{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "https://blog.keras.io/building-autoencoders-in-keras.html"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using Theano backend.\n",
      "Using gpu device 0: GeForce GT625M (CNMeM is disabled, cuDNN not available)\n"
     ]
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "from keras.layers import Input,Dense, Convolution2D, MaxPooling2D, UpSampling2D\n",
    "from keras.models import Model\n",
    "from keras.datasets import mnist\n",
    "from keras import regularizers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "(x_train,_),(x_test,_) = mnist.load_data()\n",
    "\n",
    "x_train = x_train.astype('float32') / 255\n",
    "x_test = x_test.astype('float32') / 255\n",
    "x_train = x_train.reshape((len(x_train), np.prod(x_train.shape[1:])))\n",
    "x_test = x_test.reshape((len(x_test), np.prod(x_test.shape[1:])))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "def plot_imgs(decoded_images,n=20):\n",
    "\n",
    "    plt.figure(figsize=(20,4))\n",
    "    for i in range(n):\n",
    "        # original digit\n",
    "        ax = plt.subplot(2,n,i+1)\n",
    "        plt.imshow(x_test[i].reshape(28,28))\n",
    "        plt.gray()\n",
    "        ax.get_xaxis().set_visible(False)\n",
    "        ax.get_yaxis().set_visible(False)\n",
    "\n",
    "        # reconstruction\n",
    "        ax = plt.subplot(2,n,i+1+n)\n",
    "        plt.imshow(decoded_images[i].reshape(28,28))\n",
    "        plt.gray()\n",
    "        ax.get_xaxis().set_visible(False)\n",
    "        ax.get_yaxis().set_visible(False)\n",
    "    \n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "# relu activation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "%%time\n",
    "\n",
    "encoded = None\n",
    "decoded = None\n",
    "\n",
    "encoding_dim = 32\n",
    "input_img = Input(shape=(784,))\n",
    "encoded = Dense(encoding_dim, activation='relu')(input_img)\n",
    "decoded = Dense(784,activation='sigmoid')(encoded)\n",
    "autoencoder = Model(input=input_img, output=decoded)\n",
    "encoder = Model(input=input_img, output = encoded)\n",
    "encoded_input = Input(shape=(encoding_dim,))\n",
    "decoder_layer = autoencoder.layers[-1]\n",
    "decoder = Model(input=encoded_input, output=decoder_layer(encoded_input))\n",
    "autoencoder.compile(optimizer='adadelta',loss='binary_crossentropy')\n",
    "\n",
    "\n",
    "autoencoder.fit(x_train, x_train,\n",
    "               verbose=2,  \n",
    "               nb_epoch = 50,\n",
    "               batch_size = 256,\n",
    "               shuffle=True,\n",
    "               validation_data=(x_test,x_test))\n",
    "\n",
    "encoded_imgs = encoder.predict(x_test)\n",
    "decoded_imgs = decoder.predict(encoded_imgs)\n",
    "\n",
    "plot_imgs(decoded_imgs)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# linear activation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false,
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "encoded = None\n",
    "decoded = None\n",
    "\n",
    "encoding_dim = 32\n",
    "input_img = Input(shape=(784,))\n",
    "\n",
    "encoded = Dense(encoding_dim, activation='linear')(input_img)\n",
    "decoded = Dense(784,activation='sigmoid')(encoded)\n",
    "autoencoder = Model(input=input_img, output=decoded)\n",
    "\n",
    "encoder = Model(input=input_img, output = encoded)\n",
    "encoded_input = Input(shape=(encoding_dim,))\n",
    "\n",
    "\n",
    "decoder_layer = autoencoder.layers[-1]\n",
    "decoder = Model(input=encoded_input, output=decoder_layer(encoded_input))\n",
    "\n",
    "autoencoder.compile(optimizer='adadelta',loss='binary_crossentropy')\n",
    "\n",
    "\n",
    "autoencoder.fit(x_train, x_train,\n",
    "               verbose=2,  \n",
    "               nb_epoch = 50,\n",
    "               batch_size = 256,\n",
    "               shuffle=True,\n",
    "               validation_data=(x_test,x_test))\n",
    "\n",
    "encoded_imgs = encoder.predict(x_test)\n",
    "decoded_imgs = decoder.predict(encoded_imgs)\n",
    "\n",
    "plot_imgs(decoded_imgs)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "# using regularization"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "%%time\n",
    "\n",
    "encoded = None\n",
    "decoded = None\n",
    "\n",
    "encoding_dim = 32\n",
    "\n",
    "input_img = Input(shape=(784,))\n",
    "# add a Dense layer with a L1 activity regularizer\n",
    "encoded = Dense(encoding_dim, activation='relu',\n",
    "                activity_regularizer=regularizers.activity_l1(10e-5))(input_img)\n",
    "decoded = Dense(784, activation='sigmoid')(encoded)\n",
    "\n",
    "autoencoder = Model(input=input_img, output=decoded)\n",
    "\n",
    "encoder = Model(input=input_img, output = encoded)\n",
    "encoded_input = Input(shape=(encoding_dim,))\n",
    "\n",
    "decoder_layer = autoencoder.layers[-1]\n",
    "decoder = Model(input=encoded_input, output=decoder_layer(encoded_input))\n",
    "autoencoder.compile(optimizer='adadelta',loss='binary_crossentropy')\n",
    "\n",
    "\n",
    "autoencoder.fit(x_train, x_train,\n",
    "               verbose=2,  \n",
    "               nb_epoch = 50,\n",
    "               batch_size = 256,\n",
    "               shuffle=True,\n",
    "               validation_data=(x_test,x_test))\n",
    "\n",
    "encoded_imgs = encoder.predict(x_test)\n",
    "decoded_imgs = decoder.predict(encoded_imgs)\n",
    "\n",
    "plot_imgs(decoded_imgs)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# deep autoencoder"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train on 60000 samples, validate on 10000 samples\n",
      "Epoch 1/100\n",
      "2s - loss: 0.3585 - val_loss: 0.2649\n",
      "Epoch 2/100\n",
      "2s - loss: 0.2594 - val_loss: 0.2526\n",
      "Epoch 3/100\n",
      "2s - loss: 0.2407 - val_loss: 0.2297\n",
      "Epoch 4/100\n",
      "2s - loss: 0.2256 - val_loss: 0.2202\n",
      "Epoch 5/100\n",
      "2s - loss: 0.2146 - val_loss: 0.2069\n",
      "Epoch 6/100\n",
      "3s - loss: 0.1979 - val_loss: 0.1900\n",
      "Epoch 7/100\n",
      "3s - loss: 0.1867 - val_loss: 0.1826\n",
      "Epoch 8/100\n",
      "3s - loss: 0.1803 - val_loss: 0.1770\n",
      "Epoch 9/100\n",
      "3s - loss: 0.1745 - val_loss: 0.1707\n",
      "Epoch 10/100\n",
      "2s - loss: 0.1695 - val_loss: 0.1651\n",
      "Epoch 11/100\n",
      "2s - loss: 0.1651 - val_loss: 0.1621\n",
      "Epoch 12/100\n",
      "2s - loss: 0.1616 - val_loss: 0.1592\n",
      "Epoch 13/100\n",
      "2s - loss: 0.1587 - val_loss: 0.1560\n",
      "Epoch 14/100\n",
      "2s - loss: 0.1563 - val_loss: 0.1548\n",
      "Epoch 15/100\n",
      "2s - loss: 0.1540 - val_loss: 0.1505\n",
      "Epoch 16/100\n",
      "2s - loss: 0.1518 - val_loss: 0.1500\n",
      "Epoch 17/100\n",
      "2s - loss: 0.1494 - val_loss: 0.1482\n",
      "Epoch 18/100\n",
      "3s - loss: 0.1472 - val_loss: 0.1451\n",
      "Epoch 19/100\n",
      "2s - loss: 0.1449 - val_loss: 0.1417\n",
      "Epoch 20/100\n",
      "2s - loss: 0.1427 - val_loss: 0.1420\n",
      "Epoch 21/100\n",
      "2s - loss: 0.1407 - val_loss: 0.1385\n",
      "Epoch 22/100\n",
      "2s - loss: 0.1390 - val_loss: 0.1360\n",
      "Epoch 23/100\n",
      "2s - loss: 0.1371 - val_loss: 0.1353\n",
      "Epoch 24/100\n",
      "2s - loss: 0.1357 - val_loss: 0.1337\n",
      "Epoch 25/100\n",
      "2s - loss: 0.1343 - val_loss: 0.1319\n",
      "Epoch 26/100\n",
      "2s - loss: 0.1329 - val_loss: 0.1303\n",
      "Epoch 27/100\n",
      "2s - loss: 0.1317 - val_loss: 0.1295\n",
      "Epoch 28/100\n",
      "2s - loss: 0.1306 - val_loss: 0.1283\n",
      "Epoch 29/100\n",
      "2s - loss: 0.1296 - val_loss: 0.1268\n",
      "Epoch 30/100\n",
      "2s - loss: 0.1285 - val_loss: 0.1265\n",
      "Epoch 31/100\n",
      "2s - loss: 0.1276 - val_loss: 0.1257\n",
      "Epoch 32/100\n",
      "2s - loss: 0.1269 - val_loss: 0.1261\n",
      "Epoch 33/100\n",
      "2s - loss: 0.1260 - val_loss: 0.1238\n",
      "Epoch 34/100\n",
      "2s - loss: 0.1252 - val_loss: 0.1245\n",
      "Epoch 35/100\n",
      "2s - loss: 0.1245 - val_loss: 0.1223\n",
      "Epoch 36/100\n",
      "2s - loss: 0.1237 - val_loss: 0.1219\n",
      "Epoch 37/100\n",
      "2s - loss: 0.1230 - val_loss: 0.1206\n",
      "Epoch 38/100\n",
      "2s - loss: 0.1222 - val_loss: 0.1202\n",
      "Epoch 39/100\n",
      "2s - loss: 0.1215 - val_loss: 0.1196\n",
      "Epoch 40/100\n",
      "2s - loss: 0.1209 - val_loss: 0.1206\n",
      "Epoch 41/100\n",
      "2s - loss: 0.1202 - val_loss: 0.1182\n",
      "Epoch 42/100\n",
      "2s - loss: 0.1195 - val_loss: 0.1172\n",
      "Epoch 43/100\n",
      "2s - loss: 0.1191 - val_loss: 0.1172\n",
      "Epoch 44/100\n",
      "2s - loss: 0.1185 - val_loss: 0.1158\n",
      "Epoch 45/100\n",
      "2s - loss: 0.1179 - val_loss: 0.1163\n",
      "Epoch 46/100\n",
      "2s - loss: 0.1174 - val_loss: 0.1150\n",
      "Epoch 47/100\n",
      "2s - loss: 0.1167 - val_loss: 0.1146\n",
      "Epoch 48/100\n",
      "2s - loss: 0.1162 - val_loss: 0.1145\n",
      "Epoch 49/100\n",
      "2s - loss: 0.1156 - val_loss: 0.1152\n",
      "Epoch 50/100\n",
      "2s - loss: 0.1150 - val_loss: 0.1145\n",
      "Epoch 51/100\n",
      "2s - loss: 0.1145 - val_loss: 0.1127\n",
      "Epoch 52/100\n",
      "2s - loss: 0.1140 - val_loss: 0.1129\n",
      "Epoch 53/100\n",
      "2s - loss: 0.1136 - val_loss: 0.1124\n",
      "Epoch 54/100\n",
      "2s - loss: 0.1132 - val_loss: 0.1118\n",
      "Epoch 55/100\n",
      "2s - loss: 0.1128 - val_loss: 0.1112\n",
      "Epoch 56/100\n",
      "2s - loss: 0.1123 - val_loss: 0.1114\n",
      "Epoch 57/100\n",
      "2s - loss: 0.1119 - val_loss: 0.1098\n",
      "Epoch 58/100\n",
      "2s - loss: 0.1116 - val_loss: 0.1102\n",
      "Epoch 59/100\n",
      "2s - loss: 0.1113 - val_loss: 0.1104\n",
      "Epoch 60/100\n",
      "2s - loss: 0.1108 - val_loss: 0.1096\n",
      "Epoch 61/100\n",
      "2s - loss: 0.1106 - val_loss: 0.1088\n",
      "Epoch 62/100\n",
      "2s - loss: 0.1102 - val_loss: 0.1097\n",
      "Epoch 63/100\n",
      "2s - loss: 0.1098 - val_loss: 0.1087\n",
      "Epoch 64/100\n",
      "2s - loss: 0.1096 - val_loss: 0.1098\n",
      "Epoch 65/100\n",
      "2s - loss: 0.1092 - val_loss: 0.1095\n",
      "Epoch 66/100\n",
      "2s - loss: 0.1090 - val_loss: 0.1076\n",
      "Epoch 67/100\n",
      "2s - loss: 0.1086 - val_loss: 0.1070\n",
      "Epoch 68/100\n",
      "2s - loss: 0.1084 - val_loss: 0.1070\n",
      "Epoch 69/100\n",
      "2s - loss: 0.1081 - val_loss: 0.1061\n",
      "Epoch 70/100\n",
      "3s - loss: 0.1079 - val_loss: 0.1067\n",
      "Epoch 71/100\n",
      "3s - loss: 0.1075 - val_loss: 0.1058\n",
      "Epoch 72/100\n",
      "2s - loss: 0.1072 - val_loss: 0.1066\n",
      "Epoch 73/100\n",
      "2s - loss: 0.1069 - val_loss: 0.1051\n",
      "Epoch 74/100\n",
      "2s - loss: 0.1068 - val_loss: 0.1060\n",
      "Epoch 75/100\n",
      "2s - loss: 0.1064 - val_loss: 0.1059\n",
      "Epoch 76/100\n",
      "2s - loss: 0.1062 - val_loss: 0.1046\n",
      "Epoch 77/100\n",
      "3s - loss: 0.1060 - val_loss: 0.1048\n",
      "Epoch 78/100\n",
      "3s - loss: 0.1057 - val_loss: 0.1051\n",
      "Epoch 79/100\n",
      "3s - loss: 0.1056 - val_loss: 0.1054\n",
      "Epoch 80/100\n",
      "3s - loss: 0.1052 - val_loss: 0.1045\n",
      "Epoch 81/100\n",
      "3s - loss: 0.1049 - val_loss: 0.1033\n",
      "Epoch 82/100\n",
      "2s - loss: 0.1047 - val_loss: 0.1043\n",
      "Epoch 83/100\n",
      "2s - loss: 0.1045 - val_loss: 0.1027\n",
      "Epoch 84/100\n",
      "2s - loss: 0.1043 - val_loss: 0.1031\n",
      "Epoch 85/100\n",
      "3s - loss: 0.1041 - val_loss: 0.1023\n",
      "Epoch 86/100\n",
      "2s - loss: 0.1039 - val_loss: 0.1024\n",
      "Epoch 87/100\n",
      "2s - loss: 0.1036 - val_loss: 0.1031\n",
      "Epoch 88/100\n",
      "2s - loss: 0.1035 - val_loss: 0.1019\n",
      "Epoch 89/100\n",
      "2s - loss: 0.1034 - val_loss: 0.1020\n",
      "Epoch 90/100\n",
      "2s - loss: 0.1031 - val_loss: 0.1027\n",
      "Epoch 91/100\n",
      "2s - loss: 0.1030 - val_loss: 0.1015\n",
      "Epoch 92/100\n",
      "2s - loss: 0.1027 - val_loss: 0.1029\n",
      "Epoch 93/100\n",
      "2s - loss: 0.1025 - val_loss: 0.1022\n",
      "Epoch 94/100\n",
      "2s - loss: 0.1025 - val_loss: 0.1013\n",
      "Epoch 95/100\n",
      "2s - loss: 0.1023 - val_loss: 0.1013\n",
      "Epoch 96/100\n",
      "2s - loss: 0.1021 - val_loss: 0.1018\n",
      "Epoch 97/100\n",
      "2s - loss: 0.1019 - val_loss: 0.1006\n",
      "Epoch 98/100\n",
      "2s - loss: 0.1017 - val_loss: 0.0999\n",
      "Epoch 99/100\n",
      "2s - loss: 0.1016 - val_loss: 0.0997\n",
      "Epoch 100/100\n",
      "2s - loss: 0.1016 - val_loss: 0.1012\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABiEAAAD+CAYAAABcOabWAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAAPYQAAD2EBqD+naQAAIABJREFUeJzsnddTm+mW7h/lnHMOgEgGnLrdeWZPTe2pfTX353+cq7mY\nOTtM1exUu6udDSaDEBLKOedz4bNWCxt329jYSLy/KhXdRgLp5fvesNazniUZj8cQCAQCgUAgEAgE\nAoFAIBAIBAKBQCD42Eg/9xsQCAQCgUAgEAgEAoFAIBAIBAKBQDCbiCSEQCAQCAQCgUAgEAgEAoFA\nIBAIBIIrQSQhBAKBQCAQCAQCgUAgEAgEAoFAIBBcCSIJIRAIBAKBQCAQCAQCgUAgEAgEAoHgShBJ\nCIFAIBAIBAKBQCAQCAQCgUAgEAgEV4JIQggEAoFAIBAIBAKBQCAQCAQCgUAguBJEEkIgEAgEAoFA\nIBAIBAKBQCAQCAQCwZUgkhACgUAgEAgEAoFAIBAIBAKBQCAQCK4E+bs8SSKR2AD8G4A4gM5VvqFr\njBpAGMD/HY/HxXd5gRg3AGLcLosYt8sjxu5yiHG7HGLcLocYt8shxu1yiHG7HO89boAYu/+PuOYu\nhxi3yyHG7XKIcbscYtwuhxi3yyPG7nKIcbscYtwux7uN23g8/tUHgP8DYCweGAP4P+8yZmLcxLiJ\ncfvsDzF2YtzEuF3/hxg3MW5i3K7/453HTYzd5cdOjJsYNzFuYtym7CHGTYybGLvpeIhxE+N2bcbt\nXe2Y4u/4vJtA/IqeO+vEr+i5s078ip57E4hf0XNnnfgVPXfWiV/Rc2ed+BU9d9aJX9FzZ534FT13\n1olf8fNnmfgVPXfWiV/Rc2ed+BU9d9aJX9FzZ534FT131olf0XNvAvEreu6sE7+i58468St67qwT\n/6VvvmsS4qaWk1zE+4yFGLefEeN2OcS4XR4xdpdDjNvlEON2OcS4XQ4xbpdDjNvleN+xEGP3M+Ka\nuxxi3C6HGLfLIcbtcohxuxxi3C6PGLvLIcbtcohxuxy/OBaiMbVAIBAIBAKBQCAQCAQCgUAgEAgE\ngitBJCEEAoFAIBAIBAKBQCAQCAQCgUAgEFwJ8s/9BgSXQyKRQK/Xw2AwQC6XQyKRQCqVQqFQQKFQ\nQKPRQKPRQCaTAQAGgwGazSaazSZ6vR663S46nQ5arRa63e5n/jQCgUAgEAgEAoHgKpFIJDCbzXA6\nnTAajVAqlZDL5eh0Ouh0OqhUKigWi2g0Gp/7rQoEAoFAIBAIZgyRhJhSpFIp3G43IpEItFotZDIZ\nVCoVDAYDjEYj3G433G431Go1AKDVaiGRSODk5ATlchmVSgXZbBZnZ2ciCSEQCAQCgUAgEMwwUqkU\nUqkUwWAQX3/9NZaWlmA0GqHX65HL5ZDL5bC9vY1Hjx6JJIRAIBAIBAKB4KMjkhBTBiUb9Ho9IpEI\nbt++DaPRCJlMBo1GA6vVCqvVinA4jHA4DK1WCwCo1+vY2dnBzs4OMpkMstksNBoNms0mSqUSxuMx\nxuPxZ/501xuqNlEqldBoNFAqleh2u+j1euj3+xgMBhiNRp/7bX4W5HI5lEolK+rkcjnG4zFGoxEG\ngwG63S663S5Go5G4zgQCgUBwDolEcu5xEbSmyOVyKBQKSKVSjEYjftD3xTpzc5BIJJDJZJBIJBgO\nhxiNRpBKpfzvMpkMcrmc/3s4HPLzCPp/+jqr145EIoFarYZGo0E0GsXXX3+NBw8ewGazwWw24/Dw\nEIeHh+j3+9jf3//cb1cgEFyAQqGAUqnkr7QODodDrmYaDoef+21eG2QyGaRS6bm1gL7Suf4iJtcD\nOsv2+330+33+N4FAcDGT+/nJPTtB95PYs99cRBJiyjAajYjFYojFYlhZWcHq6ip0Oh1bMWk0Gmi1\nWpjNZg4EA682LR6PBzKZDG63G8ViESqVCuVyGfl8noPEgrejUqmgVqs5+eP3+xGPx3F8fIx0Oo1M\nJoNms/m53+YnhVR1drsdwWAQgUAALpcLLpcL/X4fzWYT2WwWe3t7ODg4OJeMEAgEAoFAKpVy8lql\nUkGlUr3xnPF4zOuH3W6Hx+OBRqNBq9VCq9VCp9NBu93m/+/1ep/hkwg+NRqNBgaDATKZDPV6HY1G\ngwPtRqMRFovl3KNSqaBcLvP1MRgMUK/Xzz1m8dqhM0IwGEQ0GsXt27cRCoVgsVigVqsxHo/RarVQ\nLBZRrVZncgwEgmlHIpHA7XYjEAjA7/cjEAjAaDSiVquhVqvh5cuX2NraQqVS+dxv9Vogl8uh0+mg\n1+tZpGm1WmGz2WAymdi6+nWGwyFbWLdaLbTbbRSLRaRSKWQyGbTbbbTbbRE4FQheY1IAQuJUh8MB\nt9sNvV4PABiNRlx5Wa/X2SpecLMQSYgpw2g0Ym1tDf/8z/+M+fl5LCwsnFtAKetIwWFCqVTC7XbD\n4XCgWq2iWq1iOBzi8PAQBwcHGI/H6PV6YkH9BZRKJYxGI5aWlvDv//7vuHv3Lv72t7/h73//O2Qy\nGarV6o1MQsjlcthsNty6dQv379/H8vIylpeX0W63USgUsLOzA4VCgWw2CwCc+RYIBAKBgAKkarUa\ner0eer3+jWqI0WiEer0OiUQCl8uFW7duwWw2o1QqoVQq8b6mVCqh3++LA80NQa1Ww2azQaFQYDQa\nodVqQaPRwGQywev1IhQKIRQKIRgMIhgM4vT0FIlEgvdq3W4X2WwW2WwWmUwGnU5n5q6dySreYDCI\nL7/8EhsbGwgGgzCbzaykbrfbKJVKqFQqMzcGAsEsQEmIjY0N3Lt3D/fu3YPX68XZ2RnS6TQUCgUS\niYRIQvx/5HI5DAYDHA4HwuEwIpEIP/x+P0wmE8xm8xv7jV6vh0KhgEKhwBbWR0dHePz4MTqdDsbj\nMX8VCAQ/Q0kIpVIJnU4HnU6HSCSC1dVVOJ1OAK/EHzs7O9y3ttfriT3HDUQkIaYEp9MJn8+H5eVl\n3LlzB7FYDC6XC1qtFkqlEsCrzD01nW42m2g0GpBIJG8c7qlywu12IxgMIhwOI5lMotlsigX1F9Bo\nNLBYLLBarTAajdyLYzAYYDgc3sixk8vlHASIRqNYXV2Fz+djhYlCoUCz2eQqiXQ6jW63i8Fg8Lnf\n+idFrVZDrVbD5XLB7/dDKpXi8PAQR0dHl/6ZWq0WOp0OEomEF3AqFb7p6HQ6GAwGOJ1OPmiQDV25\nXEa5XEan0/ncb/OjQiWvSqUSPp8PPp8PwKt+QI1GA5VKBZVK5Vw5+YdCJe5qtZqt/0g9dhPnQ8Gv\nI5PJeC9iMBig1+thNBphMplgNBphNBphMBjesEgYDoes9vR6vZibm4PBYECr1WK1YqvVwsuXL/HT\nTz/dOEHATYCC6QqFAl6vF16vFz6fD16vFxqNhuc42ufabDY4nU44HA7Y7XbY7XYYjUa4XC6u/O33\n+xxkKpVKXB2cTqeRz+dRq9VQr9enWjihVqtZiXjr1i2sr6/D7Xaj3+8jkUggn88jl8uxivrw8BC1\nWu1zv23BDEL3MO0bVCoVB6qMRiNsNhtsNhsAcKCXlLK0dymXyyiVSmg2mzfSBtdgMMDr9cLj8cBq\ntcJsNmM8HkOtViMQCMDr9aLRaKDVas20w4HBYIDZbIZer4dSqYRKpYJWq4VGo+Hgplqt5jWAHk6n\nE06nEzabDVqtFlqtFtVqFZVKhef7SqXCDgeNRoOr+s/OztBoNIRoUyB4DZrb6exN52+fzwe/389n\nceDVft7pdGJ+fh6pVApnZ2c4OztDNptFLpdj20zB1TEZK2i322g0GqjX659UUC2SEFOCx+PBV199\nhbt372J1dRVzc3NQqVSQy3/+E1L5YK1WQyaTQSaTgUQigU6ng8VigdfrhV6v59eRMiCdTqPVaiGd\nTt+4zdz7oFarYbFYYDaboVQqMRqN0O120Wg00O12b+SEKZfLuRdJOBzG4uIiNBoNpFIpVCoV23/5\n/X6Ew2FWl9wkJBIJtFotLBYL1tbW8O2330KpVOI///M/PygJodPp4Ha7IZPJeOPcarUwGAxu/ObY\naDTC6/ViY2MD33//PcLhMB49eoTHjx9jf3+fbV1maZxI6arX67G6uopvvvkG4/EYxWIRZ2dnODw8\nxGg0QrPZ/GgbPPL5NJvNcDgcAIB0Os1l6mI9EbyOQqGA1Wo9F0R2u91wuVxwOBwwmUwwmUwXVkJU\nKhVUq1UYjUbY7XZotVr2kiUxwB//+EecnJzg5OTkM31CwVVBiVaNRoPFxUV89dVXmJubg8/ng9Fo\nZD908ksne1IS6yiVStjtdkQiEZ7/RqMRv44e8XgcDx8+xPPnz3F6eopmsznVc5lGo0EgEMDKygrW\n1tawuroKuVyOWq2Gs7MzbG5uYnNzk6tEyuUyWq3W537bghmEhHFKpRJmsxlmsxlOpxMejwfBYBBL\nS0tYWloC8CoJUSqVkEwmkclkOOF8cHCA3d1d7oFwkxS0EokEer0eLpcLNpsNarUacrkcJpMJWq2W\ng36lUomtlmcVk8mEaDQKj8cDg8EAk8kEu90Oh8MBhUIB4HwSgtYAslZWKpWcrCgWizg6OkIikUAq\nlUIymeRHr9fj3oZkzSTOWQLBecgZg+7L5eVlrK2tcdWyWq3m+3I8HiMWi6HdbiOZTOLo6Ai7u7t4\n/PgxarUaer2e6BNxxajVasRiMXz99dcolUo4OztDMpnEycmJSEIIXqFQKFj1RQcIyibSolir1VjJ\nRZYElMEHAL1eD6fTicFgAJPJxD+TkhN2u50V1YK3Q5UQRqMRCoUC/X4ftVoN2WwW1Wr1Rqn7J3tB\n+P1+zM/Pw+fzsYKJID9O2iBOKlRuCqQMcLvdCIfDWF5ehkwme2Os3he9Xg+PxwOtVsvJx7OzsxtZ\naULQdelyubC8vIy7d+/i7t27iEQivEEajUZ8z87CJocs+AwGA/x+P0KhEO7fv4/79++j2+0ilUpB\nJpMhn89jPB5/tAoI6jNEAWSn08nVJbVa7aNWXHwOJpM6BoOBexY0m02Uy2XU6/XP/RavLVSOLZfL\nOQhMCkWyyfF4PJyIoASEzWZj/+aLKiFIqUNl3tSUc3JN2d7ehk6n+9QfWfAJUCqVsNlscLvdWFlZ\nwb179xAOh+FwOKDX6zm5OpkApSbVlKiSSqVQq9X8MyUSCaxWK68Ng8GA12ZaT6rVKhqNxtSqrpVK\nJSwWCyf8HA4HV8fF43G8fPkSjx49QqlUQrFYFNWUgo8KNUSnuZ2SD+TP73K5uM9BLBbD4uIiv7ZU\nKsHtdnMSotVqQavVYjweQy6XI5fLoVQqfcZP92mhJITb7YbNZoNKpeI5Ta1Ww2q1wuPxIJ1Oz/we\nxWazYWVlBfPz81xBabfbYbPZIJVKeb7X6XTQarUYDAbo9XrodDool8vnGnkfHx/j+PgYiUTinCo7\nm81O5Zx/WWjfO5msoYolutbIAWJyPDudDldB39Tz502HYopzc3O4desWbt++jeXlZSwtLZ3bcwHg\ns+h4PGZxr8Fg4LNDPp9HPp9Hs9m80TGNq0SpVMLr9WJ9fR2ZTAYKhQKdToet0z8FIglxjaHFkw7t\n4XAYPp8PBoMB4/GYy2d2d3fx5MkTHB0dcdkg2RYArxoqe71e9pI0mUx8yKeF5qYFhi8DVUIYjUbI\nZDJ0Oh0UCgUcHx+jUCjMtOJkEgpAKpVKRKNRfPHFF7hz5w5cLtfnfmvXEolEAqPRCJ/PB6fTyZvh\nD0366fV6DuABr2x3Hj9+jHw+fyMX7Ek7olAohAcPHmBjYwN2ux0qlQrRaBR6vR7lchlbW1tIp9Mf\nLSj/OaH70e1247vvvsO3336LQCCAYDCISqWCdrvNVW71ev2De7JQMsdgMGBjYwPffPMNXC4XW161\n222kUqmprsqZVPTEYjHMz8+zhVA8HsejR4/YcmoaP99VI5VKodFooNPpEAgE+EFBUIPBAKPRyGNK\nSQpSJ140N1Igiw7BCoUCMplMiCduEHq9HgsLC2wpFI1G4XA42PpRJpNhPB6zPRcpWCcfk5Aqmyxg\nKHHmdDqxvr4Og8GA4XCIbDYLiUQytU2raa9PFSESiQStVgu5XA7JZJIDbu12+0ZW9AquDhJJWK1W\nhEIhhMNhhMNhhEIhtt8j+zSDwQCr1cqvHY/H0Gg0vGYMBgP0+31uJqzT6fD06dMblYQAwJUQdrv9\njeDe5PdIiDiruFwu3L17F7dv3+YzqUqlglKpZFvqVquFUqmEdrvNQfJOp4N+v88WS7lcjkWc9Xod\njUaD7axu2v6O9r1Wq5UtdUhkRJU3KpUKrVaLBbDUTykej+P4+PhGnj8FgN1uRywWw9raGu7du4db\nt27BZrOdc2yZZFJAFwgEoNVqYbVasbS0hOfPn+PZs2dIpVIoFApoNBqf+NPMPnK5HGazGT6fj6sO\nSdz1yd7DVf7wSe9HegDgSZ2+TqqXJr930yb/16EkBClFfD4fBxwHgwGq1SpyuRxevHiBP/3pT3j+\n/DlnpGmzRs1hCoUC5ufnUavV2DeRJgB6CC6GrmO9Xs92EXQgLRQKHHCbdeheJhWqVqvF/Pw8vvji\nCywvL8Nut1/4ukkfZ4VCwZtEUivOum2LVCrlJITD4YBWq+V+LR+CRqOBzWZDMBiEXq/HYDBAJpPh\neeCmQRZgWq0WgUAAt2/fxtLSEnQ6HQfpnU4nnjx5wvPftDOpMKSGo7/73e94ve12u5DJZGwd9zFK\nLGUyGV97t27dwm9/+1tWEu/t7WFzcxNqtRq9Xm/qxlgmk7GqUKvVwufzYXV1FV988QUsFgssFgse\nP36M09NTHB8f39heQL8G+X2bTCZEIhHcvn0bKysrWFpagt/v56oS4peuE9oLjkYjvsfp3y867E6r\nWv11Jvdl9N90fdKDvj8ej3kfPRqN+DEr1+bkYXVubo73HMFgEBqNhq8P6otGwZFms4l2u82WLa8n\nECQSCTQaDUaj0TmrUqvVys1Mz87OsLu7y83OpzkJodFoIJfLIZFI0G63kc/nkUqlkMlkUCwWP/fb\nnDroHqT5jO7Pi+azyXuUEmKT9ye9ju7dWYEsG51OJ5aXl7GxsYGVlRUsLy9zdRytBTTX9/t9PjeQ\nEnvyfKFUKqFWqzEYDHB6eor9/f3P9fE+C1qtFna7HWaz+Y3vkcuBVqt9a/BvVrBYLFhYWMD6+jrf\nP9Qbr1QqsVCQ7KmpZxBZKlElWDwe/9wf5bNCc5hCoWArWxK+TiYNfT4fC0ZqtRpKpRLS6TRXkQCv\nbK3Innra9h+0r6KE1uvVuBdBe4/X91s0l9F8Tz+fqjIBzNxcbzabEYlEsLy8zPt9gta+12O9wKvz\nAlUwBQIBNBoNTjpLpVIMh0P0+/1zYym4PLS2KpVKmEwmuN1udDoddiv5lOvGlf4mKg20Wq3cDGg8\nHqPX6/FhaTAYoFgsolgsot1u80afPLtvMqTS0mq1XK0wGAzQ6XTQaDTw/PlzPHnyBJubm0gkElwu\nTkrX8XjMwc/l5WWEw2FYrVZuqNxut1kJVa1Wp27B+BRQwE2j0SAYDGJlZQVutxuNRgPFYhGlUmmm\nFpFfwmQysSKCGnzdvXsXoVAIFouFg0OvQ9lWr9eLdrsNAHy/N5tN1Ot11Ov1mbz+KHBkNpvh9/vh\ndDqhVqs/Sla/0+mgVCrB6XRyKSNlsSnwPItj+jZkMhmMRiOcTic3IKUFdXKdobL+adwkT6JQKCCX\nyxEKhbC4uIg7d+4gHA5DoVDwZ93e3sbTp0/Z7/tjoNfrEYlEsLi4iGg0CovFgkajgWQyiRcvXiAe\nj6NWq6HT6UzV3CiVSmG1WmGz2djWKhKJYG5uDtFolNeBRCIBvV7P3qbT9Bk/FRQ80uv18Pv9WF1d\n5WtFLpe/0+GOGAwGHEhut9tot9u/OOZUjTPtUKCN7nOyg6RqTIPBwNdgr9dDLpdDLpdDrVY715R0\nGoPmr0MBdKfTiUAggGg0ypULZAlRLpdxdHSEo6MjDjaR6rXb7fLZgqBAsV6vx8rKCvL5PPx+P7xe\nL++RtVotgsEg7t27B4VCwWeTaWuaOBlkkkqlXC2Sy+WQyWREE/dLIJPJuKEyNd602Wwwm81viBxo\nvKnpOXnN0/1JwT+NRoNqtYparTYTyVSpVMoN0Tc2NvDFF19gZWUFLpcLWq2Wr0diMBiwEp1sEGmO\nm8RkMiEcDiOTycBoNH7Kj3TtyeVyeP78OXZ2dlAulz/327lSGo0G928wm83cF4gcIh4+fIjDw0M+\nY04mpPv9PjqdDqrV6uf+GJ8VEgcGAgHMz88jHA7D4/HA4/HAYrHAarXCaDRCp9NhOByi3W5zhaFC\noYDdbodSqYTL5eK+pCqVCsVicarswOicbjKZMDc3h4WFBVit1reKdCn5MBgMUKlUUC6XOUBOoq9u\nt8tVItSfVaVScZyO7EVnJbDe6/W4+uj1z1Qul7niiO5DGluHwwGfzwez2QyZTAa9Xo/FxUXIZDL4\n/X7s7Ozg8PAQ6XQa6XSa48eCy0GJ6sk4yetr8afiypMQLpcL0WgUS0tLWFxc5M0YTWK9Xg8HBwc4\nODhAtVrlxlPj8fjGJyEAnEtCUEaw2Wwin8/j2bNn+O///m8kEgm+sV/PMBqNRvZnC4VCsNlsXLbe\n6XREEuJXkEqlXCIWDAaxvLwMvV6P4+NjHB4e3rgkxNzcHGKxGJdVk+3LRR7eBDWu9fl8AF4p+MnH\nldR31Cx31phMQgQCATgcjjfKpy8L+ZrW63UoFApYLBZW/ZMtxU26p6mM2Ov1wuFw8OIqkUhYEUUb\nmVarNdVJGgoqqdVqhEIhfPvtt9z7QqlUolwuY3d3Fz/99BP++te/spL3Y6DX6xGNRnHnzh1EIhFY\nLBZks1lsbW3h0aNHiMfjvJ5M0/iSZcTc3Bxu376NL774AgsLC3wwIRUT9VBSKBQzOWd9DKhiwWAw\nwOfzYWVlBT6f71xF7LsyGAzQbDZRrVZRLpfZy/ltzEoSgsaPbKrMZjOvuxQk0Gg0AF4FY3Z2drC9\nvY10Oo1sNst7k1lJQlCCmZIQOp0OcrkcvV4P7XYb2WwWP/30E/7yl7+wtcZkIub1HjV0ADYajcjn\n82g0Guj1ehxwocRPKBTiCuN0Oo1SqcSJiGmB9iF00ByNRueSEMLq4P2Ry+XQ6/Ww2+1sPzE/P49A\nIAC73c7XF62D+Xwe8Xgce3t7ePjwIduZUDWOy+WC2WxGKpVCp9M555k9rVASYnFxkdfUWCzGtmfA\n+So4qvDPZDJwuVznGplOQqKbdDotkhCvkc1m8eLFC+zs7MxMcPNt1Ot1TkJQ8LLT6aBSqWB3dxd/\n/OMf8eTJkzeqAycf0zSPf2wmha5zc3P4l3/5F9y9exculwsul4v3ayR+pWT+cDhk61udTgePx8MW\nf6VSCf1+H91ud6qSEHK5HBaLBX6/H9999x3+9V//lXsJXhTboAREr9fDyckJEokEOxBQMrVer2Nr\nawvdbhcSiQROp5OvUfLeJ+eSWYAszprN5hvnzUqlgoODAyQSCZTLZV77pFIpFhcXuf+eQqGAWq3G\n4uIiQqEQQqEQ7HY79Ho9JBIJCoUCAOGW8yFQBYTNZuPqh7fZ4F41V5qEMBgMmJ+fx7179xCJRBAO\nhwGAJ7LRaIR+vw+Xy4VIJIJGo8Hq6HK5/FEy1FS+0+v1UCqVUC6XOQt33Ref0WiERqOBbDaLly9f\nQqfTwel0cqn506dPkUwmUalU0O12LwyG63Q6+P1+zM3NwW63n/NRnpxEpzkgd5WQGm6ycc5oNEI+\nn8fx8TFKpdKNGTebzYbV1VXcuXOHqyEsFsuvlm9RM8nBYMDNESmYRNf2YDBAo9GYivvyXaFsMzXS\n9Hg8rBBJJpMfZYMmkUh48aYkBFVCzMo4vgt0kPf7/VhfX4ff74dWq+XNY7/fRz6fx/7+PtLpNCds\npxX6vGazGR6PhxOCRqMR4/EY2WwWz58/x+7uLrLZ7Eexi6MSZZfLhVgshlu3bkGv16NYLOLk5AR7\ne3s4Pj5GuVyeqsQsHcQ0Gg3C4TDu37/Pyn232812EITVakUsFkM2m0UqlUIqlWJlzjRfU+8K2fLR\n9Wc0GrlCgRqHAj/bbU42jiYFGCV06PmkTux2u3xAIyVZq9VidSz1uvql6+vg4ACVSuXqB+IjIpfL\n2UqOmraSH7Ner4dGo4HBYGB/ZqowJu/WTqcDjUYDu93O+1yqBqhWq1xFUiwWOeA+Tdcq2UCSDzwp\npCUSCWq1Gk5PT7G9vY3d3V3s7+/zdUXJh8Fg8NbqhX6/j4ODAx4Pas5H6yklJs/OzrC3t4disYhK\npTIVzZspiGQ0Gnmd0Gq1qNfrKBaLyOVyXJUq+HUkEglUKhXfawsLC1hYWMDS0hKWlpbg8/lYXfh6\nEoISQHK5HO12m+e34XDIFesmkwlbW1tsoVupVKbSWpOS0Hq9HuFwGLdv30YsFuP+La9Tq9XYNufo\n6AjHx8eIRqOYn59nRbbBYODn03X9udSb1xmqHLwJIs5CoYAXL17wmcdsNmM0Gp07A1FT21kJ9H4M\nKCmtVqsRiUQQiURw9+5drKysIBAIcBI+n8+zYHCyQqvf7/M8SAJNv98PvV6PYDCIQqGAbDbLiSGD\nwcCJiUlBwHWCqnfNZjO7tzgcDg6Uvw4ltcg6TqfTseCDKkba7TbsdjtCoRAkEgksFgvUajVXZabT\naWQyGRYT0p5lcozoudNw/ZJNnN1uZ5tL2rvv7Ozg4cOH3Lu21WrxGlmv19HtdlEsFvlaor6DPp+P\nqwWpHwk1jr9JPVsme/ja7XY4HA4WklSr1fea4+iapqp0qVTKlTlUJPCpuPIkRCwWw1dffQWHwwG7\n3c5q/smM9MLCwrlNGSUhPkYpYb/fR7vdRrVaxd7eHnZ3d5HP56einHo0GnHSpNfrIZVKQavVotPp\noNVqIZPJsO/h2w7lOp2Ovf2sVqto5PieUCWE2WzmUjpS6kxjwO1DIMXXgwcPoFKpuLfDr/nHKZVK\nOBwO6PV6XlBJ2Xp2dgapVIpKpYJsNjsV9+W7QtUJPp8PXq8XHo8HnU6HD1ofOr9NBmaMRiNfozet\nYSstqFoC14NhAAAgAElEQVStFqFQCPfv30coFDp32B0MBsjn89jb20Mmk5nKg/0k1C/IbrfD5XLB\n6/XC6XRCqVRiMBggnU7j6dOnODg4QK1W+yi/U6VScTArFothZWUF1WoVyWQSh4eHrHL5WL/vUyGT\nyfizzc3N4auvvkIkEoHNZoNOp3vjAGKz2bC2toZer4fHjx+zMrrX692IDTHNO1arFfPz8wgGg2z/\nlc1mkc1mWXFOXq50ACUfdFLClkolVu3XajVUq1WUSiUOpFerVVaoTyYofmmc6/X61NlQTCr95+fn\nMTc3h2AwiGAwCLPZzIkwetDaS9fmcDiE1WrFwsICJ3AajQbbBBSLRRQKBWxtbeHZs2dTVwk26dNM\new/67JVKBXt7e3j27Bn29/c5KUh2NpNnjYvodrs4PT1FuVzma7Pf72NxcRE6nQ4WiwVarRaJRAI+\nnw+ZTAa9Xm8qVJ5ULWexWBAKhbC0tIRWq8X95PL5PCtXBb8M7TN0Oh0cDgei0Si+/vprPHjwgANW\n1FcEeLP3ITVYVqlU6PV6fPhXqVSwWq3w+/3Q6XTQ6/UYDodcrT6NexW5XM77k8mqyclEwiTlchnb\n29vY3NzEy5cvsbW1hfX1dZRKJaysrCAWi731tYKbSzabxePHjzEcDmG32xGJRACAE9VkUXuRPcxN\nhgKaBoMBKysr+OGHH7C4uMhxIhIF7u7u4vHjx4jH4ygUCiiXyxwkJ7vNUCiEb775hsWIPp8P2WwW\n+/v7bNfk8/m4BwfZRV63NYcSzCT6+DXb0ElffVLqUyxosv9PLBbjHpDUc4q+R1ZNZ2dn3JupVquh\nXq9zRQFZVk/D9UtJA6/Xy/ZdpVKJrXr/8Y9/4ODggAPmFKOgpFU6ncY333wDh8PBNmEWiwVKpZJt\nSK1WKx4+fMj2agCmZh/7IZAg2mw2Y21tDevr6zg7O8PTp08xGAw+eI4ja7pqtTo7SQi6SSnLQgo5\nGihq5KXT6djugL5P2TNqNkeZ219rYEgHDvrd9Mcpl8uQy+UcsCf13XVmPB7zJrTT6SCfz3Np3GST\nvItuQI1GA61WC7fbDa/XC7fbzeVMNAGQ9RX14hC8CZXo0eIslUrRarWQy+VwdnaGWq020xOgVCqF\nyWSC0WhENBpFNBpFKBR6p1K4yftRpVJxcFwqlaLX66HVasFqtbICYHt7m306pxnKME82aSJlfqlU\nwvHxMSsqPwRSbZBKjErqblICAjgfRPZ6vYjFYrDZbNz8HHiVjC4UCjg6OkIul5tqlRhtfKkKwuVy\n8UatXq+z7cPx8TGy2ewHf1Zap10uF/fFiUQicDqdKBaLSCQSSCQSbFUybWsJBdT9fj8ikQjm5+c5\nWHSRdZDRaMT8/Pw5T1hSnlM15ywmJGhOod5A4XCYr4XDw0NIpVI0m01uTNjpdFCv17mxL12LpFSc\nVDQVCgVUq1VUKhVOaJAQpd1uv9HQdZqZbGRLfUZsNhucTif8fj9isRhisRj8fj8CgQD0ej0fiCmh\nMxwO0el02EqBGpFSIz9qSk/VEPl8Hvl8HoPBAIVCgSsFpkUBT5+J+hqQmrzT6SCZTGJnZwdbW1tI\nJpPvnQQdDoc8TgaDAWq1mtdvj8cDpVIJrVYLp9MJn8/H9qXTgFarhcPhQCAQgMfjgcPhQDqdRrFY\nRKPR4PnqMrzeZ2Jy30FN4ycrUKZ9PlQoFFCpVOcqAW/fvo319XWez3q9HvL5PO9hJz8znWG73S5X\np1gsFjgcDrY/USgUOD4+htlshkajeW/rus8N7X1pLxYMBjE3N8f7BZVKdS5AR4lnCnY+f/4ce3t7\n2Nvb4/V10rqZziJ0jtBqtfD5fFhYWODE9U0QhZEl86Q4jqB7PpvNotlsTvVe99eo1Wro9XrQ6XQ4\nOTnhHiEmk4n7tWg0mpmwJPyYUFDX5/NxPzmfzwej0YjRaIR0Oo1kMomHDx/i4cOHiMfjKBaLqFar\nHHuiys18Ps9JHwqaj0Yj7gkbjUaxurrKNlmZTAbxePxC+/DPCfWrbTab3HCbKq7IwvB1gSQp+V8X\nhUxisVh4faSfRb/P6/Xy7zo9PUU6neaKE1qfyTEin8+fs6u/juj1evh8Png8Hk6m53I57OzsYGdn\nBwcHBzg5OXnjde12m6shyCaI+pGQiIKcDcjOqlwuQyqVolqt3oieVmT/6HA42GHIarWykISqb96F\nSVEP7TFIWPOpE4RXmoSoVqvY3NyETCbjhnrkGTYej3mRoIw1LR4ajQZ6vZ4vtm63y6VSF/lD0kRG\npVG9Xg9KpZI9rvr9Pmw227nmRPl8fmoOEgA4aUA2SjQhXjQZkfccKZ+8Xi+MRiNbSpAXaSaT4cnt\nOk9snxO1Wo1AIIA7d+7A7XZjNBpxoIQO87M8bgqFAgsLC9xUzm63A8A7BblJBTt5ECUlJwVPHA4H\n7t27B7fbDZPJhFKpxJ5/0wqpKaih5ddffw2Px8PVTC9fvsTm5iby+fwH/R5a8El1cFMhRY/NZoPd\nbofT6eTGosCr9aHf76NcLiORSKBYLE7toYQsCMiaLxqNsndyrVbD9vY2tra2sLu7ywe0DzmU00Ff\np9NhZWUFX331FTY2NuD3+zEcDlEulxGPx9lXnGwWpwnyfb916xYHfX/J5oEakdLmeHl5GYeHhzg8\nPMTJyQlOT0+Rz+enSmn+LtBBKhQK4cGDB1hdXUUwGITD4eC5jWyWBoMBarUahsMhHj58iFqtBoVC\nwQ3hSSRBfYEajQYnLqiMncQXVFUxK+NJhwmj0Qi/3w+/389VD9TPxuFw8L5YIpFwJQgFy2mMqE/E\n5P5Wq9VyoI4C6FQeT+vrcDjE6enp1CQhJgVNtAcuFotIp9N4+fIlXrx4gf39fZRKpQ/6PdTUVaFQ\nIBgMYn5+ntcSWm/9fj/S6fRH+mRXi81mw/LyMlZWVnjv9rGQSqUwGAwwmUxcnUNVsZM2DJTwmrbk\n9CQSiYRViEtLS/jhhx+wsbHBTcwpKZhKpbC7u3vOoo8gOwmJRMIBc6VSiXA4DJPJxD2GSKDzNovd\n64xKpeLq+5WVFbY1NBqNHKSjsWq1Wnj+/DlXMB0fHyOZTHJAJZPJYDweo1Qq4eTkBHNzc9jY2MD6\n+jo3tbbb7fjmm2+gVCrx448/4scff5yaOe2yjMdjZDIZPHv2DOPxGJFIBC6Xi7/vcrmwtraGwWCA\n4+PjqZmrLgPZLVUqFSQSCWxvb2N+fp6DlxqN5ty8JHiFXq9HJBLBysoK5ubmuFE8qfP//ve/489/\n/jOLRGjP0ev1OBBPtl/ZbBZPnz7lPRxVYUqlUszPz+P+/fv49ttvIZVK0el0cHh4iD/84Q9IpVIX\nBvY/F/1+H9lsFoPBAEqlEsPhEH6/n6+h16twJy1tyOnl9fgkrRvU13XS2oksnCi2SfZ+NIb0yGQy\n2N/fx8HBAY6OjnB0dHRt11K9Xg+PxwO3282imLOzMzx58gRHR0dvTRZ0Oh0Ui0UcHR1Bo9Gg0Wiw\nGId6JZGjhlKpRLPZZHurly9fIh6PX6uE1lWgUChgMpngdrvh9/sRDocxGAzg8XiQSqW40uhdoEpF\ng8HA1yXdz+12+5MKva48CbGzs4NqtQqbzQar1cpVCePxmBMT5G/ldrvZU1Ov13Ngrd/vc9Dyoqau\ndPHRINIh12g0skKg3W6zWjGdTp9TDkwD76oAJHWS2+3GrVu3sLS0xFlJqVTKjcFzuRw3MCwWi2yT\nIDiPSqWCz+fDxsYGL9K1Wo0tIyYbHc4aEokEarUaCwsL+M1vfoOlpSXYbDb2uX0blCTrdrtss0YK\nTvLKJUsFtVoNm82G9fV1VKtV/PTTTzg4OJjqgBMFJ4PBIO7cuYPvv/+eVQ2T3tUfupEgtRnd3zcV\narJEG0GbzcYJCNrkdrtdlMtlpFKpj9If4XMx2aPG6/VyEkKlUnES4i9/+Qv29vZQq9U++BqjJITV\nasXi4iJ++OEHLC8vs+qzUCicS0JclwPFu0IHgVAohI2NjXN+uADOzUOkeqJDBVly9Ho9PHnyBI8f\nP+YNMgVDp2083gb1nlGpVIhEItwI3WKxQKVS4ejo6NxBfzgc8pzXarX44ERl5SQymWxiN63z/btC\nSRzqE+R0OrG6uopbt24hFotxMIDU5bSvJZVYuVxGOp1GOp3mkn2v18uey6SYs9vtUKvVUCqV3OtE\npVLBbDaz8pgCBZlM5nMPyztB1W5UnTQajVAqlXB4eMgqu6Ojow/+PVSFo9FocO/ePdTrdfbO1uv1\n8Hq98Pl82N7e/gif6uqxWCxYWlrC8vIybDYbgJ/XxPetTqDrlxJB5E3scrlgMBig1+u5R8lgMGCh\nDp33KJE4bYF1qmAwGo1wu91YXl7Gt99+i9u3b/NnooTL8fExfvzxR+5zNnlmU6vVXLFK4xEMBqHV\navl8Rgmbae2PRnsxSkJ88cUXCIVCnNgHwK4I5XIZL168wH/913/h8PCQKwnpmqR+Jaenp9jb20M0\nGoVcLkc4HGZVsc1mw5dffgm3241ms4mtrS2uFJvV9WQ8HqNYLGJvb4/3vZNJCLvdjuXlZdTrdVZZ\nzypUKVOr1ZBMJrG/vw+bzcaB3UnR20Wiktct02YdCoKbTCZEIhFsbGwgGo1ygrparSKRSOAf//gH\n/uM//uMX7ZHpe/l8Ht1uFycnJzy/k7Xk3Nwc7t69ix9++IHnPUo6ymQydi25DuNPVaKlUondUjwe\nDwujaT9L6xeth5RIjkQib8QVKVButVqh1+uh1Wp5jQR+rq6zWCx8viCo8jeVSuHp06cwGo1sHXld\nkxCTPSEAcLx1c3MTJycnb01CUMKFEj20x6UKYKq6sdlssNlsLGKXy+XcY3OWrLwvgpIQHo8HPp8P\ngUAAzWYTDocDZrP5vZw1SAhFPV2Bn9flTy0YudIkRK/XQ7FYRL/fR7FYhFarRa/X4zIsrVbLVQ86\nnY4bwlATYKPRyIkDKh+7KNg22SCGNoMrKyu4ffs23G435HI5xuMxut0uarUaJypmDbpIrVYrVldX\ncf/+fSwtLcFqtXKmixRwz58/x9bWFpfoT9vB4KqRy+W8obZarXA6nQB+XhjoMavjRupyakBttVp5\n0v+ljDPZ3tBiXiqVUK/XOQlBanUqP6eDMfDK5iMajSKfz6NQKHywXdHnwmq1Ynl5Gaurq3A6nZBI\nJEgkEtja2sKTJ08++J4jKw+j0QiXy8XN/ma57PqX0Gq18Hg87J8+uZEjn8NZsU4zGo1YWFjA4uIi\nbt++jbW1NVitVlbMJBIJHB4ecmL5Q5lsamq322EwGDAajdjLdGdnB7u7uzg7O5s6BSKVo9psNgSD\nQSwuLsLpdEIul/M+hRKow+GQlekUTKGEv0QigdvtxsrKCprNJlKpFBKJxDnV2DRDyej5+XnMz89z\nYEmpVCKTyaBYLGJ7exsnJycoFApv2On1ej0+vNG81+12ufx9lgNGk5Cdmd/v5wo2slxyOBwwmUwA\nXllMkFVSu91GOp1GPB5HKpU612i62+1ifn6ey/1fvnyJ7e1tzM3N4datW3A6nXxIpgolp9OJO3fu\nQKFQcBUT2XteZ2w2GydsHA4Hi2moovljV7Z1u10UCgWkUilIJBIOHlutVthstgub615H9Ho93G43\nCxVGoxEKhQL29vaQTCbfy8aADr50rZpMJm6irtFo2MYKeHUmo0RZMpnEyckJkskk0un0B1eAfkpk\nMhk3hV9bW8Pdu3dx584d2Gw2Pm/WajW2P9zd3cXm5iYSicQbKl9KZFksFkQiEYTDYa6C6Pf7ODs7\n44B7IpHgXkPTBFmEulwu+Hw+BINBWCwWvi7IYjiVSuHw8BDHx8fI5XKo1+sXWhiORqM3bFx2dnYQ\nCoXg9/uhUqlgMBjgcrnY235/fx8nJydT5XTwPtB+Y2NjA7FYjNeNmw4FcelBZ1ir1crB7tcZjUZc\ndTmNCdL3xWQywWKxYHFxEWtra7yejkYjHB8fs5hmZ2fnjUqut0H3KNkuq1Qq3mdsbGwgEolALpej\n0WhwD7lqtXquN+x1gZLKpVIJR0dHKBQKnMSiatTXRUlyuRwnJyd48eLFGxU3EokEVquVHzab7VwM\nkxIUBoPhDZE1Jb8NBgPC4TD6/T4SiQTbi13HfXOz2UQ2m0Uul+PPSbHZd3m/VEENALu7u5BKpeh2\nu3wvU/KZKoCLxSL3K6F403Ubk48F9ccgZ5vXqwrfZw+v1+sRCASwsLAAs9nMe5lisYhKpfJJ9x1X\nmoSgbueVSoVLqekmB8D/RhlFynYZDAbY7XZYLBYUCgXk83mo1Wo4nU4YjcY3fs+k2pUaa/zud7+D\nx+PhADxtfuiANwvBgdehBjnk233//n34/X4uAR4MBuh0OkgkEnjy5Albwky7x/JVIJfLodVq2Z/O\n6XSyjdVkImJWJzzKzlPDW4vFAp1Ox+oF4GL1yGAwQDabxe7uLk5PT5FKpVAqldDr9dDv9+H1etm3\nn7zYCbPZjLm5ORSLRQyHw6lNQpAFwurqKlwuFyQSCU5OTvC///u/2N3dRS6Xu/S1Q5selUrF3uwO\nh4MX65vUC4LQarXwer0IhUJvHMiol84sJSEWFxfx7bffYnl5GUtLS+j3+2yvl0gkcHR09NES7b+U\nhCAFMlU7Tts6QnO8zWZDKBTC4uIitFot5HI5z/UkWuh2u/B6vdBoNG8kIWQyGZcfkwWlSqVild40\nQ8q5yYq41dVVhMNhjMdjpNNpvHjxgpMQ1A9iEvL0ntz7dTodDjrN+sGfcDqduHv3LtbX1xGJRBAM\nBllBTpYRlDTNZDLcE2N7exsPHz7EwcEB9wGjw3uj0UAgEIBKpcLTp0/xt7/9DZVKBTqdjg8tGo2G\nr0WHwwGfzweNRoOdnR2o1Wp0Op1rv5ex2+1YXV3F2traG0mISqVyZUmIZDIJg8EAj8fDVZvTlITQ\n6XTweDxcMTOZhDg9PX2vqkCv14svv/wSi4uLCAQCcLlc3NOEmkhO+l1Tomxvbw8vXrzAixcv0O/3\npyoJQbYFDocD6+vr+O1vf8uBdbJ4TKVSePLkCX766SccHR3h7OwM5XIZwPk9MlkV9ft93Lp1C2tr\nawgGgzCZTFwlSz0REokE6vX61J1T6dzgcrng9/sRCoVYNAP8HPRNpVLY3NzkHl31ev2NdYDETmTJ\nJ5FIOAlB4gFq5C2TybC6usquCZSsnUUmkxDz8/PnlNWCn1EqldzIlgK6r9Pv91GpVPhMP+uWLiaT\nCcFgEEtLSzwH0R7s8PAQv//97/H3v/+d4xzvAnnRk5+/RqOBy+XCnTt38N1338FoNLJinRowX9fz\nAv39y+UyV0FS5d9Fe1U6b9Mcd9H5m1xgSME+GfOIRCKIRqPweDxccTn5s+n8FQ6HeY+n1Wq578Z1\nu1YpCZHNZjmuS+Kjd01CUI8HqVTKfwOyLaVKWOr5Uq/X2XKOrNKnbc18Vyhe5vP5YDKZuD8wiZXe\nJwmh0+kQDAYRi8VgNpu5EX2pVEKlUpkdO6bJ5mTvAvldazQabhJHSQWlUsmHq9ehDPbkBDEejznp\nQQ0Sc7kcEokECoXC1ClM3gVSR1MDOioBo74Y1KDw5OSEN8uNRuNzv+1rCTV+JW9mpVLJ2cbrmoX+\nGNA96HA4sLKygrW1NSwsLMBqtUKj0ZxTNAHgng+dTodtSF68eIFnz57h7OwM+XyebWHIyoqaxJMK\nlBYrm82GlZUVDIdDyOVyVtPR66/zeEulUu5xEwwGEY1GOQFIwRJqnEaVYJdBoVDA6XSyL6DBYAAA\nHntSlM3q9XkR5KfpdrtZfUGbyVqthqOjI2xubiKbzU79BoXsmIxGI1cQUqCcqgCbzealK20mLYfM\nZjPb+lHARKfTodFoYHd3F//4xz9weHjI19y0Qc3MqfLSYDCg3+9zSTpVeNAcFwqFEA6HufyVLCbI\nKlImkyEYDGJtbQ31eh3xeBwnJyesQp+m+5GSD2TF5fV62Vve6/VCrVazFdfTp09Zedrtdt+4x6gR\n6a/926xBSSqz2QyTyYTV1VWsr6/j1q1bcLvdcDgcrOyipFcul8Ph4SGOjo64QWE8Hsfh4SHOzs7e\n2OeSlRUFAiqVCjdqlslkWFxchMVi4dcolUq2KlWr1W8csq8b1ITaaDSyMIlUiXRo+hi2c69DAqqz\nszP4fD6MRiNu5vdL/WKuCzSPUx8ChULBFhPZbBb7+/tvrYSYFDro9XquWt3Y2MDGxgbC4TALU0hE\nRtU0o9EIKpWK+xYB4Ouz2+1yVc80VN8Ar/oFUbP4ubk5+P1+DgKVSiXs7e1x9cP+/j4ymQxXKr0O\n3WvUsJUCUiqVCsViEblcDsfHx8jn85/cl/ljQQ3cPR4PzGYz+01LJBI+I6TTaezv7+PZs2dIJpO/\nKgqkOEKz2cTJyQlb7LhcLm7KTAGaUCiEw8PDqbNbfl+USiXbu9xkJi0OnU4nJwipD+nS0hJUKhXb\nPr7eKL7T6SCXyyGbzXKsqV6vo9VqXXgPTzsajYb75pGwkM4NxWKR+4S+795s0o2ExCV0hqBeMNTX\ntNlsXlj1dJ14n7jlr0Hi52q1imKxyGd2ACwcczqdXC1B1YVUeU2JcFoz5ufnIZfLkcvlrl1DZqpa\nq1arsFqt0Gq1fBanvzdVNZAAR6/Xs0vN5D1H4plGo4FsNot0Os0iENqHGQwGribJZrMzKcKks5hG\no4HNZoPX64VUKkU+n0c6nUYul0O5XH6v+YoqPEmsRIkiEgt/SmHYterWQ5MYqbaq1SpPauQZdlGD\nIbq4KXjqdDpZJSGVStFoNJDP55FMJnF4eIhMJjOTC4xcLmcLHbKNoA1gr9dDPp8/1wDsItWi4BUU\n2HS5XBzYHAwG6Ha7M2OzcRHUyNLn8+H+/fv4/vvv4XK5OAnxupqEfOSoSeTJyQkePXqEx48fo1Qq\nodlscoO98XiMVquFSqUCo9GIlZUVdDodKJVKyGQyOBwOrK2tnWsMe3R0xEpa4Pp6d0qlUg7WRaNR\nPqiPx2NuZJ7L5T7YOkKhUMDv92N9fR3hcBharRbdbhelUokVeJQVv65j9bEhhSrdq5OqlXK5jJ2d\nHTx69IiDeLMEHdCpoe+kSvoykPrGarUiFothaWmJy7bdbjd0Oh2SySQ2Nzfx5z//mfviTCPkkz9p\nNUf9fp4/f47/+Z//we7uLoBX4zI/P4+5uTm2w6L+ERQolUql8Pv9ePDgAbRaLf7617+iUChMxaHr\ndajhntlsxsLCwrkmozqdDhKJBOVyGUdHR3j27BlKpRJX30zT57wq6OCgUqng9/uxsLDAVRDRaJTt\na/r9PrrdLrLZLI6Pj7G/v4/NzU1sbW3xPU02BheNrVQqhVKpZCscqVSKQqGAzc1NSKVSbl5P9hQU\ntJkGqO+ATqfjpCspBamnxaTd48ek1+vxmlqr1aZu3aC/MwW+yfqs0+ng7OwMe3t7F/ZHovMC2eoE\nAgF8+eWXePDgATecJLUr2XxR4pbU/2SlS8Fiu93OVTt7e3vQ6/U8V1z3cdVqtYhEIrh37x7C4TAn\nqvP5PI6OjvDo0SPeW5Co6237O4VCAb1ez5akJHCSy+XnKm+q1eq1H5e3odPp4Ha7ea8A/Lxnb7Va\nLMbZ2dnB06dP30vAQL7zlUoFFosFc3NzsFqt3PtRp9PBbrfDaDS+0SBWMJuQg4bZbEYwGMTy8jJb\najqdTty7dw+Li4tvdS5ot9tIJBL8oOr9s7OzmYwR0bxuMpl4Le10OiiXyyxm+pCqSIoHXJREpUDn\npw5yfm6o0pTiJJNz08HBAQfkjUYjHA4HFhYWEIvFEAqFEAqFuJKEEuLr6+tcCXrdkhDUn6VWq6HX\n612YJFWr1WzVFwgE4Pf7USqVEI/Hz7lfKJVKKJVKjMdj3h9LpdJzFt6UoJns/ztr0FmM+m14PB52\ntInH4zg7O3vvWC7t8yYFSJN9uz4l1yoJQeq44XB44cbklwZZKpWy3+bi4iJ8Ph90Oh03m4nH47zA\nlMvlqVRuvg3qq+Hz+dhn1G63Q6FQ8FiSd9rm5iarbUQVxNuhsmK73Q6tVguJRMKb6GKxOJMbFACs\ngLNarYhGo7h16xb7PE5OWJQxzeVyyOVySKVSODk5wfHxMTY3N7G7u4tms/nGQbPf76PRaMDtdiOV\nSiGbzbK60WAwsLKJyq/JPoC82a9rgIsCt5FIBKFQCB6PBzqdjq2ACoUCqtXqpX3zJ5XJPp8Pa2tr\nXGlBvuGHh4fcJGwaDvgfymRvDEo+08GXLPrq9ToSiQT29vZQLpdnakwmvUknNymBQODcOkqb4LdB\nilm5XA61Wg2VSoVwOIyNjQ2sra0hFothYWEBcrn83Hq6t7f3qT7qlUBqXZpzpFIparUaEokEtre3\n8eTJE2xtbbECulKp8Fyk1+thNpt5EzzpVUr+rXt7e5xMnTaFjlKpZCuXpaUl3L17F9FoFFarFc1m\nE5lMBvv7+zg8PEQ8Hudg5HWdnz81VFpvNpsRiURw584dtrEiD+bhcMgJ6oODA2xvb2N7exsvX77E\n7u7uL1o+kgKdBCa9Xo99XikZQUIJujYBcK+Ter1+bp24jn832tM7HA7Y7XaYTCbeG1ACgnpCfOwk\nBCmvK5XKB1Uufg5on0CHcwo2NZtNlMtl5HI5pNNpThrQa0htp1ar4XA44PV6sbi4iK+++gr/9E//\nxGsD7f9IOVytVllNPB6P2fuaKiiUSiWcTif8fj97YlPlznVFrVZzVSv5p1MFWKVSwcnJCV6+fImt\nrS1sbm6iXq//qn2SRqOB0+nkvhpGoxEymYwDN9lsFqlU6r2sUK4LFNTQ6/VwOp1cpUCQACmTyeDk\n5AQnJyc4PT19r99BFq10lk0mk3C73VztQ/0hLhJLCWaLyQbLDocD8/PziEajbE04GAy4ut5kMr1h\nB0MNgQeDAaxWK68x9NVms3ECularnXv9NK0Fr0M2vkajEUqlEqPRCNVqFalUCrlcDq1W64POSJT8\npga2drYAACAASURBVD3IZLPlyQD1TRK/0t74ovXudSt6q9XKew6ZTMYW9CQOU6vV3D/iOs5xVAlB\nSYjJqkqVSoVutwuz2YxoNMr9QsLhMAqFAhwOB3K53Bs/k67VyYpyuqYoVuXxeJBMJmEymdixZFYE\nUZN/d6vVCrvdjpOTExY+FItFNBqNd/qsk31MqIJwOByi2Wz+apzgqrhWSYjLQhYyZrMZsVgM33//\nPWKxGAwGAxqNBo6OjvDkyRPuzj5rmVi3243V1VWsrKxgaWkJS0tLcLlcXIaYyWRwdHSEp0+f4scf\nf3xvL9ibCJWMWSwWPsTl83m8fPkS+/v7qFQqn/kdXh2UeSWFpUwmOxdEI+VbpVLB8+fP8ezZMyQS\nCRSLRRQKBeRyOXQ6nQsXAVIEU98Ii8WCWCzG6iWJRAKTyYSFhQXodDpUq1UcHx9zKe11TR5OJiFI\nBUbNBre2tpDJZD5ogp8Mavl8PiwsLMDtdkOlUqFQKODw8BBPnz5FKpWaOtX1ZaEgCynvHA4HKy9I\nddNut1GtVlEqlbh52qxBzUdJZWS323njS0mwX/JHVigU3GTUarXCYrHA7/djfn4eoVCIq6CoSiCZ\nTM5EApsOsjS/DYdDZLNZbG1tIR6P88aOrqVCoYDBYACj0cie/lRuTD+Pgst0aKUy9Wm77sgHPxaL\nYW1tDWtrazAajeh0Otjb28PDhw/x5MkT7lMwjZ/xKlEqlXwfxWIx3L59G+FwGDqdjhtzt1otbG9v\n49mzZ5zMSSaTKBQKv2jhRVUNpPyi4KXL5cK//du/Qa1Wsy+zx+M599p6vY5sNsuqs+tc1SmTyWC3\n2zE/P49gMMjCmnK5jEwmg1QqhXQ6fSU9IaYZqioNBAIIhUJsm0TB83K5fO5vThU7KpWKe3bNzc1h\naWkJCwsL5+Y56llCiQxSEBeLRf65Go0GZrMZd+7cwd27d7mXECVGtFot2u32tU7MulwuLC4uYn19\nHRsbG4hGo9Dr9RiPx8jn89jc3MTjx49xfHzM1a1vO1PSwd9kMmF+fh7r6+twu92QSqVcGUzq60wm\ng0ajMXXnU7lczkIiu90Oq9V6zt8ceDX3nJ6e8ph9CNVqFScnJ3A6nSw+FNwcyI5lbm4O9+7dw507\ndxCLxaDT6bgyvNVqodVqcSUhNbwFXjUE9ng8nAh0OBzQ6/WYm5tDo9HgHmtU1U+9ROk+n7b7kyAb\nL6rgHQwGyGQyeP78OY6OjlCv1z/o59NZhKrSKRkBvLr/k8kkTk9Pz/0tbjIkvJ5cP46OjjAYDLi3\nJD2n3W4jm81yFeN1TOJ3Oh2USiW2ByJrX6vVina7DblcDrfbjbW1NXz//fewWCywWCxwOBxwuVwX\nVnZQlRsF4Sf3DTqdDoFAgPclJEwpl8ts83dd97fvCtkGk0WXRqNBv99HoVDgve+7nr0otkeWrFSV\nWi6XP5urwcwkIagkLxaL4bvvvuOGfOSx++TJEyQSCbRaran02vwlXC4XHjx4gK+//hqBQACBQIDL\nsAuFAk5PT/Hy5Us8efIEP/74I6vKBW9HqVRyYI6aEBYKBWxvb+Pg4GCmm55R1p2SEK9D/UWSySQe\nPXqE3//+9zg+PuZmo78EBUcpCUH+9nNzc2wXYTQaYTAYEAwGcXh4iJ9++gn5fJ4Vn9cRSkJQkymd\nTodms4lUKsVJiA+550jpbrFYOAlhMpmgVCpRq9VuXBKCNjcOhwMej4f74BBk7UeH/EnV57QzWTJJ\n3qtarRZ6vR5GoxHz8/NsYXh4eIjNzU2cnZ299edReazX64XP54PP54PH44HL5Tp3uM9kMkgmk0gm\nkx98WLkO0Dw3mYTI5XLY3t7mJATws+UVqTAtFgs305wcH1LJDYdD7iMxrWocg8HA8wz1BCG17+7u\nLv7whz/gyZMnaDQaYi9xAeRRTkmIjY0NmM3mc9Wp1WoV29vb+NOf/oTDw0Pkcjk+nNNB6/XeA5P+\nsHq9nvd46XQa6+vrWFtbO+eT/rotSa1W44pFKuG+rgmk15MQNpsNSqWSg5mpVAqZTGamBSGXgYJq\ni4uLnIQg+7REIvHWJATtuVZXV3Hnzh188cUXiEaj/Lxms8mBkN3dXWxvb2NzcxObm5uoVqvn5gKD\nwYB6vQ6DwcB7OlqntFotarXatU9C3L17Fw8ePMDy8jIikQjP6blcDltbW2w5+msBNVpnzGYz5ufn\nsba2Brfbzb1gSChA1/O0QcpKtVoNo9HIjVhfb94+mYT40HuWkhDUAPui9ySYTSavt2g0it/85je4\ne/cuzGYztFotGo0GCoUCB0MLhQIn+IhAIIDV1VWEQiHu0UKWOLRm5nI5aLVaZLNZVp2Px+OpFrFS\nEkKv17MKOp1O48WLFzg+Pv6gfb1EIoHRaITX64Xb7YbBYOA4FCCSEBdBZwaKbZAopFwuIxaLodVq\n8fmj2+0il8vh4OAA6XT6WiYh2u02SqUS27MCr86XFosF7XYbUqkUXq8Xt27dwnfffXeuovcy6HQ6\n+P1+yOVylEoltrsiMdhFPeqmDaVSyb0LaR81mYR4n6oiqoCYtDgtl8tIpVIolUqfJb42E0kIyoYt\nLy/D6/VyqRL5sNXrdVQqFXQ6naldPC6CLDSoJJGy+VSKT0rYnZ0dPH/+HGdnZ9da+XYdILUWKYEj\nkQiUSiUf9JPJJJct3jS63S4fQqkCYmtriyev97mu2u02crkc4vE4IpEIcrkcNzKaJj/XyTJJk8kE\np9MJk8kEhUKBdruNZDKJly9fIpvNflDyk6wBlpaW4PF4WJlIDcWy2SzOzs5Qr9dnao77Jch+iJqf\nTVIqlZBKpRCPx2dqw0vNGTc3N9kuB3i1FpD6mta+4XDIPYLetRLCZDJxufak9Qb5MZNF0aR357Ri\nMBgQCAQQDAZhNBq5vN/v9yOXy701cUOCB5vN9kagpdfroVKpIJ/PT3WTeGpgSJ+RmtqS4pmUgSIB\n8XYmrQhe//vT/9McBoAbIJO1yORzqTEy+ewbjUa+Z81mM9uw0X0rl8vP9QMga6d0Oo2trS28fPkS\n+Xz+2iYggPOJfafTCZns/7H3ns9tXlnycCPnnDMI5qAcxj9P2J2Z/TBf99/dqplZb8m2rERSzCAJ\nEDlngMjA+0HvOQYoKlMSALKrWC7LFik8us+9557u0y1CpVJhW9FUKnWz/i6BUCiESqWCyWTibDhS\nUlJo8miNoFAo4PP5EAgEsL6+jo2NDQQCAbYxqdfrqNfrODs7w9nZGSKRCPunk2XJxe85OjUxjcIv\nakzSJDS9g2SrQTZU72sA0Ptvt9vhdDpx584dzM/Pj2VX5XI57O3t4eDgAMVi8Rt+wqsDrTcKVtXp\ndGOBl41GA+fn52zZSnXqVYMaLAqFgm3FrjJg9gaTAYFAwNNW9CWTyVCpVPh8293d5WmIRqPx1iRE\nMplEJBLh+kYul8Nut8PhcHDOiEAgQCAQwH//938jn89zsy4UCiEWi3FO6aSen6OgZi/l/DidTp7s\nIrEWuRd8Duge7HA4cPfuXaytrcFkMkEoFHJD+Pz8nPfNWbWz/lJQzUOTnzKZbMw+x2w2MyFO62+S\nUKvVkEgksL+/D5VKhXQ6ja2tLUQiEVSrVSbwaLp8dFLmc0AByyaTCRsbG1CpVExAJpNJtuGnYPBe\nrzd1/RH6fGQNPxwOUavVkE6nUSwWP2kNUI/K4/HAYDBALBbj/PycJ4pvJiE+E2q1Gj6fj0kICk4k\nO45arcYb3zQcGB8D8vFWKBRjJAT5U48yp0RCZLPZqWbxvwUo3Mzj8WBhYYH9vfP5PCuBc7nctTxE\n2+02yuUyYrEYtra28K9//Qu5XI5JiE9ZV61WC5lMBjKZDPF4HJlMhke6p42EoAKBNnhqio+SEF86\nfUR+38vLy7Db7RxITSREOp1GKpWaykP2c0CTECaT6VISolAo4Pj4GOFweCZJCIlEAqfTiZWVFQ52\nJ0tCjUbD55zL5cLq6uoHMyEoX4Oal/TPwWCAVquFer2OSCSCra0t7O7uIp/Pf6uP/NWg1Wrh9Xrh\n9Xqh0+nYMsPj8SAej186BQa8UaYQCXExeI32SFK1TysJQc1xk8nETbhGo4FCocDq32azeS32ms8B\nkQ+jJMTF8DeaYiJ/eLlcDqVSCZ1OB41Gw9+LLvEikYhJQoPBwJMVRFAQeTjqxUwZTt1uF61WC+l0\nGnt7e9jb22MSYlJBnt7z8/McNkp5NDs7OzckxDswejZSqCUJsi67BxEJ8ejRIyYhTCYTZDIZ+v0+\nN/c2Nzfx/PlzhEIhtjxot9uXWpQQCUFk5bQJn0i9OWorRMKuZrPJTY33vT80teRwOHD79m3cvXuX\nSQix+M3Vm0iI/f39qSUhBAIB35uoHlOpVEyA1ut1ZLNZxONxRKNRJBKJKw9UpSaoTCbjfVShUFwa\nkHuD6QaR8XQOGgwGSKVSZLNZRCIRPHnyBP/+979RLBbfssckkN0wrReFQoH5+XnMz8/D5/PB4/HA\n4XBwLiJZvezv7+Onn35CrVbj+m7Sazvah8RiMQwGA5MQdEclEoKmIj8HZGvqdDpx7949rK2twWw2\nQygUMhFIU+kfIm+vM8RiMYxGI2eHSaXSMaGjxWJBIBBgS8RJc+QgESSRDOFwGIeHh4hEIhgMBjx9\nQ190RnzuJASREJTvt7i4iFarhVarhdPTU/z66694+fIlYrEY1z3TlutCgjCz2cyCsC8hIWh/oyzW\nRqNxQ0J8LsguhgLUaMxVIpFw0bOzs8MjyNMWMPc+UHFL5Av50NOmRSOJZJ+RTqfRaDQm+tL5vUHB\naqSIoOYANdrL5TJqtRqazebUXaquArlcDjs7O9je3sbOzg4ikQiazeZnTRjRqBwpMNrtNodwjjZw\nDAYD/H4/arUaFzKTBCoQJBIJN5DI+iCTyaBQKFxJE5xU7larFRqNBmKxmL1Ly+Uy6vX6tSPGRich\nSLVPBVAul0MwGJw5EqLdbiOfz0MkEmF7e5sLU61WC5VKxSpAakjK5XK2f6nX65de/imYqtFo8N5H\npA79vlwuxyTsrEyC0eWLFP0UhGuz2WCxWLgZTLY2ZCWyuLgIq9U6NnFJlzAanaUgXQrnnDbyX61W\ns80ZiTpkMhnUajXMZjPcbjfq9TqazSaazSaPks9KffWlGA1PJhsvkUjEFgVkr+d2u1lNSc0zjUbD\n62bU4kskEkGtVkOj0fA/KZi63+9DJpONEfi0vunCQh7XJycnyGQyV94IvGpQ9oVarYZIJGK/4VQq\nhbOzs7dshW7wBmSvpNVqOaSXghpHAxtHiWuv14u1tTVuflCGRC6Xw/HxMYLBIHZ3d7G/v49UKsV1\n8PtAkwNEYlMoKQWTTvJeIRQKWRAjEonGmhejdnuX/b5RUYpWq8XGxgbu37+PlZUVWK1WJndoao5s\nmCb9fXwXaBKCajG5XM4ky+g0Pu079Xr9ym0faK9QKBTQaDR8dvf7/WtXF1cqFSZ7ZiG76yJoOpcE\nOc+ePcPp6Sk7Fezs7IzZab4PQqEQUqkUUqkU7XabA+Kj0SicTid8Ph+8Xu/YVF61WoVYLMbp6SlC\noRAajcbEO0yQIGG0jhUIBCwkoabwp+7JJGAiO1i73Q6Xy8X1MZHYlONULBZ5QvgGv4PODRJG3b59\nGx6Ph0VOdO4QkdNoNCaSXB3N/BQIBJwdVS6XuS4pFos4PT3F9vY2T/RKJBI+c+kO+zGgKQoioHU6\nHa9lqVSKTqcDuVzOk5vpdBqZTAaVSuWdZ/ikgPpgRBx4vV6o1Wqu4QaDATQaDebm5jAYDMashUdz\n40bvBD6fj3vGJpOJcww7nc53E7BONQmhUqlgMBjg8/mwvr6OW7ducQhbPB7Hzz//jJcvXyIYDKJU\nKk1dI+B9EIlECAQC+I//+A/cuXMHDocDMpmMmcVSqYRwOIyzszO2hpjkF24SQCQE+aGTiqzT6aBU\nKo0VG5N8gboqjNpJAEAikcBPP/2EX375hdcUbYZX9bNGf6ZQKITdbsft27fR6XRQrVaRy+W++Gdd\nJUZVJuS11+l0uJC9qgY4FXoajYatOqhopkv9dQOpPSkbg2xHut0ustnsTJIQ3W4X1WoVvV4PT58+\nRSwWg9lshsFggNls5oCvhYUFbkBRs4MIaWBcod1sNpFIJBCPx/GnP/0Jf/nLX7j47ff7HGaby+WY\n8JqFs4Q8qu12O7xe75iik8I1zWYztFrtWAD6rVu34HA4WIXebre5iCZryG63i5OTE5hMJv5Z01R7\naLVaeDweuFwu9nXX6XQQiURYXl5GoVCATCZDOp1mRdZFteF1RqfTQbFYhEQi4feOLgkkFhGLxVhY\nWIDVasVwOOTpI2p+XlRsUROBrNeo0SeTyfgSQqD3OpPJIBaLsX1iKBTC2dnZ1NWDZAlXLBaRzWaR\nTqdnwu/3a0EqlUKpVEImk42ti1FIJBKo1WqYTCa2HyV1HDUyg8EgXr58iRcvXiCXy6FQKHAd/LFo\ntVoolUrI5XLI5XLI5/MzO0U1mt3l9/sxNzeH+/fv4/79+3C73dDpdGMWh9VqFYVCYarVwaMkxKin\nPvCmfkgmk9jc3MTR0REKhcInT01/DKjpQvkj5J89ib7pXxuZTAY7Ozs4PDycybwcUgJ3u108e/YM\n0WgUcrmcQ6hzudxHv0tElPb7faTTadRqNUQiEQ7TXVpawtLSEpxOJxwOB/R6PR4+fAi/348nT57w\nRH+lUpmKs+jiHfsqvpdMJmOBnNVqhcFggFKphFgs5rsY1R35fH4q7fm+NujcMBqNWFhYwMOHD+Fy\nuaBUKnmqoNlsIp1O4/DwcGKDqel9ot5QLpfjKRvgTR1H50G/34fL5YLb7Wb7PrVaDavV+tEkxGUg\nYajJZMKdO3fgcrmQyWTYRvzly5ecfzLJ65B6SwaDgae01Go1Op0OW9K53W62jyPSgUROWq2Ws24I\nlBtJk/zD4ZAFdF+SzfElmHoSwmazwev1Ym5uDj6fj5sC8Xgcr169wubm5lSovj4W1PAkBuzhw4dY\nWFiA0WhkVWa32+XNKhQKIZ/PXzs1yOeCNkGz2cwK0PPzc7bX6Ha7M01A0FgbHQqjKBQKHMj3pSDm\nmtTbpJ4a9QckW4HRTXaSQUUZhY6Wy2V0Oh0mBj9n3dD3lMvl7P2tVCohFAr5cv+p4UTTjNGxe71e\nD7vdDrPZzJYxvV4PnU4HhUIB0WgUyWRyJlT7BPL2JnVpOByGXq+H0WiExWLhoo5GTyUSCVqtFgqF\nAiu3LuL8/JwJa5vNhocPH7I6irJbTk9PkUwmZ2qt1et1xONx2O12VCoVDIdD9t+kppxEIuGRf7Ju\ncrvd0Gq1aLVaPJZP3sQSiQRmsxn9fh9erxcul4unvqbJOkapVMJqtcJisbC4gfbqubk5NBoNqFQq\nRKNRnrTMZDJotVo80UZFLU0njX4RqAAmpReR2tN+xtJFTCAQIBKJ4PDwkNeL0WjkZ6nRaGAwGN6y\naRp9BlTzEYlBaifCxSYzNThzuRxOT09xeHiI3377Dc+ePUOpVJqaJsDopYjeIcp3m8Xm2lWByCq5\nXM6N2csgl8thMpngdDrhcrngcrkgEAjQbrc5fHpzcxMvXrzAs2fP+P38FJAdB/mykx3HxX1g0kCN\nH1IHUh0mFArZQs3lcr31++idtlqtWFlZwerqKtbW1rC4uAiDwQAA3FCqVqtcv02zOpjqU51Ox81H\nwmAwQDabxf7+PgdSX/XeQ383ZCdJqmyVSjX2Z7kuoNycRCLxvf8oXwVk9dZqtVAul3F8fPxF34/2\ntVKphFKpxL+uUChQKBSQz+extLSEdrvNwoy5uTlUq1Xk83neY0ul0sRnkIxOPdD5SiHfVOd9CLQP\nSiQSyGQytg/y+/3weDzQ6XRsYddsNpHL5XB0dIRoNDpT94erhEQigVar5X7m0tISOx50u100Gg2e\nAo1EIsjlchN5XlAdTxPSF0HW5oeHh6jX6/D7/cjlclCr1TzJQDlShMuIs4u2SjRlIZPJeLJJo9FA\no9HA6/XyOSuTyVCr1bjGIQJyEglEuhuRRbDb7Wbhm0ajgcvlgkKh4PwQ+tyU/0LEoF6v5+9JZyS9\n+7VajQUERNzX6/VvOiUy1Se0RqOB0+mEzWaDSqXCYDDgAKJEIoFUKoVisThTDXipVAqTyQSHwwGX\ny8WKfalUim63y6Nvh4eHePnyJY6Ojm4ubJ8AhULBTW+JRMLjxCcnJxO78V8liImnYMJ3BWp+KeRy\nOWw2G3w+H1wuF1vKiMXisYZMKpXC9vb2xK7j0cZ3rVZj2w0iT2hygYrTT31+lJOh1+vhcrng8/mg\n1+uZHBtVFl4HjDL8Pp8Pi4uLcLlcUKlUXAC12220Wi00Gg1uis4aaN0Nh0NUKhW2Tcrn8wiHwzg6\nOsKTJ0/YiqPZbKJYLI5dsghUNI5OClLhm06ncXBwgF9++QUnJyczQ+YDb/I1SFVN47lke7a2tgaV\nSoVyucwkKfnxU+g8TdsEg0FsbGzg9u3bsFgsTOS63W7cunULw+EQ1Wp1qp4dTXaNKqmpcLVYLFhf\nX4fdbke5XEapVEIoFEI4HOa8iG63y7+fSDOyvhq9hJI1QK/XQ7VaZauOab+ojo7PHx4eotlswmQy\nwWAw8EWTgqTpnKVzhAIH6eKlUCigUqnGbJjeh1KphFQqhePjY2xvb2N3dxeRSITV1pPc/AV+J5pH\nM2q+h0JrWjHaJHpf8KNOp0MgEMDKygpsNhukUilnvuzs7ODXX3/F1tYWEonEZ0280lQPkYyjJOSk\nk4w0eVutVqHT6QD8vicGAgH813/9FxYWFt76fSSs0ev1cDqdcDqdsFgsY/lCZFGSTCaRz+fRaDSm\nfkp/lCi9+K7W63Wk02megvhaoPybm73iBleBbreLTCaDXq+HYrGIcDiM5eVlFn4GAgFIJBLs7e1h\nZ2cH4XAY+Xx+Yu+pZLtCQfFEOuh0OrhcLib33zc5ThNHJLahO/zGxgY2NjYwNzfHVpLAG8KV6uRk\nMnlt7qmfCoVCwYHBRqORc/7onh+LxRAMBpFKpdg+dlrPi06ng3w+zw4j4XAYMpkMYrEYCoWC884I\n1AMhQpkyzkafgVwuZ8siu90Om83GNp4kZhUKhSzclsvlODg4wGAwQKPRQL1en7jnOSq4VKlU/HmG\nwyFWVlaY8NdqtZDJZOh2u+h0OvzPWq3G/QGCTqdjy0SyovN6vVwj0lR/Npv9ZhlVM0FC2O12qNVq\nvuwnEgkmIQqFwsQtri+BTCaD2WyG3++H2+2G1WqFVquFSCRCs9lk242joyO8fPkSiURi5hvnVwW6\n8BMjSJ5y+Xwep6en14KEMBgMWFhYwNzcHLRa7dh/u8qLIx26Pp8PTqeTPSTpInEZCTFpIUwAxhrf\nREJQMJ/BYGCGH8BnNcMpkIpICL/fz5esZrOJfD6PQqEwU0Tr+yAWi6HT6TgPZ2FhgRWcFEDX6XTQ\nbDbRaDRmuuglBQcVHNlsdixP5eJl/F3TOBKJhFUj1CAi26dUKoWDgwP8+uuvrKKeFdCEWyqV4pBB\nqVTK7+7q6upbdjgCgQCFQgHxeBwnJyd48uQJ/u///g/1ep0V7jqdDmq1mkkIatJPE6j4pawVau4Q\nCWE2m3mtdDod7O7uYnd3F/F4HJlMBs1mk9WoFOJaKpVQLBbHvJop/JVsDehnzUK+xGAwQLPZxNHR\nEY6Pj3lc2uVy4cGDB7h9+zbnbgBgoqZer6Ner/PUg1arhclkYq/+jyEhTk9PsbW1hV9//RXb29us\nHJ2GZzqaszQ6Jn7TXPx4UD6NWCx+57PT6XSYn5/H8vIyrFYrJBIJarUaB38/ffoU29vbnzXFefHv\nEMDYpNOkg0iISqUCq9UKANxkDwQCrBi8CMolUCqV3Di4+OyJhEgkEmMkxLRi1JL0IgkxHA7HQjS/\nxs+mrxsC4gZXiV6vxzYu4XAYCoUC2WwWZrMZgUAAgUCAvdWpwd9utyeOhBi9T7fbbc5/o7NVq9Wy\nZU0sFnvv9xrNlbDZbFhaWsLt27fxww8/4PHjx29ZuvR6Pc7nuyEh3g25XA6r1Qqv18tB69QYPj8/\nRzwex+HhIdLp9NRMsr4L1FMrFAo4OzsD8LvAieqWUdcLmUwGhULBdxEAPGlBz0Gr1eLx48d49OgR\nC8gUCgV/T8qZmJub454KWaYOh0Ocn59PVF1CZyrdwy6Kj1ZWVrC0tMT/3m63eZKrVquxU8JF4RtN\nT1B2MJEQTqcTvV4PlUoFrVaLRYvfAlNJQtCCJQbXarVCLpej1WohGo1ic3MToVCIfZin4eL1IVCR\np9frMT8/j3v37sHtdjOjJRAI0Ov1WA1bqVTeUrfe4HKMjiSSxYtWq+WXMplMsq3VrJMQVJA4HA6o\n1eor+Z70fEcbW36/H+vr69yIuTjCTSPrxOxO6uV1NKiQRuzVajX0ej08Hg8ePXqE4XDIwaDvWz9k\noTA6UkjWWLdv34bNZnvrgjctysKrArH3dJCOKj17vR7K5TJSqRRnAF0X0N//56wDsVgMl8uFxcVF\neDweyOVyVCoVBINBbG1tsYf8NKioPwVEHubzeWxvb0Ov1yMQCGBubo5Dp0et1EglHAqFcHBwgMPD\nQ/a5pjFpmqAjX2qPxwOHwwGDwcBK0Gm4QBSLRRwfH3MDfFSZBLxNctntdgwGAzgcDiZ0aC+jCwNd\nfkenHGh/I7//0a94PI5YLIZarfbNPvfXAJ0R5IU+Oo5O0zUAWMHUarXQbrd5rNrlcmFlZYXH1UdB\n+3+r1UIqleLJpb29PQSDQfYO/pwpvO8FuVwOs9kMr9fLPtxCofCb7uck9vF4PPzzp+2sHSUPLvsz\nk/KQrDepThsNM/zU/Z7uKWq1Gk6nE0tLS1CpVBwQPA17H/DGV//Vq1dc29FnUqvVkMvlMBgMY4pf\nAhEvvV4P4XAY6XSa1YcUVE3Nzf39/Yn19v4UjGYpXcyE+JK65GNwkSCblnfzBtMBWk+jgbs7XWjL\nAAAAIABJREFUOztQKBTweDzweDwwGAy4e/cuVCoVTCYTtFotN1knZb+ju2I4HMa///1vVCoVbGxs\nwO/3w+l04sGDB2wjR1mG1WqV7+t0zxq1JyU3Do/Hw6IU4M0eSPsmCSpqtdpY0/gGb0D3fIfDgfX1\nddy7dw9Op3NsepCy64LBIHK53ERaB30KLvaDyMVltF85WuuRswG5ZAC/18q05vr9PoLBIDqdDtLp\nNM7Ozlio7XA4+C6iUqlgt9uZjBMIBNjd3eWcxUnBReKQ3iESlVC+Frkb0ARTtVplwdFltdvjx4+Z\n0KGJJrrLkQCZQuS/FaaOhBhliHQ6HauoR0mIFy9e4PT0dCJHbD4XxBBSSAmREKM+fv1+n0kIahpN\n0+Xze4EamxQoZ7PZoNFo0O12USgUkEgkEAqF0Gq1Zr6xSV5zDofj0kvW52DUP5IuYoFAABsbG7hz\n5w4MBsNboZrdbhetVmvig8DpoOj1ejg/P0e1WmV7OLp8WywW7O/v4+Dg4L1KcqVSCb1eD61Wyxde\nUqi73W7YbLa3fvZ1wyhhSAcogUiIWCx27UiIL4FcLofX68XDhw/h8/mgUCiQyWRweHiI58+f4+zs\nbCYsIy6C1Gu5XA5bW1vodrv485//PJaDAPxu85DP55mYef78OQ4ODjgTggpfk8kEk8kEo9EIvV6P\nwWAAu90Oo9EItVo98WFohEKhgMPDQ2g0GqysrLxFQoxCJBLxRCZ5iVImhFAoZAX0ZUpoUvvThbVS\nqXAz/enTp6hUKlNPQhDoHCuVSmi1WohEIm+NmV/MzRAIBFhdXWXlYbvdHlP8ko1TvV7H8fExXr16\nxWdNKpXiEOFpOivIqnF+fp5JCFof3/LPQMpEvV7Pa/RiXsc04F1/VlIEUh4XhaPTJfVdgdbvg1Ao\nhFwuh1arhdPpxMrKCodRTlMTivJt6vU6kw5OpxMqlYqbcpedhaPTcsFgEE+ePBnLLCRiO51OMwkx\n7VOsQqEQarUaZrMZOp3urRyGb/GufIhwu8ENvgRUt2SzWezs7KDT6XDjXqvV4s6dO+zIIZVKcXh4\nOFGNTcq1oT5GtVqFRqNhJwKz2Qyn04m5uTlEIhEWgJBdJN236Gy22WxMrBIx2+l0IJFIuOZrNpvc\nPCV19izdH64CUqmUCXsiIXQ6HYseer0ekxBHR0czQUJQP0gul3N/o16vo9frsdBt9L0hYc7oXZ9q\nMNrre70eTk5OkEgkEA6H4Xa7sba2hkePHrGVLvX4RmtuuVyOarWKYDA4cVP+lN9CLhvVapVF55FI\nBHt7e5zzmE6n2Wbtfbl6MpkMCwsLMJlMb1mfkz1nPp/H3t7eN/ucU0dC0DSAwWCA2+2G0+nkyy9Z\nEZ2cnHAROStQqVSwWCyYm5tjtabBYIBEImH7kXQ6jXA4zGNbdGG9wftB7D55yRmNRkgkElQqFeTz\neZRKJVSr1WtxgJKfOW3WVwGpVAqLxQKLxQKHwwG73Y719XUsLCywFzEVLsT8p9NpJBIJVmUQGTGJ\nINIkn8/j5OSEGXcKBlIqlXwxf9+eJJfL2cKE/g7IB5zIiVGM2jFNu5ruY0F2TDStNHqQNptNZDIZ\nHB8fI5vNzvzU0peCJkocDgfm5+dx69YtOJ1OSCQSzhtJpVKclzBr+x+tm0ajgWg0il6vx6pVqivk\ncjkXxcFgEJubm3j9+jWOjo4Qj8fHLqdHR0dQq9UwGAzQaDQQCoUwGo1wuVwIBAKo1WqIx+NT8a7m\ncjns7e1hOBwyOWA0GmEwGLiAJzHIaDPzc+0wSNl0fn4Oi8UCp9OJWq2GTCaDfr+ParU6cZeEzwER\nX51O54P2grT3y2QybnbIZLKxZ0y5JtFoFNvb23j58iXC4TCi0SiHrU8bKFxWq9Vy05dIrW/xs0ns\n43Q6mYQQCoWo1+ts9ToLxFij0UAymUQsFoPJZILb7YZUKoVer4fNZoPL5WIiq9FofNQ0qlqt5mBN\nv98Pg8GAVCqFTCaDRCLxTYmkLwFl14jFYphMJohEIvacVqvVYzZTAJh8JdUmWU0Q6UzKz16vx+Km\nZDKJcrk89WIJEoYoFApuQn5LjDZSaOrsczPYbnCDy0BrrF6vIx6P8zsNgAlGt9vNnuqNRgOhUGii\nar3hcMjB0Gq1GoFAgMkEEl/K5XKYTCbY7Xa4XC4Ww5E9jkgk4sYxAFZeG43Gsdwbqm9yuRwHDY8q\n1687qHY2m81wu91YWlqC1+uFzWZjtTvZO1M9l0ql2N1lGkH3Ta1WC6vVCovFwpPA6XQax8fHSKVS\nrOQfxYd6PzQ9Ua1W0W632VYIeDNJsbi4iIWFBc730+v1cLvdGA6HODw8hNlsRq/Xm6hpTTrLstks\nXr9+jXa7zSQETeLT2sjlcmi32+/sedA9rVqtch1H9q80VVEul9k291v2zqeOhJDJZPzSLi0t8ahN\npVJBPB5HMpkca5zMCoxGI5aWlrC+vg6/389hZ0KhENVqlZtvr1+/xrNnz5DJZKY+3PFbQSqVwm63\nY2VlBS6Xi5u9RELMQvPje0KhUMDn8zHxsLCwwHkmtIYBsAVTKpXC1tYWNjc3sbu7i1wuN/E+iL1e\nD/F4HC9fvkS324VAIOBcDavVCoVCAa/X+1FEClkakPemUqnkw3MU5XIZ4XAYZ2dn7w0TmyUQoTU/\nPw+z2QyJRMJNYrog7O7uIplMzhQJ/TWgVCrhcDiwuLiI5eVlrKysQKVSQSgUsr9tq9WaWPLvqkDe\noN1uF69evUK73eZRXp1Oh263i3a7jRcvXuD58+eIRCIoFotjE1rZbBaDwQAikQh6vR4ajQZGoxFG\noxFutxu3b9/mJnsmk/nOn/jDyGaz2N7eRjabRSgUwtzcHO7evYs7d+5wdg81ni76AH8OaNKTJl11\nOh0LAIRCIU5OTq7VOSwQCGC1WhEIBHDnzh2srq7C7/e/lQdRKpWwu7uLra0tvH79Gnt7e5zdctOA\n+3TQJdlut8Pj8bBHs0gkQj6fx+7uLl6/fo1cLve9/6hfjFwuh83NTfT7fej1eiwuLvL56vf7sbGx\ngXa7jVgshmg0ytaY72uCkDXJjz/+iEAgAIFAwBNkZF03DaAGQLFYxObmJlKpFDweD9xuN4xGI3Q6\nHRQKBf//tVoN9XodhUIBuVwOrVYLJpMJXq8XgUAA8/PzMBgMAMCXfWrOzfr5+q1AdSDVLe12++bZ\n3uBKQWG6tBcmEgk8fvwYMpkMHo8HgUAAZrMZkUgET58+/d5/3LdA03ypVAqbm5sYDofcoNVoNOxU\nQHkPYrEYYrGYFdnkF5/JZJDP55HNZiGXy/HgwQPOrALARGwikUCxWLyxBh8B5WqQH/+9e/dw69Yt\nWK3WMXvhbDaLg4MDbG9vIxqNMmE9rc9QrVbDarVibm4OGxsbWF1dZbeHw8NDKBQKiESiLxaQkziz\n3+9zxtXf/vY36PV6mM1mJsvpbud2u7k3k8vlJkYoQaR6OBzGP//5T2xubrL7QzabRTab5Unx993T\nhUIhZDIZ5HI5C2IBoF6vo1QqYWtrC1tbW2yZlk6nv+kddWpICGIOtVot/H4/7t+/z42ofr/PypNk\nMolCoTBzDXiDwcAkBPkQEmq1GqLRKA4ODrC/v4/9/X1W3Nzgw6CL19zcHCvXKSCyVCqh2WzeXOg/\nA1KpdMxa4e7du1hdXcXq6uql9h6dTgeNRgOZTAZ7e3v4+eefkclkUC6XJ17V3uv1kM1m0W63ucjo\n9/vwer08BWE0Gt/brCOfQgoQHQwGfICMBlUR6vU6kskk0un0RBM0VwWhUAiFQsHryWw2sy1Cp9NB\nvV5HKpVi78xJXzPfG0RCBAIB+P1++Hw+tNttHqOmYO9ps3P5VNDIc6PRgFAo5Byg+fl5GI1GHgfe\n3t7G7u4ue/2OXgZoWk6hULB1mkwmg9PpZIK7UqkgFAqxvdMkgxpllHURCoXQ6/WgUqmg0WggEAjG\nxqnJhoSU5HSZ+lhLF7JuImIDAAKBAAdZ5/N5xOPxr/mRJwKjF1QKNt/Y2MD8/DzsdjufAWTZVCwW\ncXh4iN9++w2hUAjhcPim8fYZoPVHBARNWdvtdlawU1Pg8PBwYpvpFMRIn+d9gd7lcplr2/n5eayt\nrUGr1bKV0traGvr9PmQyGStbK5XK2N2KFP4SiQQSiQQejwd37tzBH/7wB6hUKhaVHB8fIxQKTVxo\n67tAymeyEgmFQjzFa7FYYDabOTONFMbkpRyPxyEWi/H3v/+dLQ7dbjfEYjF7OH9M8+AG7wZZco5a\nY1HuUKVSQb1e58bnLIIyhrrdLr/vBJpmVygUPKFzg6sB1Yo0HRoOh6FUKrGwsACXy8XkNSnaJw1k\niUmTro1GA/V6Hf1+n/NVKdfFYDCwapruAZShRtOXZ2dn0Gq18Hg8HEovEAjQarWQy+WQSCRQKpVu\nCMERiEQirqNJ3LOyssITd6MK+N3dXezt7SGRSLwVMjxtUCgUHOr+8OFD/OEPf4BSqYRKpYJer+c1\nNhgMUC6X2Y7oU9FutzlHIZ1OIxqN8s8lcY9CoeDzmybV6d47KSQE1SCpVAr5fJ4nZAQCAX/Gj7lH\nUt+EbL7lcjnXNolEAtvb2/j3v//N067n5+ffdNJ38nbJd0ChUMBgMLBC59GjR7Db7RCLxcjlcuzV\nnEqlZnKzMxgMWFxcxNLSEoxG49h/o2DNZ8+eIRaLMVs66Y2OSQE1ejUaDSvzKVjp5vD8fJAn8PLy\nMhYXF+H3+2Eymd5p89TpdFCtVlEoFJDJZJBKpdgrcNJBdkDD4ZBVu6FQiBsZCoUCSqXyvaPq5XIZ\nmUxmzKaDFOputxtms3mMvKFiZZIzM64KFJip1+vhcrmwtLQEk8kEqVTKoZe1Wg3FYhHZbJYL6xu8\nGwqFAlarFU6nkxXWZIETi8WQSqVQKBSujQKdCjPyL89kMlAqlXwWJJNJ1Gq1S21J6N+psAuHwxzW\np1QqYbVaYbPZoNPpIJPJ2C5iUjEayEi2Pr/++ivS6TRPLCiVSlgsFphMJp5gsFqtPOVGo/yfC/Ji\nNxgMkMlkV/XRJhoymQw2mw12ux137tzBw4cPsbi4yJZA1FAeVSRGo1FEIhG+uN3g00Ce9iqVCn6/\nH/Pz81heXobFYoFYLMb5+TnOz8+RSCQQi8XYf3fSQGQA2TmSldW7ag5qZOZyOfz2229ot9tYX1/H\n+vo6xGIxAoEANBoNHA4H/H4/dnd3sbu7y1NgQqGQ30+XywWXy4XV1VWsr69Dp9OhUCggFArh4OAA\n0Wh0qm0j6WwAwHahZFECvFFfUg3SbDZhMpngcDiwsbHBhHSz2US5XB6zKJlmZev3hE6nY+GERqNB\nr9fj7L79/X3OBZvVaVhqsJFNq1Kp5P9ms9lw+/Zt9Pt9xOPxmZjamjSQBe9wOGTboXw+D6VSCZ1O\n98WToV8bRBL0ej00Gg2Ew2HYbDZYLJYx21+qdRuNBorFIsrlMjdqm80mms0mxGIxer0ekxDAGzs7\nykkrlUo3e9wIyM1lYWGB3SHsdjtUKhWGwyHbESUSCRwcHOD4+HhqyPv3gURGJE6l+kQkEsHhcOCH\nH35gq8Nqtcp5Ip97j6f+SKPRQDAYxL/+9S88fPgQDx484KkLuVwOp9OJ27dvcz09afslTfj1+30m\nIT7FapAIReohETmdy+UQiUSQSqV4epN+1rcUcE4VCWG1WjE/P4/19XU8fPgQAoGA1WBHR0fY3t5G\nOp2eyQ2PJiGWl5ffYtiJhHjx4gWq1erMqj++Fmgz0mg0fGkjj7lJziKYdDidTvz444+4f/8+h7WS\nyuIyjJIQ2WwWqVRqLHxokjFaPNRqNYTDYc5xocA+nU73XmVwJpPB6enp2Cjcf/7nf/JFirwMCaQU\nuA7rk5Rv5OW4uLjIPoek3q9WqyiVSlxcT8O6+Z6gM3XUgq5arY412wqFwtS8g1+K4XDIxW86nX7L\nZuh9hB8VvDSdFA6HsbCwgOFwCJVKxSQE+frT95vk5zocDlltSY23Z8+e8TPRaDRwuVxwOBysvF5Z\nWcGDBw9YIf0lJIRCoeCA74tWdLMKmUwGh8OB1dVV3LlzBw8ePGCv5tG12G63US6Xkc1mEYvFEIlE\nbva8zwQpEy0WC9tGLi0tMQnRbDaRzWY5P2FSJw8pFJrUhaOX/MtAE4TZbBa//fYbgsEg/vGPf3BA\nYSAQYPHI8vIyhEIhEokE56OJRCIYDAb4fD7cvXsX9+7dw8LCAhwOB5RKJUKhEHZ2dnBwcIBYLMZn\nyTSCSIh6vX6p/Rx9LhJ/SaVSOBwOrK+vc04CTZIQCUHKzxt8OkZJCK1WyyTEq1evsLe3h3g8PrOk\nLGUTZDIZmM1mDlwlWK1W3Lp1iydDJq2pNisgIUm1WkU2m0U+n4fFYrl0yn/S0Gq1+M98dnYGsVgM\no9H4ThKiWq2yPSbdByiz0GQyMSlNoAk4IiGuwx31Y0HWXffv38fGxgYWFhZgNpu5n0k9hEQigcPD\nQ5ycnMzE8xslIRQKBVv/isViOBwOzg0tl8sIhULIZDJoNptfREL0ej2cn5/j6OgItVoNUqkU8/Pz\ncLlc3PdzuVwYDAbI5XLY2dm54k/95aCp58+FWCyGwWCA1+uFyWRiEiKfzyMSibB70Kjjwbes0yae\nhKBGk9VqxcbGBh48eMAWJxSiEYvFkEgkkE6nUavVprbQfR/oZaXL/UXQZybF8PtAzcvR5hIV1TTa\nOVpk09+BTqeD0WgcK3gu/ny6sOVyObaUmXRIJBIYjUa2uRKLxawUoDCg64bLlBykfHW5XPxrdrsd\ndrsdarX6rbW5vLyMu3fvwu/3c7aBXC5/56U4nU7j+fPneP78OWKx2FSsnVHQO0DFKf35yR/zQ5MQ\nlEEyqrKkxnqtVnuLXJRKpRxe2m63Z5p8lMlkXCSTWoLWKO05iUQClUpl4pu7k4LRSQiVSsUjwFT4\nXkcFE5EJn4vz83OkUikolUr2Aidy22w2w+/3Y2lpCel0GqlUaiKbmRcxOnE12jSji2ez2WSFDp2l\nNPZrNpv5e5BVRrlc5i+DwcBEAwURE2i64ioyJyYddDFzuVxYW1vDDz/8gMXFRRgMhrEzky4k8Xgc\nh4eHeP36NdLp9ExZptFkG6ktSQjytT6fVCqF0+nE8vIyexW73W7I5XIWFOzv7yMYDKJcLr9lxTYp\nIFKh1Wqh0WigVqvBYDBApVKxB7LdbodOp4NUKkWz2RzLU+r1enj9+jWEQiF8Ph+cTicUCgX7DycS\nCfT7fZ7ItFqtWFhYwPz8PDeElUol8vk8KpUKNjc38eLFC5yenjJxMc0YDUB+FyjbRqVSQS6X87RE\nv99HpVLB2dkZDg8Pkc1mZ6KxBLypd1OpFLa3t3mqSKFQsCWf2+3Go0ePEAwGOa/xS0F5aRRa32g0\nkEgksLe3h2g0inq9PvXr7X2gM/kyx4NRe7RvHRJ+XSAQCCCVSiGVSjnImRwqKBR30s9jqiVIYQ28\nyYwolUpv/T/UGB+dipbL5ZBIJFyfXBRJlEolnkqf9GfxLUCWOD6fD2tra7h79y68Xi9UKhVbtJJq\n//DwkKcOZ+VeT3bJJycncDgcnD1it9shkUiYFBi1c70q0DodtbUeXfPn5+dot9szeWbQXqVUKvnz\nt9ttFAoFpFIpVKvV71rTTgUJIZPJeDz98ePHcDqdEAqFrDokEiKbzV5LNZhQKOQRp4+xYbqo8B8O\nh2OezrQhAL97FEskElZEWSyWt74nFeiFQgHb29vMfk/DSy2RSGAymeD3+2E0GnkEn/xdq9XqtVtT\nAN7yE1YqlbDZbPD5fPxrDx48wMOHD+FwOKBQKMYIMI1GA71eD6VSyeTWZSQXIZFI4Oeff8bTp0+v\n5KLyvUH+oc1mEyKR6IMeoWT/NQrKOmg0Gm8VI3K5nCcsKpXKzBQrl0Eul8NoNMJsNnPRRqDGbywW\nY+uYG3wYF0mIbreLTCaDg4MDnJyczMQ7+K1BCrDhcIhAIIC5uTn2BDeZTAgEAshkMhgMBlyvTCso\n0JtIeoFAwL6vBoMBVqsVwLi/OuVL0NfCwgKWlpb4OY2SENcJMpkMBoMBHo8Ht27dwh//+Efo9Xom\n9+mS2u/30ev1EIvF8Ntvv2Fra4tD0Wdl3+v3+2g2m3x2Uq36NWpJuqC5XC7cuXMHa2trWF5ehsFg\nwGAwQKVSwenpKX799VccHR1NdDN9MBiwF3KtVmPLH71eD6lUCqPRCJfLxfZmo8RWu91Gr9fD7u4u\nYrEYHA4HvF4vtFotCoUCCoUCSqUSut0urFYrZ3utrKxgZWWFpy5IFHZ0dITnz5/jxYsXKBaLE2lf\n9TUgFouhVCqhVquZgCCip1QqsV0QvbOzgG63i2g0ypM4ZPsAvHkefr8ff/7znyGXy9Fqta6kriCy\nh7LX6vU625ckk8lrYyF5g+8DqnU0Gg1sNhu8Xi/sdjsEAgFKpRLOz8+n5v2m+owyVEZFglRTUN0x\nCsrvoqDfUXQ6Hbaem3VC8GOhUqnYSnh9fR13794d8+gnMcD+/j7+53/+h0UPs4JarcZiGbVaDZFI\nhDt37sBoNPL6oZ7T+7KsPhVCoRBKpRJGoxEajWZMwEjrnjL9ZrGHQn1dmUzGn51IiHQ6/d1Jwokm\nIQQCAcxmM4ekLS8v8xREu91GJpNBMBjE6ekpcrnc1PqNfin0ej2WlpY4rORDPph0SWm1WsyA0ZQF\npdVTAU2XNCIhyIv9MgyHQ6TTaTSbTRSLRVY9TqrihwgXmvAgvzQKImw0GqhWqzMXcv4+NJtNvnDK\nZLIxOw2j0YjV1VVIJBLetO7fv4/79+/D4XBwUBxh9CB5l6XLYDBgz+V4PI5wODwzAaTkG/olBxv5\n0V/mHUzP9DrY5UgkEt6baI2RKrFSqSAajeL09BTFYvE7/0knHzSVZLVaeXxcKBSi1+uhXC4jGo0i\nlUpdm8bRVYKCIHO5HM7OzrC3twepVMqBf4FAAI1GA+VyGWdnZ9ygmsb3lxRyo/UGBUjb7XYsLi6O\n/f+jlwsak6a98TLV0/v2vlmDQqGAxWKBx+OB1+uFz+fjmowICJrOpFDJvb09hMPhmSNeyf6LFPXN\nZpOFC0TmUxjtx66LUYGNUqmEQqFgZZzJZOIsBL/fD6vViuFwyB7zFEY96c1NIqlarRbS6TQODw+5\nUaRSqaDT6eD1erGwsIBkMskqOLI8GH2fyVKIhBESiQQOhwNut5uzH5aWluDxeOB2u9FoNFAqlRAO\nh7G7u4udnR0Eg0EkEolrVT8rlUoOp9XpdADAKmLKWopGo6hUKjOzp/X7fRQKBfT7fTgcDszNzbEF\nIU3f3Lp1C+VyGeFwGJlM5rMCk2kaX6fTsZJWJBLxfTOfzyOTyaBarU7snfNLQGJDUlRTFsQkBiB/\nK5C3vEQiYVU/5bN8LdKahLEk4PF6vbDZbJBIJEgmkzg7O0M6nZ66huan5pSNCsMuumP0+3202+2x\nGu+6giZmyGbz7t27mJ+fh81m47q3Xq9zhtL+/j52dnaQz+dnqqfZ6XTQ6XSQTqdxfHwMiUTCoea0\nt9FdaWFhgS0ly+UyGo0Gms0mZyO8D5S/SV9arRYLCwtYWFiA2+0eW6vUoxkVu8waaKJ8VADc6/VY\n6POxAddfCxN7etFmv7CwgB9//HHM1oV8+CKRCDY3N7G/v39tmk+XsYMejwd/+9vfOIzqQy9SKpVC\nMplEtVplIkImk7EfsdPpHGs+0wLWarUwGAxQKBTv/N5WqxXn5+doNps4PT3lxPVJBAVf2mw2VuyP\nsqTXEcViEScnJ9DpdJDL5WOEk91uxw8//ICVlRXetKiRScq6d21m72qU06EUj8cRi8Um+pL/PXGZ\nMqDdbrPP8Kx7C5PybVTBSQFKhUIBJycnODg4YBu4G1wOgUAAk8kEj8eD+fl5mM1myOVyvoDU63UO\noLtOzaOrBGUpJBIJbG1twWAwIBAIwGq1IhAIQCqVIpFIYH9//61m/LTj/Pyc83zq9frYnqXRaOD3\n+6FSqaBWq2E0GjE3N8fP5mJ+BD0bUujNMlQqFex2O9xuNwwGA1tpkE8w2bns7u7i5cuXeP36NcLh\nMIrF4syFr9KeLpFIkM/nUavV2MpQJpNBLpdDLpdzXsnH7PeU+6DRaODxeODxeGA2m6HVamGxWDhv\njfJHUqkUXr9+jWfPnmFvb4+zEKZhHXa7XYRCIQiFQnQ6HahUKg7wVSgUaLfbUKvVCIVC3BRut9tj\nNYTVaoXdbofD4eCAea1WC51OB4PBAJPJBJ1OB7FYjGq1ilAohGAwiKOjI7bzKxQKUz3p9TnQ6XRY\nWlrC7du3YbPZIBAIUK1WkUwmmdwvFArf/eJ/lRgOhzg/P0e/30coFMKrV68wHA5x9+5dDi4XCAQ8\nYZNIJFCr1T7Z5lYmk3Em5MbGBiwWy9jzLZfL6Ha7M2vHKRaLodfr2RKXJk6uS17SZaBMGq1Wy1Z0\nqVQKqVTqq9yJRCIRFAoFdDod5ufncevWLT43Go0Gdnd3eWpu1utntVoNl8sFj8czliNxg3Go1WrO\ndH38+DEePHgAt9s9VhsXi0Vsb2/j1atXfI+d5N7Zl6DZbCKRSGA4HDJpLRAI2MZ1dXUVIpEIyWQS\nyWQSkUgE4XCYhXEf6hEZjUZ4vV4WSLhcLq5l7Hb7zVrF7wLOSTgrJ5aEEAqFrL7/4x//iNXVVZjN\nZshksjHWcHd3lz1HZxmkmOx2u295PZIv/8ciHA4jFAqhUCjwBZ/G6vx+PwKBAKt4LsPFBTyqOJBK\npUin08hms6hUKgiFQp/3gb8ByDqCSAiFQsGfh5py02IpdVUolUo4PT3l0f3BYMCHpdlsZiuuj9m4\nRv8fsougr9FQr7OzMxwcHCAajd6QEJ+AbrfLVk2zDlJ0ElEIvGlStlotFItFnJ2dIRQKzVxD7qpB\n04VkgWM0GiGVStHpdFgZUS6XUa/X+f+nf16HiZurQqfT4Yuw3+9HsViEzWaD0+mEXq/4ydXNAAAg\nAElEQVTH7u4u7HY7iwCmobn5MWi1Wsjn80in00yOkoiBmpkGgwEajQZms5kvB6ROGvW6JnX1dSBZ\n6ULv9XqZhCCQqrBYLGJvbw///Oc/kUgkkEqlZkopR6DpF6FQiHK5jPPzc/YJViqV0Ol0MJlMaLVa\nnGvwIchkMphMJpjNZqytrWFtbQ0ej4d/jb6AN7VKPp/H3t4efvnlF6RSqamyTuv1eojH4yiVSpDL\n5XA6ndBoNLDb7Ry2TXuRVqvlumu09jKbzUzWOJ1OOJ3OMcEJWTjl83mkUikcHBzg+fPn2N/fx9nZ\nGVKp1Hd8At8eJBIxGAxYXFzE+vo629FRyGg8Hud70SyBprRarRZisRgUCgWkUinsdjvm5+ehVquh\nUqngdrtht9thMpnYqvRjQEpOlUqF+fl5/Pjjj1hZWYHBYOApjFgsxnZhs1qj0NQW2TvSvfU6gmoK\ni8WClZUVWCwWzpzqdDrIZrOf9X3pPSbV8OjPEwqFPIVisViwuLiIu3fvwuPxQCKRIJvNYn9/H0+e\nPJlpEQ/dCbRaLdxuN7xeL3Q63VhmzrvySq4bhEIh9Ho9vF4vVlZWcO/ePdy5c4fFriQaplDkp0+f\nIhwOz3SYN7nYNBoN7jdKJBLIZDLodDoEAgFYLBbOzTs8POTnVa1WP2hR5fV6sb6+zu458/PzPDk2\nmltKfT4iLzudzrXq9dEk8ff+zBNJQhDbTGo5i8XCYWqtVguhUAg7Ozs4ODhANpu91C991tBoNPil\npCDHz4VOp4PP54PZbGblCOU+kF8s8G71eqvV4uZAo9EYY2zL5TIODg6QSCQ4xG9SoVKp4HA4eHRa\nIBCg2WyOqTkzmQw35K4DqtUq22mQypJsDD4Xg8GAGws0KptOpxGJRBCJRJBIJJBIJBAOh2eeTPwS\nXHwXr1OBR/kFpMwE3jR6af+hImJWC7erAk1CLC0twefzQa1Wo9vtolgsIpVKoVgsotPp8EWMvsRi\nMVtzfO+iZRpATXTgDem/ubmJ4XDIimSPx4NHjx5BJpNhb2+P7QOm/Z3udDpspROLxXBycgKj0chk\nF/CGUKTzljxaATA5TaHVe3t7eP78OXZ3d5HP57/nx/pqoGaH2WzG8vIy1tbWuBlOoNrv+PgYZ2dn\nU1FbXTXIhsTr9eKHH36AQqHgKZmP2fPJPoMsNCgbgUb2pVIp2u02qtUqSqUSjo+PEYvFWJE4TXse\nTWIBwOnpKf73f/8XpVIJ9+7dw8bGBkQiEa8xnU7HzdvRO5RKpYJer4dOp+MLPOWTNBoNJhqDwSBO\nTk4QDocRDoeRTqevhSjiIuRyORQKBfvDe71e6PV6CAQCDuTMZrMzL7JpNptsyRcOh2Gz2WAwGKDX\n66FSqWCxWGC1Wj/pTiWTybjh6fP5EAgEYDQaMRwOkc/nEQwGsbm5iWQyOdN7IokEHz16xLbU1xEi\nkYgnbDY2NvD48WOYzWbOB43H42O5Bh+LUUtqEkoQaALMaDTyP51OJywWC0/inJ6e4vj4GOVyGc1m\nc+prucswGshtsVjg8/ng8/lYtEpW37lcjjMAJkFt/a1B9yepVIrFxUX85S9/we3bt+F0OscyV4vF\nIjKZDPb29nB8fIxoNIpyuTxV9cangur8ZrOJYDAIoVCISqUCkUjEdwGVSgWbzcY2kh6PB7lcjoUn\n7wP1CWw2GywWC/c0KexaIBCg2+2yyOn4+BivXr3CycnJTf/pO2AiSQgKuCKvfvKtFggEKJfLCIVC\n+PnnnxEMBplRm+WXFnhzEaUxQ6lU+sUkhFqt5mc2GkxNCgP69cs855vNJvub5nI55PN5vsA0Gg3E\n43HE4/GJ30xVKhWcTueYfyv5i2YyGaTTaWQymWvV2KxWq4jFYrDZbExCyOVy9qb+HIz67o02mJ48\neYKtrS0ezW42mzOp7LwKXLci7iIoFP19JAQRqjd4P6jh6ff7x0iIs7MzFAoFViET+SiTySCVStmT\nc5L39EkBZd00m02EQiFWt5OwgkiIVquFRCLBTfZpX79EQuRyOcTjcZycnGBubm4sZ4psJTQazVi9\nAYDXYiwWw/7+Pp4/f46Tk5OZPRdo4peIwbW1NX5OhEajgUQiwSREPB6/du+hVCplOyGRSASfz8ej\n+R87CWG32/liq1KpIJVKuabpdDpot9vI5XKIRCJMQuTzeQ5TnBYQCdHtdnF6esrB0kqlEl6vF2q1\nGjqdDmazGQsLC2NTqvQ8SPl72RfV+Pv7+/jtt9/w22+/8fTcdSWp5XI59Ho9bDYbN+aoSVyr1a4V\nCVEoFBCPxxEKhWAymTA/Pw+9Xg+1Ws0kRDqd/ujvKZPJYDQax0gIeraFQgHBYBCvXr2aqmmlz8FF\nEmI0f+86QSgUwmAwwO/3Y319HY8fP4bJZMLBwQG63S60Wu2lGVMfglgs5pwDv98/5i7hdrvh9/vh\ncDi4H0XT/AcHB9ja2sLu7i5OTk5QKpVmdmqYSAh6l/1+P3w+H0+yNhoNZLNZDqS+Dnlel4EyqORy\nOZaWlvD3v/8d8/PznF9CoP1rf3+fSYhpzYj7WBAJMRgMcHR0hHg8jsFgAI/HA7vdzgJ0tVoNgUCA\nQCDAGUIfo9wnQTWRilS3jPavyIWDxD0vX75EIpG4VoLjScFEkRCkuDQajVheXsbKygqWl5ehVqvR\naDSQy+VwenqK/f19JiC+VgDRpCGTyeDly5dot9tYXl5GoVB4b4I8+acrlUq+vBFI3dpsNnma4aIS\nilRh9XqdA2VoY6S/i1wuh0qlgmq1ysVfp9NBqVTitPlJbKoQ2aJWq+FwOOByuVj1QJMAFNo3y+O9\nl6HT6aBarSKRSGBnZwcqlYpVJ6RouhhC9SG0Wi0kk0nE43Ekk0kkEgmcnJxgf3+fgwvb7fZH5Zlc\nN9BapUN0lBScBeX0x0AgEEAmk0Gj0UCn00Emk43ZphH5cB0VN5+K0WA9qVTKOS40lqpQKOBwOGAw\nGGA2m6HT6XidxWIxnJ2dzfzU4VWB6pJCocAWd36/Hx6PBxqNBvPz84hEIjCZTGytM+37H029VSoV\nnJycQKVSodlsQiwWs381rTuJRMKKqPPzc1QqFeRyOfaV39nZYWX1tD+XiyDyYWFhAfPz8/jDH/4A\nt9sNhULB+zw1xuPxOHZ2drC5uYl4PH5tapJ+v88qwU6nA4/Hww10uVzOdcPH1P8SiQR6vR56vZ4v\nqcDv4erZbBbZbHYs1yCZTHJdMm3Pm2oE8uoPh8N4/vw5BoMBbDYbbDYb2+aQ4Eur1XI9f9keX61W\nUalUkEql2NKV7mFk63od7mIXIRAIoFAoeOKL7lsUGNzr9T5prU4ziITOZDIIBoN817RYLJDL5QgE\nAjxpKRQKkc/nUSgUeHKHmpwSiYSzX1wuF0+JLSwsQKvVotFosHCCbK7q9frUvacfA5lMBoVCwZlB\nPp8PRqNxLJD6/PycLb92dnawu7uLs7OzmVT2Ug1La4TuqB6Ph+0ctVot8vk8E6MEUvEDb84XsvlS\nq9X834hMJOEr8Ea4SXffwWCAQqGAdDrNTcyjoyNEIhEUi8WZfscFAgHkcjk0Gg20Wi00Gg1b5A6H\nQ+4dpNNp1Ov1a3NHvQiaFLHZbHC5XDCbzdBoNBCJRGxlSK4uL1++xO7uLmczXZfnNRgMuG4IBoP4\n6aefkM/nYbFY+O5JtZ5UKoVSqRyzU3oXqC9K7y79nFarxbaTNKl9dnaG/f19nnidRRKb8tDMZjNU\nKhX32Ynskclk3zUHd6JICPJ8dTgcePToEf7617/C5/NBpVIhl8thf38fm5ubfMDOUqDjh5BMJjnQ\nhTIXqJC7CIFAwCPUVLhcFiZNiqZEIsEvJx2go814UqpTI6Df76PVaqHdbjN5Qb9vMBgwaTGpymRS\nX5JPLnnmDodDlMtlRCIRJJNJNBqNa3MgEGh8MplM4uXLl6hWqxzuEwgEsLCw8MkkRLPZ5BD5o6Mj\nvrhWKhUuVOiif92e94dAa1UkEo0dqtfFc5P2OFLf0GUB+N0/nr5mufj/2iBSR6fTwe/3Q6lUYmlp\nCQ6HA/V6HfV6HRKJBLlcbiYvtl8TlUoFZ2dnMBqN2NjYQL1eh1KpZIsJuqDQuTrNoHeyVqshGAyi\nWq1iMBhApVKh2+3CYDBAq9VywHyn08H5+TkymQwikQhOT0+xubmJzc1N5PN5th2apX2OmihKpRK3\nbt3CP/7xD6yursJut4/VcyQSiUQiePXqFZ4/f84qy+uAbreLeDyOFy9ecCPJ4/HwiP6nnIFE+tD0\ng0AgYOK1VqshGo3i6OgIr1+/xvb2NmKxGCqVytSvPdpTkskkfvnlF4TDYfj9fvj9fhgMBqjVapjN\nZszNzUEulyOdTuPs7OxSS6WzszOcnZ0hmUwik8lwaDiRhNf5/CUSgiy+yMMaGM+5mea19DEgm4tM\nJgPgjT2L0WjEwsICVCoVFhcXodVquXm8u7vLQjfgbStmg8GA5eVlPHr0CLdu3YLNZoNSqUQ2m+Xz\nIplMolQqzex0mFwuh9lshsvl4i/KyCE0Gg28fv0aP/30EwfOkxp9lkHvmEKhgMvlgl6vh8PhwL17\n91jwRhksZP9IgsN2uw2xWMyZN6SaJnJjdCKRzop6vY5cLsf5D/v7+7wXVqvVmc2BIJAgjOz56DnR\n3wM5KSSTSb7bXzcQUWO327G8vMzqfnKT6Ha7nL0XDAbx7NkzBINBlMvla9cDIdFSOBxGrVbD69ev\nMTc3h0AgwJlUJpOJ3Vs+x5VjMBhwjnA2m0U6nUY0GmUryWQyyVkys/jsRSIR94J1Oh1PiYwKer4n\nJoqEoKBgr9eL5eVl3L17l/1aa7UaIpEIDg8PEYvFZp5xvohqtcqTCYPBALVajRXSF0EkhE6ng8Ph\nQC6Xe8tnGAByuRzC4TDi8TjbmtBLWC6XWbleq9W4mTALoMamRCKBSqWCQqFAr9dDpVJBMpnkUfxZ\nL+AuA427FYtF9sijLJJ6vY5+v49mswmZTMYFiFQqHVNuUkgdraVcLofDw0Ps7e0xCXHTyPw4yOVy\nbtyJxWJumtRqNfZynmWMKhouFmgUTH2TBfFlIDUYBUaqVCoolUr4fD4YDAYOWB4lwm7w8SA1ciwW\nQzAYhN1uh8/ng9frhc1mg8fjQTKZRLfbnYl9kYJKM5kMarUaKwgrlQrbGRChmM/nkcvlEIvFEAqF\ncHJygt3dXRwcHMykKgkAExAGgwGBQAAPHjwYm4IA3pzDpVKJrYFOT08RjUav1T43OglBe3+73Ybf\n74dGo2Gf34/Zk6gZT8QCNUur1SrXJ0dHRzg4OMDR0RGKxeI3+IRfH1TPVatVzuIqFArI5XJsiWax\nWFAsFlEoFBAKhXB6enopCUHTD9lsFrVabeathT4FNKmpUqlYWUh2LVSrkUf6LINEaLVajaeMvF4v\nXC4XvF4vjEYjbDYblpaWIJVKIRQKOZi11WpBKBTCZrPBbrdzWPzS0hJu3bqFxcVFAL/bAZ+cnCAU\nCiGfz089ef8+0D2VlMFarZbVvCQejMVieP78OZ4+fYpMJoNsNjuzz4TOgWq1ikwmg2g0ypO9EomE\nRR16vR4WiwX1ep17JUqlEiqVis8DoVAIu93O1ks0XU0CS0KlUkGpVOKsSLKj29/fR61W44m5WQdN\nKlHfhN5hEjEVi0WEw2HEYrGPDp6fJZCFLdV2FFyuUCi4V9ftdlEoFBCJRBAKhRAKhZBMJq8dAQH8\nfqcvFosoFotIJpNsH0n7GOU66HQ6zvCiKTkALHamZz8qTKQcw2w2yxbrqVQK0WgUoVAIZ2dnb/Wr\nZg1CoRAKhQJ6vZ7XIdkt0/t7Mwnx/0Or1WJubg5LS0twOp1Qq9WQSCQQCoVotVrI5/McFDyrC+ZD\nOD8/RyQSQbVa5YVz2QKSyWSQyWRQq9XQarWXTkKQRz+NX48euq1WC7VajccZZ+l504HZ7/dZ3UBq\nLlLDXfeQZLLVarfbKJfLiEajSCaTiEajmJubY/UIXRS63S43lEiBQk2karXK6hzynL/Bx4GKGSpk\n6vU6NwPe1SyYJVCR0mq1UCqVUCgUoNVqAbxRMpES82ZNfT6kUinsdjvkcjl8Ph9nGZCCluxJQqHQ\nzK+3r4F+v492u41kMomnT5+iXq/jr3/9K+x2O1tPki1iKpX63n/cKwE1o8j3tVqt4vj4GHNzc3C7\n3TCZTDAajTg+PkYwGEQ8HmeLx1wuNzOCh8sglUphMBjgcrlgsVig0Wggl8vHxsyHwyESicRYMPc0\n2gJ9Cfr9PorF4thFMpFI4P/9v//HuQ5qtfqjQkg7nQ5SqRSSySSq1SpbxkSjUcTjcW7CFwqFmWyu\nUy7XcDhEKpXC+fk5N+6USiVevXrFIdXlcvnS85TyvM7Pz2e+mf4pGPX/Hm2sn5+fcyh1OBxGIpG4\nNucniZUAYHt7G81mE6urq1hfX4fD4YBer8ft27eh0+kQCAQ4R7Db7WJlZQWrq6s8zW8ymWCz2SAS\nidjq9/j4GAcHBwiFQqx0vy4YDodIp9N8D6BQ+EgkwoLCWSXwgfFzQaFQoN/vIxQKwWw2w2g0cgOS\nLPhsNhsTFERA08RCt9tFLpfj79ftdlEul5HL5VAulwGAs0hLpRIqlQp/kd0T1TnXAQKBABKJ5FIC\notvtIpPJ8F3hOvZPZDIZ9Ho9vF4vbt++jT/96U+w2WxjavN2u41YLIZXr14hEong/Pz8WhIQl6HT\n6SCdTnNW3t7eHtt+kXWQTCaDy+WC0+kEAJRKJTSbTT4vyHaJQtKJQCSLePo1CpCf9mnXD2HUvo6E\nhP8fe3f61PaVpwv8YRNakIR2dsTmLY7TSVe6p/rNvLz/w/0/Z6puTc2LnqqZSSXpTsdxMF4wIBCg\nfd+QgPvC/RwfydjBxLIl8XyqKHfH8sLxbznnfJfD+3UQ2o0ORBCC2a5+vx9ra2umhIkHkAKvb9xC\noWB6d43yRfM+jUbDTNjkZvjSbLVapkyr0WigVqthZ2cH29vbOD09HemJ3G9hX2/2AAaAw8NDHBwc\nYG1tDffv38f9+/dxdnZmMvSPjo6wt7eHp0+f4unTp2aix0BGoVAwh1XJ9fj9fqysrCAajZqMgZcv\nX+LHH3+8NZvCDEKwnDIcDpte8mzpdZvv1Q/BZx9b5nU6HZOVPTs7azJnj4+P8fTpUxwdHWFnZwd/\n+9vfUCqVRnKDrt+4KE6lUqZVxeLiIv7lX/4Fs7Oz2NraQrlcxsHBgTkPYNhx05OH4+7u7mJ+fh7J\nZBJra2uIxWKIRqP46aef8Le//c20m7wNm5sOh8NU/DIIwUNG7fvz6OgIf//737G9vY1cLndrNjro\n4uKia8Pn1atXyGQy8Pl8mJ+fRzAYxOXlZddBj+9SLpeRSCTw9OlTZLNZ8/ttb2/j4ODgE3w3n5fd\nupCBPvl4pqam4Ha7zcHp5+fnqFaryGazJvOSLYpuA66vzs7OsL29jf39fXPNTU1NYWFhAfPz84hG\no3jw4AH29/fx9OlT1Ot1fPPNN/jjH/8Il8sFp9OJ8fFx09c7m81if38fz58/x87OjknIG2X2Jm+z\n2US1WkUikTDvzh9++AGvXr363H/NT8Z+L7TbbRQKBdNmbnl52Xxufn4ea2triEajJnOam5Gc53Jf\nqVAomIzoVCqFvb29roSQcrlsArCdTudWVD1cxa6EYOKEfaZcJpPBy5cvkUgkbt18BQDcbjei0SjW\n19fx8OFD/OlPfzI/x0AD9/B++uknHB4eotFojMSc/2PgNWTPT5go4XQ6zXle9+7dw4MHD3B5eWkS\nS6LRqKl8KpVKpuIzl8uZlsK8d2/beNut+5l8zefd5947+exBiLGxMVOaH4/Hcf/+fdy5cwehUOhz\n/9VkRHFSd3R0hL/+9a949uwZ2u02zs7O8Pz5c3P+xW17UP0W9u7udDool8vY29szfVs7nY7JVGfp\nGx9u9uGjtylr5GOo1+vI5XIYHx83maAM8iSTSZNtNqp4D+ZyOfz666+mp+b5+TkODg5wcnKCYrE4\n8r1YP5bLy0scHR3h+++/R61Ww+bmJsLhsJmksZrp9PTUVDQlEgnTzkP37s2xBUC5XEY2m0UymcT0\n9DRCoZBpzRQIBNBsNkdyYcIzqOr1OhKJBLxeL46OjpDL5W5NOwMA8Hg8WF9fx7fffovl5eWuTXS2\nGjk5OTGbbKPebuQ6OIc4Pj7Gf//3fyOTyZjS/KtakvZqtVpIpVImEFitVpHP50d+A1P6b2xsDMFg\nEOvr61haWoLH48HZ2RnS6TRevHiBVCp1K4KrV2EwutFomJZB6XQa0WgU0WgUgUDAnKPx4MEDdDod\nTE1N4dmzZ5icnMTk5CSazSaKxaJ5ZyaTSXM2yW2ogmXwZWdnBxMTE3j16hUODw/N122rBLFxTXp2\ndoZcLofd3V3zc6yi8Xq95lqy16HcOG80GuaAXPbr7303cMPutp99MzExgWAwiHg8jlgsZipRarWa\nqeYd9czyq0xMTGB8fBxLS0v4y1/+gm+//Rarq6tdn2HSK9dUbCV3W98N18VK8vPzc3M+5t7enkmI\n49o0k8lgZmbGtATnPK9er6PVapkzg27btcm9uUQiYQKG+/v7pnKuVCrd7koIBiFisZjJsL579+4H\nH34rcl2MSHMDZGpqqqs6gpMN6caHeS6Xw97enpnYTU5Omocbv1j6T8w8uY0vgd+j0Wggl8uhVCrh\n119/xc7ODhKJBBKJBJrN5sgvwIDX92sul8P29raZeExMTJgej8ViURO5a7KDENzortfrOD09RSqV\nMq2X0uk0arUa6vW62RTXvfv72EGITCaDo6MjczD1ysoK5ubmEAgETFBt1N5BbO14enqK8fFxczA1\nz3S5LdeWx+PBxsYGvv32W8zPz3cdgNlsNnF4eIhffvnFPOtzudzIXQsfigvIZDKJfD6PH3/88Z1n\nol3FrszhPITzFJHfKxAIYGNjA0tLS5iZmUG73UY6ncazZ89wenp6q5Mk2u02Op0ODg8PUSwWsbOz\ng2AwiFAohC+//BKPHj3CysoK1tbWMDExgRcvXmBnZwfA6/0B9k9PJBJIp9NIp9NmXtK7zhhF/D5L\npRL29/fxn//5n+a9ya/bqlarodVqIZvNYnJysqs138TEhDlsmt027DUov3r/G98N9jtXa9fXGIRY\nW1szLVwZhGDG+W0MQoyPj8PhcGB5eRl/+ctf8Je//AWzs7Ndn2m1WubcA54H0Xv2iLyN6yb7fEgm\nNPHnLy4uTCDIvqf5c/b9ftvYQQhWQezt7WF/fx/JZPKzJxcOVBAiGo0iGAzC6/V2lXkxk+K2LVal\nv1qt1q1eHHwoPth5P0r/HR8f48cffwTw+mBIbkqVy+Vb9RzkmRATExNwuVxoNBool8vmMPnbeIj8\nTVxeXqJUKpnWP2dnZwiFQuZgsEQigYODA1PqfluzR/qFh3Xu7e3hf//3f/Ho0SN8+eWXXYdP8tDF\nUdt4ZpD/tr9zJycn4fV6EQqF4PF4TLuR8/NzVCoVHBwc4Oeff0YikUClUtEi9Z/sNnK3OftXBk/v\nZiavVWZd3+b3p92GhAfCVyoV5HI5kwSRSCQQiUQwOTlpzjngmJVKJZyeniKdTqNUKpl3420ZU3vd\npXaY3XifyacxMTGBQCDQVQnR6XRMFfWon0dylbGxMTN3j8VimJubQyQSMcmtTGxNJBLY2dnB48eP\nzVkQt+k59nv0BhAYmJDfxvPQfv31V5ycnCAYDCKVSpmq9M+9zhyoIEQoFILb7cbk5KSJeLVaLVNW\nw/5Vn3vQREQ+hYODA3PuAw9hbjabt27iwklHJpPB2dkZEomE2ZDiIVRyPTyXpV6vm5ZA3Bzmu5Zt\n025r9kg/nZ2d4cWLFyZrLBqNwuFwwOFwwO/3o1wum/mPjJ7x8XFMTk5ienrazHVZcl4sFrG/v4+f\nf/75VrdxERkmbBlUqVQQDoevXaFzm3AjnT+yVcbe3p7p+T02NmYOFOW8g0lPDGJo407k87iqEoJB\nxUwmg2q1euv258bGxuD1erGwsIBoNAqv12sO7QZgzh199uwZ/ud//gd///vfcXh4qOeYfBKtVsu0\nXeLB3o1GA/l8fiCqlgYiCOF0OuHz+czp51yUdTod0x93b2/PHCB82yKtInI75XI55HK5z/3X+Ozs\nbDD21ZSbYfCmUql0Hb4nn0an08Hp6SlyuRxCoRCWl5fh9/uRz+dvfc/h24AB1UqlgunpaVxeXqLZ\nbCKfzyOZTOLg4AB7e3tm001EBhez+bPZLPL5PMLhsFno8x7WM/1NNqtdTa0zWUSGByua8vm8uZ9L\npRJevHiBly9fmiSx24SHdbtcLkxPT5suLjx35OTkBPv7+3jy5Al+/vlnPH36FPV6/bNv/srt0Ol0\nUCwWUSwWP/df5UqfPQhxlfPzc5MN8Y9//APff/89nj59ipcvX6JcLt/6cn4REREZTswWe/HiBf79\n3/8d09PTOD4+xunpKSqVyq3LJrtN2H7k8ePHuH//Pvx+PwqFginVTyaTqNfrt6LfucgoqFQqOD4+\nRigUgs/ng8vlMucXVCoVJc6JyNBrNBr4/vvvUavV4HK5TLuhfD6PXC6Hk5OTW9eqmYGZYrGIUqlk\n9i4bjQaq1Sp++eUX/Pjjj9jZ2TEtcJRcIvLaQAQh7AwJHpxRLpeRSqXw008/4d/+7d+QSCR0iIuI\niIgMNVb2vHz5Ei9fvvzcfx35hBqNBg4ODvD48WP4fD7E43HkcjkThOBCVUQG3+XlJcrlMpLJJGZn\nZzE7OwuPx4NUKoVUKqWgsoiMhEajgR9++AE//PDD5/6rDAxWsjLbnGcVFotF5PN5PH78GH/961+R\nSCRQrVaVRC1i+exBiIuLC5yenuLx48emrMvv95s+c48fP0Y+nze9IEVEREREhk29Xsfh4SEuLi5Q\nLBaxvb2NXC6H3d1d7O/vq/2eyJApFAp49eoVGo0Gjo+P4XA4sLe3h2QyiWKxqLFD6boAACAASURB\nVOQ5EZER1Ww2USqV8Ouvv2JiYgI//PADGo0G6vU6nj17hkwmg2azqT1MkR6fPQhxfn6O09NTFAoF\nPH/+3BzWd35+jvPzc9RqNdRqNZWmi4iIiMjQYiXE6ekptre34XQ60el00Gg00Gw2b11PZZFhdnl5\niXw+j2q1iqOjI3MoabPZRLPZRKfTUTsmEZERxEqIs7MzbG9v4+DgAA6Hw3R2qdfr5mwg7WGKdLtu\nEMLZz78EJ2ulUqmff8zH8iFj0ddxGzIat5vRuN2cxu5mNG43o3G7GY3bzWjcbuazjtv5+bk5tHZQ\nD4t7hw8dC11zb+hevZmhGLd2u412u41arfa5/gq9hmLcBpDG7WY0bjejcbu5gRk7tlctl8sol8v9\n/KM+hoEZtyGjcbuZ947F+DV/k/jv/3uMjHifPjvq4n367KiL9+mzt0G8T58ddfE+fXbUxfv02VEX\n79NnR128T58ddfE+fXbUxfv8+VEW79NnR128T58ddfE+fXbUxfv02VEX79NnR128T5+9DeJ9+uyo\ni/fps6Mu3qfPjrr4+35y7DrlQWNjYyEA/wfAPoDmx/hbDSEnXg/m/7u8vLxW016NGwCN201p3G5O\nY3czGreb0bjdjMbtZjRuN6Nxu5kPHjdAY/dPuuZuRuN2Mxq3m9G43YzG7WY0bjensbsZjdvNaNxu\n5lrjdq0ghIiIiIiIiIiIiIiIyIe6bjsmERERERERERERERGRD6IghIiIiIiIiIiIiIiI9IWCECIi\nIiIiIiIiIiIi0hcKQoiIiIiIiIiIiIiISF8oCCEiIiIiIiIiIiIiIn2hIISIiIiIiIiIiIiIiPSF\nghAiIiIiIiIiIiIiItIXCkKIiIiIiIiIiIiIiEhfKAghIiIiIiIiIiIiIiJ9oSCEiIiIiIiIiIiI\niIj0hYIQIiIiIiIiIiIiIiLSFwpCiIiIiIiIiIiIiIhIXygIISIiIiIiIiIiIiIifaEghIiIiIiI\niIiIiIiI9IWCECIiIiIiIiIiIiIi0hcKQoiIiIiIiIiIiIiISF8oCCEiIiIiIiIiIiIiIn2hIISI\niIiIiIiIiIiIiPSFghAiIiIiIiIiIiIiItIXCkKIiIiIiIiIiIiIiEhfKAghIiIiIiIiIiIiIiJ9\noSCEiIiIiIiIiIiIiIj0hYIQIiIiIiIiIiIiIiLSFwpCiIiIiIiIiIiIiIhIXygIISIiIiIiIiIi\nIiIifaEghIiIiIiIiIiIiIiI9IWCECIiIiIiIiIiIiIi0hcKQoiIiIiIiIiIiIiISF8oCCEiIiIi\nIiIiIiIiIn2hIISIiIiIiIiIiIiIiPSFghAiIiIiIiIiIiIiItIXCkKIiIiIiIiIiIiIiEhfKAgh\nIiIiIiIiIiIiIiJ9oSCEiIiIiIiIiIiIiIj0hYIQIiIiIiIiIiIiIiLSFwpCiIiIiIiIiIiIiIhI\nXygIISIiIiIiIiIiIiIifaEghIiIiIiIiIiIiIiI9IWCECIiIiIiIiIiIiIi0hcKQoiIiIiIiIiI\niIiISF8oCCEiIiIiIiIiIiIiIn2hIISIiIiIiIiIiIiIiPSFghAiIiIiIiIiIiIiItIXCkKIiIiI\niIiIiIiIiEhfKAghIiIiIiIiIiIiIiJ9oSCEiIiIiIiIiIiIiIj0hYIQIiIiIiIiIiIiIiLSFwpC\niIiIiIiIiIiIiIhIXygIISIiIiIiIiIiIiIifaEghIiIiIiIiIiIiIiI9IWCECIiIiIiIiIiIiIi\n0hcKQoiIiIiIiIiIiIiISF8oCCEiIiIiIiIiIiIiIn2hIISIiIiIiIiIiIiIiPSFghAiIiIiIiIi\nIiIiItIXCkKIiIiIiIiIiIiIiEhfKAghIiIiIiIiIiIiIiJ9oSCEiIiIiIiIiIiIiIj0hYIQIiIi\nIiIiIiIiIiLSFwpCiIiIiIiIiIiIiIhIXygIISIiIiIiIiIiIiIifaEghIiIiIiIiIiIiIiI9IWC\nECIiIiIiIiIiIiIi0hcKQoiIiIiIiIiIiIiISF8oCCEiIiIiIiIiIiIiIn2hIISIiIiIiIiIiIiI\niPSFghAiIiIiIiIiIiIiItIXCkKIiIiIiIiIiIiIiEhfKAghIiIiIiIiIiIiIiJ9oSCEiIiIiIiI\niIiIiIj0hYIQIiIiIiIiIiIiIiLSFwpCiIiIiIiIiIiIiIhIXygIISIiIiIiIiIiIiIifaEghIiI\niIiIiIiIiIiI9IWCECIiIiIiIiIiIiIi0hcKQoiIiIiIiIiIiIiISF8oCCEiIiIiIiIiIiIiIn2h\nIISIiIiIiIiIiIiIiPSFghAiIiIiIiIiIiIiItIXCkKIiIiIiIiIiIiIiEhfKAghIiIiIiIiIiIi\nIiJ9oSCEiIiIiIiIiIiIiIj0hYIQIiIiIiIiIiIiIiLSFwpCiIiIiIiIiIiIiIhIXygIISIiIiIi\nIiIiIiIifaEghIiIiIiIiIiIiIiI9IWCECIiIiIiIiIiIiIi0hcKQoiIiIiIiIiIiIiISF8oCCEi\nIiIiIiIiIiIiIn2hIISIiIiIiIiIiIiIiPSFghAiIiIiIiIiIiIiItIXCkKIiIiIiIiIiIiIiEhf\nKAghIiIiIiIiIiIiIiJ9oSCEiIiIiIiIiIiIiIj0hYIQIiIiIiIiIiIiIiLSFwpCiIiIiIiIiIiI\niIhIXygIISIiIiIiIiIiIiIifaEghIiIiIiIiIiIiIiI9IWCECIiIiIiIiIiIiIi0hcKQoiIiIiI\niIiIiIiISF9MXudDY2NjIQD/B8A+gGY//0IDzAkgDuD/XV5e5q7zCzRuADRuN6VxuzmN3c1o3G5G\n43YzGreb0bjdjMbtZj543ACN3T/pmrsZjdvNaNxuRuN2Mxq3m9G43ZzG7mY0bjejcbuZ643b5eXl\nb34B+L8ALvWFSwD/9zpjpnHTuGncPvuXxk7jpnEb/C+Nm8ZN4zb4X9ceN43dzcdO46Zx07hp3Ibs\nS+OmcdPYDceXxk3jNjDjdq1KCLyO5shr+3367Kjb79NnR91+nz57G+z36bOjbr9Pnx11+3367Kjb\n79NnR91+nz476vb79NlRt9/nz4+y/T59dtTt9+mzo26/T58ddft9+uyo2+/TZ0fdfp8+exvs9+mz\no26/T599r/HxcYyPvz4B4OLiAhcXFx/rt/5U9vv02VG3/76fvG4Q4raWk1zlQ8ZC4/aGxu1mNG43\np7G7GY3bzWjcbkbjdjMat5vRuN3Mh46Fxu4NXXM3o3G7GY3bzWjcbkbjdjMat5vT2N3MJxm3sbEx\njI2Nmf8/MTGBycnXW86dTocVA+bHIaDr7WbeOxbXDULIAOONbt/w1HuDD9ENLyIiIiIiH4m9Vrhq\n/dDTWkBERETk2jh/GBsbw8XFBTqdDgAMYxWE9ImCEEPMXjz0Rh3JXkQMYeRRRERERER+p971wtjY\nGMbHx83/Z/CBLRO0XhAREZHruioBWnMJ6aUgxIDh4oA36+TkJCYnJ00pk8PhgNvthtPphMPhwNTU\nlPl5RhsvLy+7fp+Liwucn5/j7OwMZ2dnaLfb6HQ6aDabqNfraDabWnDI79K7sLVfOLqmRETk97Ln\nNXqviFzPxMQEpqen4XA44HQ6MT09bb64jpiamjLrgHq9jnK5jGq1inq9jkajoXtOREbOVVVhAMw+\niv3/7R9F5N3sPaGr9oVEAAUhBgoPbuEie2xsDC6XC2632/wYCAQwNzeHWCwGr9cLn89nFhXj4+Mm\nyMDAxOXlJdrtNprNJsrlMkqlEmq1Gmq1GrLZLI6Pj5FOp9Fut3F2dmbKpPSgkA8xPj5urjkuZBnU\nUmm/iIjcFBcy9vyoN2lC7xeRblz8T09PIxAImK9gMGh+9Pv9Zo3R6XRwfn6O09NT7O3tIZFI4PT0\nFCcnJzg/P9c9JjJg7A0+ub6rKsF6x5L/jXONd+2P6N9A5DV7rm5/cQ/o/PxcCc9iKAgxQMbGxjAx\nMWFu2PHxcXg8Hvh8PrN4mJ+fx8bGBuLxOILBIILBIDweD5xOJyYmJtBqtdBqtUxmEysg6vU6stks\ncrkcCoUCisUiDg4OcHZ2hkqlAgBot9tdVRjSHc0dHx83/z6cdJyfn5uH6m16sHJMJicnMTU1henp\naRMMY+XN+fk5Op2OCXAxyHXb+wH2tk27LdfM53JVNpOIDJbexQuD2uPj410/PzY2Zt4p5+fnAKBg\n9y1lz8tYLTw5OfnWxpI9r7WvE85HuPnOucuwX0u9G2xOpxPBYBALCwtYXFzE4uIi5ubmMD8/j1Ao\nBK/Xi5mZGTNPe/XqFRwOB9rtNmq1GlKp1NCPicgws9+PdvUS35FcfzabTTSbTXP46225X3ufefZ6\nnXOJq36NnfzJgAPXr9TpdLreEXYwQus5ke5gnj0X4xfvK+4J8XwI4O35u+6hT+tz7pEoCDFA+NLk\nTetwOODz+TA7O4uFhQUsLy+br8XFRbNwYGum8fFxc5Pz9+ICa2ZmBm63G8FgEJlMBplMBvV6Hclk\nEtPT0zg7O/vc3/5A4YRkYmICDocDDocDXq8XXq8XLpcLU1NTGBsbQ7FYRLFYNCXr7XZ7pDfZ+aJh\n8CESiSAcDiMcDiMUCsHv95uxY7uvQqGAZDKJ4+NjNJtNNBqNrhfQbdE74SV74+NDfq+r/ptKhl/r\nDR72Zk+P8j0KXN0ejT/e9mvjQymI1V/2RvLU1BRmZmYwMzOD2dlZzM7OwuFwmGckFy/FYhH5fB61\nWs1sCvQmBMho4wYTr5VAIGDmI3arIYfDgYmJia4gQ6fTQavVQr1eR7VaRT6fRz6fR7VaRa1WQ7PZ\nRLvdRrvdHtpn5tjYGKampszcNRgMYn5+Hmtra9jY2EA0GkUwGITX6zVriLOzM7Rara457lVzDZGP\nrfeMQ/7v3jntsN6PN8WxmJycNB0RGEiMRCIIBAJwuVwmAfH58+d49uwZstksms3myK3tr2qfZG9+\nOhwOTE9PY2ZmBj6fDz6fD36/Hz6fD0D3HI6/xn4/NBoN0y2C69VarYZqtdqVSHfVxqk2U+U24r4G\nA6NOpxNOpxMul8v82Ol0zPyCgVLetwxM8Ks3gKr7qD/sPSnqfc/2ey2lIMQA4QXBm9jtdsPv95sg\nxJ07d7C2toZYLIZoNGoWWjwXghUUNrs1TjAYRLvdxszMDKanp1EsFuHz+TA9PY1GowHg9k3w3sXe\nbHe5XPB4PKYNlt/vh8fjAQAcHR3h6OgIuVzOZEuMcjUJJ22sfFhYWMDW1hbi8TiWl5cxNzdngmiV\nSgWFQgGHh4d4/Pgxms0misWiyWAd1TG6Sm/Gpp3dy8DVdcfDXqDZP/a6TePbqzejmgFauxR0lMeH\n96ntfeXkcrWrNuE0dh+PHeyfmpqCy+VCKBRCNBrF4uIilpaW4PF4zAKmXq+b5Anew3zvXpVZJaPJ\nnp/5fD4sLi5idXUVW1tb2NzcNAk6Ho8HbrcbDofjrarMSqWCfD6PTCaD/f197O/vI51OI5PJoFQq\nAbhZgsCgYKWq0+mE1+tFKBTC3Nwc1tbW8ODBA4RCIXg8HtPKdWxszGwQuN1uTE5Odv1eIv3S2z/8\nXb3E7Q2RYbwnb4r7AjMzMwgGg7h//z6+/vprbG5uYmVlBbOzs6hWq6hWq/iP//gPVKtVk+w1KkGI\nq4IPvVUPk5OTJokhEolgfn4eCwsL5qt3XmBnbfO9UCqVTNcIJhnmcjkAQL1ef6tarrcTgl1hB9yu\n61RuJ87FpqenzX6Zz+czibszMzNot9toNBom8aNer5t7l3P7RqPxVsKg5vL9Ya+9ONfrDaCen5/3\nfT9TQYgB0PtynZiYgNPphMfjMdF8tlxiqTlffADeOzmzo1r2BrLH4zG/v8fjQb1ef6tsXV4/XDle\nc3Nz2NraQigUMpFdZmVWq9WuaOIo4nU6MTFhAmSxWAwrKyuIx+NYXFxENBo1QYhGo4FwOAyfz2d+\nze7uLnZ3d1EsFtFut7tKXkeRPTlmViIX/Yz6A+i6lz8kGNG7WLvNWcB2oCcWi2Fubg6BQAAzMzOY\nmpoyC4pMJoN0Oo1arTYSzzr++zscDrjdbrjdbvNsv7i4MBlyjUYDjUbD/P8PbR/3rl6fdgXeKLWl\n672/7D7BytD5cL3VSePj43C5XHC5XKblZDgcxvz8PObn502FncvlMhPiWq2Ger2O+fl5zM3NIZvN\nmqzFUqmEYrGIVqtl3s29Zd8y3Ph8dzgc5jpZXl7G6uoqlpeXzWYTs++YqDM5OWmC0LwmGo2GCXiF\nw2EsLy+bIEQ6nUYqlUI6nUalUkGlUhmqYMRVwT2v12vehxwTPsv47C4WiyiVSjg5OUEqlUIul0Ot\nVlMfZ/ko7OAhA2Rc63q9XhM0ZBCM1fzcwMrlcsjlcuaebLVaI79JZd/LTqfTbLCvrKxgaWkJkUgE\nXq8XbrcbXq8XkUgEs7OzcLvdqNfrn/uv/9H1Bh7YDphz3lAohFAohPn5edN2LhwOIxgMAnh7jcXx\nZZY291vYPWJqaspcgwC6zs/sDTaM0qbpbyW79X6fo/A9X5eqo9+wn+l+vx+RSASRSMTMq7geZaCh\n1WqZ9SLHrlaroVKpoFwum2c7q1JbrZbZK9IcpNu7uh38FibDOhwOsxdszwVZqcLArIIQI6r3oW5v\n8nJDyW4BNDk5icvL1wdNt1ot8wIGXl94vZUQdkTR3hyygxx82ZZKJWU7XYH9NxmEuHPnDmKxGBwO\nBxqNBvL5PA4PD+FwOG5FAIeTYR6SHovFsLy8jKWlJcRiMYRCIbPhziycQCAAj8eDSCRiNoO5eBjl\nIAQzEVlNw+omt9uN8fFxs3nGB78d/b8uZjBeVU7X+79HHcd7amoKq6ur+MMf/oC1tTVEo1E4nU4k\nEgkcHBxge3vbZF2MwqSGizG32901AYxEImi32yiXyyiVSsjn8ygUCiiXy2Yz7kMCXnYlD6vvuFFw\ndnZm2tEBGPpx7Q1A2AerMaCvQMSHsa8hXjs8HHdhYQGrq6tYWVkxrSY8Hg9cLpdppwPAVEJkMhnE\n43FkMhnk83nkcjkcHR1hYmICtVrNbCjU63UFIUYErx+Hw4GZmRlsbGzg66+/Nmekzc3NmaQaPqf4\ndVU7PrsyYnl5GbVaDblcDtlsFgcHB3j27Bmmp6dxcnKCZrN5o/fz52QHIVgNwc1J3k/cXONY5HI5\nnJ6e4vDwEMlkEul0GtVqdagCMDKY+C5l8MHlcmF2dhbBYBDRaBQLCwuYm5szm8g8Y67T6aBWqyGf\nz5tWQ8lk0mxk3YZr004g5Eb7/Pw8IpGIacnMn5+dnTXV+qzoGhX2fMyuyPd6vQiHw4hEIl3n3iwt\nLZmNULfb3TVn661c4IYb5x18TvLnGo0Gzs/PTTUE12x8t9i/17AHI+w2tu9rj3bV+3BYv+d36Q04\n2Ovt27rW7sU5/ezsLJaXlxGPx7G6umoSQtjyvV6v4/z83Jxpw7lHuVxGoVBAPp83VUipVAqpVMpU\ndbVaLQAY6T2jD2E/C8fGxrruxfddi/a7xO12m/ctr+lGo4FisYhyuWz2m/u5t9n3IMS7Ntqv+qZu\n+w3dW8pnL5RarZbpS8j2IlxAMaPJvhDtzZLenl/cKGKv2NnZWeTzedMX8TZspv8W+/tnlJeTv3A4\n3NXqhRl2wz7x+C28xqanpxEMBrGysmLOJ4nFYibLjhtMDDJMT09jYmICHo/HLHI7nQ7S6XTfo6yf\nE68Pp9Np2qqxVHhsbAyFQsGUwdktma5z/9kBDm6ysLUTgJHKSL8uHsDp8XiwtLSEBw8e4M6dOyYI\nwWyxQqGAV69eDXXlF//9JyYmzLXF8nM7i7zZbJqWI9PT0+aebDabZjL3vsoZe6Jj9/t0u90mMD45\nOYlarYZyuYx6vY6zs7MPbjE2KHpLVNk67aqFJsvyb+O99lvsyTHHkJNet9uNmZkZs3EQDoexuLho\nMjuj0Sii0ag5dNP+d2CGjtfrhd/vRzAYRCqVgtvtRq1WM+8U/tntdttsII+K3o0Bshcgo/T9Am/P\nPaLRKLa2tvDll19ibW0Ni4uLCAQCXdnTvD/5rON/syu4eGaE2+1GOBzG7OwsQqGQ2bDis5EtBFhl\nM+jjy2vDbu/KszOcTqc5yJYZhsxE3Nvbw97eHp4/f45kMolisYhGozGS11Q/9G7YXRWsHvVNu6vw\n3mVFDtdR/OIh6XNzcwgGgwiFQqaKiUGIQqFgkvJ8Ph8mJydxcnKCSqUyMlWt78J1Aaua2O6E7dS4\nBuBZkkwSGcXEQjvzmnMKVuUvLS2Zr7m5OUSjUXi9XrP/wSpgzlH5/LMDWjwDwp7P2j3qe+d/dob2\nMLd6tYP8TFBlpRITQRi85jqCe1Nss8N2fsNeFW2vA+xEBnteC+CtjiRXBaFGtTsBx2NqaspUss3N\nzWF1dRXr6+uIx+NYWFgwcywmBl1eXsLpdGJqasrcj+VyGfl83pzd4vf74Xa74XQ6US6XTQU0kwev\nuu9ui6sqyvn/uaZ/1z1n3+OsjF1cXMTCwgIAmGpYzgk/RXeXvgUheidi9qS4d1EP4K2btvfnRxG/\nN7ucxi495YPv/PzctHIBYF6e3BSyf71dcs7PcKLCDSPe+OPj4/D5fAiHw8hkMnA4HF2b6bcZ/y3s\nBx1fSgBMVLdaraJUKpmMy1EfN270zs/P4/79+1hbW8PCwoJZuPOAdGYOj42Nwel0YnZ2FhMTE1hb\nW0OhUDAR1nK5PLQTlffhteJwOOByucxGm9/vN21y7Oome+LGX3/VmNgvHpYhT01NdU2y7VZPw7rJ\nfhMMdLGly+rqKpaWluD1ejExMWEqA0KhkKlcGka8BhgM2NzcxL179xCPxxGLxRCJREwmFw9d9fv9\nuLy8NIftVSoV07LmOlkTrJ5zu93w+XwIhUKYnZ01i91CoYB0Oo1CoWBalwDDUxHRO6ljhh0XX3Yl\nhP1+ZSIA39Mf63sd1vu2N2jFjQKHw2EODp6bmzOZijxQmIdQc/Hh8Xi6Nov5+7Il5eXlpfnv3Gh2\nOBwm+MXgLnvMAsO92dc7n7aDY8R7bVTL1sfHx02A+c6dO7h79y7W1tZMBYTdXoj3JTfXOT/rdDpm\nc9M+uJoVdEwYYOYr8HqjoVqtYmJiwizQgMG/nnorIWZmZuD3++FyuUwLvXa7jXq9jlQqhdPTU/z6\n66/49ddfcXh4aLIQRzlR5GOyM7TtZ5e9UdK7Sck5Mv/3KI4zx8Hr9SIajWJpaQlbW1vY2tpCIBBA\nIBAwm088JJ1zYwYUufHudDrNgcysetrf3zdth0Zx/OyNTWaw2s+tqyo2e9udDLveylS7nYjL5TJt\nYFhNE41GzbmNDCQ0m01ks1lz5g/70vO90DufaDabplqdbeo4f7ZbyvSeETGMY24HsHw+n+luwIQQ\nn89n2hsCMGvWer2OQqGA4+NjHB8fI5PJIJPJmGpUzseGZZOYz2I+exwOhwnEcP/MTvhl4Ir3p30+\nGb9vrhOGZQyui4FAZtPHYjHE43Gsra1hdXUVsVgMwWDQzK3s+4T3GhOq7fNUp6amTBeYcDhsghDl\nchnZbBbZbLarXZD9rBvGe+9D9K5T7XWA3U3jXetH/hqXy2Uq0Dc2NrC+vo5Op4NWq4Xj42MUCoVr\nV1b8Xn0JQvS+KOwJGW9ku5ypN/v/qq/eiOJvbZz0GuSL0/6H5sKA2UcsBWRG4Pn5uYk825u9dlSe\nPb1arRacTicCgYBp6eRyucyfOzU1Bb/fj2g0imQyaRbxKnd6zb4J7QUGALN4q1arKJfL5hCwUcaX\nss/nw8LCAu7evYvV1VVEo1HMzs52ZeXT5eWlibpOT09jdXXVvEBYcseX9CDfox+Czz+2YWKWTjwe\nN+0QGCDgdcRAwnU2xnkdcmNhenra/Dqe7WJvRt0WPKR0bm4OCwsLpkWYw+HAxcUFAoEAOp0OvF5v\n16Gbw4bXFg8p3NzcxJ///Gfcu3cP4XAYgUDA3EvFYhEejweXl5fIZDKmvQ3fwdwkeRe7fY5d9t57\n/gszh+33D/D+KotBYT/bGdhnpqG9GdKbJNBsNs17mM+vj7n5O0yBCHtRwYQJLt5Y/cDWffF4HJub\nm9ja2jKH1nGRZ2ec2RvvwJuWk1NTU2Zc2u02qtWqaZ3AORCrKOxn6jCMpZ1pZ1eS9H7Z1SEAzAL4\n/PzcZJz1bgAMw/f/LvyevV4vVlZW8ODBA2xsbGBlZQU+n8+8PzkG7C1cKBSQzWZRKpXMpgizPO12\npHYbJz4vLy4uMDExYQ6w5jkSw1RZ09va1efzmUoIZrGypeju7i6ePHmCf/zjH8jlcuYaGpbv9VOz\n7027PSGfPfziPIxV7famlR3Y7s3uHBV8H4RCIayuruLu3bv45ptv8NVXX5ngA5PkuL66ah7M5AcG\nErkRWKlUkMlkzEbWqOkNWHEfxd5fAbrfk3bl16ix52sMQni9XlMhFwwG4fV638q0LpVKSCQSSCQS\nyGQyJmGG9yUrSLihZ48hM/258WlXUPQGH4bt3uW7lRWqsVgMd+/exf3797G6uorV1VWEQiFTSc7v\nlXsfx8fHePbsmVmLspsG58Yc30HfJO6t2mWVDeeo/G+837jWsau+2dKLbYa4X8cKSntPk9fOp9rs\n/diYUc82wCsrK+Z6WVxcRDAYhN/v7zr4GHgTVOV4MFna6XSaBCKPx4PZ2VnEYjFTBVEoFMw+ZalU\nMpVKdmvvYRq/m7CfffYXgLeeQ+/79S6XC6FQCIuLi4jH47hz5w6azaZJOpmenv5klUwfdSfGjs7Y\n/QvZuoHZDMxo4A3Pm9X+kZlMXOizFVG9Xjc3uD04vZUWdguiYZngccLADUo+qBh1r9frSKfTmJ2d\nhdfrNb/ODj70toiwW8HEYjHEYjGz6AoEAuZhkEgk4PF4ukoU5U0ZMV/AnPw2m02z0GWZ/ihO+GzM\nZGLkNB6Pm6xre/PI1rvBFw6HzSZepVJBu902h0COSuYOJzMejwcLCwtYXuGMmwAAIABJREFUWVnB\n2toa4vF413kQ9oSm9/t+VxWEnZ3BrGIeKnR+fm4yUeyF7bt+v1HC+3RxcRGPHj3C6uoqfD6f2XDn\nc42TFj4rh21c7BZf8/Pz2NjYwJ07d7CysoJoNGo2dPmOrNVqSKVSSCQSODo6wsnJCYrFYleLkuv+\nmSx7ZzbjwsJC10SbbRHq9brZtB+GjXROpqenp017CPbJ5DVkn73UaDRQLpdRLBZND3kucvmu/r3f\n86CPmW18fLxrTsFDIFnZwJ/jPITtNpj1amew937fV7Uc4mKvWq2aCpyTkxNUq1UAr+dR1Wq1qxXT\nII+nvbHExRiDNx6Px4xpIBDo6lVtB/y5UdJqtZBMJnF0dGQ236vVateiFxi+1jDsH7+8vIyNjQ3c\nu3cP8/PzZu7B+7JUKpl2e69evUI6nUaxWEStVjPPKp7PxE15ZoTxHBK3242xsTHMzMwgGo1iZWXF\nnIvQW1UxqOzKod57khu+lUoFJycnSCQS2NnZwbNnz3B4eIhKpWIyDD/F9TBMQULgzdi6XC5EIhFT\necmKHFan8rPclLKrcri+5bzEzrZmks6wryc4X+V5D1tbW7h//z42NzexurqKcDhs+oXbZ5TYz+ur\nOilMTk4iEongzp075lw+HlbNyupR1Ol0UK1WkcvlUK1W35v9O+jvvI+FyVi8hhiUury8RKvVMp0K\neKB5JpNBOp029yHvM25+vitB1g7sc27du0k3jOM9MTGBYDCIQCCAlZUVrK+vm01kbiTzTEc+1+zx\n6q20C4VCaLfb8Hq9plKC96W9MT9I7P0J7lOyMpdt40KhkPm532rPageomBTCPY9Go2Gq0wuFgmkv\nZLcIG5Z7127rMz8/jzt37mB5eRmhUMi8B+0qcqC7wwgTiPL5PIrFojkrlPNgBsUYeG40Gqbd08nJ\niam8yefzKJVKI5fM2qs3+MquI9wr5ri+L5HVPj+SQYi5uTlTcWJX9X+qOeBHC0JclU3I3o+hUMiU\nXHIBak/WeqOHLIXmoqJUKiGbzSKVSiGXy5nARG8QovfQw8nJya5yKH4Bg/nCsKN5DA6wpLxarSKb\nzZpAjtvtNhcegzZ2OQ4XqVzIRiIRtFot02aCi1xmFbKdDh+K8hrbnthBCABdQYharWY2n0YVF+WL\ni4vY3Nw0kxUGGHsrIGxcOIyNjZlDwrhRxIzpfD4/MhvmvUGIe/fuYXl5GSsrK5iYmDBBRrYE6i3H\nf1cAgs+46elpeDweRKNRrK2tYXZ2FgDMC4Tlx61Wayg2gT8GBiEWFhbw6NEjkyHL+5WbRlx4DHsQ\nwuVymZZonPyxjz6zThi4Pj09xcHBgQlCMJu3Nwhx1f1rt/Rwu93m7ImlpSUsLy+jXq93tXiqVCoo\nlUrm3Tvo7MCex+MxgZ3FxUXMz88jGAyaRS7Hp1KpmPlIIpEwk227X/wovwtsHD8GqNkPdmVlBXNz\nc4jFYmazl+N0VVa/vZnO5xx//17cFGAQIpPJ4PT0FLVazUygGWwc9EWdXTXHrE72xOXid3V1FVtb\nW1hdXTXtS3rP+ep0OiYh4vHjx/jpp5/w4sULHB8fd7UP683AvsogjhVbGi4vL5v2c6xU4iKs2Wwi\nk8ng8PAQ33//Pb777jscHx+beQa/Ly7gmD07NzeH+/fv4+zsrKud3czMDKampkxAq1KpIJlMIp/P\nD82zzel0IhgMIhKJmMxE3l/1eh3JZBI7Ozv45Zdf8OTJE/M8/1QJIb2JK4N47fXi+LndbiwuLuLO\nnTt4+PAhvvzyS4TDYVNtyHuMWdTFYhGZTKZrDcufy2azODo6wtjYmKnaGeTn1nXwuRaLxXD//n18\n9dVX+Prrr7G+vm4qj3or1ezkmd7WE/ZXOBw2ewdHR0c4PDw0a4phuDc/FMeF+wB2EOI2sINSNrZ7\nZFIrr4F6vY58Po/T01PzxfZyzPa1298CbyoC7O4d9jXJDfTe5K5h/Tfg/RkIBBCPx/HNN9/gz3/+\nM+7evWvOXuGGsH2fchzsyi++L3nALedfTILi2msQz4lgdS2DD36/35ytx8PN5+fnu6ole59NdiDC\n3hC29+dYmZlOp7G/v4+9vT1zGHOlUgGArl8/SGN0FQYhZmZmTBBifn7e7O/arVN7AzYcEwZVeRh1\nu93uSl5iEhgr9nn/vnz5Eh6PB5OTk13XFhNDBn3sbsper9pzXz6f7Orx9yWycj69sLCAWCyGcDiM\ny8tLlMtls4b6VMnoHyUIYVdA2C8FHnocCoUQiURMqRzL5dhz2a5S6I0e2j35GPHixI3Bi8vLy67q\nC2YuTk1NmQVptVpFOp1GJpMxp60zIDEoeNHY5fPn5+emfysrQ+r1ele5TG+vUY4HM4AvLi4wMzNj\nMhYZxABeV130HsB5nZYwtwH7083OzmJpackEatg7rVKpmOtolDedONmIRCK4d+8evvzySywtLZlF\n+ruqIIjXNTfQx8bGEIlEsL6+bspbmcHIgM4w44TGPiyNhwV3Oh2Uy2VzABr7jPIa+q1SOr5AfD6f\nydKMRCKYmJgwz0JW5XCRO+o4GXQ6nQiFQlhZWTEHG/KaPDs7QzKZxM8//4yDg4OhPMiQiwa3241A\nIGAqEvhcslvFcXN2d3cXz58/x+7uLlKpVNe1Zr9neictvRtDfKewRJntiuw2PEweYAbFoG8GcM7A\n9nIMsG5ubiIWi5lsfc4nOD71et3Ma0KhEBYWFkyZ/+npqek5PMwL1Pex7zdWNsT/2QuWwZtYLGY2\nzNnj224Tx7mb3TcYeBP05xeztvlrOR88PDzE/v4+Dg4OcHBwgMPDw665HQMQg1qizfclS/w5juFw\nGNFoFJFIxGTjzc3NmdZybA3Qu3l3fn5uMsfYHpLVctls1syrWbHD9yzfGXbLU/6eg4BrC7/fj6Wl\nJdP+kesHVqVWKhVzpsH29jaePn2KRCKBfD6PZrPZNdfnfJfvx2azaTLW4/E4zs7OEIlETPZnIBDA\n0tKSCToy0DqoCU32QtXn85lsfc75ec9xM+TVq1c4OTkx31Pv5pz9+/LHq9oCMPhtL4rtquzejUT7\n9wHeHPI5aONJ9gb4wsICVldXcefOHWxtbSEejyMej3dVX9qtXBqNhkm683g8XdX+Z2dnyOVyJlHv\n4ODArPWGdaOZ15/X68XS0hK++OILs0nFs6S4acxNJCZ1cX1vB2aZBMYAIgATYFtbW0M+n8f4+LjZ\nHxjk6+imGHzn+433V+/3aW+m5nK5z/S37a/ellQXFxeo1+tmI7PdbiOfzyOVSiGTyZhNTraptqtI\nrkoCs9tv2oGxUah+AN4Eb3w+H+LxOB49emTaL0WjUTMvsQOpvC+ZGFwqlcy+HCvw+f6wDwznPcyM\ndwbQBqGSkGsXrqkDgQBCoRCi0Sjm5ubMVywWMxWU3O+we/JftQcHdJ/T2mg0zD5oJBLB8vIyyuWy\nmZOxxZD9HOR6ahATqO0gTO9mOL94b9rVREzi4vks9rydnXEuLy9NWyb+Wax6AmD2Mdku9zbsWTJY\nxs4sfr/f7A0z4dTuhvMu3Nfk+Y5MrrbPCbPPNRn4SoirJqRsw8SMrtnZWVOSyZJgRsq4kLB/L+DN\nC4GTNTtzhF+8WS8uLkzfSZarc+Og3W6bSgouTk5PT81gD6LeRSEvDi6Ums1mVxZcb8kqANPzlYv6\ny8tLs0nHwx95GBP/zN/KkLtNOJ7sZRqPxxGNRuF2u1GpVEyUf5izqq+DL+np6WnMzc3hiy++wB/+\n8AczUbnqgMz3ZfLzhcHsCwCmJI8voGE/CNFuXcPD0hiArVar5vqpVCqmtPq3St84fgxCsL3a6uoq\nFhYWzLOOGcKckA/zOF6X3aIoGAyaA2/ZEgF4fWD3wcEBvvvuO+zu7prWLcPGPguCE2V+r3xfdjod\npFIp7O7u4unTp9je3sbLly9RqVS6ys/ta8NejF31/5mhzo0Bbixzs4oVONwAHeQNYOJGSSAQwNbW\nFr788kusr69jY2MDs7Oz5iA+++yMi4sL08Jlfn4ey8vLqFQq2N3dRTQaxbNnz0zG4qB//zfFuZ7X\n68Xq6io2Njbw8OFDfPHFFyYLyu12d1WQcA7DAD4zmliGzWf+5OSkOaza5/OZHvbcFMhms0in03j2\n7Bm2t7exv7/flVzCAxHftUEzKPgs53tiaWnJVBfyUEiepTEzMwOv1wuPx2Pmufw9gDf3KjcN5ubm\ncHHx+gycXC6HUqlkPnd0dIS9vT2cnp6aygm7lSfQnQTzufHZHggEsLq6ing8btq48Jqo1WpIp9PY\n3d3F3//+d3z33Xddm069AVH72uDagoFbnqlkH+DHvsanp6dYWFhAsVg0re4GcTOKi0xuMnHTw+l0\nmr93rVbD8fExdnd3sbu7i2w2a5IXfqsS0z7fj5vCTDwDYDZM2FaI76TeILf9+/G/2dXcg4YVgUtL\nS/jTn/6ER48eYX19HWtra+ZsEW6uA2+CKtyEYgUw17x2Bm2pVDItsy4vL5HL5bp+/bCx56krKyt4\n+PAh1tbWupJDOF9hECaRSOD4+Bj5fB75fN6s5cPhsDlsmM9CXh8+n88cqlmtVpFIJEa2LYfdHcFu\nm2zPwQCYHu1MABg1dpCV7ZMYYGfLVVYe5XI5FItFc7it/Sx6XyVgb+a2vR8ziM/8D8XzlaLRKDY3\nN/HNN99gfX3d7HHYgeF2u41yuYxcLod0Oo3Dw0McHh6aoALXFXanDVa/snNHMBhEMpk0m9WNRuOz\nz8/4DGZg2O/3mwBEJBJBKBQy+5U8d8o+S8/+6r02gDfnpPHnvF4vOp2OaSnEBIhms2lahmUyGRwd\nHSGZTCKbzZrE61qtNnAVEva+Izu2sJOKvTfEZxSTN46Pj7Gzs4OXL1+aoAvn/m6321wXDGLbgQ4m\nZffe//Y9OSjj87HZ+xxsi8wzp2q1mkm4v858ge9nJtowWNR7ntynGMuPUglhX4zv+uJFYx/axc0M\nZlozm6b3TAe7HM5u1VQsFlEoFHB+fm4mwswU6w1CZDIZADATbW7UDeoFa0fp31WKyP9mH1B1VTCC\nmSQMCHExy4UC+yN+yj6wg86OvDITjq05OKEpFAoD8TLtp4mJCYRCIcRiMWxtbWFzcxMrKytwuVzm\nEC+6KpBlBxj5I6/JcDhszjEoFAoA8FbfzWFiZ7fyEL1IJIJYLGYmZAyc5nI5M4ljdvpvPfTtIAQn\nTPz9eT8nk8muwydvA04k2ZeUh3/bmwHNZhOpVAovXrxAJpMZ2mobZu2wXNrr9Zo+pQzYNxoNnJyc\n4NmzZ3jx4oU5iO+37ik78GBvEDEIyZ7/HF+77VOr1TKLP7amG8SNJGKrktnZWdNS49GjR1haWjIb\nwHZ/YODN5qVdfRgMBs2hhtPT02i326ZVE7NcRwWvBwag5ubmsLm5iS+//BL379/H/fv3MTMz01W5\nwAUnKzjZWpJZdIVCAYVCwRwuNzk52dXCMxgMmnZ/5+fnOD4+xsnJCZ4/f46dnR0kk0mTQWa3HBrE\na88OxHODjZl2a2trJqN6YWEBCwsLXW1F7X7V9oKLG1OcG3c6HVMBwXdso9Ewf/bs7KxpoZBOpwGg\n69wMfm4Q5jN2AgQ3TLgRyUVTs9lEPp/H/v6+Odfg+fPnZjF11XXAhB57U48Viszg5I+BQMD8fwYd\ni8WiCfIP4tyPyTOsVuMZcqw6r9VqyGazOD09xcnJCdLptMlOBbo31+z3gF2hxKxQ9snml501yyAX\nNybOzs7MHIe/L683+78P4niOj4+baqW7d+/i66+/xh/+8AfTtoN4T3IzlBszXF9xjPnuZlatz+cz\nz83Dw0M4nc6uqsVhwndrLBZD/J8Vcsyw5lyFgRm24kgkEtjd3cXh4aGZGzMIG4vFzGYcnwGsKnG7\n3Zifn8f5+Tn29vYQCARMB4VRevcCb/ZBOK94V6Cd9749Rxs1vYmv3KDk/hBbgzIxoff5Y7/rrqrq\nsgP8vQGIYd7o5PfL+4pJNxsbG6bl0OXlpclM5yYxAxCsBkwkEl0BCPL7/fD7/SYoy2BYq9Uy7x8m\nNNn/Hp9zLOxKvt4KUzu7n/8N6G4byrnI5eVlV0cYAF1dIuwv/j6cuzERM5PJmMrFdDqNdDr91l4B\nN5w/9xyX48MARLFY7Erc4rqQ69J8Po9sNov9/X08ffoUT58+NWsDPrNYOcNAFtvr9Z7DYSe5Deu9\n+CGYWMIExFgshoWFBYyNjaHVamFqagqVSuWt/barfp/eZEJeq2dnZ+b8lo91tuF1fJQ3lH2zAq83\nEpk1Yk8WmH3P6B+zFZhNw95fdmaMPWG1z5xg4IHZPXYlBDNSWBLLRRcPjmy1WkilUigWiwMVWbxK\nb7YqvWvzt3fzyOv1YmFhwWTyczECvG4rkUqlTCsDvlCGbdLbD7xRuVEVi8XMA7LdbqNYLCKdTg9l\nW5fr4mJ2c3MT3377Lb7++mssLi6+MwDR2x7MnuDxs/YGEc+NWV5eNi95LtY4eRymseX36vF4ujIq\ngsGgOdCrVCqZBdfJyclbD/zf+n65GcCeicw45kQIeHNQ/TCN3e/B3pRs72dvHtsl7LVazWTqD9sz\nzr6XOIGwD/3qnQweHBzg6dOnODg4MG02PmTSamcJMbAzPz+PeDyOpaUleDwek3mRzWbNBJmL40Fe\nrPF5NDs7i3g8jvv37+Pu3bvY3NzsCtL3Tnzt/w28OYdkenoasVgMAFAqlbC3t4eZmRmziTSIY/Ch\n+D6cmpoyPebX19fx4MEDPHz4ELFYzCw+mOHKzTeWXPOL2fk8hNW+HzlnYWl8MBg0rSfb7bb5Pbgg\n5iHrduBrUMfbriBhFdPW1ha2tra6WvbNzMyYDV1+39wwtxe9vOft+TQ/OzY2Zt4P9rzQPiuMFbLF\nYnEgF3VceDFji+X3dgCi2Wzi8PAQT548wePHj5FMJs0c9n3fB3+O1x0zyJLJpFmLsNqJz1te80xA\nSSaTA5l1bSdC8Bpgttvl5eu+v8fHx+b+4UHJwNUBCK6t7FYSzBJlWwQGJ5j0xQBjsVg0rRdYAXp2\ndtbVxoKbPIP4zuAYTE9PIx6Pm0rge/fuYXFxETMzM29tjrRaLZTLZSSTya4g6cXFhQmoxeNxBAIB\n857pdDrmucdK/mFsM8FNPb/fj7t37+KPf/wjtra2zIHovHdZSf706VP88ssv2NvbQzKZRDqdNu8N\nBvaTySSOj4+xtLSEjY0NtNttk6HscDjg9/vRbrexsLCApaUlk709ykEIuxLiXffLsF0712EnWfK9\nxuczn0P8Oa4j7cACnzn2utROgOU919uudNCy0G/Krizc2NjAV199hfX1dXOWzeXlJSqVCg4PD3F0\ndGTaXXJ/yG4fxL00vkPtYDJbxMZiMfPvxX0nzhF5DsLnWovZ1WgMGhcKha5Dx/l+arVa5vxLrrPs\nczF4jXCuwnOBuE7rPWuE+O9hJxjyjAVWCdsJA/z3aDQa72yb+KkwuY8Vzel02qy9eQaGw+EwgYZM\nJmP2G5PJJHK5XFcg1T4fmNXS+XwejUajq6ri/PwchULBnPVlJ9YNYhLD78Xr1Ol0IhqNYnl52bQP\n5jhfXFyYc6iAd1dp2UEy7pFzzKrVKk5PT7v2NYcuCMGHDQ9I5Q3GAZyenu7KkGHmvZ1Vw8Fk5pu9\neWkfCmQvmHqrLXjgMjNhmaXDA+ZOTk7gdrsHKuvrfX7r79e7McLFm8PhwOzsrOlfGo1G4fF4zMuC\nFx1fMtVq9Z3ZY7cNg10sKWYWDzPOecjcsG2UX5ddsbC5uYl//dd/Nb3S7RZqwNsZA/zRDhzy97M3\nirgZwsBGo9FANps1ZybU63Xz+w8DPu/cbrepHgmHwwgEAmbhUCwWkUgk8OrVK6RSKfMCue4Ed2xs\nzPSvtA9LHxsbM5shw9AK52NisJCLd7sqjO1feC4Qe20O4zOO9xAX59zAAN6U6vOgr6OjIzx79gwn\nJyddWa4f8uewuoSln4uLiyZT2+FwmHdINpvtOs9lELJ03offn9/vN329Nzc3sba21rUQBd7OhLPv\nVf5bcPLndDqRzWYRiURM6fUovB/sqkC2eojH47h79y7u3buHe/fumTO4OD6NRsO01djf38f+/r4p\n4z89PTWLWaD70EkuxFiByI06Vm1yYWaf+zDoLUvsORm/r5WVFWxubuKrr77Co0ePEI1G4ff7TWCA\nG3XMFiS+U+0+w8y25ruXVRZ8Nzgcjq7NVAZrz87OUC6XTQYUDcpGMJMgeAYNgzMTExPm3iqXyzg8\nPMT29jZ+/fVX5HK5a1e52ZveHO/T01NcXLw+Q21jY8MkO7lcLtOWs1KpYG9vzzwDB62tK9dbDEJw\n3soNuEqlguPjY6TTadNDv3e+YK+peN/Pz8+blmErKyuYn5838xBeZ5zDsQ97NpvF4eEhHA6HWWOw\n3z//He3Ep0F7b/C+dbvdiMfj+NOf/oSHDx9ic3MT0WjUvCeY0WqfP/js2TM8ffoU+XwehULBtJqb\nn583h8vbZyN4PB5z1klvFd6w4Bx4dnYWd+7cwZ///GcsLy+bszIAmEqFfD6PnZ0d/Nd//RdevXqF\nTCZjgjXn5+dm89jr9eLk5MQccMsqX44Xx2p+fh5LS0uoVqtmk2rU9Pbstu/bq5IURykQcdX3wvUl\nK9cYFGSGMDdCOa+zs6p5fXHuYn+Gcwvg7ffEMLNbkK6vr+PRo0eIx+MIBoMYHx/H2dkZ8vk8Xr16\nhV9++QW//PILnjx5gnQ6bdawwNtrcs4v+L/Pzs7gcrm6zpfg2DJQfXx83DW//hy4/8cghN3ahvca\n14981/HZdXZ2Zq4dXh9erxfBYNB0I2Awgus1Pu972xpyDevz+RAOh9FqtcwZHHyH7u/v4/Kyu1Xf\n5w5CMCDDA7e519NsNk3QmdXPqVQKJycnOD4+xvHxMQqFQldynf1c478Hg11ck/Ez5XLZJE9wD7j3\n33RU2Osirr0YhKjX66aS1b4Of+v3uiroap+bzIO+P4XfHYSwv2E+THhh8AHTbDZRq9WQSqVMMMLe\nSOMCg9lOdj9hOxrJhSiz6FiaxKylSCRieq/ZN7n9Uuq9eIf1Yn1XOSGzAkKhEMLhMDY2NhCPx02p\nHQDTj3l/fx/Pnj3Dzs4OUqnUJ+0DNuiYpc/e1B6PxwS1mGHBTM5RHK/p6Wn4/X7Mz89jYWHB9J7n\nhkbvJp2dkcn7khOe3oPi7V633Bxla4p79+4BAF68eGHu00HLNLwKH+5TU1Pw+XyIxWKIxWLweDwm\n+5Bl50dHR+bFYZf4XvfP4aKYm0lcxDJwk8/nUalUhv4Zd13MhgsGg+Y+5fXJDAxmJA5rcIb3HA9K\n5UaG0+k0G0zcTMxms2ZziYH+D/lzxsfHTSuSSCSCtbU1U7a9uLgIv99vrrVisYjDw0OkUqmuxAJg\nMIOHdnKEx+Mxpc8ulwsAuuYtzOhnxhezK+1qFI49N0vtDH6e3TSsC1d749zj8Zg2QcvLy1hdXcXq\n6irm5+fhcDjMgp99mNmSKpVKmbJylpSz+qv3uc7JsB1I5WF1nP+xtYt9gO4gXmdkZ5Kznc/KygrW\n19cRj8exvLxsDg3mpiwXn7lczpwZZLe0sTN87ec7s5vm5+exuLhoql6ZNMA2DHbLBL4/Bu09YWd/\nseqPcw9u9qZSKRwdHWF/fx+pVMpcVx+K9zufoVw0M9DFzVP2KGaPaLau4HkmgzB+vQEve47Av2uh\nUMDx8TGy2ew75692r2ze92zbsbi4iGg0ilAoZP4MO+Hk8vKyq9y/3W6b8eScj9eqvdk3iBumbDu3\ntLSEe/fudSXicG3a6XRwcnLStcHCtnEnJyfmfna5XGZeXKlUTPWHXaHE98VVz8dhwDYabBMai8XM\nepxz/2KxiJcvX+L58+fY3t5GMpk0Ga9XnQlydnZmNqTYM93j8SAcDpsgBDc8NzY2UCwWTf/5YRu/\n9+lNNrzqPD6y12ij4qogqR2EYAtcr9eLZrNpMtJnZmZMSxs7K58Beb6fudas1WomOaz3/Ihhv6Y4\nr2cbOa5RLy4uTPLSq1ev8OTJEzx58gQnJyfmzK7etVPvONh7dVyrsF2s2+02CROs1Lar9z8HVhbx\n7FQApkqPiTS5XA6Hh4emd/709LTZ4+DBy9wXYoIg9zP5xRbxbMHHJGm+I+0KYjux2j5XgXsp2WwW\nR0dHb+2ffK7x495PoVDA0dERms0myuUyZmZmTOCJ1c6ZTAanp6fI5XKo1WoA3gQR7VbKgUDAPNv5\nfOef13ttcW+OCex2xf8w36c27iuxvdnS0hLC4bCpAGTVyE3Op+W/X71eN2d2cB3/qXy0Sgig+0A7\n+0DpUqmEVCrVNdFiqyZG//jl9/vNBH9mZqYrMsYFAkvxWebDw682Njbg9XqxuLhoNj3tl3Gr1TLB\ni1HacLdfyHyhhsNhrK2tmSDEwsICvF6vKbtJpVKmj+7Tp09NK41RGI+PgUEIRrXZfoSbUrxh7X6I\no2R6ehqhUAjLy8tYXFw0QYjeEnH7Rc6DafP5PHK5nKkK4CTE6XR2BSH467k5wt664+PjqFarOD4+\nBvCmvdCg40SBQQhWHgGv27QcHR2ZIEQqlTKHBF/nnrMnwgxCMMticnLSPM+uCkKMOgYhQqHQW0EI\nTn6Ojo7MgaLDOibclGObkNnZWZM1yc25SqWCTCZjghA3acPECqVoNIp4PI6HDx/i4cOHiEajCIfD\nmJycNFUldhCiUql09b0eVJz4zszMIBwOIxKJwO12m5+3K7s4weP7kdlP3DS2KyLGxsZM4DoYDKJY\nLA5NtWUvu+qG47S1tYUvvvjCHJ5st3esVCrI5/M4ODjAy5cvsbe3Z8qurzow+qosLt6zXBQyyAV0\n/5v0vj8GlV194PF4sLKygm+++Qb37983AQhWLHAhy7lFoVBAIpHAwcEBUqmUOaSQva7tzRdmqnOx\nOz4+jkgkYs5TYIBtbGwMHo8H7XbbnOvCDWpgcCogiJWYdsCkNwjiQ1XPAAAgAElEQVTBa41BiJtW\nJdgLajsIYQdVGYSw1yhcwA3K/KS3io0ZmHYQgpu02WwWjUaja+OD7LnMysoKHjx4gM3NTWxsbJjW\npHzX2u3BuP5gaxSXy2UCaqy6YQ9oVpA1m82B3TB1uVxYWlrCw4cPTcu+3hZxrVYLh4eH+Pnnn7G9\nvY2dnR0cHh6aszF4bfn9frRaLTidTlQqFXPN2Alz3Kxn1vGwBbBZ6cV2c3Nzc109wnn9PX/+HN99\n9x2eP3+Oo6MjlEqld55xwHusXC4jnU6bzRj7HKHx8XEThDg+PobL5epKlBol/H56W7sQnwHvClCM\nCjvwyWdVJBIxcwR7A5xrJX6e56rZm8vlchnFYhH5fN4kA7wrwDOs1xQTb5hcyDaaFxcXSKfTePLk\niXmObW9vm2rM67Q35F4A5/88R47zEu7h5XI5E5j8nPPj3rW33VaK7yz7nIjeqjc+m+37jfuO7KDB\nc4SYFByNRhEIBMw1x31O7lUywYlBClaBcn80mUzi5cuXpuLuc+OciZUhPD+EVbhssdRut825Inz3\nca3JKhAmeHC8+L1zzsd1AYMQPp8PtVrNVHsyQWzQ5rG/F59vXIctLS2Z66FUKpmkaAYhrjsX5TjZ\nc16eWTp0QQgAb93MbPljt1Qi++KbmppCoVAwFxx/5MuDLxT70E0eZshFLaPWk5OTWFpaMqez86Vj\n3yiJRALpdHokNo97x9U+tHZubs4sdKPRqDn4rNPpIJvN4uXLl9jd3TWtEYZ5c+5jGxt73U95aWkJ\na2trCIVCcDgcJsOfB3nzph+lcePLdnZ2Fvfu3cM333yD9fV186K0J72X/5+972puK7uyXmAEQOSc\nCOYgtqS2PfZMzctMzc+el6nvwWWX7Q5qqSWKmcg5gwQzvgfV2ty4AimJYosXsHcVS2o1CeIenHN2\nWGuv3e9L22Cj0RDmK1vx6SyYDHN2AZ2O1nmm3AH3aK1WQ6PREA3wLynYP4WxQMwiJGWYKHtRrVZx\ncHCAdDqNRqPxxXuHnwsDE4/HM5QdqiWHzFIY+a1My+osLS1hY2MDgUBgQMu/1+vJMDXOABpFI2uG\nnz0ZX1rrm6A/GYU6ONbPrTvojIw5MnnI/NzY2JCBudQXZ0GgUCiIhiQlcozMMbOZEaxnoAxAYox+\nvy8gM5n77XZb2lc9Hs9AW7WRBMDYhfHHKCatZN+QOUc5roWFBSQSCRkQDHzorCwUCjg+PsbBwcHA\nMPRKpSLgzafuby0robtf+W/6y2hGYPypTYMEkUgE0WhUhneTTe33+6WISx1cgg/6q1KpoNlsotVq\nybBb/hyLKV6vdyBZZWJGNhpwuy4cxszfaUZCjj5PTGi11vDNzQ2KxaLMMyMz8Wvud02oYPcT14fy\nMJT9YExDANhMZ1yDhwQDKIVBEIDxqzFGoEwSz3wymcR3332H58+fIxaLIRqNwu12y96i6e4xXSQl\nS93r9Qr7UbO4b25uBATTr/XUxvvP6/ViaWkJL168wPz8vBTPer2e3Hv5fB7b29t49+4djo6OJMdk\n7sq7iVJrRo1wdoARdOVctFHq2uQZoJ55IpGQTk1+vpTTy+fzSKVS2NvbQ7FYvJMUaCQDaJITCyUE\nz9iBHIlE4PP5RLZtlNbwU6a7rQniDRumC9yy0sfl2Wm6I0F3hbOA6ff7ZS1YT3K73QgGg2g0GlLc\nJUBKKduJiQlUKhWRNGFXBHMrmvYvo7S2zAN8Ph+Wl5exvLw8QL7h4OCjoyPs7u4in8+j1Wp9cSxP\nv0OVEx2HMM/Xw3CfGiTjHWOMHXTB2xh3DjtbWuKLxJNWq4VarYZKpQK32y0sf6fTOTDDlvEEySSc\nt8T1Y8f1zc0NAoEAwuGwdBw/ZdyhQSdKHHMtT05O5KwRtKEKDTtISOpwOp2IRCKIx+OIRqMIBoMI\nBoNyRvl58HX4+9jtREJ7sVgUKSjtP43rY+b8dJhRlYRAls/nE/LQ5eWlgKf8DO57Pt09RrIm41vd\nbf0t6ySPBkIAHw9RJhAx7KJhwn9+fo6zszN0Op0BBJBfNG5unSAY27+mp6exsbGBk5MTQb/JVun1\neiiXy9jf35cW2VHaiHeZkb1EtCwWi2FpaQnz8/OCKnL9SqUStre3sbe3J6j/qBbnHtsY3LhcLils\nhkIhTE1NCYuHQ1h5eMfJeEEFAgG8fPkS//M//4NoNAqbzTYAQPCy6/V6kliQlcjBoUwwGThz6CiZ\nTOywYHDCgLLf70uXye7u7sCAVzOeWa3Zp5F8An8XFxcol8vY3d1FJpMR0O9Lz5yWSOBQSBYCNGOb\ns13GbW8aTQfWm5ub+P777xGNRgeYvScnJ1IgbTabptw/nzKCT7pllUEqi3P0c3ogo5ZAG6YrP4wt\nx0SOevUvXrwQWQWefw4hOzo6ElkPDh01+zwI4BbQYQLPRETHFvl8HsfHx8jn81IAJrNpcXERgUBA\nXgO49cOahWwsEIySkUXvcDjg8/kQDoelmM4iDxljnU5HNIQPDw+RSqUGBox+aUFEgxHD/v0+o/9+\nyj2oGW12ux3z8/N48eIFtra2sLa2hng8Lsx++jQW6HK5HN69e4c3b96InEu73RZpICZhmjVN3XSC\nsfF4XKTptOY8cKvHTsaTnuNitrtRywoRKKSfOz09lfWp1+vCiv5aM+YZHPjNHIPApdYZf2pGJ02D\nysNmQhCkJnlLJ6y8o7jekUgE3333HZ49e4aNjQ1sbGwIc1gPTdaFB/oeAh0EcKanp+HxeHB5eSmx\nIAv0lL8wG2udsWgwGMTKygpevnyJYDAozM5Op4NsNosff/wRP/74I/L5PHK5nMwiGBav6sGjGiC0\nWD5oqFNuiB09ZjyTd5kmhCSTSSSTSbjd7oFC3unpKSqViswGymQyAibc5x/omymDQjBWF0t07kuQ\nTA+MHQdjHEgWNe8hLQcD3BYHR2n/fInxvtJdhizwksg2OTkpgHwwGJT7SN/djP14Z5HMw7uMhBoj\nK9hY5zK76XgkGAxic3MT6+vrCAaDIqWpSbqpVOrBhC2bzSb5r5ZR5GfCOoCZ5t3wvOjOBg0eG+PR\nYXUI/ix/jkQPzqchaUGTozURW4Nlq6urWFtbA/BBPovrRTA/Go2iXC4PkACewjSAwzue9z27Ehij\nUmKQZHHGUcFgEJFIBAsLC1haWkIsFhO5UH4PgIHuaZIpzs/PEQwGpTPVZrMJ2KHn2xqBw1HooqZR\n4svv94sEptvtludi/Y1yhp/KszQAoYeIE9jRseG3skcFIYx2VzJJY5Hs/Px8gJGp/+QFyotiGBuO\nRUCySPSGZdJFpnY2m/2i4XWjYFpXzefzIZFIIJFIIBaLIRAIwOFwYHJyUtg7pVIJR0dHyGazMgR4\nFA7ktzCupcvlkk4IDm1iEF2tVh+FeWdGMyZfW1tbA2xh4FZujV01mUwGh4eH2NnZwcHBgQwkZRCn\nGQJerxcejwcLCwvSus2OCbKwAYi+M2WZdDHLTMY7ihI2fD7KlHAOTaFQEGkNDr76kjPHIJJaiCxC\nkyHR6/WksPQUaPZTGNfD7/dLEOPxeAa68DiLKJPJoNVqjdya0BeSOULGjJZhYiCoWzEJVl1dXUnR\nRwcfPM9GXxqPxxGPx7G+vo7NzU2sra1J8YnFpW63i2KxOCCDwi4Is7PvdGzBwJg6tRaL5aN5Sdw3\nnU4HiUQCNzc3iEQiACCdYQzAube0TIBZCpRfagRTHA6HMKJjsRiCwaAM3WPgSi3hvb09pFIpFItF\nmb/y0PP2JWuli69Gbd1vbfq8EsSJRqNYW1uTJMvn80kBm2eWg7wJfu3t7cmQOH22aEzmjcSTxcVF\nRCIR+YyMxAHKozYaDbTbbZHJMlNiZuws0kx+EpY4CJEAqJZh+pqCNv2G7hjg+vBME9xlAdBsQCOL\nc3o4OeN/zRLU9xVwK9fl8XgwPz+Pzc1NvHjxAouLi1hYWBjoXtByuyxiEQRnN8nJyYnMnOBZIKuY\n88MoQ2E2gonVaoXf75c5LgsLC9J1ysHeOzs7+Pnnn/HXv/4VrVZL5Kbu6tSanZ2Fx+ORznQWowhC\nVqtV5HI50WAfpViFZ9XlcsnQTKfTCQDClqV04+HhIbLZLKrV6kCx12iaecyiHrshOC+O8QY7kVl4\nZgcQc5VxMN1pqQEIfffoWMTssdjXGkEZssn5xS4RktsIUPf7fWGVA7cdgZRNYwGO3Ur0sQCkc4z7\nlfnbp5jHZjCuEwdSszbkcrlkDhJZ+1QyeIgcCzuhKEXsdrvFd3OfGuU0zWD683toR62+qzQIcXp6\nOsA6N3af6w4yr9eLeDwuc4h8Ph8AiN/lnnY4HNJh9pTgvRHg1YTz8/NziSu4vvxe3mFOpxPhcBgL\nCwsyczAajco5Zp6qyQ40nkPe9Q6HQ85qpVKRWWpGwjrf4yjcj5pQQuKhy+USNQLGp6z73FfT1p1j\nVCAhuYfkkaeas/ebghBfYsPQZW5obr77EiVdpCPzB4B0QLAA2G63h7Yhj6IZW9Z9Ph8WFhawtbWF\n+fl50Z4DPgT99Xod5XJZhsPoltZ/2QfTbWLU8NPs/HQ6jVwuJ7p247Z27FJgIUOztWgsmpyenmJv\nbw8//vgj9vf3kcvlUCwW5f8NY1+ziGC1WhGPxyXo051PbM+7uLhANpuF2+2W9TZjQsGA1el0wu/3\nD4B+zWZT2F+6YPKl+0YPb9JDmKempoSNrB3vOAJk2sgQIHNfAz/AbYLBGQnFYhGdTmdkz6sualOe\ngEEI/+36+lqScbfbLTNJ2FqsJQ/JGiETmoHi0tISlpaWsLq6ing8PlAs4YDRer2OXC6HVCqFer0+\nUtrVJDNQ1qVcLmNubg6tVgt2u12kHikZQSm4y8tLRKNRaRHWckx8XV0s4ZrowrjZg95hZrfbEY/H\nsbGxIUPJKXXBJKter8sgakrNfSvfqAvWTPaY5H7rtdaJJVnPHFROhq4eJM+CEc8h17DVaskdzmfQ\nhWIAUnhJJpMi87SxsYH5+Xm43W7pkNJFKUqXlkolAYr4e7Q99R7VzE09uJFxBwEILV1zV8f1l5ru\niqL0ptfrldcnqYISAPqzNIMZ104X0tgRrgvc/F6enVAohEQigdXVVSwuLsqZJ9v65uZGEl9KQuhu\naha8eA61dAJlGGw2mwA9em+axYdYLBa43W4pjlCOFYB0kr9+/RqvX78W0o3xvBpfj91K8/PzWFtb\nk64Kvma73UY2m8X+/j5qtZqcyac+i59jPA+Mx8LhMAKBwMDnfHl5iUKhIDMIS6XSZ0tO6T3Cbgij\nnKmWKuK963K5cH19PRbSy8AtuKiLkMPuHgKp45if0rTf15Kjl5eXct+RCGDsFKEECYtvBJhdLheA\n2+HNms1PSU5KlnD/8d4yc2ynOwpZ4OX+ubm5QbPZlDmFDyVX8rMgwTCRSMDtdg90PWjteQ1ymImk\n8xjvYVgtU8cnRtBQx6+cnUGig1ESivmFWfylEYjg56qBJ5ruXuIQeRJn4vE4AoGAkGe4P7Wqgf5z\nZmZGlDZIzLTb7QgEAsjn88jn8wJGNJtNASMIIDI2MXMdmOvFeiRzcXZnkTDH/PSu/aDvSafTiVAo\nJEQIq9U6ID32FHGsaUAI4GHt98At0svEgK2YAGSAHVmblHkxwwF+DNPolgYhgsEgPB7PwPC3er2O\nTCbzEQjxL7s1zTbX2nR01hqEMHPg8VBzOBwyUI4tccaLifpxjUYDu7u7+Mtf/oLd3V1hDBuZqNrR\nknXscrnQ6XQk2SL4w98ViUQwOTmJvb09uN1uVKtVUyYTOvkhI9XpdGJyclIkIw4PD5HJZAQkeAiQ\novW/NQhBXUCCEDzT45yAAB/WnfqZBCEI/gCQQr0GIUYZcOU+I5uIsx/ItGGSziCPIATZETMzM/D5\nfMKmcLvdMrS73W7LeV1fX5diJgcPAxgIfKrVKrLZLFKplEhHjNJdyHZeghBTU1OoVCryJxnoHKIG\nQIqget6DkYnEwFbPltAJstkkR+4z3tkahGBSTtkCSmw0Gg3Rvn0KFq/uPGHc963ZdsYEk1JMbrd7\noKChO5iA231DEKJSqQhQPYypzmRhbm4OXq8XyWQSv/vd7/CHP/xB/PawwjgLNNzzRhDiU13L39pY\nHGdMPzk5KWfLCEJoUP9rkygNQrDYTlCNYKIu6uiZEE9dSNFxFuNYDUIQENAAqe7smJ2dRSgUwvLy\nMlZXV6VzR+vrM/Yrl8vI5/PIZDJIp9PyeiySkiAxOTkpXQ/sxrNarcJABmCqgin3D0GIpaUlkd5j\n0bJYLOLNmzf44YcfkMvlUK/X731Nfh4c8r22toZAICAsxMvLS7RaLeRyORwcHKBarY7UrLlhIATz\nJsYqjIV3dnbw/v37Abb1fc+pWbRXV1fSHc2ZZ1pCRc8xokTH6enpN1mDb2EkmBiZ0MY7jz5llGKy\nh5i+7wAIuYRzHVjA00N/dSeDZkkDkJlnXq8XvV4PrVZLpNOq1Srq9brMCLNYLB9JAJp1vfX51DMI\npqamcH19jVarhUwmg1KpJPnpQ2InKh0sLCwIgUnPo2LM/c9CfuXzfarQrUEGksa0hJWOLQjqmsk/\nGAlWBAeMcrR6HovL5UIgEEA0GpUua8bI7DI1do0MMz2zLhAIIJlMIpVK4fj4GKlUCul0GhMTE+h2\nuwI+6vzMrGcWGOyE4IyVYfLbGoQYtlYa5GL3CTuV2CmmpZm/NRBhKhDiIWax3Or3f/fdd4hEIpiZ\nmcH19bXIRrx//x6vXr1CPp83tbb8lxgdy8zMDILBIBYXF7G+vo7FxUWEw2E4nU7YbDYBINrtNvL5\nPHZ3d1EsFkW2atTX4bGMh4/Fu2g0Co/HA6vVCovlg8Zdu91GpVJBvV43ZUH8a81iscDv92NjYwOr\nq6silUSjE2y329jf35dhfGSZM6C7C0zUDB0mX8PmajB4JGPbarUOaJ+ayTT7am5uTtj419fXUvDJ\nZDKo1+ufNZx1mLFdM5FIYH19HZFIBHa7XQoDJycnKJfLyGazIyk59CWmu79isRhevnyJjY0N+Hy+\ngaJwt9tFuVxGqVT6qKg3SsaiF89Or9cT2QYyuHSiTxDQ6/UKGM/Eg+xAsvl7vR6i0Si63a6wueLx\nuPgPHfBcXFwglUphe3sbb968QTabHdDdNNpTF+Q+ZfpOZ6I6MTGBWq2GarUq4ArBRcYYKysriMVi\ncDgcErhpm5mZgdvtHhged3p6KvroLKSa2dhl5PP5EAwGBeTjnUPjvqOkD4CP1uNbvmddfNH/9i2M\n55S/l0E/mfw2m21AfoU/w+9n0ZPx3PLysgxGJ3u93++LtnAgEJCka3V1VToXdeKni3ftdls05yuV\nChqNhkj38W40y/3I+50xgNPphMVikfum0Wig2WwKmUjH818LpnDNNKCo10X7Hz1g0yw5he7kZaLJ\nu43vmd2UFxcXwvCnpvrm5iY2NjZEhlR3FnLvUFKHQ9Or1aokwFarVVh7LATzXLCw0O/3RYJXE1bM\nsH68010ul0gTcv+xk6TT6QzILw27Z3SBlMzDZ8+eYWFhQQbC9vt9NBoNFItFHBwcCKDBLmIzrMfn\nmP7c7Xb7gMTW1dUVut2u+NV6vS7AHvB551TnD5RJIwmDhBsAA/etmXOGh9r09DTcbjei0ehAgVcb\n/dCweV/jZNqvFQoFTE1Nydl0OBxwOp3SzaaJloxluX/YOcEcTgPLlLem32UnsdPpFN15FgE1y9os\nfpRGgI4yVSwGM6+gOgGHBn/pvUPfMjs7KzMLgsGgDBXWcmzZbFYkTs1y55vBSDLUpBXKLQO3ABtn\nTGhFBTOtoY6/eBZ0nkS/yE5NkuruqkV+6v7i63IgOn2Onr/k8XjQbDbRaDTEb5+cnAC4HTKucwb9\nHE9leo3m5uYkltIdcMAt2YnEVw3ccN0J/NhsNqyuruLZs2dYWVmRPJ+gDLu09et/CxtpEIIL7Xa7\nsby8jOfPnwsIwdYvtoC+evVKNCifeoN9remDbLVaEQwGsb6+jufPn2N5eRmhUEiG4BAlo47p7u4u\nCoXCwOyMx3xfwNMf4IcYHbXb7RbNRK/XK6wtrqEGIUbxOe8y7qlAIID19XUBIfj/gFsQodPpYG9v\nD3/+859xcHAgwzPvK2RoVgAvPX4Zf0ZrGpO9oZnHZll33S5HEIItc1r6IpPJoNFo3AnSfOp3EIRg\nKz9BCDKsjSCEmVsMH8MYUMfjcbx8+RLr6+vwer0DSVen00GhUBAQwkzMkS81XRg7PT1FrVZDLpeT\nvUedd3Y9WCwW+Hw+2O120ZJ0u93Siq2l0TSr8Pr6WoaCke0PQBhMqVQKP/74I96+fSsghHHA6agY\nQYhOpzPANKIcE0EDJlbxeByLi4tYXV0VEGJYgWNmZkZ0v6vVKsrl8kAQzuDcrGvFtRgGQugiOvcj\nZ2pQk/8pix7ch0+VUBg7/1jcYAFDF2H1e2LRc3JyEoFAAMvLy+h2uwIccBDzzc2NFHdDoRDC4bDI\n0LErThemNHOOBArOUWAiq2cEmOUMG6VVHQ4HLJYPWvyUg2u1WjKID3hcEIBxDGOVYfHJMBDiqU0D\nJC6XC6FQaODcak1pj8eD6+trKdbG43HMz89LJ5yeW8DOkHK5jN3dXbx79w7v3r3DwcGBFOJY/KQU\nIItRsVhM5lJo+ap2uw3g1reZpejOved2uwWEIJBydXUlQD8Z0drnGoFILXG1ubk5AEKwm6zRaODo\n6Aj7+/vI5/Mi7WS2QuZ9xnidIAQLnBqEqFarAyDElwyS17kD55Dw6/z8XIbB0h9zdpaW4hl1s1gs\nElsQhDA+m9Gn/DOBEMzPG42G+EhKRXa7XblnCD7oLgjGxX6/XyRhqJnOGgCJPHNzc2g2m/In2erA\n/YOLn9KM8nwsNGoQgoXZhxS2uTfJSh8GQlxeXqLRaCCdTiOTyQzc/2Zbr4faQ+PNfr8vZDEt36lB\nCC0RyT1tplrmMOKpBof19+guV/rVh+bnOuYhAXRqakoUI5xOJzwejxAoisWirOWwGM9Iun1KIzBK\n0sh9IARjZAASdzAfoPyVy+USEIJ1YoK1FxcXUmf71oPjRxqEoHMIh8NYXFzEysoK/H6/tN8UCgWR\nzymVSmPB/tfJbTAYRDgclsRheXkZ4XBYWJpsGeTAQ34Nk0y4K1j53LUytvGM2hoziPX5fFhaWkIi\nkYDT6RSmsU4+jJqG42B8/rm5OWlts9lsA58pkzBqpu/v7w/oSH7OZ64vfs1GMRZwNJtUD3Yyi+lC\nEwETDkgig5zMOaPu/pcYGSZ+vx/z8/MDGsUM7k5PT9FqtUaSRfelpgtUPp8PyWRSQBld4KvVatjf\n38fx8fFYdIcwiTo/P0er1ZLiNtno1L1k4BEOhyURd7lcAkqwOw64bY+12WyyZ7TcENvXG40GCoUC\njo+PcXR0NDA805hEjMK+49llMKrv8omJCQE+CUovLCxgcXERi4uLCAQCcDgcHxWU+dxkqSeTSRkK\nzjb+er0uGqVmYp5r49rMzc0hHA6LLJ8Ofsl6ps9gPMIkl50exntoGEuHxjXUBeAvfc8aTHoq03EQ\nC2anp6fodrtwOp0fyTEBtyw4MtgJzp+cnODk5ER0qG9uboRl7vV6pUCiCyHA4HBlSgql02kcHx/j\n8PAQhUJBfIVZtIVpGthnfK9nkGjgiwCKloP72oIGf16TJTRAw38jm/YpBvndZTyPOnklAMB7zel0\nwufzIRKJwGaz4eLiAlarFclkUgYwRyIRSeg5r6DdbmN3dxe//PILdnZ2cHh4iFwuh4uLC/EDFssH\nOSZ+JixMsUjFeE7fG2YqlBrjOUog8K7XkgaRSARLS0tyRvnz+l7knfjs2TNsbW1hfX0doVBIdJjP\nz89RqVSwt7eHg4MDmT80Sjkq12Vubg6BQAAejwdzc3MDw3+73a7oczMW/lKijD5/5+fnElcb8wcz\n5wwPNeO+Msq0aNPA86h2/37K9D4giYSF9HK5LKACZ/IxJ2L8zDuLcQr9jN/vR7VahcfjkQGwtKur\nK0xNTWFubk7uM3bdNJtN+Tz0vFEznmG9Z3S8pX3YlxTTeef7fL4BYgS7sNmpTxJEqVRCtVrFycnJ\nSN1znzLeg1+aB+l4O5FIYG1tDdFoVOoIjLd5pim7zE5tM9lduaAmsWryEusWtVpN2PxWqxX9/nAZ\nW23GWiN/hnJ8em8zrvZ6vQKSz87OotVqDcjzm2kv6g4GkkbYjc5n9Hg8iMViODs7g8vlklk1ep4I\n8CG/YO6/vLw8ILXMe0zLf2lZpm+xJiMNQjDwmZ+fx+LiIpLJJObm5gAAzWZTkq5KpTKA8o6q6eDW\nbrdjfn4eW1tbMpQwmUyKjAaZrWQEHx8fI5/PiwOgAx4WyNA+1wnxT256sx3oT5kO8Px+P1ZXV2Ut\n+/2+6K9Rz5DJ+7iYTrB46TGR0A6F69BoNARVpjP83M9bO1QWaDT7wliM0g7JTKbXTIMQc3NzHzGG\njQP0viRAoR5gKBTC/Pw8kskkfD4fpqampCBCSRQOKRqn4M5oFsutHrfb7ZbZNwR+mIRVKhVsb29j\nf38frVbrid/11xn3jJ5zwc+fbGbKoQEfAhjKXdjtdvh8vo/aTo2sORZGOYieAeD5+TnK5TL29vaQ\nSqVQLBbRaDSke2KUwIf7TA/uIuNkbm4Ofr8fy8vLWF5eFmkOHeAZA2+r1YpAICBsnHg8jlKphFKp\nhFQqhX6//1W6u9/CLBYLnE4notGoyPzo4JR3G4P5ubk52Gw2YTuzIMkYhKbnNmitel305Z32uTGE\nEdQwxiBPsS/57OxQqtfrKJfLUlAnA0zHHZotxnOqAQmSRji0j19GQAO4ZR6SNUcG+9u3b5FKpQRE\n7PV6pmIi6vXQbGbNzpqenhb2Jr/0oMbHeo5+vy/rzj3MddWSPLob46lNny+unx7yyM85GAwikUiI\n1IjNZkM8HkcsFkMoFBKQFQDOzs5QKBSQSqXw5s0b/Pzzz44i4f8AACAASURBVEilUgPFPX3GGY8w\nXm61WiLVQ+AXuPXjen7JUxeMCULwnBolfVgUiEQieP78OZxOpwyHNBJntLzLysoKVlZWEI/HReKU\n57NYLAoIUa/XTSnn8inTWtN+v18KaP3+rQ485wXxvDzkGVl87vf7Aux+Kq9/6j31WMb9xTvoLhCH\n95YenjxOpnMqxruUqalWqwNgJ+fOECTVcsB6DzIWKZVKIr1JAJIAhY592CXBe417nf7HTDPSNKBg\nnBOii7dTU1MDRcjPef9cE7vdjkgkIgV0LdtJEJsxcLValXtgnOaWcM8ZaxefMu4pj8eDtbU1fP/9\n94jH4wOkH8p88S6tVqsix2RG43Mb95p+Dkr0Md+knJLVagUAAfXuIinofzeCafTj7G6iRG4oFMLc\n3Jx02BYKBXl/Os4zg+k6BwEUghAzMzOSk01PT8Pn84naw/X1tcTM9L/6uRYXFweGUnOP6S+dmwC/\nfW5vKhBiGKp/3/e63W4kk0ksLi4KukMHVa1WcXBwgKOjIxlqa5Zk6yHG5IzFN7/fj5WVFbx48QJr\na2sDbb4cBnd6eopKpYJUKoXDw0MZRsiETa+3ccN9au1p2jl/q0372KYZJl6vF/Pz8wiHw8IQJqOC\nOpCa7TAuxguILfu8oIDbYIWM62q1ilqtJnMOHrIWGrwymg4YjUwNs+0tOgIGpSzG8RxyPekY9HMN\nM2NhyufzIRwOC1OR8gYMeDQzjICOWdfqMYygD7VZOZCagBmLmbVaDUdHR8hmszJceBRNMxIIQrTb\nbZl5VKvVhFXC9nIN2HEP3tzcCDPW+JoAhMWuC6K8+4rFInZ3d5FKpSQA1nt4lPcZn4HrRADW7XaL\nrNLCwgIWFhZkzpIefm6Ul5ucnITL5YLNZhNfks/nkcvlYLFYUKvVJPg1I1jIvUE5Jp/PJ0m49vFG\npg4lEAhWGxNxBtUEKsgi5F1P1v7JyQk6nc4n5f30e72vw+KpjPNb2u02isUijo+PZf0uLi4GisUs\ngmigBxjUq+U66DZ2o+/kepN5yL2WzWaxvb2Nt2/fyjwIMhH5M0/dQUJjAqlnQuki0OTkpOwXSmr8\nFj5Pd0To+IOFLe7TYXMpnsq4/1lQ0iw3MtMBwO/3C4N8YmICdrtd5hbwjgNuZfhyuRzevXuH7e1t\n7O7uolwuS8JrNE0w0VrpXB8WUPVnZgYAgsZzOYwVSGAhHA7j+voawWBQCuHGuE2DGfF4HIlEAi6X\nS8gSBCjL5TLS6TTy+fxIsoP5zC6XawCE4NBtMqHZAfgQSVKaPo8Euk5PT4fOWjPTnnoMI4DHzudh\nXQ76jBGEGLc8FbgtaAKQmFh/3twjjM94pobFqzpuqNfrQihzOp0yW4Jyh3pGIf90u93i0xkD9no9\nALg3dvmWpvMiPZdsamoK/X4f09PT0sXP+Jffy583rhkA6fTy+/1YWFjAs2fPBkAI7sVGo4FsNoti\nsShSQuNGltPkHPo04NO5EfMOKnBsbm4iHA4PgN9XV1dCKKFcrCYSm8l4zjThgJ8zzyKl8pnLa/Lr\n5OQkTk9P4XA45DV1/KV9sVE2yEhgnZqaEmDD5XLh4uJCfob7k/Ej5yKawfiMJHmREEcAhed1dnYW\nDocD4XB4YC8w5qOvYKx3eXmJSCQCr9c7QDJ+ahKIqUCIzzUGetFoFC9fvsTW1hYCgQAmJyclQaBc\nzNHREZrNpimShK8xbshAIIDV1VWsrq5ia2sLm5ubA0MJefHX63WZh/H27Vvs7OygUqnIhaA3nTEB\nvS9INCb9esiRdvSfeh0zmT7wDEDoSC8vL2Wg0kN1/UfFhoGAugjS7XZRKpVkBsRDmAwa9fb7/cK8\n08xiLdnEAvtDWri/lRmdBh0F2cSdTkeYOZ1O5yPmpj6HGsW32Wwitfb8+XPMz88LE1ufdQbALAyM\n+l13lzHhJbufya4umuvOkE6nY9pg7UuNz0YG0dnZmfjBq6sr2VPcTzc3N2g0Gsjlctjb24Pf75d5\nD3pYLvUk2e6ph0pyv+bzebx//x65XA7dbnfkGUxGEObm5gazs7My9DESicgcBK/XC5/PJwUkrq3W\np+71epLY8fySRcthaRaLBc1mUwaG8+yabQ01aMCCvk4sdKJFiRcO4l5aWpL9wS4uq9U6wCD0er0D\nbcD0Lc1mE8ViEblcDoeHh1JkBj5O5o2gwzC231OZZmqy0HhwcICrqyvk83ksLCyIzyOI6na7B4Zi\n8hn1EHMmTvf9XiYaHBx8fHyMTCaDbDaLQqGAYrEoM2Dof/RaPTUQoYvolBPiF5NXDfwZWa2P8d41\nm50EA+A2kdaA/10F0Kc0I6ONf9dkCbfbLXEs7yruQcYVZCtWKhUcHx/j3bt3yOVyA+DVfaaL9kx2\nz8/P5b6kRB0lcs1SNKYvJYmr1+sJkYR5jt/vx+zsLMLhsMRe2njf8zMgkMufJzu4WCwKqM8zaZZ9\n9DnGZ+QMkmg0Cr/fL89Kn0HixJfItt5n3J/s/tVnUMe+ZtlTj2GM/zQTmveh9guamGQ2du9jmGZZ\n04cZ43sjG90IPhj3H/cM95HuOG42mwNFQI/Hg0AgIHGc1WqVjmHGLHpGjhnOswamKCtN4HR6ehoO\nhwOxWAytVktyVfq3YZ0jPPculwsejwfJZBLfffcdvvvuO0SjUZHUIVjITupcLicMfu0/zLBGDzUj\n+YV33ud+9mS6BwIBRCIRkULU9ZBut4t0Oo3t7W0UCgVT+wqdK+h7GLg9s4xTqaxBMMtisaDRaEj9\nRIN6LKQT4GdcyDyLHZUTExMCEvLfCXQQMOO9wToBScZmk1HTeY7xTiNRgkSuYcDq6ekpzs7OhFys\n55jp+aoavHmK3H7kQAjNWo/FYnj58iWePXuGQCCAiYkJXFxcoNlsIp/PY29vD8fHxx/NPxhFY7IQ\nCASwubmJP/3pT1heXsbS0pK0vzLA1UzgnZ0dvHv3Dru7uwPrcFfLzX1Bi7EDgpub7B49YEajoGa3\nYSAEGa+9Xk+KeeMOQmgzXkxkXpdKJeTz+QEQ4kuM6LfD4UAgEBiYYcLfq0EIMmTNDEIAg7MKNAgR\ni8WkgEJZIM1a1wk4z9Ts7KwUhdfX1/GnP/0JGxsbMqdEn8O7QIhxNc4t0Wg+k30dbBOEIGts1I1J\nKAEIFsN1oqU/90ajISxXSgyR2U52F0HAcDiMiYkJKZjz97EQlcvl8P79+4+k/EbZuF/or3jmOPw9\nFArB6/UK0KUH/moAghJ9XH9Kx/AOIPNmamoKtVpN9HLJQjYjYMj3o8/V9fX1R4wZ3ZkZiUQkCbdY\nbnWCOTB+YmICPp8PsVhMWqQprXN9fY1CoSCSQRcXF8jn8/cCqrrIp2MYHXt8LhvtMU2DEPRhBwcH\n0pGQzWYRi8UQDAYRCoUQi8UG3jNjKeD2bmecdZ/GuQYps9ksfvrpJ/z666/SDaZlroYV7e8ipHxL\n00VNzi9gsknms45HNBDxWO9Z+2ImehaLZYAUwUL6XYzkpzQNoOhn0YCKy+WSIhkLAHqQLwEDyuik\nUim8ffsW9Xr9s+9/i8UixXsNQnDYerPZ/AiEeGrjPae7XU5PT0XiV3foBAKBgfes7x4tK8eBjyzM\nA5CBusViEbVabaCjZpRMk2boAzTgwn3COVaUTvoa4zlnbHJycjIAQmhJDjN0xD2G8Zk5X4N7xgiA\n6hyeRfFR21Ofa3xufRfpdfhcf6B/hjkU42stb8dOgXA4LKCbzWZDIBBAv98X0LLdbsssBDMUinkm\nGEcQuDs5OZH41OFwIBqNynybiYkJ0cvXuSXPFjvEwuEw4vG4EGK/++67gW4vdoOWy2Xs7+8jl8uJ\nfPO45Kk6b9dSoPx/n3pGq9UKj8eDYDAoIISWc+73P3S2ZjIZbG9vo1gsmlpmTYMwNKNvpF/kHc6f\nOzs7Q7FYlJ+j/221WiIBzHqjcYg8ZyWwq4S5BwvvjBtZwAeAbrcrsYiWIDaLaR+m33+/35ezq7s2\ngcF77OrqSs4qY0LmpTq3G1br+5Z56ciBEDMzMwgGgwgGgzJEze12Y3p6GmdnZ0ilUnj37h329/dl\neKaZi5efMl5ufr8fiUQCGxsbMoQ6FArJoBU6mlqthmq1infv3uHXX3/F3t4earXaRwEJN7e+KIe1\nlQ8DHowbmwUxI1o3Kk6GhU0yXnmhMbioVCoyOO6hA4bNbDrgGoa68u96NgZZqp+D+uvXZ+FlbW1N\nNHIpEUNjEM1gibrPZnEQxsIX3y/lN8gyZycEGTOBQACNRkPYqHdpdJLtb7fbhV2mB1Xx+zVLgPfc\nOO1Lo7G4GQwGsby8jGAwCKvVKncZ779CoYByuSyzW8ZhTbj3GeTx70b2HzDIRqEGMwsrzWZTiiJ+\nv1+CO5vNhkgkIq91dXUl+5nBoNmBwM8xXWhiQZGDqcnAdrlcct50lx+D51arhUqlMiBLR5mcQCCA\nRCIhwA6ZOJR4orYn13cYgPTUZrFY5DlbrdaAzzPuNS09wqIxwXw+KwNmAmFOp1MCZz6/z+dDPB7H\n6ekpcrkcgsEgms0mut0uLi8vPwJqgY87JO56j9/a9Pnj50yg5eLiAtVqdWDAtB4ar+cwUTOfc89C\nodCAHJj2PZ1OB4eHhzg8PMT29jbevXsn8ml6MOen1sgM+3B6ehputxuBQECGolssFpGTY7yrn+Gx\n3reWGDPOxjo7O0Oj0UC9Xke73R4gE5hh3YDPOwM8R1qj+fz8fAAUrNfryOVy2N/fRzqdlvkXd93/\nPJ/sjPL7/QiHw6LDPDMzI4VRPV+NMrlmGCKsfUOlUsHOzg6sVqvIOGigRndgaUIACy1GYgjX7ebm\nBicnJ9KlnsvlcHp6+uR31kNNy8rZ7faBOTU6p6AZ2Z0PMSPoPGzPG0l2o27cmyQ+MOa/urqS+0kX\n+SgzMurx2n1mvPu/1hdw/wwrzvG8897qdrvw+XxCwggEAiI7Nj09LV2xZuiG0PdaqVTC+/fvYbFY\nsLi4iGg0KrPQ5ufnMT09jUQiMdDpp6WmCBjzZzweD6LRKLxe7wBRgnFPt9sV4JkgpM5Tn3ptvsY0\nIZp/fk5npr4bg8Eg1tbWsLS0JEVzLcN0cXGBer2Ow8NDvHv3DqVSybQAhDaeI/3fwMeSkSR40Acy\n32A9hWQPgs38OUoZaqlrLanu9/sRi8WwtLSERCIhAMXs7Kx0niQSCVxcXKDdbiOfzw8oCjy1Md5s\nt9soFArY399Hv9+H3++Hz+eT3IBfWg2CeVu9XketVsPZ2ZmoIHBWIWdG6JktWiqNa/AtzqdpQYhh\nD88iVCQSwfr6OhYXFxEOhyWpbbfbSKVS+Pvf/469vT00m82RL0LxsvL5fFhbW8Pz58+xsbGBpaUl\nKSYBt7pxhUIBR0dHePfuHX755Rdh8OtB1PpP/v1zL06jXuqwwvWogRBkOvj9fmE4EC0lCHF4eIhy\nuSxDrsbN9Gdp/Ax1iyoTSBbuaHcFW8Z9Ew6Hsbm5Kcx+trVq1hJRcbaN6iL7U5tObDSbmu3h7XZb\nhgaxhTcYDCIQCGB5efkjBqBugSN7hjIJU1NTiEaj8Hg8UgzRCZhuseX6mKkg8pjG556ZmUEoFMLK\nygqCwaAAM8AHUKZarcpZJWvTLODVQ0x/lkwAjICD8fv435RC4D5ptVoDQ4EDgYBo70cikYEzf3V1\nNdC6zXbYUV5LYFBPWIMQZMiwPZoDwDRjhK32usuyWCyiWCwKU25paUm6Knhe2fLOORNer1c+D12c\n4vt7SuM5o5RQq9Ua0L2n8e8EWli45bNyRgSHxhOYZdJgBHJdLhdisRiurq5wdHSEUCgkybMmP+hk\nz9h5eV9R6lub/v0svrJLl90yepYQ5an0AGF2w21ubkqnErtP+Fws6pZKJfz888/4y1/+gnQ6LcOn\n+bv1Z2b8u9mMTFMO0OMdf3l5KSDEb/X5sguDM4e4pwHITKxarSZFFTORI7QZ12fYmWDccn19LTE9\n5zjk83kcHBxgf38fmUwG7Xb73mdlQYbgdiwWQyQSQSgUEnCDIC6HOTPh7ff7ppnlwvukXC5je3tb\nutp8Ph8ACIuQ/tVYaGGczDuTXXNM+pmn5fN5bG9vI5/PC0ho5jM5zHRsz/udd5i+rzXhjT/zWM9q\nvPv1+7prqOmoGkEIPQtDy7rq/UZ/Y8a76Wvsrs/zMYvaxjtTx91Wq1VmQ97c3Igvpq68w+EYqI+Y\nodjOZwCAcrmM9+/fi2yNx+PB5OSkSNuEQqEBJjUBakrUXVxcDHTCsqjJeJdnnSAEQRsNnGlpp1E2\n+jzjLL1PyQuye2x6ehqhUEhACJ/PNzDngF059XpdanocQGxm0wCEcQ2MnZmsNzHuYFGcRNezs7MB\nMF9332kJJv3l9Xrh9XqxtrYmYBkL7ySW8cwCQKFQgMPhEJkr/V75PN/amDcQhNjd3cXl5SVisRhO\nT0+lk5x5A5/p5uYG+XwehUIBpVIJ5XIZl5eXiMfjiMfj0s3FWhFBmFKphFqthm63OzDH61uYqUCI\n+x6awY7T6UQymcTvfvc7QQ+npqZwfn6Oer2ObDaLvb090U4btcBOGwsYnAWxuLiI5eVl6f7QQ6hP\nTk5QqVRwdHSEN2/eYH9/H9lsFvV6XZiMxqDsvqRdXyR0qBySwkBTM/K0nqIZ0P8vMe6rQCAglxXR\n/NPTUzQaDRSLRWklHKVn+xzTyQRwKzeiUX3NsGHiqlsGhyUBDFIY7Hg8HmxtbeH58+dYXV0VlpzW\noScCTCZevV43pewB8LH+d6fTQa1WkyIcHQQdny5E8pn4xaCN7EqdyDHI0ZJVTEhYKOTrmQGo+a2M\n96HT6UQwGJSAH7hlDlB/tFQqDS2ejqJ9Duhw18/pPUrj3c3kgzr+3GeanUJJq3HqtGExiKBCp9OR\nc8QE32q1AriVGLy8vBRd/ePjY+zt7SGdTqNaraJarYrGf7/fh8/ng8fjQb/fF81O7l273Q6Xy4Vm\ns4nZ2dmBz0Z3ufB9PsXa8Lk1wMmgVK8H9anL5bJ0hfB5WSy/uroS8EHHDsYkhQwln88nbf60Xq83\nwJyamJgQphTvQu2DzHTmdQGAa6pZuhpY0YWLfr+PQCCAQCAg/mJYQbnZbCKVSmFnZwevX7/Gr7/+\nimq1KuCRWdbhS0zLBGlwhntJD+Z+zCIjSU5aus7r9WJ2dhbX19eSrOlY0Ex3Is+GZqLqwhGLkycn\nJ2i1WgPJPvcT5yAUi0WkUilkMhlUq9WPOj647kz8NeEikUggkUhIFycLdJoZq4FtM82x4nnlueJ6\nnp2dyZ6Ym5sb6Frq9/uy1vy6ubmRThreiQQgGo0GSqUSstksarWarP+omfEM3iXzRmCPUhBf0w2h\ngQwNcvDe112fZpYteYgxVtCgts6LzATCP7bd1U0D/LYdkDoeIqmHX6x1zMzMCGjNfU75IzPEIjom\najabSKfT0rnE2ofT6RRCBONYFmqZ31JjXtd5OBeMOS7P4uXl5Ud+Rs+iGqe9CXze2eOeJVHH4XAg\nHo9jbW0NyWQSLpdrIP6jrPzBwQFyuRyq1erIkeqMgARJSw6HA16vV2JfytuyjlKv1+WOGxYfDAM3\n+NXpdESOOBgMigws52mS1MLX9fv9mJubk5j5qU2fvX6/j2q1iunpafR6PVQqFQQCAQFh+MV9c319\njXK5LJ0NjUYDAARwCYVCAjIAkNc8Pj4W4ua3rrWZCoS4zzis1e/3Y3V1FX/84x8RjUbhdDpxfX2N\nVquFYrGIQqGAfD6PZrM58lrg1AKmfn4ymcT8/Dw8Hs+ATESv10OtVkMmk8HOzg5evXqFUqkkmn5a\nI3NycnKo1M5djkGjvXNzcwOdArw8WKzXBa9RcTJ8PofDIYVNrWFLdnuj0RD2+jiaBiF4+Wtda34R\nmOC60XRyyuSAA4L8fj+Wl5exvLwsQ9WpH0uHQHkAJmrFYhH7+/solUpPcjHeZcY9rfXhm80mSqWS\nFN0sFoskrP1+X2RLqKtJbWk6DzJlu92unC/97JqZc319Le2/RLDvctjjZARDmdxr59vr9UTmoFgs\njmxyP8we8zl0UsLzqvW7JycnBYRot9s4Ozt78oG/j2lcS+4P3lUzMzMIh8OIRCK4vr6WDksCFW/f\nvsXbt2+RTqdFz1sP+2u1WiIX5nK55PyTBQxA5uFQx5TgmZmKBppRqWfOsBBJhlK9XkcqlUI6nRYZ\nSALOgUBA/GW/35f1vSuRYGxitVrh9/uxsLAgSUWv15PuHcYqnU5H3iMTM2MrvBnW0mhGkgfvLi3f\nQDs5OYHVapU5GxrA4c+USiX8+OOP+Mc//oH9/X0Ui0UBpM34/J8yo49jDHx9fS1FEiMo/xjGtbda\nrfB6vTKzIxAIYHJyEpeXl2g2mzLXg11MZlpjLQ+hGYWMP8hqrdfrKBQKqFQqqNVq0uWg512w44Mt\n+kaSiS4oUCI3Go0iFoshHo8jEomIHFu9XpfXJNhLAIIghFmKK3xOzvnp9XqoVqvY3t5GJBJBOByW\nzla73S6AbL1eF8LXzc0N7Ha7xLzMh/iZlMtllMtlVCqVkWC23me8r9htyQKHLr7SJ7AIpEED4GGx\nDfeeZl9zH1FHvN1ui2zsuJielUfmsI7ljADNOHSC6CKm8RmBQZ/xJSSdLzHtk4blwixu0keRSGaW\negHXgxr83BudTkfkzf1+v8za0wCPJiISgKE0jp4JypoUY0QWgznwWg9UNpPf/BrjurCofZ+mPvft\nzMwMvF4vQqEQFhcXsbKygkQiIbOHuN6FQgGvXr3C69evUSwWRyqmM5IV+Hd24Hi9XkSjUYnru90u\nGo2G5EKaBDssN9LghvH3UTayVCrh+PhYJD3179OdA263WzqBTk9Pn7QDQj8Pz0qj0cDl5SVqtZqQ\nuTQArxVMmLcTMDw/P5euiampKSwuLg74aNbaSNx/CmnIkQEhKHEQi8WwvLyMra0tQWBrtZqgOQyu\nR5UFpo3MLM4rYHszB/nyEPV6PWGOHx0dYXd3VzYinaBuVzIe5mFJu3b0LCaTCcRC/eTkpOi1AZCA\nfNi8CTObsf2fcg9sfWWB6T5N3FE23QnBy89Y3DFK/fD7deGEr0P5DUpyxONxvHjxAs+fP5e2MHac\nUOqAQR7Z/WQAlEqlAQaFGUzvaRYYOcC8UCjIhX91dQW32w2XyyXfr2WTWNhjIbPVagmLlWvDIrAO\neglYcHBktVoViaxRl5+7y/QeJVtnGAhhnN/yLxtuLL5rLWcW+AgIEoTgXIlxSR6MLC9d4GUydnZ2\nJl2W1NZ89eoVfvjhB+TzeekQof88PT1Ft9uVYYUE6/1+PwDI2SXjn+Ck1g7X/vKp1pm/1yjtoIuG\nBGWKxSIODg5weHgoWvnUSNYEiMnJSTmvxiHf2lhY4pBTJrxsP2bHK4Nr3Rk2SlKQn/MZ817jGdW6\ntwCE0Z7L5fDmzRv8/e9/l0KoWfzkl5peF8YcPJs0zag2dtR8zWfOWNfpdCIcDsvwcI/HI6zqZrMp\n3VCU4zDTWutOCMYHnU4HDocDACTWYGdvPp9HsViUgdMnJycDQASfm8UPY4crGcB2ux3JZBKLi4sC\nQJCtx1iaBalmsyldJLqYZzZAR3/muVxOihj6y+l0SgxH9uHl5SWmpqbg9/tlXbhHzs/PZe2r1aoQ\nm0YZhNC+1AhC8C4mgVDHFw89N5pNzDxDa6hTFYAghFE2dpSNa31XTgYMxsmPDdR+S7tPyoZxAvNP\n4DYPAwZrGo9ld8UY9NG6c0+D5A/t+Hls43vV3WeXl5dotVqIRqMyFDkcDsPtdn8kZ0ZQgRJ6Z2dn\nMgdsbm4O/X5/YFYX8/harYZms/kRCDEuxliFYNRd5xIYlCzkXMOFhQXMz88jGAwK+ZUgY6FQwK+/\n/ort7W1UKpWRrz3xnNhsNng8HoTDYTkz7HxotVriH+4CIGh3/RtBMpItfD4fEomE+CLGk1xrzrEz\nDmx+StP7ipLId3WD8f0Oy38sFovMF7Xb7QN1Iub4rJ2zbv6tu7dGBoTwer3Y2trC73//eySTSdjt\ndmH112o1/Prrr/jb3/6GdDo9FheddrYcWqjbWXUwcnJyIrIIHOxiXAN9mHWydxczUet8chhKKBRC\nOBwWzXuLxSJJmUbJp6enTdeqPsx00ZwXFbtMyFriIOFxLvACH4NRen/pYItDRzmkT2v18YInWBWJ\nRBCNRjE/P4+lpSUsLi7KfAMWlfSFT9ZZLpdDKpXC4eGhSAGYsbBER3FxcYHT01Npl6Q0ms/ng9vt\nhtPpFOfGrgmd7BMwbLfbaDabODs7k6B2fn4e9XpdWgatVqv8XK1WQy6Xk6KokYE2TmYEICiLwICF\nSXC73RYJr1EP2n4r4703Nzcn51PrkZI1y8KRBiHGyfg8DH77/Q/ySb1eT1jQU1NTwlw9Pj5GqVQS\n3Ux9NxJE4JAztuf7/f6BwgvB/HK5bFqmIguOLO6QxczCYaVSQaFQQDabxeHhIdLptNxh09PTUvzk\nndbv9yVeYxFTMxkByD3KohEL7/Qx9BUcDkxJF2OsMw57lGuTSCTw/fff48WLFwiHwzL0FQCq1SpS\nqRTev38vspujLj8K3LIAmYAxaddyrJzV0Ol0HiVpJDvPZrMhkUhga2sL6+vr8Pv9mJyclA5FdvuY\n9U40xiPshHA6nUJWohwQASsCrCcnJ+h2u0JAIfioSUwEFRwOBzweD3w+nxSwQqEQgsEgvF6vAI7N\nZhP1eh35fB75fB7lclnY6Rp8GCbjYwZjDKxlEc7Pz1Gr1ZBOp0WmizEfCwUsjF9eXsrf+/2+zOxL\npVKo1+sjkSPdZ1wfFs0o/cXPVsu3krFPv/LQ59ZxoB5UTxJKq9USoIszW8bFjB1IunCm5yYxd9Bd\nIqNuRgCU8m9afo4xCO+wr+m0GWaagEdygNVqFRayBkUe8/c+pvG8Ml89Pz8XoJWEObvdPgDyMG8l\nyMiYl9J7LpdL/IQmxlarVeTzefEv4zajhM9q/Ldhwe94DAAAIABJREFUn7vuOHe5XFheXsZ//Md/\nYGVlBU6ncwBIpWQh6yD5fF6IvqNkxtoilSAoV+3xeAZUI7rdLlqtltSWAAzEBZ97nvh9ejg6JSl5\nj/B1dfeKGX2xrsvdJUFFG7ZWOjbW30v/zPWhpPdT1DhHAoSwWCzw+XzY2trCv/3bv2FhYUFkTa6u\nrlCr1fD27Vv87W9/Q61WG5viE1lgGoTQhSIdABOE0Mxp42YyAhHGS4IbXV8YLpcLkUgEiUQC8Xgc\nsVhM5lGwXZ76rrrth4G3mZ2OEYSIx+MyQZ5DlQhC6KLTuNmwS16DEPwMuVZ6wCi1mulgdWFzeXkZ\nKysrSCaT0vKpBwoZ295OT0+lsJ5KpXB0dCQzEsy2jzRjmMHV1dUVWq0WyuUy7Ha7zIBgkMHElGAd\nk/1ut4uTkxMBJ66vr6UIx4FBlDcg26vb7aJarSKbzaJYLIo8yTjuT+A2iNEgBAubTIJ5ZimLMC5+\n4DFNB2Jzc3OIRCIi8ae7kjiwShfcxtH4rCxiUNIrEAggHA5jcnJSGMNkgg0rHhEAZBcX2bDRaFRA\n1+vra9jtdni93oGivGZzP7XpGIA+nHORGKhTKo8ARC6Xk//HIki9XhdA0G63IxQKwW63C7CgB/Dp\n4imLRvw+6ivr7kT6ZM6rGEauGGXjXRePx/HHP/4Rz58/RyQSkTklAFCr1bCzs4P3798jk8mg0WiY\nspD7JcbPUCeFjENZWHO5XMK+5IDzrzWCEC6XC/F4HM+ePcPa2powQlmoYcGeA7/NttaaRalBCHZt\nUDau0Wh8BERwUPRdHUWUkdCEnYWFBayvr2NtbQ0ul0vuOcY59XodpVJJQIhSqSQghBF8MOveNUqR\n1Gq1j1jC+t6xWq1SIL26upIY+ebmBt1u9yMQYpT9qs5B2WXDz1Z30Wltby0fwdf4XNNyYxqEYC5K\nEIJflDIdJ9P5uZ6Xw5gOgAAUMzMzYwNCAIPEzLm5OXi9XiFl6hkMlA/9rYhrBMIIQgAQX2UGFvV9\nxnuWeXWj0RjQl9dzmOhfCbLwvuL3LS8vy6BbPdeMHRe1Wg35fF5IYeMIQhh9130xKPeuBiGCweAA\nCHF9fY1ut4tKpYJ8Pi8gxDiAqYzlnE6ngBCcjXFzc4NGozFQ6+See8iZYixE4gX9gFaD0TWu+6Sf\nntq+5v3cBZSxbsVaUrPZfLKuTNODEExEOROBbOqJiQkpkuRyOdFo1gP8RtU06q/1z3kYjc9HqSp2\nKlSrVWFYk33CJGJmZkaYAnTUOhGgQ6IkAr+i0SgCgQD8fj9mZmZEfkIfWiPL0ewOR68v148OgcP5\nMpkMms2maZOkxzJ+hufn51IUt9lswjaZmpqC1+vF4uIiLBYL3G43AoGAtGZyQBWHLcViMRlQSEer\n5z/oJITJSjabxc8//4xffvkF6XTaVLMg7jL93ug02e3AeSI2m03OBb9HSxEQfOC/a+SaA8Go8cdi\nO2XC2u22IP3jvD9nZmbgdDoF1NEDSsm6KRQKwtj+FwDxsRkTWN57LIprXcmzs7MBEMJsgdljGosp\nwAcdfhader0eJiYmpGhHRucwsI/rowfy8avf70vLOgNxFg90y75+rac0+oF2u41qtYqZmRl59svL\nS9HF16QHHdQzjuBa6eIQCyNkO+n9RgkUgrQ3NzdSRCHRodPpyD2oCw1mTB6+1CwWC5xOJxYWFrC4\nuIg//elPWF9fFyCLjM+rqyvk83n8+uuv2NnZQb1eH6v7TreZ63/j2fF6vQiHw7KfNBD1JcZ70OVy\nYW1tDevr63j58iUWFhbg8XhkblO5XMb+/r7MgtAdAmYyzUrXdxDBFAAC4nu93gG5JU2mYExD2Zu5\nuTkhVLjdbhkcr780+MBui0wmg6OjI2SzWVQqFZH2M864GoWzq+83YPhgTACi2cziCucgMJar1+sD\ncl6jbFwTkm9yuRy8Xi8CgYDkmf1+XyQ4qGX9UM3tqakpiVcikYhIXvX7fbTbbWSzWRwcHCCVSonk\n16ivsTYNllLOUUv08c68vr6WDpRRfX69NxgjURra4XDA7/cjHA4LyHd+fi7ScOyGYVfOY+TurGnY\nbDa43W74fD4Z5sx4hZ2ZWsrYjPeaZooTRKSMNkE+fhnlXvr9vgBcV1dXsh+tVqvkt9Slr9VqQt4Z\nVxWJz/l8NekrHA5jYWFB5B45d46v0+v1cHx8jDdv3mBvb0/yDjPGG59rw9ZI1yQZ08XjcQAfQFSr\n1SrzLglgsSPxc9achFnW9TgXQpO4jeCbJkaZYS7EYxln3lJunmRDTRTgvfUUzzsSIAT1+ufn5zE/\nPy/tO51OB9lsFplMRhLicbnsjE6XIIQOgnXgGwwGBwYB0RFTJoItjHa7XZwmUS8GktfX17DZbLDb\n7YhEIlhbW8PS0hL8fj/8fr8wT1hcaDQakhTrwqmR4WlW0yAEg2e73Y6JiQmcnp6iUCggnU5LIWkc\n9tVdpkEIMujcbrc4i4mJCfh8PkxOTsLr9SIWi2FxcVGkN+hkXS4XwuEwwuGwyHhxhggvP33B39zc\nSACXTqfxww8/4NWrVyiXy0+iT/elpoMzFokIQrD7QQ8RMoJ+RiSehTf+G1mNZJlRI50Mx263K1qb\n42wEIbxeLxwOB2w2mwTJvV4PpVJpYGCo2ffNtzZjKz8Zmywssa2cBXlqupLFNKrJ7Ocan5tMTt6D\nFotFzt2nWnY1+4YAYavVkmBXd/Mw6DWrr+TnXyqVRJeUAEG5XB5oszeCAQAkaSJLkVJzfG7+fybs\nGoTQHWEkMzC20XIuwwqZo2o8m263G99//z3++7//G6urq1hdXYXH4xEdW64VQYjd3d2RbNW/z3TM\nAUCINBqEiEQiKJVKUoT70vtJF9hDoRC+//57/Nd//ZfMNJidnZU9WSwWsbu7i2w2i06nY9ripmbs\nn5ycoNlswuPxoNPpyNDU2dlZeL1eOUN6AKQGIRjHMxfQmuH8O4solHvi7yYAkUqlcHBwIHMn2EWm\nY5VRPbPD3jf3KOMUDrwEID61Xq+jXC4LgWvUjT6v0WggnU4jEAggHo8Lu1WfMxaBjADj5xhfx+v1\nIplMIhaLIRAIYG5uTjoQU6kUXr9+jaOjI7Tb7bHrDGYR3OVySS4+rNuBPmKcuoH1/e90OmWfcU9d\nXFzImkxMTAzkjvex0z/XGDdTzz4YDErxnUQx5mKPCX78VqbXhLmoUWd+WFxKaabZ2Vn0+32R5tNd\n6ZR7qtVqKJfLY92lr5/prjuN6zk3N4dYLIalpSVEo1GJ69hxTGn1w8ND/OUvf8He3p7ks6O6dvcR\nDXQ+xNoSFSRcLhcymQwmJiZQr9fR6XQGujT16xuN603yejweh9frlZlEBG0JPjA20god/CzNfIY/\n16ampuByuRCNRuF2u6UOR0IsOxWfqnZuahBiYmICfr8fCwsLWF5eRiQSgcvlErS/XC7j/fv32Nvb\nE327cTQeCKLWRokczm3ghpqZmREgwghCzM3NSYDCTccE4urqStjvkUgEy8vLSCaTosHLA9zpdNDv\n9weG1ukLYlQOLhOyUCgkWra6sFmr1UQDfBSe56HGz48stnq9jmaziUAggH6/L4UyMj/YShcOh6UQ\nPjExAbvdLoCO1+uV/TYzM/ORg9aMPbbrv3nzBru7u0in0yIpNgo2zLnqvxu/jD83LEjmv2u0mmeU\nLXSUcDJrUeQxjTIQmj1CUIedEKVSSYKVcT6vDzXNPuHMAkqbGAMTdkR1Oh1TSo/8FqbBeOplAvio\n2H3Xz/J7tTxYt9uVtaZcoZYaMquO8Pn5uQAonI/EOTbsDKHOryYgcI+RZU7Nbq4t10bL03Gt2fFF\n5pcGaQm+agCCv9Msa/Y1xmR+fX0dL168wB/+8AfpPKU+bq/XQy6XQyaTwc7ODjKZzEgPor7LeIZY\nsGYSTnab1+tFNBpFNpuF0+kc0Km+i3WngQ0mnYxhlpaW8OLFC2xtbcHtdsvAeO71YrGIVCqFcrks\ngKwZ9xzPGO/uRqOBQCAg68MBwUz4eT49Hg+q1SpqtZo8G8+ww+FALBZDPB4XSc1AICBMYCbtp6en\nIgubyWSwv78vUhJa0lSDHMPe/6gbQZuFhQU5u7zPhnUXjrrpToh8Pi/yjmTpU1KYOSSZ65ph/SnT\ncwDC4TBWVlYQi8XgcrmEMMZZcoeHhygWi2NJnJienobP50MymUQgEBhg9mrT8xnHCYTQg23Zic+c\n9Pr6WsAZ1iYsFot0tj7kztY5GzvB2OHPwiZlNjudDprN5sCcKrP6CaPdF9cac1k9d4T+xOFwfESS\n4HqM0133KTPGwfxvkpB8Ph8WFxextraGcDgsZGrgw1y6breLUqmEVCqFnZ0dlEqlsVg7XV+iPBK/\nKO3ucDikS5Oz9Nj1VqlUBIgga18TqXU9FIAovgQCAZEFZ3feXXP4Rn2N7zJNjGDtZHp6WkgnBE11\nd+q3NtOCEAyCk8kk/vM//xO/+93vEAwGhSV8c3ODTCaDf/zjH3j9+jWq1epTv+VHNePB1UO/9DDg\nyclJkcxh18jS0pIwCjWriYUPJv56OByTPiZoLpcLPp9PBjVz4zJBpNyMbpEyDiIzewDEBCuZTMLn\n84kj1QlDs9mUYtS4mk5eyX4Nh8OIxWIAblkgGjkmI5EoKplK1MnUbF9e+lpOjDrLxWIRP/74I376\n6Sfs7u4il8uh1+uZfu/cZ8aLnF0SwN1siWEghu4Y0YwVBnnsgBh1FvDnGHXlI5GIFD8ASMLRaDRQ\nqVRwcnIy9mvxUGNwTDaX2+0WqT+yiXUnDwvD46arfJ9pJr+xJfdz9pU+p+xiurq6kju03++Lj2Wh\nhmCamYy+ALhNkHjXE/zUYACLkfQTZCx6PB44nU7Y7fYBJiw7RrQkHRMTzhiiX9LyBka5mC8pZpnR\n+P5DoRC+++47fP/999ja2kIsFpMBkcCHfdXpdPDzzz/j//2//4e3b9+i2WyO7HPfZbqbqNvtCshO\nPfmZmRl4PB4Zhuz3+4XBr+VFabpwwuSUnb2cabC6uorl5WX4fD5hw11cXKDRaKBQKKBYLKJUKonE\nh1nXnKQFnlcChTc3N1I4mp2dlQQ/GAwimUyi0+mg1WrJEHkCFtPT07Db7bJedrtdvsgcJKGpWq2i\nUqkglUphe3sbe3t7yOVyH+mBD7tLzdYF9hDjOfZ4PFhdXcXW1hZCoZBIlDDh5wBwM++jzzVNYGu3\n2ygWi8jlckin09K1QPDd4/HA5XKJzzs7O/us36EJdD6fD/Pz89jY2EAkEoHNZsPl5aXkEcViUSQ5\nx0FD3WhWqxXxeBwvXrxALBaDzWb7CIDg/aklC0fddD6kyQ0EIjweDwAI+Y35JcmqD5EZ0WSKqakp\nBINBzM/PY319HRsbG1hZWYHD4cDU1JTESpyDpcmZwwhqo3TudWGdf9KXzs7Oii9g/sDclP5kXO66\nh5hm5JPYu76+jmfPniEYDA6w7bvdLnK5HPb395HJZFCpVKTrctSNnz27q202G8rlMnw+H1wuF25u\nbmC326WD0O/34+zsDAsLC6jVahJb1Go1tNttkYllJ7ZRSoh5B+O7Z8+eIRAISDc63xNzC5KrtHzd\np0hno2D6DjPOImS3Ejvcn1I9wpQghG59W1hYwB//+Ec8e/YMPp8PACS5T6fTeP36NXZ2dj47qBkl\n01I1lF85OTkRR0wHyYIvW2603AsLoHTgdBYEHnSSf3V1NaCTxtfXhQKuPeUmWDAmw5EAhNllschu\niMfjSCaTMuSK69Dr9UROY9yZwLyQ+/0+Wq0WisUiotGoJLDcO7pVTSeTeu6IZvsbk0t+L4cvFYtF\n7O3t4YcffsD//d//DSSs42J3FTDva0nXrE39fbq4Sec7KmybrzEWzoPBICKRCBwOh+xBBr6NRgPV\nanUsZgL9FqbPJAFEj8cjIAQL4TowIzP9qbQin8q+BHQY9nMABggEeigazzOLgmwNNlvAe3FxIW30\nZ2dnsFqtsjfYmcWir7HQSwlJdswRhNCt55oAoTsdNOiluyeG3XPDustGzRhjRaNRfP/99/j3f/93\nrK+vIxwOy/ewsFyr1fDmzRv87//+L1qtlnSkjovxWZggtdttAbwovzczMzMwrywajQ50AmqCjj5n\nLJjY7XbE43EkEglsbGzg5cuXWF9fl65NXcQvl8vIZDIoFApSFDBzUW8YCMGiGN83pXHYXcMiHc8e\njXmAnm8wrJv17OwMzWYThUJBOiB2dnawv7+PZrMp3dj3sdLHYQ8zRna73VhaWsLGxgYCgcAA4Mpu\nkacaAPlbGO9ono18Po90Oi0a8QSeddfl7OysgH33xa/0K+xaYpfFysrKwGxCDRZSlnkcinfaWMyM\nRCLY2NiQeQjG/EATykZ5JsQw434gOOpyuRAIBBAKhQT0otQcyQwsUDKn/BTrX/sN1lZmZ2cRi8Ww\ntraGzc1NrKysYH5+XmKX09NT1Ot11Go1KeoNu/OMZLxRMmMdiUQaro8GIajC0el0/inzB2Awh3c4\nHPD5fIjH41hdXcXKyorM+GLtrNVqIZ1OY2dnB9lsFrVabazqTgQGr66uMD09jXK5DI/Hg0AggF6v\nNyAxx3WjDDVlvcrlsoASmUwGNpsNjUZD5mXybAeDQYTDYSwvLwvBhPVMmpa6JmFbx5HjBELo+5J+\n+erqSmaWsob7VM9qOhDCYrHAarUiFAohGo1iaWkJkUgEbrdbGBTUGz06OpJBcePkbHWxtt1uY3Jy\nEoeHh3C5XLi4uMDCwgISiYQwIYa1GA1D4LkpyXSampoa0KTXyRsLJjqArlarKJfLMishk8mgWq2i\nWq2i2WxKy/Wnko6nNkpBuN1uzM/PI5lMwu12A4AUeDnrgJIT4278vNrtNgqFAsLhMOr1Ok5OTiQR\n0BqRDKSMoMN9AES/3xcm2P7+Pn766Sf88ssv2NnZQavVGkiWx92MDBMaHbBuPeYMBK19bRwiNq7r\nxj1ltVrh9Xrh8/lgs9kA3MrTnZ6e/tO1/n6JGQtyTqcToVAIgUBA2FyasalnjbD77l/2adNJrNaz\nZzJ8c3MjHTulUkmYyk/ZCnuX6eIkYwC+Rz3Dhkkpu+A4F4hSQh6PR2Q4AEiywJ/RmqQA5I4jyKEL\nArobz6gtP4qJ/eTkJAKBAILBIDY3N7GxsYHFxUVhdtKosf/27VscHx9LYdnMMdZDjclqvV5HLpdD\noVBAuVwWv0iJoH6/j/X1dQBAJBKR4ce9Xg+9Xk8KJOzopUa/2+2WNv9wOIxoNCp3YL//YcBtvV7H\nwcEBXr16hdevXyOdTsv5NbNpZjrjtnQ6DZvNhvPzc6ysrEjnlZ6pdHNzI2ul4xJ97gDImddyfZlM\nRvKxVCqFbDYrbPSnTm6/penuYL/fLxrU/X5fZDbK5fLADJ1xMXbBn52doVqt4ujoSABDh8Mhs30o\n/eX3+z/qhDN2L1HGaXp6GrFYDBsbG9jc3MT6+jp8Ph9mZ2dxfX2NZrOJbDaL/f19VCoVkSAap/Xl\nejAfYEeJ7oKgXyZRYFTUCD7XdGyl2fiUjybA7HK5cHl5CYvFIix9u90ucw4Zu2jTXZwEHubm5kQi\nkazqpaUlzM/Pw+124+bmRnIO1kIoB3tX/WMc9iRzMRJLWNjV9apSqSRd6cazPe6m2efsdmYx/Pnz\n5wiHwwLM3tzcSD3k+PgYr169wg8//IBMJjN2dxhwG59w3irzADLz6S/4b4xRWH/yer1Sl5ufn0ex\nWESr1RKZVnZJR6NRxONxLC8vI5FICOAN3O5T1jMZN2rwkOv+KZDc7MZ9yOenj6DserVaRbFYfHK5\nedOCENFoFM+ePRsY4jI1NYV2u41MJoMff/wRR0dHpm+Rfqhphsn5+bm0vLFbwe12y+GkU/4cVqAx\nWdfyEfqLzoMHttlsIpfL4eDgQBwuW9RZ/NMtoGb+PFiM83g8SCQSWFhYGApCdLvdfxo5El5Q7XYb\nABAKhaTdzWq1AviYvf8lDFTusW63i0qlgt3dXfz5z3/GX//6V3HEo3zhP8SMz8pgWDsPBsQEITTz\nmN9D3fZxNfoEj8czAEKw4GIEIf5lH5veMxwez8GObFNlAVCDEOOkK/xbmhGA0EEvi3a9Xg/lchnZ\nbBbFYnEAhDAb64ZAANnOxq4sYHBPsQOCeq7BYFBACM28JjOReri6Fbrf7wsIQTIDgAGwlUAEz/6o\nSh0wZgsEAlhfX8fm5iY2NzextLQkRV/gg49oNBp4//49fv75ZwEhzB5jfY0RhJicnBR2M+dNUdqA\nPoBzHfx+P3K5nMhAEHCg3GY8HkcgEBBtYIJZerA1459cLoednR38/PPP+Omnn4RcY7YzOswYxzFu\nZWxwcXEhkoYsYLKbmt1Y/HnjM+rOagKTlETY29vD9vY2Dg8PcXR0hGq1KrkA1/SfwTj3gEV2Snv1\n+x9k1IrFohTmxi3OJQhxc3ODarUqBBmfz4dIJCLFJkpC+Hw+2UPD7mz6FQI70WgUL1++HJiTw+65\nVqslIES5XP6okDQupmU13G73R6QwfT61IsE4FICNAIRm45OkpWeFMl/gGk1OTqJSqQCAdHLqQcyc\necifcbvd8Pv9CAQCiEajIk0ci8Xg8XhkKHOr1RIANpvNDoAQD+2mNbvxXBKE0LkDZ2MUi0VUq9Wx\nBFw/ZYz92XHodruxsrKCP/7xj9jc3BQZOYL/HOBNEOLvf//7WHXKaeMe4RxQzpybnJyE1WpFOBwe\nIMsRqOEcK93hxY4IY556fX2NRCKBxcVFBINBmWVL00X4SqWCbDYrr6N9h87hRjWGIbBK4gnjMSMI\n8dQd1aYCIehQYrGYDOhLJBLSCnx5eYlWq4VUKoXXr18jk8mMNfPVGNxxkGOv10OlUpG2cp/PB7fb\nDbvdPpSVzv/maw77PWRQsC2bhahOp4NGo4F6vY5CoYBcLicsTg5iooQOmRdm/jwoR+JyuSRZ4KDb\nq6srdLtd1Gq1gcGbo3oJPcTIhC4Wi9jZ2YHb7cb6+roUw41JK+2ufcXCJls0Dw4OcHBwgDdv3uDg\n4AD1el3YUP+MRlBBG1lgZOOwjV1LsPHvd8k2jYtpVgllIViE1EO72VI5jsHbYxgDEp5jyuQYhxuS\noUgZoVEovJnJ9FkmoM8W9aurK9TrdWkhNvPwQhY1eD/ppF0n7/yTcjc8o2yr1oMKKfulh6Czy5LJ\nBWUQKT2p4ws9YFITJkatdZr3mdPpxPLyMn7/+99jc3MTgUBA1giADN8rFAp4//493rx5g1KpNPZ3\nnJ7JRUCAPpHFp5mZGZHQ5Cy0ZDIp+4W+0+fzCUjhcrlknpBmhlFOo9Fo4ODgAHt7e3j//j3S6TRa\nrZYpO5XuM8bzLJQBt6DXyckJlpaWsLS0BL/fLwVNLeWipVp1lzRjuHq9jnq9jnK5jKOjIxwfH0vh\niUW4cSwE32VkXesOCC1v2Gq1kMvlUCwWZWbVuK2N8SwVCgUcHx+L1jdnQ/j9/gEJXy0bxFjPZrNJ\nV4nX68Xq6io2NjYQi8WEdEcJpt3dXRwfH0sB2Oz559eY7jgx3kks7pE8NwpkwM81rQxBqaV6vY58\nPi+FXrvdLn4hGAyKn/B4PJifnxc5F86w0bMyud9Y+Jybm4PT6YTb7YbP5xPtervdLjWCTqeDg4MD\n7O/vI5VKCWFPSw+Nw9rTNOGEQ26pMa+Z1pS4YXw7jnfdXaZzLLfbjXg8jng8js3NTayurkrXJRVG\nLi8vkU6nsb29jV9++QWZTOYjEGscjQSnfr+PfD4/oOpwenqKUCgk838ZCxN4ZBwIfOg+1LNyuddI\ngmJXhZadJ0BbqVRwdHSEt2/fIpfLiXybzidGKeYbZlxXgqycp0yiYavVQqPRkO70pzJTgRB2u120\nvJ49e4aXL18iHo/DZrNJMbPZbAoIUa/X8f/bO/entO6tjT+KKKgoiiZeoiFJbU/inPucv+H9m9/f\nO50z7Uw7yYloVBQUcCMb2NwFvLw/nPdZftlqaqgIbNZnxmnaooHl/l7W7VnNZrPfb7tn8GJHGQdq\nn6fTacRiMWxvb2N7extv3rxBNBoV596UU3InJcwgiXm4M4jHDGM2m5UqNDod1Hil1h+rrcwh14Ne\neTE2NoaZmRmRI5mfn5fKsHa7jUqlglwuJ1JEXmppfQy83GazWXz+/FmGqq6srMhl474khAk3ND5b\ntVoN6XQaiUQCv/76q1R05nI5z1Tr/BEY6OPhZ1abmFq6DBizgtM9p4PZbi/BywkrYDlYiZXQ7svF\nKK3Vx2JqQ5p6/ZTJ4SWFl2M6arTlt3Q8jTJuqTozMOM4DprNZodO+yBfcB9KKrtlHXkecH0yCcHA\nZr1elyQGnX23bAzvEhzcSukEzmRiJboZEHUPqR4WZ5eOaigUwsuXL/H999/jX//6F968eYNwOCxO\n183NTUcgPhaLYWdnB5VKpc+foPeYsnCpVApTU1PS/TszMyPnI5Nd7Hhglw11h7nfBYNBOTvNQd+t\nVksKJBKJBI6OjrC7u4vd3V2cnJzAtu2hnbvE98y102g04DgOEokE/vGPf8i5ae771ElmUt9dYZhO\np0Uii53Q1EKv1Woi3zcsa/GpGB8fx8zMjATmWCnN34HjODJbhEkIr8HfOc81VgIDwMbGBl69eoWJ\niQlEIhEAtwl6JlqB2+QsA7+rq6tSgb66uirVsJeXl8hms4jFYtjd3RVfgpKBXsVMzprDRLl+2WVe\nqVTuDIIfZvhssUiScwuDwaB01zDgSNmW2dlZRCIRRKNROI4jSQgWPVxdXUk3GCWXeH9jJTblr9h1\nBwCVSkUkqff397G/v49MJoNisdjRvekFu5uYM7/YSR2JRDA7OytBYSZdTf/Ba3Z4CFNCjne7ra0t\nvH//Hj/88ANev36NSCQizxKTNfF4HD/99BN2dnZwdnY2EvMz6Bfd3NyIHBDvHI7j4P3791JcMjk5\neaeYmolCSvJxn2OgnUPTGQ8xC+v4bHL9fvziIdRdAAAW5ElEQVT4EZZldSTN+B6H/R5j+mpMRphz\nhtk13O95ygORhOBFOBwO4+3bt/jw4QO2trbw5s0bhEIh+P1+qSbkAZBOpzs2fa/ChUAnvVKpSKUJ\nFy47FZaWljokXdzJCDOzyJ/NihQuUk6itywLlmUhm81KBRQXsDnsyV2VOOjwoKBeJAPqvOBxyJ6p\naTjMG9G3wt8jHdZ2u435+XnMzc1hfX1dqkIYCDe/j84qu2I4iI9Z56OjI3z8+BGfPn2CbdsDPzvk\nOXF3RJjJBg7yc0uu3Zds9Bq8+HIQGoPmZgLVvIgonfC54OXMrPIKhUJiTz5bDABqQqc73In9arUK\nACLJZAbV3ZU3g8ZDiQgmIMxAm5l0YdATuA0o1+t10W8OBAJSdchuB94/eMfhF4cTm0OxWRhgrvlB\ntJ8b7tfBYFCKbd6+fYt3795heXlZJKtoV9u2kUwmpdrXsqyh+Jx/FPM5oowG5Zio+UtZA2p+Ly4u\nyvfyGTXvvUwM0glrNBoolUrI5/OwLAv7+/s4ODjA4eEh4vE4bNse6nk4ZhcqJc8qlQps25aKQsdx\nkMvlROOcXVu0D4c2upMQlBYqFAril4zqGUwHn4mwpaUlSbTSJo7jiM28LhfJoiLK9/J5oN43deR9\nPh9mZ2dlPzdnHrBDnes8HA6LPjh9tEwmg3g8juPjY5yfn6NarXran2CAl5r7CwsLWFlZEf+VEifx\neByWZXXMcBp2zKIF7mPZbBZjY/8d1j01NYXLy0ssLy9LNyF9BUorsSOOPillm8wkhFnkYHbn8/5C\nCZNMJoNUKoXj42Mkk0mZh+nVzmGeofQflpaWsLm5iZcvX0oSgr8f4DaGYBbXeR2eA6FQCKurq3j3\n7h1++OEHbG9v49WrVx0yiJSbtCwLh4eH2N3dxfHxsUhSjwI8Hzl3lVJqLHxoNpuYmZmRZAPlXpkU\nNAtKCOObfP5YJN1qtVAul2UIc6PRQCKRQDKZlO4Td9LWK8+sWxqYcvPsNB+EwfEDkYRg5nllZQV/\n/vOf8c9//hPR/x/Qx0xYvV5HNpvF6ekpisWip9oNHwsXLh2Dw8NDFItF7O7uIhwOS5WwKdHCP7OV\nzgxkmtXEDByz1ZByTAyYUOrEbNF2ByCG5XdhDgtl9SX1c+n4Mzs6TJ/rKWm1WnAcB1dXV/jpp5+Q\nyWTw7t07bG1tYWNjQ7RZ+SxRJqxWq8FxHHEUWAFm2zby+Tyy2aznNa27xXzW3FIj/P+mxjwvfaY2\nrJdguzQ1vN3yU8DtQHV2TJgBJ+UW6m7Ozs6KY8YBh6bduDdyfY7q/tcNprPcbDZRLpdlGGkwGBTN\neTOR7w6cDNqze997uW+IaK1WAwDRGy2VSpK85rPFQZJmdTq7IFg5Zw5GZyKiXq93VM6ad79Bs9fX\n4B1sfn5eZJhev34tjjwDSnQW4vE4fvzxR/z222/I5XJD8zmfAt51y+WydGNWKhVkMhl8+PABzWZT\nZjyY3VxmkY1bGo37mm3bOD8/RzKZxOHhIY6OjnB2doazszPp+PVaEpbPVKVSwf7+PkqlEv7zn//I\nzBYmotkRzXXGoPL19bUkBenAugcLj9LzCXTeTxYXF/H69Wusra1JBwClcxzHgW3bKJVKnu7cB27P\nQHbVM9h7dXWF9fV1rK2tSafD1taWfB87NXlGMOnImVUAUK/XkcvlxK9Ip9MjoT1vdpmcn5/j8PAQ\nc3NzWFlZkbWXz+cRi8Xwyy+/4ODgwFMBTf5ezS6DbDYrActCoYBEIoHNzU28evXqzrlwc3MjyQYW\ndpl+g8/nE3+KBZnc49gRViwWkc/n5YvqEKVSSfZBL9+V2dm0uLiIjY0NfP/993j16hXm5ubg9/vl\n85tzu0wpHa/ahfAcePHihSiUbG1t4e3btwiHw1LwyoJiFmTu7e2Jj+DluY4PwfOiXC7j5OREpMA/\nf/6McDiMhYUFkVdaXl7Gixcv8OLFC5GlNn8GnzEW0XEdU2KdHZscjH12djYQQfheYcZ/GTfm2Wx2\n07kLYJ+bvichzMr0tbU1bG9v469//SvC4bBU6HDhWpYlSQivPji/BxMA1PVKJBIAOjX7mGgwExDm\nsBczY8hqL15mWHF4n9ayVw5ZBjvY2lkoFOTCQi1TL7dOPwbOXymXy8jlcvj06RPev3+PXC6H7e1t\nbG5uSnCAA0bL5TKKxSIsyxIda2rumW3Xo2rTx2JWU5uSQ6wYu7i4EJ10UyLFi5h7GnC7B5nJGF7e\nGIhS7mKes6Yckxn8ZOCPlziva5M+JWbykEmIUqkkAXq2YZuDmGljs2rM/fMGAfe5b75P8xkxh8TP\nzMxI4QL3Mco2mfNdTHkEABLs5BcHz/F9DPPATX7+cDiMN2/e4C9/+Qs2NjZkzg2TgOwiPDo6wo8/\n/oiDgwMUi8V+v/1nh5VydCQpBdRut+H3+9FqtWRYq/tu665oMzu8KA25s7ODX3/9Fbu7u/L3eLWa\n2vz81WoVR0dHEizimgwEAiJZYCYf7tNJHqT9qV+YSYiFhQVsbGzg5cuXmJ6eBgCpOiyXyxKwbLVa\nfX7XvYeBDsrtUUpnfHwckUhEqqk5vJtr1+278tljx3+1WpU5JCcnJzg7O0M+nx9aybRvgTbNZrOI\nx+NYWlrC+vq6BMAty8LOzg7+/e9/S2LGS/uY2VnEez8VIFKpFBKJBL777jvYto2NjQ00Gg2RG2UX\nOZNcpmw0n03uee12W9Yr515S/smyLJkVyeAl/TOvnhv0Lf1+v0hcra+v47vvvpNOCAaDr6+vpRjM\nlHj18roEbiWCpqam8OLFC/zpT3/C3/72N5GhY6KL3Uz1eh2JRAI///wz4vE4crkc6vV6vz9G37i5\nuZH7vmVZODg4wOTkJJaWlrC8vIyNjQ1Eo1FEo1FcXl5KVwQL1LmeGRulT1Eul+E4jsyv4twgzi1i\nx5jZweS1Z5XxEBYJMwZnJk/7vW/1NQnBzW19fR3RaBQfPnzA2toa5ubmOlrTOdzr9PQUiUQCjuP0\n820PJGYVJnUiza/x8XG0Wq2ONnUe7GbVq6m3bP5s85/DzM3NjVxmd3d30Wq1EIvFxIHl4GTqR3px\nY/oWzMCkZVn4+PGjtAMvLCxIQIqtshwsypZZzm3xqlbmU8JnjRcVJnSSySSur69lqFU8HpfqTbYB\ne7ESjLZgVV0ikZCB8gsLC1LNsLe3J51Lo+Dkfwt8Jng5o86+4zhYWFjA5OSkVEpcXV3BsiwkEgmk\n02kZyuqliuBew/2Sji0rElkRyoDKfUUUw7R+3QFJ85zw+XzioFerVfn/DNpRuzkQCHTMkwIgBRa0\nHde0eV8x7TQsNhsbG5OBjtFoFBsbG3LXNZOAjUYDlmUhk8nIwEsGx0cVBopYiPTbb7/Btm0sLi5i\nYWGhY6ioe2aVmdBnaz7nnp2dnSGVSkl3khfP0IfgWuKfmThlIO6hTsxRvw+b0H+dmpqS6n0OpTaH\n6NL5HwSH/zmhP8pk/MTEhAScGFCanp6WrjiuYT5jlOrjwOV8Pi+yaYwDuIeSeplWqwXbtmW9ptNp\nBAIBqSQ+PDyEbduePi/M+yyLU7mXNZtN8U0XFxdllpDZJcf9zSz6MP0u3pPdMpGmnIs5TH1UusA4\nMJ4Dumlbc4YVu3UcxxGJG6/D2Bq7RFZWVrC2toaVlRWEQqGOwjgWvjIonk6n4TiOZ9dqN5iyXqVS\nSSr46f9//vwZkUgEoVBIincI1yC7qDmjypSs5/qtVqsol8v3Srt6BVMKt1wuw7Zt+e+5XA7FYlHm\ncfTzs/ctCWEOy1xfX8ff//53fPjwAaurq7J4TXmIcrksOnzFYtFzD8xTYCYRTNyDXdzf4/6zl217\nfX0tbZbVahUnJycyxGZsbEwq4ug0eNkWj4WXM1YexWKxDh154NbZ5xedfq8GyHsFL8OsZGJnTrvd\nRigUAgDE43EcHh7i7OxMLntedcTY5ZDP53F8fIxgMCjV0YeHhzg4OMDBwQEsyxrZltbfg+eCmYQo\nFotYXFxEMBiEz+eTCnMmIVKplFRuahLicbgDnqzgZ4KHhQC09UPrddDX8H1V5vzMvNexFZoV/qxO\nZ4GE2U7Nn2WeHeb5we/h3+V+H8Bgt/zz84dCIaytrSEajWJzcxPr6+sioQncDqM+OztDLBZDMpmU\noNIoBS/vg8kqym/s7OyItIZZTW3eS+g/0IfgumNXpnstjpqNze5nd0CODOqaGgTYpcm5JNSv5gBI\nBgDY+TZKSQizW5XSD+zCKZfLuL6+RiAQwNzcnCgeMKDOynJKuFIiLZvN4suXL9jb25MK9VHogiBM\nQnDQ+S+//CIBTgbrKLnh5TubGecwh3Vns9k7ZwHPXgByHjABYRZkmmeF2YFuzoe4T4aa78fr+Hw+\nTE9PY25uTvY5U3IJuC2ioILCqCQhfD6fdHetrKxgfX0dKysrCAQCklQFIHLnxWJRkhCVSkV9VgMz\nGdhut1GpVJDL5UTmizNgmGikzJXZeVOpVOA4jnTMjo2Nibwaz2EmJxiA92LsxK34Ytu2nK/n5+dw\nHEds0k/6loRgZnV+fl7au9bX1+8kINjSSl1N27ZHunXpMQxzheVzYF4qGo2GBAE422BY5R56CaUw\n+r1heR06bzwcs9msyHHMzMxgbGwMp6enSKVSMhjSlHTxGnSmbNvG/v4+6vW6VMCm02mkUilYltXR\nceNFO/xRTGctlUphZ2cH5XIZL1++xMzMjLSwxmIxHB4ewrIsVKtVz1aJ9BKzG8JkfHxcpNUeCkgN\ni53dgUoGMdkezc/fbDbvzHC5ubnpmO3CBIJb8sU94ND8e02nYdCl6FidPzs7i5WVFayuriIcDndo\nVvNOUqlUkEqlsLe3J5q1Xg4oPRb+vs07iCk1an6Zz5uZGDQDSrqn/RcvOt/PxcTEhEiUsCqTQTlT\nmo7FEaP2zPHZYiGS4zgds1ko2zczMyPyGj6fT9ZqqVSSgGatVkOxWEQ6nZYB3+5hol6HtjQLHADc\nORtHBfO+wAS1WwHCXYBp3hnMRIT58+7r8hzVxCw/q7mfOY4j8pBTU1PStVStVsUfKxaLuLi48PT5\nYj5DgUAAoVBIirrcBUntdhu5XA4nJyf48uULMpkMKpWKJFGVTuhDMZAO3N73xsfHRTLNnAfG34U5\nfJ5SbO5CADOh6FXMDuJsNotAICDrsVgsolgsdpyj/eLZkxBcuJOTk5ifn5fM4ebmJpaXl0U72Wzv\nYmaLFREXFxfP/bYVj2FKO5jtcKN0qVUGE1NWLZvNotVqIZVKSccOL4KssvNy0ozrNJ/Po9Vq4fT0\nFFNTU5icnBRt1nq9LtqOXrXDH4Xt6tfX1zg+Pkaj0UAikUAkEkEwGBRnP5vN4vz8HKVSSQYC6374\neEwn1nweb25uOgIs93UuDbOdzeQAExHmP92OvvnfvpZguK/acJiKLFgp5/f7ZZgoE39Ap4xmq9VC\nqVSSJAT3fuV+3BWx9wWe+Lr7AkuK0i2UYqIPOzc31zEo8+rqCrVaDYVCAbVabaSLI8y5ju12G7Va\nDbZtY29v704Hk9kxZ8784v7IilZTRnjUURvcct994aEihYcSFF+7e4wq7EAMBALIZDLIZDJoNBqY\nmprC5eUlCoUC8vk84vG4FMixK8fr9jPn7XFIN5OEDIrX63Wk02nEYjF8+vQJ6XRa5uZ43T5PhVmc\nRLUSU42Ddz92MZl+Bu+IDyUWvQj9ilqthnQ6jVqtJudso9GQJES/VTT60gnBC1woFEIkEsHS0hKW\nlpYQCoUwMTHR0ULNoUDZbFYCb+qYKU8BnQVFGTToZHHPY5sxh4CbGsNevujxcORwUqU7WBVxeXkp\nA7osy8L8/Dz8fr8kIaiBa1YMK98GL8oAOqSEWCFlJsC9iDrvtzxUmclZN9znqVN7fn6OTCaD09NT\n1Qv+HcznTPcppR+wMvPi4gKFQgE+n08kNlKpFE5OTlAoFDpkH0YNdxdTrVZDPp/veM19yUN+79f+\nXVHc6DPztHDtVioVGQy/tLSE+fl5TE5Oot1uI5/PSxLi/Pxc5JhGxfas2C+Xy7AsC4FAADMzM5ie\nnhYJ3P39fezs7ODLly84Pz/Xorku4FnCGIjyMLwXN5tNmf9AWSsmygbhXtK3Tgi/3y8ac8FgEBMT\nE1I5x8tKs9lEKpXC/v4+EokESqWSpwNuiqIoJtwP+U9m+s0B8rofKo+Bzwk7IliVyHZVJrZGUTri\nqTBtxgQEOx8emmegdvYmpgTV1dUVisWizLVh+z5nGnBwXDwelyQh5QwURRksGJhjMN3v96PdbmNu\nbg5+vx+tVgvpdFokIxuNhq7lr/CQbdRmitJ/WCjhOA5isRhKpZIMk7++vhYJHGrNj9IsuaurKxQK\nBbnvNZtNZDIZSUIwQcOZtul0GqVSSfe2Z2BUpeqI2WlN1RfKL/W7A4L0pROCOq7BYFAcMTrr1E/j\nppZKpRCLxXB8fCwDrfptNEVRlOeAlbKPqRJTlN/DXZGoPD3uBAPXr6k//LXvUbwFL/xMQvj9foTD\nYUQiEZGWs20bp6eniMfjMo9F9YIVZXBhEqJaraLdbktgjjO9OFS5Wq1KEkL3+YdR2yjKYMKK6maz\niVKphN3dXZF2MYd682sUMGUei8WizK5xHAeJRAKzs7OYnp5GNptFNpuV86BSqXi6E3rQ0ETEjUhU\nmf7noNjj2ZMQdMrr9ToymYy0hSSTSczOzkpHBDe8k5MTJJNJWJaFUqn04EBHRVEUr8IDw9QuVRRl\nMDHXKB0ODi01UWfEu7i7Ymq1GnK5nNxvE4kEfD4fJiYmZPBqLpdDKpVSvWBFGXDMeXKFQgHX19eY\nmJiQ/85Cumazqfu8oiiewt3ZO4pJVrPS/Pr6GqVSCWNjYyiXy5iamoLf70elUkGlUkGtVhMJTo1h\n9pZBDLYPApyTMUg26VsSolarod1uw7ZtJJNJ/Pzzz/D7/VItyApgDh01p5sPkgEVRVGeC937FGU4\n4Fr9PadD17R3MR1zBiRLpRJOTk6kA9iU7OIw1mazOZJOvaIMC+Zg9EKhgHK53FF1SRk2cyC6oijK\nsONOOozq3mba4erqCuVyGfV6HT6f704sk1+agOgt5nwhPXfvMmj2eGwSIvDUf/HV1ZUMwaxWq7Jo\ngdtqX7O9a4Aepm+xxZPbbYhRu3WH2q171HbdoXbrDrVbd3jebj26u3jebj2iL3Yzh1BXKpWn+rHP\nybfaQp+5W3StdsdA282U5Gi328/913+NgbbbAKN26w61W3cMtd36HJMbGNvRDpS9GXAGxm69gNX+\n/PMT4mm79ZCv2uKuPsD9RP/4+3gYDl29uLjAxcXFnc6HAZsDEe3Ra71OtEev9TrRHr12FIj26LVe\nJ9qj13qdaI9e63WiPXqt14n26LVeJ9qj13qdaI9f72WiPXqt14n26LVeJ9qj13qdaI9e63WiPXqt\n14n26LWjQLRHr/U60R69dmDo0WySaI9e63WiX/ufY48J7o+NjUUA/A+ABICLp3hXQ0gA/zXm/97c\n3OQf8w1qNwBqt25Ru3WP2q471G7doXbrDrVbd6jdukPt1h3fbDdAbff/6DPXHWq37lC7dYfarTvU\nbt2hdusetV13qN26Q+3WHY+y26OSEIqiKIqiKIqiKIqiKIqiKIqiKN/KY+WYFEVRFEVRFEVRFEVR\nFEVRFEVRvglNQiiKoiiKoiiKoiiKoiiKoiiK0hM0CaEoiqIoiqIoiqIoiqIoiqIoSk/QJISiKIqi\nKIqiKIqiKIqiKIqiKD1BkxCKoiiKoiiKoiiKoiiKoiiKovQETUIoiqIoiqIoiqIoiqIoiqIoitIT\nNAmhKIqiKIqiKIqiKIqiKIqiKEpP+D/AcFS215VGLAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7ff20426d350>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 3min 19s, sys: 1min 34s, total: 4min 54s\n",
      "Wall time: 4min 54s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "\n",
    "encoded = None\n",
    "decoded = None\n",
    "\n",
    "input_img = Input(shape=(784,))\n",
    "\n",
    "encoded = Dense(128,activation='relu')(input_img)\n",
    "encoded = Dense(64, activation='relu')(encoded)\n",
    "encoded = Dense(32,activation='relu')(encoded)\n",
    "\n",
    "decoded = Dense(64,activation='relu')(encoded)\n",
    "decoded = Dense(128,activation='relu')(decoded)\n",
    "decoded = Dense(784,activation='sigmoid')(decoded)\n",
    "\n",
    "autoencoder = Model(input=input_img, output=decoded)\n",
    "\n",
    "\n",
    "encoder = Model(input=input_img, output = encoded)\n",
    "encoded_input = Input(shape=(32,))\n",
    "\n",
    "decoder_layer1 = autoencoder.layers[-3]\n",
    "decoder_layer2 = autoencoder.layers[-2]\n",
    "decoder_layer3 = autoencoder.layers[-1]\n",
    "\n",
    "decoder = Model(input=encoded_input, output= decoder_layer3(decoder_layer2(decoder_layer1(encoded_input))))\n",
    "\n",
    "autoencoder.compile(optimizer='adadelta',loss='binary_crossentropy')\n",
    "\n",
    "autoencoder.fit(x_train, x_train, nb_epoch=100, verbose=2,\n",
    "               batch_size=256,shuffle=True,\n",
    "               validation_data=(x_test, x_test))\n",
    "\n",
    "encoded_imgs = encoder.predict(x_test)\n",
    "decoded_imgs = decoder.predict(encoded_imgs)\n",
    "\n",
    "plot_imgs(decoded_imgs)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "encoded = None\n",
    "decoded = None\n",
    "\n",
    "input_img = Input(shape=(1,28,28))\n",
    "x = Convolution2D(16,3,3, activation='relu', border_mode='same')(input_img)\n",
    "x = MaxPooling2D((2,2), border_mode='same')(x)\n",
    "x = Convolution2D(8,3,3, activation='relu', border_mode='same')(x)\n",
    "x = MaxPooling2D((2,2), border_mode='same')(x)\n",
    "x = Convolution2D(8,3,3, activation='relu', border_mode='same')(x)\n",
    "\n",
    "encoded = MaxPooling2D((2,2),border_mode='same')(x)\n",
    "\n",
    "# at this point the representation has 8 x 4 x 4 = 128 dimensions\n",
    "x = Convolution2D(8,3,3,activation='relu',border_mode='same')(encoded)\n",
    "x = UpSampling2D((2,2))(x)\n",
    "x = Convolution2D(8,3,3, activation='relu', border_mode='same')(x)\n",
    "x = UpSampling2D((2,2))(x)\n",
    "x = Convolution2D(16,3,3, activation='relu')(x)\n",
    "x = UpSampling2D((2,2))(x)\n",
    "\n",
    "decoded = Convolution2D(1,3,3, activation='sigmoid', border_mode='same')(x)\n",
    "\n",
    "autoencoder = Model(input_img,decoded)\n",
    "autoencoder.compile(optimizer='adadelta',loss='binary_crossentropy')\n",
    "\n",
    "(x_train, _), (x_test, _) = mnist.load_data()\n",
    "\n",
    "x_train = x_train.astype('float32') / 255.\n",
    "x_test = x_test.astype('float32') / 255.\n",
    "x_train = np.reshape(x_train, (len(x_train), 1, 28, 28))\n",
    "x_test = np.reshape(x_test, (len(x_test), 1, 28, 28))\n",
    "\n",
    "# the tutorial suggests we use tensorflow but I'll use theano\n",
    "\n",
    "autoencoder.fit(x_train,x_train,\n",
    "            verbose=2,\n",
    "            nb_epoch=50,\n",
    "            batch_size=128,\n",
    "            shuffle=True,\n",
    "            validation_data=(x_test,x_test))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "decoded_imgs = autoencoder.predict(x_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "plot_imgs(decoded_imgs)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## denoising"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "(x_train, _), (x_test, _) = mnist.load_data()\n",
    "\n",
    "x_train = x_train.astype('float32') / 255.\n",
    "x_test = x_test.astype('float32') / 255.\n",
    "x_train = np.reshape(x_train, (len(x_train), 1, 28, 28))\n",
    "x_test = np.reshape(x_test, (len(x_test), 1, 28, 28))\n",
    "\n",
    "noise_factor= 0.5\n",
    "\n",
    "x_train_noisy = x_train + noise_factor * np.random.normal(loc=0.0,scale=1.0,size=x_train.shape)\n",
    "x_test_noisy = x_test + noise_factor * np.random.normal(loc=0.0,scale=1.0,size=x_test.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "x_train_noisy = np.clip(x_train_noisy,0.,1.)\n",
    "x_test_noisy = np.clip(x_test_noisy, 0.,1.)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "plot_imgs(x_test_noisy)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "input_img = Input(shape=(1,28,28))\n",
    "\n",
    "x = Convolution2D(32,3,3,activation='relu',border_mode='same')(input_img)\n",
    "x = MaxPooling2D((2,2), border_mode='same')(x)\n",
    "x = Convolution2D(32,3,3,activation='relu',border_mode='same')(x)\n",
    "encoded = MaxPooling2D((2,2),border_mode='same')(x)\n",
    "\n",
    "# at this point the representation is (32,7,7)\n",
    "x = Convolution2D(32,3,3,activation='relu',border_mode='same')(encoded)\n",
    "x = UpSampling2D((2,2))(x)\n",
    "x = Convolution2D(32,3,3,activation='relu',border_mode='same')(x)\n",
    "x = UpSampling2D((2,2))(x)\n",
    "\n",
    "decoded = Convolution2D(1,3,3,activation='sigmoid',border_mode='same')(x)\n",
    "\n",
    "autoencoder = Model(input_img,decoded)\n",
    "\n",
    "# decoder_layer1 = autoencoder.layers[-4]\n",
    "# decoder_layer2 = autoencoder.layers[-3]\n",
    "# decoder_layer3 = autoencoder.layers[-2]\n",
    "# decoder_layer4 = autoencoder.layers[-1]\n",
    "\n",
    "# encoded_input = Input(shape=(32,7,7))\n",
    "\n",
    "# decoder = Model(\n",
    "#     input=encoded_input, \n",
    "#     output= decoder_layer4(decoder_layer3(decoder_layer2(decoder_layer1(encoded_input)))))\n",
    "\n",
    "\n",
    "autoencoder.compile(optimizer='adadelta',loss='binary_crossentropy')\n",
    "\n",
    "# the source is noisy, but the target isn't?\n",
    "autoencoder.fit(x_train_noisy,x_train,\n",
    "               nb_epoch=100,\n",
    "               batch_size=128,\n",
    "               shuffle=True,\n",
    "               validation_data=(x_test_noisy,x_test))\n",
    "\n",
    "\n",
    "decoded_imgs = autoencoder.predict(x_test)\n",
    "plot_imgs(decoded_imgs)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 2",
   "language": "python",
   "name": "python2"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}