畳み込みニューラルネットワーク.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 畳み込みニューラルネットワーク\n",
    "### im2colによる入力データの展開"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def im2col(input_data, filter_h, filter_w, stride=1, pad=0):\n",
    "    \"\"\"\n",
    "    Parameters\n",
    "    ----------\n",
    "    input_data : (データ数, チャンネル, 高さ, 幅)の4次元配列からなる入力データ\n",
    "    filter_h : フィルターの高さ\n",
    "    filter_w : フィルターの幅\n",
    "    stride : ストライド\n",
    "    pad : パディング\n",
    "    Returns\n",
    "    -------\n",
    "    col : 2次元配列\n",
    "    \"\"\"\n",
    "    N, C, H, W = input_data.shape\n",
    "    out_h = (H + 2*pad - filter_h)//stride + 1\n",
    "    out_w = (W + 2*pad - filter_w)//stride + 1\n",
    "\n",
    "    img = np.pad(input_data, [(0,0), (0,0), (pad, pad), (pad, pad)], 'constant')\n",
    "    col = np.zeros((N, C, filter_h, filter_w, out_h, out_w))\n",
    "\n",
    "    for y in range(filter_h):\n",
    "        y_max = y + stride*out_h\n",
    "        for x in range(filter_w):\n",
    "            x_max = x + stride*out_w\n",
    "            col[:, :, y, x, :, :] = img[:, :, y:y_max:stride, x:x_max:stride]\n",
    "\n",
    "    col = col.transpose(0, 4, 5, 1, 2, 3).reshape(N*out_h*out_w, -1)\n",
    "    return col"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(9, 75)\n",
      "(90, 75)\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "x1 = np.random.rand(1, 3, 7, 7)\n",
    "col1 = im2col(x1, 5, 5, stride=1, pad=0)\n",
    "print(col1.shape)\n",
    "x2 = np.random.rand(10, 3, 7, 7)\n",
    "col2 = im2col(x2, 5, 5, stride=1, pad=0)\n",
    "print(col2.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Convolutionレイヤの実装"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "class Convolution:\n",
    "    def __init__(self, W, b, stride=1, pad=0):\n",
    "        self.W = W\n",
    "        self.b = b\n",
    "        self.stride = stride\n",
    "        self.pad = pad\n",
    "        \n",
    "        # 中間データ（backward時に使用）\n",
    "        self.x = None   \n",
    "        self.col = None\n",
    "        self.col_W = None\n",
    "        \n",
    "        # 重み・バイアスパラメータの勾配\n",
    "        self.dW = None\n",
    "        self.db = None\n",
    "\n",
    "    def forward(self, x):\n",
    "        FN, C, FH, FW = self.W.shape\n",
    "        N, C, H, W = x.shape\n",
    "        out_h = 1 + int((H + 2*self.pad - FH) / self.stride)\n",
    "        out_w = 1 + int((W + 2*self.pad - FW) / self.stride)\n",
    "\n",
    "        col = im2col(x, FH, FW, self.stride, self.pad)\n",
    "        col_W = self.W.reshape(FN, -1).T\n",
    "\n",
    "        out = np.dot(col, col_W) + self.b\n",
    "        out = out.reshape(N, out_h, out_w, -1).transpose(0, 3, 1, 2)\n",
    "\n",
    "        self.x = x\n",
    "        self.col = col\n",
    "        self.col_W = col_W\n",
    "\n",
    "        return out\n",
    "\n",
    "    def backward(self, dout):\n",
    "        FN, C, FH, FW = self.W.shape\n",
    "        dout = dout.transpose(0,2,3,1).reshape(-1, FN)\n",
    "\n",
    "        self.db = np.sum(dout, axis=0)\n",
    "        self.dW = np.dot(self.col.T, dout)\n",
    "        self.dW = self.dW.transpose(1, 0).reshape(FN, C, FH, FW)\n",
    "\n",
    "        dcol = np.dot(dout, self.col_W.T)\n",
    "        dx = col2im(dcol, self.x.shape, FH, FW, self.stride, self.pad)\n",
    "\n",
    "        return dx"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### col2imによる Convolutionレイヤの逆伝播"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def col2im(col, input_shape, filter_h, filter_w, stride=1, pad=0):\n",
    "    \"\"\"\n",
    "\n",
    "    Parameters\n",
    "    ----------\n",
    "    col :\n",
    "    input_shape : 入力データの形状（例：(10, 1, 28, 28)）\n",
    "    filter_h :\n",
    "    filter_w\n",
    "    stride\n",
    "    pad\n",
    "\n",
    "    Returns\n",
    "    -------\n",
    "\n",
    "    \"\"\"\n",
    "    N, C, H, W = input_shape\n",
    "    out_h = (H + 2*pad - filter_h)//stride + 1\n",
    "    out_w = (W + 2*pad - filter_w)//stride + 1\n",
    "    col = col.reshape(N, out_h, out_w, C, filter_h, filter_w).transpose(0, 3, 4, 5, 1, 2)\n",
    "\n",
    "    img = np.zeros((N, C, H + 2*pad + stride - 1, W + 2*pad + stride - 1))\n",
    "    for y in range(filter_h):\n",
    "        y_max = y + stride*out_h\n",
    "        for x in range(filter_w):\n",
    "            x_max = x + stride*out_w\n",
    "            img[:, :, y:y_max:stride, x:x_max:stride] += col[:, :, y, x, :, :]\n",
    "\n",
    "    return img[:, :, pad:H + pad, pad:W + pad]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Poolingレイヤの実装"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "class Pooling:\n",
    "    def __init__(self, pool_h, pool_w, stride=1, pad=0):\n",
    "        self.pool_h = pool_h\n",
    "        self.pool_w = pool_w\n",
    "        self.stride = stride\n",
    "        self.pad = pad\n",
    "        \n",
    "        self.x = None\n",
    "        self.arg_max = None\n",
    "\n",
    "    def forward(self, x):\n",
    "        N, C, H, W = x.shape\n",
    "        out_h = int(1 + (H - self.pool_h) / self.stride)\n",
    "        out_w = int(1 + (W - self.pool_w) / self.stride)\n",
    "\n",
    "        col = im2col(x, self.pool_h, self.pool_w, self.stride, self.pad)\n",
    "        col = col.reshape(-1, self.pool_h*self.pool_w)\n",
    "\n",
    "        arg_max = np.argmax(col, axis=1)\n",
    "        out = np.max(col, axis=1)\n",
    "        out = out.reshape(N, out_h, out_w, C).transpose(0, 3, 1, 2)\n",
    "\n",
    "        self.x = x\n",
    "        self.arg_max = arg_max\n",
    "\n",
    "        return out\n",
    "\n",
    "    def backward(self, dout):\n",
    "        dout = dout.transpose(0, 2, 3, 1)\n",
    "        \n",
    "        pool_size = self.pool_h * self.pool_w\n",
    "        dmax = np.zeros((dout.size, pool_size))\n",
    "        dmax[np.arange(self.arg_max.size), self.arg_max.flatten()] = dout.flatten()\n",
    "        dmax = dmax.reshape(dout.shape + (pool_size,)) \n",
    "        \n",
    "        dcol = dmax.reshape(dmax.shape[0] * dmax.shape[1] * dmax.shape[2], -1)\n",
    "        dx = col2im(dcol, self.x.shape, self.pool_h, self.pool_w, self.stride, self.pad)\n",
    "        \n",
    "        return dx"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### CNN (convolutional neural network) の実装"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# coding: utf-8\n",
    "import sys, os\n",
    "sys.path.append(os.pardir)  # 親ディレクトリのファイルをインポートするための設定\n",
    "import pickle\n",
    "import numpy as np\n",
    "from collections import OrderedDict\n",
    "from mymodule.layers import *\n",
    "from mymodule.gradient import numerical_gradient\n",
    "\n",
    "\n",
    "class SimpleConvNet:\n",
    "    \"\"\"単純なConvNet\n",
    "\n",
    "    conv - relu - pool - affine - relu - affine - softmax\n",
    "    \n",
    "    Parameters\n",
    "    ----------\n",
    "    input_size : 入力サイズ（MNISTの場合は784）\n",
    "    hidden_size_list : 隠れ層のニューロンの数のリスト（e.g. [100, 100, 100]）\n",
    "    output_size : 出力サイズ（MNISTの場合は10）\n",
    "    activation : 'relu' or 'sigmoid'\n",
    "    weight_init_std : 重みの標準偏差を指定（e.g. 0.01）\n",
    "        'relu'または'he'を指定した場合は「Heの初期値」を設定\n",
    "        'sigmoid'または'xavier'を指定した場合は「Xavierの初期値」を設定\n",
    "    \"\"\"\n",
    "    def __init__(self, input_dim=(1, 28, 28), \n",
    "                 conv_param={'filter_num':30, 'filter_size':5, 'pad':0, 'stride':1},\n",
    "                 hidden_size=100, output_size=10, weight_init_std=0.01):\n",
    "        filter_num = conv_param['filter_num']\n",
    "        filter_size = conv_param['filter_size']\n",
    "        filter_pad = conv_param['pad']\n",
    "        filter_stride = conv_param['stride']\n",
    "        input_size = input_dim[1]\n",
    "        conv_output_size = (input_size - filter_size + 2*filter_pad) / filter_stride + 1\n",
    "        pool_output_size = int(filter_num * (conv_output_size/2) * (conv_output_size/2))\n",
    "\n",
    "        # 重みの初期化\n",
    "        self.params = {}\n",
    "        self.params['W1'] = weight_init_std * \\\n",
    "                            np.random.randn(filter_num, input_dim[0], filter_size, filter_size)\n",
    "        self.params['b1'] = np.zeros(filter_num)\n",
    "        self.params['W2'] = weight_init_std * \\\n",
    "                            np.random.randn(pool_output_size, hidden_size)\n",
    "        self.params['b2'] = np.zeros(hidden_size)\n",
    "        self.params['W3'] = weight_init_std * \\\n",
    "                            np.random.randn(hidden_size, output_size)\n",
    "        self.params['b3'] = np.zeros(output_size)\n",
    "\n",
    "        # レイヤの生成\n",
    "        self.layers = OrderedDict()\n",
    "        self.layers['Conv1'] = Convolution(self.params['W1'], self.params['b1'],\n",
    "                                           conv_param['stride'], conv_param['pad'])\n",
    "        self.layers['Relu1'] = Relu()\n",
    "        self.layers['Pool1'] = Pooling(pool_h=2, pool_w=2, stride=2)\n",
    "        self.layers['Affine1'] = Affine(self.params['W2'], self.params['b2'])\n",
    "        self.layers['Relu2'] = Relu()\n",
    "        self.layers['Affine2'] = Affine(self.params['W3'], self.params['b3'])\n",
    "\n",
    "        self.last_layer = SoftmaxWithLoss()\n",
    "\n",
    "    def predict(self, x):\n",
    "        for layer in self.layers.values():\n",
    "            x = layer.forward(x)\n",
    "\n",
    "        return x\n",
    "\n",
    "    def loss(self, x, t):\n",
    "        \"\"\"損失関数を求める\n",
    "        引数のxは入力データ、tは教師ラベル\n",
    "        \"\"\"\n",
    "        y = self.predict(x)\n",
    "        return self.last_layer.forward(y, t)\n",
    "\n",
    "    def accuracy(self, x, t, batch_size=100):\n",
    "        if t.ndim != 1 : t = np.argmax(t, axis=1)\n",
    "        \n",
    "        acc = 0.0\n",
    "        \n",
    "        for i in range(int(x.shape[0] / batch_size)):\n",
    "            tx = x[i*batch_size:(i+1)*batch_size]\n",
    "            tt = t[i*batch_size:(i+1)*batch_size]\n",
    "            y = self.predict(tx)\n",
    "            y = np.argmax(y, axis=1)\n",
    "            acc += np.sum(y == tt) \n",
    "        \n",
    "        return acc / x.shape[0]\n",
    "\n",
    "    def numerical_gradient(self, x, t):\n",
    "        \"\"\"勾配を求める（数値微分）\n",
    "\n",
    "        Parameters\n",
    "        ----------\n",
    "        x : 入力データ\n",
    "        t : 教師ラベル\n",
    "\n",
    "        Returns\n",
    "        -------\n",
    "        各層の勾配を持ったディクショナリ変数\n",
    "            grads['W1']、grads['W2']、...は各層の重み\n",
    "            grads['b1']、grads['b2']、...は各層のバイアス\n",
    "        \"\"\"\n",
    "        loss_w = lambda w: self.loss(x, t)\n",
    "\n",
    "        grads = {}\n",
    "        for idx in (1, 2, 3):\n",
    "            grads['W' + str(idx)] = numerical_gradient(loss_w, self.params['W' + str(idx)])\n",
    "            grads['b' + str(idx)] = numerical_gradient(loss_w, self.params['b' + str(idx)])\n",
    "\n",
    "        return grads\n",
    "\n",
    "    def gradient(self, x, t):\n",
    "        \"\"\"勾配を求める（誤差逆伝搬法）\n",
    "\n",
    "        Parameters\n",
    "        ----------\n",
    "        x : 入力データ\n",
    "        t : 教師ラベル\n",
    "\n",
    "        Returns\n",
    "        -------\n",
    "        各層の勾配を持ったディクショナリ変数\n",
    "            grads['W1']、grads['W2']、...は各層の重み\n",
    "            grads['b1']、grads['b2']、...は各層のバイアス\n",
    "        \"\"\"\n",
    "        # forward\n",
    "        self.loss(x, t)\n",
    "\n",
    "        # backward\n",
    "        dout = 1\n",
    "        dout = self.last_layer.backward(dout)\n",
    "\n",
    "        layers = list(self.layers.values())\n",
    "        layers.reverse()\n",
    "        for layer in layers:\n",
    "            dout = layer.backward(dout)\n",
    "\n",
    "        # 設定\n",
    "        grads = {}\n",
    "        grads['W1'], grads['b1'] = self.layers['Conv1'].dW, self.layers['Conv1'].db\n",
    "        grads['W2'], grads['b2'] = self.layers['Affine1'].dW, self.layers['Affine1'].db\n",
    "        grads['W3'], grads['b3'] = self.layers['Affine2'].dW, self.layers['Affine2'].db\n",
    "\n",
    "        return grads\n",
    "        \n",
    "    def save_params(self, file_name=\"params.pkl\"):\n",
    "        params = {}\n",
    "        for key, val in self.params.items():\n",
    "            params[key] = val\n",
    "        with open(file_name, 'wb') as f:\n",
    "            pickle.dump(params, f)\n",
    "\n",
    "    def load_params(self, file_name=\"params.pkl\"):\n",
    "        with open(file_name, 'rb') as f:\n",
    "            params = pickle.load(f)\n",
    "        for key, val in params.items():\n",
    "            self.params[key] = val\n",
    "\n",
    "        for i, key in enumerate(['Conv1', 'Affine1', 'Affine2']):\n",
    "            self.layers[key].W = self.params['W' + str(i+1)]\n",
    "            self.layers[key].b = self.params['b' + str(i+1)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "W1 5.16687704339e-07\n",
      "b1 9.17413822861e-13\n",
      "W2 4.03362018129e-13\n",
      "b2 6.44677367045e-13\n",
      "W3 5.97610934263e-13\n",
      "b3 1.20126120162e-10\n"
     ]
    }
   ],
   "source": [
    "# coding: utf-8\n",
    "import numpy as np\n",
    "from mymodule.simple_convnet import SimpleConvNet\n",
    "\n",
    "network = SimpleConvNet(input_dim=(1,10, 10), \n",
    "                        conv_param = {'filter_num':10, 'filter_size':3, 'pad':0, 'stride':1},\n",
    "                        hidden_size=10, output_size=10, weight_init_std=0.01)\n",
    "\n",
    "X = np.random.rand(100).reshape((1, 1, 10, 10))\n",
    "T = np.array([1]).reshape((1,1))\n",
    "\n",
    "grad_num = network.numerical_gradient(X, T)\n",
    "grad = network.gradient(X, T)\n",
    "\n",
    "for key, val in grad_num.items():\n",
    "    print(key, np.abs(grad_num[key] - grad[key]).mean())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": false,
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Thu Feb  2 14:42:28 JST 2017\n",
      "train loss:2.29855255808\n",
      "=== epoch:1, train acc:0.299, test acc:0.329 ===\n",
      "train loss:0.0792833412231\n",
      "=== epoch:2, train acc:0.952, test acc:0.962 ===\n",
      "train loss:0.0602736800449\n",
      "=== epoch:3, train acc:0.973, test acc:0.975 ===\n",
      "train loss:0.0805845238219\n",
      "=== epoch:4, train acc:0.984, test acc:0.983 ===\n",
      "train loss:0.018761358324\n",
      "=== epoch:5, train acc:0.982, test acc:0.984 ===\n",
      "train loss:0.00388361826815\n",
      "=== epoch:6, train acc:0.987, test acc:0.992 ===\n",
      "train loss:0.0449459457794\n",
      "=== epoch:7, train acc:0.984, test acc:0.985 ===\n",
      "train loss:0.0154577127933\n",
      "=== epoch:8, train acc:0.992, test acc:0.983 ===\n",
      "train loss:0.0118215261914\n",
      "=== epoch:9, train acc:0.99, test acc:0.982 ===\n",
      "train loss:0.0104199717378\n",
      "=== epoch:10, train acc:0.994, test acc:0.985 ===\n",
      "train loss:0.0199123876411\n",
      "=== epoch:11, train acc:0.99, test acc:0.984 ===\n",
      "train loss:0.00520106329608\n",
      "=== epoch:12, train acc:0.995, test acc:0.984 ===\n",
      "train loss:0.00218466190755\n",
      "=== epoch:13, train acc:0.995, test acc:0.987 ===\n",
      "train loss:0.00305702721674\n",
      "=== epoch:14, train acc:0.993, test acc:0.986 ===\n",
      "train loss:0.00414820351582\n",
      "=== epoch:15, train acc:0.998, test acc:0.986 ===\n",
      "train loss:0.00025411010528\n",
      "=== epoch:16, train acc:0.998, test acc:0.99 ===\n",
      "train loss:0.0253313695622\n",
      "=== epoch:17, train acc:0.999, test acc:0.989 ===\n",
      "train loss:0.000210100938599\n",
      "=== epoch:18, train acc:1.0, test acc:0.983 ===\n",
      "train loss:7.13242715925e-05\n",
      "=== epoch:19, train acc:0.999, test acc:0.981 ===\n",
      "train loss:0.000152278844558\n",
      "=== epoch:20, train acc:0.999, test acc:0.988 ===\n",
      "=============== Final Test Accuracy ===============\n",
      "test acc:0.9876\n",
      "Saved Network Parameters!\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYUAAAEKCAYAAAD9xUlFAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nGToKEQAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1266b9fd0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Thu Feb  2 15:46:56 JST 2017\r\n"
     ]
    }
   ],
   "source": [
    "!date\n",
    "# coding: utf-8\n",
    "import sys, os\n",
    "sys.path.append(os.pardir)  # 親ディレクトリのファイルをインポートするための設定\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "from mymodule.mnist import load_mnist\n",
    "from mymodule.simple_convnet import SimpleConvNet\n",
    "from mymodule.trainer import Trainer\n",
    "\n",
    "# データの読み込み\n",
    "(x_train, t_train), (x_test, t_test) = load_mnist(flatten=False)\n",
    "\n",
    "# 処理に時間のかかる場合はデータを削減 \n",
    "#x_train, t_train = x_train[:5000], t_train[:5000]\n",
    "#x_test, t_test = x_test[:1000], t_test[:1000]\n",
    "\n",
    "max_epochs = 20\n",
    "\n",
    "network = SimpleConvNet(input_dim=(1,28,28), \n",
    "                        conv_param = {'filter_num': 30, 'filter_size': 5, 'pad': 0, 'stride': 1},\n",
    "                        hidden_size=100, output_size=10, weight_init_std=0.01)\n",
    "                        \n",
    "trainer = Trainer(network, x_train, t_train, x_test, t_test,\n",
    "                  epochs=max_epochs, mini_batch_size=100,\n",
    "                  optimizer='Adam', optimizer_param={'lr': 0.001},\n",
    "                  evaluate_sample_num_per_epoch=1000)\n",
    "trainer.train()\n",
    "\n",
    "# パラメータの保存\n",
    "network.save_params(\"params.pkl\")\n",
    "print(\"Saved Network Parameters!\")\n",
    "\n",
    "# グラフの描画\n",
    "markers = {'train': 'o', 'test': 's'}\n",
    "x = np.arange(max_epochs)\n",
    "plt.plot(x, trainer.train_acc_list, marker='o', label='train', markevery=2)\n",
    "plt.plot(x, trainer.test_acc_list, marker='s', label='test', markevery=2)\n",
    "plt.xlabel(\"epochs\")\n",
    "plt.ylabel(\"accuracy\")\n",
    "plt.ylim(0, 1.0)\n",
    "plt.legend(loc='lower right')\n",
    "plt.show()\n",
    "!date"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### CNN の可視化"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAcUAAAEjCAYAAABD3BobAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x113e3d780>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAcUAAAEjCAYAAABD3BobAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nwKEUAQBwKEUAABxKEQAAh1IEAMD5P1+p1coizE39AAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x116379d30>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# coding: utf-8\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "from mymodule.simple_convnet import SimpleConvNet\n",
    "\n",
    "def filter_show(filters, nx=8, margin=3, scale=10):\n",
    "    \"\"\"\n",
    "    c.f. https://gist.github.com/aidiary/07d530d5e08011832b12#file-draw_weight-py\n",
    "    \"\"\"\n",
    "    FN, C, FH, FW = filters.shape\n",
    "    ny = int(np.ceil(FN / nx))\n",
    "\n",
    "    fig = plt.figure()\n",
    "    fig.subplots_adjust(left=0, right=1, bottom=0, top=1, hspace=0.05, wspace=0.05)\n",
    "\n",
    "    for i in range(FN):\n",
    "        ax = fig.add_subplot(ny, nx, i+1, xticks=[], yticks=[])\n",
    "        ax.imshow(filters[i, 0], cmap=plt.cm.gray_r, interpolation='nearest')\n",
    "    plt.show()\n",
    "\n",
    "\n",
    "network = SimpleConvNet()\n",
    "# ランダム初期化後の重み\n",
    "filter_show(network.params['W1'])\n",
    "\n",
    "# 学習後の重み\n",
    "network.load_params(\"params.pkl\")\n",
    "filter_show(network.params['W1'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAaIAAAE1CAYAAAClaOSUAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\noQAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x10f15fcf8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAaIAAAE1CAYAAAClaOSUAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\noQAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x10f15fb00>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAaIAAAE1CAYAAAClaOSUAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\noQAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x113840c88>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAUoAAADuCAYAAABf005JAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAIABJREFUeJzsvXts5Gd1Pv7M/X4fz9jj+3q9vux6k93N7oaE3DZACJDS\nNLSlAkqpQEVFom0kJCRaVYI/2qp/tIgCQkgEVJUSQqkoaUlC7tmEzW52N3v32l7fPfaMPff7/feH\nv8/xOw40a35fyFjfOZK1Xtsz8/m8n/c9l+ec8xxNo9FAW9rSlra05VeL9p2+gLa0pS1taXVpK8q2\ntKUtbXkbaSvKtrSlLW15G2kryra0pS1teRtpK8q2tKUtbXkbaSvKtrSlLW15G2kryra0pS1teRtp\nK8q2tKUtbXkbaSvKtrSlLW15G9Hv5I9tNlvD6/VCo9GAHT3VahUGgwEA0Gg0oNVq5XtVGo2GvE6j\n0TT9zfafq68BIO9Zr9ebPpuvNRqNKJVKSKfTcDqdTe/Da4rFYshms80f0EJitVobbrcb9XodOp0O\nwOa9FQoFaDSapjVSv26ms+p/+7tGo4F6vQ6DwfBL179Wq8FkMsFqtSIWi8mz4PsaDAasr6+jVCq1\n7Nra7faG1+uV/zcaDTQaDVSrVWi12qZ9qP57s/Kr1rder0Or1crz5Ger3xsMBhgMBmQymabP12g0\nSKfTMBqNiMViG41Go2NHF/VbFJvN1nC73QC21q5er7/l79R1utk1/t/0AgDodDp5ntuF657P52Ey\nmZpe22g0YDQasb6+jnw+/7YXsyNF6fP58Fd/9VdNP8tkMujo6ECpVIJer2+6aG7CWq0mf28ymeTC\ntVot6vV608Vv36z8vclkQqVSQblclt9Vq1VZjP7+fszNzSGRSGBgYADZbBZarVa+/vEf/3Ent/pb\nF7fbjU9/+tPQarXQ6/WoVCpwuVw4c+YMrFar/MxoNMJoNEKj0TQpLWDr4ZfLZdRqNeh0OtRqNej1\nemg0Gjm4tVpNnk25XEaxWERHRwcsFgtqtdpbDNny8jKOHDmCnp4ePP3009DpdKjX6xgaGsIrr7yC\nF1544be+XjsRr9eLL37xi9BoNKjVaqjX60gmk8jlcnIvWq0WBoNB/lUPOvcpsGU8uH46ne4tv+e/\ndCKsVutbDBHPSTqdxvj4OJxOJ55//nn52xdeeAE9PT2o1Wr43ve+t/BbXK4di9vtxmc/+1kAkLXL\nZDLQ6XSyvtsVFYXOj9FoRKVSAbCp4CqVijg5NByUSqUCg8GAYrEIs9kse55/W6vVZH337duHSCSC\nF154Abfddpvok1AohJmZGXzve9+7qXvcUejNmzSZTHJBer0eBoNBFsRsNjddcL1eh16vl8NdqVTk\n0FerVRSLReh0uiYPiV98vepB8bO5uLTY8XgcZrMZ6XQapVIJFosFFoul6f1aXajUVU+Hm0U1InzY\nXGP+jEqQh5evUw/9ds9er9+0ldVqVYyQarxqtRq8Xi8ikQiq1SpqtRpKpRIGBwfx4osvvmUTt6po\ntVrk83kAW54N71NdX51Oh2q1CmDLW1GF+9hkMqFarcq6AWhShlzzWq0mf8fva7UaarUaCoUCDAYD\nVldXRenWajU8//zz6OrqktfvBtHpdDCbzahWq9BoNLBYLBJpck/zSz3r3KvlclnWul6vw2g0ioEH\ntgwUjRTXl46S0WgUPUL9YzAYsLS0BLPZjLGxMWi1WhiNRgQCAczOzsrf34zsGKNUPb5Go4FSqYRi\nsSg/Vx8sb4I3yL/XaDQwm81yqJeXl3Hjxg1ks1lZbNVqUyqVivycHimVST6fx969e7Fnzx7EYjE0\nGg0UCoUmJdzK0mg0YLPZkM1m5Z65iShcW26Gcrnc5BFxbQDAbDbD6/Wiu7sbfr8fDoejaTNSIfKz\neUhVxcq11ev1WFxcxOrqKt7//vejv78fr7zyimz6VheNRoNcLodCoSD/Vw0C75cHUK/Xy3oAm/vO\nZDLB6XTCYrGgo6MDgUAAvb29uO+++3D48GHk83kUi0U0Gg3xhqrVqrwXFXG9Xke5XEalUhGYJRaL\nIZ1O48SJEzh9+jS6u7uh1WpRKpVE2bS60DDr9fomg6DCDjRE/D3vn+u0traGRCKBdDqNSqWC5eVl\nnDx5EmazuUlh8jkZDAbUajUxVCqcwuug99nZ2YnTp0/DZrMhHA7/Umjgf72/X2dB6vV6k0anR8lQ\njhqdB7hcLsPv90On08Fms8mhttvtMBgMCIVC6O/vh8FgwLlz58SacDGMRiMASFioWn2GSwBw7do1\nGI1GpFKppk1Kt7+VhQ83HA6LItRqtWIcVOsLoCm0Jo5YLBYBbBmoUqkk8ITb7cbo6KhsYCpBHli+\nb6VSgU6nQ7lclmdcr9fR2dmJ69evo7+/H6+//jqq1SqMRqN8ZqvLysqK7FOubT6fF+Oh0WhEgRHi\nIUyUTqdRr9eRSqVQLBZRLpdRKBRgNpsF7vnABz6AXC4n76d65VzbUqnUdE2MiJxOJ6anp/HSSy/B\n7XbL2eE17QYxGAxy7wy/+WUwGGSNufY6nQ5WqxXz8/Mwm82wWCwYHx+Hy+WC2+2Gx+NBX18fTpw4\nAQB49dVXYbFYRLcQIqGoOqler6NUKslzCIfDCAQCGB0dRSQSkSi2WCzetMLckaLkBZpMpiYvhp6M\nenDVi/f5fFhfX2/yYoDNA81EDN9/eHgYFy9eRDweh8ViQb1eR6VSEYtCC8EDzwPNMDIYDEpCh16r\n6qW1qmi1WjmQKnTB+wS2Ei/0WDQaDQKBAOr1OrLZrCg4HnY+EyrcRqOBw4cPw+v1ymfxMBLKaDQa\nyOVyck2qV2C1WvHEE0+gXC7DYDDIs9iOlbaaEOLhHuCeYRitJs8IRageoNPpRKlUEmegWq2iUCiI\nd6PVajE3N4fjx4/jj/7oj5BOp5uwXiZsYrGYeKcM0/kZ6XQaFy9ehM1mkzXX6/XiBbeycK8yelPX\nBYDAbcDWuvI1vb29KJfLyGQyiEQi4iTRQNDg3H///bh+/TqsVmuTE6F+BtdZjaz4HgsLC+jr65Mo\ns1ar/dIE5q+SHe1wXjS9NS4Ib6xer8NisQDY3Jy0Lul0uglXyGazcnPU/jqdDtlsFhqNBmNjY3A6\nnXIIGf7RswQAv9/fhNupG8vj8QgOajabYbfbd3Kb74hoNBpMT0/L/aq42XZFxI1Sq9UQDocFy+Vr\naGmZMTcajYIt5vN5+P1+3HPPPeJx0rvUaDSw2+2C3/Cw8/fj4+O4cOGCGDCTySTQSytLpVIRvJrK\nsFariUfHtQS2ohZGJjabTX6u1+ubDigVsMPhgMvlQqlUwoULF/DQQw8hm83K+tJ5KJVKmJ+fl2Qb\nDZ9Op8P58+fhcDgAbBkoh8OBhx9++B1bt5sVrVaLZDIpe3V79Uu9XpcIkBEScxT5fF68TpPJJK+l\nR6/RaOD1erG+vo6DBw/CZDLhwoULTQ4X8yRMgAJbECHfr6+vDzMzM+Lt8jlu9/J/5T3udEHULB9v\nkB4JDywVm0ajkWw4lVyj0YDdbkc+n5fMWKFQQD6fh9PphF6vR7FYhF6vFxyMB52HemFhAQ6HAzab\nTT6nWq0iHo9jfX1dPAUAEjK1ukdZKBRgNBpRr9ffgs+q//L3AGCxWFCtVgVaYJKByrJarUrZlN1u\nF/y3Xq/D4XDA6/U2eacA0NHRgYWFBcRisabrO3jwIH74wx+KcjEYDJIccTqdv5U1+nVFVfbMrNLD\nAdD0Mx6sfD4vxpZRFACJUgwGAwqFAoLBIIrFIpLJJKxWqyiC2267TfB4isFgQDabFUeBn3nq1Cm4\n3W6YTCbkcjk0Gg2YzWYcOHAAq6urv40l+v8lhHfU/AQNAT1oFU4gHl8qlWC1WmG1WpHP50VxEft2\nu91YWFhAsViUyCmTySAYDMJsNgs+z9xHMplEpVKBzWZ7S9R1+fLlpuoDOg1Wq/Wm7nHHMZOataJW\n56FUFaler0c2m4XVahUvVH2NyWSC2+1GOp1Gd3e3hCO5XE7e78CBA5ifn2/KQlLh5nI5OBwOAd8d\nDgfMZjNsNpt4DdVqFdlsFm63u+UVpbqJtmfqaZhopYm7JhKJpowhPREecACw2WxwOBxiwfl+LPnh\npqGxqVarCAaDmJmZEaVqtVpx6tQpKU8qFovyNTY21vJry/263TPnenDdCfGoEBIPv9VqFYijVCqh\nu7sbbrcb0WgUmUwGVqsVxWIRRqMRyWQSwWAQwWBQDiywVTozPT0tEdja2hpqtRosFotAHo1GA52d\nnUgmk+JltrpsTwSqSVcmHZmJBiDwTaPREKPcaDRknf1+PyKRCPbv3498Pg+tVotcLgej0Yi9e/di\nZWVFwnA1suWamUwmiRzW1tYEW85ms6hWqzCbzQgEAje9d3esKKnB6TXyZ0zPc6FcLhe8Xi9yuZyA\n42oNZCqVQiKRkPCNWJnZbJbPKpfLGBgYwNe//nWYzWZx4Y1GIwqFAq5evSpgbaFQgMPhkJAon8+j\nVqtJyVCr42iqkuQ9qHVoaigej8dx6NChpmytXq8Xi5lIJBCPxxEMBgVzVEMRbo5IJIK//uu/luQX\nMTiTyYTV1VVks1m4XC4cP34cMzMzsNvtkqyrVCro6upqqmVtVeE9UzEy8qEhUBMw3d3dyGQyEqmw\nzCWTySAej8PhcODQoUNYW1sTb95sNotHRK80kUhgYmKiKdPOz3/44YcxMTGBe++9F8FgEA6HQ8JH\ns9kMv98vX7spmaM2nqiiVmiopVIq9m0wGGC322GxWDA/P49KpSJYOuEROkalUglDQ0N4/fXXJQoj\nNsn62CeffBIvv/wy3G43UqmU6BVGt0zk3Kxe2LH2UC2tir/wcNMCM5xmOMJEATcS3We622azWW6W\nBad02T/zmc/AZrM1eauRSEQy5T09PXjuuecAQB6IzWYTeKDVDzKF16nWjqqHm7+fnJzE2traW8Lz\nTCaDjY0N+P1+7N27V5QAjYyKOVJJPP300zh27FhTHeHly5fxO7/zOxgcHITb7cZ9993XlM1kR8PA\nwADUjpdWFbWcTK3pZc2v0WgUDwTYqhNWjVOtVsPhw4cBAOvr67Db7XLwM5mMYKBUlsyM33fffXId\nyWQS//Iv/4JSqYRSqYSpqSkJ6elNVqtVjI6OQqvV7ookJAC5bypBCpWaWv1SKpWakmj5fF5w4HK5\njPX1dQwPDyObzSKfz4sDls/nJQ/B/x89ehSLi4tSSsUoCwDe+9734pFHHsFLL70kcBQ/JxqNwu12\n78hb33HWm5aDm4kJE4bW9ICIw5RKJTidTvHuqtUqbDYbMpmMhM0svgW2wvhCoSDJHoPBgFOnTgne\nWK1W4XA4UCgUpLOks7NTNr7aNaGGtK0u2+samTzYXrrT39+PaDQqf8uNqNPpEAqFJMFCw1Qul5HL\n5SRjTi+bG3xkZESwmjfeeAN/8Rd/gZ6eHvEg+cx5CC5fvoyHHnpo16wrsAXq01jQK1brgYlb8Xfc\ne8lkEn6/HzMzMwINJRIJwWiZ+NFqtYhEIk0VIclkEmxN/dSnPoVTp07JutvtdiwvL0sHj0ajweHD\nh+VZqdhxKwuhLhpRippZppH3eDyoVCrSFcXQ3GKxwGQywev1olAoiD5wOp2CqRPHZ1RptVrFU7TZ\nbPjWt76Fw4cPo9FoiDd69OhRUaQsBxscHGxK4N2M7Lgzhxk7KjkeIDXtDwAzMzNN2T56RMyYs4SH\n2AawlRGzWCzwer2w2+3y+8HBQUSjUfh8PrjdbvFogU0F09/fj3w+D7PZjFwuh2KxKNelWrlWlu0F\n5mrmUC3Duffee8Uo8HeFQgFWq1UqCFQjQWyXeBlLr4BNKx6JROD3+6HVajE+Pi74LiOFixcvNmV5\n9+3bh1wuJ5a61UW9RnqWwFtxtUqlglAoJB6PXq+HXq9HZ2cn0uk0TCaTYIk2mw1ms1nOAyMmv98v\nyZ1CoYBkMonOzk4EAgEsLS0hHo9Ly129XpdkTb1eh9vthtlslqhr+7W3qqhYuqobAAg2yUiPFSis\ncmF5IGsbs9ksOjs7xaiTXyCRSEiSplwuCw7f3d2NSCSCVCqFT33qUwKpqFEu8Xpel3p9vzGMkp4k\nLYSaaOGFlctlBAIBafOyWCziMrtcLqTTaeRyOdHoahsSs6rEiAiSl0oldHV1SThNDzYYDErGsFqt\nIpfLwev1NoXcu8EqA2jqVgC28GD+jp7m4OAgdDodvF6vZLUdDkdTrR9r02gwmE1Uu6j4f3XdmNyx\nWq1S+/o3f/M3YnCWl5dx4MABwYt2w9o2Go2mg6OupdrGptfrkU6nYbVaYTKZJKRmqyENTS6XQz6f\nF3ipXq8jFoshk8kgl8shlUqhVCqJUVefAz3UWCyGX/ziF+LV6vV67N+/v6l0CNg9e1fNT1DR0zEi\nrs11Yma/0WgglUrBarXCZrPJOQcgESVrth0Oh6yhxWKRmlTCdqFQCD6fT/Z3o9GAw+EQRWu1WrG8\nvAyPxyP7YCfdOTsOvdUiZbUeCoDgYAw1stlsU09rKpVqKiSl1eju7halQLebCZl0Oi1gO0Nteq9u\ntxuVSgXz8/PiTRmNRgGAGW7thhZGYMtzpKiKnveSyWSQSqWEbKFarSIUCsnfl0ol2ZTEOJPJZFNR\nf7lcRj6fRyqVArAZ3tDjZCbSbDbDbDbjwx/+cFMC50Mf+pCs/25RlDS8wJZBJx5ORUXM0uv1wmKx\nwG63IxAIIJFINFV1EFPLZDJYWFjA7OysKMBisYhCoYBSqYRgMCh7uFqtwul0ynVkMhlcvnxZsrjs\nTGHYD+wOT1IVwkI8b8BWRERYiBCH3W5HKpVCJpOB3W6HyWRCKpWCXq+Hy+XCxsYGbDabRI6xWEz2\nr8lkEs8zFovB7XYjGAwimUxKdGA2m7GxsSH7tNHYbHHev3+/1P+yo/BmZccF5+oiqO2LPEw6nQ6B\nQABarRY+n0/KVVhAC2xaC4be9P46OjrQ09ODe+65B/v370cymYTRaBQ8lKE7v0ZGRlCtVqUomPVT\ntGZqHZfqmbWybFfo2zFAeoqNRgMWiwVGoxE+nw/5fL6p/IWWslwu49VXX4XH40Eul0Mmk0EikcCT\nTz4pBb0sHeJnazQaqaX86U9/imeeeUaYm+bn5wFs1hJSce+GdQWa15aGRD3YGo1Gohx62FxXtXwq\nk8kgm81icnJSunZ+9rOf4atf/Soee+wx5PN5BINBqS+2Wq3yuRqNBv/0T/+EqakpfOITn8DMzAwq\nlQqKxaLgZrw+/v1uWF8V8+V5U4UOkV6vh8lkgtFohN/vR3d3N4xGI/L5PFwul9RULy4uIplMolwu\nY3V1FUePHoXD4cCxY8dw8OBB6PV6eDweeDweJBIJCaW1Wi3cbjeGh4exuroqcAmdAwCCV1I/3Kyh\n3xHNGhdELebe3rJUqVRw/fp1GAwGhMNhKa9g4S7rH+PxOAYHBxGLxVCr1RAMBjE9PY2FhQUMDAzA\nYDBAr9c3YW4A0Nvbi/n5ebz66qsIBAI4cOAArl+/Dr1ej1wuJz2ewFZSZCcu9jsl29fxlz1E9mwv\nLCzgwIEDeO211wSCUGsnC4VCExazvr4uGcXz58/jwQcfhMfjaUrQ1Ot12Gw2RKNRKfT/0pe+hG98\n4xvQarUwm83Yv39/E466m/BfRjDcC9uTJVyrhYUFKW2Lx+NSgcG9brVa8dprr4kXODMzg71798Jm\ns+Gzn/0sEomEhIUqRPXEE0+gv78fjz76KNxuNx599FHceuutqFQq8Pl8Ak+pUNZuMfBMGtJrV42u\nyuhTrVbhcrkkMqzX68jlcnC73ZLhZw3l0aNHsbCwgFQqhZMnT8JoNOLpp5/GnXfeCavVCovFIlEN\nXzcyMoJz587hJz/5Cb71rW/ha1/7muRBGHIz0abuhZuRXyv0ZojNg8JkAb2+er2Offv2wWQySRmF\n2jDf0dEBm82GlZUV2Gw2mEwmCbdnZ2cxPT0tpUPMapnNZng8HtncExMT2LdvH5aWlgRADgaD4lFy\nAdWe21YXei3q9avJBv77+OOPI5/PY3h4WCoR2PfOutVUKoWlpSXcdtttmJ6eRqFQQDqdxt13343e\n3l4AkH7lnp4ebGxs4Ny5c8jlcpiYmMCf/umf4qtf/aqUVJw/fx5DQ0NCTMBr2y2lV6r3yzUj1k2l\nZjKZsL6+DrPZjGw2K8+gWq3C7XbjpZdeQjweR39/PxwOB65fvy7tsrfffrt4ocBWxveJJ57A2bNn\n8a53vQt79uxBIpHAo48+isOHD4uBUzukeH27qVpD1Qvq9ypMRk+9r69P9AWhhlKphHK5jP7+fgml\nz5w5A7vdLlj41atXxVPn61ig3t3djWQyiatXr6JYLOIrX/kKvva1r0k0SlwT2PLWd2qIfi32IHoV\nqndBbwZAU5F3qVRCpVKBw+FAPp9HqVTC+vq6/G0ikRAKMI1Gg76+PmlPoiuv0Wjw0ksvQa/XY3V1\nFaFQCGtra5ifn0exWJTwPx6Pi8erLshu4PTj4VVDF1pllRSjWq3i6NGjSKVScLlcTa1cHR0dKJfL\ncDgcePPNN5HNZnHu3Dlp7yqXyxgbG5MMIw2e1WpFOByWki6Hw4FvfOMbsNvtcj0HDhyQ5I5K97Zb\nhEkcHmbikxTul3K5LJgiE2JGoxFTU1PYu3evtN5SIfwfhmz4/X4sLi4inU4D2GLe/8xnPoNKpYLV\n1VWcPn0ajz/+OA4fPiz7UsVMVVFL3FpduBY02FSONEZqCdvs7KyE4fSimTycmppCNpvFxsYGstks\nTp06Jcma7u5ujI2NNXGmZrNZXL9+HZcvX0YoFEK1WsXDDz+MJ554QkLxWq0Gm83WFPlshzhuRn4t\n4l71IKsZRLrgOp0OkUgEANDV1YVEIoFarYZMJiOeITtpFhYWsLKyghdffBEdHR1NxbosiWH5z8zM\nDMxmM55//nk4nU6p/dNqtdL3rCpvtZ+31YWbS11P9ec8SLVaDT6fDxcuXMDq6ioqlQry+Tw6Ozsx\nOzuLcDiMtbU19Pb24siRI1JIbbfb8a53vUuULdvGGJofOXIE4XAY4+PjUrzPGrhwOIzR0VF0dXU1\nwRpql89uEMIaPLTEy3ifjUZDFOS1a9fEAK+trSEej4shMZlM6OnpAQA88MADuPvuuyV60el0kgjj\nQfX7/VhYWEBvb68Q8jI51NPTA6vVikKhsCsM+i8TNfJRy9hUikBiwqlUCsFgEF6vV6oxbDYbnE6n\nVBoYDAYsLi4iFAqhXC5jeXkZgUAAAMQRIv7b0dEhnWf04AkpMSvOsi6g2Vv/jSlKYItJmwukKjOt\nViuFzmxJYrGu2+2G2+0W76arqwuxWAyFQgEzMzMYGxuTtkNaGVKGAZttTolEAh0dHXKz3Fy1Wg3p\ndFp6xVmsTSB3N5Cf0juj4lEZgXi/Op0OPp8PlUoFN27cwMzMDEqlEjo7OzE5OQmNRiPlV1qtFqur\nq7h27Rq6u7txyy23SAUC69CI81SrVQwMDDR5Nmynq9frKBaLiMViAoew4mG3CA+yihsy1KaweYLY\nOrPf6+vr0rrI+keNRoN8Po/777+/iWyZ0Y3aeAFsMl2pn6WWf7lcLuEqqFQqyGazQly73ettVaHS\nUWEitZaXmWh6eb29vYhGo6hUKkgkEiiVSlJaZbfbsba2hsOHD+P06dOwWCzo6emBTqeTZBv5VwFI\n+2gsFpNqEDZJsBqBCTMmn5mM3Am+vmNFyc2mVrbze7Yq8md2ux0ulwu1Wg3z8/NIJpPYu3cvMpmM\nVM8/+OCDCAaD8Hg8SCaTGBwclKwuM4bsY9ZoNIhEIk2JC6PRCLvdLpk1huzsLSdW1Oq93uomo+ej\nwgdUlplMBtVqFXfeeSe0Wi0GBgawsrICu90uFF5arRapVArPPfcc3vOe9wjTErE3ekqUcrkMp9OJ\nvr4+PPvss6Kcs9ksUqkUPvCBD6Czs1PgFjLocJ1b3atUDQ2wFREVCgU5yI1Go2kekcFgwPT0NHQ6\nnYwVUXk4u7q64HK5EIvFYLfb8fOf/xxWq1UytzycrP3jZ5Mhq9HYbOPlGaKSttvtUhHCyKrVhUlI\noBkyIBZJnJAR57lz50RxmUwmrKysYG5uTtiTjEYjIpEIxsbGMDQ0hGq1KqVwVH4AZB/bbDYkEgmE\nQiEJ66kEnU5nU54kk8kgk8lIY8vNwkc7Dr1ZCK4qHiqo7d0gc3Nz2NjYkAFk6XRaGG9OnjwJrXZz\nOmK5XIbRaJSMttVqhdfrFethtVqltEij0eDOO++Ew+GQB0TmIf6eVp0HgESqrS5qZwPQPKKBD5Qt\nm9xUZ8+eRTabhd/vRzAYlIZ/u92Oe++9F0NDQ4LRMCyfnp4GAKlE0Gg0ePnll9FoNHDq1CnodJvs\n0+VyGUtLS8jlcqIcuY70DnaDtw40DwijolLHNFBhWiwWjIyMiPHhwDXOZIpGoxJCp1Ip1Go1nDlz\nBgMDAygUCnjxxRelBXF9fR3f/e53USgU4Pf7peaYNYEkcWHYTgyVitVms900Ddg7LdxLLOznvRI6\nouE3mUzY2NgQUhdi6na7XTx21vR6PB7EYjEMDg7i1VdfxcjICN773vfKe1YqFbz3ve/FrbfeKtHo\n9iYCKldWbrBuk6WFv5FkjvpQKWpJC4CmAvN3v/vdGBgYgNlsRjgcxuDgIPL5PGZmZrB//37Mzc0h\nlUphdHQU4XAYmUwGMzMzGBgYkJ5ajUYDn8+HU6dOYWxsTNqdksmkKMLl5WVZJILxvDaj0QiLxbIr\nQHHWSQLNI3rVgzw4OIi+vj7YbDYsLy/D6XQKLZrP5xOP5MaNG9izZw86OjrQ0dGB06dP48yZM4hG\no3A4HEgmk8jn8wgEAnj11Vdx9uxZAJteVmdnpxjErq4uUdZqIkTlHm11YRkK15b7lweVh66jowOp\nVAo//OEPEQqFsL6+Lvfq8/mQyWSkTq9UKmF5eRlPP/00YrEYxsfHsbKygoGBAfGAvve97+EjH/kI\nrl+/LkXUTHgwHGXniZodpjeZTqebuCtbVRj10XjSwDMCIa3ZlStXUKvV8Mgjj2BlZUVe09HRgUql\ngitXrsDmCs6mAAAgAElEQVRmsyGVSiEUCiGfz+P222+HVqvFHXfcAYvFgh//+McSNR4/fhwvvvgi\nlpeX0dXVhWKxKOQi9XodHo9HFDchJhokTm/8jXiUwJaXww4QPtxCoQCv14tkMolMJoO+vj4sLy9j\nZWUF/f39klhZXFzE+Pg4BgYGpHbt8uXLiEajOHnyJEKhEE6ePAlgM+RcX1/H4uIi+vr6AADFYhFd\nXV0yA4U3rRaWE0Dm9RGfamVRWxXVn9HzYQZxdnYWoVAIzzzzDAYGBjA7OysdO8ViEYlEAlevXsXB\ngwelEmB1dRVWqxW5XA4f//jHxZqGQiG8/PLLiEajGBwclGSbVquVEP3IkSPw+/0A0LS+VOQ7bQV7\nJ4SeDPcgD7DJZILNZhNi40OHDuG1117DnXfeiampKezZs0fuc21tDXa7Xcre+Ew+9rGP4dZbb0Uw\nGMSePXtgt9uh0WzO5P7zP/9z6d9mXzhDd7XlV+09ZiUClfrExMQ7s2g7kO2lQOoeKRaLyOfz6Ojo\nwODgIEZHR/Hss88iHo9jbm4O9fomTeLIyAj6+vqQy+UwPDyMRqOBQCCA73znO7h8+TLOnj2LH/3o\nR7BYLDJ07Stf+QouXryISCQiHi3zIADEi6QhVMsZG42GGKmbkR2H3mSSoXurtt1duHBBkgLhcBjB\nYBD9/f0y4sBgMKCvr0+A19HRUdRqNQQCAXi9XvT390On08Hv9wvXn16vF9aPcDgsGay+vj44nU7B\nzmjVVHyPLWe7gTORFo+JARXeYDg9MTGB/v5+/Ou//itCoRAWFxfhdDqFgooEpUNDQxgdHYXJZMKN\nGzeE2OGhhx7CzMyM4L0cO3vnnXcC2Hy+bASoVCqYnZ2VcRKqF0ZYg+19KmNMKwoNOfE+GiUenlAo\nhAsXLuCb3/wmjh07hjNnzognubq6Co1mk5fS7/ejo6MDhUIBp06dkpK3UCgk5L30vs1mM6xWK9bW\n1tDV1dU0LI+temq3ELO0TJ719vZidHQUmUzmHVu3mxUaHqB5Jo5Wu9ljvbS0hGw2i66uLoTDYbzy\nyisANqEHDlNbWloSbs9kMol9+/YhFoth//790Ov1CAQCGBgYEIWs0+kwMTEhEQGJeWlo6Lmr1whs\nlQs2Gg2srq7edO5ix6E3O0HU7wGgr68Pt912GzY2NuDz+ZBKpZBMJhGNRjE0NIRDhw4JA7RGo8Hq\n6ip6e3vFsrrdbvh8PthsNimaZgjD3xPIZa9nLpcT/JHJHB4C4qVkdWn1EFEtvdreNcI5QOFwGGfO\nnEFPT4+Q5rIA95ZbboFer8fk5CTcbjeOHz+Os2fPNvW/Egsi9haJRKS32+/3yyHt6ekR/JgZQlVY\nZeDz+WR2USsL15KHWTWaWu0m27ter0dXV5fAPXq9HtevX0ehUMDQ0BDK5TJcLheCwSB+8IMfoLOz\nU6i78vk84vH4W+avkByDcBErQtj6qOK7arlVR0dHE9t/qwsTZaonR+MaDodhNptx5MgR3LhxA8lk\nEvfcc480Ltx1111YXl4WpiCHw4E77rgDk5OTQk7i9/vh9XqlR5sJS7/fj6GhIfHiq9WqdFNxlIdq\n4ImhktGMHAk3I79WMmd7uJBKpbC4uAibzYYbN27A4/GgWCxibm4OtVoNk5OTiEQieOqpp1AoFNDT\n04OlpSXcuHGjSYnRFSaDEAB4vV4hW6XFNhgMmJ2dlZICVcmorZS85t1QGE14gKEAQWh6GkzesJZx\ndXVVvJtLly4JjdWhQ4fwuc99Dn/3d38n61mpVAQ3ptWt1+tyuLVarfTLM8ur0+kwOjqKjo6OJkXI\n7/P5PHw+n2zaVhatVtuET3Kd6fUVi0Ux5AaDQViq3G43xsfHodPp4HQ6sWfPHjz22GPo6+sTpcD+\nb9bsZbPZpoqLWq3W1IHCQ81/gWb2c2JrrP/bDYoS2OpE0mq14j0Dmy3HiUQC//Vf/yXYuslkEvb8\n//mf/8Ho6CjMZjM6OzvR29uL559/Xvgd2CrK8iKSi3CP1ut1aWk0mUzyPR0rYCuPwqROqVSSjp/f\nWGcOAFkIhmBDQ0NwOp0oFovYs2cPAGBjYwM6nQ7JZBLA5uTFhx9+GMPDw+js7ERXVxeSyaRkWJsu\n6v8AwqVSCR6PR1iO6d0wA8ukA0NrYg8AhCSYh383iGqVaZQymQy6urqE/5HrGggERDlOTEwgFAph\nenoaDocD3/nOd+B2u+X3KnNLo9FANpuFVrs5HldlniehA7un9u/f39RpQYW4tLSE8fFxwYd3g6j4\nFCGaXC4ntXsbGxtSfkXMkqQhTz/9NEZGRvDSSy9JDR/3FpOHpVIJhUJBcDGSy1YqFanxLZVKMrOa\nRCNcX4aEgUBAWncZRu4GIUZYr9eljZPzmkKhEJxOJ1ZXV1GtVoURzGKxwGazyfpvbGxgbm5OEok0\n6FS+AGQP01NXk8lcf+777WVwhLY8Hk9TKd7NyI4VJWu7uCgGgwHxeBz1el2wiBs3buDo0aMwGo24\nfv063G43XC4XZmdnsbGxgWg0CpvNJnVsXFhKIpHA1NQUMpmMeIcMh4rFIi5evCgHdHsbGvEywgK0\n1rvBq2S7lYr39Pb2Yn19HWtra/B6vbjllluEUioajaJWq6Gnpwc///nPcdddd+HEiRNYXFxsmmFC\nz5RzQrhReLiNRiNeeOEFmM1m9Pf3IxKJYGZmpmmzAlv4JMMmhu2tLjSuPp9PlI/aPaLVaiULG4/H\nMTY2JrRn09PT+LM/+zNEo1EsLS0hEAhIIog4PbF6t9stHg2weahJ+tJoNKS+Uh1qxX3JbDefGRX6\nbpFarSa0fTS+DocDGxsbkmBhZw6ZxLjfc7kcgsEgPvShD4kR5zOivmFRejAYRFdXF5xOp6whi9bZ\nrcMWU3WNiaOSMV19bjcjO1aU6ihItejcbDaLWzwwMIB0Oo1isYgHH3wQuVwOp06dQm9vLyYnJ6WU\nR+0ZZiik1W7Ss+3duxeJRALXr1+XB6HRaIRQtlgsiiXZXjdFS07ZDeQC9Hi6urqarjeVSqFSqSCT\nyeDGjRs4ffo0fD6fjIKoVqsy48bj8eCJJ56QsIKKjOUmwGbVABlr7r33XmFuYvIsEAhgbm4Ow8PD\nTUzbDGNIoEymejUh0apCD4TC9a1WqzK1U6PRSMlZPB6XVsbf//3fx+XLl3H69GmZGc11ZXaXtF2s\nAezu7obdbpdWURo+o9GI5eVliaCIV/LZ8O9YzaHCW60s7HJS51Sx9TWdTktJoNvthtFolGgnHA7j\nF7/4BVwuF5xOJ376059KXoLGhWuk1Wql5fn1119HOByGwWAQ0hYW+vt8PtELaikjazq3G6Kb3bs7\nfgpqH7XaTbCysgKfz4eenh7JRN1///146aWXMDc3h97eXnR3d8sMX9KtqczRtAj5fB4DAwMS3nFR\nqPwMBgP27dsnGNt2DI3eABe61TE0YEtRbg91C4UCurq6ZJY5jVMul8Pv/d7v4ac//SmOHj2KF198\nEalUSoqmeUBpeXU6nSQTarWaZNJLpRKeffZZKfzV6XSYnp7G+Pj4W1r8Go1NAlSPxyPDtFq9NAjA\nWw4ckw0sGifVF/EzFox/7GMfw/nz54XZhiQt8XgcHo9HwjsqQHaXnTlzBplMBkNDQ02GhiVFPMTA\n1gRTGiK1s2037FsKPUDifjzP3d3dcq/RaFTYyAHg4sWLOHDgAC5duoSuri7Bi/kv8W86Q9y3d911\nF6rVKtbW1pBIJKQDCNjUDYwc1GsDmruGtteDv53smGaN0xL5YDn3olarIRqNolgsCovN4uIi9u3b\nh4cfflhqn3jgSEZgs9nk4svlsoRAly5dwuzsrJRnENglWO5yucQSsxRA3Vx8OFyc3eD1ENIAIA9z\nYmICDocDAwMDSKVSmJ+fR61Wg8vlwrPPPgun0wkAeN/73ifeR7FYlMy02ukBALFYDIlEAtFoFIuL\ni5iYmGgay5FOpzE9PS1Eqtsln883YT+7Bf/lHqXHYzKZMDc3B41Gg56eHhSLRbhcLkSjUeRyOezf\nvx/nzp1DKBQSxhseSNaZMoFTr9cxPDwMp9OJ22+/HY888gje97734caNG2LsGG2xyoNZeI5bVetl\ngS2HpNXrf4GthKmqnNhpx33d19cHvV4Ps9mMSCSCjY0NPPTQQ/D7/TKpstHYpE3j4DvCR6VSCadO\nncLKygpOnjyJxx57DM8884zguTTgrDVWYQvqAfXfX0cn7DjrTTyRB5nkpbTYtVoNFy9exNWrV+Hx\neLCxsYFLly7h8OHDmJ+fR6FQECby7u5uSQhwE3HOBZWqWo6kMoG4XK6m7LgaonDqo4qttXroDWwm\nwFQiBI1Gg+effx4Gg0EOcygUErKG7u5u+Hw+7N+/X6iliONyiJaakWULl81mk438ox/9CPV6Xdrl\nOETeYrE0tfxRMpmMhDvbvaNWFe5TtRmBCRN2g9FjGR0dRSAQwNmzZzE8PIzvf//70p9No0P4gtls\nJgmYJEqlUrh48aLglQzxVAYohvtqUoF7XW1bbfUaVVWYOOS9MPFotVqRTCZhtVplPMzY2BgikQgm\nJiYwOTkpESoNr9r9pdfrcejQIfh8Ppw4cQLHjh3DrbfeiqmpKWlDZWUMz46a8ea/fA5qudhvDKP0\ner2SFGB6nrRJrH0MBoM4duwYnnzySczPz8Nms2F1dVWsZzabhclkQjwelxo+bii2dTkcDrHa1WpV\nWG9Y2MuEAjEefs8aKhWsVanXWlnYqK+GtKS7N5lMQkMXjUaxf/9+2O12dHZ24rHHHpPWLCpKeiuR\nSATJZFIys5wL09nZiZGREVEYOt0mezphC2YOga2yK41Gg1QqJSH5bgi7KdlstqkpgXtldXUVfX19\nUg1w5swZdHV1obu7G48//ji6u7ubWGe4xgAwNTWFtbW1pm6qeDyORCKBRmNzjjSZmpgtV70vhqjM\noBO7JINWsViUBEmrC9sKVcWv1+uRSqWwtrYmEMX4+Dii0Simp6fxvve9D88//7wkYpjpTyaTiMfj\n6OrqksiT1QPbo1quJQl51D2psq7T0VKrQNTcxtvJr8VHqYL4Wq0W165dE9YfFu6Wy2VhBjp9+rSE\nlSzpoeRyOVy5cgU3btxAJpORLGwul8PGxgZmZ2eh022OuVTb6wics9yCYTfLBvigDAYDDhw40PKK\nkoeYCooHee/evdDr9VhbW4PNZsP09DT6+vqwsbGB1dVVoQDjAVbb40wmk2QcWfysstWEw2F5FiSc\nBSCFvslkUsZr8GCr8AATb62uMNmFRNya3mBPTw/GxsaalOjY2BjOnTuHdDqNUCgkIzNqtZo4CPz+\ntttuw/333y9z0QkfdXd346Mf/WhT2MxnarPZsLS0hNnZWSwsLEhvMhNLLDfK5XL45je/ie9+97vv\n0KrtTEhrCGwR05w8eRIrKytSZWEwGKQ11Ov14itf+YowxKuEvIx8zpw5I91UbrdbjJbL5UIikWia\n2EqdlM/npeOHa069BGxBcQaDQTqtbkZ2pD0IZqu9v2azGffccw+Wl5el9XDPnj0ol8vY2NhAsVjE\nsWPH8NRTT0koTZyIyYtgMIhgMCghIQtL+/r6MDQ0hP/8z//E5OQknE4nKpWKUF6lUimZ5sY5J2QZ\nAiDe6vnz51s+PDQYDHA4HFhdXZWHm0gkpNe9XC5jbm4Oe/bswcDAgLAuRaNRjI+Po1QqCcO7SmIR\niUTQ09MjUIVOpxOqqVwuhxMnTmB8fLyJ5CIUCiEejws2zNpCki1zRKvP58PJkyeblGwritVqFZ5D\nlaPQbDZjfn5eRgWsra1JFFKv13Hw4EGhtQPQ5E1arVbE43FcunRJStnouWcyGSwuLuLRRx/F5z73\nOYmGqGhDoRCGh4fh9XoRjUZx/fp1rK6uCtk1AJw5cwbj4+O4/fbb35E124nU63WpPaUDtbCwgNHR\nUXi9XmSzWRQKBZmqeOHCBZhMJoyMjCCZTAqRi1pFUy6X0dnZKV5fLpeT35fLZbjdbvzzP/+zlHdZ\nrVZRgOFwGLOzs9I04Pf7YTab0dHRIYaO1Q1LS0s3dY87bmGcmpqSxQE2+zXfeOMNGTfZ2dmJqakp\nKY0YGRnB5OQklpaWMDc3h2w2Kxx9PIRcXAKz/Jc45YMPPohQKIQvf/nLktFdWVmR17FVamlpCWtr\na4KjlkolRKNR4atsZdFqtTh27JjU3bGk58SJEwJXdHd3w+1245lnnoHX64XZbMbo6CguXbqEEydO\niFdNQ8QeZxY2A1uJHhY8kwLs0UcfxdraGoxGY5PRYusjsSDikwMDA/j3f/93gQVaXUZHR4XKCwAO\nHDiAxcVFuN1uRCIRLC0tQa/XY3l5GRqNRmbm+Hw+8Tq4JvTUua5syyU3QTabRT6fx7lz5zAzM4Mv\nfelLTT3b3OsWiwUDAwMYGxtDT08PhoeHodPpcOHCBej1egSDwV2TLGO7Mh2gQCCAw4cPQ6vVYu/e\nvWg0GvjZz34Gi8WC48ePo1AowGKxIBQKYWVlRTxK0toxFKdBVyNJwm7/8A//gGAwiG9/+9tC0m02\nm2Gz2aQ+c3FxEVevXkUymUQ6nZbyo2g0iuXl5Zte3x0pSmagVK/F6/UKxyQ7FbLZLG655Rb4fD4k\nk0ns2bMHX/jCF+D1euVm6O0Rb2AorY5zoHIjlvOZz3wGkUgEXq9XgHQ1E8mNzJIhEv7uhE7pnRIq\nLDV5ValUsLGxIfjLtWvXoNPp4PF45CBnMhncf//9CIfDUq7F8J1eDued0zqzIJj4L1mJvvCFLyCT\nyaCzsxNA89hUhizHjh2T9VaTaa0sTG6pUcXGxkbT6N+hoSGUSiWJTKxWK2ZmZqQdUW0pJdbFWS/E\nFFkSA2yVU9VqNVy7dg1//Md/LCQixCW59mpLHcNunrWdsHC/U8Lxs2qjBGd1azQazM7OAgC6u7vF\nC9doNtn4l5aWsLq6CoPBICVBbBDYTrrD8J5OFitdvvjFL+Lb3/62XAvxUBp2t9vdVFVAvbGT9tsd\nK0r2XVLxnD59GrfffruEJIVCAbOzs4jFYvB4PFheXkYwGMT169eFlYUKMRKJSGsiw3IeYmCru0ZN\nxqysrKDR2GRHVzNazHiptVwkAiU21crCBJQ6OyUQCEjZQzablXZQ9hR7vV5cvnwZDocDU1NTEv5Q\nyXJqHcMaYAujUck3+HksEO7s7Gyq5aSHxamafr8fTz75pBATtPraApB15X3RoDLxxV7itbU1Icko\nl8vo6enB+vp6U/KQPeIMNZnlVdd0O/0cDQuNFcNIGsjR0VFotVosLS01tUG2uoEHtjqf2EEHbJ7d\n119/HcAWH2hHRwe6u7sRCASk9bhcLuP++++X2mpCb8yQcx3UbDqfI9+7Xq/jd3/3d1GtVqVEkUkb\nOlI8O1qtFouLi6hWqwIh3ozsuI5S7RrhoLBEIoFyuQyfzwcA6OnpwdraGlZXVzE2NiYhyfj4uPR5\nErAFtmrzuMjMMqqkAdykhw4dwoEDB+ShqIsWDAblYLP7hzVzre75qIXgNDrValVKr8rlMo4ePSqH\nWKvVYmZmBh6PBy+88AL+4A/+AP39/YjH4zIK1Ov1SiZQbZXjepG4gf9fXl5Gf3//W1rA6vU6ZmZm\nMDQ0BJPJJFad411bfW0BNHF6FgoFmUpZr9fR3d0tHTOk9Lp69SqGh4extLSEAwcOyO8ajYZ4KDqd\nTnBLQh3bSY2pLGOxGO644w5RkEzW0bth5JBKpYRdXh150urCtaVjw2aHYrGIQqGAd7/73TJaRCXQ\nZWheq20OYSN/JdeP78V2Z1UnqBn23t5eLC0twWKxNM3x4mcNDg7KuaFDxn1+M7JjjJKbg4fX5XLJ\nDJzZ2Vk5RLyZmZkZTE5O4tKlS/D7/RgeHsb6+rqEIJznrWZtOW9E9RhZKFqr1fDDH/5QsAxgCy/t\n6uqC3W5HoVAQ5UwCjd2AUWazWfh8PtlAasmD3W6XXldyUJJxxe1249q1awJe5/N5mRXCigBga/aO\nuh5qEXaj0cCnPvUp6aPn35TLZXR0dCCZTMLr9eKll16Smlc1AmhVoRFiaEiWGfJyptNp8SBdLhdK\npRL27duHjY0NNBoNJJNJJJNJrK6uiqFIp9MyLpjMOcBWzSlbG3lYAWDPnj2i+Ij5mkwmhMNhrK+v\nS7tqsVhENpttSh61srAUiueR59XpdMr6zMzMyPkulUpSA5nJZPDcc8+hr69PPE32ygNbe5AeqBoN\nqfBcrVaTpBEAqc7QaDQIBAJi+OPxuLAL7WTv7rg8iBcLQA4dW8AsFovMVwE2w+RIJIITJ07g4Ycf\nRqVSQSQSQWdnZ5PlZWjMflxaEYpKPsBaPpfLhZ6eHnR0dIgFvnz5slyTxWIR0lla71YWdnB0d3eL\nd6LT6SRTRxDbarWiVCphYGAAfr8fMzMzePnll3HkyBHUapuzXTjiIRAIiJe9ncCC3/MZ8DAzmbSy\nsiL1lFqtFiMjI3jttddkWBzHiO6mg0wjz2QKDxIPl0ajkdnlLpcLMzMzqFarmJqakgFsXM9YLCY1\nr3wfYMtjVzFK7ttMJoM33ngDb7zxBs6fPy/vodPp0NXVhUgkIjwG5GHdDRgl9ysNPLDVgskaaU5H\ndTqdQixy4MABHDp0CLfeeivC4TAikYhEKowC6QVyn27nh1B/x4qQ3t5eScDV63VEo1FRnFSuq6ur\notBvRnakKGu1mvRh0sWen5+H0+lEJpORG2HKf3R0FLfeeitOnTolGAY3D8NMsohwxjIXgV8sZFUr\n6/fs2YOVlRUsLy8jmUzC4/HA6/XK+FsCy2yBYljQysL7DAQCcrjcbneTVebfkEllZmYGe/fuxYMP\nPiiYYTabhdPphM/nw09+8hPZMAS+6fnQg9xesrW8vIxMJgO9Xo9cLoeVlRW8+eabcDgcGBkZQTwe\nx+zsrGThyfDUykL8mvgtsOlx0GtjEpJ98rVaDa+99hqOHj0qmHy9XpcOnVgshp6enibvXO202f7F\nZJDRaMTdd9+NsbExBAIBXL58Ga+//jrGx8eh0Whw5coVmM1muFwu4cYkttbKwvtWo0BGn4Q4WMWR\nTqdhNptx+PBhfP3rX5c6ahoxGppkMtkUFqt4OfWMWstNpTw1NYWzZ88imUxiaGgId9xxh1RzqLwS\n6pzvm5Ed7XBmplOplCQFiN3o9Zuzia1Wq5SzFAoFbGxsQK/XY2NjQ5IK5KnLZrNi6enV0IqqwDAX\nnYfbYrHA6XTCYrGgWCxicnIS+/btg16vl8wx8bdsNvsW4oxWFBbZ9/b2iuKy2+1ilJgdBTZDx3K5\nLJMqmTGkheRIYI/HIwZGDa/VrKua+eZnPPLII1Kwbjabce7cOfh8Ptx66614/PHHodVujiImI3Wr\nr61aZMzoghn7en2LBZ/ZUqfTieHhYVy7dk1mBxmNRgQCAeEiYJaaHo+Kzan7j0X+9MLtdruwb09M\nTGB4eBgGgwHf+c534PP5ZCYMlcFuKL3iHlQdIUIIJBrhKOBarYbFxUVMTk7i4x//uHiO3OMmk0mI\nkVWGcj5DNZpV9y6wmTOZnZ2F0+lEIpHACy+8gFOnTiEQCKCjo0OiBe5rNXR/O9mRouSmUGvC2GLF\nzhmGfclkEgsLC0in0zKkSafbHOEQi8UQDocl7KFC5EWrRb9q8zq7JzQaDX7+85/Le9rtdkQiETid\nTqysrIjLzlINNSRoZaHhUK93dXVVyltYHjE0NCTtYmtra5JQKZVK8Pv9mJ6eRq1Wa/J6uMmIi6kE\nJYQ9uNF7enqQy+Uk7HvggQdQr9fx1FNPwWg0Cus5I4dWDw8LhQLS6bSUCKmHmXPhyQbkcrlQqVSw\nvr4umG+5XIbJZBJj4ff7Zb1ooHhgVZhIxX+5tmTHYptusVjEv/3bv+H222+H3W5HLpcTLLpcLr9l\nvEQrCisG1JlVvO5kMiksVXq9Xs59KBTClStXZH/r9XrY7XZMTk7i6tWrcDgcso6qMQLQZNiBZlKO\nnp4eYUUn7dsHP/hBXL16VaIyg8Eg+PNvxKPkRZELEoD0pOZyORlulclkcOTIEdx7770yToBYJC9U\nJV0AIB6kmnhQs1b84kb0eDxyQNmzC0AIVdVBUtt7QFtRGo0G1tbWJAMIbG5Aehc8tBqNBteuXZMs\nII0BO0jcbjf6+voE4yKGxv5WNZu4fZMwRLx06ZIUplutVnR2duLMmTPw+/0Sxm5sbGDv3r3yfq0s\nGo0GXq9XPOztPJTcw2rWtru7W/g5DQYDOjs7kUgkhLiECkH9DO5ZVi4QF1bxNVILsunCYDDg2LFj\nku11OBxiCNkj3epSrW5O9VQ7tKg8WSOaz+eRSqXQ39+P/v5+5HI59PT0IJlMCqNQqVTCoUOHZPCa\nume51mpSl7XcKj5Kg2e1WoWk9z/+4z8AQNY8kUjAarXuiJ1fs5NNrtFo1gEs3PQLWkv6G43GzTV2\nvgPSXtvfnOzytQXa6/ublJta2x0pyra0pS1t+X9RWh+4a0tb2tKWd1jairItbWlLW95G2oqyLW1p\nS1veRtqKsi1taUtb3kbairItbWlLW95G2oqyLW1pS1veRtqKsi1taUtb3kbairItbWlLW95GdsSR\nZbPZGh6Pp+ln5KVkuxzQzP7zf0NUTrrt5MFsEeO8bw7FUr9yuRz7TVuWLtrhcDQ6Ojqa7k8dO6Cy\nsvxva6v2b/P/29drOykvuRH5nmoTAkkEAAi1P38ObPZRZzKZll5b7tvtHAIqH+F20gW1D3j7eqgE\nsvw7vre6tmy7U3ka+R7AJhFKoVBAPp+H2+1+y/tyrs78/PxGK3fmOByOht/vb1oHtm2qxBPbWz6B\n5nWmqOv0q54Dv99O86f+Dds/OXNdfTa8nmg0inw+/7Z7d0eK0uPx4C//8i+FIYVs0aQx4kYj5b16\nEyrxBRdPHXjOg6o2uqvU7yS/IAfg9gUMh8M4evQo+vv78cQTT8ic8PPnz8Nut+PZZ5/dya3+1iUQ\nCPYt5TQAACAASURBVODv//7vZT0qlQqWlpYQjUaF+os98hwrsH1Nt29SflkslqYDrlKDZbNZOBwO\n6WlW15+vNxqNmJiYQDQaxSuvvCKMUZcvX8bGxgZeeOGFd2zdbkY8Hg8+//nPCxdlo7E5w3t4eBiZ\nTAYGg0FmqABoMiJcD4PBIITQ3JfqWA2OtOXrScxCtq1CoSB7XN2/Pp8Pc3NzsFgs8Pv9Mvfo2rVr\nOH78OAwGAz760Y+2dHtgR0cHvvzlLwt7Uq1Ww/r6+ltmmpMh3mQyCY8D+R2MRmMTfSPPPakXuce5\nv7VaLXK5nDCRqe+lcuaOj48jk8ngtddek7ElfDbT09M3PQ54x6E3LxCAcBn+MgZxLpDK7MHXAJDN\nSdZjlS6Jv1OFje9kbSFFGGcsO51OrK6uChlGuVzGuXPnhBBjN7Rqms1moedSDQ2VGjcASQHUAfAq\nryTZVkiBxiFVKsEA34MHt1wuN43X4HuSq5GUasDms3jzzTexvr6+I2KBd1LUA0KjQ4IWoNnbISWa\nOneFA8S4xlxDKr3t+1vlptRoNELrxetQn6HBYMD09LRwJs7Pz+Nd73oXAMhE0VYXjmDgcDDS2HHe\njWp0VD5aYPPZkEWM+5jjHBhZqdEAf07FyOdJOkE6ao1GAzMzM8jn8xgYGGjSK5OTk8JwfzOy45k5\nHCQPbLEYA1tEsKSa4sZSWT94Qz6fDy6XC93d3ejt7cXo6Cg+/OEPY+/evTIDmQecC0zFwe9Vqiwu\n7PLyMqLRKP7wD/8QU1NTwiSkclu2qtC7Y5jAteQmURlrqODUQ18qlWA2m2Uj+P1+9Pb2wu12Y2Rk\nBLfddhtyuVwTozmpv1TSWW5MUoiRKHhmZgY2mw133XWX8DRqNBoUi8V3eOXeXkjnp84PIvWXypDN\nvUqGbRotss6oRMfFYhEbGxuYnZ2F2+0Wg89no64tr4GeE7DJuENawqGhIfT29iKRSGB5eRkHDx6U\ns7Yb2IMAyOhjNSLi7PRfBhuR+UoNzfP5vBAVc276mTNncPDgQTnr3KOMRsnKpBorPgfqEA41C4fD\naDQaMpqYHuzNyI49yrW1taYNQEozXhwVk/rFGyoUCiiVSgiHw8hms8hkMsKMfunSJVSrVXziE58g\n5gUAsqDqOAf1dwAkDA0EArh06RLOnj0rVGLbcaNWFs5FVuGLXC7XhFNy5grJYV0uF+bm5iRkI1Eq\nLfLY2JjMM3nPe96DfD7fRCmmKkqOSt2OMfGQz83NIRaLYX19XTwFdThZqwoNA/kJfxkuSU+Ra1yr\n1WA2m2UKoNVqBQC5b41GA4fDgY6ODplZzzWn8aJSZEgPQPalyrAdiURgt9vx4x//GAcPHpRnTRLs\nVhfuH551Ghyuo1arFQo/7i1ihpFIBIVCAW63G8FgEB6PB36/H1arFfv378eDDz6IeDyOU6dONQ0N\nYzSw/RkCW8+bz/nq1auoVqu4++67MTc3J44Vo+GbkR0pynq9jnQ6LVZTnSan0WiEf297skHFd4BN\na0GvhUqPnInnz5/HAw88gAceeADJZFIWtdFoCHFtLBYTa0vlyUNfqVTwox/9SDYZMY9Wl3q9jvn5\neblPejMqlyYNDr8HNhMsTqcTGs3mkKxsNiszp0mwTE+mWq2iv78fn/70p5FIJGQzNRpb7N6xWEw2\nPMl9+awjkQj++7//+y3zjFpdUTYaDTgcjiY4iEaA7Nu8JxUTo1dns9nkZ9xznHHEfW6xWPDqq69i\naGhIDHelUhGFoBoWFYoCNvdub28vHn744abRrCQcbnVRvTT+X6PRyN5lOMyQmyG52+2G2+2G1WpF\nKpVCMpmU4YRUrNxfH/nIR7CwsIDbb79dRktwcgGdM5UgeTvfqkajwfnz51EqlZpmg/9GPMpyudwU\nCnBSIB/s9g1ATCgej8tmo0fEmwOaMSG32414PI5wOIyPfexjyGaz4lGqQ+ivX7/e5G5zY1+5cgU+\nn6/JshHwbWUhzEDvTR3SxKw015qbAgByuZwYIGaoqbgYrnCOkFarxaFDhxAOh3HPPfcgkUhIeM3X\n53I5LC0tAdhKanDjPfvss4hGowLG12o1eDweGTvcqlKr1eD1egW/JWxhMBhkgiiFe81qtUrWWXUA\nqGw1Gg3y+bwoXqPRiH379uG1116Tva4qX8I/TNioCR29Xo+rV6/KHHbixX6/H0eOHHlH1mwnQm9u\newUA9y2AJkhOhcpU3J3GnGddq92cb873u++++5DJZNDZ2SmRDwBJmgFbuQz1HBkMBkxOTmJxcVHm\ne3Hv3qzseAojhTfCxaB7zZsGNjdKsViE0WgUD4kzchh6cIhVKBSCRrM5YZEbs9FoYGRkRBI0fE+H\nw4FYLIZ4PN40r2RyclIOQD6fFy93bGys5UMYFYvkoVLBajXBooLiKvzADavVamU4FeeQMHznOI5G\no4GDBw/K2nKDlctlLC4uIpPJyHVwiFwkEoHD4ZBknl6vx9DQUMvjaDQkKmM+vR012cdxBvV6HalU\nqmmuNudFlUolFItFWCwWUaTc23xe3//+9xEIBCRMJI67tLSEUCjUNL2x0WjgzTffxL59++S8NBqb\nc6Gq1Sqee+65d2zdblbo5Kh7k5nrUqnUpLS2K0smuWj86/XNeVsAZA641WoV45xOpxGNRnHHHXc0\nTT8olUpYW1sTZ4GfQQhlbm6uacwzDZ06Gvd/kx0nc9SwSwW3VTeWWAFnZ9CCms1m8Xg49IulKSsr\nKwK4c9MuLS3h4MGDshmBrWmCRqMRV65cEaA9Ho9LFpYgMa0GlVCri1r+oIYRVKA8uJzzQqtKHJfJ\nnGq1ikKhgK6uLjgcDthsNqyvr8NisSCdTsNqtaK3t1e8QbXUxWQyIZ/PY3JyUt47nU7jypUrsFqt\nMoKiXq/D5/PB7Xa3PLTBMJghNtcTaDb+9NgzmYzMbAG29pwaQpdKJbhcLomyCCFpNBocP34cTz31\nFLq6uiQJydET5XJZohuu+dGjR8U54FpqtVqEw+GW99aBrbHValmVOiVR/bnRaITX621yqOhM0agT\nDnG5XHC5XEin06LYDAYDJiYmMDU1hZGREbkG7v1IJNKUL/F4PHjjjTcAQIa6MYJVX/928mvN9Va9\nHzXrpOI/LpdLQh0AUi6Qz+eRSCRgMBhw5MgRlEolCR9ZgMtslFarRSQSwSc/+UnxYngdHP151113\n4YMf/CC8Xi9sNltT2YfL5cLQ0BBcLtdObvMdFzUDTaVPPJeKcH19XaYJqmN/uZYMTwBIvVkul5Mo\noFqtYu/evfjgBz+IfD4v+BuxzePHj6NSqWBkZETCd1p/zhefmJhAV1fXO7xaNydqnSRFxQ15/3v2\n7AGwVcICQLKnuVwOGo1GMDW15pRYGsPHUCiEv/3bvxVvlp+VyWQwMDAAr9crFR6dnZ3yeo61TSQS\nANAUTbWqqLWhPP/UD9zL1WpVprGurq4Kvliv18WIOBwOmZ3DpCRrMYGtpG0+n0d/fz9+/OMfA9ia\npd5oNJBKpXD69GmUSiXcuHEDXq8X5XJZEpr8m/Hx8V86N+pXyY4UJQ8JrSSzTqpyUif8qXV6at3Y\n4OAgXC4X1tbWZAhWrVZDJpMRoFXFLycnJ/GBD3wAwGa4l0wm8fnPf15KKi5cuIDR0VEAkAx8o9HA\n8PAw6vX6rqhFY2KBWA5LmwiCm0wmmdA4NjYmm4bZO51Oh3w+j+7ublitVsFpGRal02mxppVKBVqt\nVhJzBw4cEIOXyWTw6KOPore3F0ePHoXf74fNZpNSFibM7rzzTjidzpafwEjZXqqj4pVcR2BzaqDD\n4ZDwDdjEjy0WCzo7O9FoNJDJZKTwn2eBM8P5tzqdDu9///vlbHBP33LLLXj11VdlEN8vfvEL1Gq1\nJoVIo0aIpNVFxbIBSPJVrX0kDNTX1weLxSLePcfcVioVZDIZDA4Oyhjm7bXEagVCNpvFJz/5SVF4\nxJbL5TL279+PgYEBPPjgg/jBD34go6DZLDA8PNxUQ3szsuPyIGYLGa6pmVjiMVRU/Bm/LxaLsNvt\nSCQSiMViMBqNyOfzgvtQCRoMBiQSCXkfgts8mH/yJ3+CaDSKWq2GfD4Pu92OGzduiNXS6/Xo6+vb\nkcVoBSG0wU1GYJ9YV6FQQK1Wg8vlEu8E2KwiSCQSMmo1l8tJkThfw9drNBpEo1Fks1lotZvzuTmn\n2mw2Y2RkBP39/TItkG1eBNlrtRr27Nkjv98tihLAW8pJ1AJwKquLFy824ZdUplqtFrFYTLxH7nPi\nj8Rz4/G4KAWLxYJnnnlGMtj33XcffvKTn4gnSWxfzaRPTEzg8uXLUuPZ6rAGAMlwc51UBUllR2+5\no6ND1prdTMTUXS6XeH2FQgG5XE6wS6PRKKWEPBtarRYzMzOSICsWi/jQhz6Effv2CS5/4MABaYrg\nVEifzwer1XrTpUHAr6koga1QGthKKnDTVatV/H/svXls3Nd1NvzMytn34TYcDhdRoqhdsiTLji3L\nm7w0cdy0TRogSYsk/idpUaRAgCL5I22aAk1bFEXeok3iug3axM7i2InjLbIdy7IkK1osyRIlUuK+\nc/Z95cz3B9/n8A7t97PoD34zxKcDCFoocmbu795zz3nOc57T0dEh0SI9udvtlijTarUKHYAfmuA4\nO22IjxGoDYVC6OzsxNjYmIy+JT46Pj4ukYLdbkdbW9u64U/S+F65Hlxffi46KqPRKC1hHB1stVql\nDYwXBHEb4rgs8DgcDpRKJSSTSSSTSdhsNvh8Png8HuzYsUNSdBbnrl+/LtxLq9WKbdu2yc9eDxcR\nL1seSgCCF6qUFUbeandUpVKB2+2WwiEvB0IYavZkMBgQDAbruL+dnZ1IJpOwWq24dOmS4GRM53t7\neyVCBYArV64ISVuNghvZVr9XtQOKe5rrQafFi1uv18vnz+VyqFQqEp2zMUCr1QoNjj6BcFBbWxvO\nnz8PnU6H7du3y3uhJRIJ2c/5fB5bt26tY8TcqK2p13s1/kDnyGoq/w8XzGq1olgsIpFIoLW1VW5f\n3tLZbFbaiHjDRKPROsAVgAgGpNNp2XRc4Gg0imQyiZmZGfnwAwMDstHX02bjgVUPtMrDY6RCTqrX\n60UymYTX6637d2A5XWTFkWuRTCZRKpXQ1NSEcrmMVCqFYDAor8Ue7mq1imKxiGg0KvxJshV27dol\nl6SKDTWyMeLhfuDno+PjXuYvUrWcTiccDgdSqRQcDoesp8ryYPcZU3kWhFiwqVQqMJvNclhJSdq4\ncSPi8Th6e3vlAgwEAjh//rxUvvne14uVy2WhRgGQteGaEFuv1WqwWq3y7y0tLVLIZXaZTqfrom12\nSJGTncvlJDDo6+uDy+VCsViUYlmhUEBHRwei0SjMZjOq1SpsNps8P7Vl+kZszRglb1BuFhUMB1bS\nR5vNBpPJBKfTCZ/PJxgMwe5sNguXy4VkMonR0VFMT0/L4pZKJenOcblcMBgMyGQyyOfzcDqdcmOl\n02lMT0/DYDAI1cBisYjzpiNv9IMMQPh0XFM1DaejA5YLMzabDS6XS6L0TCYjqQ8dJDfb1NRUXcth\nNpuVtMZisYjjJK7G152ensbRo0dx4sQJcaC1Wk3EM4CV/ulGN7X6qhZgyCzgwfF4PDCZTEKNMplM\niEQiACDteUwFmcIlk0nBJpnlkIICQBgIvIBqtRpsNhtisRjeeOMN2O12CTTOnj0r71NNxxvdmPGo\nfF5e7iq/kiRym80meDfrFgCEl8oOPpvNhtHRUUSjUWSzWSSTScEyCYu0tLTAarXWYbzRaBRjY2MY\nHBxErVaTbrRgMIimpia5fNYCa6y56q3iJmqorb5wNpuVWzKXyyGTydS1MlWrVWSzWeRyOVkIl8uF\nF198EY8//jj+9V//FZcvX4bL5RIOJjcUnd6Pf/xjnD59GgcPHsSpU6dkQxJwJ16yXkzFeWjqoebv\nJNbycuCmzOfzkq4nk0mMj4/jzJkzSCQSqFQqOHv2LL773e/i6aefRiKRgM/ng9PplFSSjhoAnnji\nCczPz+OOO+7A6dOnpQDU398vr8n3sx4iHrWwAKyIrqhMDb1ej46ODiwtLcHhcCAWiyGRSEhEzmIW\nC22kRrFaa7FY0NLSAovFItEScUi+B51Oh6mpKeh0Otx2223YvHmzPLf29naJltRovdH5vzQGJsT9\niCPy3JPET5jHYrHA6XQKZ5JwA/fj9evXJaP0+XyYm5uD2WyWCrnD4ZCIkq9PGiEvrVdeeQUGg0Ge\nCWGWD5IJrSn1Jh9NJT+rL8h/N5lMGBkZQUtLC4rFIjKZDEwmE7LZbF3UdPbsWeRyOXR3d2N8fBz9\n/f2wWCz4+Mc/jlKphFKpJBVr3q5vvPEGCoUCDh8+jFAohG984xsIhUKoVCqCs7F6rD7ARo8qCXa/\nl6lydSTTM2LJ5/MSRWezWeh0OphMJrz11lsYGBiARqPBkSNHsGfPHthsNtx///3o7e2VTcji3NLS\nEn75y1+itbUVH/nIR1AoFPDb3/62rpDW2dkpUejqLpVGttWHgoeFqR5pUZcvX0atVsPExASCwSAA\nCJWqVCoJp5FRpVarRSgUEjztoYcewi9+8QuhcrEZo1aroVAowGKx4LbbboPb7cbjjz+OBx54AMAy\nfjc0NFQX3fL71kMxB6jH1XkxqZgkI8r5+XmYTCYYDAakUilhF5AtQNbKwYMHMTExgUgkgtdee016\n9f1+P/x+v5wVBgqkAA0PD6NUKmH37t2inZDNZrFnzx4JumgfWuoNrJBk+SZ5iLn5WES4fPmypOUq\n895ms+HYsWOYn5+Hw+GA1+tFJBIR6svevXuF7EswV6fT4fXXX8cLL7yArq4u7NixA7VaDX/1V3+F\nrq4ueS8ks/N76cgb/SAD9cICKs2CEmlq98jbb78No9EoKTEjdafTibNnz+L69etobm7GuXPnMDw8\nLMTnTZs2SdREPEev1+PMmTN4++234XK54PV6sbS0hBdeeAFXrlwRDUy/31+XmgPvjtQa1bi2TPPU\nCJ0XKXHbLVu2wG63IxKJyNco4cW9bDAY0NzcjGvXrsHtdmNsbAyhUAgvv/wyEomEYHBarRZTU1PY\nv38/du7cKU7317/+NR599FE5tFu2bJH0kBmXSq1rdOP7ZOWff1aLOfx//f39sNvtEh2qHU1kG0Sj\nUbzyyivQaDTw+/1C13K73bBarRJFarVa+Hw+pFIpRCIRjI2NYefOnXC5XHjjjTeEAmS32wV3/qAX\nzwcq5rAays1HnIuLAUDK71arFbOzs5KyjI2NobOzEwCQTqdhMpmwtLSESCQCm80Gt9uNmZkZcQr8\nvk9/+tN46qmnMDIyAgDS9sXKL3FR1SkSKmD02+jGjcXCjVpNVCuzRqNRlG2I9ZhMJkxPT6NUKiEe\nj0Ov18Pj8WB4eFha5shFJQULgPAyH3/8cQQCARFioHgACxjt7e0SuaqV4/VCDyL/ToWOVFobYQ2r\n1SrRo9qfzGiJ1W2j0Yienh7Mzc0hGAxiaWkJk5OTwnsljh6NRjE7OysR5dzcHO666y6R0zMajRgc\nHJTnQV4hsfr1EFFyH/DCJrShFlOJs1+6dAk6nQ4+nw/xeFxEd00mE86dOydO0WKx4PLly9Dr9fB6\nvdi2bRusVqvQ2tiJdvz4cXg8HhgMBmEnzM7OSnCm1+sRDAbr1JqYAazF/j+BeLyRCayqMkgejwfR\naBQXLlyQKmkymcTExEQd1WLbtm2oVCq44447cPDgQSGcc4FZFc/lcujp6cHk5CSCwSB6e3vr+HDU\nuFSjLxVLa3QjXsYIBlgpngErLVparVZwx8nJSQGyK5UK5ufnAax0gdjtdni9XnziE5/ALbfcgubm\nZsHA+LxsNhuam5vR19eHSCSC5uZmXLx4sS569Pv9grup2Nl6idb5fvm5eXAZjauX8vT0NLxeL6xW\nq/BU5+bmoNPpBLLo6uoS2Ein06G1tRVjY2NCjWPBsVwuIxAIYGZmBsFgEMePH0d/f784X3ZQJRIJ\neS4A6nRX18P6qtoDpALybKr7Ra/XI5VKoaWlBel0Wkj4hDZINK9Wqzh37pw4SbJeeNmoDS99fX0I\nh8OIxWKIx+M4ffq0+BsS2cnRVjPhtdYvPpCjZITGBeEbU4s7Op0Ow8PDcDqdQo+YmZmRTg/SfObm\n5nDw4EFp6WKEqlbY+ZodHR1SoVWLHwR3dTodHA6HVHxJWl8PGOXq1JuXDhv4GWWygjoyMiLPYWZm\nBhcvXpR0ularYXJyEna7HYcOHUJraysCgYBsVK4dqRxGoxEtLS1IJBJIJBISmbMqPDAwID3jZCyQ\nNM0e5/VgKlap4ofEuFKplDAsvF4vvF4vYrEYOjo6YDQaEQ6H0dnZiYsXL6JcLsPtdkOr1WJ+fl6i\nP5PJJKLAdA7Um0wmkwJ7sHqbTCZhMplQLpelIsu1Xy/G96w6SLbWEnPV6/XIZDIwGAzo6OiQttp4\nPI5EIoGhoSERXYlGozh48CCcTic6Ojpgs9lk7+XzedRqtTqOdqFQQKFQwMTE8sQMtcDT09MjuLp6\nUaoQ143YByKcq5EaPT3xH5Xe4PP54Ha7JfXweDwCoDc1NSGbzcJkMqGlpQULCwtwOBw4efKkpJX8\nuVwIAHA6nVhcXBRSarlchtPphM1mk4fS1NQEp9MJt9stLYyNvvHUNJsPEoC0GxK/sVgssNvtmJyc\nhE6nQzQaFToVe7vT6TSam5ths9nQ1dUl0eXzzz8vDpjYJyvaJPMyrWQEyQ1JiITVTKaY66Eyy4uG\n75NwgVoFZdV0aWkJXq8X+Xweo6Oj8Pv9Mkogn8/Dbrejt7cXn/3sZ0WAgR1Q3H8qNNHU1IREIoHR\n0VH09/dLOmgymdDV1YXr169LhsACJluBOfep0W01Dsl9zHWmRgB1U69evSpRYbFYRCwWkzWZmpqC\nz+eTvnufz4dSqYSZmRm5aLj3uBedTidmZ2fR2tqKpqYm2asU23A6ncJxzeVykhkRkroR+0DFHLWi\nTMCUC8SNZzAY0NnZicuXL0sqyPY6SqsFAgEMDAwI+/7EiRPYsGEDZmdn8dRTT0lKGYvF8Pjjj2Nw\ncBA+n08EfUlFUOe2UNqJYTcXaj1QhdQ0S8V8ueb8vF1dXejs7ITVakUsFkNTUxM8Hg/i8bhoH1Yq\nFbS0tEhb44ULFxCNRmG1WjE4OCjdDXNzczh9+jQqlYqQqllQqNVqCAaD4qTpGNViznowptZqOyL3\nKrAiPqzX6+H3+xEMBmEymbB582YEg0FoNBqEw2E4HA68/PLLAIAzZ87g+vXr8homkwltbW3w+Xzi\nIHS6ZZUhck+3b98Oo9Eo1V6TySRzjZhKApCUVNUKbXRbLRBCDiWxXZL5eb67u7tx7do1JBIJRCIR\nuZxnZ2dx5coVaDQaEfY1mUx48803sWXLFvT29iKVSkm0bjQa0dnZCZPJBIfDAb/fL/uyra2tTuqN\nUS4bNzh24kZszQrnVPfh3+mwWPEql8vYvHkzCoUCfvKTnwgtyGKxIJ1Ow+/3i2MNBoPC1v/5z3+O\n8fFxaLVaXL58Gbfddpvoxf3oRz/CoUOHcP36dczNzUlUxfDZbDbLLUJsh5Ev1XAaPeqp1WpyMQAr\nkbtK3OVtOjExgYWFBUxPT0vXzMzMDLq6urC4uAir1Yr9+/ejvb0dU1NTuHjxIubn53HffffhwoUL\n0tlgMpnw9NNPw+Vy4dVXXxVFFxWfCwaDohZP56lGZ2xtbGRToxs1/SZrg1/PZrOYnp5GNBpFf38/\n5ufn8dprr0lh5cyZM7jjjjtgMpkwNzcnODwd5N133y09yzrdsuQXxXeZ1ZCOVKlUMDo6KviZuqbE\nhhkxNboxSGKKTTyYFyqzl1gshlwuh0996lOYmpqC2WxGc3Mzenp6UCqV8PTTT2Pz5s3YsGEDfD4f\nXC4XZmZmRGGptbUVr7zyikSHhJxGRkZEeWxkZEQul/b2dgk4eDGqs3w+NMI5U0C+AHFEqqcQ3Ha5\nXDh//jy2b9+OdDqNQCAg/29sbAxLS0vo6OiAyWQS3OxLX/oS7rrrLoRCIXziE5+oi2L+7M/+DG1t\nbXIzq3wzvV4Ph8MBABLJEuMh3sl0vNFNJe6rfzYajZiZmUEoFMLAwACGhoZQKpUwMjKCrq4uSc/C\n4TBcLheCwSD2798v/bNtbW0IhUJoa2uDzWYTXUm3242Ojg44HA4hsJOPppKzgRVh4Wp1RfyWmUCj\nT2KkQ2RWo3Y9MfvgmJFisYiBgQGMjIzg/vvvh8fjgdPpRCQSwe7du7Fnzx5Eo1Hhs7JY09fXhyee\neEJoW1evXsWePXvQ2dkpnSStra2YnJyUwkI8Hkcmk6krhvB9sUjZ6GsLrHCAV4ukkHJFLPKhhx5C\nPp/Hf/7nf4p+wPj4OKLRKADg8OHDKJfL2LdvH0KhEC5fvoyjR4/i2LFjuOWWW/AP//APaG5uRjab\nhd/vxyuvvAKv1ytdfTqdDoFAoC4rYnGJfoAByFrZGh8oolytkcdoIxAIYHFxEU8++aT0b1KxZmZm\nBhqNRsYyUBn7tddeg9vtRiKRQHNzMwDIrcvNQrC9tbVVDjNvBHVjAZCOCILmPp8P27Zta/g0kaRk\ntc+Xn53thZcuXcIzzzyDvr4+vPDCC8jn8+jo6KgbZ2AymRAMBpFIJHDhwgWpWnd3d0tvN7G4dDqN\nLVu24MSJE5JiptNp6dG3WCx1RGLeyNwHPp8PnZ2dDQ9r0BHxslbbQsn5Gx4eRiqVwqOPPoqxsTE0\nNzcLAd1oNKKtrQ3FYhFnzpyRS6K5uRnhcBhGoxHDw8MiA8bAYWZmBvPz84jFYlJYsNvtMhmTItWq\nSpFKr6vVauJE1oOxsMeLnqyVS5cuYWBgAOfOnUOtVsPZs2cBLA8qdLvdAheFw2F4PB7YbDYMDg5C\no9Fgx44d2LlzJ1KpFA4dOiSvkU6nceDAAZw6dQrNzc0ol8tCuWKRZ3VDhAoLsLB0o1HlmlsYHqKW\nVgAAIABJREFUVR6daiaTSfAGhs25XA5arRbXrl3D5OQkdu/eLeo+TqcTzz33HNrb21EoFKSSmkql\nZLPRQSYSCfl3ticx6nG5XHJr0JgiOhwOaZNqdJqFKsoArIwCrlQqSCQSqFar0vIGAIFAAG63G0eO\nHEGxWEQoFEKxWMT27dtxyy234IknnhAB5FKpBIvFIlE60yNGVkxR8vm8RDhmsxmdnZ3SjcJIElgZ\ni+B2u+uUwBvV+P54sXJ/VKtVUXnv6upCf38/Tp48icnJSemZv+WWWxCPx2V/pdNpDA4OorOzU3Cv\nanV5dATbcqvVKnw+H7LZLCqVCpLJpBzSVCqFZDIp84oIVxHPY3GCzI1Gv+CB+rRW3QuM6trb22E0\nGjE0NIT5+Xls3rxZxJ+9Xq8UWDZu3Ij+/n68/fbbIiZCJTC32y3MAu5hs9mMrVu3CkuDa63VaqUZ\nA1gp2qljPlpbW+FyuW74M655FARxQQLODG/5RjkljQeR5OgtW7ZIFaqvrw8nTpwQnJGLykFi2WwW\niURCKBQMmUmupgZlrVaTG4kOXFVf93g8KBaLdZJwjWpcW7XPl2lLV1eXAM9cM7/fj2QyiVAoJBCG\n1WpFW1sbLly4IAULFtLYdsjqqsokoGQVaVwExFtbW9+l2UccOBQKCZi+HhwlAXz1wPDvTU1NiMfj\nmJubQ61WQ3t7uzgpi8WCZ555RqLK73znOzh48KB01fBnASsFuHK5LNMF+fomkwlWq1WoRPPz85Jt\nEVfns2dkr3L/1oNRh4Dnj9CGy+XC+Pg4jEYjvF4vWlpapNAaiURgsVjqBKLn5uYArIyO4C/CI7Va\nTSYiEG5jZEgcv729XS4g+gY6c0KBfK83Yh/Ie9CzMy1LJpOi/BGJROroLdRObG5uxtGjRxEIBDA0\nNISRkRGZb0NqEVVvWKUiHmkymVCtrgwd0mq1IkTATorVhRyn0ykbdL20gpGbyPXj4SEtJR6PS+88\nLyCj0Yjdu3dLh0g6ncaxY8dkFg7XldQTCpSw+kq+H1v2eNsCqJtlTcoHKUOU6m90J6maOvpXq9Wi\nu7sb6XQafX19ggVOTk4Ktp3L5ZDP5/HII4/g0KFDyGQy+PKXvywjgekYmdLTtFotWlpaAECgE+oP\nRCIRdHV1iUaoSn5mB9Ts7Kxcauuh6l2r1eTi4OWs0SxPnGxra8P09DQ8Ho8QxxcWFpDNZtHV1YX9\n+/cjHo8jl8uht7cXL7zwwrtU4RkoVavLWpTEHPl61PjM5/NoaWmRTj2m4cDKxNh8Pi9BlsrTfj9b\ns6Ok02O4zciF5ORsNgu73Y5UKoWBgQEYjUYkEgnMzs7ivvvukxDc5/MJQZfYDG8Ft9stG5c9sJz7\ny1srm82+5ywctd+cP7PRo0lg5fYkXUSlVeRyObjdbrzzzjswmUzIZDIiqT88PIzXXnsNt912Gzwe\nD65evSrFCRbeePkwug4EAmhpaZHDm8vlEIlE6gSUu7q6RNFFxc4AoKOjQzIJQiSNbmy9VXm+V65c\ngd1ux9TUFCYmJjA4OIj77rsP2WxW8LPr16+js7MT//Ef/4E333wTZrNZok1VpZzR09DQEAwGgygL\nlctl9PT0IB6Pyyjl5uZmKYTxFyNTzspZ3Sfd6KZepvxMHNlgsVhw8uRJpNNplEol9PT0YGFhAZFI\nBMeOHUNLSwu8Xi8mJiaEvcI1YeaTy+Uky+zv74fL5ZJLOxwOyyVE7ikDJvUSqtVqCIVC0nK9ln27\n5mIO6QwqQEq8gFHm5OQkAMgUOZPJhIcffhiFQgGDg4Ow2Wx1E/FU4nqlUhEMcteuXdI2lkgkBCC2\n2WwYGxtDa2urOBOmqjTy5ngTrYfDXKvV5LYDlh8uCwFU8SG5fHFxERs2bIDT6cTBgwclchwaGgIA\nGarENSEOxAyAz6VWq0nRggc7n88jEAjUSWcVi0VJXcgB5Jquh7W1Wq1IpVIAIBAN2+EymQw2bNiA\nBx98EJcuXYJGs6wExCaJixcv4q233sLtt98OYEV/kXuPl5LNZsPWrVsxMTGB48ePC67L/Tc1NYVC\noYDh4WHpbwZWaHbsWGOayPfa6MbAhJkf9+/09DT0ej1yuZx0OhHL3bt3rzBgent7EQgEcPHiRQCQ\noWKqSplOp5Oe7uPHj4tzJJRhMpmkfsG9S5+gMjWYCajUsBuxNYtiACszvfnL7XYjmUwiGo3CYrGg\nra1NBAG6urpw5513YmRkBPF4XDZYrVZDJBKRSIoOl6pCCwsLOHPmDMrlMjo6OmQWCTEmhtc0dpUw\njVexs/VwkNWURX2AVqtVohgC1Jxw6XQ6sWfPHhG/uHLlirQlRqNRtLS0SKRP0QyHw4FEIoHx8XGk\nUimEQiFJXXh4U6mU6HoyVWUFkZfhak3SRja+RxZL+Dk6Oztx4cIFdHZ2wuv1IpVKIRAIoFar4ciR\nI6jVahgZGcHBgwexe/duabejYyAWT4hIJY4TGgmFQqLLqtVqsX37dpH/Us8Ri4+xWEzSStKZGt3U\nfaKeOWKVLpcLqVQKs7Oz2Lp1K3bt2oUnn3wSmUwGn//854VkbrPZkEgk0N7eXtdeSofH2U5d/7tL\nil1P6mx2ine/l3KRSmsE1ha1r7k/iguhHqBkMimcvPHxcRHa1Ov18Pl8ePPNNxEKhTA/Py+RHjct\nh4txNsnAwAASiQR27NghadGRI0fEOba2tspsaf4cLgKrsuR78v2q0nCNbsQVGaVTYaWzsxOzs7Ow\n2+2Ym5uT8Q+cq02wnEB3pVJBPB6XFkQAUpHV6/VSBT937pxEPZyBzu4FFkBUmTy2+tHWw9pqNBq5\nIFSF8tOnTyOTycBsNsPlcsHn82FiYgIOhwN79uxBoVDAtWvXcOTIEdx2223i1Dwej0AWrGSbzWY4\nHA4kk0loNBqZ7dTe3i6pPiOrRCIhbbw84KrALZ/Rekq/yYghbET/YLfbJcgJBAIiXNzR0YE/+IM/\nwK9+9SvYbDb5/NQc4NA7AMhkMrh69Sq6u7tx5coVKQapEBD1LCmio/adA5AIfTUu+aFFlExbeJgZ\nwSUSCeEmVatV9PX1IRaL4Z133sG+ffvwy1/+UvAD4jnsayXAWigUBGegxtyVK1ekMqkKGrxXOs0b\nSI021fC9kY2UHFUhmtzRbDYr+E1HRwe8Xi8ymYykjP/zP/8jfeBsU+S8m0wmI/Sp7u5u6HQ66eCZ\nn59HIpEQBSJWzklbYcTDSJcRAiuGTOUb3UjNWT3fhnoCjPCGhoZEhWpkZEQiSgoZAxCHmM1mYTab\nZWRBpVJBKpWSFHxiYgIDAwMyWoJROKeQkifMdaUsHiNPVcSh0Y0RH8VqeKEYjUbMz89LIMNCosvl\ngt1ux+nTp9He3i6DAXlWeYFwaqvZbMb27dtRKBRw9913y8/mKA0WJ1kFV4Mk1Zhyc8+qjvT9bM27\nnAeK+CJvjnA4jI6ODulvPX78ODZt2oSenh489dRTMsqBkQoxIoPBgPPnz2Nubk5SjqamJuRyOfk7\nHYgqD0bMgR+WUZH6wPg9xDsa3dQqqIrbjo6OCuZYKBRkRKfT6cSpU6ckPVMpUXq9XlRZ5ubmpHUL\nWOYCLi0t4cqVK0K7AOoHnAH1clS8HDlvnNEl089GNrXFDlhRjG9paREeJQChnBw9ehTRaBSnTp3C\nvn37pKqtXkakpvHyJ65WLpcxMTEh5yQWiwEAUqmUtN4yQqejXM1tBVB3mNeDeTweYaXQ+ZOzSEqa\n0WhER0cHjh49iqGhIWzduhXDw8MAIFF3rVaTwq3L5ZILSm0SKJVKGB0dFRhFnYLA4prKbuA68z3Q\nZ3xoxRwAQg4HVgYn7dy5E/39/YhGo3Jz7N27F2fOnMHi4iLuuOMOOZCMJvl7sVjEwYMHceDAAZkI\nyFTc6/Xi0KFD79K0q9Vqohhy7do1TE9PCxGblTA6x1wuhx/+8Iei1dioxl5v0nV4EZAjyQceDocR\nDAYRiURQKBRw6dIleL1ekZvKZrMS5Tc1NcHv92P//v3YsGEDPB6PjAzu6enBY489VreBmpqaZE5R\nNBqVajBFCJaWltDW1lY3/vZnP/uZbNRGNhXHApYvldnZWfT19UkEQprQ/v3761LASqUCl8sl36vT\nLQ8QW1hYkGgVWJmvrtVqsXPnTkmvLRaLVIUJf1CohVkaaV/qhb7Ww/y7MnaVqW2uhNaoZq7RaOD1\nemEwGNDe3o62tjZ89rOfxebNm+W5cB14kR07dgwGgwFerxcejwddXV0yPZNiGRTTUJ2j2hhBX0Ka\nlirC7PF4PpzOHKvVCpfLhampKSEyazQaRCIRXL16VSK+cDgsOBhJ0+zUYQTCxTCZTJidncXY2Jgo\ndCeTSeTzeaTTaSQSCfyv//W/8Cd/8ifI5XKCaep0OrS1taG/vx82mw1TU1N45513MDs7i0gkIpt6\ncHBQxo42srEKPTc3JwUC4mcUsOBah8NhUXHetWuXjB9gdKOmwxqNBiMjI7h27Rqi0SgSiQRyuRzS\n6TQKhQLuvPNO/Pmf/zni8ThqtZoUjVSSdCqVwuTkJMLhsFxyS0tLOH/+PDKZjFycjWo8iG63W/7N\n7/fjkUceEe1JYFmVaXJyEkNDQ6jVavj4xz+Of/zHf5TOD5WTBwBer1cmNpLz19LSApfLhdHRUTz/\n/PPSK84sKJfLCTxltVrh8XhE1IVKUMzKAoEAdu/e/X9/wdZojOJUXJUq5uFwGOFwGOVyWbrHGCR9\n4QtfwGuvvSbpteroarUaOjs7Ua1WZd8yGDKbzWhra8M///M/49577xU4iAXJWCyGSCRSx7fmmWJm\n5HA48OKLLwod6/1szZ05+/btg9lsFlC0tbUVc3NzsNvtGB0dRTKZRDAYxPXr17GwsCCl+6WlJfT1\n9dVxHCkPxv5N3uwmk0nS72w2i2eeeQbxeBxPPfWU0GXUHnC73Y5NmzZh586dCIVC6Ovrg06nw8jI\nCBYXF9HR0dHwWBqjEPLriOVQb3NqagqZTAajo6My24Y9yRQRYARqs9nQ1NSElpYWSXvoQNVOnXw+\nj3379sFms+Gb3/ymgO6srDOa5XwTp9OJlpYWobqcPHlSsL9GtnK5jMHBQRHa1el08Hq9+OEPfyhF\nAK1Wi4mJCWzYsAF+vx9GoxEnT55EIBDA9PS0ZD/k6XKcBgtbFLBg5K3RaLBlyxacOXMG3/3ud2Ew\nGJBMJhGLxUQ+rVAoSGsuq+m1Wg3j4+PYu3cvHA4HZmZmfserd2NWqVREcQqAYN9arRYHDhxAIpHA\nM888g8XFRYkItVot/H4/xsbGxAewyYHsFsJNWq1WLmSKbTz99NPI5XL4/ve/D2AlAqc+bT6fx+zs\nLIaGhmQ8Nr//mWeekUr6jdiavAdL8ozWtFotYrGYgLC8PSmnRsxyaGgIHo9H1MdpVOPO5/PC6WNq\nza9xAcrlMp577jl861vfQjQalU2pSpDl83mh0xAfIYG70bGeSqWC5uZmwanoLNm1UKlUpLshEolI\n4WxhYUE4qSoXtVgswuFwiPPkxuOtT04m185ut+OTn/wk5ufn64QEAEiF3Ol0wm63AwAuXLgAq9W6\nLsYVGAwGmdvEaEbliE5OTqKnpwfFYhFut1taODs7O7Fnzx6cPHmyrgGAGQoxMgB1rbhMO7k2W7du\nlSxLFbdWNVUplVcul2Gz2epUjRrdmMaq+DYl6CqVCs6cOQOv1ys48bVr16DVamGz2WTyKvcweboe\nj0f2FQMrwndkLjB6/clPfoKTJ08KY4NROUedtLa2Slah1+tx7do1Eef5UIo5VE3hGwUAh8MBh8Mh\n4TLD3MnJSUxMTMBqtSKbzcLn8+HChQt12pFarVYa3xklMjXnB+NBJkDMgfRUOCcmwsr5wMAA9Ho9\nFhcXJWVdD1w0OjKmFxrNstISK4KVSgV+vx9ms1kqrORTMuIjjsuHzw3IXnceSq6p2hpWrVblIlNn\ntvCScjgc2Lx5M/R6PSKRCOLx+LuGuTWqkdit7pkzZ86gpaUFBoMBDocDqVQK4XAYhUIBJpNJZO38\nfj/Gx8cFk2XzAy8X7lfiayzUcM/Scfz6179GS0uLvL5aHTabzaKgE4/HsWPHDrnoG71QBizvVbvd\nLutAZaozZ84IPsvgJRgMoqWlBalUSpT69+/fj0KhIJc/MW92PpFhw39TJ5PSYX7sYx9DKpVCKpUS\n2hbXOZ/PS3NKU1MT5ubm3gWjvJ+tOR8llYEAbjqdRiwWQ61Wg9/vx9zcnPDw3G43Lly4gL6+PoyM\njGD79u1wOp11gCwPMm9PDl7iL24UOoDp6WnceeeddQulVrri8bioiKj8rEaPKEmW5YOjiC+7ZXjI\nGcVXq1UcO3YMPp9PWu/8fr+sr8/nq1NVYucCK7OMxOk0AYiiCiuDKj+VoDwAXLt2TW73ter6/S6M\n60bHT+yKw+g6Ojpw5swZdHV1iW6B1+uFxWLB4OAgPvWpTyGTycDj8cBkMsHv90tbndo/TsoR9zGw\nUnG/4447sGnTJtnTapYTCAQwOztbVywidroeLiJiknRO7FQivs1L2OFwYHR0VCJ6imV0dXXB7XbD\n4XAgl8sJPAdAuqeIbfKc81zw2W7YsAGtra0yEoZWLpfR3Nws4yWuXbsGnU4Hi8UiNLgbsTXLrBUK\nBZk1QYdIx6lyK1ld3blzJxKJBMxmM8LhMKLRKBYXF6UVsVAoSAWL5X9uamAlimWardFosHfvXnk/\nxEoNBgPm5+cRiUSQz+elIESJq0bfcIxAVHUUAHC5XJLGhcNh1GrL0nJMHZaWlpBMJgVzIUevra0N\nS0tL8Pl8dcOrWEhTVYTUFPxzn/ucDCBjhNDU1IR0Oo2FhQVkMhnMz89LVMsUtpFN7aSh4+IsnGKx\niLGxMfT39yOVSglzgHjwxMQEbDYbbDYbxsfHZeZ3JBKRCra6BoSM1FSdl81vfvObur1IB0OGQjgc\nxsaNG+X/kMbV6MZLHliRC+RlxIIgpQ51Op3wVOfn5zE3N4fXX39dii61Wg1tbW1oamqSuU18ZuqI\nEvoC7tulpSWMjY3J3xmMkTdMiFDFUTlX/EZszRElezr5AUjwpHAmeZLhcBg6nQ5utxsnTpzA1NSU\niPfabDYRxVBxRG4uLjgfAtM/WiaTwalTp3D8+HGcO3cOiUQCFosFLpdLullisRiKxeK6UQ9aWlqS\nFkVGaSyakTGQzWalbatarWJgYAAXL15Ed3c3otGo8Puam5thNpsxOjoqwiG87VU+Jf/OlJ0FpPHx\ncbzzzju4fPkyIpEIrFYrcrmcFJUWFhaEjL4eyPzAclTJkSAsauVyOYlYiO1Wq1XJig4cOIBHH30U\nExMTyOVygrlns1mkUinJXpgqrm4/VXl9AHDixAmYTCZ4PB55zpVKBVeuXEFPT48MKMvlctJxxeJG\nIxujZMJvwEphtrW1VRgWzPAWFxcRiURw6NAhfPKTnwSw3Bcei8VEXzaVSskaqMVbvoYKM3Hv7tmz\nB9Xqst6twWCQAtHVq1dht9uRzWaFhcNhfB8KPYjgNauHqmACMQTiM+x6ePHFF/HJT35SWu5Ii7BY\nLMhms7K5+DW1A0TFzyh9xULFrbfeil27dsHn8+HEiRN46aWXsHPnTpjNZly8eFG6eWiNHvUYjUYU\nCoV3qZokEgkZu0t1Jl4o09PTaG5uhkajEYYBFbTz+TwcDodEkcCKLiOfmSoYwNvaZDLJzBeTyYSx\nsTGMjIygt7cXZrMZJ0+elJ9FRkOjp9/cY4VCQaLz4eFh5HI5ZDIZSeG4Zm1tbaLERHyLZHPSVA4f\nPox0Oi0RN7DSucZMh9EKDzrFliuVCmw2G0KhENxuN/L5vLSltra2wu12C03rRukrv0sjpEHHxTOc\nyWREWzKXy4kWRDAYRH9/P773ve9Bq9WKziyLMXa7XShFjLj5c9VAYnXtIRAIQKPRYHBwEIlEoo4n\nabFYsLCwINivxWKpoyS9n63ZURIvY0pB8Qm+adJPTCYTXC4Xtm3bhjfeeKNOksrr9SIejyOdTku/\nplo9VeebqL25dJ7FYhGbNm1COp2GyWTCbbfdhs7OTgDAc889J+BxrVYTx9LoUQ8vGj48UlBYvePo\nC/5OikSpVML169dFiKG5uVlG2JLUzM/PVIQRFb8/m80KQZ8ybHq9Xm7kpaUltLe344UXXkA0GoVG\noxGu5XqANfL5PCwWi2ilEuLw+/2S9bCgSNijXC5jcXER8XhcKtxms1loQaRSEaLgWqhNERw7yzVi\nV0+5XBbHcfjwYdRqNVy5cgUHDhxAsVhEMpmUWVI3GvH8Lm1paUn2LaM99tWzZ5uRO6G0sbEx3H77\n7bK/GYB5PB4MDQ2JKrrafqrCbwyguI8JBcZiMdjtdqTTafzmN7+RtlGtVou5uTmk02lRkFfx5fez\nNTlKasKpPclarVaEZG02mxC76bgWFxdhsVjgdDqlsutwOLBlyxb09PRI+kFsh7eHSq/ga1WrVYl8\nHA6HCKiycv7EE08gFAoJLSYej4uCTqNHlCxAUXCBh5kbi5ulUCgIQVqj0QgmyRu0u7tbquBms7lO\nao0/l73ajCzVZ7C0tISBgQHkcjnY7XaJcN566y2kUinYbDYYDAZEIhFJpRp9bUk2jsfjgicmk0lx\nllarFdVqFa2trVhYWMDs7CxmZmawZ88ewbF0Oh0ikQj8fr8UDdXPTroP03dV4ILOAABefvllcaBW\nqxUXL17EwsKCTHDM5/OC72UymTqSfKMasyG1aEYaj1owcblcmJiYkDlD0WhUMHJ+/eLFi6JHS8hN\nxctJ72IQtRo6YkDQ1NSEUCgktMVjx46hVqvB5/NJwYiveyO2JpIWyZyBQAAjIyOoVlfGQKhKPpRY\nSyQS8Hq9KBQK8Pv9uHbtGjo6OgSn+H8T1SXnCVgJ7bkx1R5vcjubmpqwd+9eZDIZiQqIg87Ozq7l\nY/5OjBglB6xxbSl95vF4pAOH2FooFEImk0FbWxsWFxdhMBiQTqdF5k5NtQHU3cq87NRUnD3L7LkF\nIMRdblRK+HMI1KVLlxoe/9Vql0Vkp6amxNERWySRmfSe9vZ29Pf3I5vNylgMZjUsDKoHkn8mOZr7\nklEkf1cnFFLQNpVKIZvNiso3C6M+nw/z8/PrYgIjUE83I8uE55LB0dTUlKjxBwIBjI+Po7OzE+fO\nnZNCY6VSQX9/v6wrI3OVX82Mlf/GvcvovqurC7Ozs9L9lE6ncfHiRaHIZbNZGQy3lplEmrVsco1G\nEwYw8QHX83dtoVqt5v9dv4n/k91c2w/P1vnaAjfX98O0G1rbNTnKm3bTbtpN+/+jNXYD9E27aTft\npjWA3XSUN+2m3bSb9j5201HetJt2027a+9hNR3nTbtpNu2nvYzcd5U27aTftpr2P3XSUN+2m3bSb\n9j5201HetJt2027a+9iaOnPsdnvN5/MBgLRosXUJWBEFYD/t+9l79QmrHQ3q39Xxnqu/xnaneDxe\np6WoNugvLCygWCw2bFOyxWKprR5gRdFRdX3fa23fb61XS3upf2Z7njpwXjUKA7Mv/L1eK51OI5fL\nNezams3mmsPhkFY4dnVYLJY6ma3VOpGq8d/eax1XmyrMy7bQ9/q/FHihCInaM66+p9nZ2UgjE85N\nJlPNbrdLuzE741TRCVU0Q13jG9UKWL3ulGxj19h7/YxCoSDdVnxfqtVqyzOiUqnU+76BNTlKn8+H\nb3zjGzI0rFAoYHx8XNqH2JjOkQPqZDR+OH5QdQEp06YqEqlS/dlsFr29vdJUD+BdfbZbt26V0ayq\nuIFer8dPf/rThp8U6HK58MUvfhHFYlHk1rRaLa5cuSJScWz1Yi8tWxJXm9oWWi6XRRZv9eZkL225\nXJYJeeqo3Gq1ikwmg0AggP7+fpnNzu/VarXo6urCX//1X/9fW6cPYk6nE5/+9KdFIg1YXu/t27dL\n369er5fRqlwrdb0A1GlPqo6Q66YKZFCYOpFIYGZmRp6T2uar0+lw7tw55PN5fPazn0U2m5X/RxWu\ne++9F3v37m3orhebzYaPf/zjMo5Fp9Ohs7MT27ZtQyqVkj53CmUUCgVZU64fHZ16kVWrVWmZpWNk\nKy2FfycnJ2UuO7/GdsWmpiY8//zz0Gg0+MpXviLaEPl8HuVyGZs2bcK3vvWtG/qMax4uRmfF/koq\nzPBDqHLw3GyqxiT7bCl+sVoR5L0OMxWO1Z+p3tq12vIs4E2bNomiMqOkn/70p+jq6mp4hRtg+TNz\n0BV7r9UDRiUZtfd99YFm/ysjGQqe0tR1UC8v9bmph51D4yYmJnDo0KE6PcDe3l4cP3684Xu9+f7Y\n082Dq86jWX3h8PPzd7V3nirm1F5VBRv4M/ns1P+jrj3Pwu7du6HT6fDKK6/Ia1D44cEHH1wXM3MA\niAPkbHT18uD+4udnMEXhF7VXnoIu1Bxgtsg1ZM8315v98OwBpx4Bo/W77roL5XIZR48elYsNAHp6\nerBr164b3rtrxigpcqo2qVNQE0CdBh8XCFhRrlEdnXqQVYl8NdVUI0gqB1GhhT+jUqng+vXraGpq\nwv333y9qOz/72c/Q3d29LoR7gZVog9E1UH9hqLcugLrInOvBS0LdpKvXVD2wagSpqkmrjsNms2Fh\nYUF+RqlUwoYNG/Dqq6/eMMzyuzZVgYkHThWF5f5QpehUNRxVqZyfl19Txz6oz4MCERx4pf5sVaiE\nI6DpZEqlEj72sY/JfKr1YJwNz3Osri0vGq6xOoQMWBmXQShkdXbK4IzPjpcX9VNtNpvI/vFi4ut7\nPB40NTXh9ddfh9vtRlNTE7q7u3H77bevaW3XrB4Uj8flga6eR8PbWpWYonAsDyxH0XK+CG8NrVaL\nixcvygdUNxoPOp2x+vPVCGhkZASdnZ0ol8v41a9+hd7e3rqBXY1stVoNVqsVkUhE0ltuMjo3dZY6\nb05uQFWKSqNZHvZOdWyuEzUUV8Mf/D46AVW5heM9FhcX4ff7cdttt2Fubg6vvfba/xHA2B9vAAAg\nAElEQVQbakSrVqtobm5GPB6Xz6rCPFxTVYSYn4+QRzKZBADMzc0hmUzKzB2PxwOfz/cuqTEVs6eW\nq3rAGVXt2LEDr7/+Ot544w3s3r0bhw8flsh9PUSUfJ+cEMpLd7UKvnqR8zImLLS0tCRCyvl8HgsL\nCygUCrDb7bj11lslGmT0uRqzZJZAJ8uIv1wu48EHH8TTTz+NN954Ax/96Eexa9cumcFzo/t3zTNz\nxsbGRCOPH1xVyKZmpDrGkzJolLViGq1KMy0tLWHbtm0IhULI5XJ1X+cBVjHM1ZJWnClDbbxAICCL\ny6FYjWy8dRcWFuoiFjXVVYdWrdY8pKK8GnWqh83lcqGnp6duRCoxSjXipONQixAUVR0dHUVXVxfe\nfPNNSYcaHfsFViTrXC4XlpaWJJJg9MF9rM5sAlYOHdeYYsb87FSfv3r1qmDJarrOkcpcX76W2Wyu\ni7aWlpbQ3d2NZ555Bvfcc09dkW29XERtbW1yMdNBcrokzyHXUF0fnmV1DLZ60edyOfziF7/A0aNH\nYbVaZY0tFovM2FELknq9HlarVSTr+CsQCOD06dPYvXs38vm8YKM3amtylOVyGdFoVCI9VaSUkYoq\n6MuNdv36dSQSCZkEqNVq64ZacXGIWfAG4UEGIOlLtVpFOBwWbEKj0ch8DoPBgJdeegmTk5OwWq1S\nGFHfV6OaVquViIUHSMVzgRXdP+ojajQatLa2olZbVnJnUYwFCa6pms5v2bIFDodDtBWpJA9AniGd\nH1MeOmej0Ygf/ehHEsmWy2VJcxrdduzYURdFAxBVeK4RsTV1zTh9kt9ntVphs9kEO+PzWVhYwIkT\nJ5BMJmXWCxXj6QwsFovsW8JO/F2v1+PHP/5xXWBgNBrrhmE1qmk0Gmzbtg3AipNcDWkw06TjqlQq\ncDqd8Hg8dWe1vb0ddrtd5sfT18RiMfzgBz8QFXT1YuP54Dhgfh/PSrlcxoEDB/CDH/xA5nqp+qE3\nYmtylOp8XLXCB9QXG9QCg1arhd/vFxFPNZ3hTQ8AqVQKZrMZpMjcfvvtuHTp0rtSbK1Wi0uXLtU9\nJDXievLJJ+FyuWQQVy6Xg8PhEEHcRjWNRoPx8XEAK9HialyS66tGItPT03JhcR14U7ICSdyrWCyi\nWCyipaUFt99+uyhFAyvULovFgvHx8TpRWkZVW7ZswdTUlMwQp4J6o0eVFMNVDwUPa7FYrMPP1YKi\n3++XfaWOoTUajRJ1c71rtRpisRiOHTuGq1evyggEAJIZmc1mUfUHVigz5XIZe/bsQTAYlOiKg/fW\nQ+rd1NQEh8NRFxGuxq6p8q7O1OLZZgGHavssSKbTaYlINZplRfK/+Zu/kZScX+MadnV1IRgMSpAA\nQH7WrbfeCofDIVmBRrNMvbtRWG7NM3N4KOt+iJIi8u+skBMTYnWc/9disch830qlgkAggKWlJWQy\nGej1euRyOXz5y1/G1NSUHFiVkjE9PV0XrhsMBvzXf/0Xdu/eLbdJPp+H2WzGvn375PUb1agIrVag\nV19EjD64/haLpe6y4SydarUq0WWpVJJh85wFs7S0BLvdDq/XW5di12o1uN1uzM7OIpFI1L2/PXv2\n4NlnnxUnYTKZkMlkUK1WZdJjoxodOoB3YYerpyZyzVVmASuuvBTozDiLJZVKyUwonU6HkydPimNU\nA4p8Po+tW7fC5XIBWKmsJxIJ3Hrrrcjn8zAajeK88/k8nn/++d/Zut2oUZ1cjSLVSI/7lhGlGgjQ\nnxACisfj4hc4LCyZTMoedblc+OY3v4menh55Rsw8c7kcHnjgAVgsFoksi8UiQqEQtm7dWsfQ0Wg0\nuHDhAjKZzA19xjXTg9RwWuU2Aai7jSuVijgqpmhWqxVWqxV6vR7pdFoqZRykxA2oDhR77LHHZPQt\nI6ympiYMDw9LGA0Ak5OT8Pl8aGpqEvzBbDajr68P8/PzDX8zqxSf96L8cPMBKw6RzkylTnGtGe3Y\nbDbY7fZ3Yb2zs7MYGBiQ9QZWij4+nw/j4+OyiU0mE1555RWhhHH2dbFYxNatW9+Ty9lIRmdFbiqw\nktKpBTDu33Q6LbNyGBzk83lJG00mE4xGIyKRCEqlEqxWqzwbYmRPPvmkYGXMAIDlZ7dp0yZ5L9PT\n0/jUpz4l0RUnD9psNpw+fbrh1xaAFBrpDwDUUXkYcav0QOKZhOEY2ZnNZng8HuRyOYyNjSGTycBu\nt9fxKf1+P/74j/8YO3fuFKfH351OJ7Zs2SKXk8FgwL333iuRJdc9FothfHz8htd3zcUcGjca0wp+\nnYvW19eH4eFhuWX4IdPptFRQd+3ahVwuJ4ugRkt8rWKxiGAwKA9ETbW7u7uxZcsW3HnnnXjyySfh\n8XhkxkxTUxMCgQC8Xq/M6Wl0UzmTXF+1Gk2Lx+PYuHGj/H+C2MTAEomEzNJhNVGtEvL7YrEYHnvs\nMYkM1Qrt3NycjFXds2cPFhYWYLPZJDKrVCoIhUKS3jey8ZJRaVUqjUXl8m7atEnWi3xGEsHL5TI8\nHg/i8TgmJiZgsVjqZuGshp7+8i//UrpuaEtLS9i/fz8SiQSGhoawfft2eDweKXYAy40dv/rVr6QK\nvJ7svehnaoReq9Ukc2LWx1SY6fTCwgIymQycTqfAPKxi83nZ7XbcfffdcDgckmUyg+ru7kYikcDr\nr7+ORx99FBaLBfl8XvD3QqGAI0eOyJz3G7E1R5Rq4USl7qi3CbCcqpEXxXSQv7Zv3w6tVov5+XmZ\nopjNZuXDMPVghHP33Xfj6tWr8nqJRALf+c535P8uLCzITZHP52Vw1N133w1gGf9cD7aae6ryU1WK\nxaVLlxCNRmU9+CuXy2F+fh4tLS0IBAKycdQCAjcxHcXJkyexZ8+eOtjk7Nmz+KM/+iOZQ7179+66\nCKGpqQnhcBh+v18mQjayqcUsri0HqfHr3Ltnz56VYkytVkOhUJBKaygUwvnz5yWyZpTY1NRUh2Gy\notvd3Y3JyUkAkMKOx+PB1772NQwNDWF+fh6vvvqqXFIsPJ4+fRqpVAq5XK7hLyEaoQdCOwAk22Na\nbTAYZHY6MV6j0SgV7mQyiaGhIYTDYdnTjBJZgOHamkwmuN1ugX04BfbFF1/E17/+dVy8eBG5XA5/\n//d/X4dXGgwGHDt2DJFIBOl0+sOhBwHv3XMJrNwcwHK0EYlEYLfbxeEZDAakUim43W5p6Vo90pJk\ndpPJhLm5OYmmUqkU7r33XmSzWWg0Gvzd3/2dcC6LxSJMJhOmp6froptHHnkEly9fBvDurotGNDVS\nBlboQiqhmRXFzZs3C1+NjpU4Wmdnp1wWAGSNWHApl8uwWq1IpVKSqodCIcF1Tp06hb/4i7+owzQ9\nHo9EThqNBufPn8fDDz8szqTRTaVKqVmQCllwr42Ojgp9R6fTwWKxoFAowO1245133oHP55M9C0Bm\nfqu0Ih5eg8GAV155BYVCAQ6HA7lcDl/96lcxNTUlB9/tdstZKpfLaG9vx9mzZ5FOp+vOVCObCrmp\nbBUyMAh7VSoVvP3223LhkhGg0WiQy+WQzWbR1tZWNwmUs7eJg7pcLrS2tkKv18PpdOLrX/+6rPeF\nCxfw7//+7+Joq9UqQqFQXRSbyWQwODiIXC4nWe6N2JqBOx5oFXdQ+ZSsKg0NDQkuwMVsb2+v4/ox\ncmQFknQMknnT6bR0ARUKBYRCIQDLKSPbpSgaQfJ6qVRCe3s7NBoN3G63fH09HGgWF3hwePC4UXig\n77zzTszPz0vKrNPpkMlk0NzcXEd6ViNT0iF0Oh1isZgUGsrlMiKRCKxWKwqFArZv3y7ryELH6dOn\nJbotFArYsWMHUqmUzE9fT6aSoVen5EajEdls9l1O1eVyYXp6Gm63G7FYTJwi8XBSgTirm5c91z0e\njyMcDsuoVLbtVSoV9Pb2olKpSBPGs88+C2CFz7oejGeLWYq6fmrKbbPZMD4+jlgsBpfLJRljsViU\nwGpychLValUKlbVaTfiWHR0dSCaTkqZrNBp4vV4cP35cOL2MYjm6+eGHH0axWITVakU6ncbLL78s\nwdtafMKaI0pGL6t7glen3tFoFG1tbVJU8Pv90odNx8oqX7lcFqcZjUbl7/9blQYmkwl2u10OMDfW\nwsICotEoTp8+jd///d+XQsY999yDdDotnM3VuEmjGquAKtZFU0VA/H4/dDodnE4n8vk8kskk7Ha7\nXAqMmDKZjESWjJ5UwYdSqYRsNiuvS4Udcv6IU377298GsPzsZ2Zm0NPTA7vdLoeh0U3do2pLLHEy\ntZsjlUohHA4jGo1KIWtiYgIulwuFQgEWiwVGo1GoJmymcLlcMJlMgutarVaJdJLJpOBh5Eb6/X4s\nLi7i937v9+R9jo6OYmFh4V2YcqObGvmqxVy1JZZVcKfTifn5eRQKBaGtFQoFWK1Wofi43W4hlJMy\nyAq4yWSC2WyGzWYT/jAA5HK5usr6xo0bMTAwgM2bN0v3z+joKGZnZ9+leHZDn3Gti0InSQekcsXo\nAEkcJf7Y1taGubk5iVAYAWm1WuRyOSwsLNSRnguFgkSezc3NMBqNyGQyWFpagtVqlYLD/Pw8JiYm\n8O1vf1tuHo/Hg2vXrsmhUNPZRrdKpVLXLaBuMkbe2WwW8XhcWj9LpRICgYBcYMViUdIVPo9UKiXU\nCI1GI3xKpndkIpAtwIzBarXic5/7nDjoUqmEhx9+WBzPeoomVQob13V1psGULpFIiOrN9PQ0fD4f\nUqkUrFYrgBWs3uVywWq1wmKxQKfTwWq1wuPxwOv1wmKxIJ1OI5/Py3ox2urq6kJLSws2btwoTsZo\nNOKtt96qo9Gt5iI2svHSITyk6hWoBUkWx6LRqOxLQnBsSOHP6+7uloIOmxyam5vh8XhgNpvfxXst\nlUqSUTY3N8Pr9UoWVC6XceLEibrgbi32gVNvPsTV1UMAEsllMhkYjUbMzc3BYrHUEXR5WKenp/HQ\nQw/h+eefR7VaRSKRQDqdxje/+U1MTEwIbkkHyb8/9dRTMJlM+NrXvobjx48jn8/DYDDgvvvuQzgc\nrlMkWi+2+uDy70zDWLwi8E+qhMrBZBpIEYhTp07h0KFDUtAqFos4duwYDh8+LCA600e+diAQwNDQ\nEF566aW6VrzR0VFs3rxZsOf1AmkQM+N7VWlW6n7mpeL3+5HJZBCPx9Hc3IxwOAyHw4FCoYBqtQqv\n14v29nZUKhUMDg5i27ZtWFxcRDAYRDqdxuXLlyUzUjtA2ASxe/dueDwe9Pf3ixNhVxafL229OEo6\nRq6j2hqqBkilUglarRaTk5PC3KhWq7DZbACWn1VPTw98Ph/S6TQsFgtCoZA0Spw/fx6nTp0SPisx\neJ7zffv2YcOGDVhYWMBHP/pRCeauXr1aVw+hz/rQijn8MCoZevUNncvl0NHRgXw+D6fTWReB8OsW\niwUjIyOYmprC8PCwYENmsxlf/epXkUql5MBzMbVaLV577TUcOXIEDz30EB577DHs2rULwWBQDnQi\nkahLt1VKUSObSkxWDy+N6+tyuTA+Po62tjbE43GJmNWblfxK4maxWAzlchkzMzN47rnncODAATgc\nDumWApbTarfbjYWFBTz77LO4dOkS/vZv/xanTp0SjHT37t1SKVZbVNeLrW4bpHHtw+Ewmpubkclk\n4HA44PP5kM1m63rEnU4nHA4HSqUSenp6cMcdd0Cj0WBgYAB9fX0YGxtDOBwWJ8m9r9Pp8IlPfAJ/\n+qd/CgA4c+YMtmzZAmA5izp16pTIwK2nLAhY4UyuLowB9ZH80tISEokEZmdnEQwGpbOGXTsGgwEO\nhwPt7e1wuVzweDy4/fbbBSdeXFzEyy+/jFwuV1fA1Ov12Lp1K77yla9gYGAA4XAYO3fuxPbt2wEs\nC0uzCq6yRNZia1YPosPiIjBVUx2Tw+GAzWZDW1sbhoaGpOuGGMXg4KCkIhaLBS+99BL6+/uh0+lw\nyy23YGlpCXNzc7IYbE3s6uqC2WzG7t27kU6n8Yd/+Id49NFHJZq8//77peSvFkLWS1Sp9l0T6FYd\nIR3n0aNH0dfXh97eXly+fLmur5V4Y6FQQDwexy233CJQRD6fx7333otQKCRVX4PBgFAohGPHjmF0\ndBQOhwObNm3C3r178U//9E+Cg7799tt45JFHpBCxmvXQyKZGkiq7gMUW7o9KpYIHHngAv/zlLzE7\nOyv4I7tyyMttaWmBz+fD5cuXpeunUqngv//7v5FIJGA0GmGxWJDJZJBIJPC5z30OqVQK169fx7PP\nPosHH3wQX/jCF6SPm+2K7EhRL6H1sL7ASvFpNb7Oc85M8Etf+hJ+/vOfC95rMBhECLyjowMejwcm\nkwnt7e2YmZmB0+mE0WjElStXMD4+DpfLJec7mUxiy5Yt2LFjBwDgxRdfRDqdxsGDB/GRj3xEKtup\nVEqaAwCIg/3QIkoWAFRwnLiLuiAmkwlPPPEENmzYIDhkNBqFTqfD22+/DaPRiJGREQSDQZjNZnR2\ndkKj0aCtrQ2BQABTU1PiHCiG8ZGPfATJZBJWqxXXrl3Dj3/8Y9x1112SYuv1esRiMXi9XrnV+Gu9\nVA/pKFXiuYpTAssbsqenB9FoVEQdrFYrKpUKfD4fcrkcvF4vLl68iEwmgwsXLkjaXSqVsHHjxrr0\ng8+OuDAvv+9///vSf6vRLEuBRSIR+T4ehPVykHnp0lQnxPXmZcNoiPAGu3I6OjqERxyJRNDR0SFd\nOqOjowiHw3V8P71eD7PZDL1eD4/Hg2w2iy984Qu45557JKsyGAw4c+ZMHUOBjnx15NvIppLBuX5q\nVxL/vbW1FePj44JNUiHI5/PBZDLBYrEInBQKhWA2m6XnfWZmRp4h8fx33nkHGzZsgNPpRCAQwGc+\n8xl87GMfE5hDo9EITVB9L2R8fCjFnNWHhC1B5PsxwmCbFyNJtnWxUycSiaCrqwsTExNSady6dSt8\nPp90lSwtLcHpdAqRt7W1FZOTk5ifn8fw8DD8fn8dlrN3717kcjnMzc0BWNn862WjASsdI6pj5IFW\nyegejwcjIyOYm5uTHtfW1lYhMZN0vnHjRtEOtVgs2L9/v0Sf8XhcBB7i8Ti2bNmC6elp9PX14ezZ\ns3LQgeX20J6eHrS1tcl6rv59PZjKm1Md5WqRCo/Hg7a2NslGcrkcNm3ahLa2NvmemZkZabVLp9O4\nevUqrFardI2w46RUKmFubg6tra0oFAro6+sTTYNKpQKr1YpkMinFTWDl8lkPjALV1PfP9ldWvwEI\nJ1qjWVa9oq+wWq3o7OzELbfcgmq1ikgkgtdffx0GgwHXr19HtVrFsWPHhBJkt9ullsH+e8Ik+/bt\nkwifqf7CwkJdrz+w9n37gVoY1bRF3Xg8zOl0GjabDdlsFl6vF2azGZcuXUIqlZJFS6fTaG9vh8fj\nwRe/+EW0t7cjmUzKB+EB58+12+0wmUyYnZ2F2+2Wm51UilAoJFgl+Wrs9FlP1Vk1mlDTbWB5I3q9\nXiwtLWF2dhbz8/MolUrweDwYGhqCw+GAy+WSim0ikcClS5fQ3t6OTZs2SbpNqEK9nTs6OqQVEljR\nBKTzjkQiQjpX1YrWi6m4lBpJqmLFpPBs3LgRra2t8r333HOPEM1tNhtyuRxSqRTGxsZw4sQJXL16\nVRofuP7UPCyVSpiYmEBra6tQZBill8tlXL58GdlsFktLSzIrSVWJWi8RO4u0KlbJc0fHSVrQZz7z\nGZFR6+jowB133IFSqSRReiKRQD6fx6lTp3D06FH89re/FdqV3W5HPB4XPmw+n8fly5exbds24RZT\nhaharWJwcFAoRG63WzQt1WDkRuwDjYJg+KoeaPmBWq3oyf3bv/0bDAYD3n77bRFRaG5uljA9n89j\n586diMfjqFQq8Hq9OHfunBSAWODhbeB0OpHNZrFr1y75wEajEd3d3RgcHEQikZAbh05cdZiNbiqW\nRshgNT+VA5p27doFjUaD9vZ2xGIxOJ1ODA8PixBJLBbDkSNH8MADD0jLV1NTEzKZjLTk0Uj0b2lp\nwcmTJwEsb+50Oo1EIoHDhw+jvb1dUhd2VLD7otHXlmuoXuqMHrmXyZkkLpnNZhGLxbBr1y6Uy2VR\n/AkGg9BoNNizZw9+85vfoFAoYGZmBm1tbTCZTHJxq0WcoaEhJBIJBIPBOm3K9vZ2XL16FfF4XAqX\nlCDjeAMGBI1s6sXD/cDsk//GSJ3dTuQ8d3d3Q6vVoq2tDS6XC/F4HBs2bMDBgwdx/PhxRCIRTE1N\nYfPmzahWq8KkUZ/puXPnkMvl5BmxFlIqlTA8PIxEIoFisYhEIoFCoSDCGx6P58PBKNUqJ9+o+ote\nnNgMb+lgMIju7m7UajWMjIwgFoshGAyKjl0mk8Hi4iKuXbsGj8eDkydPyq0CANPT03jqqafQ3NwM\nvV6P3t5eOcR6vR7bt2+Hz+eTtFKVeaczbXTj+1UxSpXuo25GjuGs1Wo4d+4cZmZmYLFYEAgEYLPZ\nkE6n4XQ6cddddyEYDAoFKBKJIBKJYGxsDMBKS6TRaMTRo0eh0Sx34bDDpFKpYHZ2VnibapSr4nDr\nwVSKlbqHyQkmXjY1NYWWlhak02kcOnRIMMtUKgWtVouFhQVs3LgRb7zxBoaGhjA1NYVgMIiOjg70\n9/dj+/btkg1RPvCee+5BsViUBgyKwHAOEbF9RviqqPV6wNd56TA4oZ9Q25e5fxOJBB544AGEw2EM\nDAzIfqQgBnHw733ve6IQ1tPTA6/Xi89//vPYuXMnDAYDmpqa0NzcjAceeECKOQcOHBD4g5Qrrp86\nqIzqV2QZ3IityVGy+sfIQu1yUHmL5XIZyWQSIyMj0Ov1mJ+fRz6fF7mknp4eXLhwAffeey/K5TLG\nxsbw05/+FFeuXEFHRwfi8TjS6TQqleV5JP/yL/+CPXv2YGpqSj40HYa60dXeZ5WmQKfSyLYaL+MB\nWt1JsmHDBnR2dsLj8eDSpUvo7OxEV1cXotEobDabwBTT09Po7e2Fw+GAyWTChQsX8NZbb8n/SyaT\nIsT86quvYmhoCMDyIaUohEajQTAYrBtpwPeo4nrrwYxGo0Qi6ntmwcHhcODKlSvIZrO4du0a7rrr\nLhSLRVy5ckVm48zNzaGpqQlXr/4/7L1pbNzXdTb+zAxn3xfOwp0SKUoUaS2U5FiK7Ti2Fdtx7MSJ\n06BJU7gp0PRF0BZBUiAf+ilFEbQJEARo+yZIkDZJnTipW6fxnjh2LNmSJVm7RElcRA63Gc6+D4ec\nmf8H9jm8Q7m1aMBvhvjrAIZFiRzO3N+9557znOc854pQ2u6//37cc889eOCBBzA4OIharSZK27t3\n78ZnP/tZHDp0SKCMfD4Ps9mMlZUVXLlyRQ6zWrRTneNmwClZAOSfaZRQ5L9PT09jdnYWkUgEjz/+\nODQaDebm5tDa2opKpYKXX34Zw8PDOH/+PObm5gAAn/zkJ/HVr34VBw8exPbt21Gr1RAKhVAul/GJ\nT3wCH/rQh/C5z31OIsrr169LYHTq1Cmk02nJ0PgfsCaQ8b5FlKqmIR8oK9OcWfPAAw/gySefxGOP\nPYaxsTEZ8qXVrs7RiEQieOihh+TAGY1GDA8P4/bbb8fu3bvh9/slNA+Hw/j85z+PgYEB5HI5SSs5\nj6NWq2F+fh65XE4KR/zwFIMgD7CZTcUm+TWNTAJiWh0dHXjhhRfQ29uL8+fPQ6/XIx6PC6n5/Pnz\n2L59OzQaDSKRiKjHZ7NZPPTQQ1JMCAaDeP3115FIJNDR0SGiyXq9XkZLDA8Pi3CD6rRZ4WQU38zG\nKEeNKBlNkoRssVgQiURgs9nw6KOPolqtwuv1SkFRq9Wira0NDocDb7/9NqampnDnnXfCaDQikUjg\n9ddfx3/+539Ke+P58+exb98+nDx5EqlUClarFV6vV3rFDQYDIpGI4GeEo4gdM9vaDPOeAEivOltl\ngTVNWpK+i8Uibr/9drS3tyMej8NoNCKVSiEUCmF6ehp33nknkskkTpw4AaPRiO3bt2NpaQnPPfcc\njh49iq997WtobW2Fy+USv3HmzBnk83nYbDb4/X7xEcBqGzWr6yoRnopF1By9GdtQ3sSQlg6SKRj/\n7eDBg/jVr36FN998Ew888IB4d7vdLo3v7N0MhUKIRqMYGxvDli1bBL85duwYBgYGpOjQ19eHeDyO\nY8eOobOzU5ri5+bm4Ha7sWfPHrhcLuTz+YbUlSl4X1+fyIw1s/HgkhSrYpMszOzduxfJZBL/9m//\nhqGhIUxNTUmrHJ0iAOzYsUOmUWazWVgsFvh8PnzsYx/DwsJCA8bY3d0tFUP22nd1dWFubg4zMzPC\niVXhAEY/6XRaxnw0s/GgqimhKg7Nuc8f+chHsGPHDly8eBG7du3C6OioSIGZTCb4/X68+OKL0vCw\ntLSEy5cvi5MolUqy3n6/H8eOHUN3dzfC4TACgQC6uroQDodhsVjgdDqlqKMKMzMAcblcuHTpUtOP\n2aBR8YcQHGGapaUlXLt2Dd3d3fijP/ojOBwOTE9PY9u2bTh9+jRsNhuMRiO2bduGSqWCX/ziFxgf\nH8eOHTtQr9dx5MgRmYag1WqRSCRgMBiQSCTw1ltvwefz4fLly9i3bx8cDoc4RlUchhelegmVSiVc\nvnz5/RkFofIomdry7+12OyYmJuByubB79260trbi6tWrqFarEm6zt/Pw4cPw+Xz4xje+IdgiKUV0\ncARumQaS+EuVkGg0CgDS3aCG0IQIyHNTRQaa1eiA1qe05PGFQiFMTU3h7NmzCAQCCAQC0vterVax\nZ88eaLVanDt3Dl6vF4cOHZIZPBqNBjabDU6nExaLRfCgWCwmF53H4xH8hrNimKavJw/X63VkMhm0\ntrZKtNnMRhxWVThXs6JCoYBAIIC2tjYcPXoUAGRcxq5du3D27FlUq1W89dZbuHz5Mh599FGZwbO8\nvIxIJIKZmRksLy8LZORwOJDNZpFIJBCJRETerlwui3o3L0X1fer1evh8Ppw/f22TlBYAACAASURB\nVP6GcRzNaoyEyZJYX8dgqqzVavH666/DarUKZa1Wq2F0dBR6vR5jY2OYnJzERz7yESnyVKtVhMNh\nUQzLZrOCJ6dSKSwvL2N+fh4ajUawZQCYmpoSYRNGjYRfVlZWcOnSpQ2t74YBEFURhPggow7eFJlM\nBg6HA16vV8r1g4ODeOihh0SI4Qtf+AIOHz4sm40pMikVVHxW5dwIfieTSfh8PgCrkmtUgeZ74n+d\nnZ0oFouw2WxN7yj52dd/BjWVyefzoqIyMzMDr9eLQqGA06dPC1Rx4MAB/Pmf/zm+9a1vNUzG5I1M\nYdV6vS4bTavVwmazyYXlcDhgMBiwa9cueDyehvdIYzsfL85mtvVNB8TTOPnPbreLY3M4HOjs7MTS\n0pIwARj5ZbNZvP766zh8+LCsFUnN1WpV6DH1+ursIVV+jQpYs7Oz0Gg0mJycRKFQkIuK0Q+xPGYJ\nm0F0msIqLPQRC+ZYC/Z0T05OSpZCWttdd92FEydOwG6349ixY/B6vbjjjjtkLIzdbhfiPx0saXJ8\njhTTqNVqeO2115DNZjE6Oir6tIwk+Wd1IN/Npt7vyVHSuEEoblGv1xGNRuF2u1GpVIRPaTabMTU1\nhR/84Af4xCc+genpaQwNDQlexMNLEi67Fpgi8eadn5+Hw+HAxMQEent7JRWnIwXW4IHdu3cjHo+L\nynqzGyMcOi6mYsViEW1tbTIfiHxG8vksFgtGRkZgtVpx7tw5uFwufPOb37zBiaktjoVCATqdDrlc\nrkFMo1QqIRAIoFAoIBaLyRzw9aT36elp9Pf3y9iNZneUAASXUq21tRWRSATd3d3Q6/VycRBXS6fT\nePXVV6HT6eBwOPDGG2/g7//+76VAwCIMZdXYWaLT6RAIBACsXt6BQECqwgsLC+js7EQsFrth3Xhh\nsRFjfcTZrKZyKIE1pX7qeFosFni9XphMJmzfvh2JRALlchkrKysYGxvD0NAQxsbGoNVq8Vd/9VeC\nX/L17HY7rFYrzGazFG79fr8EQD09PcIMoepVNBptgIT4HsvlsjRqlMvl96czB4BMnlOHCfHAtba2\nIhwOy8yL9vZ27N+/H5lMBmfOnMHDDz8Mm82G73//+0LOXS9UwAg1mUw2tD7OzMxg69atAIDp6Wm0\ntbXhjjvukM2v0hFqtZq0jqkVuWY39r6rkEZbW5voboZCIQwMDCCVSsFmswmnsqenB8ePH8eDDz6I\nkZER5HI5Kcqw95t4EdVveMGxaPPqq6/CYDCgvb0dyWQSY2NjDc+Ff2ZVHYA0BGwG02q1QvGhxWIx\nOXyMSgKBAE6fPg2Px4PbbrsNXq8X+/fvxw9/+EPccccdcLvdSKfTcjGTd0nh6FKpBIvFIrO/C4UC\nhoeHkc1mMT09DbfbjVQqJe+FuCQ73CYmJgBAsMnNoFNAyKZcLguNjFKKoVBInBfrC4lEQrppWltb\n0dXVhW9+85sYGRmB0WjE5OQkgLWmB0aCDIK8Xi/6+vqg1+vhdruxY8cOVCoVpFIp5PP5hhlQqpCv\nRqPB2NiYBGMbWdsNO0oeMH4QgtFqr6bFYpG2RDq6r33ta7hw4QJ++tOfYtu2bdLJAKwRVfn/lpYW\nhEIh7Nu3T3CEU6dOyWLlcjl4vV75GbUdTaPRyDxvDq7fLJutWq3K/HE6IOK6+Xwe169fx6VLl+Dz\n+RCJRISkfOLECfzBH/wBgsEgXnjhBRnIxM2h9pDzJvV4PDh48CCq1Sra2toAQPpsw+Ewdu7c2QB7\nMCqvVCqCj6oMg2Y3ts5yLywvL0uvtcPhQCKRQH9/PxYWFhAIBHDt2jXBFn/yk58gEolgcHBQMEiV\nN6jOz6nX6zhx4gROnDiBlpYW7NixQxxooVCA3W7HyZMnpVeZz4d4M5svqLizGfYuPwPrBgBkVlMu\nl0MgEMDOnTvR09MDi8UioxqMRiNee+01PPPMMzh8+DCGh4dl1rY6l4gz0in28uMf/xhHjhxBrVYT\ngZdMJiO6EmfPnhUJNjpLFVtXZdlu9qLfUNVb5Xmp7XVUL2dBxuPxIB6PS9Hlq1/9qsh7LSwsyGbj\nYrKSzjSbG6tSqSAajaJer0vfMsm6Xq9XWpnUD0yMRL3ZNqIS8vsyRn6ENpiKkx8Wi8WkB57c0E99\n6lN49tlnceDAATz11FOiBk8VbeI6KpkdgIiN8HVeeOEFeRYajQbz8/Po6+t7x5kifD+b6RJSq8kq\nY0Or1SKXy8FgMEhUEgqFoNPpMDExgWw2K+v7xBNPSF+3Vrs6BoJiDfX66piDUqkkUZXL5cLS0pIM\n0iPZeWBgAHNzcw3tfXxvnAvDeTyktDS7qWurZnZLS0swm81IJBJSQAwEAujt7cWrr76KQqGASCSC\nkZERmYmVTCaxZcsWLC4uynA1Qk2En0KhEIDVTr29e/dKoYi6rOPj41haWhIZQWaWRqNR+u+BjUXr\nG44oST6mszSZTHj77bcRDoeRTqdhMpkwMDCAQCCAVCqFe+65B7Ozs2hvb8ePfvQjzMzMoFqtwmw2\nS4cJb2i1y6e/vx9dXV0IhUL4+c9/DmCVl2UwGODz+RAMBhvapdZX4NXZM2o3RrMa0xdGGXzfBw4c\nQGdnp4gqhMNhVKtVOBwOPP/887BarUgkEvjEJz4hmA0vIXXMKgn5qVQKiUQC0WgU4XAYIyMjQucA\nVgnn165dg91uf0fiM/FNOtDN0DkCQEbtskhGnc5sNotwOCwwQr1ex8LCAvr7+3Hq1Cns3r0bVqu1\nYXCdy+WC2+0WB0EBl46ODpFi27NnD7Zs2SKiL1qtVopHpVJJLnE6QxWT5l7dDNg6sDYIjwVJtmlS\nLzWVSqFcLsPn8yEajSIWi4kuA9kIzIA4VsPr9YqDY9Grp6cHpVIJbrcbg4OD2L59u/xcsVhEJpPB\ntm3bGlqWubb0BcViUfbARnzChjtzOjo6RD2bh/vw4cOIx+OYmpqC1WqVauHWrVtx7do1eL1e/N3f\n/R327dsn6bZOp0MikcD8/LwsLlvq9Ho9kskkkskkzp07J5uV41fVFkk6SbUrh5ueD2+9uESzGgVf\nVbXo559/XiKRpaUltLe3Q6/XI5FIoLu7G4FAAIcOHcIzzzwDg8Egun+MDvP5vMwlMRqNsFgssFqt\n0l/785//HLVaTTA1YDXdJx2LGxFYdeb5fF6G12+ks+H3acwyjEajOHxqP3Z0dMjneuutt0TAYn5+\nHjabDXq9Hvv27QMAKV6R2uLxeKRdkd1kjFCvXLmCQ4cOAVhdJz5Xk8kkEeN6nQRqWDLSpyNpdmPx\nURXpNhgM6OnpkdlA5XIZ//zP/4ypqSm5mH/7299i9+7dOHDgABKJhJx/u92Ojo4OeL1eGbeRz+cR\niURgsVjgcrlQKpVw6NAhof5xBApbpdmMQgdJn8H1ZY3lfenMAXCDJ67X68Ihq9frWFxcxNzcnGAG\nbW1t+Md//EfcfffdDa1afK1SqYRwOCzUGBZ5dDodstks3njjDRFHJSbH9Fud8sjUslKpIBgMyusx\npd8MtzPxVRWb8nq9WFxcFGyQ8497enrklv3Xf/1XuSWpLcnCWDKZlK6l5eVlGI1GGWUwODgoqvKs\n7DLFYasqsDYPiXQVHt7NsKYA5MLgmvKSZ5cHLxiPx4POzk6Mjo4KeZ/OjgVMQg42mw29vb3w+XxC\nEYpGo1hYWMD8/Dx27dol69XS0iKFI0ZJvLx5UImjkVJD57MZUm9gdS9wnLQauPj9fgwMDKBQKKCr\nqwttbW24evUqxsfH8dhjj+H222+XAiFxY1K2RkZG4Ha75aKZmprC5OQk5ubmJLok9svIVaUmce34\nnrhfVZ3a98VR1ut1acGiw2TIm8vlZM5Fd3c3zp8/L2Mp7777biSTSTgcDgmDeatSap83da1Wk7nU\nbrcbBw8elJ/h+AJ+3+TkpPAoWeVaWVlBd3e3iBkUCgV85zvfaQCam9HUogsPMqfJGY1GES0dGxtD\nV1eXDBmbnZ2Fy+WSKJTOkriYzWbD0tISMpmMjNUAViOqK1euSMGGfL2VlRVYLBYYjUZks1m5mLjJ\n1JkuWq1WCjvNbuTgAWtK5xcvXpRCQb1eRzAYRC6XQ89/TwLs7Oxs0DEkFw+ADA1zOBwiAcjXYSca\n5ezUS6ZUKsFkMiEUCsFqtcprUnCDkBaw2o/8qU996vezYBswjUYDp9PZgFWyoYHEcZfLBYfDgW3b\ntqFer6O9vR2RSET66MlJBVb3GBX6bTYbAoEAnE6nND9Q7o48awZS7Izy+Xzo6ekRXJQZxcTEhBSI\n9Ho9PvzhDzfI6f1vtiFH2dLSgnw+Lx5co9HItDOXywW9Xo+RkRE8+eSTcDqd8Hg8mJubg06nw969\ne6HRaKRPk2YwGNDW1iYflFVtyjI98sgjeOKJJ8TxUVmFDpmgOhvu8/m8KFLX63U8++yzCIVCMryo\nWY0CpvF4XC4h0nSmpqZQqVQQDofR29uLUCiEdDqNSqWCmZkZ9PT0YGVlRWTm6DBrtVURVM4gIVk3\nl8vJ7X/77bejv7+/Idpqb29HKpUCAOk2KRQKMgOJOJrX68XLL78shP9mNXIjOzo65JCUy2V0dXXB\n7/ejUCjAZDIhEAggFothZmYGbrcbfr9fLhNG+tVqVTo8yuUyJiYmRA6wpaVFmiyOHj2Kp556ChMT\nE3JgtVot4vE4HA4H9Ho9Wltb0dnZiUAgAJvNJu2PxCq/8pWvYMeOHb/v5XtXY4W6v79fnL6a2uZy\nOej1etxxxx04d+4c7HY7fD4f2tra8IMf/EAuYl7AfI1UKoXx8XHJqCwWCzo6OlAoFPDtb38bTzzx\nhNQ4dDodbDYbJiYmBOvv7e3Fnj17sG3bNni9XiwsLEhUPzAwgI997GM3DW1syFFarVa5BemIOjo6\n8PLLL8vGIel5+/btmJubkwp3X18fksmkYIh8nZ6eHgCQwhDTT7PZLLO5S6USnnjiCXzgAx+QKDaR\nSECj0QhW1NraCo/HIzeNVqvFc889h2Aw2KDM3aym1WqxZ88eabsCVtOZhx56SDZcX18fdDodnn/+\neXGs27dvx6VLl7Bv374bhB9I5aJkl5qWU2SD4g+f//znsbi4CK1WC5/PJ9QZRqXEnYhDdXd34yc/\n+Yl0UDSzFYtFHDt2DD6fT0j9w8PDchn4/X7Mz8+jWq1i+/btGBgYgEazKjlnsVjw6quvijwaL3kK\ns6jiICxQUrBhbm4OTz31FP72b/8WHo9HRnAkEgmk02kpJhFXLpVKWF5eRiwWwze/+U1pdWx2S6fT\neOWVV4RkD6x2kXFtSMtZWFhAd3e3tDTOzs7C6XSKMC8zSRWqIFbOAhyFW1hs/OIXv4gvfvGL8Pl8\nMJvNMmo4Go1iZmYG4XAYxWIR6XRaikwdHR34i7/4CxQKBckY3s025ChZAADW8n5uNobFIyMjCIfD\nOHv2LPR6vUi1f//735fBYUxJcrmc0DGYPtK4MRkZ1et1oQ0sLy8jkUjcUABihBuLxUSwgAvf7EUH\nVgXp5LVardCj+DnffvttGbFBDCaXy+Hee++VEb2ckcxqIXlj6yfeEQ/j68RiMfzJn/yJKM+rbX90\n1CsrKxgaGpJ0hkKszX4JGQwGjI+PNyi0z87OIhQKCc49NDQErVYrn8lgMKC1tRUnTpwQ2IMHjf33\nFoulgZTP/5tMJkn12Y3zve99Dx6PR7BzkqeZLmq1WiFD33///UilUohGo5tC71Ov1+P48eMSSZIj\narFYUCgUJFjp6uqS9thAIACz2SyRtcvlkgvZZDKJ/BqN+5WBlNrOuLi4iKeffhqhUKhBLYjQW7FY\nRLFYxPT0NDQaDT70oQ8hFoshnU6/P505ahWOOA9vWDajd3d3w+l0CrG5paUFgUBAxlB6PB5JoV0u\nVwNRVOX60UFyQ3Gx/vqv/xoajUZem5SEcrmMzs5ObNu2DcvLyzh79iz27Nkj0vHN7iipYsKxAMBq\npMNLpVwuIx6PQ6fTCfXH5XJhbGwMer1eondKgq2srCAQCECn08HpdMrvIdCuVlzVW79YLMLhcDQ4\nbNIqOAfc4/Hgueeekza1ZneUJD6rog28wKkFMDExAY1Gg0qlArPZjIWFBZhMJmmkYDSuyomRhqXu\nT5Wrqeofzs7Ooru7G4lEQt4X17a1tRVXrlwRvH9oaEgiJzZ3NLNZrVbptAFWoY6lpSV0dXWJrBkF\nep1OJ/x+PyYnJ+VsXrx4Uc60RqORjjPCaCT3MxDgBaTqTD7zzDMSgfLCoupTtVrFyZMnsby8LAU7\nwoc3axumB5GzxE1SLpdlUJDb7caJEyfg8/kaSvCZTAYf+chH8Itf/ALRaBQulwvlchm9vb3y4YE1\nTh4rtORXqoebhQcqcKucw8HBQelkmZyclNDf7XY3ffVQpTapOA95p+VyGQcPHkQ4HAawimlOT0/D\n7/fj6NGjePTRR9HZ2SnjUulImbqoZGBuENI0iDezmkjMUXWkExMT0hPNKq/RaEQ8Hm9oC2xGY6WZ\nRUhiuMFgUA5eOp2GxWKRed6dnZ1yEZRKJZG040VFKhdhDrUNkWvD30t6ikr7Uqvd5XIZ09PTWF5e\nxq5du6DX60VDdDMUyhiVqwRv8q1NJhPy+TyuXbuG/v5+2T8jIyNC2aLwC7vwUqmU0K0YPBFmYxRJ\nmI5fW61WKQpzfbnvV1ZWMDc3B61Wi4GBAanQb0R+cUM7nN0brPiRmwSs9rxygiJblJxOpxQlisWi\ndN6Uy2VR/0in07J4KjbHg03Kj5recPQn3wN7Qjkv/Pr161KNJUbR7MZU2ePxyOFgBbxWq8HpdCKX\ny8Hv98sUOoPBIOTnM2fOQKfTwe12izwdtT/Xk8NVzqkauQPAfffdJ8+U37e8vCwFJJfLhTfeeAMa\njUbECprdeKAYpQEQtSS2NrLCTeyMXxuNRuzfv1+Ej202G65fv45YLIZIJHJDqyFbGtlrr04o/b//\n9/8Km0GlCFENq1gsYteuXahWq3C5XCgUCk1fhATW9hXnaDPLI8WppaUFe/bsEUywVqvJ+Y3FYvjD\nP/xDgYjq9bpADrFYTM6uOt6FnXfkA3ON//iP/7ghC+JzYVXeZrPhtttukyh+I8pXG3KUaj8qsKaV\nmM/nheRcKpXEKcbjcXg8HoTDYRQKBUxOTopD9Pv9KJfLuHDhQsOAKhaJeIDZ1qdy+e6//36cPn0a\nU1NT0uLocrkwMzODbdu2YXR0VEiunZ2dAJp/ZAFTO7/fL5GeVquVsbyklVB/sr+/Hy6XC1NTUzh5\n8iQ++MEPolQqIZVKYWlpCa2trWhra5OfJw2IG0ONFlV4g5kCx3dw3Xt7e3Hq1ClJtWu1mozmaHbj\n4eKMJfIq0+m0zEGv1WoilOBwOHDhwgXs3bsXPT09En20tbWhWCwilUqhs7MTCwsLgp0xqlQ5eqrD\n1Ol0eO2119DW1ibUFYr9VioVdHZ2CtWI0Rnnqje7sV7BjhhivGwm4Z6t1+tYWlpCIpHAmTNn8MAD\nD+CjH/0oXnrpJRiNRvT09IjAt8/nQzKZlEBJbUWmY2SqzaCKcBv1cFlDWV5eRltbG/x+P/x+v0Ap\ndKg3YxvmUZrNZpGvB9BwEzDddrlcojLMKOfIkSPo6+uTEbZ2ux3hcBiPPPJIgyME1lJvkqjpLJmK\nd3Z2SovT/Pw8jhw5Ao1GI/SOCxcuwOv1IhgMYnZ2VsiszWzkUHq9XnHqDodDdA1Z0SaNqlQqYXx8\nHJ2dnTh8+DB+9atfoaWlBYVCAQ6HAw6HA08++SRcLpek8WqBhqbRrM0Sr9VWNT+J/5ZKJSwsLODK\nlSuw2+0YGBhAMpnE+Pg4UqkU2tvbG+ZRN6sxPWavMGk67JRhN41OpxPHRuHp+fl5DAwMSBQzNTWF\nSCSC22+/XST+eLkz1WMKTnyOEaTH45GRqX19fThw4AAOHjyI2dlZTE5OypiEYDDY8NrNbsxM2GLM\naI60HTItlpaW4HQ64XK50Nvbi9dffx2RSAQf+9jHJEUeGxtDS0uLwEiEeXj2+Rz4vDg5k2fE4/HA\narVi//79+MxnPoOdO3fi6tWrCIfDePjhhxEMBsWfqBjyu9mGdjhvSDpAVTOPgKua0lmtVkxPT8Ns\nNkvLmF6vRyAQwMWLF0WYgUZnwBuY0Q4fgprmXLp0Sbp4gsGgcDm/853vSAUzHo+LbmOzH2ZiVW1t\nbeK4HA6HSKYRp9RqtdKK5Xa7AUCmLBLP7e3tRSKRwJYtWwCstZipUSQvIrXyzb8/fPiwpDR6/eq4\nYaYt//7v/y6iJPF4HE6ns+nxXxbE2HRAShTJ+Kygcj/m83kEAgFxpCSTJxIJTE1N4a677sK5c+cA\nNI7G4AWnOl1ilYxkT506hXw+L/hbV1cXEokEstksHn74YZjNZszNzQn5vdmpV8Ba///8/DwAiJoP\naUAU/FCdU3d3N6ampqTyTfWhTCaDQCCAy5cvC6asdoKRVUB8k3/mWnP87+LiIpLJJPbv34+JiQkM\nDQ1h3759UoSjiMnN2oauK4KqBEyp6uzxeIQqQUKz2+1GT0+PbB62ITHN2bVr1w3dC+tTQ7XVUU3F\nDQYDbrvtNlHZbm1txblz53D58mV0d3ejXq8jl8uhXC5jeHgYo6OjTV/1BlaZA62trUgkElJwmJ+f\nlwiPhQefzyfy+fF4HHa7XaqGxNBCoZBUu1XYgWmo6txYBaZz7OjoECUWALj77rvR0tKC559/Hmaz\nGW63W0YatLW1NVx2zWi8dMgaqNfrmJubk15tteXzrbfews6dO6VN1mq1NhQZBwYGcOHCBZRKJbl8\nVSxX7U7h8zKbzZJCvvDCCzhw4AByuRyi0ShmZ2dljvXS0hLC4TDa29uF/8rLsJlNp9PBarVicnJS\nqGc2m00c3PLysqzBsWPHpEo+NDSE3t7eBvYMVeA5SnZ9O6KaGdEBM1CoVCr41re+hW984xtSLzEY\nDHjwwQdFFzeTySAYDIpw8s1q1W7IUa6srGDr1q2SJtfrdRF+JQF6fn4eFosFfX19CIfDQlnp7OyE\nXq+XKIejZplasIChgt8qhsCNxkW655578Oqrr8rPVatV7N69G3Nzc1Lxoh4m+3ab2dgnf/jwYczO\nzgKAUBuIV7IAMTExIZcGsU3Sq7xeL7RaraQsNLV4wK/pQPl3xEbHxsZEpspms8HhcODUqVPCUSXh\nn6yFZjdGLFevXkUgEJA+b0bp3ItUrero6EA8HkdfXx+mp6exZcsWxONxXLlyRWhvKtuCKTIjfrZ5\nqu18dMhut1vOC5XSh4eHMT8/L91ser0ely5dEiyz2c1oNCKXy+Hs2bNSSaZqEqPCbDYLvV6P9vZ2\n9PX1CU+aM+knJiZw9epVifiYPRKTBNYufP6bwWCQQIuqTPV6XfRwqUvA56nRaERti/34N1u70Gyk\nyKHRaGIApje2jE1j3fV6vfXdv+33Y7fW9v2zTb62wK31fT/tptZ2Q47ylt2yW3bL/v9ozV3huGW3\n7JbdsiawW47ylt2yW3bL3sVuOcpbdstu2S17F7vlKG/ZLbtlt+xd7JajvGW37JbdsnexW47ylt2y\nW3bL3sVuOcpbdstu2S17F9tQZ47Vaq2rLVVU+VG7FFStvXcytRtkfWvdO30v1Uj+p58DIH2kpVJJ\n2qbWf99/K8U0bR+j3W6vezwe+bw0to2u76JR13n9erzT2vLZqNP/uL7sglq/xvx+CkmwbU99jVqt\nRhm9pl1bs9lcpxixKjnHlkMAN6zvelPba/+3v6OpHWX/07Q/rj8nX77T2anX64jH4/FmJpxbrda6\nx+MBANkT7Bpb35q8/mvazfoFrpl6TtbvZ/6dyWSS+VqUDlStVqshnU6jVCq9697dkKN0u9340pe+\nBGBNFoyDvDgnhD2Z6sxnVauPC8neWnVR1A+/fnA528HYzrdeiMHpdKJUKsFms4keJt9juVzGk08+\nuZGP+v/cvF4v/uZv/qahrVOn02FmZgYmk6lBnYY9qlw/dZNYLBbR8FPXiX/mSAM+M4pjcIqequLE\n1rxEIoG+vj6YTCZcu3ZNXqtUKiGTyeBf/uVffi9rdrNmt9vx+OOPy4VQqVTg9XoRi8UAQEaWULvg\nf3JY6s9zL6rqPqpKE9DYxscxt/w+/o5z586hr68Pn//85/GjH/1IZtKrgjH//M//3NRdLx6PB1/+\n8pflc5XLZZl5w7EQqoCx2jsPrCnBUyFIPec07nVV3IUCGdSRUOdF8Xt6e3uRSqVQKBREkFqn0yGR\nSKBUKuHpp5++qc+44dT7naIPOjIqCFEwQ+3ZpK2/uVUnqn5A/hsdAx2nejuzn5aRQqVSwejoqLwe\nhVipatTsRvko1cGptyYAEcfg96gRtEajEU1F9UJR14tyc9Q55Kakw6RSEZ8xZ4HTOVB8lSNwNzIb\n+fdp3A/cU+oMba6Ruq40db9yv1FVCbhxtjkdpxqp0rnya1X7c/fu3YjH4/jlL38pkmQc1AfghoCg\nGW392q0/p1wDnmMKWFCxSl0TYG1NqYalqvDz9fjv3Ouqn6C4C1XTi8UiZmZmJNiik+RkzZuxDTtK\n9c2pDpEfgH/PN8sDzAhGVQCyWCwid1QsFht0LLm4/BkuAn8P5+Xw96VSKbS1tck4Vw4xU8PxzWD5\nfF7Wk05MTUe4Kejc6Mwo8EBlJzq4trY2WX/1tdT0njc+ny2flTreYGFhAfl8XobZ83beDE6Sn1cd\nVseDu17hnVElTZ2zzUuJqkrValXEfTn/XD0f3LdcX6BRgITOmunhZz/7WTgcDrnENoOTpKnDAHkp\nrY/yaNSspKA0140ziTwejyj9R6NRdHZ2QqfTYWlpSdaUzpQ+B0DD3uX+TSaTGBgYwNDQkMgWqhnB\nzdqGHSUHB60/IOqbV1NsVSaJG5aLQ2fmcrng9XpRLpdFrl/dvKrquZoaAmtpDADRs/vNb34jkABT\nzc1g1KTkJgIga8cDp64nZdaAtehar9fLIDc6UqfTKaM3VPk69San42TEYQ8L/QAAIABJREFUqWJs\njFwTiQQ+8IEPSJZALcBmv4i4ltQf5AWgDgXj9zCC4d7jGi4tLSGVSomif6FQgNlsxrlz5zA2NoaH\nH364YSywerHT1BEnalQ6MDCAixcv4siRI8jlcg2ZUrOvLY2jZtULmFGgOkkVWMNtmZnUajUJmgCI\n2nxrayt27NiBfD6P2dnZhrlPwNrFxf2oQnVq9Hnx4kWk02mcOHFCZoyvF7B+N9uwwjkPDd8UD5ZW\nuzqhUV0UOjan0yljOvnBVCyT6Yper4ff78fMzIxMcmRUyd/B96EWFtRbpr29Hd3d3bJh1d/bzEY8\nVXWKPMD8/LwkVEiipaVFZihTMJUXDiW/eCn5fD50dHQgk8k0jC/gGqkZAv+O30PpttHR0RuA981w\nEalRDfcM14wRjvp9Go1GZj7xUuIUR35u7kej0Yjf/va3MJvNuOuuu2QiKZ0BLxWKAhuNxgY8s1Kp\n4I477hCxWmANGtnIAKzfl9VqNdF+5F7g4D9GjYyyAYjf4FgRfk3HSYyWUZ/RaMRtt92G8fFxdHV1\niZwd/69mqxTpVgMzAGhvb4fNZmuAXjYiEbjhKYwUFFUl7wHIYeSb5KHW6/WCZXHBarWa4F4ajUaq\n1cRxbrvtNtRqNRw8eFAWDoBsbFaz1CmN3LQLCwsiurq0tCQbvNkl9ev1OhYWFhpSbB5IKmyvxy2r\n1SpMJhN0Oh0ymYxcOupkSw5hYsrB2dwcvkbnzE1lMBgahkQx+tFoNOIQgDXHTkyt2U0ddM99CDRm\nJCqWq9FoZICbCvMYjUaJYKg7yZQPAF5++WX85V/+JcrlcgMkxbk44XBYHIOKvVPclgVRKq43+wgT\nYA1TrFQqUqwhZrm+RsF1Vyd4cnoj/QLXhrPTuc533303qtUqDh8+3KBOzkiTRTnin4Q8dDodjh07\nBmD14uGoWpfLddN7d8OpNw8NUzJuEKbHfMNqMUDF2LgwHFGppuwc+sMP+eMf/xg9PT2yiHTE4XBY\nBhapKtMcGs+IVx3s1OxRpZr+qdENbz5GH6xU8/OtrKzAbrfDZrPJZ2TFVKPR4OzZs5iZmUEgEAAA\nmasTDAZRKBQEb+PP2mw2zMzMiGPhxUb1b2KdjEqprt7MptKZuNd42ej1ekn/1CKlimHW63UYjUaB\nPpgyc6SAChNZrVZ897vfxY4dOxqcM0e4arVaLC4uNlzcXq8XmUymQYj2nZxMMxuDFWaRvFx49hhZ\n8nwXi8UbZg3RYXK8DDNKBgN8ne9+97vYu3evzNHiMx0dHYXb7RaIgz5qamoKxWJRnDBV5zlw76Y+\n30YXhM5JBbmBRol2YhXqjAzVcfLWYRXV7XZLCqcqG9955504efKkALuMeOg4OO+F70uv1wvOob6W\nw+Fo+sNMx0bcUK3E8rJRb2uXyyXR9tLSkgxdAtbgDI/Hg+HhYZkeaDKZJKVLJBK48847ZZIiNyLX\nkemjVquVgU0c5kQaEmelbBTv+X9tTAHXX0AAJPrWaDQy18lkMslhVRX0GSCoqXkymRRHqY6nvXjx\nosyKUgOLvr4+zMzMoFAooFZbnXsdj8clrWfRgurcm8VR0i+olz1ZFmowo15Qqm9QZ57rdDrk83l4\nPB4peKln4qMf/SiOHDmCrVu3ymtaLBZUq1UkEgnJfFVKEf9d9Vs+n6+BgvS/2YYdpYqZMW1QsR+O\na2hvb0dXV1fDuE2m3ysrK8hmswBWJw0yXFYxSOKQt912G15++eWGypbBYEA2m0VLSwvcbjdqtRoG\nBgZuqJTr9foG59DMRke4nkpVr69OviTVSafTIRaLweFwAIBcSkyXdTodrl+/jvPnzws1IplMAoCs\nM9O7xcVFPPbYYyiVSrIJualefPFFvPnmmzCbzbDb7Xj77bcl+rJarVJht1gsTX8JqZG5il8Ba0VI\nOsxEIgGfzycXgFpY4zOwWCzyNbF5lYrGsSiPPPKIpIhqcWdoaAhtbW1ob2/Hrl27GiIvADKpVMVB\nm92Ih6sVfxVbZyAQDAbh9/vlImKGyecSj8fhcrng8/mkcEznyourUChgZGQEX//61xsuPKPRiHQ6\njUKhgK6uLuh0OmzdulUuQhV79/v973/Vm56aTolRDgAJZy9cuIBKpSKbaP3NwCl4TG3UVEQlRVer\nVRw6dAgHDx6Um5v/8ZC2trYim802pCpMuRm2N/thBta6cNQKNteNOCUJvJzOaDabhXM3Pz8Pv9+P\nzs5O3HfffQiFQnA4HLBarcjn8+LY1KJQPB6XqJXPtKOjA1u3bsW+ffsESAcg6Sed7W233SYRZjOb\n2rCg/pk4GtM+jUaDQqEg84d4SFXeo8PhkMmAZFZUKhVZB3UQ1o9+9CN8/OMfFydMB22z2VAoFDA1\nNYVDhw7J3lQnlfKsbBYjPgms1RIYYTNoqdVqKBQKgq1zH3IiZbVaxZYtW8Qhqpkoo2/ixJVKBV/5\nylfw6KOPij/gRUVKXEdHB95++21x1Cz8tLW1vb9Vb41GI+Dn+gFgaum/UqlgcnJSOIFqOb5er8Nq\ntQJYA4FXVlawtLQkaUqhUJDbiaArq4rLy8vYtm0bOjs7BdNYXl6GxWKBxWKRyMhut8Nut0ua1OyH\nGVhdUzVdUKuEpVJJ1j8QCCCbzcooVKvVikKhgFAohJmZGYTDYYyOjiISicgmVbtPFhcXMT8/L054\nZGQEpVIJS0tLSCaT2LlzJ+677z50dnYCAGKxmODKxWIRuVwO7e3tQm7fDMZ9wD2l0oBY1FLZBuq+\nIRHcbDYjnU43kMoZGZFloEZSDocDb7zxBhwOB1ZWVmA2m6XKTc7glStX5LVaWlpkBDDwzu26zWga\njUZaBNc3SLCOwTV+8803JahhlVyr1aJQKMDlcjVQqNj5x+Dsv9sNZc85HA787Gc/g8vlQrFYxP33\n34/t27fLGGJOaWUGoNPp0NHR0cBrvVnbMD1IbdFSMRUuCh0px1CqbHxgFdQulUpCBeDNzgNdKpWk\n+kfsghXDer0Or9fbwKVklMX0k9GCiiVtls0G3Ni5xLUmhWVpaUn4i9lstmFaIrC6GQYHBxEKhQAA\nra2t8noajQYWiwV2ux1OpxPValWcY29vLxYXF+F2uxGJRCQT0Gq1uH79ukRfKjDPwfObwVSO73p+\nHy8gRiG8WNR2WovFIpF9Op2GxWKB0WiUNWDEbzAYhApEmovL5YLdbpe517SVlRXMz88LY4StlcTX\nAGyK9V0fATJK55nn+azX6/D5fDdg8QAk62HAxP1GzmWxWITZbBbnWi6XsbS0hN7eXqTTaXR2djbA\nd4zK29vbG7B0Zq8bdZQb4szQMaoVWvUGUBelra2tYWyn+v1ms1kiQRYgeBObzWY5jNzQTIUYIar8\nSq1Wi5mZGUxOTkoq73A4Gjidm8VIZwLWqoi8aHghmEwmBINBxONxuN1uIewXi0WZXR2LxRr4kby9\n5+fnhXhO+o/X68Xo6ChqtRoGBwclhWHvt8lkwmuvvSbrWCqVMDg42ED+3QwXkfo+1cuIuDD3Notk\nDodDsF0AUjUl7YzQEKlypABxX3LWtNFolNGpjGTYoqjT6QQHZjrKLG0z7VtgDetdH0mqnUgajQZb\nt26Vmd0c48vii8ViwcrKiqTQfB3uewBygZN50dLSAp/PBwDib/g+5ufncf36dQkKenp6JE1Xays3\nY+8JKeahVSM2/vJarYZgMCgOjQdVBXQLhcLqL9euzUBOpVJCXWHYnMvlYDQaYTabJe1WAW46VR78\npaUlmYOsHobNAIgzxaODVImxjLS1Wq3gObx1bTYbcrkcvF6v8MOKxSKq1Sqo9DQ7O4t6vQ6PxyMX\n0OLiInK5HCKRCNxuN6xWq2xuFpBWVlaQy+XkdxaLRbS3tzcU8zYD/ksnqHZsAGhIsQE0tM6SysMi\ngMlkkijGarWiWCxicXFR9i1pMZz7TUoRiw+kYfH35nI57Ny5U9JCla/KrzeLqVmPyv9lpEl4o729\nvYH2VKvVBPLR6XTI5XKyDmyiiEajwmnNZrMoFovIZDICrfFCWs+nzufzOH36NIxGowQThK/4zDfi\nFzaceqsbjoUbFcRtaWkRzh75YarQAlMdRoXVahWzs7MIhUKStmcyGRw/flyKDOoD4c9XKhXMzs4K\n3Yjvj+9DdeKboXOEpm40tZtBJZwXi0Xo9Xr4fD6JtmOxGAwGAwqFAnw+H7Zu3Sotdzt37sTY2Bhy\nuRx8Ph9mZ2flQiI3UFVuIeB+6dIlOcxMXRwOh+B4myEtBNZEGRitq0UdlWNJdoFOp0MqlWoopKmc\nXabmV69exZ49exAOhxEOh3HmzBn88pe/hMlkkkiIlx//nEwmEY1GodfrcfHiRQkkyEslVqmm/c1u\nvMxpalbEr9l119LSgtbWViSTSUmp6R/U/Z7NZhGPx+H3+8WJZjIZnDhxQiJvlfNKP5RMJnH16lXY\n7Xa0t7fLs7ZarVIkVqPem72Q3nO7Cp0dsMah5IE7ffq0YBQssKhN6Cqxt1qtCnl8amoK09PTOHjw\nYENYDqw5CoPBgGg0KsWLlpYWKQLpdDqUSiVYrdYGmSoehmY2FXNVbzwecv7ZbDbDarXCaDTi0qVL\nEoUHAgG5ad1uNxKJhDyD2dlZ4eS53W588pOfxIkTJyT95sWSzWYRDAYxOjqK69evo16v43vf+55U\nGdVuKx58lVTdzMaoA8ANQhjqHpuensbevXuRyWRQr692RREnY5dINBqFyWTCwMAAxsfHsXPnTkQi\nEQwODiIWiwl1So1W0+k0arUaAoGAFMZI12IlXXWOmy0F5/kn5AOgIVhZXl7GW2+9BaPRCJPJBLvd\nLv5hvS4DHRuLOVeuXMH4+Dgef/zxhiiSvxdYPT/Xrl1DuVxGf38/5ubmGuA+ZmCES9bXAt71821k\nMXjzqmx4lQzNiJM9rQS2+XOMBIvFIrLZLBKJhBR34vE4pqensW/fPuj1erhcLjnEKh7J2ykQCCAW\niyEajUr05Xa7BVtTD8FmkVlTPydvOzoxciQB4Gc/+xlKpRKCwSAAoKurS3ilwKpTSCQSmJmZgdVq\nxenTp/H444+jUqng6aefxpkzZ+T7iC0NDw+jWCxiamoKIyMj+OQnPykYG7sltmzZ0iCjp6rDNLvx\noua+UA8ID+fKygpefPFFpNNpBIPBhkPY0tKCS5cuyZpYLBap9JJbaTAY4Ha7G4IDq9WKdDqNnp4e\niczD4bCsPQBMT0/Le1Tf02aAjAA0OEOVSqXilPxsDodDKtGUOWN6XCgUEI1GMTk5iWw2i1QqhXK5\njGQyiYceeggulwtOp1MCI+5DQktutxsDAwM4duwYksmkNKRs375dIA41s33fUm/+ovVemJuMf6/X\n67Fv3z4UCgW0tbUBQEN1kKTmfD4v3j0ajaK/v1/AWi40Pxg36fT0NLRaLRYWFgCs3b4mk0kEIdTQ\nmtHPZjA+OMIJapSsanuSrNze3g6NRoPFxUVppWtpaUE6nUY4HIbRaEQ0GoXf78ePf/xjXL58Gfv2\n7RPeH7Hg1tZWPPvssyL4u3fvXjz99NOw2WzQarXIZrPSGMB0VcXTmj1aBxpxdaBRJEPtGvvyl78s\nVVJGLSzidHZ2IhgMSmGGdKkrV66gWCwinU5LlsVLLhAISK9xtVrFlStX4HQ6JSOr1+tCH6Ixulyv\nxtPMpp5//l/loAKr69jR0YF0Oo2tW7cKrU+j0cBmswn1inuTrAAS9NPptBTSuBdfeeUV+Hw+RCIR\nmM1mvPnmm/D5fJKKm81mLC4uNkTp7D5730QxAAhBXL0lWG7nA9Zqtbh48SJyuRxKpZLQgcLhMKLR\nKLxeL/L5vFSh6vU6QqEQdu3a1SBrpeKfoVAI6XQaRqMRFy9eFHyUEZjD4YDH45FbXo3G1BSomU2F\nJNTqHLCGq9XrdQQCAVy+fBkAhNdoMBjQ1dWF+fl5ScH379+PfD6Pubk5xONx3HHHHdIEwOIQgXO/\n349cLoe2tjZ8+9vflvdDDNPtdgtWtx6wb3b7n/YA9yqr0aykjo2N4fr169KCaLFYMDc3B6PRiEwm\ng87OTmzbtg0GgwE9PT1CvCe8oSrYxONx7N+/H06nE/39/eju7pZIXK/XI5/Pw263S9rJiHczrKtq\n9AvcU6p/UNuez5w5g0KhgGQyKVj45cuXEYlE0NbWJoT8dDqNlZUVdHZ24uDBg0in01JvYGAErHY5\nhcNhrKys4KWXXoLT6ZQqOgAEg0HYbLZ3nBKwEXtPLYxAI6bGr7nZ2Hcci8WQyWQE57Hb7bJ59Ho9\nxsfH8corr8DpdMJut6NarSKbzTZ8GG5ypi1WqxWZTEYWjFX0WCwmkIBaOaSz3Qwbj+ksAOlr52ZT\nqVY6nQ7nz5/HhQsXsLy8jI6ODhSLRZw4cUKwHbfbjXg8Dp1Oh9bWVnzoQx9qiCLJx1y/vjzwxIyI\nHxGbVLFK4lLNbqRYqXgfGyFUEjqx2NOnT8vnCgQCwirgPpqbm8Pp06eF2MzohBErX7daXdVb9Xg8\ncLvdkj2pKvaENRwOh0RgjJqAGxXUm9VUorkKmZGPyuzOarWipaUFuVwOLpcL8XgcXq8XMzMzojk5\nMTGB3/zmN9i+fTt8Ph8ymQwymYz8LpU6SJK6w+HAzMxMg7xboVBAoVCQfatWulVM+GZsw46SoLha\nZgfQ4K0ZFt99992C0RQKBdjtdrS2tsrhNhgMuPfee9HS0gKj0YhYLCbEXjoK/g5yzTKZDBYWFiTd\nZz+nx+O5obNFTQWa/UDz/XJjMXJTyd109vl8HnfddRcSiQT27duHyclJRCIR9Pb2Cg8tkUggk8mg\nUChgaGgIVqtVUniKaKjOLhAIwGazYWJiQhxjpVIR+MRsNkurX0tLi1BlNoMMGLBWwFEPMonedEaU\nnHviiSfgdDolgvZ4PCiVSqJ/SIWalpYWRKNRVKtVmb+jClmwal4qleDz+XDixAl5L7ys/H6/UNtW\nVlZgtVrhdDrldTaDTgGw5hfW1we4viyiAEBHRwdsNhvC4TASiQRsNht27NghmWFHRwe+9KUvQavV\nwmKxYH5+HvV6XdafGDAA8ReRSETkGpla9/f3C82L+1yFRjbiEzZMD1KxKVIgVGoJb8xgMIiFhQVc\nunRJOhOYDlssFvz6179GMBgUgi9blJLJpIyH4KHWaFb7Qanm7Xa7GygIhUKhQXxVfVjEM5q9eqhG\n5Co7QJWt5224srKCCxcuoFar4Y033sCePXswPDyMq1evIplMorW1VRzn3r17sWXLFuzcuRNzc3M4\nd+6cbBz2IE9OTgoPc3x8XKS++F7IQOBlyLRKhTaa3QjVqGwClepGGMdsNmP79u2YmppCJpMR1SWy\nDebn51GtVuFyuaDRaODz+XD16lV4PB74/X48++yzDU0RCwsLeOONN+B0OhEOhwFAHHA6nRbiOdeR\nnSnsbWbHVTMbU26VskPetFpMNZlMGB4eRjKZxLVr1+ByudDR0SEtshaLBa+++iqA1YFwzGbS6TSi\n0ShsNpvweVUaF3m+g4ODDXiv2WyW9VNZBHyfqtbtu9mGq968DVXaCrDGXdRqV2W50uk0AoEAvF4v\nFhcX8fbbbwvGODY2hp07dwrH0uPxCK8vk8kgGAw2dOywiOP1epHNZqHVamGz2YTPF4lEhOS7XsOP\n3M5mjyiBtfksXFfexPxMhUIBCwsLGBoaQn9/P7Zt24ZSqYRcLofR0VF0dXVh3759KJfLiEQiWF5e\nxqFDh9DR0YH/+q//EsA7k8kgn89jeXkZCwsLcLlceOqpp4RixE4Hpkp0kLwY1Yip2QWRgTXh3PXZ\nED8Po2fOA/rd734nTRAGgwGXLl0SZe3Ozk7B3snbo4L/tWvX8OCDD0oUWCwW4ff7odFoMDc3h5aW\nFlGtoYMmd1jF43mJLS0tieBJMxsjNToeQkRkC6j6C7OzsxgcHJTRIkePHpXI/ne/+x12794tF7HP\n58PY2Bjy+Tyy2Sy2bdsmqmPVahUzMzM4ffo0AoEA0um0aLLSMU9NTaFUKt1AB1Ox1Ju19zyFkQea\n6W6lUhH8cHx8HE6nE5cuXYLH48Htt98uEeX09DQ6OzuFS0Wy9PLyMqLRKPbs2YNkMim/Z3p6GtFo\nFKFQSPq/Dxw4IOE41UDUD80bRSWcN3vUwyhkPZGb8EJXVxcSiQSCwSCOHz+OxcVFwRZjsRjy+TxS\nqRQmJiZQLBZx8OBB5HI5GdW7uLiInTt34p577kGxWEQqlYLD4cD8/Lz0I1NbkU6ErYzkW66HXAh/\nNLtpNBqhh6jVT2Y/1WoVNpsNJ0+eREdHh2Bfvb29sNvt8Hq9OHv2rLQjktqyuLiIRx55BMFgEGaz\nGffee6/glT6fD3v27IHZbEZPT49wA8nPJAGbgcY7cSb5fDeLqY0lVA5iO+jCwoI0R7z88stwOp34\n4Ac/KPoMV65cwZ49e+B0OuH3+2E0GnH9+nXkcjnMzMzg05/+NE6ePCn1j+npaSwuLuLOO+8Ux3rn\nnXcikUgAWF3PYDAoDpsBl3pJboQD/J5m5pA7SSO2ViwW4XQ6YbPZ5CZZWVnBxYsXZTORahEMBuFw\nOOTW1mg0UtmOxWIyTpKcSuIQNpsNg4ODCAaDsFgs6OzsbCC7cpEYXZIK0+w0CzUd4PryP4LfHR0d\nKJVKiMViiEQimJ6eRj6fRzweF/GQhYUF1Go1DA8P4/7778err76KLVu2YNu2bSLo4PF4BAvdunUr\ncrmcZAT5fF4wOI5CUJWpgTXgnuMOmp3vp0aUwI3DqKiCFAqFZD9y/127dg21Wk1SQZPJBK/Xi7m5\nORgMBhkx4PV6RQnIaDRiYWEByWRS+MCkxwWDQdRqNczNzQmsxKBDhTHovDdDMYfvm7UF1TdQIzIQ\nCCASiUiE3dLSglOnTmH//v0wm83wer3QarXYsmULfD4fRkdHYbFY4Pf7MTQ0hAsXLghOyWIZ2R+L\ni4twuVwYHByE3++HxWJpaA9VGzfok8rlstQ1bsbeU+qtijXwwZpMJlitVmSzWZRKJZw/f16qeHa7\nHR6PBzabDQaDQXpk2XnDwo3T6ZQNxcVmZZZYA7l+LpcL2WwWY2NjDb3kauTDljsVO2lWY3SsOh5C\nC3a7Hel0Gtu2bUM8HsfOnTuFIdDZ2Yne3l7phff7/TCbzSgWi/jJT34iz0lVRwdWnR2jI0Y33PB+\nv1/oF9yUKr7HPnymks0Oa/B9vxMthBXThYUFaYaw2+0wm824evUqpqamsH//fgkEKpUKSqUS7HY7\nlpeXkclkkM1mRbFcp9PBbDajUCggnU4jnU5LUaxarcro1Uql0iD3pXYH0SETTtoMpl7uzODq9dVO\nMtUhvfnmm9KQ4nA4JHokoyWXy2FqakqUgqxWK9xut+CTzBApRceokVivyWRCqVTC7OysjKtm6q1i\nqaFQaENru+GIcj3fix+6XC7DarWKfFo+n5eyfCQSgcPhwJEjR6DX63Hw4EFJF7l5CbarhGAOF+KH\nY4TpdDoxNjYmD4d0CwAN0Y1a/W72w8yoh3gqP3+tVsPo6Ch27doFnU6Hnp4edHd3S0Qei8Vw8uRJ\n+P1+qTJ2dXXhwoULklbymXBYFde8WCw2wBTEbRh9cSgbLyJiqMTPGC1tBiMOyUOs7mGTyYTu7m4A\nqyOPuXdbW1vR39+Pn/70pzKkbnBwUAqY3Lfq/Hhiv9lsVv6OToIQBtdOjXiANZxfFfndLBiwWlPg\n+nKsr8ViQTqdRi6XQzKZFPHo69evo62tDUePHoXVasX+/ftx7tw5OQMqjxdYWydmroTi+B4sFgve\nfPNN5HI5aVnkv/E11Eh9I3zK90w45ybTarXo6uqC3W5HV1eX9FhTzp44mc/nw5/+6Z+io6MD4XBY\nAOB3ohQwBSXOQPY+1W8cDgeuXbuGjo4OdHR0NKhK00wmkyi2bIbNBqwJGBM24ETEkZERqQbm83lM\nTk5i+/btcDqdKJfLAmR7PB74fD7ph2UkSYfHzcM15kYm348RJvHKrq6uG9aOajgkBwObozOHzAFV\nEIPrzHUEVuGEQCAAjUaDZDKJXbt24eDBg5K2j46ONmigMgNgOl+tVuUAcw4O6VRtbW1IJBLCm2Rx\nScX7qZgFrKlrbQZbn83V63Xs3LkTdrtdBKCphWqxWJBMJmEwGOD3+/GFL3wBPT09Mr+JWaOK2zNj\nJKbZ0dEhSmF2u1242vF4HFu2bEFHR8cN+HmtVoPT6ZRijyrk8W72nkZBqBLvJpNJhABIgF5eXsYj\njzyCEydOYHFxEX6/H1evXsVbb70ljPt6vY5UKiVRCheG0l7JZFJu7nK5jEKhgEwmg3g8jrfffhtd\nXV3QaDQNghhqJYutjKweN7upFVBgbZP09vZienoaBoMBo6Oj2L17N1paWnDs2DEcP34cdrsdDocD\nZ86cgdPpxJ49e3Ds2LEG7CWfz8vlRjEGdZKjw+FANBqF2+2G1+tFsVgUUQ06VW4qlev5XpSif1+2\nsrIiFy0dE/u2KVRss9mkwHLt2jWZN0Tsl6kdRS9UpR+upcvlEiiJdKPFxUVpZ4xEIg1TH4G1Qg4j\n0PV8z81gvNh5EVksFkxOTqJWq0nnTalUwv33349z584hkUjA7XbjpZdewrlz51AsFjExMQG9Xo9I\nJNKwr1ZWVgTmuH79umQDHNtRKpWwuLiIF198EUNDQ2htbW1gFgBrYjP0BcyMb9Y2HGq5XC4AayV2\nVS6N7V6tra04deoUrFYr7rrrLly6dEkEZycnJ+F2u2E0GpHP5+V1edu3tLQIJpHJZIS/x6/NZjPm\n5+cxNjYGn88nYyfVTgYC78DmIJsDEKDa4/EAWEsz4vE4BgcHcfbsWZw5cwbpdBo63arArNVqxdTU\nFEKhkIyZHRsbg9VqFRhkZWVFZlEzuuFGDoVCDe1k6ixph8Mh36t2nRBz5nvcDBgaI531HGDqCnAI\nHZ3Y8ePHsXv3buRyOVSrVVgsFnR1dYkuJVsUSWzmbPNCoSD7l+SiF/SFAAAUlUlEQVR/s9ksjs9g\nMGBqakoGw9EIh/BZ0UFuFvoVgIZLCFibispIsKenBzqdDidOnEAgEMCHP/xhHD16VObXjI+PCxSR\nz+flmfF19Ho9gsEgnE4notGo6H6ST2k0GjE/P4+zZ8+ip6dHlIKYAVcqlYYpAOr/b8beU9WbuAv5\nftwkHGuaSqWEFnT8+HEkEgmRRtq6datgmIVCoUFogYvKDcYKeS6Xk1Ykttd5vV5YLBZJg9QPzjYp\nYHNIrNFUp0OnRuB/bm4O+/btg1arFSC7p6cHADAwMID77rsPPp9PIIdUKiVRDonjFPTl+gGQfmZe\nfAaDAZOTk+jr6xNMjdkDKUN8BpvJ6ORVNgGj9kgkIjNvQqEQrFYrPB4PlpeXYbVaEY1GMTs7K1kK\nZf5U1XmVB/zyyy/j0qVLqFZX5xOp0fvi4qLMzuFZYibBNF7lJG8GU3m2bERg5lGpVOB2u2E2m0XG\n78EHH8QPf/hDpNNptLW1wel0oqurSxgdqt6pmglRNNnlconPoZo8I0YKuVDLVsWIqaWq1i5u1ja0\n24m7rB9vSqD88uXL8Hq9Mu9Yr9fjwIEDSCQSIu9OnCydTstcXbYVsceVfeIcQuZwOJBOpwGsduWM\nj4/LYrC7hAUHFTtVgd5mNxZJeGmwEJPL5XDlyhX09/djcXFRKqIstgwODgoJmp1LKysrQrSlcjl/\nRzweh81mw/z8PLLZLAKBAFKpVEOLVzabFT4hsT2VN0knvlkOMyEaYG2ONyuyGo0Gfr8fc3NzKBQK\ngnFFo1FcvnwZBoMBZrNZqFn5fF7gCfZuk2aVSCRgt9vR09MjODufq81mQzKZlAKm+t6YDVGfkX8P\nbA7hXrYVAmuFE17MS0tLuHbtGtra2rBv3z6cOnUKLS0t+MxnPgOPx4Nz586JTB2pRK2trbIXNZpV\nLQcWiyORCDwej7ADmOnYbDahIbJnX51myQtJnR/+vlW9a7Wa0EJ4g9RqNUxPT0ur2/T0NF555RWJ\nKg0GA8LhsBQPSPp0Op1SZOCtQf1KUmII+jIt5O1BfIhal0CjOAfJ7fx6MzhKABJNA40SceyOMZvN\nGBoaArB6YdTrqxMtY7EYjh8/DovFIhuCEQsnYQKrkTbVlVwuF26//XYhnAOQWUPRaPSGfnMAcvjV\nPvHN4CiB1VnZPHw8zJSiY4W6Wq3C5/MhlUohGo1Kuk1cnV1ptVpNngn3XyKRkLVg1bVQKDT0wpPJ\noOLyKtuAKT1N5QY3s5G8zQyEf1ev1xGPx5HP5zE7O4t/+qd/Qi6Xg9/vh8/nw7PPPitpNsc0eL1e\nwTq51haLRWQEg8Eg3G63NEew4GmxWOB0OuHxeGAymQRTp7IZsMp1VTPMjVxCG34K68NsVr2DwSC6\nu7ulAhUIBHDmzBlcuHABi4uLyGQygkEuLS1JZKpWIBmpECtbWloScdpcLicpTigUEqcArFXigbUR\nogyxN4vRoZM3SgyMhQFKRaXTabhcLpGgqtfrmJiYQL1eRyaTgcVikVG2xODUDgR1YiOjdK1W24Bj\nMj1nGqiqrquDn+hcNgO0wYOsRhNutxu1Wg1dXV2wWq1wOBw4ffo0ent74Xa7MT093VCR5sVMfHhu\nbg6pVEoELer1VZV4fg/TTwYA6uWjNmTQqEPKCJMNE5vBWBxjhE0Oc09PDw4cOCCfs6enBy+99BKO\nHz+OSCQiKk50eISCqtW12d+8/NlSyxG2AKQtmu2i5E2rgRwv++Xl5QZnvpF9u2FHyZsUWLs1VlZW\nMDAwgFwuB6fTCa/XC5/Ph/3798Pj8SAWi4kyMUFfYK2ym0wmG3Qu1dEDLBiQAM0F58/SuFGXl5eF\nZsCoymq1Nn2rnVqRU2WpqIVIjHZpaQkTExMwmUwYHR1FNBptAL0pn8bNUiwWUSgUkM/nG+gxZCfQ\nOPZWJfdTTUelqKhDssgtbHa8kg6L0/+413bu3IktW7ZgcXERVqsVly9fxqc+9SksLi7i+vXr2Lp1\nq1zsAASfpZbnyMgIhoeHhfTMA2g0GjE8PCxRaLlcloyLBzkej8u4YZWbyn3K6JVCGs1uvGiBNaWm\npaUlHDhwAOVyGcFgENu2bYPVasXIyAg6Oztx+fJlhEIhVCoV+Hw+yRqXl5eRz+cxPj4u66PX6+Fw\nOJDL5VCpVBAKheTyUp8RLys6SdKWKpUKZmZmxJHTEbPo+2624c4c1eFoNKvqKZyzsrKyglgsho9/\n/OOIRCK4dOkStFot/uzP/gzHjx8XoicPMbEj8iS5yCRes8k9Eomgs7NTquUqjYibXu0WIh2GDv3q\n1asb+Zi/FyPdJpvNSsqRz+exe/duKd6kUimYTCYZaFUqleD3+2UEKNAYXbPy5/P5JAJnxM0qbF9f\n3w2sASo6FYtFaVPk81IriV6vVyLZZjby9WKxWAOUkEgkMD4+DrvdjqmpKWzfvl00BAKBAOx2O2Zm\nZoRjSsiBe3VsbAyzs7OIRqMCm5DL99xzz+Gee+7BY489Jo5yaWlJ9nRXVxd8Ph9qtRpSqZQopDOC\n5YjVrVu3/j6X7qaMNQpGyQDQ19eHjo4O/O53v5NJnp/73OcwPT2N+fl5FAoF/MM//AOeeeYZmRbK\nPcYiUFtbm1DQKFdnNpvh9/sFkxwZGZEisl6vl4YBBlosiFarVUQikYZI8+jRozcNbWy46s2pafza\n5XIhEolIytHa2orJyUlJdTQaDX79619jYWEBqVQK2WxW2sAAYG5uroEOof4uOj4qily8eFG4bWaz\nuQGjZIqZTqdljO38/Dyi0ajQWZrZdDqdqCOpt6Hb7ZZoIxaLYWhoCLVaDXv27EF3d7fQgnp7ewWz\n0Wq1ot1HyINQBC8XduC0tLRgy5Yt2LFjh6TigUBAOIFc30KhIE68Xq/D6XTi6tWrItvWzFapVNDf\n3y+amhqNBqlUSlLd06dPIxQKwePxIBKJSBTOVs2DBw829Lezokr9Q64tL2tGkr/+9a/xy1/+El//\n+teF2ma1WqHRaASbc7vd6O7uFh1GUr9mZmakvbTZrV6viy4ATatdHd9SKBTQ2tqK1tZWHDlyRObj\nmEwmPPfcc5ifn8fk5CQymYx0kAHA/Py8XBrvpC2xsrKCT3/609DpdPiP//gP5PN5xGIxaDQaURZj\nXSSVSmFxcVECjrGxMVy5ckWYHzdj7ymi5Jsmiddut0On00lqXa1Wce+99wru4HK58H/+v/au5aep\nr4subnn0RR9cS4EKhEBFwRh0AEGMiTrzMTQxceaI/AY+Ro78D5yZODEk/gE4cKROfCSIiZAYYoMN\n71dR2qKURwst0N+gWbunfH4BvsRY8p01IqSUe889d5+991p773/+wfv37yVETCaTME1TwmKV8VXl\nQlx0AOjq6oLf78fOzg78fn8BmcDEt5r49Xg8R0brt729jerqagnrmP9dXl6W2SJerxcjIyOYmZmR\nEk7TNGGaJgKBgFR60Lu0WHLzdsg+Utaj6iPpeTscDnR1daG0tFSIDxpezjEpKSlBS0uLtBebnp6W\nOt1iBofXqemN8vJyyefW1dUhHo/D6/Xi3Llz0o0qGo2it7cX4XC4gC0vLS2VF5tryDQPDyG12OHx\n48d49uyZVFAxNKUHSplLVVWVeO1kkfksixkqx8B1qK6ulkF/P378QG1tLXZ3d3HhwgVhn/1+Px48\neICJiQkhK5PJJI4dOybMN/OTXH+uG0PobDaLu3fviufJGUUAhCRzOp1wuVwFY4epzTyobTiUoeSJ\nSY+trKwMHz9+lIve3d3F1NQU5ufnpSVSJBLB5uYmEokEQqGQ5HnKysoQjUaFpaKXquoiVbkPkDu5\nXr9+Da/XK4l4ID/g3mazSS9LyjdUvVoxg7IpVb5QXl6OsbEx8cDJ+JPg4XqrYSVlFpRVkEHk5qHB\nU3PCXG+r1YrV1VVZWxpLegwkfSorK7G8vCy198UuYaH+k4cC9aJs/sGDPx6PY2pqCouLi7BarVIQ\nMTAwIHIsICeIdrvdcLvdUqWkavP4ffzZYrGgt7cXFktuLId6sLAklN/ldDoRjUZ/27ugWMEO/G63\nW2zBhw8fRNNL9nt0dBQjIyOw2+0Ih8PY2tpCJBLB6OioOEo8oDgagyC5S/E4STG+1+FwGA6HQxwp\nAOJwVFRUyIwjdspiocAfqfVm/oD6M8PINdD99euXzCdOp9PiVTJsqK+vx+TkJPr6+iR0czqdksCl\nqFk1jPxbitB5Q1evXsWtW7cKJB0ARFKkdmNRa3KPwobby95ns7kyTx5OTU1NiEaj2NnJjR4gSROP\nxzE4OFjA8DEtws1HFlutL95bNheNRmXuyN7nwqa1JIxohI9KY9mdnR3ZtwzlFhYWhDykd0dPc3h4\nGMePH8eLFy/w8OFD1NXVYWNjQ5wEVtBw/6maYuA/h8OVl5fj7Nmzsm8BiJGlvpJ7lMqHYicgCR4G\nTCvQLlCY73Q6EYvFcPLkSdFXV1RUIBgMIplM4smTJ7JeVMzQ+7PZbBJx0hME8rpjrtnNmzdx//59\n2d/c19vb2wgGg3KtDM9ZBfVHDCW9PvVi/X5/AWvHxKzFYkEikUAwGMTS0hJCoRBCoRAaGhoQi8VE\nNK3We6vt+skA7t14JHtIANHroRgegExsAwCn01kw8LxYwReNqgIeGh6PR1QAc3Nz8Pl8ME0TLper\nYLZ3Z2dnAZvN8M3hcIhHqb58XDsy4UBuje/du/dbL7GyshJbW1tCfKittIodFotFph2qqR3qQq1W\nq5TNcpQs19Nut0tVDlt3sdcBc54Mx9XOT0BeoUHv/PLly3IAqddAj53zpahGOAqHO4CCewdy91tb\nW1uwx1gjz/tubGzE+Pg4Xr16hVAoJIPGVOma1WoV+RD3I5UYQD4VyENubGysoGM8o152K1Krovjd\nB42G/qcO56xfNQxDSBm73S4NKqh3TKVSGBoaQlNTE65cuQKPx4NIJCIVOXSLCdWgMf+4twIIAJ4+\nfVogbaFRiUQiSCaTYmgovzgKoyBo8EmgcB0ormc9MD0/9uTb2trCzMwMTp8+Da/XKzkvai3V8b1q\nXT4AEfWqwnKSdXyu3Egej0dmqVP/WlpaeiTIBu41NiQGUECsUH9qt9uFwbfZbAiFQtjd3ZUBYuwq\nRO+c78HeA4OEEXOaRH9/P1KpFObm5jA7OyvEXX19PZLJpBhRVcmhvh/FCr6jrCajXSDLz73EfrXT\n09N4+fIlenp6cOfOHSSTScRiMVRVVck0UL7HQKEkkNHS3t8BwPPnz0XRAeTtyZcvX2AYhpRMs3eo\nqoDYD4cOvWkImUebnJwUL4/5iUQiIVUgZ86cEQZ2eXm5INRLp9NygjPxql64Gj5TVpTNZqV3IJlJ\ndZg5r49hFHMkxe75kKlWm/fa7Xbxqmn01Rwbw+ru7m68ffsW6XQam5ubcLlcqK2tBQDx8FXxOJAv\nF90rzk0kEhgbG8PKyoqoGdgFhqMMAEiao9jXFYD0jGQXbe5Tzr1hJJNIJGT64cDAAG7fvo22tjbJ\nZdEgsJSOoRv3r5pT4888tLPZLObn5+Hz+dDW1obW1lak02m8e/cOgUBAWr1tbm5KpQnTVMUOem5s\nL2cYBiYnJ5FKpbCxsSGe+69fv+DxeGCz2XD9+nVMT08LKUYbAeT2ZkNDg0SZDO35/jMKosaS6Ozs\nhN1ulwmkhpFvf7eXQWdhy0E7CB3KUFLyA+RPEdM0UV1dLc0BbDabJGJZrE5jyptl9+L19XW5aIYw\nakjC/6m68Kyo6OrqApCrZojFYujo6BDDwiQ8GeDDuNh/C3yIlCxw87GEkawdiYlMJoNIJIITJ07g\n69evBdKU1tZWKScF8gQZ9W4ARNenVtbs7uZKyXp6eqSEdH19HTMzM2hubobP58Pk5CQymYx4tUep\nMsftdhc0e+YMdO4rCphN00R3dzf6+/vx8+dPAPkIp6KiAvF4XJoW/y6loXro3Hf87I0bN5BIJJDJ\nZNDS0iLNICwWi5A4LpdLZF1HZW2ZSuP9smcki0WojikrK4Pf75eBgUA+lUe7sbCwIJVlQP6Q30tA\n8tDjepumifb2dvns1NQUrl27JpECw3bDMESm+EdYb3qB8XhcjNfExARWV1fh8XhQW1sLwzDQ1NSE\npaUlDA8PS/6BJzUJhXQ6LcPF6AbvJSO4SOrJys/39fWJka2srMTq6irm5+fF4+FpokoXihlss8WQ\nGch3EGKIRjkKB341Njbi8+fP4lWvrKzg1KlTGBoaQjgcFq+SHjk3LU9/rg2rQGhMa2pqRDHgcDjQ\n0dGBVCqF+fl5uYbKykqpaCl20iGdTmN1dVXuF8hr/0g80Ns2TROzs7MIhUIwTbNAXWG32+FyuUSP\np0pV1FI7RkbqnuXPzHNS89ve3i7zYejlsAqLhrzYwcP7+/fvcmDPzc0Jmx8IBFBaWorW1laEw2G8\nefMGnz59gs/nk74M3L9bW1uor68v4CS4zmoJKIACYpLG89GjR+KMkUQOh8NYXFwUMprCdZUM2g+H\nrj1Tu4nTuyPxMj4+Lp1pampqcPHiRXno3759K1DK84RWWWuVdOD38wRQLT83JpO8hmFgZmZGqhn4\nnZR+cDMXM8gsNzc3I5FIoKQkVy9M8oCqAh4Us7OzojLwer0itF9bW0NDQ0NBmA1A1pybTu28xNQE\nFQYrKytIJpMyp8TtdmNqakpymlVVVdJh6KChy98EZVeBQADDw8MA8jl1CqC5jwcHB2G323H+/Hks\nLS3B4/GIpI0dcVT9HcNu7tW95Jja1ovVICUlJdL8N5vNIhAISKcbq9WKtbU1OBwOCVuPAlhJxnun\nIiKTychkxIqKCly6dEkIXnWON+dp0YtkPh2A2Az1f7EklEaS70UwGBSixjAMjI6Owu/3y8QEtmqj\nY3LQSLPkMA+ipKQkBmD2wH9QXGjMZrO+v30R/w16bf8cjvjaAnp9/yQOtLaHMpQaGhoa/484GkIt\nDQ0Njb8IbSg1NDQ09oE2lBoaGhr7QBtKDQ0NjX2gDaWGhobGPtCGUkNDQ2MfaEOpoaGhsQ+0odTQ\n0NDYB9pQamhoaOyDfwGS9jciTRYYtgAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x113840cf8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# coding: utf-8\n",
    "import sys, os\n",
    "sys.path.append(os.pardir)  # 親ディレクトリのファイルをインポートするための設定\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "from mymodule.simple_convnet import SimpleConvNet\n",
    "from matplotlib.image import imread\n",
    "from mymodule.layers import Convolution\n",
    "\n",
    "def filter_show(filters, nx=4, show_num=16):\n",
    "    \"\"\"\n",
    "    c.f. https://gist.github.com/aidiary/07d530d5e08011832b12#file-draw_weight-py\n",
    "    \"\"\"\n",
    "    FN, C, FH, FW = filters.shape\n",
    "    ny = int(np.ceil(show_num / nx))\n",
    "\n",
    "    fig = plt.figure()\n",
    "    fig.subplots_adjust(left=0, right=1, bottom=0, top=1, hspace=0.05, wspace=0.05)\n",
    "\n",
    "    for i in range(show_num):\n",
    "        ax = fig.add_subplot(4, 4, i+1, xticks=[], yticks=[])\n",
    "        ax.imshow(filters[i, 0], cmap=plt.cm.gray_r, interpolation='nearest')\n",
    "\n",
    "\n",
    "network = SimpleConvNet(input_dim=(1,28,28), \n",
    "                        conv_param = {'filter_num':30, 'filter_size':5, 'pad':0, 'stride':1},\n",
    "                        hidden_size=100, output_size=10, weight_init_std=0.01)\n",
    "\n",
    "# 学習後の重み\n",
    "network.load_params(\"params.pkl\")\n",
    "\n",
    "filter_show(network.params['W1'], 16)\n",
    "\n",
    "img = imread('dataset/lena_gray.png')\n",
    "img = img.reshape(1, 1, *img.shape)\n",
    "\n",
    "fig = plt.figure()\n",
    "\n",
    "w_idx = 1\n",
    "\n",
    "for i in range(16):\n",
    "    w = network.params['W1'][i]\n",
    "    b = 0  # network.params['b1'][i]\n",
    "\n",
    "    w = w.reshape(1, *w.shape)\n",
    "    #b = b.reshape(1, *b.shape)\n",
    "    conv_layer = Convolution(w, b) \n",
    "    out = conv_layer.forward(img)\n",
    "    out = out.reshape(out.shape[2], out.shape[3])\n",
    "    \n",
    "    ax = fig.add_subplot(4, 4, i+1, xticks=[], yticks=[])\n",
    "    ax.imshow(out, cmap=plt.cm.gray_r, interpolation='nearest')\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}