju0yabb5EHPfyRSZIHPWy3XHVlN3DR6lWcccXm/772mjzvkIPyhP3+PPs+cb+hc9aq4oyrNVfrTTTPpt/c8Mtc/JaXZbuHPybb/Z89kySb3u0eeeI/nJ0/fOQTctmH3j5s4BpUmXFlFWdcrbla70rVji8AZssGuzTZWts+yU1d1z2767ptkzwjyRlJHtR13feTpOu6d3Rd940k91rlbI+XJHnwWm7+s13X7bnyLcn71iPt40m+lORfkxyfZMV6XPcOscsuu+aLl1ySJLn8W5dlhx0e0HfCeqnWm2juQ7XepF5ztd5Ecx+q9Saa+1CtN6nZ/IkLP5Qz3vS6JMl1v/xFfrPo5my55VYDV61exRlrnn3VepN6zdV6E819qNabaO5Dtd6qqs25Wm9Sr7lab6K5T/d74I654htfSZL8+PuX5w+2mbtnCKs442rN119/XQ7/u0PygiOOyv5PesrQOeuk2oyTes3VehPNs2XxTTfk82/7h+z05wfl/n+yT5Lki2cfn//5xc+TJJN32jgTE3N3FaDCjKurOONqzdV6k5rHFwCzZe6vuf/udkpyaGtt/67rliT5QZJfJ7mqtbZD13X/2Vp7ycznr26t7dR13eVJ9pj5XDK90LjyaHLiDuraM8k1Xdft01r70ySvTvKYO+i218lej9s7l176pRz49AMyNTWV4098dZ93v96q9Saa+1CtN6nXXK030dyHar2J5j5U601qNj9h/yfnlBNeniMOPTCZmMiLXn7CnH1p7qTmjDXPvmq9Sb3mar2J5j5U600096Fab1XV5lytN6nXXK030dyn/Z71vHzwbafkK5/81yzcZNP89QtfOXTSalWccbXmc99+Vm666cacc9YZOeesM5Ikbzj9zCxcuHDgstWrNuOkXnO13kTzbPn+pz+Qpb+5Od/71Pn53qfOT5Ls+GcH5mvve2PmzZ/M5II75f8cMHdfdrfCjKurOONqzdV6k5rHFwCzZWJqamrohlnTWnt5kqcmuTnTy4+vTfLzJK/P9ELkNUmeleRBSf4x04uRy5Ic3HXdj1pr70qyXaaXGj87c923Jzms67oDVrmfk5NckeS/Vn6ttXZt13Vbz7x892FJDsj0y31fkOT8JBtlemn1+K7rPrWm72Pxsmy4v0kAABug6/5nydAJ622rzRcMnQAAbECWLa/311mT8++ofzMNbOg++f1rh05Yb/v+0dZDJ2zwFi9dPnTCelu40fyhE2CUTrroP4dOWC8vf9wOQyesN89H2FBUPL64y8YezKybH1+3uN4Pa3pzv60WzvrPkg16aXJDYWkSAKAWS5MAwNj5n5TAhszSJLen4lKDpUkYhqXJ2ef5CBuKiscXliZZV5YmWZM+libnrf0iAAAAAAAAAAAAAPVZmgQAAAAAAAAAAABGwdIkAAAAAAAAAAAAMAqWJgEAAAAAAAAAAIBRsDQJAAAAAAAAAAAAjIKlSQAAAAAAAAAAAGAULE0CAAAAAAAAAAAAo2BpEgAAAAAAAAAAABgFS5MAAAAAAAAAAADAKEwOHQAAAAAAAAAAAMA4TAwdwOg50yQAAAAAAAAAAAAwCpYmAQAAAAAAAAAAgFGwNAkAAAAAAAAAAACMgqVJAAAAAAAAAAAAYBQsTQIAAAAAAAAAAACjYGkSAAAAAAAAAAAAGAVLkwAAAAAAAAAAAMAoWJoEAAAAAAAAAAAARsHSJAAAAAAAAAAAADAKliYBAAAAAAAAAACAUZgcOgAAAAAAAAAAAICRmBg6gLFzpkkAAAAAAAAAAABgFCxNAgAAAAAAAAAAAKMwMTU1NXQDa7F4WfwmAQBAcb/+ThrGjgAAIABJREFUzdKhE9bbXTbZaOgEAAAAAAZ01yeeOnTCervhYy8aOgFGa+GkF11m3fzX9YvtQrFa991y4az/LHGmSQAAAAAAAAAAAGAULE0CAAAAAAAAAAAAo2BpEgAAAAAAAAAAABgFS5MAAAAAAAAAAADAKFiaBAAAAAAAAAAAAEbB0iQAAAAAAAAAAAAwCpNDBwAAAAAAAAAAADAOE5kYOoGRc6ZJAAAAAAAAAAAAYBQsTQIAAAAAAAAAAACjYGkSAAAAAAAAAAAAGAVLkwAAAAAAAAAAAMAoWJoEAAAAAAAAAAAARsHSJAAAAAAAAAAAADAKliYBAAAAAAAAAACAUbA0CQAAAAAAAAAAAIyCpUkAAAAAAAAAAABgFCxNAgAAAAAAAAAAAKMwOXQAAAAAAAAAAAAA4zAxMXQBY+dMkwAAAAAAAAAAAMAoWJoEAAAAAAAAAAAARsHSJAAAAAAAAAAAADAKk0MH0L8VK1bkpBOOyw+6LgsWLMixrzox22633dBZq1WtN9Hch2q9Sb3mar2J5j5U600096Fab6K5D9V6k5rNSXLDr67PIc98ak47/exsd9/7D52zRhVnXK25Wm+iuQ/VehPNfajWm2juQ7XepF5ztd5Ecx+q9Saa+1CtN9Hch2q9ieY+VOu9+102yZdPf2b+7JgLcsP/LM7pL9wnd918YebPm5eDT/lYfnzNr4dO/F+qzTjR3IdqvUnNZoDZUuJMk621Y1prF7XWPt9a+1xr7WGttTe21radhfua11p7WWvtC621i2fub8c7+D4Oaq2d3Fq7b2vtK3fkba+Lz37moiy5ZUnOe9/7c8SRR+e0U0/uO2G9VOtNNPehWm9Sr7lab6K5D9V6E819qNabaO5Dtd6kZvOyZUvzute8KndauHDolHVSccbVmqv1Jpr7UK030dyHar2J5j5U603qNVfrTTT3oVpvorkP1XoTzX2o1pto7kOl3sn58/KWI/bJb29ZliQ56Tl75P2f/X72Pvr8HHfuF9K2udvAhbev0oxX0jz7qvUmNZsBZsucX5psrT0oyf5J9u66bo8kRyY5p+u6F3Zd99NZuMsXJ9kqyR5d1+058/G/ttY2moX7GsQ3v/H17Lb7o5IkOz1053z3u98ZuGjNqvUmmvtQrTep11ytN9Hch2q9ieY+VOtNNPehWm9Ss/mtb3xd/uLJT8tWW9196JR1UnHG1Zqr9Saa+1CtN9Hch2q9ieY+VOtN6jVX600096Fab6K5D9V6E819qNabaO5Dpd6TD90zZ3/kslxz/c1Jkj998L1z77tvno+e/NQcsNeDcsnlPxu48PZVmvFKmmdftd6kZjPAbKnw8tw3Jtk2ybNba5/ouu6y1tojWmsXJzksyQFJts/0ouOWSU5P8pQkD0jyrK7rvtJaO3rmcsuSXNJ13Utaa8cluV+SeyTZLsmRXdd9MsmhSR7Wdd2KJOm67t9baw/vum5pa22XJG9OsjzJ4iSHZHqJ80td132wtfaJJJ/quu71rbWzk7wzyUZJTpq5zpVJnjur01oHixbdnM033+zWj+fPm59ly5ZlcnJuPhyq9Saa+1CtN6nXXK030dyHar2J5j5U600096Fab1Kv+eMXfjh3uevd8og/fWTee+7ZQ+esk2ozTuo1V+tNNPehWm+iuQ/VehPNfajWm9RrrtabaO5Dtd5Ecx+q9Saa+1CtN9Hchyq9z9j7wfnljb/JRV//r7zogD9Okmz3B3fODf+zOH92zAfy0qf/aY5+6iNywru/NHDp/1ZlxqvSPPuq9SY1mwFmy5w/02TXdT/P9JkmH5nk0tbaFUn2u83Fftt13eOT/HOSJ3Zd9+dJTk5ywMxLaz81yW4zbzu01lZe/5au656Q5IhMLz8mySZd191wm4brZ949O8kLZs54+dYkr0/yoSRPaK1tnOSuSR7bWptI8rAkl85c58kz1/l5koN+35n8vjbddLMsWrTo1o9XTK2Y038IVutNNPehWm9Sr7lab6K5D9V6E819qNabaO5Dtd6kXvPHLvxQ/uOrl+aI5x6UH/6gy6uPfVmuv+66obPWqNqMk3rN1XoTzX2o1pto7kO13kRzH6r1JvWaq/UmmvtQrTfR3IdqvYnmPlTrTTT3oUrvsx6/Yx67633zyVOflp3+8B55x4uemOUrpvLRS3+YJPnYV67Mrg/YeuDK21dlxqvSPPuq9SY1mwFmy5xfmmytbZ/kpq7rnt113bZJnpHkjCR3W+Vi35j59ddJvjfz/g1JFiZ5YJKvdF23tOu6qSRfSPLgmct8c+bXn81cNkluaK3d+TYNfznzuXt1XXfZzKcvmbmdLybZNcljMr20efckj8r0wuTdk9wzyQdmzoy5T6bPajmoXXbZNV+85JIkyeXfuiw77PCAgYvWrFpvorkP1XqTes3VehPNfajWm2juQ7XeRHMfqvUm9ZrffNa78qazzs0/nnlutn9Ay8te9epsudVWQ2etUbUZJ/Waq/UmmvtQrTfR3IdqvYnmPlTrTeo1V+tNNPehWm+iuQ/VehPNfajWm2juQ5XevY8+P/v8/fnZ90Xvz+VX/iIHn/qxfPyrV2bfR9w/SbL7jvfJ938yN/9BcJUZr0rz7KvWm9RsBpgtFVbGd0pyaGtt/67rliT5QaaXI5evcpmpNVz/iiRHt9YmZ67z6CTvTvLQ1VzvXUmOba39fdd1U6213TJ9RsmW5OrW2k5d112eZI8kP+i6bkVr7T+SvDjJC5NsneSUJC9Pcl2Sq5L8Rdd1N7bW9k9yc6Zfbnwwez1u71x66Zdy4NMPyNTUVI4/8dVD5qxVtd5Ecx+q9Sb1mqv1Jpr7UK030dyHar2J5j5U601qNldTccbVmqv1Jpr7UK030dyHar2J5j5U603qNVfrTTT3oVpvorkP1XoTzX2o1pto7kO13lUdc+bFeetR++bQ/XbOjb+5JQe95iNDJ92uijPWPPuq9SY1mwFmy8TU1Jr2DeeG1trLM/0S2zdn+uyYr830guJhSQ5Icm3XdWe01g5LsnXXdce11p6U5PFd1x3WWjsqydNmrvvFJEclOXaV6z0wyRld1+3ZWpuf5IQkeyVZOvN2RNd1326t7ZLkH5NMJFmW5OCu637UWntCknMzvTC5T5L3znQsa63tk+SVM/d9U5IDkzwx02fAPCPJ+V3X/cmavv/Fy9a4FAoAABTw698sHTphvd1lk42GTgAAAABgQHd94qlDJ6y3Gz72oqETYLQWTmZi6AZq+NmvbrELxWptc7c7zfrPkhJLk2NnaRIAAOqzNAkAAABANZYmgfVhaZJ1ZWmSNeljaXLebN8BAAAAAAAAAAAAwFxgaRIAAAAAAAAAAAAYBUuTAAAAAAAAAAAAwChYmgQAAAAAAAAAAABGwdIkAAAAAAAAAAAAMAqWJgEAAAAAAAAAAIBRsDQJAAAAAAAAAAAAjIKlSQAAAAAAAAAAAGAUJocOAAAAAAAAAAAAYBwmJoYuYOycaRIAAAAAAAAAAAAYBUuTAAAAAAAAAAAAwChYmgQAAAAAAAAAAABGwdIkAAAAAAAAAAAAMAqWJgEAAAAAAAAAAIBRsDQJAAAAAAAAAAAAjIKlSQAAAAAAAAAAAGAULE0CAAAAAAAAAAAAo2BpEgAAAAAAAAAAABgFS5MAAAAAAAAAAADAKEwOHQAAAAAAAAAAAMBYTAwdwMg50yQAAAAAAAAAAAAwCpYmAQAAAAAAAAAAgFGwNAkAAAAAAAAAAACMwsTU1NTQDazF4mXxmwQAAAAAADBHffmH1w+dsN52237LoRPg9/atn9w4dMJ6a/fabOiE9bJwo/lDJ6y30z7/w6ET1tvRe2w/dALcIRZOZmLoBmq46oYldqFYrfvcdcGs/yxxpkkAAAAAAAAAAABgFCxNAgAAAAAAAAAAAKNgaRIAAAAAAAAAAAAYBUuTAAAAAAAAAAAAwChYmgQAAAAAAAAAAABGYXLoAAAAAAAAAAAAAMZhYmLoAsbOmSYBAAAAAAAAAACAUbA0CQAAAAAAAAAAAIyCpUkAAAAAAAAAAABgFCxNAgAAAAAAAAAAAKNgaRIAAAAAAAAAAAAYBUuTAAAAAAAAAAAAwChYmgQAAAAAAAAAAABGwdIkAAAAAAAAAAAAMAqWJgEAAAAAAAAAAIBRsDQJAAAAAAAAAAAAjMLk0AEAAAAAAAAAAACMw8TQAYyeM00CAAAAAAAAAAAAo2BpEgAAAAAAAAAAABgFS5MAAAAAAAAAAADAKFiaBAAAAAAAAAAAAEbB0iQAAAAAAAAAAAAwCpNDB9C/FStW5KQTjssPui4LFizIsa86Mdtut93QWatVrTfR3IdqvUm95mq9ieY+VOtNNPehWm+iuQ/VehPNfajWm9RrrtabaO5Dtd5Ecx+q9Saa+1CtN6nXXK030dyHar2J5tn24+67+dC735qjTjr91s997fOfysUfvSAvPuXsAcvWrNKMV6rWXK03qdV85RXfyfvf+Za87LVn5CdX/iDnnXFq5s2bn8mNNspzjz4uW9x1y6ETb9eypUtzwnGvyDVX/zxLlyzN3x7y3Dx6z72GzlqjCo+LFcuX5cvnvTE3X/+LLF+2NDs94WnZ9K53z9c+cGYm5s3LvMmNsvuzjsrGd77r0Km3q8KMb6tac7XepGYzwGzZ4M402Vo7rbV2cWvtitbaT2fev2A9b+Oq1trnZ6678m3nNVz+i6217X//+n589jMXZcktS3Le+96fI448OqedevLQSWtUrTfR3IdqvUm95mq9ieY+VOtNNPehWm+iuQ/VehPNfajWm9RrrtabaO5Dtd5Ecx+q9Saa+1CtN6nXXK030dyHar2J5tn0qX95T95z+muydMmSWz/3sx91+fJFF2ZqasCwdVBlxquq1lytN6nT/NEL3p13/ONJt/63954zT8szD3tRXvbaM/J/dntMPnLBuwcuXL2Pf/TCbLHFXXLWO9+TN771zLzu5BOHTlqrCo+LH33tc7nTpnfO448+JY97wfH56vvPyL9fcFYe8dTnZt8jT852O++W73zqg0NnrlaFGd9WteZqvUnNZoDZssGdabLruqOTpLV2UJIHdl13zO94U4/tum7ZHRY2h3zzG1/Pbrs/Kkmy00N3zne/+52Bi9asWm+iuQ/VepN6zdV6E819qNabaO5Dtd5Ecx+q9Saa+1CtN6nXXK030dyHar2J5j5U600096Fab1KvuVpvorkP1XoTzbNpq63vnece85q88w3HJ0luvunGfPi8M/NXB78w7zl9bi82VJnxqqo1V+tN6jTf4573yeGveG3OfN1xSZLnH3NS7nK3rZIkK5Yvz4IFdxqwbs0eu8++2Wvvfac/mErmz5/7KwAVHhfb7bJ7ttvlkUmSqampzJs3L486+CXZZIu7JUlWrFie+RstGDJxjSrM+LaqNVfrTWo2A8yWuX/EdAdorU0mOSvJvZPcM8m/dF13XGvtPUluTnK/JFsnObDrum+t4XYWJDkzyf2TzE/y0q7rvrCay34gyTu6rvtka+0hSU5KcmGSvZNsNnMbr+667rw76NtcZ4sW3ZzNN9/s1o/nz5ufZcuWZXJybj4cqvUmmvtQrTep11ytN9Hch2q9ieY+VOtNNPehWm+iuQ/VepN6zdV6E819qNabaO5Dtd5Ecx+q9Sb1mqv1Jpr7UK030Tybdt3tMbn+v69JMr2o9Z63vDr/37MPn9MLWytVmfGqqjVX603qND98973yy/+++taPVy5M/uf3Ls+nP3JBXn7KmUOlrdUmm2yaJFm0aFGO+fsX5rDnHz5w0dpVeFxstHDjJMnSxb/J589+dXbe/8BbFyZ/ceX30l18YfY96pQhE9eowoxvq1pztd6kZjPAbNngXp57NbZN8sWu6/ZN8sdJnr/K13408/kzkhyyyuc/s8pLc39q5nPPTXJ113V7JHlSkret4T7PTvKsmfcPTvKOmfc377ruz5I8OclLfp9v6ne16aabZdGiRbd+vGJqxZz+Q7Bab6K5D9V6k3rN1XoTzX2o1pto7kO13kRzH6r1Jpr7UK03qddcrTfR3IdqvYnmPlTrTTT3oVpvUq+5Wm+iuQ/VehPNffnplV1+cfVV+aczTs3bX/fKXPuzH+cDb3/j0FmrVXHG1Zqr9SY1m1f6yuc/nXPfcnKOPu4NufMWdx06Z43++9pr8rxDDsoT9vvz7PvE/YbOWasqj4tFv/plPvXGl+b+f7xX7v/wPZMkP/6PS/KVfzo9ez3/uCzcfIthA9egyoxXVa25Wm9SsxlgtoxlafL6JH/aWntvktOSrHqe7G/O/PqzJAtX+fxju67bc+Ztn5nP7Zjkz1trFyf5YJIFrbW7rOY+L0ry0NbaVkn2SvLRtdxfb3bZZdd88ZJLkiSXf+uy7LDDA4bIWGfVehPNfajWm9RrrtabaO5Dtd5Ecx+q9Saa+1CtN9Hch2q9Sb3mar2J5j5U600096Fab6K5D9V6k3rN1XoTzX2o1pto7st9H/CgvPIt781RJ52e5/z98dl6m/vlqc954dBZq1VxxtWaq/UmNZuT5Euf/Xgu+sgH8tLXvi33uOe9h85Zo+uvvy6H/90hecERR2X/Jz1l6Jx1UuFx8dubbshFb35Fdn3S32aH3abXBX701c+m+/xHsu+RJ2fzre45cOGaVZjxbVVrrtab1GxmwzUx4c3b6t/6MJaV8YOT/KLrupe31lqSQ1f52tR63M4VSX7Ydd0prbVNkrwsyY23d8Gu66Zaa+9L8uYkH+u6bvn0Xa/X/c2KvR63dy699Es58OkHZGpqKsef+Oqhk9aoWm+iuQ/VepN6zdV6E819qNabaO5Dtd5Ecx+q9Saa+1CtN6nXXK030dyHar2J5j5U600096Fab1KvuVpvorkP1XoTzdy+ijOu1lytN6nZvGL58rznjNOy5T3+IG86cfqFAx+446558jMOXcs1h3Hu28/KTTfdmHPOOiPnnHVGkuQNp5+ZhQsHOX/POqnwuPj2Jz6QW357cy7/+Pm5/OPnZ2rFivz66v/KplveIxefdVKS5A92eEh23u8ZA5fevgozvq1qzdV6k5rNALNlYmpq8B2+WdFaOyjJA7uuO6a1tlOS85L8OsktSe6b5FGZPuvkuV3XXdRa2y/Jk7que05r7aokV+b/v+D4+iSfzvTLbm+T5M5J3tx13TmttS8m2WLmtpPkM13XvaS1dq8kP0nyR13X/bC19pwk9+267hWttc2SXNZ13fZr+14WLxt+0RIAAAAAAIDb9+UfXj90wnrbbfsth06A39u3fnK757eZ09q9Nhs6Yb0s3Gj+0Anr7bTP/3DohPV29B5rXRuAEhZOpqdzxFHdNTcusQvFat1ziwWz/rNkg12anAtaa9skeccqL+/9O7E0CQAAAAAAMHdZmoRhWJqcfZYm+2Fpkg2FpUnWlaVJ1qSPpcl5s30HY9Va+6skH0vyyqFbAAAAAAAAAAAAgGRy6IANVdd1FyS5YOgOAAAAAAAAAAAAYJozTQIAAAAAAAAAAACjYGkSAAAAAAAAAAAAGAVLkwAAAAAAAAAAAMAoWJoEAAAAAAAAAAAARmFy6AAAAAAAAAAAAADGYSITQycwcs40CQAAAAAAAAAAAIyCpUkAAAAAAAAAAABgFCxNAgAAAAAAAAAAAKNgaRIAAAAAAAAAAAAYBUuTAAAAAAAAAAAAwChYmgQAAAAAAAAAAABGwdIkAAAAAAAAAAAAMAqWJgEAAAAAAAAAAIBRsDQJAAAAAAAAAAAAjIKlSQAAAAAAAAAAAGAUJocOAAAAAAAAAAAAYCQmhg5g7JxpEgAAAAAAAAAAABgFS5MAAAAAAAAAAADAKFiaBAAAAAAAAAAAAEbB0iQAAAAAAAAAAAAwChNTU1NDN7AWi5fFbxIAv5ObFy8bOmG9bbZwcugEAAA2cI6TgXW1bHm9v5qdnD8xdAJzULXHsscxDMNxMgxn+8M/PHTCevnhm540dAJz1MLJOJBjnVx709JaT1Lo1dZ33mjWf5Y40yQAAAAAAAAAAAAwCpYmAQAAAAAAAAAAgFGwNAkAAAAAAAAAAACMgqVJAAAAAAAAAAAAYBQmhw4AAAAAAAAAAABgHCaGDmD0nGkSAAAAAAAAAAAAGAVLkwAAAAAAAAAAAMAoWJoEAAAAAAAAAAAARsHSJAAAAAAAAAAAADAKliYBAAAAAAAAAACAUbA0CQAAAAAAAAAAAIyCpUkAAAAAAAAAAABgFCxNAgAAAAAAAAAAAKNgaRIAAAAAAAAAAAAYBUuTAAAAAAAAAAAAwChMDh0AAAAAAAAAAADAOExMDF3A2DnTJAAAAAAAAAAAADAKliYBAAAAAAAAAACAUbA0CQAAAAAAAAAAAIyCpUkAAAAAAAAAAABgFCxNAgAAAAAAAAAAAKNgaRIAAAAAAAAAAAAYhcmhA+jfihUrctIJx+UHXZcFCxbk2FedmG23227orNWq1pto7kO13qRec7XeRHMfli9fnteeeGx++pMfZyITedHLjs39t99h6Kw1qjbjpF5ztd5Ecx+q9Saa+1CtN6nXXK030dyHar1JzeZqx8oVZ6x59lXrTWo2J8m3L/9W3vzG1+Wsc84bOmWtKs64WnO13lV5LM+ear2J5j5U603qHScn9eZcrTep11yld95EcsrTd8kf/sFmmZpKXvpPl2WjyXk5+a93zi3Llud7V92UV15weaamhi69fVXmvFK13qRmM8BsmRNnmmytndZau7i1dkVr7acz7/+ytXb+QD3nttYun+n4XGvtktbag9dw+eNaa4et4es7ttYePfP+o1trO828/y93fP3affYzF2XJLUty3vvenyOOPDqnnXryEBnrrFpvorkP1XqTes3VehPNffjSJRcnSc4457055HmH58y3/uOwQeug2oyTes3VehPNfajWm2juQ7XepF5ztd5Ecx+q9SY1m6sdK1ecsebZV603qdn8rnPenhOOe0VuuWXJ0CnrpOKMqzVX613JY3l2VetNNPehWm9S7zg5qTfnar1JveYqvXvvdM8kyV+e9oWceuH38+L9H5TX/s3OOfaD385TXv/F3PTbpfnLh99n4MrVqzLnlar1JjWbAWbLnFia7Lru6K7r9kxycpL3zbz/V4NGJS/uum7Prusek+Q1SU74PW7rKUkeNPP+s5PcK0m6rnvy75f4u/nmN76e3XZ/VJJkp4funO9+9ztDZKyzar2J5j5U603qNVfrTTT34dGPeWxe/PLjkiTXXnN1Ntts82GD1kG1GSf1mqv1Jpr7UK030dyHar1JveZqvYnmPlTrTWo2VztWrjhjzbOvWm9Ss/k+22yT173hzUNnrLOKM67WXK13JY/l2VWtN9Hch2q9Sb3j5KTenKv1JvWaq/R+8lvX5CXvuyxJcu+7bZybfrs097zLxvn6j36VJPmPK6/Pw/9wyyET16jKnFeq1pvUbAaYLXP95bl3aK19PMk9klzYdd1xrbUdk7wpyUSS6zO9hLhLktcmWZLkrCQ/TXJSkuVJrkzy3CT3S/LOJMsyvSz6N13X/ay19pokj0oyP8nru6674HY67pbk5iRprR2d5ICZ27mk67qXrHrB295eki8nOSjJktbaN5M8PsmurbXvJfla13Vb/75DWl+LFt2czTff7NaP58+bn2XLlmVycm4+HKr1Jpr7UK03qddcrTfR3JfJycmc8MqX5pKLP5MTX/uGoXPWquKMqzVX600096Fab6K5D9V6k3rN1XoTzX2o1pvUbE5qHStXnLHm2VetN6nZ/Ni9983VP79q6Ix1VnHG1Zqr9a7ksTy7qvUmmvtQrXelSsfJSb05V+tN6jVX6l2+YipvOHDXPP6h98xz3/613O8em+ZPdtgyX/nP6/O4nbbOJgvmXvNKleac1OtNajYDzJa5/pNvYZInZXoB8adJjktydpJnd133vdbawUlenOTTSRZ2XffHrbWJJF2S3buu+0Vr7YRMLy0uSPK1mcs/KskWrbWHJLlf13W7t9YWJvlKa+3TM/d9SmvtmEwvXl6d5MUzC5tPTbJbppcm/7m1tt/K2NbaE257e0n2THJukmu7rvtqa+0TSc7vuu6nrbVZGNnabbrpZlm0aNGtH6+YWjGn/xCs1pto7kO13qRec7XeRHOf/uH41+T6636ZQ57113nvB/8tG2+8ydBJq1VxxtWaq/UmmvtQrTfR3IdqvUm95mq9ieY+VOtNajavVOVYueKMNc++ar1JzeZqKs64WnO13qqqzblab6K5D9V6V1XlODmpN+dqvUm95mq9R777G3n1ne+UC1+0R/72jK/k5U96cF74hHn52pXXZ8nSFUPnrVa1OVfrTWo2s+GayMTQCYzcnHh57jX4Ttd1t3Rd95tMLykmyR8leWtr7eJMn2Xy3jOf72Z+vXuSeyb5wMxl9kmyXZJ3JPl1kk8kecHM7e2Y5GEzl/tEko2S3Hfmdla+PPdju657Ztd11yR5YJKvdF23tOu6qSRfSPLgVXrXdHtzxi677JovXnJJkuTyb12WHXZ4wMBFa1atN9Hch2q9Sb3mar2J5j584qP/lnefc3aSZOHCjTNv3rzMm5jbhxPVZpzUa67Wm2juQ7XeRHMfqvUm9Zqr9Saa+1CtN6nZXO1YueKMNc++ar1JzeZqKs64WnO13qqqzblab6K5D9V6k3rHyUm9OVfrTeo1V+l9yiO2yfP33SFJ8tsly7NiaiqP23HrvODc/8gBb/pS7rrpgnzhil8OXLl6Vea8UrXepGYzwGyZ6yvjU7fzuS7JgTNnanxkphckk2TlP4m4LslVSf6i67obW2v7Z/qltf8iyRe6rntVa+2vk7wkyYeSfK7rukNba/OS/EOmX857da5IcnRrbTLTZ6B8dJJ3J3noKl+/vdtbkf+3oLrq+4PY63F759JLv5QDn35ApqamcvyJrx4yZ62q9Saa+1CtN6mq3VUYAAAgAElEQVTXXK030dyHPfZ6XF593CvyvOccmGXLluWIo4/JnRYuHDprjarNOKnXXK030dyHar2J5j5U603qNVfrTTT3oVpvUrO52rFyxRlrnn3VepOazdVUnHG15mq9VVWbc7XeRHMfqvUm9Y6Tk3pzrtab1Guu0vuxy67O65+5az545O7ZaP68HPfBb2fFVPL+w3fPb5cuy5d/cF0++93/HjpztarMeaVqvUnNZoDZMjE1dXt7icNorR2U5IFd1x3TWtszyWFd1x0w87Vru67burX2sCSnZXrhcyrJwUnudZvL7pPklZleTrwpyYFJNk/yriRLMv1y30cm+ebMbT08yWZJPtR13fGttXMz/RLan7idxqOSPG3mtr+Y5Kgkxya5NsmZq7m9P0tyapLnJ3nAzK9Py/SC5dZrm8viZbe7PAoAa3Xz4mVrv9Acs9nCuf5vOgAAqM5xMrCuli2v91ezk/O9xBn/W7XHsscxDMNxMgxn+8M/PHTCevnhm540dAJz1MJJr7nMuvnl/yyr9SSFXt1988lZ/1kyp5YmuX2WJgH4XflLLgAA+N8cJwPrqtqiWWLZjNtX7bHscQzDcJwMw7E0yYbC0iTrytIka9LH0uSgLxMNAAAAAAAAAAAA0BdLkwAAAAAAAAAAAMAoWJoEAAAAAAAAAAAARsHSJAAAAAAAAAAAADAKliYBAAAAAAAAAACAUZgcOgAAAAAAAAAAAICRmBg6gLFzpkkAAAAAAAAAAABgFCxNAgAAAAAAAAAAAKNgaRIAAAAAAAAAAAAYBUuTAAAAAAAAAAAAwChYmgQAAAAAAAAAAABGwdIkAAAAAAAAAAAAMAqWJgEAAAAAAAAAAIBRsDQJAAAAAAAAAAAAjIKlSQAAAAAAAAAAAGAULE0CAAAAAAAAAAAAozA5dAAAAAAAAAAAAADjMDF0AKPnTJMAAAAAAAAAAADAKFiaBAAAAAAAAAAAAEbB0iQAAAAAAAAAAAAwCpYmAQAAAAAAAAAAgFGwNAkAAAAAAAAAAACMgqVJAAAAAAAAAAAAYBQmpqamhm5gLW6+pdZv0uT8iaETAGbFzYuXDZ2w3jZbODl0AnPQsuWlDi2SOL4A1s3ipcuHTlhvCzeaP3TCeqs254ozrsjjAoDfleeoADC3VXu+l3jO14ddX/mpoRPW2zeO32fohFFYOBkHy6yT625eVu/JIL3ZarPJWf9Z4kyTAAAAAAAAAAAAwChYmgQAAAAAAAAAAABGwdIkAAAAAAAAAAAAMAqTQwcAAAAAAAAAAAAwDhMTQxcwds40CQAAAAAAAAAAAIyCpUkAAAAAAAAAAABgFCxNAgAAAAAAAAAAAKNgaRIAAAAAAAAAAAAYBUuTAAAAAAAAAAAAwChYmgQAAAAAAAAAAABGwdIkAAAAAAAAAAAAMAqWJgEAAAAAAAAAAIBRsDQJAAAAAAAAAAAAjMLk0AEAAAAAAAAAAACMw0Qmhk5g5JxpEgAAAAAAAAAAABgFS5MAAAAAAAAAAADAKFiaBAAAAAAAAAAAAEbB0iQAAAAAAAAAAAAwCpYmAQAAAAAAAAAAgFGwNAkAAAAAAAAAAACMgqVJAAAAAAAAAAAAYBQmhw5gGN++/Ft58xtfl7POOW/olLVasWJFTjrhuPyg67JgwYIc+6oTs+122w2dtUaaZ1+13qRec7XepGbz8uXL89oTj81Pf/LjTGQiL3rZsbn/9jsMnbVaFWesuT+OL2ZXteZqvYnmPlTrTZJlS5fmhONekWuu/nmWLlmavz3kuXn0nnsNnbVaZtyPanOu1pt4XPShWm+iuQ/VepN6zdV6k5rNieeos03z7KvWm2juQ7XeRHMfqvUm9Z7zVZxxleZ5E8nxf/ng3Pfum2RqKnnVv34/v7p5SY7/ywflzhtvlHnzkpde8J387Fe/HTr1f6ky41VVbAaYLWXPNNlaO621dnFr7YrW2k9n3v90a+2V63j9g1a53sq3/We7+zYNJ7f2f9m783hd53r/46+199ps2xaiyBAJHw0nUoYc85A46XDqnJxy/ChTGiQNphAyFAkl446kSNNJ4hSSTI0iykdCMmWmve1p7b1+f1zXYtn2XoNa93V99/16Ph7rsdZ9r3vd672/+1rX8L0+3+83duvk7wQ4b8rZHHXEocycOavTv/pFuerKK5g1cxbnf+Mi9tv/AE78/HFNRxqWmcdeaXmhvMyl5YUyM193zdUAnD7lAvbc9yOccdrJzQYaRoltbObO8Pxi7JWWubS8YOZOKC0vwGWXXsKSSy7FmV/9Ol887QxOOO7opiMNyTbujNLaubS84HbRCaXlBTN3Qml5obzMpeWFMjN7jTr2zDz2SssLZu6E0vKCmTuhtLxQ3jVfiW1cSuYt1noZALuc8StO+cmdfHSb1TngbWvww5sfZNezqude9bLFG045f6W08WAlZpaksVLsTJOZeQBUxY/AWpl54It4m2+8yJ8r2korr8wJJ53Kpw/+VNNRRuSm3/6GjTbeBIA3rL0Ot912a8OJhmfmsVdaXigvc2l5oczMm26xFRttshkADz34AJMnL9FwoqGV2MZm7gzPL8ZeaZlLywtm7oTS8gJs9dZt2XKbbasH/TB+fLsvo23jziitnUvLC24XnVBaXjBzJ5SWF8rLXFpeKDOz16hjz8xjr7S8YOZOKC0vmLkTSssL5V3zldjGpWS+8o+PcHU+CsAKSy3G0zP6WHeVpbjjoamc8743cf8T0zn2h9lwyvkrpY0HKzGzJI2Vdp99jFJEbA7sk5k7R8SdwPXAmsCVwJLA+kBm5v8M8R4rAV8BJgKvAA7NzO9HxK3AHcAs4HZgI2Ay8H5gV+DNwDLAzZm5e0QcAbwKeDmwCrB/Zv5fRLwTOBR4BFikfq+O2mqbbXng/vs6/WtftGnTprLEEpOffTx+3Hj6+vro7W3v5mvmsVdaXigvc2l5oczMAL29vRx12EFcc/WVHH38SU3HGVKJbWzmzvD8YuyVlrm0vGDmTigtL8CkSdVI9mnTpnHgxz/KPh/8SMOJhmYbd0Zp7VxaXnC76ITS8oKZO6G0vFBe5tLyQpmZvUYde2Yee6XlBTN3Qml5wcydUFpeKO+ar8Q2LinznLn9HPOu17P1a1/OR79xM/+29vI8PX0275/yGz6w5Wq8f7NV+dIVf2465guU1MYDSsyshVdPT9MJ1O2KXZ57BFalKk7cBPgIcBqwAbBxRCxVv+Y9g5bmvrh+bi3gxMzcBtgL+GD9/GTgqMzcuX78x8zcCLgfeKJ+/ZuBDSNixfo1MzNzO2A/YP+ImAB8Adga2BZ4Ziz+4QubxRefzLRp0559PLd/busP2mYee6XlhfIyl5YXysw84NNHHsuF372U448+nOnT23t4KLGNzaz5KbGNS8tcWl4wcyeUlnfA3x56kH333I3t3r4D227/9qbjDMk27ozS2rm0vAPcLsZWaXnBzJ1QWl4oL3NpeaHMzKUpsY3NPPZKywtm7oTS8oKZO6G0vANKuuYrsY1Ly3zwt29l+y9cy5E7vZa/z+jjqj8+AsDVf3yE16/4kobTzV9pbQxlZpaksbIwF00+lpn3ZuZsYFpm/iEz+4GnqGaRhGp57s3rj/+sn3sQ2Dsizgf2ASYMes+cz9fTgZdHxDeBM6iKKwd+5qb681/r3/ky4PHMfKzOcv0/7V+7EHvjG9fl2muuAeCWm3/HGmus2XCi4Zl57JWWF8rLXFpeKDPz5Zf+gK9NOQuAiRMXY9y4cYzrae/hucQ2NrPmp8Q2Li1zaXnBzJ1QWl6Axx57lI98YE8+tN/HeMeO72w6zrBs484orZ1LywtuF51QWl4wcyeUlhfKy1xaXigzc2lKbGMzj73S8oKZO6G0vGDmTigtL5R3zVdiG5eSeYd1XsGem70KgOmz5zC3H3599xNsGssC8KZXLc2df5s21Fs0ppQ2HqzEzJI0VhbmkvH+F/lzRwFnZeZlEbE7sNug782dz9fbAStn5rsj4mXATsDAJLLzZngYWCoiXpaZjwDrAeWsEdKQLbfehhtuuI5d37sz/f39HHn0MU1HGpaZx15peaG8zKXlhTIzb7bl1hxzxKHsu8eu9PX1sd8BB7LoxInD/2BDSmxjM2t+Smzj0jKXlhfM3Aml5QU49+wzefrpp5hy5ulMOfN0AE768hlMbOnx2jbujNLaubS84HbRCaXlBTN3Qml5obzMpeWFMjOXpsQ2NvPYKy0vmLkTSssLZu6E0vJCedd8JbZxKZmvuO1hPvuu1/G1Pdejd3wPx116O7c/+HeO3Ol17LzBykyd0ccnLrql6ZjzVUobD1ZiZkkaKz39/S+2trAdImI3YK3MPDAiNgf2ycydI+KhzFy+fs3gr38HvK3+WCszD5zn/f6balnvx6gKGtfOzNdFxD3162dExBHAQ5l5ekQsD1xCNeNkP7AYsD+wzaDXrAWcnpmbR8S/URVmPg7MBi7KzHOH+jdOnVnWf1Lv+J7hXyRJBZo6o6/pCKM2eeLCPD5CL1bfnKJOLQDPLySNzIzZc5qOMGoTJ4xvOsKoldbOJbZxidwuJEkvlteokiS1W2nXe+A1Xyese9iPm44war898q1NR+gKE3vxZFkj8sQzBV4MqmOWnjT2F97FF012A4smJakdLJrUwsIbUpIWVnbid0Zp7VxiG5fI7UKS9GJ5jSpJUruVdr0HXvN1gkWTWhCLJjVSFk1qKJ0omhw31r9AkiRJkiRJkiRJkiRJkiSpDSyalCRJkiRJkiRJkiRJkiRJXcGiSUmSJEmSJEmSJEmSJEmS1BUsmpQkSZIkSZIkSZIkSZIkSV3BoklJkiRJkiRJkiRJkiRJktQVLJqUJEmSJEmSJEmSJEmSJEldwaJJSZIkSZIkSZIkSZIkSZLUFSyalCRJkiRJkiRJkiRJkiRJXcGiSUmSJEmSJEmSJEmSJEmS1BUsmpQkSZIkSZIkSZIkSZIkSV3BoklJkiRJkiRJkiRJkiRJktQVLJqUJEmSJEmSJEmSJEmSJEldwaJJSZIkSZIkSZIkSZIkSZLUFSyalCRJkiRJkiRJkiRJkiRJXcGiSUmSJEmSJEmSJEmSJEmS1BV6mw4gSZIkSZIkSZIkSZIkSeoOPT1NJ1C3c6ZJSZIkSZIkSZIkSZIkSZLUFSyalCRJkiRJkiRJkiRJkiRJXcGiSUmSJEmSJEmSJEmSJEmS1BUsmpQkSZIkSZIkSZIkSZIkSV3BoklJkiRJkiRJkiRJkiRJktQVLJqUJEmSJEmSJEmSJEmSJEldwaJJSZIkSZIkSZIkSZIkSZLUFXr6+/ubzqBhzOjD/ySpw6bO6Gs6wqhNntjbdIRRu/NvU5uOMCqrLze56QiSJGkh43nn2OubU94lde/4nqYjSJIKVtqxz+Oe1Bz3F1Iz/NuTmrHD6Tc2HWHULtlnw6YjjNrEXtxpaESenF7YAVEdtdRiY38C4kyTkiRJkiRJkiRJkiRJkiSpK1g0KUmSJEmSJEmSJEmSJEmSukJZa2pJkiRJkiRJkiRJkiRJkorV40ruapgzTUqSJEmSJEmSJEmSJEmSpK5g0aQkSZIkSZIkSZIkSZIkSeoKFk1KkiRJkiRJkiRJkiRJkqSuYNGkJEmSJEmSJEmSJEmSJEnqChZNSpIkSZIkSZIkSZIkSZKkrmDRpCRJkiRJkiRJkiRJkiRJ6goWTUqSJEmSJEmSJEmSJEmSpK5g0aQkSZIkSZIkSZIkSZIkSeoKFk1KkiRJkiRJkiRJkiRJkqSu0Nt0AEmSJEmSJEmSJEmSJElSd+jpaTqBup0zTUqSJEmSJEmSJEmSJEmSpK5g0aQkSZIkSZIkSZIkSZIkSeoKFk1KkiRJkiRJkiRJkiRJkqSuYNGkJEmSJEmSJEmSJEmSJEnqChZNSpIkSZIkSZIkSZIkSZKkrmDRpCRJkiRJkiRJkiRJkiRJ6goWTUqSJEmSJEmSJEmSJEmSpK5g0aQkSZIkSZIkSZIkSZIkSeoKFk1KkiRJkiRJkiRJkiRJkqSu0Nt0AHXe3Llz+exRR3BHJosssgiHf+ZoXrnKKk3HWqDS8oKZO6G0vABz5szh+KMP596/3E0PPXzi4MNZbfU1mo61QCW28exZs/jS547gbw/ez6RJi7PHfgeywkqvbDrWkEpr59Lygpk7obS8YOZOKC0vmLkTSssL5WUu7ZwTymvjAb+/5WZO/eIJnDnl/KajDKvENjbz2CstL5i5E0rLC+VlLi3vYB77xk5pecHMnVBa3sHcX4yd0vKCmTvJv72xU1peMPNYGtcD+2+xGisvvRj9/f2cfPXd3PP4dAD22XgV7ntiOj+87eGGU85fKW0sSZ1Q3EyTEXFiRFwdEbdHxL311xeP4OfeHhFnz+f5b0fExhGxdUQ8ERErDvreCRGxS0SsGBGn/rP/LU256sormDVzFud/4yL22/8ATvz8cU1HGlJpecHMnVBaXoDrrrkagNOnXMCe+36EM047udlAwyixjX9y6feYuNgkjvvyebz/w5/k7FOObzrSsEpr59Lygpk7obS8YOZOKC0vmLkTSssL5WUu7ZwTymtjgPOmnM1RRxzKzJmzmo4yIiW2sZnHXml5wcydUFpeKC9zaXkHeOwbW6XlBTN3Qml5B7i/GFul5QUzd4p/e2OrtLxg5rG04apLA/DR79zGV39xH7tvuDJLTuzlszusxVtetXTD6YZWShurO/T44ccQH51QXNFkZh6QmZsDxwHfyMzNM/M//0lvPxs4Zz6/8/7M/PA/6Xc07qbf/oaNNt4EgDesvQ633XZrw4mGVlpeMHMnlJYXYNMttuKThxwBwEMPPsDkyUs0G2gYJbbxfX+5i3XX3wiAFV+5Kvffe3fDiYZXWjuXlhfM3Aml5QUzd0JpecHMnVBaXigvc2nnnFBeGwOstPLKnHBSOWMrS2xjM4+90vKCmTuhtLxQXubS8g7w2De2SssLZu6E0vIOcH8xtkrLC2buFP/2xlZpecHMY+n6u5/gpJ/eBcBySyzC1FlzWGzCeM7/5X1ckY82nG5opbSxJHXCQrM8d0R8EXhL/fD8zPxSRLyOqghyKvAM8HD92o8AuwMPAssPepsfA5MiYp/MPH3Qe68OnJuZG0fEccCmwATgosw8oX6/9wJzgRsy82MRsQpwBrAoMB3YE1gMOA94AHg1cF0TxZjTpk1liSUmP/t4/Ljx9PX10dvbzs2htLxg5k4oLe+A3t5ejjrsIK65+kqOPv6kpuMMqcQ2XnX14Nc3Xsv6G2/Bn/54K48/+ghz5sxh/PjxTUdboNLaubS8YOZOKC0vmLkTSssLZu6E0vJCmZlLOueEMtt4q2225YH772s6xoiV2MZmHnul5QUzd0JpeaG8zKXlHeCxb2yVlhfM3Aml5R3g/mJslZYXzNwp/u2NrdLygpnH2tx++MTWr+ZfV1uaoy77Ew/9fSYP/X0m662yVNPRhlRSG0vSWCtupsn5iYgdgRWADYFNgN0i4rXAicBBmbk18Iv6tSsC+wIbADtSFTUOtjfwiYhYbQG/7j3AzvXveap+bndgH2Aj4M6IGA98ATgxM7cATgaOqV+7BrAbsD6wU0Qs++L/5S/O4otPZtq0ac8+nts/t9UHwdLygpk7obS8g336yGO58LuXcvzRhzN9+jNNx1mgEtt4q+3ewaRJi3Pofu/nF9f+lNXWeE2rCyahvHYuLS+YuRNKywtm7oTS8oKZO6G0vFBmZijnnBPKbeOSlNjGZh57peUFM3dCaXmhvMyl5S1Vae1cWl4wcyeUlrdUpbVzaXnBzJq/0tq4tLxg5k74/BV/Zvev38z+W6zGxN4ySm9Ka2NJGktl7LmH9xrg55nZn5mzqAokXwOsCfyyfs119edXA7/PzFn1a381+I0y8xHg48C5zL99dgE+D1wOLFk/tyvwUeBqYCWq5dX/Bfh0RFwNHAIsV7/2T5k5LTP7gIeAiS/6X/0ivfGN63LtNdcAcMvNv2ONNdbsdIRRKS0vmLkTSssLcPmlP+BrU84CYOLExRg3bhzjetq7Gy6xje+8/Q/8y7rr8dlTpvCWzbZmuRVWbDrSsEpr59Lygpk7obS8YOZOKC0vmLkTSssL5WUu7ZwTymvjEpXYxmYee6XlBTN3Qml5obzMpeUtVWntXFpeMHMnlJa3VKW1c2l5wcyav9LauLS8YOaxtHUsy85vWgGAmbPnMpd+5vb3N5xqZEppY0nqhIWlZPyPwH8Dp0bEBKplus8A/lB/fQWwXv3aO4A3RMSiwBzgjfO+WWZ+LyJ2oiqQ/O3A8xGxGLAT1UyTPUBGxDeplt7eKzNnRsSVVLNY3g4cnZm/rJcJ36h+m8aPlltuvQ033HAdu753Z/r7+zny6GOG/6EGlZYXzNwJpeUF2GzLrTnmiEPZd49d6evrY78DDmTRiR2vmx6xEtv4FSutzDeP+grfuWAKi09egn0/fljTkYZVWjuXlhfM3Aml5QUzd0JpecHMnVBaXigvc2nnnFBeG5eoxDY289grLS+YuRNKywvlZS4tb6lKa+fS8oKZO6G0vKUqrZ1Lywtm1vyV1sal5QUzj6Vr//w4H9/q1Zy402vpHd/DV37+F2bNabwMZERKaWNJ6oSe/kIq3ucVEbsBa2XmgRHRQ7Uc9gZUy21/MzNPiIg1qGaMnAE8Cvw9M/eIiL2ADwAPA5OAg6hmfNwtM3ep338p4Pf1924Ezs3MjSPiM8DbgOnAbzPzYxGxD/A+YCrwV2AvYGXgtPp9FwU+DDw+8D717/g1sGNm3jfUv3VGX/OFllK3mTqjr+kIozZ5Ynl18Hf+bWrTEUZl9eUmNx1BkiQtZDzvHHt9hXRaD9Y7vqfpCJKkgpV27PO4JzXH/YXUDP/2pGbscPqNTUcYtUv22bDpCKM2sRd3GhqRv8+YW9YBUR21xMRxY74vKbZosptYNCl1njevO8OiSUmS1O087xx7pd2MAm9ISZL+MaUd+zzuSc1xfyE1w789qRkWTXaGRZMaKYsmNZROFE2OG+tfIEmSJEmSJEmSJEmSJEmS1AYWTUqSJEmSJEmSJEmSJEmSpK5Q1ppakiRJkiRJkiRJkiRJkqRyuZC7GuZMk5IkSZIkSZIkSZIkSZIkqStYNClJkiRJkiRJkiRJkiRJkrqCRZOSJEmSJEmSJEmSJEmSJKkrWDQpSZIkSZIkSZIkSZIkSZK6gkWTkiRJkiRJkiRJkiRJkiSpK1g0KUmSJEmSJEmSJEmSJEmSuoJFk5IkSZIkSZIkSZIkSZIkqStYNClJkiRJkiRJkiRJkiRJkrqCRZOSJEmSJEmSJEmSJEmSJKkrWDQpSZIkSZIkSZIkSZIkSZK6Qm/TASRJkiRJkiRJkiRJkiRJ3aGHnqYjqMs506QkSZIkSZIkSZIkSZIkSeoKFk1KkiRJkiRJkiRJkiRJkqSuYNGkJEmSJEmSJEmSJEmSJEnqChZNSpIkSZIkSZIkSZIkSZKkrmDRpCRJkiRJkiRJkiRJkiRJ6goWTUqSJEmSJEmSJEmSJEmSpK5g0aQkSZIkSZIkSZIkSZIkSeoKFk1KkiRJkiRJkiRJkiRJkqSu0NPf3990Bg1jRh/+J+l5ps7oazrCqE2e2Nt0BEkaEzNmz2k6wqhNnDC+6QgLvSefmd10hFFbatKEpiNIkqSFSN+c8rqzesf3NB1BLeS2rPm5//HpTUcYtRVfuljTERZ67i8kjVRpx5ESjyH222thceLP7mw6wqgdstXqnmBoRKbOtGBNCzZ50Z4x35c406QkSZIkSZIkSZIkSZIkSeoKTv0mSZIkSZIkSZIkSZIkSeqIsZ9HUBqaM01KkiRJkiRJkiRJkiRJkqSuYNGkJEmSJEmSJEmSJEmSJEnqChZNSpIkSZIkSZIkSZIkSZKkrmDRpCRJkiRJkiRJkiRJkiRJ6goWTUqSJEmSJEmSJEmSJEmSpK5g0aQkSZIkSZIkSZIkSZIkSeoKFk1KkiRJkiRJkiRJkiRJkqSuYNGkJEmSJEmSJEmSJEmSJEnqChZNSpIkSZIkSZIkSZIkSZKkrtDbdABJkiRJkiRJkiRJkiRJUnfoaTqAup4zTUqSJEmSJEmSJEmSJEmSpK5g0aQkSZIkSZIkSZIkSZIkSeoKFk1KkiRJkiRJkiRJkiRJkqSuYNGkJEmSJEmSJEmSJEmSJEnqChZNSpIkSZIkSZIkSZIkSZKkrmDRpCRJkiRJkiRJkiRJkiRJ6goWTUqSJEmSJEmSJEmSJEmSpK5g0aQkSZIkSZIkSZIkSZIkSeoKFk1KkiRJkiRJkiRJkiRJkqSuYNGkJEmSJEmSJEmSJEmSJEnqCr1NB1DnzZ07l88edQR3ZLLIIotw+GeO5pWrrNJ0rAUqLS+UmXnOnDkcf/Th3PuXu+mhh08cfDirrb5G07EWqMQ2Li1zaXnBzJ1QWl4oM3Pf7NkcdcShPPjA/cyeNZvd99ybTTffsulYC1RiG5eYGeCJxx9jz//5L0788lmssupqTccZUoltbOaxV1peKC9zaXnBzJ1QWl4wcyeUlnew399yM6d+8QTOnHJ+01GGVVo7l5YXyswMbsdjrcTMP/nR/3LFZT8AYNasWdx1Z3LB969g8hIvaTjZ/JXYxiVmBvcXY620zKXlBTN3QmnHECivjUvrs4fy2hjKy1xK3rlz+rj+/C8y9bGHmdM3mzds924WX/pl/PJbZ9Azbhzjeiew8f/7GIu9ZOmmo6rb9DQdQN2u1TNNRsSVEbF+/fUiEfFURHxi0PevjojL6++dGxFvi4jdIuK4iFg+Ik77B3//1RGxVv315Ij4WUQc+I/9q5p31ZVXMGvmLM7/xkXst/8BnPj545qONKTS8kKZma+75moATp9yAXvu+xHOOO3kZgMNo8Q2Li1zaXnBzJ1QWl4oM/Nll17CkksuxZlf/TpfPO0MTjju6KYjDanENi4xc1/fbE449jMsOvi/veoAACAASURBVHFi01FGpMQ2NvPYKy0vlJe5tLxg5k4oLS+YuRNKyzvgvClnc9QRhzJz5qymo4xIae1cWl4oM7Pb8dgrMfM22/87x596Dsefeg6rx2vYZ79PtrrYpcQ2LjGz+4uxV1rm0vKCmTuhtGMIlNfGpfXZQ3ltDOVlLiXvXb/8KYsu/hLedsDn2PpDR/KLi07nVxefyfr/tTfb7n8cq6yzEbf++NtNx5Skjmt10STwE2CT+utNgP8DtgeIiInAKsB2mfmCq8XMfCgz9/1nhIiIlwCXAxdlZjuPdKNw029/w0YbV836hrXX4bbbbm040dBKywtlZt50i6345CFHAPDQgw8wefISzQYaRoltXFrm0vKCmTuhtLxQZuat3rote3/wI9WDfhg/vt2Tg5fYxiVmPu2LJ/Dv//Full32ZU1HGZES29jMY6+0vFBe5tLygpk7obS8YOZOKC3vgJVWXpkTTjq16RgjVlo7l5YXyszsdjz2Ssw84I7bb+Peu//Mdu94V9NRhlRiG5eY2f3F2Cstc2l5wcydVMoxBMpr49L67KG8NobyMpeSd5U3bsw6O+wCQH9/P+PGjWOT93+Kl678agDmzp3D+AmLNBlRkhrR9qP5T4BPAydSFUueDRwfEUsC6wI/A+4emA1ysIhYFbgwMzeMiHcBHwQmAP3ATsDrgYOAmcDKwOnAlsDawMmZ+ZX6rZYCzgNOy8zz6vceD5xR/9wrgB9k5qERcQYQ9c+9uf6dVwFnAosB04G9gPHAJcBjwI8y83P/cEuNwrRpU1liicnPPh4/bjx9fX309rZzcygtL5SZGaC3t5ejDjuIa66+kqOPP6npOEMqsY1Ly1xaXjBzJ5SWF8rMPGnS4gBMmzaNAz/+UfYZ6IxpqRLbuLTMl13yfZZa+qWs/5Z/5YJzz2o6zoiU1sZg5k4oLS+Ul7m0vGDmTigtL5i5E0rLO2CrbbblgfvvazrGiJXWzqXlhTIzux2PvRIzD/jW+efwnt33aTrGsEps4xIzu78Ye6VlLi0vmLmTSjmGQHltXFqfPZTXxlBe5lLyTpi4GACzZzzDz846hnXesSuTlnwpAA//+Q/k1Zew7cc6WrIiSa3Q9pkmbwLWiogeYFOqIskrgK2BzalmfxyJNYF/y8yNgT8A29bPrwS8E/gAcCjwP8B2wN6DfvbrwGxgxUHPrQzcmJnbAusD+wBk5t6ZuTnwTeB7wNeAE4BT6udPAAZmqlweeGunCyYBFl98MtOmTXv28dz+ua07cA9WWl4oM/OATx95LBd+91KOP/pwpk9/puk4C1RiG5eWubS8YOZOKC0vlJkZ4G8PPci+e+7Gdm/fgW23f3vTcYZUYhuXlvlHl3yPX//iBvbbezfuvCM55vCDeezRR5uONaTS2hjM3Aml5YXyMpeWF8zcCaXlBTN3Qml5S1VaO5eWF8rMXJoS27jEzABT//409917D2uvu17TUYZVYhuXmLk0JbZxaZlLywtm7pSSjiFQZhuX1GcPZbZxaZlLyjvt8Uf48RcPYrUNtmS19TYH4O5fX8ON3/wyW37wCCYusWSzASWpAa0umszMucDNwNuAhzJzJnAZ8K/AxsCPR/hWDwPnRcRXgTdQzTgJcGtmzgaeBP5cL/P9BDBx0M8eSDXL5W4RsVn93OPAehFxAXASsOjAiyPi3cC/A+/LzH7gX4CDI+Jq4DBgufqld89vWfFOeOMb1+Xaa64B4Jabf8caa6zZRIwRKy0vlJn58kt/wNemVDNXTZy4GOPGjWNcT3t3ESW2cWmZS8sLZu6E0vJCmZkfe+xRPvKBPfnQfh/jHTu+s+k4wyqxjUvLfOqZ53HKmedy8hnnsvqawcGfOYZlll226VhDKq2NwcydUFpeKC9zaXnBzJ1QWl4wcyeUlrdUpbVzaXmhzMylKbGNS8wMcOvNv2WdN23QdIwRKbGNS8xcmhLbuLTMpeUFM3dKSccQKK+NS+uzh/LaGMrLXEre6U8/wRWnHsq6O+7OGhu9FYC7fnEV+bMfsu3+x7HEsq9oOKEkNaOdZe7P9xPgYKrZGwGupSo+7M/MxyNigT8IUC/l/RnglYPer6f+un8Ev//WzHw6InYFvhURbwZ2Bp7MzL0jYnVgr3o2zG2BDwPb1sWYALcDJ2Tm9fUy4gOFl3NH8LvHxJZbb8MNN1zHru/dmf7+fo48+pimooxIaXmhzMybbbk1xxxxKPvusSt9fX3sd8CBLDpx4vA/2JAS27i0zKXlBTN3Qml5oczM5559Jk8//RRTzjydKWeeDsBJXz6DiS3dL5fYxiVmLk2JbWzmsVdaXigvc2l5wcydUFpeMHMnlJa3VKW1c2l5oczMpSmxjUvMDHDfvfew/AorDv/CFiixjUvMXJoS27i0zKXlBTN3SknHECivjUvrs4fy2hjKy1xK3t9f/i1mTp/KLZddyC2XXUj/3Lk8+cA9LL7My7n6zM8CsNwar2edt+/ScFJJ6qye/v6R1A02JyJWAe4BXpWZ99TPfR/4XWYeERH3AGsBpwMXUi17PfjxW4CLgFWBPqqZJK8HrgP2ycyd62LG0zNz84hYimrp7bXq2SH3yczb6997KNWy4PsB3wCeAmZSLde9FXAH8HtgOlVh5g+A7wNfoZq9crH6Zx8ELszMDUfSBjP6RlTcqS4ydUZf0xFGbfLEEmq0JWn0Zsye03SEUZs4YXzTERZ6Tz4ze/gXtcxSkyYM/yJJkqQR6ptTXndW7/ie4V+kruO2rPm5//HpTUcYtRVfuljTERZ67i8kjVRpx5ESjyH222thceLP7mw6wqgdstXqnmBoRJ6Z3fKCNTVq0oSeMd+XtL5oUhZN6oUsmpSk9rDzRfNj0aQkSep2Fo5oYeG2rPkprdgFyix4KY37C0kjVdpxpMRjiP32WlhYNKmFmUWTGkoniibHjfUvkCRJkiRJkiRJkiRJkiRJagOLJiVJkiRJkiRJkiRJkiRJUldwvVxJkiRJkiRJkiRJkiRJUkf04EruapYzTUqSJEmSJEmSJEmSJEmSpK5g0aQkSZIkSZIkSZIkSZIkSeoKFk1KkiRJkiRJkiRJkiRJkqSuYNGkJEmSJEmSJEmSJEmSJEnqChZNSpIkSZIkSZIkSZIkSZKkrmDRpCRJkiRJkiRJkiRJkiRJ6goWTUqSJEmSJEmSJEmSJEmSpK5g0aQkSZIkSZIkSZIkSZIkSeoKFk1KkiRJkiRJkiRJkiRJkqSu0Nt0AEmSJEmSJEmSJEmSJElSd+jpaTqBup0zTUqSJEmSJEmSJEmSJEmSpK5g0aQkSZIkSZIkSZIkSZIkSeoKFk1KkiRJkiRJkiRJkiRJkqSuYNGkJEmSJEmSJEmSJEmSJEnqChZNSpIkSZIkSZIkSZIkSZKkrmDRpCRJkiRJkiRJkiRJkiRJ6gq9TQeQJEmSJEmSJEmSJEmSJEkaSkSMA04D1gZmAntk5p2jfR9nmpQkSZIkSZIkSZIkSZIkSW23IzAxM98CHAic+GLexKJJSZIkSZIkSZIkSZIkSZLUdhsDlwNk5o3Am1/Mm7g8dwEm9tLTdAa1y8TJ/ulKUltM7B3fdAS10PIvmdB0BEmSpGb12p2lhYTbsubj1S9frOkIaiP3F5JGyOPI2LPfXguLQ7ZavekI0pixFkr/gJcATw16PCciejOzbzRv4kyTkiRJkiRJkiRJkiRJkiSp7Z4Glhj0eNxoCybBoklJkiRJkiRJkiRJkiRJktR+1wHbA0TEhsDvX8ybuMavJEmSJEmSJEmSJEmSJElqu+8B20TE9UAPsPuLeZOe/v7+f2oqSZIkSZIkSZIkSZIkSZKkNnJ5bkmSJEmSJEmSJEmSJEmS1BUsmpQkSZIkSZIkSZIkSZIkSV3BoklJkiRJkiRJkiRJkiRJktQVepsOIC2MImJZYNLA48y8t8E4kiRJkiRJkiRJkiRJkiScabLrRMTKEfHJiDhs4KPpTEOJiC9FxDpN5xiNiDgT+AVwIXBR/Vn/ZBHxroiw8HsMRcSJTWcYrYh4b9MZXqyIWLnpDJLay+OeRiIilmo6w8ImIt48z+PNmsqyMIuILZrOIP2j5t1fSNKCRMSEeR6/uqksC6uS+4ckSZKksRIRH286gyS1iTdeu8/FwBXAX5sOMkI/BA6OiBWBrwMXZObTDWcazhuA1TOzv+kgIxURewP7AIsCPUB/Zr622VTDejPw6Yj4CXBOZv6x6UDDiYiXAxMHHhcwA+lrI2KpzHyy6SCjsBdwQdMhRioiPgE8CSwF7B4Rl2fmxxqONV8RcTcweL82G5gAzMzM1zSTamgRsUdmnh0Rx/L87GTmwQ3FGlZEXA58F/heZj7SdJ6hRMSuC/peZn6tk1lGKyJeQbUN9wArZOYNDUcaTlHHvYg4NDOPHvT42Mw8qMlMwynxfCgi/gz8R2beXD/1XWDLBiPNV0S8PTN/GBF7zfu9zDyziUzDiYhNgNcC+0fEF+qnxwMfBF7fWLAhRMSDVMe7nnm+1Z+ZKzQQaTQ+A/y06RAjVRfH7cbzVxd4X2OBRqAejLgXz78eaXvmLwFnZ+bvms4yQh+PiFWp+i6+XsJ1VERsAGyQmadExAXAiZn526ZzzU9EvHVB38vMH3cyy2hExJ+ojh8DZlP1yX2yxW1d1P4iIpYEtuH5++RWX4sA34yI/8zM/voc9ABgzaZDDSUiXg98BViaaj93a2b+sNlUQyqqfwggIsZTnV+sAlxF1caPNhpqGBFxOPAhqn3bwDVUa88763sMSwJ9wKeAU9t+nlHg+VAxx5GI+Cbz9BcOyMz3dDjOqNT9cQfx/P6L1ZpNtWCFbsdLANvx/O241ecXEfEGYHFgLnAMcExmXtlsqheKiBt44d/ewHa8UQORRqwelHEgz//ba905XER8OzPfNaifCFp+nC75PgNARLyU5+8vHmgwznC2j4iTMnNO00EkqQ0smuw+f8/MQ5sOMVKZeTlweUS8DDgZ+HxEfBs4KjP/3Gy6BXoAWAJoe3HnYPsB2wNPNB1kpDLzwIg4mOrC9eiIWB44i6qwdnaz6V4oIk6jauMHqC9OgFZfAFIVCjwWEY9Q5W3tBdUgi0bETUBSdQ60vZPrncCmwOWZ+dqIuKrpQENYi2rb/TJwRmb+MiLeCOzbbKwhDQwQuH2e59te1P5+4B3AlIhYFPhhZp7ScKYFGSiY3RB4BrgeWI+qGLG1nRkRcQ7wFqqOxEnAn6n+Da1VynEvIt4P7AG8JiK2r58eT7VNtLpokgLPh6j+7r4cEZ/OzJ/ywmK5tlim/vyKRlOMzhPA8lQd4QO55wKfbCzRMDKzpPadV39EfI/nn8O1doADVdHIl4CHmg4yCudSZS5lACUUNogyM3eOiKWB9wAXR8TDwFmZeXWzyYb0JWDn+utPU20nmzaWZmj/vYDn+4HWFk1SFT5dDPyc6vxzD+CrwCnAxg3mGsq5lLW/+B7wF57bJ7f9eg+qAe1fq2cJfwLYoOE8I3EysDvVNcg5wGVU++m2Kq1/COAMqn7DbYBfUV1Tbz/kTzTv7cArM3N600FG6BvAEVQDob4NnAS0fcbzos6HaudSxnHk9KYD/AM+BexA+9t4QInb8f9S7ZMH2riE84vTqQrZPwMcAnwOaF3RJM9df5ToYKp7Oq3+28vMd9Vfrp+Zz2aNiLUaijQSRd5nAIiIU4CdgAd57v7v+o2GGtqywAODJkppfcGyJI0liya7z60RsTNwE/VJfmbe0WykBYuI11CNsN0BuBrYhGq7/RbwpsaCzceg0VEvB/4UEXfV3yrhZOMW4K8ljSqJiB7grcCuVCOwL6A60bsEeFuD0RZkfWC1zJzbdJCRysxVms7wInyq6QCjNIeqIONv9eNJQ7y2UZk5E6plwzLzl/VzN0VENJtswTLz/+ov762LiYiIScAXaPGFdmbeHxG/oprBY0fg3VQ3VVtnYObAepbUfxt4PiLafOMaYG3gdVQ3pQ6mulnSagUd975O1SF7MPDZ+rm5wMONJRq54s6HgEeoOmu/HxHLUs2Y0jqZeV79+TPzzrzdVpl5K9W1U39mHtl0ntGIiCnzPtfGmV3m8YLMLff0wHZdkIcy8+ymQ4xGoYMolwNeSXWM/gPwrnr2812ajbVAswfaMjPviog2X6/u3XSAF2nNzLyi/vrqeqDDlfXsbG1V2v6iJzN3bzrESETEIvWXU4DJwFZUhbRFyMw763OjRyLi703nGUZp/UMAr87MPSJi48y8JCIObDrQCDxMNctkKeYC1wCHZOaFEbFn04GGU+j5UBHHkcz8GTw7o+CngBWoivtuaTLXCN2VmXc2HWKkCt2Ox7X4HH5BZgC3AYtk5o0R0db+raHO69s8eBLg7sycd4KG1qlnCF8ROL5e7awHGAccB6zTZLYFKfg+A8C/AqsW1Ke8Q9MBJKlNLJrsPuvw/BOiflq4hN8gZ9Ufn8nMZwaenN+NwBYYGB21CDBr0PMvbSDLaF0F3FUv8TgwRXubtwuAP1HN1HBKZl438GREvK65SEO6k6pA4JnhXtgWdVueTjlLLwH8lrI6ua6uP3aJiJOASxtNMzJPRsRRwC+pZkt9sOE8I3FURHyUara7c4DzG84zpIh4nGqmlOOAbTLzqYYjjcTLI2KpzHwyIpbhuVnl2uqxeim8xTPz0RbX/g5WxHGvLrC+JyL2pzp+zKZalutrVNt1m5V4PtSTmU9ExL9RFf++oelAQ4mIL1PNljN45HXbBxdtHhGfLajjE+Ci+nMPsC7VeVErRcSbM/PXlHE+MXh54Kfq2X9/w3ODEdvekX9PXXwxeABlqzOXNIgSICJ+QXW9dzZw2KBBR/835A826y8RcQxwA9VAv/sbzjOUZAHL+AGtXY4SmBUR+1DNlLIRMDMi3kS7+2SL2F8MKkC8KyLeQtUfMJB31gJ/sFkD23HPfJ5r83YM8Hi9lPji9YD8J5sONIybqGbQfS1wB3BUs3FGpLceCDVQxNXaQvZByxovB9wUEbfy3N9fm2f0nEA189o1EbEFVT9+q5V2PlQr4jgyyBSq2XM3o5q1+Jz66zZ7JiIuA37Hc23c2mKzebbjn1LGdnxLRGzA89u4recXA/qp+t9+FBH/RXuL2rPpAP+AqRHxQ56/fzus2UjztTTVPevlqFZCgOq84rTGEo1cafcZoLr/O4FqkpQS9AHHU00CdTHVfdS299tL0phpcwedxkBmPm+5iUEdjK2UmRtHxA7AByLi1oFZwzLzyw1Hm5+ZwEuoLkr+h+dG7pxBu6fhhmpk13/R/g7PwdatP69aF71MA2jx7AKvpLohNTACtIQZSE+hrKWXoLBOrsw8BDgkIl4KfKqAjheoZrnbk2oJpj9QLWvUdjsCP6DqDP/PzPxjw3mGsz3VzIHvp5qd6IrMPKPhTMP5LPC7uuBzSeDDDecZzm8i4uNUy1BcSItneR2ktOPet6kK799Jta84E9i20UTDK/F86CCAzJwaEe8APtJwnuFsQDWDTmtvAM/HyyhsyZpBMy1DNZtHm29QbgX8mhcuu9vW5XYHcj4FrFF/QHvzDrYoEPUHlJG5pEGUALtk5p/mfTIz23z82x3Yh+r88w/A0c3GWbDMfNXgx/UNtMczs+1LJb6HannEfwd+T9VftD7Q5hmAS9lfDC5A3HLQ160tQJx3Oy7M+6lmf3oUeHP9uM2mAD+jmqF/M6rlgt/RZKAROBS4DngFcCPw0WbjDKnUZY13p1r+/Byq/fL/azbOiJR2PgTlHEcGLJOZUyJil8y8PiLGNR1oBH7UdIBROpNqYFFJ2/FmPH82ttaeXwzybqrlmH9UF4a3chnsgVUbIqKXqpj2lVQDmW9tMNZI/aTpACORmT8Hfh4R6w+sGgYQEa29VzZIafcZoJrV856IGOgP6M/MTZsMNIwzgROpBhhdA5xHtSy6JHUliya7TD0i+GNUIx56qEYardloqCHUs9EsQzXrwR4RsXVmfqLhWAuyIbAfVWfAmfVzc4E2zyox4D7gV4XdwN6GqjOxF/hWvTxQa2/w8MIbwUUobOklKKyTKyI2pRrdNx64OCL+kpnnNBxrON/OzLcO/7LmRcSxPDcbze1UhYj/ExGtHn1dL5/yV+ABqpusu1EV4LdWZn4nIv6Xqrjo4bbPyJaZB9czd0wHtqMaNd52pR33JlEVK++XmbtGxNZNBxqBYs6H6qVezwb+vS6WLEVxM29TDRIoyqDZEKG66b5cU1mGk5nH1593j4g1gdWpRrg/0GiwBRgoVI+IV87zrdkRMSEz2zqTx0Abv556xq3MbP2xrx5EuTXVrOw3UuWe0dJBlFANbDiZaj/X6hmLB83yuilVseQf6m9tFhGzgJvaOtt5addQmflYRBxHtV0ALJ6ZlzWZaTil7C8GChAjogdYKTP/GhHrZeavGo62QBFxAy+cMRWAtg/KoLrWuzkzD6y36ZcAjzecaSjLZOap9de/i4h3NZpmBOqlgqNevvbRlheFX0u1H76Qqkinp358Ke1e1epOqtm3NqAabL0ScFejiYZR4PnQCwZ3RsQrmsoyUhGxVv15JaoZuNruAqqBnwOz6X6l2TjDum+gUG6wlm/HawNExMupVqxpbV9nRLy9XiHsP+rHe9XfWoPn7lW20elU1/7bAL+impBm+0YTDW8HqgLgS9t8nI6Ijan2Dx+LiC/UT48DPgS8vrFgI1DafYZaCYMwBlssM6+KiEMzMyNiRtOBJKlJFk12nw8Cm1PddL+Ydo9YBVg7Mzeuvz45Iq5vNM0QMvP7wPcjYvvMLG2k36LAzQUtpQJV8e+GwOVUs2H8mhbPikFVqPyfPFewvAJVx0ablbb0ElBcJ9fRVDcqvwMcQzWjQGtv+NWeqAt07qBeKioz72g20gLdPujrpJplovUi4iaq2Tu+B7w3M1u7TOKCbvrVhamtvekXEZ/MzM/VX98D/C/PzeTYVqUd9xahGkzym4h4LbB4w3lGoqTzob/Wn2+f5/nWdtjWipl5e1Bh6j68sF1bW3hfGzxYZwbtns0MgIj4ELAT8FKqmaDWoOrMb6sfUt1kv51qEOIzVEtqfjIzv95osgWIiA9TDcb4BfDxiPhWZp7QcKwh1ctGrwS8hmplh4No92C0k6j6WP463AtbYEGzvEJ1zfoa2rtUYlHXUBFxGtUgnQd5bhbEVh77BhS4v/gKVSHUCVRFRe/NzLb2d7ZyxqcR+hpwQP31j6j+7rZqLs6wFouI5TPzoYhYjqqgr5Ui4kuZ+aF5r69bfl39Pqpz4uWp+lt6qJakvLbJUCPwHaoijPvqx/1UMyy1VoHnQ0TEkcAHqPoFJlH1Ib6u0VBD2w/4KlUbfxvYt9k4I3IGVV/9T6hmRDybanWgtpoQEW/g+f3JrV5xKSI2p5q1+Clg6YjYMzPbOsvgwPLF8xYot72P6NWZuUdEbJKZl0TEgU0HGoFDqGbbPioifgScnZl3N5xpfp6k2h4W5bntYi7wycYSjVBEbAPsTz3orD4favOADKj6hI6lWu76O8DNzcYZ1oyI2BYYHxEbUvUdSlLXsmiy+zyQmQ9GxBKZeXVEHN50oGHcGxErZeZ9dQdXa28+1CMyjqaaxWyXwd9r8Q33Acc2HeBFmJOZM+uZtvojYlrTgYbxDaoCqI2pRs9NbjbOiJS29BKU18k1NzMfr7fjGYXM5vlyqovWAf20dBaBeZb6WI/nFy232VaZ+eyMHRHxisx8sMlAQyj1pt/rI2Ifqn3xrlSd+W1X2nHvAGBHqiVVdqHaP7ddMedDg5ZfXi8zny0si4ivUd3QbqtW39ybx4IKU1uvxJldqI4nmwJXZubJEdHaWcJqdwNbZuajEbE01U3KPYHLgFYWTVIVQG2SmX0RMQG4nqrAqM02zsxNI+KnmXleRLT9eH1vZl7RdIiRGDzL6+DnB847W37DsrRrqPWpbgi3fibrQUrbX6ybmfsAZOZ+EdHaAqjM/As8O8jzJJ6bIWz/oX6uLTLzxvrzNW1f2YNqucHrI+JpYAlgr2Fe36Sj6s/zXl8v0ukgI5WZZwFnRcT7MrPNy+vOa/kWF6IuSGnnQwDvoCr0PAn4AtUM0a2Vmb8H3tJ0jlFaY9DSr99v82QjtaAasNzPc4NI2r7U9dFUf38PRMSKwHdp79LMN9QrN3yz6SCj1BsRywL99YpArT9fzszbqGZvXAY4FciIuAo4LActg920zLwVuDUizgKmAqsCf87MtvcnQ1mDEQecBZxCdT/1BqrBRW0+ruxFdX23LPBxqgHjktS1LJrsPk9FxI5UJ6F7Ux0QWyciHqS6cJoI7BQR91JdaD/aaLChXVJ/Pr3RFKMwaNr+mM+32z4j27UR8Q1gpYg4nWr6/jabmpnHRsQamfm+iPh504EWpL7AHjC443NZ2r30UomdXHdGtYT0MvWNyb80HWg4mbnF4McR0dpO/EG+R1UwuSLV7BIP0O5OpI/WBX2LUo3I/xPVzbTWGXTTb0XgeKqi2oupllZt8/a8G9VSRi+jKjqb2WycEbk2Ir5Jy497A4NdqM7ZzqbaJn7cbKoRu4nq5urAzeujhn55cyLig1Qzx780Iv6jfrqH55ZXbZUSZ23MzP+rO+8vzsxnIuI9VDeuL2g42rAKnNkFqmWi+nlu+2j7fnm5zHwUIDOfiIjl6iKuNt/k6cnMPoDMnB0RrV1KfJDeiJhI1X8xnmoGqzZ7uD5G38RzMxa3eTm8Be4vMvO4RoMNrbRrqDup+raeaTrIKBS3v4iIZeql0JeijP7us6hmyLyGakWgts/aCPBkVMt93kBVDNzqguV6NrDVImLZgWN2W2Xm3+ov3z1oRYTXUw2GavuKCCtHxGGDn8jMI5sKMwK3R8QKmflA00FGobTzIYAH60GfS2TmnW3tO4yIu3n+9elsqv7DmZn5mmZSjdjEiJhUX6suRotn060d39YZ+YcwZ2Bfkf+fvfOOkqwqt/ivRMXl3wAAIABJREFUB4YMKgwZQUmbICKIBCVIkiAgPkGCoCQJAhIEFUQkjICoIAhKkCgZJEclKUiWLLKRPMAAEhwkp3l/fOdO3y66uqp5wjn3Tf3W6lW3ipq1NrWqzj3h+/a2n1TZ8bVHt3m9WLODxI8I1/hZgZtoQMN1ckHcDPg0YZTyPWLueTGwaD5lbVmG2D+cFDgrNZ6VnFwEDWpGrDGl7T9K+qHt+yQV7aQLrGh7QrOOpJ2BX2XU06NHjx5ZacImUo//LlsB8xIxDt+j0Mgz201wQxmA7cpu+3FgbZJ1eKLUAsSm2vZje09JqxMHUvfbvqjTv8nMeEmzANNKmpqynSarRXbV+UnteZGL7AZvcm1LjMvXEx1/384rpzOp4H5X+l0b3yJiKUtmlO1lJP0O2JFyu4Ir1gE+TkM68hPHAL8kCs7+ApxEREkXRUvc2UhiM+uawmPPgAH3vduBf6SmhxLZNf0dzcD7SLH3kBrHE3O2U4l4qxOJ32Nx2D4SOFLSnrYPyK2nC6ru8H/SjEM+JFWd1m+msWNO4BlgFcI9tWQa5eySOI24f8yVIq7Oz6ynE7enQvYbiUOIOyVtQHxHSuV6SecA1wHLEYdTpXMo8DeiyeHm9Lxkqli2WdJj8etqmjleNG0NNSfwmKQH0/Pxpc87ad54sR9wm6QXgY8A22fW0w1T2L4wXZ8vadesarrjW8Sh+1eJRp0t8soZmnq0oxS94g2IdmxiIsLT6bGPKPAs3YF0OSLZ6l/p+XjbpaeRHMLA+dAheeV0xROStgBeSY0OH80tqA0LEN/dI4Gjbd8iaTHKTy4COAy4S9K9ROPnPnnldOTblOvI346XJO1IrFOXp2AziVaTgwbxpm1JmhF4znYT1k9bAce2FvVJKrXxeldij/5ywj31tvRYMo1rRiT2DlcGRkhagvLjrn+T9G6REhHWoVc02aNHj4mYXtHkxMcIohBjPmKRPXNeOUMjaRXiezqCsDr/se3T8qrqyAWEVf+LuYV0ooqutb1v/fXUVVI0kjayfTpwuaRZJV1me43cuoZgX2Jj+ffAw+mxSKpFtqS168Wokr6eT1VHGrXJJWn52tP76HcGW5rYiCmZ7QknjL0IR8Hixwv6XV2mtv2apNI3YBrRkd/ClLavlrSXY7er1I2BqoNyMqDe8Tl9Bi3DIsXUTEcU5Ewv6Zu2i4thtl0d+F5q++dZxQyfGWz/Ol3fKWm9rGq6Y02g+KLJWpz4hra/lFVM93ydmF9MA/wD+HiKKS39Pg0NvI/YPkLSVcCn4qnvzq1pKGx/R9I6wILAKbYvUVRkFNvIZXs3SV8mNB9v+9Lcmjph+2xJVxKNn480wClsX0mz0t9cVHoRBjRwvCCK72+nOWuojXILGC5NGy9sXyzpMiId49mGHLhPKmkR2/dIWoQGFFnbfk7SAfQ3iU+VU08XNDHacTMalohge4C7WfotFovt+XJreB88DyxLQ+ZDie8T+xdnE9/rjbOqaUP1G5M0TxWra/sOVZXWBWP71PR7m5v4XjyfW1MHJpd0B2BSBLPtIr8XNTYh9r9/SuHNAupP7qvTR/mF4VtIOoJoRjxX0p9TAVex2N6g/lzSgbb3sH1OLk0deCet98bbHi+pCfHcTWxG3IZoapiVGDeKPJescRtwA3ChpPVzi+nRo0eP3PSKJic+/kgc+lUFfeOBs/LJ6chPiUX1kcAXCK2lF02Osb1PbhH/Rzam/K6STSX9h4ivPQD4SWY9Q2L7L/Qf5Fw41HtzI2kt4PPAxpKqqOsRwFcodLxo4CZX5RYwD1G8dSuwGOGU8sVMmrrlKdtj08HqtZKK/u0lzpX0Y6ID+ybicy6ZpnTk13ld0mrAJJKWptxuyjeIjfuTgU2JDcQRhCvikhl1dcMFRLR8dehX+obRGpIOsd0IV8HElJJmsf20pJkpP94K4AVJOzHw4KHkOPQXU6HZA/TrfSCvpLa8mqJJ/y3JVUwp4bBcOo25j0h6gpgjXw5cYfvszJK6QtI30+VYIiK4yEL2OpJmJw4fHgS+L+kp23dmljUoycXzPfe55Axd7OGqpOMI59GpgSmJZrninLdbaMx4UeNcojhuDOkwmAKLJiVtZft3hDNm6/d5zwySuqYp44WkI2zv0OImX40Vpbt5fhc4XtJswJOU75iKpN8AaxD3vuq3V/Ln3JhoxyYnIkiqJ4/MCsyVS8tQpAbP0YPNMUqeWyT2tb08sXfYFC6yvWy6/vWQ7yyDfyeXuFuIcW1sZj1tafddLn2eDPwgt4DhYnucpNHE3sW6ufUMRROT+wBsbw0gaTngYOKsZKasoobPyrkFdOD6NF7Mkdwbi7+XNLEZ0fbjyXSmjzhjeDyzpE6Mt32MpHFE3UgT9sB79OjR4wOjVzQ58THO9ma5RQyDVwlHpbfTAXbpBQIAF0k6iH7XA0o/QBuEvs5vyc7XCCeXKYBlbf+rw/uzIumnwJakAgGAgrv87iIOoV4jijAgdJ+RTVH3NGKTy/ZGAJIuAb6SnKsmAS7Jq6wrxklal4ic34b4rhSN7SMl9aVuykuIw7+S2YaISaw68pvgTrM18Avi+7Ab5caILQ3sBIiIFIcY365o+y/KYYTt0iOB68wIPCXpEWIjvwlRlD8GbkgbRtPRgMNrwnXkM+kP4rMuuWhyJiImsaLo2HZJI0kRg7XrJmwkNuk+Mg9RZPZF4AxJkwPXApenpqNSWTA99hG/vxeIgvySOY2I7dseOIdw4Co1xu2o3ALeJ4sCCxPNGHsSn3PpbEOkkTRhvKiYuQFzCuhvdLk/q4r3R1PGiyoGccMh31UQqXnkKtt3AJ/LrWeYLAnMU7oDVI0mRTvWv8NVQerkRNNf6dSdJl8HvpdLSAcqN/AmzjHGSzqPgY1yRRff07zmvm8QTQ5rAX+n7Kjrpn6XbycKJ2cDLgaKThYAkHQGofXzxF7A/xBJYsXR1MLwlHi3MrGH+FcKN0ZpIrb3lLQ68Rv8h+2Lc2vqRBObESX9gjhzmoPQOgbYPKuooXkAwPaZaR+8CXsXPXr06PGB0SuanPi4QtK2DCzoK/kw6iXCdeQYSdsDz2bW0w0bEm6e1UFasYWeDYneGkDLwu81YtP2sAZ0U34ZmKsh0TpjgBMljawi3AEkfRcozl2ihfom132UvckF0YVfMSnN6KTciogE2oPYDN8hr5zOSFqFiD8bQXS4/5iyXYvnBNYj4s76gPWB/bIqaoOkSZP72rPANzu9Pze2zwfOl7Rm6TGDg3C3pKWIcbg69Htz6H+SlbVyC3gfvGh7bkmjUvzgCrkFdcE+hJvL47YfzSulK35Z36BNXdilMhdxyFc1E1WOmMXO7Ws05j6S5sbXAtdKGkUUT+5EFC3PmE/Z0Njeo7qW1EccppXOu4Qb349snyGp5MLwIl2quuD51KQzdbqP5NbTluRqtovtm4DH0su/lnQhsE4+ZV1xv6TZbD+VW8hQ2K6ack4miuOmGOLtpdGI8cL2M+nybeBnxHr6bKIQ47F2/y4zXwV+kZyWLyeaBIovHEk8SHyPX80tpEtaox2LxfZjAOm3Nr/t3SX9Efh9+isW2ytKmoFohHm41Oho23ely8eBtRk4Jv/5w1c0LI5ved6E9UgjmvskLWH7NiLh7J70B7ACBepNLJX2hgaj5O/y8cBlxGf7NHBcui6Z2WyfImnLNNaV7F7c1GLa1YCPAX8gkieKnRNJmnuQl/uIJofikLR1y0vjgNkkbV1wE0lFE5sRl7G9m6Rr0nhxVW5BQ2F729r15S3O4T169Ogx0dErmpz4WI6YxFULkiJjjGp8nehivk/SwsDvcgvqgjdsl+qw1YqJ70ATnCUrWhd+v8iiYvjcQWzIFV80KWkj4qBsRUmVm8QkwKeAw7MJ644/EHFte5XuPpo4Dvi7pHuJheDPMuvphhHEZvgixGFUEw61fwpsDBxJbISeRdlFk6cTB2hP5xbSBScTn211P4F+Z4zBNpNK4QVJR1OL+bC9WmZNnViBONypKP0znpFwrJqq9toWeaQMTYoBWgjYRdIh6bURRFH4p3Jqa4ekaYixYgbgUWBeSf8CNrL9Uk5tgyFpLWL83UhS5Q42AvgKMSYXh+1P5tbwf6Ax9xFJnwXWTH8Qzr+7E67hxdLSfDYr0ITvy0gi8uwvaY5fcgNd1YC4FNEodwNRdDaSsh09/yZpN8Jp+QwG3gNLYxRwkqSDbR9Xe33aXIKGwbLA4+m+B+FmXWqKA8QadUbgifS89H04aNZ4AeEg/0uiOe4vwEkU6kZje3MASZ8gGgV2ljQv8E/bW2aU1g1zAo9JqpIbinaST9GOM9GsguXtiOZwiObrv1B40aSk9YHRhIHApyTtY/uUzLKG4gJi3/DF3EKGwfnAqpQ9rwBA0ua2T6jGugawMnAb73XaLrLIM9Euhrn0YtoZbB8vaRPbN6Q9l9KZTNL/APelBr+S58mLSlo0t4jhYnsNSVMQjuaHSVqg4Kjxkxj8LLW4PbhE6+fYpHPgxjQj1pgk/QYfS0k10+QWNBSS9iPmnZMR84sHiPPJHj169Jgo6RVNTnxMY3uV3CKGwYzAvpIWIm7auxAHwyXzmKQ9CLvzygmqyEV2Ew+Dbf8ZQNLawBK2fyLpciIuqmTuBcZKeppUUGS71GKXy4lY6xnoj9l5F3gom6Lu2ZIo+Dw+xTtebLvYQs8UG302UYT4z1I78ls4j3DtqAoxSt+Ug3DCeAZ42/bTkkrX/KrtfXOL6JLKvWOX5OLYFH5LHASvRzgJlH4QDLCV7VurJ5K+mFFLN/wWOIIGFG0RB2azEI1F1abiu8D3synqzEHA2bYnFA9J2gr4ORG1Whp3EfOK14gia4jP+IxsirpE0meArakduNsusgC4RpPuI7cQhbMbNcQttaLefPYa8dsrnc2JA/fjiILlb+WV057KyVPS5ba/XL2eXLeKJUWfTUPEk64B3JxZ0lCMIVIyzknFyzukyN3S58nYbpoLxiwlF5a1oTHjRWJK21enaEpLej23oC6YApgemI5wyiz1wL1Oa1FR0Ug6kig8fIr+xr7Sf4vvpCQHbL/VgL0LgF2Bz9p+WdK0wNVAyUWTY2zvk1vEMDmPOA+p3HVL/l5sCpyQW0S32P5ZeqwK2hcG3rT9z6zChub03ALeL5IWSI9zEPe+0jmYmC/vCnwX2D+vnCFZsHa9EWEWUN37iiUVpa4JLE4UMBdrKGF7ufpzSR+xPS6Xnk7U94QkfZkoiLPtC/Kp6prWZsQpcwvqgtMIR91vEUY/rS7RpbEOESV+KHAI8Ju8cnr06NEjL72iyYmPeyVtSLjeVQV9Dwz9T7JyLHHo/heiA/s4ogOwZEYC86c/KLszEQBJV7e89BZxiDK64MPLfYkONIANiHiHK9q/PTsbEA40/84tpBO2X6Q/JnFloqDvJuCFnLq6wfaTkm4lYh3WJT73YosmWwsxUsx86YUYfQ3qGK/4D1EMfIyk7Yko6eKoxSA8kxxf68X3pd6rvy7pKWDH5OIxgcKjPp6zfbqkL9neR1KxEUaDuSASDn3FuiAmXrJ9Um4R3WD7XmKO/LDtCU4uhUdHL2p7h/oLtn8nqUiHIttjCEez6vMdASwD3JdPVdecSBQAj8msoyMNvY8sSxSXnSrpZeJ+fZnt+/PKGpomNZ/VYgc/SUSrrkCsSeYFHs6prQtmkvRR2/9O0Z8z5BY0FJKWYKDL8lcp1GUZwPazklYi1kvXSFovt6ZuaGAxeyPixAEkfan2tEnjxeuSViMcXpYmCpeLRNKvic/1UWIfa3fbjwz5j8phsOLZ/T50Fd2zFDB3KghvChdIuo5oKlkcuDCznm541/bLALb/04Ci5YskHURtHVJvRCuUvsLvc3WmkjQfg7iZlbgekVQ1CMxDzNm+D/xL0rEtTtwlcTSDO8aNB1b68OV0zU5EQe2CRNTud/LKaY+kSVMB+8XpD8JRt1iqpjMASUvb3jOnnmGwLHCi7a1yC+kWSV8g0vBGSjoLeNj2iXlVtUfSgcB8wPXAtyQtZ3u3zLI6cRLR9PIasV9UdBIJQDJuORxA0kG2x2aW1Imxtt+QNK3tB1vSVHr06NFjoqNXNDnxsWj6qyh9MTWF7WqD6HxJu2ZV0wUN60yseAz4K3AdcYC9NnAjZRepvlV1ctkeJ+md3II68Bjwiu3i47krJB1AdBstSMSK70HhzgKSXiA+64OAVUvu9kucSHMKMaqF08OSlmFgIcab2YR1x/rAPLbvk/QpoiC/RI6uXW9duy75Xv0NYDUGOvQ1gXfTfXoqRcbH9LkFDUGjXBBrh+3jJO0J/I3CnbdboqOr7vyio6OJBpfBKN2t4RAium8u4iD4Gcp3r3ra9u9yi+iSxt1HbN9IrDv2TsX3qxNNDnMU7MqOpHWA7YmGuT5glO1F8qpqSxNjByt+CtyZ5vgfAXbMrKcTTXJZ7gNIB8LfkbQ5sSfQhMi2E2nIGirRpDjxduv90seLrQlHl1HAbkTUXKmsTKR4XEg0CTyZWc9wqFzu+oh5XOnRqg8SxdWv5hbSLbZHS7oYEHCy7btya+qChyX9kjA8WJ7yU2o2JNYj1bqvWBe2hu7DiViTNKWgb29gyeTs+kPCZXkMYSZQZNGk7RU7v6ssJE0HPGR7mdxauuRkYGMGpguQrotdo9YodlwbhEOBQyQdTUoaLNjEpeJAwuDnnHR9PbE+KZXlbX8BQNJhhDlK6Rxne9l0fVFWJV2SahfGEfsW20q6wPbumWUNxROStgBeSYW1H80tqEePHj1y0iuanMioL6okfTy5vpTMpJIWsX2PpEUoeMLf0M7EijlrznGW9A3bx0n6ZlZVQ3OLpNOIQ9YlCffUkvk48JCkyp1hfAMiupa1vbyka2yfJKnkg4eKNYnD9i2B9SRdafvoDv8mJ00qxKhvFK1Uuy52w0jSVunz3QcYH7VxEyiu43awjc8G3Ku3s725pD7bJbuMtLIrEUtyOP3xGUVSc0Ec35DPuDpsH0d0Ms+Xnpd82N7E6OgXau5xwASHs9JdoT9ne+c0t1hR0lW5BXXBo+kAre7UX+R3uaEHaH3AZ4DliIP2+YC7gdLnR6OBbYBtgWuAVfLKaU8VOwi8YPt7WcUME9t/kHQBMCPwrO3SG+Ua47IM/KT+xPYJku4lClVLp0lrKIAvNsFlEvqbgGGCe/G8xJhctH7bTxCFUMVjeyFJcxN7F8ckF91rgUtt/yWruA607q1IuiyXli6ZE3hM0oPpebH7cNXeRTq0rva+F5W0QQPcwjYn5kSrEsWIP8grpyNv2G7C/ia8dx+uoth9OOBO2yUWR7bjLdtPp3H5LdsPApRsziDpHNvrSRpLy1lZiU0ZknYAvge8LWkH2yUnhQFge+N0+VXbd2YV8/+fY2he0uA7tp9P+7SvSfpPbkEdGClpRHLeLj6yPfGKpEOJ++C7UHyiFUTi3fKEk7yA0vc7tyHOrM8m0jKKNsvp0aNHjw+aXtHkRIak3YlonY8Cm0u63HbJ7o07AsdLmpXYpP12Zj1D0bjOxBqTpSijG4HPExPpuemPFSsO2ztKWpeYgJ5lu8iOI0nr2z6b2MBvwoKkzqSSpiCKzSYBit0wqrB9k6QxxHixMTHhL7loskmFGJ+ECcUNc9geI+lztm/NLG0oqmLDoiM+W2nYvXppST8H1k/jxQRKPtyx/XdJI4l7yGa278mtqQu+KOmnpReM1By3RwGL2f5T2iQ/Ja+y9rRERy9IxKH/s/DN8d2ACyVdS7i5fJIo2lo7p6gumETSZ4n732TAtLkFdcHkxFhRVd6XXAAMQGp8+iEDo2tLPVh9ipgHXQns25DxGCLK6EZJ29o+UdJmuQV1wUJKUde5hXRC0hG2d5B0I7U1lCRKLHhpostyVZwlaVLgc/S7pp6QU1eXNGYNlThB0uSES8q5TYhiTnO3rxJu7CcSBe075NQ0GLWCkcmJ/asxwOzAv2x/IqO0IbH9MHCEpJOJ+dvOxOc7TVZhHUiFtBWzEs7hJdOkw99G7l3ABMfiI6vnkg6m0FSExGOS9mCga2OR95BqHw4g7ctWTSRNipwvnfFpLvRl4AoASdNQ9pnIeumxKWkvGxNr6emA35M+54awf2puOAE4zfYruQW1Q9Lp9BdZL5zMRoABRaAl0po0uEtWNd3xiKT9gekl7Ub57vdnAn+VdBOwFGU3iFfckB5nzqpieLxN/316vKQi7yOSvm/74DSXeCy9dgZwKvClIf9xjx49evw/plc0OfHxNaLb4fLU3XxNbkEdWMX253KL6JLGdSbW2Az4OfAr4B7CKXNpwomrKCStZftiSVXk4IvArJK2LrTbaB9J9xGFs5vSjLizikOJA78ZgZvT86KRdAfwHHAe8I0GRF01rhCD6P58kIg/2yQ50+6cWVM7nk+PY1v/g6QvAncUGuHepHv1mkTk4Fr0O/QVT9rcWgm4BdhJ0nm2f55ZVidmBJ6S9AgxVhTrlJI4HTgsXb9AFE2ulU9OV2xPRM7fBOwu6Szbv8isaVBsPyppSeJwZ27iPv2jkjfxEycDvyHmmgdTdmMDMNB1CyA1c5XOD4B1KH/zHmCuKl5Q0ihJc1b/wfbj+WR15A1JyxPNZqsRkbClsyDwnKTn6L+PFOdGk9g/PbY6x5UaBdtEl+WK84iCydmBSYhC5tOzKupMo9ZQtldTxFKuAZwqaUrbi+XW1YENifXIVbYPk1Rko1xVMCLpFGCP1Ng3GwXvXUhaj3BX/gLhnHMVsC8R7Vg69Xnb64RzWMm8Q3wXFiJFfuaV0x7bV0iaiTirmQF4gnAfLb7RYRC+mFtAB0YC86c/KPweAiDpq8AhxB74dJK2s/2nzLLasX51ke59nyBimUtdp55EOKSOBFaS9Cli7+LwrKqGoHKaTNdr2C7d9ff1tN57Tv2R843A9tqSZiHOdP4o6R+2t8qtqw1Htbku3cSjNWmwCWyT/m4mCuVKNvrB9i8lXQEsQMRe35tbUxeo8GLfwbiOmM9vKukQ4PLMetqxpqRxlYO8pOWIgvYSz9Z79OjR40OjVzQ58fEOMAvwTHo+ZUYt3bCmpENLd1VKNK4zscL2Q8D/tLz88GDvLYAZ0mMTDqwhCswOJw50jqa/aHI8A6NVisP22ZKuJCK5HrH9XG5NXbCy7QnRpJJmtf2egrlSaGghxuK2twWwvZOkkiPEVgJuY3CHiZFE8cBnP1RF3dGYe3Vyynkkud1NRzMc+iAOrZe0/W5ybLiRaB4omdILDluZ2vbFALZPk1T0JmJiY2BZ228nJ9IbiALxIrH9uiK2djNiPBsr6d6S79e2f0MUTQLsnD7nopG0H7AdMBkxr38AWDirqM48XDVwlU6tYPJowm3rGfojo0ouDN+OmN+PJgr8RueV0xVb2L46t4gueV3SrkRxwEnpfr0IsZ4q7ntRc1meI8UEk543IS54lO1lJP2OSPootQhjArY3T0UNCwEPlD7vTAkZqxDOLo/TDIelEaTi6vT8jYxaumHu5ByO7afqBfgFsjrxO9u/5DnbYNheESA5p2K79O/FsTQk8lPSioTWc4BniX2KfSVtZvuvWcX9PyPdQ6aj5sjeAPYGlrL9rKSZCefiIu/X1biWCsR/RJw/nqWIsS1uvmz7ZEnnE7Htb6R92c1t35Fb2xDMULvenYiCbQpNMpOoGEk07ExCFMgVie0/Q79jf/V6crUuee++NWlw6w7vz4akT9u+2/ZbwBG119cCLs6nbHAkLUjsU/wH+EFDiiUrJpP0aWL/rYrnfjOvpKGx/UMi8QVJtxQ8T14LuEzSq0Tj5ObABrZvziurR48ePfLSK5qc+Lg2/W0i6VDgkqxqOjOK5rgqNa4zsSJFiH0feJV0SFmw68g1aQO8CbFh2D6CiF0q1QmzLZJO4L1xeFtklNQNO0valv54rn8Sh2lF0tBCDCTNYPt5SR+l4LmE7YPT4+aD/fcU61ci19KsezXAujTEoS/xBBELPI64bz8z9NuLYCTh2lDFZ85GdDaXypuSViW+E0sSxcCl05fi5bD9lqS3cgvqgqOIjeVVgVsJJ8c1syoaAknbEE7m1ff4LfpdXkplHWAOwqnoEPqLPkvmVUmXAXfSHzu4Z15JHVkUmNd26U4YANh+UtJiRBH+MbabUAS1D9CUosmzicaXxYCPS3oG+Anlu5qdnQ7N3iaKXz5G+fFnr6bHqW2/Jqn436CkHYlGh5uB3Row7zyQKDo8iHCSb4Jz3GnEAftcki4Fzs+spxP3Sfo94SL/eSIxo0gqhypJ00r6OrXCLdsnZxM2BJIWJQ7dnyHGtDOJxvFdbP8+q7ihaY38LC5Np8bewPK2n61ekPQLwvlnlWyqhkAD49or+ii8GFHSSURaxjj6G3UWzyqqM89X3w3bz0h6KbegLtiFSLG6nBg/bqPQJiPbL6Ux+Suk729yvityTG6hCUWIVVR002KjkXQ1ccZwHGHUUKpjKpK2B/YiIqMrc5Q+4L58qoZG0iSp+elzkqYFXqv25ArlVyQTFEmX2149vb4rBRZNEuvRg4DpiaSXb+WVMywEXFB7Pp5I2SkOSYclY5HreO9Z6vIZpQ2K7ZclrUk0X7wJLGH7P5ll9ejRo0d2ii106PHBYPtHRJcfkm5NXTEls3ZuAd1S60ycKsV0N6EzsWIDYDbbr3Z8Z37OJCaffYSr0n00w41mE5pncV4d8PURG4ilFtLWWQf4OM0pbGhiIcZ+wG2SXgQ+QsTZFo2ksfSPG9MTDlwL2j4or7LBaeC9Ghrm0EeMZw9IuosorH5T0g0ABTdnnEZEaC5LFMlNk1dOR7YivgOHE/fqkgs8K66XdA4RqbIc0ARXl3lsbyVpOdsXFVwMXrE94fSzF1EUtXNWNd0xNrmOTGv7wYZEil2aW8D74CmimL0Jh8BIOpJwebkR2ErSyra/n1lWJ8ZLOg8w/W4NpRbTTmt7T0l9hN5Hgc8/7nWRAAAgAElEQVTUi0kKZSficOcjwKG2j8+spxvOlbQ3cJekm4CXcwvqgo2B5Zoy77S9oKRPAKsRn/dUtpfOLGtIbB8h6SrgU8D9tu/JrakDWwNfJRoxzrB9QYf3l8AFxL1vTHpecsHyb4nC9emJAtrFgH8RxVAlF022Rn6W/BmPaL3H2R4rKZeebji6zevPf6gqhs8CtufJLWKY/CdFq/4ZWAKYStIBUPRc7p20hhpve7ykYovNEk0ak/vS/GdE7boPinVi+3rt+qi27yqTnRowBwLA9pHAkZL2tH1Abj2dSIYz50v6nO0XCSfoX0pa23aphZ71IuXJ27xeEu/avhxAUulGKAOw3ZSodoCfpcfNcorollrTy06E6dMqkv4OYPuBbMJ69OjRIzO9osmJDEnfIJx+JgcOlvTzEjvy02JvX2C/FD24FlEksFdeZUOTOhMvJQpHxgLFxgK38AjwWm4R3WB7mepa0jVVPFADaNIhJQAtjjmXS/pjNjHd07TChqbpxfbFyb1qFPBsExyhbE+IPZc0F+G0VCzJnW8X+jvcsb1SXlUdaZpD3/rpsSqmbQIv2z5Q0ny2t0gdrMVRG8ceJzbHq8aG4rG9m6QvAwsAx9tuQuHZpJJGEfOMaUlzjIJ5Kh0AT2v7Wkk/yS2oC55IG8yvSDoQ+GhuQV1wKrFhOyfhLFhsDJOkG4kxYibgn5IeTv+p5IQBgEVtL5uuD6sK7wunCQV8Fa8DpEP214B1bL+eWVNbJH2p9vQq4EvE2PEl20WvodIBKwCSLiGc+kunUfNOSYsTLtCrEs6eZ+VV1BlJSxD3kamANRqQOjE1EZv5JPARSd9sgEPYCNub5BbRJW/a/hOApJ1s/zNdl15k3ZjIT9q78o/4UFUMA/fHtU9WL9SSVHpB4i2SZNu5hQyDutvvk9lUDI/rk6PgHJKOIlIRSqZJY/JcxPkCxH6L6d93KdGJbfLObymWTybX36owdVSpxVySJiHmQkvUCmlHAJcWuqd8GLBhKpjE9vmSniUar4t0WGbg3ma761Ipdj4xGJLWIZquq+/yDLY/nVdVW744RJPLQx+mkC6pN708AXw3XY8nOan26NGjx8RIr2hy4mMnIkLsDOIQ7Y+U2ZF/KBHXVx383kAcPBxC/028VF6QtBMDi+OKPighooHvkVR1zo0vPR4h0YQFSUWTDimB9xz+zQrMnEvLMGhaYUPT9CLpGt4bNdCYBZXtxyQtkFtHBw4l3NfGdHpjQTTNoe8d4nNeCHgA2MX2o1kVdWa8pFmAaSVNTblOk6a/GLUaK0rewJ+ApBmIDSIBoyT91fa4zLI6sRfxe5uViEIv3blxnKR1ie/zNkQBfulsQ7hCn00UkGyUVU13NCm2fcP0OBkRDVQxfQYtw+FxSXPYfkLSzDTjnn0q8X2u7n2/zStnSOrrvOdLLphMtI4L96fXxhN7LsUhaS/boyWdznvX1aXvBfy1Nu9clvLnnXsB5xLFv6XPKyp+CxwBPJ1bSJc0ySGs4m5JSwF3kvQW6hAGA5ty6uNxsQfwDYz8nKdyDqzRR+Hrp8RpwHoAaX7/PcL1tVTGAbemot8+Yg+89GSd84EVqEWf2y69AP9nwDLAHYRj8UWZ9XSiMWOy7U8CSFrA9v259XRBu/VzsfPkGqOJ9dO2wDWUW8wHsAWwJzALsdaDuH8X2XBNFCrfVn/B9g2FG0r0pSSEEa3XeWW1ZYZ0vtdHxLZPOOtrwHl1629v1bxyhmSxlud9wKbEnPm0D1/O0DTIhKhHjx49PlR6RZMTH5Wb4H+Su1mp34HP1h0FbVeFiDdn1NQtzwOfSX/QjAXgzzq/pcf/kSYdUlbUNzVeJxbfpbMNEc/dlMKGVr2lH1BCLFYhFoCfpX+sK5aWw+BZgWcyyumGx21fmVvEcKg59C1IMxz6jiXG4b8QUcHHEVEwJbMvsC4RgfcwhUbhVRv4DeVk4KL0uBxwEvGZF4vtPwOSNBPwQuGHwRCx7fMCexAHqjvmldMVcxIHwVMR9771gf2yKupMk2Lb3wCmI353m9LviHE0sGRGXYMiaSwxp5gC+Kqkx4mi2ueyCuuOo4F/A38iDt5/B3wzq6L2fDa5d/YBC9Wui3Qgtb15dS1pOmpFDQVTFS80LSYR29+rzTtPtH1Jbk0d2JwonNxI0gPA/rZfyKypEy/ZPim3iGHQJIewihWAtWvPS24wWjg5xvW1XC+UV9bgNDTyc+82rzfBlf1KSb8nmoD/DSyVWU8nVgKmb8C6qc4fgfuIzxdivCi9aPKS5Mp+eW4hXdKkMbnid0TzSNG0zJPnJ/YD7iaaHUpnrO0bJW1r+0RJm+UW1A7bxwLHStrCdhPMOyZp8/rID1XF8JgXeIz+IsnH02OpzTq30382dkftugnn1U367e1eXUv6JHAice8r2vwpNbpsw8CGjCLn9j169OjxYVBqwVyPD46HCReaXVIc3t2Z9bTjPVHRKZrrlRxihoPtzRu4ALwHWI1+u/PZgD9nVdQGSfU4ndnrz20fk0FStzTpkBKY8F0eRRQJNALb7xKLV4Bfp43bTTNK6sTUxIbyFES3+xLERmixtEQY3S9py2xiuqd+GPw6cFu7NxbCsym+6A76O9xLHt+QtDawhO2fSLpc0tuFd61OYfvCdH2+pF2zqumOm4kD7DsljQCKLBCQdITtHWqRuxMosdClhSlsV+PFXZK+llVNF0hakSj6HQd8TNK3q/jEkpC0fMtL0xKuUE3gdGLDsyluW9Cs2PaliTQEAdW97l3gimyKhsD2rACSZrPdhHVenflsV7/F8wuPFC81emtIJJ1EHF6Po99lefGsotqzqKRFc4sYLpI+TRSyjyLixB4Y+l8UwXHE/sqpxF7AicA6OQW1o+ZAM07SnsDf6F+PlDy3b4xDWIXtJv3+vl67PqrNdUk0LvKzXqQsaRWiWOsmCh7jam5gxxMpCCsDTdgfeoBI0mlKzDXAuHrhWUNoVBJXw8bkilckHcrAz7jYvUNJOwBfJRIFTgTmA3bIqakL3kh7GSMlrUYzkjL+JOlsyk/WuSxFn+9ve5ykaYB9gKvzymqP7Y/n1jAcBmns+wTwkO3iz9hp4G8vFSHuBuxmuwn7nTsRiTQv5hbSo0ePHiXQK5qcyEhFUNPYflnSbbZLPfz7l6Ql6hbtkpYAXs2oqSsaugA8D/gHsAhRUFTy5zxr7fq0lucl06RDSgAkHU1sej5L/6Ff6QUvrSi3gA40LkaspXB5NsqNCJ4QO2j7z5JmtT02t6YueSQ9zpJVxfDYF6jiHTYALqPsrtVJJS1i+x5Ji9CA3x5x0H4JcRg8P3F4WaI77f7pccMh31UQqdkF4DlJ6xPxRUvS/1ssmf2BZW0/JWl2Iv6zuKJJ4FrgISIqGvo788cTjq8l86rtfXOLGCatse075ZXTHtvnE3PjNRvgUlznBEmTE25959puwngxhaSpbL8qaUrau3tkx/ZjAKnodw0GOjeenEVUdyxge57cIrpkwfS4FNG0egPwOaKRssjPON2jf0A0JN4GzAWcI2nvwg+mZrD963R9p6T1sqoZmsqBZhyxlzVfel66I01jHMKa2GCUnM2RdDkx1zzP9r/yqhqSJkZ+ApAiuucgxug3CHf2UhNUTHyH67GkVVRwkb+/xBeARyU9T+hvQjz3FZK2pdZkbbv0NVQjkriaOCbXqM4WZs6qons2BJYHrrJ9mKRbO/2DAtiOOFsYTey9jM4rpyuOoRnJOgcB3wdulzQV8AKxBvl5VlVdkPaRv81Ah76t2/+LvKSG8L2IepCzJI23Xfp3eTtgARrw25M0K/E7exVY2vbzmSV1y93AGNvv5BbSo0ePHiXQK5qcyJC0MHCUpI8Bp0i61/bFuXUNwveAC1Lc2cNELN4niDi80mniArDP9raSjidiE6/LLagdDTy0rmjMIWWNRYlizyYUEzWVJsaI1QuVX6PscXkl+hfVp6bnTUC2SyyGG4q3bI8DSB3CpS+4dwSOTxsbTxGbXaUzu+0TAGwfLOma3IIGw/YzklaxfaWkg4lu4PHEgV+pHF27/k76g2YU075Tud3ZflLS67kFtWEJosh3ccI54JRC3Q4mUCumfUbSRkS0UeVeVazrDwyIbZ8ReK7kuVzV4ABsKmnAnKjke6Ht1ZJbwxrAqZKmtL1Ybl0dOIxw0b2XcB7ZJ6+crmhag9EtktTizF4ktveAKIKy/eXqdUnFFTTU2AlYoe6Oktw9L6Bs9+IpJc1i+2lJM1PwXkDlSJPcihez/afUGHxKXmVDUzmESZoBeKHk+x79DUbfBOpumNNn0DJctiRcUo9PjQMX2z48s6bBaGLkZ8WytpeXdI3tkyRtl1tQO2x/EkBSHzCH7TGKSPSi98Ftz9f5XcWxHDA5USAOzWg8e8H293KL6ILGNX1W2N63Kc60iRGkQuX0/I2MWjqSCu1F7Gm9DqyfEq5KpzVZZ5esatqQ5mo/S39N42RiH3FMpzcWwq5EwsflxDnJbRRahChpLdsX236Sfkfo0lOA/g68RTQGHCr1e7jYLjlp8GrgYUkPkQxzbDfl7KxHjx49/uv0iiYnPg4HNgeOJbofLgOKK5q0/YSkzxHRVrMB5wA3Fb7xWdGoBWDibUlTEFHB4yl4bJA0ltA4OREbPQaYHfiX7U9klNaJ1kPKn2TW0w1PEfGZL+UW0olajFidPmC6D1vLMGlijNiEwuUUyXw0sHo+RUPS1+a6dCZL0YMP0B+xU/T3gigSOA24kXDouyOzniGxfSfhpgSApOUyyumW8ZLmt/2ApHko9MBd0l7Ap4AriSaSnxCHPD8milWLw/aKAJJ2t118V3sLL0nakTg0W57ozi8O27cTDgJ9RAH7jyXNAlxo++ih/3U26rrqrgHjKbQIP7mwHUJ0uG9S+qF14qL0WGrE56BIWpeI+FwKeJxC48Tr2D5V0mXEweojDXFAaFqD0TjgVkkv03/wULp71UySPmr736ngbIbcgobg7dY4OdsvNaBZZy/gBknjiPVpE5p1Tif2MCDmFqcAa+WTMzQpvu83xPz4bEmP2T4us6x29KXGjJOBTYmxYgQx71gyp7BOpAadW4GPAesSCQMlFk02LvKzxqRpf3a8pEmA0sc3CEezB4FfAJtI+obtnTNraktyCDuecPR8GtjCdtH7F8A0touMlh+Char5RW4hQ2H7mXQ5CeFwNz9wL+GAVzQNc6aFSAz7CzCXpEuB8zPraYukzxBzob8RCVxzAQtK+prtf2QV15nWZJ2ikbQqsAsDXRuL3G+p8YztJu1fvGP7jeQwOV5SyfHcu5LqFSSdaXuDzHq6oWRDkaHYhkixKvo+3aNHjx4fFsUWRvX44LD9YJog/UvSf3LraYftdyXdQcSTzgfMJwnbRUZF1WjMArDGkcDORDfMGOD6vHLaY3tWAEmnAHukTubZgEPzKhuaJh1S1iJJZgL+Kenh9J/GFxxN0m5T6MYPVcXwqWLEqkijYmPEKiRNTzjSbk1Erf4ur6IhGd/munTEQLec4r8XtndMBSQCzrJ9Uad/Uxi/pPBDSmIT8cxUaPYksG1mPe1Yhf7on9dsXyHpSuDmjJq6ZQ1JhzQhmqTmgjga2I0oXh9LNEcVS9qgvYGYY3yTuJ8UWTRZK6Zdq+7ML+nr+VR1ZGfg00Qxw68IN6iisX1XurydiN2djdgkvzubqO44kDicPAi4vOQD4ZQmMNjr2N7iw9YzTJrWYLQSML3tt3MLGQY/JSKjnwc+CuyQWc9QtHP3GfGhqhg+M9ueW9Io28/lFtMlU1f3PtunSSq90HM00TzyB+AA4K9Eo3iJLE24poqYA/UR3+3ii+8lvQA8Rtz7Vq2SBgrkIGJO0bjIT6L55W/AjMT66ZC8crpicdvbAtjeSVLpDoiHA1vZvisVRh1JRHaXzL2SNiSaUxvhfE8YBjwv6TlijCu9keQ44GAi8np5orB21ayKOtMYZ1oA20dIuoposrXtktd7BwHr1t3jU4LfL4E1s6nqju8yMFmn2NjoxKHEPkZTXBsBHpG0GwPH5JIbM65PRgdzSDoKKLm5tm56MVM2FcPjtdwC3idPALc2xEG3R48ePT5wekWTEx8vSNoGmDottos93Ek0LY6rUQtASR+xPc72H2qvnU04N5bO3LbHANh+StKcuQW1I8VZbUC4djxBdCqekFXU0NQjSapCvskp2DW1ihFrID+3XXTUWYWkzxKHqJ8HzgKesL1aXlUd+Wwqzukjutyr65ILgLFdfCdwHUlfsX0BcA3hEL2UpKtbnYAKp3gnUts3AxOiXyUVGy9XKzo8rHqe3JVKZ0bgKUmPkFzDCx4rBis0XIY4hN93kP+WlfR9XYOI6K4Kw3cuOcJW0lrEPW9jSdX3YATwFeI+WCJv2H4ReFHS1LnFDJPjiRSEFQjXn+PojyAsDtsLSvoEsBpwrqSpbC+dWVY7liAc+k8hDoKLv+fVqBqMKkpvJHkAmJn+KLEm8BgwL3EPfLbwxoGF04FfnT6iMKNktgZObVDBJMCbyfnnJqKxqOTvBcC7tl9IDeKvF94gfj4Rl7mm7Utz6xkmaxKNOlsC60m6skTH8JRSdFD6axrPE2vqeYmG60aMG5JmsP28pI9S/nlTX9W0Y/tOSU1odFg0/VUU63xfYXuu3BqGyTu2L0vXF0kq1i21RqOcaSXtXXu6oKR1be+XTdDQTNW6V2H77ymyu3TuA7a2fUdqbv97bkEdeNz2lblFDJNpGDguj6dgN2vbe0panSjy/Ee9KbhAmmiAsUt6/CRxhnobsX8/joL3tAitVTJiVfy7cV5JPXr06JGP0hexPf77bAnsCTxHHJ5smVdORxoTx9Wy8KsofQF4iaRVbL9ee21tIlJl1kyauuU+Sb8HbiGKBP6WWc+gSNqHcEvdAniGiHPYTdLstkfn1NYO248BJDeJ+W3vLumPwO/TX3HUClwq3gJGAq/bLvkQ7dvEAXYTuIEYGxax/WZy0i2dT+cW8H6QtA6wPfEd7gNmsF3k/4ukgwgn6EuAXwOvEIUCvyVc5JpC8RsxqellV/q/F28TTtylMZmkyWy/mQ6FSRvLTVh3FBs72UrlglhH0giiuKG4okki0uop+htHxhOu7HPZ/mNWZe25i2h4eQ2oDkzeBc7Ipmh4NKkwDuJed7ykTWzfkL7PxSJpcaJ4ZFUiDr3UQlpsf1rSp4BNgB8SqQin2H4wr7LO2F6087uKYlng0eSqBOW7KkG4FX+CWJOcQtmNte2cfkuPx5s8paiY/qaM0g+ktiLWfocTh+/b5JXTkQclHQjMIOmHRDFw6fwQaMKaegK2b5I0hpjTbQxsRoGO4bX0lPdQcENUxb62l6dsF6hW9gNuk/Qi8BFiL6Nk3knNUdcRjoLFNohXDLb2K53kyncU4YB/CnBviYU6kr6ULl+R9H1inrwksYdfOocy0Jm26BQu+j/TPmBxynYKb1dMXbLmilOBS4gCufmJ+XPJ885nk/th3bXxmLyShsb2pvXnkmbMpaUbJM0OPA48Anxf0hO278wsqx3zSDqAGCeqayCKP/PJao/t9QHS2ci6tt+SNClQegrXgbkF9OjRo0dJNOHwssd/l6MasDlbp0lxXNXCb11iAvpX4HNAsQ6IwNnAhZLWJBZ9RwKfoewOmIrvAysSi78zk8tZiaxme5na83skbU44shVZNFljO/rjar9MbBwVWTQJLEAspo4EjrZ9i6TFgO/kldWR+gHau1B0R9dyxOHZ3yWdS3RVFk2tAHgS4qBvYcIB6Lc5dXXBaELvtsRYUXIk0PK2P582A9YC5rD9qqTrcwsbjDYHaH3EGFI62wNfBPYi7t+lOh+cSkQB7Wj7xeQ28iug1RmqRN4GfkZEwJxNxAM34dC9GueWpdxCuQuI39486a9iPFBk0WRyND9J0u/rcTUp5qpUGrfBXEfSAulxDtofVJXCXsC5wFdKjuausH0vUaCDpOWBAyV9vFR3TElH2N5hsPt2yQUvtufNrWG42N5Q0seIw9SzJT0LHGv72rzKBmUFoljrVNtNikH7QW4B3SJp0hQv/zhxyF4lT5TOtsRa9XrgZaI5sXTGSzqPgXsBRd+r097Fc8B5wDdsl+qqu2HntxRL474Xti+WdBkwinAsLn3M2IIoCj+IKAovfrwYpFF8nO3F2r2/EA4HNgeOJRzkLwOKK5oENkqPLwAL0r+/3IRi2rMlXUk40z5s+/ncmoai1Zk4jRulMruk1ljrPqD0ZiiA2W2fAGD7YEnX5BbUgUfS4yxZVQyDZJ6zA9HUPjXwEDF+lMppwD7EvvI5RIF1qcX4e7e5bgKzkuZuxHgxc0Yt3XAPkZxSmTPMBvw5q6IePXr0yEivaHLiY3JJnyaKRqrNl1KLEKFBcVzVwk/S12xXhVqnSvpTRllDYvuw5P50AbEwuRr4vO238irrigttL5tbRBe83vqC7XcbEv/yTjowIXVIFbvxafsNAEnz2L4lvXaHJOVV1pHGHKClz/WWFPe5IbC8pJuB39s+Iq+6jhxDRCL8ibiv/I6yXRDH2r5R0ra2T5S0WW5BQ/BSelwSuMf2q+l5qZE1TT5Ae8r2WEnT2r5W0k9yCxoM20dKehf4i6Tpie/Ib0qM7huEY4BfAj8mGgVOAoosKBqEKYAd019x2N4st4b/A/tI2o4Y16Yi1lEL55XUlsZtMEuaw/YTwHcJF9IFiY38M7MK68y3gS8B60jqA2azXXSnvqRpgf8hDoenpmy38/3T4zeB+n7F9Bm0dETSXrZHSzqd9xZ5ltoQVWdmotlzFFFAsp6krQpM/eijGQ4/E0iutI8BTxCNn5MRzSSlcjJRQFu5YkJ/4WSRe3GSRgGv2j5K0jeIg79i9y5qHJ9bwPtgZWKcmBfok9RXYoFcrXlydprXENWY70W7BgdJxTY4SJo+fT/WlzQL8HZDItCrJs8+4LPA+hm1dI3tByWNt/0vSf/JrWcwbG9eXSdn9oWABwp2YZtAKoar//beAsYAo20/mktXOyTNX3s6K5HGVSqnMXgK2+kftpD3wXhJ89t+QNK8wCS5BQ2F7X0lfZnYY3HBxih1/odYO/2SKBA/PK+cjrxL7HH+yPYZKV2uSGyfVF1Lmo7Y62wKJwL3SrqbSD87JK+cjpwH/ANYhDjDfnXot/fo0aPH/296RZMTH6Lf6aXojU9oZBwXwPSpcOuhVLD1kdyChsL2zyWNBFayvXtuPcPgBUk7MbD7ukSnonYbyKU6QdW5QNJ1RAT64sTYUTr/lrQ/ofnzwNjMejpxO1E4ORvRcX13Xjmdsf0K0SV+XCrC3zKzpG6YL0VcAZwv6YasajrzRnKBGilpNeJgqlTeTnFGmwN/gAkOVkW6btUO0KYF1mDg5svJWUR1zzhJ6xIboNtQ8PfC9m8p39F1MKa0fXUqfrGk9zQ+lEoam9fLraMTkr4J7AFMXr1mu9i1SGIdYA6iG/8Q4Dd55QzJjbkFvA8ulbRSckJcJhUg/gjYhbIj5s4lNpg/TUS4F7vBLOnrRNPAXMS9etsSD1Jb6EsHqycDm9JfLHc0/U74JVFFb50JvJhTyHBJTVCvEk5QewMfT0UOV+RV9l7Soeq0wNclzUW4IZ5r+6UO/zQLkg4h0kdGEnPjscCTRMHyWhmltaUq8rX9ydxaukHS7oRD/5upcGtOIgVmFaC0ot9WTqX/+9EU96qNga8SBewnEcWTO2RVNDRNbIh6pPNbimFMmtsf1fJ6cYW0AJJWAE6W9BnbLxJzuGMlfcN2kUkZFVWjeOKvkopu1Em8kPYtppa0IYXuEVVI2pEY424GdpN0lu1fZJbViUeJpLPrgGUI45EbiT3blfPJaku9kfZ14Hu5hHTC9r7VdWrOmCqjnOGyM3CmpAWBvwOtjplFkcaz+Qi38G9JWs72bplldWKs7dclTZf2Dks1DqgYCRxMNLevSLlGBxOQdBKRevZv+usYFs8qqgO2D5d0DpGu84DtZzr9m8z02d5W0vGEY/91uQX16NGjR056RZMTGbY/lVvDcJC0DmEbXm0izmD703lVdWRn4DxJMxNuAttm1tOWtCipCmjnl3Q+4SxRfPwL8DwRJf6Z9LzUeMdlJT1Ff5Fk9Xl/LJ+k7khuKRcTxdYn2S6+oA/4BvGbW4vYGNgnq5rOHE9E1KwAPE1sbK2QVVEHUjfi/KnI+heU7VJUMYWkqVJs9JQU3mULbEc4CYwmnJb2H/rtWdkJOID4/v42FXkeTMT5lcwFRLzjmPS8yIOdFrYiDib3IDaXi3QUrEgHJNsQhal9wHjbC+VV1ZHX03d4EklLM4hbdI//Mz8gDnPGdHpjQYy1/UZyeX2w8A3xoxnoDFYxHljpw5fTFfsBl0lamVjznUI4Cxa9IU6zNpjPAO4H7iJcBA6ozNgLdkFcmphjiPhe9xGNcsUV8gHYvitd7taQNIQ6m9j+Z/VE0mnAkrZXy6hpUCTNB5wPXEgUFi0M/EDSuradVdzgfN720pKmIJxzVocJzlBFIunqdv/Ndon3kfWJddM0xDg3h+23Jf0lr6yuOI+4781OrE+fonwHqw2B5YGrbP9K0q25BXWgiQ1R26XHPmKMe5Qo+CyRj6a/ij6imfJVymxIHA2skAomsf1HSasS+3DLZVXWgdr+PUSB9btDvL0UtgT2BJ4DlqD8huuNgeXSPWQkcAOx51kyc9acMp0KgI9LxczFYbvUOOC2SDqaKEB9lv7CrVKddBcnxrMliX3ko4BpiXnG3zJK68Tytr8AIOkw4KbMerrhqZQK9Voy7yj9nG9zYFUieWtd4Ft55XTFAg1osAZA0rEMcq6QnLdLLlp+O61Tpyb09+qFevToMVHTGwQnEiStTzijvEpsipe+sVUxmjhw3xa4hpjcFU3qTi29sLPi/tr1j7OpeB/U4zMAJA0WmZAd2yUfrA9KmixvQ0QLPAfsRRSS7Gb76azi2kOck6EAACAASURBVCBpCdu3AV8A7kl/EAWIJRbTVsxg+3hJm9i+QVIT4ua2o9/h58vEBn6JG+J1DgPuknQvEbVTZKyxpFltj7X9JOFCA/A1SUvl1DUUth8CNkibs28RxQxFFjS0MKLAyMlBSc6ddaalGc6/OwFr0izHra2Jg5FRwG4U3PjSYB62/WBuEcPkCUlbAK+kA8uPdvoHuagOoiTtbvvnufV0g+1z0sHkn4jDhsNsH5lZVjc0aYO5cQeUts8n3MHXtH1pbj3DoClpCBOoF0wmSk5E+AWwUb2ZTxGJ/guiIL80XgNITjQP114vuVnnZaJJ5yyiQPW1vHI68qrtt4nEifvTNcBbOUV1ySjby0j6HdEM9afcgrpgBPH9rb7Dbwzx3hJoXEOU7Y2q69Soc1ZGOUNie4/qWtI8hJPnxYSZQIm83eq0neJrm1CAWN+/vwu4PJeQTkiaw/YTwCwMjJsfBbyQR1VX9FX3ENtvpajr0pksjXE3EoV8IyXNTaGuiJJGE8WzE+ZBtkt3WV6USC8qee5W8XPgW+n7OxpYHXiQMGu4MKuyoRkpaYTtd+kvTC0SSWfa3oDYO5yTSJ/YkkhGKJmHicbUvYCrgSJd+lu4RZIKbYxr5fz0uDXhVvxXwk3+s9kUdceRRMLLH4nG9qJdt3v06NHjg6bkjf0e/112Jgr5Pgb8ioiZawJjbd8oaVvbJ6YOnqJJ3Xw/pBb5WWpXjO2TqmtJMzEwprRoJO1HFG9NRmwGPEB0YReFpL3b/Tfb+32YWobB4cRhyQhi8nwr4dr4WyKKqURWBm4DNmp5vVQH0glIWiA9zgG83eHtJfBOy0ZisZsZFbZPlXQZMDfhSlPqpvipJCcwSb+3XW26HEi5DmEVWxP6m8LdqRj1TtKGnO0380pqy7XAQ8RYDAOdi0t1HQG4Gxhj+53cQobB6rY3rJ5I+i5xT+zx3+PVNB7Xf3ulu5tvA3wcOBvYjHAhKZ01JB3SlN+f7dMlTQJ8m4gIbgJHEmvs4jeYbf8ZIEWff44GrfmIyN3ViXXJr4Ef2z4ts6ahaEoawlCUPLf/SGv6ge3bJZXq7jJlcscc0XJdZDEDgO11JE1PuMYfSESKnwZclVXYEKTC+xEt16UnC0A0tQNMbfu1ygG4cE4j1h9zSbqU/kPiUmltiNpu6LcXx6RExGPRSNqemBPtYvvi3HqGYEStMAeANP8sttm91kDZGtu+KOXuBewqaR/CPf7l9FpVCFXyvtb1KVb1OsJ5tNi5fY3NiEK5Q4F7gS0It/ZdM2oairWAT7TEzZfOU0TjchOKzCaxfbek2Yi5xe0ADSgMPxP4q6SbgKXS81KZESDts1Tj8qH55HTN0cR3eVVib/lkosm9ZMYBt0p6mf70oiKLrG1fAiBpJ9sHpJf/LKnohijbf6iuJZ1tuwnjXI8ePXp8YPSKJice3kjxEy9Kmjq3mGHwRtogGJk650blFtQFPyCKUhsTOyjpSMIxroqRLjZqoMY6wBzEwuQQ4Dd55bTlmfS4LrGYqjqN5symqDML2/5Ccs9ZDlgvFcd9L7ewdtj+Wbq8i4gSb4qz2XeBE4AFgXOA7+SV0xUXSLoOuIWIziy5WxUASb+2vSPh/rMaceg+f2ZZg1F39pmjzeulMrmkOxjorFRyYdEKDHQkGk8U1ZbIEkSR1uJER/AprQ4ZhXI18LCkh+jf4CrykETSRsS8YkVJlcYRRIxtr2jyv0urY1zJxTkVUxOb91MQG7dLAPdlVdSZGYnIqEdIjlC2i5zbJ5e48cQ4MQ9xYPkgFH8fmcL2QdCoDeY/ADPRv04tvfge4KfEPfBIwlH+LKJop0iakoYAA357dfoodz4E7efEpe5tvgYc0+a6WGy/QEQ6HiVpLuBg4EQGrk9KYS5i/VF9N6rrJswvzpX0YyIR4UbgldyCOmH7CElXAZ8C7rd9T6d/k5PkdldviCp2TK6QNJb+7+9IwvigSCTNTuxnvQAs2YB9uFOA0yUdQLhufZxIXSq5QKde6LsiscauxrhS53APE3uz7wB72y7WFRNA0qTEXsAlRFPDgkRBeIkO1gOw/VAya1gIeMD2w8TnXyp3EGvq4osm0315PLF2+mfNMbzYdTX9LturA1fChGaSabMp6gLbv5R0BbAAcJzte3NrGoJ50j3kPRTeDDyP7a0kLWv7Ikk/zC2oC1YCpq+5yDeBaVMtw63E2fqUmfUMiqQjbO9QG+eq1yl4fOvRo0ePD5xSNxZ7fLA0ofiiYjtiwjwa2D89lk4TYweXAuaud9s2gLG235A0re0HU2xNcdg+GkDS12xXBXGnFt5p9J/0+AXglhS5C4VO9FuYBLhS0v3AsbavzaxnUKq4mrQRsExuPcPB9mhJFwMCTrZ9V25NXfCSpIOAaYgDnjUy6xkuTTj0+0FuAcPB9qK5NXRL6g6/PTmErQT8WNIswIXVPaZQtiFciv6dW0gXXE44Kc1AdGBDFP8+lE3R/1/OJ7rbi3XZGoQLiMaieqFZ6ayVW8AwOKrNdelMcFhuSMEk/9venUdZVpZn//92QyMBcWBQRlERL8CBaADlDTI5xoH4SwCBEAcCCiKCQhxwRHlFjL7ExIFBUGRUUAIO4KwIqFFBEZVLGgXUBgSJEEBApH5/PPtQp5uuqa3u59mnrs9avc4+h6q1rlWc2nX2s+/nvoF1e7gIfidlE9q9tm9ovcN5X6YhdCb6fWv59/AySQfa/vDgBUkHAD+smGlCtncCkLS17e9P9fUtUWl7ONhUYmD/uomWzvZjlvZ6w91HkTQ8rnYlymfORfRgpLikrSidzVajdLXG9j51U01M0rsp793mz8lD74slC8xa7jT5U0rx09eBDw93S21x44vtEyTdRtl8vz5wDfBx280WTS4xrv0bLf5cl2IvynrhQ4BTaHiUeOc0ytSfdYFzKKPQPwZ8sGao6egmY+xJGQd7mKRP235/5ViTuQK4XtINjG+sbXWzzh5Tf0lzvirpYkpB+C6SNgE+RNuF4UjahvLzXhXYufts0WpTiTspn4v7ZmVJawNIWoN2J3AN+wXwSOC3tYPMwL8AHwA2pWy0fnnVNBN7d/fYx/NcRMRyk6LJuWOwC2YeS+yIaXEXjKThDmCDG5SH04+blH0cO7iQcmFy51Rf2JDfSNoHuEPSUcDDageawpqSNul2gQp4aO1Ak7hd0iuBXYHTJc0H/gm4rm6sqdn+APABSVsD/yrpeNstdhT8JONjmN9s+6jKeaYkaV/bH+t+3wbn4i0lvaT1c5ztt0j6N+BxtnesnWcSYxMc98GllMLJ9YHPU0YzN0vSLsCBlO4d84C1bD+5bqrJ2R6TdAllp/tLgX0ZL/Br0W+A7/dhQ0bXFeWbwDclPYLx0bW5Vpp951BuUA46cffhXDff9t61Q8zQAmA3xs9x61MKmZszGB3dQ33rsAxwpaT1bS+qHWQGbqPcbD++G//5u8p5ptKXaQh9/d17C3CCpFdRNjY8mrKW8dKaoabhUEmPpnQ5O9V2sxtKJL0B+EfK79oZwHa2m18nGposgKTnUIoEWlwHgNKxejXK++ES+rWx/aOUn+0NtYNM04voyTmZB74voP33xt/XDjBTXYHkYgVEkhYMbRZvWR+umwDusn0PcHOrDQ6WsIntrbqsP6QUAu9k++eVc03HnsAzbN/bdRS8BGi5aPIlwGPowcZa29cCSHoc/bmuPlrSecCtthd1RZPH2z6ndrYpnAwcDbTerRjgBtsn1w6xDN5CmX63HvBd4OC6caZlO+AaSTd3z5sdzz1g+2eSXk/pWPyLhhsrHTLJZtSm7/FFRCxPuRE4d7x9guNWDRcBDP6Ar0PZpbHqA7+8KUuOHeyDRwHXDkbh0faogYFXUXbOnUXZtdP6TcpDgHMkPZJSSNJkp4bO/sC/Um5Qnkwp7tuVtjMDIOmvKDd5XkZZzHhH3UQTGl78fjbQfNEk4wXsV1HG7DRvidFW84BHDl5r9EJ7O0mLKFnXHDputlPKkJOA8yljr28ATuyOW3Uk5e/I/sA3KL+HTeoWv/+O8Y4N5wKH2G59d/ODKOMGr2B8E0nTf6slfRh4AaXbz2DsWeufh/pmXssdiSZwuaSnsfiGqHvqRprS6ZQC1e0o7+cH140zknrVYbnzDOA6STd1z1v9PDTsjZTC5Z9JeiKl80/LejENoa9s3wHsJWktyhjxRbab7z5ie4+u8+FewFmSfke7UxHeSylIvQ84CHjNoHtc42tEvZksYPvJ3flsb+BNlBG7pzZ8Y3XYbT0rFujNObmP74ueFt/TFd6/nvEiqHsp9xti9rVe+Atlgw627+maBjzH9i2VM03XvMHoWtt/ktR68e+1wB22mx/PPaRX19XDxb62r6Yf01Ousv2J2iGmqcnu9tOwkW1JWge42XbzRfi2H1c7w0x1mzxfBvw3cLik02wfUznW0lxZO0BERItSNDlHDC9qLdFBp0mDMUZwf4v21wBbUIowWncasDWL70Br3Z5Tf0lzVqeMFV8VuJWyK/tnVRNNwvZFwP1dzLoimCbZvpnFbwR/rfvXB5cDZwMHtLy4TH92iN/P9pe6wz1sP6dqmGmyvV7tDDNhu9mbONOwlu2TJO1t+5Jusbll19v+jqT9bX9C0strB5rE7yiLs2cAH6ecPzaWtLHtL1dNNrk+FIMv6WnAY/vQHbNvhm5S/1LStpTutH0pQNyB0qloYIxSrNOy220fJWlT2/tI+nbtQKNE0pOBZwJrUzZDnWX7qrqppma7jwUBH7O9HYDtK2qHmYa+TUPoq1UpHVIeIeks4HLb36ucaSqPpGxWXZuybrFr18m/tU7GSx133boeTRYA7j+fvQlA0vbAUZI2sv30usmWruveCXCrpMMpRQODz3EtX4/06pzct/dFjx0I7Ai8lbIR/5CqaSYxNOmlF5PDOk+QdDol8+AYaH8TJXBjjwomAS6SdDbwbcrmqIsq55nKRsDVkn7ZPe9D045cVy9/n5F0JkP39Wy/q2KeCdk+rHaGZfRK4DTbN035lZVJeqvtIyWdwRL3z3rwN2Rv4G+7IvZB998Wiyavrx0gIqJFKZqcY/rSQae7sbon8GrgHuAhwGNs/7FqsOk5h1IwuQGwEuOFDs0ZjNuldNpasoir1cWXgXMpP9tB97umi9D6uJNZ0kuBN1O6hQFgu/Uigc0Hu2wBJK1nu8ULgbUkPRuYT+koeH8RYuM3HQD+R9Lfs/g4yl/UjbR0fb3Q7jYL7MHQBgfbr66XaHokbdY9bkg5x7Xs7u5G1AJJz6XcwG7VuZT37ybdv4ExoLnzhaQX2v48pSvmklrvRrKQ8nvX/BjKHjLjN/x2Hnq9+QJE21vC/RvPfm+7D92WxyStC6whaXUa74jRJ5J2o2wuOg74AbAxcLakt9s+t2q4CfT181DnDknHsPjnzuPrRprUktMQ+rg5sQ+OBz4AvI3Sje1koNmiIknfo3y2OAF4+6DDkqQvTfqNdTx3kv/W3O/eBJMFFgG03klX0hrAP1DOE6tTxjK3anAuu5WyjjVYy2ryemTIqyjjuXtzTu7Z+6KvFtm+vutA+k1JrU6ogcW7QfVhchjA7kPHx1ZLMX29K/KUtDKwC/AFYDXKKNj/YvGNfi16Se0AyyDX1cvfgcBn6MHY9h57kKTLWPyausnzG6UAHPrx92NJ82z/Ce7v/tvqBvGJPg+3/rk+ImK5StHk3NOXDjrXUAoN97Z9laTze1IwCbC27W0lfYwyzugrtQNNYlBwuGRL7qYLEDvzG+zKMJne7GQe8kbKgsuvp/rChrxd0gHAKpSFo18AT6gbaakuZXyk/GWMX6z04eLkEZTOLoMxO6sC29aLM6nPdY99u9A+GTga+J/aQWbgtZQuiJtTur02WeQp6fHd4dHAlpQRhP8PeHe1UFOw/fLBsaSHUgp0ru5GVLZo/e7nvBPlnDYG3AT0oWPDo4BrJQ06Ffeh80Ev2F6sc1WfChAl7QicRCkUeLik/Ww3+fle0oa2fwMcAbwYOAX4JaUbYsyOg4Edhs/Bkk6mFLg3WTRJfz8PQenOAKVLHzR6nSrpDbbf162zXNu9diZlCkUvOrT3zF/Z/npXEGxJd9UONIXButaawyMpbU9WoFhL3zr1rwcgaWfbX6+dZzok7U7ZILcxpUhgf9vXVA01BduvgMU2XtM9f229VNPyKGBXytrQPGA3oMnuVX18X/TYrZJeTCmGehUNb6AcTA6TdHrDBS6L6eHY9r4VeUL5fHkvsC6leceVwMeAD9YMNQ0LKOfh4elsr6qaaGpLXlefUjfOSPq97aNrhxhxb5z6S5qxAHr5twTgO90awLeB7YBWJyG8sivq7PPEs4iIWZeiybmnLx10/h34J+DRXfHhvCm+viWDn+3qtv8oqckbO7DYuN3/oowebHps+xIul/Q04Ef0Y7xjn3YyD/yy8RHXS7MLpZPAMZRCqI/UjbN0g5sOfSLpU7ZfYnsnSYfZfn/3+jdqZ5uI7R93h9dRCoCHz3EtX3xfZfsTtUPMRDdKrNXi2WHHLfF8N0oh8OYVssyIpH+kFN6vDHxa0pjtIyvHWpo9eeCu1b+mbBhoXfPdZ/quTwWIQ44EtrO9SNIGwGdpd1PUF7vCkQsp3dcGo6RbvxnVJ/cuWbRu+zZJzRYAD30e+gDl9+9U27dVjDSloQLgJic2LMXzJd1q+zgASc+g3FRtrjPfiLir6xS+kqSnA60XTa4n6RxK3rOAa22fWDvU0tg+YnAs6VmUbtDfpWxGbNk7gV4UTQJnUgpcfgw8CXiPVBq0t1oUJWlPylrLTpIGHcPnU/L/R7VgUzsDuAC4oXaQaejd+6LH9gUeR5mscyil4UHrVuk+0/+C8S5hLa+B90ZPC3M2sb1VV/DyQ+BuYCfbP6+cayqnU4o8t6NMD+tD18ZtBuvfwHlVk4yumyUdR2kwMbjHl2uo2XUppXByfeDzwOV140xqE0nvWdp/sN3kZMShe2av66azbQ6c2eokEspm3+czPhEIxqeSNj0JKCJieUrR5NzTiw46tt8HvE/SDpTFjK0lHQ2c0hVmtOyzkt4G/FjSd4Hbaweahi8DP2O8Df4Y8Ol6caZlB0oR1GDUY+sf6nqzk3nInZLOZ/HC1CYvTobcYPvurjh1Yes7piTtQulCOthlu5btJ9dNNaFHDB0/HxgsGjVbGD7kXEqRS186N36m25n4s8ELtpvsiDHQFYIfyNBY7hZH4tneacnXJM2n3Ax+5woPNDOvp4yevIBSxPWD7rEpffwZD3XN2Z8HntNa/7vXN30qQBz4s+3BqM/fNt7R7F3A+ZKeSflscRrlJtpTqqYaLRNNbJi/QlMsmxcA/wx8TdJPgRNsX1w500TeSVkHOI7Fr/cAdp7ge2p6IeV3705gA+AVwEtst9pdou9eSbkWWRs4DDigbpwpHQlsT+ke9x7gYqDJosmB7mblhpSbfndTiota3lwy1hWmDo8dbPUz3AM+K/fABcD1wFqMb0K7D7i6WqLpuXO4ELhxfXxf9IqkrWz/gPHNnutQ1sObXjfsiMU7mre+Bh7L121QCme7tZbn2O7DZI/bbR8laVPb+0j6du1A0/B8Scf0YUJGjw3uU6/bPfbhPkPfnAScT7mfegPlOmSHqokmdifl83yfrDM46AolWy2WHFhN0tXANymf8b9suy/3zCIilpsUTc49LS9yPkC32+9bkh5GucFzCu3f9PsopRh1TNIXKKPGW3drDzvf/ZvtU2uHmIH9gE3o107mL9YOsAx+LWkf4A5J7wUeVjvQFI6kdH/aH/gG8Oy6caatT91/AX5t+521Q8zAgZSbqn+Y6gsb8kJgY9t/rB1kJiStRNnl3of39J+7ovDBZ4xWx3Mvpic/4+u6xyurppgb+lSAOHCbpIMonRu3p+FR87bPlrSAUoj6cOCDtj9cOdaoeYKk05d4bR6wRY0wM2H7RuD9kj4NvI8ytnvNuqkm9DRJaw8K8SXNA95Co11Tbd8u6fmU3717gK1s/2/lWCOr60K6R+0cM3Cf7Vu6z3B3SerDe2M729tL+obtkyW1Xph6Uu0A09XHrmbdjdRvStpuOL+koyhrXE2R9Pju8EZJe1E6sQ02AjfZNbWP74seeiZl4+GePHBDxpdrhZoO208CkLQWcIvtFBXFwI09KZiEssFhXWANSavTj06T6wCLJP2Kcr5osgFNH/VwskCfrWX7JEl7276kK7Zu1Q22T64dYoZ61R3T9o6SHkTZRLIjsG/3nvim7XdXDRcRUVGKJueeP1PG1m5BGenwurpxpsf2H4D/7P617huUBfzrgdWA7wBPrJpoal+StD+LdzW7sGKe6dgP6FPR5Nm2n9MdH1o1yfSdBryc0qH260DTXV67UTU3UcbtXkwZ/dn6GKPrbX9H0v62PyHp5bUDTWJsguM++FxXRDt8jvtkxTxT+b3to2uHmKHfAX+qHWIZrEopYu9DIftFks4ANpR0LPD92oGmqQ8/4wMkPRq4wPZ1U31x/EV6U4A4dMP9SEons+dRPt83vdHI9hldsfJ+wAm184yg3Sd4/dgVmmIZSHop8DJgJUp3iZbfy0fQo66pQ+eLgynXqM/qunk2W6DTZ5IOB95A6UQyj3IDu7kO50MWdsVla0l6E3Bt7UDTsLKkVSkFDitR1hJbdhqwNeMTHFp+P/SOpH+hdP/dvCsQh/K3ZAENFk0y3g0Tyueh/brjMdrsVhwrwNAay78CT7H9FUmvoQdry5K2Bz5C+b07S9K1tpvuWBzL1WAT1zyW2NBlu+V18COAF1M2bl0HfLxunGl5Ye0AI+z13b/BZIE1KZ83byV/q2edpM26xw0ZmhLVoB/WDrAMetcds2vK8EPK790awFNpdK0lImJFSdHk3HMCpRPihZRdBCdSdlrG7DkC+KKkbwFbAbtWzjMdzwAexHhb9jHKe6RlD5J0GYuPX2p5YeB/JP09i+dt/QbascAiSvfD7wOfpIxlbo6k3YA3UjJ/F9iYsjB+HUOFcg26u1v8XCDpubQ9tn1pi3K96KxEKWT/OWW8HLRf9HmzpOOASxnviHF83UhL1xXxjQGPBC6TdAXjmVs+JwNg+w768Xca24dLeh7lffFz25+vnWk6evIz/nfKSLyTJT2U8REl37J9d81go6KnBYjHLeW1bSlFUU2Oehw6J8+jdDi/SNJC6Mc5uQ8GnaC6zodbUwrD+2JL4EDbzXfV7WHX1OHzxW+A13bHKdBZPl4CrG/7ztpBpml/SsHZRcDtjBdwtewYyk3LdYDvdc9bdg6lgG8DSlHRItK5aDadCnwNOBz4v91r91E2zjVnqEvxqsDmti+T9GLgC3WTRSPOAD7YHd9CeX+3Xhh1JGXD2WeA91A2iqdocu4a3sTVh41bT6W8X7ehrHsfC/ye9u89QflssRuLb8posvN9D53a3dt7GuUcfCxl4lKTay0991pKkfLmwNnAq+vGmZjtw2pnWAa96o4p6VDKPd6HAV8FPg+8yXYfm2FERMyaFE3OPavaPq87/i9Jr6+aZjT9lLJw+GzgS8DVdeNMy4NtP6t2iBl6Y+0AM/QI4BDGb6z+kfZvoG1ie19Jz7D9ua4rRqsOBnboinMAkHQycG73r1UHAJtRFkDf3T22aqJFueYX6IC7bbc+Um7Ywu5x3e6x5SLPPvz/HwmS1gAeAtwIrCnppY13TO2NrghqUAj1IEpB39uAz9KPkVF90LsCxMEN92HdyJrv0mhm+vf3uc8+Q/l8/+vueR82nR0BvEHS+pSF8cttL5zie6rpU9fUpZ0vYrn6FeV6ummSnjP09JfdPygbmFsfBXuWpO9Srkdu7EEn7rVtbyvpY5Tu5l+pHWiUdJuIrpF0IGVz+KB4ZDvaLk49lVIoeRnweMqaRjaRxOqDDYi2T5fUh0L2+2zfImnM9l2S/rd2oKhnsImrR/4NeJntP0k6krLeshA4Hzhv0u+s73TKxoztKBsysj40ewbvi3t6+L7oFdtXUNbfYvnoW3fMt1EaBRxFaRaQYsmICFI0ORetLOlJtn8i6Um0XYjRV98G/tX2uZIOo4zn3qpypqlcIWkPykLioENY610QL6UUTt5/069unKWTtCWlEO5q4FPAmd1/el21UNO3sqS1KSO51qDrkNmoe4cLJgFs3yap6TFitn8r6U+UYtqm3xM9XJQbdq2kN7N458Zmb1TaPkLSevRjvNxFlG4uZ1K6/szrnn+B9gvD++ZcyiLtcIFOzIKuEO5vKbvbn0UZB/QF4DU1c42SnhYgLqYr3tqOcp5rUs//VvfNurb/T+0QM3Qi5SbUDsAN3fMdJv2OSvraNVXSvpSNcqsNXrP92HqJRtYqwE8k/aR7Ptbo+2LPCV4fo/GiSUnvAB7UdTo/S9IPhkbbtmjQdXR123+UVDXMCPss/erouYHtjwPYfp+kb9QOFE24R9KzKdch21DGwbZuoaSjgLW6De3X1g4UMQMr2b6827i1uu1LASS1fJ9h4HbbR0na1PY+kr5dO9AImeh9kbXOWdZ9rj+QobHctlu+19ArPeyOuQ5l8uTzgfdIup6yTvTFHmyUi4hYblI0OfccBJzUFWMsoh9jgfpmZ9u/AbD9/p4sym3Z/Rvowxixk+jHTb+PAu+gjJU7B3gKcBNlN0/rHcLeQhn5sh5lMfHgunEmNdFCy/wVmmKGJH2EcoGyiHJTeAzo2w34PlhA6SwxGA/b9I1KSSdSdoCuDvwVpSvN06uGmtg+lDFt6wKmvI//TCmmjNk13/betUOMqJsoIwfPBP6v7dsq5xl5fShAXIpVKddSB9UOEk24UtL6thfVDjIDa9k+SdLeti/pCpdb1deuqQcAL6Bcn8by03Lx3v1sv6J2hr/ALrb/BsD2bpIupu2f+2clvQ34saTvAHdM9Q2xTPrW0XNM0uNt/0LSJpRCz4h9gfcD/wH8jH6M2t2fkvsiyvlt37px4Y2mLAAAFbxJREFUImZk0MXseZRRsEhaAKxRLdH0jUlaF1hD0uqk0+Rsmuh9kZ/x7HshsLHt5jv1x/LXdZb8evcPSc+j3Nv5MPmsHBFzWIom555n2d66dogR99CuM8bDKaNgrqicZ0pLdv+RtEqtLDPQl5t+99j+CoCkQ2xf1R3fXjfWtGxkW5LWAW623fJOvydIOn2J1+YBW9QIMwPbAI+13Yfdtb3VwxuWWwJPoIyzPRw4u26cidk+AThB0j62T6qdZ8RdLulpwI8Y75h6T91II+P9wHOA1wKbS/qi7csqZxp1vStA7Dpa71o7RzTjGcB1km7qno/1oVuDpM26xw0Z6jTRmh53Tb3Zdro/LX+XAn9H+VvSvK7j/Rsp3RDn0Y/zxX2SVulGJi6g0c2IkoavP1aibKZcxPiN+Jhdfevo+TrgU13By28phWcxR0la2fa9wHWUUe2Djct98O+275+CIOmTwEsr5omYia92my82Anbpitg/RJnI1bojgBcDn6OcOz5eN85I6fP7om9+Rz4bR0fSVpT1rGcAmwE/Bk4G0qghIua0FE3OPc+XdIztPoye6Kv/AF4BnMD4CLTPV000BUmvAl7P+CjYPzHeka1ZPbnpN1wMd9fQcZM3HZbwSuA02zdN+ZX17T7B6613pllIudl351RfGMuuGzMwGPG4JvBL25vXTTWp39sek7S67Zt7cDMK4MLuhvD9I8Vt96FjQ5/sALxo6PkYkJGfs8D2UcBRkh5KKZ58jaQnAj+1vU/ddKMpBYjRd7Y3rZ1huiRt2E1CeC3lRt/mlA0ZuSE1SyS9pztcRdKXKEV9gw0Oh1cLNrrOpRTG/bp73nrRyx6Uz8Z9uub7KHBFNwJ9M9rtMrkVsBplw/Il9KuDdR/1qqOn7e9Rpr1EQJn2sxdlQsbg78agcLLJ62pJBwJvBdaU9A/dy/MoHTIjesH20ZLOA261vagrjjve9jm1s01E0lMp9/W2Adam3F/4PXBhzVyjpI/vi74ZajDySOAySVcwfo26V7VgUdt7KRPYjgQua7xRTkTECpOiyblnHWCRpF9RPiCN2c4o2Flme6GkMds3Sfrf2nmm4UBgR8pCzFnAIVXTTM/gpt9fU276HVA3zoQGHRDnLXHcegdEgAdJuoyyoHgftHtB1eNuNI8CrpW0sHuec/JyYHu9wbGkjYF31kszLT+UdBjl7/WZlBuBrTsdOIcybncRGacy62xvWTvDHLAxZTHxwcA9LL7xISICSW+1fWQ3WWCxxeVWPycDX5S0s+0rgG0lzQPeQunAdUzdaCPDSzzG8jXfdp86YfwK6Ns4vF8Bf0spJLra9s2V8yyV7Sd3G132Bt5EKWY41fbCyb8zZqJvHT0lnW1716HNkwNjtjeolSvqGvqc9jbbp1YNM022Pwx8WNLhtt8z5TdENMr2z4eOrwaurhhnOv4NeJntP0k6kjJCeiGlOcp5VZONkB6+L/rmWcButUNEW2w/q3aGiIgWpWhy7nlh7QBzwC1d58bVJe0B/KF2oGlYZPt6SWvY/qakd9QONBFJW1J2wdwIvJ3SIWVT4IlAi6M0hzsgHjvBcaveC/xP7RAjbs/aAeYa29cOutS2yvbhkh5M6U77d8D3KkeajtttHyVpU9v7SPp27UCjRtIulE0Og26ea9l+ct1Uo0HSF4AnUT5HfBV4h+0r66aKiEZ9rnvsw2f5gXcB50t6JuVvyKmUwvCnVk01QmyfDCBpdeDhlCkI+1G6WsXsu1zS04AfMd4t5Z66kSa1CvCTrmsjlMKtVousB46wvT3Q/NSJriD8TQCStqd0D9/I9tPrJhspfevo+b9doecFtYNEk/ajvJf75EOS3g1sQJlmdXmKwyOWq5VsXy5pfWB125cCSMrG2uiTn/a40UhERMQKlaLJuWcBZXfJ/eMzgYzPnF3/AhwO3ExZWPyXunGm5VZJLwbGuoLPtWsHmsRHgXdQRuyeQxm1cxNlMfSUirmWqucXJofZ3q52iFEk6e2T/Od3rbAgc8QS3aDWoxRdN0vSVsDLGe8w+f8BrY8IHpO0LrBGVzCQTpOz70jKZ7b9gW8Az64bZ6QcBXzX9r0AktaRNN92FsQjYjG2f9wdfgA4idLR7LaKkaZk+2xJC4CvUAr6Pth1LorZdzblenVXyujM44HnVk00mnYAXjT0vNnRqp1WR1tPZkzSOSw+daLZUfOS1gD+gbIpcXX6VxDVtB529Hwq4++DS7rXWi/0jBWnN1N1hpxI6XC3I3BD93yHmoEiRtygk/LzKBtr6a6n1qiWKGLmHitpqV2KW/5cHxERUUOKJueejM9cTiQ9fujp8OiatYFbVnCcmdoXeBzwZuBQ4KC6cSZ1j+2vAEg62PZV3fHtdWONpFskHcziC4lfrhtpZAyK9l5MGX12MbA1ZVx3zJLBCE1KN6gNgN9Sujf+oGqwqX0U+BBlMbwvjqAUd54C/JIGi9hHwPW2vyNpf9ufkPTy2oFGhe2LACTtSPkMdxvwMEn7DT5zREQs4QXAPwNfk/RT4ATbF1fONCHbZ0haidJd6YTaeUbYapRupIfYfqmkjL5aPva1/f3aIWbgUkr3+FVrB5mBk6b+kvok7Q7sAWwMfAbY3/Y1VUONqD519LS9Zc+KPGPFemPtAMtgLdsnSdrb9iWS5tcOFDHivirpYmAjYBdJm1DWaT9VN1bEjNxJua8XERERU0jR5NyT8ZnLz3FDx4OOZutQRkc3uTguaSvbPwC27V5aB/gSsEpXOHCZ7Vtr5ZvAcNenu4aOs2A0+34P/HX3D8r7OkWTs8D2cQCS/tH2q7uXT5OU4pzZtTNwpO1vSfq67Z1rB5qm2wZjHvvC9oWSfgQ8GtjEdgrZZ9/d3Q3KBZKeS9tdofvqSGA724skbQB8ltKZLSJiMbZvBN4v6dPA+yiFcmvWTbV0Qx235wGbABdJWgi96KzUN6sABwM/lLQFpdNZzL5DJT2a0kXuVNt/qJxnKudSNi3/uns+NsnXtuI0yqa+4Sk1LToTuBL4MfAk4D2SgJzfloc+dfTsU5FnrBjd5pGVKH+nX0I5t60EfIGydtQ0SZt1jxsC91aOEzHSbB8t6Tzg1m59aBPgeNvn1M4WMQM39O3+QkRERC0pmpx7lhyfmUX8WWJ7p8GxpG2A1wBbUEZmtOqZlI5rey7lvy0ANgf+ZoUmmtoTJJ1OWdwaPt6ibqzRY/sVw88lrVcrywhbU9Imtq9Wubvz0NqBRsy8CY6bJOk53eGtkg4Hfkh3U7X1Lq+S/hF4K+Wz5acljXVdPuMvNNTJ+mhgS+C9wP8D3l0t1Oj6s+1FALZ/K+muqb4hIuYmSS8FXka52X4i8IrJv6OqYyc4jtl3KKWT/JGUTqQH140zmmzvIenhwF7AWZJ+R+n2+s26ySY03/betUPM0DmUNaENKOe5RcAZVRMt3U5Tf0n8pfra0bNPRZ6xQuwDHA6sS+m8NQ/4M3BRzVBTkfQQ4LXAxynr9GcDr570myLiL2b750PHVwNXV4wTsSx+WDtAREREX6Rocu45grKIfwpl9GAfdrj3gqRVKAtxrwbuAR4CPMb2H6sGm4Tto7vHpd7kk/SmFZtoWnYfOs4NwOVI0ruAAygdU1YDfgE8oWqo0XMIcI6kR1BGRx9SOc+oGZvguFWDAvZbKV2KN+2e96HL6+uBpwMXUAoFftA9xl/uuCWe7wY8gnLDJGbXbZIOoozw2x64pXKeiGjXlsCBtq+sHWQqtr9VO8Nc0Y3MfDjwSuAXtv+7dqYR9kjgUZTO2z8DdpW0b6PFiZdLehrwI8Y3RN1TN9KU1ra9raSPAQfRaOftnN9WmF519OxrkWcsX7ZPAE6QtI/tk2rnmQ5Jr6FsyLgXOMj2BZUjRURET9g+rHaGiIiIvkjR5BzTjc+8i9IF8SbK4lHMjmsoO+/3tn2VpPNbLpgcJunNwBuBOyk7bcdsr2/7vXWTPVAWxVeoXYANgWMoXc0+UjfO6JD0KdsvsX2RpE/afn/3+tfpwVigHvkbSZfQdaMdOh6z/X/qRnugSQrY+9Dl9T7bd3cdJsck3VE70KgY7mQ9IGk+8F3gnSs80Aga6uZ5JHAY8DzgetruHBcRdR0BvEHS+sDngcttL6ycKSqTdBRl08tFwMskbW/70MqxRo6k71HWLk4A3g5sZHuhpC/VTTahHYAXDT0fAx5bKct03dk9rm77j4MCuZiz+tbRs1dFnrHCXditgy+grA+tb/tVlTNNZC9AlMYMp1A2qUZERERERMQsStHkHDHUBfFA4G560AWxh/4d+Cfg0d1u/OZHwQ7Zg7JIdOeUXxlzyQ1dEdQa3U2oVWoHGiGPGDp+PvD+7rhP540+eHLtAMuip11evy3pdGBDSccC6ay0nEhaCdiOnC9m05LdPAG2pYxVPWIFZ4mIfjgROJ9SDHVD93yHqomiBdvb/lsASR+kbHCI2be37asGT7rPoNvYfm7FTJPZ1/b3a4eYoc9KehvwY0nfAbIhag7r4eblvhV5xop1OnAO5Zp6EfDgunEmdVfXmfjmrMlGREREREQsHymanDuuoXRB/Ke+dUHsC9vvA94naQdgX2BrSUcDp9i+om66Kf0KyPshlvRrSfsAd0h6L/Cw2oFG1HDhUx9GSPeG7WtrZ1hGvenyKuml3eGVwHXAZcBdwG3VQo2+VSljEg+qHWRUTNHNM0WTEbE0a9k+SdLe3Ujm+bUDRRMWSJpv+z667ua1A42i4YLJTusbSQ6V9GjgVOBU23+onGdCkoZH1q4E3EcpKvpTnUQRM9fDIs9YsW63fZSkTW3vI+nbtQNNU+t/6yIiIiIiInopRZNzR5+7IPZKtzj3LUkPA/6ZMj7jKXVTTWkV4CeSfkJ3Yycja+Y2SU8GbgJ2Ay4GbqWMhYnZMTbBcQTA9T3q8rr50PGelK4N84A+jBTvJdt3ALvWzjHK0s0zIqZD0mbd44bAvZXjRBs+BVws6bvA0ygjYmP5a/p6yvYekh5OuZ4+S9LvgBNsf7NusqXaitLp/lTgEvJZKCJGz5ikdYE1JK1O250mn9B1U543dAxk3T4iIiIiImK2pGhyjuh5F8Re6roH/Gf3r3VH1w4Q7ZC0G/BG4FhKl62Ngf0oXeR+VjHaKFnawuc8YIu6saIRvxnq8noUDXd5tf3mwbGkp9s+vGaeiFmSbp4RsVSSNrT9G+C1wMcpmwfOphTLRXwe+BKwGXBi1lpml6QzeGCB5DzgsRXizNQjgUcBa1OuqXeVtK/tvevGWpztJ0t6IrA38CbgQkp3zIV1k0VE/OUkPYQySeDFlCYHv+weW7X70PGx1VJERERERESMsBRNzjE97YIYy9+llCK59Sk3ei6vGycqOxjYoetmBoCkk4Fzu3/xl5to4TOLoAHwBuAhwFnAy+lPl9emu/xETFe6eUbEJL4oaeeuGG5bSfOAtwCvA46pGy0acKLt7YAUSy4fE10rNX0NJel7wJ3ACcDbgY26bvJfqpts6brz25sAJG0PHCVpI9tPr5ssImLZSXoNcCilO/hBti8AzqubanIZNR8REREREbH8pWhyjupZF8RY/k4Czgd2AG4ATuyOY266d7hgEsD2bZL+XCvQqMnCZ0zhc90Nd8jf6YiIiJa8Czhf0jOBBZQRtvcAT62aKlpxh6RjAAP3Adg+vm6k0dHja6i9bV81eNJNGdjG9nMrZpqUpDWAfwD2BFannOsiIvpsL0CUDaqnABfUjRMREREREREtSNFkRACsZfskSXvbvkTS/NqBoqr7Jng974uIFeMWSQez+A33L9eNtHRDYxKHR80DYLsvHTIjIiKmxfbZkhYAXwEeDnzQ9ocrx4p2XNI9PrJ7TBfuYLhgsjOvSpBpkLQ7sAewMfAZYH/b11QNFRExO+6yfQ9ws6RVaoeJiIiIiIiINqRoMiIAkLRZ97ghZVRJzF2LFT515gFb1AgTMQf9HngBsCXlhuW1QJNFk2S8fEREzDG2z5C0ErAfZdxuxMDHaweIXmi5mPZM4Ergx8CTgPdIArIhKiJGSrPF6xEREREREbFipWgyIgBeS7nB89fA2cABdeNEZbtP8HoKoiKWI0lbAB+yvbOkK4E1gA2Bj9RNNrEej0mMiIiYsSU6LG8CXCRpIaSgKAD4FOX9MR94DHAVsF3VRFHN0Pli2DzgsRXiTNdOtQNERCwngw3imZIRERERERER90vRZMQcJmlL4EjgRuDtlJs8mwJPBC6rGC0qShFURDVHA2/ojq+3vZOkxwEfo4zHi4iIiLrSYTkmZHvbwbGkhwHHV4wT9U10jmj23JG1gIgYYcMbxJs9D0dERERERMSKlaLJiLnto8A7gDWBc4CnADcBFwCnVMwVETEXrWb7B93xrQC2F0rK57WIiIgGpKAoZuBW2u4oGMtZzhcREe3IOTkiIiIiIiKWJjfhI+a2e2x/BUDSwbav6o5vrxsrImJO+qvBge0XD73+pwpZIiIiImIGJH2H8fHt6wBfqZsoIiIiIiIiIiIiIiaSosmIue2+oeO7ho7nr+ggERHBbyVtY/u/By9I2ga4oWKmiIiIiJiePYaO77J9Y7UkERERERERERERETGpFE1GzG1PkHQ6pRPG8PEWdWNFRMxJbwDOk/Q1YCFlpOMzgRdVTRURERER0/Fn4BjK9fQvJL3O9jV1I0VERERERERERETE0qRoMmJu233o+NgJjiMiYgWw/auus+SLgMcAPwDeZvuOuskiIiIiYhpOAD4KXAjsCJxI2QATEREREREREREREY2ZNzY2VjtDREREREREREREb0n6hu2dhp5faHv7mpkiIiIiIiIiIiIiYunm1w4QERERERERERHRcytLehJA95hdyhERERERERERERGNStFkRERERERERETEMpC0YXd4EHCSpN9SRnN/tl6qiIiIiIiIiIiIiJhMiiYjIiIiIiIiIiKWzRclrW37R7a3BjYEzgMOrZwrIiIiIiIiIiIiIiawcu0AERERERERERERPfUu4HxJzwQWAKcC9wBPrZoqIiIiIiIiIiIiIiY0b2xsrHaGiIiIiIiIiIiIXpK0J3AI8HDgg7Y/XDlSREREREREREREREwi47kjIiIiIiIiIiKWke0zgP8ErgdOqBwnIiIiIiIiIiIiIqaQTpMRERERERERERHLQNIZwBgwD3gGsAhYCGB7r4rRIiIiIiIiIiIiImICK9cOEBERERERERER0VPHTnAcEREREREREREREY1Kp8mIiIiIiIiIiIiIiIiIiIiIiIiImBPm1w4QEREREREREREREREREREREREREbEipGgyIiIiIiIiIiIiIiIiIiIiIiIiIuaEFE1GRERERERERERERERERERERERExJyQosmIiIiIiIiIiIiIiIiIiIiIiIiImBNSNBkRERERERERERERERERERERERERc8L/D9KLsjVvdWeOAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 3600x3600 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import seaborn as sns\n",
    "# Need to uncomment below for the word and char n-gram (merged layers)\n",
    "# pred = np.argmax(model,axis=1)\n",
    "con_mat_df =  confusion_matrix(y_ts, pred)\n",
    "\n",
    "###CCAT50###\n",
    "# con_mat_df = pd.DataFrame(con_mat_df, index = [i for i in target_names_50],\n",
    "#                                               columns =[i for i in target_names_50])\n",
    "# plt.figure(figsize=(50,50))\n",
    "\n",
    "###CCAT10###\n",
    "con_mat_df = pd.DataFrame(con_mat_df, index = [i for i in target_names_10],\n",
    "                                              columns =[i for i in target_names_10])\n",
    "plt.figure(figsize=(10,10))\n",
    "sns.heatmap(con_mat_df, annot=True, cmap=plt.cm.Blues)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 457,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "\"                   precision    recall  f1-score   support\\n\\n    AaronPressman       0.81      0.88      0.85        50\\n       AlanCrosby       0.96      0.54      0.69        50\\n   AlexanderSmith       0.78      0.58      0.67        50\\n  BenjaminKangLim       0.57      0.16      0.25        50\\n    BernardHickey       0.76      0.62      0.68        50\\n      BradDorfman       0.55      0.92      0.69        50\\n DarrenSchuettler       0.48      0.22      0.30        50\\n      DavidLawder       0.60      0.30      0.40        50\\n    EdnaFernandes       0.76      0.62      0.68        50\\n      EricAuchard       0.61      0.56      0.58        50\\n   FumikoFujisaki       0.94      0.96      0.95        50\\n   GrahamEarnshaw       0.80      0.88      0.84        50\\n HeatherScoffield       0.35      0.46      0.40        50\\n       JanLopatka       0.67      0.76      0.71        50\\n    JaneMacartney       0.23      0.28      0.25        50\\n     JimGilchrist       0.94      1.00      0.97        50\\n   JoWinterbottom       0.93      0.80      0.86        50\\n         JoeOrtiz       0.59      0.64      0.62        50\\n     JohnMastrini       0.59      0.74      0.65        50\\n     JonathanBirt       0.83      0.86      0.84        50\\n      KarlPenhaul       0.85      1.00      0.92        50\\n        KeithWeir       0.74      0.86      0.80        50\\n   KevinDrawbaugh       0.78      0.70      0.74        50\\n    KevinMorrison       0.70      0.74      0.72        50\\n    KirstinRidley       0.71      0.70      0.71        50\\nKouroshKarimkhany       0.86      0.96      0.91        50\\n        LydiaZajc       1.00      0.72      0.84        50\\n   LynneO'Donnell       0.91      0.80      0.85        50\\n  LynnleyBrowning       0.92      0.98      0.95        50\\n  MarcelMichelson       0.74      0.62      0.67        50\\n     MarkBendeich       0.88      0.88      0.88        50\\n       MartinWolk       0.83      0.76      0.79        50\\n     MatthewBunce       0.98      0.94      0.96        50\\n    MichaelConnor       0.92      0.92      0.92        50\\n       MureDickie       0.33      0.38      0.35        50\\n        NickLouth       0.75      0.86      0.80        50\\n  PatriciaCommins       0.76      0.64      0.70        50\\n    PeterHumphrey       0.52      0.74      0.61        50\\n       PierreTran       0.76      0.76      0.76        50\\n       RobinSidel       0.90      0.88      0.89        50\\n     RogerFillion       0.87      0.82      0.85        50\\n      SamuelPerry       0.62      0.76      0.68        50\\n     SarahDavison       0.60      0.66      0.63        50\\n      ScottHillis       0.28      0.46      0.35        50\\n      SimonCowell       0.72      0.92      0.81        50\\n         TanEeLyn       0.59      0.44      0.51        50\\n   TheresePoletti       0.87      0.78      0.82        50\\n       TimFarrand       0.82      0.72      0.77        50\\n       ToddNissen       0.54      0.62      0.58        50\\n     WilliamKazer       0.49      0.44      0.46        50\\n\\n         accuracy                           0.70      2500\\n        macro avg       0.72      0.70      0.70      2500\\n     weighted avg       0.72      0.70      0.70      2500\\n\""
      ]
     },
     "execution_count": 457,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.metrics import classification_report\n",
    "classification_report(y_ts, pred, target_names=target_names_10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 436,
   "metadata": {},
   "outputs": [],
   "source": [
    "w50_3gram_hist = hist\n",
    "w50_3gram_pred = pred\n",
    "w50_3gram_acc = acc\n",
    "w50_3gram_model = model"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Experiment Results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 311,
   "metadata": {},
   "outputs": [],
   "source": [
    "c_4gram_results = [c_4gram_hist, c_4gram_pred, c_4gram_acc, 0.765999972820282] \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 318,
   "metadata": {},
   "outputs": [],
   "source": [
    "w_2gram_results = [w_2gram_hist, w_2gram_pred, w_2gram_acc, 0.7540000081062317]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 328,
   "metadata": {},
   "outputs": [],
   "source": [
    "w_3gram_results = [w_3gram_hist, w_3gram_pred, w_3gram_acc,  0.7379999756813049] "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 338,
   "metadata": {},
   "outputs": [],
   "source": [
    "c_2gram_results = [c_2gram_hist, c_2gram_pred, c_2gram_acc, 0.722000002861023]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 355,
   "metadata": {},
   "outputs": [],
   "source": [
    "c_3gram_results = [c_3gram_hist, c_3gram_pred, c_3gram_acc, 0.7599999904632568]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 366,
   "metadata": {},
   "outputs": [],
   "source": [
    "w_4gram_results = [c_4gram_hist, c_4gram_pred, c_4gram_acc, 0.7599999904632568]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 411,
   "metadata": {},
   "outputs": [],
   "source": [
    "cw_gram_results = [cw_gram_hist, cw_gram_pred, cw_gram_acc, 0.7720000147819519]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 431,
   "metadata": {},
   "outputs": [],
   "source": [
    "w50_2gram_results = [w50_2gram_hist, w50_2gram_pred, w50_2gram_acc, 0.6611999869346619 ]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 437,
   "metadata": {},
   "outputs": [],
   "source": [
    "w50_3gram_hist = [w50_3gram_hist, w50_3gram_pred, w50_3gram_acc,w50_3gram_model, 0.6416000127792358] "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 454,
   "metadata": {},
   "outputs": [],
   "source": [
    "w50_unigram = [0.6535999774932861]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 462,
   "metadata": {},
   "outputs": [],
   "source": [
    "c50_4gram = [0.704800009727478]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "cw50_2_4gram = []"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}