cicy3o_nn_logh21.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Machine Learning for Complete Intersection Calabi-Yau Manifolds\n",
    "\n",
    "We then separately consider another ML approach to predict the **Hodge numbers** from the configuration matrix of CICY 3-folds: we implement **neural networks** (NN) in order to improve previous predictions using different kinds of architectures. In particular we will use **convolutional** NN (CNN) to process the configuration matrix using kernel convolutions (as if we were considering a \"computer vision\" task). We try different architectures such as a CNN with only convolutional hidden layers, as opposed to a CNN which uses a **fully connected** (FC) network to process the output.\n",
    "\n",
    "In this notebook we try to implement an [_inception-like_](https://arxiv.org/abs/1409.4842) network to predict a logarithmic estimate of the Hodge number $h_{21}$. We will use the [Tensorflow](https://www.tensorflow.org/) backend to implement NNs on the GPU, using the [Keras](https://www.tensorflow.org/api_docs/python/tf/keras) high level API."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Infrastructure\n",
    "\n",
    "We print information about the current OS:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Current OS:                  Linux (kernel release: 5.6.12-arch1-1, architecture: x86_64)\n",
      "Number of available threads: 8\n",
      "Current CPU frequency:       2830 MHz (max: 3800 MHz)\n",
      "Available RAM memory:        6983 MB (tot: 15758 MB)\n",
      "Mon May 18 09:51:03 2020       \n",
      "+-----------------------------------------------------------------------------+\n",
      "| NVIDIA-SMI 440.82       Driver Version: 440.82       CUDA Version: 10.2     |\n",
      "|-------------------------------+----------------------+----------------------+\n",
      "| GPU  Name        Persistence-M| Bus-Id        Disp.A | Volatile Uncorr. ECC |\n",
      "| Fan  Temp  Perf  Pwr:Usage/Cap|         Memory-Usage | GPU-Util  Compute M. |\n",
      "|===============================+======================+======================|\n",
      "|   0  GeForce 940MX       Off  | 00000000:02:00.0 Off |                  N/A |\n",
      "| N/A   69C    P5    N/A /  N/A |      5MiB /  2004MiB |      0%      Default |\n",
      "+-------------------------------+----------------------+----------------------+\n",
      "                                                                               \n",
      "+-----------------------------------------------------------------------------+\n",
      "| Processes:                                                       GPU Memory |\n",
      "|  GPU       PID   Type   Process name                             Usage      |\n",
      "|=============================================================================|\n",
      "|    0   1400850      G   /usr/lib/Xorg                                  4MiB |\n",
      "+-----------------------------------------------------------------------------+\n"
     ]
    }
   ],
   "source": [
    "from mltools.libos import InfoOS\n",
    "\n",
    "print('Current OS:                  {} (kernel release: {}, architecture: {})'.format(InfoOS().os, InfoOS().kernel, InfoOS().arch))\n",
    "print('Number of available threads: {:d}'.format(InfoOS().threads))\n",
    "print('Current CPU frequency:       {:.0f} MHz (max: {:.0f} MHz)'.format(InfoOS().freq, InfoOS().freqm))\n",
    "print('Available RAM memory:        {:d} MB (tot: {:d} MB)'.format(InfoOS().vmav, InfoOS().vmtot))\n",
    "\n",
    "# print info on GPU\n",
    "!nvidia-smi"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Setup\n",
    "\n",
    "We import the Python modules we use and print their versions to keep track of changes."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Python version: 3.7\n",
      "Matplot version: 3.2.1\n",
      "Numpy version: 1.18.4\n",
      "Pandas version: 1.0.3\n",
      "Scikit-learn version: 0.22.2.post1\n",
      "Tensorflow version: 2.0.0\n",
      "Keras version: 2.2.4-tf (backend: tensorflow)\n"
     ]
    }
   ],
   "source": [
    "import json\n",
    "import sys\n",
    "\n",
    "import matplotlib as mpl\n",
    "import random     as rnd\n",
    "import sklearn    as skl\n",
    "import numpy      as np\n",
    "import pandas     as pd\n",
    "import tensorflow as tf\n",
    "\n",
    "from tensorflow       import keras\n",
    "from tensorflow.keras import backend as K\n",
    "\n",
    "import warnings\n",
    "warnings.simplefilter(action='ignore', category=UserWarning) # ignore user warnings: nothing that I can really do anything about it...\n",
    "\n",
    "\n",
    "%matplotlib inline\n",
    "mpl.rc('axes', labelsize=12)\n",
    "mpl.rc('xtick', labelsize=12)\n",
    "mpl.rc('ytick', labelsize=12)\n",
    "\n",
    "# print the version of the modules\n",
    "print('Python version: {:d}.{:d}'      .format(sys.version_info.major, sys.version_info.minor))\n",
    "print('Matplot version: {}'            .format(mpl.__version__))\n",
    "print('Numpy version: {}'              .format(np.__version__))\n",
    "print('Pandas version: {}'             .format(pd.__version__))\n",
    "print('Scikit-learn version: {}'       .format(skl.__version__))\n",
    "print('Tensorflow version: {}'         .format(tf.__version__))\n",
    "print('Keras version: {} (backend: {})'.format(keras.__version__, K.backend()))\n",
    "\n",
    "# fix random_seed\n",
    "RAND = 42\n",
    "rnd.seed(RAND)\n",
    "np.random.seed(RAND)\n",
    "tf.random.set_seed(RAND)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Session Preparation\n",
    "\n",
    "in order to save the results of the analysis, we define where to store images, log files and models:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "from os import path, makedirs\n",
    "\n",
    "ROOT_DIR = '.' #-------------------------------------------------- root directory\n",
    "IMG_DIR  = 'img' #------------------------------------------------ directory of images\n",
    "MOD_DIR  = 'models' #--------------------------------------------- directory of saved models\n",
    "LOG_DIR  = 'log' #------------------------------------------------ directory of logs\n",
    "OUT_DIR  = 'output' #--------------------------------------------- directory of outputs, data, csv files, etc.\n",
    "\n",
    "DB_NAME = 'cicy3o' #---------------------------------------------- name of the dataset\n",
    "DB_FILE = DB_NAME + '_analysis.h5' #------------------------------ full name with extension\n",
    "DB_PATH = path.join(ROOT_DIR, DB_FILE) #-------------------------- full path of the dataset\n",
    "DB_DIR  = 'original' if DB_NAME == 'cicy3o' else 'favourable' #--- subdir where to store images, models, logs\n",
    "\n",
    "# define full paths\n",
    "IMG_PATH = path.join(ROOT_DIR, IMG_DIR, DB_DIR)\n",
    "MOD_PATH = path.join(ROOT_DIR, MOD_DIR, DB_DIR)\n",
    "LOG_PATH = path.join(ROOT_DIR, LOG_DIR, DB_DIR)\n",
    "OUT_PATH = path.join(ROOT_DIR, OUT_DIR, DB_DIR)\n",
    "\n",
    "# create directories if non existent\n",
    "if not path.isdir(IMG_PATH):\n",
    "    makedirs(IMG_PATH, exist_ok=True)\n",
    "if not path.isdir(MOD_PATH):\n",
    "    makedirs(MOD_PATH, exist_ok=True)\n",
    "if not path.isdir(LOG_PATH):\n",
    "    makedirs(LOG_PATH, exist_ok=True)\n",
    "if not path.isdir(OUT_PATH):\n",
    "    makedirs(OUT_PATH, exist_ok=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We also create a log file to store debug and related information:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Rotating existing logs...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2020-05-18 09:51:06,479: INFO ==> New logging session started. Log is at ./log/original/cicy3o_nn_logh21.log.\n"
     ]
    }
   ],
   "source": [
    "import logging\n",
    "\n",
    "from mltools.liblog import create_logfile\n",
    "\n",
    "path_to_log = path.join(LOG_PATH,\n",
    "                        DB_NAME + '_nn_logh21.log'\n",
    "                       ) #----------------------------------------------------- path to the log file\n",
    "log = create_logfile(path_to_log,\n",
    "                     name=DB_NAME + '_nn_logh21',\n",
    "                     level=logging.DEBUG\n",
    "                    ) #-------------------------------------------------------- create a log file and a log session\n",
    "\n",
    "# these lines provide the same setup also for the Jupyter logging\n",
    "logger = logging.getLogger() #------------------------------------------------- get the current logging session\n",
    "\n",
    "fmt = logging.Formatter('%(asctime)s: %(levelname)s ==> %(message)s') #-------- customise the formatting options\n",
    "\n",
    "handler = logging.StreamHandler() #-------------------------------------------- handle the stream to the default (stderr)\n",
    "handler.setLevel(logging.DEBUG) #---------------------------------------------- print everything\n",
    "handler.setFormatter(fmt) #---------------------------------------------------- set the formatting options\n",
    "\n",
    "logger.handlers = [handler] #-------------------------------------------------- override the default stream\n",
    "\n",
    "# we are ready to go!\n",
    "log.info('New logging session started. Log is at {}.'.format(path_to_log))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We finally set the _memory growth_ property of the GPU in order to avoid overflowing its RAM memory:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "GPU setup: 1 physical GPUs, 1 logical GPUs.\n"
     ]
    }
   ],
   "source": [
    "gpus = tf.config.experimental.list_physical_devices('GPU') #--------------------------------------- list of physical GPUs\n",
    "\n",
    "if gpus: #----------------------------------------------------------------------------------------- set memory growth only if GPU is active\n",
    "    try:\n",
    "        for gpu in gpus:\n",
    "            tf.config.experimental.set_memory_growth(gpu, True) #---------------------------------- set memory growth\n",
    "            \n",
    "        logical_gpus = tf.config.experimental.list_logical_devices('GPU') #------------------------ list of logical devices\n",
    "        print('GPU setup: {:d} physical GPUs, {:d} logical GPUs.'.format(len(gpus),\n",
    "                                                                         len(logical_gpus)\n",
    "                                                                        )\n",
    "             )\n",
    "    except RuntimeError as e:\n",
    "        print(e)\n",
    "else:\n",
    "    print('No GPUs in the setup!')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Loading the Dataset\n",
    "\n",
    "We first load the dataset we built during the preanalysis."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2020-05-18 09:51:07,081: DEBUG ==> Database loaded.\n",
      "2020-05-18 09:51:07,082: INFO ==> Shape is 7851 rows x 7 columns.\n"
     ]
    }
   ],
   "source": [
    "import pandas as pd\n",
    "\n",
    "# load the dataset\n",
    "if path.isfile(DB_PATH):\n",
    "    df = pd.read_hdf(DB_PATH)\n",
    "    log.debug('Database loaded.')\n",
    "    log.info('Shape is {:d} rows x {:d} columns.'.format(df.shape[0], df.shape[1]))\n",
    "else:\n",
    "    log.error('Cannot load database from {}!'.format(DB_PATH))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We print the `dtypes` and the name of the keys inside the dataframe as a reference:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "h11            int16\n",
       "h21            int16\n",
       "num_cp          int8\n",
       "dim_cp        object\n",
       "dim_h0_amb    object\n",
       "matrix        object\n",
       "pca           object\n",
       "dtype: object"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.dtypes"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Dense Format Extraction\n",
    "\n",
    "We now extract the matrix from the available features and the labels. Notice that in this case we do not flatten the output as it will be processed by _Conv2D_ layers in _Keras_: we reshape it as if it was a black and white image, that is we take each matrix of shape (width, height) and output a matrix of shape (width, height, 1):"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2020-05-18 09:51:08,438: DEBUG ==> Matrix shape: (7851, 12, 15, 1)\n"
     ]
    }
   ],
   "source": [
    "from mltools.libtransformer import ExtractTensor\n",
    "\n",
    "# extract the labels\n",
    "h21    = df['h21'].values\n",
    "logh21 = np.log1p(h21) #------------------------------------------------------------- extract the log value\n",
    "\n",
    "# we also extract num_cp for the plots (not for training)\n",
    "num_cp = np.array(ExtractTensor(flatten=True).fit_transform(df['num_cp']))\n",
    "\n",
    "# extract the tensor features\n",
    "matrix = np.array(ExtractTensor(flatten=False).fit_transform(df['matrix'])) #-------- old shape is (batch, width, height)\n",
    "matrix = np.reshape(matrix, (-1, np.shape(matrix)[1], np.shape(matrix)[2], 1)) #----- new shape is (batch, width, height, 1)\n",
    "\n",
    "log.debug('Matrix shape: {}'.format(np.shape(matrix)))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Training and Validation Strategy\n",
    "\n",
    "We now define the validation strategy and split the dataset into training, validation and test sets.\n",
    "\n",
    "We will take out 10% of the dataset to be our **test set**. For the remaining 90% of the dataset, we will use **holdout validation** with $\\frac{1}{9}$ of the training samples as validation (effectively we use 80% of the total set for training and 10% as validation)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2020-05-18 09:51:08,514: DEBUG ==> Train set size: 6280\n",
      "2020-05-18 09:51:08,515: DEBUG ==> Validation set size: 785\n",
      "2020-05-18 09:51:08,516: DEBUG ==> Test set size: 786\n"
     ]
    }
   ],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "# divide into training and test sets\n",
    "matrix_train, matrix_test, \\\n",
    "num_cp_train, num_cp_test, \\\n",
    "h21_train, h21_test = train_test_split(matrix, num_cp, logh21, test_size=0.1, shuffle=False)\n",
    "\n",
    "# keep a holdout validation set\n",
    "matrix_train, matrix_val, \\\n",
    "num_cp_train, num_cp_val, \\\n",
    "h21_train, h21_val = train_test_split(matrix_train, num_cp_train, h21_train, test_size=1.0/9.0, shuffle=False)\n",
    "\n",
    "log.debug('Train set size: {:d}'.format(np.shape(matrix_train)[0]))\n",
    "log.debug('Validation set size: {:d}'.format(np.shape(matrix_val)[0]))\n",
    "log.debug('Test set size: {:d}'.format(np.shape(matrix_test)[0]))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We also provide a way to visualise the predictions by comparing the predictions as function of a (scalar) base feature (e.g.: `num_cp`) with respect to the ground truth:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_counts(base_feature, label):\n",
    "    '''\n",
    "    Returns unique values and counts of the label as a function of the base_feature.\n",
    "    \n",
    "    Required arguments:\n",
    "        base_feature: the feature considered as base for the comparison,\n",
    "        label:        the label we are interested in comparing.\n",
    "    \n",
    "    Yields:\n",
    "        [ unique value of base_feature, unique value of label, count ].\n",
    "    '''\n",
    "    \n",
    "    for n in np.sort(np.unique(base_feature)):\n",
    "        uniques, counts = np.unique(label[np.argwhere(base_feature == n)[:,0]], return_counts=True)\n",
    "        \n",
    "        for u, c in np.c_[uniques, counts]:\n",
    "            yield [n, u, c]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Notice that the use of `num_cp` will only be necessary to show in a simple way how well the algorithm can predict $h_{11}$ and $h_{21}$. `num_cp` will be taken as a representative of the matrices in the test set and $h_{11}$ and $h_{21}$ will be shown in a 2D plot."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Feature Rescaling\n",
    "\n",
    "Before proceeding we rescale the input matrices to have entries in $\\left[ 0, 1 \\right]$: we divide each entry of each matrix by the range of variation of its entries (since they are all positive, we divide by the maximum) in the **training** set. We then apply the **same scaling** to validation and test sets."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "# compute the scale factor\n",
    "scale_factor = np.max(matrix_train) - np.min(matrix_train)\n",
    "\n",
    "# rescale the sets\n",
    "matrix_train = matrix_train / scale_factor\n",
    "matrix_val   = matrix_val   / scale_factor\n",
    "matrix_test  = matrix_test  / scale_factor\n",
    "\n",
    "# define the shape of the tensors (batch axis excluded) since we will use it later one\n",
    "input_shape  = np.shape(matrix_train)[1:]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Architecture Building\n",
    "\n",
    "We then move to the problem of building the correct architecture. We will first implement a function which will allow us to build different models using similar structure blocks: the idea is to first process the images through convolutional layers, then add a FC network at the end (we will also add dropout and normalization layers)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "from tensorflow.keras.models       import Model\n",
    "from tensorflow.keras.layers       import Input, Conv2D, Dense, LeakyReLU, Dropout, ZeroPadding2D, BatchNormalization, Flatten, concatenate\n",
    "from tensorflow.keras.regularizers import l1_l2\n",
    "from tensorflow.keras.optimizers   import Adam\n",
    "from tensorflow.keras.utils        import model_to_dot\n",
    "\n",
    "def cnn_model(input_shape,\n",
    "              model_name='cnn_model',\n",
    "              learning_rate=0.01,\n",
    "              conv_layers=[32],\n",
    "              conv_paddings=['same'],\n",
    "              conv_kernels=[(2,2)],\n",
    "              conv_alpha=0.0,\n",
    "              fc_layers=[],\n",
    "              fc_alpha=0.0,\n",
    "              dropout=0.2,\n",
    "              full_dropout=0.0,\n",
    "              normalization=0.99,\n",
    "              last_relu=True,\n",
    "              out_name='output',\n",
    "              l1_reg=0.0,\n",
    "              l2_reg=0.0\n",
    "             ):\n",
    "    '''\n",
    "    Build a CNN model.\n",
    "    \n",
    "    Required arguments:\n",
    "        input_size:    the size of the input tensor.\n",
    "        \n",
    "    Optional arguments:\n",
    "        model_name:    the name of the model,\n",
    "        learning_rate: the learning rate of the gradient descent,\n",
    "        conv_layers:   a list-like object with the no. of filters for each hiddend convolution layer,\n",
    "        conv_paddings: a list-like object with the corresponding padding for each layer,\n",
    "        conv_kernels:  a list-like object with the kernel size of each layer,\n",
    "        conv_alpha:    the slope of the LeakyReLU activation (ReLU if 0.0) of the convolution layers,\n",
    "        fc_layers:     a list-like object with the no. of units for each hidden dense layer,\n",
    "        fc_alpha:      the slope of the LeakyReLU activation (ReLU if 0.0) of the FC network,\n",
    "        dropout:       the dropout rate (do not use dropout if <= 0.0),\n",
    "        full_dropout:  use this dropout rate after every layer (disabled if <= 0.0),\n",
    "        normalization: the momentum of the batch normalization (do not use normalization if <= 0.0),\n",
    "        last_relu:     whether to use ReLU activation in the output layer (force positive output),\n",
    "        out_name:      the name of the output layer,\n",
    "        l1_reg:        the L1 kernel regularization factor,\n",
    "        l2_reg:        the L2 kernel regularization factor.\n",
    "        \n",
    "    Returns:\n",
    "        the compiled model.\n",
    "    '''\n",
    "    \n",
    "    # define the regularizer\n",
    "    regularizer = l1_l2(l1=l1_reg, l2=l2_reg) #--------------------------------------------- regularizer\n",
    " \n",
    "    # build the model\n",
    "    I = Input(shape=input_shape, name=model_name + '_input') #------------------------------ input layer\n",
    "    x = I\n",
    "    \n",
    "    # build convolutional layers\n",
    "    for n in range(np.shape(conv_layers)[0]): #--------------------------------------------- loop through the conv. layers\n",
    "        x = Conv2D(filters=conv_layers[n],\n",
    "                   kernel_size=conv_kernels[n],\n",
    "                   padding=conv_paddings[n],\n",
    "                   kernel_regularizer=regularizer,\n",
    "                   name=model_name + '_conv2d_' + str(n)\n",
    "                  )(x) #-------------------------------------------------------------------- add conv. layer\n",
    "        x = LeakyReLU(alpha=conv_alpha,\n",
    "                      name=model_name + '_conv2d_' + str(n) + '_activation'\n",
    "                     )(x) #----------------------------------------------------------------- add activation\n",
    "        \n",
    "        if normalization > 0.0:\n",
    "            x = BatchNormalization(momentum=normalization,\n",
    "                                   name=model_name + '_conv2d_' + str(n) + '_normalization'\n",
    "                                  )(x) #---------------------------------------------------- add batch normalization (if requested)\n",
    "        if full_dropout > 0.0:\n",
    "            x = Dropout(rate=full_dropout,\n",
    "                        name=model_name + '_conv2d_' + str(n) + '_full_dropout'\n",
    "                       )(x) #--------------------------------------------------------------- add dropout (if requested)\n",
    "        \n",
    "    # add dropout\n",
    "    if dropout > 0.0 and full_dropout <= 0.0:\n",
    "        x = Dropout(rate=dropout,\n",
    "                    name=model_name + '_dropout'\n",
    "                   )(x) #------------------------------------------------------------------- add dropout (if requested)\n",
    "        \n",
    "    # flatten the output\n",
    "    x = Flatten(name=model_name + '_flatten')(x) #------------------------------------------ flatten the output\n",
    "    \n",
    "    # build FC network\n",
    "    for n in range(np.shape(fc_layers)[0]):\n",
    "        x = Dense(units=fc_layers[n],\n",
    "                  kernel_regularizer=regularizer,\n",
    "                  name=model_name + '_fc_' + str(n)\n",
    "                 )(x) #--------------------------------------------------------------------- add dense layers\n",
    "        x = LeakyReLU(alpha=fc_alpha,\n",
    "                      name=model_name + '_fc_' + str(n) + '_activation'\n",
    "                     )(x) #----------------------------------------------------------------- add activation\n",
    "        \n",
    "        if normalization > 0.0:\n",
    "            x = BatchNormalization(momentum=normalization,\n",
    "                                   name=model_name + '_fc_' + str(n) + '_normalization'\n",
    "                                  )(x) #---------------------------------------------------- add batch normalization (if requested)\n",
    "        if full_dropout > 0.0:\n",
    "            x = Dropout(rate=full_dropout,\n",
    "                        name=model_name + '_fc_' + str(n) + '_full_dropout'\n",
    "                       )(x) #--------------------------------------------------------------- add dropout (if requested)\n",
    "        \n",
    "    if last_relu: #------------------------------------------------------------------------- output layer\n",
    "        F = Dense(1, activation='relu', name=model_name + '_' + out_name)(x)\n",
    "    else:\n",
    "        F = Dense(1, name=model_name + '_' + out_name)(x)\n",
    "    \n",
    "    # define the model\n",
    "    model = Model(inputs=I, outputs=F, name=model_name)\n",
    "    \n",
    "    # compile the model\n",
    "    model.compile(optimizer=Adam(learning_rate=learning_rate),\n",
    "                  loss='mean_squared_error',\n",
    "                  metrics=['mean_squared_error']\n",
    "                 )\n",
    "    \n",
    "    # return the compiled model\n",
    "    return model\n",
    "\n",
    "def scan_inception_model(input_shape,\n",
    "                         model_name='inception_model',\n",
    "                         learning_rate=0.01,\n",
    "                         conv_layers=[32],\n",
    "                         conv_padding='same',\n",
    "                         conv_alpha=0.0,\n",
    "                         fc_layers=[],\n",
    "                         fc_alpha=0.0,\n",
    "                         dropout=0.2,\n",
    "                         full_dropout=0.0,\n",
    "                         normalization=0.99,\n",
    "                         last_relu=True,\n",
    "                         out_name='output',\n",
    "                         l1_reg=0.0,\n",
    "                         l2_reg=0.0\n",
    "                        ):\n",
    "    '''\n",
    "    Build a CNN 'scan-inception' model: scan over rows and columns and merge.\n",
    "    \n",
    "    Required arguments:\n",
    "        input_size:    the size of the input tensor.\n",
    "        \n",
    "    Optional arguments:\n",
    "        model_name:    the name of the model,\n",
    "        learning_rate: the learning rate of the gradient descent,\n",
    "        conv_layers:   a list-like object with the no. of filters for each 'inception' modules,\n",
    "        conv_padding:  the padding to use for the convolutional scans,\n",
    "        conv_alpha:    the slope of the LeakyReLU activation (ReLU if 0.0) of the convolution layers,\n",
    "        fc_layers:     a list-like object with the no. of units for each hidden dense layer,\n",
    "        fc_alpha:      the slope of the LeakyReLU activation (ReLU if 0.0) of the FC network,\n",
    "        dropout:       the dropout rate (do not use dropout if <= 0.0),\n",
    "        full_dropout:  use this dropout rate after every layer (disabled if <= 0.0),\n",
    "        normalization: the momentum of the batch normalization (do not use normalization if <= 0.0),\n",
    "        last_relu:     whether to use ReLU activation in the output layer (force positive output),\n",
    "        out_name:      the name of the output layer,\n",
    "        l1_reg:        the L1 kernel regularization factor,\n",
    "        l2_reg:        the L2 kernel regularization factor.\n",
    "        \n",
    "    Returns:\n",
    "        the compiled model.\n",
    "    '''\n",
    "    \n",
    "    # define the regularizer\n",
    "    regularizer = l1_l2(l1=l1_reg, l2=l2_reg) #--------------------------------------------- regularizer\n",
    " \n",
    "    # build the model\n",
    "    I = Input(shape=input_shape, name=model_name + '_input') #------------------------------ input layer\n",
    "    if conv_padding == 'same':\n",
    "        x = I\n",
    "    else:\n",
    "        x = ZeroPadding2D(padding=((0,3),(0,0)), data_format='channels_last')(I)\n",
    "    \n",
    "    # build convolutional layers\n",
    "    for n in range(np.shape(conv_layers)[0]): #--------------------------------------------- loop through the conv. layers\n",
    "        a = Conv2D(filters=conv_layers[n],\n",
    "                   kernel_size=(x.shape[1],1),\n",
    "                   padding=conv_padding,\n",
    "                   kernel_regularizer=regularizer,\n",
    "                   name=model_name + '_conv2d_rows_' + str(n)\n",
    "                  )(x) #-------------------------------------------------------------------- add conv. layer over rows\n",
    "        a = LeakyReLU(alpha=conv_alpha,\n",
    "                      name=model_name + '_conv2d_rows_' + str(n) + '_activation'\n",
    "                     )(a) #----------------------------------------------------------------- add activation\n",
    "        b = Conv2D(filters=conv_layers[n],\n",
    "                   kernel_size=(1,x.shape[2]),\n",
    "                   padding=conv_padding,\n",
    "                   kernel_regularizer=regularizer,\n",
    "                   name=model_name + '_conv2d_columns_' + str(n)\n",
    "                  )(x) #-------------------------------------------------------------------- add conv. layer over columns\n",
    "        b = LeakyReLU(alpha=conv_alpha,\n",
    "                      name=model_name + '_conv2d_columns_' + str(n) + '_activation'\n",
    "                     )(b) #----------------------------------------------------------------- add activation\n",
    "        \n",
    "        x = concatenate([a, b],\n",
    "                        name=model_name + '_concatenation_' + str(n)\n",
    "                       ) if conv_padding == 'same' \\\n",
    "                         else concatenate([a, tf.einsum('bij...->bji...',b)], #------------- swap columns and rows\n",
    "                                          name=model_name + '_concatenation_' + str(n)\n",
    "                                         ) #------------------------------------------------ concatenate layers\n",
    "        \n",
    "        if normalization > 0.0:\n",
    "            x = BatchNormalization(momentum=normalization,\n",
    "                                   name=model_name + '_conv2d_' + str(n) + '_normalization'\n",
    "                                  )(x) #---------------------------------------------------- add batch normalization (if requested)\n",
    "        if full_dropout > 0.0:\n",
    "            x = Dropout(rate=full_dropout,\n",
    "                        name=model_name + '_conv2d_' + str(n) + '_full_dropout'\n",
    "                       )(x) #--------------------------------------------------------------- add dropout (if requested)\n",
    "        \n",
    "    # add dropout\n",
    "    if dropout > 0.0 and full_dropout <= 0.0:\n",
    "        x = Dropout(rate=dropout,\n",
    "                    name=model_name + '_dropout'\n",
    "                   )(x) #------------------------------------------------------------------- add dropout (if requested)\n",
    "        \n",
    "    # flatten the output\n",
    "    x = Flatten(name=model_name + '_flatten')(x) #------------------------------------------ flatten the output\n",
    "    \n",
    "    # build FC network\n",
    "    for n in range(np.shape(fc_layers)[0]):\n",
    "        x = Dense(units=fc_layers[n],\n",
    "                  kernel_regularizer=regularizer,\n",
    "                  name=model_name + '_fc_' + str(n)\n",
    "                 )(x) #--------------------------------------------------------------------- add dense layers\n",
    "        x = LeakyReLU(alpha=fc_alpha,\n",
    "                      name=model_name + '_fc_' + str(n) + '_activation'\n",
    "                     )(x) #----------------------------------------------------------------- add activation\n",
    "        \n",
    "        if normalization > 0.0:\n",
    "            x = BatchNormalization(momentum=normalization,\n",
    "                                   name=model_name + '_fc_' + str(n) + '_normalization'\n",
    "                                  )(x) #---------------------------------------------------- add batch normalization (if requested)\n",
    "        if full_dropout > 0.0:\n",
    "            x = Dropout(rate=full_dropout,\n",
    "                        name=model_name + '_fc_' + str(n) + '_full_dropout'\n",
    "                       )(x) #--------------------------------------------------------------- add dropout (if requested)\n",
    "        \n",
    "    if last_relu: #------------------------------------------------------------------------- output layer\n",
    "        F = Dense(1, activation='relu', name=model_name + '_' + out_name)(x)\n",
    "    else:\n",
    "        F = Dense(1, name=model_name + '_' + out_name)(x)\n",
    "    \n",
    "    # define the model\n",
    "    model = Model(inputs=I, outputs=F, name=model_name)\n",
    "    \n",
    "    # compile the model\n",
    "    model.compile(optimizer=Adam(learning_rate=learning_rate),\n",
    "                  loss='mean_squared_error',\n",
    "                  metrics=['mean_squared_error']\n",
    "                 )\n",
    "    \n",
    "    # return the compiled model\n",
    "    return model"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We then define a function which contains all fit information:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "from tensorflow.keras.backend   import clear_session\n",
    "from tensorflow.keras.callbacks import EarlyStopping, ModelCheckpoint, ReduceLROnPlateau\n",
    "from mltools.libscore           import Score\n",
    "\n",
    "def cnn_fit(model,\n",
    "            X,\n",
    "            y,\n",
    "            X_val,\n",
    "            y_val,\n",
    "            batch_size=32,\n",
    "            epochs=100,\n",
    "            early_stopping=10,\n",
    "            reduce_lr=5,\n",
    "            verbose=0\n",
    "           ):\n",
    "    '''\n",
    "    Fit the model.\n",
    "    \n",
    "    Required arguments:\n",
    "        model:          the model to fit,\n",
    "        X:              the training features,\n",
    "        y:              the training labels,\n",
    "        X_val:          the validation features,\n",
    "        y_val:          the validation labels.\n",
    "        \n",
    "    Optional arguments:\n",
    "        batch_size:     the size of the batch used in forward pass,\n",
    "        epochs:         the number of epochs,\n",
    "        early_stopping: patience of early stopping,\n",
    "        reduce_lr:      patience to reduce learning rate,\n",
    "        verbose:        verbosity level (debug)\n",
    "    \n",
    "    Returns:\n",
    "        the fitted model, the history of the model.\n",
    "    '''\n",
    "    \n",
    "    # clear the TF graph\n",
    "    clear_session()\n",
    "    \n",
    "    # define callbacks\n",
    "    callbacks = [EarlyStopping(monitor='val_mean_squared_error',\n",
    "                               patience=early_stopping,\n",
    "                               verbose=verbose,\n",
    "                               restore_best_weights=True\n",
    "                              ),\n",
    "                 ReduceLROnPlateau(monitor='val_mean_squared_error',\n",
    "                                   factor=0.3,\n",
    "                                   patience=reduce_lr,\n",
    "                                   verbose=verbose\n",
    "                                  ),\n",
    "                 ModelCheckpoint(filepath=path.join(MOD_PATH, model.name + '.h5'),\n",
    "                                 monitor='val_mean_squared_error',\n",
    "                                 verbose=verbose,\n",
    "                                 save_best_only=True,\n",
    "                                 save_format='h5'\n",
    "                                )\n",
    "                ]\n",
    "    \n",
    "    # fit the model\n",
    "    history = model.fit(x=X,\n",
    "                        y=y,\n",
    "                        batch_size=batch_size,\n",
    "                        epochs=epochs,\n",
    "                        verbose=verbose,\n",
    "                        callbacks=callbacks,\n",
    "                        validation_data=(X_val, y_val)\n",
    "                       )\n",
    "    \n",
    "    # return the fitted model and its history\n",
    "    return model, history.history"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The test results will then be saved to file along with other quantities which can be reused later on."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "test_results = {'h21_true': h21_test.tolist(),\n",
    "                'num_cp':   num_cp_test.tolist(),\n",
    "                'matrix':   matrix_test.tolist()\n",
    "               } #--------- dictionary with the predictions"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Analysis for $h_{21}$\n",
    "\n",
    "We can then follow the same line of ideas and implement models of neural networks to predict $h_{21}$. In general they will have to be deeper with respect to the previous models since the dependences on the features are more complicated in this case."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2020-05-18 09:51:11,763: DEBUG ==> Training model h21_large_kernel_same_conv_no_fc...\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"h21_large_kernel_same_conv_no_fc\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "h21_large_kernel_same_conv_n [(None, 12, 15, 1)]       0         \n",
      "_________________________________________________________________\n",
      "h21_large_kernel_same_conv_n (None, 12, 15, 250)       9250      \n",
      "_________________________________________________________________\n",
      "h21_large_kernel_same_conv_n (None, 12, 15, 250)       0         \n",
      "_________________________________________________________________\n",
      "h21_large_kernel_same_conv_n (None, 12, 15, 250)       1000      \n",
      "_________________________________________________________________\n",
      "h21_large_kernel_same_conv_n (None, 12, 15, 150)       1350150   \n",
      "_________________________________________________________________\n",
      "h21_large_kernel_same_conv_n (None, 12, 15, 150)       0         \n",
      "_________________________________________________________________\n",
      "h21_large_kernel_same_conv_n (None, 12, 15, 150)       600       \n",
      "_________________________________________________________________\n",
      "h21_large_kernel_same_conv_n (None, 12, 15, 100)       540100    \n",
      "_________________________________________________________________\n",
      "h21_large_kernel_same_conv_n (None, 12, 15, 100)       0         \n",
      "_________________________________________________________________\n",
      "h21_large_kernel_same_conv_n (None, 12, 15, 100)       400       \n",
      "_________________________________________________________________\n",
      "h21_large_kernel_same_conv_n (None, 12, 15, 50)        180050    \n",
      "_________________________________________________________________\n",
      "h21_large_kernel_same_conv_n (None, 12, 15, 50)        0         \n",
      "_________________________________________________________________\n",
      "h21_large_kernel_same_conv_n (None, 12, 15, 50)        200       \n",
      "_________________________________________________________________\n",
      "h21_large_kernel_same_conv_n (None, 12, 15, 50)        0         \n",
      "_________________________________________________________________\n",
      "h21_large_kernel_same_conv_n (None, 9000)              0         \n",
      "_________________________________________________________________\n",
      "h21_large_kernel_same_conv_n (None, 1)                 9001      \n",
      "=================================================================\n",
      "Total params: 2,090,751\n",
      "Trainable params: 2,089,651\n",
      "Non-trainable params: 1,100\n",
      "_________________________________________________________________\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2020-05-18 16:51:08,834: DEBUG ==> Model h21_large_kernel_same_conv_no_fc has been trained.\n",
      "2020-05-18 16:51:11,803: DEBUG ==> History of h21_large_kernel_same_conv_no_fc has been saved to ./output/original/h21_large_kernel_same_conv_no_fc.json.\n",
      "2020-05-18 16:51:13,794: DEBUG ==> History of h21_large_kernel_same_conv_no_fc has been saved to ./output/original/convnet_test_results_h11_h21.json.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy of h21_large_kernel_same_conv_no_fc on the validation set: 26.879%\n",
      "Accuracy of h21_large_kernel_same_conv_no_fc on the test set: 27.354%\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 2160x360 with 5 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2020-05-18 16:51:18,408: DEBUG ==> Plot saved to ./img/original/h21_large_kernel_same_conv_no_fc.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"logh21_inception_same_conv_no_fc\"\n",
      "__________________________________________________________________________________________________\n",
      "Layer (type)                    Output Shape         Param #     Connected to                     \n",
      "==================================================================================================\n",
      "logh21_inception_same_conv_no_f [(None, 12, 15, 1)]  0                                            \n",
      "__________________________________________________________________________________________________\n",
      "logh21_inception_same_conv_no_f (None, 12, 15, 128)  1664        logh21_inception_same_conv_no_fc_\n",
      "__________________________________________________________________________________________________\n",
      "logh21_inception_same_conv_no_f (None, 12, 15, 128)  2048        logh21_inception_same_conv_no_fc_\n",
      "__________________________________________________________________________________________________\n",
      "logh21_inception_same_conv_no_f (None, 12, 15, 128)  0           logh21_inception_same_conv_no_fc_\n",
      "__________________________________________________________________________________________________\n",
      "logh21_inception_same_conv_no_f (None, 12, 15, 128)  0           logh21_inception_same_conv_no_fc_\n",
      "__________________________________________________________________________________________________\n",
      "logh21_inception_same_conv_no_f (None, 12, 15, 256)  0           logh21_inception_same_conv_no_fc_\n",
      "                                                                 logh21_inception_same_conv_no_fc_\n",
      "__________________________________________________________________________________________________\n",
      "logh21_inception_same_conv_no_f (None, 12, 15, 256)  1024        logh21_inception_same_conv_no_fc_\n",
      "__________________________________________________________________________________________________\n",
      "logh21_inception_same_conv_no_f (None, 12, 15, 128)  393344      logh21_inception_same_conv_no_fc_\n",
      "__________________________________________________________________________________________________\n",
      "logh21_inception_same_conv_no_f (None, 12, 15, 128)  491648      logh21_inception_same_conv_no_fc_\n",
      "__________________________________________________________________________________________________\n",
      "logh21_inception_same_conv_no_f (None, 12, 15, 128)  0           logh21_inception_same_conv_no_fc_\n",
      "__________________________________________________________________________________________________\n",
      "logh21_inception_same_conv_no_f (None, 12, 15, 128)  0           logh21_inception_same_conv_no_fc_\n",
      "__________________________________________________________________________________________________\n",
      "logh21_inception_same_conv_no_f (None, 12, 15, 256)  0           logh21_inception_same_conv_no_fc_\n",
      "                                                                 logh21_inception_same_conv_no_fc_\n",
      "__________________________________________________________________________________________________\n",
      "logh21_inception_same_conv_no_f (None, 12, 15, 256)  1024        logh21_inception_same_conv_no_fc_\n",
      "__________________________________________________________________________________________________\n",
      "logh21_inception_same_conv_no_f (None, 12, 15, 32)   98336       logh21_inception_same_conv_no_fc_\n",
      "__________________________________________________________________________________________________\n",
      "logh21_inception_same_conv_no_f (None, 12, 15, 32)   122912      logh21_inception_same_conv_no_fc_\n",
      "__________________________________________________________________________________________________\n",
      "logh21_inception_same_conv_no_f (None, 12, 15, 32)   0           logh21_inception_same_conv_no_fc_\n",
      "__________________________________________________________________________________________________\n",
      "logh21_inception_same_conv_no_f (None, 12, 15, 32)   0           logh21_inception_same_conv_no_fc_\n",
      "__________________________________________________________________________________________________\n",
      "logh21_inception_same_conv_no_f (None, 12, 15, 64)   0           logh21_inception_same_conv_no_fc_\n",
      "                                                                 logh21_inception_same_conv_no_fc_\n",
      "__________________________________________________________________________________________________\n",
      "logh21_inception_same_conv_no_f (None, 12, 15, 64)   256         logh21_inception_same_conv_no_fc_\n",
      "__________________________________________________________________________________________________\n",
      "logh21_inception_same_conv_no_f (None, 12, 15, 64)   0           logh21_inception_same_conv_no_fc_\n",
      "__________________________________________________________________________________________________\n",
      "logh21_inception_same_conv_no_f (None, 11520)        0           logh21_inception_same_conv_no_fc_\n",
      "__________________________________________________________________________________________________\n",
      "logh21_inception_same_conv_no_f (None, 1)            11521       logh21_inception_same_conv_no_fc_\n",
      "==================================================================================================\n",
      "Total params: 1,123,777\n",
      "Trainable params: 1,122,625\n",
      "Non-trainable params: 1,152\n",
      "__________________________________________________________________________________________________\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2020-05-18 16:51:19,201: DEBUG ==> Training model logh21_inception_same_conv_no_fc...\n",
      "2020-05-18 23:19:52,606: DEBUG ==> Model logh21_inception_same_conv_no_fc has been trained.\n",
      "2020-05-18 23:19:54,422: DEBUG ==> History of logh21_inception_same_conv_no_fc has been saved to ./output/original/logh21_inception_same_conv_no_fc.json.\n",
      "2020-05-18 23:19:55,823: DEBUG ==> History of logh21_inception_same_conv_no_fc has been saved to ./output/original/convnet_test_results_h11_h21.json.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy of logh21_inception_same_conv_no_fc on the validation set: 36.178%\n",
      "Accuracy of logh21_inception_same_conv_no_fc on the test set: 35.751%\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 2160x360 with 5 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2020-05-18 23:19:59,754: DEBUG ==> Plot saved to ./img/original/logh21_inception_same_conv_no_fc.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"logh21_inception_valid_conv_fc\"\n",
      "__________________________________________________________________________________________________\n",
      "Layer (type)                    Output Shape         Param #     Connected to                     \n",
      "==================================================================================================\n",
      "logh21_inception_valid_conv_fc_ [(None, 12, 15, 1)]  0                                            \n",
      "__________________________________________________________________________________________________\n",
      "zero_padding2d (ZeroPadding2D)  (None, 15, 15, 1)    0           logh21_inception_valid_conv_fc_in\n",
      "__________________________________________________________________________________________________\n",
      "logh21_inception_valid_conv_fc_ (None, 15, 1, 120)   1920        zero_padding2d[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "logh21_inception_valid_conv_fc_ (None, 1, 15, 120)   1920        zero_padding2d[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "logh21_inception_valid_conv_fc_ (None, 15, 1, 120)   0           logh21_inception_valid_conv_fc_co\n",
      "__________________________________________________________________________________________________\n",
      "logh21_inception_valid_conv_fc_ (None, 1, 15, 120)   0           logh21_inception_valid_conv_fc_co\n",
      "__________________________________________________________________________________________________\n",
      "tf_op_layer_transpose (TensorFl [(None, 1, 15, 120)] 0           logh21_inception_valid_conv_fc_co\n",
      "__________________________________________________________________________________________________\n",
      "logh21_inception_valid_conv_fc_ (None, 1, 15, 240)   0           logh21_inception_valid_conv_fc_co\n",
      "                                                                 tf_op_layer_transpose[0][0]      \n",
      "__________________________________________________________________________________________________\n",
      "logh21_inception_valid_conv_fc_ (None, 1, 15, 240)   960         logh21_inception_valid_conv_fc_co\n",
      "__________________________________________________________________________________________________\n",
      "logh21_inception_valid_conv_fc_ (None, 1, 15, 240)   0           logh21_inception_valid_conv_fc_co\n",
      "__________________________________________________________________________________________________\n",
      "logh21_inception_valid_conv_fc_ (None, 3600)         0           logh21_inception_valid_conv_fc_dr\n",
      "__________________________________________________________________________________________________\n",
      "logh21_inception_valid_conv_fc_ (None, 500)          1800500     logh21_inception_valid_conv_fc_fl\n",
      "__________________________________________________________________________________________________\n",
      "logh21_inception_valid_conv_fc_ (None, 500)          0           logh21_inception_valid_conv_fc_fc\n",
      "__________________________________________________________________________________________________\n",
      "logh21_inception_valid_conv_fc_ (None, 500)          2000        logh21_inception_valid_conv_fc_fc\n",
      "__________________________________________________________________________________________________\n",
      "logh21_inception_valid_conv_fc_ (None, 100)          50100       logh21_inception_valid_conv_fc_fc\n",
      "__________________________________________________________________________________________________\n",
      "logh21_inception_valid_conv_fc_ (None, 100)          0           logh21_inception_valid_conv_fc_fc\n",
      "__________________________________________________________________________________________________\n",
      "logh21_inception_valid_conv_fc_ (None, 100)          400         logh21_inception_valid_conv_fc_fc\n",
      "__________________________________________________________________________________________________\n",
      "logh21_inception_valid_conv_fc_ (None, 1)            101         logh21_inception_valid_conv_fc_fc\n",
      "==================================================================================================\n",
      "Total params: 1,857,901\n",
      "Trainable params: 1,856,221\n",
      "Non-trainable params: 1,680\n",
      "__________________________________________________________________________________________________\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2020-05-18 23:20:00,233: DEBUG ==> Training model logh21_inception_valid_conv_fc...\n",
      "2020-05-19 00:03:30,495: DEBUG ==> Model logh21_inception_valid_conv_fc has been trained.\n",
      "2020-05-19 00:03:30,833: DEBUG ==> History of logh21_inception_valid_conv_fc has been saved to ./output/original/logh21_inception_valid_conv_fc.json.\n",
      "2020-05-19 00:03:31,354: DEBUG ==> History of logh21_inception_valid_conv_fc has been saved to ./output/original/convnet_test_results_h11_h21.json.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy of logh21_inception_valid_conv_fc on the validation set: 17.707%\n",
      "Accuracy of logh21_inception_valid_conv_fc on the test set: 18.957%\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAACGgAAAFgCAYAAADt+909AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+j8jraAAAgAElEQVR4nOzde3wU9bnH8c+TBAwQCFHQAiLBgnINAUGxIKgoeAEUpIB3bBHFW/UcFLzUouIptVQ5omK9gfWOVFCrIqXCQQShYoFyU0CDXCo3uSThIiS/88dM4mbJJtlkk90N3/frNS+y85v9zbOzYZ7MzDO/MeccIiIiIiIiIiIiIiIiIiIiIlJ5EqIdgIiIiIiIiIiIiIiIiIiIiEh1pwINERERERERERERERERERERkUqmAg0RERERERERERERERERERGRSqYCDREREREREREREREREREREZFKpgINERERERERERERERERERERkUqmAg0RERERERERERERERERERGRSqYCDREREREREREREREREREREZFKpgINCYuZTTWzcQGvV5nZueXop1zvqwgzO93M/mVm2WZ2RyWtI8vMLqiMvssbR2nbOvg7DXM9Vf49RkJl/C6Y2SYz6xSJvkQk9ij/lbqOmMh/saiytr/yjoiIiIiIiIiIiEj8UYFGNeRfJDlgZjlmts3MpphZSmWsyznX1jk3r4wxFV64Kev7IuweYJ5zrq5z7smKdlZZF6PM7GMze7iY+ZeZ2fdmlhROf5Hc1lX1PZpZdzNbaGZ7zewHM/vMzLqUJ8YQIv27kAY0BtZWtC8RKT/lv5DiJf/lBEz5Ad9ljpldXYF+S423kvNORLe/v07lHREpEzN71sx+G+04ysvMRvo5PcfMToiBeMpd3F5VzCzdzFy4x41l7NuZWYtI9ysiIrHNz8OnRjsOERGJPjO7z8xeiHYc4TKzU/x8lhjtWERABRrVWT/nXArQCegCPBC8QGWcsIlxzYBV4b4pCttpKnCtmVnQ/GuB15xzR6o4niplZvWAvwGTgOOBJsBDwKEIrqZcvwsQ8vehPfCtc25/haISkUhQ/jtaXOQ/51xKwQR8h/9d+tNrlbXeKsg7kc45oLwjIhQpTMw2sz1+odnNZlZ4nO+cu9k590gZ+4qpkZDMrAbwONDbzwW7qnj9w8xsQSX2P8/MDvonCXea2Ttm1qiy1uevM6I3A4iIHEuilSv94sAfgwrah1Ti+uaZ2fDAeX4e/qay1ikiIhXj56gfzaxB0PxlfpF1ehn6ONfMNpe2nHPuf5xzw0tbrpj+A49/Cqazw+0njPUF3zT3nZ/P8iprnSLhUIFGNeec2wJ8BLSDwp3SaDNbAeSaWZKZNTazv5rZDjP71gKG3zazjmb2pX/S7y0gObB/K/oojab+SaUdZrbLzJ7y578CnAK87+907wneOZpZa38Hvce84d/7B61jlJmt8O9sfcvMkgPaR5vZFj/Gr8ysV/B2MLNPgPOAp/wYTivDOotsp6D+jvpMAc2ZxcVa0nYOMhPvAtE5AetLA/oCfwmYN8bMNvife7WZDSius2K2dcjvtKQ+K/t7DHAagHPuDedcnnPugHNutnNuRUBfxW7LUr6XgveG9bsQEHvI3wcgA9hgZv/rx7TVzC4s7vsQkaqh/Fe4TDzlv1KV8p0dtT3KkheoxLxT3Pb355f7O/Ap74hIgX7Oubp4xWDjgdHAi9ENKWJOwsu/xRa5hdg/xpvb/MLEFkAKMKGS1zeVY/hmABGROPZYYEG7c+6taAckIiIx51vgyoIXZtYeqBXJFUTgGOy2oHy2KCKBicQj55ymajYBWcAF/s9N8U5oPRLQtsyfXwuvSGcp8CBQEzgV+Abo47/eCNwF1AAGAYeBccHrAhKB5cATQB28E2ndi4upmBhrAOuB+/x1ng9kA6cHLLsEbyjv44E1wM1+2+nAJqCx/zod+HmI7TIPGB7GOgu3U2nbOWjeUbGWtJ1D9P088ELA65uAZUHL/NJfTwIwBMgFGhWzfQN/LvE7LanPyvwegz5XPWAX8DJwMZAW1F7itizue6nI70JZfh+A5/yY+/jx/Q6YE+19gSZNx9qE8l/c579S1lHSdxZyexTXV1C/lZp3Ard/BL8D5R1NmjSF2leeCeQD7fzXU/npb/0GeCMG7QF+AD719yGv+O85AOQA9/jLvw18D+wF5gNtA9YzFXga+MDfhy0mIA8BbYG/++vZBtznz08AxgAb/P3YNOD4Yj7baXjHIs6P6RN/vgNuBdbhjSQEcKO/X/0BeK8gFwQsf4u/fDbwCPBzYBGwz19/zWLW3xo4COT5699Txs/dKuBzfwUMLuH7C84PtwCrytIXcCnwL/8zbALGBrSl+587qZh11vK/zx4B89L8z9rB//1Z5P+O/Ad4KnD7+P22CBH/MGBBebaFJk2aNMXDRIi/+4HjgInAVn+aCBzntxWbe/220cAWP598BfQKsd6pBByLhpoPnAtsDop3FLDC3/e/BSQHtF+Gd9yxDy8vXwQ86ue+g37+e8pfNnD/n4p3A9kOvOPmBwI+0zBgAV7B4W68i4UXR/u706RJk6bqPvn7/AeAfwbMmwDc7+/D0/15x/nzv8M7TnvWP0aog3c8mO/v/3PwzvGNBaYDr/r5Yrg/79WA9XQHFvq5bhMwLESMRY4fQs3n6OMKh3eOcZ2fW54GLKD9RrzzkNnAarxRlY86xiXoOMn/fO/h5ef1wI0BfY7FO1b8i9/vKqBztL9nTdVr0gga1ddMM9uD90fx/wH/E9D2pHNuk3PuAN7w7w2dcw8753503nB1zwNDga54FxEmOucOO+emA/8Msb4z8XZodzvncp1zB51zZR0Otive3ULj/Rg+wTt4uTJgmSedc1udcz8A7wOZ/vw8vKTSxsxqOOeynHMbIrjOgu0UjuJiLWk7F+dl4JdmVlDheJ0/r5Bz7m1/PfnOq5xfh/c9lKTE77ScfQb2Xd7vMfBz7cNL6g5vG+0ws/fM7CR/kXC3ZSTiLog91O9De+BR59zHzrl8vD8EADCzVDNb4t853S7UPBGJGOW/yKwzWvmvJCX1Vd7tEat5p7TvoNi8Eyq/KO+IHDucc0uAzQSMxhfgv/22hnijU9znvcVdS9FHSz3mL/8R0BI4EfgSCH7c1JV4j4RKwzuh9SiAmdUF5gCz8HJkC+Af/nvuAC4HevptBSfYgj/H13hFHgD1nXPnBzRfDpyFt88/H/g9MBhohHeh6M2g7i4CzsDb/96DV+R2NV4hXDuO/rsf51xBoeEif5vUL8PnroNXkPC6v82uBJ4xs7aUwsxOAAb6/ZWlr1y8Y8T6eMUaI83s8tLW4+eVaf57CwwG1jrnluPl07vwLiieDfTCKxwJS0W2hYhIHLofL8dk8lOxW8GjNovNvWZ2OnAb0MV5I2H1wbu4FmmD8fJgc7xR+IYBmNmZeBed7sbLJT2ALOfc/XhFJAV3ON9WTJ+T8Io0TsXL59cBNwS0n4VXcNIAeAx4sZiRm0REJPI+B+r5o7Ym4t2A+2rQMn/AK4bPxDtOawI86JzLxbtpaav7aXSLrf57LsMr0qhP0DGhmZ2Cd9w4CS/XZeIV/0VaX7zzcx3wclsff/2/xCumuA7vBqz+wK4SjnEDvYGXoxvj3Zz3P0GjE/fHO7asj1fI8VTkP5Ycy1SgUX1d7pyr75xr5py7JegE/6aAn5sBjf0htvf4F7XuwztoaAxscc65gOU3hlhfU2CjK9+QqI2BTf4FhsD1NAl4/X3Az/vxLmzgnFsP3Im3E95uZm+aWeMIrXMT5VNcrCVt56P4F/d2AJeZ2al4yef1wGXM7DrzniFW0F87vIOfkpT4nZazz8C+y/U9BnPOrXHODXPOnezH0BjvDgQIc1tGKG4I8fvgH2S2w7sYWaAdPxVp7Mc7aTo9oL24eSISGcp/FV9n1PJfKUL2VYHtAcRk3gn5HZSSd0LlF+UdkWPLVrzRjIIdxitiaOYXIH4alOuKcM695JzLds4dwtu/djCz1IBF3nHOLfFz4Gv8VETYF/jeOfcnv3Ax2zm32G+7CbjfObc5oN9BYQ6V+3vn3A9+jr8aeMk596Xf373A2Vb0Gct/cM7tc86tAlYCs51z3zjn9uKdTOwYxrpL+9xZzrkpzrkjzrkvgb/inewL5Ukz2wvsxDvuur0sfTnn5jnn/u28wvoVeCcXe5Yx/pA3AzjnljrnPvfXmQX8OYx+A5VnW4iIxKurgYedc9udczvwiviu9dtC5d5wC8xHBRyP7AwjtlAF/7/Gy59/93PJFufc2tI6C7jgd6+f37OAPwV8XvCOj593zuXh5ZdGlP/YSUREwvMK3t/3FwJr8UZqAgrPJ90I3OUfT2Xj3dhW2g1Ii5xzM/18EXwj0dV4I7q+4ee5Xc65kgo0ngzIZ1+G8bnGO+f2OOe+A+byUz4bjvcYsH86z3rnXKhzuIXMrCnezVqj/WPWZcALFM1nC5xzH/r57BW84hCRiFGBxrEp8CTcJryhYesHTHWdc5fgDWnaJKjK+ZQQfW4CTinhxFrIE394JxCbmlng7+MpBCSPkjjnXnfOdce7gOLwqgBLU5Z1lhRzWdoDlbSdQ/kLXjK9Fu8k4raCBjNrhnf37m3ACc67o2slUFpFesjvtIx9Vtr3GIp/gDgV7+ITlL4tw/lewok7VL/N/X/XB8zriF8p6v9hsqNIR8XME5EqofwXH/mvXH2VsD3CygsxkndK6jNk3gmVX5R3RI45TfCGaQ32R7x9x2wz+8bMxoTqwMwSzWy8mW0ws338dFdvYPF2qOLrpnhDpRenGTAjoOBtDd5FqnAu3AQWsTUmoIjSOZeD9+iUwKK3bQE/HyjmdbFF4yUI9bmbAWcFFfRdDfyshL7ucM6l4t3VnAacXJa+zOwsM5trZjv8Ao+bKWNhvSvhZgAzO83M/mZm3/vf+/+Utd8g5dkWIiLxqkgu8n8uKBYvNve68AvMJwQci4SzXy5Pri5JA356JGiBkDcZOOf2+z+Gm2tFRKR8XgGuwhsx6S9BbQ2B2sDSgL/RZ/nzS1LSjVzh5pM7AvJZpzDeF+l81hgoKFIpUNpNc8lh3lggUiIVaMgSYJ+ZjTazWv6JuHZm1gXv2bNHgDvMLMnMBhL6cRdL8C5ojTezOmaWbGbdAtq34Q19V5zFeEO03mNmNczsXKAfRw9NexQzO93Mzjez4/Cej3gA7wRfacq9zgAlfaZgJW3nUP4CXIBX1fhyUFsdvIs3OwDM7AZ+upBUkpK+07L0WSnfYyAza2Vm/21mJ/uvm+INifu5v0hp2zKc7yUScWcA//bvgCjQEVgeRgwiUvWU/2I3/4XdVynbo8R4lXdEpDrx901N8B71VYR/p+t/O+dOxdvv/Jf9NIRrcGHYVXhD2V6AN4x5esEqyhDGJuDnJbRdHFT0luycC6eoOzDWrXjFAF5w3qM1TqCCReLFrKcsNgH/F/TZUpxzI0tdkXP/BsYBT/sFoqX19TreMLtN/QKPZynbd1Mg1M0Ak/HutGvpnKuHN2JUqH5z8U7wFggsvij3thARiUNFchFe8fVWKDn3lrPgPlBJ++HSlJSrS8p/O/FGBQn+vJHIuyIiUkH+6BHfApcA7wQ178Q7X9Y24G/0VOdcQbFDqP1/SXmhpHxSVtHIZ1uB4817PGcB5TOpUirQOMb5w/P0wxsS6Fu8nfQLQKpz7ke85+AOw3s28BCO3qkH99MC79lOm/3lC/weeMCvzBsV9N4f8Z7ndLG//meA61wZhtbDGw5wvP++7/Geb3tfGT53RdZZIORnKmZ9IbdzCe/JAhbiFU68F9S2Gm8IwUV4F4baA5+VFnBJ32kZ+6ys7zFQNt7zKhebWS7eBbKVeM/tLMu2LPP3EqG42xNwUcy850f/zI9ZRGKU8l/s5r9y9lXS9igtXuUdEYl7ZlbPzPriFXu96l/wD16mr5m18AsA9uEVsoUqZqsLHMIbjaI23kgKZfU34GdmdqeZHWdmdc3sLL/tWeBR80bvw8wamtllYfQd7HXgBjPL9Iv0/gdY7B9LVdQ24GQzq1nG5f8GnGZm1/oFeDX8IsLWZXz/y3j5q38Z+qqLd8fXQTM7E6+gJhyhbgaoi/e7kWNmrYCSCiqWAQPNrLaZtcAbLr9ARbeFiEisquEX5RdMSXiPmXrAz2kNgAeBVyF07q1AwX2gZcAlZna8mf0Mb0SOsnoRL3/2MrMEM2vi7/ehhAJ0/9hoGl4ur+vn8/8q+LwiIhITfg2c75zLDZzpvEftPg88YWYnAvj7/z7+ItuAE6zoYy1L8xpwgZkN9m9yO8HMMkt9V1ElHVeU5gW8x4CdYZ4WBcealJzPNuFde/u9n88z/PW+FmbsIuXnnNOkSZMmTZU84Q+XX9o8TZo0adKkqaJTqPyivKNJU/Wa8B47cgCv0GwvXqH1rUBiwDJTgXH+z3f578nFKyj8bcByl+EVGu4BRuENGfuu3/dGvNEWHNAiuF//9bnA5oDX7YB/4BU6fg+M8ecn4F3I+crvewPwPyE+X7q/zqSAeYUxBMy72e/nB7zCgJNDLY83ssiwgNfjgBdCrL8m8IHf784yfu7T/ffswCtu+QTIDNH/PGB40LzRwBel9QUM8r+XbP8zP4VXmFPsdith/buB4wLm9cAbQSMH+BR4GO/Zy0dtT7xh7mf7MXyGN0z/gvJsC02aNGmKhwkvh7qgaRyQDDyJN7Lif/yfk/33FJt78UbFW+LvQwvyV+MQ6y2SewLmJwNv4RV+rPDXtTko3gsCXo8tyBX+6wH++7LxHsPSx59/NvC1nyOe9OcF7v/T8AoyduDdufwgkOC3DQvMBcHv1aRJkyZNlTMF7/MD5if5++F0/3UyXlH7N37+WIP32JGC5V/y/3bfg/cYkCK5w18mOJ+cgzda7D4/L1wfIsZ5BB3/+PNLO64IPqYrkhfxjge/wjuGWQl09OcHH+OmE3CchPd4yb/5eXgDcHMJn7HIezVpisRkzoU7aqeIiITDzD7Eu/N6I/Bn59zU4uZFMUQREakmQuUX5R0RERERERERERGR6FOBhoiIiIiIiIiIiIiIiIiIiEglS4h2ACIiIiIiIiIiIiIiIiIiIiLVnQo0RERERERERERERERERERERCqZCjREREREREREREREREREREREKllStAOIlAYNGrj09PQK9ZGfn09CQuzXrCjOyIuXWBVn5MVLrPESJ0Qm1qVLl+50zjWMUEgx6VjKW1VF26MobY+itD2K0vb4ifJW2VV27oqHtliJo7q3xUoc1aEtVuKo7m3l7S+alLvKLha+w1iIQXHEZhyxEIPiiM04YiGGSMeh3FU21fG7VxzVJwbFEZtxxEIM1TWOULmr2hRopKen88UXX1Soj+zsbOrWrRuhiCqP4oy8eIlVcUZevMQaL3FCZGI1s40RCidmHUt5q6poexSl7VGUtkdR2h4/Ud4qu8rOXfHQFitxVPe2WImjOrTFShzVva28/UWTclfZxcJ3GAsxKI7YjCMWYlAcsRlHLMQQ6TiUu8qmOn73iqP6xKA4YjOOWIihusYRKndFvwxFREREREREREREREREREREpJpTgYaIiIiIiIiIiIiIiIiIiIhIJVOBhoiIiIiIiIiIiIiIiIiIiEglS4p2ACIisebw4cNs3ryZgwcPRjuUQvn5+SQkxEdNXTixJicnc/LJJ1OjRo1KjkpEpHoLN3fFU16pbMpbIiIVV5CH9u/fH3KfWtL+Nt7byttfVVDuEhHxVOX5vmjv+2MlhvLGUZ1yl5k9CvQAtgHXOef2RzkkEYkDVX2NKp5zRqzEEW7uUoGGiEiQzZs3U7duXdLT0zGzaIcDQF5eHomJidEOo0zKGqtzjl27drF582aaN29eBZGJiFRf4eaueMorlU15S0Sk4gryUNOmTUlKKv5UU0n723hvK29/lU25S0TkJ1V5vi8WjrdiIYbyxFGdcpeZtQN+7pw7x8xuBn4FPBXlsEQkDlT1Nap4zRmxEkd5clf0y1BERGLMwYMHOeGEE2KmOKO6MjNOOOGEmBqpREQkXil3VT7lLRGR0JSHYpNyl4jIT5Sr4kM1y13nAB/5P38EdI9iLCISR5Sz4kt5cpcKNEREiqHEVzW0nUVEIkf71MqnbSwiEpr2kbFJ34uIyE+0T4wPsfY9mdltZvaFmR0ys6lBbceb2QwzyzWzjWZ2VUBzGrDX/3kvcHwVhSwi1UCs7QulZOF+X3rEiYiIiIiIiIiIiIiIiMjRtgLjgD5AraC2p4EfgZOATOADM1vunFsF7AZS/eVSgR+qJlwREYl1KtAQEamG0tPT+eKLL2jQoEFM9CMiIlIa5S4Rkeqj2/hP2LLnQMT6a1K/Fp+NOT9i/QGkpKSQk5PD1q1bueOOO5g+ffpRy5x77rlMmDCBzp07h+xn4sSJ/PrXv6Zu3boAXHLJJbz++uvUr18/ovGWJCsri4ULF3LVVVeVvvAxwsz6Af1atGgR7VBERADYs2cPr7/+OrfcckvY7y1LbnnwwQfp0aMHF1xwQUXCBHRMFcw59w6AmXUGTi6Yb2Z1gCuAds65HGCBmb0HXAuMARYA9wIv4xV3fFbFoYuIlEu85aymTZvy6aefFs7LzMzkyJEjrFy5kv3793PjjTeyYsUKnHPUr1+fWbNmkZKSQmJiIu3bty9839ChQxkzZkyFYyoLFWiIiMQw5xzOuWiHISIiUmYFuSshQU9TFBE5Vm3Zc4Cs8ZcWmZeXl0diYmKxy5fW9vP7Z0U8xgKNGzcutjijrCZOnMiVV15ZWKDx4YcfRiq0MsvKyuL1119XgUYA59z7wPudO3e+MdqxiIiAd7HrmWeeKfZiV0l5EMqWWx5++OEKxSflchqQ55z7OmDecqAngHPu3/5jTz4FtgPXherIzEYAIwCaNm1KdnZ2uYPKzc0t93sjSXEUFQtxxEIMoDiCFRdHfn4+eXl5VRZDfn5+kde7du3imWee4aabbjpq2dJy1vvvv1+4XCi/+93vil0mOI6yys7OJisri6ZNm7JmzZoisT7xxBM0bNiQZcuWAfDVV1+RkJBAXl4etWrVYunSpUX6ysvLK3cc+fn5Zd5/q0ADYN9W+Ner0HIA+AfUIiLhcM5x8HA+yTUSKvxssKysLC6++GLOO+88Fi1axMyZM3nzzTeZPn06hw4dYsCAATz00EMAXH755WzatImDBw/ym9/8hhEjRoTsd/LkyXz77bc89thjAEydOpWlS5cyadKkUvvJysqib9++rFy5EoAJEyaQk5PD2LFj2bBhA7feeis7duygdu3aPPvss7Rt25a3336bhx56iMTERFJTU5k/f36FtovEpgXrdvD1thyu/0U6iQl6Lp7Isaq43DVt2jSmTZsW87mrVq1avPDCC7Rq1Uq5S0QkTo0ePZpmzZoVXvQaO3YsderUYeTIkVx22WXs3r2bw4cPM27cOC677LIi7w3MFwcOHGDYsGGsWbOG1q1bc+DAT6OAjBw5kn/+858cOHCAQYMG8dBDD/Hkk0+ydetWLrjgAho0aMDcuXOL3HX8+OOP89JLLwEwfPhw7rzzzsKc2b17dxYuXEiTJk149913qVWr6IjpxeWkvLw8xowZw7x58zh06BC33norN910E2PGjGHNmjVkZmZy/fXXc9ddd1XyFj825Oc7np2/gbYnJtOzjc4XikjFjBkzhg0bNpCZmcmFF17IpZdeykMPPUSjRo1YtmwZq1evDnmMU5BbcnJyQuaQYcOG0bdvXwYNGkR6ejrXXnstH3zwAYcPH+btt9+mVatW7Nixg6uuuopdu3bRpUsXZs2axdKlS0scKaO4XJabm8vgwYPZvHkzeXl5/Pa3v2XIkCGMGTOG9957j6SkJHr37s2ECROqZNtGUQqwN2jeXqAwaTjn7i1LR86554DnADp37uzqlvM61aYf9jN9+S6GnJVK4/rBT2OpeuX9HJGmOGIrBlAcwYLjSEhIKLEIojIEru/+++9nw4YNnHHGGVWSs66//nref//9cueswYMHM336dEaNGsW0adO48soreeWVV0hMTGTbtm00a9as8PO1adMm5Ocuy/ySJCQklPl3SgUaANvXwNoP4IRMaNwy2tGISJzZ9MN+xn+0lh05B2mYksyYi1vR9PjaFerzq6++YsqUKTzzzDPMnj2b9evXs2TJEpxz9O/fn/nz59OjRw9eeukljj/+eA4cOECXLl244oorOOGEE4rtc9CgQZx99tmFF7neeust7r//foCw+gk2YsQInn32WVq2bMnixYu57bbbmDt3Lg8//DAff/wxTZo0Yc+ePRXaHhK7Plm7nVVb9zGwUxPq164Z7XBEpIwO/JjHa0uyWLl1L+0ap3Ld2enUqlmxA7/g3LVu3bq4yF0LFy7klltu4ZNPPlHuEhGJU0OHDuXOO+8sLNCYNm0aH3zwAcnJycyYMYN69eqxc+dOunbtSv/+/UP2M3nyZGrXrs2KFStYsWIFnTp1Kmx79NFHSU31HiPfq1cvVqxYwR133MHjjz/OnDlzOOmkk4r0tXTpUqZMmcLChQtJSEjgrLPOomfPnqSlpbFu3TreeOMNnn/+eQYPHsxf//pXrrnmmiLvLy4nvfjii6SmpvLPf/6TQ4cO0a1bN3r37s348eOZMGECf/vb3yKyPeUnj836ilt6NKNnm5NLX1hE4sZD769i9dZ9Ee2zTeN6/K5f25Dt48ePZ+XKlYV38M6bN48lS5awcuVKmjdvDpTtGKcsOQSgQYMGfPnllzzzzDNMmDCBF154gYceeojzzz+fe++9l1mzZvHcc8+V+JkKctnixYtxzhXmsm+++YbGjRvzwQcfALB3715++OEHZsyYwdq1azGzY+V4KgeoFzSvHlD+4S8qaOOu/fzv3Cy6n94oJgo0RKTijqWc9dRTT5UrZw0aNIhhw4YxatQo3n//fV577TVeeeUVAH71q1/Ru3dvpk+fTq9evbj++utp2dKrBThw4ACZmZmF/dx7770MGTKkjFuxYlSgAXDquZDaFGqeGO1IRCTOOOcY/9Fa9h04zM/q1WJ37o+M/2gtT13VsUIjaTRr1oyuXbsCMHv2bP7+97/Tsa7+b34AACAASURBVGNHAHJycli3bh09evTgySefZMaMGQBs2rSJdevWhbw41bBhQ0499VQ+//xzWrZsyVdffUW3bt0AwuonUE5ODgsXLuSXv/xl4bxDhw4B0K1bN4YNG8bgwYMZOHBgObeExLo7LzyN7INHVJwhEmde+Xwj//hqB2m1avL3NdsAuKnnzyvUZ3Dumj17tnKXiEiBJ9rD3u+8n49rDIe2Fr9cpNsK5qeeAnf9O/y440THjh3Zvn07W7duZceOHaSlpXHKKaeQn5/Pfffdx/z580lISGDLli1s27aNhg0bFtvP/PnzufXWWwHIyMggIyOjsG3atGk899xzHDlyhP/85z+sXr26SHuwBQsWMGDAAOrUqUNiYiIDBw7k008/pX///jRv3rzwROAZZ5xBVlbWUe8vLifNnj2bFStWFD6SZe/evaxbt46aNfW3eGUoOKTXY0dFpLKceeaZhRe6oGzHOGXJIQADBgwoXOadd94BvNxU0P9FF11EWlpaifEF5jKgMJdddNFFjBo1itGjR9O3b1/OOeccjhw5QnJyMsOHD+fSSy+lb9++YW6NuPQ1kGRmLZ1z6/x5HYBVUYwJAKUuEYm0ysxZBcc7nTp1YubMmUB4Oev4448nLS2NN998k9atW1O79k83UGdmZvLNN98we/Zs5syZQ5cuXVi0aBGtW7emVq1ahUUoVU0FGgAJidDwNKjAc71E5Nh08HA+O3IO8rN6XkVyWp2afL/vAAcP51foTuSCAx/wTgaNHj2akSNHFllm3rx5zJkzh0WLFlG7dm3OPfdcDh48WGK/Q4YMYdq0abRq1YoBAwZgZmXqJykpqchztwra8/PzqV+/fpEkVvDcsGeffZbFixfzwQcfkJmZybJly8p8Z7PEj3rJNaiXXCPaYYhImFb/J5u0WjX9XFWTlVuDR2UNX3Duuvfee496VmUs5q7AZ2cqd4lIpdn7HYz197XZ2aEfrxrptoL5Y1PDjznODBo0iOnTp/P9998zdOhQAF577TV27NjB0qVLqVGjBunp6aXmneIK7b/99lsmTJjA559/ToMGDRg2bFip/ZR0Uf+4444r/DkxMbHIo1QKFJeTnHNMmjSJPn36FFl23rx5JcYi5WNmmEG+LnKJVDsl3TVclQKPocp6rFSWHBK4XGJiIkeOHAHCLzgLtfxpp53G0qVL+fDDD7n33nvp3bs3Dz74IEuWLOEf//gHb775Jk899RSffPJJWOuLVWaWhHctLRFINLNk4IhzLtfM3gEeNrPhQCZwGfCL6MUarTWLSGVRziqbIUOGcOuttzJ16tSj2lJSUhg4cCADBw4kISGBDz/8kNatW4fVf6QlRHXtIiJxLrlGAg1Tktmd+yMAu3N/5MSUZJJrRG732qdPH6ZMmUJOTg4AW7ZsYfv27ezdu5e0tDRq167N2rVr+fzzz0vta+DAgcycOZM33nijcKimsvRz0kknsX37dnbt2sWhQ4cKh86tV68ezZs35+233wa8pLl8+XIANmzYwFlnncXDDz9MgwYN2LRpU0S2h4iIVFybRnXZfeBHDvyYx+4DP9KucWQv3PXp04eXXnpJuUtERKrM0KFDefPNN5k+fTqDBg0CvHxx4oknUqNGDebOncvGjRtL7KNHjx688cYbAKxcuZIVK1YAsG/fPurUqUNqairbtm3jo48+KnxP3bp1yS7mhp8ePXowc+ZM9u/fT25uLjNmzOCcc84p8+cpLif16dOHyZMnc/jwYQC+/vprcnNzQ8YgFZdgRr5uQxaRCChtX12eY6Vwde/enWnTpgHeqEy7d+8ucflQuWzr1q3Url2ba665hlGjRvHll1+Sk5PD3r17ueSSS5g4cWLU7kiuJA8AB4AxwDX+zw/4bbcAtYDtwBvASOdcDIygodwlIuUXjzlrwIAB3HPPPUcVs3/22WeF7/3xxx9ZvXo1zZo1i3i84dIIGiIiFWBmjLm4FeM/Wsv3+w5wYkoyoy9uVaHHmwTr3bs3q1at4uyzzwa8ar9XX32Viy66iGeffZaMjAxOP/30wmHlS5KWlkabNm1YvXo1Z555JkCZ+qlRowYPPvggZ511Fs2bN6dVq1aFba+99hojR45k3LhxHD58mMGDB9OpUyfuvvtu1q1bh3OOXr160aFDhwhtERERqahruzYjISGBlVv3cmH6SVx3dnpE++/duzdr1qyJm9w1dOhQOnTooNwlIhIhTerXIn3MBxHtrzRt27YlOzubJk2a0KhRI/Ly8rj66qvp168fnTt3JjMzs0guKM7IkSMZNmwYGRkZZGZmFuadDh060LFjRzIyMjj11FMLH7cFMGLECPr27UujRo2YO3du4fxOnToxbNiwwlw4fPhwOnbsGHJY32DF5aSMjAyysrLo1KkTzjkaNmzIzJkzycjIICkpiQ4dOjBs2DDuuuuuMq1DSpdgGiZeRCLjhBNOoFu3brRr146LL76YSy+9tEh7eY6VwvW73/2OK6+8krfeeouePXvSqFEj6oYanYufcllBPizIZR9//DF33303CQkJ1KhRg8mTJ5Odnc1ll13GwYMHcc7xxBNPRDz+aHHOjQXGhmj7Abg8Eusxs35AvxYtWpS/j0gEIiLHvHjMWXXr1mX06NFHzd+wYQMjR47EOUd+fj6XXnopV1xxBQAHDhwofARLwecaP358xD9Lcay6VNJ17tzZffHFFxXqIzs7u8QvN1YozsiLl1gVZ+QVF+uaNWvCHt7IOcfBw/kk10iIaHFGgcDh12NduLEWt73NbKlzrnOkY4slx1LeqiraHkVpexRV3bdHuLkrnvJKZVPeKrvKzl3x0BYrcVT3tliJI6JtY1MLH3ESlW0ZsP7KWFfBvrGkfWp1bitvf1VFuav8ueu0Bz7imi5NePCyjAhGFb5Y+VtWccReHLEQQ7zEUZ7zfeUVC/v+4mI4dOgQiYmJJCUlsWjRIkaOHFnpI12Ud1sod5Uvdy1cv5OrXljMmyO60vXU6D6uMx72C8daHLEQg+IoWxxVmbMgNvJWcBzRyFnFxRGOcHKXRtAQEYkAM6NWzegnMBERERERERGJDG8Ejepxc5uIyHfffcfgwYPJz8+nZs2aPP/889EOSSLNv29QqUtE4l11z1kq0BAREYlH+Xnw434g+lW+IiIiIiIi1VGCGfm6yCUi1UTLli3517/+Fe0wpBKZHnIiItVEdc9ZCdEOQERERMrh88nwwX9D9vfRjkRERERERKRa8go0VKEhIiLxxaHcJSISy1SgIZXi41X/4alP1vH93oPRDkVEpHo6/lQ4Ph1q1I52JCIiIiIiItWSmYaJFxGR+GEaQENEJC7oESdSKd5b9h+2/7CbLlv28rPU5GiHIyJS/bS6BJqcA8l6xImIiIiIiEhl0AgaIiJSVcysH9CvRYsWFe9MqUtEJKZpBA2pFHf0asmQLk05++cnRDsUERERERERERGRsCUY5Osil4iIVAHn3PvOuRGpqanl7kMDaIiIxAcVaEilOP1ndenTthF1jtMgLSLRNm/ePPr27QvAe++9x/jx40Muu2fPHp555pnC11u3bmXQoEGVHqOEb+7a7Twzdz25h45EOxQRkYhT7hIRiXNPtIexqUWmxEeOP2pemdueaF/i6oJzQbgmTpzI/v37y/3+AvPmzWPhwoUV7kdiR4IZTiNoiEiUpKSkACUf45x77rl88cUXJfYTnOcuueQS9uzZU+H4xo4dy4QJEyrcj0SeMpeIVLV4yFlmxvr16wvnPfHEE5hZYUwvvfQS7du3p2PHjrRr1453330XgGHDhtG8eXMyMzPJzMzkF7/4RYXj0dVzEZE4lZeXR2JiYljv6d+/P/379w/ZXnBi85ZbbgGgcePGTJ8+vUJxSuX4els263fkkHPoiIrhRCRuKHeJiBwj9n4HY/cWmVVSDii17ZHjS1xdcC4I18SJE7nmmmuoXbt2ud5fYN68eaSkpETkhJ3EBjPTCBoiEnUVPcYJznMffvhhpEKTGGOmMTREJLpiOWe1b9+eN998kwceeACA6dOn06ZNGwA2b97Mo48+ypdffklKSgoHDhxgx44dhe/94x//GNEbwjSChohIJOzOgk1LvH8jICsri1atWnH99deTkZHB4MGD2b9/P+np6Tz88MN0796dt99+m9mzZ3P22WfTqVMnfvnLX5KTkwPArFmzaNWqFd27d+edd94p7Hfq1KncdtttAGzbto0BAwbQoUMHOnTowMKFCxkzZgwbNmwgMzOTu+++m6ysLNq1awfAwYMHueGGGworCOfOnVvY58CBA7noooto2bIlo0ePBrwTqcOGDaNdu3a0b9+eJ554IiLbRjy/7t6ch/q146R6ydEORUTi1cF98MM33r8REJy7Bg0aFBe5q1WrVtxzzz2AcpeISLwKzgUAEyZMoEuXLmRkZPC73/0OgNzcXC699FI6depEu3bteOutt3jyySfZunUr5513Huedd16xfbdp04aMjIzCvnfs2MEVV1xBly5d6NKlC5999hlZWVk8++yzPPHEE2RmZvLpp59W3QaQSuM94kQVGiJScaNHjy4y2tPYsWP505/+RE5ODr169aJTp060b9++8G7dQIHHOAcOHGDo0KFkZGQwZMgQDhw4ULjcyJEj6dy5M23btmXs2LEAxea59PR0du7cCcDjjz9Ou3btaNeuHRMnTixcX+vWrbnxxhtp27YtvXv3LrKe4ixbtoyuXbuSkZHBgAED2L17NwCTJk0qzKNDhw4F4P/+7/8K70Lu2LEj2dnZ5dmkUgKlLhGpiKrOWQXHa5MmTarUnHX55ZcXxvzNN9+QmppKw4YNAdi+fTt169YtHAkkJSWF5s2bl38jlkK33IqIVNSKabB0KlgCuHw44wbI+GWFu/3qq6948cUX6datGzfccENhQkxOTmbBggXs3LmTgQMHMmfOHOrUqcMf/vAHHn/8ce655x5uvPFGPvnkE1q0aMGQIUOK7f+OO+6gZ8+ezJgxg7y8PHJychg/fjwrV65k2bJlgJfcCjz99NMA/Pvf/2bt2rX07t2br7/+GvAOwv71r39x3HHHcfrpp/Ob3/yG7du3s2XLFlauXAkQkWGo5CdJiQnUraU0LiLltGkxzH8MXB5YIpx7L5xyVoW7Dcxdv/rVr+IidyUlJdGmTRtuv/125S4RkTgVnAtmz57N+vXrWbJkCc45+vfvz/z589mxYweNGzfmvffeIzExkb1795Kamsrjjz/O3LlzadCgAXl5eYX9/vDDD8yYMYO1a9diZuzatQuA3/zmN9x11110796d7777jj59+rBmzRpuvvlmUlJSGDVqVFS2g0Se94iTaEchIhH30Rj4/t+R7fNn7eHi0I9mHDp0KHfeeWfhaE/Tpk1j1qxZJCcnM2PGDOrVq8fOnTvp2rUr/fv3DzkSwuTJk6lduzYrVqxgxYoVdOrUqbDt0Ucf5fjjjycvL4/zzz+fFStWcMcddxTJc4GWLl3KlClTWLx4Mc45zjrrLHr27ElaWhrr1q3jjTfe4Pnnn2fw4MH89a9/5Zprrgn5+a677jomTZpEz549efDBB3nooYeYOHEijz32GN9++y3HHXdc4fHVhAkTePrpp+nWrRs5OTkkJ+vmo0jRABoi1dAxkLN69erFihUruP3225k4cWKl5ax69erRtGlTVq5cybvvvsuQIUOYMmUKAB06dOCkk06iefPmnH/++VxxxRX069ev8L13330348aNA6Bt27a89tprZdzYxdMIGiIiFbE7yyvOqN0AUk7y/l06JSIjaTRt2pRu3boBcNVVV7FgwQKAwotWn3/+OatXr6Zbt25kZmby8ssvs3HjRtauXUvz5s1p2bIlZhby4OmTTz5h5MiRACQmJpKamlpiPAsWLODaa68FoFWrVjRr1qzwIlevXr1ITU0lOTmZ1q1bs3HjRk499VS++eYbbr/9dmbNmkW9evUqvE1ERCQCDu7D5j8GSbWgzonev/N+H5GRNAJz1zXXXBM3uatNmzbKXSIi1cjs2bP5+9//TseOHenUqRNr165l3bp1tG/fnjlz5jBmzBg+/fTTUvNIvXr1SE5OZvjw4bzzzjuFw+zOmTOH2267jczMTPr378++fft09281pRE0RCRSOnbsyPbt29m6dSvLly8nLS2NU045Becc9913HxkZGVxwwQVs2bKFbdu2hexn/vz5hcdLGRkZZGRkFLZNmzaNTp060bFjR1avXs3q1atLjGnBggUMGDCAOnXqkJKSwsCBAwtHgGrevDmZmZkAnHHGGUUK4YPt3buXPXv20LNnTwCuv/565s+fD3jDyV999dW8+uqrJCV5Nxp169aN//qv/+LJJ59kz549hfMlchzKXSJSflWds1atWlVlOWvo0KG8+eabzJw5kwEDBhTOT0xMZNasWUyfPp2WLVty1113FY5GBd4jTpYtW8ayZcsqXJwBGkFDRKRicrZ7I2ck1vBeJ9bwXudsh7T0CnUdXHVY8LpOnToAOOe48MILeeONN4ost2zZskp53qAr4aTUcccdV/hzYmIiR44cIS0tjeXLl/Pxxx/z9NNPM23aNF566aWIxyUiImHavxPLPwI1vYtM1KwNP2bD/p2QXLGCBOUuERGJBc45Ro8eXVjUF2jp0qX87W9/495776V37948+OCDIftJSkpiyZIl/OMf/+DNN99k0qRJzJ07l/z8fBYtWkStWrUA7xFZiYmJlfZ5JHrMjPxoByEikVfCXcOVadCgQUyfPp3vv/++8HEfr732Gjt27GDp0qXUqFGD9PR0Dh48WGI/xR07ffvtt0yYMIF//vOfpKWlcf3115faTzjHS6U94iSU999/n88++4z33nuPRx55hFWrVjFmzBguvfRSPvzwQ7p27cqcOXNo1apVufqvTsysH9CvRYsW5e/D/1e1hSLVyDGQs4YNG1ZlOatfv37cfffddO7c+agbs8yMM888kzPOOIM+ffpwww03FCnSiCSNoCEiUhEpJ3qPNck77L3OO+y9Tjmxwl1/9913LFq0CIC33nqL7t27F2nv2rUrn332GevXrwdg//79fP3117Rq1Ypvv/2WDRs2ABx1EaxAr169mDx5shd2Xh779u2jbt26Ie/86tGjR2Fl4Ndff813333H6aefHjL+nTt3kp+fzxVXXMEjjzzCl19+GcanFxGRSlO7AS4hCX7c773+cT8kJHmjQFVQYO564403lLtERKRKBOeCPn36MGXKFHJycgDYsmVL4R1gtWvX5uqrr2bUqFGF+/lQuSQnJ4e9e/dyySWXMHHiRJYvXw5A7969eeqppwqXK3i0Skk5SeJTQkLJJ4NFRMJRcNfu9OnTGTRoEOCNPnHiiSdSo0YN5s6dy8aNG0vsI/AYZ+XKlaxYsQKAffv2UadOHVJTU9m2bRuzZs0qfE+o/NSjRw9mzpzJ/v37yc3NZcaMGZxzzjlhf67U1FTS0tIK72R+5ZVX6NmzJ/n5+WzatInzzjuPxx57jD179pCTk8OGDRto3749o0ePpnPnzqxduzbsdVZHzrn3nXMjShvhS0SkKlRlzvroo48K31PZOatWrVr84Q9/4P777y8yf+vWrUXOAy5btoxmzZqF3X9ZaQQNEZGKSEuHM27wHmtiCV5xxhk3VHj0DIDWrVvz8ssvc9NNN9GiRQtGjhzJpEmTCtsbNmzI1KlTufLKKzl06BAA48aN47TTTuO5557j0ksvpUGDBnTv3p2VK1ce1f///u//MmLECF588UUSExOZPHkyZ599Nt26daNdu3ZcfPHF3HrrrYXL33LLLdx88820b9+epKQkpk6dWqQyMdiWLVu44YYbyM/37jf6/e9/X+FtEi1mlhM0qxbwjHPudr+9F/A0cAqwGBjmnCv5rxMRkWhJrofrMRrm/8EbOSMhCc69t8KjZ0DR3NWyZUvlLhGRY1XqKTC26MWFksaXKLUt9ZQSV3fCCScUyQV//OMfWbVqFWeffTYAKSkpvPrqq6xfv567776bhIQEatSoUVj0N2LECC6++GIaNWrEnDlzCvvNzs7msssu4+DBgzjn+NOf/gTAk08+ya233kpGRgZHjhzhnHPO4c9//jP9+vVj0KBBvPvuu0yaNKlcJwwltiSYka/6DBGJkLZt25KdnU2TJk1o1KgRAFdffTX9+vWjc+fOZGZmljqSxMiRI7nhhhvIyMggMzOTM888E4AOHTrQsWNH2rZty6mnnsovfvGLwvcE5rm5c+cWzu/UqRPDhg0r7GP48OF07NixxKHhQ3n55Ze5+eab2b9/P6eeeipTpkwhLy+P6667jn379uGc46677qJ+/fr89re/Ze7cuSQmJtKmTRsuvvjisNcnxSu4UV2pS0QqqipzVsHjkqFqclbBiCCBDh8+zKhRo9i6dSvJyck0bNiQZ599trD97rvvZty4cYWvlyxZQs2aNcNedwGrLlXgnTt3dl988UWF+sjOzqZu3boRiqjyKM7Ii5dYFWfkFRfrmjVraN26dXgd7c7yHmuScmJEijOysrLo27dv4cWpeBoyN9xYi9veZrbUOdc50rFFgpnVAbYBlzjn5ptZA2ADMBx4H3gEOMc517Wkfo6lvFUlDuwhe99e6p5UeVWt8Ua/H0VV9+0Rbu7Ky8sj8XCu91iT2g0iUpwRnLviRXXPW5FU2bkrHtpiJY7q3hYrcUS0bWwqjN1b5bEUzg9Yf2Wsq2DfWNI+tTq3lbe/qqLcVf7cdd6EebQ6qTaTrz0zglGFL1b+llUcsRdHLMQQL3GU63xfOcXCvj8WYqhIHMpd5ctdSzfu5orJC3n5V2fS87SGEY4sPPGwXzjW4oiFGBRH2eKoypwF8Z8zYiWOcHKXRtAQEYmEtPSIFGZIXBgEbAc+9V8PBFY5594GMLOxwE4za+Wc0xiNVeXj+yA7G658AZJC3x0vIgGS60WkMENERESkujKDanJvm4iIHEOqy43ZIiLVlQo0RERiUHp6etzdgXwMuR74i/vpSKctsLyg0TmXa2Yb/PlFCjTMbAQwAqBp06YVfj51bm5uhd5frZxyIbn7dsP+Q2A/RjuamKDfj6Kq+/bIz88nLy8vrOUjrWnTpixfvjysOGJBuNsiPz+/wvtvERERiQ/eI050kUtEROJDwSNOREQktqlAQ0SkGM45TH/RVrp4q+Y2s1OAnsCvA2anADuCFt0LHDU+mnPuOeA58IYrjMQQarEwDFtM6DQAYmRYulii7VFUdd4eCQkJJCQkhJW7YmHIwFhR1m3hnCMhIaFa/y6JiJRXvP1tf6zQ91IxCRpBQ6Ra0fm++KDcVXHagiLxTzkrvoSbuxIqKQ6pJMs37WHTD/ujHYZItZacnMyuXbt0MFDJnHPs2rWL5OTkaIcSjuuABc65bwPm5QDBzwioB+j2ahGpMspdlS9O85aISJVQHopNyl0VpxE0RKoP5ar4oNxVMbqUK1I9KGfFl/LkLo2gEUcO/JjHuA9W0yItid83Oyna4YhUWyeffDKbN29mx47gQRGiJz8/n4SE+KipCyfW5ORkTj755EqOKKKuA8YHzVuF99gTAMysDvBzf36195dFWazYvIdHLmtPrZq6G18kWsLNXfGUVypbNc9bIiJVoiAPbdu2LeQ+taT9bby3lbe/qlDdcpeZPQr0ALYB1znnKvUuJjMjX+fFRaqFqjzfF+19f6zEUN44qlvuigrlLpG4VtXXqOI5Z8RKHOHmLhVoxJFaNRO584KW1LbD0Q5FpFqrUaMGzZs3j3YYRWTH0aMb4inWcJjZL4AmwNtBTTOAP5rZFcAHwIPACufc2ioOMSqyDx5mz/7DuqtMJMrCzV3VdV9dHvG8LcxsHtAVOOLP2uKcO91v6wU8DZwCLAaGOec2+m2GV3A43H/fi8Bop1szRKScCvJQSfvU6txW3v4kPGbWDvi5c+4cM7sZ+BXwVGWu03vEidKjSHVQlef7YmHfHwsxxFIc8cDM+gH9WrRoUZE+IheQiERNVV+jipV99bEUR9TKUMxsqJmtMbNcM9tgZuf483uZ2Voz229mc82sWbRijEXdWjSkxYnR/+UUETkGXQ+845wr8ugS59wO4ArgUWA3cBYwtOrDi45bzm3Bn6/tTJ3jVPMpIhIltznnUvypoDijAfAO8FvgeOAL4K2A94wALgc6ABlAX+CmKo1aREQkfOcAH/k/fwR0r+wVJmgEDRERqSLOufedcyNSU1Mr3peG0BARiWlRKdAwswuBPwA3AHXxhib8pgwnEkVERKLCOXeTc+7aEG1znHOtnHO1nHPnOueyqji8qDEzEhNUnS8iEmMGAqucc2875w4CY4EOZtbKb78e+JNzbrNzbgvwJ2BYVCIVEZFjjpndZmZfmNkhM5sa1Ha8mc3wb+jaaGZXBTSnAXv9n/finTusVAmGRgsUEZG4oTN0IiLxIVq3uz4EPOyc+9x/vQXAzEbgn0j0X48FdppZq2NlqPhqY/8P8J81kPIL0LBaIiIRt2rLXpZ/+z2DurakZlL0n8sWbfO/3sEPe/ZyWZcUDecoIsea35vZeOAr4H7n3DygLbC8YAHnXK6ZbfDnrw1u939uW1zn/jHaCICmTZuSnZ1d3GJllpubG9dtsRJHdW+LlTgi2nZcY/D//0RlWwasv7LWdSy3lbe/Y9hWYBzQB6gV1PY08CNwEpAJfGBmy51zq/BGLCy4rTgV+KGyAzUzVJ8hIiLxRrlLRCS2VXmBhpklAp2B98xsPZAMzATupvQTiRIvlk6FrxdAWgNoeHq0oxERqXZmr/6eFd9+zzltmtL0+NrRDifq3vlyM/tzc7i4U3OOS0qMdjgiIlVlNLAa70LWUOB9M8sEUoAdQcvuxRu9EL99b1BbipmZc0VP5TnnngOeA+jcubOLxDM4S+ojHtpiJY7q3hYrcUSs7dBWCHhd5dsyaP3RiKO6t5W3v2ORc+4dADPrDJxcMN/MKTegmAAAIABJREFU6uA9PrKdcy4HWGBm7wHXAmOABcC9wMt4xR2fVXasGkFDRETiScE9S0pdIiKxLRojaJwE1AAG4T078jDwLvAApZ9ILKIq7+aKJXERZ3ofchObQlLDo+7SiUVxsU1RnJUhXmKNlzghvmKNZ7/qfirftUxVcYbvvktas2fvPhVniMgxxTm3OODly2Z2JXAJkAPUC1q8HlDwh3lwez0gJ7g4Q0REpIqdBuQ5574OmLcc6AngnPu3/9iTT4HtwHXFdRLJ84UuP5/DzlX4nGNFxcpxtuIoKhbiiIUYQHEEi4U4YiEGiJ04jhWmh5yIiMSFaBRoHPD/neSc+w+AmT2OV6Axn5JPJBZR1XdzxZKYj7NuW6h/SuzHGSBeYlWckRcvscZLnBBfscar1Fo1SD+hTrTDiBkn1kumlh2OdhgiItHm8B47vAq4vmCmf0fyz/35+P92AJb4rzsEtImIiERL8AhPEHTjlnPu3tI6ieT5whpJSeDyYuIYNxZiAMURLBbiiIUYQHEEi4U4YiEGiJ04jiWqvBcRiW1V/tB659xuYDPF54iCE4VAsScSRURERI6y+JtdLNywk/x8HYKKyLHBzOqbWR8zSzazJDO7GugBfAzMANqZ2RVmlgw8CKxwzhU8NvIvwH+ZWRMzawz8NzA1Ch9DREQkUGkjQFU5M8jTIYaIiMQJ0wAaIiJxocoLNHxTgNvN7EQzSwPuBP5G6ScSJU7sO3iY9duz0SjJIiJS2bIPHuaPH3/F64s3smFHTrTDERGpKjWAcXiPiNwJ3A5c7pz7yjm3A7gCeBTYDZwFDA1475+B94F/AyuBD/x5IiIi0fQ1kGRmLQPmRXWUpwQzndsSEZG4o9wlIhLbovGIE4BHgAZ4B14HgWnAo865g2Z2BfAU8CqwmKInEiVOvP75Rhau3cyDA+vR4sSUaIcjIiLVWMpxSVzY5iRyc7M5Oa12tMMREakSfhFGlxLa5wCtQrQ54B5/EhERqVJmloR3TjIRSPRv0jrinMs1s3eAh81sOJAJXAb8IlqxJiSABukTEREREZFIikqBhnPuMHCLPwW3hTyRKPGjV+uTqFcjj5PTakU7FBERqebMjJt6/pzs7Gxq1UyMdjgiIiIiIlKyB4DfBby+BngIGIt3rvAlYDuwCxjpnAt7BA0z6wf0a9GiRfmjdI5kd4gD+UfK34eIiEgZRSR3+VRbKCIS26L1iBOp5lqeVJd+GY1JrqELZSIiIiIiIiIi4nHOjXXOWdA01m/7wTl3uXOujnPuFOfc6+Vcx/vOuRGpqanlDzQ/jxe39Gfggb+Wvw8REZEyikTuMotgQCIiUmlUoCEiIiIiIiIiIiISKCGRfIxEpxE0REQkvrj/Z+/O4+Oq6/2Pvz6zZbJ3X2jThbZQWmgrhB0RWYQrt4IsigsX7hV6xZ+7XvF33RBBvSrXexUU+0ME9SoIwpXKXgHZBIFCN9rSna60Tdtksmdmvr8/ZgJpmk7SZpLvmeT9fDzySOacyTnvnrb5Zub7OZ+vWmiIiASaCjSAPQ2tPLZ8O7VNbb6jiIiIiIiIiIiIiG9mpIioQENERAqGoRYaIiKFQAUawO9fepP/eGQlf1681XcUERERERERERERCYCUqUBDREQKkVpoiIgEmQo0gBMmDWP6mAqOGd+LdSlFRERERERERERkwFCBhoiIFBLLNtDQEiciIsEW8R0gCKonDaN60jASiYTvKCIiIiIiIiIiItILZjYXmDt16tReHSdpEcKoQENERERERPJHHTRERERERERERERkwHDOLXDOzaus7F233JRFiKiDhoiIFIi3O2j4jSEiIt1QgUaB2ZloIdGsF4YiIiIiIiIiIiJ9KY2WOBERkcJhmO8IIiLSAyrQKCDNbSk+d9erzP/rWt9RREREREREREREBrSkRYhoiRMRESkwTi00REQCLeI7gPRcUSTEOTNGMyqu0VVERAJi9zqo2wYTT3mnj6KIiIiIiMgAkNYSJyIiUkD01pyISGFQgUYBMTP++dTJJBIJ31FEREQynr8Zti+Fj/weSkf4TiMiIiIiIpI3KYsSVgcNEREpMA7d5CsiEmQq0BAREZFDd9I1ULcFSob7TiIiIiIiIgKAmc0F5k6dOrVXx0mH1EFDRET6Rz7GLjXQEBEpDCHfAURERKSAjZgGh5+hHooiIiIiIhIYzrkFzrl5lZWVvTpOWh00RESkn+Rr7MocKw+BRESkz6hAQ0RERERERERERKSTdChCWB00RESkQOj+KRGRwqACDREREREREREREZFOXCiqJU5ERKTgqIGGiEiwqUBDREREREREREREpBMXimiJExERKSBqoSEiUghUoCEiIiIiIiIiIiLSiQvFiLgkzuleZBERKRwat0REgk0FGgDOQePuzOeAW7MjwVt1zb5jiIiIiIiIiIiIDGzhCBFSJNPBf89QRETE1EBDRKQgqEADYOPz8PvLYMsi30ly2rCrga/dv4wfP/4GzW0p33FERMSn1x+AJ26A5lrfSURERERERAYkF44RsyRtqbTvKCIiIt1SfYaISGFQgQZA2WgYOR1KhvtOklN5PEJ5PMqIshiRkIZaEZFBrW4L1G6FZIvvJCIiIiIiIoFiZnPNbH5tbS8L2kNRIqRoS6qDhoiIFI4CaBYvIjKoRXwHCIQRU+EDP4FEwneSnIaXFfGLy4+jPpEgElZtjYjIoHbiJ+HIS6BspO8kIiIiIiIigeKcWwAsqK6uvrpXBwpHiZKkJZUConnJJiIi0ldMa5yIiBQEzfIXmHDICKl7hoiImEEk7juFiIiIiIjIgGXhKFFStKV0K7KIiBQOh8YtEZEgU4GGDHrNbSlcAfT8SqUdrcmU7xgiEhQ7VsLKhyHV5juJiIiIiIjIgGSRGFGStCXTvqOIiIh0S7f2iogUBhVoyKC2eU8jV/7q7zy6fLvvKN264c+vc8ODK2hL6U0BEQGW/RGW3g21m30nERERERERGZhCUSKk9F6MiIgUlAK4H1VEZFBTgYYMasXRMOOHljC0NOY7SrfGDS3msCHFhLSOnIg3ZnaZma0wswYzW2tm785uP8vMVppZo5k9aWYT+zzMif8Kp/8bDJ3U56cSEREREREZjCwSI2opWtrU0VRERIJPUwciIoUh4juAiE/Dy4r40aWzSSQSvqN066p3H04ikSAc0m9ZIj6Y2TnAfwAfBv4OjM1uHwHcB1wFLAC+A9wNnNSngUpHwOgivfISERERERHpI6FIFIBkW4vnJCIiIj2nDhoiIsGmDhoiIiI9823geufcC865tHNui3NuC3ARsNw5d49zrhm4DphtZtN9hhUREREREZHeCYUzHVeTba2ek4iIyEBnZnPNbH5tbe2hHwPdyCUiUghUoCF9wzlIq/2jiAwMZhYGqoGRZrbGzDab2c1mVgzMBBa3P9c51wCszW4XERERERGRAmWRTIFGW6sKNEREpG855xY45+ZVVlb2/lh5yCMiIn1HS5xI31h8Fyx7GD7wfag4zHcaEZHeGg1EgUuAdwNtwJ+ArwNlwM5Oz68FyjsfxMzmAfMAqqqqer28UkNDQ6++f6DR9diXrse+dD32pevxDl0LERGRgcfM5gJzp06d2qvjhLIFGqmkCjRERCT42ldCdlrjREQk0FSgIX0jVgpFlRCK+k4iIpIPTdnPP3XObQMws/8kU6DxNFDR6fkVwH7VF865+cB8gOrqaldevl8Nx0HLxzEGEl2Pfel67EvXY1+6Hu/QtRARERlYnHMLgAXV1dVX9+Y44Ujmfa1kW0s+YomIiPjzyp2w4LPwte0QLfadRkRkUFOBhvSNGR+AqvdCmd7sFpHC55zbY2ab6bpD4HLgivYHZlYKTMlul37inNPdASIiIiIiklehSBEAyTZ10BARkcLR5TtkT30/87mxBirH92ccERHpJOQ7gIiISIH4FfAZMxtlZkOBzwN/Bu4Hjjazi80sDnwTWOKcW+kx66Dzw0dX8b2HV5JMpX1HERERERGRAaK9g0Y6qQ4aIiISfO1LnHTzrL6OISIi3VCBhoiISM98B3gJeANYAbwK3Oic2wlcDNwI7AFOBC7zFXKwioaNaEgvMEVEREREJH/CsUwHjXRSHTRERKSA5Gwyqw60IiK+aYkT6RMLFm9l4eINfO2DxzKqPO47johIrznn2oBPZT8671sITO/3UPK2L5xzJIlEgkhYtaciIiIiIpIf4UgMgJSWOBERkQJguVpo9Ky9hoiI9APNYkifSDtHGqdiTBERERERERERKUiRaKaDRirZ5jmJiIhIz7muJmacJmtERIJCHTSkT1wwZxxnTqmgXN0zRESkj6WSSf5++xdIJ5PM+tiNlFcO8x1JREREREQGgEg000EjnWzxnERERKR7PeuRoU4aIiK+qYOGiIiIFLTmpnoqdr5KWcNG9rz1pu84IiIiIiIyQITjpQBYa4PnJCIiIj2nZhkiIsGmAo0Ckk47Xlm9ibXbanxHERERCYzS8iHE5/4Ad/KnqZo6y3ccEZF+Z2bTzKzZzH7bYdtZZrbSzBrN7Ekzm9hhn5nZf5hZTfbjB5ZzsWIREZHBKVw6HIBo617PSURERLqX+1WdqjZERIJCBRoF5OlFSwndcwW1f/4Gq7bs9h1HREQkMKbMOoUps9+NhfSrjYgMSrcAL7U/MLMRwH3AN4BhwMvA3R2ePw+4EJgNzAL+EfjX/gorIiJSKKwkU6ARa9njOYmIiEjP5SzFUG2+iIh3msUoIMm2FkIuSTTdSlsy6TuOiIgIW/c2sXhTAO4me+NRWPkwpFO+k4iI9CszuwzYC/ylw+aLgOXOuXucc83AdcBsM5ue3X8FcJNzbrNzbgtwE3Bl/6UWERHpW2Y218zm19bW9u5A8SGknBFTBw0RESkARj8WXzTuznyIiMhBi/gOID13+vHH8Vzkp8QiIWZOGOk7joiICD9/ai3Lt9Zy2xXHM6w05idEcx08cxNQBhNnwajp3X6LiMhAYGYVwPXAWcAnOuyaCSxuf+CcazCztdntKzvvz3498wDnmEem4wZVVVUkEoleZW5oaCjofUHJMdD3BSVHXvcVHQbZ/z9ermWH8/fVuQbzvkM9nvQd59wCYEF1dfXVvTpQKESdlVHUpgINEREpHK4/VjP5weTM5+t6WQwpIjIIeSnQMLOngJOA9jYQW5xzR2b3nUWmRe8E4EXgSufcRh85gyYWCfHe42eTSCTQEtEiIhIEHz1xApt2NzK0JOovRFE5zPow1NfDsMn+coiI9L/vAL90zm3q9PqgDNjZ6bm1QHmH/bWd9pWZmTm371t5zrn5wHyA6upqV15eTm/lOkYh7AtKjoG+Lyg58ravZSt0eNzv17LT+X3kGOj7DvV4EnyNFBNNNvqOISIi0q32l4Wuq0VO+qVqQ0REesJnB41PO+du67ihw1rJVwELyLzheDeZYg4REREJmKPGVnDU2Aq/Iczg+E9k7gqNFPnNIiLST8xsDnA28K4udtcDnX84VwCJA+yvAOo7F2eIiIgIpCwC6TbfMURERLrVs9t6dfOviIhvId8BOulureQ+sXHlIl76z0vZsnZZX56m99IpePTrsORu30m69eTKHdz06Er2NLT6jiIiMiCl0o6m1mT3TxwkttU2sWm37moTkUHlDGAS8KaZbQe+DFxsZouA5cDs9ieaWSkwJbudzvuzXy9HRERE9pO2MLiU7xgiIiI9lrv0XnX5IiK++eyg8T0z+z6wCviac+4pul8reR/5Wg95b80OoulmEnt393pN5T6VbIXdb9GQLNtv7dig2Vazh7r6BnbvrSWSjvuO061CWRNWOfOvULIWSk4orKyF7FfPreelN7bwnUuPZ1RF8H/O9qWm1hTX3ruEOK1844PlVA0r8R1JRKQ/zAfu6vD4y2QKNq7JPv6hmV0MPAh8E1jinGt/TfVr4Itm9hCZd+e+BPy0P0KLiIgUmhQRLKUOGiIiUgDUHENEpCD4KtC4FngdaAUuAxZkW/R2t1byPvK1HvLsU8+jcfbJpFwo+OuCXnoz1DcGPudHTjmCc44azchhQ3xH6bGgX9N2ypl/hZK1UHJCYWUtVBOHl/DW8FKKY2HfUbyLhI0RZUWkW9OUFvmsPRUR6T/OuUbg7dZBZlYPNDvndmYfXwzcDPwWeJHM6652vwAOB5ZmH9+W3SYiIiKdpEMRzHXRvbC1AWKl/R9IRESkG7l7ZKiKQ0TENy+zGM65Fzs8vNPMPgK8n+7XSu4zJWWVwe6e0S5SBOHgLxtiZsSjmjQUEekr58wYw0lVpZTHo76jeBcNh/jhpbOpq6tjaGnMdxwRES+cc9d1erwQ6HKpSOecA76S/RARz079/hNs2dsEwOhix1tNXb9pfij7jhkVY8EXz8lfWJFByFkYS3cq0Ni+FG49DS75FRx9kZ9gIiIinViPii+0xImIiG9Buc3UkSnbWw5c0b6xi7WSJZ3qbgExOVi6piIiBS8cMiLhkO8YIiIiIgdty94mNnz/fAASicQBu9Edyr4Tv/3n/AUVGaTSFtm/QGNbdoXm1Y+rQENERIIn13yH5kJERLzr95kMMxtiZueaWdzMImb2MeB04FHgfuBoM7vYzOLsv1by4NZQA3dfDk/eCKkuWisGyNJn/sTi//katXt2+Y6S2/Zl8OsLYNFvfScRESlID7y2le8/vIKWZMprjhfuuYlX//gD0im/Of72m2/xt99+y2sGERERERHJH2dhzHV+naH28CIikl9mNtfM5tfW1vbiGLn2qjBDRCQofNxqGgVuAHYCu4DPABc651Zl10u+GLgR2AOcyL5rJQ9uzXuheQ/UvwXpNt9pcmrcsoKy+g001Nb4jpJb/XZorYfaN30nEREpSCu317Fkcy0tybTXHLZ9CcW7V5FM+h0fY2+9Rmz7a14ziIiIiIhI/rhQhJAL9o1SIiJS+JxzC5xz8yorK3t/rF7sFRGRvtfvS5xkizCOz7H/gGslD3rDp8D5P4a2EESLfafJqfqSL/PW9i0cNn6i7yi5Hf5eKB4K0dG+k4iIFKTPn30ELckU5fGo1xyzr/oZdbV7iRXFveaYcY06MomIiIiIDCQuFCWsAg0RESkAPervpCVORES86/cCDemlUdMhkfCdolvhSITyymG+Y3QvFIZxxxXENRURCaJYJEQs4qMh177ixaW0ee7iAVBc2vW68CIiIiIiUqBCYUL7LXHSTpNcIiISPF3WYKgwQ0QkMPzPqIiIiIj0QnNbiu8+tIL/fGwVNfUtvuOIiIiIiMhA0lUHDcveo6zJLhERCRDLjk8u5/iksUtExDcVaBSSdBqe+TGsesR3ku5tfB6evxma63wnyalubw1/+8WnWbfked9RRETkEG3a3cjLG3azdmc9y7cGe9wREREREZG+Z2ZzzWx+bW1tr4/lQhHCLtnNZJeIiEiB0HgmIuKdCjQKSaoV1j8FWxb5TtK9ra/B9sXQuMt3kpzqdm1jyK5XSGxa6juKiIgcokkjSjlnxhjmTBjCnAlDfMcRERERERHPnHMLnHPzKisre30sC0cIkyaZ1oSWiIgEW7a/Uzc9MjSeiYj4FvEdQA5CNA6X/Aqa2nwn6d4JV8Pkf4Bhk30nyWn81KPZ9c93MSZe5juKiIgcomg4xDVnTCGRSFAej/qOIyIiIiIiA0koQsRStCTTRMPt97pZzm8RERHxwTQ8iYgUBHXQKDQlwyBW4jtF98JRKB3hO0WPjDhsIpFozHcMEZGC1JDYy86tG3zHEBERERER6RMWjhIhSUtbqou9ugtZRESCJ+cqJlriRETEOxVoSN9Ip6GtyXcKERHpY0t+cy077vg4tbt3+o4iIiIiIiKSd6FwhAhpWpLpdzbqFmUREQkg61GHJxVoiIj4pgKNAlLX1Mr9t36LxxfcRTrI6146B498FRZ8DtY97TtNTuu3bOdP//15nnv+Gd9RREQKUsn0s6mrOpPS8t6v7SwiIiIiIhI44SgRUvsWaIiIiARY17NHAZ5TEhEZZFSgUUB27qljSs0TlL/1YrBfFKbaYMdySLbAW8t8p8lp25YNTK17nuY3X/EdRUSkIM0+81JOvvx6LRUlIiIiIiIDUihboNHc1RIn+WoT39YENWvzcywRERm8etRAQ4UaIiK+RXwHkJ47fOxwFn/w/zEmHKI4FvYd58AiMTjrOti0HGZ/wHeanI479ngWR27lmBGjfUcREREREREREZGAySxx0scdNO79BKx6EL6+M/O+moiISC84FWGIiASaCjQKiJkxZ8Z0EomE7yjdqzoehkyHknLfSXIqioQ54dhjC+OaiojIAf3tV18l3ZTgpHn/TTiiX29ERERERCQ/LBwlailaWpN9d5K1f8l8dl106RAREekhUwcNEZGCoCVOREREpPC1JAi3JUin9YamiIiIiIjkj0WLAWhraey4NftZk1wiIlIgVJghIhIYKtAoJC0JeOjL8PfbAj+YvvSnn7H0js+zdcMq31FyampNcdsz61i3q953FBER6YWT5v2Uo674CdFYke8oIiIiIiIygFi8EoB0096+P1nA3+8TEZFg60kDDRUXioj4pwKNQlK7BbYsgs0vQltj98/3yNYsZFjjOnasfdV3lJy27G3kz0u28cqGPb6jiIhIL1goRCgc9h1DREREREQGmC4LNHrUQ15ERMSPnPV+KgYUEfFOi7QXkpFHwpnfAFcCsVLfaXIac8H1bFu7lGNOmes7Sk5TRpbxw0tmURntw3VERUQGsNdffJTE5tepvuAzhCMef61Y8xfYWwPHXer3zdIVCzKfjwr2+CciIiIiIj1jxRUAuKa6fjibJs1EROTQWfY9MafxREQk0FSgUUjMYMp7IZHwnaRb46ceTeXoicSLg11IYmZMG11OogCuqYhIENW9dDcVtStJvOcjDBkxxl+QxXdBQxPMuQAiHpc5WfTrzGcVaIiIiIiIDAjh4kwHDVq6WOIk33ch665mERHpBS1xIiJSGFSgUWAaWpK0JlO+Y4iIDDpm9hRwEtDecmeLc+7I7L6zgFuACcCLwJXOuY0+cva3Iz98I/V7d/gtzgB43w1Qu9tvcQbAP/7Y7/lFRERERCSvIiVDALDmjh002qfA8j3JpUkzERHpPS1xIiISbCHfAaTnlm/Zy2233sRv/ngfOxLNvuPktGZHPQ8u2UZzW7CLSbbubeJr9y/lmdU7fUcRkcLwaedcWfajvThjBHAf8A1gGPAycLfHjP1q6MixVE2b7TsGlI+GyvG+U2QyBCGHiIiIiIjkRaQks8RJqKUfljjRpJmIiPRCz1b91VgjIuKbCjQKyOtv7uCsxAKOqXuaDbsafcfJaeHr23lo6VY272nyHSWnZVtqeXFdDU+/oQINETlkFwHLnXP3OOeageuA2WY2vU/P2rgbdq7y/wZe/Q5463W/GYBFS5ayaMlrpNMer0eyld1P/ITdT/wEkq3+coiIiIiISN5Esx00Qq39UKChSTMREcmDrkcTjTEiIkGhJU4KyHtmTuD3G/8vJcWlfHxcpe84OX3kxImcML6EKSNLfUfJ6dSpI/j0mdOYNjTsO4qIFIbvmdn3gVXA15xzTwEzgcXtT3DONZjZ2uz2lX2W5MVbYf0iGPojGDqpz07Trb/+ALYvhct+B6XDvURINLdR/+DXKYs61k2aztRxI73kYPsSal74HQDDjjgFxlf7ySEiIiIiInkTihbT6iJE2hLvbOzZLcoHz3cBvoiIFDSjB+OTxhoREe9UoFFARlXE+dxl55NIJCiOBbugoLI4yrTR5VhfvWDNk9KiCBe+axyJRKL7J4vIYHct8DrQClwGLDCzOUAZ0LkNTy1Q3vkAZjYPmAdQVVXVu589k99PQ2wSWCX4/Bl2xCUw6gRIRrzlcM6x56iPUtOU4LR4xN/P9NgYwuPeBUAiNsbv3wvQ0NDg9fxBo+uxL12Pd+haiIiISE5mJCgh0lUHjbxNcmXfP3PpPB1PREQGs9zDkwo0RER8U4GG9AnnHC3J1P6zkyIiBco592KHh3ea2UeA9wP1QEWnp1cA+83OO+fmA/MBqqurXXl5L35Kls+B4VPo1THyofwk4CS/GYALLvwwiUTC7/UoL6f8ip/7O38XvP/7CBhdj33perxD10JERERyqbdSosm+LMDWZJmIiPRezvtl1TlDRCQwQr4DyEFwDt54FLYv850kp7ZUmh//4VH+8D+38cSyTb7jDAjOOR5+6AGeeOQ+mttSvuPk1lADKx+Epj2+k4j0NUfmNqflwOz2jWZWCkzJbh/wnHO0pfzf5ZVKJkm2tXrNsGl3I9f+cQnX/nEJm3Y3es0iIiIiIjKYmdlcM5tfW1ubl+M1WCmxtvq8HCsnTZ6JiEgeuFyFfxprRES8U4FGIWlrguf+C5bc7TtJTjsTLYza9DDvaV7IsmWv+o4zINS3JClbfBvjN9zL5h27fcfJbdML8Or/wOZXfCcRyRszG2Jm55pZ3MwiZvYx4HTgUeB+4Ggzu9jM4sA3gSXOuZU+M/eXnzyxhqt//TL1LUmvOV6a/yne+M3naGtt8Zbh9mfXs7GmgY01DfzqufXecoiIiIiIDHbOuQXOuXmVlZV5OV5zqIRYqqsCjXxPcmnSTERE+prGGhER37TESSGJlcA//BBSwf5rG1MRp23W5SzcsoIzTjnNd5wBoTweJXLON9hdt5c5Y0f4jpPb4e8FymDSib6TiORTFLgBmA6kgJXAhc65VQBmdjFwM/Bb4EXgMk85+93IshhjK+NEQrl6KPaDsjG0NSUIhcLeIpQUhUmlMi9yS2LBHqtFRERERKTnkhYl7PYvBl+2pZaj83ki3dUsIiJ5kHM40VgjIuKdZg8KzZijIdGXa172Xihk/PP7qkkkjtR63nl0cvXxJBIJwr4nQbsTK4Fxx0I07juJSN4453YCx+fYv5BM8cag89ETJ/qOAMBJ/3R95mdn6YZRAAAgAElEQVRkxN+vNleddjglsUyByEdPCMZ1ERERERGR3nOhKKF0W4cNmWUeN9Q05rdAQ3c1i4hIL1jApw5ERCRDBRoiIiJy6JItkGyGeH5aBxeyoaUx/s97p/mOISIiIiIieZYORQklOyzr+Pbdx3kuqNBdzSIi0uc01oiI+BbyHUAO0vqnYdcbvlN068W7/4Nlv/ocG1e84jtKTmvWb+Chn36W559+1HeUnJxz3PvKZu59ZROtybTvOCIiGa2N8L/XwF0fg22LvUZ56YGf89qCm3Fpzz8jX/olvHS71wi1TW189vevcvMTq0mlPb7oTrXBX38AKx/yl0FEREREJA9cKEa4iw4a+T+R3vMREZFDZ2RaaLhcBX8qBhQR8U4FGoWkrQme/C4s+o3vJDm1NDdSseZ/KW/ewraXH/AdJ6c1Lz/GrD0Lia/8E22p4L4ITrQk+f3fN/LEyh1sqGnwHUdEJKO1Huq2ZT7v3eQ1ilv/LCXbXqK1tdlrDtY+AWv/4jXC7oZWtu5tYsOuRlqSKX9BWuszhaVvPu8vg4iIiIhIPkSihF2HAo28333c3pNek2YiInLoci9xojFGRCQotMRJIYkWw/tugHTcd5KciuIl1M/8GK3bVjDx5Et9x8lp+kn/wOKaDcQPP5VoOLj1SuVFET5x2uHU1SU4fESp7zgiIhllo+DMr0NiG0w922uUmf9yC3v37KEoXuI1Bxfc4vf8wKThJXxr7kxCqWZKYh5/1SseCpfeAVHPfyciIiIiIr1k4Rhh19USJ3mmu5pFRAYUM6sEHgdmACc555b1x3lzDycaa0REfFOBRqEZXw2JhO8U3Tr+wk+TSCQoLy/3HSWnSVVVTPrkd0kE/JqaGe8/ZiyJRBmRABeSiMggNOlU3wkAKC0fQpqw7xhQPMR3AsyMY8ZXkkgEYLwoHeE7gYiIiIhIr1kkRoT9lzixvE1yuU6fRURkgGgEzgd+2B8ny9lAo52KAUVEvAvAO/cyEDW1ptiZaM691pmIiBS8zWuW8frfHsalPS4TlWyFp/4DnvlPaNrjLUY67Xj8L4+x8C+PkU57Hv82vQQ167xGcM7xwroaNuzS0lwiIiIiUtgsHCPSsYNGtpCiRxNhB0Pvo4mIDCjOuTbn3M5+P2/XYfo7hoiIHIAKNArJ7nVw9+XwxA3QXOs7zQHV1Ldw1/zvsfruf+euR57yHSenVMNu1v/pe9SuX+Q7SrfW7EiwYlud7xjda0nAxr9Ba6PvJCLSDzY//ENCf7mOPbu2+Quxex2seRy2LYYt/n6eb69rZuQLNzL8he+yva7ZWw7amuHxb8LLv/SXAdjd0MoPHlnJL59d7zWHiIiIiEhvhaJFREnSlsoWpru+KlDX5JmIiC9m9mkze9nMWszsjk77hpnZ/WbWYGYbzeyjnmLmZNaT0kGNNSIivmmJk0Ky7mlIbIemVti+LDBt5TtbtWUXJ9U+iMUreWv5n3HnndHDXwz63/YlC4ks/g2129cwftZ7fMfJ6b8XrqapsZ5Zk8cQjwagjf+BrH8anvsZFBlMO8d3GhHpY+PO+yK1b21k6Iix/kIMmwyTz4CmFjjsXd5ijKmIs6T6K5gZsyvi3nIQjcNZ34R0kb8MwLDSGF9635GMrfR4LURERERE8iAciRKzFLUtSSpLYn13F7LubhYR8WkrcANwLlDcad8tQCswGpgDPGhmi51zy81sDHBvF8e7xDm3vS8DH0jO4URDjYiIdyrQKCSTToNVD0J8LIw6yneaA5p62HAeKT+HquRGQkecG9jiDIBRM8/gzc0XUX54sIszAK45Yyp79tYGuzgDYOKpcGIKqk70nURE+kHVtNlUTZvtN0SkCM76OiQSUFLuLUYoZJx37vnezr+PiSdnrodHZsapU0d4zSAy0JnZb4GzgFJgO/AD59xt2X1nkXkTcQLwInClc25jdp8B3weuyh7ql8C1TusTioiIdCkcyRQ/NzU3Zwo0+mx2S0OxiIgvzrn7AMysGhjfvt3MSoGLgaOdc/XAs2b2AHA58NVsEcZpHiLvp2czMRprRER8y1uBhpn9k3Pu1/k6nnRhxFT46B+yE1AVvtMc0KjyOB+c9w227tzD6eNG+o6TU7RiFFMuvYGE50msnphxWAWJ8uAWu7yteAhMeS/E/U2SivSUxq7ea25qoKWxnsrho71lSKbS3P/qFhrr67nk5GmUFnmsP23JjidFnn8GblsCyTCUz/SbQ0T62veATzjnWsxsOvCUmb0KbATuI1OAsQD4DnA3cFL2++YBFwKzybw79ziwDri1f+OLiIgUhkgsBkBjUxNQ0QdLnGTf7+mzpVNERKQXjgBSzrk3OmxbDPTork8ze4hM140jzewXzrk7unjOPDKv06iqqjrk+Yp0tua+paVlv2OU4TCgobGBdB7mQ9rf+TpQ1oaGhl6fIx+UI1gZQDk6C0KOIGSAwZXjoGcwzGxGV5uBfwU0ydXXzDIfAVdeFOGwihjhUPCzisjAp7Gr77x651co370UrrmPyqF+uiW8tmkvd720ibJQKxWVFXzwXeO85ADgfz8FGHzY4z+rmrXw8LUQGwkf+jnESv1lEZE+5Zxb3vFh9mMKcByw3Dl3D4CZXQfsMrPpzrmVwBXATc65zdn9NwFXowINERGRLoWjmQ4azU3NmQ1a4kREZDApA2o7bavlnRqFnJxz7+/Bc+YD8wGqq6tdefmh3fjT3hQxVhRj/2Nk5mpKi0vgEI/flVxZD/XPkW/KEawMoBydBSFHEDLA4MlxKLeYvkBmPa3OM+8Tex9HBoS2Zmru+zKJmm2ET/snSmZd4DvRAS1fvoRtj/2Y1vGnct7FnyAU4IKSJ1a+RW1tHR84vizQhS+bdjfy1LItnD17ImMrOy/VJ+KNxq4+UjTpBBLROMUlZd4yjCgroigSImphxlbGveUAMsuR+VY8FMpGQfF4CBf5TiMifczMfgZcSWaN5FeBh4AbydzRBYBzrsHM1gIzgZXZz4s7HGZxdpuIiIh0IRrLvM5obmnKbMhOgJnaxIuIDAb1QOeW5hVAYNty567309glIuLboRRorAD+zTlX03GjmT2Yn0hS8HatYvf619gWOoySF38T6AKNN5/5H45oXEpi4y527P0wY4YFozKrs0RzG7c8uYYSa+PoSWOYNjqYOQHueWUTz76+mUQ6wqfOmOo7jkg7jV195NjzrvAdgUkjSrnpQ7OprU0wfcJwv2FO/Fe/5wcoGQaX3pFZEi3scbkXEekRM5vonNt4qN/vnPuUmX0GOBk4A2ghc4fXzk5P7XiHV+c7wGqBMjMz5/Z9Ky9frXbb5WoTWQj7gpJjoO8LSo687is6LDM2d7FvdLF7+/9WvrMML8oeu8P5e3M87cvvv1kpHJFYtoNGc7aDBu0FGvmSHX7VQUNEJIjeACJmNs05tzq7bTawPMf3eGHZ7utdjyYaY0REguJQ3rU/B9jv1aVz7vzex5EBYegkKkeOw9W3EJ9+nu80OQ095jx2Pb2YmmEnMrMyuC3gy4oiXDhnHPX19VQNK/EdJ6dzZ46hramRc44a7TuKSEcauwa4sZXFlIWSvmOwY3dmvnPUsEq/QQpkSTQRATJdL4aZ2Wedcz85lAM451LAs2b2ceAaur/Dq/P+CqC+c3FG9th5abXb0aG2wA3KvqDkGOj7gpIjb/tatu7TSrrjvreabJ/H+cxS05I9dqfz98W5Bvu+QmjvLYculi3QaG1uzGzos0IKTZ6JiPhiZhEyc2ZhIGxmcSCZ7Uh4H3C9mV0FzAEuAE7xl7YXVAwoIuLdQRdoOOfqutpuZv/knDuoBdfNbBqwFLjXOffx7LazgFuACcCLwJW9uaNMPIhXMuqf7qS4Zhvloyf5TpPTSae+l5YTT6elsYFoOOQ7zgGZGZefPIlEIkE8GvYdJ6eZh1Uy4b1T9QaUBEo+xy4JptueWcfe2jq+8P7Z3paB2rx9B1vvvBKA1ivuZPyYkV5yBEZbMzz2NRg7B4693HcakUAxs1eAV8gUZ7T/cncdcEgFGh1EgClk7uR6u8WRmZV22E7282zg79nHgbz7S0REJCgiZSMASDVmmzK6dN+cSJNmIiI+fR34VofHHwe+Tea12qeA24EdQA1wjXMuuK+hco4nGmtERHw76BlpM5vRxcdM4FB6et8CvNTh2COA+4BvAMOAl4G7D+G44lnKIjSHigvi7t1U2pHW7yQiA1qexy4JoM17mti8p5G0xzc0GxsbiacaKUo10tSkVtakWmH3eqjd7DuJSBBdAjwGTARKzGwRUGRm7zWzHrXgMbNRZnaZmZWZWdjMzgU+AjwB3A8cbWYXZ+/6+iawxDm3Mvvtvwa+aGbjzOww4EvAHXn9E4qIiAwg0coxAFhD+wpi6qAhIjLQOOeuc85Zp4/rsvt2O+cudM6VOucmOOd+l+/zm9lcM5tfW1vb/ZNzHae7J6gYUETEu0NZ4uQF4F72/zk/8WAOYmaXAXuB54Gp2c0XAcudc/dkn3MdsMvMpnd4M3FQW7OjnlCyOdDdCWr37GL1rz5JJN3C+slnc8LFn/Md6YDueX4l21/4A22jZnH1JedTHo/6jlTwWpIpVmyrY05xKbFIcLuSyKCTl7FLguvr5x9FbV2d125I0yZPZPmFt4AZUyfpnxbxCrjsdxDW2CrShZBz7l7gXjObR6Y97irgM8C7zCzpnJvWzTEcmeVMbiVTeL8R+Lxz7k8AZnYxcDPwWzKdCS/r8L2/AA4n080Q4LbsNhEREelCfOhYAMINOzIbsh00LG8FFbbPcUVEZPBxzi0AFlRXV1/d22NN2/U4tIyDorIOJ+jtUUVEJF8OpUBjBfBvzrmajhvN7MGeHsDMKoDrgbOAT3TYNRNY3P4gu7bX2uz2QV+g0dyW4mv3L2Xq0Ag3fii4bdM3LX+ekqat1JVNoWLlPbj0Z7BQ8Cbq02nHur8/xBVNf2DljtUs2/JuTp4y3Hesgvf0G7u4/ak3+OTZMd5zRHD/ncqg0+uxS4It7fy/zjQzjp4x03OKgInGfScQCarfmdkE4HUgDgwFmp1zFwGY2bDuDuCc2wm8J8f+hcD0A+xzwFeyHyIiItKNoopRpJ0RadqV2dBXdx/rrmYREemlY0LrmPvG1+GhZfDBW7t4hsYaERHfDqVA42ygsfNG59z5B3GM7wC/dM5tsn2XwCgDdnZ6bi3QZbuI7N1m8wCqqqpIJBIHEWF/DQ3BbkfunOPy6lGUh1O9/rP2pZKRk9gTH0MYqBl1EvUBvq5lVcew6M3TeLNsNufE04G+rhD8f6MARwyL8IGZw5kyJBz46wmFcU2hcHJCYLPmY+ySgNpR18QTd1xPEW3MvOhajhjvqTisaQ888u+Zr8/7LhQP9RKjuTXJo3+8nVDpCP5x7gexAlhuTGSwcc6daGYR4BjgWTKdLsrN7OfAouzHbo8RRUREpAOLxGiyGNbS/j6HljgREZFgKqMp88WBlpxVMaCIiHc9LtAws7OAa7MPXzez18h0u1jmnGs7iOPMITNR9q4udtcDFZ22VQBdzvI65+YD8wGqq6tdPpb9CPLSIQD/eFwFiUQi0DnLj5zFrvIfs33zOmbNOpGieInvSAd01QVnsnLb8cyIJJk0doTvOD0S5L97gPJyGFYaC3zOjgola6HkhOBkzdfYJcG2YctbzE48STI2hHWrl3PE+DP8BKlZCzWr3/l6fLWXGJt37uHwtXeyq3gK9S1ztXyXSEA555LAq2bW6pw73cz2Ak8BxwEfJvOaSURERAKihSLSbdlJr7wvcZKlSTMREemlA9+mozFGRCQoDqaDxp1kiiFeB44GPkhmmZLRQNFBHOcMYBLwZvaOzjIgbGYzyKyffEX7E82sFJgCLD+I40sAjDhsIkXlwwJdnAFQFAkzu2pIQXR6EJFDkq+xSwJs5uHjuX/yJwm1NXLWsSf4CzLmGDjqA4BlvvZk8tgRPHvG94jFilWcIVIYvpD97JxzdwN3+wwjIiIiXWsLFcHbBRrqoCEiIoVKY42IiG8HU6ARc85dn/363vaNPVkfuZP5wF0dHn+ZTMHGNdnHPzSzi4EHgW8CS5xzKw/yHOKTc+x6+jZ2b1pB7KyrKBo7w3eiA1q9vY6HnnySqqrJfPDUYwLbBj7d2sTK+26kuSXJjAu+QHzIaN+RRApFvsYuCbDSoggf/+iV2Q5THgsDI0Vw6mf9nT8rHDLec9rpKj4UKRDOuTuyXx7uM4eIiIjk1haKY8nmzINsB41Ky9cSn9nJMnXQEBEZtMxsLjB36tSph36Qra/ym+h3D3SGzCeNNSIi3oVy7TSzOWYWzz78nZnt12bXOXdQayM75xqdc9vbP8gsa9LsnNvpnNsJXAzcCOwBTgQuO5jjSwDUbWXvC3dSu2U1bz3xM99pcnr4qb/y/nU30vDKH9iZaPEd54C2Ln+W0NqFRHYs5s0X7vcdRyTQ+mLsCqJEcxtv7m70HYNV6zfywssvkUr7e3HnnON/X93CgsVbaW5LecsB8PoLj7DixUe9ZgDYvKeRvY2tvmOwsaYhEDlECoFzbo/vDCIiInJgEXOc2vI0NO6mvaDixNBKaK7L41k0aSYiMlg55xY45+ZVVlYe+kHamglZd0V/GmtERHzLWaABPAAkzOwNYDrwezP7nJlNyVcA59x1zrmPd3i80Dk33TlX7Jw7wzm3IV/nkn5SMpzY8EmUxEKUTjnJd5qcJh4+nWfjZ1Az7FgqS4LbBn7Y+CNIxSpIh4sYMnmO7zgiQdfnY1cQ/PLZ9fzo0ZVs8likkUo71t37Dcoe+Ryvr9voLcfanQ389oWNPLp8Oy9v8De/uXPrBkILr4PHv8XOrRu85WhuS/Fv9yzh1r+u85YBYE9DK/92z2L+a+FqrzlERERERPJhdNvmzBfP/3Tfua1kPm74ab+rOZ2HY4mIyKBVPCTHThVmiIgERc4lTpxzE8xsCHAMMAvYCHwY+I6ZOWApsNg593/6PKkUjmicCVf+kkTNNspHTfSdJqcPHD+VXUd9h1CyiaJI2HecAyoZOZEZn/0jdbV1VI4a5zuOSKANlrHrzOmjGBpNMaqiyFuGkEHyiPN5461VnDHa39JLYyrjjBtajGuFySNLveWoGDaKNSOOBYzDh43ylqMoEuLiY8dRGfXbTaSiOMoH5oxj2qgyrzlEREREZGAws0rgcWAGcJJzbpmXINHiToUUeZzwUtt5ERHpjXiuAo0sjTUiIt7lLNAAcM7tBZ7JfrzNzCaTmfg6pm+iSUGLFEHJcDDznSQnA0ayhwTBLc5oZ0XlhIp9pxApDINh7Jo1fgiTK8Nei8vMjLkXfNjb+duVFUX48YfmUJdIMKTS3w/KongJJ3/S/9JeZsYl1VUkEgmvOcIh4+MnBbtQU0REREQKSiNwPvBDnyGSkVIibR1+107nszBak2YiItILOTtotNNYIyLiW7cFGgfinFsPrAf+lL84kpNz8ObfIB2H8mN9pzmgVEsDy3/9JVzDTsqOvYQpp3/Ed6SupdPw5I00r32OVFkVvP+bMKTKd6ouOed4/rmnaKqv491nnU9R9JD/64oMahq7+kB71b3ngjzbvZZQ3R6oqPabpXF35nPJMG8Rkqk0z67ZRTTdyinTy7CAF0uKiIiIiPSEc64N2On799smi1OW3svbKVweCzQ0ZyYiIr0R7XDjUufxUp0zREQCI+Q7gByEXW/A49+Ep2+C1kbfaQ5o09JniO5YTIMrIvm3W4M78Ndtpmnts/x9V5Tdu2tgzV98JzqgDVt3MOSv32DE0l+wfMki33FERN7x3E/g7suhtcFfhtotvPnreWz60w0kN73sL0fTHrjv6sxH0x5vMR5//S3+a+FqfvnsOpZvrfOWQ0REREQGLzP7tJm9bGYtZnZHp33DzOx+M2sws41m9lFPMQ9Jc2sS1/G9LnXQEBGRQhLU+RoRkUFEBRqFpHwMjJwOY4+BSNx3mgOqGDUBF4pSkqqjpXKy97uqDyg+hGi8lMPj9ZQVhaByvO9EBzR8aCWbS4+mJjaOUYepXbyIBEg4CtE44PFnvYXAwpnxJuRzyars+UMRfF6PcMgIWeZyBHUIFhEREZEBbytwA3B7F/tuAVqB0cDHgJ+b2UwAMxtjZs928TGm35IfwLJTfwJAc3PTPgUaLp8FGpo0ExEZtMxsrpnNr62t7eMzaawREfFN6yQUknglXHAzJBIQCm5tzbAJM3Af/Rk7Nq1j6rHv8R3nwOIVRN7/A8avfoxE6SSYepbvRAdUURLnzM/8jLq6OoYP7ck6ciIi/eTkT/lOABVjmXjlbSTqaomMn+UvR/EQuOi2zNfxCm8xzjpqNGVFESLpFmaM9ZdDRERERAYv59x9AGZWDbx9R4yZlQIXA0c75+qBZ83sAeBy4KvOue3AaR4idys5+Qx4Dpqam3Cl6be3u3Qqj+XZmjQTERmsnHMLgAXV1dVX5+mAB7ddRET6jQo0pE8MH38ksZKhRIuDPTG0qGE4L7edy6gWx3nJNPGozzuvc4uGQ8Qiwc0nIuJVxTigzHcK1iYyBZRTPDa6CoeMU6aOIJFI4Ht9bhERERGRTo4AUs65NzpsWwz06A4fM3sImAMcaWa/cM7d0cVz5gHzAKqqqkgkEr0K3NCQWc5xVGkMgLdq9jI+En77TdW6RB3heO/OUYbDgMaGelJd5G3P4Jty7CsIOYKQAZSjsyDkCEIGCE4OARUBiogEhwo0JP/qttL8wJeob0xRcuQJhE//UiB7rL+4roYfPbaKkW4PreESVuxs46v/MD2Qk1nOOV5++UUaErWc+p5ziIaD20ElUbub1a89y5HHnk5pubp9iEg/aK5j22+vpiEZoexDN2DDJnuJ8fyaXfzXwtUAfP7saZwydYSXHIna3bx+x2doLRvPKVfcSDji6de9VBL+/gsYdRRMOdNPBhEREREJmjKgc+/2WqC8J9/snHt/D54zH5gPUF1d7crLe3TonMrLyykvLQWgrrGZaLTy7X3F8SKKen2OzHtRJcXFcIBj5ePPkQ/Ksa8g5AhCBlCOzoKQIwgZIDg5pJ0KNUREfAvuLK8UrjceYcuWzayoKyKx7FGo2+o7UZeeXLWDCWznS3Xf5Vz3LK9u2kt9S9J3rC69uW0n8ce+wpBFP2XpkkW+4+S07P4fEX/hv1n6wE99RxGRwaJ+B001W2iu30NbzQZvMdbXNJByjpRzrK/xd4fInrfepDKxhrI9K2huqveWg5Y6WPkgrH7MXwYRERERCZp6oHO71Qqgdy0o+kMoRJIwTU2NONdxiZN0jm86WJo0ExGRvpK9MVVDjYiId+qgIflXOpqymDEsXU80XgxFwayQHVYaY6WrYG3xbN4MTyLqjFgkmDVLQyoqWFM8mag5Jo46zHecnEomzKKpZiWlE2b7jiLSJ8xsGrAUuNc59/HstrOAW4AJwIvAlc65jf5SDjLDpzDq3C9SV1dPbPIp3mK8b8YYVmyre/trX6qmzmLF2d8mFinx28moZBhcNB+Kgr3cmYiIiIj0qzeAiJlNc86tzm6bDSz3mKnHUhZjdPM6Yi/84e1t6VQeb/ZxmjUTEZG+ojFGRCQoVKAh+XfkeYxONlGyczOlR58L8WBOzFxyXBXLt9Zxe93lVIZSfPo90yiKhH3H6lJlWZxTP3M7dXV1jBo+1HecnGafeRmJ489X6zoZyG4BXmp/YGYjgPuAq4AFwHeAu4GTvKQbjMwom3Iqrq4WwjFvMUaWF3HDhcd4O387C4WYcdJ5vV5rOy8qx/tOICIiIiIemFmEzPuOYSBsZnEg6ZxrMLP7gOvN7CpgDnAB4K/S+iCkw1HenXx53235LNDQ5JmIiPQ5jTUiIr6pQEPyLxSGYy6BROKA62YGwbDSGD+6ZBY7t2/GokWMHT3cd6Sc4tEwbTH9lxXxycwuA/YCzwNTs5svApY75+7JPuc6YJeZTXfOrfQStD/Vbob6t2Dccf4ybPwbTY9dT5NVUn78hTD7Mi8xmluTPHH/fADO/OA84h5/Zq9d8jwuUkz5jHd5yyAiIiIig9rXgW91ePxx4NvAdcCngNuBHUANcI1zLq8dNMxsLjB36tSp3T73YLjQ/gXhqXwucaIOGiIi0teaa30nEBEZ9IK5noMUvCV/vZ/FC25m59YNvqMcWDqFW/ht4v/7L4Qe/XfYs8F3ogNqak3xm4f/yn0Ln2ZXfYvvOLk5B017CuJNhURzG63JlO8YUiDMrAK4HvhSp10zgcXtD5xzDcDa7PaB79kfwyP/Fxpq/GV442E27GlldS0kX/+ztxir16xiyuo7OXz1naxZ+4a3HM1NDbQs+Ao7F/63twwiIiIiMrg5565zzlmnj+uy+3Y75y50zpU65yY4537XB+df4JybV1lZmd/jhqP7bWttbcvnGfJ4LBERkS788RO+E4iIDHq6HV/ybtf2TYSfu4nyaAVrHniTkZ/8ue9IXdu5it0rn2VVQymVLTWMWHof4dO/6DtVl55bsZFjFn0LV1TOo/FSPvYPZ/iOdGDrnoKnfw5nfh4mBrdDaVNrimt+u4hjRsW49gO6w1x65DvAL51zm8ys4/YyYGen59YC+7UQMrN5wDyAqqqqXi9B0dDQ0Kvvz4sZH4Px2yAVzXRO8mHUCQwbuoKoldA0/gRvOUaMGM0rY87HMI4dPsrrEiMNJ36OaFFlMJY5CYhA/H8JEF2Pd+haiIiISI91saRia1seCzRcHrtxiIhIQcl796eNz8LLt0P1v2QeF8ANlSIig4UKNCTvikvLaQuXgYWxigCvPV9UTiwSptSaiIeKCZUGd4mToeWl7AqPIk6Y8iEjfMfJrWIcjDgCysf4TpJTNGy8a8IQJleqkZB0z8zmAGcDXVXz1AMVnbZVAPvNjDvn5gPzAaqrq115HpaByscxehfgWL/nB5h1PuWjJ1GSqKd8yvGZpbY8KC8v5x+v+qaXcx0JfeQAACAASURBVHd2/NmXkkgk/P/7CBhdj33perxD10JERER6IhkfDvUb99nWktcCDU2eiYgMVs65BcCC6urqq/N20D9/4Z0CDRERCQwVaEjelZYPYeK//IqtG97gmHed6jvOgQ2pYuh5X2XGa/fSMnwGNutDvhMd0LGTR7HsY/Opq6vl/KMm+I6T28gj4IxrIeATHZFwiC+970jdXS49dQYwCXgz2z2jDAib2QzgVuCK9ieaWSkwBcjrGsqSg0uzfuWrNDS2MGPCLEJFJV5i7Khr5n8eeRoz46PnvptRFXEvOQCat62iLWmB/1ksIiIiIlJIkhVVsGvRPtvK1jwAx70P9u20eIhUoCEiInkULvKdQEREuqACDekTQ0eO/f/s3Xd4HOXV9/HvvV29WpKLXHAvuGE6oYNDMUmAAElIIAVIIeEJSQhvkichhYTkSSGNYkggAULHAds0gzHFVPeC5d5kyaqWdrWrrXO/f6y6tJJlS3uv7fO5Ll3a3Zmd/VmWNKuZM+fg8GRid6T2t5gaOoN0u4uYygJnmuk4CSmlOL40F5/Pjt02EH/wCyH6aT7wRIf73ydesPGNlvv/p5S6AlgM/BRYp7UuS2rCY1jVpuWElt+DduWyw+Ng3LnXG8nx/PJ1nLftV/Hbmb/nhotOMpIjXLOTnQ/fQCyjhPyvPwwuMwUrQgghhBBCHG10bveLZorKHoVNF8KUT3Ve8MjlMHQGnP+zfryAFGgIIYQYQI6OFw/JPkYIIVKF9PYXA09r9PrniC7/O9RtN50mscZyqh/9KmVP/pjKF35OaM3TphMlFIzEeOGlxbz22os0+EOm4/RqW3UTi1duZ1etzHMXRw+tdUBrvb/1g/hYk6DWukZrXQNcAdwJHABOBq4xGPeY48rMR9ucWMqBJ9vcGKisrCy8KhuvyjY6LsGRloUjPQeVUQB2p7EcQgghhBBCmKKUmqeUmt/Y2Dig27Xljmq7fWP4u+0L/LXdV97+Orzzx36+gpw8E0IIMYAc/eyg4a8DyxqcLEIIIdqkdnsDcWSq2Uz10r9SEcnC3bCF9Gv+aTpRj4JlS6irrcXrKiIU8VDw3kO4Z37WdKwefbR6FWNW/oaYJ4e3Xrdx2WWpmVNrzTPPPM4n/c/z5NbP8sMbruv7SUIcgbTWd3S5/xowyUwakTdqGo4vPUBjYwMjJpvpWgHwmVMmsCTrXhTwmanDjOWwZRYy/ubn8DX5pUBDCCGEEEIck7TWC4GFc+bMuWEgt5tefFzb7Wqd175gQMabIPUZQgghBlanDhp97Ksa98GfpsC5P4EzfzCosYQQ4lgnHTTEwHNl4HI6SbdrbOkFptMkZHOlYVOaaCyGQ8fAlWE6UkJpmTlElIsYDpyZ+abj9EplFFBrH4IjI3X/74UQRx9fWOMPa7TBlsBuh50pw/KYPCwPt8NuLEcsGmXj+6+yq2wl2vBVDzHL7P+JEEIIIYQQA8kzZEzb7Vinw6oDNY5W3jsLIYQYQA7Xwa/rrYh/3vzy4GQRQgjRRjpoiIGXW0reFXfjqNqFZ8LpptMk5JpyKSVb3iR9XxmxzEzy5n7fdKSETpg8njW2+QR8Xs6fPdN0nISUUnz9qnl8vPskbhxdbDqOEOIYsWfLGrzPfIeoK4eNvq8x7ROf6vtJg2BdeQPrnvwFAP6rf8bxI3KM5Fi18F6yNj5GxDOEMqJMPnmukRwVtQ2s/tcPaC6YyhXX3YIaqKsKhRBCCCGEMCVnZNtNazCue9PSVl4IIcQA6tRBo68iwJblcvxGCCEGnRRoiMFRPAXSS8GTZTpJYu5Mci6/m5yq9fjsBdiGjun7OYYopZg1aRw+nw+nPbUb3+SkOZk2PIcsj7TVF0IkR6ChBocOY+kowYb9xnLU+8OMDm8BoM4fMpYj6q1CK4VNRwk29jALO0kam3yUBrewp96O1vL3vRBCCCGEOArY2w+lduqg0d83u6/8GPatgq+81Pn50n1OCCHEQOpUoNGHtiJBOYAjhBCDTQo0jjDRmIVlpf4fa817VlK3s4zMEy9CpafoSI5wgNhLt9NUUUYkvZisC38AQ6ebTtWjSKCBsuf/QChiMfGim8gaMrLvJ5kSqCe6ZzWMnQMeM1ePCyGOLeNnn8Pqqq8QCjYz+9zPG8tx2thCFp37RwAuHVtoLMfEi77B5uf9RNOHcOIZZrqJAEwaNYIt1/6HE7Ozsdnkj3shhBBCCHGUmPsbfrdozeGNOHnvb53vtxVmHMIxv1AT6JgcgxFCCNGdw33w62rpoCGEEMmS2pfii07CUYtv/Wc1D76zw3SU3vnrqHjq+3g/fJR9L/3edJrE9n5A3a51rD3gZlddM9ZH/zCdKKFtSx/Buf1V7FXr2Lnw/0zH6VXVCz9l34v/R/WiX5mOIoQ4RtgdDuZcegPTL7yOtAxznZsaAmFqgnZqgnYaAmFjOfKLhnPSV//IjHk340nPNJZDKcXEUcMpzkvhblpCCCGEEOKopJSap5Sa39jYOPAbP/Wb1Mz8FlbHooyBOpl1KB00fj8B7urlQp4/z4S7U/OCJCGEEO0GZd9l63CNdp/7mNblUqAhhBCDTQo0jiA2BUVZbnLSUnx0hM2OsrvQyobdnWE6TWJ2JzalUErhVFb/qkmTzOZKA0BhgTPNcJreKXcWyubAlsrjbYQQRxd/HTz9ZVh8K9RuNRIhErP4+QvrcK2Yj2vFfH7xwnoiMTPzo7fvr+PZP3+PRU/cT32TuVEr2rJY8/qT7Nq0wlgGIYQQQghxbNJaL9Ra35iTMzhdJTLcDqzeOmgkOgn23t/7KJY4hAKNiL/35Qd2QsPu/m9XCCFEUg3Kvkv14xRgWwcNOW0ohBCDTUacHEEcdhu//PQ0fD6f6Si9S8tl1Jfuo7Z8G8VTPmE6TWKlJ5M/9VxO2PYWoezjsJ3yDdOJEhp/9rVssaIQDDPx/GtNx+lV0aU/I638Y7JKp5qOIoQ4VlR/DL4KsOXC3g+gcHzSIzQ2R8C7n7MjbwKKj7yn422OUJCZ/OK/LR+vZZb/bfyxIrbt/SwnTR6d9AwAB2or8bz3J/bnTGL05IeNZBBCCCGEEGIwpLnsnUecdO2goRMUa7/yo943fCgdNIQQQogBISNOhBAiWaRAQwwKe8EY0l2F4PSYjpKY3Ylv6hepdk5AZxaTlzfadKKEfDE724ddRjQYoDiWxlDTgXrj9MRPjqZwRxIhxFFm6HQomgohC8acaSRCXroLd0Ep/w1eBgo8BaXkpruMZDl+xhyWbf4MKqOQeWOGG8kAkFc4lOq5v2LkkFJjGYQQQgghhBgMGS57lw4aXcQih7hlKdAQQggxgPpTbKFlxIkQQiSLFGiIQbFj5w4qd21m+slnk5WemiM5qnasp+apW7BZYcKuPNbuWsmMy79nOlY3vmCEXz/zDidXPobH6eSPqy/klqvmMiIv3XS0HsUszX5vMxkZmdhsKf5mTq5MEeLo4MmBeX8Cnw+yzIxXstsUP7tsGks+LgTgK1NKsBv6HTiiIJurbrgdf5OPbI+5sWjKZmPSiecbe30hhBBCCCEGS7rLQUy3F2hYVpdyDesQCzRaj1M01cBrd8Alv0/5UbdCCCGOFH0dC5cOGkIIkSwyTEoMuIamAPVPfpOMFX9j+YJ7TcdJaN9b/0bHogTcRTQ7c3GWLSAWbDIdq5s1exuYXrWA6bqMEVRxccNjLC2rNh0rof98sJtfLdrEc6vKTUfpXTQEz3wFPvqH6SRCHNH8oShV3qDRDLpxHxX//CLVT36bSMV6YzlyrEYuLb+bS8vvJsdqNJbjve11fOfBV/j1fz9kW7W5/VpzOMbCd1ay/ONdxjK0qmhoJhSNmY4hhBBCCCGOEiPy0rA6XGG8vyEQv1GzJf65vx00Wk+GtY5GWfJTWPMobFxwmEmFEEIc2w6lg0YCe96Hhr2HF0cIIQQgBRpiUCh0645fpfC3WIcrm9tupWB1qEKhsYGOf2UtbcOWgjlb2WwKpUClcMY2NgfY7KZTCHFE++OSLXzn8dU0Bg61he/hq13xLI2V26hraKR22X3GcrDpBfZufJe9G9+FTQuNxXj83S18p+G3XNTwBM+tNlcs996GLYxe9h3qF91BjS9kLEdFQzPfeXw1T69I8cJBIYQQQghxxLhgSjG/uXJm2/16XwA2PAd/PxG2vALWoRYH6y6fhRBCiMPQ6Rh9H8frW4sEE53T+edc+PP0AYklhBDHOhlxIgZcbmYahZ+/j8pdW/jEiWeajpNQ6ZnXU/XkStJC1UR0HtFpn8XuzjAdq5uZI3NZWPJpXBUhPNh5K/9Cbp1UZDpWQp87cSSnlWYwemiB6Si9c7jhigfiIxGEEIfsE+MLKc52k+E2V+yUXnQcTruNNCK4iycYy0HeceRmtrQfzhtjLMbwwhzerTqZJnsJo/LNjcMaWlTEqvRT8GaPJ8tj7i1nfoaLsyYOYfpQaQ0thBBCCCEGhlKKcycPhUXx+w3+ZqhYFb9TsxlKjm9fuWYLDEnwd4rWnU+etV69LCNZhRBCDIiORRkDMOKktYjjcO39MH7sLHNIz8vLV0DBWEjLG5jXE0KIFCMFGmJQjBk5isK8fDLS3KajJDRk9BTcX/031dvX4knLZezxp5qO1KNMt4OfXHk6H20qIRJo4rZpUynKSd2TTDabojDLfWR00BBCHLazJxZx9kSzRWMZ0y5hVFYhvqYm8iefZS7IcWcx5Jq/x28XTTYW4+ZzJ7Bs+C04rTDnHD/CWI7JIwoYffPvcNoVDru5jloep51vnzsenxTkCSGEEEIcM5RS84B548aNG8QXaX+P29AUgPwoANrmoN7rp+2ylb+fCHckGIFoRcHu7FCQ0fXkWZdjK7XboGk/jD7jcNMPDH8tPHcjXD4fMgpNpxFCCNFV6zH69++FaIcRxZXrYGiXbhht+6J+HtfvWmx4MP5xAeSMhO/2MKpYa3jwPBh+AtywtH/bFUKII0QKz58QR7JdtX5W7KqnOZza894zsnJxqhjp6ZkpOd6kVfq2Rcx899tMXncXeWvvT+krKfZsWcO6JY9Qvm2D6Sh9amyOEIqm9veoEOIgKIUzbwTO7KHxg5uGbN/wASv/dRsr/3UbOzZ8YCyHx2lnhLWPYtWA02BhRCQcYvuHi9lb9pGxDEIIIYQQ4tiktV6otb4xJydn8F6kw8jUDXtqeXT5VgDe3eXlir+/2TVQz9uIRbqvt/s9qNnU8/p/OwEevuRQEw+8D+fD9tfhwwdMJxFCCNGbl2/vfP/+T/Sw0kF00OjJoZ6raNyTYHstXTr2rTy07QohxBFACjSOJLEIvPAdWPmQ6SS9agxEeP4/92J/9Uc8tnSF6Ti92vzcnTS/eidVC3+BFWgwHadnoSZ8b93DpkYnO/we6la/ALVbTafqUbPfR8Ozt5Kx40Xqnvku4VCw7ycZsq+hmW89torfv7KFaGyAWrMJIczw7WfPQ1+mfMFPiO1bYyyG96U7sMeasceaaXzxDmM5PnpvKfbF/0N4yS/Yss3c/mLN4vk4lt1J4Pnvs29HggPMQohDppRyK6X+oZTarZTyKaVWK6Uu6rD8PKVUmVIqoJR6Qyk1qsMypZT6rVKqruXjd0ranwkhhBD948psu3m78wmudbwOQGzvSk6xdXn/G6jveRtWtPtjD30SKtcOVMq+aQ3R8CE+Wd4+CCFESmvYAx89eHDrttVZdPndvuUV2LV8IFP1kSN1L04VQoiBIgUaR5pgI4QDplP0KqY1DquZNB3Aikb6foJBOhpGYYG20AM1P22gaQutNRoV/9B64Ga9DTCtLWw6hsaG0tHU/ZoC0ZiFpTXhWKzP6XtCiBRnRbGsGFYsho4d6oHFw2ezYmjlQCs7Nt3DgdYk0ZEwSmtsWFhRczmsaDi+59IWlsH/FyGOYg5gL3AWkAP8L/CUUmq0UqoQeK7lsXxgBfBkh+feCHwamAFMBy4FbkpedCGEEOIo0KGDRkdnBl7lLmeXk2FN+3veRtcCjWSdlAr52k+2vX8P/GoI+OsOY4NyZEUIIVJSTRks/t7Brdt6LL9r7f5/roKHL+7tif3L1Ne+ru2cghQBCiGOXg7TAUQ/2J1w5UPQ1GQ6Sa/yM1yce/Ut7Ni3n6unjjEdp1fjP/Njdn80k4Ki8dgz8k3H6Zknm6yTr2Xie4/QZA9QMO4cKBxvOlWP0jNz8FzyaxrWv8mwmefh9qSbjpTQqIIM7r56JtFQs9ERAEKIAZAzglFfvBdvQyOO0hONxXCd8wMiS++K3z7vR8ZyzDnjQlY7XeDwMHHiJGM5Zlz6dT7OKSY9r4TS8TOM5RDiaKW19gN3dHhokVJqJ3ACUABs1Fo/DaCUugOoVUpN0lqXAdcBf9Bal7cs/wNwA3Bf8v4FQgghxDEkEoRlv+3+eGuBRuvJsEgvF2XV7xi4PM/dCJtfhB9sh7WPxx/zlkNGwcC9hhBCiH5RSs0D5o0bN85QgtbCiUEecWL1MXI8hS/6FEKIgSIFGkcam63/M8AMmDw0mxGZiqw0p+kovXK5PYyfdiI+y206Sq8cM64my54OgRCumRclvEojFYybcTrFx00nKyvLdJQ+FWV78PlSu8uLEOLgqCETsXl8RveRk7NC6KEZ8TxZ5kY8BaMWjaRjsxxEYhqXw8zXxJOWwewLrzXy2kIci5RSxcAEYCPwDaCtN7rW2q+U2g5MBcpaPnfsnb625bGetnsj8Y4blJaW4vP5Diun3+8/opelSo6jfVmq5BjQZe5h0PLz03VZcZpu+9ka6CwF7pZtd3j9w9meLBvY71lxFBhxEpR/2Pd69dth2a+7Px7rckzCX935vlKwaRHO6m3wxs8OPWdX+zfEP0cCoFqOMfV1wqwnrX9/STt6IYQ4bFrrhcDCOXPm3GAoQPzzYB9b01KgIYQQUqAhBoVlaZrDUVL5FL1uqmH3o9/EX18FGYWMOO8mcqbNNR2ru0gz+/5zM959ZYRc+TSveZbSL92Pyio2naxHe+sDrNtZzcyxDobnppmOk5BladaWN+Cywkw9AopJhBCJWZZm4XOPEAs08smrv06620BxYN12rPfvZYvPBcCE9+7BVjwF8o9LepT/PvlPZpf/m6C7kEUHvsnlF5vZt8UszdMr9nLckExOGpOiXaqEOEoopZzAY8C/tNZlSqlMoKbLao3Q9vY8s+V+x2WZSimldeczLFrr+cB8gDlz5uiBKMLtbRtHwrJUyXG0L0uVHAO2LFQBHe53XFbVrDrdH8gsdaGWbXd5/cF4rWN92aFuTxzhvvoq/Dy37/W8+3p+fP5ZcMkf2u83dd19A09+AU+i7R5KUQW0XxytdftFQId0QqxtQ4eWQwghRAo5xA4a/d0H9Lnvkn2KEOLoJ339xaD402tb+NGCDWzY19j3yobsf/cxAnUVBFwFNETdNL72e7BSrzqzecd7ePeV4XMVEXAW0NRYQ+3K503H6lEwEmP+0y+QsewO7n1qMZFY6n09Wy3ZsJeKx7/DioX3sbM2ta9oemlDJf/3chmNAen2IURP/OEoudsWMKp6CRX1XjMhAnVELKgOaKoDmogF+GuNRPHUrsdpt+EhSlNlmZEMAI3NEZ5dVc6L6yuMZRDiWKCUsgGPAGHg5paHm4DsLqtmA74Ey7OBpq7FGUIIIYTow8FeZfzaHT0/7q+Bp77UPtqk+cDBba91l92xA0e/jmm15Lai7R00oqF+PL91M6nf5VcIIcRBOtQOGv0ecRLtY3vWoeUQQogjiHTQEIOi3h8mGrPwBlP3hHIsEsTChlKglT3+R622SLW6JSsaQev4n85KgdYKK9JsOlaPYpbGHvaRpxtQkSZilsaZotNYmprDlMb2E7IKCIT7eFNoWI0vRJUvRCgaA1J7bJAQJmR5nOR9+i58DQeYXWKoU0PBOFxpGczIiRcmutNyocDMzFDP7Guo/WA3fvtQJpx0kZEMAPkZLn57xXRy013GMghxtFNKKeAfQDFwsda69c33RuC6DutlAGNbHm9dPgNo7ck+o8MyIYQQQpgS7HKhU6KuFrEIOFwQC7c/ZkXB1vLee/H3wOGBC3/V8/NbT3pFQ+0dNKKHcaxJajyFEOLIo3XnIojW0SNqkM+PyIgTIYSQAg0xOG6/aBJ79tcxZVSB6SgJFZ90JaHNr2NrrsblKSDthM+BPfV+JDLGnEhG7hBorCZEjHSPm4IZF5uO1aMMt4N5865g+cezuPL4sXhStToDuGT2GF5xzCfXrpkytOsFpqnli6eM4uJJuRRmJ2xqKsQxb/qkifh8PpSp6vr0fDjrNpyv/TZeUXfWbfHHDLjkrNP4YOTD5OsQJ4wbZiQDgNaaTI8Dp12ueBBiEN0LTAbO11p3PKuyAPg/pdQVwGLgp8A6rXVrW51/A7cqpV4k3j/2e8BfkxdbCCGEGFxKqXnAvHHjzBRNH7Ktr3S+H0tw4ZMVAVydr0IO1IE7E9xZ8NGD8cfO/3nPz289+RYLtXfQSNGLgYQQQgwSbbXvA1rvA4M/4qSPAgwp+hNCHAOMtApQSj2qlKpUSnmVUluUUl/rsOw8pVSZUiqglHpDKTXKREZxeLI8TkYWZJg7UXYQnEPGMeYrDzNs3k8YPu9HDDnrRtOReubJpWT2PErcUYqcQYZMOwdHXqnpVAmdMDqfL35iIjNK80xH6ZXHaedTJ4zhxLFDUvr7FEAphduRusUuQhgX9uNb9BN8r/wKfFVmMoR8VC76NTv3VbKzvJKKRXdCyNf38wbBwteWYX/q83hf+BHvrDF3QfxTH+zgsX/8mT/88zG2Vpn5WgAQCcK6p6F2q7kMQgyClr+TbgJmAvuVUk0tH1/QWtcAVwB3AgeAk4FrOjz9fmAhsB7YQLyI4/5k5hdCCCEGk9Z6odb6xpycnMF/sa8uAdvgdLv89/JtPS9o7ZzRsajij5Pg7uM7r5ewjXxrB40w2Gzdt3XQUvt4ihBCiBYlx3d/rGunCqu1g0aKjDgRQoijmKlZDr8BRmuts4HLgF8ppU5QShUCzwH/C+QDK4AnDWUURzutCW1dSuT9+QQ3Ljr4OZ9JFlrzBPve+ifbwrnsjBZQseolmpb+3nSsnlkxmt+dT+MLPyL4wUOpX+3qqzJ2AlUIMYD2raJmw1Lqdq0nvPmVvtcfDAd2422op8lVRJOrCG9DPTTsMRIlvO45clQzRVYNO1a+ZiQDwMcr3+KKwNNc1fAgb5ZVGsvB7uXwzh/hAzn3LI4uWuvdWmultfZorTM7fDzWsvw1rfUkrXWa1vpsrfWuDs/VWuvbtNb5LR+3aZ3qb9yEEEKIFFV6UryD3iCoqKruecGfZ1C54gWCgS7HNLoe2+p4Eizsh1+PgLIX20++deqgEWhfd/H3YeXDnbfVsDe+jR7J2wghhEhtClxZnR/qWgiRrMKIPkectO5TpAhQCHH0MlKgobXeqLUOtd5t+RgLXA5s1Fo/rbUOAncAM5RSk0zkFIfurS01/OX1rTQEwn2vbErVRqrffIBttc3UbF+Ld/kDphN1pzW+j56g2srG5srA7vRQQz7eTUtTsqBE73qb/W//i92V1ex/85+w9wPTkRKK7lvL7vnXUPXcD6G5wXScXi3dVMVfX9+KN5igtakQx7rC8eQWlpCdlY1zxCwzGTKLSHc7UWEfKuwj3e2EjCFGogSLZxGJWQRwkVU61UgGgKwRU1jjmMbb7rOYMNRgV6WS42H0GTDufHMZhBBCCCHE0a0pQSHFYcpWgZ4XBBvJWfg1rvhL94Ls//fcuvY7sQ7H5Rr2QtgHr/6Ezh00ehhx8tEDsPAW2LSo/bG7p5H+5JWdX6z13JnUeQohRGqLRdo7JrVKVKDR727TLfuA6k2w/pm+V5cOGkIIgcPUCyul7gGuB9KA1cCLxFvwrm1dR2vtV0ptB6YCZT1s40bgRoDS0lJ8vsO7Et7vT1QFnlqOhJx7quqoPeCl5kAD9lia6Tg98zbis+UR9mSj7Q58/mbUYX4PDTitadYeQq5stNNJ1JGJdkfxO1z4GhsgauxHuEfa10STPZ+gqwBfNIDP64VU+5q2CNdV06TTiEUd+BrqIJq6I0T2VNVT2+Clrr4RleU2HadPR8LvKHGUySoh//rHcXobUflFZjJkFlEy93tkLvlN/O4F/w8yzWS5/MprWbc8D8uRxrwzzjGSAeDmi+ewaurvOS7NydRh2cZykFkEF/9fyu6PhBBCCCHEUcBbEf88fi5sPbyufndHL+d/HM8B8E3HCwnXC+EknVC3x3eteBlc8dt79+2ldUjui2v3cDHEu2B4Wt6fx0Lt41nqehin8uZdBMddhE0pXIC9en2CNF0KNAL14MoAR+ofwxBCiGOCFWnvmNRKWxD0gncfFE3uMOKkn9d1txbp3XNK/PPxVyZeF9pfJ/EG+/f6/XVgd/zC0ulXDe7rCCFEL4yd3dVaf1Mp9W3gVOBsIARkAjVdVm0EuvReatvGfGA+wJw5c3RWVo+r9ctAbCMZUj3n50+fyHmTixlaaPCK2b6kzWFE6Uiy9qxHe4Yx7NSrUCn4ddVjZlK08kXqI/lod4yc4B4KiseRVTz6EKpZB9nETzBy4wIyDlRRWDKcrPGngTvTdKqeTTuXDBc027LIGjrWdJpeffHMSdRMK6EklX+eukj131Hi6KK15s1tjfiafFyYXYDbYaDgKlBP/TsP0tAUvwog/M6DlIw+GdLzk5vDilH93x+Rs/MDwu58qkJbGD73luRmaKFjMaxdy4nkl6CGn2wkA0AkZvHRrnoKXZb8bhJCCCGEEIPjgp9Deh6UnnLYBRrf+Ol98Ovn+lyvXmeRrroXaDzuurPt9n0PP8ydLfUXDy9dx8VuCDd7abLSyQf+9PIGzmquYzbARw9inX4rd3/k59bWDVgWU376MkrH2O7pIUTLSTb94YN4Vz1L84k3U3LOTfC7MTD2PPhi3/8OIYQQSRCLtHdMaqUtePQK776X3gAAIABJREFUKP8Q7mjs0LlikM859PU6h9zJ4yA9eD74q6VAQwhhlNHL77XWMeAdpdS1wDeAJqDrJZbZgFzyeISx2RSZ7tTq7tCNw03u5X8i17sPX8yFKhxqOlGPss/6FtTvpHj3ewSsQjKL8si99OepV5wB4MnB/pl7iO7djb10FLhT+EoJmx3PpAuIHAFXVNttioxU/3kSwqC15Y389Y1tZNoiKFcal04flvQMkd0fcKC2kiZ3MQC6tpKC3R/gnHxRUnOE939MYOcH+J2FRGxZZKx5Bs66Hjw5Sc0B8O6rTzN83V+I2DzsKHyU40aW9v2kQbBoXQX/fnc3wzPh958vwONM3Y5JQgghhBDiCFU4Hj71dyhb3H3ZzSvgb3MOelNu18EdSynJUNx8/DD4MPE6s21b224fb9sBgCsWID8Qvz28djmzHe0bCP1pJo+G/sKtLcUYTaEwloYMuo8wjsYsQs3NZAAq7CMn7CPnzdvgrK/FV9j++kH9O4QQQiSBFe25g0Z5yz5Aa9CtHTQOccRJf7JA4k4dgz3ixN8ylsyyuo99EUKIJEmV3z4OYCywEZjR+qBSKqPD40IMuAOVO9i47CnKyz5M3XmZgXq0t4K6WAYHdDqxZh807jOdqkeb99Xzwv3/S9WiX7LggZ+zvarRdCQhxDEgzWnHYVPYbYp0l5liJpvdiQIsrbG0RqGwOVxJz6EcbuyAtizQMWw2O9jMfE0caZlYyk5EuXC7kv+1aJXhcmC3KdxOO7ZULG4UQgghhBBHj9xR3R9zpg/KS2U0V3Dih713y7vC/nbb7f91PtZt+VWONzvdT1NhVnm+3nY/s3ErY1QlaR0KNE688zV+93IZNzz8Ho8v39xtm5uWt49l2fV+4hEtQgghkigW6X58qGMhhLba729cAC/edvDb7u95lb5GqQx2gUayX0cIIXqQ9AINpVSRUuoapVSmUsqulJoLfA5YCiwApimlrlBKeYCfAuu01mXJzpmKLEvz9ze2suTj/aaj9OmDnXXc/9Z2vMGI6SgJ6UiQ8iduQX+8gPD7D1K+bpnpSD2KrHmS8jof9a5h+JxF7PIp/MvvNx2rR28tfpQ5gbfx2C1OanqDZS89ZTpSYlYM3vwtbDm81qNCCPMmlmTx809N5etnjeWciUOMZLCPPo3C0gkUh/ZQHNpNYel47KNOTXoO55DxeGZ8hrxYDdk0kXnWd+Kznw0447zLiM37G3lf+AfDS4qNZAC4YEoxv71yOt+7YCIuR6rUJgshhBBCiKNSyTRwdRmrZz+IYuVzfzI4eQbAG+7vMatDJ44aX4h7lm3n7j1X8jXHS93Wn/z69W23R7/8RdaVNyQjphBCiN7Ewt27RXQsrLCi7YUTAB8O4vmHvjp1JOtCWinQEEIYZOIotSY+zqQcOAD8HvgfrfXzWusa4ArgzpZlJwPXGMiYksIxi/e217Oxwms6Sp8+rvCyqdJLgz+FCzSiIexhHwFnPlpbBBprTEfqUay5kTAO7DZQShFRTmKhJtOxemQPNqBsdqLKhVI2rOYU/iM8GoJdy2H/OtNJhBADYFJJNpOHZqMMdUiwYhHqvM1ordEa6nzNWLHk7wN9wTDrd+3HsjQxy+LdzeVEY2b+4NRhP2neHbgathvtUqWUYuyQTLI8MipKCCGEEEIkwSV/iH+2t4wqsTv7fs6ZP+h1sZVZcpihDs8Drj+23b7hpEJu+sRoclTgoJ77/o66wYolhBDiYPU04uTZr7Xf3roEmqoOfnudjvMM1oiTQT7Gp2N9ryOEEIMk6UeqW4owzupl+WvApOQlOnJ4nHb+/oXZhJv9pqP06dpTRnHeuGxGFgxOG8eBYEvLwX7Cl8hZ/SjewhmMPuFC05F65B57BkVly9kTDhMmyki7j7QJl5qO1aOCGZ+kYfk7uMMHqCeL4TMuMB0pMVc6fPZhCMkbMSHE4dvz/gKshr3UpY8GIOPAXvZ8sIDRZ1+f1BwfvrOEsQ3L8XtKiDqzmLDnKdZvOo9Z06YmNQfAx0/+DHf5OzRhY2/sl5TOOj/pGQD27K9h7eM/wxp9Jp/6jNT9CiGEEEKIQTbj6vhH/Q7Y8iqk5fa+/uzr4p8/Mx+izfHb5/4Elv6qbRWdNQyaOnTUHToDKtfGb489D7a/3v+c+cfFM/bTjzd9GobNOuj1TRWMCyHEkUYpNQ+YN27cuIHfeCwMti4FGh33HU9+oX/b61igoTXEogf/XKtlv9DfESd12yHSHO9Wte4pOLAbzuq9wLFX0kFDCGGQ9Hk+wuSkOfE47X2vaJjTbiMv3dy8+YM1ce6NTPzhm0y6+te4Mvr4g9kQNeEiMsafzlRrMxOtrWSXHIfzhOtNx+rRZWeeSOiye9k95SasT9/HRafOMB2pd+n54PCYTiGEOArEAgewsBGv7ldY2Ij5G5OeI+JvBKXQyoalHGggHEh+DgDVuIeAyoh3qaorN5IBoLp6P2P8a6Bqg7EMQgghhBAiuZRS85RS8xsbzbwXBuIFEKd8vedlo85ov33ZX+KfZ1wNrcd7WjtqlJ4CX36J5nn3xS8yAcgbA7mj4rc/+zDMu7t9W9c+2377+M/2nk9b8Lkn4ts49ea+/z2tIgHYvfygVx9fufDgty2EEMcwrfVCrfWNOTk5A7/xWKTluFU/1G6D+z7R93rNB+CXBYmXP/45ePdv7ffbRpz0s0Djr7PhvtPjt5+7Ad74Vc/rHSwp0BBCGCS9nsUxTWvN7roA6SpKVt+rG1H50X/xbngTPyMJqwJy927Ft+QeRl96GNWhg8RqqqXk/V+S33QA1/4h6JG/Q2Xkm47VI28wwtJN1WSoCOdOz8RuMzMWQRw5lFKPAucBGcB+4Hda6wdblp0H/B0YCXwAXK+13m0qq0i+wilns3/ds9hCjSjAYYPCKWcmPceo408jvHE+Rf7NBN1DqLXnM3Zc8rtnABSefyv1r/yGcMZYJp5ymZEMALOmTWOd8yGmZ6bqnl4IIUSqO/2upexraO72eHGapqq5578jui7b5YHRty/ucdnw3LQBTiyE0FovBBbOmTPnBtNZevTlxfDhAzCklybCP60HFNhsaJ8Phk2Mj2sddRos+Vl8HWcGZI9of87Y89pvxyJgc7S3kgcomd4+6tXmhIkXxW8XToD3Opw8G0DuUP2gbFcIIUR/aKLY6NcltW/9rpfx4B06aHj39b6dzS/GP05rKQbsa8RJskiBhhDCICnQEIPikfd3sXpbJTfPPZ7jhmSajpPQfz7Yw3/X7GN0to1ffvbElOxO4nv/IXy2bHCmo52ZNFqFqA0L4fybwJNtOl4nu176E8H922jIPI68yk3sWXIPoz/9E9OxenT3K5so3fIvtntG4tdX8ulZw01HEqnvN8BXtdYhpdQkYJlSajWwG3gO+BqwEPgl8CRwirGkIulyxswidvY3yHnrTwC4zvwuOWMOvu3vQMnybceng1iWRcyyyMaLq2kf5A/C1Rd9GHrcVPLnfhN7ejaOjLykv34ru00xa/JEfD6fsQxCCCGObPsamtl11yXdHvf5fGRl9VwA2G3ZHbRto7fnDbSiLDejb1/cqUCkVX8KTBItG56bxvLbzx3Y0EIcjUZ/Ana9Hb89fE7880l91I50bUUPMKNlZN9Fv4XckTD2XLDZ4EvPQ3UZKAXXLYJ/XQo5I+Brr0PFKnjnTzD317DttfaTbYXj27dbPBV+VAG/HgY5pdC49/D+vQCzroXVj3YuEBFCCGHOYBVExCL9W99qHTme4ILJ1sIJNcgXVFoy+lwIYY4UaIhBsWr3AfZ7g+yuD6R0gcbW6iYADgQi+ILR1CvQ0BpHuJGYyqU1mc3mIBbVEG5KuQIN7dtPyObBbrMTUm483krTkRLaX9fAVbEV7Iz52X4gYDqOOAJorTd2vNvyMRY4AdiotX4aQCl1B1CrlJqktS5LetBj1JYqH/UNXk6ZaKhLQqCewIpHqYvF93kFKx4lf8ZF8VFKyRKLcmDJ7wk78vCnDSfizCGjaSc7X7qbvJvmJy9HS5aqJ2/hQHkZdruN4rNvIPuU65OboYUVaGD3knuwjZhD1gkXGskghBBCmPKfG06JF4PcQbcik34VmCRY1rXoQwiRwDX/gfIPoWQGuDIOf3uZRXDBz9vvH3d2/ANgzCfgqkdg/IXg9MCwmTDnK/Flo88AVyY0lsMnf9N5m64M+HFV/ITY4u/B6kfij8/6Isz7C1SsIvLG73Buf6XnTMNmweUPwN/mxItH5v0lXqCh5QSYEEKkAksN4OlA3aGDRtdCPK3j+xJtwd3H9/Dc1hEnfRRoDLaO/wYhhEgyKdAQg+K2T05ia3kNp44rNB2lV18+fTRPfrSXaUUuhmS5TcfpTilCJSeSvudDQvZCIF6wYcsqhswS0+m6yZn5aWKv/pbmcC3pKkTerM+YjpTQNWdM4u7Xbycn3cU3ZwwzHUccIZRS9wDXA2nAauBF4E5gbes6Wmu/Umo7MBWQAo0k+cvrWwkGmph5XImZYjtvBT5/M0FnvFNEk78evBVJLdCwmhtQ4Sa0Z0jbYyFHNlmN25KWoU1zPf6anRxwFuOI+snc8raxAo2asnewrXucpn0bQQo0hBBCCCGECZ5sGHd+8l5vSoIRg2l5MPfOxM9zeuKfL70bLvhFvGjD5ox36Rgxh+Bl9+OsWA4v3w5fXBAvxpjzFZjwyXiXEFc6fH8bxMLtV2rHpIOGEEIkk5U5FFtTJdxaBn9sH6UVo78dNHrrYNGhuKFrQYW2QNnjY7ka9vQQsI8RJ8kqnJARJ0IIg6RAQwyKoTlpZNrycNoNzxHrw6iCDG775KSUbns+7rIfsPOx75BWv52QLsTlcTLs8l/E/zhOMUNOuIyY3U3zjvXkTphN/rTUbXV75oQiTh07F39TE7k5A3D1ijgmaK2/qZT6NnAqcDYQAjKBmi6rNgLdLjlUSt0I3AhQWlp62L97/H7/YT3/aPK1U4bi8zURCQaIBA0EsOVizx6KLRACwJ49DJ8tF5K5f4nZCWWOIBYDlx1iDifK5qAxZ3ry93MxJ7aiqaRV78bp9sCYs4ztaz0jZlE39Qs4So5P6f19ssnvj3bytRBCCCGE6MLu6LnY3OaAyZfGPwB+sB08OWB3tq+T2V4wHsEuI06EECLJAp97nszGMsge2unxzVVNzB6MUwpdR5xoC+jl4imrjxEmSeugIQUaQghzpEBDHNO01lR5QzisWPczqSnCmZ6LIy2H1rcLypmOJ3tIr88xZfOKNwgvuQvLmcP+zS/iszyMnX6a6Vg9CoSjLN9Wi1uHOSMrC5ttkGfaHYZgJEZdU4jMzEzUYM/eE33SWseAd5RS1wLfAJqArvOGsoFuZ4K11vOB+QBz5szRAzF7PFnzy1PdzKyspM5z7yYri/QrfkrVG/cBipJzbsJWNCLpMYbNvoj0N36C3YoScJfgjHph3veS/nXxh6IsLPgyobrXCTlzmJ1zHp809TssK4ucy283+/2RouTr0U6+FkIIIYQQhyCj9865FjaUjDgRQoik0tnDYPhEAFaUXM2c/U/GH++1I0YPejuG07HLRSzU87JEv//bRpwk6qDRdiakz4iHRfZPQgiDUu8SfJGQLxjhxwvW8693d2JZMh9rICz4YAuvPvBjHnlhCcFIau6Qq1YtJlyxgQPuUrzuoQS9tZQvnW86VjdWLEbwtTuJ2lyEXHlYNgfel3+JtlKzEvXuJZvZ/tJfeX/ZIl7cUGk6TkKBcJTvP7WWXy7+mEXrUjfnMcoBjAU2AjNaH1RKZXR4XBxDNuytp3LnJip3fsz6vfXJD6A1Q6vfJpQ7jt22kZTbhuHIH0lexZtJj3L/W9uxtrzEZeHFzAs8x0tvLmfFLgNfE+LFIi+s2cfWKumeIYQQQgghRDLFsKOt1DzeJoQQx4LZ877RdrvfBRq96nB+Ktq1QMPq/LmrPkecDPb5BJWk1xFCiMSkQOMIsr8xyMYKL2v2NhKMyh83A2Hvrm2cEnqHgsZ1eJsjfT/BgJCvBkvZUDYboAgrF1ZTtelY3YTDQZyxZmL2NACi9nTcES9WihZolFfVcU7kbWZF1rCzNnVbm9c1hanyxec1rN/XaDjNsUspVaSUukYplamUsiul5gKfA5YCC4BpSqkrlFIe4KfAOq11mcnMIvka37oXl4riUlEa3rov+QF8+8FXybZANv70YfjdhewM58Ku5cmb30m8O9WqHVVcFFpCkz0Xl45wTuh1Vuw+kLQMHS1eV8mDb+/k/jd30ByW909CCCGEEEIkSwwbSsuIEyGEMMVmb2+i3+8jQwdbYBdp7ny/tfAh0fPbHk9UMJKkY1hSoCGEMEhGnBxBxhVlctvciXgIk+6S/7qBcMk5Z/KS7WeMHD6KIVlu03F6lDf2BIIfPYKO+LGUh3SayRiXemNDPGkZeHMnk92wiSa7h4xQNQ2Fs7A7UvN79dMnT+Avy24lMz2dG6YN7fsJhozIS+PKE0aws6KOz5080nScY5kmPs7kPuLFjbuB/9FaPw+glLoC+BvwKPABcI2hnMKgSMYwnM3bQEM038DvFWcaaEhzKvwRi5hNk+6wwJ3Ve1vKAaaUwuV04rdlkhlrxK4j1NnyyfI4+37yIBiRl4bHaWdItgOnXcZECSGEEEIIkSwWNpR00BBCCHOUve1mvzto9DYCRPfdQUMluliopxEnWsPCW2D2dWBrefxgjmVZMbDZ+14v0XOFEMKQ1DxzKnqklOK0cYX4fNKie6CMK87i21ddjM/nQyXx5FV/ZI05kfAnb8f29nxCNovsmddRdOJnTcfq0dTP/4ad/7mVtOYGwvkTmP65X5mOlNCFU0s4fdylBANNFORlmo6TkFKKq08cic+XR1ZWhuk4xyytdQ1wVi/LXwMmJS+RSEWzrvp/rHllGACz516X/ABpuTDufKateYKmsJ+AKqLEEYYZ3096lC+dPpYnXrmKi5tfoNGWw+qcufxsanHScwCcNq6QSUOz0eEADrs0jxNCCCGEECJZYtjlBJgQQphkO5wCjYPsMBEN9vy8RAUeVg8FGiEvrPoXbHgOrnv+4DNa0f4XaCgVLwiRDhpCCIOkQEOIFKe1ZlW9i4wA4IRttQ4u0Sn4w2tZeJf9GVfjdoKuAtIObKHxzb+TcemPk3rl9sF6bdVmtr73PI60XM6YeyWTh+ebjiSEOMKlZ2YTKD0TgLTMbDMhhkwiGAxCxAe2DIJKk1E4Iekxzi4OMtX9HHX+ekbYqjmnYCFZ6acnPQfE96Pe5gieZLXIFEIIIYQQQgBgKTuqtyuwhRBCDK4OHTTstn5etFK7tefHF/4PbHm5/X7XAo3W4y+JCiDaCjQ6nDNoLdawov0b0xuLgOMQO6MncRywEEJ0JZcRikGxtcrHwrUVBCOp/UdYeU0dC+7/OR+tXGk6SkLb9lZS9MFdZNlCOB02xm59iJWrUi9vuGItvk1v0OgootlVSIOjEO+Gl4lVl5mO1s3WKh8s+Snne//LqfUv8N4zf07t79X9G2D9s93bxQkhUsriN9+ldNl3KV32XV588z0jGYKrHqcsUsy29Nns8kxmZ3M61sbnkh9k/TPUNzRSZytgXyyP4J7VULUh+TmA1zdV8YOn1/KHVzcTiZm7OqLaF+TWp9bwztYaYxmEEEIIIYRIJgsbSkdNxxBCiGNXhyKImWzu33MTHcdZ+RD4Ktvv97eDhu6hQKP1OVakf4UTVuTg1+2WQzpoCCHMkQINMSheL6vmlY2VlB9oNh2lV7u3bWJy1UIaN79pOkpC3gO1OIgSsmcSVW5A0VRfYTpWNyHfAaLahr2lfbzdbidqKcJNDYaTdVfTGGBYtByvs4iII41hoR14g4fxZm6wlS2GskXg3Wc6iRCiF+HmJpxEcBIh1GxmHJllReJXqSlQQFQrdDSc/CDRZqKWwq4UymYjplX8qgYDvMEoEcvCH44Rs8xdHdEcjlHR0Extk4H/DyGEEEIIIQywlB0lI06EEMKcDuM/HAzS7+OuFxW2FlgkKrSwWgr3Oo44aSvQiHYonDiIrtyHtI9p2a50eBJCGJRyUxLE0eELJ4/klBHpjB2SYTpKr2bPPoWPwr9l0vAxpqMkVDpyNFvshRSG9hPSMULKzbBx003H6ia9aAxuhyIQbsbmyMAKN+Nx2vEUjjQdrZtxQ3N4NeMMZvvfoZlCdgy5lAsyDrEVWjKcfBOMugDyUvf7VAgBF5x9DouDTQBcevY5RjKkTbmI0uoH2RsCcFOaCfZxFyQ/yKRLGbbhNeoay3EoyC6cDEVTkp8DuHT6MIqyPBS6LTzOfs4lHUCjCjJ4+Msn4XZIfbQQQgghxGBTSs0D5o0bN850lGNaDLucABNCCJM6jDjBmcErwcnMta8Y2NfobweNtqKKDgUYVoduFv3pbHEoFwMpFZ/CIh00hBAGSYGGGBRZHidjizJR6iCqHA1Kczs486zz8PnMXOl8MIrysmm4+m4+fv0x0FA65xImj029E/X2/NEMPf5shq54CH9gCBnRWtScb6Kyh5mO1k1RloczP/dDNr4/A7s7m2vPuhC7LYW/V9NyoXB857ZvQoiUU5jp5rqzZ8R/VrM8RjKomV/AVbmH49Y+jl+V4Dj5KzDmjKTnWF9vQ3mj5FtNaBTvVzs5IWInx0AtXNSysNkgZpn/w9tkgYgQQgghxLFEa70QWDhnzpwbTGc5llnKjk0KNIQQwhxb5+MQQ2dfDGsHukCjawcNC8pexBZK0EG0tTCipw4a8TsH/9oy4kQIcYSSS/iEOAKML87hxNIsphSnMWNUgek4PdvyKsFtb1PmmMhO1zg228cTKnsFdrxlOlk3OlCPWnQLoz++j6EfP0Bgya8PsR2aECIVaK35x9s7eWj5ToIRcz/L/nWL2P7Pr7L9H1/Bv26hkQw+XyP71i9jP4UcsDLYuvy/6P7M7hwAWmsqX/4jNrudiowpVKZPYlhgEx8uXZDUHK1Z7l6wnPpnv8fSxY+zvrwx6RlaBSMxnlqxlz11AWMZhBBCCCGESCat7Cgp0BBCCHPsna+UmT4id+BfI9JlzL224InPkb7gup7X73HESYd9RX8KJ1q3dSjkfIAQwiAp0BDHtE2by3j1D1/m7VcXJP0EVn/seuYneD96HP/mpex57NsQTb359c0r/s3OJifalYV2ZmC5stnV5CC44hHT0brZv/RevPt34HMW0KiyadzwKv4ty0zHSmhbdROvbdpPJCZVvUL0pCkU5cUNlazYXc/eenMnv73b38UXVfiiCu+294xkaKypwGGFCbkLCDrzSG/eTzgc7PuJAygcs8gOVxG2t4w5UwqtbIQaq5KaAyAUtYjtX8fs6FpmN7/PlmpzHbO21zTx7/d28XpZ8r8OQgghhBBCmGBhx6YP4+SZEEKIw+PObL9tRTsXRQyUrh00mg/0vn5rYUSiDhptHTYUfPwC3JEDDXs6LO9wHifWso9pqomvt/bJgwisur+mEEIkmRRoiGParvXLmez/kPTdS4haqVugoWo+ptFeQMCRT7SpFoINpiN1pjWWr4qwzdM21kYpRVC5sbwVhsN1F6vfTVClY1MKu92BBQTqK03HSmjRugoWrNrHvgPNfa8sxDEoy+Pku+dP4Oo5Ixk7JLPvJwySgpOuojDTTWGmm4KTPmskQ+HwMTS7C8loriQtUkdjwQzcnvSkZnA77NRlTcQdacBhhXBYQbTWZA2bmNQcEB8pUjz1bJ5Nu5y3cj/FSaPzk56h1eSSbH75qWlcccIIYxmEEEIIIYRIJkvZUXICTAghzHF06KARCw3OCO3Nizvfbyzveb1ld4FltXfLUAqW/wUa9nbuZtFxv7H2ifjnijXtj3Vct3XESe2W+OeVD3V+zVgvRYIpfMGuEOLoJwUa4pg2/RPzWDfqS9jmfBmnPXV/HJzjzyU3VkdGtA5XyURIT7ExJ0phHzKBDB0gZsXfQEUti0wdwFE8yXC47twjTyBD+wlHIsSiIRw2G9nDUy9nqy+dOpqbzx3PqILknmQV4khyxvhCzpwwBJttEP7QPEjO4TMJX/U44asexzl8ppEMHk86k0+9mMJ0G3npLk449wojOWZdciOZMS9j/asY619NOHskp515oZEsl588nqFnfpkzTj2DEXlpRjIA2GyKWSPzyPY4jWUQQgghhBAimbSyyYgTIYRIKUk4bubd1/Pjy34Du5e3jyXxVsCS/4XHr+nSQaND4YTdEf/cWogBncehdB1xsuc98LV0Lq36GH5ZgGPryz3nkQJCIYRBDtMBhDBpeHERw6/7IT6fuZbnB2P4xbfjHXc6Pn8zw6efDTa76UjdeE65gbHVP2T3gXr8ykZmpI5RBem4Tv6q6WjdDJl5MfYNTzOybiU+NRTX6JNwD5tmOlZC+RkuJg/NbutOIoRITf/5cA9vr1wHwJknTOfzJ49KfohtrxP46BF2hLLQ2knua3fhyh0ORcktQstY8w/2qhx2O0eAhmGBctSOpTAxuUUaDYEwf3rqFc6ofZrNaROpCV7P5SeUJjWDEEIIIYQQxypL2bF1PXkmhBDCDFfm4Iw46eKNDXs4J9HCSHO8iwa0F0gEvZ2LLjoWTthaLnLp1DWjw+1Ya+FGh6KO/eshqxgqVgHg2P4qzO6h023XAsJAPaSb67wqhDi2pG7LACFEm6oDjSxftZrVH29i274UnV0/fBa1E66mKFpBaXQPhbEq6iZ/CYqnmk7WWThA7XPfp9IbYo1zJlvtE6jZsY7GJb8xnSyh8gMBNlV6CUbkqhMhUtnidZV8q+mvfKvpryxaZ2hsUvmHVAQUNmcaUbuHuqYg1JQlN4PWRMrX0GDPw+NJw5OWhj9mJ7BndXJzAJsqfZxYv4jxejenRD5i9brkZxBCCCGEEOJYpZVDRpwIIcRBUkrNU0rNb2xsHJwXuOmtwRlx0sXyLfsTL4wE2gojgtGWogpF564ZbbcV2FsKNGJHSi8xAAAgAElEQVQJOmj4a3t4ka6jS7rcb/0adNw/la+E342B9c8kzi6EEANICjTEMa1sv5dvPbaK59ckaLuVAiIxi3WP/pAxu56mtHoZNU/dQl1jk+lY3TT7fTR9+Cj73OPYnX48Fa7RNLz7EOFQ0HS0TmI736Guah9NzkJcbg92l4caWwHesjcTz8czaPN+H4898g/2vPRH/vzCe2iZjSdEyhpVkM6rzvN41XmeuZFEmcVkO2IEoxaRmEW60wZpucnNoBQqPY8s7SUnUkNGpA6XDuPKKU5uDmBIlpttzgnEtKZO5ZA7ZETSMwghhBBCCHGssjucxKKRwT+WUbURPrh/cF9DCCEGmdZ6odb6xpycnMF5gYKxJGPEiZNeLjIM1LWNJYnFWjosaTp3xejUQaOHEScd1/W1XCDVqcDDIhCO9r3v6fg6Vevjn3cs6/05QggxQKRAQxzT3iirYU99gKVl1URiqVnR7wtGKQlswe8uIuAqJDt2gJqaXqpQDWmsr8KmY8Qc8ZOSMUcmjlgIX0NPVazmBGt3E7XAbmt/M+qw2/FHgKbU606yqdLL2f7FzI6sIVa+kmAkNb9PhRDwg7mTyDvpKvJOuorb5iZ3pEib4z9LSel4Jtr2MdZRQ/6kM2DUGcnNoDXZpVMZH9lKSXArI4KbKVXVeIZOSW4OoDTPQ0HBEFZEx7LNNorzxg/SQQ4hhBBCCCFEN26XE6VjvLv9/7N33uFRVekf/9zp6b1XCIFADDVIkyJFKWLHvqvuoquuq6uuq2tl7X33t/a2a1dAsYCCAoogTUroBEggjfQ+mUmm3t8fd5LMpILCTJDzeZ48c+9p952byUzmnO/5vjUn90Kvjoflf/dcoBMIBAKBgsFt444XUpxo6D61VVFJMbXGZgB0bu1kN1eM+mYLAE7kbhw0nKB3ze/kLgOHHXeXDGOLjSEPfcsPB7pbl3CtC7S4OZW0CUGEg7VAIPAOQqAhOK2ZdUYswxJDuGB4PFp13/xzCPHTUhI8ggBLBf7WKuq0scTF9r0dwKGRcdhVerQ2IwBaWwNWbSDBYVE+jswTQ0w6WjU4HXb8HUYMcjMOu40gnQTBfe++Dk8KZUXw5azTTSJgwAQM2r75OhUIfI0syz53mPHXqYlUNRGhMuGnU/skhmarnYNHq2m0yhitsOdIOU6nl4Vdlfup2ruG/drBlOn6c1SfzkE5mYZVz3n9i+6qJf9lbNHrDNcWMcq+E/MXd1BSXuXVGFqRZZmSOjMWu/iyLxAIBAKBQCA4PQj0N6DGwdVvbWbZrlIe/GIPi7YWA3CwwojTeZzf4WryYUEIqqNbu6532tvG7O37YW55I2sP+ua7QSvbi+pIvfdrCmtMPo1DIBD8xrkrF+4rVY69kOJEI3U/D/XVtsOs2K24mWslZX7E6nDy6g+H2tpsylfem81WJ1/vVUQWsqODg0br08hbBWs8U5fXmhSBR05RXddBtN6DT65yK3PN48lizkYgEHgHja8DEAh8SWpkAI9dlIXRaPR1KN2iVkmM/f3jbF3zBTarjaETZhMSaPB1WJ0w+AUQfP4T1C99EIO1BqdKT8SFT6DV6X0dmgfq1HHEhQYQX7EVB2rMzmiCbBWoMq+CIO/b7/dGamQAf/ndZZRVVJDRPxXJC/9ECwSnIs9+e4Da+kYenZftM8Hdqu25JHx/OyCxSvUfzhuT6fUYCravQt9SRb1fCjZNEMF1ByjLyyFh8BivxeAo+Ik6sx1ZH0ajpOxosNodNNZVEVKTD1EDvRKHxe4g/PAXWHWh2FR+2LQhhBgPkb97A4mxF3glBnc25Ffz/HcHGdM/nHtmDvb69QUCgUAgEAgEAm8TGuCHOtwA5XDrRzlt5X//dJdHu/FpEVwzNgWVpLjd/m5cCnEhBiIClTmlez/bxeC4YK5Rf48a0O5bjGXAJJpa7Px4sIqLXeO8vWYfz60pY+YZsXyec5SbJqdx1oBIUiL8SQpvT0NZUmdm5r/XAXDbtHTunHF831FsDicalURxXTM15S3YHE4qGi1ceWYy3+0tJzs1nPAAHS02Bzq1CpXLxXVvaQPBBm1bLB9vLgJg3aFqIgP1+GnVbW0BTBY7MmCzO/lsewlzh8UTE9z35gQFAkEfR+vnduIFgUYPDhoanJhabB4rkxVGC0t3lHCLaxlhydYiZuqU41KjHTSweN0uLmvtIDvAfTNSxR7oN6nt1OFySm9zou5BsJf/7BQir32PEJVLoOHsPnaBQCA4kQiBhkBwCuDn54ej/3Ss5iYCQ8J8HU636CL7UeSfSaillPqAZOIiUn0dUmdKdxCkcWAOS8FurMKpCUAfnIq27gA0VUFg33L8IH8NweteQHb6ozoQB9MeBr/Q3vsJBKcZWo0Kg0blha+Z3aOSVDhdinuVFywju0RqNXWUW0+9sjvCIwRJjeSKoNOVvS4yU6F2WtDJDbRodCDL3g/BhVqlvCY0KuGEJBAIBAIfEJIMCzqk+tLHg6W06/bHWFdgABYcQ7+QZJi/4XijFggEpzoqNUFaWHzTOOa9trHbZhvyazzSoCx0uWwAjEoJY1uhsgs6V72fJ7WQX2nk1cW7+Gqn8n5zsUuz8MrKvTQTwuc5yu7s137M57Uf8wEI0KnRalTcPDmNxdtK2sb/z+pDbMirZnicgdtnDEKjNbBwSxELlu5j1hmxBOg1jE+LYHtRHSv2lHP79IF8tq0Ei91JVWML1SZr21hatYq/Ld7J+LQIRiaH8dIPefzxrH5cODyBFruj7R48eXEWK/dV8H1uJQAWu5PMh7/llilp/H1mRpv7x5gnVtNkaV8s3FnSwItXjjjOX4JAIBC44ZUUJ927UKhxoMbTYUNCRuVWFkBL27EDZZ7tMvPH7R2cDiXNiYtak5XV20qY5zp/+YdDwDDqmq2gBff0Jx1JM+WwfenzjBwzqX1sgUAg8AJCoCE4rbE7nPyUV01qsIqgoCBfh9Mt/1l9iNJD2whTW1lzeDjPzhvW51KyyLLMtkVPMsi4EWNACjH1a9m8WMMFNz3u69A82fkxNl0Qu+oCsaujsatCqLWZyXKaIW8lDL+q9zG8RWMp1jXPsLsGjGotAxt2EB/wCpx9n68jEwj6HHdMH4jRaETjw/fG6SMH8b36ZeV42ACfxNBv1Az25yzC31yOVbbTEp3F4HTvTuCp+k0kNOADjGYTQSobDiQ0TgiOjYPw/l6LQ69WERCZSL+jXyIjYXLUIskmItLSvBaDO2P7R/D677IJ9df65PoCgUAgOM25Y3fnMqMRuvsefIx1qfd+TcFTc3rv11EcIhAITg/0wWAxkp0SxrDEEHaWNLRVfXrTOK5/ZwvGlq53K/ePCuBwlalNnAFgdy3U7T1az1cFncVgemydyloxWR1gdfDk8txOdVsL6/i04ip2b09lrvWJtvLle8qVWN0EHQ9+safba/xt8U7AU3Dy9k9HePunIx7t/rFkN1HUc4V6O584pvLosn0AvLImnxA/LVsKalm1XxFvhNCEGQM2NOworqPBbCNEfKcQCAS/FC/sWkkK0UI3mZs0OJC6EEyo3MrCpca2YxudUwj/6b3NvGS30fpOuL2ojs+PlDDP5bphbFaEc7J8bM9VpVKJFCcCgcDrCIGG4LQmv8rEs98e4KLMcObHRfo6nG7JKarnLsti/NQyTzX0p95sIyqob6UOsdidRBoP0KINwy4ZsOpCCKnfh8Mpo1b1obQcDcU044fDacKgVWPTqGgw25DDdEh1R3rv700ayzBZ7Jidfuj0akqa9cRXHfB1VAKBoBt0GhUzR3knfUd3GALDKBjzGAXrP0GlD2XMlMuR1F6evIsaSPToSwn/4WnsNhsqQBUQgWHWy6Dq/MX6pFGwjgH2Q+RKafjLRkzEEG5oJnXbvyDlTfCBk0WgXoOmL30mCgQCgUAgEAgEJxNDKDTXIUkSX956FjaHE4td2fUcqNew6+Fz2JBfw//WH+HB84Zgd8r8kFuJQavmmrEp1JutBOg11JqshPhpWfjWLqjAY6f19eOSwJU9Zf74eC49ZwoF1SaKa5vZfKSG9zYWMiQumCPVJppt7QtvN01Oo8Xm4J0NBVw8IgH2Q5aqoNunEhmoZ9YZsWwrrGNfWfvi4WXZiSzaqgg4DFhoQZmvu3hkAku2H+12vNd0/2KU6hA/OoZRRkRbeUcByU7DjXzryObPjrsorm3m2v/9zBd/ntDLjRcIBALfMXtIJGzpui4pVMfEARGww7Pc3VUjQlLS0UuSxB/9fwKrZ9t9pfXIOoeHbat7/9biVslHfpWZgU4ZlUoir7KJNCQPx9fhhf+FEJeiRDhoCAQCLyEEGoLTmrSoAP527iBSg/qWG0VHRiSH8sHBy4jQtBAeEtwnd9/qNSpqAgcSYdyETReCztpIQ/ioviXOAAhJwq+uELVKosXmwC45ifDTIjnMENbP19F5EhSLv16Dv8qG0e4gVW+BqEG+jkog6JMs31NGXX0DV04YhOSrHBbAZ9tLkICLRyb65Pp5lUbyN3/FBfZVNEuRfPJdBIMTryDEz7ufG5qmMop1KTTZwImKZL2EwVQFDPZeEHs/p8ikxuIXiVMdjU0bzFFzHVHVJQRW5ULMEO/FAqzbnU/Jt//G3O8c/njJXK9eWyAQCAQCgUAg8Al+YWAzgd0KGh1atcrDEVaSJCYMiGTCgPZNU2lRgcqBw0aov7IdOiZYyWFy7YQ0WAJzh0Qwae50ggwa9A5zm0DjDyPDYevLDB17M0MTQ5kzNI7fjU0hNTIArVrF8t1lHKkx8dOhau6ZqXx3XHB+ptJ5gfLw6U3juPS1jUxMj+ThuUNICPWnvLGFfpEBADy5fH+bQOOT60cwdlA8UwZF8/iH37LecDtvh/2VtUFzeOGy4Tw/bxhrDlaRERtEQbWZK9/cxO3T0vnz2QPQ/eduaITnL81kdbkf980ejMlq57NtJQxNDGFUSjivrtoDP8G56q1ckJXAZ9tL2FFcf5J+WQKB4LTAGyl5nV07IwFMHxQBcmcHjS/0D7UdB7vsNwJoBmtzp7ZqnB6CDBkJtVtalbnqDXzrHN12vrfcyMz7vuHB84bw6LJ95BocGDoOumshADY3Zw6BQCA4mfTtVWmB4CSjUas4e1A0kX3MjaIjt01L55yzz+aM7In88/zMPpfeBJQv1SMuu48DweMwOdQcDJvEmZfd5euwOjPsClRWE9H+Eg6n8s9gjN6KrNbBgOk+Dq4DIQnop9zNsCgVZ4RaiRswHMbe7OuoBII+ycp9Ffx4sBqrw9l745PIN7vK+Hp3mc+uX2+2kWQvRkJCK9kJt5V75Cz2GuZqah3+NOsjMapDsTgAi9G7MTTXYXao3YSCyg4Jhyx7PxagpPAgI5o3IxVtasspLRD8GiRJulWSpK2SJFkkSXqnQ900SZJyJUkyS5L0gyRJKW51kiRJT0uSVOP6eUbypbJNIBAIBALBbxe/UOVx9+LOdQ4bNJR4llUdhB0fwcZX4NFIWHo7NNfDtnfg2XRwuLZRyw4iA/XoNWqwmtv7vzsXVj0M3/ytrSg9JqhtHm1WVhy3TBnARzeMVYT9a59TUjC57ZjOTg1n6wPTef+PYxgQHYSfTt0mzgC4bWo6f5k6gH2PnEtmvJLSaXZWHI9MVVww/mh6i3f/cCagzJWdPSiauBA/xqVF8PP907hjxkB0klN5/sD4fqE8eN4Q1CqJYIOW6yf0Y1RKOAA3ZIe1XXdqRjSgpH4RCASCX8zJ+uo3YEb7saOHeSinvVMakahAz7WZxICe5/aGJwSjljznVTRugo3z1JsZJuV16teaTqqnKZn1Byux+3huUSAQnB4IBw2B4BTAoFUzOysOo9FIkGv3QF8kMSaSpose4PCRPIYMSCcuItTXIXXCHj+aRarZTKp9k1HOeppUsVSbdfx85j85LzDa1+F1RuuPWq1FixNJ6w8q8bYtEHTFQ+cNob6hUZmg8yHPXDrUp9dPjwni07CZRFRVYJJiaIofT4wvRIipE0gpyuVAkwo/lZ2QAANEZXg3hoRRxJV8Rp25nnAaaZTjMclO/LQ6iEjzbizAjMln86Pen+x037q8CH5TlAKPAecCfq2FkiRFAkuA+cBS4FFgITDW1eRG4EJgGIrr60rgMPCatwIXCAQCgUBwmuDnEhh8eYuS7tBiBLUW9n4OpTugpR6Sx0H2H2HoPHh1nOfO623vgLkW9n+lnBtdYnj3NjZT5+Pt78GsZ0FrgOpDEJYKR7dDwihQu+ZVZBm+f1Q5rj3sEXZk62Lhz2/Ctnfh5p+Uc4edAL2Gu85R3E2NlvY+0waGwwbA2gSvjIerF0GIp7NidJBrz/ajkbSZ79tbur535lo0/253/ZszNI6HvtSRERvUdXuBQCA4Jk7CfEREOqTPgLyVyvmOD7pv20UKkY5zeWMCysHSqVkb95wzAD7yLHN31AD4Uv8Qd9tuBDo/Y7mHexAvVTPo/mXsfXQOBq1v5xgFAsFvm763DV8gEJyy7C2qovDDPxO/6REOvnc7B8vqfB1SJ3YebaSk0YFVE0yxYSAV+hQ0Gg0r84zUm629D+BNjOU4vn+MvVVWcmoN1OZthM2v+zqqbqlvamb50kUcLi7pvbFAcIIJ9dcREeh7N6SIQL1P4wjx03LHpdM4eOYj1I+4mbsvGIPGB65LBfFz2ewYRH9nAWGOapb4zcMSkurdIAZfQIxcRZqjEL3NSLC9lqHOXHTJ2RAQ2Xv/E0yg8Qijpf0kNe7sebuGQHCMyLK8RJblL4CaDlUXA3tlWV4sy3ILimH3MEmSWlVS1wLPy7JcIsvyUeB54DovhS0QCAQCgeB0wj1N65onFWeLpbfD4TWKOAOgaCMsma8cd2WLX7G3/fj7xwDQ5q1QnDNkGZq7mXtqPApNlfBStiKI+O857YuHAFvfbj9+Kbv9uHy34toBSrwVu5Xd4KYaeCIOvrpNqev4P73dzYa/cq8i7qjMVRw6qg50CM6tr60LgUbeakVQ4s6H8+gXImOxiZ3dAoHgV9C6YSRpLKRM6L5dxnnHPqZKA1q/3tsB1BcpwjsPPN9P/Wr39zhEXKCnwGLqgGBejFnaqd2fxypzP2mR/hx+YjZ/mToAAInu52QGqo7ygOaDtlRWAoFAcLIQW7EFAsEJIy8vl3RbEQ2BA0htOsThw3kMjBvde0cvUlxrIsl2BKvKD5MmFJs6BH+pmmhHBWUNLW35TfsEDUcxW2zU2/Ro/FQUNxsIr9jj66i6Zde29aTlPMnesosZNuRBX4cjEPiEwv3bAEgZPMpnMaz96QfG73yGJn0cm/3/ztQzh3s9hndXrGOeZT8NujjU2Igv+ZplO2ZwSXay94JoKEYOiKbSosfP1kCz5I8pqD/B9cXKZKoXXSwslXkUfXALdpuVZsmJpaGM2InXee367ljtTrYV1tEvREWQ2Hz3WyUT2Nl6IsuySZKkfFd5bsd613Fmd4NJknQjiusGSUlJGI2/LkWQyWQ6pev6Shy/9bq+EkdrXYyf3OVr/7jG1MeDa4zf0r3seG+67aeP73O/1xNZ90vHEwh+88RmwcP18Nl82PNpz21zPuy6vDa/6/In4noeryoXfnjSsyxvNaRNBbUOvu4mLe9rZ0FIEqSd3V7WUg8NxUqKle3vQmMp5K3EPywNbncJKToKLeoLYc9nyvGeJcru8sr9ioOIO29NhYvfVOL69n445zH44OLOcR36julBg/jJPrfn5y0QCAQ9Ibk28UgSGEK6bzfzKchddmxjqtSgOUaBRuH6Y2vXE2W7PE7VR9bQlddFaqDi1jEkNhCVSuKucwYxb1QSuhd7TgV8rnoLB5ttvz5OgUAg6AEh0BAIBCeMhKT+1GyOIMhWS4U6loRELy7EHSMxwX5s1sSjteSALKOW7YBMjRROlC/SAPREYBQGjYS/2kGjQyZWb4Ww/r6OqlsGZY1mY8kfiRs0ztehCAQ+o+GLu5WDwd/7LAZp/zIMKicauZGcXWvAywINi92Btj4frSTToFFSXcVaKtlVVgZ48XOh+iBWSUeRKgl9QCrN6iDMsonMpgqwNYPO32uh1BXswmGzYDbEgsWI9uBa8JFAY3tRHY8s3csN4+K4ODKs9w6CU5FAoKpDWQMQ5Fbf0KEuUJIkSZY727vIsvwG8AZAdna2HHQClD09jXEq1PWVOH7rdX0lDoCKZunXx2kpxV0Z91u5l13dmy77WUohIKBP/V77yn0WnBwkSZoLzB0wYICvQxFIEgTGdC4fOAsOLm8///KWE3vdH59R3C/c2fImlO+CjDk9920oVtKktGKu9XTqcDlxqOvy4cVsmPUUlHZwvDj4HUQVKMeyE96a1v31ltzQftzD95QI6rHYO6cHEAgEgmPG5Pqq6BcOhuDu23UUk/WEJB27g0ZXNBQfX/ultx1bu5aGTkXJEf5Az05EGpw0eEOgUV+kuI8Ex5/8awkEgj6HSHEiEAhOGKMGxKO/5GXyh9xK8KUvkpXSxRdwHzMyJZSi2Ols1wwjxF6F3tHEV7rZDMjMbs8x2lcITUY74S9khjnJCLYSkZAG4/7s66i6JSY8hAuvvpXMQQN9HYpA4DM0k+9GM/lun8bQnDQeu8OJGR2had538tCpVbT4x+OUZfROM0H2OhqkYCIjvJxWJDQZnVpCrZJostixOZyE6x1KHmyNwauhBMcNQK1SYWipJFA2ok/0vqtJK1kJIdw0JY1RKeE+i0Fw0mkCOs60BQPGbuqDgaauxBkCgUAgEJyqyLK8VJblG0NCetgdLPAeQR3mh/zC4bJ3T+41y3d1XV68GVY+dHxjvTxaSVfSFTWH4INLYN3znuU2E5TmKMdrnzn2a239b7dVwZix2EWKE4FA8CuIH6E8Tr4bInuYw9Ucxzy5w/brBBonC3eBhsMluLD3nuJchZPGlp5dNk4I/86CFwaf/OucYB76cg+p937t6zAEglMeIdAQCAQnDEmSOHNgIudMHMeoAX1T+anXqHng/OFsT7mBx+zX8Ip8KYEj53HDpL7nTCHLMktKQ/mxJoTDDfBRaRyF5uNQLwsEAq8zZPwshoyf5dMY5l18BY3nv41jxhPMmXyW168vSRLzZp3LZwGXU2EzkOtM5Luk2zh/pJddlZLHYdRGEm0tJsl2hBh7KbKpGvuI34HKu/8C+xv0pEaHkKapJjXAQUJ8XOec1V4iQK/hguEJhAf0oZReghPNXmBY64kkSQFAmqu8U73r2C25u0AgEAgEAsEJJjbL8zwwuuvFv4vegOHXwG05nevOebzna8xfDSN+B1cuVM5ll5AhaUz3fc5/secx3fn2H+3H/hGKyMTLnGv8jMSWQ16/rkAg+A0RP0JJPRU/AsbfBhe/Bdev6NxO34O7Rkfslr4p0GgoAUBd9BM8GgnFP8NjUb12U+PkjbX5mCyKSKPFdnKdi9YcqOQfS3YhyzKbD9fQ096Jb3aXsWxXKU6n7/ZXvLexEMCnMQgEvwWEQEMgOAUwWew8/vU+Xlh5gIrGlt47+JAjP35I/vt/oXBjL7lFfUhZQzOZhe/xD/UH3CwvomHvSszWvmcRuT2/jITNj5BEOVqNijHG71j96es9/pPmU+oKlF0ohRt8HYlA4DtsLZ1zD3uZKqOFZXktrC9uobHFNzkzB0QHYgxJp1KKolgVT79+aQQZvCswa0bPG41jCXCaiJYaCKOJYos/37ZkejUObC3w3YOotHoaokbhiBgAW96Gsp3ejaM1HIeTPfmFNFu9sBtEcFKRJEkjSZIBUANqSZIMkiRpgM+BMyRJusRV/xCwS5bl1m2f7wF3SpKUIElSPHAX8I4PnoJAIBAIBILThX6TFeHFlQtBUsOMR7puF94fLnwZwvq1l8W4xB3xXbjQZZzXfpwwCi54CQbNbHfMix4Cf/gW9F04qVy/HEb+Hm76CVInetZNexiuXQZXdzG3NeF2uGYJ3HME451F3T/n3pi+oP346s9cYxYqz6MH7jU9+8uvKRAIBKCkJAFQa2DoPEhxS1c98S4YcxOo1Mc+Xl910ChYB4DKVKmc7z629Qo1Toprm3l6RS5H65vJeHAFi7YUY7U7OVrfTGl9MwXVprY5elmWO6VEsTucLPhqL8W15i6v8coP7WK76/63hY9/LuaiVzZw+RubWLy1BEcX4odPfi7ilg+3c+tHObz6Y/4xPRd3WuPNLW/kaH0z728sYM/RBpqtDt5ad7jLawIcrW9uF2OUbGWSSpnParLaMfYw77h8dxlp931DZR9fzxIIfIUQaAgEpwC7ShrYXlTHkSoTPx7smFK8DyHL2De9ToupkZafXvN1NN1yoKyekZbNmLSRyBo9w0wbKK1v9nVYnSg/WkCAbMakCcMu6bFrAog27qXJ0kcX1XYtgiNrYdu7PtsZLhD4nC9uPvG5k4+T/60/wuFKE/vKGli2q8wnMewtbWBM5WLGSPuZJa/j8LbV2BzeteI9WGFkUPNOavUJHAjIptAwmBDJzM69XjYKqNoPNjMFJi0Hyo3sLmtGVmngyDrvxuHixzUrkT6+go2rlvjk+oITygNAM3AvcI3r+AFZlquAS4DHgTpgDHCFW7/XgaXAbmAP8LWrTCAQCAQCgeDkoFIrwotBM+HhWhh4rlI+82nPdoHRymPr4iHAOY/A5HsgeTxkzWsvN4TCFR+2n7v3mfaw8jj2ZqX8H0Xw1z2Q4uYwGDvU9ZgF13wG97qJLfpNgn4TIX2GZ3waP0Vc0ioWkbqYWr9lE0z8W+fyIRfAggYY6HJc1AW216VPhwHTwC9UEYYkuxZLQ5Phjn3wYE1b00rCOo8tEAgEJ4ppD8Gsp3tv547Dorw/nggm33tixumKmmNzIAqWzBiw8N7GQjblK++/f/9sF7//72YmPPU945/6ninPreGTLcX8b/0R+v3jG4b98zs+2lzUJljYWdLAOxsKuOezXRz68ilYEAIOKwu+2ssL3x3g1W93dHOAy6gAACAASURBVLrujuL6tmul3fcNz317gCe/2c+GvGoA7l2yu63tpsM1FFSbPPoX1pg4UG70KGsVjxhbbGQ/toqF20qZ+e91THjqex78ci/nvfgTz313gMe+3s+3e8vZX9bINW9tpspoaRtzwlPf8/QK136Pt6bxnk55fXy9q4ysBd+xrbDO45oPf7mHFXvKuPnD7TicctvzEggEnmh8HYBAIOidtOgAwgP0qB0ywxJDfR1O90gS9tQp+BfnQP8pvo6mW+JCAynUpDLQXkCLFM4RTSYjAvue1XtQeAwOWULrbMYmB6N1mKnxi8df10ffuvufreRyTZ7kOTEiEJxOtE7y+ZDwAB0WuxM/LYT6+yYtkkGrpkEVimR3YEONXReM2svvC3qtihbJgNqpiNoknMjI6PUGr8aBxg9kJwaNCpVKwqBVg2wFfWDvfU8CIeGxlGmT0If1zVRkgmNHluUFwIJu6lYBGd3UycDfXT8CwenDuxdA9bau6/TxYCnttq7AUNr1X1sv/TzqQryc6ksgEAhOBcbeBIFRsPpRSD0LQpI864MTIG2q8gNw8Zsw6xl4ph8MvVwpu+sgWJs8+427RflxJzQJksdA4U8QNdjz/3GN3jPlir9b+pKzHwBjKSSOhqHumtduiMqA0fNh3XPtZQ/VtadZbN2V3voYnOjZX+evPLeijZA0FkISPKpL6d2eXyAQCLyK3Xr8DhqJo6FkS+dyXcCx9Z9yH6x54viu2djN/+1d8FBGGfflpnLX4nb3002Haz3a/MNNMAFw3+e7ue/z3SSG+TFlkPJevSG/huiS/wMJFm06xDsbFMFHqtTQawwv/ZAHwOtrD3PpKM/PinWHqpny3BoyYoP45raJqFQSk59dA8CgmCAkCc7NjOX/ViuilMuzk6gxWXlsRV6n62wpUJ7XHQt3YLErm6vuXLSD84bGcc9nu9tiqDJaeMGt37Jdyv3cfKSGUSlhVDS2UN1k4d2NhbzrSoMCYBIOrgJBl3h9lU+SJD3wCjAdCAfygPtkWV7uqp8GvAwkA5uB62RZLuxmOIHgtCA6yMDLV42k0dhIZFiQr8PpkcGX/RNjTTlBEbG+DqVbxvYP56Nx9/NpzndoDEGMO3su0UFeXrA7BsYO7sf7u28mu+B1/KQa8tRpnDH7JtSqPip+SBqt7DoxGntvKxD8Vpl4h68j4A9n9SM53B89VqZl+ua9ODM+mJ+zb+CjHalYdGFcPHsWKi+/d6VHB7E4fi79Cw8SZKvBgp11uklMGd2FNfLJJHIgRKQTWbYPKdhAgMoMal3n3XheYvSIYVizPsHSbOq9sUAgEPyWMJYpO5e7rDNCUDffs4xGUh9fS8FTc467X7d1AoFAIGjnjEuUn47cvlNxyXBHksA/nKYbNhEYm66UBcUAMcd2rQCXuGHy3T23cxeK9Nb21q2QtwpW3NseY3AcPFwP394PmRe1izOgXZhhCIGpD3T93B0uy3j3hcorPoZPrgSHjRV7ygnQq5mYHsVr6wqZkZXI8KQ+vKFLIBCcmgTGQlN57+0clt4FGmlTIf/79vOU8d0INPyPLbYO4rVjoiq39zYurjorg3WaUJbvOYbn34GSumY+2NTuyiSjzIe9uKYAUL4fGDi+lMSfbivpsjy33MhfF+7wcAc/UGFsq2tl4dbiLvurcFJUUgIEtYkzQBGArDtU7dF2Sc5RXnBbQlmfp4hNnllxgMz4EK77389dmmpXNFp6fG4CwemKL7Zha4BiYDJQBMwGFkmSlAU0AUuA+Si2u48CC4GxPohTIOhT6DQq9JrjyP3mI0obWsirsDFQayE2pO+JHgAkSeLqrACcOn+aVP4EJ/fBHHkov/Or583jk3WZWOrLGTF0BKMGRPo6LIFA0MeRkFCrJFROCclHbjaSJHH9mbEUOoPwCw0nxgeThWqVxM3nT+a+D2UiytbSok5g3OQ5jErxsiWwSsWO9FtR7bmBFNsBjKog1mbez9zgBHzy2zHVoC7ZBoGpYuFQIBAIBAKBQNB3CUvttkoOTgT1L5jWHn0DhKbAoFk9t1MfhxNhZLry028ylOa0l0sSzOxid3d4mvLoFw6TuhBnAES42qSMby/LmE1DQCqqRjs3faA4Qj15cRYvry3k5bWFvHr1SHQaFVMzorE7ZbRqRRQiyzKLt5aQmRBMZnyIx2UqGltYn1dNcW0z49IiOLNfODuK61mfV01yuD+jU8OJDTHQYLbR2GIjKVxZOF13qIpQPx1ZiZ7jHSs5RXWsOVDFHTMG/qL+AoHAS9y+E3Leh2+6SNvkjv0YBBrp53oKNLTdOGXojtFt1PDL3n+OGauJ9OhElnco/vSmcVz62sa28xA/LSOSQ1lzQElL/9/rsnlmxQEOV5nISgzhsuxEWKa01dAugMiM9VMSkgKLMzcwOv8lPpq5CyfwwBd7ug1rdGoYWwqUjunRgZQ3tPDVzp6dQSamR3qILcIDdEhAjcnK3ZqF3KxZytCWN2gkEL1G5SHUOFZueG9rtxnPn1qeS0ZsEFMGRR/3uALBbxmvCzRkWTbhaQ66TJKkI8AoIALYK8vyYgBJkhYA1ZIkZciyfOzyNoHPMbbYyK9qYlhgoM8Wh35rVDa2UNfQzKA+vJiSV9nEC5+vZYh9P59ozuCeS88iJeIYbcm8SVMlli//SmlVPQ5DGAFHf0J9wYvtuxj6EEvWbGHwlofR6XTsKxxM/HULSAw7RiWxlzlwMJeCVW+i7jeBaTMvEn/7AoGP+O/6w6zaV0mo1o5NrWd2VpxP4ji8ZAH2gk1YJAmtRkt4lvcdI/IrmxjZtI7phh+pU8ex6EAml52Z4vU4Nq5dwXTJwqGAUQQ66tHlfU1J3YVtE5xeQ5apWnw7NUfz0UWlEXTtaz5LtSIQCAQCgUAgEHgdtQYyZndfP+d5aK7/ZWPHDFF+euPs+yAxG9LO7r7NgGnw558hapBHsUarQ4Oj7dzdXv/mD7e3HUsSpEYEcKTaRGqEPwU1ZkL8tHx8w1h0GomPfy7GbLXz8c/tO6r/tQqOPDmbC19e31YWGahj4Z/GMe35HwF4/KIzGJYYyu/e/hmAvMdn8dqP+UzNiMFsNpHtmrNssTmoaGxpmxPckFeNv15Dcrg/IX5aLnplAwBLckp46LxMBsYEEuqvw+FU9pm32B1Y7U40ahV5lU2M6x9BWUMzapVEbLABlSQhScrGgC0FtVhsTvx0KkaluKWmEQgEvx6t4RhTl8hKetcOmC/5CP/PrnI16bDo3zpuxADFaWjts67yY5wn6Tfp2Nr9UqxN3DCpP4u2lhAZpGN2VhxTM6LJiA1m5R2TmPGvtQCsvmsyoX5arv3fz8w/qz9nZ0QzNcPT1al5hRoc8Prlg3lyUzPPzhtGv+a98LZSPzr/JQCuGp0Aag2TB0bx1PJcnp03lIVbivnn0n1tYy2+SRHuVRpbCPHTsq+0kZd/yGPNgSrsTkUhEeavxV+n4Wh9c5sLYOq9X7eNccuUNCx2J89+e4BzVYqLSYRkpFEOZOfD5/D41/t5f1Mhr10zkr98nIPNITMyOZS/TE2HTzxvU/4Ts/lwcyFPfpPLxSMTeH7eMPr94xsA3r42mz++uxWADzcrjiL9IwMJ0Ku597N9PHrRsF+0wVd2KUFW7qtgakY0GrWKH3Ir0WtVNLXYmT44BpVKosliR+NKNSzLMp9uK+HsjGgiA/W9XKFr5r22AZ1GxYfzhZ+A4MTgCwcNDyRJigEGAnuBm4G2pE6yLJskScoHMgEh0DiF+HBzERtzS3jwoiAGRIuJ/xPBE9/sx2xq4j+/j1Dy1/dBthypZUrjMs6U9hNCCTlFmX1ToFG5n7qGRortwTitgcRU5BNkqnbZYvYt6nLXECEZqdamM6JpPQeKK0nsYReJLylY9SZpdWtpNB+i9qzZRPTBtDECwelAtdGKXqtCkiRqTVafxSE3VdGMHj+5heaGKp/EYLY6CHfWoMaBASv2Zt+kYFJZmwAVSBI2lZ5AR9Mv2pHwq5GdOJuqMMl6VBYz2MxCoCEQCAQCgUAgELQyev7Jv4ZaCxldpM3qSAdxBoCf3oAGe69dZRmOVCspDQtqzAA0NNuY/Z91aFRS2yJeR5799oDHeXWTtU2cAXD/5567uq96czM/F9Ty3HcHAZgwIILIQD27jzZwuMqETq3C6mj/3pMU7scZbi4exbXN3PDe1l6fT0f0GhWyDJePTuL9Te2Z0VfeMYnYvrmnSSA4helqA54EuL2PjLmpS1cjR+okGPtn2PQyyA7PylaBhix7ijL8exFaTV8AWZedfAcNSxNBBi2b7pvWqSo9Joif75vGjwer2hb7u1y0dyrvfwatChwwKErPpzePUOoKupivc1hArSEp3J+Xrx4JwHXjUxmZHMadi3aQ7SZCiw4ygNXEiM/P5q0LXmJGTQCHKpv40+T+3HNuBlaHE5uj87zT+UNj+N24FLQqFePTIrC/pawzabGjdokZHp47hBsn9Scp3J9rxtbyv/UFfHLjOHSa9nRd/70uG3+dBrVK4vfjUvn9uFSP6/STypiW0e6YsXJfBSv3VQDwr8uHsTK3muoPt/GXqQMYEhfSSahRb7by9093cd/swaRGtq8vWe1Oxj25mhrXfOd9szO4cVIa17/Tni5nwdwhXDehH2c8/C0AOQ/OYE9pA3d/uosZQ2J46aoRFNaYGRjT80Zou0MRCrbS6lwy7snVLPrTOKwOJ2lRypxaRaMimDlQbiQySE9FYwsjk8OQZRmHU/YYx+GUWbilmDB/LTEhBkYme9npV9Bn8KlAQ5IkLfAh8K4sy7mSJAUCHWfwG2hNzNS5/43AjQBJSUkYjb9u0t1kOjVycZ8KcY5PDiDYGUqo1v6rfy/e4FS4pxdmhlPfqMPabMLW0jedCeICZNYFZBMhq8mTRnCOgb75+1eFgH8kaocarU6P1T8Oo12l5InuY+gj02iyRKCTnJT6DSLEoOmb9xRQ95tAo/kQZeFn0s9hwWg8vlx6fRlJkvTAK8B0IBzIA+6TZXm5q34a8DKQDGwGrpNlubCb4QSCk8p1E1J5ZU0+ETonc3zkngEQO/seypc/ixQcR/zouT6JYWRyGI9HX4qpOphqQyrTxo70SRyRWTOo3/ADgdYqnE6Z3bEXMzrcB+m1VGrCz3sEacsnqNKmQqCwlxQIBAKBQCAQCE4VVBotZ6eHQPtmas7Piuar3ZUe7Z6fN4zPtpdQ3WQhKkhPTZOV3HJlLqlVnHHzlDReXZPv0e+VDue98XNBrcf5+rwaj3Nrh8XB4tpmimubj+saXdEqdncXZwCsPVTNZcNEamCB4Bcx/jbI/6FzeatD8rAr4aLXoPhnUOvgjclKef+zYdbT3Y+rci1Md3LQcBNl6Nw2eAZ1mMca+XvY/l77eUAUhCT0/FwANAawt/TerjusPc+/RwcbmJed1PMY/xkGDnu7y7TTba7c0YVAw27xvBcoTkHDkkJZdefkzm7Vlfuh7gh89yCv/24pX+0s5fZp6UiShEGl9tjkO2doHF/vKuOROQPRa5TyEclh7HUtET930WDC00cDoFGr2hxfH5gzhDtnDPQQZwCdXELc2XCZiviv7oKd/kBnIc33ucoScE5RPX94Z6srllBC/bQcqmyixeagxeakyWInUK/hiYuzWHeomiXbS1i+p9xjrNL6FvYcbfAo+3ZvBXtKG9vOn16RyydbFMeolfsqeGTpPj7cXESov5ZzB0fy9DxlrvCfS/dSWGPmpslpfLS5kC92KKljnps3DIez/fVb1tDCxGe6+FvpgfAAHRcOT2Dj4RrmDovjmRXtgsgjTyrOXg1mG0W15l+cPqyVisYWXvo+jwfOG9z2uxb0TXwm0JAkSQW8D1iBW13FTUBwh6bBQJfvhrIsvwG8AZCdnS0HnYDUDydiDG/Q1+McGhREv8iAPh+nO3091ilZQRiNxj4d5+TMQEzyTHYUjWRCaizjM2L6ZpqLoOEETbiGiG3v02TwI2zinyAsytdRdckF553PF8u1yMZKkoZOJDs9oW/eU2DazIuomziHfvYWwkJPsorZ+2iAYmAyUATMBhZJkpSF8tm1BJgPLAUeBRYCwu9M4BNSIgJ4+pKhymdGgM5ncQQkZqG+8GUCDTokvW8cdUL8tdw/K52DP+djCE0gy0eClfPPGskz1Q9TlZ+DKjSev14402dfkpojM6kaeAWB4fGIqUuBQCAQCAQCgeAUQqVFLduRJGXT+cjkUB45bxCTM+J4c91h/jo9HbVKxYwhMVwyKrGt27sbCnj4q70APHlxFpdnJ6FSSUwfHI3VLpMWHcDDX+5l+Z5yXrhsGMnh/gT7abE5nDy6bB+7ShowW5Xd7/fOyuCasSn8kFvJbZ/kIHcw4xgcF8z+skaPsu/umESdycrlb2xiyqAoZmfF8fdPd7XVz86K5ZvdnotuAFeemcxFIxK47PWNDIwJ5IXLhvN/qw+17cJODvenuM7cFkNl469YjBUITnfOebTn+tY/tKQzPcvn/rv9+MpP4OMrPOsl18K+081B47YcKN3hqpd6TqMy61lPgYbOzQX0qkXw0WVd9+tKANEdsUOhfBfEZsHZD8AnV4H1BGzorVfSeuCnOF9ITjcHJEcXbkg9xNzzeoBM/6hA/jp9YLct/nXZcB4+bwhqyXNDZVx4ENTD0Dg/6CKtulolEWTQ9nDtzsTbXM/76HZunjK/kxhw6c7STn1yirpOL7Yk5yhmq4MVezt/RgBsLaxl7UHPPf8bD3uKBVvFGa20plupN9tYuK2Mhdu+Jsigwdii/E6+z/UUPT7xzf5f7VBca7Ly3/VHAMgt9/yMzCmuJz1Mw9Pf5vLR5iK+vu0sMl1uU/VmK/VmG346Nf+3+hAPzhmCn06ZT2yxOXA4ZQL0nsv8jyzdx9e7y5gyKIppg2N4d0MBCaF+TBwY2TYXWW+28uy3B/j7zAwsNgcVjRb0WKm2qOgX2dkR3+5wUt9so9ZkpbjWzLTBXQt0lu4sZURyKIldvJa8SYvNQb3ZRnSQHhnlddzYYuORpft4YM5gQv19N1/ujk8EGpLybvI2EAPMlmW59V1hL3CtW7sAIM1VLhAI+jiSJDEnK55JqUF9WkgCUJ86k5/lM9E6LZwVlcTx/ZvhPaKC9Nxw2YV9XpwDyu8/PFCP0ei7lAonC1mWTcACt6JlkiQdAUYBEcBeWZYXA0iStAColiQpQ5ZlkZ5L4BP2rPsSDCGcMXqKz2JY/t0KErc9TbE6nIarXmZAkg/EEZYmahb+hcDGSuz6SKq1RiLHXNF7vxPMT3nVqIo3cYvzc/YbB/L+xnTuPy/L63E0t1jZ+cZNRLQUUOKXiv/v/k1UpJBpCAQCgUAgEAgEpwRqLTjsbLx3Gk0WGwOilc1cl4xK9BBkdOTa8alMzYgmMczPY5FvlJtd/r+vGM78ow0eZQCf3DgOgD1HG1iy/Sg3TuyPSiUxd1g8I5JD0ahUlDU046+y0+RQMyIpjOV7yhmfFsGIR1eikmBgTBCyLPPilSOYMSQGg1bN/rJGWmwO7p8zhEC9BrvDiVqlxOYeo9Mp89iFZ3D+8HiCDVoWnJ9JQqgf980ejFYt0WJzsnxPGXcu2kmV0XJCbrNAIHCn9e+xgxrrthzQh0BARHtZV6KGUdfBzoWQNQ9W/1MpC+8PVUpqJAwhEDesvb3shD+uhLdnKOfaDht+9G7z4wPPhakPwvddiEsiB0JVLly7DN49r3O9PhgsjTBoDqTPgGV/Bf8IGDRTSQWb/wMMuRDihnbu25GSrYojSMyQntu5izK6c9A4Seg0KqKDDZ0cr8ODAqGeEyNIacXW7pR0z8wM8iub+M4lrPsluIszhsQFs89NBLjnqKfYYVBMEAcqunY/SYnwp9CV9qsjreIMd6ZlRFNntrK9G/FIT0zLiGa1S+gxOjWsLT0K0EnYePnrG9lw13j2ulw/1hyo4n/rC5gzNI7r/6ekbrk8O4mFW4vJiA2iyWLnaF0zByuMbCmoY9Wdk/lsewkqCe4+N4PSBuX+W+xOHvxiT5vbVHp0IIv+NI4Ve8tZ8NVeLHYnZySE8MAXe3C4pT479Pgs7vl0F78bl8IIV/qV19ce9kiDdujxWWjd0raUNTSTU1TPXz7OAWD/IzMxWmyE+uk6ua8cC/VmK7IMof5aCmvMJIf7o1Id+6blP72/jR8PVjExPZJ1h6opeGoOL3x3kE+3lRAZqOfucwe1iTbqTFbuWrSTxy/KIjbEgEGr8tqmOl85aLwKDAamy7Ls7mv2OfCsJEmXAF8DDwG7xAKX4GQhyzIFNWYCVY6u8+j0IeqqymhoqCcoaLCvQ+mRvPJ6crZvZnT2OFKjOxri9A2aLHaeWLyOMRULadLH8p/iy7hzVmafdKawOZx8+M0aLJX5jJ16PsP6dW8f5muqy4vJW/0O/v3HMnTcDF+Hc1KRJCkGGIgiILwZ2NlaJ8uySZKkfCATyO3Q77RMzeUtxP1QsLQ0Y9/8Fkb/VIwZo3wXx+ENSPoggp02Cg7sIiY0sPdOJ5q6IhosEubAQYAKXcEe9EO8nyaqoLyaM9mH2S+WAc4q1leWYDSmej2OsrIygpwNNAQMQIeDwrw9GPS+e430FcR7h0AgEAgEAoHglEClAXsLsSEG4PhcClvt6rtDr1F3Eme4c0ZCCGckuNxSa/JB60diWDwAsSEGj41Fc4Yq4vzHLzqDbNeYkqSIOtj7BaRM4OG5mR7ja9RdL+CoVBLXjE1pO08I9WPB+e19/XRqLh6ZyHsbC6lqUhY3rXYn3+wuY1ZW7K9eZFm2q5SJ6VGE+PXVrV0CwUmmVaCQfo5neXj/zm1jzuhcFt4f/nagc7nsctTwj1AEGvoQsDQoAo2kM+HufGjuYmFc32EVZ+TvuxZoDJgO81crYouu8I9QBBpqbXtakda0K7pAKN0Or0+EBQ1d93fnrWnKY3dtO6Y4MddCQ3Hndsfq+mGqUcZ0dyXpDosRDq+Bwd2kHla73ttsvz79FAA5H8Cqhz2KFpyf2Umg8dCsdB5ZfsijbMkt42mxOdBr1IT4aZj+wtq2ukExQbw//0wCdBrO+ddaksL9uG58Kjd9sB2Ad64fTa3JyqSBUby57jCv/3jYY+wv/zyB/lEBLNl+tM1RqicGRAfy1rXZ3PpRDlDPO9ePprHFzm0uAcKwpFCuGJ3EhvwamlpsqFUqSurMjEwJ45HzM3nIdY3PbxnPiOQwUu/9GoCXrxrJir3lVBlb2HRYSRNmc8i8sb4I2aXceGvdYerMNj7dVtIWT6sryENfdo59+gs/th2XN1ja3Ei+3l3G17vK2uoOVTYx4tGVHn1f+j7PQ5wBcOlrG9lZXM+SnKN8cuNYxvaPYHthnUebez7dxb2zM3h02X4C9Ro+/rnIo37wQysAuGJ0Ek9dMpRKYwu3fpTDs5cO5f9WHyIiQMfY/hG8vvYwMzNjiQ0xYLY62HO0gVvOSuDMx5U4541KZLHrPsw/qx+ZCcFcNMJTkGq1OymuM+OvUxNk0KKWJH50uaqsO1QNgNlq550NBQC89mM+ByuMzJ/Yj6ve3Nw2zrn/Vl5vCaF+3DljIAPCtQw7yRumvS7QkCQpBfgTYAHK3RZE/yTL8ocuccZLwAfAZsD72xwFpw35VSbuXryTuZlh/GFKqK/D6ZG8D25H7bQR9+cP0Rt8axHUEz8s+5jxtV+wqupy5l/7B1+H0yWHKowMrPmekc7dmB1lfJCXitma0ckOqi+QW2Zk4J4XiNS0sGq1zLD5N/o6pG45tOJVIkq+o6lsG/KYaUiq41dHngpIkqQFPgTelWU5V5KkQKCqQ7MG6Kz7Op1Tc3kLcT+Ue1A/4x78NX4+vR+hWefStG47DboYMoeO9k0shv6EBunRVx/CYYggevClPoljcHIM3+zI5sKWL9iuGUJcYj+fxKEz+LFen0iccReVgYMZmjFC/M24EPdBIBAIBAKBQNDnUWvBYeu93cnmxZHKYy8Ll1ePSfEsMNfC4msh8UyYv7LrTr+Q6CA9O4rrsTmcbDhQyV8X7mDzkSQevzALSVI2a32ec5TLspP42+Kd3DipP0MTlbnghmYb+8saOTNVEZNYHU4MWjX5VU3c+lEOs7NieeXqUeQU1XGg3Mg5mbGEu6UT/WBTIePSIkiNCGBnST2Z8cEcrjZ3u7BjtTvRqCSO1jf3KpwRCHxOTCbcV9ouYuiJ6Ax4sBoe7capc9ytcOAb5bg13UdYP+UxKFYRaLSKDgIilZ+OdIzD4BKOxY+AUmXxnKFXwMS72sQZ1uHXodvxjlKXcR7kLmsXbqi1kDwO0s+F0fNd1/iFG4zK90DBT4poZM0Tnetbn/PzGeDowi2jKhci07sfX5ahch+8Ol45//2X7eXd8cMTsOkVuH45pIzvXK9yrYc0HoWynZ5uJu5UH4KXst2eixPc5/5NNbDkBmiu7dQ1PtSPgqfmtIkUAGZlRrUJNP40qT+jUsIY6XJraGVEcig5RfVkJYTw9rXZRAcpwsT1905ta3PDxH5sL6pnYnpUmwvTP2YN5o7pA2mxORj+iPJZMyxJeb+/akwyNoeTYUmhrD1YRWqolsjQICICdCzfU0ZyuD/3fLabQbFBSJLE/XMGMzIljMkDo7DYnQBMGRTFO9crqX6uPDO5y9t1z8wMJqVHtTlQvHL1SPaVNjJnaBxzhsYhyzLvbCjgn0v3AfDm+nbBTp258+d8Ua0ZrVriitHJbC2s65RKrJXPtreLOtzFGd1xtL6zMGdncbsw6ukVucwdGt/mBtLKkpyjLMk52uv4n2wppt5swyHL/HyklsnPrmmre3OdkvLl5yOer5lWIQXQJs4AeOsnpf1raw5zoMLIxSMSMFntfLtXEf9kJYSQX9XUlpLNnY5pa346VI2qm83aR+ubuWuxshe34Kk5vT7HX4PXVyNlWS6k3Repq/pVQIb3IhKczsSFGDg3M5YRiT3kglaWeQAAIABJREFUOesjqM+4gOaGWnS641PIe5vw/sPZZSogOq2bD/M+QHiAjmJNClarBjN+WPzjMWi9Y1t0vMSE6NniP5L+jgKCEvu2e0pAv9HYj/6IKXzwb1mcoQLeB6zAra7iJqCjXUww4P1t+gKBiwHDzvrVDi2/lrPPOovaEZ/jp1W35Uf0Olo/Un//Go2Ht2DXhROWnt17n5PAmH7hNM+8mBWHJ5MUHsBNI7u3Hz6Z6LUaJtz0H0qPFjPYP4iw0BCfxCEQCAQCgUAgEAh+ASpt+wJfX8LpRJ2/CoZd2L5TvCtad2jXF3Xf5heSEuHPd/squPvz/YzuHwXAgXIjgx9aQbCfti39SVGNmWW7yli2q4xRKWEcqTYRqNdQVGsmLSqAsf0j+HBzERv/MZVPXDuCC2vMPPTlHt7bqNjE37tkN6vvmszXu8pYurOUQ5VNZKeE4adTs+5QNQmhfhytb2bHQzP4709H0GvV3DIlDUmSOFhh5Jx/te8KX377RAbHdXYgbjDbkFQQbOibzh2VjS2E+Gu9ZgMv8DHHIs5oRd3Da/bcx5UfUFKLTP8nnHmDcj79YVh8HYSldNsdAE2HtRGNHq75DOKGQ0OJ4syR4OkUapn2WLtAI0B5f8Ck7KwneSyEJsHVi9o7dOe60RuvTVAev3ug3S3DDam1rCtxBsDCa3oWvu1aBJ+7bd48lpQoraK+4s2KEKUjrb+vb/6mPC5ogKoDiuvGmD+1tyvd4dnPaQOVvv18y1uQv7rXcAxaFZ/eNJ5AvYqCp+Ygy3K3ruaf3DiWsvoWUiO7f/3dP6frtDIGrRqDVs2W+6e3OVMAaNUq5k9U3F9Gp4Z7uD+dkRDC6v3KYn92iiKsiA/1449n9Wsbc9Wdk0gM611YF+KnZeYZsW3ns7PimJ3VnvpZkiSun9CPswZEMsPtM6GVUH8tOQ/O4IWVB3nx/9k77zgpiuyBf2tmNu9sIiywsCxJchAREBAQCaIiKuphznp6euqZ8HfqoWL29M5wnjnnLGA6AygiiCg5KTmnhd1lYXP9/qienZ7Zmdm8M7u87+fTn+nu6qp+XdNdr7vq1Xvf/gHAh1cNpXfbZApLSul6+xcV8rhjXOQVlhDtcjCpfwZv/ew1+rjgmPblOiwxxsWlwzqwfFsuX68MHXrmt037yz1yAHRskcC63dXzQmsPUVMXeELY+BuILN0a/NkxnlC8FJWW8ceuyvvNRz0yi69uGB7Uy1dtibzp4oLQgCTEuPjzyE5hH8SqCv3HnEteXl7ED3xPGnUMe47sTvPUyPVI0r5ZAsPHTOTZn7JIiovh6uP7l1tYRhot3bFMuOTvbNm1l2FZrSrPEEb6jDiNwkHjaFsYAbM56gFl3hZfANKBE7XWngtdDlxoOy4B6GTtFw4zsvONO0L7bJrDld15hXw0+2dSUpsx6ZjuYWtnt+UUsuyXhUS37snIzsE//OoTpRTt0+JZuXIZ6fFdwmewAmxZuYBdP75GdLcTaNlyYtjkKN04jwO/fYC7/5k4MgeGTQ5Ki41r0cTuoTuTBUEQBEEQBCHcOF3h96ARaKb2Ly8Q/9lN4Hge+pwZPG+RNaij6r5v0zOz+pvVe/lmtXEF/6s1qOQxzgDvDFyAhZbLds93/Nrd+ay1Bp6Ouf/b8uOWb8tl+Tbf2crH/3O2z/YvNvfvnhnJnpnbAAs2ZPP8BQN49acNPvnOeHou957Wm1P6tsHhUJSWab5btYvn56xj3rpsfrtjTPmkspp8R2qtKSnTRDkdlJSW4VAKRzW+zTdnH+TGdxdz6/huvL9wC3dP7IlDKQbe9w0n9GzFf8+XkJlCDXG6YNj13u1uJ8Ed/g6KAxAbYKJJ59HmN5DHDQ/jHzaeJDwhUo48D5LbmV9/7B40ig5C/i5IzapcNg8BjDPM/iqEJPFHa29fxe6VvmnlYUk0rP8e2g/z9WoBEGeN0RzMhudGEZvUDs56CRxWe+LwM6gpLoDnxxhvJgMuNf8TVKz3shIghuqwZOpYYlwOYlzO8vG4UH10MS5nSOOMqtDCXT0ZR3VryZuXD+KYjs0CpnduWbfeV7uku5l980iuf+tXbhzXnce+XsPCjft48uz+KKW4cWxXTujVik8WbaNnG2PMF+Ny0r5ZPFv3HeK4bi1pkxzL6f3b0qdtMo99/Ts9Wrs5oVdrpp3amw1787n/s5XcNK4rfzmuM/fOXMk9p/YiOS6K3IJiznt+Ptcd34WhnZvzxtw/+GTpbkZ3T6d9s3hmLNnO/2yhaW4aewTXjOrC7zvzmLFkO1cf14ndeYVszyngmdnrfIw9LhqSxZZ9B9mbX8SGPfm0TY0vN6AY2yO9Qsgbf369YwyTnp7L+j2BjUFS46M4UFhCcaku3w7kecSnrlsm8vuuA+XbG/YeDHk8wLo9+Szesj9kCLjaIAYagiDUOY3Bevr47ukc332sj5VkpNLCHUMs7nqz1KtLYmLjKSqOfIOnGvI00B0YrbW2+//6CHjYCtE1E7gTWKK1XhUGGYUwc8v7xgXa8xceHWZJws/nP69gyKLbWB/dmdUdHqdHm4qzghqCpV++TKfNH3Jg909s7TuEtukhPtjrkU+//obx6+9n9rrj2ddpKqlhMuLZ/9UDuIv2UbxwK4XHjAlP2DStyfnsLjZl55O5cyVpV85oeBk8LP8IfnkHxt4GbcPjYUUQBEEQBEEQqoQjyjvwt3s1xDcHgnxXFBeYWdzDbwZ3um/aof1msK0mBsrFFd2hk2PN0s2pxDNGkdVf5PAbksjfC/FpweXZNA8+uxku/QqiAntBPv+Y9qzckcuHv5oZte5YF3kFxtvIHSf3YOPe/PLZw+N6pvPA6X3ILyrhilcXsiKIq3iApFgXuQW191oya/Vuxv3r+3IDEA/5RaVc/84irn9nEa2SYlEKtucUlKcfeY8x8shIiePHKaPQWnPNW79xYq/W7DlQyKLN+8sHuzLT4mmWGE1Ldww//L6Hq4ZmcPvMRXyyaBsXDckqdxt/8dAsMtPiOWdQpk8fbnFpGdv2H+JAYQmdWyZyoKCEp2ev5ecN2Ux6ei4Ak/pncPvHywAzK3rysz/x6iWD2JFTwKodubwwZz2DOjZj8tHt2H+wOGz9AEKYadYF9v5ed+Wd+IjxHjHy1tBGGKEYdIVZDmZD3g4Ycm1gYw/wNdD4dhrMewqu/RUKciC+ma+Xj7KyKpzctG2u1dPhvT+FPjRvJ/zzCDj9OZj9ILhbw0UzTDu5/GPfYz0eNLYvhlcmwPiHfL1eABRaA9J71sC2X4na9ivcnQbnvg9dxlRsUwtyjHEGQHE+OD115GecZzcW3LsW9m8MfV1EgEegPb9DSntwBe+PU0oxpJPfPVZ4AH56Co79W2gPMTWkfbMEXr2wH263m+6t3XyxfAdDO3sNRHq2SaZnG997deZfj2VfflGFMFl/G3NE+brToejUIrG8jzopNorHzz6yPD0pNopPrxlWvn1W/zZcOqJr+fYpfdtw1LSv6dA8gQ+u8obH6ZLu5oYxZjytbWo8bVPj6Z+ZytOz/iA2ysmlwzoENLwpLClFoVixPZevVuzkzpN7MLhjMzJS4nA6Fc9+v44dOYe4bkQmaQnRvHn5IP71v9+JcincsVE8PWstYLywzJ1yPAXFpXz/+27mr89m6oSefLViB0u25PDK3A3cd1pvbv1gCSVlmoW3jyYpLorSMs2u3EJm/76b2at3V/Ae8u/J/bjuba+nmOuOy+Lf321g7a58MdAQBEGoS/YfLOKD738lKS6W04/tG7EeNNCa7AXvsmfzGuKOuxBXWuC4ZpFAUUkZS5Yvo3nzlhFv9FJdlFLtgSuBQmCH7SXjSq31G5ZxxpPA68B8YHJYBBXCzrieke3ppiHJbN2CX+MGsTuuI8ckVc9ivS6Jy+hF0ZYZ7ItpQ2d3Dd1U1gEtMzqyaOtR7EvtS3xM+AwZi9K6E7/9e3Li24QvbJpSRLXpRfOD84lq3Ss8Mnho3Q8yN1VvNowgCIIgCIIghANnFJRaxgJPDQTlxDX2YdizBI6+3Lj+b97ZpK/9BhY8Bwf3wJkve8vI2QKP9YRx98MxV1dfhoL9Ffd5DC7ss8M9cjptww8eDxr2Gd77N8G/esPYe2HINQTks5thxxLYtaJC6AIPsVFOHj2rHyf3aEZ6WhK7cgu5+OUFnDMos9w9/Y1junLX9OXcckI3UhOiSU2IZmjnZqzYnkvvjGSiXQ5O6NmKC4a059nZ69iWU8A/JvTgh9/3kBDj5Jb3lzB1Qk+KS8u4/eNlXD68IxcNyaKotIzTnvqRkjLNPRN7cfvHy9iUbWbmjurWkm9X7QLwMc546pz+JMdFcd4L88v37cgtIBhb9x8ia8pMemcks3RrDjOXbC9P+8jPzbuHN+Z7DWY8xhkAL/1o1u+avgKAaJeDozJT+Wnd3vJjBmal8fOG7Aplrt+Tz6od3olZ89Zl897Czfz9o2Xl++avz+bxb8zgvDvGxZhuzXngzCOJdkX+xDOhjrhqLugyKKgjjz8DL/eGQqkt8Wkw6bnQxzhsfTbbfjW/f3wDn99s2rsbVhiDiHPehsT0wGXYsYwZope8Xvmxaz43v3Mfh71/mAVgzqOwb73vsfZwJwCf3wJf3wXXLoQkK5yGxzDOU46HOY8ZAw1/CmxhIgoPGI9Hjigo8Wuf7OG2nugf5GICeFwKF3k74ckBcPRlcNI/q5d31v3w05MmFE6/c+pHPotmiTGcO6iSMD+YMCWJMfU7vK+U4sdbR1XJltPpUFwzqkvIYzwGgf3apTDn1uPISInzMeTwGJd4vKu0To7jwTP6lKdfc1xnnA5FlNOB06GIi3YysV8GE/tlAHBynzac3KcN/3didwDG926FQ6lyL1RRTshsFs/5zdpz7sBMNmUfJDU+mvgYJ1HWxOjN2Qd55Ks1XDw0i8uGZHLVqG7l+esDMdAQBOGw5Mvf1nL0wpvZEt+TpR0foF+7CA3JkrOF7O+eIF8ls700j3ZnPRRuiYLy889zSf3mRha1OZ2sS28Ntzh1itZ6Ix5z58DpXwPdGk4iIVI5c0C7cIsQMYzs0ZZtbaaREOMiOS58VvLDjz+Jrb0H09yhcMeHySABmHxsb7b0eoRT3DFh9TQ18IJ72bFpNS3cLcIaNs198n2492+ClDAbHrY4wswwSWhahoWCIAiCIAhCE8ThMh40PDOidSlxX/7NrC982fxe/p05btVnZjvbGtDbvwmUE/ZtMNsrPq7cQGPhK/DdvXDjari/HXQYDqP+HlgugMJcI1tMInx6DSx5B25ZB3GpVrolt3KaAcsdS4w3D4DfvzQGGmVlFV30ewZwSgrhwyvNrPcTbf1jWsOiN6HzaI5un4Lb7aZnG/jy+uF0buk10k+Oj+LRP/XzKfq28d05f3AWmc18ZyFfe7x3oGlMDzMAO+fWUeX7xvduXb4eG+Vk5l+PJdrpwOFQfH/LceTk5pKc5PUe8efXFvLF8h3MvnkkLqeDjBQza33NtPFszznE9e8sYv2efPZX4qLd4yK+LikqKfMxzgACGmcA3Pz+kgr7PLOaA5FXWMKHi3eQW7SQJ87uH9Zwn+FCKXUM8ChQBGwDLrCFTW6aeDwU1JWBRjjxeIrYYTzmUlYCP/4L9qyG+c/AsTdWXkZhNZ7b9d+bX3vokTVfQXQVQ30U5xvDjaHXQ+fjodAy0Ni/2fe4XStg5k2w5kvf/XYDjaJ8eKwHtOzpG44GvPUSyoNISWHwtIYm3xjKsXFu9fMWWl6WIul6Goj6arPbplbfm29CNQ1S4qODH+9wqIDhczyeMlLjTRtWn8YZIAYagiAcpmQ0T2NR7EAOxLbnyDDO7K6UuBRcic1xFZQS1zqyx//TM7JYlXgUsW17h1sUQRAihDYpgd3PNiSFJWVs2pVNcmI86S3CJ8eh/Bz2L/kaV1Yv4jqErz13OJ1Et+hMVFl4Pyy1w8We6LY0d7iCW78JgiAIgiAIguDFGQV522Hpu8GPee443+3ti+CXl2DG9ZDeG46/w+zfPB8+uQYmPhnYKGLzApj+V7N+MNvMwl49E448r+I5lTWAMfcJ+PVVmLIJFr9llfMztOoNn9/qHdBTDjPT+5cXjScPgKh4M5B4f1vjTaN1H1jwApz4MGy3jAJeGu89p8dAY8McmPc0rJoBff4EY7wzo7u2qtwI2+FQFYwzaoL/II7Db8rxk+ccyf5DxTRP9O2DjHY5aN8sgY+uHgpAXkExN7yzmGVbc5jQtzXnD85CKWOY0a2Vm9lrdtM13c3Mpdu5eGgWox/9vrycsT3SSYqL4k3Lc8bEPuk8Nvko5q3fyznPzScQFw/NIjnOuI4vLCljyvhuPPa/NRSWBB50PX9we/KLSjimYzNufn8JzROj2bIvQNgbP75euYvX5m3giuGdKj22CbIRGKW1PqSUuheYCLwfZpmEUMR7w0uw9Rfz+5vN+8W8/5hfhwv2VR7ao1ps+8382r1lvHmm8f5ZVdZ/b5apOV7DuFK/PqBD+4yXJX98DDQs445dy+FDPw8mr040xilbFgSXo2C/MfKoqnFJfVJSZH5dNRgH0pYnkJqEBRMaFcd0asZblw+mf/sUig4drPfziYGGIAiHJcO6ppPV8h500SFaJ4d/ADEoMW6yLn2FvN1bcLftEW5pQtKlfVu63PCfcjdUgiCEh5LSMu74ZBnt3A6uHhO+8BGHikr5+0dL6dMuhYuGZIVNjplffUm3RfeyMbEf6Zc+WqFDrKFY8tY/SN01j93zE0n+ywckuMPjuWnO/HlEfzuVXb2uZMIpp4dFBoD3F27hzZ83cd6g9kw6qm3Y5Ji3di9vz13D9eP7BLSeFwRBEARBEISIYY8JG8GMG6qXb4Y183nnUnjzLO/+314zof7mPg5j7ob+F8IPj8Dyj2GnN2QFD3f0rr99tnf9f3easCb2Aa+CHFjwPEQlmJncM2+EHL+Z23t/Nwt4wwdkrzPGGQBf2bx0rPg48DV9dx/MftB3XyQMAgbB5XRU6VvUHRvF8xcOqLC/XZoxIunYwngEGdK5OQD/Obc/nVsmckS61xjlnom9KC4to7jgIA6HYkin5vxx73g2Zh8kIyWOt3/eRJTLQVp8dLknkGtHdeH3XXl0a5XEn0d04lBRKX/sOkBctBOtNb9s3MfZA329H47r1Yq8ghKGPvCtz/77TuvN/320tHz7T/1b886v21m4cV9VqqrJobXeZtssAUK4HBAigsFXw8KXKj9uxScm9EVt6DwG/vifdzt7nfk95Pe8bF9U/bKXfeD1oFFVPJ4mIHTePasrhljxZ+V0WP8DTKljI5aaUGIZkjlt7fCiN4mf91/48/eh83oMNGR60WHBMZ2MgVZRA5xLDDQEQThsaZsaT15eaeUHhpvYZEhxiJWmIAhVQgN5BSUUxIY3vmuZ1uQWFHOoqKTyg+sRXVKE0mW4dDFlZWGMf1lWhELj0KWUhXIBWc+UFBeSrA9S5h8/tIEpKC6lrExTVBrevqmi0jIOFZVSEs57QxAEQRAEQRCqQmp72PCDWbcMIIr6nEt0i85mICxrmDG28OCK8w5KBePbe8zv9OvMUh1+/Hfg/TNtLv/9jTP8Wfqe+d2zpnrn9jfOAIhJqriviXOiLdSKB6dD4XQ4KbZ98rmcDjpZxh0XDe0QME+3Vt76i4t20rttcvl2l/SK3kiSYqNIio1i2V3j2JVbQGp8NHHRTmKjnAzr3JzhD38HwO3ju3Dd2O6kJ4Uv5GhVUUpdA1wE9Abe0lpfZEtLA14AxgJ7gNu01m9Wo+wOwHjg3joUWagPWhxhvE882gNytxpPP0dfarwAfXmb8RQEpn1L7eDr7aK6FFuz9ONSvUYZaZ0gO3joIB/6ngOLg9yG718CaR0DpwVj53LvuidUlp2T/+U1+qsKBfvhwC5IbFk9Oeqaonzz6wm/A/DxVTjBGBo6qhDKorpjMyVFJiROdO29NAlNEzHQEIRGwqGiUg4UluCWEOl1Rs6hYopLSon0Ki0uLSPnULH894IgVIkop4PHJx/JgQPh9WaTEOPi2fMHhN22bPwJJ7OgdTtaJKfRMowdQj3OnMranz+jWVYf3MlpYZNj5LARbOr+GcNd4TWMOGdQe4Yf0aJGcSfrkuFHtKBvegzJyYmVHywIgiAIgiAI4WT8wzDyNlj7LXQ8DhLTKTxUSLTbDcf+DYoLIKkN9DwNVs2EHhPhvYvg4F7YtcK3rNb9zICVLqv6QGBVOOtVePeCuinr2JuMEUjbAbDpp8qPD7MR+uFKYoyLxBa+31OZzeJpnhjDpKMygMgIf1pFtgHTgHGAv9BPYSZVpwP9gJlKqcVa6+VKqVYEDltyhtZ6h1IqCXgFOF9r3RATs4W6wN3KGGg062w8BblijLchj4EGwOh/wMKXYd0s774xdxsPQ1VBWZOrTnjAGJ5lr4PMwVVvlzsdF9xAAyB/T9XK8WD3CPLzMxXTY2owQPHUILi1ikYsqz4zBiDXLYYo2yNYfAj++Bq6T6j++cHrDcQZwJNR0QEzQTYoATxorJwB75wLN6+FhOaBsz1/POxYYox9BCEA4Z1aKQhClbl7xnKmzVxBYUkj8PjQCDhUVMrVry/k+e9rYeHaQLz043qmfrqcbfsrj+koCIIAJo6uCrdlRITIkRDjYuTAo+jcJsgHUwOR3Cyd/uMvpn33o8Iqh1KK9s3duJzh/QxwOhTtmyXgdETGfSoIgiAIgiAIEU90PCS3hf4XQEo735nAAFGxMPgqM6h49KVm0OiiGXDVXLh6PvT5E1y/zAwWXTkbrvoRrl0IF073lpExAC77Frqd7N3ncMH5H0Pzrma7z58qiKaj4uGK2cYopMs438Tup8CIW+H4f8Dou8y+XpPg/7bBqf+teJ1jp0GfyTDgErhzD1zyhQnFAtCqD/xlAVz2DVz7K1xpc01fXP/x4oWq88vto7ltfPdwi1EttNYfaq0/Bvba9yulEoBJwB1a6wNa6znAp8D5Vr4dWuthAZYdSikX8BYwVWu9uoEvSagNZ71q2q2OI7374lKh3WDvdmIrSMrwbg+8AobavBF1HhO8fGc0nPQotOkPHUbASf80+3udDsNvCZ7vtGe96/GVTAAqzIWYUMYH1cTjqaj3mTDy/0If+7dV5vdQNuRs9UmKe28yzArgCemrv8OBnbDPLyzK57fAO+fBtmqGetm71njIKMw1264ABhqFB7zrJYUVw8tozwQnm+fVn54yv7tXBT/3jiXe9V9eNLL4lKth/6aQ4gtNG/GgIQiNhCMzU8ne5yDKIXZVdUG0y8GADml0SIr8QZkerZPIzc0jKS4q3KIIgiA0OgpLSnlnwWZ6tYiiv7gigoPZsPgtyBwN7q5hE2NXbgHTF29jQr82tHRHvqtbQRAEQRAEQWi0KAUtu8Hpz/ru9wxUdRjuda9/4kOQcRRMfsPMCHfFGnf7iS3gmp/N4BUKRk6Bxe/A7AcAOHDtKtxJ1iDgue/CutkmJMs39xiDiyRbKI5hNvf4/c6Gj//sK1ffs2GIn4H9n+eYcweaqXzJV/DiWDPDWhDqhyOAUq21PRbPYmBEFfKeDQwC7lRK3Qk8rbV+x/8gpdQVwBUA7dq1Iy+v5l5R8/Pza5y3Lmn0cjiSod/lcKgQKPTun/Qm7n+Z0CEHHInoY+8gfusinLuXUxDbguK8POIyBuLa+jN5E57D/VgWADsv/ZWkXT8TN91q80qLyIttDWd/arZbHo26agk6Po3oDQsIYEoAQF7HE+G6P6C0CEf2WhIquYyy6AQchdXz4lDaogfO3Ssq7M9rOQDnWe9S2nYwzm0L8fhE1a4YVEmh77Eqsdxz+aG1cynJGk7cp1dQeOxtJGyaA5vmkHfU1T554lUUTuDg3s2UxnkNX+J2rsIFHMzZQ6m7as+GytlM4vPHUDjoWnRMErFAsXZQYD1bHtny9+2kzGEMT+JfOwHnrmXk3bilvJzY4iKigIL8PNTnt+PcsRhVXGjkLCii1P9Z1RqUKi8/b8823DNuoMydQf4V88sPi1r0KrHf/B/5531BfmLF0FMNTaN/XuuYhpBDDDQEoZFw1oB25OWlyCzPOsLpUNww+ohavew2FMO6tKBvq1gSY6TJFgRBqC67cgv5+Let5HVOon/njMozNHV2r4Il70BUS8gIn4HG8m25fPDrVjq1TKRlVzHQEARBEARBEISw4jHQiLPNxu44suJxHqOOtI5w3G2weyXEN/e66S/Pa41bn/FC5ece/5AJt9KmH/z8nK8MHmLcwV3rZw6C9F5QJB40hHojEfAf4c6ByiNna61fA16rwnHPAs8CDBgwQLtrOcGktvnriqYuR2KrziYUx8mPwEvjie15ErFuN1zwIWSvw52cag5UTuJTWhLX7mzoPxnuSoEep1aUy7OdmGJ+B14BP9uM6058xMpjHaeChHaKii/3KuQ46kKYdT+c/jwHmvUisWg3vBIkVEi3k8EVi/OkR4xeKMiFtd+YcFmAOyUNUiwvSe7jYOCVkDkY1et0+M8Q2LXcdiluE6rk332JU4VwcDNsmIVri9dIwe0qhbgUY9RQVgrRpn8oviQHEhPhh38aL0zKeK+Ij0+gynHg95sB9pjNP0LWMFMtMfFE+eVP+PJ6o4faDYJdy4xc8bHgjDIhYlxmTCbWpWD+EyZTm/5Gnrg4iHHCyunQ5yz48v9gww9w+XfeazxkjD0ch7J9/++N35rzl+6nLCEhIp6VSJABDh85ZLRPEARBEARBaLK0S4vnwUl9SHSWhFuUyKDdIDjlSYgNr7HKsC7NaZkUwxHpkfHRJQiCIAiCIAiHNZPfgoUvQUr76uU761XzW5sJUIOu9K53GlWzMqLiJMSJUJ8cAJL89iUBkT/zT6gfTn7MGDBExZnt9kNM+CgPMW5o3desX7MQYhK9aUrBlM3evIHoOh6+mAL9zvEaaIy4FQZe7ntcbIp3/fh/mHD7SGhTAAAgAElEQVQinUbB71/DvKdg1O1w7E1mX9uj0QcOgNsWesjd2hgbrJ4JLboZ70k+5SdBz9OMwd7BbN80pYzXJQ/nvA3rfzDGIIMtzxiekCgFud78JTZvRw+2N8dmr4dtv3rDWeVtN8YR394Dv71m6hpgxvVwyZfGqKO0BJwBhri/vsuEJYmK9+47aEUtKi2qePz2xfDiODj2Ru++vO2wYxm8fTZgTZj2hEkBY1TouZb5T8M3d8NHV3jTN/zgXX/nPPPrijZlvj4JRv0dcrdbB8iE7MMVMdAQBEEQBEEQmjRd0t2NwmNSg+BwQus+tetArQOinA56tqnDOKiCIAiCIDR5lFLHAI8CRcA24AKtdXF4pRKEJkLLbjD+wXBLUXOi4kyIE10GKz7xhm2pCgezYd96E9rFn2/ugYL9MO5+M7hWHQrz4MAuaNYpcHpZGehSM0tbiHTWAC6lVBet9e/Wvr7A8hB5hKbMgEuqfmzzzubX3g8T62/v40dqlq/BB8Bx/1fxuFhbv8qxf/OuK6cx0Og8xhhStBvom++8D4zXoR6nmO3s9RAfwHuRh7jUytvUlEw48lyzePB4PirMg4N7Aueb9x/v+oGd5ver22HVTLNeUgRl1qSr3avgx39B1xPhhTHGWCNzMOzbaNrSwjyY86hv+fvWew1k7AYaymnaYA8//NO7/u00430WAOO9g1n3e9NzrBAoxYfAEWCY/bXTKl5TQQ7MuAEO7IBPr/Wm2w0/6oqCXGOgEsiARYgY5N8RBEEQBEGoIzbtNTN2MpvFV3KkIAiCIAiCIAjVZCMwSmt9SCl1LzAReD/MMgmCEAnEJMH6GbgfzTTb7jYwcgqsmmFmhw/9KzhjzCBfVKwZ8Fs1w7juz9sBf3wN4x827vmd0TB2Gnz/CPzwiClvwfPQfhik94TkDFPON3fD0OvMrPXlH0HbASZPv3MgrTe8cSZs+gnu2FtxkGzvWnh+tJnt/o/9ZgBVCDtKKRdmzMwJOJVSsUCJ1jpfKfUhcLdS6jKgH0YHDQmftMJhw59/hC0LAqc5XcZYpLtfyJKOI+DOfeBwBM7XebTvdlqH2ssZUL4oYyiQs9kYu1WHTT95y9ix1Lt/3wavUcfGuSb9uVHGG0je9grFcGgfrP/erK+aAS+MNZ5F7MYZ/pQbZwSh2PKgsfoz+O31Kl0OAFt+rrivwD96Ui0pK4UH2sGR58PEJ+u2bKFOEQMNQRAEQRCEOuL2T5ahgFcuGVjpsYIgCIIgCIIgVB2t9TbbZglQFi5ZBEGIMIb81Qy8ecjbBtP/atZ//wpmP2DW41LNYF0gPr/Zu77ghYqDdxvnmMXOj/82i50l7xLXqi9stQbiHu0GJz0Krlhj2LH0PVj+off49y82s9zts86FcHE78A/b9nnAXcBU4GrgRWAXsBe4SmstHjSE+qdVL7ME4+THAu8PZpzR0DijTZiSmrJ/o+/28o+862u/hV0rzXog44xAbJ4Pr0yo/LiqUFXjjNhkyBwCaz6vmFZTDxr5e0A5jOeTwgPGY25UHOyxnPz89hqsmwWT3zSedMEYqpSVQMfjvIaBpcUmJEvz/mb7qzsgobkxQPSnpMgYNKa2NwaLQq0QAw1BEARBEIQ64qIh1YzXKwiCIAiCIAhNEKXUNcBFQG/gLa31Rba0NOAFYCywB7hNa/1mNcruAIwH7q1DkQVBaMxkDoKzXqV4+UyisgbCxp/MoOBiv6YlmHHG4L+YQSc0DLsBZvwNSg75HhOV4J01DZDey7iuz9/t3dfrDNiyANdW2yzp/N3w7vnBZV/+Eaz+IriBRmkJbPsNWvU23j9qg9beQbmCXLPuCUEAZpBPKYhOqN15Gila66kYY4xAadnAqfV5fqXUBGBC586d6/M0gtCw2D1nDLoK5j9t1if8G5p3BVcMLHoDep9pvGP8+LjxZlQVNvxQNzKm94adSwOnnfyYCU0CMOoO4w1k9efe0CV2Jv4HPrkakttBm36wcrrZf8KDkNTGa6Ax5m74351mvSAHx9Zf4JspxnDiqIusECralLNvPXQdb449tA8+vMIYYuTtMCFf0joaHREVD1fMhv8M8sqTsxmeOdas9zoDllmO54bfAv3ONqG2Pr8Z1n5L1Ig7YcQNMPdxc8zQ62D3ahPCpVkn4+njgUxv2WPuhkF/Nv9fIIoOQlE+JLYIVutVoyDXGJ949FLuNkhoWdEzVUFu5aGDIgwx0BAEQRAEQagjRnVLD7cIgiAIgnBYMvSBbykpPMjOQ4FdhC9NhqwpMwOmpcfpoPnS4zQZKRK6TBBqwDZgGjAOiPNLewooAtIxbuJnKqUWa62XK6VaEThsyRla6x1KqSTgFeB8rXVRgOMEQThc6TGRgnajiHK74ejLzL5TnoCyYti+GNytYP4zxjX+pV9Du6Nh1WfG28bRl0HZNECbgaAu44zhgnJAaaHxfgHw87PQrIsJH+BweY0dNs41A4vdJ4DDRemzo3Dm74TJb8H3D0NGf0hsaQbrmnU2RiAb5sDa72DJ28YYZM5jkLvdDLwV5JgBseUfw9ZfvNc44BLYNM+sJ2WYgbmep5oBsvhmZnb0ik9g2yJiOo+DODfEpcGq6cY4o+gApGaZmdc7l5lyep9pXOIntTF1o8ug28kweqqZgd26HyS0MIN/zigTFiZnizEY2TAHTv1vxYEyoUZoracD0wcMGHB5uGURhDpj8F9MaI/MwTDyNhj6V/L3bCaho82QIMPy3pA5GJofAR/9Gc7/0HiomHW/b3lHngd718HAy+D9S8y+Ft1h90rIOAq2LjT7TnsG+k6GdbPh1VMCilba+iicBXvhlH+bMCkAaZ0gey1cMcu0dd0nQIcRxqtEbLI5Ztx9cF8b38JuWGFCYLXpZ8KtxKcZ+dM6QftjoPiQCS0z6g5zTJdxxoPSj/8mAcsT0+5VXgMWO33PMe33yk8rpm37zfwWH4Snjg54nYDXOAPg+4fMYiN29t2w2Wbw8sxwozsB2hzpPY+H/91pDDCG/c3orGUfwJ7VRk/Nfsh47yg6YIxO9q41hjpj7jJGJ+7WRheWFZv/a+UMyN0KmcfgyBgKhUlG/62b5fVc1fdsKC0y53HGwICLYcOP0PdP8NXt5phx90PP04xxZvOu0LqvdXHJpu52LDX1lNASjr/DeGDZvhiS2kJcCmz+2eRP72F05uYFRhfHpwWv11ogmlMQhMOWzdkHUSXFuN2VHxtOtNYUlYSIiSYIgiA0CgpLSol2OlASX1gQBKHO2br/EEv/PgJ3kJf7vPtvYsMDJwVOy8sLni9EmiAIwdFafwiglBoAtPXsV0olAJOAXlrrA8AcpdSnwPnAFK31DmBYoDKVUi7gLWCq1np1PV+CIAhNAafLLJmDzfa4+2DErWYgBqDbid5j7eEAEpr5luFh8FWBz9N+iFksDl7wlff94TzboJjHcATMwGHfydD1BHjvIvh6qk2WKDNw5Yo1RiLaiuj0y4veY3atML8/PxNQpOglbwSWdd8G3+2l71U8ZtUM35AxoVjxCYy5xwx0JbU2A285myElC7qMwbV9FcTFmwGyjiOqVqYgCE2HE+7z3Y5OoEyF+L7K6A/XWF6IRk4xA/47l5lB/g4joMUR3mM9Bhp//sFrNLdzBbToaoztwLQ7N6wwXjqWvGvawCPPgwGXcDC+He7EROOVAeDYm2DINVBcYNqzNkea/c06+crosmyPY1O8HkKSM8yvPfTHked516Pi4LwPvNstuxmDlXfPR0e7UUV51nF+3prA1xvUiCnesF12Txxjp3kNFUIx6nb4dprvvjNfoXTWQzjXfefd5zHOAD/jDAVoszr7QbPYWfiy77a9TI8nkmCs/oyg/psWv+VdLy2E+f8161/ZPJ98eZtZqsL62bBlQcX9luFKoivOGE+e8CAM/nPVyqwmYqAhCMJhyYY9+Uz5cAmt4jQPnzOYKGeExGQLwOvzNjJn5RbuOWMALZNq6cpQEARBCAub9h7klg8WM/noTE49MiPc4giCIAiCIISLI4BSrfUa277FQFVG7c4GBgF3KqXuBJ7WWr/jf5BS6grgCoB27dqRl5dXK4Hz8/MrP6ieiQQZQOTwJxLkiAQZoLHJ4YRatgt1I4dFxnBcJz6OKj5IacYgVMF+Slv3x7npR8qSMtDuNsZAIyoOSgrBGUXUsrcpS+2EY+/v4IzCsXMpzl3LKOkwitIWPShr1gW99AOcHY/FuXUBZQktcGT/QckRJ6GT2qJdcVBWgnP3ClTOJko7HIdr1aeUdDkB7c7AuWUezk1zcGSvpbRVX1wb56BjU9AJzVGFeZS26IGOSSJq1Se4Ns6GL24FQCsnStsmmX03rdyFUnGncRS06F93lSwIwuFBYktIHAWdRgU/xhnlXU/vUTE9OQNG3GIWO3l5Jm9qe7j2V0jtYIz1/H2/+eNwwKQXoN1AKCkynh1qQo9T4PJvydexJMbGGOPAuFST9vmtxovEhdPh27uNB6aBV8Jxt5mwXsmZ0LwzpLSHtA7Gu1NJgTkmNgkO7DJ1l7sdfnkBBl9tPEGUlUH3U4ynkr1/wP6N0Hk0B1sPwf3R+cY4cNgNMO+/cMxfjHeMnC2wwpIlPg12LDOGGAueM94oht9kdNTu1cZAb+CVxrih3SAT5qsgB1p0gzn/MjIufhtSMo1nFWc09DkL+l8IW36haO8GojMHwKH9xlBlzVdw/J3GKCMl01yX1uY/Ky02cnQdb65103xTZmoHk/7JNcaApmUPc/9k9Dd51n5nrqnfedDnTOMJ5GA2pLQzIXaKD1LcvCfRbftBj4k1+2+rgBhoCIJwWBIT5SDa6SQxRuGI8JnMzRJjaOGOIdoVuUYkgiAIQmhioxy0SoolLSE63KIIgiAIgiCEk0Qgx29fDlCpqxqt9WvAa1U47lngWYABAwbouvCCEwmedCJBBhA5/IkEOSJBBhA5/KmWHAMvrLivd2DPXwAM9XjyGOuz22lbz0vtQLzbDd1Hl++L8S+nRVvveivbzOYeY80CmGFP30HN8qHQYy41YU5yt0On41AxSbBvPSSmmxnva74kP2s0CfEJRLXqTZSrggSCIAg159z3zcB6XeDvJaMyep9RN+fNOAqdl0cFN+/jHzQLwJ9eN4YgDmtI326s0vNU7/rwm73riS3Nb1Jr4zXDg8NhPIwANO9iFjAGFpd+6T2u+wRfeUbYym7VC058GAZdaUKVhBpfi7Fd13GWd4vRU00eT+gtzzHtBlKYl0e0vS6G+XndSOvouz36H9711CwT8sRD5zEmBJi/fH0nB5e340iAinLUA2KgIQjCYUnr5DieOf8oCg4ewOmIbAONE3u35tisRNzxMqgnCILQWGmZFMu/Jh8ZbjEEQRAEQRDCzQEgyW9fElC/09kFQRCE+iHLLyqVZ+AvaxhkDaMs0MCjIAhCXdBlTLglaDhcETY2pJTXuKMmeT2/MfWoH6Ii2xu9TMcWBOGwJS7aiSuCQ5sIgiAIgiAIgiAIQhNjDeBSStl7dPsCy8MkjyAIgiBEBEqpCUqpZ3Ny/B1NCYIgCE0NGZkUBEEQBEEQBEEQBEEQBKHOUEq5lFKxGI/7TqVUrFLKpbXOBz4E7lZKJSilhgITqULoEkEQBEFoymitp2utr0hOTg63KIIgCEI9IyFOBEEQBEEQBEEQBEEQhGqTkRJH1pSZ5dvpcZqdhyqGkNwQC6MfnR0wLVS+jJQ4vvjL0XUnsNCQ3A7YgkJzHnAXMBW4GngR2AXsBa7SWtepBw2l1ARgQufOneuyWEEQBEEQBEEQhFojBhqCIAiCIAiCIAiCIAhCtflxyiif7by8PNyB4sw/lsnXBTfj1tsClpMX3SZwWgHkPdkGCgPnI6aKacmZcMPSYJch1ANa66kYY4xAadnAqfV8/unA9AEDBlxen+cRBEEQBEEQBEGoLmKgIQiCIAiCIAiCIAiCINQfNyyFvDwIZLwBQdOypsxk6Y3Dq52vQtpUcRUuCIIgCIIgCIIgRAaOcAsgCIIgCIIgCIIgCIIgCIIgCIIgCIIgCILQ1BEDDUEQBEEQBEEQBEEQBEEQBEEQBEEQBEEQhHpGDDQEQRAEQRAEQRAEQRAEQWgyKKUmKKWezcnJCbcogiAIglAlRHcJgiAcPrjCLYAgCIIgCIIgCILQsCil0oAXgLHAHuA2rfWb4ZVKaNI81htyNvnui2kDhdsqHhtsf4i0DbGQ92SIfM2PqqbAQiSQkRLH6Edns/OQCpieHqerlLYhFrKmzPRJc8XE8+OUUXUvtBARaK2nA9MHDBhwebhlEQRBEISqILpLEATh8EEMNARBEARBEARBEA4/ngKKgHSgHzBTKbVYa708vGIJ4eSc5+axdFdRwLSqDoQHS9sQu4msAl8boHSl2VlQMV+w/aHSMlLi+OIvR4PbHTAfeXmB9wsRzY9TRpGXl4c7yP9a5bSpsOGBk3zSet/7fV2LKwiCIAiCIAiCIAiVIgYagiAIgiAIgiAIhxFKqQRgEtBLa30AmKOU+hQ4H5hSryd/ZSLsWRg4rQZeE6qdlpwJNyyturxNjcd6Q0FJ0Pp6M6YN7tjAaXnRbXDrWqQlZ7Jh6km+aUEG12szIC8IAUnOhKnJ3u2YNmyI3QZTAxxbk/YmRJ4dMT3onXN7tUUORUZKnHj/EARBEARBEARBaKQorXW4ZagTlFK7gY21LKY5xr1vpCNy1j2NRVaRs+5pLLI2FjmhbmRtr7VuURfCRCqHmd5qKKQ+fJH68EXqwxepDy+Hnd5SSh0JzNVax9n23QSM0FpP8Dv2CuAKa7MrsLqWpw9V340hLVLkaOppkSJHU0iLFDmaelpNywsnjUp31ZQm9N0VCTKAyOFPJMgRCTKAyOFPJMgRCTJA3cohuqtqNMX/vjaIHJElA4gc/kSCHJEgAzRNOQLqribjQaMuFLNS6het9YC6kKc+ETnrnsYiq8hZ9zQWWRuLnNC4ZA0nh5PeaiikPnyR+vBF6sMXqQ8vh2ldJAI5fvtygApuCbTWzwLP1tWJQ9V3Y0iLFDmaelqkyNEU0iJFjqaeVtPyhPqnqXx3RYIMIkdkyhEJMogckSlHJMgQSXI0JmqruyKlzkWOyJMjEmQQOSJTjkiQ4XCTw1GfhQuCIAiCIAiCIAgRxwEgyW9fEiDxIQRBEARBEARBEARBEAShHhEDDUEQBEEQBEEQhMOLNYBLKdXFtq8vsDxM8giCIAiCIAiCIAiCIAjCYYEYaPhSZ6576xmRs+5pLLKKnHVPY5G1scgJjUvWxo7UtS9SH75Iffgi9eGL1IeXw64utNb5wIfA3UqpBKXUUGAi8FoDnD5UfTeGtEiRo6mnRYocTSEtUuRo6mk1LU9oHETCfxgJMoDI4U8kyBEJMoDI4U8kyBEJMkDkyHE4ESl1LnL4EglyRIIMIHL4EwlyRIIMcBjJobTW9X0OQRAEQRAEQRAEIYJQSqUBLwJjgL3AFK31m+GVShAEQRAEQRAEQRAEQRCaNmKgIQiCIAiCIAiCIAiCIAiCIAiCIAiCIAiCUM9IiBNBEARBEARBEARBEARBEARBEARBEARBEIR6Rgw0BEEQBEEQBEEQBEEQBEEQBEEQBEEQBEEQ6pnD3kBDKRWjlHpBKbVRKZWnlPpNKTU+3HIFQin1ulJqu1IqVym1Ril1WbhlCoVSqotSqkAp9Xq4ZQmGUmqWJeMBa1kdbpmCoZSarJRaqZTKV0qtVUodG26Z/LHVo2cpVUo9EW65AqGUylJKfaaU2qeU2qGUelIp5Qq3XP4opborpb5VSuUopf5QSp0WbpkAlFLXKKV+UUoVKqVe9ks7Xim1Sil1UCn1nVKqfZjEbJI0Jr3V0DQGvdNQNAad0VA0lva+PpC22pdg9aGUGqyU+p9SKlsptVsp9Z5SqnUYRW2SBPuWsem1nda7Y5lSarHVpnv0XYH1vx1SSn1hPdcvWP9XqbVsVEqdapU3yypHW2kF1vpgK98mpVSRUqrEaie/t/K9Yx3nyXdIKXVHAN273TpmtJX2gSWftsr8TSnVw0r71CpHK6WKlVI/KKVaW2kvWef3yHqLrb6uUUqtttJK7M+plbbBVuYXSqk2trS1Vr5ipdQKpdSpVtrL1nV58h201h+0PRurlFJ7lHn3/F4pdaetToqVece/w3YuT77fPHVipU2zrk1b6V8rpXpYaY9Y5Wvrvyl/5pRS1ynzLHrOOdLvPrK3XfOVUj+qAO/JKkTbryr5ZlEBdKiqwnejCvAeoiq/7z33lM/7nF/dvqyUSlNKfWTJtFEpdU6g4/zkCXqOYHJV4XpC1kOQuqusvkP9V/Zvod3W/+zfhgd8VpRS0dY1e57NL2x5Blv3UYmVtsF2D/aw6syTtk1Z965fnXvSR9uuw9N2eJ6XO2z54pVS/1G258u/zoXIItTz1QDnDtoPoOrpHS7Y9SqlzvV7hj264ygrPUUp9YpSape1TK2FDJW1jUGfIyvvf5V5n8hWSk1XSmXUQpZg7bfnWbfXif1Z/9wvrUgptbSmclhlBv3OVUr9w94W2fb3V0aPH7Dq5LpayhCw/bfazF+UacP3KZu+t9KnKu/7g2fpWAs5AumZKr3LK6MXVimlttTi/EF1mlX++8r7jjbSL+/1Sql11j21TSn1mKrFd6kKrT+DthNKqZuVUsusZ2y9UurmepLhMmXarwPK9q5qpR1nyZWjlNpQ0/MLFVFh0l0qDHrLKlt0l68soru8ZUWE3rLKFN1FZOitKsjRoLrrsDfQAFzAZmAEkAzcAbyrlMoKo0zBuB/I0lonAacA05SlVCKUp4AF4RaiClyjtU60lq7hFiYQSqkxwIPAxYAbGA6sC6tQAbDVYyKQDhwC3guzWMH4D7ALaA30w7QBV4dVIj8sxfAJMANIA64AXldKHRFWwQzbgGnAi/adSqnmwIeYtjQN+AV4p8Gla9o0Jr3V0DQWvVOvNBad0YBEfHtfj0hb7UvA+gBSgWeBLKA9kAe81KCSHR4E+5ZxAXuBOOBsYBLQHXgfo+9uAXKBG4ASYA/wGrAfowcnABOBFsBbmGd9FtARcFppGtgELLXKXIh5Bi4AyoCHLDm2W7JGW/lKrHPZde+RQKFVZrqV9jtwvHW+0y35P7DScoEbgRRgMjAYeNtK2wL80yp3HzDVps/zgLbA/6zrsj+nKdbyjpW23voFKAAygc+AN4CbgTeVUi2BT636/a+VdjVGP8zHPBsbgRhL/jSrznda5T4DvGG9699j7fM8U+9jnh9P/QGssvL/17reT61fMN8IzwAvYL4V7M/cduAj4GPruHL82q4WQBcgg8DvyUHb/lDfLJXo0Mq+GwO9h4S670O9z/m3V08BRZa85wJPK6V6BjjOTqhzVKVvIdh7VcB6CFZ3VfhGDPhfBfgWegLzTHzgJ0+oZ+U3YCpw0C9PKjAdOB94HijGew9uA+4DzsDcp5vx3rue9OeBHPzuUYszsZ4x27MCRs+k4ft8CZFNqOer3gjVD1DP73ABr1dr/Ybfc+zRHb9ahzwGxGP0wEDgfKXUxTWUobK2MdRzdB1wDNAHaIN5T6jNhKHK2skUW72UP+ta6/F+9TWX2veLBWyPlVKdMG3Vdr/9zYEvMG1YM6Az8FUtZYDA7f82S4Y0oDm++t7DO/Y60VrX6Ns0hI6u6rv8zRh9U2OqoNPmAOcBOwJknw70t+6pXkBf4K+1ECeY/qysnVCYd+BU4ATgGqXU5DqWYQRGl060ZLC/qwLkY9qaWg2yCQFpcN0VRr0Forv8Ed3lS1j1Foju8iMS9FYoORped2mtZfFbgCXApHDLUYmMXTGN2FnhliWIfJOBdzGdIa+HW54Qcs4CLgu3HFWQcy5wabjlqKbMF2KUjQq3LEHkWwmcaNt+GHgm3HL5ydgLOGCvQ8yLyT3hls0mzzTgZdv2FcBc23YCRul3C7esTXlpDHqrAeqgUeidBqqLRqcz6rk+Ir69b4A6kLY6RH0ESO8P5IVbzqa8+H/LBLgnl2KMILoBjwBPWfuXAJdhjCNOB3bZ8izBDJgeE+B8B4C3befOBZJs+SZZ61lW2S7/NFtZnwMnYgbLpwY4lwvYChQGufY1wKEA+7cAa22yXGG159OAl+3PqadObGltLLk7AYMwnQ3l9zmw214vtnzfAf+w1UshZlDZLpenTu4N9twAf2AGxzcAo/3SpgGvAH8BDgZIeznQM2el5QMjbfvK7xO878nlbRe292Sq2Pbj981CEB1KJd+NVOE9hEq+4YPcb9OA16377Qjb/teAB/zrsgrPXqBzVJAr2PWEqodgdReqvkP9VwT/FvqBADqNAM+K7Zhc4IsgMk3DdMxWaPcx9/28APfu55jOugNY9zy29sP//8Cv3ZGlcS1Vfb7q8HxB+wFogHe4yq4Xm+6wtvcAR9u2/w/4oQ7lWYIxLgz5HAFPAw/Ztk8CVteRDOXtJH7vCpXkywJKgQ61OHdQ/YL3nWQDNv2LGVx4rY7vy6Dtv+0YF376PpDctZChSt+5BH6v6IDRN+OBLXUkT9B+T8x73cgQeZsBXwP/qcX5g+nParUTwOPAE3UsQ/n7u7W//F3VL/9oYENd3quylNdtg+kuwqy3qnK9iO7K4jDTXUSA3rLKE93lzR92vVWJHA2uu8SDhh9KqXTgCGB5uGUJhDLukA5iZiRtx8yMiiiUUknA3ZiZYo2B+5VxMfWjvwufSEAp5QQGAC0s9zpbLLc7ceGWrRIuBF7VVqsVgfwbmKyMm7EMjKL7opI8DY0Ksq9XQwtSDXoCiz0bWut8zGBDz7BJ1MSJdL3VEDRCvVNvNGKdUZ80hva+oZG2OjTDOYzb1PokxLdM+T1p6bUuGG8OPTHvPsqm79ZYeTSwUil1iuUKtBvmI36J3zmPwnzgP2ftGmSVfZdSai/mvcrfjexGpdQ2oAfG2MJT1pmYgfKFQJRVjv1c+zEeLNpgOtv8rz8d0+m2JEkAABzISURBVMkSKKyiA2iH994L9ZwqfN8TPeu9MLNNVmLqUCkT3qQQv3rB1Mlw4FVrexDGwKWf9W20VCk1yXb8tcCZyoRlae5XJ6XY6smPmzBeCp7AdLoFoqrPnL1OPNdsb7vs78lVbfvLv1mqoEMDfjdW9h5SlW/4St7nkoBSrfUa277FVLPN9j9HMLmq8F5VoR6q8f4R6Bsx2H8V7Fuohd++utBprfCre+t5vg3zbNxn2+9pB9YGKWsjZlbZUNuzYm93Aj1fgmAnVD9AWN/hlHE3bdcddvns63XSZ+HXblX2HL2Aee7aKKXiMd6GPq/l+UO13xutts5HL/pxAWbAb30Nzx+0Pfa0RVrrQP3Cg4FspdRcZVz3T1dKZdZEBj+C9p/a3oEC6fsJyrhwX66UuqomJ67md26g94onMAOwgTwf1ZRq93sqpc5RSuViBof7YgaEakow/VnldkIppYBjqfm3TygdHuxdVWh6RKzeAtFdHN66K2x6yzqH6C5fIkFvhZKjwXWXGGjYUEpFYVy9vqK1XhVueQKhtb4a4wrnWIzbl8LwShSQe4AXtNabwy1IFbgV4/44A+NSaLoybpYiiXRMB/AZmP+9H8a18u3hFCoUlvIcgZktF6nMxjTyuRjrwF8wrpQjiVWYGZA3K6WilFJjMfUaH16xQpKI6dy3k4Npt4Q6pjHorQaiMemd+qbR6YwGoDG09w2NtNVBUEr1Ae5E3O3WCyG+ZRKBHLtew3h9cGM6lM7CuM19HdNppYFYTCfbmxjjAIWZmZLvOZ9V3rvAdq31N9butpgP7DxgEcYV731Kqe5YM6kwnij+wIRFucsqKxHTcXOTJeMBjCtbOy2A7zEuQv9nT7Bk+QRjzHBtgLQ0zAx/jz4P9Zx66qQlJqzKnVadxGutS616OQ1jGPEmcKW9Xiw64dvx1taSvxhjYHIN5n9oZtXJExj3pm7r+u11Esr17COWDNdgwk34k0rVnzl7nXjekxOBlADvyZW2/QG+WULp0FDfjSHfQyr7hq/C+5yLWrbZgc4RQq5Q1xOsHip9/wjxjRjsvwr2LRTll7+2Oq0lpqPR5x7UWqdgZlPNw7p3bff89QHK8bQf7TGd7Z46B2+7k4Pt+bLaHUHwJ1Q/QLjf4QIN2nwBTFFKuZVSnYFLqIM+iwDtVmXP0RqM3t6KaU+6YwaIakyQdtL+rB+FTS8G4AKMZ5+aErA9rqQtAlNXF2Jc52dS0UV3TQjZf2q1mclU1PfvYv6LFsDlwJ1KqbNrcP4qfecGepdXSp2GmTX+UQ3OG5Ca9ntqrd/Uxk38EZhwWDsryRKKYPqzOu3EVMz40Et1LMNnwFlKqT7WQGT5u2oNzyNENpGst0B01+Gqu8Ktt0B0lz+RoLdCydHwuqsu3HA0hQXzp75t/QlR4ZanijL/F/hruOXwk6kfxnop2tqeSiNyNY9RzteGWw4/mVKthuBC275JwG/hli2EzLcDs8MtRwj5HJiXn79j4mw3w3SYPxRu2QLI2gejNPYCX2IGJl4It1w2+fzd9/4bPzdXGDflh3X4jXqq+0ant+qpHhq13qmH+mh0OqOe66PRtPf1XA/SVoeoD9v+zpiOkfPDLWNjWzDuQ3WQJSdEWk3yFIRIK6thmcU1kFFjjDTqWpZQaaU1lDPYcrCO/7fa1Mkcv3tqYw1lKwqRVliD68kPkXaohjKGyudfD57wG/4hNm4EplfWrllplb4zYvUtUM33KqzvZ6rw/kGAb0Qq0dME/hZaRACdhm9YHx+dRpAQJ5h2Pxf4Psj1ecKm7MUYcvwTuNOWVh7iJEC+d6w6ScLEGi/C5lYaY/B0Xbjbb1kqX0I9X/V4zoD9ADTAO1wl7cnvwMV++9Iwgzw7rPZjGrC2ljJUaLcqe44sGT6y5InBxDGfX4f1ErAPFuOBR+Pnvh4YZrURiTU8X9D22N4WWdsb8HUTvxh4ybbdzJIxuQ7rI2D/qfXf7QVaBsk3BfigBuerip6p8C6P8Rj2O9DF2h5JHbiJp5J+TypxE28dMxn4sIbnD6o/q9pOYAYl1wNt61oGK/0vVt3vwnikygGO9StDQpzU00LDh+cKm96q7HoR3SW6Sze83rLyiu7yreew6q3K5LDSG1R3iQcNyt2ivICxaJqktS4Os0hVxYWZ+RRJjMTEqdqklNqBmWE2SSn1aziFqgaawG65wobWeh+mcdThlqUaXEBke89Iw7iQflJrXai13ouxujsxvGJVRGu9RGs9QmvdTGs9DmP5+XO45QrBcswMNACUUgmYdkpcxdchjVhv1Qcjadx6p05ppDqjPmk07X0DI221H5bb06+Be7TWr4VbnsaG1nqk1loFWZID7cfoscet9SsxH8CzMLMT2mAGsbtjBkdnYbxDKEzn30HMbIZNfmmfYmadOKx8v1rHJnnkwHxMl2E6ED35vsDM5HHYz4fpuAJIwXQYFWI62HZaZRRhOmz880VZx/a30t7DGJb81V4nfvm2AsfZ0q7ExKu9F/NenWiV2d1Wh560rtZ1plnX/7EnzTruE+BmW75XgBIg3bbPUy+vemTEeBe53u9c/nWyF2PEUGCvkwAylteJtT8L09kxzzp2GL68Zl3TcbayrsTEpvXUUSLWfWJt/4QxFvCEwEix5T0N+N3vHvwd43XFvm8LpgPNs30BsCjA/fsFZlbWDZacO63FUxdrQ9z3zwb4/+OD1AMYAwKXUqqLbV9fqtBmV+Od0dO3MJLqvVdpTPziqrx/BPpGDKmng3wL+YfTqZFOs7X7czBxmIMeinmuM4Djgb9adXO9tf9dpdStAfJpW37/EEOCEJIQ/QBhe4dTSg3F6Of3/WTN1lqfq7VupbXuiWnbatxnEaLdquw56osZnMvWWhdiPD8NDOHCvboE64O1P+t2LsQMYByo4flGErw9Lm+LrLR2+LZFS/Btj4PJWBuC9Z868LaZ1ckX+mSV6JkQ7/JdMPX4g1VXHwKtrbrLqq4cNuqi37M2/fqh9Gel7YRS6hLMO+zxWust9SADWuuntNZdtNYtgQ8w17ushucSIpxI1FvW+UR3ie6yl9dgegtEd/kRCXqrMjkaXnfpOrDyaOwLxpJsHjW0DGsgGVtirJMSMW5sx2E6fyaGWzY/OT0dmZ7lEYwCbBFu2QLImmLVY6z1oJ1r1WnXcMsWQNa7gQXWfZAK/IBpuMMuWwBZh1j16A63LJXIuQ7TqLuse+Ej4I1wyxVAzj7WPRqPealZD8REgFwuS677MR3onueoBaazfZK170FMx3vY67IpLY1BbzVgXTQavdOAddJodEYD1UejaO/r6dqlra5afWRg4lveHG4Zm+pS2bcMZpC6xHofL78nLX03H+OOVWFcnc7CuEb9BOP54hirjCMxxgJjbXryRYzBgV2WZzED6PdY//9QTLiTbpiOj0UYV5rNMDPgv7PyvQQsxHQ2tsKENznTuqYZmA6lJGt5HNiGNwxLAfD3APXiub5mmI6bsVYeZZ0jx5LpDUu/zbPyJWCMP+7HdBp8D9xvpY3CuLB9wbrPB9nqxfMM/Gbd855nwGVdR7Z1/QmYGLh51v/W2zrXOxhjk1m2/7W9db3vWXXyJ6uscda5H8DMJHvKVieZmLb5W2zPou05TcLMpMkHTrLkUVRsu1606q/CezKVtP0E+WYhsA59iCDfjYR4D6Hy+z7o+xwV26t3rXpMwNyzORi3rAHbNb97zOccoeSq5HpCfj8Hqbt7QtV3Zf8Vvt9Ct2Bm2j2Ebxse6lmJx7guzsV0fibj2+7f5l93wBhMrOh4zEy/VZh7NwHzrGbgve/zgXOscgcBPYA44DFL1tlWmVGYsEl34NfuhLt9liWk7gr5fNXzuQP2A1CP73CVXS9Gf74aIF8n69lwYmJ47wF61kKOgG1jZc8RRk9/YD2PUZiY7VtrKEOodnIQpv134PeuYMsfB+wHRtWiHv6/vbuPtmu+8zj+/iRCgpASLSGSalqD1sN0eZw+KJ2SFjEZDxMdRMvQGYxVrbZ0VdrVqE5JVY1hGeqhJCHF0DI6s4xWUUzN0CFRT4kgQciVRBIi+c4fv9/J3ffknH3Ofcq59+bzWuuse89++O3v3ufe3/fsvX/79yurj7eumrf2O0le9yBgMelJ5iG5Trq/G7HUrf9JdeZe+Th1+A6U151AygkC9iE1sjuxi3HUzDOUfJenPU9UXhNzfNsCg7sYR1lO2yQfpw7f6/K8k8lPaJPyxZPAtG58LjXzJw3qifz5LSQ1cO1uvVEvhqGkYR06fH8vrDcoLzOe1GvaUPIT9351+zNpSe6iBXmrmf3FuWuDzF30kbyVy3Puat9Oy/NWgzjWe+7q9s709xfpxDpIF82WFV5fbHVsVXFuQzq5byNdXPgjcEqr42oi7in00a7m8zF9lJQU20hJ9C9bHVedWIeQum1ty5XRpZWk0ddewJXADa2Oo4k498yV7GLSl7BbqNOVVYvj/FGOcRlwNzCu1THluKawbpfMU/K8z5IuZK7Ix3hsq+MdSK/+krdaeHz6bN5Zj8eg3+SM9XQ8+kV930v77rq6ieMBnE/70AxrX62OdyC9ys5lCnntXdqHwFhd+LxW0j60xxpSLxkfzO9XFdZZQzqpLubJyH/jy0gn9ZV571SV+ePCvGJ5q4C/r5N715AailSvF6TGJt8szKt+rSyZF6SnZWr9vV6Sj9mFNeZ9t+Tv/J4G/wO1pr9O6nViVo15F5WUd32ed3ONef+S5/1XjXnvlpQZ5DqKjnXXi6QLOet8T6ZB3U+dcxZq59AdaPK8kY7d+Dbzd1/z+1yd4zCHdFHtReC4ss+0wTZOqxdXE/tT9zjUOXZDy453o8+KjudCz9Ta1zrHoPK/0lZrHu31fq3yjs5x1DuuNY85MCnHWm+93Ui9vLwNPAX8VavrZr8a5q66/1/rYdt1rwPQS9/hyvaXdAG6jfTUYvV6x5BuGiwnNfI7pBsxNKob6/4fkW783EjqkauN1DvOPl2Mo6z+nkS68fg2sIDUEHPbqvUnkS7cq4f/Hmue51LVTXye9hXSTaXFpO70R3dj23Xrf1KdOSd/Tq+TuvbfvbDudFJD0WV5uS4P1U2dPEMnvsvTA93EU57T5tb4Pxqb5/2M9t625pL+z7t8nk55/qxbT+S/31VVx+uKnoyBdMPribyvC0k3zwcX1juwxnG6r6f+XzbkFy3KXbQgbzXaX5y7NtjcRR/JW7k856727bQ8b5XFQQtyV6UVjJmZmZmZmZmZmZmZmZmZmZn1kkGtDsDMzMzMzMzMzMzMzMzMzMxsoHMDDTMzMzMzMzMzMzMzMzMzM7Ne5gYaZmZmZmZmZmZmZmZmZmZmZr3MDTTMzMzMzMzMzMzMzMzMzMzMepkbaJiZmZmZmZmZmZmZmZmZmZn1MjfQMDMzMzMzMzMzMzMzMzMzM+tlbqBhth5I+oGks3qgnGslfT///klJTzezbBe3tUzSTl1dv7dJOkLSjFbHYWZm5XoqB7aSpGmSTmt1HGZm1n1l51SSdpb0P5KWSjpT0jBJd0p6S9ItrYvazMysZ0gaKykkbdTqWMzMzJrh3GUDkRtomPUySdsAJwBX9mS5EXF/ROzcE2VJuk/SyVXlbx4Rz/dE+d1VKwFHxB3ARyXt3sLQzMysRHUOlHSgpJd6qOx1clcPlTtZ0u+qJv8IOE/Sxj29PTMza50a51TnAPdFxPCIuBQ4CvgAsHVEHN2SIM3MzMzMzMxsQHEDDbPeNxm4KyJWtDqQAWg68HetDsLMzOqazADIgRGxAJgDHNHqWMzMBqrqp6GUNH3NorPL1zEGeLLq/Z8i4r3OFuSnu8zMzMzMzMysFjfQsD5J0lxJX5P0RO5OdqakoXneOk+25t4VxuXfr5V0uaS78zAdD0jaVtIlkhZLmiNpryZiGC3pVkmvS3pD0mWF7T8g6ac5tjmSDi4pajzwm0K5syUdVni/kaRFkv48v79F0sJc9m8l7VYnvg5PIUvaS9JjuTvemcDQwrz3Sfpl3pfF+fcd8rypwCeBy/Lxquxn8ZhuKen6vP48Sd+uXPysfB6SLsplvyBpfMlx/Yakl3OcT1eOnaRBkr4p6bl8vG+WtFVe7bf5Z1uOcf/8/j7gCyXH3sys3xmoOVDSZsDdwKgc2zJJo8rqf0lDJf08T2+T9KikD9TLXVX7UHPdPG9LSVdLWpBz0vclDZa0C3AFsH8ut61Q5H0455iZdUqu53+R88kLks4szJsiaVauq5cAk5V6R5oq6QFgObCTpANyHf5W/nlAoYx1lq8RQ9l50tpzKkn3Ap+hPbdMB74DHJvffzkv9yWlc7rFku6RNKZQXkj6B0nPAM/kaYdJ+t+cix5UoQfAspyf50/I6y7JefLQPL1mHuvWh2VmZi1TL18WcuXMnMcek7RHYb1dci5sk/SkpCMK84ZJuljpOt5bStfuhhU2+0VJLypdkzxvPe6umZkNAM5dZt3jBhrWlx0DHAp8ENid9BRuZ9b9NjASeAd4CHgsv58FTCtbOV/c+iUwDxgLbA/MKCyyL/B8Lu984Fa1Nyao9jHg6cL76cCkwvtDgEUR8Vh+fzfwYeD9OeYby2LN8W4M3A7cAGwF3AL8dWGRQcDPSE+A7QisAC4DiIjzgPuB0/OwJqfX2MRPgS1JFzw/Tequ/qTC/H3zPo4E/gm4WpJqxLkzcDqwd0QMz/s+N88+Ezgylz8KWAz8c573qfxzRI7xofx+NjBW0hYlh8fMrD8acDkwIt4mNdh4Jdflm0fEK5TX/yeS8s9oYGvgNGBFk7mr5rp53nXAe8A4YC/gc8DJETE7L/dQLndEobzZwB6YmVlTlBp03wk8TsolBwNnSTqksNgEUm4aQft5z/GkXvKGA0uBXwGXkuryacCvJG1dKKO4/LyqGBqdJ60VEQfRMbdMAi4AZub3V0s6EjgXmAhsk5efXlXUkaRcuatSI/xrgFNz/FcCd0japLB8zZwvaR/geuDr+fh8ivZzp5p5rNZ+mZlZ39ZEvpxAyl9bATcBt0saImlIXu/XpGuIZwA35mtvABcBHwcOyOueA6wpbPoTwM55e99RaqxuZmbWkHOXWfe5gYb1ZZdGxCsR8Sap0t6zE+veFhF/iIiVwG3Ayoi4PiJWAzNJF7HK7EO6SfT1iHg7IlZGRPGJ5deASyJiVUTMJN18qvdU7QjShcWKm4AjJG2a3x+XpwEQEddExNKIeAeYAuwhacsG8e4HDCnENAt4tFDmGxHxi4hYHhFLgamkG2EN5Rt1xwLfynHNBS4mXQitmBcRV+Xjex2wHWms5mqrgU1IFyuHRMTciHguzzsVOC8iXirs+1Eq7xq4clxHlCxjZtYfDdQcWEtZ/b+KdENrXESszvu1pEF5FTXXVepFYzxwVt6/14AfA3/ToLylON+YmXXG3sA2EfG9iHg3Ip4HrqJjfftQRNweEWsKw2FdGxFP5mFFPgc8ExE3RMR7ETGdNOTU4YUy1i4fEauqYig9T+qCU4EfRMTsHN8FwJ7FXjTy/Dfz/pwCXBkRD+dcdB2p8eR+heXr5fwvA9dExH/k4/NyRMzpRh4zM7O+qVG+/ENEzMo5bhqpJ6j98mtz4MK83r2khvaT8o2zLwH/mPPH6oh4MJ9vVXw3IlZExOOkG2xujG5mZs1y7jLrJo+Jan3ZwsLvy0k3i5r1auH3FTXeb95g/dGkRgf1xhp+OSKi8H5eSXyLSU9zARARz0qaDRwu6U7SePZ7wdrGEFOBo0lPZFVaB44E3iqJd1SdmMjlbkq6aHco8L48ebikwfmGXZmRwMZ0fBptHqllZMXazyoilufOM9Y5xnnfzyLdfNtN0j3AV/MT1GOA2yQVW0SupnZDj4rKcW0rWcbMrD8akDmwjrL6/4YczwxJI4CfkxpzVN+Aq6Xmunl7Q4AFhc6eBgHzG5Q3HOcbM7POGEMa1qpYdw4m9TpRUavuLU4bRVWvGKx7LlJWf5eeJ3XBGOAnki4uTFOOp1Lu/KrlT5R0RmHaxnTMm/Vy/mjgrjoxdCWPmZlZ31SWL+dRqN8jYo3S0FyVXDE/IornUZUcOZJ0M+w56qvOP43OE83MzCqcu8y6yT1oWH/0NlDpfQJJ2/bCNuYDO5b03rB91RAeOwKv1Fn2CeAjVdMqw5xMAJ6KiGfz9OPytM+SumUfm6evM1xIlQV1Yqo4m9T1074RsQXtQ4ZUli9esKy2iPQUcvGpsB2BlxvEVFNE3BQRn8jlBfDDPGs+MD4iRhReQyPi5ZL4dgHmduJpajOz/q6/58Ba9Xnd+j8/7fzdiNiV1L3hYaRhtuqV1b6h+uvOJz29PLKwvS0iYrcG5e5Cap1vZmbNmQ+8UFW/D4+IzxeWqVXnFqdVGnIXVZ+LlOWDRudJnTUfOLVqn4ZFxIN14pkPTK1aftPcE0gz2/pQnelleczMzPqXRvlydGXB/HTxDqT8+AowOk+rqOTIRcBKaucRMzOz7nLuMusmN9Cw/uhxUu8Le0oaSuqNoac9QrqYd6GkzSQNlfQXhfnvB87M42YdTbppU+vpJvL06uFEZpC66/0KheFNSE/nvgO8QboBd0GT8T5EGoP4TEkbSZpI6qK+WO4KoE3SVsD5Veu/CuxUq+Dcw8bNwFRJw3P3vV8lPYncKZJ2lnRQHnN5ZY6p0oPHFXkbY/Ky20iakOe9TupNpDrGTwN3dzYOM7N+rL/nwFeBrauG7qpb/0v6jKSP5R6mlpAaDK4ulFUzd5WtGxELSGNdXixpC0mDJH1IUiXOV4EdJG1cVaRzjplZ5zwCLJH0DUnDJA2W9FFJe3eijLuAj0g6Lp/nHAvsSuoGtxmNzpM66wrgW5J2A5C0Zc6F9VwFnCZpXyWbSfqCpEa9SwFcDZwk6eCcq7aX9GdN5DEzM+tfGuXLj0uamBvQn0W6bvh74GFSA/5z8rnZgaQhwGbkJ5OvAaZJGpXL3D9fjzMzM+su5y6zbnIDDet3IuJPwPeA/wSeAX7XC9tYTUoM44AXgZeAYwuLPAx8mNSqbypwVES8Uae464HPSxpWKH8B6WLhAcDMqmXnkVoMPkVKWs3E+y4wEZhM6k7+WODWwiKXAMNyvL8H/r2qiJ8AR0laLOnSGps4g5Q4nycd75tIybKzNgEuzHEsJN3kO7cQwx3AryUtzXHum/dvOek4PyCpTVJlzOZJwJVdiMPMrF/q7zkwIuaQepF6Ptfnoyip/4FtgVmkBhazgd/Q3kCwUe4qW/cEUhfzT5Hy5ixguzzvXuBJYKGkRQCStiPdELy9zn6amVmVQj7ZE3iBlDf+ldRTYLNlvEHqAelsUiP2c4DDImJRk+s3Ok/qlIi4jdQD4AxJS4D/A8aXLP/fwCnAZXn7z+ZYmtnWI8BJpKEq3yLlsUpvImV5zMzM+pEm8uW/kfLXYuB4YGLuLfBd0rDJ4/M6lwMn5HMugK8BfwQeBd4k5S/fCzAzs25z7jLrPnUcitXMGpE0GTg5D9PR7DoXAK9FxCW9FtgGRtLhwPERcUyrYzEz21BsqDlQ0sXAcxFxeatjMTMzMzOzDYOkKcC4iPjbVsdiZmbWDOcus+bUG1vczHpQRJzbeCnrjIi4E7iz1XGYmVm5gZADI+LsVsdgZmZmZmZmZmZmZv2fG2jYBkvSjqQuYWvZNSJeXJ/xmJmZrS/OgWZmZmZmZmZmZmZm65+HODEzMzMzMzMzMzMzMzMzMzPrZYNaHYCZmZmZmZmZmZmZmZmZmZnZQOcGGmZmZmZmZmZmZmZmZmZmZma9zA00zMzMzMzMzMzMzMzMzMzMzHqZG2iYmZmZmZmZmZmZmZmZmZmZ9TI30DAzMzMzMzMzMzMzMzMzMzPrZf8PJPlNvciL1s4AAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 2160x360 with 5 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2020-05-19 00:03:34,114: DEBUG ==> Plot saved to ./img/original/logh21_inception_valid_conv_fc.\n"
     ]
    }
   ],
   "source": [
    "from mltools.libplot import Plot\n",
    "\n",
    "h21_deep_model_no_fc = cnn_model(input_shape=input_shape,\n",
    "                                 model_name='h21_large_kernel_same_conv_no_fc',\n",
    "                                 learning_rate=1.0e-3,\n",
    "                                 conv_layers=[250, 150, 100, 50],\n",
    "                                 conv_paddings=['same']*4,\n",
    "                                 conv_kernels=[(6,6)]*4,\n",
    "                                 conv_alpha=0.0,\n",
    "                                 fc_layers=[],\n",
    "                                 fc_alpha=0.0,\n",
    "                                 dropout=0.4,\n",
    "                                 full_dropout=0.0,\n",
    "                                 normalization=0.99,\n",
    "                                 last_relu=False,\n",
    "                                 l1_reg=1.0e-5,\n",
    "                                 l2_reg=1.0e-5\n",
    "                                )\n",
    "\n",
    "h21_inception_model_padding = scan_inception_model(input_shape=input_shape,\n",
    "                                                   model_name='logh21_inception_valid_conv_fc',\n",
    "                                                   learning_rate=1.0e-2,\n",
    "                                                   conv_layers=[120],\n",
    "                                                   conv_padding='valid',\n",
    "                                                   conv_alpha=0.0,\n",
    "                                                   fc_layers=[500, 100],\n",
    "                                                   fc_alpha=0.0,\n",
    "                                                   dropout=0.2,\n",
    "                                                   full_dropout=0.0,\n",
    "                                                   normalization=0.99,\n",
    "                                                   last_relu=False,\n",
    "                                                   l1_reg=1.0e-5,\n",
    "                                                   l2_reg=1.0e-5\n",
    "                                                  )\n",
    "\n",
    "h21_deep_inception_model = scan_inception_model(input_shape=input_shape,\n",
    "                                                model_name='logh21_inception_same_conv_no_fc',\n",
    "                                                learning_rate=1.0e-3,\n",
    "                                                conv_layers=[128, 128, 32],\n",
    "                                                conv_padding='same',\n",
    "                                                conv_alpha=0.0,\n",
    "                                                fc_layers=[],\n",
    "                                                fc_alpha=0.0,\n",
    "                                                dropout=0.2,\n",
    "                                                full_dropout=0.0,\n",
    "                                                normalization=0.99,\n",
    "                                                last_relu=False,\n",
    "                                                l1_reg=1.0e-4,\n",
    "                                                l2_reg=1.0e-4\n",
    "                                               )\n",
    "\n",
    "# create a list of models to train\n",
    "models = [h21_deep_model_no_fc,\n",
    "          h21_deep_inception_model,\n",
    "          h21_inception_model_padding\n",
    "         ]\n",
    "\n",
    "# train and display properties for each of them\n",
    "rounding = np.rint\n",
    "for model in models:\n",
    "\n",
    "    # print characteristics\n",
    "    model.summary()\n",
    "\n",
    "    # print architecture\n",
    "    model_dot = model_to_dot(model=model, show_shapes=True, show_layer_names=False)\n",
    "    model_dot.write_pdf(path.join(IMG_PATH, model.name + '_model.pdf'))\n",
    "\n",
    "    # train the model\n",
    "    log.debug('Training model {}...'.format(model.name))\n",
    "    model, model_history = cnn_fit(model=model,\n",
    "                                   X=matrix_train, y=h21_train,\n",
    "                                   X_val=matrix_val, y_val=h21_val,\n",
    "                                   batch_size=32,\n",
    "                                   epochs=5000,\n",
    "                                   early_stopping=400,\n",
    "                                   reduce_lr=125,\n",
    "                                   verbose=0\n",
    "                                  )\n",
    "    log.debug('Model {} has been trained.'.format(model.name))\n",
    "\n",
    "    # compute the predictions\n",
    "    model_val_predictions  = np.reshape(model.predict(matrix_val), (-1,)) #---- reshape to match the labels\n",
    "    model_test_predictions = np.reshape(model.predict(matrix_test), (-1,))\n",
    "    test_results[str(model.name)] = model_test_predictions.tolist()\n",
    "\n",
    "    # save the history of the model\n",
    "    with open(path.join(OUT_PATH, model.name + '.json'), 'w') as f:\n",
    "        json.dump(str(model_history), f)\n",
    "        log.debug('History of {} has been saved to {}.'.format(model.name, path.join(OUT_PATH, model.name + '.json')))\n",
    "\n",
    "    # save results to file\n",
    "    with open(path.join(OUT_PATH, 'convnet_test_results_h11_h21.json'), 'w') as f:\n",
    "        json.dump(test_results, f)\n",
    "        log.debug('History of {} has been saved to {}.'.format(model.name, path.join(OUT_PATH, 'convnet_test_results_h11_h21.json')))\n",
    "\n",
    "    # compute the accuracy and score of the predictions\n",
    "    model_score_val  = Score(y_true=np.rint(np.exp(h21_val) - 1.0), y_pred=np.exp(model_val_predictions) - 1.0, rounding=rounding)\n",
    "    model_score_test = Score(y_true=np.rint(np.exp(h21_test) - 1.0), y_pred=np.exp(model_test_predictions) - 1.0, rounding=rounding)\n",
    "    print('Accuracy of {} on the validation set: {:.3f}%'.format(model.name, model_score_val.accuracy()*100))\n",
    "    print('Accuracy of {} on the test set: {:.3f}%'.format(model.name, model_score_test.accuracy()*100))\n",
    "\n",
    "    # plot the distribution of the predictions and the error difference\n",
    "    plot = Plot(rows=1, columns=5)\n",
    "\n",
    "    plot.scatter2D(np.array(list(get_counts(num_cp_val, np.rint(np.exp(h21_val) - 1.0)))).T,\n",
    "                   axis=0,\n",
    "                   title='Predictions for the Validation Set for $h_{11}$',\n",
    "                   legend='real values',\n",
    "                   xlabel='num_cp (validation set)',\n",
    "                   ylabel='$h_{11}$',\n",
    "                   colour=False,\n",
    "                   alpha=0.65\n",
    "                  )\n",
    "    plot.scatter2D(np.array(list(get_counts(num_cp_val, rounding(np.exp(model_val_predictions) - 1.0)))).T,\n",
    "                   axis=0,\n",
    "                   title='Predictions for the Validation Set for $h_{11}$',\n",
    "                   legend='predictions',\n",
    "                   xlabel='num_cp (validation set)',\n",
    "                   ylabel='$h_{11}$',\n",
    "                   colour=False,\n",
    "                   alpha=0.65\n",
    "                  )\n",
    "\n",
    "    plot.scatter2D(np.array(list(get_counts(num_cp_test, np.rint(np.exp(h21_test) - 1.0)))).T,\n",
    "                   axis=1,\n",
    "                   title='Predictions for the Test Set for $h_{11}$',\n",
    "                   legend='real values',\n",
    "                   xlabel='num_cp (test set)',\n",
    "                   ylabel='$h_{11}$',\n",
    "                   colour=False,\n",
    "                   alpha=0.65\n",
    "                  )\n",
    "    plot.scatter2D(np.array(list(get_counts(num_cp_test, rounding(np.exp(model_test_predictions) - 1.0)))).T,\n",
    "                   axis=1,\n",
    "                   title='Predictions for the Test Set for $h_{11}$',\n",
    "                   legend='predictions',\n",
    "                   xlabel='num_cp (test set)',\n",
    "                   ylabel='$h_{11}$',\n",
    "                   colour=False,\n",
    "                   alpha=0.65\n",
    "                  )\n",
    "\n",
    "    plot.hist2D(model_score_val.error(),\n",
    "                axis=2,\n",
    "                title='Distance from the Real Value',\n",
    "                legend='validation set',\n",
    "                xlabel='error difference',\n",
    "                ylabel='#',\n",
    "                binstep=1\n",
    "               )\n",
    "    plot.hist2D(model_score_test.error(),\n",
    "                axis=2,\n",
    "                title='Distance from the Real Value',\n",
    "                legend='test set',\n",
    "                xlabel='error difference',\n",
    "                ylabel='#',\n",
    "                binstep=1\n",
    "               )\n",
    "\n",
    "    plot.series2D(data=model_history['loss'],\n",
    "                  axis=3,\n",
    "                  title='Loss Function',\n",
    "                  xlabel='epoch',\n",
    "                  legend='training loss',\n",
    "                  ylog=True,\n",
    "                  binstep=int(np.shape(model_history['loss'])[0] / 10)\n",
    "                 )\n",
    "    plot.series2D(data=model_history['val_loss'],\n",
    "                  axis=3,\n",
    "                  title='Loss Function',\n",
    "                  xlabel='epoch',\n",
    "                  legend='validation loss',\n",
    "                  ylog=True,\n",
    "                  binstep=int(np.shape(model_history['val_loss'])[0] / 10)\n",
    "                 )\n",
    "\n",
    "    plot.series2D(data=model_history['mean_squared_error'],\n",
    "                  axis=4,\n",
    "                  title='Metric Function',\n",
    "                  xlabel='epoch',\n",
    "                  legend='training MSE',\n",
    "                  ylog=True,\n",
    "                  binstep=int(np.shape(model_history['mean_squared_error'])[0] / 10)\n",
    "                 )\n",
    "    plot.series2D(data=model_history['val_mean_squared_error'],\n",
    "                  axis=4,\n",
    "                  title='Metric Function',\n",
    "                  xlabel='epoch',\n",
    "                  legend='validation MSE',\n",
    "                  ylog=True,\n",
    "                  binstep=int(np.shape(model_history['val_mean_squared_error'])[0] / 10)\n",
    "                 )\n",
    "\n",
    "    plot.save_and_close(path.join(IMG_PATH, model.name))\n",
    "    log.debug('Plot saved to {}.'.format(path.join(IMG_PATH, model.name)))"
   ]
  }
 ],
 "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": "3.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}