dlnd_face_generation.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Face Generation\n",
    "\n",
    "In this project, you'll define and train a DCGAN on a dataset of faces. Your goal is to get a generator network to generate *new* images of faces that look as realistic as possible!\n",
    "\n",
    "The project will be broken down into a series of tasks from **loading in data to defining and training adversarial networks**. At the end of the notebook, you'll be able to visualize the results of your trained Generator to see how it performs; your generated samples should look like fairly realistic faces with small amounts of noise.\n",
    "\n",
    "### Get the Data\n",
    "\n",
    "You'll be using the [CelebFaces Attributes Dataset (CelebA)](http://mmlab.ie.cuhk.edu.hk/projects/CelebA.html) to train your adversarial networks.\n",
    "\n",
    "This dataset is more complex than the number datasets (like MNIST or SVHN) you've been working with, and so, you should prepare to define deeper networks and train them for a longer time to get good results. It is suggested that you utilize a GPU for training.\n",
    "\n",
    "### Pre-processed Data\n",
    "\n",
    "Since the project's main focus is on building the GANs, we've done *some* of the pre-processing for you. Each of the CelebA images has been cropped to remove parts of the image that don't include a face, then resized down to 64x64x3 NumPy images. Some sample data is show below.\n",
    "\n",
    "<img src='assets/processed_face_data.png' width=60% />\n",
    "\n",
    "> If you are working locally, you can download this data [by clicking here](https://s3.amazonaws.com/video.udacity-data.com/topher/2018/November/5be7eb6f_processed-celeba-small/processed-celeba-small.zip)\n",
    "\n",
    "This is a zip file that you'll need to extract in the home directory of this notebook for further loading and processing. After extracting the data, you should be left with a directory of data `processed_celeba_small/`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "# can comment out after executing\n",
    "#!unzip processed_celeba_small.zip"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "data_dir = 'processed_celeba_small/'\n",
    "\n",
    "\"\"\"\n",
    "DON'T MODIFY ANYTHING IN THIS CELL\n",
    "\"\"\"\n",
    "import pickle as pkl\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import problem_unittests as tests\n",
    "#import helper\n",
    "\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Visualize the CelebA Data\n",
    "\n",
    "The [CelebA](http://mmlab.ie.cuhk.edu.hk/projects/CelebA.html) dataset contains over 200,000 celebrity images with annotations. Since you're going to be generating faces, you won't need the annotations, you'll only need the images. Note that these are color images with [3 color channels (RGB)](https://en.wikipedia.org/wiki/Channel_(digital_image)#RGB_Images) each.\n",
    "\n",
    "### Pre-process and Load the Data\n",
    "\n",
    "Since the project's main focus is on building the GANs, we've done *some* of the pre-processing for you. Each of the CelebA images has been cropped to remove parts of the image that don't include a face, then resized down to 64x64x3 NumPy images. This *pre-processed* dataset is a smaller subset of the very large CelebA data.\n",
    "\n",
    "> There are a few other steps that you'll need to **transform** this data and create a **DataLoader**.\n",
    "\n",
    "#### Exercise: Complete the following `get_dataloader` function, such that it satisfies these requirements:\n",
    "\n",
    "* Your images should be square, Tensor images of size `image_size x image_size` in the x and y dimension.\n",
    "* Your function should return a DataLoader that shuffles and batches these Tensor images.\n",
    "\n",
    "#### ImageFolder\n",
    "\n",
    "To create a dataset given a directory of images, it's recommended that you use PyTorch's [ImageFolder](https://pytorch.org/docs/stable/torchvision/datasets.html#imagefolder) wrapper, with a root directory `processed_celeba_small/` and data transformation passed in."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "# necessary imports\n",
    "import torch\n",
    "from torchvision import datasets\n",
    "from torchvision import transforms"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_dataloader(batch_size, image_size, data_dir='processed_celeba_small/'):\n",
    "    \"\"\"\n",
    "    Batch the neural network data using DataLoader\n",
    "    :param batch_size: The size of each batch; the number of images in a batch\n",
    "    :param img_size: The square size of the image data (x, y)\n",
    "    :param data_dir: Directory where image data is located\n",
    "    :return: DataLoader with batched data\n",
    "    \"\"\"\n",
    "    # Define the transformation for size, square nd tensor transformation\n",
    "    transformation = transforms.Compose([transforms.Resize(image_size),\n",
    "                                         transforms.CenterCrop(image_size),\n",
    "                                         transforms.ToTensor()])\n",
    "    # ImageFolder wrapper for the folder with the content and the defined transformation\n",
    "    image_data = datasets.ImageFolder(data_dir, transform=transformation)\n",
    "    # Data Loader with pre-defined batch size and shuffling the batches\n",
    "    data_loader = torch.utils.data.DataLoader(image_data, batch_size=batch_size, shuffle=True)\n",
    "    return data_loader"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Create a DataLoader\n",
    "\n",
    "#### Exercise: Create a DataLoader `celeba_train_loader` with appropriate hyperparameters.\n",
    "\n",
    "Call the above function and create a dataloader to view images. \n",
    "* You can decide on any reasonable `batch_size` parameter\n",
    "* Your `image_size` **must be** `32`. Resizing the data to a smaller size will make for faster training, while still creating convincing images of faces!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Define function hyperparameters\n",
    "batch_size = 128\n",
    "img_size = 32\n",
    "\n",
    "\"\"\"\n",
    "DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE\n",
    "\"\"\"\n",
    "# Call your function and get a dataloader\n",
    "celeba_train_loader = get_dataloader(batch_size, img_size)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Next, you can view some images! You should see square images of somewhat-centered faces.\n",
    "\n",
    "Note: You'll need to convert the Tensor images into a NumPy type and transpose the dimensions to correctly display an image, suggested `imshow` code is below, but it may not be perfect."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABHEAAADqCAYAAAAlBtnSAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMS4wLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvpW3flQAAIABJREFUeJzsvVmvJdmd3bdjPvM5d86psrIyayBZJFskuyXZDWvofhDc8teQv4E/gG0BBgzBgB8sPfSLYRi24QcLkN2C0RAgUWq67WZ3c2iyWKwps3Ie7nzmGP1w4sb6HfGmVVl1CnYa//VQ2BUZN4Y9/Pfecdb6L6+qKmcwGAwGg8FgMBgMBoPBYPj/Nvz/tx/AYDAYDAaDwWAwGAwGg8Hw74Z9xDEYDAaDwWAwGAwGg8FgeA1gH3EMBoPBYDAYDAaDwWAwGF4D2Eccg8FgMBgMBoPBYDAYDIbXAPYRx2AwGAwGg8FgMBgMBoPhNYB9xDEYDAaDwWAwGAwGg8FgeA1gH3EMBoPBYDAYDAaDwWAwGF4D2Eccg8FgMBgMBoPBYDAYDIbXAPYRx2AwGAwGg8FgMBgMBoPhNUD4Kie3WknV63acc86VZaF/KCuc5eEwj7P8ksOed+kpBa5f4Zo828PfVpc8jr92dnVJaf0PK/eSe+I+nq9vYD4u73k6zrsWZekux+XX9+vrT2cLt0zTyyvnFbE9GlQ3ru6v7lq9pB74jmt1rzYv0rnKi1lTLvO8KfuhulcQBE15OV825bxSnVSVh/N9lFfHS9RfWajs+QHO1T1xOVf6Ou5HbZyjv82yVOegRoIowd+2cFx/G+Jd+R4l3u+jD355WFXVntsA4jCoWtHqnt2u3qcVRzoJ7ZVlWVPOcx3nWPHRn0v03AJ17bNd2Ol5IbwzB8DF9cNQz8h+F6Df5bmel+0Y4Bl5nQBl9hO+d4m+XBb5ped7L4lfF3U2WWRukeUbGYuxH1atMKnvpeOMEowH4Vq8QfvwfdFWRcn2ufyaPmMVx/3aOfrbi3HsYzyzXpepyozb67H98ljsYSw6T+OsCDo6jvFXBoj5L/k5wsc05S8Rd5abG4vDQb862Ntd3QOtV5XoYxgTHGcs56meryovH6Mpxq4XXN4fygIxm/MiO0FRX2dtfCJG4tpr8wT6WhTrfMa/td+GPLYv+hpidoF3KjDuXzrI6nd6fnTmzsezjYzFwWBY7R/sX3Jf9GfEkvOz06Y8n2kudGvzt94xjtVv4yTG6Rxzl4/FwGNbqK7Sur+kmfpNp9Ntyv1+vykvF5rb5ijzZVOMjyLXOWz/KLo8djPOMkYnSYKz9BeMO08ODzc2Fnd3d6tbt2594fNLjNHx2ZnK4zOe1BTXl6toI//yAMRxXxTFpeXLlsacc9f61EtiNJEXfF6sYxDIo5eszdgheHxtJY/+cBEnTk7Hbjqbb2Qsep53SY28fC7k87C+X7LjWJvjWSd835e1bfmS/cfFM7x0r3D5lLf+XIzhXOuu9SEe19++bGfBlvVfMgWvjeOi2uhYvHnzZn2/l9wQ73Z2rpj6+MmTplxgT/GFtpEsv6SfrF+GbepdnHzp363V29rxl1Uur435GmsX/mVZXv6CEeaMvd3dpjwcjnTNOqbef3DfHR0dbWQs7mz1qxvX9n7jOb2X7KGx9HDLpdptgfJ4ov1itrYX4R6C4wj3fUnMW2vbushvFNwvEi+53L91DmP7v/u7AGuqekmHXV8D/uae3znnxpPZFxqLr/QRp9ftuL//B7/vnHNuOT3Xgy40+XuVLjlfciOGSQuoCgZLLRByxPEJFhdpqsVFhBAVYQGac1NWTzKxxwDNjQY3c3rGHIuYLNM5QazBFLe1MGvHun4S6RxOANM5PnasTTy6bxTpPTqt1fX/xZ/8udsUblzdd//sv/tHzjnnMtRlgV1QFagdwlTvnqcnTXly/6+a8tmvftKUpycvmnIHwWbQ227Kn3zwcVM+Rd/J8AzDvhajnd6qbpf4WDQ5VbnVHjbl7uhA1wtVx2lnqynHV7+nd4q00H3x9EFTbjv13cGVO005uf5OUx5d1Tvt7+ldl/jetsy0wP873/nW525DaEWh++23bzjnnPsbP/h2c/y9W/s6KZ80xaePHjblk5NxU06x6U5aqvMFxtYJgm6vP2jK/QQfUcpFU65Sne+HqotWt+ecc257R21UIPjxeofPNYlXgZ6xi03K3sHVpjzY1XvzI+GTx7oON1vjUy0YFjgeYOO1TDUunx4eOeec+6O/vOc2hVaYuN/Z++bq2Ur1/Tkm8jjU8+xiTLTw0ep8pvd9ejxtymczXZN7hihUfGq39AEwxrhPsFlLsLgdDVft393SuJlMFW8/f6C6PJspdlTc/HmYdhD74lDjuIzf1XVGv6Pzr2kszob4cJdwpsfm+VzP0Pv4blN+/snmxuLB3q77x//lf7q6n9P7ZxPFwjRTu7R7PZU7mkOe3/+sKRdTbSKzpd7t4aHGbjjS33ZitfXiTLEri9W+S6d+Up6vruNjXu70dppya6BxXpS6XjpVTLl+XeuLHcS/Ch/gKnxFC/HxvN3R9c9ONBaPnj9tynHIxZDg1RvQ/+Q/+2/dprB/sO/+0X/1XzvnnAsQb7xCdXZ++Kwp//H/+r815V/89OdN2Q8wtrrqz2/cUn++cfONphxhPdFKVD8h1jODto5nC7X/gwefOuece/j4XnPs+7/9g6b8t//W323Kn36sue2Xv9b5JRaQDz//pCmfHypuLvGDzdVr15sy1t/u0eNHTXnYV/++c/stnYSFdDtRH/nP//APNzYWb9265X784x/X//eSFTpiznyqNc2/+uM/aso//Bf/vCkX+LiaYZcSY6026KmN+OFyhjXLKT4STc4VDy72sVWuv+NGx+GDahLroxjjON/17Fx9JMOatt/R+fu7GuujofopNyyDkdZMXL0vsDbu1HPSf/OH/7PbJC6WBRV/tMEPdHs9zT9FqrXHGH0150dDflDFmqOHsTVCXI4xX/UHOr5c6F7cfOX1HoFran7YZl+p8FGCH2hDzLPTue4zXap8iradYe+0WPvFEm2OqbYTXf4BsET5weliY2Px5s2b7kc/+tHFDZvj3KuXWB//0f/+z5ryP/wv/mFTPjnUGHVL7JvWfnDDHipUncYtxZkMi6C1jwfoJ2HdHtz0s009fHzJMC86HA+wjqrwvLGPWN/VOGYfmKPdK/SvN97UnPEP/sF/3JT//h/8R7pmPdf/3u//ntsUblzbc3/8P67aAtXqfEQE7rcn+H3g48+Pm/KHH2st9MM/+WlTfnSoub8q8Mf4QT1C/AsxX1Zo/yVi9MUH0PlEMXYynuDv0N5rH7AFzhyco6uKPxDrnlHEH2bwAxW/RaA9k7bO7/bUR1tYD/7LH/7FFxqLJqcyGAwGg8FgMBgMBoPBYHgN8EpMnKqqXJ6uvj7mYMeEaxRyfZ30HWhwHmjY+FLJL6VUG81x/a4PtgC+JvPLYAJqt49fr/36632rpS+fCahpGb7gkc5epKQf85cYflnDFz2fX7n1XKSdJfgyP1uAgr1G2Vadpf6qLl9GsfsyqFzpinz1tZc09ijRrw0FWBgB2nABJlGOX//5q2HUFlugO9CvPUUGahu+4vIrZ4D7xi0dT+pfMbKF+lbU1S8xbdyz01E7z0h9S3T+YEtMkGdnYGE49UWK1+KRKIv9odg3rtTzLlJKjtS/iynbeXOIoshdyOJGYGiQBt/BmNjawq9t+M7sVRgrqP8CbIlt/Mq49ssxvpBHCCWhL/lLiLF20ZOovIhisglUPvjWd5vyk6PHTZlfwgvIAnL8cpmAIbe1rfZqt9QWva7q7KRm2Ti3Tt0NY5V3ylUfCMLLv9x/GVRV1chjSowzxrXRQHXZQztMzvUuJyf6xWE50xiJwe7p4sv/7khjZNjT8Q6YUFEE2RwabDhanT/o69xlpnE2CNQmnz4SW/NYPyC6tFLdV/gdYZnql+oy/bX+AL9QRy2d32nf1Cl4p+UWfqFL8IvbXfWLTcL3Shd7q5haZuqHnq++2iezEOXTI/U9ToDdruJxGuodtjAVtEa6DolIJzPVe6cDhkBLMTCrf8XPwdaLcJHIUzvu4lf7x5gXj4/1C+kuxlmnr3su8OtaDPYFZTY9MDfGp+prBX7parc1Dryaheb5L2FafAmUZekWy9U4inHZDLKan/7pnzXlzz8RaypEvOcv6AFYUGQVTY/V5vOZzm+BFXftun55XWK+PHkGhmLNiorARBhRqkWZ5VLtQJbdMSQM6UJ9IUGMjkO123KuuLO9IyZWsS8m5NmZ+sUYv4Du7YittSYn2jguBslLqPdgKjx8cK8p/+Kv9Asx54EK7JsALJuYv76vyRIhdcUvyhkYFSGJE/WctoSUsNUCiwPjpoWyj9h5eqYx77AGb6Gt21gDlxXeD2vzGPP7fAa2cwdzOlgf8/NV+75MrvDlsXo3b+0IZVyo45fIUxPMoy2yYnB80NbxfcTTLhg6a6kAsOavsC7v1GymHP2Ge44R2DzcW3QT7ktUnkw1p997IBZdBObk6UJlH3NkCoJIVuh/5pTOU84TvNI28BXguaYFyRTCfvHw+LAp/58XrB3nXArGUYUxlFPKB8YDp4K1d6MUHE8Q+JeP46QemIsczFYPjHMwLhbLy5kYa2wNHykasB6gjIzrK7K35ogXhy/EZPmn//R/acpbYMv97r/3Hzjn1uV3Xx2Vc/6qTzNikwV1NlUfe/Rc/fYnP/+0Kf/4J1JfTMDQ7pLlAvm/hwmIzJkC7Z9RZofnieu28NtIh5BeLvflGoIqH5a5Pw+x/k/w7NwXUTaXM+0M7sXUHefnTPnw6m1nTByDwWAwGAwGg8FgMBgMhtcA9hHHYDAYDAaDwWAwGAwGg+E1wCvz6C4ouqQ2kv5POiEp+UzwPICbjiv0DxNQkZMAybnguNNOROvsdlkWFTKOmCxwRT/tgLLdhtyCtNkClOMFqKQvjpSg6Rno7+egFjO5k4PzDSlULVLEKj3jEs4OdDZa1PKhtazdXxFlUbjJZEWhDiPQ8SGJWkt1j8SO8zO9e8jM30jY1R6Jet8eiGI/fipJTAg+ay+AtAJ9od8Hfbfuawmel3S7Xk/nxpBTZSUcGAZXVIasoHqBZIOfK/Fp5qttfVBrh1febMqDrq5JaVVRgc9aqv42iSj03f52v34OvXOnzeS0asi9XbWFDyq7BxpugL6XQE5FKQOpxQnkVC04dcVI8h2A3u9qane2JpvU/8xmkIKFeqdvflvSqs/u32vKfSRinoF+3AZVtotEYRGSAfP880iSgrVkZXiPfi1lepkLyZdB5ZTYmclUh31In7bV92aQ5h2fnOM4k71DBjfSdd68LinDG/tKKrvdQzu3kCgeUrwK9XYhf2u9xAXuYKhr7wz1Tr/8RM/74hyJRdEXlnRVKSTJiCeSVuUPdd8W6OfRLSVQPTmgQwlcdhKMyw2iKgtXLVaxMaXDHfjbCWRNFSju+VLzTAf0e0ph4kDlHciP25AthZzH2nqGw4nq/fqB5Gde3Z+nMySxx3jOMBePMF8X12805dNnSmY7hiRxCBltBWc/JnwMmcAfdbO1Jenq8eFzXYfGFRcxa3PTosvz1L14vnqfBH3yo79U0v5f/LkMBhZjxY/TiSRDGQZDZ6ALPXygucVHP48RZ2/c1NxS5br+559JQnXvVzIU6NXU/9u31SY39kSvP32hJNETyMKoCD3DnF6WkHkgCW6EWFxClDBF4siDA0mUc/TvMRJhX7mq+TL26Fq1SVSuuljArCV7hVEG6uIv/uxPm/IL9Gc6kYWQbbDuEix8fZpikIqP63TQ5x1i6oXUvMO5KmL969xuS3Px5EySk2SkNU0+1Fp3Agm8j+SvbbpYYsHHOS/F3JwioS8TDBcX5iUbXKOu4Nf/1TOEkL3QkW2BMuUOOwOtD66ifrYRNwfYQ4wgYWXi7ZT7AkhomFi1U0shO1grDSGh6rFtEbjakGGUWJedjhW3k0LzfuTBaOS55sgF0xXQTQuS6nRNqkHXMvc1wsN/V6Dz6A//zQ+b8p/8m3+tZ2KqB/xxgfHH+l93PqRDq4p0P21hLHYxLHu1jHGJsTJewH0zv9w1zkPbMcly4V9eZlL44CWpOLi+YWLen/9Uss9/8k/+sf6gfswJ5qNNoImnTMCM9BE/++BeU/7TH3/UlD97iL3VTC+cYA+/g7UuvyOkkKVPkRx+SZMhcFAqyFyjeo+SMg0J9v8FYvW6G/Tl5QAPtr7/4ecTJgqH5A/xaM3xGktRynXnU8XZLwpj4hgMBoPBYDAYDAaDwWAwvAawjzgGg8FgMBgMBoPBYDAYDK8BXs2dqihdWssQPFDGfNCNItA0U1LiIlLckGEaVMiyEJVoCMnTlZGcda4dyBGBEpFeT3SphO5XtUwoBD01xLMwG3QKF6pz0KW3QGFllvuncJk4m4vmlVGuwmzqjvITVD3OKeAWlG+corrKXL6oqbHbgz38C3iKlFCNT1DW+w6Yqxy0v3go2moQq95K0P17cNwpCjiWBLpOe6Djee3gkYGBnQRw7RlIwpEz27/T8WD7qs6BG0N1KAq1e/xhU5yPlQ3+0yeiwiexrv/9v/MHTTmMRGNf4l1PjiUJ2CSCwHejWkbWgvSlDdpuEqn/eLn6/DiUPLALR6IW6JIxaKM4xbVhqwG15JqbCWWUhUcK76peygoOQ3BbaMVwWJiJKj4aSSqzs1TdlmOcs6XxT6eO7V3R/Ml+JLWRDnnHcEsgnbWopQaV22Dmf89rOKQhqLY7Q/XtCNKw0xONxRlcbeg2c2VPcpQ339BYvHFFsXK3r/cdtBG7QVf3QRtlxXneKo5GTveEIsSNtkRhH21rTISx6vVnH+k9Hp2JKst+UWSQHznRcqvJr5pydk9tHr2hsR7hvsUc8gBf99okqjJ36XQlyUsh4ehu6Tnijp71/EgylwTyhRDTcQYZRox+2MY8RsexGZyiMki0ypRyCsX1nb1V7O/OEVQh/Zxkilse6O+UzSRwVPMwXtO56NztLbkWgYHtfFLbISPrQLqVp4jrcEALa0kfKfRfFYv53H34i58555ybvlB8/Pgnki/RkfMUMqEp2qqNdp5DblSgDmPIHfa2NC7Tpfr55x/rGV48kTvNAJLH73/nfeecc7/1vd9qju1el2Tp8FTP2EVsPTkCzf1E43Iy0/MO9yWJa2Ee90O4DYKuHkLys7unNn/0WM/+AvK4KweQIm8cF8+lDldC4v75p3JK+fjDX+icDJIh2t2A7h5ibkNoXnMVpWtRjHqJUXdcR1yw70tcMMf1dtFHIjp47mn9Rpp/BneaMWTDz54+0zNibdmjoxldObGumEwk6eljrd2q5UCUJXx1eK5xp6q4DuHaGLIirOF38GxvwlXv9oHmxRsHqs+tvubIEeTZLcRZqjLoPkdDrgtnuR7WEgOkBKB7mYe+WMI1Zw7JWsA9Ctp5CnncAlL4RaExncPFtYKml+9BByVXfn1OcfKJ083HZ3rW/+tH/0dTXiA9Bd8/riitguQf+7zO2hwJWThibYDr8JpDpCDYr6WIHtZjh+eQzsL9czqHCxHWyzOUx5CzUShDmSBdnDmKmBaELsos//rXWg/9D//Tf++cc+4Y+9KvDs951ap+OM7uP9Qa5sc/UQz985/JtbGAC+lwpDnh+pVrTfnGtsaoDxnuAg6bk4n6C12rM2rSKFWqY9tkoXabTOggqTFE+TuddOnUyqCYc12G9YcfUGqmflygn3Hn4DvKteK1f3lVGBPHYDAYDAaDwWAwGAwGg+E1gH3EMRgMBoPBYDAYDAaDwWB4DfBqcqqqcmVNCy5zZGnPRV9rd+D8Axoqs7rnkC2VqahkI7jsvH1VtO13b77RlK/heA/Z32NkeY/gtHRRJmOQNMgSHMOiDUcsOBi1IlEkPdDzKtDLAk/UZVKt56instS7+sig7lXMTq7yhQvAJqmqvhe4dmtFU/chX6rITAOtbXomurUHmZVHOmBLdR/3RYH3QCWka8BgpHOWC8h5fGTwT3S+V1PYOqCvhbHaZziS8840ADU41vEANNtyArrh40+a4m6o9xtCZvLgRM5aT38ON4tbbzfla9/8gZ4dVOwCjkKbhOc8FzTSJmRmhxuMD3r2GBnr07kohFBkuMpjhnX2Q1yfBhuQSvkYFx6ys8d0qKv7QInn9Uo49VR6rrLU/WdHco8adSStfPJYlM7hFdGl845ozKToJm3ROwcD9Z8UtOspnEuOjyVp8C6+d29U4Vg5V7tNtDCGem1Iw04lU5mB0p6gXq/uajy9/abeZX9HfWE4gMNJB2OuDSkkxqjvLo9JXiN5gsQKVNIAMoEDyDDeuaXYd3SqOj6HC2BZ6LkmGEPLQv2iKFUf7kz3mv5U96qKv6ln89UvskrxZZOoSufSC5lfrHsMtxR/cjpW4G87XUpONRY8DLRWW+0bor1C0LDPUrqq6W/3OvrbbCGKcpis6M3bnQP8u2KER5cEuO30IHOOQPPPNERdgTnPX5Ncqn2pFA4gUW63IKdq6zozp/Kgls8GG7RVKbLMndRyk1/8hdw/HFxfSOU+X8AxBlT+GSRJ1VLtWeZwONyjVEm3mh1J2juEJcv3bqqef/A732/K77z3jnPOue5Q/czDOojuGcW5xs3zuyr30IfoGphO9ewXji3OOedDntGHdHqO/hdDUs95/8GD+025ixi9edR1B8nIHPT8v/rZj5vy9FwxPoZUuFiCBo+5KPAul8FTzkRRRAhJc5fuR4iTF+vINNWYaw3071t9zXkhxk0PMn86rlBO1Yez5ihR+RHcQhO0KdcABe1iMKfP5urj/VqOtFk5FYD3Zd9jPA3xnHvok9exztxHnN1CvNlBTNrDOR24UyWISQ4SHig8XVhLInyu8bFGrnLK7XS8wDt5cEHL4AiXYd1ybUdjvcAe4hRrullKt1s9I9UnlFOVxQYl4v8WvEukjU8ePWzK9z76VM+ExwghZ+lD2rQL6e2dPbXvdbjy9fC3HbqmYjIqlnRChgS6juUR6uQMa5QzSNXGcFM9Poeb8Vyx8BniyOECbQQZTwFpYID1wxINtuZshP5DWeRnn63qcrncrCPuxS3meK+nz+HSDGl/hBgTlirvDtRv79y405QHGIsFJNkRJEYRXI7ZjwqEHMor03r/Sje2s3PF//NzyIkhN6WzNuktHEMp6pb7qAD5JHJf8foc0nI+zwx9wYe7s2dyKoPBYDAYDAaDwWAwGAyG/3/CPuIYDAaDwWAwGAwGg8FgMLwGeCU5leeJOr+EGxMzcztfNKElaNhFJtpS4jGbvOiP37khB6H335NU5dbNm025BxlECOpwBfo9WEsN3SwlVxgyLx+OKFBhuSAUra3Ct64U9OoU9DKHOsiQnXoCytUc50SQaPmgaBbF15cp3rkVJbVTy6l4qxCylmqp5x8fSU5FZyIHaVMF6mkL1N/liahkazQ7OjOA8k3HqbiDrPo1PTSAxCrqij7ZBt0uL9QnCpxDSdDkviRU8Vy0QL8r2moOScIALgDpc0l4ju/K5aJ3RZK/7pYkHLtdUT43iaoqXVa7aTDD+hI0vRJygwky7OegeKY5HAHoTNFRXdCFg1IpSqh8ZNL3KOkCPdSvqZOUYQWgiRa+joeQ4s2O5BRW9lCf6MATZJwfjtTuz1+ofStmhAf9vd9TvxsO1X9PT0TBXNZ9sNygY5znOZfUrjHb27pvjvh0eqZ2Yztc2df5d95U/7+yrXobdVRuJyhDHpG04AIBCU+Adl6jgtdU1YJWF3RVgsyEVgvbkKq+fVWxYDrW3z54BsmtD5kd5DmFB0lOIde4+BlkOC25H7jrmi9K/+uRU5Vl2cT5nSHcFkCVn4KSHSH+JXAbSnFODFloHGssJpBTzc/VtzPIkrd24GaDV57D/dGrO1MH82+OsVrBhWUK+nEJevAIroCTgq4tkI5hLuzCdY1UaLpMtSEdyJaKZZyDL5wgNinhSJepe3J3Jfehu9ACTjLHoF57oOl7lFWjHMV65gPEpG04CJaQK7+5o3d//0314TdvqHznzi1dp3YKCzq6dolxGyFWzTvqT29t6z6LU0l8+ojnJef3VG0ehvrbEm3bgbPH8aniZglJwOGh+uuj3hP39cADJV3958lDyTbu3/2oKfsQN1aYT+gAE2MtWMCdJPT0znRoLUD5pxQnwnp1TRpUSzcoIRgNuHbBXNXGGgltlFOKifFHk60Icq75SGuUOSRAlL8VWK/GkKicw/0xruePstqkJKdqXCA5xjM8T4F2a6PutyD37GH908ZklEAe14bbVwt7gRaOR4g9Ida6IWKVX69jl0ilQJdAutQWkOp4KIe4z5oLGWRT20gLkOOdxpD5nE9135MZ0jlwreWEzXvgruB5MvZhTL13V/P28ZFiwgJS1F5b73ZzqHf+7lVJS793S3PttT3NeW3Eoghjh46bBeJSDmnVRaqPEtL1Idalc+xBxpDCPsd6JYL8i65FPmr6HC0wxdCZY21cYO7hnBogTkeUI30Ne8eqLN1isepPDx7JXXAMue1sAZkQ1g3b2JdRBtjBerHkmgdxaIC26kFaGFHyjz2ah3Zx9T6V3wFSfIvIECuXWGNk2Ee5NRminotpKU7hjhtjTp9iL333iSTSRxOdX1XqXxm+TVBa/kVhTByDwWAwGAwGg8FgMBgMhtcA9hHHYDAYDAaDwWAwGAwGg+E1wKu5UznPlRduT9Ae5cyGDgor6YSxL1rfLmjV778lqvB337nVlO+8LQnV9u6+rg/KYwHZAV2mmLH74jiJ1x44phWoleSeBsxY3hEldhtyruUcNOZMNLJ5JqrUC7jdUF42X4ii1YJ0JQgprUp/432+KjznubB2WFjAhaoElSydyGqkAD2uta1280ArDkArjVqg7ObnOK5ziljvE9JJLB7gfLVz0q3rAW5aQU904GCA8hSUWMgQpseSQc3uiU6d+JCKJHo/H3KGUU80xacTvdPZw3tN+eixnDcCuHN0Ol+PnMp5zoU1bTAF9fv0RG1XZaABvsBxUBi7kNNEoEyTqu2t9T84FVE2BTe3AOVorW/XsQOOUR7cPjqVxlkBeUi2FM32GBloZv8gAAAgAElEQVTsS7hkHB+rDkZbihcxpCtleTlVtdUSFZLuOF1ItxaNtGpzEo7A91y/vapzxpjTE73jYgF66khj6/qB+vzeSO/YS9RWSQBaP+SELTjftCD5oMsH64c07Kbswb2lIuUdroW4Rgjnsb2e7nP7it4pXaiPnqGP+HDQSRAvSqd6CtN7OudM0olypPmlBWozPK6+OjzPubpvdyDNyyFL9eGmEqOPsc4zxKIQFHqfdH44mMwh72lR0ooYHARwXFiAkl3T9UO4nAWR/q6C/MflcFyaaz4LIclI+nTBUnzJUpWrQucEiDXrskzEeEg0u6AZVxexaYOGOFVZuWUtiQsgbz4bnzTlkpIM9EP6Eg36incHu6KQj/pwnHsmd5bbO+oLv/3Ojab81lX97e6BysMh5shkda8A8auiJhbj9gDXKBboN5hnn76QU9MJJFHjDI4ckG2Aze5aaLcM1PIC1PXxRH3hwaOvS07lnKvjESVRn3zyy6Z8fiZZQIF3KxZ09YEDjKOclE59qgC6knEd6dGVDnKHDtrsQkHRg9wpQozowTWpC7cp2g15WFtWa+txPYsPSW0f69hsDNc6yBimdF/hGEUfm9QSgbLcrJTjwv2Uqt0lUhk4yE46kFDtwPWsF8ElDfK1OET8pTMMJNx0IVtzkkVsiDtqr6SW4gd0qjuH6xfmgrWXyi+X8zlIS+gWG6Nf7nT1Hm9BZnR0phg9XTxryqw+9ouiYszYYDtWct5dok9++LHSEJxhPdem4ybkMT+4LlnO79y+onOwH+li3o2RwiDE+tsLX7IWxBy1qKXDi0ixsEI/CrAe4zW6kY53EBd7KC8xUQQInlBfraX9WJMoUjqM8ib3hpdhmabu03srSdAPf/ST5vhHd1U/nz6QJLjC+oRzSxwozi7H6pMtSjmpiML/RJDBJ3SkRH4BpgLwL1xwEZO8SGPFDzRuC6QKyLFfKjEWc0rvhlrzLDC3pZhrjyE5n9GVmy5Xqdo2z9V3qi8hiTMmjsFgMBgMBoPBYDAYDAbDawD7iGMwGAwGg8FgMBgMBoPB8BrgFeVUzuX1dx8fthfpEtIayGyGoIFug/L4rbdEiftr37jdlK9ekQyivy3KkwP9P4MrVrFU2aPdBSlmdTmigRb+uagowwLFEBQqfunqwR1keyhKaroUhepoKjpjDBp4ACemlHQt0LgCOBRcSMGqDeaPr6rK5TX1tgB9PyBlGnKqFp6/DUpaXkC2BJcSDzIkNgPlVFVwuZwqboMGGcLlo6atFgGkLn3R/Ss4X/k5ZFggun/+uTLiZxNR5FugGJeg8JE23YUkx0Eed/JMmccHz0URHOzKZS3dJOcf8Dw5zCwX6nvzc1Gj04loeiEouT4ooRXkHH4bfc/RCQJjhNn24ZThgcIa9OUakKCdvLp+s6mesZqp7QJQjmNPFNcC4pfjc1E3J8/VjgvQ00+31H/boJ8fHepvHd6PHbWFtg4xFlut1btS+vFVEQa+2xqsxkUJWukMDk8BXHD29uRItbul8dSBJCpZk8dAHhrSVQzZ8H0eR/wrIZdFH85qiRTY7K5YkwzAJZCSLFBbE4y57S212xup+tCzmfrryaFikw9ZVkhnhlxtHi8lV4knGq+tjuK1Ru5Xhx8Erl/LXCjDzTO4k2AMBb76WFXB5QjxyqNrIu5V5nr/BeRUcXC5/ApN4yLc98LBh85fdM3qIKZXc7VRBlo8XRijFpynKO2odP4cNOMe5bVwaFlzwkO/i1p6tovTN+lOVValm9dx8fBYcSXDsyWQd8WgjbdB5d6GxCzE8c/vf6ZzMO9+5+13m/LtG5JEHOwpbvb3EU/hJubX814ZUuIK57lI58aY27oDxcRrt7QWm6E/PXzwoCn//JcfNOVP7mlsrTmPDbRe60EaSqlqCons4bHG5deFKaTPjx5o/q9K9U9KhXM6TyG2MDZzfeYQ06aYdx3mVLqmepg7s4XqZVg7eiZYx/QhT9zaUl9YiwUYZ25NHnOJ/NWtO461MHcnuA6lsaT2040wgbxkXEuxSoyTzWB1vQrPzGdgnE3Qt1uoe65dw4DyKMRZyKZoTxtAehPGcBhD6gDKGOPuaqz5iKEFtyFIk+CjrmZYizk6y2Hej9DMmOqdh7XITk/PeAPSyeOprr+ATJtS+9L7muRUq6d0zjk3xvj45DO5w2a43zbWn99GLPw+YtRNyE/bXdV/hJgTDyW/cl2VC0jnSkhh/Ax/W/eBlHsy7HUCSCizpdqxHev8dqRr9xBTcrQppVgLyKZoABwiXiToa2vSRrTdxb6u2qBT3DLN3b37q73Nhx9q1fTJ5+pLUyj/egP187LU/JCnciFLM52TcPwhrpRwIS2wFnKIvwHGeoC6iuvr0JCark8RYgElrkwLs4Ccn2vXCGPa596+YuoCpIigzJ3un5XWq0T1JZY0xsQxGAwGg8FgMBgMBoPBYHgNYB9xDAaDwWAwGAwGg8FgMBheA7ySnMp5nvNqeiOpmb5PCq+Of/dtuS3cuSkpwJ0rB035+r7cQwLQ6XxQqKoUTihwUcpBLa/IXXS/KYup1uj/OrdYO073GlC8ceUIfLctZPjPCz3L+VLUwcMT0VaZXb+EpMEHNTtAXc6bd9ocbbwqC5dePB/1TsjQ7sFdpJ2oTUhDS0Nx6PyO2haMNFeCvtiGzMpDPaeg0LV6omRnqaj3Lr6QsoAei77iICXwYtXlYiLK3+SJqIBtD9n+PbVP6YlWOYWDllfomlmurOxPH4ha3nsqOdXWDd3XB4VukyiLws1rWVIGevDsXO82g5xqt4N6xhgqM9Vjho4OAdka/Zi0ZJbpJBP31B/8vuisVU1R9uEO4UK4TY1VtwUo7xHorH1QUh9Cwjg+Ez3xAfrarXfebsqUguQlOiooqVFCOSMlV/U5mzQD8Jzz6niSoU3o6NIGf3rY07N12qCSxnQgQfvAAYH0cw+0dLqA0X2nWovvemS/qYa1wIlrU4aKcyIU8ewjT2M+B1X5+qGkHY+PJL1IKQ+KMV/EmBcWkq605qLxJrHG+ibh+77r1o48FeYBzgmkvuclRhfqtsL/0DXBVZAWok+mkB9y3qC0waMrC1yDgtpFsoIkpnR0bVFdRV3JOfJC9ZmdaswlA41zSnp8zNFZrnsVpWKT71NyA8cH9E26vvgXawxvc79BlWXpprXcawn5TMy4hvHRgUzmYF/vHqC+Xzx8rL/FmP7+t95syt+4drnzVAKHvWRItz21S1U7CZWU79DiA1K9oCsZSA/OYK2Rjm+jP40oVYA74OJYzk4fPPi8KT8719/efPv9prw7ZN3A1Ygykg2iqiqX1nPH0aEcKU+P1G8pN1xmdJ6ibIr1SGmjYt0U8jPOLWUGiSrahk50Idyn+vW6Zwh3qt2BxlwcUEoIVxOsIT3KhejKhKVjgfESI11AG+unWUrpKu4FOTZjSll9Pb8DNzXoUV8JGTB1B5yKOOVgLm9RpoJ1r/+SpXWFNQFjt8NanXugi9qh3HrNGRPxvII8ywXqQ4VDfMRL+ZSKJJTEQi4N2Wofrpq7O5JiHs/gQMu9DuqAAr2visqrXOGt3vvoRHHj9EhlysO2unq3O1fgvrkPt9MB1v1wgaWcJqTkBeOlrLBnoc4Ya6OwHjtpwD6uC2aI4+yaBVyOPUiBglD3aWFvEuO5uAmPETvoEFvhXSn5d3A/chdr5g06VuV55o6OXjjnnBufq68uIXcPIkoFsf6DA1MxxXobmqQS9UO5aVBxv4hOgnbxuL6NfnNf4r3M0YvrWcSyAGMuWCItAGJ+Tmc5pplAf6o8vpPOLhG/UqyTma6gKF697YyJYzAYDAaDwWAwGAwGg8HwGsA+4hgMBoPBYDAYDAaDwWAwvAZ4JTmV75y7YDBNxjMcF02oA8r2FhxUrl0VbXh3R/S4FmQeFSiheQ4HI2aTBg2N5TVOJYp5TX/LSatbo0TRTeVyOVUFiqQH+ww6Zhzs7uI4aI6Rzpn+1UdNeQlqGjOhV8jW7r2M6/kVUJW5K2p3porUfMg5wGpzEanafBxkCY+ROX06kcQlgjOQh0ahI0QAKm/Q0jVTuExdXD8FvdCDu0IB15UAzldnz0Vn96aS7bRxnRCSrwWodQWpeJDwzFKVHzwVXXv//r2mvHtNMsLRgTLrbxRV5ao68/10obZ7+gKUR7jGdSFhStDH1piXqHM65QSQJwVrDjcYuyXppxivoJZ69VgIIDOpUsgq0Beqlzj1dODgsbMlyu3zz+415Sf3RfNPQKnG67mMtE/0h+1txCb87enZpH6uzVFVy7J0y9qJqsghiQLVsgPpQxtuUwkkVC3QSmOcz74doK1Cti07AGQk1FCtsVKD1TjO15iqoKoyDINiTOp5GxIVyqwq9JVr++rH7Uf3mvJyzpgPFwBIq5aZpI35XM46Xgf33SA8z7mwfj86JhaQIAR4Pi/HXISxWEDO4grKqTj/6HxKdBzo9BWsGMM23dbQ1vmkvg3GaqXYncOdxQOF33Euhova8vRFU056Oh8GHq7i/ArXjgjzxwJOOcsFYwMkeHsriY6/QXeqoijdeb2moUtRDKo4pVK7O5rvOeaePn7SlDO4cd3ckcTh5j6cpzpaIyUtyVBbbckg4kTHPboj1XVCKV2B3+XKtep5iSNPCMcMULk7kPO8fUMOok8f6P0eHUk2/pefS8L4EPLH974padXOrqRV82fqLxtFVblquepDTx/cbw6n8xlOUsWkdKSCrIJSWlLu6b63mF/uDkXpDOVXvY7WLDt9temodt/c3VeagRBub1yvMlx7iH+Brz4YBZenAqBsv8T8RykKXXsoL6MEPoDTUhJdIl3YAPyL+uSik25PWBtfvpp3roWYG1NmTNnG2j1xfbrWepef5AV8hlX9ZJDs0IU1w/p3gf5R0mkI86K/1j56dsrCCuwbMsTNdKm4yXZpI84uuXeCdHA22Zygyqs8F9RzF9eTYXB5TO1ir7QLV9MunAxDyhyxCFni/TM4IZVHkvxSIhe0MHYSpN+o+38AF8yCLkvVEschl0Y5hLQmqLjOhMsRU5CsOSepTUPIJUvUGV3UHNM+XMi1NuigWuSlOztZvb+PeaPTQRxEX+UcUuUs673WnNcwuEJK5dFfEqxjYw9rWsrW4KZa1ZI0utmVqHsslVwIvWmFdCdrKs61NAPYx9IZDPLUHHGzqJjCAcexxufe26P+6gvCmDgGg8FgMBgMBoPBYDAYDK8B7COOwWAwGAwGg8FgMBgMBsNrgFdzp3JVQzMsQR9K6FID2le6pHsU6Ez55W5TaxQmOg7hH0JINWhO4YHmRGePCwodM+ozJT2lGjkztjPbNKiVfG8HKnyM++/AtSq9ImeJj+6Liny0AF2eJFBcJ6xdODbKVM1zV56sZECk5tItgc4MJeRjlBh5iei+HrP0w42E7jEZKGZgs7mgxUz9oMShjcKaZhqAauaRblyAEp5LNjV7/qAptwvJM2L0BTrClMwkDppfBhrhOUw1Tk9El39+75OmfPXqXlMedl5xiH1BVJVzxWLVV5ZTUTxnucpLUDzTXOOpYOZ91HNA9wdkkI8itgUp3GgPjinQQB2o3Rf0Yo5bUtUj/3IJFWnjdNK4CueFx5AxPDiSy9WDz0TzH+6rXTJ3OUV9b1eU9hbkknm2rM/dnJyqqiqX1e4CFWj6YPq6Dtz+uqD9dkEPb6McYez6oHsHkEXGpBKj3Qp3eczj+PYbCYee0UeMWHMBxDuF6HN0W4lgT+F7ihH723Jq6XZ0/ZM1qrjKbbrcBRoD0+WHeoZCEpiNoqqcq+nXpLXT1YIUcgcafE6nuIWeuwINt1yLV4IHWjW7ZQH5Y7sjyWECueqidhzyCrgSQqlTwgUpBN3b8yinFM3chwyjHMP9pYW4Q1dCOPt04NZCaXaBulnM6BZ4IUXZoLSxqtyslta0QPduYU7Y26LcSXPO8+eSBmVz1UkfY/fatuqt34W0AJLNsCPJVdDWvdyaIxVcyOqx61eMm5evcyoMWA9zG53qKod1U6T36+8qbr55S85a2/fu6ZxIEoZPH0nGlED2vL+reP3ihRxqNomyLFxaS6efPtT8nyFuLCADLLAuTbjuAbWebbQkbZ6uJVhzUjrewTpl2Ne6cNBT+er+Veecc/222p/zH6VdFdqL54SMBZx/gQqunFwO0zWrwPqNwdxjOgJI0Prd1fsFG5RwOCcHsTW1L9fJXG/wOKSqMWRfISSPJdYzhX/5mr+ALCqnxAySV39GOUdZPwvaCvufkmVIKfhOdA8rWJ+YO1qIxQvE7gxOfvOpjqeQpPbg/lpwrYW5+cULyY++MjxJ0QYj9e0BZPBcN7JPtjuKl3QI9FFfIdbuFdrag7TQy+CwRuc1rJPjFNev+4mXaHzOW6rDAnEhgVyoB8nXGM5GPjY7aUlZDp3qIIXEWsr39Iwh1sCR/xJpVS0N8jboZhyGkdutU4V0e5KpDzmXn0hWW+SXrwWZGYSy1WBtE4/2XNuvc88PFz6f6RewB61l20XOB6A8ks9IGSzG5ZqrGcb0S/YNXItwv75mbIf1NV2uK8isuH/+ojAmjsFgMBgMBoPBYDAYDAbDawD7iGMwGAwGg8FgMBgMBoPB8BrglbQelfNcWX/38UGJj+GYAKagm09JXwN9DLoVr8R3JMp7QK0K4GrBVPGUX+WgmRZrriirVyQNj24Oa24qpBPrsKtAswQjau147oGiCyqmzzIzW69JqFQfS1D+vMaJZ3P0uKrMXTU+XJUhE1pzQ1h7NMiKWIcd0Q3ZPMHa38LVgd8LcTzifdesF+iOsypHEdyuAtG9vQKuVuei9s1eiE7dcarXKBT1MQd1PgC1NmAfwrNPUrX5AhTZkye61/FdZdbfgfvaJlGWpVvUfYX1loKOx8zooUeKNWm7OhpRohOTzsqxrnoPQWek80XIQQXaqlfF9T0vl3P5pBiC/ujg8kDqdyvW/a/ALebxofrAM8isjseSvyWkto/UphNQkSPUwQaHYIOqrNyillO10LfjkBIzvDvKAcqkJKeg9c9AJc4gxQrbkFZhLLJfZDkp/Hz5C9cQOC/5HCukqiImI0hkmejJk6n663ii44HT8TbmlwruEEGgZ+ig73ql2m02O8Jx0YE3i9IVtVyzXJOhgcIPGvYihwsVJEMZHHSqEhR+yJy6cIqiFJEy5ihRfKOcihKOSbmizY+nqhM6onAiL0JIqyCz4VihQ950Knei0b5o9EkfcXepsZgu4KDlX74kmeJes9mqn6y7TH41eL7vwmT1blt9veNBX+/ehTTo6EzvWKaY1yGVG23p/NFI1+z09b6tAdpqJAmV35f0IW/THRD1U8+LdPKjJNmLKKHSKX7Ga+BPKSdhfNnSs+y+ea0p7+/peYd3Nf+10I8e3pVT4Hvf/k5T7sOpybkjtymURe7G5ys57fGxZG4FJDGUPpHmvyYtK+hUovadQJaVw32OrqI9OLR20b6DbblzXb3+hs7pr8ZoiDmUjolFcPlvrZT/Z3CFoQsj5ToV5QV4V84A3tqCmA55lN1B8l/LWDbtTnUxjVRrDUSnKshOGAfWpipVRES3veAleoc1F1o6vgmUeFYTyEhq2ZLHdS4kGQHW2i2kIqCDTwE5B92W6M7LfuFh/uNam7KQdK75MulpvHaQxWABydVmUbkLq+AOpLyjLlI3YB3hY/8Vcy2IPlkWkBbTBbincRbHilcx0m8sIeHN4FZbYM1S1OtL1tU26nyKep75usYCe9oO3JTakZ6rKjVnLLAe640kM018xY7WudYJR6fqa4xfBZxek3qdxD7yVZEkkbt1ZxXzP33wqDk+KyW7K0/UhnTEZVqFADJAP7x8vg+wnvfoQkYp8EtiIZ2lL5z36N7GFCERUkLwamXFeKeiRwc5xOgc9+QanHKxBPeKONcihsZ0uWOw+YIwJo7BYDAYDAaDwWAwGAwGw2sA+4hjMBgMBoPBYDAYDAaDwfAa4NWsc3zPeXX2bh+0piQm3U88pDbciVqgezPTtu+QYRquGhVoS6S2MjM+s8ZnoFdmuH5WXmTsFlUqAv0yW4KaBirYWurpNWkHs9kjgz2otZS05MUSZbhtgC6ZQSKWgxYbN3T8DcqpiszNT1dyqqgr2Q8peDNk4CdN1kMG+BjZ2NMlHE5AIc0z0BRJ2kW2+RB9pAStk1KdC1psAJmA47+jPadHcr3Ix3IpakGGE4DOx1zgC9Atj2dqt/FC98ogc/BA+UxnkgeMnz5sys9jkpU3h7Iq3bSud8+JPjqf65n6cFChqIvyFw/yqDIQzTcPRXnNQE8t4JriwzXFg8yCNgMlqJ9BTke2+hgp/wkcORLI+CDnyaD/yjFuhiP15X5PEo5n90UBTWeQl03Uvgu6DAzhNAM65iZdqS5QVpWb1y51ESRjCWRcnk/HKI0t0q05XqHacWmh61SV+nbgHTZlsD1dv6OxOOipbdvgXl/IObw1hw2hQPxaQs51fKL7P3mmcXk+1ZjvdCX3KaEJiNdcy3SvwEO2f/YjxP8OZUHp5mQbRFmWblE7S8WQxfmY85aIJxWc+iLIHNOl6oiywSrT+2clqfKg84ag53bVjnR5rBAz272VhGOc6p7zuSjSEZ7R9+GsQikm3MRmBVzDIFGdThUXI1g9eXRZmYmWHsUauz6o1mdnerbhqJZTUcf7FRF6ntuuXbLe3JEbUyvQOx7jGZapOuIYUq8O5svuQHGzM1B86gxVHsJhrwO5md9C+9OMhDTv+vXpnkE+9tqqgXKLNbm5ykuupyCDbWEOa8NZZmdX8qB2rOtsDTQXPJ3qeQ4PNe4ZozeJvMjd8ekqhlPWV2Gc0R2Isnb2Jw/xfjlTfF1ijHKNOMT7bEMWd7Avt8PrV2805XYH0sKL9QvWFhHc9tYc/8DzTyHtOZ9onM3mKns4v82fbDFRUEbG86lkWqszSBPy+ni1Qac455Q1gU4ydH4KMPe3IDGlG1oVXu5kQ9layLQQLY3XIMHaqatr+nAqY+6Aou5HASz+Ys7jXD/M1Ic87G24Poqxh1hyv0LpPOogivW8o4HGaAS3qTFicYZrTs4k1dkoKufCOtYEXP9zT4H1whAys0ECOWmGeoYjWxxgrugoZgd9xdQQe0pvgPn1VLGoXCh+F/U6ooXxGWCPMsPzHk3UXo+wv3yMfd5z7GkOZ7rPCeblt/uaD95665u6F9ZA9x5pHfuzD37elJeZYtOwXsxtUtmYF4U7rt2nKNmu6OpcXB6TcrjhOcwhPtqZqQPodlti3NM1ji54a6kAmMHhwkGbSkl8rwgRL9bnAjjy4nzK/Cq6iVKW6ziO4SiLGDDqqk895D55iZQTX4JXY0wcg8FgMBgMBoPBYDAYDIbXAPYRx2AwGAwGg8FgMBgMBoPhNcAryak8z7kgXn33KUGfXs5FMQqQgrmdXNVx0qNIJQKd1QPlbjIR9SwDnTUFJa2g8w2ok3QuKmvqZAqqcAWafwqaewHnIWaqj9rIpk2qekS+lmj0NPlg1vgINPMcGeR9uK+s+8Bs3hKnLEuX1nRbvy06tEP2/HyhuorbcD2J+Jyg9UPOAbWZSyFPIz015Pvi/AwSETqe5TU/0IN8J2dGeVCcT55IykTXgrTU9Y5P1ObPJqKbnmeQPCRqz9HWlabcR/u3KOEAh/HsSHKVwdckp1qRi1dts1hCTjZVne/AQSx0dJDTM52c6Z3ns8dNuXOqehntqF26Q1EIux1RXkdD3WsLjGMPco0wXY1dH1TMglI8uPYsMBYnC73fDBTWxVjvOh3rGSnDWKY6/xzOb24JGiWoyI8fP23KVfmbLiabHJFlWbp5HX/aMZzXoEwLIVUlxXSKOBhA5tiCJKmDODgHZfMUbbuADPA5aMA7kILcuK7+361dZVjHKV01IE89OVN73v1c9TqDhnEI6Uq7L3p0BrejOKarH/oLdCaeR5mjzmlXqoPlUhTqzcJzXrl6lhBOTpSTjs8lGfLhyEEnpFYLboF0YYA9Fd1MlnNdZ7LQ9UNQ0bsDONiAfh+2V+cMR5qjj5+rj8yOVVdhiAaDU0qWaw3Qb6EPRqqDFH1qPtYzduC+VCHuFmjHCDT2czjLvXixerb8Ennml0UchO7N0UqGMAJt/AxSgynWOS8mkDBC4u1h/fMCa5gnxzr/NFXbLiG/oBTRp0slXS0rLtnqdQlknx6MZmh0McFcfFF/zjl3+kISw2yhOg4gz9sfwWEJLjP9kRwBEzho9QeqgxOs9U4gR7uyt+++DuR53rhSLVO4wNEpiI5NkJFzHUPZflFdLunpdxWvWpBtjHqSR1zdVezsYJ5xdJOqHfdySGjmU4xztN0ZXGqOnmudsYRsowMZdQ/jbAJZgE+3LkiryuIlsihKq4r8N/5h83LjVT2HuHGIWySQXnQhQ/Qgdxpz/emrbbstSMUjpEdAH+lhXdqh3BqS0C5cli7qoaggCcG/no21Dnr6QpJ/ykQXcCdM59jz0D2qUH0sKMVFPY0GkvNdP1D/+/Cp5uAF1g9lxqQCm4Tniot9BWQlQRvthT3F3q6kh+22xtbTU0ncn0I2GLQgbWxpXHT7J015iLFwBTLPawM+g+L9Il3VdYR9G5af7u5TXftPfvZxU753qONTxOMI+5QE+sRtzAGTR1p3t7/7g6a8eyAHu61rmqc/f3ivKS+XkoxebIE3KadKl5m7d3fl8Do+U585PVS/zeZ0daLEUP3zHOv8EO5dGdYNlApHGN99rJdKrJGGoda6jOlZterPKeJU5bD/R4oTul35aB+mLlhgfcI0BtT2p3R2RBxJ8F1iZ6hxOYBc7+hU66L0S6xpjIljMBgMBoPBYDAYDAaDwfAawD7iGAwGg8FgMBgMBoPBYDC8BnglOZXv+a5d0+I6oDr7oP6RnjQciO7kkBH+8FSUuNPnKjPb+nwOmQfcH0pIqBJftKUtuKkMkKE+qZ2IZpAIzZbI6pOdhzQAACAASURBVA9K+gko0jNQTIOWqIBtZKpvgVrcBS2wkzCLN+iXoFC5UpTKAhQtr6RzSX3MbRg19bWAhIpZsf3ycpcGH/S4cskM+8hCDjpuAEcAZjb3QJUroD1j2YP86sIRJQJl2Uc9pWeiFZ89E2WUzjd3T0Q7PBJ7zQWg4g72ROfs9NR326CCboH+R2p5FzKWFLKg5Uy0w83Cc35dLzPQ/bIS0kZ8oqWsbw6XpjbcGXrbege6PDwFVfTwM1E/55DiDEFz/M47bzXld++o3KlptCWoh2mqZ3kGmv9Hnysb/7MTNRidWAaQOQ4gOxrBqWoECuPho2dNmRn1x3RnoHMQ3Me6F5nlN0kbr6qGtp8u1VfLLqjH6Ht06qgQFfJcdTI5V5uUvsYolD2uoLMgqNd0L6FSNAMtvYhX9VOhP5Fiegxq6NEZ3OY6us9opH62gLT1DE4adCSks0zo6V4hHbIKjTk/RP9Gv3CQ/2wSnpPjm4f+mYOmPocjSYeVu+aIQ9cXngK5DOUFLTiSwPHBg7wtRX/II8iC63jchvNcvyNXj5OnktkcHmvc+C3Mf5hzE8gSqraeN4HUL09V//NSY649hOQUTlgh5gw66t1/cHf1bunmZABh4Lu90aouJmeaK86n6lfHkCAsCkps9GydHhyp4EL1yWNdc/yjv2zKR2jn3/6G4vitq6LPD7exbsA6I/NWz5ZDn3h+qHZ79Ehz4SdPJUmACtXtbWvOu3JFEqedHTncxHDHWkBmFCca0104R/YjxYCtrsbf84nafwFHmE0iz1L3/NlqjkoXajuusfjLJRTXa5J7upZUWK+0u0McV8VMFzr//FzveXoGOTHWLxH6T1jLOXLU7fEx1iuYf8/PdT3G/cFA8pPBSOO4jzU41wbjY/WHOdbGXAcGkJrRRYmpAIKvSS5+sV3YQ1zrQPYyxhqrhJPf0yP1/2ePJa3v9CR9WpPzwx2nhzXcLhyerm0p7cA7NyVxuXZNcsKLqqrg/DZFn/vxT3/WlH/x8SdN+ehccXCB+YLyNDoyJpBTtQfY86Cde3BI3N2Ca+dYfeoUjoBhsGkp3AqV51xWyzIxhNwe+yr2Vlu7qvMJJC9HU6z/sF59caQ+fHp+rynn2L8MEX9+647c4Xpvv6n7Yo1YzVZjd/xcY+7xM+0vDuE824K+792b15pyBLmejzjSpnQH7ob3IVN68vCzpnwbcwDHcb+nWHtyrP5wYe66yf1iq5249795xznn3J07t/A8HzXljz5+0pQfPdG+toKu6wVkg+djxGWsbWZwT+O6aIh9850bN5vyuzfxPJDNXdj5jSHBfoIUF2dIzzKZY02I540RaxJINFvoryHWcdxrMXVMiHXyCOkCtuDy+BRS+y/jFGdMHIPBYDAYDAaDwWAwGAyG1wD2EcdgMBgMBoPBYDAYDAaD4TXAq8mpfN91kxWta+/Wu83xClSpk1NReBO4Cc2XokcdHYkydIjjL8CwPT4TxWiG7Ox5JirUoCX609WeKE/vXBFFeK+/+k61gNvJ01PRCh9MRE99NCF1GhInVFMfMp8RaF5boGWOIIfY2dLfBnD2iJC1egk5WjvRO/VquQ6p5F8Vnue5IFrdO8Hz8w4FaO8Rs4eDGrgYq83LghIC0pBFK8whM4iRsT2HFMgPIBHhQ9c05LDQN8cK7mjHD0VxPAWd9rOHcAFI4XYViR47PgW17qmyzQeQjn3nhqiSbbiWJag1H8c9nw4uX1fm/9K5cvXsy4w+JJC/Id27D4cfZt4fwoUog+Tx3mPV41NQwheQPMbgEM4zDd7k88+bcheSs+s1Xd8LVCfPjjQWf/LBvab8wQPdfx6oH80gO4rRS67BQeXmjuix/aFoi+4RJIzIAj+dQnYEmWAXcq2qdtmh5O+roqqcK2pq+pyue0PFzRAOUyncVpZwIUtBzYXyyZ3BVmEJmvcM7gmUfr51TU5RCZwcUmgOqrqfB4h3KRzDThFPUzjppHAevH9P4/U5ZANL0Ml3hnCqyinR1HVCMvlLOA5g7LYgiatonbdRVM736vuXeufFTPPcbKrycEc0fAenuPOx+sDxWP38E0ibAtB8d0CV3wVF/UpHbXd+DjcvSPP8ZFUOMOanmCMnmcb2WaG/++CX95vyHGPo+hW90/VduKP09LctxKBiQccr0JV9Okuq3QdwE3l+uBrHRbm5sZiXuTuerOScU0jfziCJYOxbYs7rok26kH+kWNuczfWs9z5SHf7quWLSz37+YVP+e78tl5K//oNvNeXtqxqjy3oNcX4qqvgnH/yyKT97quMfPZVU9d6R3i+Ak9lopFi5t6e+9c13RGG/fe16Uy7R1/tY6+219a5TSIQmkNyF4dcj4SiKwk3Ozy7+R/9AiTjjWXW5RJUS+hLyG0rCPMiTtnrq80dwPPWfSPLBmDmEpNSvafkV+tQca94l3mOJteAMce7uffWpP/tMcp0hpCK3DjRGtyA/8MPLHUhTzBMhpFUF6q+4mHDKzbWn5zkX1nX1NvpbjHv86khy6yXiwBkc8A5GevfWNtYBGdI2IBXAfTjyPX6hsfMMLpyUlfWw5ujV0qYK6R6OIfc5PtH1UoyJCr+jxy1IoiDLHKDsZ5xrtQZ/9lz3yiELa8FBrZ2oT8eRxu42rv94LGnMV0flXO02GEE+eQVd5QCxcwDn1yhRvXz3G+/piog5ZyeSPJ0hRcezF9iDBqrrKx5kvnBNLZd6hlm9DvvZz/6qOXb/iVIIVNi3Xcd6KeooXt59qPt/dv9BU+Y+r4993u6e9heTB/ea8umO5FR7N+405RD61gDPc1q71RUbdG1st2L3rfdXEsIJZG2t1tt6BuwDllifJ5Bvs8/7kLhTEh5jD1XAnXGGv32MdBldpHS5CelZUvepaao+9xxz5CFixBhyqhLzO8ffDuSUZQn3OaSlYKqRAdZfnY72LnTQ6nQgRee4RH18URgTx2AwGAwGg8FgMBgMBoPhNcArMXE857mw/mrfwlfxq9f1Fex5S1+nFvjyeIQv/Ee5vkg9QaK2z57oa+qLU/36OMZXXH7N2u/rK1cr0Jf2Cr8KdbdWvz4sfSQPwq+/D5BM85On+rI7XiKBJH4p7LX0LKd9vd9iS1/9s0pf93ywUXqJvorvbiMB4lx1+cauvuhW7dUXwCh6pWb6f4TnBS5qra4bIxmtK0GDQnJXJsTNl0gwDfZCWeFbIJLtZTESecWqhwgJpJa4ZtzW8+T4uuuFdXsW+vfZRL+aPLmPX/ZP9QV3kusL5/Ybt5ryPFI7fPbBr5vyAr+69vGrvYfyoKV2c0gytyzV7zv4RcH39QwbReVcUTOTPCSma6HeQnw1bvn61ciP9NwPjjUufn2IZIpIdniIMRpG+lL83g39SsakgGoB554gKePu9mqMREiG+PiZxvk5fsHsbCk543iu+r//BL9o4Rfzx/h1mSyibSSBc06/YJTo4wt8ja/wa3GChIlpumrHaoOJjauqcmmd2DcI8OsJni1Bewa49xFYG09PyOBAksoYbBanfrjALwjDlu41iBUL+23VfwdsvItQ1MFYPZ2gXkOwlxC3slTlp0i2OwOLyJV619kE8QhspABl3MqFYJ2B8OGiWO+3LNgzN4uyjgUFkn4+fahf8CZzjTPvmhLIxuifh8/1zj/+UP384+cqM0H1zpbmn3dukHWl9tgdIOE+frEO6zFYVqqfcyQqP5krFt59objw+EzHHx5pbH/wWPH4CpLFv39N/ej9d8SQ7SVgBoK9FTPpaEvntMFuTepY4/ubS6paFKU7q+McE4ymuMcCrEqaMPSQ4HC2UF09PQJT44qSanaQnJNJ439S6VfGUfhBU766ozrcvqbEqu3610IPrKkbt3XPo7H6++Hx3aacY174xUdibSRgHOxuax3yi1+I3fP3fvd3mvKbQ73H3o769BIMpE6IJJZIPu4KtfkmUZaFm9UMBbImczJFkCidLOc2kjOniEUPH2v8zTC2XKhYNAfj7xbZrVg/Tcdi420h0W5S97EcTORWjF+o8eNs4MCUqVSfp7mufQx28X2M3SePFI++fVvsqjd2kGi2q+dKkXA15K/t+KX/olxucF70PN8FdXB/+46eM8A+4PO55vgB1li33xI74gBJ+zMwQSdTjYtpqvpJ0V8GMEfYQnmM5KunSJba71xZ3QdzzBQxvw3WwNtv6J1OZlp7fPS5WBvHYAL5C9X3PhKOv3vrdlMukMR7TJY8GFTsUy0oCt6CcuEndzfJxBEv3DtTex1UqqObB2AfgbW93QODCOvbhw/E8r53rLb44Ezv//AT1eNNJIXfG4qJdu0A+5QTPdvJk1X5rz4QK/JjMBqvgTVTxupr29fUpvceqd1/8kLjcod74Jn6XRZr3G911QdPHyh58AES+rYSjdEYLLp2zaoku/arwvM9l9T7xLLUOHvntgxLPvlILLAB1jNDrANymAnNoXjpRIqVAZg4jIVX9sWEicBWHCOelgeaf4aD1XjNodbYB7O03UPcXsJMJ4KZyynMXMCi7pa65hZiZQuMmwiJkNtg1izBKEoQDzy0V1C+etsZE8dgMBgMBoPBYDAYDAaD4TWAfcQxGAwGg8FgMBgMBoPBYHgN8Eo6ncrzXF4nDWLCyH4X0pdANLjREAlJkZTzGWiID49EPQuReLE/QFJcXxQqKF7c2VzPcDxHIsZUlKRlLWci3fQxfOqfIPHjEgk0W6BIM4FdVYl+NUlFc3xxpuNXkeSOSY79UO90BYko+1dFkXzvDdGuPzxa0Q7pNf+V4XmuqhPv8rpMMOz5lycgXCJRZzFTHTLxnwMNrkCS2DaStpWotxI017JSuQD1LAxX9Z9VOnb8RDTwk8eiWE6PIYkbi26ZIFmzB0lQP9H7XUOCqfdBN377xhVcE5RwJCpcQKpQ+qI4Fv7X853UCwLXHq7qtBeqXpYT9VXS+kiljTB2T8aidbbQdnfeVD9MIOcIPb3bX//O+025GktO4WOsTdAHzhercguU8HMkGYshudtKNCb23xTd9wwU7wztdWNfkoP5qSjBlNn0UR85EgTmkACSzVggGXdYxz0PUsBNoKrlLGWBZJFIttgCxXg0EN0Tj+zOF0iICb3NzVtqw1mufvgIMp8b+0iOC+pxG3TrOGQfXl2fidl7XUhD2xofIRK5h4kqdrurPjREG/aQEK6NhMvHM7XVM9BcfYwz5+lePn6biCBTSzBONonKVa6sk2yf41lPTtm39f4J6P+dlvr8ASi/793RO7QhC6CMZ29fFOJbSLK43QE9u4f7IlF3XI+FOebQEpK7HqjnVwo94+6BEizeeC4a+gKJzSvIHnLIn5lYudNDUvSJ2rTyFEdbkL20PPX9pkk3mBu3KEt3Op3VzwB5FGQiGZLjVpDnxOjnV0d6zjaSHe5h7F6/ornlARI1jiAPH081X50uIIPb0lwU1Ymtk1D9LF3o2RfpL5ry++8qVr/5jhIl//N/9a+bcgDq9ze+KeOKXkt96LvvikbfLSC/QzLRp0jWHCI5dYC+uxbANoiyLJtE9SXai/HVo40D2jFBO87mWg+FiCfv3brVlJGrek12vr+vsTiKGNc1LlJIki4SeC8qrqn07z30qXSqZ+xsa85L8X6MKW3Oeecar20kNu4jiXzoaf1wDhlBC3W5oHlFtfnB6HleM6Yo6wuXMBXpqk52hpp/rl9R3fcghbyPBNMe1jAHSObd7+r6d2692ZT3IRWfH+o6YyTWXdTPwLXEBHKfUUcxPGkjaSqGQQCZ5WKp/hegvvO5+lB3XzG6TWkH5PLHkF1T0tVdWztpfbVJeM65qJElqn+8t6M2PX1HCXJ30W9bmC/LXOPixbnqPO+orYdtvcOLF4qdg131baZ0mGN8dykznq76fw5Z9CnODWDyMYC0Z4C9wE3UbQopUFBqTB9hb3JhmuGccxHkzSnXxjBL2d/VvP/wPtZY9Xtsconqe16zd2hBHn8Kyabvq646HaQe6Wo9GcJE57yU3G8H0tNWGzIxJD/eRXxKIWfMUIcBYnSr3t/1YTJwFekZBj3V9xgGRv2R+lOAuLnImMwehiKQ/O/29Yz7A8WULuRuJfYccchPL+pHyy9hhGNMHIPBYDAYDAaDwWAwGAyG1wD2EcdgMBgMBoPBYDAYDAaD4TXAq9keeZ6ratroNmhTO4loQl1k+N/qiVbWSkUZIhXSdUQJrJCl+fhcMo8oUnkO6QDlETsjUSGvXxWFrV1n/u4io3evLXrkzpaoaS4UPYsOMZ1Iz9WGv3sLmbhDUKJGcH3qJyrTIGEb1LQh6I9dyLUunGk2KeConHNFfcUE9i4pKPA0/aiQbb+Yid7nUtHKYrrQ4I9D0AdjT9dJ8bcRaHMVpCAl6sHVDjoZaPezu3KVCiaiCXu4z+6eKG7XkJG8D3r4u98VPXwXFOP9bVHifFClPQyZuK2+cA7ZUFlAXuZ/PbRxP/Bde7jqf+FC9MQETkIB+mGLYwXUfjovvJGB1gfq+/cgywlB+e1DcpjACanM9LezqWRW2WJFS/Qr9gude21X43bpi2aZQ9L3+z+QFCBwkNyhXGWiLuNP3V3IgmYfS4JHKmYI2i+dhvwLCdoGB2NVVa6qx0iJe2VLSpIgjYEDXpvSI8TT07Gon/0e+kJHMe9gG44GuP4AGfYHvQ6Oq346dZ8KIcnrdCmJUnmOgIfT3bdvy81wCaeQGOOsN9S7PjtVbHr8WBeqHNtHf+sjXvvoX3719YxF55yr6nF+DNe+2RLOXy3VZ8fTWEwQL29e11y4vyvZzHfoBAiHixD08wTv3IV0LQ4YUyEvqcd3SUeWgeazEaj6N6ExXE70TtW3RIUvIYPinJEgRgzaaosAbohncAiaQrKbwi2rBTefQd3H6Cz0VVE5z2X1+oOyqRmkDGsGR4gTMc6/AmfCO1saczHkDrsDxeUffOvfb8o9vKO/UB1eu616Dodar1Td1Rjx4IDZh8Twd//DP9B7nGnuXkxF2f5bf+N7ein0m9vvaV58E246I7zH7Pmjptz9VE4qzlMfoQFmAmlxsfh6xuLK8S9tys0j4RwPcxuP99ucF/dQhlQQ65JtyEjbcNO5PtScihDsJseQrOPOVT0u8+rydVEHjpiFp7nhBKkArm/r/ttY30QYI9s3NaYvnCKdW5fLlamumUAWModjWwSnqrKRF2xQ21hVrqzXIqcnWttcHVwuoRr2dXyI/rlN+WZbsc3DuncEqUYcU26t983hXllQWgxJXD5dtcuCbjepytx/dChFxtr54Lta21ToH3PIRjK4ufp4Fs6vDvLXGdyv2txzQLry1jU5Lm0WXuO2w5QYia96u70nmUuFYMG5HRkX3Bs3FYtuBOoD347Vn/82nIqCXHW3FaBtIBuMW+obQXv1bN2urucH2l8sMT7iXPvS4Vix8Peu6nrvRXrez54+a8rPIHnuxJScqkzJJffGV6/LoTD5mY6fPF09Q55vUjbuNfJ0D7Gk21Fge+ddSQ/vfa7UC5SA7fQ1F25Dqh/4HIs6p4v1Lft20tL54yndDlFv89X7eyUcJDH+W4jbSQRnWqwVb1/TnodyJx9yqpjOhniubqy6qZD/JcSYjrFeC5Dywfdeve2MiWMwGAwGg8FgMBgMBoPB8BrAPuIYDAaDwWAwGAwGg8FgMLwGeCU5leecu2B+9eA6EEGK0xuJthSDf8ws+beRvf8hFEznoB+2QT/c2RFVKYILSd8X9ej2ru57a0cUvcStqE1lW9e+vSNaaYbvWL2B6FwTuC85yEy6oIJ1I91zu6vnur4Hh4oEFGJQs/eQlfv8WDS7E2TO9xPRcjcHzxV1s3twgyrhBuX5pKaJyl/MVA6RmT0Blbgs4AwDKn2aixbrIJsIctVhnur8AJnkXe1mtXgmZ4D5IzkQDUEHHoGq7sHZ59093ec6sofHXci/6LIFDud0eXm/7ILmfHYqGYCjs1bMa24Ovu+7Vk3hG8JVJI9FD6SUwQfduUOqNqQ4N9C3sxL9Fu8Q4ByPDiNwnzuFK0M61TlRHQ8iuMAFqKs4VP+6dk1j2EVw+yr1vF4J1wCEstBTf5xDluBB9vD8TM/4/Fh9MwB10kMfv6iNTUobPVc578KJioYpyIZPmWY7pmRG/bkDJ6fdoeibga92q3DO9khU4QqSxwRSjG4bkivIqS5MzkhHbqF9BpCTeAs4sIAS3dqGbGSg/tQCzZ0UZx/S1m2d7o5Bba4gvwsgA6gwjrMc2oZNonKuNqdyZ5A4oJu7NmVQkDDlGDdhRAkb5J+x5qUCDjol5QIYxwld8yDFWoDaWxWrc4pM9wdD2vUHuraD/GsZa7xmGP8+ZFsJ5D0hKPwB1gxVhnVCoXOK7L7utdQ48Nsq37q9ov8nyQbb03PNIE8pK4XMsXiJZASKRLeDdvvGDckUDt643ZSHcBXrDCUh7ULiEGEu2npD9HkfkscLaR1Y4y7GXHgDcvYZnHSmh5IB7NyAww3GYh/ynFZb8SJCoEohP/Jx3IfDRieG9AY0/+IVlfyvgqpuJ7pTrf07JAttxK4DrAsTxJ+dbV2nQLz0IripwHGIzjpzSND5NGyzrJZNeCWk2pBP+LlO3oFTEuVfDlLJam1ehKQVsYOSeYdxWaBu2Ad9SNAixJeolgVs0rWxrCqX1k4uD57ISfH9O3+zKY9eaP3XhQx+AAe+3Y7aZBeyDTqiMoL4cI8psS84g4SuQB+uStVPVjsYLZf6dx/r6NBxnoUbEqRpVaV2o5yqizFdtTUuC1Q57+XhXseQUSaQX+3BlesqJE0bReVcUa3uOUN/W2Df1u2pDmd4vgJBtYV1ydV9yYzZeAEcBSusL1K4RVZjrfPofhkUkL/Ushwfleuh7+By7vETufBdD5HOY0tttJvoOvk2xi7ScuRwrTtbYO+FdXqBfnLrjuaS64hZ5x/8fPW8L4l7XxZe3Yac/iLEuNt3IHH7lWRlszPsm7lWwRwWw7GZ47gDWX4IGe4Jxt8MMWdNIlvvXUqMoQjuvKOR2uH6VR2fzTk/YT5DfIwRH0tE9ILpHOAuu4A6qlhL1aCOFCK+HEDS/oH7YjAmjsFgMBgMBoPBYDAYDAbDawD7iGMwGAwGg8FgMBgMBoPB8Brg1eRUVdU4DnVBJd3viJpXtPFdCBQnUu5vXxF9z3shKtmjMbhqPmj2oG17uWjVN0E9+mt3RF0eQuZSlavrdOGMcbAt2uIClKgo1rOMQVstMtKpdJ0RaGG3d1UH2129K2UHaSEa8wHcPzqg/+eQlHk1jXCTVFXPC1xcu3ORdZeDNk7Xj+UUmfFnqp8IEpsSblPrWeVB/QfFuh2ASojM/xnoxF3SDccrqdL8seiLLdIhIVm7stB9fErlEsi/OsxaDvkH6MMFKMxLuC0lEWh+Ac7HfdMlKHe9r0fCEfi+G9Z06oVTv1qCeggmuytKtAvKYOq6PqQwPuRslG2EkFNVoA6fHUq2MYYcqBdiLNQyBSgiXRsPuVyI8u+WGsP7I1FoPY9jRadHiBfEMaSedNO6clWShgmGVwRZSLVQ36/qvl9tkKrqOeeimjZMxy6XKj65jGU4KkCe2oU7h+srJlIGk4GOm3l6B2bGD0Dhj9BurQgSAn9ZPwqkLgEcCfvqNx4opnQ6CXboUgSXFlBP55QwUkYG97VTyKk8xJ3Qo1xpgfLX5E5VOVemq/c4PJKsssBcEYMaXZZwPoGTXQRKeBdSiQ6c3xzGXwmHAw9tupxDwoFB4tFloaYIhxjnFeRvAeJfAolQbwdzfUUpAFwVIRerQO3P8SxLtIUPujrp1ZOJxl/hK760L1xqNmiI45xz3gVtmtJG/Dtv5yPOlqiHNmjbN/Y0x996+1ZT7h5ITh5BntvCfFJBWtjC2sJL6HwY1M8CCXEAV48hxjPGX68vmTYlo1A5Ow+xko5ojPlQhTiH4wHkjz3IjArUZr7pxrt4vqpyZU1Vp5xqrQzZTNimnBNujpB2dDBGPDiJugBzO+QUJd4/z+hIRWm9zlnWkl8otZyDnDmAtKRN6RBcUEr8cQl3OFeQz486QP1z7GZ09EI84vrQQ9+/iOubXKNWlXNZvUY5w/wdr0lGsVbBvOXBpcbBHTDiWhTvC2Mg14EsvUJsnegR1uqN0rML+VeKOZqBpIDcLcbc2kW8WEBanmINFaHf+NwroMq5vqNpX7jmrKrnHcHdi8q6TaJykIXBETjqwbWvggweY7TCOKMLVxRpLvTQ//01Vyc4DsE1tgXnsgjti6p2WR3U2F45+pfDfbjVWWBvNMf+rz9SPV9FfA/OJWM+m2m8nmP92Yb7cYn7Jth7DXFOFGxe2uic51w9v1HiRwesCPZR774rqdevfy5pNOuqDalUDxJivksH8xzXDce4b4A45CNWBfW8GCKdQI51CN1oe1hn9ZEShXvBCmtz7oFTrNNn+EYxR/yd4m/nWKPl6F9beIZ3ruo7xr/8y0/cF4ExcQwGg8FgMBgMBsP/zd6bxEqSnVl61yY3n/3NMWZEZCaZTBZZE0Gyq3vT0ACoBAnaCGhBO620FSRAagiCtgIECK21NhIgQAs1VFIDakANtVpdg7q6SFZVkkkyB+YUkTG8F29+PpvbpIXbs/O5+KIZwfBU4xH/Wd3wsGd+7Q7//a/5OfcYDAaD4RrAXuIYDAaDwWAwGAwGg8FgMFwDvJKcqnTO5ZdURNATu6CMnSWik9OZIgQluwVJytfuikIVH4gGvn+q+5yBthSDZvc7b96ty3sDOGLgNOnkkrUEuuugI7eF+6CIhcVJXT4Zi2PnQbrTBeVydyBq3x7kV0Go788y0fmSuU5HD3HC//aWpB0zWJo8Gy3rUK7TE8f3XFDR8UgVz0G79cAvpCNVkohK1gatd4prXAtuJJkoaa0AlGSwfWfo29YGZCGp7rk4XLp3xWN91tuUq8cZ6GujCejeoNP2QENtttVvcRMuWKDZZqDlRrByCDB2CzIrvfKaJQAAIABJREFUQWfO8B9Bc53URqCUYwRPfo8HGts+xlgCmm8C2i6p5aTbxqA8enhmD8+Wo+8aie7fAf00Ie25emccQRISQ5KY4ej/OaRMlH9tbkiikMNBgK4R8/mp/hRyGjeH6xNeX2/d1vy7d/9OXT5/+EVdblZtFgVySHtdkG7sgV4ZgHuawx1uPlZ8bAYawyun/WMseKDvJnSVwTj3KUPySWHGaf+kgVbtWYI+GkHittHVvMwRRzKMuRjOVz6kAgvQ/VNQaMuSc0jfldDNbqKx2AadmGM9TTV21oksK9zZ6bKfJog/pN4v0M4ZypRYuJJlfAH6hetGwbaD89RspLUzhVw1xhrlV5pGH+PFD9WGi4Xmot+AVAo09ABSgBWZIWJBCc5/Wqgus0x08jTX56T2Z1iDj/a1NhfBsXPOuTme7XXhOdHQA8hqSsibSaWn69kCkkAq9hpon25PkqjOJl2Q4NS20P1TxM0IbooOn182OdId569qvupi0NTaGsJVs8Q6x9hOCrkr1Z8F5SKQuqRDSXpDtBPzBEoCkgVlJ+uD55wLqn5k1Ci5WGNyFVhbKIOi41iIxaIBx8sAMrMCGmGOyzldo1COIIELq/g2g8xygXU5inEtZG5RE7lIAQdHuPDRdbQstZaUdFmCU0qA8cv2W+CaKeSSUdW/lF69LjxXOs+r5EmZYtLx6bHqM4Mkta3+nOPzaaKH6eKYhw6cx3rI51vIES+wdjG2Uk3jKOGojovwU0pJFSNyyIZyWDW24PzWQJydTSHlxnEFOeJyyjUF+SclXZQFpcjxW5H2bAnm8TrhOefC6uubodaeKY5TKI9xRALkNIydEfaOfgxnzRJyGcim6MjWbyp2Uv5ZwIGwgNTcr/a1IY8QoPQKbo8uxf4GMu8GxlSTxy9gbl00qB2H1B3raItunbp65ZiEFDn4xu5Sphucr29ddK6sXZhKrHM51o3x6AKXa+yFAfNzjFs6ITcgbcW3+lyLkGsWdIylVHAld102UJxDeoh4ypwohZsvpeUuxNxCvpyUlK0iT8c6wvx6nvNz3b4JCdW7b3+tLr95E+5r/+s/di8DY+IYDAaDwWAwGAwGg8FgMFwD2Escg8FgMBgMBoPBYDAYDIZrgFeWU2XVe58cJ/P7pJVO4Q4EVxFSeynzaIPKdn8XtHy4NhwciMIWgWK228YJ66DceygXyZIuVYDO1UBdBvieGU7EzkFn7HV5grWocm3QXCNIxwJV0XlT1AU0P8oLtkDv9Dr649H58rl54P5aUDX/IiftEjIAuJtkkHPwiPEFqL8BxkITbdigIwDcZiaQmDV6kIXAZWp2Ivr87GxJo93s4PTyhuiZ6VTU/LANenQmml8MaivdASLI7DxHGQAkJzChIEV2ntIJS2hiTHOMrBOlK11a0QwjyP12u2rPpICUZEEXMNDa56JjUnrSxryIMc8ySJLmhdo3BW2QNic+pC3N7vI+XcgZBz3Jb1JQkZtwqVnMNHfzLq5pazxMIMVarLhwAD7lAvqfTkv32d2Vc0yHkoJqHgTRI7c2eM4VVfuQGkqpV4k+pNMQqap0k2uA+hvDEakJB6AZ5C4LzPUcspaEkh+PkqtluYO2b0SKj3kGijlc3cZDxVMvhySHMSKHbAA9F4K2XuZwM0NM8UrFgCFkEXkg6cq00LhbJ9IscwfHS5ldApr20ZlkfW/sag69sbJW8ZkZK+AEAaouZRApvuvsWPFyeoG4R1p6oDq4cvl5gLg/x70nlNFi3A1ICYbegktUjjFbsu4ly5DoYJ5RrhBAojI80zOdDA+cc84tkvXRxstS8kz2YYG+oqyvQP/Mcf0Y+U8OaXQIWRnXRUqYkrHGcNFDfgU3EiruvCrmFoVieAlnlgKSkBDfGay4LXFxQ3yEbMMtVC9vJqleMZWbYDLWWO9A/jBDbsj+zClLWCvKmuZOujvnE8cYqf0ZXdVwkb8SayGbQNuViMEZYlSB2BlBitWE01JU5cMJ5h/lwbOpYnSbDk2QpdBNxUVq/yzA/Ea+R4lOifHrsAYv0DYJ8qEpcoZenbuuTzbu+8714+p+ucbebKJ2GHQYyygrRSxparw1MIfacMRpQvpCDV2Zsz+5FtJxEDL2Ks4mM5wVACAlcTnd5ODy22pjXYSkM5tJ7jSfqz3Yh5R8UDmYIv6OEdPbG3CLir8aB1XPORdU+UIjhNtsV+uwdwAZOXIyn/OVe0d/RXRTlyLImSh/nENuNMZGKu4oT27EGkudk2X7DuCO20UMO3e63zn0MV5XeWMXx2O0YX+28CFtntItDXIqHt2AcU2Xu5D7MMzFvZvLowDChwduXShdWe8NaXoVY+xvbao/F4me68vPJJVbkT9S+slcjbGY84WSX7RPiL0m3RQvt2gBnXQh355hv0hHyHb3ankW3ew8rAuFY57AsSvQNavAcwQYx9s7Gos39jSOXhbGxDEYDAaDwWAwGAwGg8FguAawlzgGg8FgMBgMBoPBYDAYDNcAr6T18DzPeRWdlCe/k9ZNZwIHGvbKSc6gZga8HOXxuZycJic6lX5z80Zdfvjoqe7vRMXqw61lMV3SzeYpaFCgZKWkwYL6OpuJvr2Yi0Lsl/r+GHTaBJTmBqm4oI3npOvyhH/QshLIXpKKF1m6deqpRI+bzUnNB00zIYVVzx73IPuCS1gXp8pHoPImoK0l4MfRQanZFp0tHarP0xEoxL0l3XGwq5O7Q1ANGweidcdnau+jiWiFCVwOfDho+ZTBgS6Y0a1LH7vRWNTWMSR3bdB1N+HI0Q2+GqpqXhRuOF3WpYtxeOe23JUuIJXZf6a5kkBaWIB7S4MZL8EYBuPYx9whWT1zdEEg/VjXNyvqcBzohhu97bo8PFN9xxegnoZo84Zo+4tU7b9ArEnoDgIe86zkPNN46G+oDh1PbRlsaJwepss6lOtjjTvnec6v5E8BpF4N0EQjnsC/Ig6jUw4cw+iw4sMZDIN7hvg7h+PbbKr5x/nd74vueantnEHitoC7RTJRPxwfwCUJMo8sofMKKLEOAw3mNRH6MAetOIecMYVFz5id1FLdvf7X3VcBP/Bdu7+MUfPHasNzxIrDY32+uHtfZUjC4gRyCjpFgW5Nd6AR6PFzxCIMGVdiXSxKzItKFlOiPWMPcmIf3z9RHB8vJNvyYMUUQrpHl7+S7jiQRjjSqBEv6AroQTrd7sHRpJLr0P3yteF5rqwozgvICssXLr2gyWNtu4Akaj7Dmgr3pnKsuJJQTnWhvu1vbNblAPPegyTGrxImD7mHj7rPz9VXC3xP3IOcAa5VlIowByincB9BjEghocpTjfUW5OdHcBhrQE7F8jpRFqXLKjehLLvaSYROJTPIh+iIR3o83b9COLREKFMOW0IK4+F7fchCmpAjBBXl3oPkLkC5gHR9PlRMDdCnbUieG5g3DnkMpc18Jod2yuDEtEg0rjLEYEodyrUuiEtEge92B8sx1MC6ODxXntduagzTLYvSU64tAWKFB2mhC1mGrGWGow/w7CHyuRjt3KkkWpMxHGixuZlibI1xREEG58cG6hJCHpSlqleIcRbhWIAZ+opOYgu0DbMHH5L/BGNknSi90uWVnKqEJKl9+1Zdzs7kANpoMrdEu7D9ma8i5/awzkyxdyznkN935FQV4NiBBiT9ve1l3NtsaT7twUEwamM9nasfQ9i6bW/gSAM4Ei8gG26iPZgjRzgyIsexI1NIWumEWyCXOz5afleWMjN/TZSly6r9oo88hLloCCfTZowyHOGGUx7bgHjzgthDR7BV+Sv26DjaYUUKWc3jEXL8C6xnTdR3NNSeA1PRtdEndHB1CdcUrBHYJQY4coNHAdBZjn/bgFPVSjx6SRgTx2AwGAwGg8FgMBgMBoPhGsBe4hgMBoPBYDAYDAaDwWAwXAO8GnendM6raED7I1HJno7hGuREPStJFSwp3QGdEFW4gAPF/oGkMDGcUKK+qEfvffR+XR5B9rO3CcebyZJStQBlcAI+6BRSsJs3JUXpt0W9G56KNvzl4yd1eQEq+g5O6C4WPOX+aqeOAt87As3qZApqaPX5iyndvw7K2vGmoPPUGNIH0ATpQtXydaI9TyfPQdnPxap2i1zt0+rhNHzQE2cTUNcxpmK4tmzfvuucc66NU98LUOnCU7VfuwHXhw5OlZ/q+e5DZhI2RZssQEX3QnDrFnrXeXQoWVIKyd1uR9TWLdDJ6Q6wVpTOXbIqDy9E2fQL9QtpwzOMySEcDgZwW+tgTM4u1F6UKmVObT3F+BnCbQQKERfRWaVynAogcSnPJQOZzTXQz0b6PMdzZBiPdI2jg9toqkF4ASemC8ytBM5J212NTQ/07RAnyLd6S5ejFSr2a8J3pYuDZXs2QdPuthVL4lj9GYIDT5o+JTaZp/6ZzDAvC7Xn8Fz9fwHZIh38SnTiFBKRs/PluKCEttVQfedwCTs90fzoDdTn27twqoCTkQ8JQwY5wRR08gliSkk6K43EQPFP4CYzaL3tvgo0osjdvnXbOefcTbTtcKY23x5oPYFSzU3nav+gQRccuCnCZSjz6XYAhzXE1ykowofPtF7tIGY2qjg5g3w3g9sJdYMbm6o719Gzi+f6ftD/aRSRo44pqN8FaO4FZFY5JM0+coMu3PIaletlQN3Ya6IoSzer6NxzSKApFuEvXhnkVAlo5keIMZORxv/wRE4dUUfPldLhCO5hiwuMC0hhw45yEVcun99D2zi0n4NziaOTDeau11X/0+XOBxXdm0CegGcqEGcDUPwXntaUQ0gKe5BOTyit06OuBX6VMFEexlWYzmIzUPjHeOYe4muHcYYuQJAEuoBSA0ihMOaPjtV3Pp1yqjXy6PCo/izEuNjeldw3gJx5gpyN0tIm1gxaFVFykiPW061wBtlBins6yN9StMGiOqZgnTmq5znXqmQZOXKPz58+rsubu/rCCGtnXmoO5TyGAOMzLeiCg/wE82ICSfYE84gOsz6c/+JKqpOhj2eJ5gqWMzdD/jWHJNaHIxndM3kUQb6iecfnkOokfFbcJ8B8aCOPOT5HTFkrPFdUMTxA7tK7cbcujw4lLU7nqkeZKShQEpXBNcyhPEMOnF5IdrcBNzdKSrOR4lKWqi1G82oNwBjZ6Cu+UtY7OVVM97EHopoRpkzuFHLAs3PIieno2VB5wbUZ83LCowN8HlOx3Pf6a8xRi6Jw09Hyuz1qMAvmXsgbUM8IuTQdcecZxyrkmzjyIkScm2LyTCCnenK4X5eH2Is/riSnc0i4HCRcd+FA22gpdgxxtIAPORffkjAm8pko45zju0ZT9eEMcvl0oet/+sEndfnH2c/dq8KYOAaDwWAwGAwGg8FgMBgM1wD2EsdgMBgMBoPBYDAYDAaD4RrglXhXZVm6vKKB7sN56M+/0Anj726JevYWpCol6ZulqGw5KHGHx6KTgkHv9m6I/sQTu3fvyq0oOheH7Rx08klFVyxAJaQ7VQM09BL07LApqmR/S/Ks5/uipz989FB/m+zU5c2B/jbAqdkFKKn7C7XBR2PR0i8gh5lWtDieKv/aKJ0LK3pwcib6YsAT80G7bzZVH56en4MnWCY4PR8uAJ2+HDZ8yGpSOJvFDqfWg6rY6ao9o51b1T3UrmUoqYLXUn83+5J23PZE1fQLOmWp7UNf31nQmQE01BGe9eHnn9blBiiUA0hyYlD05qAXrhWe57xKt3ABeu7jn39Yl/sDjdubG5qLU5ywfjYEJT6AhAkUXn8GKjUo5CnogRO00RBUVc+Ha0C6lAX0LlTfkwNJBfaPNP9bfUgoB4opKeQCw6Hqm8FN6yzR95+PNTZGC11zCmedTqLxTrmID1r8pevems2paveNXltje2tL8pUW5G5Qbbg53J48yP3SAnT7meikM8y50RiOeWDIbm0qzi7ghjCBVG5jZ+nORxeh81O4U2HedLY0F1twXUghV5mB8uqXpPsr7pyN9V0j9FUCansKSu8C9NsMNPNwvtZIWqMoy9op4e37t+vPtzHn7sGRL4w1nxag6l/AVaOAc2AzhmsOxoAPGU0U6z9ixOwclPuHT7ROB1XcLUCRjlu6Xx+SZId4FqJ/Z3BccphzkU8ZCyjHlAvQQQNjMChpraW+8z21QaeSG/r++mZjURRuXo0tb8UNEtLoKz9dXZ8v0J8jliHJjlta8zrIbZp9ff7si1/oez/4q7p8967cMadV++Sxxtn2G1rzorbWpCDUvYuZ+qo8g/4ZVH468lATQBlQDjp5CNniMSRUnx7LWTRG7jaA/NwdShb0+ihdWfcI1jDKEJELpgUlNMgR6PYTQyoRgk7P+4NOT7e1HuQcJfKLKVymXHXNYFtxP4bDZQvun3RQcrnKUzgELhAjoxX5JaXQeFbkewnd5CCp9OG4MoL8yq/2ASsyn9eE5zznV64ujB/PjjSWTsZ6lrfvP1B9IGubI97QHSrA8QVzB7dDuiBCTkyZ1RnyiU+CZ3X58oiGE0ji6MIb+pTMwB0XOdQ45RiCkw3GWQapFKXtpc95iWs4LtFHnz58VJd/9oHyqHXjUlqYI1aHcINq3XmrLg+f4ZlPJD3imKSMuzlVX0wxNtotzBdI//Kh8pgcyVTWpuR+OS89OH95U4x3yGC24ZA2GWoOnRwq1neaXOt1zYdPJAWaYV/Y3FaZOUCJseRh2/6v/eG/XZffuP2mc865v/xP/mO3LiyShXv6cJk30FE5Rr4dYX1o4JiNbeQnTzONsQny1VmicTGHw+Ucz3400hw9guPuCdaxQ0h7m+fLa7rIOXs4wuMUsZJzq4ExWoR6vibWa+Ywc8zRBebZEPH3HHF2Ds3pBP35/i+01j+BZPRlYUwcg8FgMBgMBoPBYDAYDIZrAHuJYzAYDAaDwWAwGAwGg8FwDfBqcirnuayiWHqQwZyDPnQE2v59nLodgkKPM/3dFCePT0E3bW+IqjbGqdU8wb0PJyIwil3HgQpXORqVcJoZzyGbKUCMBs2WtEnSuosWTjKHg048xgnvqFcH9LI01jMdgaq6j9OsWR05AqyPNl4WuVtU0qnTIzmANeH64kES1WpBbgQpSwmaYjOAgw6cAhboWx8ShyZkdnQe67T1edCWFGsWL+tAupsHWnEIyiLbr4TDlZfjtHNQNQP8QQlHCLoaffJIkp9nj0VD3elqDmxtiP48X4i2+dUIOJb08E4lP+ttqa1OID0pGmpbD1K0AnLGEajvCSjEow316aCnfmniPtQXdHtyH/IjXZNiLJ1WJ8g/25esI4c7z8Yd0WwHfYyF8moHowWo/YdwNrugRAcRLsRYm3s4uR7yyjHcV1bmbnUqfblGGw7fOdepxvHNHVHpb+wqTrS7sDoIcMI/ZF8FJHslHL5yOHAxzjLmbW1r7DQwnlVyrrMJN7dqfhdT3aPbgDwL984hjZlCnpCkkg1NQJUGC9wVkL5dzFSeQpabUrYDqnheUmZFty70+RqRF4UbV3ReMKPddhcOICeKta6n/t1G2YO0eDxRn45Bwy4x6QaQSzYpnQFd//7XRVefwX1hVlGUI0qL4fbmQ5LlQsjTIEPM6DIBOW5ImRPjMeSnEeQZnqPkQ39AaaML9L3tzrJu65RTeZ7nwkpO3cU6wFyB48p7gbZqBIr1EIqxswRuU1O1bdOXbNhDm3uIPc8+kETWUQK9uaSIb9y/p7+jS1JHMYX2TAHmSjnTXMxHlGdo7JbIi3xIATNIxXPIqU4Qf48mWl/ehESv14ajlnt1CvkL4TlXKXFcgLYIQ9ZVl/PZ5sg/Z8hp6SbE+UdpUxjD5QjViSCL2gzU1yGk6Y1qvAWQ3NDtyvmUy0JiyHwIfUF51AJ9XUAOsUiulmpAIed8jAE6ruxjDF7KJ4pifeti9e3OOecyZFBT1HMBmQLla00cg0C5W0KzGTQt3flmOCJiinKvjdwm1P3PIc8ZJctjFppwtRl0kDd5iHeeKnB8oXwjbGgONWKuwBxzyO+whwggRYlxH5+SY4yFz77UnBtNvyLJP1DimfNYeVjz/jt1OcP6MD3RsRXZXNJDr6l2zEaIXdgjugjut4i7UQ7ZD3LUAm13687SoTjAtSUcPPefQgYF+dujfR2JwWl8f0f5VQ8yr6KttfYAjmZfGyBnxx4kh7wo8NS/f+tv/+t1eefG8tiJdheb4dfEIpm7J79YHiHBvUIId62NC7gDY68w3FW8G07VV40Csu2IMkPk8Nh/PzlR/8Nw1WXYa/Y7yn/m1ZqdYbxT+nSCHH8Mh78u1t859sMbuZ6JMXcK11DK/Id0PEzTq8uozwX+1q3M+5eDMXEMBoPBYDAYDAaDwWAwGK4B7CWOwWAwGAwGg8FgMBgMBsM1wCvJqZxzNVfVh04kg3zpQkwiN8FJ4h3Qkwo4ZixmOEmdVHnS70HR3wWFvAVqU4rvbXVBJ6xckTJQD5u+6EspuNBQ/LgZbjgE3XAC1xYH2noegZZLRxRICkjpGoM2ngSi9oVwl2lUp+V7a2SqFlnmRifLk9zHE1HcIpzA3SV9FBRyyjYyR2mH6HEcF2EsyuAAsoFGX44c3qYo3zlo7B6ovIvqxPgEp6B3UtUlHOo53AQOS+hzr6E6TkG/7ZAGjBPp98902v2PP/hI9wT99uYWJD+eqHIztFMPJ/GvE77vuXZF/7t15279+dMTnLQ/U133+nC7AJW9hwiQZ2qLGeZiH1RgSq4WmKMt0Iy3m3DkAFX1xs7ymoLuQXC4SUGhzIag1uZ6jgx02gjyuyJSfXs7+jwK9Pn5KeSAiCmYum6KOkQDuEhU1OwVOv1rIgoDd7NyablzU7TbzU31T7NJpx/Q3lFeZHRvoGMYHEggT+p09V1BwPkNSjKkLPmCMosqJi1QrzFiO5YUutcUlFZCStN0cB7DbwozUvlzPUcEunobMStJGY90DV1p3BrlN0Sapm7/cEmn7sAlio6MT58pniR35CC00Vd82OhqDuWQCk6xnsRNuAyBinwGWn4DToAN1KffJbV72b75lC41oJDDzWXmUy6HtQ3SmrKgjFb3DCAbCahjoPwN8XjOMhxEog4khr3lmuEHr56+vAiB77tuFU9C0KpnC1HdSc/mukiJ5RnacH+iefGNhq5JMrXP2dmJPm+qf7q7ct58+3t/UJebHqTX/rLcv/2g/izsYL0JNLcoVaDzho/57xBHPNrWQcqWJBOUIduJ4KwFCRUlB5Tk9BuUU60PnvNqqVIjopwa4w0uWZTQpxh7lJ4sMI9dqj4tkcPFkIv4yDsc1lQf1nIRBKtVNzoP4yhHP2dIqsroaglNhGcKIcUr0ac5JPABnrvEvAwK3TPEd40Xus8Qxwgk25XMeJ3Ccc9zfiVtpBw2Q66+khNjnWlCjhCg3Xy4P45WXMiQtyEWlhj/IdaiLmQbcV+566Jaj0NULKRkFEOCTjZTuPqVkD+3S+VQDUjlWk3IJZGXRXCfi+AmSoesFOVTrNkjrO/rRlnL3ymx1VqRxtprBFty3gtxJIWXam3j84yGeoY+jt9wzPtLjnONgQx9w772q1g3gCvozR31uQdX4f0hXFYhIX12qJgeY6/bbauPbt6+U5e/PFfO2ShwFADi0QRudnPE14sLzcXBrep51pjmZNPMHfz00Dnn3FM4omZTjZlbE7Vfu6n6nEGGdo7YF8DZqhOqn6eQsJ1eqA2n0FCFPo7faKpfdpBHuV7lJItJx3KJIxkS5s6Q5I1meg4eXUC3wSHcsSh5nyGOz+FmlnAtTBiXdX3D5FQGg8FgMBgMBoPBYDAYDL+ZsJc4BoPBYDAYDAaDwWAwGAzXAK/ER/acc0ElJ8hB8UtB6xuBPnoBKlErFv2KJ4kX+NsWKIHHoGymhSi8w+eiWZUTUZgaPP0dp7NfMrdK1IWUvNxdTSvNQTPOQPPyILPqgNo+CCEpAhWdLZygDhltlOAQUpJmXksm1ifhyLPUXRwsT1hvwAGhAQp5itO1fVBwQ1Cy446etxXj2T3REJ8+OazLgS/q297eG3U5wUnyga9yCCeZxumStrjASeJJob6fg+IYgWLsJuhPUOg62zo1vXtD0ob9M7lWffC5HKl48vyNDUl1Om04dWCMRDHcR3AS/1pROldU9LwQNNFeT+18fg43Gkhl/D5Oh99Q2SP9F/T4ApLAk1O10RBuN13QX+kax7aI8srZCJMC7HyXzTVvh5AiXExEy+wPQEltiUK7ty2JXoq5mySKI+OZYgf52BnmawLZZwoKvlfLctbHVY3C0O3tbjvnnNvdUr+1IIGJQGn38d2UKpUo+z6lRBoXng8q9Ynac3SuedRrQeIAinECOmsUVf2J75wn+ZXl8VD3vjjX3N3c1pi4AUeyLhzJqFqdHEjS0o41RnodUbGPQSv2E8iywC3udr8aCUfUiNytu7edc86lIz3nGVwV4raek3KvOSi2eQsyzwFpw2qvBeLxaKRGOj+QXCs906TyIX8qYJ0VVHaOBWm9lNOVlHNgDRioLnv39ury9qb6olOq7gtIOrMC3/WCOTfMNOc+eKS4u3tD97z5tUpOhbH+uvD9oHbYu8A6Q6kX576HNZlSkiHcOT/cV7z53juSCtyBa+foiVzLHj3U9QtVwbUQL++9Jelsu7Xsi6dncpppDNT3rR2NuaCNfC1TDG+F6pM2XB5dAQfPc82tBeSmUBC7fbipPYTbCmXRp8grPMhz1o1LujwNnsqrFVQrVjIz5oWUYULCSQk9ZRgpZHRTzL98pM+budpihrU2uPwtFRXOsIbN4OyahnBp29Sc2Lul9S+GzNRDfYsZ1jMMaygdXIx1cYacZoy+a0LGeClv9bz1rYue51xUyakS3JfKd4q3CuR2q6odSMMgU0jnyhfp+LiAC2YO6d8GrG/bkeJfA86t7f4ydlBCOx1r3mTY89AZc7qgFFnP0e/BeRfP0YQkJ8UiWWDfkGJNSSGVyjBe6ey54tC7xr2Gc6XzKjkRcxcfR0akGEt+G04FwUsAAAAgAElEQVSNkM/6ieIi1PGu4akvuLZ9+aUcpIIATsGbt/Q5ZDxNmANdHscwOoTDbA6JHlwgvRSuppBWeZDWPEB/dbYkv9uaw/ETOQr3Wwn2NWfnWt+7B9qbOOzJbr+93Mus00E1TXN3+GS5GC1iOL9hn3cEl6bQw7ox1rPE2OclWP9OsVegg+CC5rGQGRY4RqPwKaGH5LDa/7MZCsqp8O4ig9yNzn8h4qY3QTxHjJ7CqZNHvtA1LkspUceeA+sF+z8MTU5lMBgMBoPBYDAYDAaDwfAbCXuJYzAYDAaDwWAwGAwGg8FwDfCKcqrShRVtryRVlZRU0IqOcYL1gDQ9UKlLuBS0u5BewGXIB8WzBXlGD9Q20hwT8KhGFUW8BD2SXEwf7hk5nikF37QEzYuU8ABuJ12USYssQZudzOHyAn4vzCJciZPrs3L9jjiuLJ1XLtukDXnBHDRQD/TUVkz5kMqDPdG6G1uSR7V7oiy6HVEAx6Bhj0dwxYpFm0zhqpKCKp9WjmQzcmUzuKHAvagYqXz29PO63IOTSvNNfc8EFLcPvhC1/S/f/1hfheZv9yGV6kEWAWpdsyWac68jCuU6kWe5Oztdtul0IXp8iHYLIQ8cTkXNnOdwr4AjV4gx7KEvYITkSrhw5JAOTCDLiLc1riI4kU0uacFgp/c31D4ZaIujJ09Ul0TaAhqLBAHmIijk2Vz1msIJa/8ENH/QeAuHsQ+6JCnVzWB5f3+NtHE/CFy7f+mIwweDkxPolT7fueeUZmqAxnAmaQSKj2O4AI5w8r8HSeDOm2/X5f62pBh5oDHSqpyBqDKZQvo2H2mcffpTzaH9n3yg72lrfsR9uFBgLC7gRHeOtWACad0m5mILEtoCAyydXb3urBNxI3L37i7jXpko/u231c5fPFZsuRhpnJ9DNtxvwbUL60zs9JwhKM0wO3TxtsbJGFK0FN9FyrlfVpRjSLUirMUeqMWlLnGNPa0Brb76MWqpXg1KlEtSiOE+iTVtgfKXh1oz/vLDz+ryd1vv6J7Vc5RrlDZ6nueCavyN4DpRrMimrn4urs90wfwYcqqfPlPsud3drstdyCY6JeY6JDm9Um07PdYYzioXzqCleFdkipXjkVxd4k24T8KSsIA+IQGHPYETCZTwLkA8OoSE4CcPRfGnU9wG1sKLmep2sS+p9dpRNQcdjDxHyrpiRUDXMNQ7yShrh9wBa2SAtTaHE2sGNxOODR9S7xjy+7Bai+gq58F5bphqLHiQvMWIf3TeW1nDILPxkOtyvUlTHhcAB9UF3Vo0J3b6Wlf6laNe4K/x9+BSRx4U7AdIgDzIIBIEwjPkmXuQeFJKtLennGQ2pzufnvHw8KguFw6a7yacW1sqz6u8k04zKeRoM8hARpB4LyCV29yWO90G9jkdyJxTOG6dQ6pKR6Ix1s4FpFWLFWkVYkb5VcmpcD+0i4eEMigo4Vcs9Dck1c2HWjsbkPkGyCOguHGn54q1hdMY/s73/nZd7t5XrlPi+SfDZS49QB7jjiVXLZ98WZef/+yTujx+pvGyGTPWaoz04Kw8PVH8KzHGp4gj57nGXWOoHL/x5AvV4VBx91KCOYak+3UR+M5ttJZzexMyJLrkNRGTOl19PoBLdI4c5udjjc/RXPO109P4L31dTzkhwqLrY/+/s0nnzWVMigL1A99RjOAePEOeNUN+eDFVvRz2xj7KCR0PV0wedc0MTmKex2Mk4K4MyXH0a7g2GhPHYDAYDAaDwWAwGAwGg+EawF7iGAwGg8FgMBgMBoPBYDBcA7ySnKp0nisq2mYIOqMPCl6OWx7NRJvbo4QCTlIJZBhNOI+89bbkOnQhaeJE8gg03wz3OR3CnWO2/K4pnDc8UPu6XdHD447ofORtUVqVgnI1BXUzB0XLgSI4W/C0f12SQ4/mF6Trgupc8YHXefK/H3iu21u286UDgHPOzSC36IEuG3XUPlt3HtTl/p5cnbyBnDdcV85Pt26qD1PQ5GegmeeTZ3U50J+6cA/0uPtL55dRogs6Y1FJhz9TeTT6VHWEC1NnR9TLUaSxcvBU9Mg//cv36/L+oSRfG13Kj9ROURfSBnR/s6nv6qL91onSOZdWY2j/QC4ucQ+OJKAhHp2KWv/0TDTDrQGo0R09pw9KeK+ra2756oM2pErpWO3SBW1wAxRZr3Jz4AHsIRwWLi4gmxqrvv2W5spgm1Ictf8Yp70Pp4pNj451/2dwZbp5W5RqsERXXJ8yOKiEl64gazz5v/ScSyvJGx21UugAQsRWSuUCunZAgsk+j5qgvLbVD4NN9SdUgK4FJwUfzntRX3OxvbVsN9KxS7gkpaAms3z7liSXN3bVh3TEmcABZQjnwaNzjYUko7YDEokWJBKIpxGkEAmkS+vEfDZzv/jgZ8455+KG5IH7h6r3s33Rnhv46eToUH263QKFGJOkKNS+zb7+uEs6OSQF/g4ovzHklXCI6Gwu57EPF7qMa9gKbR1jakNzzm+qLgX6bg6afwFqcY6YUmCejacqv/+BqOJHcDRbcY2rXDtKutK8JhZp6p4dLNci0qqLFckW6OQrSzKlVYi5E93nz98XJf/NnqQCv4X4e+u2pHidPTo1qp3jLcT3XrW2dOEyCVmbQ987yIY8SDtmR1o7FtDhNbEWRNCwPnyi6/8vSCQ/hlxsd+NmXaZrDozw3Bjuk2tH1TmkqbtA9WhAWh3BKW4ECv0ZZIibyIdiSK4izJ0cEpF4ADnxBqSoDUqo9HmtOMFvqhFy4a1AfUpHlhRuSpRwpdnV+SIlxBnPFMD1C8TXo6HiF6fBHciRtjeWbRME6/s9uCgKN60cfpIEEiAYmtFtdIx1eh/umZQh5cfI525AEo88/OaG5laEfqYMNG4pn2vgeIHL/uxBzkw5lRtj3CCfieCC0x/oftHKcRV61tlMc3SO507RQWeQcEwhoRqN9fnK+u19Vb/le/WYLpmvYO/IoyQ8jPNo605dnu5LVutKHLOxpX1hG+v/733zW3X5+UfaXwx/qPw+hBNq84H2Mo3OUuoaeFpbZweSc82fak6cfa7PB5DCbvgqUxb+HPvSz/floEUn2DPsEc8T9VebDklfPqzLBxca1//z//4PnHPOPXumZ35dtELPfXtvGTdyyDEncDWMPY2xO109y72urk/hzDcstRC8dwpZGfajhaMUC458SJ5OU/XF+Eht1anksnSvpJQ0hqtc3NOcKyAhn+O8gCKFPHaq+Ud5dYZrJiOtI8xzWvjeFBK6sKGx1uA7iJeEMXEMBoPBYDAYDAaDwWAwGK4B7CWOwWAwGAwGg8FgMBgMBsM1wKvJqTznkuq1Tw6JUQhusYf3QmegGO0nopVFDV6PU/0LUWz7HVFYWx1Rq3yw6ROcLD3ORDc7WcAFoTp9fgouZgSaVZmBclyAcgVKXAyaVwRXqQI0xCmo/RlP1CdrFW2TgGbFk61RG5dV91/nefGe51yjkm6U6JNeJCoj5Qh0oWl1RDcNm5IN+JH6qoTcrQBluLMDenahz+egCc5OJDk4fvjzutzsLOszgzPNZK6K5WeSakUL0S23d+GwA+nTeaJW/ov3dML8Tz+UmxVdEcpS3+WB/5mD5lrCHSKGs4DvsUfXCN93QbykC4dwtWhEatspBtYpJC8fJ6IzbnVF3wt8SNjoIAApTtDRd/V29PnYqd3Ph6KKFpBlbHSW9/chSZrAseTsRDTQEhKS7oa+px1DngXbHLqfPT1V+cMncg1YQOa429f4bYFmXqK+VDEuKrp6sUY5VV4U7mK2HP/bOfoQ86+BuYKwtTKuIlBMGX8z0D0j0EkHEWQ4oPvOnqnfhs9Ec21tSkKXbVSyiYQn8EOqCvnoG4gL/h5c2kJdTye/c1DFj880LuisVsBxzcHtxWfkBP2dbmIrDoVrRLpYuGePHznnnEsWasMZKMTzueZHG7Ho4lQylCO6LUKSmcFhowDPvgPqctjUOIkwX7KQjiTqm2k1vwJI8bwGJNKQ0IT4Hh/0f1KIc4yjbArdDB3BPNrcqbj/RPLaRw8l16EzlI+FP6/kVG6NcqqiLNy0yhVyn/IooLz6c/4SRvp0gvzg432N5z/9SJLf3g1JPno7kgcETm14cq71LdlXm7eby35uDjSebty8XZc3b0uSkCG3mc5VFz9Un7TgfMYnHMEp6wd/8eO6/E9/9BPVBS6ENzdVhyfHWtNnkP+cTb8apzhXurrqQXB1bke3Kecjj0CnHhyrzXcgc2mFmpceHUkxFktIYwO4QDWxdsaQ+lzKAik58SktgCQth2wxxRECHJAFnpVxn1LhOeZ9jnk8xDp6CpebTbhIDnAEwUZF/w/XKacqS5dUUqEUsibmohnqP4I07wyS2RHiUDJTGx4gtg066s8An3fw+TygbEmYwtWyrFygppBQUc0XtjW3mp5idUxJMPp/gftMpnq+MaRSGeLREDIPSstHkLaej9U2K2nM+k5ruALLcZHj2XLMjwhuTB6kbc0NxZMhHF4XqZ4hxj4ygCNUsKN85eYd3XP8UNKj8Y8krbr4REcqXMr3YuyNTr/4qC4ffqryHO5/HqTFOZSSp2jzDNr1J6eKL16osXGKOU1n1WyouTjKJBk8g7va2dkyl8g5UV4TUZC73e7yOYel8oqTOZ2cND7LTIN+O9Ycuo2l5Rsd1e/H+2rDkxPEv5b63MM860J+34QjdT7XOD+t5mKA2NeC9HwbUuEUxzmcYX85g1RxhvbkPGtA8prB4W+BPtnbkbQ4bChuNiHF3Otq3xU0uQa/HIyJYzAYDAaDwWAwGAwGg8FwDWAvcQwGg8FgMBgMBoPBYDAYrgG88hXkAZ7nHTnnHn111TG8APfLstz91Zf9algf/kuF9eP1h/XhbwasH68/rA9/M2D9eP1hffibAevH6w/rw98MvFQ/vtJLHIPBYDAYDAaDwWAwGAwGw78cmJzKYDAYDAaDwWAwGAwGg+EawF7iGAwGg8FgMBgMBoPBYDBcA9hLHIPBYDAYDAaDwWAwGAyGawB7iWMwGAwGg8FgMBgMBoPBcA1gL3EMBoPBYDAYDAaDwWAwGK4B7CWOwWAwGAwGg8FgMBgMBsM1gL3EMRgMBoPBYDAYDAaDwWC4BrCXOAaDwWAwGAwGg8FgMBgM1wD2EsdgMBgMBoPBYDAYDAaD4RrAXuIYDAaDwWAwGAwGg8FgMFwD2Escg8FgMBgMBoPBYDAYDIZrAHuJYzAYDAaDwWAwGAwGg8FwDWAvcQwGg8FgMBgMBoPBYDAYrgHsJY7BYDAYDAaDwWAwGAwGwzWAvcQxGAwGg8FgMBgMBoPBYLgGsJc4BoPBYDAYDAaDwWAwGAzXAPYSx2AwGAwGg8FgMBgMBoPhGsBe4hgMBoPBYDAYDAaDwWAwXAPYSxyDwWAwGAwGg8FgMBgMhmsAe4ljMBgMBoPBYDAYDAaDwXANYC9xDAaDwWAwGAwGg8FgMBiuAewljsFgMBgMBoPBYDAYDAbDNUD4SheHfhlHyz/x/KD+vCxLlN3Vn+M+WZnX5aL06rLvVPY8/AHu4/u45gX19D29myqrb86Kglfg+1kzXROHuqYZ6lkDfH+aZXU5x30K3D/F92Y52+NFtb8CWe7KoniFP3gx+t12ubu9cVmJGh4aPCvUP+zDAs+SJIu6HIVRXY4bKrNl+bhhpGsajVifo53zLK3Li2TmnHMuTfWdWZajjL5F3weBvscLNNQLVGal+zEWy5dobc97mf7U5wdPnxyXZbn7q+/8q7GxsVneun3HOedcUaitsvmsLqfTcV2eTid1OUnVXqw3hrxrROoLtmmeYzxnnMcYJ2jTDP+4LJerI+NKrEz/F/6Pe+FVvy44D64q53nuijXNxW4rLrd6beecc0FwdTzNcrUxYyLnnO9xPOtvPbRVlmdXXsPvTTmn8L1pqvF1WYlm3Kw/ylfiQqJ7+xxbGkMcWxHiRYD57+N6TsYC4ynPr67vAvXl9QFiwLOT87XNxX6/X+7d2HPOOTe8uFA9FqoH160iv/oZilLt+KJxGKLfGQ+9lTUPWJmXaItq/c6whhEcF6wjsbp2Yh1HbpBjLQnwuYf2yFEHD/1eriYTdfFynUiSucvSxVrmYrPZKjvd3i/VzXvBMxZo5QDrQOixzupP31OZfej7SMG4BmMusg1X429WfQ2+B/3WwDrLecAnYj9zPWNM8YOr+4TllbmYXr0ucE7z+pPz0drmYhgEZVQ99+oa/oK8kWsVYmS+End/dc7J2My5yC9g/sS2uIzTq3EZOWpxdZuzXlxT+Tmfb3W+4pKVGFFceQ3nLuP05d9mWe7yPF/LXBx0W+XNreVc9F/wM3NRXF3PMESe94KUgPlngLnITuScYqdzDSnRVov53DnnXIb1b3VOaGylWBfY3gEeNgyCKz/3AqwFWDujuFGXGQM8p3KBPDkZKh9Mp9O6/MUkWdtcjKKojJvxL33+orixmrvomaOGnq3T6dRl9jVz1PlcfVAgXrZaylkWzFNWcqBlP7Xabf0/xvt8Nrvy7164Ri+0Rgeob4YcZTJRnp4kc9zTCSVzdsH3fnmCpGnqsixbT47abpVbg/4vfe79C/51CcYk7pXHE4095h+Micwd3UrMc1eWV+LcFfsM3jvGeGo0VeZ85v6TX7MyF1GBKOIeWPdkJdkGfG6u6czZHx++XI76Si9x4ih033qwTFYb/UH9+QwDMp0jyVigE0Ms2skQf6uJEPututxgQlFoInRaaiCvuLpz201N1rTqjNOZJkfh6//nSJY8p2D21raueXd3Q9/fUJMdnp3W5fO56jgP9Bz7I31+PEE7FUy6OUx+eUNWHp26dWF3e8P91//Ff+iccy7HZr4Rql3Pxud1eZ7juWYqf/Hpo7p8a/dWXb73xu26XHjYOOINwfaern/w5tfq8tZGry6Pzw/0XZ/81Dnn3MH+k/qzo6Mz1fdcgbXwFHz726pLs79Tl2do+wX2J16mvk2xWOY+kitd7nyM0YKbNCxAXqm/+K/+8/9UjfaauHX7jvsf/6e/75xzbjo+qj8//uQndfnJe/9PXf7JX/1VXf5sX4tG5hR8NrsaD2/sbtblMNJcOL9QWx+daJxMkLhPcz3/8Vhj5my6LKe+PiswcQMEYh8BtfDUhpmnPvK4kSo1t140n14EBmMGeyYJjSown51p3L0utnpt95/9nX/VOedcr6exnyLhO7tQrAwwh27sajy3sWjkeLnqB/r84lz15ua/izj+/FT9eXSmlxFPDzQXLxPgb3ztXd0bScjnX3xel/stJXC7HY2hOze36vKtO3t1eWNHY66DJKrAfJ1MVPeLob739EIx8smz53V5NNX1G5s36vJ/+d//0drm4t6NPff3/t5/45xz7h/9H/9I9XiieMWXXrOp1qIh+neG5JOJYByrvLWjfn/2dL8uN17wUo2J8QwJaL/q95Nj9Tl/XBlsaFxcnGss8AeQJOHLPdWx1dVYZh/1+t263MQ6fnqivmtx7cZYLrEObW8t1+MPfvYjty50uj33h//O33HOORfFeBGCjQ9f/i98zdFeoPJ2Ey+t0lFdjj3Uf1fjsN3TmC99teH5mf72Yqgxcj7U58dVu11uIJ1zrtdRG9++rXV2jr5KM42JwZb6OQr1eTPWs/a6SuIzJLdJovE0HCkxf36g/pzNdX0X/X+Bl53/wx/9k7XNxSiK3FsP7jvnnMv5Qs3jphirOPr35OS4Lg/HamcPL1Qi7JlCrl3YTMWx4h5fNkxmyi/bLeS61VzPkYx0uop/M/4wg00vX0ZwMxxh08kXQMx1uetJsNFcpNgA4297PfXd9ua26lPl/k/2tUa8Lm5u9dx/93f/XeecczHybc9T/Wf4UYqb2u0drSfzBV7E4GUp88/+nvZIIfptNtb9Hfp/sK29QI7x//jDj51zzh0+1Po3whp6dKCx9eyJ4vYcfdtra0zsbmpe9joaCw2so7tvKo7cevNOXW731T+Rr7k7P9T4++T//ud1+fl779flf/+ffby2uRg3Y/d7v/+7v/T5AuONcYA/IvmB+uLuG/fq8ne/9/26vIu10I/URj/74NO6PJvomb/9TeUsjx89rMv9vvp0//myn37vu79ffzbY1Nj/4Gcf1uXNLbUtf4De2b5Zl58+fVyXt7c1Ng+ePq3LP/rhn9flzz/7qC43Ir7s0rzkC8YY6/4lvvjs4S999utia9B3f/c/+PeqfyFP9q5+gcV0O8XMfHp4WJf//Ic/rMvHWPu7iIm9puarj9w+whuLkC+6+YKy2qcu8FK+01Y7vfmWxtODt1VepBorp8eqb4y9XQ95aRyrMjd31bdvPXiguqPCB8cndfk5yuMRXgwiv/qP/tv/7aXm4iu9xCl957L2srHGFwpy46kaK2jg7RR2ZW28nMr5QzY2a9NcjVjiYbpoCP7KN1vgrTcW7Axvui+DferUUN2GgsVeRwthD4Pid7+uQH/3pjrow08/q8sJNvqbfS0Gzy+QuOLlSI4F6WKCl0oFXwCUv1TOXvCLwq+D2Txx7//8E+ecc4dYWJg8ewhIx9j8jfAGlQnH+VDP8smXCk5vvXm3Lv/2t9+py+880Oc39xRASzBKFliw+9Xi9hw/9Hgl34Kqvqcj1ff5J3q5sYMXTTdv6fsDNG5SgEVUoN/woqNAsuZjfPOFuLfy89HVvxi9Lsoyd+liOc7Gx0qgHv/i47r86POHdXmOl5jdhtqr1dYCtbOlhbPTwq8JCV50YjPVWZnreuYOwkoDc6RIl3PwLOMYR8KJ5/PwL99hY1Re/dMc32C/EsvNvdwvQ/+iz14Hl+tfgo1YHGsx62BhS8i84Bjjr0CYC6Gv+XRrT0lhjuT82XMkl0g4zkfYcOAXv+3tZYK4g01+mSkODzC2hiMlaKd44f0cG6SvIY68jZd/5bbqGCG5dmA0dPtaUFs9lds9xZQvHyth5mZlnTg/O3P/4I/+F+fc/4cphBebw/kYf6HnnOFFBTeag4HaN8XG6uxU/TXHJoIvHyOwG/mLT4wXQ2WV4DQbjG2atwtsLmL8ytuIdO/Ax4bP48ZRz93EL11N3GeKtYSMkRLtR1aXjxdMlyzQ8kU/tf8aKMvSFVU7Z2BwBiV+eUfuEfl44RUjVwnBvkIcanX0ssbhPk/3NT4Pj5TYHWFezseKDX7JXxSXfdFEMhkh9PHXZjKn23hBMMJLoQZe4rS2Nf6SGfoB7z8Ys0rEzS42nZMJklW8jGIyvFZ4nivrsfiC8YGcM13oGZKZxmSEdYtLToBxzvkUgqFBCkgJ1l2nqVje4kuc6m+9Jn5cwpwnU40MqSbmjcdff7FGLZAjB5hDbJmQP0CBXbAo+Mvx1S8z/aoNVjZyawTXY/bVAj908AeQOeIpGce9gfIcH3FuOlWcC/G8XbxQ4Uux0bHm5dPHekk/rjajJYiBh0/w49qBym28oGhzY4Q8ZzEDW6dx9Uv58wPNrRwvQHZu6jm2dvSsAfY6N95WDnz2cG3vbV4K6QpTFiwEBJebt5Sv/xt/+G/W5c0tvaB6ipdh97Gn4MvrH/+1frz84z/5x3X56LlyZo9rV7U2f1T9cOycc+2O+msxxxjp8QWK+v2dd76JZ1J/8aXT1pZ+yOr19TJolQGLfM8xNpCBC0VExLG0PlxuT1cyb+9Xr71T5Hmff6oXa8Nz/XBUZsh/QAoeNLVe3sYPlnvbmust5BNkes+r9eroVHvB8UR1yYaaN9Mj3eMNjKGdjub/+alyV8Z5Mm64/+CPLlu7Gq83buil64pCB3liyh/GXhJ2Jo7BYDAYDAaDwWAwGAwGwzWAvcQxGAwGg8FgMBgMBoPBYLgGeCU5VZaX7vlkSQNKhjN8DokJ3gttQm5xE/TEd26IqnuIMws+2hf9qYSercEDESG52cCZCw70JB402Gos6xPFOhBrCzT8m6BZ7kA3F4NDe34mmu1wqu85PVdd4ljlUQJeGChXvb7aYIizHcqVc3BAnaylA+ujjQ9HU/dP/vivnXPOJaDMt3AAmQdt/sVINLQk5ZkNuufmhq75/velWf2Dv/k36/LeptrcAx3wdP+Z7r9Qu/GAr8l02Rd+LJlE3AZ9bXa1LKwoJAmZ4Iyd1h3R887PdH5G0dAYKZzGRVao7rkPOUEDh7WCih5SuBl8Ne9JyzxzyXCp23z8wXv155/+WPTR4ZF0ndTRDkDh3eipPMB8iiCLSUAP70M60PFJ4QbdGvrzGejke72l5PAnmOdHUxz+9wJJlFfy3CqefcMTiq7W5b4OrpJZrVNOVZZlTS0mHTPCGRuUZLShf454gClope2m6KZRRIqp5tMxzi+YQirRBN0/mmKug26/XZ1rcRfnhG23EZ8TSag+e6q+2j/luR6KO8+fi9ra9ihX0Xfu3dL863T0rEPo3bNU/UI52ht3dVbAEDTXdSJLU3dYyWL6A9GAebDxgppYnv1EmQtkYzx7owQle4Z+bOGiFmMRPo94ECTNCKo53eG5Gpg3PCh65WBzPAeljZy7C0gvfMyX6UjtX0AuGUFynOBcIJ4zUiAfuDwYdJ1z0fN0mCLl2KunAEOeEaq8BVZ924coFM+VlWrnJ08Vl3kOC+VJXcgib97QerXZ0fzutpZ9u0C+Q4n5DHK7KQ//D3GGTk9U/mSqdXyI8zy6kKtQzshDNReUXOHzzQ3d//mB1lrmd2tF6Vx5OS5RV8q9SrRXDikol2pKaPyVQ+evPsuQh5+uuCXwcHfI3hqY65eyuBu7kuRfQHJwDAkPz9vhYZocpzz7irJz5sUZDtMsIX9cPShVf0uZ0gjnxXTq8bhmOVVVvQRyC56lwnagYcgk0ZoQt5jPqX5TzBGuMy3cs4Q86RTn4I1HksXGkFqm+bL/H32h4wRinBP2B7/3nbrcRq7NuDaFnG+O85MiHMNQNnHILmLl+VNJPtKJ7jk81ZzevvtGXW7e11gbQEbi3A/dWlE1Lw8Y5rrIZYaGKnxZ/P4AACAASURBVJOJxvA51u12V3vHL7/U3uFHf6WzIA+PJF2bTvW3l8cPOOecV1KiiPatcpC80GfJnBJplc9Orz781kMu/M43dObnbKKY+sGHkhd98fCh6sKDuSGjbED2TIn0jHG32retW/Ivuf6LDlJHXMG8+fLLL+vyPqT6JfaRXWwhbm1pvn7jno4w+a23lMPtbugaHtWwEq+rPpxhP3l4rDExx95yiljpxhorb9zRd27i3QUl+cxhOI8pJ5/i0PDehnLmO3d0Xitl4+MhZfcvB2PiGAwGg8FgMBgMBoPBYDBcA9hLHIPBYDAYDAaDwWAwGAyGa4BXk1MVzh1X7LwSlC6eVB2k4kdthbAMA93pVkfl7Z7ocX4ietLhENangWhr/ZZkLnf6OuG7CQZZG7TVtLInL3H0dQuypvMLfc/zc9Iydf3xRBTJDPTUyVi0qbMUJ+eDThfgxPsFKIUpKb042b8gjf7ymrUyVX2Xl0tKmo8+mcGBiCd99zs6XTuMdE2vr37+/vdFFf1bf/DdukyJ0XQkuuc+HK9IlTyFq9fxmfrlsy++qK4FdZAyE0jW+j3R7fyVU+dV9zlodhdw5xjPRffe231Qlxdz0VwT2NNHoOv6oMhSfuIFa6Y2VkjnM/f0o58555x78rEsIs/3NVZTyvow0/vgMHZaaqMmLKxXbLdhedLZkLQlpMU3ygtQKgNQlJuD5XzduKX2/JOffFKXn4/hyLHyfhnUb3c11s0gXd4TLlqQo60LRVG4UUXP3hzAuhTOGymolp2uxlsT7boBJ6McFsDPjjTnHldzyDnnTp5rnNPSsNNVbG3d0Lz3abFaOTVswuHm7fuinr7xQDTRbzyTe8QvPtb3H0FWUUDycXYm2UAAWUoOCe32nqQlPdS3WWgcn59D/ppAoluoLdeJ0mlcUu5EuUUGuu0CFN427KzbEaxDnWi7vS5iC+xOPUgf2i3Y0FLCizbKMK4uab5FDqc6XDsHbXgICcH5ueJl3oJcDw4LU9D8fUiUSV2egdKcYQw4SCMacP9JkG9MKhldvuY5WVYxrwEHCp+OiHCk2kbc7Dcgv0KVJqDhr1D8QdvuQRLz9l1JHHZAvd6A29MA8rhLiSxzFa6nR1hnC/TbEBTvBdaqwYb6//QUUlw8FKXCAdaLADKAOe5PFyZa8j5HDFonSidJOuP3SizHKsIxRFkqF5oU8pvSp6QAcqO55ksQUGYDmj2k2CWkXknlutTEvKWDD+WGEeYEafuUMST4W0ohGyHdVFCvlWe92h0lf4EcJgqTqi7rlRlfSkYStv3KIk+HP0hjKDGDrDujNBouV11ITOdwWR2yDSFtbLUkLcxmylEfPVw6Ve3CtfGte7Iv7mB++JDy0IVslqh/ZnPFzRX5DOJLCRlQCjfJ00Ot+4dYa8eYl7feld327Xcl+VkvvNplOMUzzDF+aCtO67sx3Au/+OJhXZ5inj19JnewR7AMT1dcPCHF9zi26YqFjy9jB+YWY4QHCb+HPQWH5sG+6jXCvoc22OeQS2aQDK7mmYovOeWwGPt0NppN59U91psMX+6jPEqoyqtz8tMLrTkPHynnm8AdqofjHN66q3zuW7D7/jacje/uaE7F2E+FK/uVX37mEC5wt7e1tjGmjKea80dwoSpmqi+dWOe57omlwI0nypFy5JmF0zyeTHVNu6s4cuOGpGOL+avnqMbEMRgMBoPBYDAYDAaDwWC4BrCXOAaDwWAwGAwGg8FgMBgM1wCvJKcqnXNZRSHzQHfnKdEhXWpAcTqEE1J2JMrvdlduHg+2RBtuwkFhDCpeBFeIdKbTvknDjjzdp1XR/0l3PcJJ1Ydnqst8gRPvQcU6w8n8K++9clE3fZxmnsNNp1iQBo77QDdFZxpvhU13SQt2a0Mzbrl33/ntZT0z0NghWWhBBheDndpoiOp1557kFm++fb8unx5LQnECV4URZEsZqIGHp+rD56CB7oMGeknlDWkfAdrhCOXzY9EUWzhVfGcHjlQneu7JGCf5n2ss+Omjutze0LPOp7p/UYoG1wzUUCXcYegOsU4k04n74r2/dM45d3EgCRVpjhHo282eaH2dASRhIemhmEMw2whBUW6Fumcbf1uA1kmaeX9LbdeqnHve/JZkA/3tG3X5H/7Zj+ryARwhVlyoPDqFcGKw7L3g86uvIF401S4pqus9+b/U/egMBEkUT72nVPAGZEVzUDk//vCDuvzokRwBtuEw8/tff7Mu72JedCDLCtqin8ZdySDalZtNrycpawzJxA1wTN9+RzH8u99WzN1/rLn14ceq7yeff16Xjw4VO7p9PXero+/KQRtu97SO3NhR3bpwszo8lCvXOlGWzqXV+pOBqsulwkdg77S0Vmy0NbfuwPHrzTckS9vAHNpEOW5oHrcgpwopB4Is8hiSnmEVjwNf7bmxhTEFOcHJiebiGWJkkuqZJlhfKdcpMF/OIF0+xNpwCiekBWL5AnUomFdcWpqscy56nvMruUkDwS/CutiBGeatAdzhSvX5EIz9L/e1no3hFHh3V+Pz229rLt7Y1BhuQ6rbgYTKR6516WDThISvQ9fIpv4xaKjyX0JmeT5Uuezp+hbmDa+hK1AIeQ7dlvxQdUwzOmFpjI7HGnfrRbkihbiE94J/eHCeiiB5CTDeupHq3W1pvvYhb/WxRiWQi9AJKed4vWKJSsbKkQrks+kUcm5I8ShnY67dQF+XyD+iQuUm1vQmpWaQTmYvkivSLbL6Xu+FK+eroyyLWlZA5xZK9pIFZThqhxB52IpMDLHEowwNTqzHY8WwRhsucE31fzlXbDt4KPed29vLuLx362b9WYsSKqzpORy0XInPi6ulcisSG0hiKbGZznWfsyni6UTjZca23FIMeoAcbL0o67yJcipK/7g+hNA19QfKOQLk/R/8/Gd1+csvH9ZlSqjofFfiOAv/auO1FZehy/WyLBlDqGFb0V6hpPIcLoMp3O8+/0L5jYOzaxfS9DFiLdehAjZenrv6QS6/lw7HrwvP6bQUznFuaxao2+NncgyjtCqCs+Nbd7Vv+u5vPajL70KWf3dX/T9oK541IOcNI8hDuWG5/H+83tikex+k3IwvW1vKfw/wjqDMdf3GAG6f2P904PI7Rv9D0ekmM8VWSss3NpR37cKh8GVhTByDwWAwGAwGg8FgMBgMhmsAe4ljMBgMBoPBYDAYDAaDwXAN8EpyKs85d8nmbUSiw26DEl6AbjgFbenLoajfo0jcpjgQDe6tW6Iz3dy6VZd5gPnRqeiBx6eiJ41BYR3NRUvdzJa0rACUuGwhemob9P8I9NQM9NQm6PwwpHLnIzgCQHIVQU5Twt2iKMihpcsBTv2mK1DVPdka7akaUeTeuLWkrZWggfugb0Y49T5w6s/BQPXcuy065uefiCb40WeSUz17KvkCT0xvNjUWSGWkdKQbS87RqGiWrYg86BdQk6l2g6PCBK4dT2fqq0unE+ecm6MPT1LR6ZJU10RdUd8iR1ox6Hyg6PpfjZrKlXnm0vMT55xz2UzjOaczGul+oORznPusoI8+iumIoUu6TT1bA/RXD+5A7Y5kIe0NjZONiiroNVWXf+tfEYWyDbe5//NP/qwuP97XmJpmpLCis72rnQJe1Pw8zf7FqoyrnEjW16Ge59dOQpz3HIetFk7Yvy2qdhOdMkL/JyNRqe+AHvqd3/3dunz/vlwAuuifEO0fQC7lwxHQqyQ8HmnrlNLB1ciHC0gfLoR7N0XfvvsW5CTvy2Xtr374w7r86POHuk8fjlSQEyWQlA02JEsJIUGLYzkerBely4pfllORAp2hLTbhMPVbX39Ql7/xQHPh3m1RjvduqN69nmjGcYw4SsY3xjPrMGjBKWG0HDM+3PYCyDBSxM49UO8RomsJmXPOzUHzHyKmjiEFoXPSoyeS+n2C/j3G+p5BihNCmpCm66OLC17Nt6frVxMOi7tdlfsBZDKoz6PHilWPH6n87i31+d/49lt1+e6O6NO9pq7xI9C2Y9Ht6XaUVvOuGS5wLejmkHsXoOznGBMlHKySseRubdDGZ5AcQ7WwQkVnKG4iZtGFpQH6+wByiXXCc875l+sy+pFzgq6VMWRFASSPWy3V9e6OJIz3drS23b6heLwxUMw5uVAMPr2A/BDtTqlgUNWhQN41GYtuv8B64CPniOBsN0Xe/XRf0r0zuJK1IB1rwoluhgQ7XYlfkO74V+cM3lfgoFoUpZtOl8+fIS+N6fqFOjeQk3DuepTM4Fnagfp2MlL/0D3MzxAL4Zp6+lCOO104hd25sxwLIXK/kHou5P6Fr7p4yFt8n4kI3JFy9UmG9bXEujuGw994jJwWdR8nGnPdzyEF+5bi+3rhOb8aNwHG3soeh52EfRBlMw8//6wunw81t5JEz+awXlFYw9tTZUTJP+Vql0dhUOblrZxx8YIyvojHY6SQ/X2BsRMjppeMx3TFglyOjsFcC318fuli9lU4tTrnnIOszcceYnymNf7xgfZ8KZzXbmA9+fodrXkPEE/3eppzfUjOW4hzfgSpKOafh7XFr8ZawLnI9isllYzhBtVGbhU09LdPj7CPxTM1O8jF4MJZoG3o5tjFURTPD7S/9Eo9Xxdx+WVhTByDwWAwGAwGg8FgMBgMhmsAe4ljMBgMBoPBYDAYDAaDwXAN8GpyqtK5qGJ7RaBU9nAy9HZfFKPTM1H8ZqBb02SoEcARJ9f17cbVp8z7bf1x5EShOpuA5w1q22ZneX23rWsd6JRBQ7Kw/qYcQW7ilPnBtihf7/1Czip//x/+aV0+mYAGVzRQ1tfCqMrlpHqunH6Ox/gKnI1833f95rJ+Be5PqmGABo9CVXpjm44VojX+4K9/WpePQK0bQELRAfU/jkit1PfyVHm+XbwcIkHBk/x1beHxtHvI5kCDLHFK/Mrp8aCE854pvssDzfXutqQNrY4okQWphqDQ0fFinfB9z/Uqp4xhSPcmFZubojAGDVDIcVEASUQT9iuUBLZAVw5D9gycOnqQBWxp7nS3JQsJKwedBiiREU77/8M/+H5d3sT3//EP5Fr13qcP6/IogQuBI66eN557CRkGuKil99W+4/Y8r5ZTZXBvKCH1unvvjbq8san5NB9r/uWg0t/ANW+/9fW6/NY736jL3R31iQ/ZlGsrznlNUU5d+Mt09VXJmije3sp/oL0xDQLQ4vdixd/vIRY3G7rmj//pH9flD97/sC7/9u98uy5v7sAdZi6adbMLl5Ffg6r6MijL0mWVPILzPcvh+Ad5xmZP9fjaA0nbfu933q3LbdgCcr40KH8EDbvIKO1FnMQCRPePZuUmRIlvnqEfMYcoY/AQa7i2hYjpPsYv3Xx2tzS+GLLO4Xg1GsM5AjkDy7VGda3LYwlnP1LX9ZCDlp6LTlVPTyQZ+vijj+ryVkPx9zvf0Fy8hXboQXoUQ87mYS74WDtL6JmixbJfihzOmAX6kApfzL8mxtYANPAh5o2PNbKDuTiear2EmsNdYO3e2ZXkOICb1hjuS634q3GnKh2UKysuLYjljPGgxw+auuZ37mkN++Y9yRzvbCq/3dnWc27B2S3JdP0IzmETSP7pFOT5y44qkDc0AsYLxfQIuQjdRYeQ0X7yqWQbjyFpzyHhojzqHOtoiPs30O9z5qJcD+qJvL7JWBalW1wmyytyG+RYXnBlmaBUhxIwygAz7ktCyN2QL14cyDGohzrc+9oD1bmx/Nyn3JsutZDJMK8oV+pOydrV47WAy5XDeCrQt3RHm3Frgdh69lTur6M3FL/WjaLKzCinilYkjHBggqxpH3LbDOucu0rK55zz2UbIabmOsd093Cd3WFuqOtBJjv246nKn+nor9dIlpXe1m1iGPgrZNtwcr8i1VFwd13Co+wp0VKXnubI6xqLRhDMh3Aiff6bjNM7gQtxALnRvF27GcGfcaWsP1cOev4V8gnGIsacItXaFWE8uHQf9Ffcq5DB4vmZT62UGV81NJDcJJIwXUz1fmeFdRFPf3+mqjjnmdDtWPjA61z55/5nchW/e1DEyLwtj4hgMBoPBYDAYDAaDwWAwXAPYSxyDwWAwGAwGg8FgMBgMhmuAV9J6BL7nNisnqgyODPlCNKSNLZxCvSfaPjlmPlyRbm+Jnvo2Tq2mewGlLedwIVmAfpdCjtAEtep25YTSBa0+ALW9uyenlFv3RXke9ODmAJrxrXuf1OUf/PXP63LyVJRwDzKvIVwGXEnZC112rpZT1ZqrdbLkysK5fFEVQfUDY7EAlazZAtUPdLr33vvrunw+Un/ubKkPNzqSDcQ8nb4EDRR9W/KEd8hsLruZtEdKwXjyvw8KNYxFeIkrA+oAILmC3m1O5w286zw/kptH4YnmHO9ovEag8TVAHV0nPM+5S7ZiDkpwBFcpr4F3tJAtxnDBoDyDVM4GqI0RygGuaTf1zJu7kiL2dyUB8uhsVLmSRR7cu0DD3+upjt/5ulyL+m24RkBa8qMPf1GXJ+wv9vXK3KGDAP/janc4t3LFV2MzdknzTkGH7g0Uq27fVbsyZnig+M4gbbyxLdrqG/cl/etCZhW01Sc5nMIcHBM8SGg4py7tibwcYwsSjsKHnCOkNA33WzF40D+6kBm99daDuny4/826/Bf//Ad1+Qd/IQer3//u79TlB29r7LCdFouvxrbB8zy53GDAkSjfwhza2VBfbG1IWtNBv0SQoVAqtQC1N8ScThED5pBw5DmfmY4cl/8PCSniPv/Mg5sKXbAWcN5IsP5muIh09pCxA3RsOq2RTZ5DgpAVlD1cXr++Oel7nosrqSia3rWaiqetWHO02dBFB3DkGMGN6W98Ry5UNwaikDcDOILBsaLkWgH5a0nXGtDzL/s/o9zNh9yN0rsQElpQ1dn2yQzXT+ZXXs8xnSEujM4hYYRcq4lxPx0hd4Nz6f8fYFhP4OTjFirvwIXq67clj3pjoP7a7Go8bKO82cGaj5jqQSKXo59yzotKUptCfkm9fZvyJch8piPJfIYN9cV2oHH3LpwAFzlkAYgRzy5E7acU3D/S/c+RGyXIveqYscblsXRlHZcCSDJyxJgQc4W5f1iqH5hX0w0shXxsJTubaiycHZ/U5Y6ve968p7bl9EqrHMIr6e5F+T9iGT/PmQszzupzylOZt6y4h0EKkuHzBSREMWLDZKy5ODxXP68Xpbtq80L5EPN4HjHBIw8oLw8c/5auTvpbn/sBn5rSq+VUbNPL5gpYrxW3RzwIJVeQ0RWQl5eUakHTSulYEChGBDiugPswylJf5MZ7KV0r17hh9H3PNSrJbRdHpQzHihkPH0sOtEAestdSPnlrS+59u3AM7UEq3Ao1z8IXOJj5dObCPt+PdB9Xja8SLsge1iE2Xw45X4h3Gm04Re5t6T0G9xNposHQ7KsP6QRM184Ie5cd5OkHcN998kSucS8LY+IYDAaDwWAwGAwGg8FgMFwD2Escg8FgMBgMBoPBYDAYDIZrgFeSU4VB4Lb7Ff270Puft/ZEg/rON0Xf/NY35R6ysyU6cRSIYtQDtXz7hqQD3Z7cASiFWZW58NRyPUrcEkW92Q6r/4ZsJtDf+T5ciAqceJ/ThUE0y9u39Rw335B06LMDnVrdAc1rPJPMqihUxxLSKh/USb5XKy6faZ1U1bJws8W4ui0ohRnf5+nZ211R4vafH9XlLx+qPOiqTbb6avsY9Dg+AlijzpUZrqFbFviR1QBY+WzVHkeX4h9kProXnOgeQ/rEE+mnkDCMQO1sTySn6hUaryHofAFcSRrNr0aGUxa5S6ZLGc0cNiFhA45Z+LyBflyApkkJFZ2qGqC8hmivZlP32d7W829tQUIVaU6TQh5WlEbKDUmLLj3Ng81t3ePbkPR1OnTtED3xBz+Va9HzC1GLFxwPoE56GB0BaLk+KNiUUBVfhVNVWbrs0h0F9N0bkFB14YbipqKwzmai7M6Gokbv7iomBTj538Nc5IQJ6OqAiVmC27wiyLl0roOcasWRir8LkAFMpzg2JSWUqOOgL0nZu5BHPXoouumfv/+zujzBfSj5e/vBA91zQxTWdSIMQ7dVuS9MJ4qdM7j8xZDfbA70nP02nN/QXgHo3iW44ilp5j7o5CtOg+hTUsHxj0vXKjp5lPnVMijS/1PExVmyuPJzjzEFLjsFNbuQQDQwfiKPEi2N9xz922wsYy0p8a+N0jm/GphRqefqgI0PxrRLIWs7QbyJQ0q5RSFvQx7lo31Sjn9Q5tkO3gvY8WUl86A73Kp8FO4tmHMR5MRN1GsB6RgdYTw6yKANFrn6p8A6fgoZzq1bcttoNnT/6Ux/u254lXQzgJShpFNehLEEKcONHeUuGx2t503ImVbM4TCG6fzkQpVXXHkg7fAgIwiqfJEupTmd5+hChO/kI1Eh3oy4XkI6negi39M4LSAXej7UWH58oXUlmejz4QzOWtX6Xq7TGacsXV7leh6el5Z2yVx1oMSGbpe0T5tNJDn2kesWc82dydFxXW5izL/xtfu6Z6CxkKDPg+p7faybGeItj4TIKD8ur57nPE7AYdxQHrTqeEUpHvKZkhI+rhe6fjrS3mXd8Kt+5HM6lHOUU44hxCWuG3zmgtJS7P98n/JSOisy8YBEizG4niLMf2itqTnEfYyPtZNSqRW5GPc62HfmGIOU80eIOzxxgzKrleeoEogXHQnw68DzfdesjtGIkas/P5C72emJpIc+nn3QUg63CSlW3FLsCSE3dQGlb3T7glsk8gk/VH7r4OboXcqWAu4tcG/0A/N9h6MAuC9pwUFysy+Z1REcAT3kaE0cS+B89o/GUW+gZ7qJNfLZc0mzXxbGxDEYDAaDwWAwGAwGg8FguAawlzgGg8FgMBgMBoPBYDAYDNcArySnysvcjSunpj4oRnt7okp94917dfn3v/edujzACc+NDqQtLdFK/Ybo7l4gKnJJOQeoVaT5eqANeiVpVPPqWjgSONFEXS7qb7EgPdGhDHoqHJq+fk/uL3/2AznlTOHW1UALJ6B3ZtCIrdClKUGoKGXr9lS5pI03VmjC/GJRwxqgQH/wYz1jCkrqZkf9H8J1JPBJB2SDqkjacJ6zj0Bzu/zTFckALuUNy6vLK4+HV5ch3mPGocZiA5T3JMFJ/nAC6k1EI+yHVzsnRKDXrxNlUbrFYkkpjpv6jhkkezlpnaT2gqJcQMISUPpEdyK4SfX6mse9LdEASWcs6UQGCmFwSW8kxRX97OWgMGKeuUR033duSrrX+9vfq8tfv6XY8c/e+6Auf/hIJ7+fwZ0oReijVKp0lDZSUnDpwrG+2Vi40i2y5X03NxTv7r8pB4wQJ9oXgaQ047HaJJlp3kR0+sHYKzEXKU1ZmS+QxHC+rvh4VX9bepRBYp4VlFmxzAmL8Qe6aQG3pQhyg+1N0VPfvC+p2V98+qwu//Bj9XMLMjK64+zm646kS3h+4OLKiY9K0XSm72tiHnTwbO021j+0hU+Xqxe5DOGe1Kh5mOtehvm1Eg/9qu5w84ODGIOkj/tRttVYyR5AD+eYKq6mwlPOGMe6Z7+vMdubaeyfQdoRVOuxt0YJR1kWLq1kxt2W6rYJh8UWHvjoVOvAcKT1vgUq9aAreQ4d/vwVuRnzGcShlb6ADBtrVxks25yyYc6ywL9aytMEJT2FzDxFDJ+nkquwlUMunhgXTaxBZ2ibfn+Ga9Q24ylysHXCc6KwMxXAWkh3vA7qvdmXhLcJmj/dJkP6c9GRJoHLZoQeibAGI9eJ0HZ+5YqyIklCrGJ9qaQoV2xzrnb2DP9f9t7sWY7sMPM7uVbWXnX3i62B3rh1c5HY0mikkUIRdtjhV/+LfrIj7AiHw46xHoZyeDQeUmSTbDbZDQKNpQHc/datvSpXP1Te/H4lou2GUHjAxPleeFhdtyrz7Fn4fufjFt/lPI1xSSwMyUZzJBjFI+FvCTDKtKzkNVTyNVUY1AX67QLrU260FkZYc0yqdluOdJ3xCP15jiS/ifpqCITxzvvCw9stzdFBgLHLwVYm1cSYqxPsLTP0Gwf7KY7RnImEQPIM00EDjSHP1T0FGNMh6sbF9XD+DdDX0gX3PJuUUx0JwTUhR10w4S4hHobxSszXxfpTD5h+i70j1sUACF4YvXx/6/M5pewaxPXmiepzgaMmFtivxC6fY14+jnlcQY65I0O/+xfRSSouuRclas3ym0H+r/GwIZIXv3rwoCrPppob6th/dpE22mroiIkA66gXYE7yWMYzApDXwkH63Boqh2f+EifNmQbGeuXDPbH1bzq3BM3ZwN4tmmvdX2Cu9BvCyHysHVxHuAfYO9SeNs64g/x2sk4cKysrKysrKysrKysrKysrq7dA9kccKysrKysrKysrKysrKysrq7dAr4RTFbkx89nKWtWJ9PtPAOSm0xcW4OKU/LWTp4Fw5I6sUo6RPYmJHA6TnNwm3g8rOm2+sKcV116otQQavBXWvow2f+ACLm3++J7v3BQ61ox0LYOpLMSdmqo4hV1wydgQ83LUqNigRVXf5JigvGc2/jLWNTuwco7GaocXR6d4TxtlJC34tI2yptk+sr7lwOCY2FWwvarqod3t5Tjaer+BJZP2zG/g15h24q6heuqv06mslecXqo+DDAgd+r0P++smVRihLexKc9hAA5zwnySyFtdQF0RhPNh5U7y/2daY7m7frMpuTRhdxpP0iQsAhbz+WpexGmgLulCJExSGFmb97R6Sqn76vq7roKf++NmDr1V+pPZ6+EJJFFOkdqTogynqpviTwuvLcVzjlbb9O3eFUPW3dvQmpFowDSNGuVZTPbSAEtU94pKY574hacsrXp58QpfuNXqWM9EBn7H2yWt2f36gikxBcmGhLWBbrTWQYLitutnZ0ev3T9Se//S52rzX1ft/Wpe9d5PKityMlqt5YTKXrdZBAkgNluA7SCPoIu2A1neXYwGvE1d1sY66WMdCJjEs1aYJ2jcIy5ZyOBfQV6dE5gAAIABJREFUmq2PWF+G0I/QpCnal8lNBSI8MpQ9MCJbPa3plxPdX2eiPjCfyLJ9PQdsEqdyTG68Er3m3ma7hfQmtO1wqOsZwUrdqmldbKO/EbElQhVwHBGnwqArmGrk8Z6L8r/rFSJ2tG+vTblryK1ej4GrJEgk87lesIxLqWO8DoxQz7NzIcc3bqrfN+pq803res+XrW2mVOTYioCrtnBNxM/WsXO0BQN3kpfjSWvO/YBoE9PKygRVIjeF1nEmhTF5julYafry94Q+sGTMzimS3xaxvivGHJFz/UtfznSl5T5tk+lUruOYoFy7UtRrhv1cyFQ/QISLidJgh6dI1wJalS0wXpFUdu9dIVT9rvqCF+LeSN4QWy3LhSFmDHyJCDMwj9Tw/tCemIsnSCebYz5fJkSlsM/CulvDXFPD0RUB9mXL5NURjm+tck5j7yDaT1SqwJ6GmD0TNJkUtwXM7BDPne8dKiH51p7w+90dvafe0jwdIUUpLxOKz4EOPf36eVV++ETpmE/OlNB7sdR4mhR6lsqwf+MzzVraFO5v7RgP9AfuzXnsgcHr1/2n2OAmNc9zMy3XtwlSUC/Oz/Em3Vezrf1nr60x1MTxKxGP3EBoqouxaLgXxJsKn2XUA5LHius6QWWu7WeBsnEOd/ksyGM28CzCtM12Tfd3PiNuqmdEv4U9GjAyJmE3kcTax/7228o6caysrKysrKysrKysrKysrKzeAtkfcaysrKysrKysrKysrKysrKzeAr0aTlU4JktXfxI4OtW/2ZBlrd6WxT0BYmISnAKf6j1MXkgc2TodT/ZHx+i7HAc2R2A5a1Zwo781zqrsFHgtfzkr4KLME8YnsHJ/9VCncl8Bp+kDY5jD5go3tslQ3ROiPkSK9PbKLr7JTJVVCseqXWbAvoajQVUOG7roEVKoZgsgOU3aw5mSAlt9KltZltA/yHun/ZQXSmtbWQ/0b6+V8Xc8hnzt5HEmBfBv8R43e9nLa2hBjISjy4HshbQa0gqa52/GqpoXhZnOS4Rjpnaco5uHsPO7aBcXdZ6hjZIYFkJggO2uxrcX6jNTF6kdwOgYpxGQoivr3aWtFLZelvNA1l8nkt1wMtP1so8wwGcbl/KDG0qt+uCu7NL3n2nsfv7wUVV+fqK0OmJWcbr6gqGDCn5NuZ5nWr1Vgs3egVADF1b7NIXVPea8pmIDCTqdjhJxDDDX3P8GtBWfkzPBhZEo+Z/OkWv4KhPGirV8JnwPPo+NhbQBopjEtdy2LKbNtu5vH5Nr01efejFQnf3DL5SoR+vxJlUUuVmU4yiGPbwVaa1qw1q81xeGGKLPs93XIvSY7IF6yWCDp4U6wFznMKGBbVp+l+sTeVMxRtpOQowvfvnr+Rq58g2TJ2zXPuYjJl12YPmvIYkrhHU+KucGx93cv0G5jjGNMm5mrwPrt6t7nI60n7kYCNuYzvR6B21LzNwDBuDTGo91bG3dQP1kQJtSpEaZdPW673yTDR1tD6u6C2TUR8MF2LeE+M4Y6XdriDfKXk1tEQLvvBoKS+j2NI8zlW2jKoxxyrk6J57o0lr/ciRpHcXWe3wgKUzTywwRJpTTl6/5Hvqwg75RXGOvGfcrKKOec2DOecpUIWKLwFKY4IJ+Qrw6RvrOYKx9+nCqPc0U75/he9Ng8wmqRaF7niP5JgSSWCvQfxaqnwukhx4fay2fXCFBB2219d7dqpzgeYJrv4O0sQDPEblL/LHc22R4buFjBvYqGdbT4Uzj+fnJSVU+wrPFaKq5Jl8Cl8+JkCDFi/M4+mgDyXkhxl/hval/yy+MVyLFTq56cYEZG7zuoQ/7qNs6xk2/pb3Ojz68V5X/8uOPqvL3bik1eK+pZ8c6jjkI8bzGNCGnXFtImI2RjPjia+wb7z+syv/8xZdV+bPH2k+eYjzN8UzrYm/kvgRPNMaYDOOsWEv9xFETaPfr5XKTu5y8yE18jYqPkIiKZ44Q+7YW1m+i4nUcNxL4TFvD/AS01cH+32DcGx7XwddZLvc2Hn/eIMJH3Bx7JabUMlXXxXVxzQs9HCNS4DmK62VInArPS0Cda3X8ltJWX/y2sk4cKysrKysrKysrKysrKysrq7dA9kccKysrKysrKysrKysrKysrq7dAr4RTGVMYx135zOohLIl92YFqNZm55hOdYD2IZUm7vDyuypMZ0jNwAr7n6zOZghPWhWL1ezqFvAvkI2wAi/FKq3MGuymPpM5lsyxgy3z+5EVV/u0vPq3K06FOJG/CHnXQ03VNcKJ+Dda3Jiy93lzfm9E6u5ZWsaqPzVpVM7OMV/bM0UQI1WAs23PDRaoGcLdaQ0hAryesoYvTtecj2Q2XCySK0MGGtiAV5dPOht8Xr0u5KyspT3pnMgrrMoE9OYOdFW5T48GGFyKVpA3bdMLklaGsctOJyiPY63NgBpmB/X2DKoqiuu8l7NvLmBZyWC1h4Q/YD83Lbfvrx8bLThgDM+Tk4aMhizWkhjzh6j0J0Dq/pvoPYGGM0Y4xGmwBK/ToUsknDrCjYoKUlYEsoEVL9/fDDw+qMrqvefAA+OAEc1Npr/7PT4FzvaZc1zW1+srW24C918UckBKZ4USAAdUCwtpA6kKGunU4AIuXowVryTdEC3L3T19fwxnRhzC35kRFkB6wlpTD64JVlf+64IWw4mLObUVYd4gZoR8/PJJFffEzzUeblOu6plVf1bWbaY7s41rbQL9C1LOPRvVcWILxt0wSyWC9Xk7UFxdAY31gwTkT/zB2rrGQAuM5ifXf58AWR2N99gT2f86jfqg5gklMLhMkQAO6aOE6EnS2OurLzbr2D66v+bViNDfoG/dcY7rRqq76DdV9Ats46+Ec8/0StvcEKMsIiRXTBfBQYr60wyesW4wpDLUkVrukJQaTzZFqgzaO0bfGWHMnSCOaAs/g9ziwextgnEtgNZwjfNxTq6l7HY60r7gaar/R6ShNcJNyCsf4+ar/JR5xF6QuEQNE8lDGusN7UqJVa4koTPki/o3kqYxrLROp/tT+T8x87V9XsZ8skE6UxjxygJeFVEqiWEyw4uv4nARHB6zhE8mfoph/cp0bUpal5nKwWtuJwdWQ7gJKyMwwD54DoeL+gLvoZkMpRU+eaz9//9lRVd4GFvmdO8Kwb+0L7a0S/owx14QGE254agP3PAMk231x/6uqfHKsayFuyGSfoAMMj0lV6KMZ9m5nV3ru6uIZpY011X9jOJXWnAzzzwJzWIy5k/u/GhAqIjoHe7tVeQ9pUw7uYYrxMsb4q2NtCZtIx2N/LvFDJit7XcxtdV3XrZt65vz44/eq8j/+8rdV+R9+8auq/AB7ESa/EXNkchJT6fKM+3Q83wKTDcodubPJhbEoTF623XCgsZVjX8pHuwDPGVGNqb5Cjp8+05rT29I429tX+/t1JEIBCSWK7/h8ViYWtXp/xvfieY7PlvFSdUkseoa1ajHGsx3nXB7ngHKKhODUxzMw9qthTWWi1mGNaYLfTtaJY2VlZWVlZWVlZWVlZWVlZfUWyP6IY2VlZWVlZWVlZWVlZWVlZfUW6JVwKtc1pl5f2ZI+fFe2wo+/I2usk8pu9PUzIUnzuSxGCaxqfkt2qlu771blhi9cZzmSJenZw8+q8pcT2Zz6O+9U5Q9++H5V3r65QoMC2MppjzVLWEbPVH706T9X5Tosfz/8kU5BdyNhR797olPIP334rCpPYTlewCqXAx1bx6WIFG0+TSXPM7Ocr651gXtPwTi4OC17Dye9FzXZBydj2eke8TT2E917jpP6Ox3YQyOkDMBKVgNOUPP+NFUhhlV9Dhv4HLzNJVIIroa6v3itvlXjtK91cPJ9pwt0bEd9tLut1+cJ0SJYH5ey5bneq9vjvo08zzWd1upaji5k8yfCwrQnl/VJZzRsqxnqfwQs6wHGcTSQFfbwlmzGO9tKgWJoTLzWhVdjsADCM5+rrs4vZCc+PRVymS6BijhMNmA6CCzyQDpToCuDU9lZt5HUc4D+vuwKR7rKhAKkZdUEG/7Z+/rjPFg/UyIOtPWvpZ4Qm1PbXp7LBuojYcjzo5eWfViVo7aQrhrTG0KVi9KqypP8OU1l8JDnmO9iJNslcyAhS5VdoFgB8asFTvsHtpC5ard4LUFL9RQDITi61By9STlFYbzye3ZRn72AyJDKHjEPWKPHI61np2PNY/Ol6s6H9dvjv8EAhSJOxenAAEfIS9szbdpToEMDJHJcYf0lVpGsWbzJyKnYRD/qwMLuecQoZbuOMB9HQHpo0y4qdHJzoLHnOmarvbqmGq35I43FAdKpBmgf7idmsPJ/+eTrqsyUpt097Z329jRvtupIZ1lLJwJCBzv3ouwXi6n+bgbc6Xyq9e9ooO8/B2LKFDIiDBHKDE1brOE8+ts69jnNSG1IBGYGvGyGvrZZOZVdnulRcN6bAO1VELNHd5piPA2Aq3mYf+pIxJs6tNCjzwdYl1z1Hx9T+TUuWqDvzMcafzFQhNlU8+UECUqknRa49hTzX7rgWgIUDn97b1dHFDw9VT9ZnGvuLDD37++s+u/DpVCk11VeFFUKXh1oXobpboo1YT5U3S+QGNpBeuYhMKhmS5+5xLp0jrXz5METfe+V5mXv4+9W5Z1tPaOEZT/3mACG9XoyURt+9fBxVT5+ob1VGylJTYybbAFsFv2mg71lwI0J+uhgrL+dIbZqh+sl5+4NKyvH4gJ9OyV6iOMawpB7cY2hNuaiAnujoyfPq/Llqea337fVdn2kdf7w+9+ryu/e0jNrF8dEROV67BqOLV37Esc4TDKk72I/efNQfe2Du8L2rybCkc5HSFnCePIw2XLvlyQ8dgTPLzwa4TpFydkkTmVMXvbj0ZXmgADrt4e5r1hLfmYiFJP39Lcn5xrHR6d6jtzeUV3dvafnjP2bwLb5XUTMSrQpR9pZjv8+OtdnP7n/x6p8iXS4Amtbu4b9MvZcxKhzrAU1g702fusgZpVFeF7EfawlPX9LWSeOlZWVlZWVlZWVlZWVlZWV1Vsg+yOOlZWVlZWVlZWVlZWVlZWV1VugV/LuRKFvvnt3Zbf8+7/4YfV6HykoZ1/LVjm+lB3z5m2c8P69D6ty98ahPh+pSBn8oYuZbFxb+7J7zkY+3iMb5fxCJ757vRWi5QIFon00hl16fCxrY1NOfZPidPjf3hfONY9llTq7EgqyoIUVtrMlUlPyb3AwEqG6LucbPm3clDawGlJEdu8IZfvJX/9dVT4GgvDLz39WlR8/fFqVJ2PZUCfAY7IMyUzPZVttoz4Pd2U93O0imaSG08lL/meO9A4mTz05kj3u5FKW7Ri2Q4cYAiyIIayk81jXfjVRfzosdAr+J5/8uCr3t3Tt7Z5OzS8c2fhCD2keG5RjVC+4HeM4sO+toSchyqrbqK/rrqP+G929qryAtfAUiQ9E8OAaNB2kADCFwytTM0gzDq9kgz15DgwRKEB/S2O31xXaVoOF2IWddbpU2/nbarvswcOqfHms++i09Zm9SDbevAkbc4n9MC3jdeW6jmmUYyEjHpgzkQoW+7mspxmQtBSW8BnSgyJM70yie34kVG2OlLydXbXbux8KSd0FNueXlm8H/clhwtFYY+j5I7Xngz/IthqjP+0gYeLG7RtVudbRHJHDij4D2sPklQXGMce9t5Y8gUiTDarIC5POVp/913+hervd0yJyegzcAGlvVyPNr/dhsz8CFnAx0D2fD9RviZTWkexxc1tj+uae1stmR7ZxtxyLCZDQ58dKFnz8XH3k9FJjdApLMOfUKJKFuI7kEw/pS/2+rO13YWePcO1rSQ342z2grncOV4jv86+FK72ufM8xW53VdxfAIIbAEa6AsgyRBraMGbGoNo/6qvtzjLOn93XdvTPV7d1DoVU7Dcyb7OdAZcbz1Tg6wz6L8/PzM5WHmCO8QNcYAK1M+W96xJwDtU9Q0zgbjdRHaWf30W4B7OFDIEJXV7rmjcopqnUpwzrcAo7XQ1/KY2Bx2HM+OdGe4v4zpaQRzwzrWiuYLnjQ03r1wU3tb+8Ao+sAewvK/XMK5OQSyNsl5tRTzBcXU2E+A+B646FedzAXEulME31miLrZ62g+/uD2HV3D4kFVzmPV65/dWR1jcHTGJKjXk2OEjbrAbRJ8L9PbUrShAVbdagpPGgGzP0FiWrumvrDb1Xjdb2sszuaq87MXmhcjzH+9zmoP4fiq7yXG/Dnm1vlUr+/uq3+4GJdzjNeTob5/gT31e321VQeIGFHkAmz7DHNHgUyhzR/aUH5HYUxejimmLvGoBIMxuoVEru/cVr0cAHH/AOvGIebXDEjSGeqLiXjnL/TM0sN+tbatcZmXz0Q595PYU704VT//8sHjqnx0que/BPu3Ho6O+JsfCef6PfZGj19obwBC0vjYpxdIeiJeHUZ6/Tpp13E3580oirxKwVsgvWkOxDNJXo57t7GvvntPz5dtHF8wnqtPnp5pjLx4rvl3NgUq5aqP7GMPWmsgTbVcf1zskSdAIo8fC7cbYC9MRKyHuXqnq7kgQmroAkfHHJ+pPYcjXXsTKGCMYwRqdTwXAjXz12Kcv52sE8fKysrKysrKysrKysrKysrqLZD9EcfKysrKysrKysrKysrKysrqLdAr4VStRmT+7Y8/MMYYc9CVrWl8IbvpDFjTjQPZSm/uysrdD3HC87n+9uzr31flU5zaPrzUZ2ZAJe7ek+Vu94YsT06AFKjypGqDBAwnla1pOpY97hLWu8ulrvE//kYYxq8/0zUWuaxPJ7SXwaA4g9M6ZaIM7H8OUzZY3Fz4hr6rMMYr8YcaToO/cUPW2b3Du1X5f/n3/0NVvo/UrRQpFXOmITiw98Hqx6SqMeyA2YmsrQGup4HUsqI8Zdw3suENUd/zVHUZNGTZRzjF2gngrquKvcYKjFlP4Qhq+j8TIHcvXsgqN57o9fdxgvnWtq4zL2D13aBcxzH10oZ9nVJljDFnSGyj7bLW1Fipw+IXAgObAIs5OtK4aAB/awC5imiLHMv+HQQ4QT5QvbilVTAFFpfDekrcot5gCpja7tGR+gvHRwspFg0mG+wA48x1389dnYQ/vlKbuojzCWFLjsLVvTJZ6HXle57pl1bh5QKJZujPGdJT5rCzpkgrGMIqejm5X5WPpmp/pq1k+NsISNwctn0Xvl4nA9Zyd2WLDdfSM4SWXDyTVfXRH35XlT///MuqnMAi3xpo3n4BhO5DJBJs1zWPuOgvDrAXWo9TJpXBKx69IeN4FNXN9z5cpRY6SKBAYI856Mn6HiNNLy60FuWw5zaRzDRE+sNiqXZ8Ahv2BJjFH2H5/fCO5vXvvI/0xzIpajjWZ3/5UBjyo+f67CGSJWlb79POD8yxjfkoBLLrukx2UNu5HnlQjb9uU+P47r5SPva7q+/9P2Exf125jjG1MsluDNRnDITqHONsjPVnOld/e/pClvDPvlB9OpisPCSYPYQ9/2vgpD95T2jhHupzAfztuMTsnhzLyn+FeZjzZlgTkuDX9Hkj4o9nSufgLMe5tRmqrZjskwEFCtB3iTNyzlpDsTaowhiTe6v7drDOE9n74KYSNy/P1M9jRP9cIM2L89UYKOjlM9XXAkklTVjif3BT/fbv/+xHVfnjD+5W5XqyWi/ZFp/dV9/59IHWqsdAGy+Q7JcCJ91qaCzWMS+aTJ9f95BU1NA6vt3XmIswdgvMuxH2Sd2yjr0NTq2O4xi/RBscpCg5wCMCtK0X4MuRgjoBBvPga42Rc+wjgybuq6bx/S7Q4vf7WqOIoV2ca+w2r/er6O8joJiDIfC1tlCNo4m+/+ELfd4FkvAuLrWm7LR034dYU1qoAtJKHvYwkwXTlPQH+QYRcSpPUzMerK6dKYw++pKLtfq9A9Xzhze0L73RE/J20EVqJm40xt51CymIXTxr1pBUmi+wT54judVbfReRwSnm/cmF2sgFWrmPVLdWD8cSIKnPx/p390D7m1/8+tOq/PRE81GI9S9HvzrDHFBva16/TrkaHOnZ+fXlGLfEffgMRYTKw33VgVU32mq3MfaupydYr6ZIR6wBT21pjCSYWx8+EBLH57XdfY2FWvncGU9xVMqJfmfIsHZHeOaMsT88GWkdfXGltSpEn+vj/ppttTn3dESoTIDPqSOpE79NMLn328o6caysrKysrKysrKysrKysrKzeAr2SEyfwPXO4u/rFOV3i8LREboMafuF3US5i/WvtCQ4W+uWX+ten+39U+cW5/mVxjF+0Ozic8hP8WvZ3+3+h68Qvy/Ni9YshDBfG1Y+sxuCw4QwOghyHsz690qGcTwaqsh5Oc8VZcibDYaQ5HCsOfjMr8C/6Hg7p5WFMSWXL2Zwlpyhyk5VOiLTQ9U/xL3jHOKQuyfSvMe++852qPLzUvzg+X6r9fVxr4fOQZvyLbKp/CfHxL69TOEfGOFi4VTqrcnzGDL8EpzjMN6ipDXkgZIRDcHGOlEnxC76PQ9b4i3I70vUWC/XpX38lV9aDL+SA+K/+5q+q8vbBO+ZNyPU80y5//e110T+e6xdnOgHqHf0rRw2uoWc4CPpyjHHc1i/bD46eV+W9Lf0r3/fuaYwEcBfMcOhnC46q6wtK4MSZpyqnGE8z/CvHxVL397un+leGCf61tI92P+jpV+57cHQEbTkDd2VQMPWmflGfj+TGm+FfD7zy2hxns/9qde3CG+HwxAYOWBsPdD3LqcbEAnPf6YXq4csnqp8z/Kv9AofpttEm7x/qAOsTHKDrJvqX4Kilv91+5z1jjDFepH8B4oF3p2f6V5YHT3WA6/M5DmrEgbzOQOXmH3XtX/z286r8t598XJXp4qzh0EAfzrk6DkXnv5DPlipvUr4fmJ3ywP36kdrLwSGU9T4O4fbwLzVwM0xmKj96pDF3ygNh4ZyZjOFia+tfAiNYEJdwiXR6mgMOy8Oq88f6HrfAAbZwmfbRXxy4SNjWKebuIeaXbkP/yrSzI3dJDKddHW69oND1buH1bbhHFsPV+pRnXMhfT1maVf/iOhqqvsex7usC/yI3wwHPMEeZ6VDjkgdV/9Un2p98+CM5Mv7hH3+m9z+Vs+MGDpw/7Kv+myEO6y+vM8Za6GDu6OJA+N1bWodcHN7+j//3P1XlIcYHQw9SOET8Nlw22MOEOPyYB1KzzNCG2Qz/QrlJOYXJyj2ik3NfhUNo4SZNZxqvjbru+eaeXGvnaNMQ/7rrYFM5wHuacMJNcJ9f4SDuWzc174alc3QQ67O/givr+FR7rQncQsZFG8EJVWDhD0NdY7Mmp+s7u+ob+UzrdYD1jYelGzhituDCv3d3dR+1377So8T/j5zqcNIc+2HeVwsu4Aj7hn5D69IXTzSezjlecWjxMzhkFnD5B+gLhx19Vw//gs79VVHaAiK41nhEaTvSGD5JVJe/fKw+8Sn2NhHcVEGOZ45UbVVHO7g4oLxb0/W20UfjBO4bOCrp8tik8jyvAmccOMJ8zC0RbF19zHM9HGYcYv/95ETjdTyBc2miesnxnPXugebRnaZeH7vab4XYO7Tbq/ZlX0uWcJkjqKTb1b7oAm79L373RVX2YFHrdDTvttoq//UncOiNtTHlM9CjE/Xl6QJjsa+963U/+fJfcTjuN8lx5LSpN3W/Lp6nQsxDfThSQjhlf3tfdfIYfb7AQcUkMQ4PND9++L7qZJar/s8utf/o9fA55aHUk6naeDpW2YFDNYbj/Bn2WQ9x4PHxiebffKE18p1DuYV+9H2FNR3uqB9nMSiimfbX+VLzTo5n0MJ59XnUOnGsrKysrKysrKysrKysrKys3gLZH3GsrKysrKysrKysrKysrKys3gK9knfHdQvTilb2I9fIPhbAvu7WZO8KYW0c4gCjX336WVX+D78SqhT3dPjq84FsSD4Qqr/6r/+2Kr9/T1a57g39bQ0HtTnZ6neqPNf1+rCJRjhkrJvq9b1tHUjXQN79n/1A2EwO+9X//rNfVOXjF7KoO4XsdB6wkCSQjWsJ7AhuaeOlq/ekmz7guLShF0DJRueyj40Hsineu/d+Vd776c2q/NtP/3NVvhjI1jaf6m9jHA6WpbrfAN2ugYO5PByKvJjjgNmy+RO04RKHbS5wUFXm4MAtYG2NDg7bxMG+dB56sKT6wAaitv6WVukM1r4YVrnLX/+2KgcfvZlD43zPqyyn/bnqtoYDZ3nyrw+bfQf9mYhJhEMbt7ZkvU5x4OrWlqzIbRxmmk50/zEwtgQHLvqlNTomTkO7NA6+9PH78s6ursU/1sFvvoODxWCRP3+m8Udb7uGh5ot2W5/ZBv4xQrtfXglTWsYrWzwt1K+rPMvNYryyAXtNtc8kU12enWlchrC6O8BO3v9QyJiBDfzjQPfY2NH82MKBtO/e0d8uL2XZ9ReyLYc4oNy5tnvCsp2ib/W21T/+/Cd/XpU/wrrw8FLWVp7CmE1kW16c6wC7OuzOtLnvow23gItNcd5t5BCtkuXamM2hOLPpxPzzz1dYyo+7uta776tuY+Bk7S29JwK2+957qMee7m0Wa4xcXukevrz/WBeBjpkDb75zS7bkH/xAOOzNmyuLsuPoOx98pQNUe0AsMtjtI2BQTV913mlyjtTrhofL4oBq7hmCQH0pdnlYq+b1otB79ndXfTnwde2vqyzLzKDEMxewWBOHmc54IDvmVuwnCqxz2z3NPT/83ntV+QbmoQ9uCIOLcWjn7dvqO/1dvT8AtnRZYqu9ntpqy1Md12Ftv/uh1vE6+tzxM42zBTDYkPcEXKuHeSqLxZC3m0COMU+FDA7A63Mc5L5RFcaY67bB9y2wtqU4VJu4V6+h+j/cFm5ksJ71upo7P/5Q9XiFg5rfv3u3Kgc4uNWLcc9YR5rlmAoi1ee9u0II3r2rtlvikHm3oTn1d1/8oSqnCfAo4O11V196Z0f34RW618lI11DDAd88aP3mtvpjqzzwn0Edm9D11o2HqRJNqWEfVsdGS7aFAAAgAElEQVRauL2jNml2tD/poD1zo3Y+HmgtGs3VPh+/L/zwLlCNfk33mRY4oLe2que1ObGlOTyIVHaA5P8QaAdDFfoYo72axtxuR++/va/PXFxpnmI/3kZYxADfa3AYLQ8i36SKIq/QeQfPcET8apgf6sDi2kCleT8FcEa/o/fU25p/DPYCdw41vzZd9RMXzwkLHD47L+fAFBh2DKS2joOKSb5EdY3FDH+bZkS4EbgQa43e3dJ42sbB3KcX2oO9ADLv4fiIZMl9zGp8FJtMxHEc45VjbRfr0MU5AmmwP9jeUr+tAZWbzTSXdPA8tbWl/QkP8G7UNNbbwPl9h3iu+vNiwbVr1bcXOPg9Q4BHARQsQGjLnS3NiceYB3nud62h9rnCbxTHL3SMxSE+h0jvAs+Iy7n+1q3hOTl/9edF68SxsrKysrKysrKysrKysrKyegtkf8SxsrKysrKysrKysrKysrKyegv0SjiVUxTGLa21zDbvwmJa3xOG5CMF4Qon7NeQr/7Jv/lpVf7uj/+6Kp9cyp6UZrJFffKj71Xl24f63mZbtqgCGE1SohvxUnYrr66/Czuyc20jDcOHnesvP1aqxoeHskhfHOkanwKh+g3QJISDmALWclPQNiWLWAZLpXttQdwgkZPnhZnOV/atFDhQjqSRAZCVZK7f+e7jhPEXX8uG7dFWCvbLQcJQAUtyjFSRAXGeQChGC7a8dmnLG17J+urBzu7AsujCYucgpWU8gMUY9tQaUo182Epz/O0C5bQOtCPQNXTQd1o4fd8PN2f5pxzXMfUSLem09N1N2FBpAyWT1wA+dPtA47XXRspYqPfc/FAnr3eBJzWR4OXVNNYT4FI8ed0pU9hyWIjra6kKslnOgJ/Mkf7yk1s4qR5j2oWF1X9X6MI2rI1tWF6LDCgc/JIh3hNGup6rMn4u3yRPVeTGJKu5zYl1k5dD2UqXQBkS2H5vwA7/3R8rven9H35UlScXQF494C5I2GjDqtw61OcQlYiRmhKW6EuBRAK/rs+48Y5s6DdvCAkZIL1hewf9gykQC/W/zidCf250db2TiSzGB2cq30YSz+lUKR/EJV13kwkqUpKm5qRc3x4Z9Z8/b2gOy5H+FwAbrbX0nkZHaO/BDa05p8eam7OF1pmbSPOYoe5u3tZ4/cFHH1Tlw0PNr41oVRdtpGTsA0lm4t+CiTgYz52e5k7a4vt93VMHeB2tyA5QKBfWfg/zZQ1z7SJBQlI5T28yKS4vCjMt16IMaZET2Pe5Truon3pIHFp1dWtPY7QTaHxfPXtYlT9+R+1cA05wY19/G8CWHhBd7K5QiZuawtcSPurAbXLY2a/GssLfRZLGCOmMNbTP7t6tqtxuqu9enqsvMtmHfaeOhBKiVeMx63XDKqu6wD9RJtgXuEDFmsBC+9iv7rQ05+xgv5oipcnFHjgA/tYDxtNuqD3SmeooTzSXXyeS+rn6yId3VOe72xqXw4n+bgKUYqephJs5rPoGKHTbwx4A+4QEY3qBecQHdtAFZvzdu0ruunu4Wo9rm0RyCuFUMfaHXeCJKfC4zAB38dXO+7c1n948BDI21D2Op2rnBOO7gX67v6X2ryMdcQoksCQ4TIYUNA/JfBEice8gBbeD54/v7qhfJrjvVkv9chdoVwHs+vRKn9/AHNDq6NrHjq6XCBXn302qMIVZlvNhhIW43dKYC4EYBXj24ft7Te4RNdklGC8O9o6NJjBybgCAM06mmgPn2K/OJqv3ZPjsBZLN2h31wQ7w1wJr+i4SXEcT7aMSpFWGwMgDrHlztDvnegdJwsuZ3rOM9UzULPdhRBBfV57vm87Oahy5qMsxnsUc4Kk7GCt726qHv/9bJfYm2NBneI5j2iGTytpI9cqQbJsCSWNSZZ6n14XqtQ4+LwjVV3aBk5M2rOEIgY8+0POEi4fxyLCPqk+3sG8ZAHOcTXW9M+BaUQfHrOA5+dvKOnGsrKysrKysrKysrKysrKys3gLZH3GsrKysrKysrKysrKysrKys3gK9mr88L4wzW1kXfWAo7aasSt0tWRgjJMYc7Olk63vf/UFVHsLD5PmyJP1lRyfyhzgRvIayQapTDiuwC9tarUwHyOlSgt3XjZBkBYtdwAQDWE/9HaAIsFb+u09kYf/3v1I60dUIVnR+L04QZ4IVDeJw0G1MaV6Y89KOx+SCfCJL6iVwqtFM7XN6IszDXai85QMBg1XZA+aUwLLJ9CKPFkrYrVtANEJv9X5aXBtIYeog6WSZEqfSdxLnqjE9A99fA15WQ7JWI9HnO54+ZxfWynqi1+ewHrfrGhubVWFMmcpQC3Q/Hdj6Lq54Gr7uzUUdbcEqfmNXY9cP9TkBxnqAVAgmHNByOkfdJSAIF2UbxLBNukTPgGf1mkiewtjtI23BgYXS4PtjoFUcTzkmgQxWSNooM/4FEjeK0va5yYP/TZGbLF1hDucnQsOGsMkPrmTBdIdCIj56/15V7vWFrOx0lcaQvA9MEw0Bx7kJMf8VHk/SV125CRI5SoSQSRo5cDQHFnJa7Hf6SKQjY5oR4WBqANCbTHWTYp738V3djvpoE/PROdaCFnCeTSrPCzMukaOnl2qvCZJK3nkHaTDo/yHShIKW5oqMluhU9ZuhLbpdIQJMfrp1EwlySO1ooK1lF9fnbTMFCfMWCCGTL4AtriXH6D01ICQRUqsC4CoJbNSzpeqMsT23gKjEwG+ych7Z5FgsjGOKcn2eY66kvd0HMstkrBr2Cg62J3eAr90GtjSd6fOXwFwj7DMaQF+8XJ/vwYZ/ndCz1ULCI9ZfBgbltGkDsTlAUsgBxlwXCYYdpBB6DuaFpeajF8C1EFZiAnSMCGtHDWNjoyqMcUuOqgCqmaJfzYGo3tkVzoIqN12MlR7WSA9IZoZ9WwicuIEENwMUe4B1yUPjXG8pogVt9ZjnMNfu7WvMtzH+2iPVZ7ZUfwhRBwH2NzMk2J0PhZnE6GsZ6ozIdr+JvVk5DjaJNhZFYeISl0qRWMkxt0CSlO8AsQByHDT1+gH68LKpulosiHYAhUK/rSFNqcAeIgCK4aar9xBkcWovR9sbWOd2gLP2sI5yX1wH2hEgZedsgHmTfR3ftZbag/mFCa0mfXWE49vIcZwKCXJwbxHwqAR44AyoDKhFE+C5oIm68PA6cf4aytzrjGJhhglwPIeNVqKF6RqeA+Qbddtu4mgBYOlZjrkD6NgCCCWfRxKsH0tg9TH2bDGeL5bo47NU826SLsrv31zypuf5plMm0jbwHLB8VzjVEmjQFtD3BpLctnY0z/qFPscDWujguJFaxJRd3Q+GvRkskCDF9lqs6iHg0Qccz3V8NsZwhLd3byE1ck/reJKo3xQ45sXH/Fdgng9wTzxeJAeinoUo/yuOa7BOHCsrKysrKysrKysrKysrK6u3QPZHHCsrKysrKysrKysrKysrK6u3QK+EU+VZaqZXF8YYY+Z0JMHKRDwFtJFxYUUOPeBRW7LHFS6SDGD9o22uAO6Q4rtol/Lh53WWpT0OFlqTI9UmhG3Z6HWz0PvTAKk2SCQIYcv86D2l5twC8vH0BVK2YLnK4PMrYML0cloqgz/5u9eXY0xlCda9Z0uVpxcXVTmq6x732roOhkuEETEnIChIahlOZP28mqie63VZErdxqnxEX3q6ej9cqKaJ9LCttvoWreprSALT1JAe4QAJyozscXWkM23DihvCrt6BtTqBz2+O/upEtFZvUEVuitIW78Gy3akjdSJGW8CymcPWF2A81XliPu7Tx+se6muJz1nihP8U9b5GhZQYFw2DOfo2Rqhpwu7L8Zyjz7oFLMFIdonmupbFTH1tNlP7prCcxhjr87nen3DOuL7XDSIcaZaai8vVWMtwYv+0UF/9/f1nVfkW+rnLPobPDANY3YHE5MAPc876SCbJJ0g1GQF5gj00L1O9OOfTysoGL4ANBcBqOj21W4F2cJhyB2yoAJ7n498dshSJV0B397dkf332TPXnLF7dqvptVBhV4xyo2CXSCL73nhJmpqcnVTmcvpyZLYBZ1SLNo7ff033eg228DYt6UGDtWmpeijAXTeNVXTBlstPS/BvU1B9DF+s1UskiTMgZ0t6WtMVna4BBVWIKQ4LxmsBOboBchZiP83I+KDa4LBZFYZalZX00VrsluH5i3X6A+RFW9xD7kG0kw7Tw/npf95IC6/RcoMBr6UV4HXPxNR4acJ+FCcrH+GMKo+cDv8L1hphnA5cICezsc9WNh7EYAwE1sNE7aPMCcVG+t8E0I8pxKrTHxdrCXngG+/+P7t2tyimS75iasoZ5ox8Sj68BFeTcPMK+Z4H5oAFc3CnZkQDYJNeq5UxjeKun8V/H2G2jX2T4Tgf4xxLL2QJz5wL9dwJc0qD+7hwIL9hD4mO+0b1p+ZlFUc0h3HsU4CfnwKd9g7WcyT1T1WET6Avn0xa6IfeLTDbLwNtwf5mj/U25b+cRCESOXSCMtRzoLzAcF2lyDhJ2HXB+OfoQWXUSUQk7O/Yr20h8jZBglSdvKCmuMMYvr9HDhRBbnGfse9xD4v2O+kCExKYAyD9RVxcLQ4x9eYLPz9YQMuw1yrFA/JTYbgwMi3veiM+3QEV9rNFrOOBSfTMD8p3nQKXwXbOFXo9wrMQUqY2j6aD8vA3icY5jnLKf56iIXl9zgMH1NIEQ+lg3AvRnYvMuU9KwL3WRWsZk6Xit/2NfiHU6u97Do30cDgSsrTjNoRrDxhiTYU70MC8EwIlTF2gVMKglPp9rOjGreMbfIHQR2b9iPrVOHCsrKysrKysrKysrKysrK6u3QPZHHCsrKysrKysrKysrKysrK6u3QK+EU6Vpbq7K9I2w0J/u7+mE7O6urKoh0A5TQwoUUgoiJIw4Hsu0UfKkamJTsME5SN8pZFtKSyt+CkyiAIrjwArmInkoiXHydA7LI+1XsM1ub8su/8M7H1TlX/9BKU5jV9fYQspKgKSCGe41dfhlm5HjFKZW+hl5cjsTEHxgDQ1aFnF6/wI+tO19oAITnVpuYOvrIO2ijpQdpjps92B5heUuKa3/LmytdeAzeZ0Wf9XfHNZgB3XcBdrBU/NnQMoaSEeqIanCxERs9Pkufw5Fyk8ebr4NjVmlJcVlnw4c9eF+S/XJE/Yj1FcGm6GBbZD9IYCtj5NECvQhXTttHckbsLMyncQpbYlMA8iIW9B+TCwAaUP5N/zsXDjf8B9gucxmsJDnsKpirE+JUyGZJir7u7tB+3iWFWY8XtXh/raSgUYXwppmib7vCtczGCpJbzlWudbW3zo19HPgiR5t4LCiO0CrlqgHF3Xll3hEAVvpGhCE5DMPhIXHdwGbzTFWaGfPiZPQTgvMwKBuOg3Ze+fLr6vyaKyLmLyhf7JwHMfUyv7ho3MPrjQX+rTNA61ZIBExBOIQNpDy2FEijd/QHOkCkeH4Luaw/wL/cAPMY6XVuYFkyWYDayQSMNbs7EylxD3FSEHJQeUlOVFF/Yf5krgI2tTQLg8Etqf2LbJyLHqvtH35/1Se52ZRXseCcxzWsAipQ7S0M13EBwbAJJsUqRZBnVgWcHIgaSk/E2sa8bTr3p/Rpp2oHZpY5xpEgmvqEx7Rc3z2Yibb+nyK+Qh4wmKOFJwl10IMNMyXWGo3mmb0bUQ86hTjcrrO9lbFIRDbXaBiuYv1knw32pcYboK5K8N6zP3IdYxYhtcyIicT1XmGcoNHAWD9TZF8wiSbJa5rin3MeKL+PpnqbxdL3esW8InuluajcTlnbzLfqMhzsyznh6it+W6tnxO9QVXOp+iTNaAvDb2/GWBc4CiAYi29UvUwJoYTA9Hgs0jZdYiqe8RBie9gPHEtDvGMZIDkMG1ojnmhwPqHrzI55sUOElR3DpUQ5GD/npKz26BcxzXhNTbIvSg2cfWars+ZAE/BXjwnQg3mhutSCobMXWP4sadAEmCKfS/npRTzWHUtaLolejrnFKJ+a0mAWM88g2dN7BMKPG8tgBNPgFFyvdxHimyG+eiixEGLDTL/RZaZeYlnzoEA5c7LUV0+T7hEklAnfoB642EA6CNLYGKLmcYLx72HvyV2eY38J8TdgLJ5HIsFnoXQL/MFxwSeZ5yX48HJNxxXsZY2jXGcAHP3Y4xF1+JUVlZWVlZWVlZWVlZWVlZWVv9Fyv6IY2VlZWVlZWVlZWVlZWVlZfUW6NXSqfLCjGZZ+Yey256cnFXlZk8Wa7+uj2eqjwvLEE9mLoAnZbg0pwAKANyIln6moqSwAs8GK+vskikdsESFW8IYHJxav5giqepKdq5eB8kLTNlqyKL8w0N95l1P1xI2ZSP76R6SmHBS/H98pmSoJxeraz4pNmhWLUwV1UDbX+IwlUflMFN71oFWhUCMxkAv5kjZIW7kBExy0N/S7tmoI4EkQUJIiVYwjIzv5enxeQYMoMEEDCRj4MRwInmNBu4PaT4xTzNHWxRMEMD7a3XZdXN3c5Z/Ks9lcwwi3UO/qz6cwhLcgIWf7c6T/H1YJB0mzyDbIytebiGNYSd0kIqSu3x/+R6011rbMW4GNmMmiNFtyO/3XVoS0e/onIS1lvbHObCAGO+p1dUf9stEs/D3z82mFAY1c+PmXWOMMY1I99uYM1VD4+9sqLnh2YvTqnxn/4U+syFswodlmniah3nLwan6REhnsca0Bytqw1/VCTHBtdP1QcawrzClhnO1k9MGTbupkMsC6NjRqRJk/vmPT6vybz5/UpU/e6r6mDI5yH8z/2bhmJV13BhjYnTQ5+fCNuJCc0LQO6jK4+OvqrILy7TjIn2lCSsy10606Rr+Aryuhf7DBMfrREQvQqoGEKEMyEnCNEfYyUGFmQw4CRNB5kjVmMdI6sG9EvusIyFra1f1FAETPS/TE/NN2sbzwsQlRlWknOPVP5nUGDVUngOxoIV8giQNJpkQcfORtrKevInxgvGXEUsscY6YdY/65kf0MtWrZ4gtwsI+VbtNkOA0nmosMiVljPvj66TcmOyzRhA5m2u7dRWmKD/b4VqFd5yPdT+fffW4Kh92NEaPR9rzHQIFb0eq1JpHVBrJawkToYBWoZ97jEW5XuuQ7JLN9HlLIFkJEIsc+Be2NGv76wx3Pkc/ISJ0BXxwHOuDxkQjsGd4cKK15+eff2mMMWaAOn1dFWaFNxqznkxZ4PoL1DETXWZAPpZIAJtPgUHVkN7FPok2XCIZaDok2oG5EMcdhPXVdQZAdogwTzK9dwD0LQOqvNXB/hOITRqrLy4WuhaiKzlR5FB1ttUGBtdTP06ACCbfhKK/phzfM0F5REIKhIyY7Ds3dRRDBmxmCaxpyecU9G0m6Lkoh9hHci96OdactgS21MIaGJXIMZ+4pr762tVM62KK+SVqqu08YDY8LSDNXr6PTjAHTzEHLzB3cEzv9JW+GSP1bDRZ3d/aXus1lWW5mV6NymsGaraW6su9nYohnrlqaB+i3ymeDxLcS5ERW3x5+ikTAYNAY21ZPpexvh0e8QAMLwC2yEQ0BMQaDwuay7ECDJGpmsSsGKHJZF9i2gvs953o1VMbrRPHysrKysrKysrKysrKysrK6i2Q/RHHysrKysrKysrKysrKysrK6i3QK7EehaMDnGNYq87PT6pyb1vWtM6erHy1LnEq2VazXDY0N2dSFMqwkjnw62cxUgaGuoZsItucV1qraFv2UyRgOLIYOrAzLmaySh19+aAqFy3ZRvs3buN7ZLO821Xd/Pcf6vPvNWUj+yCSnSr2dU/vIX3kl1+ubPH/83hz9jjHcY3jlveJtCzj6ZpzhxiEfudbwgYawYJIS5wDe2g213XPYX/1YH3zYGe7vFR7FjFshYtVnXsO06ZgkHZoWVM5XrOw06YGWzMsbkzEyoAQJLgnD/WRA/9ot5XYsLt/qyqHRvW0SeVZYWYl8hek6lf1SP0twqHxtOG6hjZwGs158jqToiDgRrTfL2HzrgN9cHlqf2mHdmHlJc4To30XbC+gVUwEiYBQOWjTOJddmpbKJb53CWSCltcaLNjRjtIcrl/dZCKO4xoThm55nUgdaegee33hcVNgDUdHSr073jqqykGo9/eRJuf3YL2ONMd4wFwLWDyZ0nB1rvnUba4sz+0AOOMY6R0TWIBD1WXCJDf0uQCJTA4QrmSiex1faF74+W++qMr/43/4p6r8Aolec2IpSDZzcyQRbFKOaxx/VR+zQvf/x+fC3754pPKdm7JDLzzVYx3oCa3iOdC/GPUynRNtUR3VGxoXza7W2hkQpmmJFxD5iTFHG1iYBxe69tkMqSnLrr4T87jDlBfsE5L05RZyhywI7O/GV1/OXH3+o6NV+tgy+dMkkdfRtQ09B/aCpc3UgCnUm2q3KRCLArFCQyA5yTckVblIuGEKzdr85Ks8R2LQtEziYRgLU/qYbjIDqkAkLwEGNRupb80xt08XLKMPIZFsASt8lGN9IS/LIBJe9KZVfg/brsAegcjsp7//Q1We3BHa0cVRAFdA6/stpFa+fOlcQzjY7mR3mEqWl3W0hg0DfWQ9p+AVUibcAABZIvl0hvfM8bdDoLMTzC9L1NMSN/X5Q2GfP/tP/09VPp6v/naEMfC6cowxQbnOhky3BL7OPUmBvrQEJjZF/w+n2AwFwM0wzhhaM8RYGCFNNUIiX4RU1OS6nwOfd33spwLso7EvPTp5VpXjmebqThfPIvHL8dSYibtYU1t97QFdHF3AMWCwv3KTN4M2Oo5TIbrJUt/94Fj3PAY2vYd02hGS+tj/O0DccwxAD2xvDfvFtf0n9pp8ZiCK0+2u9vFEZAN8v4fn0oQpVMSc8dk51gOmFnGtZeLfFAmOPK6h3cJzKnr/Aphglb61yeYscpOV2BCWKvMvJryq6KN+atgThGtIkkEZz2IoF9gTzBIm7KncQnJd2OAxEuXf4vlgjnnBFKrjVhfzCxOscMQJExzX8E7il7guVj/XOWJWMXBNpmLXaxansrKysrKysrKysrKysrKysvovUvZHHCsrKysrKysrKysrKysrK6u3QK/GBhSOyUqLPk+Ep514cCHr/d4ENjjYbf21U7qBp8A+FtCqBhdcvJT9e3jydVV++sVnVbnbkc1792CFtmQ46fvyVPhBDd/T7MgqmcKe7g6Pq/JkIMtVr6UUKnep9+/nKv/tgexRLVgEfeBXNdi3/yJSk7z/4apufiZS7PXlOMYvrYeB+3J7r4HVPQdKtOSJ6rDgRrDQGSTVDGHDpg14p6ekjBCWuwVs5osZ0oNKK5wPhCqC3TmAzb2B070TYGgL2OngbDUOLHe0WdOL7RhgXEgWyZl41dvR56Ns3M1a/q+VZrm5HK76UBPJRnWgUrnRPX/94o9V+fb+B1U5Tvar8hwnwpsFETnYQ2G9XcJymsBO2HBhOa7BFlz+LVE1ogtzoCIubeNLpNzBlunABuvQFo20KWIbC4y/eUq0BtZaYJcJ+tvZ5QopYZLO6yrPUjMdrz63Aztovab6a7V0jz6s5YOxxtazY6WFREj38ZuaByOMywh1GyCRL0R9jq9UV5///NOqfOtolQ618+5dvfe50glrTVnC730inDZEvTmo+xTJKxmsxItL4WJHz5U29Zs/yNZ/fAVsDjiBB7tugDHKPrVJz7HjOMYr+00Im+wE69+v/iBso1n/SNeXYjLCPBaQBYFtfjDU2sK0jVqD9l+Nuc8/E352Dlw1K73RJ8dKspojpaaJdqyjb8ZIGbmcaV3kmuswBQu3lwPRymHBLrD2xGg8v6t72r9xoyr//f5/Z4wx5n/6X/83szE5Qu9IdNVRDoDmdVE/C4zFKeahE+BUY8xJjoskJ64t6J9TWOyXQ/Xzs6E+89nz1cZgOnk57tFqA61cAGd2kPzIhKnxy9OmxpjbpxivE+KP2Ec1MJ8yqcXNNXdn6Zv798Nr1Jqj3VlLNUUCU6b6PwGuegXc6AT1u92FhR8JmQH6MyhWs0yJqeOCkMRalGtagf2Yg3UuRgpKgWSVAnPEMlF/ARRplpj/Rth3T7GMxUBAU8Q5dpFi2MSe5skzbUinJQK4yaQ413FMeI2eof6Y2EasxUmAXiAlZoAx5E+ATWBdn2OOnsaoFOwLDb7r7FJzbqODdMzy+cLD2l0DbrWY6z5mE13vddqjMcacXSh5cUb8HchzMtc1Xs4wLsGoNDAHJEizy2PMX8ACM4d4/ebkGFO149LXWBlxDXkitCovNMfPsY4Ox3p/v6nrjhrAb4hKcewQWwGK76Gc4LlmVvaxKeaFOTDkCMhNHf2CyLMBEsPUsBSfuQCuPpwBY0WCY4H18vRcSPPXR1/qbzF/X4/BPN8wHld2LS93/+VLq7LDZCY8N2Qs616yDOsGjkHIgLvneF5LMEbnWH+mC917D2h3fD25FUhQddgn8BwwR1Kd9/JnwRrbvKZ+7AB/dnEfhgg51pEE8/gS+wGmxdbNGrP2rWSdOFZWVlZWVlZWVlZWVlZWVlZvgeyPOFZWVlZWVlZWVlZWVlZWVlZvgV4Jp8qNMcvSUpXCDjsHpnCFlJDBsSzxXeImTdnsHZ4CncFKhBOkU5zmvpzLehbC1JXMZN36+lL2/pPLlZ3p2ZMn1Wvnj1TeuyMk6gc/FmayvSvb7PaB3jM5kZ1yeP6oKscDWSGnL/T59VR2rQBWs7TAyfWwUPmwnW2VyTXEAzai0nrIU/0NUhRyWMloB3Nhu0wSJCYAWRlPZM8fIiWsDYQqqMtu6MH66BfAsmL1r7q/sq4HsESH6Ll+8PIT6DsNWVVHie6PaJULa60Hi7PB6ehMJCpgm64hoWTv4KauB8hVstGj4qU8z81osmqbJMYp6ehXHq6PJ90zbeQKSIYDlM/AKh86apd6qDp1aEOFrdMFj7Bo6lT9tEw3myIphUktTP4KkBLlow7n+Nt5gdQkWNELzBczJDvEQANTWFVdYCyjscbx4y+FouSldTOON5dwlBfGxCWG4CPtqYV2CEEo1LgAAB12SURBVIPzqlzAWh4Eagcm2VzBetxGm3iwIQeO5twCVk4HyNXO7YOqfHtwtypfJymcHGuOa6FPtLY1bwaB6tVbSykD4gaEZwL85PREqM7Jherg4krzS47PJAITwHYfYryGeNMVk5heU77vm/29VeJUnqjfFuhvs7n66gSobhfzbgy7+8JTv00SteMImEsAS/z2gdrLD9SOg/vCzz77vfrzpLTzTsaw9foa57duv1OVe9tau+sYl+dHwtySge6p2dR7MozdDPZqJk76SEr5/o+Emt39znf0mUCD/BIvaNQ3mPznOMYt538f+ALTlXz0HweW+TpiAIm+nKLfHp+rzE1XDegZkbgICX/nwCW/fiaE/PozU2AgXdT3zrYSEwOsbTnm3DmQkzHWhQkQvmmiMUrkOV5LGAOqAFu6A4TAQTrTG1oWr7+o/F+8hHXRsIzrJjbB+xzj9RHs/K0cKX8FW1V1ncbaRzx58lxfixTBMmTMPPxaaMnpC733BsZfDWtxaFTOMI/Pc+DE2NJkTPQEIoDQOJNh/xQixandED4Y1l6G3m9wk+o4Fd4/XgBHQT/camjsE7MnnjvG+hchvYipMinxD8yb/V3hGW6q93/5tea844ePq/I1HvruDSWcRU193vMjtedOT+mE33/nnr4f/WxyrPmUSPoo1RgdYK13dvRMxbTNi6HWy/6+0jZrNaz7wRvCqRzHeOWcyXTYINF3pzHTxF4+/iZMx1sixTPHPhvY5hx4XYI+42F/wz48Q/rU83LcsQ4z7Esj7M0QVGRiPA8xwBTbTJNhnp4zHQ5ITwoUisc1nF+pPxyjnOH9FUZabHZyvb5Nh5gWxlyRMz0Ra0WmOSPG/MjnBiYe8zgVj3s49FUf2NrlpfbqBebZ60TACEd1cL8R4jeHAPOChzkxxxw6nSBlEmMUS4dJgGjl6ctxqphl7D+DVON7rVN9S1knjpWVlZWVlZWVlZWVlZWVldVbIPsjjpWVlZWVlZWVlZWVlZWVldVboFfDqfLCTJYrO1MEy6gDb+xgIFvZOVJL+n3ZEwNgHh5OpDawihZIRFjCEufj5PWoL1viD/76b6tyCqtSVqY/tPdkNzzeleXxzjuyGN7c0TUaJBhksIcHDf3udXYii+TyvtJ/iitZp11Y6osMdmzY9dZO+iZm9QZ+YnOMMd41n1W8HEPiFc1xcnrBNCCH6WHAVIjHwD7dgCXURcJRMsXJ4yN9V47T/K/xL6YnLXACOJyaJghpDUVdwgo/A1qVAV9rEK1asxgDUYpkEbxxQ8jBzi6Sypy1FjVvQnlhzPUB5yntew5Sz4Cz+DXVS+YwzQF205lsnbWGpgaEipgM1mIXn9ntauzMkObx5LFwjuwa+wEf2AXmU0eCAdEq9DSTwxY5meBkftj2CyANY+CXsfNyhArD0gxHsqrOYeON3sRYdJwKYWGqzBzXT5Qs8nXvTdjbO31hE0w4mTFVATZXz1UfKYA+MNWo1Rf++P0f/0jXNl7VZ46lo9fV3Oq1dV3s+jn6WYq5fbFEIsyV1gteu4HldQbMbx0zRVIcrbOIouvgc76ebg6najUb5t/+5Z8ZY4yJ57wfrQMLYIDDkZIm6uhYS+TpEHlZYr5KA445jR0DLAducnPn3rtVeTxD4tgX91fX3tbk2WwDeY3UF5j4EqJ/hU3hJKOB0GmOVwdtkeCeOI5d9PF2h6lYutcC6Q/XCXHFWv7Q68lxHFMrbdaZr75H5JjrWQyEPMI8mADhnY40P/3xqVCZHaBhIf42554KuGR/S/ucG0y1KP82xBhuIkmsh8QwD4Mxw3qxwLicAEUZMXkKGADxuBTMAbHoNbS1KF76+vp+Y4NyHPU5uv/xlsJ5OW6whFV/hMTVBTCCBfsh1oQULvgE48VFWlEKpOCff/mbqhyX7x8hHWu7p3ZsYy5mqtsCAy3B63PgDSkS1ZiaNQeCuwBPlaBNa57KzZb64913tO+5fLzCXpfYx722HMcUJf737Ejzpg82ZRfXE4KtL3C/a88QuEcf6IXBHqnR0XyWI1kqQoLUTz75aVUeDXXPD796aowx5g8PhRn3+prL3ntX2NSNfaGvTEE1SKKMF9iDxy/HHAl2t9oa60RuScA0kKgXIMU12HSa0bUcx/glogIiZh0BIk6MdW6M8TcDojxeaIz0lhojC4ytJRN1seaM8Bw5QLJRvNT3XverHA9fF+dCoT1sG4a52qUNzK3b1hEC9UDXPkOC3xSIdIz52FtLSMIxAjziAs8XGdZAt1yrNtmajsGz0Np+S99LbJzHM5yPdC81bAt9JEUFBh0DSnHvoa863AVmnMR8ttb1tMpnkUYIrBu/CfAIDW5WCtRciMRMp85nDt1fisRtcnMp965oc6J9XDsD7N2YWvZtZZ04VlZWVlZWVlZWVlZWVlZWVm+B7I84VlZWVlZWVlZWVlZWVlZWVm+BXg2nKorqBP8ZGIScoQMoHx3JHt/p4YRpIEkN2JM8WKtyWO5cIBdM3IlhyzJNWLRgkfJKe9KdfeEut7dvVOX5Ut//9e/+oPsAftDfVzpA1NL3j45goXqhU+sD4BAuEreKTLYsh3a0NU8XUxS8P33tdeWIWiuQFpI7L/ce86RtnurvuDhtHMkXAeyAOcimAp3k+Ew4wWIO+y5ODXcjWRLzEqNKYVnMYDubTIETDWS/dXFdDMagx5RngcPhv1Z2cN8B7NHNtuy3Gez1THYyLlivDSovjFmW/SmG1TL1iRUhGQTtMofNuAZbcgPYYIS2GwAx4v00Grr/vW1Z/r0dlZkskJc4UKurtg0wYYB2MgnuaQGMLkE9e7BCu8DiRinej35dB36ygHV+OdM1Dq904j3t/9e0gONsbiy6rmvqJRaV4noCoH919NvdLvpeQ+3Q6ag+a8QJgV7kM9XJPNUYGY107wvYigukBoQZka7V93pNjdWxA9x1be4FwoWpOk2BKiDNLgW62QIWcn6h/soUtHpApFdicl0bSM4WUBdzOjabUpGnZjld9ZsY6RkOkM8h0KqTE9VdY1uoRB3X5+TEh4D2ok8+e6Y1J2rqftrAnLi0bLfUT27v7BtjjPGQirYknoE2HwJzjZHkYTCPMPlrATsxrd0ZrPOeiwQj3FMDWOEaGuty7Sz/d9NjsV6OxTquf6l2QHiGyRKOV11HvyesAW5v8+hM88pH4GDqrDemzYTEWlQnt/eEgvfLxMcc1nOivxms3BdIz3SAMi1wkYs1hA/9D3NTTizLSAHW7mwN08ab0Ib5G0I4HIO5AP2DXcXFXiRHD02AGA3GmmfGqItl8XK0ajZFys4YuCrw4xuHSrBs1bTvnF6tyneA+W/vaawShX50dqR7wriJsP9NDToerPpTtPUQ1zsHXkR0k9NlVNPnvAtE09t5zxhjzP/1s//DbEpZlpmrEj99dKxniN2mLihDW3lYC2vYkzmx6q1e03rSwNrlY45xMY49pM2E2KO0cSzAwU3tc+7ur5Di8++9r7/zgZ5iAYxx7SPgJ3MgH7UtzSMp5iBikc2Wrn2BNWKOdt7Z1XERPpN4sHYG+ZtBGx2jpFKmqrmY+31G/GDfPEAi0DzTWJjERKJULxOsu1cXQHsZM4WxvsDeqBNpfm2XdVRgjpxjQzFGmubFQtf44ljPNJ2O+trNGxrzOdY/pqauHZuBB5XBEHWAuZkpczmeIx3nTbSjY4prrwefEZm0jH6bAPcbjIG+bWHvir5XYJwxWXU6AZK01HexHzXqQK9R55G/6i/TqT5jiqMXMvShtUdrrMX+WpIZnmnXSCzslzOul3gGQz/iUSPE9RygsOfn6kffVtaJY2VlZWVlZWVlZWVlZWVlZfUWyP6IY2VlZWVlZWVlZWVlZWVlZfUW6JVwKlMUxpSWMNqtaVPLYuALOG3/9Py0KjPhqdeThSpqIU0hUnl6Jpxj+lCfkyM1IZnCegZE6zrlw1s79FmWrKUry9XZV59X5dsfyBa5j3IjQHLMlWx7HuypAbAQA4SqoNUZP58x8cFjmkNpBXQ2eN544RRm6cTXX6DvglXOoVUOWItLrCgC+gZrmN/Q69Ox6uTZqWzApxdqn2ZLqNo+0jQaSClJk5X10Jvr+2e0qRWyo12iHyyAyu3vCltowydcD5GSgqQAbw2t4ntgi4UtL5zruwysdYl5czjVdWoM26iOvhI00b7gowZDWeuZWOEjsaCOdIbxRNbPYyIcSG1wkKDSRfrKHtKK/NKWmi1x0j7Qw+VC5clc43I8lxUyAWbSBh7VB6LFPjhG4pbj656WI9jZYd2dwXZZ82E/fiNWVaF687nquNNSnfVg/Xb21IcbsHgSK4lqSOpgPwQGs0R80f0vH1flz3+p+a/m6nvffV/zX6dbXkMkPGiKef7yHImEGHPvffRhVe5uyfK+hK059PSdTM06OtF3LXEfvQYt2urIERKfuk19zjZwwU0qy1IzGayukQkHnB84p05xz89ONBbbt3T/Tcw/GdbakwvV71cYi3WmlaEcMAkCNvat2gqjygCUXp1obb2cqJ5jXjtSPTysGTf2NP4DWOQz2MOJpeZACkK0XQ+Jk3UgdTks/3mFCW0Qp3JcU6+v5rME/SQvVA8psBqzlroEnGpH2HYeaE6aYg59dok9j6f3DEbq50zqjICZO/Bzz0qcjekqzFxbYK6czZHyg8WtWdccvoawo98UmPvW4QS0LT6TyFWBti3WUlXeDE5VFIXJs+uUlmLt9Wu5a8i6iiQvBmivs6Hmt+FUc9r2QuPbBXJ2eXKs1xPdf6emee9mU5/T2l5hVMTfJkBeJ1dCTs+BeaW4+P1D9bt2F1EwSJi6QFoay0vs2XkNWx1dYwdj4twIwfxv/+a/McYY85tP/5PZlFzPNVG5bygC3fsM428GZKGF62nVkbpUVx9u43mi2db7Z8BjTs6RjmqQgojUTB4d4AExC8v9sI+98NVEo3EExO4S3zMb69nm1oHq+/YNzYNFgDUFDw4L7E+mQHV29vercrOlOdT3uUaqnrxNJoutyalQEQ97aweUrFOoX8Xoz6cD7c8uR2q7Do7TaAGDn16NUdbfbnWU3NkANtXi0QF4fnCuVn0sByLUydCnfJUXiLCdeMC5rtSmfsjjBzTXMjUrxpgjRnZyrvVgTkQZbc1nlrDsD8RfX1eFMSar5nk+CyLtDeu9g/opsLfw8azQqqs9L8ca34+eaD9zNdb9NmpaIzvY3/YjHK3iac5bOLXyNd1HAjw55RhCfaepvjMBKh5F+s7bt4AzdzGPIJE3ifW33JeCejYp1qME7Xlyqv3dt5V14lhZWVlZWVlZWVlZWVlZWVm9BbI/4lhZWVlZWVlZWVlZWVlZWVm9BXrldKpFiTyEwDOCGmztoexDuU+bFWyI+EymqTR7spvGdZV//+nvq/IXP/91VX6n16/KvY7eX9+StXTr5h1jjDE+Ejv8FDav4xNd47HsXHe++5Oq3N8TThA//HlVzl48rcoRcC3iT6nPE6xVH0W+diy2igXSMNxVeZPG4yzPzbhMjVlLYCr4LbpmWuB92JDDBk/y1vU3HNnmYtjTaGf2EfNx9uJZVU6A0HTasjsuliurWuAwKUSft0QKEy3hu7tCtbZ31P6hp89xkSBjYNn3UAcekgVSfD6TWtxAf4vD101ebM7aSBVFYRYlLpMyNQWpQm1gM/2+7IYnSMo5v9Bp6P2u7LxMgTvsqh7jZ7KKHj9+XpWTE33OJWyjLaAdzfqqTTN0vFkiO3MMtCpGslGK17u7ssd29pW84MIe7CxgMY2QaIH+nsCOfX6qOcDA2hpFmCfKUehuMBEny3MzLk/k7yIlLAMGdbCv+62jj5GbWOKE/yxXX3UDWFtd/W0Iy/nde+9U5ShFP0e3dYDHxaU9GC59k85Vlz6Sd4JY5RraPAY6tmY9DZDMcCl76hd/fFCVacvuN4lT6drrDbVbHVGFfP8m1Wg0zJ/95IfGGGPCSP39FGPryZNHVbnb1jhbAOs7m6jPGyBhGazXY6SZoLpMDfUyHyLxC0v8AMjvcLxqgynmiznSPlLYgBO0XYJ+FGDeH8Pa3qgBtUXyQrGGriAFx2gdD2A/Z5oD+3haJujkG0xVcVynSqRwMYFHwI1iYKAG9UPUkmkoHvvCsdrk1w+0b+i3v1+VkxQpXaHKxNPWdg1lypKDekgwny5yJE9hQIdoE83y67g31/01DArrH5OwyFkR+WKCVY6kpOwN/fOh4zgmKLFZJrcQEyqAH3NzxXV0ONPafgKc6sUlUI2exvEtJATe3tb+8/KZ1tpkBuQauNr4+rswQArsGx3se3pIR9o5FDbjIYWPCT5DYFNnQFTmSDxK0DfqSDDa6wP5AqJ80BLqc43a1oDXv67CKDL3Pljht1891/WPTjWfjnD9bRyr0AW+6WP9c3FfNTxbJK72meMzzTGfP/i6Ks+QMHb7PSVz3f5Aa+fu4Wov4gAxdTH+xo815p8/VZ/Yx7XcPFCSUburax9MNXf4QK2ZdtRpClHp7gLpBIIW4lkrQn1wfdm0rhNtuW3i85+Da3IM2mKmtji7EvKyv621ogDy1MZatHegfaGH8wKmSOI0eAbEV5lRmeJZkK0MdPFpyNRkJM8hTW4b95FgfokTJgGq7a5m2gM9ORFOc4T9A/+Wz21BoPpr1Fez+WJBqPZ15Zhrr0eB+TQvuAFUH8vw3Quk3hVItY0i7SfbWxqvW3tq56Cvv71z97tVuRGp/Zvbei4JgO5fHx1BxLXAs1qBBK0E+6kYSNT5iZ5t3EL3tLWt/bgbYO+EIx+4N+D+x+A4hwJs1Xyh9o+B6H5bWSeOlZWVlZWVlZWVlZWVlZWV1Vsg+yOOlZWVlZWVlZWVlZWVlZWV1VugV8KpisKYRWn5dZGI0IR9vVlHssoWyrAh9XZkx2wBeXG3ZJVqdWWJ+zd/KatgfyKb1ej+l1U5HcoWNRrJluWUlv4OTsQe4PTy8cPPqrKfChUJ+7I+pTOlUI0ffFWVY6RTNZAEkcJenbhEd2DBzmGRho12zcZsNq/COCYxpd0MX0BLNq8nwAnsnke7NWz1sIm5sBK2It17p6G2DV3Z6Z49E8qSwh6Zxfqu6zSpwCOSAXscrPa7O7LV7ewh7cylnVrlDFhbDts6Kz+HbZx9x0U790PZKa9tjcYYE3o0j25OhSmqU80z/Babwyq+BNriwsTKE9MfA/O4cXBQldvA2XiqfveGxuUEJ+bTCt67qc/xajhBvqxf4mkRsKYQCBVPjWe/o4W81ZUVOQeuOcmQSMU+g+sdIqHraiB7cwQENAJWUStf3yROVeSFWZZW8POBLNNtWKabLSYq6X7TGccr7h1pXDs4vT9CPYSwHgdbSK+4o3YzSNno7arNa2V7OpnaMIEddAxUz8fp/T4SQcZDzZsmVvtPYD//1ed/qMpPjpT20mzhnkKkh8Gu67j6nH5Lc1On/mZwqloYmnfu3TXGGJPjOjIfKWlAlXLMOQHw0xHS9DzgHEGmcdHtqj+898G9qtwGLrxEslQKBGg0UjuNJqvyeKrXpignGM8ZxpaLRLBFrO9ZwBI+h6U6Q8oYUzWYoOFgPpgg6WySyP4+B2p7nVCWbBADKApZxOMMiWlhtPaea+XAeSMkxc1hh0Y4ozkfIr0Elvl7d4RQvL+vtasHXN1nEijGelCiXiOM+XyB/47lrOOof7hYu5lgt8iYmAKLPNMqgRnw9QRIS4H+kmE/4wDhcMM3MxZd162w3RxoC9E74lTEw5h8SNr9cqy2e3LKeRrpiEi7OdhDUlSkfUE8Uf3Wa3r/NeaS4vsXCcYWcSccG1Drau64WiDNCglWT461zl0gwccgCTDP1d8PdrRP3+1q/8Q+7nbVZ3cPVvfqYw/4uvI837S6q2eE3S2NiZOnQu+PLnS/bSCPPcyzDYzdlOl2juYhYmLvYd9SQ6roV0+0R33xhdDeJ1890TXslc8xwCeWwOcKIMQf7N+oyrf76kP1OpJssF9d4vgBsvohsNU60L41tA17vQDo0lqCYPzmcKprAHQtYXctCVcv+6GuO5nrnp8fAT/r6znisK9nx5tIR2wwTRY8UxOfbxZYg8F25q3yCAsmCYMdz4C0ey20RQd9DXutOZDyM6CYw5HG9xPc35dP1MdP8Uyb4pmFe1APGPn1XttxN+vNYO6gihpn8VL3kmNNvhjqeXqM/cwCz2tMSuw0VIfzocb3FJhTgMTHHHsLJ9R8er1cZqj7FP3JzHXtw0td49W59pkOzivY39U86OOIhcUSxwKgfXI8a81xpMUMZYMUr8Gl9sP+v6LprBPHysrKysrKysrKysrKysrK6i2Q/RHHysrKysrKysrKysrKysrK6i3QK+FUxvGMG66s/otEls05jvd2OkIB6rA5NmE9DRuyPrl4TxDgdVjReweyCv7k7/5dVX4KUuX5b5Ra5T7VyfKt85UljTcawZ58OXlclfsf6uT5FuzM6ZlOlh88+WNVzoH05GvwE5IFYNctkJBA66yBddKDXdotkYXNARwrq2qvu7LAMjHKzXnNSIxx1LYBrZz0atOamTO1TNY3H7a/vR31BTfV+zOc2N5uqf2j+upvmdqzCGS3S3FCe7Ol7wmAsqVIiUqAQaWwJhZrNkW+jtScK9nvAh82876uN4QnrtkgDrM5Oa5rwvrqOxe0CsKxt5gj+QmIg4f7OTuTtf/JY9mDG0gkcnpACtpAq5AUNX2uz8kHmhvqB5oPwjL9qg0r+RDpcBPYj5uwBNdgZzTASQrgNDns/BkQrcLTvRKhevTood6DfkVEpxGp3C4xEs/b3O/eaZZVeF4D6SIF7bCwmzaAVmUYf/kMJ90DR1kumQ7HtAUgknUkVmypbc+fq13mj9Xnu/6qL7jAhmYJvgd11u2p3ZaYgAugpMVS9/fosazEP//0N3oPMIxuU9cYMi0Bp/37wHNuAN2tmTeDNuZFYWZlXcewvg+nWh8GQDKmSPNiCkaUqO1yX3W6X0c/bGM8hbwffS8xpz9++bgqPz+SbXdZJq4wqMeD9b7eErbRbmr99WHhXwtVxBSUIPGKuGqMMcoUlxTj9be/VbtfToSuXFxofrm6XKV2XA2A5b2mClNUiXVpQfRGc2iRASlBXYWYM9xAc+ViTAu5KiieqC88O1MCyV3gv0kao6z6mSLBbF4i7TMkbyyxx/CAT3hYk4inppj7EmA1MfYnGdsWaE+K/7DENbpALTOsqcQlvGBz+A1VFIWJS7s+92SkYB30SeIc7to+Rn1giPZ6fq4+2eVa2ETfEPFr2kCrmQh1CaQ0LnHUlEmZWGcaQKjYL5bnwHVg878AuvDiXHN3spYaBrQLe91bh0Jnt4Ar+9j7tW8LB2qV+313g+uiKYwpSpw3xOcWaM8TYBC7QL+7wD2vsTpjjPGx/yOa0uJREEgh2qohHe7/be/uWtqIgjAA75rUBKWoFC0tiLf9/z+oIkh6oQ0FS2uM2fTC1nkLhVgIyAnPcxVk1eyZ5HwMmcxZ7XM+fah7v46yyNnl09q1ir3Vx+MofT0/f348mcRiOKo4RHPILrbj3eNDrPWx53x7UvvrPt5b+9EtOLvlHcTPx7G/HRbb7GaUqrNRLhB/vRdjvPr4TMEoyk1u51Vu9PmyznYX76v87fQoylujUnN/Wv9sMq14DNkEKNal4eHp76xiYhi/iTVvnLGIkqsoIV3G+nG//Pc8MoturpfX1RX55q72AD9jbs7TZZ4H+1iH9p7HbHsnxn697vZ+388Q56ZFlBINMff38br6GmVCt/H4JN6vhzFuhzG2k5h/v81qfKZD/G58RcdjnCn+DFaWMA73dSb5Ma+x/3JVZ55RdAR8dxZzX3QKXXb1wllEZ9XVOkt066ncRY7ge5Srjw5qnObxfI6P6p5eyidxAAAAABogiQMAAADQgD5LajZe3Pc3XdddbbyQbbtYr9enmy/bTAxflTi2Twx3gzi2Twx3gzi2Twx3gzi2Twx3w4vi+F9JHAAAAABeh3IqAAAAgAZI4gAAAAA0QBIHAAAAoAGSOAAAAAANkMQBAAAAaIAkDgAAAEADJHEAAAAAGiCJAwAAANAASRwAAACABvwC/9ZFwZ3+2U4AAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f5320a1fe10>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# helper display function\n",
    "def imshow(img):\n",
    "    npimg = img.numpy()\n",
    "    plt.imshow(np.transpose(npimg, (1, 2, 0)))\n",
    "\n",
    "\"\"\"\n",
    "DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE\n",
    "\"\"\"\n",
    "# obtain one batch of training images\n",
    "dataiter = iter(celeba_train_loader)\n",
    "images, _ = dataiter.next() # _ for no labels\n",
    "\n",
    "# plot the images in the batch, along with the corresponding labels\n",
    "fig = plt.figure(figsize=(20, 4))\n",
    "plot_size=20\n",
    "for idx in np.arange(plot_size):\n",
    "    ax = fig.add_subplot(2, plot_size/2, idx+1, xticks=[], yticks=[])\n",
    "    imshow(images[idx])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Exercise: Pre-process your image data and scale it to a pixel range of -1 to 1\n",
    "\n",
    "You need to do a bit of pre-processing; you know that the output of a `tanh` activated generator will contain pixel values in a range from -1 to 1, and so, we need to rescale our training images to a range of -1 to 1. (Right now, they are in a range from 0-1.)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "# TODO: Complete the scale function\n",
    "def scale(x, feature_range=(-1, 1)):\n",
    "    ''' Scale takes in an image x and returns that image, scaled\n",
    "       with a feature_range of pixel values from -1 to 1. \n",
    "       This function assumes that the input x is already scaled from 0-1.'''\n",
    "    # assume x is scaled to (0, 1)\n",
    "    # scale to feature_range and return scaled x\n",
    "    min, max = feature_range\n",
    "    x = x * (max - min) + min\n",
    "    return x"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Min:  tensor(-0.7098)\n",
      "Max:  tensor(0.9294)\n"
     ]
    }
   ],
   "source": [
    "\"\"\"\n",
    "DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE\n",
    "\"\"\"\n",
    "# check scaled range\n",
    "# should be close to -1 to 1\n",
    "img = images[0]\n",
    "scaled_img = scale(img)\n",
    "\n",
    "print('Min: ', scaled_img.min())\n",
    "print('Max: ', scaled_img.max())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---\n",
    "# Define the Model\n",
    "\n",
    "A GAN is comprised of two adversarial networks, a discriminator and a generator.\n",
    "\n",
    "## Discriminator\n",
    "\n",
    "Your first task will be to define the discriminator. This is a convolutional classifier like you've built before, only without any maxpooling layers. To deal with this complex data, it's suggested you use a deep network with **normalization**. You are also allowed to create any helper functions that may be useful.\n",
    "\n",
    "#### Exercise: Complete the Discriminator class\n",
    "* The inputs to the discriminator are 32x32x3 tensor images\n",
    "* The output should be a single value that will indicate whether a given image is real or fake\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "import torch.nn as nn\n",
    "import torch.nn.functional as F"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Creating a conv helper function\n",
    "def conv(in_channels, out_channels, kernel_size=4, stride=2, padding=1, batch_norm=True):\n",
    "    layers = []\n",
    "    conv_layer = nn.Conv2d(in_channels, out_channels, kernel_size, stride, padding, bias=False)\n",
    "    \n",
    "    # append conv layer\n",
    "    layers.append(conv_layer)\n",
    "\n",
    "    if batch_norm:\n",
    "        # append batchnorm layer\n",
    "        layers.append(nn.BatchNorm2d(out_channels))\n",
    "     \n",
    "    # using Sequential container\n",
    "    return nn.Sequential(*layers)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Tests Passed\n"
     ]
    }
   ],
   "source": [
    "class Discriminator(nn.Module):\n",
    "\n",
    "    def __init__(self, conv_dim):\n",
    "        \"\"\"\n",
    "        Initialize the Discriminator Module\n",
    "        :param conv_dim: The depth of the first convolutional layer\n",
    "        \"\"\"\n",
    "        super(Discriminator, self).__init__()\n",
    "\n",
    "        # complete init function\n",
    "        self.conv_dim = conv_dim\n",
    "        \n",
    "        # 32x32 input\n",
    "        self.conv1 = conv(3, conv_dim, batch_norm = False)\n",
    "        # 16x16 input\n",
    "        self.conv2 = conv(conv_dim, conv_dim*2)\n",
    "        # 8x8 input\n",
    "        self.conv3 = conv(conv_dim*2, conv_dim*4)\n",
    "        # 4x4 out\n",
    "        \n",
    "        # fully connected layer Image: 4x4 * conv_dim*4\n",
    "        self.fc = nn.Linear(conv_dim*4*4*4, 1)        \n",
    "\n",
    "    def forward(self, x):\n",
    "        \"\"\"\n",
    "        Forward propagation of the neural network\n",
    "        :param x: The input to the neural network     \n",
    "        :return: Discriminator logits; the output of the neural network\n",
    "        \"\"\"\n",
    "        # Applying Relus on the 3 conv layers\n",
    "        x = F.leaky_relu(self.conv1(x), 0.2)\n",
    "        x = F.leaky_relu(self.conv2(x), 0.2)\n",
    "        x = F.leaky_relu(self.conv3(x), 0.2)\n",
    "        \n",
    "        # Flattening\n",
    "        x = x.view(-1, self.conv_dim*4*4*4)\n",
    "        \n",
    "        # FC Layer\n",
    "        x = self.fc(x)\n",
    "        \n",
    "        return x\n",
    "    \n",
    "\"\"\"\n",
    "DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE\n",
    "\"\"\"\n",
    "tests.test_discriminator(Discriminator)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Generator\n",
    "\n",
    "The generator should upsample an input and generate a *new* image of the same size as our training data `32x32x3`. This should be mostly transpose convolutional layers with normalization applied to the outputs.\n",
    "\n",
    "#### Exercise: Complete the Generator class\n",
    "* The inputs to the generator are vectors of some length `z_size`\n",
    "* The output should be a image of shape `32x32x3`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "# deconv helper function\n",
    "def deconv(in_channels, out_channels, kernel_size = 4, stride = 2, padding = 1, batch_norm = True):\n",
    "    # create a sequence of transpose + optional batch norm layers\n",
    "    layers = []\n",
    "    transpose_conv_layer = nn.ConvTranspose2d(in_channels, out_channels, kernel_size, stride, padding, bias=False)\n",
    "    # append transpose convolutional layer\n",
    "    layers.append(transpose_conv_layer)\n",
    "    \n",
    "    if batch_norm:\n",
    "        # append batchnorm layer\n",
    "        layers.append(nn.BatchNorm2d(out_channels))\n",
    "        \n",
    "    return nn.Sequential(*layers)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Tests Passed\n"
     ]
    }
   ],
   "source": [
    "class Generator(nn.Module):\n",
    "    \n",
    "    def __init__(self, z_size, conv_dim):\n",
    "        \"\"\"\n",
    "        Initialize the Generator Module\n",
    "        :param z_size: The length of the input latent vector, z\n",
    "        :param conv_dim: The depth of the inputs to the *last* transpose convolutional layer\n",
    "        \"\"\"\n",
    "        super(Generator, self).__init__()\n",
    "        \n",
    "        self.conv_dim = conv_dim\n",
    "        \n",
    "        # Fully connected layer from z values:\n",
    "        self.fc = nn.Linear(z_size, conv_dim*4*4*4)\n",
    "        \n",
    "        # Transposed convolutional layers with helper function\n",
    "        self.t_conv1 = deconv(conv_dim*4, conv_dim*2)\n",
    "        self.t_conv2 = deconv(conv_dim*2, conv_dim)\n",
    "        self.t_conv3 = deconv(conv_dim, 3, batch_norm=False)\n",
    "\n",
    "    def forward(self, x):\n",
    "        \"\"\"\n",
    "        Forward propagation of the neural network\n",
    "        :param x: The input to the neural network     \n",
    "        :return: A 32x32x3 Tensor image as output\n",
    "        \"\"\"\n",
    "        # FC Layer\n",
    "        x = self.fc(x)\n",
    "        # Resepahing to Tensor \n",
    "        x = x.view(-1, self.conv_dim*4, 4, 4)\n",
    "        \n",
    "        # Transpose conv layers and relu activation\n",
    "        x = F.relu(self.t_conv1(x))\n",
    "        x = F.relu(self.t_conv2(x))\n",
    "        \n",
    "        # Last Layer and tanh activation\n",
    "        x = self.t_conv3(x)\n",
    "        x = F.tanh(x)\n",
    "        \n",
    "        return x \n",
    "    \n",
    "\"\"\"\n",
    "DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE\n",
    "\"\"\"\n",
    "tests.test_generator(Generator)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Initialize the weights of your networks\n",
    "\n",
    "To help your models converge, you should initialize the weights of the convolutional and linear layers in your model. From reading the [original DCGAN paper](https://arxiv.org/pdf/1511.06434.pdf), they say:\n",
    "> All weights were initialized from a zero-centered Normal distribution with standard deviation 0.02.\n",
    "\n",
    "So, your next task will be to define a weight initialization function that does just this!\n",
    "\n",
    "You can refer back to the lesson on weight initialization or even consult existing model code, such as that from [the `networks.py` file in CycleGAN Github repository](https://github.com/junyanz/pytorch-CycleGAN-and-pix2pix/blob/master/models/networks.py) to help you complete this function.\n",
    "\n",
    "#### Exercise: Complete the weight initialization function\n",
    "\n",
    "* This should initialize only **convolutional** and **linear** layers\n",
    "* Initialize the weights to a normal distribution, centered around 0, with a standard deviation of 0.02.\n",
    "* The bias terms, if they exist, may be left alone or set to 0."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "def weights_init_normal(m):\n",
    "    \"\"\"\n",
    "    Applies initial weights to certain layers in a model .\n",
    "    The weights are taken from a normal distribution \n",
    "    with mean = 0, std dev = 0.02.\n",
    "    :param m: A module or layer in a network    \n",
    "    \"\"\"\n",
    "    # classname will be something like:\n",
    "    # `Conv`, `BatchNorm2d`, `Linear`, etc.\n",
    "    classname = m.__class__.__name__\n",
    "    \n",
    "    if hasattr(m, 'weight') and (classname.find('Conv') != -1 or classname.find('Linear') != -1):\n",
    "        m.weight.data.normal_(0, 0.02)\n",
    "        \n",
    "        # Bias term to 0\n",
    "        if hasattr(m, 'bias') and m.bias is not None:\n",
    "            m.bias.data.zero_()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "## Build complete network\n",
    "\n",
    "Define your models' hyperparameters and instantiate the discriminator and generator from the classes defined above. Make sure you've passed in the correct input arguments."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "\"\"\"\n",
    "DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE\n",
    "\"\"\"\n",
    "def build_network(d_conv_dim, g_conv_dim, z_size):\n",
    "    # define discriminator and generator\n",
    "    D = Discriminator(d_conv_dim)\n",
    "    G = Generator(z_size=z_size, conv_dim=g_conv_dim)\n",
    "\n",
    "    # initialize model weights\n",
    "    D.apply(weights_init_normal)\n",
    "    G.apply(weights_init_normal)\n",
    "\n",
    "    print(D)\n",
    "    print()\n",
    "    print(G)\n",
    "    \n",
    "    return D, G\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Exercise: Define model hyperparameters"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Discriminator(\n",
      "  (conv1): Sequential(\n",
      "    (0): Conv2d(3, 64, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1), bias=False)\n",
      "  )\n",
      "  (conv2): Sequential(\n",
      "    (0): Conv2d(64, 128, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1), bias=False)\n",
      "    (1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "  )\n",
      "  (conv3): Sequential(\n",
      "    (0): Conv2d(128, 256, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1), bias=False)\n",
      "    (1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "  )\n",
      "  (fc): Linear(in_features=4096, out_features=1, bias=True)\n",
      ")\n",
      "\n",
      "Generator(\n",
      "  (fc): Linear(in_features=100, out_features=4096, bias=True)\n",
      "  (t_conv1): Sequential(\n",
      "    (0): ConvTranspose2d(256, 128, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1), bias=False)\n",
      "    (1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "  )\n",
      "  (t_conv2): Sequential(\n",
      "    (0): ConvTranspose2d(128, 64, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1), bias=False)\n",
      "    (1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "  )\n",
      "  (t_conv3): Sequential(\n",
      "    (0): ConvTranspose2d(64, 3, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1), bias=False)\n",
      "  )\n",
      ")\n"
     ]
    }
   ],
   "source": [
    "# Define model hyperparams\n",
    "d_conv_dim = 64\n",
    "g_conv_dim = 64\n",
    "z_size = 100\n",
    "\n",
    "\"\"\"\n",
    "DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE\n",
    "\"\"\"\n",
    "D, G = build_network(d_conv_dim, g_conv_dim, z_size)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Training on GPU\n",
    "\n",
    "Check if you can train on GPU. Here, we'll set this as a boolean variable `train_on_gpu`. Later, you'll be responsible for making sure that \n",
    ">* Models,\n",
    "* Model inputs, and\n",
    "* Loss function arguments\n",
    "\n",
    "Are moved to GPU, where appropriate."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training on GPU!\n"
     ]
    }
   ],
   "source": [
    "\"\"\"\n",
    "DON'T MODIFY ANYTHING IN THIS CELL\n",
    "\"\"\"\n",
    "import torch\n",
    "\n",
    "# Check for a GPU\n",
    "train_on_gpu = torch.cuda.is_available()\n",
    "if not train_on_gpu:\n",
    "    print('No GPU found. Please use a GPU to train your neural network.')\n",
    "else:\n",
    "    print('Training on GPU!')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---\n",
    "## Discriminator and Generator Losses\n",
    "\n",
    "Now we need to calculate the losses for both types of adversarial networks.\n",
    "\n",
    "### Discriminator Losses\n",
    "\n",
    "> * For the discriminator, the total loss is the sum of the losses for real and fake images, `d_loss = d_real_loss + d_fake_loss`. \n",
    "* Remember that we want the discriminator to output 1 for real images and 0 for fake images, so we need to set up the losses to reflect that.\n",
    "\n",
    "\n",
    "### Generator Loss\n",
    "\n",
    "The generator loss will look similar only with flipped labels. The generator's goal is to get the discriminator to *think* its generated images are *real*.\n",
    "\n",
    "#### Exercise: Complete real and fake loss functions\n",
    "\n",
    "**You may choose to use either cross entropy or a least squares error loss to complete the following `real_loss` and `fake_loss` functions.**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "def real_loss(D_out):\n",
    "    '''Calculates how close discriminator outputs are to being real.\n",
    "       param, D_out: discriminator logits\n",
    "       return: real loss'''\n",
    "    batch_size = D_out.size(0)\n",
    "    labels = torch.ones(batch_size)\n",
    "    \n",
    "    # move labels to GPU if available     \n",
    "    if train_on_gpu:\n",
    "        labels = labels.cuda()\n",
    "        \n",
    "    # binary cross entropy with logits loss\n",
    "    criterion = nn.BCEWithLogitsLoss()\n",
    "    # calculate loss\n",
    "    loss = criterion(D_out.squeeze(), labels)\n",
    "    return loss\n",
    "\n",
    "def fake_loss(D_out):\n",
    "    '''Calculates how close discriminator outputs are to being fake.\n",
    "       param, D_out: discriminator logits\n",
    "       return: fake loss'''\n",
    "    batch_size = D_out.size(0)\n",
    "    labels = torch.zeros(batch_size)\n",
    "    \n",
    "    if train_on_gpu:\n",
    "        labels = labels.cuda()\n",
    "    \n",
    "    criterion = nn.BCEWithLogitsLoss()\n",
    "    # calculate loss\n",
    "    loss = criterion(D_out.squeeze(), labels)\n",
    "    return loss"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Optimizers\n",
    "\n",
    "#### Exercise: Define optimizers for your Discriminator (D) and Generator (G)\n",
    "\n",
    "Define optimizers for your models with appropriate hyperparameters."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "import torch.optim as optim\n",
    "\n",
    "# params\n",
    "lr = 0.0002\n",
    "beta1 = 0.5\n",
    "beta2 = 0.999\n",
    "\n",
    "# Create optimizers for the discriminator D and generator G\n",
    "d_optimizer = optim.Adam(D.parameters(), lr, [beta1, beta2])\n",
    "g_optimizer = optim.Adam(G.parameters(), lr, [beta1, beta2])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---\n",
    "## Training\n",
    "\n",
    "Training will involve alternating between training the discriminator and the generator. You'll use your functions `real_loss` and `fake_loss` to help you calculate the discriminator losses.\n",
    "\n",
    "* You should train the discriminator by alternating on real and fake images\n",
    "* Then the generator, which tries to trick the discriminator and should have an opposing loss function\n",
    "\n",
    "\n",
    "#### Saving Samples\n",
    "\n",
    "You've been given some code to print out some loss statistics and save some generated \"fake\" samples."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Exercise: Complete the training function\n",
    "\n",
    "Keep in mind that, if you've moved your models to GPU, you'll also have to move any model inputs to GPU."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "def train(D, G, n_epochs, print_every=50):\n",
    "    '''Trains adversarial networks for some number of epochs\n",
    "       param, D: the discriminator network\n",
    "       param, G: the generator network\n",
    "       param, n_epochs: number of epochs to train for\n",
    "       param, print_every: when to print and record the models' losses\n",
    "       return: D and G losses'''\n",
    "    \n",
    "    # move models to GPU\n",
    "    if train_on_gpu:\n",
    "        D.cuda()\n",
    "        G.cuda()\n",
    "\n",
    "    # keep track of loss and generated, \"fake\" samples\n",
    "    samples = []\n",
    "    losses = []\n",
    "\n",
    "    # Get some fixed data for sampling. These are images that are held\n",
    "    # constant throughout training, and allow us to inspect the model's performance\n",
    "    sample_size=16\n",
    "    fixed_z = np.random.uniform(-1, 1, size=(sample_size, z_size))\n",
    "    fixed_z = torch.from_numpy(fixed_z).float()\n",
    "    # move z to GPU if available\n",
    "    if train_on_gpu:\n",
    "        fixed_z = fixed_z.cuda()\n",
    "\n",
    "    # epoch training loop\n",
    "    for epoch in range(n_epochs):\n",
    "\n",
    "        # batch training loop\n",
    "        for batch_i, (real_images, _) in enumerate(celeba_train_loader):\n",
    "\n",
    "            batch_size = real_images.size(0)\n",
    "            real_images = scale(real_images)\n",
    "\n",
    "            # ===============================================\n",
    "            #         YOUR CODE HERE: TRAIN THE NETWORKS\n",
    "            # ===============================================\n",
    "            \n",
    "            # 1. Train the discriminator on real and fake images\n",
    "            \n",
    "            # Set grad to zero \n",
    "            d_optimizer.zero_grad()\n",
    "            \n",
    "            # Train on GPU\n",
    "            if train_on_gpu:\n",
    "                real_images = real_images.cuda()\n",
    "            \n",
    "            # Apply Discriminator on the real images\n",
    "            D_real = D(real_images)\n",
    "            d_real_loss = real_loss(D_real)\n",
    "            \n",
    "            # Generate Fake images\n",
    "            z = np.random.uniform(-1, 1, size=(batch_size, z_size))\n",
    "            z = torch.from_numpy(z).float()\n",
    "            \n",
    "            # Move to GPU \n",
    "            if train_on_gpu:\n",
    "                z = z.cuda()\n",
    "            \n",
    "            # Apply Generator on z to generate fake images\n",
    "            fake_images = G(z)\n",
    "            \n",
    "            # Running the fake images on the Discriminator\n",
    "            D_fake = D(fake_images)\n",
    "            d_fake_loss = fake_loss(D_fake)\n",
    "            \n",
    "            d_loss = d_real_loss + d_fake_loss\n",
    "            d_loss.backward()\n",
    "            d_optimizer.step()\n",
    "\n",
    "            \n",
    "            # 2. Train the generator with an adversarial loss\n",
    "            \n",
    "            # Set the Gradient to zero\n",
    "            g_optimizer.zero_grad()\n",
    "            \n",
    "            # generate z\n",
    "            z = np.random.uniform(-1, 1, size=(batch_size, z_size))\n",
    "            z = torch.from_numpy(z).float()\n",
    "            \n",
    "            # Move to GPU\n",
    "            if train_on_gpu:\n",
    "                z = z.cuda()\n",
    "            \n",
    "            # create fake immages with the Generator\n",
    "            fake_images = G(z)\n",
    "            \n",
    "            # Apply Discriminator and calculate loss\n",
    "            D_fake = D(fake_images)\n",
    "            g_loss = real_loss(D_fake)\n",
    "            \n",
    "            # back-prop and optimizer\n",
    "            g_loss.backward()\n",
    "            g_optimizer.step()            \n",
    "            \n",
    "            # ===============================================\n",
    "            #              END OF YOUR CODE\n",
    "            # ===============================================\n",
    "\n",
    "            # Print some loss stats\n",
    "            if batch_i % print_every == 0:\n",
    "                # append discriminator loss and generator loss\n",
    "                losses.append((d_loss.item(), g_loss.item()))\n",
    "                # print discriminator and generator loss\n",
    "                print('Epoch [{:5d}/{:5d}] | d_loss: {:6.4f} | g_loss: {:6.4f}'.format(\n",
    "                        epoch+1, n_epochs, d_loss.item(), g_loss.item()))\n",
    "\n",
    "\n",
    "        ## AFTER EACH EPOCH##    \n",
    "        # this code assumes your generator is named G, feel free to change the name\n",
    "        # generate and save sample, fake images\n",
    "        G.eval() # for generating samples\n",
    "        samples_z = G(fixed_z)\n",
    "        samples.append(samples_z)\n",
    "        G.train() # back to training mode\n",
    "\n",
    "    # Save training generator samples\n",
    "    with open('train_samples.pkl', 'wb') as f:\n",
    "        pkl.dump(samples, f)\n",
    "    \n",
    "    # finally return losses\n",
    "    return losses"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Set your number of training epochs and train your GAN!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch [    1/   10] | d_loss: 1.3848 | g_loss: 1.0317\n",
      "Epoch [    1/   10] | d_loss: 0.0865 | g_loss: 4.0679\n",
      "Epoch [    1/   10] | d_loss: 0.2049 | g_loss: 3.2902\n",
      "Epoch [    1/   10] | d_loss: 0.3237 | g_loss: 3.4050\n",
      "Epoch [    1/   10] | d_loss: 0.5329 | g_loss: 3.7378\n",
      "Epoch [    1/   10] | d_loss: 2.0889 | g_loss: 4.2348\n",
      "Epoch [    1/   10] | d_loss: 0.4855 | g_loss: 1.8386\n",
      "Epoch [    1/   10] | d_loss: 0.7826 | g_loss: 1.5993\n",
      "Epoch [    1/   10] | d_loss: 0.7872 | g_loss: 1.5260\n",
      "Epoch [    1/   10] | d_loss: 0.6386 | g_loss: 2.4055\n",
      "Epoch [    1/   10] | d_loss: 0.7807 | g_loss: 1.4434\n",
      "Epoch [    1/   10] | d_loss: 0.8968 | g_loss: 0.7620\n",
      "Epoch [    1/   10] | d_loss: 0.9718 | g_loss: 1.6253\n",
      "Epoch [    1/   10] | d_loss: 0.9436 | g_loss: 1.3210\n",
      "Epoch [    1/   10] | d_loss: 1.0262 | g_loss: 0.9289\n",
      "Epoch [    2/   10] | d_loss: 0.9623 | g_loss: 1.2973\n",
      "Epoch [    2/   10] | d_loss: 0.9433 | g_loss: 0.9757\n",
      "Epoch [    2/   10] | d_loss: 0.9030 | g_loss: 1.1766\n",
      "Epoch [    2/   10] | d_loss: 1.1302 | g_loss: 1.5276\n",
      "Epoch [    2/   10] | d_loss: 1.1149 | g_loss: 1.1764\n",
      "Epoch [    2/   10] | d_loss: 1.1557 | g_loss: 1.4719\n",
      "Epoch [    2/   10] | d_loss: 0.8660 | g_loss: 1.3539\n",
      "Epoch [    2/   10] | d_loss: 1.0941 | g_loss: 1.4021\n",
      "Epoch [    2/   10] | d_loss: 1.0820 | g_loss: 1.1755\n",
      "Epoch [    2/   10] | d_loss: 1.1008 | g_loss: 0.9431\n",
      "Epoch [    2/   10] | d_loss: 1.0564 | g_loss: 1.4423\n",
      "Epoch [    2/   10] | d_loss: 1.1263 | g_loss: 1.4151\n",
      "Epoch [    2/   10] | d_loss: 1.0669 | g_loss: 1.6489\n",
      "Epoch [    2/   10] | d_loss: 1.1321 | g_loss: 1.0313\n",
      "Epoch [    2/   10] | d_loss: 1.0512 | g_loss: 1.5004\n",
      "Epoch [    3/   10] | d_loss: 0.9489 | g_loss: 1.5805\n",
      "Epoch [    3/   10] | d_loss: 0.9582 | g_loss: 1.3813\n",
      "Epoch [    3/   10] | d_loss: 1.2932 | g_loss: 1.6055\n",
      "Epoch [    3/   10] | d_loss: 1.1473 | g_loss: 0.9118\n",
      "Epoch [    3/   10] | d_loss: 1.1042 | g_loss: 0.9264\n",
      "Epoch [    3/   10] | d_loss: 1.1654 | g_loss: 0.6028\n",
      "Epoch [    3/   10] | d_loss: 1.0934 | g_loss: 0.8811\n",
      "Epoch [    3/   10] | d_loss: 0.9749 | g_loss: 1.2898\n",
      "Epoch [    3/   10] | d_loss: 1.1713 | g_loss: 0.8371\n",
      "Epoch [    3/   10] | d_loss: 0.9626 | g_loss: 1.0160\n",
      "Epoch [    3/   10] | d_loss: 0.8344 | g_loss: 1.5120\n",
      "Epoch [    3/   10] | d_loss: 1.1922 | g_loss: 1.6833\n",
      "Epoch [    3/   10] | d_loss: 1.3811 | g_loss: 0.5569\n",
      "Epoch [    3/   10] | d_loss: 1.1223 | g_loss: 1.1463\n",
      "Epoch [    3/   10] | d_loss: 1.0987 | g_loss: 1.4966\n",
      "Epoch [    4/   10] | d_loss: 1.0906 | g_loss: 0.8383\n",
      "Epoch [    4/   10] | d_loss: 1.0489 | g_loss: 1.6813\n",
      "Epoch [    4/   10] | d_loss: 0.9405 | g_loss: 1.2938\n",
      "Epoch [    4/   10] | d_loss: 0.8860 | g_loss: 1.0859\n",
      "Epoch [    4/   10] | d_loss: 1.2250 | g_loss: 0.7602\n",
      "Epoch [    4/   10] | d_loss: 1.0216 | g_loss: 1.0612\n",
      "Epoch [    4/   10] | d_loss: 0.9219 | g_loss: 1.5956\n",
      "Epoch [    4/   10] | d_loss: 1.0511 | g_loss: 0.7719\n",
      "Epoch [    4/   10] | d_loss: 0.9834 | g_loss: 0.7999\n",
      "Epoch [    4/   10] | d_loss: 1.4256 | g_loss: 0.4232\n",
      "Epoch [    4/   10] | d_loss: 1.1274 | g_loss: 1.7267\n",
      "Epoch [    4/   10] | d_loss: 0.9195 | g_loss: 2.0237\n",
      "Epoch [    4/   10] | d_loss: 1.1150 | g_loss: 0.7380\n",
      "Epoch [    4/   10] | d_loss: 1.1390 | g_loss: 0.5704\n",
      "Epoch [    4/   10] | d_loss: 0.9500 | g_loss: 0.8966\n",
      "Epoch [    5/   10] | d_loss: 0.9530 | g_loss: 1.5228\n",
      "Epoch [    5/   10] | d_loss: 1.0534 | g_loss: 1.5873\n",
      "Epoch [    5/   10] | d_loss: 1.3852 | g_loss: 0.4986\n",
      "Epoch [    5/   10] | d_loss: 0.9195 | g_loss: 1.4738\n",
      "Epoch [    5/   10] | d_loss: 1.0895 | g_loss: 0.9799\n",
      "Epoch [    5/   10] | d_loss: 0.7773 | g_loss: 1.4730\n",
      "Epoch [    5/   10] | d_loss: 0.9868 | g_loss: 1.5019\n",
      "Epoch [    5/   10] | d_loss: 0.8869 | g_loss: 1.0780\n",
      "Epoch [    5/   10] | d_loss: 0.9954 | g_loss: 1.1470\n",
      "Epoch [    5/   10] | d_loss: 0.9349 | g_loss: 1.1140\n",
      "Epoch [    5/   10] | d_loss: 1.0955 | g_loss: 1.1504\n",
      "Epoch [    5/   10] | d_loss: 0.9404 | g_loss: 1.0557\n",
      "Epoch [    5/   10] | d_loss: 1.0428 | g_loss: 1.3134\n",
      "Epoch [    5/   10] | d_loss: 0.8455 | g_loss: 1.2000\n",
      "Epoch [    5/   10] | d_loss: 1.1956 | g_loss: 0.7354\n",
      "Epoch [    6/   10] | d_loss: 0.8416 | g_loss: 1.1681\n",
      "Epoch [    6/   10] | d_loss: 1.0153 | g_loss: 0.8833\n",
      "Epoch [    6/   10] | d_loss: 1.5240 | g_loss: 2.8041\n",
      "Epoch [    6/   10] | d_loss: 0.7251 | g_loss: 1.7810\n",
      "Epoch [    6/   10] | d_loss: 1.0628 | g_loss: 0.9683\n",
      "Epoch [    6/   10] | d_loss: 0.9565 | g_loss: 1.1104\n",
      "Epoch [    6/   10] | d_loss: 1.1749 | g_loss: 1.7561\n",
      "Epoch [    6/   10] | d_loss: 1.0282 | g_loss: 1.7698\n",
      "Epoch [    6/   10] | d_loss: 1.0964 | g_loss: 1.2815\n",
      "Epoch [    6/   10] | d_loss: 1.1126 | g_loss: 0.9641\n",
      "Epoch [    6/   10] | d_loss: 1.4566 | g_loss: 1.9784\n",
      "Epoch [    6/   10] | d_loss: 1.0194 | g_loss: 0.8733\n",
      "Epoch [    6/   10] | d_loss: 0.9961 | g_loss: 1.0358\n",
      "Epoch [    6/   10] | d_loss: 1.6942 | g_loss: 0.6681\n",
      "Epoch [    6/   10] | d_loss: 0.7524 | g_loss: 2.1149\n",
      "Epoch [    7/   10] | d_loss: 0.9397 | g_loss: 1.5187\n",
      "Epoch [    7/   10] | d_loss: 1.0464 | g_loss: 1.3926\n",
      "Epoch [    7/   10] | d_loss: 0.8807 | g_loss: 1.3250\n",
      "Epoch [    7/   10] | d_loss: 0.8966 | g_loss: 1.6015\n",
      "Epoch [    7/   10] | d_loss: 0.9429 | g_loss: 1.1847\n",
      "Epoch [    7/   10] | d_loss: 0.8048 | g_loss: 1.6319\n",
      "Epoch [    7/   10] | d_loss: 0.9552 | g_loss: 1.3203\n",
      "Epoch [    7/   10] | d_loss: 1.0162 | g_loss: 0.9360\n",
      "Epoch [    7/   10] | d_loss: 0.9814 | g_loss: 1.3738\n",
      "Epoch [    7/   10] | d_loss: 0.9114 | g_loss: 1.9042\n",
      "Epoch [    7/   10] | d_loss: 1.0784 | g_loss: 0.8970\n",
      "Epoch [    7/   10] | d_loss: 0.7878 | g_loss: 1.2650\n",
      "Epoch [    7/   10] | d_loss: 1.0364 | g_loss: 0.8628\n",
      "Epoch [    7/   10] | d_loss: 0.9953 | g_loss: 1.3935\n",
      "Epoch [    7/   10] | d_loss: 1.2202 | g_loss: 0.7879\n",
      "Epoch [    8/   10] | d_loss: 1.1796 | g_loss: 1.2244\n",
      "Epoch [    8/   10] | d_loss: 1.1412 | g_loss: 0.9225\n",
      "Epoch [    8/   10] | d_loss: 0.9066 | g_loss: 1.3896\n",
      "Epoch [    8/   10] | d_loss: 1.0572 | g_loss: 0.7858\n",
      "Epoch [    8/   10] | d_loss: 1.0933 | g_loss: 0.9334\n",
      "Epoch [    8/   10] | d_loss: 1.3175 | g_loss: 0.9971\n",
      "Epoch [    8/   10] | d_loss: 1.0659 | g_loss: 1.0750\n",
      "Epoch [    8/   10] | d_loss: 0.9292 | g_loss: 1.6820\n",
      "Epoch [    8/   10] | d_loss: 1.3219 | g_loss: 1.0629\n",
      "Epoch [    8/   10] | d_loss: 0.9993 | g_loss: 1.3302\n",
      "Epoch [    8/   10] | d_loss: 0.9065 | g_loss: 1.3843\n",
      "Epoch [    8/   10] | d_loss: 0.9728 | g_loss: 1.1199\n",
      "Epoch [    8/   10] | d_loss: 0.9195 | g_loss: 1.7747\n",
      "Epoch [    8/   10] | d_loss: 0.7410 | g_loss: 1.6200\n",
      "Epoch [    8/   10] | d_loss: 0.9098 | g_loss: 0.7428\n",
      "Epoch [    9/   10] | d_loss: 1.7157 | g_loss: 3.1344\n",
      "Epoch [    9/   10] | d_loss: 0.9292 | g_loss: 1.3809\n",
      "Epoch [    9/   10] | d_loss: 0.9104 | g_loss: 0.9060\n",
      "Epoch [    9/   10] | d_loss: 0.8020 | g_loss: 1.3174\n",
      "Epoch [    9/   10] | d_loss: 0.9076 | g_loss: 1.0880\n",
      "Epoch [    9/   10] | d_loss: 0.9653 | g_loss: 0.7784\n",
      "Epoch [    9/   10] | d_loss: 0.7974 | g_loss: 1.1021\n",
      "Epoch [    9/   10] | d_loss: 0.9970 | g_loss: 0.9002\n",
      "Epoch [    9/   10] | d_loss: 0.9096 | g_loss: 1.2509\n",
      "Epoch [    9/   10] | d_loss: 1.0104 | g_loss: 1.6557\n",
      "Epoch [    9/   10] | d_loss: 1.0152 | g_loss: 0.9667\n",
      "Epoch [    9/   10] | d_loss: 1.0969 | g_loss: 1.6327\n",
      "Epoch [    9/   10] | d_loss: 0.9415 | g_loss: 1.3703\n",
      "Epoch [    9/   10] | d_loss: 0.8479 | g_loss: 2.1367\n",
      "Epoch [    9/   10] | d_loss: 1.1096 | g_loss: 0.8890\n",
      "Epoch [   10/   10] | d_loss: 0.8526 | g_loss: 1.3131\n",
      "Epoch [   10/   10] | d_loss: 1.0300 | g_loss: 0.6714\n",
      "Epoch [   10/   10] | d_loss: 0.8223 | g_loss: 1.0999\n",
      "Epoch [   10/   10] | d_loss: 0.9820 | g_loss: 1.7601\n",
      "Epoch [   10/   10] | d_loss: 0.8686 | g_loss: 1.7095\n",
      "Epoch [   10/   10] | d_loss: 0.9627 | g_loss: 0.8961\n",
      "Epoch [   10/   10] | d_loss: 1.0138 | g_loss: 1.6701\n",
      "Epoch [   10/   10] | d_loss: 0.8379 | g_loss: 1.3998\n",
      "Epoch [   10/   10] | d_loss: 1.0188 | g_loss: 1.0016\n",
      "Epoch [   10/   10] | d_loss: 1.0127 | g_loss: 0.8766\n",
      "Epoch [   10/   10] | d_loss: 0.9614 | g_loss: 1.4913\n",
      "Epoch [   10/   10] | d_loss: 0.8748 | g_loss: 1.2173\n",
      "Epoch [   10/   10] | d_loss: 0.9916 | g_loss: 1.5193\n",
      "Epoch [   10/   10] | d_loss: 0.8429 | g_loss: 1.6373\n",
      "Epoch [   10/   10] | d_loss: 0.9534 | g_loss: 0.9294\n"
     ]
    }
   ],
   "source": [
    "# set number of epochs \n",
    "n_epochs = 10\n",
    "\n",
    "\n",
    "\"\"\"\n",
    "DON'T MODIFY ANYTHING IN THIS CELL\n",
    "\"\"\"\n",
    "# call training function\n",
    "losses = train(D, G, n_epochs=n_epochs)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "## Training loss\n",
    "\n",
    "Plot the training losses for the generator and discriminator, recorded after each epoch."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x7f531c3e6978>"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f531cd8d978>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots()\n",
    "losses = np.array(losses)\n",
    "plt.plot(losses.T[0], label='Discriminator', alpha=0.5)\n",
    "plt.plot(losses.T[1], label='Generator', alpha=0.5)\n",
    "plt.title(\"Training Losses\")\n",
    "plt.legend()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "## Generator samples from training\n",
    "\n",
    "View samples of images from the generator, and answer a question about the strengths and weaknesses of your trained models."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "# helper function for viewing a list of passed in sample images\n",
    "def view_samples(epoch, samples):\n",
    "    fig, axes = plt.subplots(figsize=(16,4), nrows=2, ncols=8, sharey=True, sharex=True)\n",
    "    for ax, img in zip(axes.flatten(), samples[epoch]):\n",
    "        img = img.detach().cpu().numpy()\n",
    "        img = np.transpose(img, (1, 2, 0))\n",
    "        img = ((img + 1)*255 / (2)).astype(np.uint8)\n",
    "        ax.xaxis.set_visible(False)\n",
    "        ax.yaxis.set_visible(False)\n",
    "        im = ax.imshow(img.reshape((32,32,3)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Load samples from generator, taken while training\n",
    "with open('train_samples.pkl', 'rb') as f:\n",
    "    samples = pkl.load(f)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f5320a1f710>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "_ = view_samples(-1, samples)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Question: What do you notice about your generated samples and how might you improve this model?\n",
    "When you answer this question, consider the following factors:\n",
    "* The dataset is biased; it is made of \"celebrity\" faces that are mostly white\n",
    "* Model size; larger models have the opportunity to learn more features in a data feature space\n",
    "* Optimization strategy; optimizers and number of epochs affect your final result\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Answer:**\n",
    "\n",
    "I am quite surpised by the output of the network. Most of the images look like true faces. I do see some slight glithces, even though those could still be real people.\n",
    "\n",
    "All my faces in that sample seem to be good in detecting the features of a normal face, hence my convolutional network is doing a fine job in feature detection.\n",
    "\n",
    "It seems to only clash when it is about face vs. background and face vs. hair/hat. That seems to be the areas for error. \n",
    "\n",
    "Obvisouyly, we could deeopen the network by adding another layer or adjusting the depth of the layers. Since the feature detection is good, I would be inclined to rather work on changing the training to more epochs, increase the `batch_size` to 256, I also could try out a smaller learning rate as well as smoothing."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Submitting This Project\n",
    "When submitting this project, make sure to run all the cells before saving the notebook. Save the notebook file as \"dlnd_face_generation.ipynb\" and save it as a HTML file under \"File\" -> \"Download as\". Include the \"problem_unittests.py\" files in your submission."
   ]
  }
 ],
 "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": "3.6.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}