Keras_EDSR.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using TensorFlow backend.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:From C:\\Users\\jgolb\\Anaconda3\\lib\\site-packages\\tensorflow\\python\\ops\\init_ops.py:1184: calling VarianceScaling.__init__ (from tensorflow.python.ops.init_ops) with distribution=normal is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "`normal` is a deprecated alias for `truncated_normal`\n"
     ]
    }
   ],
   "source": [
    "import cv2\n",
    "import glob\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "from keras_tqdm import TQDMNotebookCallback\n",
    "from keras import regularizers\n",
    "from keras.models import load_model, Model\n",
    "from keras.applications.vgg16 import VGG16\n",
    "from keras.layers import Conv2D, Input, Lambda, add\n",
    "from subpixel import *\n",
    "from keras.callbacks import ModelCheckpoint, TensorBoard\n",
    "from keras.optimizers import Adam, RMSprop\n",
    "from utilities import random_crop, ssim, content_fn, test_edsr, ImageLoader\n",
    "%matplotlib inline\n",
    "\n",
    "dtype = 'float32'\n",
    "K.set_floatx(dtype)\n",
    "\n",
    "if dtype == 'float16':\n",
    "    K.set_epsilon(1e-4)\n",
    "\n",
    "def vgg_preprocess_input(x):\n",
    "    # BGR Mean values\n",
    "    vgg_mean = np.array([103.939, 116.779, 123.68], dtype=dtype).reshape((1,1,3))\n",
    "    return x - vgg_mean\n",
    "\n",
    "\n",
    "# get output of vgg conv layers function\n",
    "\n",
    "path = './images/'\n",
    "\n",
    "\n",
    "bs = 8\n",
    "scale = 2 # 3 or 4\n",
    "patch_size = 96\n",
    "img_size = int(patch_size * scale)\n",
    "\n",
    "IMAGES = sorted(glob.glob(path + '*'))\n",
    "sr_trn_gen = ImageLoader(IMAGES, patch_size, scale, batchSize = bs)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "# # Dummy data\n",
    "\n",
    "# x = np.random.randint(0, 255, size=(1000, 48, 48, 3)).astype(dtype)\n",
    "# y = np.random.randint(0, 255, size=(1000, 48*scale, 48*scale, 3)).astype(dtype)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "def EDSR(scale = 4, input_shape = (48, 48, 3), n_feats = 256, n_resblocks = 32):\n",
    "    ''' \n",
    "        According to the paper scale can be 2,3 or 4. \n",
    "        However this code supports scale to be 3 or any of 2^n for n>0\n",
    "    '''\n",
    "    def res_block(input_tensor, nf, res_scale = 1.0):\n",
    "        x = Conv2D(nf, (3, 3), padding='same', activation = 'relu', \n",
    "                   activity_regularizer=regularizers.l1(10e-10))(input_tensor)\n",
    "        x = Conv2D(nf, (3, 3), padding='same', activity_regularizer=regularizers.l1(10e-10))(x)\n",
    "        x = Lambda(lambda x: x * res_scale)(x)\n",
    "        x = add([x, input_tensor])\n",
    "        return x\n",
    "    inp = Input(shape = input_shape)\n",
    "    \n",
    "    x = Conv2D(n_feats, 3, padding='same', activity_regularizer=regularizers.l1(10e-10))(inp)\n",
    "    conv1 = x\n",
    "    if n_feats == 256:\n",
    "        res_scale = 0.1\n",
    "    else:\n",
    "        res_scale = 1.0\n",
    "    for i in range(n_resblocks): x = res_block(x, n_feats, res_scale)\n",
    "    x = Conv2D(n_feats, 3, padding='same', activity_regularizer=regularizers.l1(10e-10))(x)\n",
    "    x = add([x, conv1])\n",
    "    \n",
    "    if not scale%2:\n",
    "        for i in range(int(np.log2(scale))):\n",
    "            x = Subpixel(n_feats, 3, 2, padding='same', activity_regularizer=regularizers.l1(10e-10))(x)\n",
    "    else: # scale = 3\n",
    "        x = Subpixel(n_feats, 3, 3, padding='same', activity_regularizer=regularizers.l1(10e-10))(x)\n",
    "    sr = Conv2D(input_shape[-1], 1, padding='same', \n",
    "                activity_regularizer=regularizers.l1(10e-10))(x)\n",
    "            \n",
    "    model = Model(inputs=inp, outputs=sr, name = 'SR')\n",
    "    \"\"\"ICNR initializer for checkerboard artifact free sub pixel convolution\n",
    "        Ref:\n",
    "         [1] Andrew Aitken et al. Checkerboard artifact free sub-pixel convolution\n",
    "         https://arxiv.org/pdf/1707.02937.pdf)\n",
    "    \"\"\"\n",
    "    for layer in model.layers:\n",
    "        if type(layer) == Subpixel:\n",
    "            c, b = layer.get_weights()\n",
    "            if scale == 3:\n",
    "                w = icnr_weights(scale=3, shape=c.shape, dtype = dtype)\n",
    "            else:\n",
    "                w = icnr_weights(scale=2, shape=c.shape, dtype = dtype)\n",
    "            layer.set_weights([w, b])\n",
    "    return model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "edsr_model = EDSR(scale = scale, input_shape = (patch_size, patch_size, 3), \n",
    "                  n_feats = 256, n_resblocks = 32)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### EDSR paper's implementation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "d00af7fa23c84468b822b07ed7ca41d7",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Training', max=10), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Epoch 0', max=98), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Epoch 00001: loss improved from inf to 77.75383, saving model to sr_x2.h5\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Epoch 1', max=98), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Epoch 00002: loss improved from 77.75383 to 44.29911, saving model to sr_x2.h5\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Epoch 2', max=98), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Epoch 00003: loss improved from 44.29911 to 34.35395, saving model to sr_x2.h5\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Epoch 3', max=98), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Epoch 00004: loss improved from 34.35395 to 27.60140, saving model to sr_x2.h5\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Epoch 4', max=98), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Epoch 00005: loss improved from 27.60140 to 24.19707, saving model to sr_x2.h5\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Epoch 5', max=98), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Epoch 00006: loss improved from 24.19707 to 22.08004, saving model to sr_x2.h5\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Epoch 6', max=98), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Epoch 00007: loss improved from 22.08004 to 20.35746, saving model to sr_x2.h5\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Epoch 7', max=98), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Epoch 00008: loss improved from 20.35746 to 18.86288, saving model to sr_x2.h5\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Epoch 8', max=98), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Epoch 00009: loss improved from 18.86288 to 17.86511, saving model to sr_x2.h5\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Epoch 9', max=98), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Epoch 00010: loss improved from 17.86511 to 17.28348, saving model to sr_x2.h5\n",
      "\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<keras.callbacks.History at 0x207cea52c88>"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "edsr_model.load_weights('sr_x{}.h5'.format(scale))\n",
    "edsr_model.compile(Adam(lr=1e-3, decay=0.9), loss='mae', metrics=[ssim])\n",
    "\n",
    "checkpointer = ModelCheckpoint(filepath = 'sr_x{}.h5'.format(scale), verbose=1, \n",
    "                               save_best_only=True, save_weights_only=True,\n",
    "                               monitor = 'loss')\n",
    "\n",
    "callbacks = [TQDMNotebookCallback(), checkpointer, \n",
    "             TensorBoard()]\n",
    "\n",
    "edsr_model.fit_generator(sr_trn_gen, len(sr_trn_gen), epochs=3, verbose=0, \n",
    "                         callbacks=callbacks, workers=8, use_multiprocessing=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAlMAAAEtCAYAAAAsgeXEAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvhp/UCwAAIABJREFUeJzsvWe4ZddVJTrWyfncc27Ot3IOkko5WrIs29iWAWPjIDBuMOYBzes2/QHv0aQHbXjERwMfuMHY2LRtHDDGkoMkW8HKUkkqVanSraqbczr3npz2+zF37bFLa1+ppAuSJdb4vvpq3XnOTmutvfY+c4w5p7IsCwYGBgYGBgYGBq8Mvtf6BAwMDAwMDAwMXs8wL1MGBgYGBgYGBhuAeZkyMDAwMDAwMNgAzMuUgYGBgYGBgcEGYF6mDAwMDAwMDAw2APMyZWBgYGBgYGCwAZiXKQMDAwMDAwODDcC8TBkAAJRSI0qpN79Kx7pJKdVUSuWVUmtKqZNKqZ96NY5tYGDw+oZS6jql1MNKqZxSakkp9ZBS6vLX+rwAZx0t2WvbjFLq00qpxGt9Xgb//jAvUwavFaYsy0oASAH4LwD+l1Jqx2t8TgYGBj/AUEqlAHwDwP8EkAXQC+C3AVRe5fNQSqn1np/vtNe2gwAuAfBrr96ZGbxWMC9TBi8JpdTPKKWG7V+BX1dK9dj231ZK/U+7HVRKFZRS/6/9d1QpVVZKZV5s35bgLgBLAPa7jrlTKXW3fcyTSqn3uj57u1LqedurNamU+uV/j+s2MDD4gcN2ALAs6/OWZTUsyypZlvUdy7KOAIBS6reUUp87/2Wl1JBSylJKBey/71NKfUIp9bjt2foXpVTW9f2rbK/XilLqWaXUTa7P7lNK/Z5S6iEARQCbX+xELcuaAfBtyEvV+X2ElVJ/pJQaU0rNKqX+WikVtT9rU0p9wz72klLqwRd5YTP4AYMZKIMXhVLqZgCfAPBeAN0ARgF8wf74fgA32e3LAcwAuNH++2oAJy3LWn6J/fuUUu8C0AZg2LbFAdwN4H8D6ADwfgB/pZTaY2/2dwB+1rKsJIC9AL67sas0MDB4neAUgIZS6jNKqbe91I+1dfATAD4CoAdAHcCfA4BSqhfAnQB+F+L1+mUAX1FKtbu2vQPARwEkIWvhulBK9QF4G+x1zcYfQF4IDwLYCvGs/Yb92ccBTABoB9AJ4P8CYOq9vU5gXqYMXgofBPApy7IOW5ZVgbisr1ZKDQF4BMA2pVQrgBsgLzm9tkbgRsjL1nroUUqtACgB+GcA/9WyrKftz94BYMSyrL+3LKtuWdZhAF8B8B778xqA3UqplGVZy/bnBgYGb3BYlrUK4DrIS8b/AjBve8s7X8ZuPmtZ1lHLsgoA/juA9yql/AA+BOAuy7LusiyraVnW3QCeBPB217aftizrmL0u1dbZ/9eUUmsAxgHMAfhNQKhBAD8D4L9YlrVkWdYagP8B4Mft7WqQH6yDlmXVLMt60DLFc183MC9TBi+FHrh+gVmWlQewCKDXsqwSZLG5EfIydT+AhwFci5d+mZqyLKsFopn6cwA3uz4bBHCl7e5esV+6Pgigy/78RyEL3KhS6n6l1NUbv0wDA4PXAyzLOm5Z1octy+qDeKZ7APzZy9jFuKs9CiAI8YwPAvixF6w710FecLy2XQ/vtr3mNwHYae8bEI9TDMBTrv1/y7YDwB9CvFjfUUqdVUr96su4JoPXGOZlyuClMAVZZAA4FFwrgEnbdD/kRegSAE/Yf98G4AoAD7zUzm1v168A2KeUerdtHgdwv2VZLa5/Ccuyfs7e5gnLsm6HUIBfA/BPG79MAwOD1xssyzoB4NOQlyoAKEBeWM6j64XbAOh3tQcgHqEFyLrz2ResO3HLsn7ffciXcW732+f2R7ZpAeKJ3+Paf9oWq8OyrDXLsj5uWdZmAO8E8F+VUrdc7PEMXluYlykDN4JKqYjrXwCiW/oppdRBpVQY4pZ+zLKsEXub+yEahOcty6oCuA/ATwM4Z1nW/MUc1N7uj0HtwDcAbFdK3WEL24NKqcuVUruUUiGl1AeVUmnbzb4KoPFvc/kGBgY/yLADUz5u65GglOqHaCoftb/yDIAblFIDSqk0vCPpPqSU2q2UigH4HQBftiyrAeBzAN6plLpNKeW318Cbzh/rFeLPANyqlDpoWVYTQk3+qVKqwz7/XqXUbXb7HUqprTYdeH5dM2vb6wTmZcrAjbsgv5zO//sty7LuhegKvgJgGsAWkOMHhNaLgl6o5wGUcRFeqRfgUwAGlFLvtLUEb7GPMwURtv8BgLD93TsAjCilVgF8DKJ1MDAweONjDcCVAB5TShUgL1FHIeJt2DqnLwI4AuApyA+zF+KzEI/RDIAIgP9sbzsO4HaI8Hse4qn6b9jAc9L+QfkPkDUUEC/8MIBH7fXrHgDnU8Jss//OQ/Sof2VZ1n2v9NgGry6U0bcZGBgYGPxHgFLqPgCfsyzrb1/rczF4Y8F4pgwMDAwMDAwMNgDzMmVgYGBgYGBgsAEYms/AwMDAwMDAYAMwnikDA4P/sFBKvdUuVzRs8voYGBi8UmzoZcosRAYGBq9X2Fmv/xJS8mM3gPcrpXa/tmdlYGDwekTglW7oWohuhdQTekIp9XXLsp5fb5tENGJlk4kLbOV60/O7DUu3e9V8VD7d5vf7PfcZDgZ1WySq2YLBkOf2UEozhUJh/Ws+/XvrwWrqNGu1VtVstYp3UfRKVf9uoVjUbM2mdz83vY7vsc9mU093YnmM0ctBOOQ9ThZ0e7FU1mzqZZStUh5j5/OYO4GA9y3hNaf8Ht8tl/Vxqte8q05YHuevoJ9nIODdTyGP47e3t2q2RNK7fNm5c+c0mz+g94kP+jh7HRsAfL4Lz3U5l0OhVLr4G+LVxRUAhi3LOgsASqkvQELjPdewiF9ZSXsJsapxAEAhxbH1oe60wzX5YlBxfahb/Lxu918E/LwG2VfAcvWta80LN8Ve9XPeJMD1tB6UeyRm0VYN8b6JNhL29rSFmymn3QzJuuGvM+eliojNqtDmi+addqMk+/S7bQXXOhuWY1XKvI5AmOtLOW9fS5T9WF/j5rWQ9FmjzHnVCLjWp4rMzaarTxo1e5sQbb4ip2A1LMcK1djPNYvrm9/2MdR9HC/V4PGr9hyPulJAuZfCgE/+qPhd41jnutCEPF8suK6jIecXCHBHtRr37wtIn7plOQHFc7LsOeNT3L7hWvP89rri/tzf5Od1+7ju51E9IOesKq7rcK2ZQfurVdfzTinXWleXuR0OsB+rTc7NuL3WVOqu+RzjNaUa0k9+i/tfSrDtq5QAALWQ69m0Judajpe4nxKf8SV7TGsNnmcG/LxQl/laj3AS5lexYFmWuz6jJ17xyxRe5kIEANlkAh9/7+0X2Ibn9Qc/ACzWSpotHIxoNn9UfxnKtng/PDb36P2xedtezdbTO+S5vfLrL2O9Q/p3A1H9PNeTpjUq+oN2cnxCs42PnvHc/tyYXt3giSf1UnWFQsFz+4rHS9rYhL7PfH5Vs1Urec0GAPB40PpcD5Lz2NyrP/gBoO5LarannzuuH8anv+Ct99IY9HiRjsfjmq2trU2zAUA6ndZsLVl9Pp04eVazzc5Oee6zael9H1L6i3ympcVz+/6urGb7+f/0Yc121c0/7Ln9HT/5E5otm9H7JKr0e7S31Xvs4rEL++kvPvdZz+/9gKAXF5YHmYDkMHKglPoopLAt4gHgdjt9Y33iIADgiRs5trHGgtPeNN0BAOgLbnNsszXmsF1ul3m23dri2Cb9MwCAtirnoC/AtWRzSebbWIbz5jrreqe90HkKAHCocY1jG+k56bT35aTq0nj6BPdZeavTLg4+BQBombvMsQV2ia1xhrbwnu877dUj1wEAMvsfcmwrj/c4bWyTY505xpex1u1jTvvEg7IuNPdPO7bF7/K+nhuUPl05xpe+tQ5ubw1LX5Sy7JOVcamtvjZIW+hZrklTQzIOPVO8r+ZrXMsSfjnX5ciiY/Ot8fjjUWnvBNdEq8x1JxOVZ9eZFMfRP8f1uwiZRBXFqedblfNryfIFa26eNeKjWdlXrc7nRXuQ92ClLM+8ZJDXvBThi0eL/QIX9fE5kCpyTVzpkO1qea7Ts21yzpER1/oT58tQZ032PxLmmhWOTDrt5rzM7a2tvC8mipybl2fl2X1uheN9zSVcP25ZGgIAtLp+XHz+Wr5shYePAQBmepccm/q+JLw/cekRx3br0T1O+1hMvjuxOuvYfsy/02k/MSc1qed23OPYvv/tFy9ofR4bofm8FqLeDezPwMDA4NWEl8fsgp89lmV90rKsQ5ZlHYp6OwgNDAwMNuSZesmFCLjwl10mof/qNTAwMHiNMIEL67T1QTLue6IW92HmKlnDKofFK/h8O73AdwT4q32kT36VL57g783Nm2512jPRuwAAoTF6lvwR8URE2/lLfzXmouE6xHtiBfc7tlCcp7sSEu/E1EDOsSUWeXmj28Semma5uoWd/FXfMiteiTGXI3PnGdnnM9fQY3HZGD0V535EPBrRPD1Ph9/Kc96xKr+vj+4ntbjDRRU9uFc8b3tBj8l92+lx2VsTL9RYB70X3b45pz3RLtvHLHo/Jm3nRmeJ+xzu5Ztwp83g5OMuL06A7aAl3qFEgR7yyQwfdz1+8aI1y9x/qYPXPJ+Wc4ktcRxPDdBbd2hePIjPNbY6tkyreESa+R2OrdnFx6k/L+dUUy4Pe5B9UsuIt60S4TFDLuoyXxPKLeQiCI73cf99SzImp0K8jv6yzPVp33bHFmnQqzmZ2AcAWA4+69h2+Elbn+qWeRbZw89jE1fx8wOfBgDsLP2eY6tu/brTPru0CQDwiWs5x+94hGP/yQO7AADX5D7p2Kb7LgEAJGrfcWzR4PucdiL6JAAgPs05/GTHM0772KUrAAC1So8x4M0KvRAb8Uxd1ELk/mWX8KC/DAwMDF4jPAFgm1Jqk1IqBClf9PWX2MbAwMBAw0Y8U85CBGASshB94MU2SCTiuPGayy+wHfn0P3l+15fSNSpra2uaLeRRBrJe9Rb8BqDrdpJZvah4e0ePZgOAgIfwKZ9b0Wwt4Q7N5iWUBwCfXx+Clqyu+cqteutmUivLmq3FQ2OztLSk2QBvjVEmox+/WtUF4NXaeu/iut1q6rbxqVnNBgB3fFjX+Dxz9IRmazS8ghS8NVP1uj72XrZiUdfqAUDEI1Ahn9c1YzmP+bBeLrcXirUBIOAxH9YTxXtpvry0XV7nCQAHDx7UbNNTY5qtuKbPsTWPgAAA8Acu/MHU/AHOY2dZVl0p9QsAvg3AD+BTlmUdW+/7EVjY4RMPRmtWdBZNl1h71k9N1KXz0o6sMDgwpzi3rvGJVik9wPGKrYl+864AtTpvs/ir+OG4eKTeOspf+tNXvt9p7zouusLYZW/iOS8/6rTD3YMAgMUl3ksDGeruqqviKdnhkg2WlsU7cOUAPTMnLKo59mTEE/HULMf5ugg9S48uiNzk3e28/+9Z6Xba746IPvTx4qBj+2GXPvOBhmy3ucrjH3Fp+LZV5GSPJbnmD5VkDk500ouTGZ1x2pWIfc0uyWLVJcYuQbRpvSWOzXL2Qac9HhUvUSjHbRLT1E++Kf5eAMCd259wbHvOPOK0Z6syZxKgvqhii9ZTLiX7phL7bCwpz5SeAj+fTdMT2rUo/V90XbN/jfdoa1I8To0yPXiXVzmOx+LibevK0x8yX5bzTFYedmzda9QXncb9AIAbstc5tu8leM03PmPPjWu5zgw+eafTHn2LvAukJ77i2K5pcv0IlKTP/qhEndWnrqLn7tbjouc7XbuF22yVtbLnBPspH2U/hnpGAACrijKom9Y+7LSDs18FADzf634vuDjP1Ct+mXq5C5GBgYHBDxosy7oLUuDbwMDA4BVjI54psxAZGBgYGBgY/IfHhl6mDAwMDP6joGEFkK8K3RLqEOrz/sQXnc9/YYWC5VzyCgDAbPxpx7Z58wGn/cSkpCzoerbPscUtobd6Yi4KNXej0xwclKwz83OXOLaOzm877ZXjQq+V2v/ZsXU/zn2tKWlvy1MbUY6QlklVxZ4LkY5M2jmdAiDtcUjxsWGFhaq6wWIKiCBcdGdyAACweo6U/qWunFKnhoVq2po47dieXyVV1d8jVNbUU6TUureTh5ybESoq20YJSHFJxqGnjWLlpUVSdrV2oeIray6NiIteq8Tk+qZamKJmpbbLaQ/VROQdrrPvG9t4fd/vk/xtAxOkx8fVgNPuicu5HF/k3FgLCnVZjA07tmiV0oLGgpzTcox9mzzDcy4n5frzWVfqhCblKjPJowCAvib79vstTMNwyZTQa2cLpH0H4nKssxGKsWsNBjeshmSOPzjGc+7YQxn1cKtQaZcwOwcGmu9w2kPqLwEAuyLvcWynf5Ri8JsflZQGD93C43/ka6Stp8IyJrE2jkP4GemzIy5x/4kw8+ktn30nACAZP+rY8u72kIxN+RFd2vBSMOVkDAwMDAwMDAw2APMyZWBgYGBgYGCwAbyqNJ8PQOQF2akObB/0/O59w5OazWrqp1sq6dFXFvTIKwBYLugRSFNTelqZ9qx3due2Vj3yb3VpUbOlMno0nQrrZWcAoOmRrcvvEdGVSqX0LwJIpfRjeUXjRT0yxQPekWb5op4tPeBRZmS9KLN6XY+m9Czx40/oNgCPP/64Zmv36NOVXE6z1Rve0Xhe5WSqHlGfXvMJAMplfe4Ew3qfemWU9zo24N1/XtF8iZh3fraQVySoRyTneuWV+vr6NdvTzzyh2dqS+nFW1rwrFyRjL8he/wMczfdyEfTX0ZMW6mPbKYm8e/8lv+F8fraHtMmO5yR6diruyr78XKfTLuyWMe13lZM5GZe+ezLEiNBSkhxJqCrUxWwvs+xXhkkfFeLS17eeIi3y/A7SQjuqcg8/sIfz6cAa58tCu8yTrhXO15k2iearg+tcLkGaLRSWc8618V6KucrV5JflHikOcB4UJkk/FXbIOm9NMpeRbysjStcm5flQdE3VtllSdjNb7HI2rjWreECOWamR0lro4foZqsn6WAxzm2qU0c7Bquy/1uC93JN1lQIqSWRgM8ys8Aslfr6pLPTgQh+jEruGOc5nUkIvdq4w6ju6Ks+W2Tipuw4/zy+TlmjBZoV9O9vLdT5bkTGLTnJN8hf5eadf1rW1Kp9tV63wu0/F5HN/jZGex2pCQasIabCqYo6zZkPW6dgmzvG5MqPkhqzNAACrm1VGFtOM3n8kIvv/QMdNju3AUxzbswNCi7+zyrXmzuvYZ+oRuU+mRhhNuKAkSnag+1uO7WRys9PeF/me2CZJy9ZypM1rI0IZDrczghB4FhcD45kyMDAwMDAwMNgAjADdwMDA4CLg80URjciv8NHrxdPyXODvnM9/Kv3jTvvz2+XXfP/T9Ar2baWn4UxdPE7laXpCp/eLp/XNT9HDlUgzT1XWkl/bam6fY9vWSs+V1XwzAGA5yADrrSPvdNrRpHgdbltkrr/yIDO0D1XFC1Uq08uSsGu5+Qr0RvVk6TmrL4hnK5Amk3BqbMhp7+sVz9hjT9Kj0LaZdfjOfFv6MdVPj8b8GQrHEwnxCpRO07NU72adzqWj4pkLU8ePtWfFcxfJ0vPSNsdzXknIY6+3SO+EP8dcbOGIHH+lRM95c4neuo4F8SKtxk45tt4kj1UcEU9Ieys9Q6slCquHUnLOs1Hmb8qdT1e2xgzmlTjnwZk1O/9SgkxKYpwiaV9A+ncuRtF8W4o50Ipl2VcwwOs47Mp7tq8i3synChybtoRsM1OlRzXoo5cqHLU9T2eYEantGno9F85JPyeKdzu2LU/+ldN+zzX/t+ynwTm4vI/exC5rCADwrQq9mh+cpsfoG0sSaLGtxPOrF/8GAPBEip7G0r3ss/rbZG6vzbvqX7oKXg9lfkb2OcP56NLPvyiMZ8rAwMDAwMDAYAMwL1MGBgYGBgYGBhvAq0rzKQDBF5QvufrAfs/vPnBGF4ZbHkLeYlkXwlrrvCP6w1XNNj8/r9lWFrzLnLS26MLuWlXvwopHSZJIwFsEHPIFNZvyqGEYCIQ0GwAkE3r5kFgsptm8ysYAQLWq94mXMDoa1ve5lFvVbACgPAToVaXbmg1vYfW5ET19/6233KzZPvf5L2g2n99b7O0Frz7xKjEDeJdkCYT0cWo09PpG61QSAvwXL+r3Qmtru2bLZrOareJxToC3WN5LrO5T+jkVPcoLAXo/NdaZd69H1FDGlE+ospseFXHsLx38XefzRyLMRfShEaGiJoMsf3H4DHPf+CMixK10cwzXzgmlNpGm2noBFJMfqgilNpngHA260uFYJck1dPkwcz7lVpl/aUtBaJlpV6Hg3css7TIWFPvgHMt/VMMyzg13laQU58ic3exZZGkSV2UVpGdkfVvcxHVm2wneN6H9ImxPP8djRtpYPDo0I4LimZYHHNv+pSt5LLtAsFqmED84JPPav0Thcb7vfqetbMpy1OL8z4dIJbWF5B6asUg/XVvi+RWTQiGFAhRzn4nyHukIPQcAqCyxI/aEf9JpP9ciuZKyY1c7toWIlE5JWRTiFyYYYDPQLWvFBNh3VhepxfGiDMSeBgNA5l1Ldvu8zJlzPaQBr1i+1Gl/p+9rAICt0y5xf0zmTqxEyqzM6YL6iNCplV6u45X7GFxQb5ftF0IfdmzNg8857dzyTwAAVIblYPY8w3xjU5vlfrh8lffQv2b/3mlPT9wBAChGf9GxHckKDa+6Wf5n13WcG2srQkcvTvC+GO3n2J0pfxYAMNhFGtDQfAYGBgYGBgYGrwLMy5SBgYGBgYGBwQZgovkMDAwMLgKRYCt2dkrpi5mmRDB99SBLgvwf0+RVvnSJ0B3tZxn9tC17xGlvnhA69OgY+bOhGyXK7rl/us+x/dh2UhSNI5LXp23wVsfm9z/mtNNbhDbK1Z9ybNEkaZcTTaHPB/ws3zHTzkis9pxQWYU2SifqNbmmwCDp2sYUqaSe86n3lihX2ONKiZfOCTXZHaVsw7fDRUUvSU6pxhAprdw8ZRKpptBSrjRWKMbZj9V5oaVU24hjWx2R8w92MeovcpKUWDMuFE8lSY50sMKxU/kFAEA0RupvuMp+bmtIfqlyK2nAwRle9EBC2nP9lBw8Xb3XafcvS+6sZ7spJzm0LGP3RIkRgv4W9sPxsOzL5yM1GTs75LR7OuWaV0Kk3zPLjB492yv2Dmsnz6mV+7o+J/TisMvWUZUxz7lkJ90T5PmWu4XyXh0jNbltpytH4ZrQ2lecI1Hm72PE68GddwIAQmd4HffuJc13AHI//Plmyjl+/aGfctqPXCI5oZ4b4zVtGRfaOHyWUX9nG5zj2cT1AIDUDpal8YU5zjcUhfr8wiLH4WJhPFMGBgYGBgYGBhvAq+uZUj5YkQszJAfhLawe7N2s2YbPTWg2y0Ps3KjromoAqHlk9i6FdRHx1Jy3AL2tvVOzhSr68YsrujA7FPHO9q08EpMrny4CDsWT+hcBBMN6FvCOtg7Nlkx6Z1AvemT8LlV0cXHEIwt3a4t3pvhlzxqRej/Fgt7v8mt5/ZwqHucZ9ND019bRn/ubeiZun0dW91rDO2N3paaff35Vv9CgTxew+3zeJ+Wr6scKxfVbMhT0FpBvGhzQbM2gPs5rayuaDQDOTY5rtrLHPQKPsa+W9f4AgLUXCNPXC3x4PUI1CvDnxeuT9H8AAPCbYf4q/3aU4/ljq+KR+VYv1yL/ExST3wPJMfT2gxTs//OyeBduzbIg8n1ZriWb++QX+LObv+fYfjz3Nqc90RQxdjbH+3IqzTG+akY8Qmd76GW5dGrEaT+fkjHducz5OpmSfR08ymubaKNAfnBOvDjTAQp2A9Qr46y9q+4S58XEEhe9q5flHjxdoPD5phLn8LBf+mx3nvfVcJretk018TLNNLkYHITYTlZ4IsluelSmlfRT5yqDn4bjLIbbYnuWSkU+G7YFuf7O1sTTMjBP2+lNzEu0WLKLQ48xM/elSWYJfzolHpntk5wbx+sy9t1xrinn6sxOPzQqa8D0bj4v97TQg3gmKMfvWOGacG4TM3f3z8m5Dvc+5Nh2T1KAfnqzeA77nmLfr/RLn6cWOUfzfSNOu2rHCVV7XQWbx+h5CnSJR+ns/t9ybPvTf+m0M7kfBQD4DtHzdINL4b64S8Tmtz3GvGh/FWXFgdaHbwIAzE8879hG4zI3pzo43y6r8PxGYjJf06MUoKdaGNAwd05E9QcH6dn6LvSAKC8Yz5SBgYGBgYGBwQZgXqYMDAwMDAwMDDYAI0A3MDAwuAj4Yu2I7v9pAEAqLwWhJ5ubnM9/OsTCt/8jIXTKdXOUK9xzFUXYH/q+iMmfff6XHFvnOyXXz/DhNzm27WUW/e5/VATi6RFSp/lO5u0JLgsVNNXPEjSxCUoj7r9UaJuOp2cc27OX8/yzp4QaGdtDSioxKjKCx3eQ2jvINFU42Sv03o4F2p5y6csvs5mobybYD7e4UgN+MSznujtKWuazLawNc6AgFPV3+0m5bZqiDOPpqBx/a57C5Udtem6wk9Th5DiF1fGwyDVKIHXY46qkrKoSNLCpRHp7Mc4LjJaEppxNsnjxwCn2eRDS5752dtSJKqUXA5Djj/hZsLq3di0A4FTiK44tM0macqlHBP7ZUZ7Td12Ufl9QOn0lShqwz1X02MoIzTewyFJE0xnKB/pHJJfTTIxibd/SuwAA6eTXHFt59CanvS0hORqPWBxQy8853sjdBgCIDlPsvdZG+qzlMrm+2nFex9EBnnNtRAItJjLDju0t//oxpz1myTx4FqRb23KSX6pWIF06rpgjrK3vEAAgGaQoPhfm/mMdki9tflaXz7wUjGfKwMDAwMDAwGADeFU9U/5oAql911xgy1h6BnAA+Niu6zTbL//qxzWbz6eLY+vwFhH7PbJLl6u6sHl8Uhe6A0A606LZhjZt02yLi3pW9aRHtmoA8If16/fSK6fi3gL2nEe280xGz9S+adMmzQYAx0+c0GyhkB4UUPMQYAeD3mPnBS8hslffA0DA4/iPPvqoZrviyqs02yNPPK7ZACDg18feK9u5WidbuFXTv1sp6GLtUECeNZRIAAAgAElEQVTvk+Y689HnkRrdKwN6PO6dKb6nt0uzxZP6d8em5zQbAFQ8ROTFsj4m0bge0FCueGeKr76gn5qW97W/HhEoz6L95J8DAIKtIqS9ZSvn4H1qxGn/yczvAAA+feAzjq3jLno/JpryCzqecnkSzop4OOIK1pgLse9PvUPGpq7o+tmcYtb1YFm8VCshBhy0BblmRcoyFuEwvSj+CgsUJ6LitchVmXqg2ydh/OFx7rOR4D4HZ8UrkY9wHdrqSqy/Yg//fte0Go3zHrvEzsA+76OI+IYm53U+Khtetkhv2UyanojMktwvKy6hfX9ctq+5giE6IjypUko8GoUq79ViG0PhV8/JORU62Dc9K7zmiaR4X7YWhnhRSe4rZ2eIn85TlN7rioxZDouof190r2M73CbPjL45Zvs+20/hczAn+1yN837aGqKwum7JWqL8vM5zrkLG4bJ46ZKKxZWnzzBrfW+HCLInmwx42BkQAfh4jWtKpknP2dz0TQCAXfEnHVuq7f9x2rtDnwAAPL37I47tJ6qcRw9XxJP7m1tvc2zPt/B5VA5IUED+HNfEb3RwzU2cluLRySWe35G9MrfCJzj2kVZWHggvy/2LBoXye2r0AD61IKk6dvW/xbHR9/viMJ4pAwMDAwMDA4MNwLxMGRgYGBgYGBhsAEaAbmBgYHAR8Kf7kfqhPwQADJyV3DsjddJ0t13CnFOfeF7ExT+5Skrtd6/+a6c9cFgKd98b+rpjW+wQqqizj5RQPMncdv3PCp0R6maOnOzI95122iffXU1R2FydoZg9tVnOeWaWRX9bdlHs7V+1c/DEKQKeGhYaL3QbczuFh0m1hHfLOSUWHRPWmHIKMVujvZgmddk3TUrssZQIfTvnSV2eCJDeykJoq6k6RfPRPA+WiAn9Fa1y/zlLqKRggnTi4iqF0XW7QHmsQpGyvzHIk24K5dgzz76vNkecdqQhNN1EgHKQLVXmR4qvyWN1d5XSDFUg1TTdKrTYtIuS21EQ+cv3g6RYt85RQN8o1+3z4DVNZLj/nqrQg+cWSO3tB0X/qz6hMRWYj+sKH6Un9WmZx/tylIjMtMjnXQ0K/hcjFItv75N5eLpC+not80WnXfaL2HvbDOfLQx28pg/Y2fWfGubYjF5JOUh0WfKqXZFi4eszoyxgPJsRCnwm+M+OTc2IfyjTyfmaWyJ1aLX9vHzez4CF70YpQL80J7KdrwRIXV4sjGfKwMDAwMDAwGADMC9TBgYGBgYGBgYbwKtL8/kCQOzCqLbiOuU/WhNtmm3ztr2a7dgpXWtfKBQ1GwAov/7uaEGPhmt6lAQBgInxMc3W09On2Qq+vG4re5TpAOCPhjVbwKucjEfUHwC0tOgRhjNT05rNK8IPAMJh/fi+gt5PSukD5WUDvKP8GpWKZrPWifRqeETUVSw9GrCnq1uzrbfPalUvMeQP6tN/nUuCL6Rfk9XUo+Esr98n60W0eZW48Yjw8xpjANiyZYtm8+q7hYUlzQYAJY8SPYuLi5qtJX2JZsvn9PEEgJWlC0vXeJ3P6xWB1SV03i2lL05e+isAgJtdVa+OKq4PvxaQD47uZWTv9i+9y2n/i+8eAMD18x91bGNrfwcAWLJGHNvoFOf91pDQHmctHmek5ZDTLielfMjBVUYiFaP3OO3VaYme2pfm2lqZ5/gE0jL5V5eYa6gvfiMAoHqGtAjCjDirj0rEVdNitHA0yEiqwoJQanHX/TPrWp6HQkL7TFd4L+wIkxKbtIRy3OEjPXQsQkowFRL6b6SFlFyLHbW43ODNnHTRU8WmrHmrWV77gmvsrGWJciukWGImA1JZTUvs/XkW2C3EeM3NpNBfY3VGyw0mWdplYFrup47tnDzjSuix/izpqdFzjHBsbZVn30KTUYdtE3x2lbrlvht0lbUZP8acUov7ZK3Y9yxpxMez7JMr5oQKO9zHqMlDMxIV+WQ7nyeDy5xvZ+dlTGMR0sIdXTc67VRC5kxlkOvs3sr7nfbdPQ8AAG73fYj7r/H8Di9JIeOFCZb9SYVcUe2zsqZHlxn9OZuRHGHdz/M9o/NyUrjL+As5zzmWqNkSZ/mlZ/JSqmlP6Occ2wy8I8RfCOOZMjAwMDAwMDDYAIwA3cDAwOAiUGvJYPL29wAA9p2RX8BLR5jnade7WGT2gcfEE33dCIuo/nHyt532u4OSn+qfrqUovX1JhLCn7qfHNX0zl+jZO8XLFUvyl3j/UXq8g503AADqvSxkHGu8x2nH+yUX08Tz/HWfGqR3NXxcPFbZW+gFWpgUj8/5bNUAUGTyaNSulmvOuDKgNzLMVp60m7Ougu4pV9qy0/bhO3so5H9gmt7R7pR45u6vUoDe0WR26hO2yLzN5XmeLNui9CzzRC2OMxdRpFUOWpqld7U15fLW+cXLFCjQc7RYp4i5LSSejNkQL7qjyO39EI/VQJ1j027x80avZM9/IM/t20PixRqdpQfwmsbPOO3hua8CAKoJzrGpDgqn+1alaPB88QHHht1kHbaPyzFHEwyS2BHjNc2Fb5fvVSm8rm+Xftx+hnnqClsZvDA4Jv0z28t+tur0YtULMnaDa8zFOJ56ymlfb2egL0XJ5Dwe4eSK5MVjlbK9TQAwUvi80/Y15ZxLe+lBTN4nc3u4n/m0UGHOyq0+yS9V62EQx/ICWZsDWfH2fbF8L14ujGfKwMDAwMDAwGADMC9TBgYGBgYGBgYbwKtK8ykFBF+grfYp71OoeYh73/O+92q2p//7M5ot6PcWa1c9SmDklS5Wj0S8Bb8vLJUBAHPzeumY3u6oZqvkdVE6ADTT+rGUXxegKw9hMgAkEnqZmWw2q9lm5mY123rfrdV1sbaXsLvgUU4F8BYde5VJaXj0JwAoj0sNerz3T3uU/cm2pDUbAFTLumBal7QDDWsdBbrHt72E9vWy3nde1w4A3R2dmm3brp2abe/2HZoNAEJBPXhibVWfz6urK5oNABoe4+xVSqhQ0sfTF4poNgAIxy4sZ+MlqH+9IphfQs8jQjNUd/4jAKDXFdfx0HHSLjdcKmO74BqO967e4bT/dFIolh+ZIJVzck5yWH2zi/f0juOcj129cr9NN0ml3LuL8zLSKiLnfaF3OLb5JovUZnMixN3azfxKmGRwgb9N6JT8AkXM2cRxud4p0nyJGNeC4AlZ19bCpI8q4PaRJTlm0rXMFV3VjXrs/sstcM076OeaNFIWOmY/SH1O+Li+brIL646HOe96kzKHz9XZ+V2u4081ZPtGB9eEYIPreEnJfTHT4yrnMkvKb6xD7HtnKVyeiJGG3B6UOe/6GEU/xywwLyLvA52ka3N5Wcu6+nlth8FnW/mQfa+Pc45tdy11gVMSTBI4RJpw5OS7nXZnl1Cn+RkGcfUHWY5mukvm2Z4UhfQjKzKP0vvYN/40xeLRmsyX/m3XO7bdLRynoF3KqO8tHNstR3hOiIrYPOAqa3Wdj8XCH2rI3K0p3mRbl2522k/GZf/JUd5D4d4/AgB0+P/AsXUFSU3O9Ah9d+mZmxzb82muz/Up6b9ruz/o2L4OiuJfDG+clc7AwMDAwMDA4DWAeZkyMDAwMDAwMNgATDSfgYGBwUWgkerA2pt/EQCQPPIZAMAiftL5/IqrmR9pQdLVIM5qLvjKOeYI+oVxoS4+u5lRQ9vDkhun4+knHNvaDsoDFg+/DQDgH2Ik0tZ7SW+1XCphct3LLDETiDLXUKxPoseCkzyp5iDpu7lpoXg6NjHcrrCaBAD0RZhPrzxHKri5X64jPM7rDLUxr0/CZoNXXGqEVhflNmKzmEMBRp7dlyP9PFSXXEjfrK/yOuukGZ+NCb3VVuP2wzartAmkoscKpKqTnUJ5lSdI0/sTzLkUCghl17VKyr0cYImetgnp88Uoqb1AlXmNJirSP+ka6alMkv08WpVoxKUy6a96Vq55PsR8V9sTP+S058p/AwCo1kmDNVo4N/x9cn0jrrSB+wbuc9q5tPRf9tAjji0+S5nADpvizZwlrZzpk/lamiANt6mDOcjmq3L87Gb2Q8sSr2l1v9CIwbUhx9a9mccfHZBx3FPiMR9r5dzeeUiuf22aeZ6W9nHss4/LOKZ3Mfpz7JsyDjfGPuvYHm3wOt9RkMi+5TSlD0nc6rQvaRVK889WON8uFsYzZWBgYGBgYGCwAbyqnikLupDZZ+lCcwBo6hpsbNumZ3zu79UzkI+MndVsABDw2Kdl6SLiUsU7Y/X8mp4xOr2wrNkSsaRmW16c02wAkGrTM70rpR/fH9SFwQAQ8BBBpzO6qD2VSmk2AMh4ZVCf1TOol8tlzRaL6QJoAGhr1YXVi4v6Pq2AlwQc8HlkOy819HkyMq6P8zVXHtJsAPDU4cc020c+8vOa7bHHn9ZsADByZliz7dq7TbO99aa3abah/iHPffqC+m8ZL8H2epnma03dvuAxz7Zu2aTZAOA/b/+oZmvUPqLZ8kU90OA73/q25z5HXd4XALCwnqD/9YdgcQFdhz8FAPBlPgkAiLCOMWaZsBp9th47TycTPpZ9q9P+Ztt9AIC3+ijcPrwsIuRw7kccW3r1PqethuTX9Nki7+XM9iGnnVRyjz65xeU5inzHaW/O3QYAaFHfdGwtM7c47VRQxqo0w3WuFpDAldlV5h9qrzMgwn9K1rqVJjN3N2Zc93pOhNs1V26pGtMSIW4v3zOu+JitcbqxFkriddiX5Do5meec2h2R754JMMP4YFCE5SfC9Jgk0uyT+bp4rFIpzuvFGD0i9aqtHLe4/coEnz1znSIMH2pQlD7XpOcpXZQ12QozwGamyedEPSMC+8XiVx3bdYviHYnve7tjQ4o5o1IZe01M8P5+s6Lw+swV8nDbVj/u2KbCH3bah1b/HgCQP/lTjm2Lax7sankfAKBrD715y35Zx6N76eWpJf7Cabfs/gUAQKHtdxxbR+w3nHaiLPtP9XHwl/N8Zgw8J/0408nBj53jmCQD4gXLWPSu9i4zz9aWrGSlH5+iW/TAtXL+M3NXO7Z3tdBzNeOTdevQA8z0PrzlX512/X6Z+295C72ST/MWelEYz5SBgYGBgYGBwQZgXqYMDAwMDAwMDDYAI0A3MDAwuAj4o/1I7/kTAEDjMRGOL2eYY6evj1RSPid0xcLl/L06/ARzCP1Im1AXnznykGMLdQlFnb/h1x3b7jFqExZHhNJq+SEu25s/zbxC+ctEPLvrEeYvanSQ9k7uElolWmYeqq4ghbgLC3Kuwa2kYiK2LVKknGJ1jQmUYvuEfmvMkx4Kd5H+itgqBJ8rH5dypSI7z9i1U7uPY4uu3FqDcs3PjpH63xQn5X84L4LtthWWmDlu+wh66sxXtehKaZdIC+VWqNMYcxXQDa7IsayUq5BwmDmretbkolZKpP7TUQrQS7bAPJhz6Ur62AGrs1I65kA7z+9UTPaV2UrZxZbJ9znt5djvAwDKR3/csS3dTgH61iPSqXO3kCK+fZTjkLxc5tuJMunIqzaT6nouIJReuytPXIsl9FygRhot5hL115RQYi1F5pYKZCmgb6nKtax1kWbbWSRFfHyHiLz7TpBSW97GMjBBO/gg3k0asNnDMVm0q8zs2sJ+PndUgh/e66dc464677t2S3j5o/tI7eXrlDskr5D7+6lxveD7S8F4pgwMDAwMDAwMNgDzMmVgYGBgYGBgsAG8utF8loVq9cISFqrhHdFV84hoq9T1iK63vU2Pnvr7z/yd9z6rekRapaKXGQkG9ag9AFjL6WU1pif0KLvOrB6ht5rLaTYAKOb1c4p6RAN6dAcAIPDC+jwAUkk9Qq+1tVWzAUAkppcFiUR0m1dJlFrNOxLT65yCHmVKmlXvcjI1jzIzQY991hr699YK3mV7OjJ6hGEmpEeafeR9P+a5fcAf1myJjF7KJ53Q+77e9J7jXqVbvCL31ovmC4f1c/JZF//7yCtysOFRxgkBPWJ0dOuY/j0AE2PnLvhbvYGi+VRpDP7nJQI01yNRTQ1XRNjkKsezNy90ycpx0iqHIqQbnq5In954M8trfOspCQfcc5S2Uv9TTnu6W+i78gRpF6vzNqftzwmN9+AhzstYlDTg1oZE7uVxl2NbXGV5kS1RmU8ti0OOrRmW6ys0uZ+4aymozkvknlXp5zmNck1ds2RfjQDXOf+sixK02cM8K7egj6wRFiZlPl6RYN+eWGA0466wfHk8zDm6RQl1eTq+4NiSrlDuqZJ8Nx3ihSy4IvcqHdJ/9QAp0OAqI9KW4kLp1UOkxAoW1/f2qoxzyVVVLFZhtGCwVyjDUw3mXDpUvB0AkD/KaLzx1AmnncnJNo2DLI2yM/cep+17h9x3++cYXhrbTfqsJSaljALLjEpUVd7DP7td1tKVGUbLXXaFRECfevZ2x9Z6GZ+Bk8/IBab2klY+nef2+YaEbdabjCY+5+f63H1MBn8iy3HqmbzGaTeHhGoLuury9Jx0mgjbj9ngMZ5TqlvKzZyMPOrYdrVwviTXRgAAmVlG1j44wLm9/KBQj3uz7Md7cHEwnikDAwMDAwMDgw3ACNANDAzesFBK9QP4BwBdkGrVn7Qs6/9TSmUBfBHAEIARAO+1LEtPGudCIzqEtV2S+bzrke8CAE4kDjif7+SPcsy1HAYAVNP02MRcNdkv2yNeoL8+Ri/RtrTknDp2Oz0eb/0mPSJfnZMcXs9fSS/I2sPM92X1i6C57x6KpRP76Z2I+u4EAHSnmPE5VqTQtrggedOWOumlOV9TOdFPz+7SCr2v/dvk+qoLdMM0qWtGxk6LpNL0RimXx6ZpO5xCrlrerlRD6NwsnqeHJlhst9+1rxNlYRGSgSnHdiwpnoo+V77AE4pe7N6w+BDOFekR8bW4PN95sUcidJGFXVnXkyURQZdWuX0hzj4v+uXz1Qb3mc1TDN6YkQvct4nej+lJ6ajiIQr924/Q25a83j6Xp97p2PZe1uu0H/NJnq+b3z3k2EYfoJf52svEO/7NGWbHH9pPMffZR8R7FNk34ti+Oy5j29nOOfLEEvM3XWW/PQzneB176vRKBhry3dWES6gfdwVH+GQeNxU9S219zDy+smB7jHz0EDbauf+5mPRva5KTpzwvualSKd6MVVcuyMeqBwEAmaG7Hdtantu37ZVIiHvnvfPovRiMZ8rAwOCNjDqAj1uWtQvAVQB+Xim1G8CvArjXsqxtAO61/zYwMDB4RXjJlyml1KeUUnNKqaMuW1YpdbdS6rT9f+bF9mFgYGDwWsCyrGnLsg7b7TUAxwH0ArgdwGfsr30GwLu992BgYGDw0rgYmu/TAP4C4io/j/O/6n5fKfWr9t+/8pJ7sixYjQtF3MpD2CzQRbvRSFyzXXPdtZrtH/7h0557LNX1fTYaugDcSxgMAGEPEfTyqi4sX1lZ0WzZFm8BeGltTbPV03rpl1BEFxsDgM9DnByL6wLydDrtuX00GtVsXmVimh4i6nrdW0DutU8vsXSl6i309/v1fvY6ls9jPCbOjnju88Du/fo5KT2gYLC3S7MBQCitBxUEI7ow29fQx6PR0I8DeAvAvYT+LyzBdB5ewnSfR80ky3fxInC/0o+/lq9qti3btms2AHj26Scv+Nvrel4rKKWGAFwC4DEAnZZlTQPywqWU6niRTQEAgfIpZE6IiHti2xcAAOVOzuFZ1zB3rkpen1JlxLFFE8xFdOppuR8+dsV/c2xffEKow/hjpDKe2E3O69yw5B3KnmaATaqHguLjy7IWqX2kp5JRCn5XU1cAAIYbLBy7uUJOLpsQqW147V2Ozd8pY98sksYKRbm+lctCdcUivD/UCGnAvE+6tZkj1RJdZlf77dutTA0w2smIYWVE/t8b5xw+U2R7MC5r1dwS52N/Q/phJspj9oa4zk5Y0qfhDNf5coVrXrxN2oUAt1eu9W9JyfjUCqQ+K1nSjFvnY/b3SI8tVRgU0NYp6+NwfsKx+Q7I3Mh+n+e5qZvnN5yXz6/ew/tr1OLYXReWMakd5Tp7x+08//ExofGuOcRn6OyRq5z2gcsl0OFo8TrHdigr+xyucj3f1EUarlCTfoi5aEDkOR8X/ULNJhaYRGzevRT55ZlUCPEZGqrz/Md6ZSIMWnx2LU/zc//5ISEDjJaS9Ml0O/fZ2cd1+tqg5J9qeZT0+/19LH8U/2dZay/13+HY7sbf4GLwkp4py7IeALD0ArP5VWdgYPC6gVIqAeArAP5Py7IuuiS8UuqjSqknlVJPzuf0l0oDAwMD4JVrpi74VQfgJX/VGRgYGLwWUEoFIS9S/2hZ1vnqsrNKqW77824AnpXILcv6pGVZhyzLOtSe9vZYGxgYGPy7++CVUh8F8FEA6O3pfYlvGxgYGPzbQQkX+ncAjluW9Seuj74O4CcB/L79/7+81L5qoe2Y2SxlZNrPSDmKTORNzufxdteXO4T+61DbHFMQpMr2vkVoob86yoirS4NCZxzpJOW0b+4jTvvEjR8HACRK5BMXl5/ntV4nv2m7vsw8VC07mbOu5/RZOc/MPsdW2X3MaYeLHwAARPykn3KnZftGryt/2gwlC8lBoV3Ki6T5mgdckW/TQrGk0vy9HXY9dfK2IiHoKiez4GK1u+wURY8tsE96ukmvPTMuuZLiMZYceT4ujsdohfzPsGI/dEDac0VSar4OvkvXzsrn0U7SQ5U8aaOWkIxTPUIHp2+GXsu5hlCfyYCrRE2O+1qsCvW6p4O+jLNTQv60h5n/6OHFJ5z2HXWh72tzpPZ2XnLWaT80Kef39msGHdt9M6T8O30SUTdaeItju3WA1/wte5x+yJU67ztFmRvXL1BKc0+KdOah52VuVw9wboR4SshmbBpvhRRpt+ualxdlX6WuHsfWXOX+A7bUJ1Jn7q1GnHPv0m6hF0czpLVLJZHIdAc4tmPnGEZbTwodPBnnfNn0DCni+A1S3umrcxebXYp4pZ6pi/pVB1z4y641660bMjAwMPh3wrUA7gBws1LqGfvf2yEvUbcqpU4DuNX+28DAwOAV4ZV6pl72rzoAUD4fgokLxc1eglkAiNR10a3loaP1eYjKLz+ki9IB4Lv3363ZanVdBJ0vFjSbHEw/gXBIF8XPLcxrtq4ub2Hz0vyEZmtr9xCgh72Hyuch1vYScHuJygEgldSF6ZGwLiBvSemZvcvlomYD1hFG+/TzXyexN5pNfey99llv6N/zyooOAMGE3idTc3oxy7d2eDPW9YAuoG96TMiGx/GDQV2oDlx8tvP1RNxe4+x1P9U8+hMAah73TtOjIkE6pWd637Nzl+c+n+jvv+Bvr8z3ryYsy/o+sG4a9ltezr6ClWH0nBFx9sgOYQvL2/lTfOsCf+EGqiLOdddXWF3iPfjQGXHF/PJl/PxbUyIuTvfRu3B2jR4RdfcvAQByV3yO5+QKbFHDQ3LsHt6riw2mFj+8ReZRqvUxx7bXx7xBqMs1FZZYHHlbRtbCdI6//psppqGurMl8iQXJOjTPck1p+OSalxdHHFu0QDeUsqdrZZSnEXMJ0Gdsx9umLgbVTKxOOu0dWbnW2UX2fbdfsoEXQswBtjnA3FmjCfEoJROc1wuuag7xVlkD8gEep9nJ+79QF89TZdklQG9nzqlwTjxbqkpfRUVxJsQiMg5n5nj8UKeM+ZngYcd2ReyU054uSxbyZoRK/XuK7JP39V8CAEjMMYP4/h3c/sEWWZe2gB6bUYtrXWuXrFGlBoMLtrRIdnxXui4MLjBjfsNOYWZFuM14H/s5lRZvXcH17LJa6IVajkqfuJLXo9bBfYWbkr8q5PJGHa2wT2cmZa1pdUkZq7bmvtrDfF1XJNgeDosAvSNNof3Dt9/ptDf9sezsEv97HduX8Vu4GFxMaoTPA3gEwA6l1IRS6j/B/KozMDAwMDAwMABwEZ4py7Lev85HL+tXnYGBgYGBgYHBGxE/OElgDAwMDH6A0QhvxfKmrwMA2kf+CQCgWkkHBFys8do2oTejTGGDQDfLauzuFYrjqSMUFLdtF3ot962vObaB9puddudHfhMAsOkblKg+NEf6yPd2oUhmn6UtvZelPi5/WNjOwK5LHFtru5OLGfHo9QCArI+5qdZGpUTN1BClC23nmLMt3in7bKyR3s1e6ioAPG6LiFtJ7aVcTHTBZqpaXNTevIsv6bE3e8bFlw4M8MvnxoW2ysZILz2XlHOKr444thMZ0nSDNi00DAr5szFXmZJZoTaDcUpArFnS1fGkUJpJH2UOjQVyVVZVxOTLSdKASfCiV4sSqRABBeyVBRFZtwSHHNtShNRlpVvE4C0LpGX3biZl94UloQc/eIjU35fmKea+SgnlmfVRiD9Z5TwatAsxn0iSIq7YgQ69rtrx+T4y5ptshrsWIl0YXCTF67ML1ldB2Uy0Qjq1GJSghWqcVHa8wNIunRHZrtkghXogQMqvpVs+X8tnHVvDHuamIh2aj1A2k2+X/OL1sxy7tz9Prr2yR8rtfHnlwoLtFwNTTsbAwMDAwMDAYAN4dT1TSmki7nWzQyv9PW9lSa9DWqzoYvF3vP2tmg0AHnrkEc1WLeoi6kajptkAoOaRhbtYzGu2GQ8Bes+Cd8BjKqmLvfN5PSt6MJbUbACgfLrguOmRMTue0EXEAJDJ6JWA2tr0bN/Lq3pW94WlBc0GAPVqRbNlMrrQvVjSs8cDgM9DL1yte42J/r1SzTux4snh05qtVtDHvrmOVNnnoeEO+T2yunsIuBvrZDD3yrTvlel9vXvEC16Z6r3mCOAtYPcSm9crHvO+oc9RABgYGLjg7/BrLED/t4SvfA6Jkz8BACjt/lsAwEKIguFEG6+9Y0VcUmsxpiEYP03vzP2Lco/8nEvH/7Attn7v1vc4tkePP+20849IGoaxy/mrefhrNzjt0FlJE3BoE++1kyWuj6f3y70+mKGweiHGc25UpLhrb37IsfVGJUAmO/N2x1ZzhZWXKnIPt7pExNUp3kSZunhXipNcExeq9MY1A/DN4SUAACAASURBVOLFKU3SuxDi4TFv37Zb+vg8GJ/jc2CL7f0Zm+b6NJiU+VosX+7Y9jboIhzJSp90TXOtmHQ9bjIJ8e6shdiP4S7ewyW/zP1Gnvd/KcEOKM7JORVbuD62nuHn1bhU3QiscZ3Nh8Xj0pfieOUa9KjEypIjYiJFb9f3Mkxr8aa8qMF7W5hfu3/e5cFTEgjQ5op3Ct7AgIhzT8o1hy32yVBN1sfhQe7z6jzv59ND8n8qwHMqtvJ5lkzIOCQLfHb5XPMkU5Fj1SquZ1uQc8vKyX4raT5jFxXPeXZcPKAx19gF7EdPaTPn08EperPOP7N2p+g9/faeB5z2gWfk+nb7XNW6LxLGM2VgYGBgYGBgsAGYlykDAwMDAwMDgw3ACNANDAwMLgL18CbMDUm999iElCaNpj/ofJ5Z5HK6ulPU6POTpOR6NpOuvWSbUD2TJ8jz5a+RvEBfuo8Zn6+PXeG0J6+QjNibnyUVO9z3eacdtVmpFTvfFACoS0lXbL5L6KnWg6RyOsZdBYjbZLvzxY0BIDgrkoniEGky6xzpqfkBoaIiOdq6dlCQvLQqdIr/AGkVV7Jy+BNCx7ipvYKLPe+y0xq5EmtjcDOlCY/NybluaadwenhBrjMVZ46tI+28pu6C0E+nQqSkBn2kr+bLwhVFfKTcihWObWtcjr+EMcfWUuE5+ZVQmuEat3HnWkrEJNeSmi87tnhVPrd8pNTKTabUX5sQgX3alWPr8i2k1xp5ofSmt3P70ykXzZiQPIczCfZD/1HOoyGbfjtX5vZrgSEAwKBrvCZ7OA9a5mX7WIbcYT3HjP8ByIb5MudDIsg+nfDL9qkIbcFZXlM1K+PUv0pbi0uFkOoUSUPVVVx+1Kb5ItMMHvhe7YTTrmwVucjJWc7Xq85QgB5rl3tvuuwtzXgxGM+UgYGBgYGBgcEGYF6mDAwMDAwMDAw2gFeX5rMs+F5QAiSf16PhAGC5qJd5CXmU1YiG9DIpqX7vkiDd3XpJl9KIXlKkXC5rNgAIhSKarerXo8dKHhGCS0tLmg0A2j0i5/J5PUKxZZ2yhnrsFuARCIloVI8aBLzL3Cwt6ZF7yyv6+c9OT2s2ACgp3UXq9+thctGEXooHAAorq572F8Ir6tLvUfIHAFZX9EjQ1j16xMbx4895br9j50HNVqp5lB3yCJwLBrwj2rxKx3iVnvH63nqoVPRIykjM+/jBkB6NqLwuwKfPsnLZO2qys7Pzgr8D65TSeT3CVxpF8thPyx+7/xAAsATOF18r1524fW+kE4y2G57j/XJyViL7rnItVRN3C+X3e7sYRfaFua847eiS5Ap6bivnff5+Ri4v9Ulx1l0t3GlllFRPcVByFKV8LDPyvUGudZ0hsfcWeV8kgkJrBnI/wvNo47yPFYS6DCdJeZUKzKfVEpRrWhvlcVYarmg+O59PMsfIs4arT4r2rrpdQaKT01xfttrRfFMWcyrtbJHSNwt50qr7m66SJzZ7tn2KEdZnXXmoMvNCGc6F+AzKJki5rYSFagu3cu1eA8e2vCLUZ9nHY7YpPrsmS/bcCLJPAnbd2kadFz8f4ELeaJf7aLOrmvbINpZJCU1KMd9sK0sB7TnLKFxVkeLXfXGWc3l6671OO3pW8jvFfdxnMCfjPNHHWj+7SsxbNmVHK5ZqLH9U7eV8C9XkmRN0PQ9CFa75UZvaTLoelz7Xs6vFknWn4H5MuJan0ox8Oemy+exhLG7h+rRnZqfTPmYHQG52LUvPtfGcdz4s57y5rD/rXwrGM2VgYGBgYGBgsAEYAbqBgYHBRaASHcDZvX8BAOib+DIAIBK83fncl+RyWmuTX+uzq/TYHIjyJ3ZqixQLHo/Qu9H80ccBAHd+g0VYr838otM+OSX5p7aDIt/Jn6YAfdc94n14YpqK4eA73+S0Y197FABQTtKzdeVj9IL5N0kuo8AgPUsJSGHnSJ0Zxssn6R05s0XcAoExCsA39/Pz3Jp4WsLU0cM3QU9r3E59nnL9rK+5nOhRW4C+7HJ2D+zk9k8viQdisJuerRPzRwAAXSl+73CV3tWeFWFDjsTp0uiv0ou1VJHtWnz07Ezm6b1oD0pepJU1lwA9Qe9MKCiC7ECZfbLcIAOSbem1P2cupERTXCXFxrOOLRalh953SvJ1DeefcmxXjztNbD8uc6Z9gEWov2dRrb2/2xbNL7MfusucZ+GUXNNUB68psyrbtC8xK/lMhAxFc0Hm7mqafWPNcT6rpLAB1Qbvi5UAvZrNHgkUqNbILgUr9Aaez2TXWaYbqeDqk067+PSCj16k8ykOUzOcg0uVEadd2S77L5wccmwH11jEu2KbzyzqTMZLwXimDAwMDAwMDAw2APMyZWBgYGBgYGCwAbyqNF+j3kBu0VuI/UJEIroAzOfTT9dL2FzxKMkBAB/60Ac022/85q95fNP7HfNiy3p4ZahY8hBAA0C1pouo11Z0AXi101sUH4jo4mK/3+P8Q95D3dqqK9u7u7s120pOF+rPzeplcwBgalYXpvuV3nde5VQAoOTTz99LhB3wCEhoWN5j5FVmpegRKPCdb33Tc/vtO3Sxeqmkj1085V32xwte1+R1nvWm9zXVq/rxYx6ifhXwLifjJXZv1vVjLa/oY+/zGCMAaG1tv+DvgP+NoyQIFs+i++n3AwAie/4RALBUI/2V8PNeCtrlZHZkWaz28CTv4eKsUCfXUg+Mc8+9DQDwwQNcK+48/C9O+7atfwoA+OoCS9hYn+YcHtn1LTmnEsvWrI5TPLxvUObwvJ804APbWdJpICPESr/FYItFnxR2blYobN7Uz/u7f+4qAEC4lWLuXIVZoQaCUlS5PsY1uVkjbbQ8L7mg/BWKhAs+Uj2haeH8wi4Bet5Vg3ZHVtaQ2Qqv4/KUXP+JVe7nYIylWWbt9X3PPOmpM1meX2Je2vOu26M7RfoqFxEar+IqkdWss08qEHptMs78RpkC17qVstCMmSLXz0V7fUzEtji2+QU+B3puEvqt9wzXl4lmr9M+2SPzZE+FAUVXKbarU7LdfIrP1ZPFh532NfM/LA1XsMrYpFCo4V7Ol07LNcdbxR5YdiV/cjXXLOlAd/xKHByT0rhQwNEw17GSn+cXsJeynCumxm+xzyftV4mwa8k6vyxN9zL31cDSkNPuPiHzqcUV3zYV4xO797T009aQdxDbi8F4pgwMDAwMDAwMNgDzMmVgYGBgYGBgsAG8cXzwBgYGBv+OqEe3YH7vV6V97nMAAN/OO5zPSz7mtim0Cf21NEcabkuQVNT40FEAwPzgXseW2CJU2cI3STFs38FowU8+/bsAgCszjByrf5D00sGvCV1yyneXY8v2MS/Q6FGJcgteSip2992kVcKXyXG7ws84tkD0ctlPk1RvaexKpz2/VeiU5hgjDENDzFW0nBfapD3LyLbaBPmzcLvQeyF2DeIJhvOVbNar4frZH+Ql4XheKJqBAdJP989JSZCtadJg36wxSi5ZloiyI3HmOMyUOHYzeYmyi2dIj0/lSAV1hOWaKosMp6u1kAaMByQPVXp1k2OrlkgHFzslei2VIidmBSRflpqiDCYTZp+3HpOOWHapZNoWOM5bHxFqNtHOjjpc5Dy6oVP6olQjJ7ZtgNRtOCI0aFuQ/FcyZtOdLnlLCSyb47PZuWqEdGmlRJqwKy65v+bBcaiF2I+xbtnO56LuMi7lRa0uf/gDjPaLuOg5X3rJ3p5RkefsN5q2nKssTZ6dNtcj+2yf4+dDTZ6/b7/sa2KGtouF8UwZGBgYGBgYGGwAr6pnSvkUAi8QHfv962RI9shk7fPIJO336+Jaq65nTweA3Tt1EXE4lNJs1YZ3Vva6hwDdS9zrZStWvAXklcqaZqvX9CzW5aL+PQCIJhOazVLegmMvxOJ6ZvRMVu+TVEq3xVt0GwAElnXBcn5Ff9MPwHvsm5aHAN2vj32trvdp0Cv9O4BKXQ8LOH32nGYb6OrRbAAwPTWh2TKtHt/1OHesk8C87hEoUWvotqqHDQBaWlo87S+EP7hOQEVTn2d1jwzq1TU9ICKd8RZohhIXzkflFQzxOkWwfAbdJ38UAODfJgL0QpMejWzIdY/mxGPU1k8B+pMjvC8LM+KRGnAVcFg8LX16KZ08uHt4xGl/rO/XAQBfmqRHonqYv/qf2SUel9y/UEAeWTjmtA/YjpJnXNUMCrvoGduWkF/tx8P0PNVSDwEAetUBx9bX8pjT7p99NwDAn+L9Fc7RC3M+IGJthgL19gY9V4UpWatDTSrMa+D6kV4Sj1OFl4k602Chp1duruUFHn9vULxdp1zeqC0pdursoqwl24Nc+875uX17j7jJJlxBO8lWju2C7WEstXLwmg3mTyo0Zb/lNleAjqtYwrz/NAAgm6OYfDQqXq6eMO+fQoj3fTUp20zHGBw0Nc1rmt9jF0JWXBMPtTP44Gxa3Hn7V3jOsyN0A/UsSdDD6i7mqZoZluz1qRrX+ZUo55PfJ/uK5Cl0D0S4pq+W5VxXq1wDWsF9TS9IP4Va+Vytupb0WFL2734VYJ57oGVKvEhV16PBZ2dbX+jmO8CeFa5VUelGhH1c58618fPBZ78DANiUvhYvF2+clc7AwMDAwMDA4DWAeZkyMDAwMDAwMNgAjADdwMDA4CJQD2/B7FYpPJyc/t8AAGvbB53PVxukUOopEQyvzpGy2pEiBZILCG3caCfltd3W0U6eIL3T/yYqsz/1wP0AgBsbzCU23secVAOjQrc8eMunHNsuF+M+NSHnVHsb8zwNfZX0m2XTUtumSQOGbaqmHpt0bOk1iuIrW0RI35wccmyT+0lf+WdFmN09ROFyrez63C4p4nepABJpiskrNisUd1E9ZZcAfc1mczp76Rd4eFbOdVMHhcknZyiqT6aFczu1TDoyVOUJTK8JfZ4OMv/R+Bzpt1SvPDZry6SnAlFek8/OPZZepdC/0GCZlmDRLicTIs3W3SH9kzrHffpdOfMaJaGlQoucY5kAgw+qZ+T8+vpJXW77viuHYkKo+rEKzzO8n/PMX5Tjd6xe7tg6bV344dYzju2yGmnhpj01i1lXDkWXAD3YJva2KYq9Q2FKaGIJub5Oi9c06nojSUXk86Irn972PAMmptpl/91+zq3ZmkyUngJ3NF8gOZjbLttnl0m57yuRws1fJmWJcmMXX1z+PIxnysDAwMDAwMBgAzAvUwYGBgYGBgYGG8CrG82nlFbCwh9Yx52m9Egvr5impqVbwx5lZwCgAb38xlVXXaXZHnz4Xs/traZ+rFJJjxzMpNKarVyuajYAmJ2d1Wx9/Ts0W7Ve0GwA0Gzo1+RZemW9iC5Lj5JLp/Xz9yrvEw3rNsC7TIxX1OV65XksS4+884p8g+Uxd9Ypc1Jv6P0/s6xHqV156HLNBgDHjx3RbNffrEfz1Wr6uXtUbQEAlMt6OZtITI+ubG3Rx2M9eJV58Xn0PeDdz9PTeimgdFwvUeODXvYGAMKRC8/ft0505esRoeoZDIxKNF9j6z8AAOqurk0pRgjly0ItDHWTkxqeZJmViaJErx1w5f0JDsvOOktcv47ezzn20W6JDHywxCjULenrnfZzPbIWhb/0M45t9s2POO3OmtyXkbOk7Pytb3baAXvuPDDAeZHNyj3SrhghuFr/V6e9tSLH70zz/uqYJZUUzEhEWbXIedVqMf/SwqT0WdHivFmb4/m1liWnVsEVPNpkeick7JRb864wr70ZodFOzpHG25bY5bRP2+PQG+Ycnlak8bqzpwAA52KMlk10cp1fs+S7zSjXnxUf1+fFmNCtwTSj+ZorvL4VO/9SfIX7H50VqmnQtXTmo1xffZDPT/jJ28bTpMdu3r8KAJjPvNWxfee633faffH3AACuqvE6/M+RBq3FpK9O+x9wbMFFiWgL5Rg1eNI6ze2zcvz+ImnZhmKf+nIyx2uux0QuwnVcrcq6koux7zIhrjX1NbkfQq7160SY+0pNCL130rU8VgryPLVcUe5tbaQZW8aEZlTgfD0W5/l3HJNovv7I1Xi5eOOsdAYGBgYGBgYGrwGMAN3AwMDgIlANbcHEwJcBAJmVvwUAWJGPOp+vVOhpKPrsX93zzE+WSdM7kayLV6NR5u/ZFdshHZmnSHfTQQqjn7hTBM17Bylav/OZx532jozkh5r8WeYKuulL9Ag9ZGfxzv84iwqXTn3XaVt2PqHLnqE3zL9NBOT+DD2zPdabnHZXXkTAZZfnZXQPRe3dk/sA4P9n7z2DJLuu9MDvpfeZlZnlunxXe3igCYAEDUgOyTEcQ42TRGkcR5xd7WpHsRuxkjY21sX+kGIjVtrQSNrljuNoqKBmODsih0NyAIIEARAECNtoor2prq7q8pXem7c/zuv3XeLeAguqGNDM/SIQeH3qmevezczzfeccREfpBam06AkIjEpiodAmP4oiY8zwPvSctznVu6Hk4Qp6jvlJdgkvb4u36+gY3Rh/uUkv0cyoeAOf7VV9W7JHgfjVoXhC5pQcYpfbbPNURrxtS216Wabi9J5EG+LlaQUVUXqHDEKuLuMb7nFus2Ny/2KN3q6EUsi915Y+Lfbpjcpf4zPXW/Ksqfd+07ed/son/OPKx6XNZ3vMUzU1Rq/pib48KzW837fdM+F5+IpM7JXd5trren0K5OiN6tZ5nM7L5FWWOQ6Hsvz7Rc9jP5Xk5G4pzvLbzsjNCMfpRIuZ5lem5f6LSob0V4bikerFue4rDXp/AzNyblpZr3cpheT7p8T7u0Ln6L5hPVMWFhYWFhYWFgeA/TJlYWFhYWFhYXEAvO00nyaOdvdoQlC3B4a6YNY12NpDs9jbNZS2+Plf/AXN9sy3nzRePzTUBekNdRF1r6eLwlMRs1i7WtEF7M2GLjbvNM0lcrpdvaRKNBLVbMOBPk4AEDaIk9NRXQSdz+ilS/JZczmTYr6g2bpNvURPu20usfPGIAXAPM6DgS6C7nbM4xRw9PU0HOjrodU0t+mVF76t2T74oR/XbANHH+eeoTwQAGTTac0WjuvrpLeHUN8UFGASoPcNYwcAFYMAPxHT5z6VMpQsMpXNAYA9+vqjgHD3Kg7dlP0ieuIPAQBdpaJSLkzaKNiUPSB9mLTI6hLpr6WeUER3x7inTCyJ+rha4Xi/foHz8bEpWe/f3OQa/W/u/5/94z86e1ba9Phdvu1b76Bg+fpzkkMndOO8b5ubZgDO9axQds/Pcf86kfboH7Ao7mspUoPNoOQdmpokFZTf+Tn/OJERQXC4rVBaSjDM5rq8r80+qb3gplJ4tiVi7uoocwEl1/jedOfl/xvsEk6Oyv53vsZ99IEYg3peaYvA/Ej8sG9bUSbyeF4osSsh0niFIt+VnaDMWUQpsVUD130l59GkMc73IM6cU7sJUdCPJPj+ljxBdHmHc7ujFMZOF48AAK67HOfpw1Sr37EjbYlU/p5v+w/v+SP/eLYxDwD4caXsT2yba6NSkDV3o/mSb0uuCSVWC7KdG5vMKRUfEfusso/GIqTcGt6QFbL8+44iUE94S7+h7NmZPtd+3fsYT7RJ7Z1Vyi8lb8g/NpRtMOLdP+Ec4XN4ObAifW7FOM6vOPy8XLh0DgBwLHo/3iqsZ8rCwsLCwsLC4gCwX6YsLCwsLCwsLA4AG81nYWFhsQ/0IwvYmvtjAECm8hkAQGfInE67Y6R9PCYI2xXS26Ec6bMjG/I7tlknfbRckL8nEoyme+AEacLlp4TqCj3EpEqf/ssL/vHds3Ld7o+TbjzyOBM0bY99EQDQHSENf+NFJm1qTgndcedZlucI3SG2sdh3fFsqRWpwtCWUXP/mu33bxsI5/7izIvRdZ5r9CHT5sROPSy6iYJU2pzDhH0e9rsQSpPaGi/4hbrM+cSWab8ljx05OU4bw1NaSf3xHXuxP7dI23uE8vAyhxWddSiAutkm3jqaEllpukl/KpFmCp1eWBvTzvKY/ZDRhvibzpObeGksJpxUZIe0bTzDaLlESynJModFzuwolvyZjttr+LPv57Q/5x8m/Jbm1Ll951Le9Y3reP+52hP4bTU75tiMZaedmmrRrv0mabj0iVFk8peSrYjeRnJS1U1Ii46YUBcjtILtikNTessv+5Ufk+lWl1M+JDjm77RF5X+IxykLW23LczpzxbZUm25edlDmbqlP6MF3iPbunZMzPL5nz6L0ZrGfKwsLCwsLCwuIA+L5nQN8LA0O2cQz1b4uuwdbvmgXo/a4uxB01ZPvOJ80Zp8tVXbDbbeti82pDF1snEgnNBgDNpt7PndK2Zpvt6Nm2AaBtEHZHI3rG6nDUPO6mJOImEfTY2LhmKxZ0oTkA7O7uarayof87O1uaDQCCQb2tkaAh+7wh+3toj/U1NAQqdDr69WfOfkezAcDRmWnNNuzq14dTuvg/mdZF3QAQCejrsdXTBfDxpC5UB/afKd50HgA0a4a1E9az1/f7+vUDQ+UBABi+4d3b69k/jAh3bmDi6m8BACJ3/d8AgPqEki28T5F0qy9zns3wXTy/xF/d61G57l7lV/2hG+IJKfUonL5+hlv0SS8T/to36Ib5zWMU2n7pjGSn7i3xXb04xfuvvSQemdjhZ33bqWmuwZubXjHcQ8xQfiwr6/FCdN63dYpKbivnHgDASPEvfVtx85f942Ra1lOhyjxOuQQ9Ee2qiLTDffajtcVx2h1Kn50Q95SxXfap6qVNqt8A/+5pvVdLfD/vS1Js/kpPRNZ3JO/2bVe3mVX+VFa8PJe6Z33bYprzvByTeU6kWcGgGeE8VQvioQwrWbY7Mc7JWk5cNf0G94p2XLxdxS2Ksbd26WFMpWVtDJT9pV2nQP3GtIyvu64Ip4/RQxiPyTz89DjzhQVmOOa5HfEgtm7w/hvr8vzSUXrQdpv0GgZc+UwqR5nXbJjjPbte8+PKR0e7y8/WsPfRvdmhqD0XoVe07A3veIWeo5eUWJjsruz1AeUzLJiWfSnYYjKyXIafPc6KrKO1LN+xG3G6zo6clYLYdyQfwFuF9UxZWFhYWFhYWBwA9suUhYWFhYWFhcUBYAXoFhYWFvtAPzqP7cXfAwCMbv0HAMDQ/TX/781pCtBb3s/U1gapppiS527BEykPFdqinhHavN0gNXiCGmSs7UjR4shPk376i8/9tH/8Xq/w7atjLGEzdZH5lSbfK+L5+RvML1S7Skrf/akZ6dvrFLi3xoX6O1EmtRYckvLOjC0BALKlj/m24RyL4a6uCsfTPj7r2wZ18jKHwkLHOD3SS8ECqXov9RWcNJ8fUJjognecUErMlDymfHGK93xhl5T23V6ZlMe2OY7jirLjWy3p02SbEpLXlNyFowFp37Ue6ffxKCm5WlOoU2eEcx9pkRIMhyUoIBAiTdepCI2XDLIhEwukFp11oT6jStXg2gzlINPLQkuFFIrY2eK5dx8ROnXk/N/ybbE7KCk4X5f79+/l4L7DkfUcLzFv2cIsaeFXHBG1F5JcY4Mt9jkwKu1vblPIn57kPCx3ZfGPB0jtLbfZ/rmYvAc3XVK8dw/4lWUrJs8a5hlo0BgIBRuJLvm2ZpPUY96b6Ika34FYi8/H3dLXC5esAN3CwsLCwsLC4m2F/TJlYWFhYWFhYXEAfN9pvqEhGg8AYIgCGhoilbqmyL097hkyfHU0lep4+OF3Ga//2te+qtk6hnIuJgTC5qFudfXyG7vVTc3WbuklZgDAMQRVHTSCKhzVI7qSGT2iLL9HNF989aZmyxjK0SSTetQfAPR6evujhhI3vd01zdYyROgBgKFLGECf+62SuU0ffM+7NdvS1Sua7fjpCc22Fzp9va3RqKEU0B7rOWgoBTQwrOeqoWwMYI7cM5WoCRlKIfXbTc0GAAF899wZgkV/aBHq3sDoDYnmi538twCA4ASjrxIVvsvxnlAYzjQpq2sr3IC2Dsnx6RHuX6HXhQJJhUlBLN1imZXR4DsBAKtfeqdv+3vzpHqeuibzdP/kj/m2C7t8pvPYrwIALn7kT33bRFXJEXRL1kli6iFekxLK79ks21mc4Hs33hN6KZL8c992ss38RhN5adPEBtdascA12m1L9Gymc4dva25yHHdceR/CO7RFB6TCSilZb8F1JT+Tx0Juk+3E8QL51HNebq87YjO+7XyHfVpMzAMArvUZbXwsRUpuLSBzPpbhe1mOsFxMNCOfCUGXY1Y9NFDOlfuGqtzTKhNy/9B1XrO+xTC4ibbYHaUcTGSF1Om5hLQ/qiRy+tAI196tCSkHlIj9f74tkPgg++dFWya/yUjRblr2pwsnGeGcvqz0MywRkmslhlL2D7HNU22Z+4ASXF5WSiXNeV2pKKV4im2evOTKPEUqXKOXYhyf/NCjIclKI5UWyq4X5edNbITjHL4u71g9xr9fTvAdOv6y7OlHC/fircJ6piwsLCwsLCwsDoDvu2fKwsLC4q8bjuMEAbwIYNV13Y86jrMA4LMA8gBeBvD3Xdc1J6jz0IvMYWPmUwCAsS0Rcwf6v+7/vTzDX8Cu53RsrVNwO5Kihy+9Jb/mG71J31abFO9vS8nOPDbL37tuWH41T84xV9Brn6Wg+M5jsp0//TT/Pp59n388/5uSLX3+c/Qqvr5Dj0zsEfGqls6y2G24KG25b4MK736f3td4/mkAwOHOL/m2TJf3rK1In67MMwfXYIuC36lDks68pwirwwUeZz2nQUexBekkQs6rlttlGincnsUJ6o5xhqmMcHJSBOhfbbCfC2P04nxlR04eVXK+nYvSe5IKCB1wrc/8R9Nx5qm66cjkh2OKGHyZSytaFC9dY0jP9OSW51Gqb/i24iTnqRCV/u+CGcobI8zyPV8RD2bxLL12GwMO1FRdghfu+cb/5tuax+hZf74jHqniXfS2jUHGZKROb9TRAoMPLuWkzxMDesjau/SWBz2HUk1x9o8rRYeved27M0xv/DmXEzXqTXmtSU/tPPiO1bKyzvpKIqvOQJ4/NbzGfpboQkyB/AAAIABJREFUhQqOyrkFpfB2ocL+4T4R01+89tb9TNYzZWFh8TcBvw3gvPLvfwHgX7quexRACcAnvi+tsrCw+JHA9/wy5TjOjOM4X3cc57zjOK87jvPbnj3vOM7jjuNc9v4/8r3uZWFhYfF2w3GcaQA/BeB3vX87AD4A4HPeKZ8G8HPfn9ZZWFj8KGA/NF8fwH/nuu7LjuOkAbzkOM7jAH4NwBOu6/5zx3H+KYB/CuCfvOmdXNcokDWh19PFufstlREO7Z+9NIl7H330UeO5X/vqY5qtaxA8d9p6O4c9c/mNcEgXAdd2S5qtaihlAwCtji5g14vJAK6zf1F6wCAbThrKwaRSKc0GAIV8UbNVq1X9vMKoZgPM4+dCn6fBUBe1w1DKRs7V7xkwrJNAwLx2arWaZnvqW9/UbKdOv0OztVpm9iiZ0WfKMYjKh0PzO9PrGcbE8H71OubnZ9N6UIFJ6j40jF0kYlD0A2g03hgo8QNRTuZfAfjvAdzucAFA2XX9mjgrgMKf7IFwbwnFW78mx0c/DQDojSn5akqKmNyRd2NknHP8zGt8h7uT8o4cHeP7m7goo+/EaCu1eU3Ky5ezfJ781U/Mcm6euiLv7a9M/2Pf9udVFkJu/qmIvK/ewwCRa9+iCLu5/DIAYGbhPt92Pi3nlpR9djpF6rIzELH5s7mnfNvxMCnB8Wmhoha2fsK3xUcU+qv6PAAgEjjt2xpl7qm7rkxZfIcrs9Pnuq2HhcqKlklvud72U1fiQw6TTcXVqjz/WIi//6+2OM4nHOEMr4P78J2uUuInLv2fH2E7ttLMMRYpy17nxLiXDBZJT+0GZEzHt1kibKMo+1ZIof46GdJf21dlnZVGSKFObN7D9veFtho+yLI9U2Wux0TrVwAAf3T6D33b8U1yo+9MypobW2FwQ2laxuS6e8m3JVe5Z9cjsnYubJBOzCgJu2LevpFPcU9rl7iP5j3/y3qHY1vscU62u55Qv8tnbivbc6gt1KajbDmRkNy/kuU7ksny+dGrcnI1yrG/lWPA1+EnJEfa3XmWGtovvqdnynXdNdd1X/aOaxBX+RSAn4X8ogPsLzsLC4sfQDiO81EAm67rvqSaDacav/U5jvNJx3FedBznxe3q/iJ3LSws/ubhLQnQHceZB3AfgOcBjLuuuwbIFy7Hccb2uOaTAD4JANNTerFYCwsLi79GPALgZxzH+UkAMQAZiKcq5zhOyPNOTQO4ZbrYdd1PAfgUANx14gF3e+IPAQDFjT8CAISHv+Kf25zh97FOXcS71W3etjDGX8Px5hIAoLtK70bPK2TcDNH7MNllaoTarHgn7kqv+7byc/QuFO6Xgq2feY7i25OjFOeu/4IIhk88RsHw8tRf+ceponhXSs/RSxOYENv8ZX7/DClFfceT8h01B+7t05sUTndLkmZm9f3MU1BdphOwPyftDyreppySxyToOenqShaWnOIkziRkzAaGOvLpO3m8rDisj0yI1/DZKufmrlF6ub/kubRmBhyHlwt8aNYRL9lVJaQ/B3retxzxKMUj9Aw5K7yXm5f+uV2O00xNOthqsKFumtdH5mQ9TQaZjbzWYUqCXEu+6EdX+DEcrNJbtll5AgBw33M/5dvGfpHr6KWOFEh+zwQ9nRMN8XJnQHH+oSK9ZaGIeMNSCXrdrjXpn4mPyrmbytjH8vQ87XhO3ZM59vN6hZ7cvCMK9KbLFBWjSh370oj0eQB6roZDeeYo6D3tNvgjqD4t78NMjX1PVbj23HvFq/r61bfuUd+3AN1xnBSAPwPwj13X1TmbPeC67qdc1z3tuu7pQt6cl8jCwsLirwOu6/4z13WnXdedB/C3AXzNdd2PA/g6gF/wTvtVAJ//PjXRwsLiRwD7+jLlOE4Y8kXqM67r3s76teE4zqT390kAeqZJCwsLix9M/BMA/63jOFcgGqrf+z63x8LC4ocY35Pm8yJffg/Aedd1/0/lT1+A/KL759j3LzsHcjvCJDQHzEJak1jclAV6r3uazu0aRO2FnDkwMZnOaraeQZzb6dY1W7NjzhhdyOmZvat1/fqtbbMAfWKypdk6Ld0WjZqn2gnq36eHrj7O8aTeztFRs4D85k09A7pJrJ5MmqTyQKGY12zb29uaLR7X29Q0COUBoNvThfo9Qz+bPX3sAOD1ixc12+E5nbbutfVM9amskQHHwNHHftDX171p3QKAa3gfGlVdKL+XWLxvuD5kEOWb3sW9AknC4fB3/fuN7/v3E67rPgngSe/4GoAH38r1od4Scuu/AQAIHv59AEBL4ZyKuzxOBYU6mJwhhfDMsxyzwLRXcHWSa7i/KoLnZJfvynabFEXA02zdOM9s4XcFec/et4Uy+6VJUn9fPcdsEIMbCwCA62OkVVbPcr5DsxJQcXqBe9Vru0KHpMc+4tvGEqRdzkckT1RknNTieviYfzwalY+FkyXm48pmue5u59vKpbhG1YCNhCcyDuxwn93u8x13QsIVZXbIegy8wxpZMIwr4QXbt+T5d8UWfdvLPYqsZyDi5XWF+b0LigA9IhTQ7CjHfjfJoJuxbXmvt6Okirrj3NO2kjIn8wXSvls5oRYrXY5DaYzzUFyR/q/1yJlN97n/1LxxCAc5N1uHuOcFS1LBoXXi22znkG3+ubQEKoRmmf1+0JR1mChzbGub7HNzUvp5tULqLzzG9327ItTusKjMzRb3vPG+XLfa5PUO+Nld6sqeHwsu+LYNRaA/3JC9xlW2rEBIPjtred4zG+HndvCmfI7uhpm3bDvJcV48cw4AcCzHgIj9Yj+aqUcA/H0AZx3HedWz/Q+QL1F/4jjOJwAsA/jFt/x0CwsLCwsLC4sfcnzPL1Ou6z6DvUtsfXAPu4WFhYWFhYXF3wjYcjIWFhYW+0A/NIfdcSknU9j4QwBAJPgb/t87Y6Sf6h510VSq7WYmSfklKkIJOopstRMUyi8cZjTfrEMq53ad5GN3kL7ZusUitLFHhML4wl+ySOsjo6RYXqxIDqJDLeaWmnnv0/7x3AWhSLaWWdIk6ZWYSZxb9W3OJP9+R1OO4+1534bpy/7hxJaUmemMM9/V8irbVFqUyMK+QvVMDhnBOPD67ORJf6UUxUMwLvcaKIqDmHerIWv2oqmw37PzctPntimneHCc7f/SjhS0L7aYP+lslsexgVBpKwodG4/xXhthWQcTIEW7Uic9ddijwnsKDxlPe9SswxIx8V1GYhbiQjMmxkizubc4D6mB3P+Ww8GZX2KupJGgUJaBHSbcuvsUz22+IvnChjOc57WhyBM6h9n2u11KDgZloS7n530TznR5z3Gv4nRJCVcLKikIr3rs2nyEEo7VG+Rj0wmh4vpRluoZD9KvU/PycEXAPgV7Mra5Lh/a7zMXXHtS1vNEifm44h3KTYYPSAjo9Yt/jdF8FhYWFhYWFhYWOt5Wz5QLPQN6x5DBey+YxLEmxGIxo13PzqwLZgEAAzOr+a5HHtVsj3/1S5qt3dFFwO3OHoLdmC6i7jb1jNUb28Y0OJivr2m2kbYulO93zcLsYEL/Pu2EdJtJAp0b0Z8DALlsRrOZxOplQ6Z3AGjHdBF4q6XPXSKpC6sLI/qzAWDt1orBqv/6GATMc7+6vaHZ3v/eRzRbeXdHs40ldUE9AMCwnoMGUbrTNa+druHdMQVpxAxrDDCLw00VBUz3NJ0H6GL3HyQB+kER7N1A7tYnAQChw78LAOgn+K5H6vReBL11NJZlMMu3b9BTUcmIl2pKWcPJtsxzEPR4XO5f9Y+LA7FvX+I9E/0H/OPyF+T/v6wIl5+7yffmwaRkgrimuHacv6Iw/PyHJXfW9IDv0G5dAi9yk+/2bY04vUhPRuT+cwXm7ckNT/rHG9nPAACO9ihgP5TnOCysSVtH0vQk9OpL/nGqKfmP6pvsx7DLnFBuQK4fa9O7UE5578smd604azOj6jnO7ilynM80KexecKWKwXKXovSjIb6DN72kVkcyiucpzuMJryjyTSXop55k/5qegD05y0CBflA8QrUMP7vKaSUb+oZ8Jmxs8p6TTR4P07Jv5Mr0ZF6J85mxMRmLB5WiEZeURFzpk88BALrFR33bXE0C9GvfOe7bunEGAuwuisdo/TLdTYEw1+alhnh/ghkOfqrMOcl5wTbbLa6nlBKTFOrJOPXaFKCvZPnZGNySdyetbEXhmNyznOaemwrzmYE1WfuNAD+PSgH2af6MeAZncnyv9gvrmbKwsLCwsLCwOADslykLCwsLCwsLiwPACtAtLCws9oF+eB7bE/8vAGBs5w8BACFXEaBPUzIwrAv3cKVNejijlJNxvMK9/Sh/z9YjYov0SUscS7L0y9WgUH6JeSU3VI/0ff79kiPn8udJUdy/yOuf3pK2jyZP+bbR33jFP37Xf5L2L9fO+bbEI0IbVc6+5tui46Q2H1z1ysEcJV0Zzr/gHy+0flae0yLlVLlGAfrrHxLaZrJGSmu+T1op4kibAgXSR1mmUkI7K7yQmootlZR/9MkOwVUY8ZTHQp5XypycVAr0PrMpYzI9zps+v0WZQCQsx2cVMXQuyDFZh7S12KVweljn2sh4iZGGZUo3eqMiok5XOA6LXeana7bOAgCiBY7ziNLpDQiVvxOnqL3YoTB7+oL0r1Llehv7KMtVHv+KZDbaOs58es+2JN/W2Gm26U6FWsw0pfRKURGVnwlQdrIQlbV3bVfJ+6jkdKp5ZVzyMfapvMRTs2mhdvspVqw+0iMFXS5I/1NBrqedgdwr61LU3mxSpuBOybmpddLSmS6p1f7dItq/ftWcb/DNYD1TFhYWFhYWFhYHgP0yZWFhYWFhYWFxALy90XxDV4ve2ytCz2Q3RQaZylqYyrHsdb1jiFSKhs3lN973vvdpti9/+YuaLRI2lIipmmtDV8q6PZfVoxF3dvQoMQCoKxFEt9Fp6LZB1lz6JeDq36ddwzgFDBF+6XRaswFAPq9Hr+3u7mq2TMYceWc6NxTQ4wnrhnkeQJ/PPWHo+9Axr8eBYZ00u3rZoqe/+axm+9jPL2g2AEDAFCWn24YDc58qFX2es1k9wnKv0i+mMjX7jb4LBMy/w/Z61o8CQt0lFG7+AwBA+IRQZp1R0gGJFve2fkIoiiMx5gp6vsq1VYvJ2B91lLIXXqmQ3pCRYZear/rHaVfyLwVbpLwqyiuwfU7ovXcrkWsv3eQJHz/yCQDAF/rM+dT7K0beXT0m566+zjxU/RWhl2ammf/n+RTpoddHZV+6N8rIuOTwQ/7xN0e+BgBYCDHn0dwcc2PN7EiUXz7LkKxYmTRhPPwoAKC1zv2v0ueY9nalzY5C/3QDQrllS9yf+sr2V/ei+Y7yNrhcY5TYXFDK4VxskF6aiTHybqMknw93xzm2N1xGhM2EZf6uJDlPnQJpwGFM2tpRSlxth2ROggnunS+lGE2Yjws9VnX52dBR3/8R2SOyFc7nap9rs3mP0F737vCZ4dov+cd//P7fAQAcrb/Dt90F6V/x6hHf1shyz7uQlNI0hZv8u6NE7p1bEjo0ppTSQYdtinUksi7Q4udtosg9KejKXhZsMATxSpLPD5SF4g0oW07U+0epQIo0F+GeFrwmY1ZTItG3wnzfDr8i630xz7I6+4X1TFlYWFhYWFhYHABWgG5hYWGxDwwi8yhP/x4AILMlOZmSg9/0/96e5q/+hisep1KdudQSGQplE14+sk6Kv7pv576J1ClMnggxm3lzTDxCTpO//gdF/gI/clKU2dWz9F4U72f7P/2CCKJP5diO9Q9v+seHn5Ln707/hW+LJ6V9y9dYvDw0RmHzvRfFCx9L8Jk55+tsU1fanwjR29wpUyC/NC2C5coa3QuVaWZAn/G8dOkUn5lWMqCX8+LlSSuO+05CPFKuklpPSROFlKe/v6U4wGdnyCa8sCZBA8eL9Cw9UVYykxfFy3NVKcicGZJhuBYQT0xxwHlcryrel4GIxOvrbGB4Qvo3LFNdP9Wnl6nTfl4OcvRkHhqj8rsdFE/LrS7nKTeg56p4WYIfNhVdtdOiF/0DT/689OPn+fxXHKkDHpmgF2e0x4EO7ko/T2bo4ruUplC/4BX73iyxn60UPZCNoayt4Dg9Q51r9PDlUrI2qwrRc0IpztxOSVtDQbodVyDjMzZg4MdWn+9l4rCM8/TWkm/Ld5R38B4vA/o1CuX3C+uZsrCwsLCwsLA4AOyXKQsLCwsLCwuLA+DtpfkcXeC6lwDdVK6i19MFv912W7P1+33NBgCpVEqzmQS3pvIZADA9NabZslldbF2t6mVS9iq/0R/ofXLcpGbrdMx5L3Z2dLH2aF53UWZG9ecAQKCvi+2DYcN37KHefpOAGQDShnIy8aReziaXy2k2AEgk9HNjMd3mGioR1RU3uYpIUL++A32d9FyzANtUDej82Qua7VZuVbP97Mf+jvGeg76+zoIGAXq1bC67Y1rPJryxxMubwfSOmbDXO/KjjFBvCaO3JK/U8Pi/AQBUpygsnqhxTBJREe9GgtwfbuzyfenEZU5mlcpRI32hhVaU96oaP8u/VyWvT6DD9yvW5L5Q+o5QOUeUpb78Mo8/PvV+AMBj15gzCreO+YdX07Jv3LymrKtpEZDffYxterHFEjfJCRGbz4T5glyNssLwa+Ny7uKAovbxiSf846nSz8h9lLI6xSpzJaVCQmk2Gsx9Nezy/sNt6f9Wi3ummxAqanSXdGZDicvoea9okWm/sK28totRoYKuNEjTHYswX9fStvBO9wZZ4uaisj8eD8r7+p0Inx8bUwT0ARHzd4ocs42k9DmR4uRVchTquxV55iBJjnNrmdTiVkjamgdp4brLnE9FL2dTMsOB6O+8yz/+/H1SFm2iw/X64bTsj6kYy870g/xsqDWESrtxnXRk427SvSs35LMpVeBGXehyzKpemav0urLnJri2664X9NBXStQk+b6NbAm9p36FiHnBOrVRUo9qnFTmqsxNWRGgb0TYpvkzkmPscPY03iqsZ8rCwsLCwsLC4gCwX6YsLCwsLCwsLA4AG81nYWFhsQ/0w3PYOfQpAEBiXaL5MPgv/L83Z0nBNHpCOw1aLGtRTDDqaNAWumEQJf11KykUy9gaqbtAl/mNAhNCK2cdUnPLQ0Z3HblL/l+ukh5K3Uf66RtPCE34aJz81guFZf94oStU1qF3f5ltviAUSHmJ1NvIhxlNOPK69M+ZYRTXzDapnpQr5wazl3nPzb/rH7enhIqqrpLqac6Rgml1ZZzmhiyBk1I+tWojMs6jCpPdigu9NlQUGHElIizkVavpkN3B2DyPz64JLXVKyY/09RJ5wMmMSEte6jNaLt0j5XbOi4DMO5RgbDZJQxbDEqVWqXLuYkEZs10lH+FMgo0qB78JAOi17vBthTQ7lYwLl9Xc4TwkXUZq7jaE/iqWuAY7xef94yNXZG1M/QRlMzccyau4UKTPZTvJuW3NypzNKTm84h32abYg43OtyWtqk4zGC7WFVu6kSQMOyUwiHxSqfBDnmCy0GNVZHZH3LR0nNXq9JgthVClrs9Pk9YM5WeMTOxybfIfrrXPPw3KfK/uTO6iwnikLCwsLCwsLiwPgbfVMOdBFy+4eGZP7BoGrSRwbjkb3ZQPMYneTWH2v7M79vt7Wdz70Ls322Fe/otmGrlkU32jqYvHRgp7F2gmYBeymLOCu4VmDjjlvRiiri933i70E6PnciGZLJfTnxGJ6pnfALKyOxPQ5bbT0vjth85LudruazbTGgkHz9YGQ3tfLN65rtgdG7tFs1coeAvIR/VmNlh5osNc4mYTlxiz/e2Q1N639/QZk7HXPN2bqN70zP6wI925g9NZvAQDcY/8aAFAap/chVuU76oRlbWZy9G584xbX4CAu9mPKMEZrMh8bIRbFdfNL/vHUUFwqK02uxWSM7/XmJVkP0SAFw1vUr+Mnp2QuXlvhPvpQ7if94ysb0n7nr/6Bb7v5oc8DAA4P+U5vlc77x8XRdwIAGkn2/VklJfXshHgPss13+7btFPNYner+BAAgP8rrDymvdTYq9+o3mXV92GL/uiUZ080ux2To+QjGlQK2O0O2Kbou50bp5MAGHTpY9ETQl1sUOx+J0xt3wVviJ4K8/7UWn38qLHNyZcD3eyxNj88VR9oXHqFrbDsif3eV6hGbYRbjrXlutlCUHsLdAT+nru7IvnGyzzZV4rzXTEv2z+URjnMlOe8fB06K8LqVuc+3/URCRP/NKe69yT738bGL4pKqdDlOO5NcGzs3xavajVEBPrbD9rc9Dx6UvGHq1l/ycrWFOwz8uprimKXX5Pqesg0mvKCiVojusmyOEx2+IntxLcN73gxzzzryylMAgFNZZnXfL6xnysLCwsLCwsLiALBfpiwsLCwsLCwsDgArQLewsLDYB7rhWaxOSUHY0bU/BgA4+KT/9/qkwk+1RGh7o8e8Z2NpUiRtRwTNzeG0b9tMCr2UUnLnZSsUHNeKIkA/lmKepQuguHbMYya6IVLFU++h0PbKn0iZlvkjpD2eqFz0j6dH5L6lX1cKEQvLh2vl131b7kFSXptnnwMARBrkZ965ToF7oCd9GSYe823HHRbYHWkL7bO7wmLgypBgsiw03qxDAXo0QNrJycvxuJJebstLVaRUeEEuSRquv+gdKHnqJuZ5vOLRTqcmOF/PbpNeOpYWGvapHuepGKYY/ZWS0OL5DjncpSYpt0kvV1JVUV7MhOWelRops2yc9FjIlXGI9RiwkHBZOubIIWlrcYdtGgwpzF5OiyTh0A7X0/wZllypVyQvU+E0abr000L31uZIZXdDzLc1fNgrKF1hO4ddFrROTkv/l5U+hUfn/WN0ZIIiiiynvE1JQWFM5qzVYTsXW3xWZUTGPJmkLGa5KV9pChFKK3aqDLLIz8s7UNwitZfrkid07xMB+tlLb/2rkfVMWVhYWFhYWFgcAPbLlIWFhYWFhYXFAfD2RvM5DkJviJRrN81lUkxVPaKGKL29oor2ev4bYYrwG+wRYWiK8vuxD35Ysz3x9a9qtl7P3E9TlF7P8HxDlREAwNrahmZbGV3SbPmCXgoHAGJpPcouEtdLrziGr917jb2pHMz4qP78erWm2QBzlFo8rs99raY3KhQIazYACET1iLhAT4/wG+4RfdY1RL41HH3upmamNdurL33LeM97T79Hs8UMZXdCe0QYDmGI8BzoC2Wvkk2m9bxX2aM3whQdCQCZzHeXEgoGf3R+r4V6yxhb/4cAAGf234mxQC4p3mCUXC8ix3Mx0jJPt0kVDSNCa00rwxWqyB5RAc/bzZB2GQnKvS62GP6UC5E+cr0tpl9m/qHdF+b84/d6KavO3eBe9PHp3/aP/7T5NACg9uW7fNuVe4S6vH6W+a66t1iOZn7mHQCA5wu0XXN4/yNpeZ+y7gd921PDZ/3jEwN5VujQt33b0d2P+sfxqKyzcIvXhHoP+seddRn/Wz1GGwZ2vIiuAQe36SjUaUn2Apcpj1BjkBlyXhWYlbpC7eVZnuS8RxHdHZ7ybZfrHPOTkCi8y0o03mSKNOANL+J4kCDPWPYidhsTyv4VZ4jhICr3XwuStp1M8F2/5rFaToP0Vj1F2rlYknVwJczPi5VjvP50TTjHweCXfdtL7/iiXBslrTvfUcq5vCzU63aIHOvKuFJi65pHKaa4RktXOdD9aaGGew22I5pg/2s1aX/c5RpeVkrDBL3h7ba5Z90mm9su99HxJCOsqysyPptK1HcjyM+xw2deBQDclWdU437xo7PTWVhYWFhYWFh8H2AF6BYWFhb7QD88i/UJKXA8uSsC9ETwV/2/l2f5CztUF4/vaodFgUdjFMr2wpKhea1Kj+1OVn5h50DbhJJjpxu4BACYC1CsvV2k9yPZEM9VTcmfNKc4edte0WPnQXosPv/iGf/47qJ4VUsfphdn8esiJi9lvuDb0kpx91tPXwMA9Kf5UXL3d3gcWZTjSOFJ33Yy9A7/eLQu7e8vP+TbbubpfUntyu/94OK8b+uBXtF0WrxDGUVs3vE8SyNKyrNImp7p8G0nkuJsTh3ncdlz/M1N0A3ynW16ZI56hdy/NqAXKJelx+dbDfHMZZUM6+cHnPsRR+611aP3ZKIj/Syts++NcYr6o11ZW5MtFk9uNZnnbnxKHjYRoQd/F/P+cSUpbV2ssh1j5zmO1bqIvVd//D/6toe++D8BAHpTzF5/o8/ghyNHvTxRbY5TvMSBLM7IOlkeco1mRuhBLDvS1mCOi7S/ynFOTEj/h9tcr7NKcMZqROY0niK7csOVv48Oec2u4smNTYmHcWyDfQ/0OSbuafGUnr2kpMzfJ6xnysLCwsLCwsLiALBfpiwsLCwsLCwsDoC3leYbDodovaFcRmiP8h/hsC4kNgmzTSLovcqc7CWafSP2KifjGFTxiwtzmm1kRC+nsltqazYAaBrKydQbHc2WiJhL5OyWKrptt6zZGhX9PAAYOTSu2UzdfytCf1OgQO4NwmQAiO9R9sckQE8mdaG86TmRtnmcBwZdeqOlj30gYO7nGwMnAMA1/BYpGebj2pWrmg0A3vkePXghAH3tmsq5AEDQNaxzwzyZysbIqfs7t9PR16OptBMAjI6Ofte/93oXfxgR6t7A2IoUNg4e/n0AQGeM41WoUwgb9ET/sTgpuScHXJvBvuwbc8rrl9yVea60eN6tDPNApZyTAIDNNKmQ0RJzOg287S1Y5Xop3eIazXtsRvolztGvz/2Yf/xn10XsPrhFnvBWUfqxvfy3fVu391f+8b2npK0bDRYCTk4v+sczeXkfboVJ/7QzpI2mW0JNpsb+1Ledrv+8f5yNyDsaaVzybWOh+/3jak1yCDUazH/Uq0qbNrtc330lBVi+KfuGkhoLwys8TnpzsrnJ+ZzPkQo6syXFnY+kuU9dLVGkfadXXuVaiJ9Xc3EWvF5OymY0kuM8l8LyrOpczrfVgyzW60SFStvIkbtMVLnR2UutAAAgAElEQVSn7gbk/ju7fN82whyz0b6IsK+3SE2WjpMbXazJfUd2WLj78++Rsj+jAwrtPwjO3egtEZgP49xHVyPM6TQ8L+2rHeZn/HCVNGZ/Stahq5TtCWY4zuWGCNuzLtfTcoSTdruO9G6be24iKHO7m+E4JpXSMsMl6fNWlGujE+GYLZwR2vuEVybprcB6piwsLCwsLCwsDgD7ZcrCwsLCwsLC4gCw0XwWFhYW+8AgMo/a1O8CADIVieaLhhnN15sldVBpCMUTrpL+moyRqg56kWC1BiUBlXG5JrHDyLOxDnPkdHuSo2daoazdHEt9pJpST2Y7wt/IU2Sf0N2W69unSfU8/cS9/vH7IkKxvD7J/EaFhkQTTtz/J74tfIOU1dYrIikI/hwfNP48/46UUDgLXVJa0Q65zXBCxie583O+rTXBPm2syPMHp0j/DLqkVqcCQu9lg6TkdjIyfjmFse4qlN5txl5N6dY7xuOm19T5w5zPl7dJdd8zJlFqT26ShpvI82HfGIiUpdBguRclwBFzXn66C0Maix4l2N0h5ZXJk8brNYWnjJaZHKvTv8b+DWScJtOU0QwypOfaPYn2m22RZnOUaYrUJErvZusZ3/aR5z4uf/sZlhK6FCStHF6QAYw4HJuKSz71feNetF2UdGL4EKMRN7zuB0LsU2+TlF1uUvrUbVJSMNtho1e93IPpBKNLt4dy/fiQUpdWn9f3p2UdTZQ5X+Em37feaXnfLl7cI7Hjm8B6piwsLCwsLCwsDoC3PQN6NPzd4uJAxJyx2ih4NgjQTdmd9ys03+s5e4lmAwFdGN3r6eLcu+64U7N97elVzQYArYp+fSPV0GyhPb73moTAFYPYvLfHmPTqugg7HNPF3iHDkAyG5mzZpvFLp9OaLV8sajYAmJ2d1WyrayuaLZXWs4WbxNKAWVgdqJsygJt/kTTb+vglI3o/L1/SxeaDuSnNBgCOIYN5u6u3MxQyz70pUMK0nvfKar7ftW86Lx4352HRz91/4MIPOkLdZeRX/iv5x4l/BQDojHNPirW4RpLePhcuMEDlMh0uGEJ+IU8rnqPojozVYJfv9FaG6ynsyL0qWf7Sz3XosVnznBrZENdQY5f7Y867fvAVtuknlVftuRvy/HsTP8E2x+RZrad+xbetf+hz/nGkLu9L7eYS+3aMuYg6aRmTZxyOTbbALNhTXckvFcn9OfsZ+IB/nJ8QYfRshWOSitE70q6JYLjfZNb21rZ4JTZ73MfcCseh0JX3puEqc1fhuo94DqESHUs4MsJ7Xbgl+/M9cXpZXtrkPvrIQIr9XlI+2k7GuCcvh+X6fJZepK2QjGMow/dqK8EF4+YlKCA0znxW2y2eO0hKX1bbHKelCtfJsaDsQVsOXXTjQ3oLd+JLAIDk4Nd925dP/pmcF6Db7mN5ZuSfcmXMwzHuuZEe96RrnuNudZQZ0iPXKAYPjcv+3Y0yX1YwT09tqypetESHAvTVBPOu9ZsyZ1XFWxUeivB8Ocv5KsQ495lVGZOa4s2qJrgOJp6RgI9DUw/grcJ6piwsLCwsLCwsDgD7ZcrCwsLCwsLC4gCwAnQLCwuLfaAfmcXmzO8AAPLbIkAPgXl53AXSa+WaUKtO+ZZvyyoSh4xX7qJdJ4WxU/C24y635XGXlNnQEbok0SStklD493GPnrqUoEygqOTtuc3KZx8hZX7u8yzMfXJOqKRvbLFo8Uxcnr/zyy/5tge+wH5u7wjtlP4YReXu587xmQl56IO3SKv0lVxK0aAUAR9zme9qpEoqq74lPOjlY6T0s2X2eSogovuOS79AsihUUZE6eygVQxDyhieV5H36irLBq62MwFHalneY0+7UjBw/v0PpxpEJjuPTEHu6TIH6txWBfN5bBxtd0kuHPLnGeotjeyjORvW9/FBunTRXxCV12m3K+GZDLN0yGyMN2G5Jp1XKf9Mh/bZQEcoufvEV33Z4SYr9Tk4y19ml9Z/2j0fyMubrIJ9ZUWjvRY/RS1Ypc+jOk9pcDwq1mU0wF1tljWs35An928o4zXZIl95MyfW5GOUiq0O5/2GXHG21TWqzPC9r73iJayxf5UT3HpLcWZtkM/cN65mysLCwsLCwsDgA3nYBeuANSua9xLEDg9jclBnbmN15ryzWQb27byVDc98gTnYNWdE/8pGPaLYnn/6q8Z4dV+9n25DFu5DOajYAgEGAv1vRM6Df2tzQbAAwMjGm2ZLFguFMfZz2SBQP1yBYDhuylZtE6YA5qCCf18XqJqH9cLij2QDzOssYsrJvb29pNgAIDvU+tYa6+H9lY02znTymZ8kHgHqtpNmiybxm2yv7vCkzukmUvldG/70yo+/nvFwuZzjzRxvh3jKmVv8RAMA5IR6q6jjXQLDG43RE1lssyblf2uUaLHvFeucTnJvwlsxnXQlCqIUp1k60xfvRHuE10QHfy5ueAD3X416hxjP0vNdl+xy9KO+Z4d/PbYjH6zcnP+Hbfn/3Rennl5h1fOs+emRqZyUre+MGf8qPzFC8ezYrnpKLTXocjqboZUo7D8s9g9/wbe3Qu/3jzIx4uUZ37/Nt0QyF2dGKjM/AZYqHipehPNhm5/phzk28JZ6UrjI4mR73nKY3fK4SMzSmbFWra7I/3xHm3L5QW/KP74Tsn5cCfP49afb/5kDaMqb0Y8lrX7BGb1I5Ru9Kb1IaNcywUdUy+9SIyr7nKN6ma0r/52oy+dUR/n3uFqMPNnKyb60oFQxCUUmT0I896NveU1Syso/LHlBwuB4jDe5f6xvy2bZ+mgLz4ascs/6I7L/brQu+zR2lAD1Ylf4l+7Qtx5nGYLgt62gnxrENd7wC4yOH2Y8k53b2hvS/kWc/t5V0EmNfehUAMDXBlCT7hfVMWVhYWFhYWFgcAPbLlIWFhYWFhYXFAWAF6BYWFj/ScBwnB+B3AdwJwAXwGwAuAviPAOYBLAH4Jdd1dd5VQTc8ixVPgD62+0fevf+h//fBLGUILU+w3IiSqskoMoOEK3mJqnXmTNqaFNopvkRBbyp0xD92PaZ8RMnoPBgwv9G09/gbh0jzZ8qkYwceUz6rMLTl53gcuUN+W//ZCyxme9eYCMCrH2Eun6mvkT56Pfl1AECxwHZuPkPR/fCQUNT33CI9kwwzuVa/IPKHhf67fNuIQpnVvWK6y9OUPsRWmT8pUhTKcuCQ8s6MSPuyCuPfSnFMc96p9Yz54y/tKS8GZEOxwsTkmJuRvry8Q8rr8AQlA4/flNxgmSzbfG6HQQPxvtivKQqPw5A5Xa3yvGKA9xxuyN+rERahjroch9mK0Gsdl3M/V1DykXlZ19sOqb1zBQYKnChLbqwjr5AG7NeEkhv90Dd9W/25j/nH5TmhDuuJO3xbYJoDdciThgRVgfci35Hrjpw7EWaQRXmNonrnkIxTvUY5x7xCx17Py5qccikwX43JoM4FuIaVFGW4NSfPWqyxCPThnpJs7RHJe7ZxRs+1+L3wPT1TjuPEHMf5tuM4ZxzHed1xnP/Vsy84jvO84ziXHcf5j47j6IImCwsLi+8//i8AX3Fd9wSAewCcB/BPATzhuu5RAE94/7awsLD4z8J+aL4OgA+4rnsPgHsB/LjjOA8D+BcA/qW3GZUAfOJN7mFhYWHxtsNxnAyA9wL4PQBwXbfrum4ZwM8C+LR32qcB/Jz5DhYWFhbfG9+T5nMlDOp29cKw958L4AMA/q5n/zSA/wXAv3vTmzkOgmFz+Zg3whSptF/sFb1kCMiCKVBqr+ipoKNHtLlDPRpvcnJCs6VT5mi8rfKmZtvd3dVs+azZ7Rg1RMk1DSVuKg098g34bhfqbRS7+vWBmF4+xA3uESVmGJNQTHdcmqL2ACCZ1PuazerjF4+lNNte9zTNaa+//1JCDvT16ET0Z3VcPcKvUBzRbABw48Z1zXb8Tv3ctxJxanpv9lrPJpgi98KGd3avcW40vrsU0nCP8jxvIw4D2ALwB47j3APgJQC/DWDcdd01AHBdd81xHD2s9Q2IKNF8keP/FgBQmiItEa1wnEIZsYdjpMSuK6/FhhcJuqDQT9EVoTUqDk/su1f840BNqLRGgmssNGDk7bL3OkbXyeOpRZAC3qveZNogTCsBra1XpC1/7/DDvu3T35G8Q+2rpCM3lGjCc+uSX6rZesK3PaSs4Z22cG3DNKPtphzST4GQROltB/guzChRciMzEk0YrfwibQn2KtSTzhSHvH9tU9jaVp08XSOiFABuyp7ZVcph5Xt8xypesGFUCcxNMY0Wdrw0XYeD3NPOKoWQ74rJoN6szvu2O8A+vz6U9k1FuadejMj18VEWCh4OuTZuHZLjfoaNCg74bpVHhNYK7fJdX3f52dCsCvXqRPl+HgHHrNST65rH+Nlzn0eTVjof9G2JU+SFR5KnAQCF9su+LXqLdG3ZK6+0NcaI1NBVfjZ207JOb4CUmzPOaMBUTSjsVIc03KUkaczYTeGtS1mOQ7gr1OFOirRzJM1xHF+R/lVy/DxZT3DMF7/6NABgrnAcbxX7EqA7jhN0HOdVAJsAHgdwFUDZdd3bO+8KAHMBMgsLC4vvH0IA7gfw71zXvQ9AA2+B0nMc55OO47zoOM6L2zU9ZYmFhYUFsE8Buuu6AwD3ekLOPwdw0nSa6VrHcT4J4JMAMD01bTrFwsLC4q8LKwBWXNd93vv35yBfpjYcx5n0vFKTkB+KGlzX/RSATwHA3ScecLdmPwUAGN36DAAg4Pyaf25njh6TZtnzOri8bS5A70fBEXu7zCKyG9Py2zSyxPNiEebLGcbEnu7xl3ZQKcA95sh2vjtNAXhylx6lXc+RO0INMzorSrZxT2T+zGP8Vf5IQjwJ60qB3bFdPr947E8AAJkyf923XqB3M/hR8TRMv8QvouOKRybkipsnrngCsi2K9gM7khm9vEDvxnCZXof2rHgY+krR3mJcvIHBLj14sRE+00vcjbri7FZTxmW87vX4GAzo2MKoN36vrNHzMzPFPj+34t0gR9H+pQ7XQaounsebSs7FxbZ4VNZ7vE8oynxc+S1xK7oheuNbfXrDpurykdwCxeIdxYO5mJd1cCtEL85qgm06tiODkb3AZ9bbIrSvhp70bQ9/83/0j525JQDA9cCjvm2yqHgNa+K5cxTv6eQsjzdjMj7j4BqvrLFPvbw8vz1c9m0LHXpdbxZELF8McMz6UVlvowkK0OtdfjUpTcv7Nt1Q1nNFyav4kHhlV14zswlvhreUGsHTGjwJ4GEAOcdxbn8ZmwZwa49rPuW67mnXdU8XCnriRQsLC4u/Lriuuw7gpuM4tz+tPwjgHIAvAPhVz/arAD7/fWiehYXFjwi+p2fKcZxRAD3XdcuO48QB/BhEfP51AL8A4LOwm5GFhcUPLv4RgM94EcfXAPw65IfknziO8wkAywB+8U2ut7CwsHhT7IfmmwTwacdxgvA2INd1v+g4zjkAn3Uc538H8Aq8aJk3heseSFje6+niXmNJDdfscAuEdCGtYxBRO+YKN3ANCnaTONgk2J2aNFOctbqe2sbUp+2qLkoHgHRcL8kSDsQ0W6etjx0A1CpVzWYqZ5OM6veEQZQNmAXPUcOYJBIJzQbsf0xN62Gve9brdc0Wj+uiepOgHwDa7aZm63a7mi0R1MdpfXNbswHApUuXNNuJU/cZztxjPe/Tr7xXySbT+JkE6Cbxf6ejByn8oMJ13VcBnDb86YMG254IdW9gbEUKGzvH/jUAoD9GQW+wxAnJxuUdCmUpqL2iLIM2hK6ZGeVan16RtbMT5D17AwrQ4zUpcdHK8/0Kgut9wxOTj5QUak/Jj5SIyPtaUdo5FWRZjdRXRWX9M7wcj3lsSSrDXEDNGNd95aW/DQDYet9jvu3oHXyvdjZeBwBsjyvlOTIUmy+nJNgkEmOB3bvxTv84m5I8VIXu3/Jtg3FyboWGrMNChKxHZ/NZAEC4y/uUNpl0yunKelZ3ubEOPwq3PCYtocxXVMnNtemlYlpQ9o9zm7x+KiV79dZVtulUk/Tcyx5nmApw71/Nyf6WLvNdbShlWuoTss9XRq76tpEUqaj1vNBWtTb3n1aGa2/1sgjXS3G+36N1qnVW20I9Nk6TYl3YkPaFyixu/Nidf8jrw1I26OHuX/q2sSX+JukGZKB2J1lYO3SLA9kel7bc2iSplZhim4Ne8EJ8QF76QpT7eHJD1m4lxT65HlvfdRTaOMt3LLgmL8RmRinZlONn4OG/EDXA4ohpu3hz7Cea7zUA2g7vuu41AA/qV1hYWFhYWFhY/M2BLSdjYWFhYWFhYXEA2HIyFhYWFvtALzKHjWlJpTe++ccAgMDw7/Pvs0pUUU0ooJZDiiMbSCjH63LeLqmWzUmhksOrpIdi8bv8Y9eLxIr3SAUNwJxUU569nSZtMRFUqB7vspRCWfWapK/790vE2Jm/ZJabuyeF83vuMvP75EeZO2v68L8HAEw+TtJsZ430dfpji9KnrzAXUUCJ5ju2IWOWO0QqJzDCXEbJ7ofl+taSb2usMw/V6rRH2ZVIO08ckzFrdkiH9ouM2Mp6+bbCCnvdUxj7nJc+qstuAgr7PTYv/z+3znx4xyfJpz5zU+jzRJHU4iuK8mGsLG1dV3KwzXq5+TaGvGc2wHnu1+SaRIyJwYIKe1+AjEnCZZRaRZFrjOdlTtNK5NsySOmlHJnf5DLX06C3AADYrJ31be99lnTreEbW8I3QA74tN8dGRWNCwzUH5EsPjyo0ZE4W4owi4djYIYUbToo0pgK+QwsOx2c7K23OxXnNsit0aCDJZGpd5ZqMlwNypKbQ6zU1mk9IuMtndAnI94L1TFlYWFhYWFhYHABvr2fKcbTs5CYRLGAWYZuEtCaxczhqLhMYjBgE6IbrBwM9gzcAIGB6vi6WdgwZwB955BHjLS+cf0WzDVxdBNxsmhMG5tSfmbfPbeli663NdeP1hyZHNVuz2dBs8bSebTwAs1jbJEx3DUm4kzGzWDyRMGRAz+j9jET0eQ4bBOAA4Dh6n6Jh/fp0Whf0A0C3q4+/A0NWdUMG9GvXloz3RMAwfkH9nu7A/I64QcP1hrW31y8m0/sUi+nj5xjO2yuQJPuG8QsF9p+9/Qcd4d4NFG/9FgAgcvjfAADccb5rTpljH/NyCGWT9LhcVjJZDQaynhcOcU+aWZNfyGsxrrVmm/mVir1TAIBGius2ofzCLnuvcqylZHdW4ltyaVkJdSXmJNHge129KMcfWuDzX7ki3pWPLnzctz22/bR/XPuyaPhLj1KEPHiGnrWVTfFqPJxkgdzlIveijaqso/E4PV/HB/TWXY49Lu0c/JRvS83QC1Tsigg7nadnKtQQr4TTo+jdWaN3ZLsrc+IqEvRsh+u+6m0/EWXLUJwb6HrOn5kcq0cs3+LcH8/InFzYpGvrHiUQ4MWB9HXe4dhfi4p3JRvgONxMsQGdjLzrOzkuotn4nWxzTITplSTfv50C2xS5Io2+0VdyhDlsX8eraNALMtCp05fjiEMx9lP3/wf/uOBIBvWfibFg8nTvfv84HpE2xVyO86qSQKmUEo9Ta4n7y6FxBmr1hhL9EMeCb7sCFmJOd+XcfofrPRry5iTJ3FVqQNZwV/6+o1QZaCrjvPD0BQDA0dH34a3CeqYsLCwsLCwsLA4A+2XKwsLCwsLCwuIAsAJ0CwsLi32gF57D9qHbAvRPAwAc5zf8v/dnSMe2SkL17DYp8p0E8wL1o5JrqbN1t2/bmRWKJ7BJ+qcQPeUf171SI6kAnzPIkOod87Tk9Tg5pbks6fFraaHCkkleU6E+HfMPeGVOniUtMnJCBOif/9Z/8m1Ts6QuNzNfBAAc/gY/Sp7tvOgfn8iJAL20ueTb+qOkot6xLMrv9JCi99YiiyYv9CTHUbJHUXv3OjP1rBaFtilUKMyOjskzq4qEIzDBOtYFTxfeHmE/kwoNB695Q4WxHioC9a7HIG1uk16amlGo0SWZv5FRUlJnlNtPb4v92TDprUlH6LcrASVfVZ1U1E5VqKhkmIWCnYEiIA9Io4Yu81Clypz71O1C9UkKq+ttnhttyHpd75AOjbWEbh25qJRzufph//hIXJ7/eoRBEmPzCt2aFbp3Z8ggh/cpqpJ2VPoyPsl5WFYkKoWU5DZrB/j8xTD9P9WoTFo4S9667ZXbmQxc820dkHYO5YWvLTQVKc22wuE+JNTlpdfeej5M65mysLCwsLCwsDgA7JcpCwsLCwsLC4sD4O2l+VxXi5R7Y3QfT9UjiPYbfWSK8gKAoSFSar/Plj/owzVwDeVsHD2i6vS9d2k2APgDR4/Icrv6PXt65RIAQKejPyuR1nNkmMqhAEC9VtFs3UZLszkDU9kec5uGBrtpmp2IOdIrk9IjB2EY50RS72ekpkdsAoBrsA8NIYbNpnmcBgO9UyFDNB8MJYc2SuZSQHcbIkkDbcPYR/XoRuC7cwzdRjCg37PfNUcDmqJW09mMfqKh5JKpPBAAtN9wzyH2WCQ/hAj1lpBf/QdyvPgpAMBwhJFAISVyLuSVRCokSV89q4zjoC/03swM96rJDaGitpQhK7dZTmYkcESuDXEfSihLq+oxQNG2Qu0pUx+ryF6j7mIRJUj10gVp871kxHD5vLwPf+foz/i2z1Yf94/bL90hz1GomqX2Mf+4WxfKb/LoIvuU+Y5//Hr6YQDAWJLjmOu8xz9+zcvT1RnO+rbM9AX/eLovUVeRFHk6B3L/yIDtaK5xUHc7EgU2iDFyrl9Typx4DFBWWcp15RUMe6zTZIH33Fnl8WJcRvj8Nvenox3uXxddmajDIfZ5FcI9LsR43vkIacTb+89uhtF8E3VGrDU92nhTodR2k1yQc13ZV1aVcjOj4NqrRoSey9e5XiteRGpzjPQ0Uoza7BVl7H8yyhC9cJrXJx2h33LK5+mWEtG6m5O1XVvl4CZyR9mn3qsAgJjL6M6lHp8V7QkPO71BPjYdl8+z2iRrIsWGnNv4rvSznGCfWnmunckvyto5PsaoxP3CeqYsLCwsLCwsLA4AK0C3sLCw2Af64TnsHBKP1NjGZwAAIecT/t870xTCBsvya/lW87JvG3f5C3oQ9jKgl/iru+ZlQHcVAfpY/Ih/XO+IGykZpMi3V6CnIeeZ6wF6N2eC9I6UCiLujSq5pbaVVG+LnlOhdo0em/qPyS/9bzxO78J94/zVv3m/FNCdP0M3zneKX/GPp9ri/dh9lcJm51F6VOZWxTszqnhfExEKq9NB8VSkWkp+oW3mANqdXgIAuEsU6kdmxaMTbLPvTpb3T5ekL10ld92I4oXqeE4LV7Gp/tqg52TbqNELXJzls17w0mgVRzhPT7foyR2DzPMNJZFVNOSJ/wP0Rs12mI9r13NXpvr8e19JGJaLSaMi/VXfdmiHz3Q8EXerQAV4tkIBej8mQQW34lyvC7viojz22ppvC5RZjvfw/ZJD7PIV5gCLnmKf6kFZUIMoPWRK7AHq6XkAQHKac7PU4DqZTYj3tt5iYey5EBdsOyxrolak+3WYlFxto70V5TzOjTsu3tlcl9npc1tqBnR55uUzb93PZD1TFhYWFhYWFhYHgP0yZWFhYWFhYWFxALytNJ/r6qJXUzkXwCyONYnNg0FdxLyXgHwvXfkbsZcofmAQHDsDg4jYUE4lkTSXOZmdntNsV65d1myDnl5iBgAq1apmGxrU6tGQWezd7uqC52qlpNkK9aJmS0TMxSAdgwjaoB9HLGG+PpnUBdfJuF56xlT6ZadsLnHT7evjFwnpc9LbY5zDBsG1Y1hPJsF1cw8BeCKl93O3sqnZ8qOzmk0aoM9pzzVEKuwRvRA0vHv7Lbm01zv2xvdxr/f7hxGh7g0UbooAHcf+HwBAJ0taI97gOA9CQpUdSrAUxgthbrc9CIVyKM+9ZmRD1q5SIQa7TZaTyQZkr+goazlJRg5ljwEKNvheXVS0w9NNoffUalHpMNfm+hWZ+7tipL9ufl2OP0xmDl9+7Vv+caohuYCaDttUXyIVdfP0l+XZx7nWd/uv+8e3xn8MAFArMi9Qy+W5TuLbAIA7B+/0bROJr/vHh/s/DgAIpLgPRjsypvEhqZ6W8lr1e1KGpbpB+srtkbq8XUc6rexZNaUocsqbkoLCDlXXOPd3hqT959vjio0NuNYS+mtxwH32RlCov8NB5ll6Psb1kh2Vz8NajJPn5Cj2rqbPAwAaIbajq8Tx5ANyXbXNfXRYpwC9N1gCAEw0uNfUgvLM1xd5z/ku576En5Q2z3KNtmPMg5XvCz03rCnUXJWfN7tlKftTXudnSzyqBCpUJHghEDvh21Y6nOdYXyhLpZ40AnXpZ3meYx/o8TMhtC3j3FJo5VaKY55/UoIb5qwA3cLCwsLCwsLi7YX9MmVhYWFhYWFhcQDYaD4LCwuLfaAfmcPmjFdOxovmi+OT/t/rM6RDwlWhNm50SNnnwQROTkjy5bjbzIVUHRNaxdlQosRSlAHUO0JRJBQmu6PQT1mPniorafYOK1RVLy7XFxzSLjd7pHUnj8v/d1qkbQb3Cq3y0ueYJ++dY6TEzqU3pE0TpGrmD5GGmz8jNOKty0u+LXycVFLyVcnrk46yU0cClHh0B3JuLsrr0xuP+sedBclVVNrkOIYXhXJsKXRn4Tj5zuGm+BCiSlRiXqFLux7L6Sg5uOKK28HxmNuWwp5n5jjor98U+cBklvzTk03W7RlJS19ezjInVAjSgEsDfiQfCXLySmVpQCrN+7gVhe6NSv/bfT5zUGGbQo5EFnayXFujfXZqKyJrs5QgjVjclrUxfZHPcaqMLg28/y/knt/+P3wbTjHyrtwXqiw8ybkNKrkBAynJ9zU3RbrybIl5DxeTEqHptphbKqdIfXoJifrsRRhr2c3ICzFWY0ToDrjGexOy9lMNUrzxEssj4aF7AADLZ81ykTeD9UxZWFhYWFhYWBwAb69nytEFqb2eWZxrEvyabPvNlBAcpL8AACAASURBVA4AgYAu2DUJZE3i970QNGRVHxpE8b2u+Z7veOhBzWYSoO+VSdo0fq4hg3iprgvVAWB1bVWzTYzrgufRel2zRZQ8Ld/1/LDef5P8PWDKIA4gZWi/ae4TMV3A7hifBAQNWbwHA33sQiHzK9Hp6ML0nkHUPgjpzwn1zHNXbbc125Xr1zTbQ6PzxuuHprVvyFTf6+8hgE/o4+z29XXaN6w903wA+ruzZzWBH0KEussYvflfAwCcI/8GANDM85f8RJ3roRcRoe1cnHmkXlaWRtsrejw7RmPE+7FcUjxLO8ov6JGg/IJu9OjRUBKwo+LFYwSVV/2G8opO9eVX+bKSNX0kw7WxuSRzOhVUvGEvyP/fd4Tz+uJ3+ND3Zz8AAPjOFr0HO8884h9vvf9pAEChx1//5QrPzeTl+vICxeJXyhyTdEb2p26LnrHNkb/wj081HgUA5HLMT5Rpyx4QiDE3Va9GsXfT25/r67Q5XSrsB654cUYU4XJTSTSV9KYkoRSJrq7STXU0Le/V1W16607Fuc+urolIe27Ifl5y5dy5OPfZ18NKwemstLUV5p6xO8J3q5aQ+wcV/8gwwf5HPM9UQ/HSuNu8VzMnaypVoaq+4+W8ujXDfFczrRf840ZX8kttHGVG/E78Uf/4joYED+Sqd/KZiui9UxMP6PJN7tmZGPNgrVSeBQAkE3yHAh3m7sq1xbPVVbbRQEfGaWOak+NAEaCvieerHeHYVtIUxY8//jIA4EiBxbT3C+uZsrCwsLCwsLA4AOyXKQsLCwsLCwuLA8AK0C0sLCz2gX54FltTQu+N7/x7AEDK+S/9v5fmSBc4XmHtSpe0bSbEvGhjjtANw61531YfF5owuEb6O58hPVbpyzXpBLftMlkPX4BeUUTpM8rf3bC073Ce9Pgth7RPbkb+31CY2am7JT/Syn+ieH7qFCmUP+h/FQBwaJbU4MJHWTj2xJPyrAtqWZ1pisGbz74GAGiPk0c7vcYOOC1Re7dnz/i2w9UP+8fBGaGKKtcpTbi5INRlbpPjnZpUChm3pc+RU8wBVlDyULVGhBZSUjoho6SkC3uXNRVGPb9I+uqCp30eG+MNHr/JcR6ZEPu2cn2hKjThax3OzXyf1OFOQ65PKvkKB4povuhIXqV+gLRwv03qMjCQcjJNJTfdlLIONiPS550M/z6yIzRh4RJp5ZbD/E27NZmTwy/+M982Mc4SNduOULg4zPleS3Fuq1FZ2yenGRBxocR3aDIuovp+nxRwWCknEyxInxIBrs1wRJ4106YQfhOUMwwmhEZMNkiBxqqKXOSdQidfe+2tyxOsZ8rCwsLCwsLC4gB4ez1TrqsJVPf6NheJ7S800SQgHxpE4QAAg7A8YMgMvlcGdJgyQRsyoIeciGYLBMxD/fA7HtZsf/7nf6bZejBnse4P9T4NDOnGW3sI/Uu1mm6r7Gq2Rl0/L7dHZu+ASYBu0oXvkZXdJAIfK+oZ2JeWdQH6oGvOYB6L6XPSaTQ1215zHzCsVCdkyH5vWA9u0Dz3Wzvbmm11aVmzBd9pHqdhX//15Az1uR/sEbzgRPQ+uYbs9eE92v83DaHeMkZXRYAeOiopElqj9D5MNjhOw4iMYyhB78CL/DGMjiPreXGC453bkOsragb01oZ/nA6IV6DT4K/3lPL6e84LpJRlVaIjAcWhvC9XFIF6USksUPNeh6zizbr0svzqP6XEpCxd4rv0WxOSBfuzO6/5tuYLDJ8/e6c05sYLzGLdW6WX6tCiiItfHKNH49US16XnvEChTs/X86lv+8cn2+K9yE0wq3q+9i4AQCRHd1Pf4aBEhpIDorLKCVkd8P5OSdIL9HsUQ7eVObk9vmFFlN5kJD4OjYuIfOsKPXinCxz062uSCuBYnV6klzriPRkP8/29lqA3a84Tm69FuL9tZfhed3LivQlvcA2GkhznZkQmtZni2G4ucU+vtr1CykN6joIh6WB5gotsXEk54Lh/FwBw5vinfduFzEP+8Ud68vypTfZjNMLUB8u7Imy/uMaB7Ixzba3uLgEAMgmunS6UPbsj86M4EOEGRIC+OcWFHx5wkcd35Pp6gLZOnAu+8Jis44URPTDse8F6piwsLCwsLCwsDgD7ZcrCwsLCwsLC4gCw/nsLCwuLfaAfmUVp5ncAAMXtPwIARAKf8P9eXSAV3/fUyYEOhbBFpYBwYiBU0LA049sqk3L9YIWc0miGdEW5I4rjEZVeUgTLaY8xbCgJnccUdUDbo/8Xw/wNvRNnkp5kUMTNJep5ccQTWw9eIqVffe+Sf/zVs0JfHZ+lmHvn0Sf84/lnRQTeH3/Wtw3T7MDS14XeC45RJDxTVjJ7r0reo3iez5xoPOAfR5tC5cXWPuDbriwKfTR+jUVzY1Okr7Z7Mg6FUbY5oygbnBGhj7JKvfeGIkAPSHqj7yLPM0yFhCsr0pfsBKnFZ7bZ5/ykUPnnlfxHh/7/9r40SLLqOvO7uS+VmZVZlbVXV1XvO3TTNI0EEkJIRgZJGGuzVsuS8Yw3jWfGlj0xM9aM7XF4wuFlHJZkbSPCsoQYJFkIawXRCCRo6G6g6X2pfd+33Jc7P87rdy68V1CtEtVUxfkiCG6ffPneXc699fJ8Z5miPj1luGg0G/mThq2ogpo4z039NPcgXKJFyxWZ2iwVmB6LaaI0w4YbRMrHuuerISpuNsB5riLjpFDZCXYAz6o2u52foez1+4/exfeM8bU9YYsm3cZzX4ry90cbiK6+vci5rfpLRuHwGH2/oHr5/l4jOCFKwQchH3PQfg85oDcWz9my8SpnWM+laW1qppnirZk3XEiup/xS3Wed7hKvBLFMCQQCgUAgEKwA8jIlEAgEAoFAsAKsMs2nHNFS3qAzygoA4BIRpl3Kj1xJuYoritxzga64la5xXucWYbhUNF9rW4tDlkwmHLLRqVGHDAACXuf8BQJOma46S5cAwPy8s8zMYs5ZOsbtusYlovmCNSGHrFpdftkft/5HwlGnLOKUuX0XABIJ55xO5pxzEgo5+w4A+WzOKXQJGnWLRFxqnGNjzjUdiUUcMqXc5xkulmhdcEZ9+l30cSm4Rf551lFJmJXAX+xD/dBvU3vLZwEA2oiWq100zpfL5SoiTCX1cBUVjHqJjugyvl83Sro7YdxmLsu0TUgRlZOf5+iqoLG0i1Y7bKjVpFHyJKnprDtnVAJqMPirsqXiMWP7D/bS/zekWdk6j+y32x+0ysx8a+5ZW5YdbbLbIyn6fOAM779Ekq9t2EnUZybKBXr7YkzLtNfR98e9zKOdCz1nt3cWdgEA6up/wrLZOwEA/jhHq+ki799EmajHqVGOAIxgp93OTFHEWbHAtGzFWJO4NT8FoxqTj9ktNFjdn7rANN7uEJ81F/JEze6d5f49VaJ93+jjyLK5KD9/Y4L04HSEabpCPd9zrI44ydpF/ntSCp+329NBOpfmY3w+eEv8rN55ancaxbizUTrzG6sc1VjU3Odp714AwNGD99uyYPwau/2eMCl86xxT0e0+pj7Hx2giz88wx7pYz3rmt/Jc1dTssmVTEU6uFcnQWP3gecgpognn2nkePMZChaeo//kqn7MLUd5P6ceonMym+AFcKcQyJRAIBAKBQLACiAO6QCAQLAPlQAemW8kilRz/CgDA6/1N+/NCB5svCjkrm3meHV1TYf6FHgf96i/OshPwTAtZIKv9Rgb0BFsK5gr0qzplJJeanmLrRcIyyi6yYQcpIz1c0TIwthnWLGXk/fGEyXpkWsaaLvvmDrK5a+R2zup+5LGbAAAHDCvxhUa2NNQWyPTWduhBWxbpM6wC58mkU93J5rL6DOfWqs1RB5JVtiI1ldnqoCLk4O/v5+LK2W1kEfFd2mbLApvZbtBfIGtca4zn1mvk5vJ00jykDQudNqxQZWuoJtHhY8MWpi1jYn0HW/0eHWKLeVMtrfOTbbx2raNkLRsxAgIimtemt0y60+hhy83MrDFnmiwy81XOXeUFj7+tTJa/7DR3ui7AzypYOauCHnYAr06R9XSw9gV+TpmVK9F/FgDQeZZzJW55C+v76TjNY2QLO4jP+NhCOZ44DQB4Q4bNp/15tsRmozTR5Qjn3uus8JoVaqm4czjG1t+ItWax3FlbNlZlp/fFJM1zfJajC0KLRr7CA+SAfuGMe77Cl4NYpgQCgUAgEAhWAHmZEggEAoFAIFgBVpXmU0rB7/e/SOZ3rTMCVFwcluFS6sINbo7mS6HqUn5j6WtdnKhdvu5Tzucv5RjtdRl/V1eXQzY2Pe6QXQmqVff35kUXJ+zR8TGHbHP7Zocsn3d3avdXaxwy5eLY7PW69ykQcJYSisWcDuQBv9NZPBx2lpgBgIrHpeyOS3mhpRzQXR3LXb7vpk+BJX6zLMxlHDK34IXSEqWAymXnteWyy7VLOKC79cqvlxfk4aa3gPucrhf4Sv1oGLYc0Lf9EwBg3ogfCWZ47P4g6aGnhp2xZ4f42skA6fMmw0E8PUZn45ihglMG5RUNEGU2l+E1jhqLmLH8iWNGfqQpI0bjsu9xX61RHqTMe61sOcinja02bDlWtwSYXkk/zKVX7rRSOT1xgXUk7dlrt30lcvyePHa7LUtcd9Ru1+ykHEOlojHOenZi1imip87k2VP/Ug2XNOnMXgsACKUf5TFNUt6jUj3fs67AEx1RRENmZjmopq7CZ25mhpycB3PG4hoBPPFFWiDF9XsBI7ggbdG1Cxd5jxyIcNHlc1O0UNcucmmUp6z7JxRTfzrMCcP2RInye9rH58tCmqnXYoT43FCWv+8Lc4mewRCdX/kE6051lBVlyKLHfHnWh/kEnU8dszwP+aBBl6aJ8suneD3nOnfb7TusEja1PqaFU0W+f9EqanzWKF+WbeFAhOIwrU8c19qykTD32V8iCjg8yHysx/o7M7vBoIJL/PcovkDzmPfwe0gmxvuh7mEqJ7O57npcKcQyJRAIBAKBQLACyMuUQCAQCAQCwQog0XwCgUCwDJT9HRhvI3ovPfbPAAAd+A3+vJUpjPwC0aUqy1RMMsCcW42faPv8JJc8mW8hWkT3GRFXKaa3pjNEPzUY9NKLovms288ZDGzCSDtWtE77xhJfEC1yrqHLlUL6FFN2iS00Du8s00P5/UzbnP0B5QDak+IIvcfrmbZJ5Yne6ujo5TH9lOmtsbNE6/ivY3p58elj3Kc6orp2TTMlXlZMOYaDFGnWNMW5r6YbqX/VC0zV9G5lWqemQu3N4PxJWYPeqkkSRdRmMObzceZeg9ZQM8ZfzwAvIwoZun9sM0/+C71cMqU9RRF3P1jgOdugqC8zXqYw80WmtM5poo3rYETbTbPrSL2icjWVjJFzyb/VbqfK9MzqPI855eXnN1vUbk2WwxpLiqjR0RTTpeEcU2Ydp2ieA5NM7e1uY2rxTPaXAQChzVxKZyRi6LNVDuatAaZgR2eYxhxtpLGW1QV+pm+L3c5adK8/zS4o3hzpdqp0mu9TYFo6k6K+RGeYTgzNGrz6jTSWC2clmk8gEAgEAoFgVbHKDuhOx1U3h9uXky/nuivKau7iXLtUxmqfz+lEXnVJQ61cnJCXOx4AOHDgoEP25NGnXK8tumQhz2edWbDjcadTOADkF52ZzacnZ5yy6WmHrHOJDOhet9TgLg7obg79gLuzfjweX5ZsKQf0fMXpLO+2JpmM0ymcsLws4Lrq/EVTrS6xzTzOe9bUONepWHSuJwAoj1ume+d1obB78IPPZZ08bkEWLgEdS+nzcvf3WoS/1IfG4XsAAIGtnwcAeIyMzd4Mt6PWlKsoO6B3G8s45bEsNvwx0iNkHTEd0GcyHHgSCZJFJmvkPwob01205HFDhTNGLEfM2gKTKV7jgGGdyFlO1PX1hpXIMqyVcrzXqqc4I/WNVn6lM+dZ7++uudlu/wxkVcg8wt8pHOC9uFgkc1h27qIta2rcZ7dn02QpOWlkrI4mOJt53Tw5u0952RKxy3MjPbPtpC1Lz3Fx5GiILGOlPKct9wTYcpUdo7H0FbkArjaOv5TV/awRPBAyggvClu9zrpv3/dYUWwD7LpFp8ZCf5+G5BVr0oM+YZ6Oob7uV+fx0kPvpS7G+Za1jbzLMZ7I3ynOyECJFKcWMvGKjbDUdz9D8Vozs+p4IWaTiY2wJ9BkO6Bf30FnVNc/zPBG7wW7vjdPaBQPs3N9p5Au7MErP7M7y2o9u4gzn3mGas2iMs5EPVdlK5p2j+4YmWPe0ok0w3crWKH+VLcLRCdocC0H+O1GJ8ZhqD1MG9J0JHsdyIZYpgUAgEAgEghVAXqYEAoFAIBAIVgBxQBcIBOsaSqk/APBxEFf7AoCPAmgGcB+AFIDjAD6ktXbnUy2U/B2YaCUH9IbJrwEAPJ6P2J+XO5lzK80RVaYNmi4eYqoq7CNOrTjGTsIZy+PZ08tUTkOaKZapeaIw0mmmcqYn2Bs9ZvmiTxtMbY2R1q9ssUKpHNN4yih2W5ciTnDAYM8vkyFl9hlH4np2Axh+gp5fs4MppwdPcEmTDRvI4bn9Lcw9po8wTTc8RzTgQpLnaW7IoKfaiAfdNcqDioC50SyeBgA0e5gaDGTJMdk3xFTNaBdTaqFZsiFsrfBAiwtMrwd205/FZoO6KxmU3mUmrWzMrTYc0C9rUWgH2yqGephWqq+j8T+yyGtX10JrMjLL81gqMCV3fpGoqqSfabAJo1SQ1+pL0igxAy+XUanJEU+Zn2HKv8ZwCUjXE6XZXGE+cyRM85BLMI0Wn2Ld63ienLh9izz46P7Ddru3x3JA32rksyqwA3llG61JQ4EdyNU4685CmnRiXp+xZbu8G/nzJJWZKRrjDBSJGo2Xztmy+RJTftk0zWNsmh3QkTHcIK4nF5vT51/2KHDFsi1TSimvUupZpdRD1r+7lFJHlFIXlFJfV0q5O2YIBALBVYJSqhXA7wM4oLXeDcAL4H0A/grA32qttwCYAfCxq9dLgUCw1nElNN8nAJwx/i2HkUAgWAvwAQgrpXwAIgBGANwK4AHr83sB3HWV+iYQCNYBlkXzKaXaANwB4C8A/EdFITq3Ani/dcm9AD4F4DMvdx8NZ6ScWiLybqVResuF23OWikCqupTKUC7RfBUXmS/sd8gAACXnPXfu3O6QJSNJhwwAsiVn9Fko4lxW7xIrXVZuUXbOiLSyS0TZTGbKIQOA2mqdQ+b1OMevXKLZAMDrd66zx1lhBtEaZ+Se1+s+z25lXvxB5/d13j2ar+J16oTbPd3KqURcou4AYLHiMqf5nEM20Hfe9fst7VsdsqpLmRfl8hwA0Mo5/qJy9t/vcd5zuWWYlhcD+epBaz2klPprAP0AcgB+COAYgFmt9WVFHwTQ+kr38pd60ThA0Xz+7V8AAFQbeW49OZ67iLL0MMo03UA/32ssTGt3LafdgW+SvlM0VHggz3us0U/7as5QkTpjGeYtxq7RiPabMo6NZku1h+O8Ks3gjTVkfT/JTBIK1rNChmz6HN90V7t1z/M8zj+sZdrmK9PfBwBUXmB6Z3gTc4bdBdKt6gyfOV2buBzNUaukyYjX0MswT0CsSvTeSX+vLdvrvQkAUG5kWfPsTfz9BN3LU+D8RdUA67jHCi4bCW6wZcEJnuiiFZ1bZNYWAWbkELIYw0mjAthGTgeG3gvEGd5kTOrxQYueMvI8zfk4mq/DKrF1yvg70pDgSOxFP+3lqSTTtqUQK8JUgOZcx1mf2gb5XjM5movgonEmVGju/fM7bJHXz+fgeYs921g4bstmKxyJnqp/HAAw7WVZs+cw3/800YMLxtzPb+c8W7lhiniNedtt2ckYb6LEBOUgq6uw7kSs7ZjbxBGEVSMKPzJPC5X1cHmfDDeRfvg5AMDWOEd/LhfLfTv5OwB/BNhvCXX4OQ4jgUAgWE0opZIA3gmgC0ALgCiAt7lc6vrep5S6Ryl1VCl1dHK+4HaJQCAQvLJlSil1J4BxrfUxpdQtl8Uuly55GAG4BwDaWtvdLhEIBIJXC7cB6NFaTwCAUuqbAF4HoFYp5bN+ELbhRaVqGVrrzwH4HABcs2O/nuggB/TWmfvpAs8H7Wu9XfwLeW7ROiJznMsoGGMv5o2avJsnptlik2uin9Wlfv4l3RJmK+/CLFkKEim2ns4uskUoZhkVRoxUZUYKHcxY/tZxwwE9W2LLWoufnjsbZytQ2CraPm1Yhuu2cXvaShvUcyM7/B45zn3eYJ35Pdf9xJZ5z/fY7WqKLAXDIfamPnuGnZCL7eSkHC9wrqB0ni1j+fLzAIDGxB5bVlukzNyZUXZM7tvITtCxQTJFeLt47gqazUyt20leP8HjLLXzREat6S8ZRaT9hnVj2pq+hGGNOsdGMERaaaw/6OVJTTZR/+aNbNyBElup+rM0lrosm8AGC4bFxcr/lMixNctnWHxqSqSbeoad7kt+nvNUgtZp8zz/GT8TIcuVNvQpP8Frt7Wb7pU1iiur/ewJpLt/le5tBBdcSLIDecwya7YYjvjBUR6Tt5asVNNezkN1MMdrVqqj+cmA13ne8o9P5wwdq3DR47kIrXNimsceKRvMwQHKpH/y4pX/cFqOZer1AN6hlOoFRb/cCrJU1Vo+CMArHEZa6wNa6wN1dU76RyAQCF5F9AM4pJSKWO4JbwZwGsCjAN5lXfMRAN++Sv0TCATrAK/4MqW1/hOtdZvWuhMUBfNjrfUHIIeRQCB4jUNrfQTkaH4clBbBA7I0fRLk/3kR5LbwxavWSYFAsOaxkjxTnwRwn1LqzwE8i2UcRgrKUW5ipbiSchXLvXapcjJQbmVinNe6yZb0nXeRp+vqHbKG+pTzQgCXBp3lYBYXFhwyzxWU7cm6OEFPTDmdzVsWFh0yACjmnSbSSNTpQa7dap8AcKsyEww6v59IJByySNhZYgYAcsN9DpnyOic/Go06ZAAwO+8sp+M2p1UX3SmV3MvuIOy8dtGlvM+l7gsOGQA0t3c5hU7/cZQr7vNcdikH5Pc757lScgYkeDzLPDqW2kurCK31nwL405eIuwE46za9DLzFftQO/C4AoLr7b0hoOHOjyPRbjUVLeUNMQYwya4OzMaL3rjEcxEvDRDeY8R+jc/ylJOhcWJgxymMY1+Yt1UkaKpQ31Dlq+TtnjGeGc0xxDFlqUj/JZ3TBGlK9saWHznK71mJ4rnuYczod2MFK+MAJyhlVnWJn7pogFyUe6iYKxtvyWVtWt40dnnsS/wYAmFP8naKPz7dMLc1JL7j48oEylR8pN3G+qnCGn+9J0qR5izy3kQrzdAsDNOiAj91SPEN8JuZ8dG3JmOeccWTWWExd0fh8S4z1pK+fHKffGmJn6hNDnQCA7RNMaT1dywEm/hqiYyfT7DS+s5vvORIhanS4hiMaSnF25q5M0jwFokx/VWaYvxvOUV8CFb7/nBV90JjjvR5WnLOpO0XnR1ecxzG6eKfdLjaRfWWylQMKDuQ4mKbhAilcPMHrOZLm5F6VUTrfvEGmEZ8K8LM6Bij4IBlgqjpiBXQttvDaaT9TuDULRJMuBlhHR4L8+ZZHyZl+d+BaXCmu6GVKa30YwGGrfcWHkUAgEAgEAsF6g5STEQgEAoFAIFgBpJyMQCAQLAOVQAcW2j4NAIj2fQsA4Gl6H3++iWnnssXATBY4mi8UYwp6Y5A4t0KGM8ostBP9FLvI3F29QfkXxynGJ1THVGx5kXm8SJL4pTmD6vUYTLjP+ukcMzwt/Ibrgi9KFwSqTGmH0+ReMMnVWFDPAVkIX6D8RcfvPmbL+p642W5vbqOHPdfXa8sys/xnp+naBwEAlX7u1Mzzp+z23H6icGrA9E+DEZ1VHDoJAPDWddqyWJj6kh3gfFfzW3lMPotSq9vCOZW6jfxwOwNEL+kC99PTwjRg0qL0ckaeKb/BpM9arJWPu4mhs+wS4NtNffnpybQtq03TOPo28zOTcxzVOJYlbjYxw9Tk6SwvblLtBgBEi0wTRsaNsjtZouQKfqYBPV6jxI+f9KzRyKdXsHID6lrWsblpjsxrmx+l8c7usmX1B39qt5vP/CEAoCnEbgrn6pj2XrieIvPiFY7QAwddoraJ+pz1cmzb1kybcQHdt5Qwovmscjmx4Kgt82ZYYfMWNVkzxdTglrLhwrGX6PcXzr460XwCgUAgEAgEgiWwqpYppZxZzNWr4KB6JU7pSzqb/4KffyVZ1WtqahyyhoYGhwwALvR3O2TlstNh2KuW/9686OKAnqs6nZWzSzigF3LOjNshF2drfQWv8m7Z792cxeNxdwd0t9RoCTMRjIXzF4YcMsBdT7RLajW3fi6lY8olGKPg4uw9PDTqkAFAyHCcvIxc3sVZfAkV97nMiafiMk6X7O9LwbG/r2AvvtbhL/aiaZiqZs1d+9cAgKDifeENsvmmmiKrR6OPfzWfM7b6SJH09LoN/At4cZAsTpEY33NgnK0HXTVk6hg3YkHaDA/0y/Imzf24aOQtai3Ssy4ZwRybvGxSOeGhtdprZEXvsbZDVwM/56kX+M/GgVqyTCW/ydaom17HOnTfURrnRs2OywNRzvw9c5SyYOevv2TLGnO8L0NlOkviFa4KMeblDOrFFF07YtwTRbL4ZMLsbL05yw7JxTiZjsIzPE8BD1txPPN0ppbDvHaFEXaSHvPSmJpLbNmaNWJ+kpYRa46HhL2tfMGJi3TfX02xBe5Hs5Q2aG+Jv/QdP69dOkI6MRHlfh5I8Tn9dICiAioVnpuLLTz+4Dg5vWf97MDdkGMr1USZzvKkYYEcXqR/dHnZupqpsjVszkdzHwtyjrFShi21T3dQtSbVyLnU7lb8NyN9gRQ20cTjmOjkQKHpaQpqiEbZk/+n4O9vKZAF0TfBf286K3RWTpbYGpU2/pwuWI740TSbwE4u8h7cfZic9nd5DuBKIZYpgUAgEAgEghVAXqYEAoFAIBAIVgBxQBcIBIJloBTswlDXVwAAzRf/HwCgp/3dXgyUiwAAG+VJREFU9ucRI0ePx2JT5raxE7EvxLRQm0UJTsxx/ripNqL0yr1M72xo4iM6Y1F+La3MxcwNMu3SREwORhb4+/UGEz4xQ/Rd0vgJPWdUQG+z5DkjOVWbVTpmyMg11PU6/n7mDDn8Zu9+wpb95DEuKrx9O1FET/Qzdeg18svV/9ILAADfMaZ6Zrs5P1R6D9FbfZrvv7vAeajGF8gh+7oy51xL+ombrF9kmi7nZyoncZHmvO8apowis0yZ+8Pk7N1SZlkgxe1t1jxpowqJmfLO8vVGZBPLeubYgb26j5Tj8ee4z7U7DgMALo0xHdqomAacL9GalktM052a5+CF1kb6/Nw801ddPs6ttWBRdbVZptzKPh7/Rh+tecSYpw01pM9Rxfcpe5iGC1kF1EtVdkGpBF6w2/uf/ggAIFD7jC0b8PNELeyj8furrG/5edbdzc1E15YMWve6MSMHY5gc0yNp7t/cEK3TXAM7vccv8Jirmy3e+iK7H+yOGJXFt9E6PdfjWtDlZSGWKYFAIBAIBIIVQF6mBAKBQCAQCFaAVaX5tAaq1ReXtvAtUWfFLQpo2ZF3K4zQWzL6yq1PLhFdbnCL8gKAssv7rNvzN2/d7pABwDMnjjtk+YIzR0a+6J43o+ISTVgoOZ8/6VJOZnpi0iEDAF123vOl6w4svfYuQWZwK8gSDocdsmgk5HIl4PM5I+ey2YxDViq61GOB+5qUtfNatwg5t7EDQDGfd8jKLuuUzblHTbqh4hLJGY445wlwLxPjCznH6XOJOqzo9ROlt1z4sz1ofe79AIDijf8EADDjQUPGdJatmu4J3WzLZoyAswGLrbjRYBgmxog3qsuzXvVzIBXaY0SH9E4wLZIO8kP7rO3YXGXKrDvDa7+pQrrVbUTzbc3xHnghQffdF2Bqsu8S9b+tmamWF44x5dUZpEir8QeY2rv+JqaafthH/Tu4gaPQeh7jifjhApXtqKtjfRrdw2PKWmVcbggwZ3bKqJmUidBYSiWOlrumiaiaAc1RYKkMfz6xgaLPbhzlMK9gnBfCO9NL92zl/EmeS/z9/ma6vxn5NmkcO2lrC88Y+tBuKErfMaKlbmjmxf3xaVKYjpo5W3ZxgPfd7Gaah9pJI4K3k7/fM0vn82QLr1OpypHNhRo6g/xFLjFTDPHnEwuUa+lar1H2Zpzmp7ONc1fN+Vh3Il6agESE/w7ksq+3208cpHIyySBTc2+v5XVO9VDUZOvO3basd4YjlweqnQCAmBEJPh7jSd3YTfPYv8Cfp0JE43V1M7UX5i2I8gWLGjX6fDzE+rj/3KMAgN1zt+FKIZYpgUAgEAgEghVAHNAFAoFgGSgFuzDa9VUAQE3hvwEA6qJ/Zn+ei5202+F5yglVrfAv6dJu/tW97zQVfx3aylbHUoAsAXOGdTBq5mcrjgEAUgG+py5ymu1EkKwSs372jI4aKd9yVbIqRA1DZ6mFrVwJy4Ba8fFP+TorFVE1yRaROqO+dl2RnLzb3/C0LcuOccnW9g5yqP7BdB0/ZzPnANoy/z0AQPwJI9v4FJt8ZjaRmee4jy0uyUYef+giWT1qNrF1IW9Z0d/gZafvSpA/7+yjeczv4Xt6Zp6325kmcqhuWmBn79xuoxhu0bpvir8fLxkWI8sKFTIM935jHZosf+mFWXaQT9WTta0ny1bsQPqQ3W64SLp1ysdVqudnuO2tJctdXZmrUCeKzCaMTZLlLVnH+jQ4xxbMtjjpTNEoALxhM81zTZGdxlVwggdiJbBLVPk7KsdMyY5uKk695Q38nfMFdmDfvmsPAGBUcfHjbCPn+dtbS5bFqJ/XLnKGxzy9jZzMW0tshapY+vhMkB3I35jjPTJkLV1nmHVo1yA7sI/vo0LNR3vdc/u9HMQyJRAIBAKBQLACyMuUQCAQCAQCwQqwug7o0CjqF7sSK4/f9Vrl4tftUc7uurn2Llm6xcUR+EpKv7g5a2s352KPi6N61d1RXVed77NVF9nWbTtdvx8KO0uiTGeczuKFktPZGQACwYRDVixkHTI3J+hs0d0xemZyxCFLpJIO2RJ+2fDA6fDs8wUcsoByuc7vPs9Bv/P7nZ0bHLL+gYsOGQCUK07dK5acZXe0S9keXXXX8YCLs/ds3rl2bcq9lJAqO93yvUHnOGezzvUEgHjSWXrHVcdLzoXyBtzHtJ790lVhFP6evwQA1DS8CQDQZ6S96RhnR9rLJ6snzBRFeZzpiDFFvE9HgJ2g6630Srkk61punOmIagM5cXsMpgXGEnqGiWqKG/EGc8YW9VpMlNegnErGeoWt7ZQ1clNdZmBG/Xwj7yD3eaBMhW+Lg0ztbTa21XAfnU8fruNyLo9d+JHdLqoPAwDy275ny0ZPcwerftKzvX7+/tQk63MiS30pDbNjdbKRqJ4T6vu2bM/AHXY7m6RCyPVDv2HLttTx92dGKblWew07W5+cNs4aqy7uaBvLmriiCiYaiJ7yj/PieDmdGPIWA9VhHL2D/UStbvTy5PfMcDmYUSv4IFliR/xAOyvCwiydJSHFlFbVy3OmLcf27ihzj6E5Dj6Ys/4ON3pYdv5yeaIYr302yI74G4pU7iWhnzWe+ct2O7rpuwCA4/EP2rJ/5+GiwrOTdH4ejPA59ljdNXY7do5K65zbxbRwZ4nLp40NX0d9vsTn8PFDRIe+ceB67vNeDkTomLLWpI8DIs538tqnvkTz5K0uP+jnMsQyJRAIBAKBQLACiAO6QCAQLAM60IR8658AACp5+rXd0XsrX7DvX7n97F30/wY2WSzmttrtHSmyAIyMsBOyfyc5Qc+f5V/nyTb+JZ+bICtPuI2dYxd72QxUX0+/qifzbPKIGxaRnGVUqImyZacyz47TMUXO3t4RtsjUpMmiHcuxNapoGMMzZbKadjazRWO0wtaV7SEyw3xpjK0PH6nttNv/NvYZAMDuHrbSXqqwI3+7JuvPmSm2HkTa2XqbA3nIbzIc1C+epSzc27ez5Whuutdu11vMQ26a53Y0yI7Pvimax+OJI7YsnnmP3c7Uk9Uimjcq6G7jZmiRrB8xNkSi3zD47rEsh8cNw/KeFnIcvy/H1qbbvJxv4Vut1JeNZ9ja9B0PW4yvV7SOD1XZSnQoz1ascz6y8uwcZVPqCSNdRHucrDtjBZ6zqFXcujC22ZYV40YR7iytQ8YIkoie+6ndHmuj+TkwxH16tIuv3ZMjK/qjLfx5ZvAtdju3gfQoPW5kzE+x8g1aARk4yDrsXaTUB0/Hed+Vfbw4uyyz7sJ2vs+eEXaAP7GJ9Gi7Czv0ShDLlEAgEAgEAsEKIC9TAoFAIBAIBCvAqtJ8SjmzXrs5mtMHTufiqotzq5vD67IzpV8h3DOgu1zn5jzvkkUaAOB1ZqHWLpm1N7S0OGQAkA47M36HqjGHrJhzd0DPlRccMl11OjYXKs5+Li6Rmbvg8qxS3pnZO1BT45AtBTf/fa/LnNbUOMcOAAGv02F6ft7Z/3jc6dAPAAtZ57Veb9Ahq5aca1d2DZMAyi6Z5nNZ59xH4+5jmpqZcMiitU5n9eoSARWFonNNvSGX31dukQIujuqAc++9WnvxasCbGUbtsU8BAKbvoUzoJ/s5r8+b5u+y27qOqDilOHCkwViHwWZyBG7LMc03DaL3UoaqGkm60bRItEjezwVuQ4YT80SS/pFkxgrjtay36Sm62UQ778/UFFMk412UubzhHDvn9nWSPgZO8V7z7OIO9p2n8QfAFRq6vDwn52dI/j4jJuVze4/Z7cnvEAVz/i4+x+q+zTRhdif179pFpuTGduy128k+kg/UMH3VmST6qWczT96t/Xx+lrZTZ7LD/Oevcg0HyJTOE+W1p6XTlvU8z5O6+RbiTrtf4D5F9vCcjPdQ/zMbeC9Xcuz4/YSP+tJU4Hl4MEx6kupnuvKhLdw/71HiBo/eyH2qP3LJbh+5hvZZWw/v/8FDnME9foyorP5O3rehHo5UmA1SPrCN2qBQtXU+eJliDTcbhZSHDwAAFts5p5My9CDd1QsAOK24H7fU8/lSLpFuHoqzg/mxc+N2O+KjMV0An1OpLp7TxMNEY07vYqf03DFa2223HLBlc2PP2e3RWsq4XzvDedFOdLKepH9C4/viRtbh5UIsUwKBQCAQCAQrgLxMCQQCgUAgEKwAEs0nEAgEy0A+0oKzez8FAGj8wf8CAFyTvsH+PGdQbh4QVVYxmNiJ+I/t9oGLFAU4X2dQqGEqtRHSXK+lA0zbzFoFjGPGsZ1fYPqt1arYwuQT0GlE4WWsSxvmOMTPa6QlaxykPvsMZrZjjCjmvOFNEDLGFMsTrRM0mGhvL1M9266hKL9vzDMl/lsZzr90fjPllOr8GVM545s4MnDHj6mY73NFjoRMDnD01XyMKLO2CFNR8xajVtvL1N1chWmbxeOUg6ius9eW9X3v7XZ7wx6S9zzKOZOK156y2yWrWG61hfvszfMENFhTHovymHODTM12WqzW8Dmmou5qJarqW16mXe8aZGrx/nqK4nv9T5mG+1dw2Z9DR2jOfhbj6M6G7tN2Ox4gajQ2yMWRpwOsRzVzpDVjjZyTqayJhvM2cATh/FyH3e7toog5PbrPlu1p5dxYz/+MnvmWN56xZRePcVRk68FnAABPHn2zLRuPM/3WPvBeAEA0/SQ//wLnjzrbdAIA8LZhpn0jozS5yVHeBWqeo2NrrSVZLHOE397zTBM+X6J1etNxph7/D5YHsUwJBAKBQCAQrACra5lSCh7/i52Gq0ukTHZzXHVz7HbLLK6XcLhdKdyyQ3vd/GvdMqAv4Yjrgcu1Llm0E8mUQwYAsaQzg3muMOmQBVwc1QGgBKdj9vRUxiHLzjsd1Ut5p7M0AExOO7N4t1edc+evLJECfSlnfcdlzuuSCXcH8ljM6cQ96lLMsrxEpni3ZxWLzvH7XLL06yWGuTg375DFI851Khfc53lx3pnZPJZwydTuEtAAAKWyU/fKFec4g77l76f15HD+UlRnepB94EMAgG99kjJ3/+rgmP352RgX8904RHtwrIEdo+vP7bDbX91JRVrv7GEH9FNJskjdaJiWnixyNvCbxshh+IgRi7LLiCF43Fq6W7hL+Ekrr8frJmgdn2jnL91yltf7p9aP/uuP8udPb6Pz4SAbfnDaKHTcbvX1MUNFGpq4QPCpEbKUvHmSdfBfWjhj9c4+Kob7mJG/qHmA9+r9b3gKAPDuB+ds2cmtN9ntNz5LOaW62znfV8Mk5S0abOGONk+zU/vZnTSnLQNcSLjnLZyXKHGRrIFjhzj/0Z5jbEU6cYjOws3PcD/PvsFuYpPlj/2UkZ1+s7GFH7N8uDf62XH76Xk6t9rBnvpPNbFFKPkU5d76YRfLvOf42u9up3Xe9gxbo3INb7PbkTOPAgBmGtlqWOljB/jhJJlsmk7x+TG0gwbgPcVzn93FfT5wnDKQn93COtqe579T3btpDwRm2XI12MkO7M391Je+LWyx3X/hbrvdv4/yriWe5L9Rp/ZxQMXucbJIPbyblXN3G83Pz4Lcj11GlYLPjdCZ+dauJ2zZ10f4nncM0Pn7yI3stL5ciGVKIBAIBAKBYAWQlymBQCAQCASCFUAc0AUCgWAZCMZasfnNfw4AiD35CQBAy21cQmbmMBfwbXoPlcVoGjVyKh3otNt3PU50cuRmzvWze4bodd9BpqQOZNhDXFlsySGunfyiMiY3X2ZQ9rPspgnm3zwWjXfTlHHss/88Dlm+w96DTKtcdzmVEbNcsBjKFz1/F6c8QuMmdoKugCi//jjL3nuaqaov3PQDAMDdn+VcROfbHrLbH3icqKb5jUxXtvYyjThaS5xnQ7HXlnVvJFqqUX/Dlk1su9FuN03QQF+4gfmfhhNM+ffcTpRh6gLT8APv5nnccphorcXbbRG2H+d2yZrTg4YXgceogb3LYueD1zBfu3uQqNWBnexUfsMjvE7P7CE69h0/YWrxaA07q3d0k86c8zPV3DLDitKjaB43DbG7xohhSmleoH/MppiP3JSjOY+3cBBEusAuIJM7KCdUbZYpUmV4nSTO0jr03M00YbqPacbS26n/ifPcpxOHeG3fNEgKfeY3mCJ+00Ps4D5xJ5USuu0wO/qPWmndrjE8VaLMpOOeMXpmcepaW/beUZ6Ib3jJGb3+m/+CK4VYpgQCgUAgEAhWAHmZEggEAoFAIFgBVpfmq1agsy+OFPOEoq6XKrcoP49blJczUqnqVv4CgMezvHdHt7IxS8ndLnUtMXMF9/S6XFpdou5OU1urQ3a694xDlgq4R8hFXB5WDDsj/LIF5zxPu0SjAcDi/JxDVsg4owFDkbBDBgA+l1JCHpd+hl0i3wIhZ4kXAKitc0ZDqkHnOD1m4p1X6BP8LiWPSi6656q3QNXjfH7VRUcrHve1HxoZcMia2jY5ZP4l5iSXdUZt1tQmHbKySziiH+5jWu4eW4uYymRx75PE54xsfhcAIPRNjp56p58r3v/aP9wPAPjyEOfV+a+38X75lSdoPx3J89y+91nSvU9zKh2863GOnvr+jUQLffAx1od7b+V98eEfkvyrtxiyh/leX7AC3j76Y45e+vybOQ/Vxw7THv3KTawvH3+U7vXlX+HI0Y/dx1zOp3+dwvk+9DneX//lXY/Z7Q988Y0AgK/dfNiWvedh3rfHmuj8uv3EUVv23SJHPe4o0r1eGOJcRFsmn7fbs5NEQe2e5X077qfvbxhlavHZfVvs9s4z1O4tcO6q9CjzcI/k6V4fvsQLcd8E009/8Dz1+TOLXPrkk99lKuwvf5P4vU/cy2V5/veHOM/V736e1vHP3vGULbvn8xSZ9uW3cJ6lO77LZ+53thANeMMJjkJ7LMHrfCBDOcqmZ3hMzYoj5xYXaU3bjQjqkSL/7Wiv0LUjiuds4zit+ZMhDktsmeH2bISiIbfNMm37TD3TqR0TpBuDT3C+LjXNfxuOWPXhfq3nZlv2jW4+Vzaepjn950scYd16lJ/1JT/p8W89xPr0UJFCWT92f6Mtu+/f203c+bdE4z348UdsWfNjHLl35Aj9TdoxwDq4XKzfk08gEAgEAoFgFSAO6AKBQLAM+MoVpCfJuvT2RygN+K4d7Hz7fJrzGt3RQDmABnZxDp6PPnrYbt8QvIU+v/QFW5ZuezcAYO9z37Fljdvfb7djF/8KAOC//o9sWf3Zf+DP9/8eACDZ+zlbFrzuHr5X/z8CAEL7f8eWtQ1+0W77934MAJAa+0dbNn/g1wEAzSc/b8smbvsVu50681kAwOIv/Xdb1nqKnXcz7ycrTOpsny0buI2tRHrm6wCA529jS0BkmC025TLNd23hK7bME+E8VQMFWodwA98z1UeZwwONRjbwS2zNakqR9cHTz6nc0wH22t94hnI6hevZmXv7qe/Z7UDb7wIA9h3/Z1tWPPhbdvuOZ74FAPDf9AFb9uZzX7XbldsoeGHb2GEehzVP7cNHbNn0O9ky1jRGJsaJ/SyLT/zQbo97yIo0VGQrTjXMVp6+DFmBdMCwclfYlnKZzIlm2cG9bDEHWyfYehmO8DpVF2ltqhFmHdrOnbTbLXVkeS8N/b0t2+LlPFKhY5T7Ktn1e7as48QDdrt2528DAG4e5TxUta/jPIK3PHkfAMDf+VZ+5vPfBgCcvvXDtqz42KfsdvCjfwkAeP4BljW8nnX3Gz/6NADghg1sdQQP6WUhlimBQCAQCASCFUBepgQCgUAgEAhWALWa5R+UUhMA+gDUA3DWPFnbkDGtDciYVhcdWuv01e7ELwLr+Pxab+MBZExrBWthTMs6w1b1Zcp+qFJHtdYHXvnKtQMZ09qAjEmwUqy3+V5v4wFkTGsF62lMQvMJBAKBQCAQrADyMiUQCAQCgUCwAlytl6nPvfIlaw4yprUBGZNgpVhv873exgPImNYK1s2YrorPlEAgEAgEAsF6gdB8AoFAIBAIBCvAqr9MKaVuV0qdU0pdVEr98Wo//xcBpdSXlFLjSqmThiyllPqRUuqC9X9nkbPXMJRS7UqpR5VSZ5RSp5RSn7Dka3JcSqmQUupppdTz1nj+hyXvUkodscbzdaWUezG+1zCUUl6l1LNKqYesf6/5Ma0FrJOza13t88tYb3tCKVWrlHpAKXXWWqsb18Ea/YGlcyeVUl+zzug1vU4mVvVlSinlBfCPAN4GYCeAX1NK7VzNPvyC8GUAt79E9scAHtFabwHwiPXvtYQygP+ktd4B4BCA37HWZq2OqwDgVq31NQCuBXC7UuoQgL8C8LfWeGYAfOwq9vHnxScAmNWs18OYXtNYR2fXetvnl7He9sTfA/i+1no7gGtAY1uza6SUagXw+wAOaK13A/ACeB/W/jrZWG3L1EEAF7XW3VrrIoD7ALxzlfuwYmitfwJg+iXidwK412rfC+CuVe3UCqG1HtFaH7faC6DN24o1Oi5NWLT+6bf+0wBuBXC5ANSaGc9lKKXaANwB4AvWvxXW+JjWCNbL2bWu9jmw/vaEUioO4A0AvggAWuui1noWa3iNLPgAhJVSPgARACNYw+v0Uqz2y1QrgAHj34OWbD2gUWs9AtCBBaDhKvfn54ZSqhPAPgBHsIbHZZn+nwMwDuBHAC4BmNVal61L1qL+/R2APwJglSZFHdb+mNYC1t3ZtV72OdbfntgIYALA/7Woyy8opaJYw2uktR4C8NcA+kEvUXMAjmFtr9OLsNovU8pFJuGEryEopWoAfAPAf9Baz1/t/qwEWuuK1vpaAG0gy8IOt8tWt1c/P5RSdwIY11ofM8Uul66ZMa0hrKt5Xi/7fJ3uCR+A/QA+o7XeByCDNUTpucHy73ongC4ALQCiIMr8pVhL6/QirPbL1CCAduPfbQCGV7kPrxbGlFLNAGD9f/wq9+eKoZTygw7Yf9Faf9MSr/lxWSbywyAfkVrLzAysPf17PYB3KKV6QTTTraBf5Wt5TGsF6+bsWmf7fD3uiUEAg1rrI9a/HwC9XK3VNQKA2wD0aK0ntNYlAN8E8Dqs7XV6EVb7ZeoZAFssD/4AyAHtwVXuw6uFBwF8xGp/BMC3r2JfrhiWn8EXAZzRWv+N8dGaHJdSKq2UqrXaYdBmPgPgUQDvsi5bM+MBAK31n2it27TWnaC982Ot9Qewhse0hrAuzq71ts/X457QWo8CGFBKbbNEbwZwGmt0jSz0AziklIpYOnh5TGt2nV6KVU/aqZT6ZdAvBy+AL2mt/2JVO/ALgFLqawBuAVW8HgPwpwD+FcD9ADaAFOfdWuuXOqm/ZqGUugnA4wBeAPse/BeQP8WaG5dSai/IodEL+tFwv9b6fyqlNoJ+waYAPAvgg1rrwtXr6c8HpdQtAP6z1vrO9TKm1zrWydm1rva5ifW0J5RS14Ic6gMAugF8FNY5hjW6RorS07wXFFH6LICPg3yk1uw6mZAM6AKBQCAQCAQrgGRAFwgEAoFAIFgB5GVKIBAIBAKBYAWQlymBQCAQCASCFUBepgQCgUAgEAhWAHmZEggEAoFAIFgB5GVKIBAIBAKBYAWQlymBQCAQCASCFUBepgQCgUAgEAhWgP8PcLQglKlbexcAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 720x720 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "test_edsr(edsr_model, x[8][:,:,::-1])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### EDSR With MAE and Perceptual Loss (VGG16)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "bf4af4065eb8447289f11328e40c4a5c",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Training', max=10, style=ProgressStyle(description_width='ini…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Epoch 0', max=501, style=ProgressStyle(description_width='ini…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Epoch 00001: loss improved from inf to 34.24871, saving model to sr_multi_loss_x2.h5\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Epoch 1', max=501, style=ProgressStyle(description_width='ini…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Epoch 00002: loss improved from 34.24871 to 27.99718, saving model to sr_multi_loss_x2.h5\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Epoch 2', max=501, style=ProgressStyle(description_width='ini…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Epoch 00003: loss improved from 27.99718 to 26.93084, saving model to sr_multi_loss_x2.h5\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Epoch 3', max=501, style=ProgressStyle(description_width='ini…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Epoch 00004: loss improved from 26.93084 to 25.30946, saving model to sr_multi_loss_x2.h5\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Epoch 4', max=501, style=ProgressStyle(description_width='ini…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Epoch 00005: loss improved from 25.30946 to 24.81123, saving model to sr_multi_loss_x2.h5\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Epoch 5', max=501, style=ProgressStyle(description_width='ini…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Epoch 00006: loss improved from 24.81123 to 24.43220, saving model to sr_multi_loss_x2.h5\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Epoch 6', max=501, style=ProgressStyle(description_width='ini…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Epoch 00007: loss improved from 24.43220 to 23.98002, saving model to sr_multi_loss_x2.h5\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Epoch 7', max=501, style=ProgressStyle(description_width='ini…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Epoch 00008: loss did not improve from 23.98002\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Epoch 8', max=501, style=ProgressStyle(description_width='ini…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Epoch 00009: loss improved from 23.98002 to 23.76163, saving model to sr_multi_loss_x2.h5\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Epoch 9', max=501, style=ProgressStyle(description_width='ini…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Epoch 00010: loss improved from 23.76163 to 23.53646, saving model to sr_multi_loss_x2.h5\n",
      "\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<keras.callbacks.History at 0x7f094ca6bac8>"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "edsr_model = EDSR(scale = scale, input_shape = (patch_size, patch_size, 3), \n",
    "                  n_feats = 256, n_resblocks = 32)\n",
    "\n",
    "HR = Input((img_size, img_size, 3), name = 'HR')\n",
    "LR = Input((patch_size, patch_size, 3), name = 'LR')\n",
    "SR = edsr_model(LR)\n",
    "\n",
    "vgg = VGG16(include_top=False, weights='imagenet', \n",
    "            input_tensor=Lambda(vgg_preprocess_input)(HR))\n",
    "for l in vgg.layers: l.trainable = False\n",
    "\n",
    "\n",
    "# get perceptual\\content outputs\n",
    "\n",
    "def get_output(m, ln): return m.get_layer('block{}_conv2'.format(ln)).output\n",
    "\n",
    "vgg_content = Model(HR, [get_output(vgg, o) for o in [1,2,3]], name = 'Partial-VGG16')\n",
    "vgg1 = vgg_content(HR) # content of HR\n",
    "vgg2 = vgg_content(SR) # content of SR\n",
    "\n",
    "\n",
    "# Create Perceptual Loss\n",
    "loss_con = Lambda(content_fn, name = 'Content')(vgg1+vgg2)\n",
    "\n",
    "\n",
    "model = Model([LR, HR], [SR, loss_con])\n",
    "losses = ['SR', 'Content']\n",
    "weights = [1., 1.]\n",
    "\n",
    "checkpointer = ModelCheckpoint(filepath = 'sr_multi_loss_x{}.h5'.format(scale), verbose=1, \n",
    "                               save_best_only=True, save_weights_only=True,\n",
    "                               monitor = 'loss')\n",
    "\n",
    "callbacks = [TQDMNotebookCallback(), checkpointer, TensorBoard()]\n",
    "\n",
    "model.compile(Adam(lr=1e-4, decay=0.9), loss = {loss : 'mae' for loss in losses}, \n",
    "              loss_weights = weights, metrics={'SR' : ssim})\n",
    "\n",
    "\n",
    "sr_trn_gen = ImageLoader(IMAGES, patch_size, scale, batchSize = bs, multi_loss=True)\n",
    "model.fit_generator(sr_trn_gen, len(sr_trn_gen), epochs=30, verbose=0, \n",
    "                    callbacks=callbacks, workers=8, use_multiprocessing=True)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "test_edsr(edsr_model, x[25])"
   ]
  }
 ],
 "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": "3.6.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}