microsoft/AI-For-Beginners

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Pre-trained models and transfer learning\n",
    "\n",
    "Training CNNs can take a lot of time, and a lot of data is required for that task. However, much of the time is spent to learn the best low-level filters that a network is using to extract patterns from images. A natural question arises - can we use a neural network trained on one dataset and adapt it to classifying different images without full training process?\n",
    "\n",
    "This approach is called **transfer learning**, because we transfer some knowledge from one neural network model to another. In transfer learning, we typically start with a pre-trained model, which has been trained on some large image dataset, such as **ImageNet**. Those models can already do a good job extracting different features from generic images, and in many cases just building a classifier on top of those extracted features can yield a good result."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import torch\n",
    "import torch.nn as nn\n",
    "import torchvision\n",
    "import torchvision.transforms as transforms\n",
    "import matplotlib.pyplot as plt\n",
    "from torchinfo import summary\n",
    "import numpy as np\n",
    "import os\n",
    "\n",
    "from pytorchcv import train, plot_results, display_dataset, train_long, check_image_dir"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Cats vs. Dogs Dataset\n",
    "\n",
    "In this unit, we will solve a real-life problem of classifying images of cats and dogs. For this reason, we will use [Kaggle Cats vs. Dogs Dataset](https://www.kaggle.com/c/dogs-vs-cats), which can also be downloaded [from Microsoft](https://www.microsoft.com/en-us/download/details.aspx?id=54765).\n",
    "\n",
    "Let's download this dataset and extract it into `data` directory (this process may take some time!):"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "if not os.path.exists('data/kagglecatsanddogs_5340.zip'):\n",
    "    !wget -P data https://download.microsoft.com/download/3/E/1/3E1C3F21-ECDB-4869-8368-6DEBA77B919F/kagglecatsanddogs_5340.zip"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "import zipfile\n",
    "if not os.path.exists('data/PetImages'):\n",
    "    with zipfile.ZipFile('data/kagglecatsanddogs_5340.zip', 'r') as zip_ref:\n",
    "        zip_ref.extractall('data')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Unfortunately, there are some corrupt image files in the dataset. We need to do quick cleaning to check for corrupted files. In order not to clobber this tutorial, we moved the code to verify dataset into a module."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Corrupt image: data/PetImages/Cat/666.jpg\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/anaconda/envs/py38_pytorch/lib/python3.8/site-packages/PIL/TiffImagePlugin.py:793: UserWarning: Truncated File Read\n",
      "  warnings.warn(str(msg))\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Corrupt image: data/PetImages/Dog/11702.jpg\n"
     ]
    }
   ],
   "source": [
    "check_image_dir('data/PetImages/Cat/*.jpg')\n",
    "check_image_dir('data/PetImages/Dog/*.jpg')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Next, let's load the images into PyTorch dataset, converting them to tensors and doing some normalization. We will apply `std_normalize` transform to bring images to the range expected by pre-trained VGG network:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1080x216 with 10 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "std_normalize = transforms.Normalize(mean=[0.485, 0.456, 0.406],\n",
    "                          std=[0.229, 0.224, 0.225])\n",
    "trans = transforms.Compose([\n",
    "        transforms.Resize(256),\n",
    "        transforms.CenterCrop(224),\n",
    "        transforms.ToTensor(), \n",
    "        std_normalize])\n",
    "dataset = torchvision.datasets.ImageFolder('data/PetImages',transform=trans)\n",
    "trainset, testset = torch.utils.data.random_split(dataset,[20000,len(dataset)-20000])\n",
    "\n",
    "display_dataset(dataset)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Pre-trained models\n",
    "\n",
    "There are many different pre-trained models available inside `torchvision` module, and even more models can be found on the Internet. Let's see how simplest VGG-16 model can be loaded and used:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Downloading: \"https://download.pytorch.org/models/vgg16-397923af.pth\" to /home/cathy/.cache/torch/hub/checkpoints/vgg16-397923af.pth\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "16e2fa013a884411acaf74d836bc8f65",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0.00/528M [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "tensor(282)\n"
     ]
    }
   ],
   "source": [
    "vgg = torchvision.models.vgg16(pretrained=True)\n",
    "sample_image = dataset[0][0].unsqueeze(0)\n",
    "res = vgg(sample_image)\n",
    "print(res[0].argmax())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The result that we have received is a number of an `ImageNet` class, which can be looked up [here](https://gist.github.com/yrevar/942d3a0ac09ec9e5eb3a). We can use the following code to automatically load this class table and return the result:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['n02123159', 'tiger_cat']"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import json, requests\n",
    "class_map = json.loads(requests.get(\"https://s3.amazonaws.com/deep-learning-models/image-models/imagenet_class_index.json\").text)\n",
    "class_map = { int(k) : v for k,v in class_map.items() }\n",
    "\n",
    "class_map[res[0].argmax().item()]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's also see the architecture of the VGG-16 network:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "==========================================================================================\n",
       "Layer (type:depth-idx)                   Output Shape              Param #\n",
       "==========================================================================================\n",
       "VGG                                      --                        --\n",
       "├─Sequential: 1-1                        [1, 512, 7, 7]            --\n",
       "│    └─Conv2d: 2-1                       [1, 64, 224, 224]         1,792\n",
       "│    └─ReLU: 2-2                         [1, 64, 224, 224]         --\n",
       "│    └─Conv2d: 2-3                       [1, 64, 224, 224]         36,928\n",
       "│    └─ReLU: 2-4                         [1, 64, 224, 224]         --\n",
       "│    └─MaxPool2d: 2-5                    [1, 64, 112, 112]         --\n",
       "│    └─Conv2d: 2-6                       [1, 128, 112, 112]        73,856\n",
       "│    └─ReLU: 2-7                         [1, 128, 112, 112]        --\n",
       "│    └─Conv2d: 2-8                       [1, 128, 112, 112]        147,584\n",
       "│    └─ReLU: 2-9                         [1, 128, 112, 112]        --\n",
       "│    └─MaxPool2d: 2-10                   [1, 128, 56, 56]          --\n",
       "│    └─Conv2d: 2-11                      [1, 256, 56, 56]          295,168\n",
       "│    └─ReLU: 2-12                        [1, 256, 56, 56]          --\n",
       "│    └─Conv2d: 2-13                      [1, 256, 56, 56]          590,080\n",
       "│    └─ReLU: 2-14                        [1, 256, 56, 56]          --\n",
       "│    └─Conv2d: 2-15                      [1, 256, 56, 56]          590,080\n",
       "│    └─ReLU: 2-16                        [1, 256, 56, 56]          --\n",
       "│    └─MaxPool2d: 2-17                   [1, 256, 28, 28]          --\n",
       "│    └─Conv2d: 2-18                      [1, 512, 28, 28]          1,180,160\n",
       "│    └─ReLU: 2-19                        [1, 512, 28, 28]          --\n",
       "│    └─Conv2d: 2-20                      [1, 512, 28, 28]          2,359,808\n",
       "│    └─ReLU: 2-21                        [1, 512, 28, 28]          --\n",
       "│    └─Conv2d: 2-22                      [1, 512, 28, 28]          2,359,808\n",
       "│    └─ReLU: 2-23                        [1, 512, 28, 28]          --\n",
       "│    └─MaxPool2d: 2-24                   [1, 512, 14, 14]          --\n",
       "│    └─Conv2d: 2-25                      [1, 512, 14, 14]          2,359,808\n",
       "│    └─ReLU: 2-26                        [1, 512, 14, 14]          --\n",
       "│    └─Conv2d: 2-27                      [1, 512, 14, 14]          2,359,808\n",
       "│    └─ReLU: 2-28                        [1, 512, 14, 14]          --\n",
       "│    └─Conv2d: 2-29                      [1, 512, 14, 14]          2,359,808\n",
       "│    └─ReLU: 2-30                        [1, 512, 14, 14]          --\n",
       "│    └─MaxPool2d: 2-31                   [1, 512, 7, 7]            --\n",
       "├─AdaptiveAvgPool2d: 1-2                 [1, 512, 7, 7]            --\n",
       "├─Sequential: 1-3                        [1, 1000]                 --\n",
       "│    └─Linear: 2-32                      [1, 4096]                 102,764,544\n",
       "│    └─ReLU: 2-33                        [1, 4096]                 --\n",
       "│    └─Dropout: 2-34                     [1, 4096]                 --\n",
       "│    └─Linear: 2-35                      [1, 4096]                 16,781,312\n",
       "│    └─ReLU: 2-36                        [1, 4096]                 --\n",
       "│    └─Dropout: 2-37                     [1, 4096]                 --\n",
       "│    └─Linear: 2-38                      [1, 1000]                 4,097,000\n",
       "==========================================================================================\n",
       "Total params: 138,357,544\n",
       "Trainable params: 138,357,544\n",
       "Non-trainable params: 0\n",
       "Total mult-adds (G): 15.48\n",
       "==========================================================================================\n",
       "Input size (MB): 0.60\n",
       "Forward/backward pass size (MB): 108.45\n",
       "Params size (MB): 553.43\n",
       "Estimated Total Size (MB): 662.49\n",
       "=========================================================================================="
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "summary(vgg,input_size=(1,3,224,224))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In addition to the layer we already know, there is also another layer type called **Dropout**. These layers act as **regularization** technique. Regularization makes slight modifications to the learning algorithm so the model generalizes better. During training, dropout layers discard some proportion (around 30%) of the neurons in the previous layer, and training happens without them. This helps to get the optimization process out of local minima, and to distribute decisive power between different neural paths, which improves overall stability of the network.\n",
    "\n",
    "## GPU computations\n",
    "\n",
    "Deep neural networks, such as VGG-16 and other more modern architectures require quite a lot of computational power to run. It makes sense to use GPU acceleration, if it is available. In order to do so, we need to explicitly move all tensors involved in the computation to GPU.\n",
    "\n",
    "The way it is normally done is to check the availability of GPU in the code, and define `device` variable that points to the computational device - either GPU or CPU.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Doing computations on device = cuda\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "tensor(282, device='cuda:0')"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "device = 'cuda' if torch.cuda.is_available() else 'cpu'\n",
    "\n",
    "print('Doing computations on device = {}'.format(device))\n",
    "\n",
    "vgg.to(device)\n",
    "sample_image = sample_image.to(device)\n",
    "\n",
    "vgg(sample_image).argmax()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "\n",
    "## Extracting VGG features\n",
    "\n",
    "If we want to use VGG-16 to extract features from our images, we need the model without final classification layers. In fact, this \"feature extractor\" can be obtained using `vgg.features` method:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([1, 512, 7, 7])\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA2cAAAByCAYAAADXuKnQAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuNCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8QVMy6AAAACXBIWXMAAAsTAAALEwEAmpwYAAChX0lEQVR4nO39a8w125YWhj1jzqpV6/Ku9/bd9+Wcvc+t+3RjTMcOdIyTYEIMxgQiRVgQ2SJRpP4RRyKKI7vhT5QfSEiREIkS/0AxCZFJMMFYIBthQwMOlhA0NODuPqcv57Jv397f7b2ud93qMkd+jDnnqlWrqlat9a738u1dj3TO/t61alXNqpqXMcd4xjOImdGiRYsWLVq0aNGiRYsWLe4W6q4b0KJFixYtWrRo0aJFixYt2s1ZixYtWrRo0aJFixYtWtwLtJuzFi1atGjRokWLFi1atLgHaDdnLVq0aNGiRYsWLVq0aHEP0G7OWrRo0aJFixYtWrRo0eIeoN2ctWjRokWLFi1atGjRosU9wLU2Z0T0+4jo14noB0T087tqVIsWLVq0aNGiRYsWLVp81UDb1jkjIg3gNwD8jwF8BuAXAfxRZv7e7prXokWLFi1atGjRokWLFl8NXCdy9tsB/ICZf8TMMYC/BOAP7aZZLVq0aNGiRYsWLVq0aPHVQnCN374L4NPc358B+B11P+hQxF0MrnHJ2wVpBYAAZjAzUIgykpK9LbMBtgtAXqNxBHvxNcfZ/9syQlp7aq3l/AywMQCzfyaA/cy1oenliUDu90DzdtPN3GOLGmzyXm8aRNL3FEl/TNO7btF6EIHsOPZjZcPfy83utln3CnY+2Jbh8WWHzMG2Hxiu70cEECnkJ1h5rl/2PgQQyN5iw3utm9vs+dyaU9o3d7ke2TW88TjI3287br4cWDKKtj3H9fokKVVqB1cdCyJwlm19vS8VKuaTEc7eMPOjsp9cZ3NGJZ+tXJ6Ifg7AzwFAF338DvofXeOStwgi6J/4FqAU6OIKPJ2CZ3OYyUQ6JxGCd98B9yLg5RuY+Rw8n99Ku6jTgep1gSBAdnJaO1jUcAjSCpyk4Pl8t0Zr0Q4odkCS9qp+Hxwn4CRee0r94BjU7QIAeDxGdn6x9jeq3wd1QpjxVC6rFUycAKZ6YqCwA+qE4DgB2Ny+MZ+bKCnsACwP863YVORRNgvcFQwWffKa7aIoAoDNxzQRSGuw4fr+F0VQ+/ugQQ8AkH786fYL5316B8DNOEry99jUMbWrS0cRkGWrY1NpeddpcqvtWWrCb/lJmH4HlGTQry5gTk5hptPStqjhEOrhMbgTAopA0zl4OgNfjWGms9r+CqVBiqRfs7nXRj8FgRiFzH6eV8M9cJIASYpsNKpuPxFUrwdoDRhTumbq/X2g1wUN+sBkiuzNib/e8rkW5/TY4rmpwQDU6YC6EczFpdggVfceWtug1wUMI3v9euPr1Z2bwqCyf701sHM0gOX+fBsO3ute47pz/ba/VxrBsyfg2UxszsrzE1QUQT15BN7rw/zGjxvZfV8JlDz7v81/5eOqw6+zOfsMwPu5v98D8HnxIGb+cwD+HADs03GjXklBAJB4wbfa8Ci9fgFxk7BFfsKhUCZCGAYUwN0OyA5e3esCpGTDM56ALy7lksM90MMHi8HnFvPMiPGfpDDj8fXvhRkcxzBwXlMFoPpeeToFIzcJ3TaYZeFveG1zeQUay+JTulGxBhF1Qv+Rm2gplEV53cZMzp3YBXVNP3FGSb4tStvGZtJXtfb9NG8YSJtksTSTyfI5ctf0xh2plU0bhfb8cVw5FiiSvuonwZoFQHW7oOEQPBqB0xTU6YgRYvvV0u+IoI+P5NxxboLNMvGgGZa2F66lul0gDEFRB2Z01WwMlzxn1e1Kuwq/V90u1OEBeDYDx0mtsVK8xrp+4bB0v5tCa8DUb7I5jmHOL0BuTnibjZ0ibvpe1pxfDYegqAOKomUHjzUcOE3LDe4oQvbmjT+/Gg5BQYDs4tLPX/o734QZdmG6IdQkgZrFoOkcmMdiOM/mvo/pwwPwe8/A3QCsCJRrtrqcgsZTpJ89X3w2GMh4JyXjPU3Xjh366HPoMAAMw0ynMvdVPB8zngBZJv0TAGeZ/C9J148Lk4H5hozX3HzqoI+OgCxDdnm59ud6fx90fAg+vxAH6nwOFUWgXg94eASOOkj7IYJPXtl3mX8RMu9QpwNoLfN5EAAmQ3Z2UbpumckEmM9BV2N5R2Ubszy2zesPAlCnI47N+RyYTORd1YDTBGZigPlc7JfC+dY5jdade+29viVQD45BvS7MXh/0+hTm5LQ0yqMGA9nYH+2LEyQ1oB/bMasI2eVV9fMsW4cbPLvgvXdhDofgbgB1Pkb2gx/X3IiG3huA9ofgfhfmx5/uZjOkNFQ3gjo+gjk9g5lM5BlZh4d7Vnp/T+YTpUFRB0hTpK/eAK9PQGcXC9sGMqapG4kNO5k2GttvLawDgKKocWCiiOtszn4RwLeJ6EMAzwH8EQD/80a/bOBJIq0ApcBNNlqlJ1AA10xCpPwitXK8Irl+cSIiBQo0oGzbLkcw0ylUvy+/jzqLY7NMvKqxdE4yW07QisCmsAFjtgsqy/c1t3kvIjFlk1fFBoKTGJysHu5/Zt8NaS0UNsqlTbp+ZbLVBb94PWZ530S1XmFy13DnJiWfwa7bJG3h3PfA4pqkFRCuGWb+emYp6kJaLQyGLKtm2RABmmqfm4fd2LJSizFABGIuP38gbacsk+cNCY7K8RUdTykxLMJQ7qdBs9wzXWlr2X1rLWPNOT82gesH6+aVbQ2Qpr9jtn299SruGqQVKAyBTgjMC2NP6/I+EwZiXOTWAeqE0v9z/YT7EdJhhLSnEdo+qyyllmZzUJKC3ZwTdmD2Osh6AYwmUC6qG6YGqmAMUmg3BqRkvFWNyRw2MnBM1tyRUYaC42Y1UrRmXaeSMU5K7tvw4rkB8hzK5oSya4cBOOoAYQdIUvlOa6ATgnsdcCeA6dgNaX49JJJ5R+uljRlpJW2pMLo5TYHixnmTaG4+ul43B7n5OUnl2CxuPL9wksq588+KFEiZej9pXWTHrZl1vyW1/NyckWrXEWOZKj5S5c7rYNdtsWuy5eOK594WzKAwBEcyPoPzcHndLrQdUQdmECHd60AlBoFbz13kfBP7tEHkjLsdZMMIpqsRzurtN1Ik/b/bAfej6jGzBSiQ8zo72cxmcs9hAGT22nadRycEd0LQjKrnmSBY2MhJE2Nl0wYTKAil3+yin1yrLcrPbdva4FurNQIAEf1+AH8WgAbw55n5T9Udv0/H/LPh7wVFke3UDDMalZ/b0hEoDLaPOl0XKrd5K5vRiosVYBf3Ncd+1WG9PWY6u5ZxSmEH6mDoN2jm8tJ7gYNnT4FOiPSTzyoNBdIa1OuB+j3g+AB0dgkznqz2SSKovT3bH0NkJ2elEaPib/RwCE7TjWggajiUSFaaes8naVUbqdoYm1ArrkMhu0mayJbtcpFOfXQoFI0GtNkWbyE2pZLl+5Ojjkdi7Cz1EevIyc/z+tlTIBAanHlzulircvQpf7ydq7yXvmhE3DJlcxu4iM4K28Su1z7SbvOPzWSymKvdPBbHIEsN5MkMSOLl51zyHNzYJa2XIpR+U5Az+qnTEeNyPl8YSVlWTk9durkFo6bRRtZeW+/vAaSQnZ2t/YkaDEDPHgMXV+DZrNIGcucP3n0HIBL66WRSG03VhwfAoweg0RicJOIcThKJ8K3ZDKpuF9B6a1tLDQZQD49hXr4WQx6SpoCHx4jf2QcZRueHr8DjCcxkArU3EJqpczAQIXj6RAz9MADenMJMZ9bRF4iz4+xsNw5n+96WNoE1x1LOkZ+PsOnDQ4BNo3WEggDU60kEu8JhTUG4ZD/WRTrdhpezbOGkbropabo25+a7UlvN2jk03BN6b5F9U7wmKahBH0gS30cALKJxGzjylw4JO1B7A5hvvAP94gzp8xUSX+W93WUA42/zX/knzPwvl313ncgZmPlvAPgbG/0my6zHew3tLMuA1Cbc3lVSocn8ZO1oJgAWO3RrKLvNJgWBGNFZjYcyF+4EADDLQsNGrpOkjTYsFATyv70BPI0yTgBjZFG0i9vaXIK7ABt5ntd8r5xl4OnMGwHeWwiAp1PxztQMajYMShJgSqDLwNLkCs/eDeA4lmebZVD7e/J5vwu+uCynNjDLgmg9uarf93kMzkCgSEL8+b7ir59lMjHb5+XauxOjbZNzlBzbmB5zkwbmJud2UTLYd47MUyK/NLAeTbU3AE9niz6lNPTBvvw7y5bybSjsrFDHnYHm++59mzuaYtO+Z4/X+/syr8zn5dRWk3nbSQ0GoH5f5hp3mtxC75w/sPRBZKg2QJSGPj60lH5a5DjnDJj7AjYsEar8Z7n1GrDPIc9qYAOe2znUfT+X58vzeNWTXsaqcGPXns85zRzMeCKRnRyzxL2PlbXGRa+Khpn7rT3GGfB+rs5vCn07jcwlZZHBEnCagkZjoR6uiyAwS/8iO05z90GBHe9x4u0GjhOo0VjmtyS1fc80SmvgrC6Kl9vMVBiznKaS6pGLTPNsDnU1QeeV9vfi7oNn8xU70K/bjnWUiQ3GAODuo9guZr/xUQNZZ7M3J37+cnMaAE8x08Oh5OUNB7JJHo9LbTaKIstMsZuzwuZ+k7Qbb2/UvIdGqRa588k/1kQ03QbLn38DO8LRmanieGahESvytk51g6WdHMer793dNwD93W+DtYa6moBPz2vZAWow8CkOIII+vQJPppXHL1/UWFbaDlEWPd4S19qcbQWbQ7K2a5gMPK+mct0WKAihjg5hzs79oCStoQY9mCsDzjIRpOhGkiScZZJ/ULE5k3ypDtTxkXyQpuDsDJykUPtD8eZfrI+OUBSB9gbgpw+AjEFuUk4zoeO4hSI7gYlxv4ws5t0YHSar9PKt9WTZicLMMmA2AyomANUJAaWWPJbBNz5AdjBA/KiH7kcd0HQGnq8+33wfUAf7QDcCUrsoTSZQR4dAmi5vzkom+8apgreU0ExRZGk996hP1cB7861hzKZB/3jLoDqhRCGePYY6W4gGqE4IvPNYjOIklb6axNbp1AU6IbKct5MO9iXPZTIRD/fs7XjHO4HSwOMHUNM50uefr/WoqqNDZI8OQd//Yel8JvP8ITB3+WMxEMer48ZurPndx8h6IUygEL6+groYwbyc348oWj6SZbJlCiKwdr3mNAUXIkQ8nwM1QaMVuI2xW4eDAOrBkZ/3OP9si21ciVAq6eclTibnHPWRuiCAengMpBlwcrq6nhYcbOvA8zmyl6/KvyzJv6sSYKAoAh498Jux7Fzsjm2pq7VOYRt9JK0rI008nyMrrF9mPJY1+vnK4aUb46p5mecl4yC3waZOB+rhMZKnhzCRRmc28/MXDYeLXKfLEbIkBh4/gDnaw/j9PgafDKCfv1ndeFsxM3TChdMkScRecPe3ybNmU5sTunaTVURDm87ZMNAa7BwYm2BNu8ree+3pyo5196I0Tv87D5B2CXufDzH4PiptMwBQhweIv/EE4RfnwGiM9MeV+holDdnweedRwXBw9v0uRHOuRWvcFPvqAf9s8K/fjzyoOuRpEo4Skfciu0khc7kJnYV3xYolVHpUnAcqDEBEyAsxVKqClcEJY3RtBM55Zw0v8Y5rw8xrnoHzLlE3Eq/VPJaFok6F7ppJx03bVpbUzTYHD0DlAuJPEUULcYC6aKV912pvADx6gPTxPvQ4Bk1j4PRCPG6FgehoP/TuU9Ashjk9s0aN9Rpnsqkny11fm/Rv1Ryb9AsnpnOjuUzFfC3rwebZvPa6qt+HenCM7PGBLBb/7W+s759Kyyaiblytyx/LTaRewMVFWRv8phJE0I8fAdNZfe6Pjb6rB8eStP3Dj2rfpYvEQ+vSZ1oUoXHXcJGaJfEYF12wpS6W+qrLmcidn8KO0GjzQgfXpNpREEhi9Hy+Oi43yCl2Y7ZWbe+aUP2+j5xduz1KQ3VCH0mo9Yq7+dbm0vJsvnUi+ZcVrm+6dbYMjv7t/657fnUCQY4xk8g4UJ0QpaJJZT8toX3eCKxog5vLbsquUt0u6BtfA41EAG0p+l4yF6luV9a/fm8lTaAomAWloQZ9iXgFgdDRap6Z6vdlfjzcB19NFhEvO9ao1wMUwVxc+uvowwPARsBce5zCM/V64OlUUixKxnxenA7AsvBWZSOtyEwQLIuTFRhXtwZnzwL3y0lfgeC9d4Um7kSW6tRJo8izRZqyznaBWlt9A+GxG6M13jhuIxJQdWknxMFmddAyL1EcOUkt1aKBl9OFdku8jBspU1rP4NrOSNT8ObpkT4gHAJ1QvIuRKPp5ZWBLh8ovkAtDbjWxdmVCvgnYZGf375Xk3uLhuc10bRIts2w27fNIBwH0aAaazZHmZYpLBEg46oAyoZmW8cwbR54Urbud1eNvEmVeaK3AxesWDW+tgTCA6YZgRQgcXWUdLG21VBwEWCuKs5wbqsQp0uS660AifMJ5YaGK40A2cTsKFgtlPqe14PghK4iw8kydc6d4D45GW8xfqcnrLYuArBPk2Qo2T4irxuU68SZ3mBuzDY/fGESbR0Dq5myTNY88Mn+51ct2AUUiNuREhrJsOW/FGuDclKnUxIBis9l7BBbj3f9dMR9fdz1sIvBSI7rRyCmiNBCGMFEIPSl5sGVzkas3acVVltpCSihyTnTHia8FgQhnrRVyI0ArEZ/QSq5LtHhHJRFsTlIQKbATeYONeE0ArGFQbL3hJeXXrc1+VxA5a/KO1tl3xQjRpvm4m6KpMJBz/Dm6rKXepp899xtZIGc/ut84Z7x1om0StdspqOL97mgDfLuRMzrmpnXOXE7VrXDum25e8ipLhfD3vaCelECUJEm8sGs8NvrhA5ivPwVlDEoy4I0kN5MN53OcwIxGCL7+PrKH+9BvLm2pAIY5Oy/nbIcd6Peegc8vGyVL32uUqWzZDa0+PgInCbI3J0vHA7i3feMmQWEH+sERzPnF8hi+CS9eXuHrDuvVNTrW8tEp7PjcRbCRiDSpBQXFGJgclWepvMhblgvm5/IyalIZtlXovQZc+QdzdXW/xuuXdA65FYbFTWFDZUbV68lvrIqtK61zrbqjTdpg8005jn0UlsIO9MNjoeel6WppgQL0g2PZOM3nldGl2jaWUBHXGux1x1Udf0/hRTuavGeSUh806ItTPM0aCcYsaSK4/DJbfmeFdpcrz+PYVmtZJA2RpwF7HYDZfJFn61hIwyF4LjWD1WAA6krZC57Hcr9XV7IW9rpQ+0OgG4HPL/096uMjUdA8OVthiVCnU78+ltCGN7/RggDRNXB/I2c1A5ANe275vYGLohSTgG5ggvDFMx8eg88u6r2qlvpFewMAWOKzc5KKQddAfIOnM9lw2WdvJlNRDlRqQZsEwFdjaPtf54mqqr/CWQZeV+tqk4nWUrEkmXMHi/sm185HTP1nVuDDmFWJ7qpzltDJStuVo354rLlf1RUaqhOHWaH5XFdmNqdmVtxA5MFZJom5Skm9MisQsKQ8WdEOR13JR2ZX6sQtXczRxra/rWuh6QYt58H0YjZOJIYZpMWA8/RkLrw3YhDT6vxzT1CluLXk9XSoGwPb3F+Zt7aEflu53jCDjPE5AzTcA6YiEHTnohwVlEkKg7U0O0eRrXKcrbyvNX3ZGZvAGoOTFgq3ZZSjO6u7WQalETx+6PtQ9uZ0ff5VQ0eQiiLQu09F0l9r0KcvwNN4QRsuIHj6BOj3wFEIOruUmlFl82Sj+cZ4KrFfu7PMinZkq6WCyk5h5yiO47U11hq1sS7f6rrnvivkojwAlvt5Xkmx5Ddq0AN1uzDnF5LuwAx68hDxowEoNghOx6AfTqEfPgCijthcztl1crpIy8iLh1G42BQWNiq+z2UZiGhRpseURCztZtFRM5cchFYsSQ+Hi5JGbOQe4sL9u3Fu7ee8GBynKTBlL97jf8sisuMEYnw/TtLFZ8U+yWz7dM28sos5hxmoqS3cGESoo/Dc3ebM7z4rLKqcKtaNY5OHfEuePupGUAf7mH/jEaIfU31SZCeE2h/CPDmWeluvT3w7N+Hg+uTdNchOToG6KvFLJ82Wo0klcPKxjXKqXJ5dklj1putuzuzEua1l7zZs8/mSSlXtJbVsprO6jVJubChbcFsm7/pJgYZD4GBPlI7SFJST5He5LNcSeSAl4ibM4MyASgr6ApD3fnkpnrFBH5jNF1RZZpFQrpDmpeEecLgPmueEKupyMe9woV5EUjd8plWUJKIlR4iHo0PfE3u2DNSNQEmy2r9KnBoUBpKvVRbN3+J9Oqnr/LuQejyZ9yjLARXvyVJmKQhAwz3ws4dQ51fAeCqqenfVxyquS2Egc0iS1vY9cXSU5D8RyTw6my/NoXWCDwCWVIZrj9Ma6mAfvD+AOr2QorP58X6PImaqEyJ7/zFYy+ZMXY1r101fG63JetXrYf61I8wPA2QR4eizV7W5hNn7jzF72EUy1Bj+sAM6Odt+jeMSoRLTrLC3P/w6dfGug/u0+VoHG+WBIsDw8pxWpBQCS3oBdHyE7HgPKk2ByRScpJi9M8Tlhx3oOTD4PED0WYTsvUdIhx10XnTBUShF1eMEfHlZnivXjQDrU1pKn8hT2aMIytXhMqt9mQIRmSIeg2OzKNROBGM3nXR8CA5srdRZDEyn9ZG+wrrnxQFL+hknMbLz1XFSJ+a11obcVb/axcasihbpDrmvtMYbQ11Y3Q6yXSViU9iRTdbhgVAIXr/xC9pa8Q87gNXewB9XlWBMUQQ13BOjlhnpx5/Wdh794Fg8FdPpcsJq5Y3cMG2zeH6rksTFQp/u2HwtOVII3n0GHvbBP/qk2f3cBDYJl28qt1rB4S6ji7nPRBa5RHq4RpCgMU1nb8+WbojXT4b5At9AroB3ddSTokgWgiSBm5/eNirf1tggmXgb6MMDMSDWGGgURV7YILu4bDZHbEL12KHkcO01Npm37DijIJSIjy35ci8FrBo+v1qBoG3ooxu8Z0ehcqIaTd613t+XmkkvX6197hR2oD58H3R5hfTFy+XzHB2BDobg8XQtNYzCDtS3P5A/mME//nR9tHSDPG59sC9OKSKReK/Z1KrhEBRJjTYznqzSbJ1d4DYDpGCuxuA0gT48BB3uI320D/2bn/lxq4+OQIf7wqSYiVKk6velzly/L5/l8qc3ESYLvvEBkicHiA87CCcpwldX4M9erDxvLxJydADuRuAogDq7Ak8mq2qUZekDdcizX+6qdtUG7BYAfhzp/T3QnthtPJ8jOzmFPj4SYRNmUeE8u4Aa9GVdT1Ofe2tGIy+CttSnmj6/dWM5L3yXyw8DxDnkFDJ5MgVfXMKVkbrT+fIOKPGNoTT00QHMh++I5P94gv/yxX94T2mNDjkFQwBLste32gxliw3qDJw2mHwdNSdb9kwsndMmsIJ1/sNFUnNVR7Yejuws9hTHSlgaoq/v0GQysxSqe4FS2k5F2/KeKDcpawUOtkzA3RVKislWeqFdWNypOuVr3pVtwDeYaLgqkrUlVLdrm8ALmqExlpaxEHqhsJNrxPKmMD8+OK1fKJyKKZJklTJZMIg8zYLNSjsaO1eaboa2TaJ2vysWp686x03PeVqjsbqMS+xvik3aXuZRLmIXTqFNzuEdAXFlwVVHX1pSuHQUoqIxtIk626b32uT5YY0neZu+1vC6QM1zrIOi5bG2DtZrX3oe239p3fnYgCZ2M5YZmCYMiKbvymQb5Vqb0Wh9eQGVm1Py41MRONCyFmonsmA/1woEgNwa6YTC9OoYLxVMsutZcaPvrmc6BBMrK9iifP7RkrGutbyTQINDLe/OOxML66TWIGKACcUcb3lQeSruYm71BeJLjPRSSm6uBuZG489Fv7QSRssm/dxvJm19vOlUnJEWZaUT/Ga3bB3SHeRFVkrTL8rasG4sF8+T38gZ7cuzYDpbcWY4596STWZcgcjFZzydgg1LWo7LhQPEprUKzTBm1VFWHNNNHYLF4zZd16/hVCRFgJLxwg3mpXsROVNdmy/14EgmlVkMPj27k1pETomQiNYmsOtHjyS/xxaRBrNIuBYHRq5T+KRJGxm483yGtxRL8vL3LDnYS+nbvJVilMJ/bylUyQePAcNQSQb88m9ulnTdBNsmwSoN/d1vgTWB5inoYgROElC3K5Lwez3QDz4Bz+bQ778jv2GWXJ3UFqS0+VPMXNvf1WAA6nWBowNQkgrtMec9L8ud0U8eC31jHgtt6mqM4P13AMNIP/1s7e1RFEH1++KBrJO2twYEwrC+wHzh2bnFG2EIijpLxXaLC/pGZTS+CrAy4c5ZsbHiK4k0/a5yxhwLgp48BBlG+qOP/HfBs6eSKxgnMDmakRoMoB4/hHn5ul4Ouqw0QotmqNvUbirWA9ybNWQr1KkwVt1fk2dk84zVwT540IP5+Pli7ipRF9QH+yLwcDiUvLkXLxfON++wNFC9nvT9Xhd8cbkYp9apQZ1QNoTKCprFsbCQmKtrv9lorSvd4cYdBQH0o4f+GWRvTsCGoff3fDBgcex6h4o+OgIdHcD0u1CT2dJ8UPccKRCVSaicoqdSS8XIt2b/bNDfN15viKAPD2U+bZD6or/9DSTvHEiaDQAQQIkBGUbWC8D2496vPgfHMZLvfg3JQYj5vvRhlTLCKwMdG6jYIDwZgy7HSD977udLFUULu2KNre5ZRnHix4cXRyFa5LuteX4usp2dnG4/V+TG4r0SBHGV3GGMl8KVpMAZ6PxyoQw4k4VKHx0BsBGB6XTnxguFHaEypqkIHMQJiMnmHxRyoZSG/vaHIqP+yWdSlTxNxfvT7YJ7EWg88TkO+sExaDhA9unnfjLTjx6CD4cwe12oqxnUZy+EvlXmfVF6afIgrcHMooBTMnmo4VDoOHYi27rG2bawdEQ36Jt4b0hrSXa1wgjm7FwmaTeRdaOFDL2liDjhEyQJspOzhdes0/Hyqq6OCQCbZzWDfvhAjLwkkYVMEbLXkg+n9waL6Euva41rtdhkOBqobdsKXzi3ILJhII6Bq/FS3/H0PnKiDwYYXSH8PLQbGSN5aLvGtolKbEAn5zImjRGBmCwTb9l8Dj2dI7NqTHyx2IA6MRI4r5frgzX3xrEsTgTZuBTzrjhJV37PkykQJ7mJ1YAvrxrfr1dMW/PMlwQtmr4fNuKUZCsUk/P+ydfL49IrZrUQcFli+WaRJTOfN39f606XZUAcg84vUXRomquxjBEnsuB+Eyfgy5EVZdJLlHnV79tj4tW5BCilZznHYVU9prL6URzHzYwwF8V3HnxjZK7L1WprwmbZSJ1uW+Sj3QXKH4Dl75puut7mTZlD3btxzuGwA3V4AJ5MYKazRgJTUAqcpNLPk2R5viyJvohCnwEZ6TP68ECol1kGTnPrpBWGoFyERg0Gi/PaNYB6PbEPtBaBkpp35QUojFnKAXdCKHJqXghSWDvTl2gw9bmU/nzTKaAV1GQzm9S3r8x+KPv3JtikD5flNa85N0+n5Xn1jqK5NwCzLd1ydoGwwHohq+wdhLkIprWXO5+fIzyL0O1ZhljGUPMESDNQmkk+nmWGcZaBAJnf7b2sfV9OYDBvp+XfRUMqpJtPrzVfNPzt7W/OOh2o/aF06DQF7OaM0xQo7siJhKOsCGoWw9yAZ5k6IWh/KJ6LWKq/u7EhlC618N5rjfF3jtE5i6GffyF8cEgOF3c7MIcD0IsQmM9FxOHREWbPhui8PvEbL/PoENN3h5gfanRPI/RHE9nkzVcpIBQGoKND20HtJjBNYYxZ3jRaqIN9cLcDNYvB48naZPGdg5S82ySRjcm6TmzrRJknx0JzYIaKY0nAJyX5ensDmezZSJHJowMkTw6gUgM1iReTozGg4R7I1r2gwQDU79rrXEn9kwdHML0OaDIXaoVSoPMLUS06OgDNYzG89veAMEAWhVCzWJ5/ZoQCMpt5D+BSYcoM8Ao+Vsxm2bAKJJLiFo0sg7H9v07sZSfYdiJhXsnlqEKVJ7PxpRwFqsorVyIQVJZHslG5hiZ1goAFvWOTuccaLZ4xs4728lXIp9sERUrNFn14p5EoI7Upy2rqVOUzcRLbEgnW0bQ3sEn/MahvDU7r1Chu+DxlKvcMXJ7RShscxbgghqL2h5IP0kCyXVQqbRsdjWwuIj60NwDmMRDHa2sz3vjmzNHFS4wplxaRb6MXimkj0gDE3sHjY9AbEjGnde9zIKV4zBtrw6yjXAKyqZ/NgNEI+vBAbJjpbGX+dHPqkhS663826sWGofcGIJZaaGYyqY+Ulqy9crHyPFszmy1LwDekKPp73ASFNeEusc14qGQguHSEowNQZmBGIxGBWyMEt3TukshjZc/cZj0uExjcQnTQC5jcAm6f1qh+zyIi1UAK3efjGL6Zyuouj6CsPUWvG5HUWEjSpYGeH9zeG2VpNTToixCI8ww5mfAgAJIYZjxdRH5cvYpC3Qa/WFpvi38OhVAzhR3he3cjL01qplO4xNO1dUoslYiCYOEBWfPslp4PcotzU+8C5erAAMvePJfTl7tf6nTsM7EbHBemBpZq0Pl8JNcWk3kvYJ77bKZT2+5w4bGyikSeM5+L/CxzsGu8siUiJ0twXt1t+rN9ZkueelILEQPDXjkKbHzfVIcH/j4QhovNpY+0Jp566DjjPgE5T8NgXpHGbrEdXH2tfD0YNRiIoWlYxmyTQvNFL2UQVv+uQK8xk8mC0QDI+44iIIkl6lxCGaHIzjG2dlKd8IOjj/ooTl0Upk4MyD0zW7sRQONaUW4duUvqYL6219I86bCOclZHT67qA847vI6uY/sh9Xvgq/HSc10pAlt7k3dID7yDa6tuF3SwD3N67sebY+NkV+OVd6X6IuyATghzMWo0tvV3vw3TC6X+6DQGzWOknzwXleZHD2FOz/y6qfp9UW6+HIGzTKTQ53PwfA69v79EEWy0Rl9XYKOs5ENN3k4+f1MfHQBKw5ydrRW30EdHyL79HoI3I4mQWWp9lWhV8MHXwJ1QbIGzC5irMdT+vgi3nZ+XX8eV8bnjWpMURQuhuCStpBk6e3BlU+Xm2LzwHTWoEVbbqPJ83GvjrgRe1mHbuSaXuwlF+Fuzv3h/aI1i2DV/eTe+mNa1p/jgK7jOKx4ge6yZzsQ7lS02G2Y6LZUNBRGWzHfXKZvcvzWcOYnBKUHdoTDGxoOIS+R+HfKyr/ZYtgvNumuXtWNpkvL10myIO98H0lS8I8VcQfs7X1DRGbj5TVyWNauF57nYa3ImnFBOw6ixFDnlZpQFInAYSFTQQetlB8KaDbqjWuXl+olIqCPF92QXav+MAJ/AfSOOl/uAdRtwN1YVCZ36HnhV7xxKSR+qE2txThab/O8pw/cxAkmuj9fPTysoOnca0Nbyf2+0zrLUefM04U3bWtWOTdDUWVWWxF+2OXX5TY7psNTOAvMkJyDhcskRBp5OlV2Nl+d0ty4MhzDvPIKypT/M6xM/pkkRmIvqwssCHo2wzW8qwJmpXm/LDOFNowvF92AycFyxqbLOhrza4PK1GVCwTtYmG0m7noVSW28jFFWNvwoo6Ut+87lBLppEBDebb5Z+u/Y4tawxsCtsssGygQvnEF9Xp5GiyNtBXuAQlqEQNtt23QtBkOvgVjjuOwAFgafoNUn41A+OAaJmiYc2L2BdUuSXAtdQy8kPRn10BNrfQ/b8i/K+4wQJXO7j8aFfZM35hUQj33kiCdK9EGqWAnECGo1hzi9qRQic+uY6qW4nd4yjA2A0XiouXgU1HIK6Xek3DaLS+t1n4ItL73RQwyHw4bvyfcYwv/Hj2kk3eP89KZg6nQNhIDVPog5oOpdcy9ymSw2HoKePgPMReDaTjW0UgXpdmNcnjaMgbwXyHrIGi4qPRN2SSu3GAhu7Rp1H1HpxVa9rZcJrxFps2QU16IOTBOby6t5t9NVgANJ6lW2BhutWfi66j5vPXcDltq3JVaZAxB7c5txMZ77uWD464wSGeDIVuubhAZDa/h5oof072lVePGY+x+Uf/R24ekdh/oBx8JvA4W/OEH7vYzH4XQ6tM85+5rt4/j8cYn7M0DHha39jhODzUxEucEyWTmdRLDg/l25iHO5CuTSPkvGnHxwDQYDs1evtrlUWJSuyX3LQ+/uyAQ6CZUGQHCgIoJ88Bo+u1tdnI4J++BDUjZA+/2LjsZJnA3xVoYZDUL8nufgNnl9pIfsGcOO4Sd4jlEbwzlNwkjRrV0MbsUxkrAp6fx/8jfegLsbSF/N2uXf0L9h3wTc+AHc7QJKCJjPJl2MjqUlRBJ7NwHGC/+ry/3GPImcWjnbmd5FWMWVFLrNBoh9d05t0G+Ask464roK5O348kQWoySTpjL+SY29r81oZQi/A1zs52Bc6RgPln6VEeGYAzV15S0Zo7vmYyQSqLinWCRLYd+XLFNgkZxCLCMZkBu02WmkKM1u/wRBxASwoR1XHxbHk0hleEkeoPfds3ryPJSn47EKSm3O/16/ObDvN6vMpjEm+uARCUUHyUURLhyxrG52eg6czubYiUWacz4Vi0TDaJycrHxd5GjRZim4VTUV1u1K0O4nF2OqE4Ml0dbLecKPl2uf6SdN3If+4ndDZTW/M1HBoo7jGC/IAEEqNe0eAV1v1ylvWOOIkhcFsvVhLIjReY3IiMk3vK69u50+4+2fCcQJWhXybNfe1pOBYmItWjg070A+scFZ+42A3BE54y0EfHkhU3NZC9Gqqear4FsqupLUYtgU6flHUhLSj58tGzNFUm4jzUKcDdXQIvhxJzq4T3jHL87tfK4gkx/f8wt8naeUFiOxDW4jHMOPwe5fYex4hGQTovppCv7mEcUJfwELMhw2Cz0/w9BcDJIMAlDGCz0/BSQL98IFs5ty6X1Zzsvj8ihubPHbQL72SYa8LHk8WggruEuOJzN+5a+UFQeSDGjYPGytsxMufZcvt985JJ8BlUwh8bUW7YYObD0Y2t//oyAuLlPZPZvB4LDnrWzgxvvKiTEQ5wYuK51Dop+tsmCqIDZSU/tbV4fX542xEWKl4LcdeCoPlEkQNbcQykbEqmOkM+uWpiKLE1tFgHUrq6WNwoGE++nRhL55dSB82dj7NMlHu3esjftRH9INXMJf1zvbb35w5qpaTUQ0CH16laU4cJE9LABb5RMDKwKud9O4LymheNdhI/rlIB8jh1jZnHRkk65JkqRuB9vaQPTkU8Y8mm7NuIV9kk4VKa1EALCYjW+GQShSe6YrBztn2pR6KdM2qJrixsEFf2Ki2UEkNHk7iahEQb8wu2l7pzXQURlL+eC+QkL/ehrTlRXHN8udHnY43vtSgD3QjUIVUPnUj4PhAahwZI2qrzCu0Y3LJ4mEoyqpNxxJXtLOMllXXH/LqoLsqsHnDkSX37OEos35zlsuJZV6ITnQjcUI4eowV4FgLO5a28XZ7AYkk9z5vQECptG1rnr/rb26TURtNCgOYh0dCS8wWKqk0T0Q1Nk7EYHVUvL096etpJtEkY0A2FxBpKup5TZ59vg1WUITTRN5xLi94RRipE0p/tpFyiYRRM0ddEICHfeBytLyRKSoG5jd+SYJsVD+H5uch88++BwUgAsAA6kZ7+vxzqOefw7kbUliF6QeHIt6VNsvNdUqZfIMiXtQR9WPaH4qBWVhXSiNXvZ4IYU1nnq5Zu4Estr3kM+qEwpYYXcnGzG7ISGthUkQd6Z+Z1NFKP3suKs0HQxFGi2OhSrrz5++hicBTFb6sUemmILVW8IKshL0X59v2mdXQIGl/aHMBz/yxZXaGFzHqdUHF3MIm69sGlF1OYqRfvFi9fjdC9ugApqNBP/rEX3dFmIwINOghedjH6L0InefR2nnh1mmN/8pP/BxO/7uPMH2swAo4+HGG3hczhJ+8FmWpfB0FpaG/9QGSJ/s4+04XnTEjHBvs/cpL8OXo2gpxlbhuYrHzBDeRk79p3FaSdFO6YU6yuSpZt/TcwHb3sGs6yC5hHRBeAKGByIaLOK+jbql+f6WsAGfGC854L6VTnXQTs+NU5yIQjpILS/vI3rwpv7bzqiUpVDeCevQA5s1pvdGV8355+e41z6xWOMH1QzalRVPzeXzOY8vOg6f1IiHaHedymdZFV/IbqLoxoDT0T34TJgrBkUbw6hI8Glc+0+DZU5gnx5g/7gOG0TmfQ338Etnr19WPyNVvq/MyN4Dq90GDgSTkNyiLoR8cg+cxzGhUKsiTfwZOjdC/77oaTddByby0VPPmvhpkG1LefOTRIddni7l4vkBwLgLkjwO2fybFcehQzJ0rOho2qcOYm2OaHr/zPtXgmr6Nm1y7kON8nei2o2xRFPkSMD4PJjOl64dfW/Lz5U1El926tzdYsBGSAmtKS+kJV5SY9gYwwwEoToBAIz3oITifgM5HSF++9n3BCduY0cgzI2ivDzAj/fjT+2sLvE2oywXeASiUch6NHG41dqc+OgKybD0d1p6nSX27FVjKOWi9c8mLLkUdmNEVeD6/X3XOYAx0AugZgxVBzwzULBXqR1oIM9p6FZQZ6BhQCYNS9uHuG4OlQwDw9WkWHm5Tu+lyIXu1Lxx2ns6WQ663jU0WBlckcpMOmo9wNhmwVVK3dbjOs7vrybgsH7C4QchhJ8WIrbS2oylQJwTCDiiE9N98tMrSEJ1sNxEBnY6I07hjXTJ9lWKbvQdPL7bjFmU1Ucrgjm907Jr3mfPU1kYbchLbvr9XHZ83sJSG6nfXc+Xte5cmFaLmtgYQUpL79rVWVs/HxgCpkXkPWDZ6m+A6NB3XtsbXYv8ea/uvyVa9sze12DMDVY/sJihMeXrfNUC2rtNS3bSK3FjSWhb9XKSqThxFvttw47DUOFpEk5GLuCzR2SrOWxZd2eTd10RUpfRNIQJUce7GOS/b5DjXiWDUYQebHn8ew4DGch/PZNw5ZlKjjau9721zi0phI8GcGRDZCLpTcXV9PE1hYN9TpyPlbOYxKDO5wsa0WJuW2mznIGaQW1sKtNdbg3VGO2dIpW1l+5nqhDKWnRptHJeqLTrHofy9cHbcSjDghh1aG82dVewUYLv5fdM+0rQcD5ArMdHs1HcjCFLlVavDFtGTXSS9q8EA+PB9oTsxg1wNsQpxhuDZU/D+HmbvHyAYpwgupjA//PhOJZybgMKOUB76vSVZ4Ca/U/t7kkfkImFuw3YbE8VdeEabXtsWEafjI5gXr/wgzj9rpOmiFpLWkvwcJ7WRkXVtkhpH+2IkTGdS/63TkQTVy6tF31Ua+vgQ1O1KvtWbU1ENPDwA97vgQIFenkoeVoVqoy8jYXNG7zxS3ARWAMBFDE0F7bEM+vAA/PV3QJ+8qK2ppo+OQIf74ECDZjHSTz+rbQ91OtWbcpenpTWo05GI2A6T1vMS7zf5m68C9OEBEEXbiyq48zx8ABr0wVdjuNIY5uJytbblYAA13EP2/mPQPAHNEuDsEjybVY5ZNRyKMyZJl+W0G4KiCGq4BxzugwwjLalRVH+CDdgMG3jpgw++BhAh/fHHa4/VR0fgrz0F/fh5rWfdOVtr62uVtHnjyNm2nvv8Zfv9xiVOnCIvT6eNRFj0wwcwo6tG1NOtsIF9R2EH6vhQHBeZqZa+vwdwwjQ06Eu5lNHVqrPevnvV6wKPH8AM+zC9AJQa6PMJzI8+Xmay2FINCAPpa4HkCXJmGgmBfamQj8Q73JU9uAXuV+QM2PzBKS1qXHG8RDuqnMxsKL7YoZc8C64ug03UdV7uvJfCeQbV2aVEEWwkYSmZuABzOQLFMXqxFLXm+Xw5p8FeW9QYxSNnLq9uTCGoaYiY0wSIrbdmg8rxnCa+ynu+Fs5G1eevg9sciMVN57prs6iJqdOzJUEPzjJQkgBTWlCR7DMz5xeNklQrjWNLEWFrmHGaApOp5J3Y+kVLx16NwfNYKGbTGaAIyjBwFYJIgcfj1f6bv1yWiUfQPY9tsIGxVnbfvswBKqILuaiwp3C55P46URiHXDTEjKdQn79Z6y0zkwkUG6ldmGZr73GRcG9pEsjlT7icIxeJtPfg630pBTOeVC/KlPOwWrondTqiBBoEyzWB8tFRG/2jbrRQmbLsBjO6kugLCjQ5RT7qQ0Si4Onq5zVUk90keqy6XVA38mJLd+0cMNOZUK+uOSfx1XhRf9CKJpT1U57PYZihPyPpZ2kqBndN1Jpnc7CiBX25Drn+6GjHnKTgq7GUidnq5jb4VZM5xZUqsMIRTeYTM5nIOF4juc5ZBsyr53kX2VlS0vTCGBusTXz9Oq4cx83z77WWXLIwrGcNQCKtPI8BpaC6XRF/UiTzQYnCrhMRaRRRsMykSqXekhQRThN51045k9k7id27z07Pap+ldyp2OjIebsj+Yic040pUlOXr2XdvpoB6cwo1GkOHoYxPV/8y9xsnLqfCPRHX6Ubgq4mIodymoAkR9MG+3xxlF5cbbwz1wwcAqbXOaCdiRFEkNfxmc694zXG8SpV2TbT1XXkeL2oI2z5HHbGNl5ysJRRrNRiIM73Xgzk7FwfFpkqrTY/N4c7UGjeBJN51fXia2UieSjcCj7OVsKYL93Kae0mdcLF4e2NNAVBCM7IvCsXNGTPM+cVa2XMHMx6LyEVdPhwp0NBOJMaIEMpNbc46DQsDMq9X6qn4XWme0G1tmHZBBWm6MbBUnjL1x6q2lQqP5JwES5Fr5uaeSa1ByMBcaL8N85tZbkxMp+BiLoG73ny+khheKpRS9ZzqaAVN4HPEmp2DrMqcF1RxSl+UcywUcyncZiEIfD06n1tXN54dXdMmqzuxlSZRTffetavn5qi/FZTQpdp5Ts0wZ9x45cfU5sCRGEewCpkUx7UiDnmanKN00d7AFmJlX2uIAC+44jdxg4FEVvtdkQZOUtDV2BvmXiwlyxbP2VJlEXVA81jUTuOk+j27/uWiiE54pYxlkVvsqNcD9XtCd4rjWiPzNrAuob4pPAXGnbfqetZo3UQIYZO1xuUtwW7mMBMaoJllGwkVbY1Gxo84XszlVeO1i+fzZuyEdaIBpGQsxfHiuG3nxOtu6Dfo+14hUdsaqbUnXtDbqRuJKIfWovpoVYqXzt3p2Bpx6/skaQ2yOddVY5e6ETDL3R+v1kalMFiyqXBazWoAFptq6vdE3OKmNmd1ol75NZVFICM7b9AOJyiWZSIaFAYb0et2BdJa1gYlBZXpatxI6GzpHMM9Kb9TlccOyJowGABRB6bfFedXlolzcTqzDudygRgKAhEYGY1lMx/H0lf6fS80h1zk1dddzTt/+31QvwtzuAeaToHxeK0w2cotbMEkuxd1znztEmc4lRmHjkJGCsHjh75jm8sROEmFpmG51qUeqLLda1kCc5Pf7QA+YZ5NudG2SZJ07YVW2++k6Yv5Tl46/ZbD4m8VReo+C4xsAH14IHlonRDmclRNgbKRCdrb8zKyVcfeNryoSa8ruZ2TiUTDnXPGRtn1g2MxGGw9tdr2281i8Pih0EScAbfte99m/mg69lUu38IdmxfbyCv02Xft2pKPhnqJ4opcCF/A10XHinNWWR4c5QrxOnGKkvM7ig49OJIN3HS2KPjLjOCDr4ly2ywWEajTMwRfe0/kva8m4MlkIWSzCUU+L/ZyA/NO1ZzmKM3m6urtmUduSqzly4IbFki4EeSEkwCsf69lTpKq+940T2/d82sy9+bnKcimyM0t2dV49fwFNsEu+rUaDECdznqapXVA6aePwWcXzQQrKs7jdQJu025zea7AkjOkcQms3PfKbu7W2RT50lue4dHk3TmGnV3jlFOJD4NFlHWdsGAub7CxEFFD3D9aYxnsxmxF6CPvWZZ/LCgbeSEBK5nu6QEFeApjldxm1Qu+ocXo1ug3pdEOAxhRyyz9rkU1viTGCWcGpDMgU7ViHMwMMpb2EgQgpaFsVGNdnoJXjEzTZYeLm+yQi3SVfZY7Xr4oTMQuydsJCTH7vr2ELAOnmfe41T8YBgirdQO3fe/b/K7p5F9a68cATBJRLX5u791HDptch3My7pWCKWVzTOYvW39+oSaRU2vjwoYmzYAklf+6d5dmABnxIBf7Yc7Qc6qMCMNS2mcpBazKiVd1n5si9x7eGrxtG4867PJdOiOz6fPZwGi9Nay5Xq06cMmGZxFRMMsR7pLzVDplC5s76nTEkaxUtbhaMZVCiQpxZfDBR6wyn7d17eiZW3+aIs2q21eHfHoFrynL45gITYpr21xJ0rr0OTunPoAlO70quODX8+I7zlM0a1KE8sinISzoww3Xr/x9m0JN4DohstyGnw0D6RrdiDJnKRaOUV/LteF8cS8iZ3VQ3W51NM0fpIX7aotbZqfnyw+ACPrxIyBOapP4d4JtJv+y39R5zzfM0QFucTP4NuBLEv26bQTPnoIHPVCSgi+v6gUx9vdhvvM1qPEcdDVF+vwL35f1/j7QkVwHF/FyIgWA5O348a409N7A0xA3qv/X4u3CdQ1nmx9lZlK4WT84Fgnu/T7ww0/Xl3TwtGWztKCrbnfls6/0HFLMvb1LUaYN4HMjgZ3kJzZmfDgjr4xa7lCm6nuDUF3JuclGo9rrSW48+TJHtQ45KxTD81hoyY7JlGXQRweA0pLDbM+jHxwDSboUOcrngpnxBKQI+sljyV0NNMynn5fbgk5DwDKCXKmeJuuFGg5B/R6y1ydvhSNCdbsyr7l0mzxy85KjbqoHx+DpFNmbk+qTEkH1+6IyPujBfPTp8rmJoB8+9PRmczECpwn0cAgzn6+8E8e4AYTBcmNCMjcI1e1KnT9gucRXGYig9vY8Gy0frQ2ePgEfHwAn58IMyfX3a0XOiOh9AP8vAE8hxM4/x8z/ZyI6BvCfAPgAwEcA/i1m3mzn41TTLMrCz6bJzjrHtyWiVe8FM/jicjsvxabY5hplv6mbJDa4RmnC9y69h1asRWS3JWcEWVZZQ8jXX+l1PUUqu7hcpkhtKl28KXbZD/JG3TonAuT+ndepkadOaQTPnkiE4Gq8VvrZJa9mZxfbPUMnvjObg9NEkrGtYWouLkHjCQyzFzipEpwx0xnUp6+AOIFJ06UxaaYzKYoLeHod27w4oNBnraiKE+4pa69+cCz5p90OaDoHz2arC5EzKK2Xz4tpuPY0eFY+F6xQB64Urp6f7RdVvymtLXTbKHME5eeIPKWpWPutCR2q6Xi77rhkm3dm22VGVyK/fTUWo3fNtTnLVmvawa5BJWvKjSC/wSzrk0pDP3rgFVT1/j6A6mLwwfvvgfd6wIs38kEnLFXjVV1bU0opr+BXiWKUwmQLY7+MQlaCnZQM2RBOvEj+uD5DpLHolXteVNNnTAYTY7t+ZWlmatDzmyF1dAgEGjyegsdjmMlk0VdGI/uuF5GBYt0/XyMs0CKElhMLciIOPJst+opTlD0+hIoTiQrNcyJYk8nCLnDPXmkgLDwq944sDY0zwJydgwYD0KAnbIwgWI1COMYU2fwnpyHQADybA0lNzUOnvllVfLtwnH74QHIa6zZDDaAPD6R9mVmK/Js4AY2ulvofhR0pRj+dLu4ry4A4FpGxZI0dbW0Xc8Gg8WS1bzN7kTEAPgJqpjN/rOr3LfVd2AFmbNtSM9Y2qmm2Y7g5r2rTZeJEVNotKIqqaY3MSzZM/hgznsh5rKJ5UzShNaYA/j1m/iUiGgL4J0T0twD8LwD8AjP/aSL6eQA/D+A/aHxlQAZSr7v4czJZfY8NDUxnFFcNna+8xz0Xcl0YV9ffAHmxFmeAay2iAZkBF403t5GJIlBfCkMiy4DLK98W72G8Sw9WU0+wE3DohNLudeplRF5kBkA5VaT4E0Xggz3QLAbiBDSf19oVkuTcB11cbmV/iAqXFWTIlK0xI/LMZQnHFNrIbGFy5SSuLDdRNhFXGmlFWkLx+loDh0Nkez1kgxDB5QxqFAArmzO15DmnwUA+N9naZ+pP4VQhac34IfIbM4Sh0K2TpPw3ejd1sbaG8+qzWX4G+XyU3N8unw1pCjYNhFxIAbilqEqegglZE3g+b1xXRmhJJfdznblow7IxpEjmkyQtnQNJa2A4AGWZ3JdbPys2Z+ZgD+mDHjqXVmGsE8rcULDVvIgD0UKMpQ7Fe9FahFnK1vCVm7BOC+B2BVyq3q9t09JxTc+3i2s7bNnPSFs6Wq8n10kS8MEeONRQSsYqJhPpK4aBq5wxaoXRqNNZrNmw68igJxRiU1C53pM8IdJKxDScIR8G4F4HCLTUJSOy5UoIPBotHCQ+Z428jkPxOeX7kJlMoLSWdakTCl3Obd6YVn/LNXTGEojQU83zdU62JK19h/64/QFoolfXoE1AJHaVHY80my/mg5Jaf/IOQyC3OXPzITfMEV+XqlBmQ+fXLeqEov7LLAre69Y0okr7oRHyeYNNxk4+z5J5MedVRcTyz9nVO86qBYKqnh1Pp4tI8gaU9o1pjUT01wD8X+3/fhczf0FEzwD8PWb+ibrfltIaK3iaLeqxLvnUJ9oP+qJyczAAMgZlGcwPPl4ZDD6iwwvlysbqP24zk0dVv8oPKIctIzyNf9uUMuKkw99/BzSZIX3+eeWhFARQD46BB4fI9rtQv/zDlVo4ZTkv23iMfRmIJqIHm0Qe88myxahJPpIJ1F/3DileqtsF//Q3YToBOFAIT8ag0WSlrhgFAdTeANlPfh3zBxHU3CB6MwX96LPG3v6dUZZzn6nBAKT1WnpRLVyelWUiNKXzOEGVjWo4vY3YhAa+Znyui3S6yLW5Gsv435N5GswynzRpx5r2LtHp1syDFHbEiLYMFFLk721JeXabzUmxzZsYSdtepyE8JcnlvLrIT9k1bfRdNgvsHY2cmRuVWd8p3PMHZA0PQv/vfF4vAD9XyPeLeV4/fCgMhDgGLN3csxUKpUvU3gD83jOoyQyYzpC+fL16XadUq0iog1EHvNcHTs6FEr8JHbZwf/l2B08eg5NkWdihYgxtVbeu5nxlx5XqG2yDTYV4iut20zlvV9GrTTdLsGsQUfPgiVvrLIOAel2pyfrRZ7X2qj46Au3vwRwNoS4nUptx03noBuycnQmCENEHAH4GwD8E8ISZvwAAu0F7XPGbnwPwcwDQRX/1gHZDth3W7cLdpJwkIAA0DQFjQJkpDzM7FTbDoLokyTJs8g7zibgWpbWrLIWMiKo3Vc7Ll88lgHgSi7SgKieE25Q6ERnODNQsXpuoyoYlkjWdQym1KiDhjssMSOWeT5PaQsVzJHHO67Pmt0UPbd2E4hKY6zbWTSajOzTsmRlqNIMKtCyMo4l4qorHGVEoVFcxwkBBxRloPFvQ4JpdbJsG1n+eZfXRVkuPrKeZ5RSzsg2SzD19aLv3R0EANRzK9Y3IW9c6EJySpFPcamK85AxK1e+D+j1Pl6IgsHUnS87lNqqWgusTu8uM9FzyOhEtqCm5+1xyJrFZNizzi3yWiYCJe7aW8uXqMdWiobGwRL1cM/dyKhFwd1y+qy/RhK/rfGkSGWpyrR2B2W6yjIiGsal5/myWI3huDG0wT7vaXrWbwJtE4fmXGtoVogz+7ySW+ndxTvCpRImV0xQ8m0ONp/ZYm1Pmvh9PwKlZ6nOcpkI3nMWQUkjBkpPAja9KNklV/2IGJ4mMueLxZdhUkGdTA34N02MjbGobb9vn8nYUsDLXNe7PJbbdWmTZynxbf42cGGCcSARXKfA6mzXLgHkMmgoDaam9ja/Ni/V42zFeiN7VHtp0ISeiPQD/NYA/xcx/lYjOmfkw9/0ZMx/VnWMbQZCtcAueuRa7gz48WE4adV4RJTWcstdvanO59P6+GG3MXibVvFpfKBiAJBsbV8vu+sVjbwqu3ETT/Chn1F5rIrnv2FW5ibvEmrmKokjkoPN5mbeJvMe68Jz14QHMt94HmEFJBvrki2UxlwKc95z2BuDTc8mTWnP/+YK2wQdfQ/zeMcIzURIz/Q70m0vwxWiFQUBhB/RT3wRN5sDpudS6AcAnZ9JGZ0C5gqQ2B8flsvjcArKFVrUGlIY5O5OcHlcYN8sWnvJrjjPvKKoz7mziPsdJIyPQR862pfUXcjW/cmgYEQzee1dqAb45LRVIeFvhnKR190NBAPy2n4TpyHwcPj8FX1wiO78oPd6JRWQXl5Ir1++LQBQpmMvLncuVXwd3kRt5o6haM22uOdm6mc4hw3GyPgf1KwTV74sdNhptPic652Q38s/02pEzIgoB/KcA/iIz/1X78UsiepajNZYnmNwFcg/MCyRUVIzXDx+Anz4Ch9JpVZwCb85W8mX00RHocB/jn3wMDghgILxKEYzm4H/6a+XiF/kkzVte2PKFZgFIUmhZjaLi8f2+8MPjRCbOXlfqQ52crlVqcvlUnKRLxo8+2AdPpzCz2UKONVePzoyn3iPiEthpZGkHzNLuGqPETCYgm2jJSopqmlmzxdHYBcRvzKwxVtwsBs+e+AnLPDqUhUgR9OkYuBiVKj1REEA/eoj064+hzybCG7+88s+i9DlWKIBxlgHzOVQ3AoxZ/L7Cq62PDuS7zMBcjW+OmuNqrYTB7S+qu7rWXW7y1kVJkhTGXHP+uI6aHjOqcsbMeAr10RfyR2ZgxtPaqLDUyYtB+Q3cmjblBaHM6xN0ZiL4AgAqCGSjFSeSI2PzWc1IlMTUJy8kMX06kzIKrg0FMQtfOJrU4rtc/Sczni5q59m50wsbcK50yzXn+EbJ4syLum5NzukiZxvAbaKzk1O5XpURcg0n6BKlsrYxW1DfGjqvKAwWa2JtmQiz9l7N6ZnUQSr2r02RT/PYdMw2oabSsljGuk1kkz7JWQb90RdQNjptrsa1zBMTJ5bBI33LTCbA3Ea2k4V4lBNtKgo2UNiBOhiKQFZ+Hc1tMHg2l5JLbKyACSH94sXSc5ALsk8/yJdocn2z8v7VLeYK75JOl6e0Fj73ghZ5uXz3LFoAAMxsLvnA2zirjLBkyOWfr0ETtUYC8B8B+D4z/5ncV38dwB8D8Kftf//aZi29HVAY2MTKAiXMGZXdLpIHPWShdEo9zdAZr9KiqBvB7Pdx9W6ArEMgw+ieK0RdjUhROTPKiQLcBawcLfUlSZiuxpZSBwAlg80d343AkwlIG/GiRJEoK64JPTvZVmQ2gdjOzaRIzpmmwGwm51QKICNGA2dLExwP+0gPegivpjb50rgT+fMXk3fXJbLWoWxxkv4SS6KoTVrlQc9PkvOHPWRdDdZAzzCCOLEiK4UTaQ0e9DB71EUXgB5pkKWMoMqTrTWIeFUMgCW5l6JInsNsVk11dAuwo1uqksl4V7DqfaR1dbLsNvks11mQmhiNBYEcAOVtv2uYNXVsGkCip9c4QcVz5CTeSJGM01SERDaJKiypXo0rlbVUt2sdTB1gIuI82fm5vXB9NGrd3LExRWxbNDzPRnPdJnRDC+qEkq/l5pZdOxaJfF3S2nuxwjqNm78BPcoL9mQiIFErTFLMyStB4/zsBu1ytKetxqzPy6oSPlFWdMvOeQ1UTNeCeXkeWPOs/Jxm16+VGlJWMIb6PekneYVHEuEL6naF3jybLdGlqdsVhUmbTsBJCvS6YK2W15S84JHWi/7IDKmT6Zw0FYwGRbIJTK+5TtVsopfbusMNUtV8vovo2Cbrdl6rIP+bDfL7PG4r+FG2Hm+SG+gcQQ2MjbW0RiL6VwH8fQC/jIW2zp+E5J39ZQBfA/AJgD/MzLWltktpjTvyClSKJpR405yYBs/nspGwx5PNDSiNADhpbFv3AIDfiFRSRqqMxE3u+bpGagUlqfb4ojjEdRO9N/B++oT7vAfbT1I78iCta0/uPvTDB6AwhLkcyaDKMruBcp50U19Y0IqMOC9eYwnyOjR5DpsmE2/blDW0F9Xvg3pdWUyblgOwvzGXVxtTTV29NM6s/HOZ/LHSCN595mvn8PMX974OS+OaSneJu653lZ9/XAmK2aySNdGiApvM+fZ4sgbuWmPfnlt1o2WGxU1gV+IMdyh4tDXchsXWjvR2XpNixNeEfvQI1AmRfv5F7RobvPuOKDsyw7w59XOwPjoCv/sE3A3AikCJARTEqWsM1CRG9us/8gau/s43wZE4wdXVDJjHSL94uei/TfJ2dwDV7Yr419VVo/VZ7++V0sBVvw/qhM2o7E7m/x5QQfX+PqDI0sbr26P394HHD0BXE/A8RnZ2JhT+KJJ5JE1hnMqkdTY7p4KZir2tHz3wAj7Cllgoijo2z1b93dZWXaK/l8G2Sx0egMIQfDnamNJ8LVojM/83AKpcIddPINuZ17Ei6bfMc+gUnGxhW199vO78RqICTWVJ/bU3+XzX2NRrWnb8Jueouq8NJo119XV8FXu3kd6G27/OgMx/F4sEKsfJYqOQl8zPb4DJ1noBFm0zGcys5P7t4tnIcHCCBVb2GMBa5UnvgVUN6rMU7qVRnpp7D7auSdWRUmsl8RRLJ4FcB84yUNLMu7SCJAEbieJVJn6zqLBRmgo1xfYrpwJVq+zpJHWtlDSiyOdGSsmBgmoY4I1Ezt970+fsTtHriWfX1p/b6RxS3FQRQfVspNjVvappp3N2wbEE2Ph3vaJCacfIWqfGNshfh40VC2lQJ/OriqpNR8mcX1sTiFmEpMoYGWXnhlkeCzeJqvqIXwUUo2/XfBZL9D93rrLxm4joh6xXyyqLfq1jBs9jUKpQFF/hNIUaT8GxLe1gLK3U9lea5sS+iECTGSi2zsjpTCJqS1r81fLnu4QX/2oUbTR2A7N6rAi5rXcIqOFwIZi2Ds5JbNNEbmKD7gVkmqzxaQo1nYPzZTucLV7oDy61hZF5p7gXBHGqqmX9uqRmZbMbaR7dgrVv2LV9hxTQjaX0r4NbEwS5D3gbvW03jR08EzUYiDy1LUSZnZyuN/A29QRvALdgmfkc1OlAHx0CkIk6e/Om2sB31CE2lYnT/vB+XyKKnRAUhgAzspOzWiNdDYf+N+b8otEm1hUIB5EYtjWbRtXvg4Z7XtnLbOK0uGfIi06owQBquIfszUnp/avBAPTssXhxwwDJUQ8m0si6CsEkQzCKwf/kVxfvxW5i1d4A7DyKlgZKnVDyMxp4dIP33xPBiqux/GaHggOq2/WFxgHp0/rZ00Vk8YtXtUnhwQdfgzmUukpMkPubZ1CzWMp25PophR3oh8eyyZxOl3K3tsINju0vOzaJxgZPnwC9LtKPP2uf9VcUajiEGvRl82ON0ZVi6bk13uWh+xy3orPmvgi3vaVzSPDh1wEipB99urbtqt+HOjoUZ1WaSimD+4r7LPS14z67Myn924JfNBos2o52xnEMHk/uT7Hpu55w1oCiCPr4CJxmi/yMukiMFfLwCbuFY9fV/wGw4pkvRi6X4FTU9gYwp+f+vGY6k4iHzVvQRwfAw2NwrwP+1R/444J33wEPepLjNbpCdnomfaXbFVWts4vFBGXD03mFNrI5eGXSyG7j4+rniIEpycPZ2Xnp/S49L5PBxKgua1CAmc1BKgami8jZWs9ULrm3aVV6juOFrG3J+fM1ScxsLoVwDb/1CcNLohPT2SKqTqt14Mx0BvXFK4mcESF8FXmFK05TIE6QFd41xwbmkpeiBUxKRGxscey1eU8XlxKds7L1dVBdqW9oLi4bbfxMPqcD4rk1J6erkbOq3796A7oY+dy9fJQ0vzFTg4HUpRn0xAkQBMDJqc0jKY/wqsHAFyp1FCDV7S4i0zWRfSfItBAJsdLdjj5jiy5TGPp5UKiwy6JDrg6QH0s20qq6EagT2uPXU/T8dXvdpTmtFrmopp+DKzbmmwpQcZ1nuTBnmfMLYHR1fYPpvhheLvJvo735iOBK3SeSWlqwebXm/MLbGRRFUMM9q/ab+t9TN5L8R2bwdLZwru0PpS5YXpwi3ywrYEZ7g1rFw5V72ZS2V/abPHOiTJRqOoVJkvrIWa7PcByDk9SL8azYDbl/u3I4jZxOjq7q74UaKSpWOiMasoNcpCqrKPhe1s5r24F2nlnZBEPmXflH5hUw8fghKE6QPX+xmHuVFmGpk1MfOQPgN86bCNZRIMXO3btdeZb23VAo75O6XZ/zXuckXhIIajo3uDm435O529UlLGFkuPp29M4TEfQ4v7AslBq9Aif64tZtZnFK2zqWtdTNHWzi7uXmTCaIZgmxFEVA1BHKT5xUCy20WAJpDXQj6agNBCO8kIcXFSkeIBuHolhH+cmUXxTZKjGWdWJJ8I2WBS1s1XaGDDg13EO2FyHb60DnjuNeBDPoQgUaNBMjjqII3IvAvQ5onEvgdknSTtCCxNj2E4ZZnrypEwJBIMZokiwSRDehWW4i9JCjZfjnm1f1KsFW/Po1ixQRAWEIWLomz9/uTZlH/j3k74ts0ndB3W/j/DQuoZGwvFNfo2/N+zJTK6YThrXHAfDiPtB6sZGr62slxsomAgdmMgEaHO+MYQYArUBBTpSFbH0xLi6qAWCNZYoT+1sNYq6noRNZ4YBIDISrnPGttRjK7tl3QlCaAUbGNVQh10DRIsfUnUOJcwhhZ0HXXXf/REDZnFb3G0VgI8/FzcGV921FLhpL31c5Pym3ybbf79LpWSqe5K63SY70ddthRSRWP7d5Wvm1rBuBAy39pFBTk8JwaQyTtqIVbqwmqfSxKILpd6HqnKCdjuTn9rtAE8Vhl++n1oiaFH+mcs86fx7/2SJfC4DMi02Etxw9HPAUVkC7i1YbrW4NDnNKkjmjeKX9Otd2NybWtY1yQh9bQOaLhqIzOypB4eaZsjHj52hazEXe5tHKC66RIoANzLygKmvZG7RJbT77G0D6ftk64ZyVXqAp0OAwAE2m1fYRKYBq7I9cLUrXVifqQ1Ekzk5KwSoGwZZFyQd4lBN26wJxCoynMi6zaie3F33Jze3OqUhE9hqVP16kLQBb9YG3n9boFevuMBH9NmFzNkyc7MaLWVTLyQ+ChkViVSds3p6Sybm2vs8671NuUVk2qPTCA5gvFpuPDLnPCu2pFDbI9zVSUJ3QJymvywFrcX/hHQVfplouRNAPH4pymTHg0dXyQnoXtGtv9KvVeWfdb5ocm4fSCN5/B2w9uPzp56v37/+toI8OQMM9ZEdDqNEE2Q9+3Ow6mz7HGo/qUkRwWxBBHx8JhaxJ1OWegIIA6sGxjEWlkH3x4nbGY1MPd94ZVhItKF0rIMaccuICNlJeeg73M+swUFG0VlyAwo6c+5qlUigQ2jWsqJoroQBYOitRZaQvD9XvQx0fwbw58Zt5/egR+NkDcKilFuLLU/BkKlGLQpspCEA/+S1Ay/NUJ5fgsaVx33KdPVcf0LMorHMXQLPxae0iKJsj3sTZZRlFZVGy0ktY8SsYlr4Sx8tlEnKRHRfxksYYKT+05MQzixIdDebk4MOvg8MA2W/8sPp4a5upoyOYD55C/ej5aj52ExAh+Np7Mo9rBf785bKDtLBG6EePgOMDkC1b5OdBt2G+rihbgznfpYhQJ/TsIjOZrPzuraM1boQyb/994TI3QUPlvXxNDW5Ih1uLsgmAjXhqm8ImZDZuT4HK4ELrnKalmytYap4TAPE0uny0yi3ieYXCsiTgsr5SmhBfMXCdJ9Ceh9P8Zu8t6GvrUFegskitKHhab0QBzNFsnHBKXX5M3vB399BwHtgqSd6poXVC6bs3XHQ2X5Jj3QLqaI042AMTgUZj8fY5RdtNEqV3KVfsab7WA+k+du+17H1te0024MnMe3FNki5UuIpGL2fg2Vw2NkSAUwM7PACiSCgz01k5/WfT9tUcf9153VPD4mSniemLCxD0cCgbyCol1PzhlmYJY5ZyGsvAhkXMwVJNb03Eo+n7qzOWK9cKGas8m1uRhPX0OcfUMLDG9ZpjPe3+GmAjgldkWNbYPDVx3nxe58zWyspHI+ZzqMsJzOGe/xtJUtpmNgw9GoO7wm6RtIFmNaF2DTXoLRSGpzOYsdDYqGHEe2EXLWiEjX62iRCWE79SyrJ72DsAXLRs0RwGJbbWo33HbBhIUpBVLG/8nJnB44nk2B8fyWcuepdlsvmOY7+x5dkM+mQEhCH00ZFEH63oB2cGMGZRcxBW7MywbI6ZJUoVJ0DAoARgpZZpkMXxN5+LCuRsIQjlNtvrGABe3KoTSn53hQDSWmTZghWWZSvjqgne/s1ZCRa743tMu9pAGU9Zek52YXObblIKtyH/On/8tu2hnkyA1AmFjprzLqlOCIShfKckTE2GAeshMvOSvDdHUcmy63na6n6XX7juS6RlR84I1RXPYNHLpzoh1PER8ObEK09SYCljtlZQdgN90slzq54IVpDzPq00XAx9R3OQ6EOGpeLYdZ6yLSLQqhuJd2x/D3w1RnbTm7Mo8rRi2IWvvGEadLAPfnKMbBBBpQbqfASEAZTqecO9aW7H0uZ4R3VwVEdoYJ72k6aL91WMdm8LZmSvXy9fOoqAp4+gzi5XxHrMeCx1DfP1mt55gvSwB301hzodgV/UPPcd4LrzOtmalNnp+c2IH2ktzy9OxHDJ5VmVQe0NQIM+OElA8zmy85r7M1nzXJ63CcybUUKd468JZXwbinXVeSoiO5sIR/B8vjIPZpeXwOUl9He+KX/XRXNNhvTjT6EfHAPvPgZPpjurH7cp6PAAZr8PGECdXcpz3kT1cRu7aMPfGFvrVEXRMhOBS9pZphptUysapaIUkL18JfPNtz8EAgXWJNHReQb94kSineMJwAZmNIIZjRC8/x744RE40qDUgOIUaiYOHu5FQjE0BurkXNa5XldsPjYwOVo6jJENlCsGXZjrXJ/Lw222zat6hpd69ABmr49sP0L46cnW48srwF9jTn97N2c1CjulC8ZNRtOa1PdxNRH6fSAIvMpgcRBRFEH1usguF8nXZi6y31+K6EwOZjyRfDCb0Lv0XZwAcQKaLiTkqStJ/BRFVka1kLCcF77wXvq3KIq6La67KRsOofaHcqr5fCV/yMQJ+PWbpefLaQIeZ809iZvAReUAScC/Gtd79ew4YYOV/DCOb4bubGZzUByDip5ioNl8sOn1pjkDb03+mDk9h5rOEISycTbjqV+8N4qauYW+6niXkB0EgKJmdXmY5XnZchRL7TG2IPANjVWOY/DzF96bqg8PxNubpKU1bfjTLxC8DMUI6ITQT59IJCGJ7w9lMBftNtOZiC/UUOYoiqrrDBaEmuQz5esGAgDOLqU/FNfYkhInZjSS8WG95LcJCjtQvS7Sf+EbgGGEn53AnJ6VG1s2Ioj3n2HytX10zmOoWQo1T0CjiThfygSz8tH6uxY4uefg5+tpkQ7m4hIqSZGN72ZjBgDmxSvQaQfMjKyhoFYliKD6/RWhoZXDLL2UBgMbuZmAZ7N6JWTm3aS4XAOmGyDrBoh+8BJ8dYWsos4Zx7HYe7oLNZrCvD7xwltquAeex7b8iX1GuTWvNDpclY/oGQSLyL65Gsv5Sp4ThR2/yTMvXgFhiKATwtxh/wPe5s0ZUJ1YXH7wtSgji/OsGg8S/WryU1pJhC89xopN+ObmqXsVbdoo+bSpYlbTjc066lORcuYGVZUoBpGnOXEGgFNwpqH0orBm6bsvvfaO3vs2qMtTKHzv7ssrYW2ovFV5jTycCqWV/nfPS4QmAiDNFu8of25bM2oJVR66wvWWjl/TNneczyGsSQivxHVpZ03h6h5u8p6u4yjYpFZgEu82krnG6eRUDxufru693qQThXnZOHeCIIZLBTrM1ZWP5ik9BAIN4rBys+rV4IDG78sbBk36ucujrZrn3XxaNd/aSDLNqwQ5RLDBi7MA8ly09vQsnkx8Hk3eYUJq4VAhW/ewkYDETcG2OxmGoNQgiBb5P6XQGqYXIt7XUEkArQlQgIpDIA7tOiLPloJQzp+7fwaWoxc3jRKBhFrccW7+JhEITtPqKOquaNZ5B1q+5qilWJvZbCEuR7SkqEt50ZKmNSdVQ8E0q+rJHcsaajKPl+VAAs3atUnObPG8hkFxAko0KFAwF5f1G8k4kRp1WgPT2XKRabuZ2lXuPmm9JDDFaSoOpRzLxo8bRSCWOXHpvddeIFcywlImPa3TRc3q0jDc+K3pD2+/IMjbhutGcpy8d16iejiUcOyLV+tpAEQI3nkmNcIKtJ/icWpvDzCiRFY1aCjsQO1bPrlh8Z7nO6TS0AdSOd7zktNUZL6Li4oL0YchkCSymJKV+7WCIZvKRe8Um2yClUbwzlP5d5pKXbLCRBs8fQIEgRRifHSM9LCLYDSHOr9C+vGnjZulHz4QGlddsq19n+rRA8TvHCL8lR8ve//LFj0i6MePRB6+it5S1Z9zxZoBa+gWJjPnUec0lSgGWdl2ZwjeJH13W9gIeOPC3jmofh/UCZtFmFpsh7xaXBny/XXNXKyPjiRab/MPstMzvwEpzgEURTJWjDhWslev4USDqmq5qW4X9LV3QZOZeJzX9Ivg6RMg6ogMf4mggm/3/r4wDMLQGyX+3HchBLMO20jBN4QX+2k4ZzsjHRDjikJRvzTTqRf3wJNHMP0I+vwKfDWBOb8QirdTeZzNb7akDxGCd98BAsk34rMLmMmkVqacIpHnl7VoC1GGewInZ8+Wmr1tn8mvmcHTJ+Cjfcm/ms6lxmWuZIbeG4AO9mFei3Q9ffPrYDd3/PjTUqGHa+EaG2kvRGEdsHVy767kQ5X8vEcuvxrAsj2YT88pvou8AEcuV73q3lx7YMs1NN74NgXZkhhBAAQa5tWbrUWy1HAIJAnMbAb93W8jPR5gftRBdDpH+PkZkBkprfTy1UobqNOBOjwA9br4mz/+M/dHEITCDtSg571OrlbUTidmpxKzN5BIRFWuSslvqNdbLHz5yIedsJVT7EnTglRnrm4KWylq91mns2jDdb0ZZOl900V0heMYPBovJxDnoiTAshQyT6oLyi43w3rFw6CURghgkfAM2GjKwiXrilAC8MmfToJUHR2BR6PlRYwlIZlcoqiNorjCzumLl9tFU3aI5kmzRjjXwCIHrnjIbCYbkjSFOh8htJ4lnk4rT+sKJvsacYbFI7SuXSSqkpjOEJ5NV5PNyzZXnRDU69oJtaLcQSAePtobiDdsOpNJ1d53WTKsy1XjHF2R48RGSo14srQSwZX7ZkgCW1O0ZKHcou/mvXT5ROg10Pv7nt7RhFLj6tc0qRl0b9EgSu7q8Kxz8EgdNSM5ftOZzE9V7z7LJKrk5i7bFk5L8h3t++Q0hbq8EufMGvEHQOZtJIn8r+Y+3TtEki4ohe7693E8cXnZHAoCqIP95kXXS1IdlsSi1tFbSpgcnGAhtmAYPJtDXYyg57HMX3MryBHHXmJ/Z2OnTF0YkLaMJyKsQEqeTbFf5hwPbFjW78lU5p+iwrG1U2DtmTsd+2scJiJ4YamzdWvemsiiFzohAs9moEvtBUiWniWLkY3JRJ4LKajzEVxpjewmnpVlpazAOq+p05F8rI7U0TNvThbUxiyTNfudJ9CTGTCPxQHci2AO+lBnV6CribelyImz1M2bbADObb7yz30Ng2aFRl91b/nb7EYyd6UbJsNZ+qibp3k+Fztm0PfrLs9mMk6VbDYVDRZRXeblEk+uvqQrpu7alWWygXR2/2SGQCmoWQJ1OQW7/Li6vuHKwtTg1jdnatADHh5Lop8xoNkcZjwB14VDNwUpyUt6eAw1T+wOt16e1BmZ6vAAPBohO48lcR3wnnywAR3sy+Cd2DoJttNSaPnCvZ7fiIlx2wPt7wHzuFlyq1NmrPAakNZSdC9JFpuz+Xw1+d3KpNJwT9rrNkHMzZJ8ba0vBAEo0KCrcfkArElKVnsD8d7OZr6ALCCKOHjvGchkKyFkTtKFcag0iAjm8ZFM2i9fbZaUu2s0mFjyx657zvnIVS0dwIEI5J5pHMumOE5sDkmNVzznmTYXl8DVeDmHqQSu73IvAgELWk+xSXZjZt5/DH0ygjo5g5kCTq2qrB87ERjO5XHkxwaFHaBzTxnX6yjGdT9N4s2Tr5330Y2HKBJRnCbOlQdH4E4INZlJQdviYleI/KgHR+DR1cKTeh8N+XVokPPmnQmzee14lvptzc7PFRGJUm9yEEpfSFOkL14uDMkqSrj9vKlgxo1GbW4CFc+Uogh4/ADKcCPBHVeYeIUqT7bGUYLNc3PyGzYrqmBerD5fV3tzl3DO1TImwdo1PB/N8MIPse9/1AlBWiNLE3FkD/fEYLxjx8w6MTeezxs9ZyduBSKZLwsiQ56yHISS33+RG1u8vPkoUt3S559velvXh3O2D4eg4QDZ0QDJQRcqzhDOY6jRCGZm310UYfLBAcLLPvQsRTrsYPowxNW7Ggc/7qH3xRR49aZcJKQMbs3btF9sYi/52ySpSxmkItC0yW+1hjo6lHzmJEGWWLv8+BBIswWjwR374Bhqb28paEKdjvSZNLUOMyOMl34PvD8AzSRvLn0p51GdEOb0HHR2AUwmyFzZhaDGfjEsOXBr8vVvndb4s+HvLchCm/qQ6rawnv/GFCkXnnXqbnnufv6wILSGp21vjgLm1OX89co+a4LayNkqrbEMajAQkYdIVG3yE4o+PJD2xDFMLJMz/0s/CROIly781Y+RnZ15qk728AD06z9eKnpIWvsE+ez1awTPnoKHA5j9HpCx1DR5/lI6oXtezBI5PT5E/FPvofNiBLw5B1uZ2tK6J50O8I2vAQowv/zr5RtWW/0dSvlQc/WD0dB5GuZodG+NUAqChXez8FmpJHrTvAOgeX5aFImzoeqZWmeC6nV9f1rkuFRE28KOFHCsouHcJXV1R3DCPs7ZkH9++bp+qtuVfJemC5GN5uu9AV7/z34apgPoOfDwH7wGf/zZ0nVUtwsaDpF98xk+/T176P2ON/jw8BQfXxzj8p8+wLv/dYzuL38KHl2JZxEAlILaH+L8f/Ahkh4hi4DOiNG5Muh/fAl1OoI5O989lec2YCklnrVxV7TZ+0gtdLgBMZvrtEX1us3pa3VRl9xa7vrAvaRNF7GDHCtXssblDprpdHkdyNPP3Gd3hV2NjWLUkRTUoC9Ozfkc+ie+Be6GoCQD0gyUpMg+fe77mX74QJyToxH4g3dx/lP7yEICGaB7nmHwgzNk3/9NuZS1aXh0BTObI3j3mRjecYLs9GynZVby5Ydgo51mNFpyplEotDmksvY4ijN6XYm+T2c3poxKUQT99DH47KLyGqrblaDFoA8e9oE083UlSWvo994RW24ylRQXV/h7Hos9aaNgZjLxa5ybs+T7XAS4EDlbcQJbOyR4+gTZ+49hQgWVGOhX5/4e3POm4Z5Ez7SWUgJhKMXizy4lAJFL3Wg6F/1t8/+9P7TGW0sSbuoVcHDh2byEbcmDzXufPDc9TUFBKDk2kDC6HGcrpRMBoa2JZAv9uaRiIlqEpOv4u7k2+XBvk4nMGsmq281dRwEaiwRpRTCRRhZpGE3o7O9BOdUcsxqm8nTJTghy19caCAOYjpaoKGC9DxULIMMuCmrB3y4eYhhkDMhRfiqiN1C0SLjNGtRo8yIAZnHOJs8ypyJYm+yZO76UluK+A9Zfs+BdaTx2ysRIaugSVYWB2Xkdq2C9sktRmTUUKt9/qwzBvPPjLQXZ/EpfO6jwHTv/r1L1QgVFmMyXlkgHhKwDGA2pD1U4j5nNoGx/TfcYP3H8Gv/ywcfoBzH+m/1jZD0NUkrEG2wdKpfsnvQIaR/IugTKADIK3AlkoVRqdSzm5sS1Uvg5ylEe7p17pU5S5ZFX56BQ5MskuKiJvz6wKK2R/6179myW+1/dWG2KpmP6GkZ20VlT15YmTjx/eI66w7Flm5TlGbuciSiyjk8xghzNK1+099rYVCq+7pr5/Jj8+KsCiXIl9bqgMIS5HNWmX+j9fSCKQIMeeDoTQ30XzqVdbpJdKQ5gee26Bhtg59jV/a6sdca+d0g/CDQ41PbfylNEPbSWz7SG6WgkfULWkXI+Olby2xxI5eyYwJ5X5665I3j7ucYBzUlcrydwgyAiuxbVRIWUfd72HVBu3uQ0FUl9pXxuGNvNGWWBX1e9fWDtPpAGpyUlYtjk0m6WbdmleTQIkEUaHCqwImil/D24FCalZb1ERPJfrRZz/iZzXsPjWkGQLaCGQ9CzxyJbPI+RvngJ/c0PMP7uI5iQEF6mCH/hl6AfPgQeHcEMIphAgTsKnY/ewLw+Ab37FNzrwHRD6N/8DNn5uQhn2ChQ9upN+SJcFFpYQ4dz8qz89XdAL05WExRz59VHB8DDY6QP90CZAf7Rry4b7I4S0Y0W9cecrHSVQlhVmw72weNJIwqOS3IFsPv8RGsoUqezlvqq9/dBx4fCO55L4nB1o0nep5aJmidTHx30nq98cdY6sZESoy+fe+QM6jx3PHjyWD6z3iafQ1kmTNDvi3xvIrlJZjKBPjwQii4Ans3LpaRz7VP9fqn4xwps/6UwEC68LfBZdaycMGtu+H6F0Fg1zLIIEIbwxdwtLbpqLOXPLRdY3nT5gtbOO9kJoZ49kT49mws90o1VR9W2fV89OLby+7YYqTXyzdUYSBKhprik7S9ewsxmS/09eO9dcFcKheLkHHxxKdcGgNl8sRg/fbjYsDKDZjHSH38swjiDvuSGzefILi9lbPd70l47BjbZZPh5dja/MQdk8PX3wZ0Q5pPntYn+gDAn6Nlj8Ms3jSjT6rf+JNKDHmYPO4jOEoTnM+CHn4pAj7sXKypE7z/D1XcOoRKGmhtEr8aSB0sE/eoMPB7fn1IDW0Lv7yP5rd/A+Xd6mDwlvPd3rhA8P0X66Welx8/+J78dFx8GuPjpBPu/FuLJPxxD//KP5Nnfp0ipHas7l1/P53hdt+/ncuBuhFUFeFtG7Q1AUQcIApg3J4uxm19vnFBMDrXt2kG0861G0/5e9ZwqnPUrx+Wg9/cBlFC/nYDXmvnSieao4R6YGdnrk/r365w9NW1qgr/Nf+X+RM5uBI6u0pAvvTGVqACOY9DZpXDRE1t/7PwS/Y+lUCzNE2Q2+VCdXkCPO9BKgQMNvrySSvPnI6hJCBUuZM15NpfzJUF15ICNeDYb1pfiTOrf6JNLmEm10ARMBp7OoK4m0imMQVY05G2iJWzCuaNnURQt1I3WcIxVvy+blfFEFoiq47pdv3Fx1DCQqo7ibLsAsgEbZd9jfTIbxzEwupJ+ti6B371PpzjpEklho272evn7qTQG85syF4p3/c79zt6L/8wlvgYB1KAHTkLJMyMFEGQhteIypCWJXkQjbB7jbL4k2uN+B6B00loadyX3oIZDMdxjmyuZZYBOfc2p8me4AWWziCb9wRoB6ybu+4olumsUSRS+zNnhWAQ5aWgVRbUR0XXRmaUkcrtJ48uRGFSWr+/PnxN8YMNCrXGeal/jjBdiJOOJTdpeMAjyY4OvroC5KH4aK27El3YDkiw2RurscmkBZVcr0q0TcS5v19bWcTk36zZmRWEWJ4zkvKyeup9zVlWeMy8m5R9wLqndfXR5JZvWdQn8ADhOQBdXy7Ufa6BOLhFO5ghGPdB4Jqp1cby8DrG8I3V2icFHgf+MJnNx5qWZUNTjZKHw5iKt1hHl/s4LZgFYphnmGQq5eUUNh83EvQDvANs2emXmc4QvLnAIoP+qg+DlxSLRvwT9jy+hpwN0LjvY+zxG+OJiUR8rd31nRHKarnUG1sGXbVjz+xUBIZNZoaUSRswmkdkiKkRdyhu1KtxSPJcXu7qpBHPLQHLOByJajHv7vUdVyZ+ac7+1uIZyqktHoH5PitTXON4dHZH2+gCA7PMXi3V4y/FaCmbPrqgDZxkwn8PY39SOqzXMoF3hyxE5s3LtPJ02isToB8egTkeSsd/mgbRjOO8BABT5uWVQg4EsoookOtNgQOsnjyXXa42Mv35wLFGfJkIZNyjFfN/gpITXcsZtgWB1sO+No/TFSzh5bygl73u4J8bUfF6df5dT9wLWG+5lvw/efUdKKFyOrmWUNL0eBeFaeWAneiJ9d8eyvbcMJ5/eRCZ7YxW8LxkoiqCiSBb0bQQQLOWtLorvHICktaf/memsPHLd7YqTpNv1m0lOM5E9r4tY3wVyTg/V7Uo0NuqAJ9PlnGRLffQ18Jhlwxh1AFLI3rxZpALY+kDZ5aU33FWvC58fZb3UwXvvijz1q9eLZ1LhhFGDAajfl/FwXxwvRAi+9p4466YzmPOLrcefGgzqc4Ht9ZoUP/bn7PfFab1LcbayZrkiwV/BuedO0cBh6fL3t8kp9uyD4wPg1Ukts0g/OAY/e4z50wHIANEv/QBmPH078kCrsEUkrS5y9tZuzoobCXZePku5kcK6VtUujpfD0tsamUsNsBS/Qr2m68CJL6ycJ0cL8seWeLmC998Tdb3p3CpObkAtKSkb0CTy4FFRZLm4afJSpWUqZi5Z39IMHb94pXZa7vxq0F/+bJ0gyHVxGxSVOopj04LT7jwuFwdY8kzLf9XCu631+gm5IV1D9SUBd0kEw1EvN90E1eT5+dwjwOdvukRp6nZhzi8qo+P5vCV3/kZqqvcVGzooNpHkb4ycAJKPpFfk0aquSEEvOQRyokzr5lOXeG6Ohkj3OtBjG4ENFNTFBDSZiQiSPbd+9Ai010f2cB9MADGgJjFoMhPq5KY0r3XzwAY0b3/fenkObpTXui1c9LQhBVN/+xtIHw6R9QNEX4yQfe83lt9tSc1K950DG/YRe/vBcvQ2f69lVO48lbbu1vKCY+uOXSNOtBXc/F1ybW+zuPIqm9C/tslZ3mS9yr/PTfqepfBzkjZnIzVsFwUB9HvvgC9GjVSm9dERKOogffnKn780IpgXvHJ5S47KzUak9gu2hH5wLFT8yRSuJJDPX7cMpSY2ZfD+e+AoBGUGfHZeb6fV2QI3gXX9qk5NPJci4jfgNjJZHAteZCOK5Dla5sFbpzRbgBMCo6NDcLcD0+1AXU1BVxPwQNJEKElh3pzCjMf3i9boQp/em5ZlOfrU9t7r0omQbH2mIl+47Dq5BXItzzmfe0UkdXBIYW0B1OL1gOsNuCK1UZFN2LcLYN6Yzie7VhkK+eT4JtjCcGhkPBqbExOGIIgsrq/rVSz+6nJojJF3t4m4wk2hhLbh6pOV0mwKCxUFIdT+Hszl1epkvyG9onSx8GF5S9fgAI3eeN6IrkNhvAE1jpDiOChbtPP0tyIULdddccXO10CMwgzI7Dxx3X5TNFRzhmmxCGfeCPf92eUDAtvNCRuWmLjpKDN1FjlrXiyiyuCzUV733jxFvaj+mTveL/6BAjTJ/wAwwb6Lkj5ABNYugTt/vvJ376miZW0oGcNLAhy5MbYWlg6+EXXKXnOTnLhrbfZsgrwJaDHH+naXHF/xnftbdbtgJsAsKEcUdhZrb8l9NRVCAamby1Vad23Ar7X5jZhHntqeo5hv/G6azhGbzCUuLUHlzMLiWrYr5+SGDrqmaRzbgErmCtIKbFbXsRWBDysuAkXAvGEbrYAZXB9YBxtZbnbuGrXVJs7V67xbuwZxmkoee4kd4C/jNmyTCVwu69pAUXGOXXOs2tsTOvX8dkrDeFXUIBARE7LrkrMvXI51Q/v6ViNnB+Ej/lfe+3eQHe+DQyXey7Mr4asbA/Pm9FpV38vgF3HHO624XzUYyE5+byA1gWpoYxR2oPb3vGBC+vxzMagHPamZ0WCSVf3+Sk7BTmAnT99RAPEM9XKUGUtBrKXk4I7r5eQmGedpwsUVeDyujnDck8Rr1e+DgmApKqB/6jswUQj6/g+XvfQ2+ucLn0O8/PFveR/Rr33eiHpbGY28I+Sl4muRK9LOMyn4S1HUbA7IJWnfORXC3oca7slcczWWDUYntI4GEc1x3kK1P5SNgjFCbYpj8TxnZlUlr+iFd4n32+aGrMO6fJCGCJ49BXpdyQGLY2A6s+wGs+LB1oeHMB+8A3IFut0iljHoxeuV/AV9eCD374prlxUrrYtSNTkOgP7Wh+BBF/TJiwVlvsKpprpdqKePYV6fbJ3LvAk2oh8RiTiVFT/Z/qI7Ejoggh4OvROR4xhQSkSiLkcLeqO7TglzxEeK88qgawoPF9tQGpFosoaURfWsneFzl9IUwTc+AIcB6GriVY/ZGJkTzMIpTYM+eDxpVn+07FncIPJCC18lOqITIVvHpvAqp52OiAjt+tlYAS2pZVozJ7uSNoNeeQTTiULZFId1tt21nTk3cG7V7UI9fABzcbksVAQsMUco7IC6EbLf+k2AGeGPXgiL5ibtWaUlssoGyMyivAFQO5/cn8hZZsAXl9Bp5hdgnk79BMu2COIuwVkmUYE18si+MnyWVScX5s7J05mIOgDW42SFOhq6rzmO6yXKt0UuQb3ykHzia8X3d468uMVoBEy0bGDqEtzvwcYMwBLF1uPkHDoIkKWF5FQr9Zp/HzyZoPPpma0Rt/6e7lJ2Xg2H4mWME5+Ds1YsxYHN8rGkGs0BFHag9gayGCa5TUuZuID9Tk6/oF/5BQK4/gJk74MnU9l0xYnP6ePZDEjs5sHVPPS3q6DffSbSw4GGurhCVtyMW3aBV5N6eCzzaJIs597sEjtIwjdXY5AT+LAiHSqKFnNlXkxjPod+fb6QI84Zv2a+Ot7NdCZGcEWR8/p72+DYswuo8RQmbwhU/J7TFHwxWii9NbmO0tB7g8W4aUrBc/NLbFVZrUiIuRyVbwyZRagjy6rbVvL5Co2w4T15KmNV+gAzzHy+iKhmduNCo8XauzQGSsQm8p/5KGXJZ1VwrJ3icU3uscT7LcZmKu/HRccuLgElmxnvBDc5O8SVqbFR4kZoKLzhaeQNNnEURVD7++IoKhyft4WcLeXnzutE+u85vAhZ0+OaiIRt1ZCGdqXtF7KJs5t+u3EEsJ4mHgRSG806GMRpWCPWBcApfHOcLDmX197SlvaKE34q2kvyZU50yorWhc9PZa4ZT9bbtDVU5KXDcs+U03SxGWdR/kWeLQPkSo5sXlfxVjdnzCz82tuU2eVmNTxEFRC19SM8zKpilFcxbDhR3fgGqHDfm1zvfmzOFs/xbZNl5jSVApA5VJYwKKEemvEYsEUZm13w7hZHtT+UpH+rZMbpBsqHjvo0zzYzJjohaH8PNJ16KhhpLcY/sKgz6I63ojVEztMumx24SfO6RXftfSwVuNRSp8WMJ/55ENm6fk61Tmskzw4R74dQiUHXGPBnJWPPbibVcA/psyOoaQKaxqCT092P1R31pVJRgb2BGORJ4t8DAJjpFObzcpEYtyFZauJ8vtPaQVVoIq7iwGkqkQ8XeWxgQZPWoOGeFFudTtcrRFplVWPp3ZxkMA+PwL0QFKcSeayI2uXFOspPrgAsjwPqhNVGXcUmj8JgQRP24iarG+iyd5hV1sTk1edZSsvfsFdcd8znYRxdcfFR4/6ziUe/7Fk45N4J2ShJ7Xxq34/qdYFHR1Dz+co7WIoEWZtCokpYKCpizfx5TxgtG6EpTTxH5bsRVKUmlBxXrNOrokgEOpiB6aw2ckRBABwdAJ0QRpOoUqN+c0ZaA0cH4sAyGzjKtuwLnKbVkf/8OY08h/SjT5qfnFSjPGz3TAGAZzNkOWeSmc/FiZqr00mDPsg7KDdzJL61giBvNSyVh/o98F4fePm6sdrhnaOO0/wVwXVpny7sjiRZpXlV/cYJ4PCqxHaT6xWT3vXDByK2UqjnsamSn35wDDhP23V43XnDwlJyqRvBiXYstaeMe+6MYqC8fxaoSE3FBbZGmSeuRCRB7Q2ktmGghdpUtfjY86l+H06A6C4FTBpRU6yKrjPwfTmJTeePt9G4a4ilKEQTOl1ZLqvWcAWGd0qV3yYvumzz535vI4XQujGFT3Wj3de2bALb1qYiDxsJNTWEmwddeYalc+euR2EHatCD+eZ7UBcT8Gdf+Egdae0LaWdvTiT61elAffA+sqM+1C//UMridCNfkxOkwLPZwsHicoJcdGQX+fL3Ca7sg3VGrF3XSVQw750aaBF5mu+6uZpyImxAeQH6sp85EbvbEiy5STRZZ+qeaT7nNJe2Yg8uPff9oTW+DbhJQ6CSUpKrTt8wvLpRbohV5mqsSPZlm3zfdlgBBFeTyycl5419p3LnqDI2iXiFout+41SqwkAixvl+sSSwUb8Zd9SHaxtP6/pavk1l3sQ6j3LZ+YtKcbuGo005OLGggkhCdnnlFa5gzIIysTLxS42/zBtMCvroyNNebrw8wbbI3exGnsOmxXJvW82sKRquIxuNm5I+7g1Jl4OZF8RxCpBWNGetSIbNb3HtXooElzh4fL5NGPgaia7Pr3t3tV7qTdbgXToLc8+wTCTiPsDlffmolRPLMAwYgAzL2uDmHqcmWIRlDQAQB+F0lit9gMU6Uob7NM6+jLhG3q/PfXSb+gY1PJ1YGWfZxqq1b0v+YSNV1ryjN/93HvmIKtHCaa7Uok5nfk63zhPq9qWMCACex8KcIQXUBNe/mpGzmoe/pBi142tSEC4ZheSkWAHPSVX9PtT+ULxcNQu36vdBnVBk5te8Q9Xtgt5/B3hzVu+xpIUU9lqO7JfYm32nKD5XlzwfWhWgq7GvW+MicNTviajGsL8wrLodqPEM/NkXS4Y7hR2Rej3YBwINnkxFcKGCDqX396Vg6tssM3/XKBF9cZ8XoyA0HMpnyTKFwxXxdu+Jwg7mv/u3QqWMzusx6KPPryf2cM+gDw+AJ4/An79cooYWoQYDUK+7VuTottFYFOcmru2KwUYRaG8ganLTGXgyrY3Mqr094JvvC402yWB+8yPf/uDZU6AbCVXIzjHBNz6A2e8jPYgQnE2hTi5hLkdiID48Al6flFPS7ZzGcbwapdhE8MKex8znOzES1XAoNO1OCMSJlGK4R6AggDo6AvW7yD5/KZtSJ3Z2Hx0zXybcor2ziW1XRPD0CdDvITveg359AT67qK5faqG/801wNwRN5sCb07cujaQW1u7WD46Afg/pR5/WjhVXdoE64dr6gBRFUIcH4KcPYKIA+vQKOLtYoTMH77+H9NkRxu/3oVJG99UcHChwQPi7f+dPXj9yRkQawD8G8JyZ/wARHQP4TwB8AOAjAP8WM9dyFagbQX/wLdH5fyE5ONnPfAfpIEAWKfS+mEBNEyBJYQ76SAchiBnByRTmV35Nalj0e2KozmPweOIVwAA031TVCYM42s01sbLJ8wnIiyRahoYKbGK7SyKOY5jR1ZLX3RXUNJeXfiIuFZ0AShcsEyfQp+fg6bS+0bxc2gDIeeqAZa9Ak0nDbvao01lLTaEogj46FIU3ImTPv1i/6JAUg/XNdyIrTT2pm0hRV9Wgs3DeJxiz1vvkNkjohLbQbG4yzLVH7+8Dva54weIEPLoSj6aiRYRlJoV0EQRAHMvnSkGPJuAkAbT2pQhEBTGBmQIKsOICiRecWKo/ZPuWL5y7dLOrAhuNYGlKZVEepz5H/T5gjWyf+Ft8lo6CoXUj5ak7hxN9KaoKFp4rp6nkDBmDotOMrOiDPzbL0PvhidTKmUzlPVm6tH92ayJVK17Wu/SCFsaimc6g3pxKjUpFoI4teBwE4ihwVEkr5LRO5Oi24dWB7+LauagJKQIHWijUdRtXZvBsDv3iZJHAnnue5nIEKpQB4dMzqMkUnfNIiipPptLvVAz1xooTVFzLTGfl76uh4MXa82wBns1hAJ8XetNwkUfJ010f2WQjwi7w0uA55cdbxI1Twu8jbtERXWnbNYAZXYHmMfR0JmJ7TXQQTs5kLYiT6jG7De7Yge+KqhOR2EKXDbQnsgwgBuZm7bzCSQoeXYEABLb+XZmNzReXCIzB/tUcyAxoPJWSV2ui85vQGv84gO8D2Ld//zyAX2DmP01EP2///g/qTmA6GrP3D6BnGcKLS8Awzr7TQ3xISPaAw/4eOqMMwTTD5EkHs0MCGWD4PED0KwDt78EcDGC6IdQkgVJKjBmnkhNj81oxK43cwURHJBQzNsvrc2mNHyXiEfbA0oJ9/T5wOATNRImHTVYqOiEHK9BwDwpYJCuarHlycp46QxLZ84oz6xb4krbAct5h6idzFUXgo32kR30xMl+fWOoGKvn8jk/vm55lIGQAGtCcNkjcd7TQug0m9XpwxRQpM7X9kDohaNCXjWicVArk0HBPchKTVCbOycSH571BnaY+D4AmE1Cv5w1YX5S5E4JMIP3BUgKzoix1mCu94CgznJV6jnwhWUenaWhN+fyHJF15PtQJ5V0eDkGzWMR1fE2gkvfuCl0CmyXU3wWqkrqLIgnFsZ9f3BRBinhZmAxZQTSGwo4IpczjRYJ2DZwENKIINJ8v5ovbRslYZNcelw/SjaSd3QjKqtu5vs/F27xOrtSujInrriPXaY/JROQmpgV1zdIO68BJLKU7yk5ZElkXp1L53LWuL1V6pNfRk5ueZwtIX7rhSGfOIFNRJHNzvFhXK8WJrKPTzOYA5zbJdc/qhoxjR5//Sm3ObhGVtl0DmPF4IQqUz8OuwSbCR41xD2jB1Ot62jPPY1+upxZOWKWhGIyZTKRWWw2yy0vg8hJ43rztQENaIxG9B+AvAPhTAP53NnL26wB+FzN/QUTPAPw9Zv6JuvMcBA/5Zwd/ULzddoeujw8X9TMmU59cSJ2ORAMAX59F9fviber3JGH14nJ5d1t2L7n6B7eKphNj0yTEqqrsJefTB/u+2nptDawcxz4vtlCk1S1h08neG16y6OijI7meq3vk2pWPhhDBTCaijDPoy/dZZpOzC9E0t3Fzl3NRpZxaWD7itXXtjnXGUr4d9vxqMBDaYaeD7PRsWdBCa/9cKjd8LmpZ9KoW25JPRM23If/5umjiNu95m8W/qlC7y4XTekH9rWtzfuG5Q0rPTXiRHV1VPTwGn18KFbnKQREEMi/uDwGtkL14tT5BOy+0YAtmLz3r+5hzWozWApXtU92ud0rwdLa8sSjkKFHYgTo+lM2fUkg//mx5nFXlftyAMJI+PPDiMObsvD4i3LBukYuIb1xy4G3GdQu6N73MYOD/XVuHys2TSkP/5DetI20GPjkTJ5plZZQyLizdVO0NkH79MfTlDDSaIPvihZ9zyvJpgnffAcIA5uRsmU5t2SbXElnZ0fyg+v3tmQ83FZXJCS81Fq6pO50rZL/mHn2d3agjG4k0LVe8xcKpZgqR7K8Mmubk5W2ymxaxsyIxaxkodu2lfg8URfibn/1frk1r/LMA/n0Aw9xnT5j5CwCwG7TH5W2mnwPwcwDQRX+lw2VvToo/sB1vdbJb1DQiS9tpMLmwAaCXjbnbUBtsev4mx5lso4igUNVc0csm518WW9i4fbWNKXhC2QBGrbaNeUVemTMDcoaYYZQmNhfUcthI/TwyRgwXmzy9cr1t7qMOVe8nyxaFR3PnalTewXnQcsbGkrG18oNCv67zrhaN/S3fsxoOrWyzVbsKA6kBOJ0t50w5ioGVkPfH242zOb+wqmQN+/mGHvZFQ0qK126BPCWJk8I5Sgz3FQEEZ1wDcOqL/ns2Ql9NswWVsY7uxCzjhLkZjcWeX57Bls/xtlEiVV4HeSYN3q+jhmXyX+kbhe+rQAog7G49sY4jNs28zmznuToH61cyunFbdNJN3znbMU3kI0+cpnZuMPJ9WcTBGLAxoLnkovqxmz+vKURIssV1ys+3RX91zlzvPFufcsBZtrz+LglP1ffduwQzg3bRjwwDusF49muAWcz/VYcaa9+0qMdtO225wkZdOsa+W/u+67A2ckZEfwDA72fm/zUR/S4A/3sbOTtn5sPccWfMfFR3rrWCIHZXqR4cS0XvoufAVUHvdcuTievuw1Z8p0749sjWf1lwh574dbli9x4uUT9JYGYzka4PApjT82t5w9VwCCLaXkTCFX3+Ld9CctBF1lOYPggwPyAMP8vQezED/YN/7o/VP/ENcBQi64XI+gGyrsLsSEPHQHiVof8Pf3gz9Ipis4dD8U5OZysJv03qnDgEX38f3O8i+/UfrTgIVL+/dG6pUTaUotnWq6YGA6jHD+U3xoAvr8DT6f3Pn/uyY4O5SnW7oG7kHYVv7Rxzn/EVE57yub+WIn+fVEj1/r5Q7WczYeZUiEgBNiL94ftC2bwayz0kNXWqWrS4SzRVSd8xriul/zsB/EEi+v0AugD2ieg/BvCSiJ7laI0VVXY3ABvZdJ1fAEmySJbNf58aGJtz53NOHLRs3krDvYbByMDTbPMEYlefJYrA47EXzfDGcWEBobAD/eTR4nOXGzSdSTQHgLm6ujnq2H1Dji7lBDxuS/2vMjHfUQtxDarjbYBZEnutR8ZcjZtTXOtOO7teIV9HH1WvLqDPZJHu9bvgbgB1MQFdTeBHLhvgxRtQECAMNMJOCGiNfi8C0kwEgqygReMxQUoo0VEkGx5n0Exn4vUkWqpRpPp9KDsmWVmv7mwOHo9lAw8gOz9v/EzN6TnoKlztWy4CXBDvMJPJUs4Pp5JMjDQVQaA4XpsTdB2ofh/U6yI7u1grVkNhABoMwJNJrQF2n0FhB/gXvwNKMqjRFOZzoYHpJ49FvZAI2fMXgCLop4+FVq8INIvBiRiUIqCz/E704YEcO51Z9UxXPHp1jqEgkBxQIjBzJU3JH28pUAhDL4ajet1qyhsW9QqdWBDb/lcG1e2Kutg8lnU29259se/chkAfHvhxZS6vFrmtQQjqRqKiyUbycTfcnDZVRSatF1EeNj7f1Lj8kRwlt5bC/5ZA5o0SqvGmaOhkUN2uFMrdH8K8fO3XZYoiy3ZQdg2aAe8/w+S9IZKhRvckgf67v1R9H2kCvD6FSVMgseVWthFZcTXW4riZYNFdpbG0eLvBXLkv8Okp3a7QzRvYrmo4tGkpStSwt3C6rt2cMfOfAPAnACAXOfu3iej/BOCPAfjT9r9/baMrl00elu7Fo9FSAcb89wBkkSjSgoCFSMBsvhpyZyP0o21qRygSlch+T34fJ7aAsE1iJwUgl8egFczhUMLZzKB5CrL1oLzy4Xgiv1lqY1ki8PK5t0ZVgmYxzwRYVDgHFvSvqvZtcn1SVhyEgel0OUfqpqimVXV2XMkAAER8fSGZG0R+DBSpn1uf87qJ76QApWBOz1YKd648SWbZ+Nh/l0Lpjfo6aS1e3F4EmoZ+0wUAZBikFciJpUCMjezhPihOgYy9kh2SWPIamYEN8gvqDO0Vw6BoMJM1gq/GG9eV2RZOVp0uLuuZelaCHQd7Qut5WzdnnRBXXxtAzw06px0EZxdAHIMPhzCdANAEevEKCENkx/vgUIG1QnA5A03nluq4GrmggZWmN2w33DXjSGtQtytU3swA6zZntmYYcsqcUqw+lb7Mq84LGu6BtZL1xVLIqqit1I3A+3sLtbDJZLFmhYGlhTNADMCIyJHd8JBzdNh6itTtgqZTsFFy3ilvJGJA2lHb1whakAJpWCqdFZjqhCIqlBpb4w6yDjtltlKhrIrNSlm6w5rCsTeKdVT0pnOFv6c1m9+uzAvZgyHU+aUXN6AgkHkx0J7ymxz1MH4aYvKUMOgrHK65j9J8rXzuqNt01/YBBeqK6FcTyrav69duzlpsytiqOI56XVCvB94fgCbTtQIgAGR+jDoyfuJkK9GyjeqcFWiNDwD8ZQBfA/AJgD/MzLW8JE9rJBJaUBiAP/0cXrr3uknL29Lnmopy1CWjuzpmKrfJ6Yg8uFtseTL1kRypMN8DDwfy2zQDvzn1Xv/saryamO48h5aiCSXRgSXanlMWzHkyfZ0zl4cxm8vGpBPCvD6BGU8QvPtM8oWiEMnDPWSRAhkguIqhL6agi5Gvk+OEWeQ55Irg2vsS7q2RpHzHUU9y7Yki0PEhzOEesn4IPUmgzq+khs49gRNaMJPJl8MLZ/u484ia0ahePdNStszVuL5YLACQkjpecVy+gBJBP34k/05TEfLZwTP1FKB8oeNiLkbO+UAdKQKJnCfeC2IAtx89vc2oeFPahn2nPmJxR04LiiKo4Z7PRdwYSiN49xm4F4F78t5plsB89Kk/hGO7ce904IoOc2b8BtVcjVc3X/l1oOTdqa5I/ntmRIlQUP5cS/XvygR81qxpPlq0ThgL8GuUP7ZAx135rb1Xf/68Iy+fkH/b7I4Nr6e6Xainj6X8TpIIfdr+3tVqAwD15gLpZ8+hf/onYLoB1CQGXp0s061LWDJqb7B2Pq1qF7ReH512deE+eAf0+evVPP1rIE+jXJq73cY49/71cCg2S2aAJBa1TttHXN5cnbNC9fvAt78O0wthQg0yDH0xA//aD2rpmxQEUAdCqTSvT0oj2i3uCZz96QIpTaKduxCpKUHw7CnYGGQvr0nqc/PgJsJKVXNnAdelNXow898D8Pfsv08AbF9Reh6DmGFcEt01vVSejhMEi41NxSROkSjoUb8nwgWj0fKiaycltSdqTBwnXjkvOz/3GySnlMVpCrW/DxztA2eXwHwOM5/b6FoG9eAY0ArU68GcikITOuLtpzixnvRcrbOa+2Qjyf8MAKystH4h+li4Z2aGihP7nBmcZrJ5sh58sPHUA2KGnsSgTLqGslG/fB0hzjIv1MF2I+aVw+zfzAx1dADuRuAwgLqagC9GPt+PmEFxCg2A5gk46kB/99vgUAMGoBevgTjxBlQlXecGlbnYScrfBpoqEDWFkyDPK7rlaUxO0MIao9BiKALwQiSehuKED9xvc1L6LlpN/R54dIUsLhRZhvVmOoMwSa9FP3IRder1hO5ZQ6cKPvga0qeHoNSIofX6zNYKLBgQnImR9c2vywbWsNTZu2lp+ds0aLkgQlNl4LpF9a6NH7NZJGYFbMAXl0DchUp64EBLtGU+zzm7FgYEF37rGQ7F52TqZZa9OEvu+Lo2+vlXfmz/22Cj5b4ueUbOcVZaJzLfeGtI+TqUzrguGOarF+WN2rhzrLte/p0R2XI1max7eXXgIJRouzGg1Cy+m8dQzEJxdRFMVxPQFn32lEz3DkltvGkU0Ylmx3OWgWaJRLFy99lE9dcf7vLug0DqoLr7QInx6N6xCvzf2dUYVCz4bQXXmBm1AwOQ5zy1z7YTABmD5rGMvRrWBGeZ2GBJuijzsiNskme8/MMdr9dl2EIR1jvN87R6w94WcKwoM5mUb8pzCt4wUgKGwo6wwZpEf1z5JCdu0hA3UaNS7NwdnHfNnL+z3xSwUeTsulgrCLItlJb8k/2heMF+4yOY6VQWnZIFKnjvXZjDIWbv7qH7xRXMf/trXk2OZ3PZ5HVCmG+9L16rN5fIDvfEm/ZLvyb1Sd594jdVNJ4i/vAxTr/bxaNfvIQ+vQTSDOZyBHN1BfptP4Vsv4OkH2DwvRdIP/kMwbvvgGeznXrBGmGDvB7V68kE3oACp588BnUjMWpdAeP//s9g+qSD2aHC3vMU/R+ewnz8GThJoff3JF/D5QT91p/EJ//mMZIhgzLg/V+YIzybQl3YZOI4QfrFi+WLqkWBZTeRbIwKifLbhivCbHJUo+vARWZxfOhpdNmPPhEvZ68HtT8EAg2ezEB7fZhhD9P3hgAB4VWKzg9eIn3+uT1ZTrbbRaCCABRo8PGB0MQUoD8/Wbwj+26ckcvTqe8XajgEabVcfLvRQxIPMh0fIn18gOCLM/BoVHme0//lfw9v/tUENNXofqHx7B/M0f3+89V+BEA/eoTP/8i35d9zxtP/4mOkX7y8836xc+RKFnyl5NVzcKUKKvN+CyyIrTbpNyC1X4qSshrB0ycy/16Olgq+U9gRMS1njBoD9fABzMMDqJNL8GSC7OTUS3WLA1LO73PO3gLk89l8Ll0hR1xFEehgf5EPeH5Rvpm10EdHUpOsG4HHE/A8XqI2l0qm5za5jSh8m96nZXcglA3UOkGl4OkToN+D2etCvTpDdna+tm+7vMPrSsqvgy+ndJu1FpVesUOa/s4FApbW6x1v2FS/LxvTDZ5J8OwpoJSUpbLg+VxsgaMDcD8SB+SnLyVgEIbI3pwu7LDBQOjvQSD26ckp9KNHoH4X6aefN5Kv1/t7ku99V3Uz3yLsLHK2U1xTwY+iCGpPOhIAmJcSaaHLEbKZLbab8wg4mhYNh0CSQL05Q38yA48kmuAjNK7OWpJCfSwFOc1sBjW6gtIaWZLCGIb64pX3WJk4QSfL8PjsEPTitVANc9dXn76ADkN0OiHMG6FUmLPzDQs674g6sgH/1jQp2ucOv7gEXy0Xf+784AuEX/SwH3VAVxM5xtI6zXi6JAdPz1/h/b8dIos0iIHwo1dSyy5OFtKjRRhbjNY08NpV4ZoTqYvkXFdlTwyH3RlzZjYHZZlETF2xaJNJ3sp0CuOklrMMPJuBLkMMLsby2TyGucipapkMxpXJySVbk9ag2RxKK4DUMj3Hivs4ymDe28nTKbhQHNPTa+BoACXPghlmOoM6PUeQpDCXIyksWYFH//AEe18cQCUpgvEUwWcny/eVP/VohGd//1z+SA3M2fn2feq2sM0caj3iXKB8uu+8lzQu74/66EhyVOwGVx3sL/Kz0nQhEW69ter4EDyZrhXDuHXUSVGzNeRtf/QGPuD7tN4b2JIQVC6ykvvbPVNfMmGXUcmyMXJxKZFy9w7dV2nixbQAO87OzkGzmRhTrhZkloFncyjYcZnbpL0NyDsd/HvM92VmmDiBurj0EdOqfCbVlXwT2LIf5s2pGOZaLSkBc5KuMl6YQYGGfnDk3312er5g9eTp1M4hYKO2kh5gwJkRJwIp2VhbSp/e3wdC2USzkegdRVHtBtOcXwDjCegihBkvCxRVwYyn8u5vmLrqHKzbwNt2bi4yBunnX6xvb4kd0qyxkqaBYsStGFHeEo4FBtgNeBRVp1dYsTpjI5o8m4OiDuhgaHMFDczVGLgaC1vpTN6juRrLNTrpUtSK53Pp00otmFKjEXgyaWYr1T1Tx+bRqlW3bYA73Jw1S1it/LlNtjbDgdAjP5mvJt3lB04Ygrpd8P4AeHUKHo2AXD7DsvCI1BXLXr9efJazLdhkyM6XPS1mMgFKPPJASS03oLkK2lLC8i178TfYuJRtTsoiFA5FT1V2cgqcnMKZ7E2H7Z0PcK2BMJSkz+ts9Iq0sybI85pXashk4HmGLOe9yqubLikOuXdX53nN15lyYwaoTo5lXlBm7N/+q9KEfeVFAgAskvyLp01iZGkiAgVrIj/Z934Dne8t/q57umY2A/7Z92qOyLf1niioFvLtKlGW95Rvf+48juJKmSl9ByIqMQBeWFprr+dr+Tn6M5hBxFb1sX8/k/PXGm4LQ8tTgy2ljRQBva51KNCqyEouR1g2vMLEIMNCd7/h51GpJsbsxbTkwEzWoeJa5DbwSSr3t8vGFYWpbmoc5R0X7j3a6AwAEXRZM2c7UQyyxZI5TWHG40XB4CBYHxFzfYXZUhLP7clVTi048/Mfs1AVqS/jCmkKjCX6Rp2OjK8sA6IIFGjJZ7cbS9LapjlUbM5ms42FCfw6rTR2IkxWeaFrbGysbWcOBpISkaR2/K05H2/JtnHj44Z8d26+4NlcbIteF5hWFJl3/SsV5wDHsfSJTgiydTI5y8BpInN6fs30NbfyDpzVTdNGjueaZ0qKZI0PQ2nDfVwX7hHeblpjvihik0R3RzG4wdohXg7Zhugb5XlUQWnog/3FAJuLkIeKoiWZ8K8SHOWmtFzCXcHSZBztb9cJy/l8BzMV17eThk+eHIAyA301R/a936g+RxSBf+YnRC7cMPT3PmprzmwB1e1CPXwAvhzZCLkt9n6P6hHl6TWq2wV9+D5Ya0AB/Os/Xk83sTTWUu+mcnk3eaPNoiT3c+ucjrtGca7OU5Zc7UFAPNM5apOyogk4GILGU6TPP/eUb8QJeDypH3eWDpynI94pdlWjUkl9Uup1QWEIdEJ5FqfnEoUEqp9LXh7dUsoqFSmDQHK8rbMqH9Wkf+mnMXlvAMosdfvVFfjTL0qjumowAH/3Q68AG3zyCmY8kWPzG8wGz6XRJq7inkFqMdaKuXS3oXZ8T+AZKvP56n3mqNoAmtmD9x3uXTcZf8V+AdzvMkH3xbl5D3A/aY1Nc33quPsmK/USqYGlOi7RrAreDlss1oslTGe1bXHesuzisr7NWQa2CZEraNohXed1k0yWLWg4hivr6nxVcO+MPWZJfmVR2ty1iAgbiURwXqQ+y4B5jOBSKLQ0WePdyjLo0RxMBOLqmh51cNTDRk4Ba4x92ZS1ODNSJyoVeu6tCcYAXt2PHI3U1U5bEX1guHAHZwZqOgfZ+opZk/aaDJzp8nsrzrlr5u9tosHBk0fi/dUK5sWr1Tye/OYvisQx0usCcSK5MdZYU72ubDCb1kfKbzqZl9coZiBX+sTR4J3IkkeWCb1+mhNOcIIjTag8bLYamzeGXc6zNhLkRKRgaZdN7jdf66yU/mb7BRv250WB7q6u5ohOOyDDUJMENJmBkwpKW5ZBje37Mwyezbxo1qbPZKv5zwkKVG0Ed0She1vAxkYey569t+2+RM8jR8FtfGzh343TPXblgGmK+2S73WPcWeTMCXCsK8asul3UFdYsg/7Wh6DMIP3xx5XHUBBAffD+YlL/4lWtGEPw4deRHe+Bvv/jm/VqrvEMtmhxKyhJbnZ5DusSzwEbYXrnqZRquG/5Rm8pVL8Peu8ZuB+BAwX16SuplfaW1iIrg+p2Mf3X/gVkPYWsQzj6+58sCdMU58bg6RPwg0PMn+whPJuB//mveQoYvfMEGI0bSSmr4RDUCZGdni2iYN2u1PJrUNemxR2iaZmIFi2+CrB5aE5he53tTKGUGvkqMrHuGvcycmZmcwnbl4SofS0Wk60kNVci51Hll28kSbYGbBj88s2iPVbEo7K9r95AXY5g4vhmo1YmA8eLYpildXAq2ukS17dWybHP3uX+VIkCOLg6LaR143pgSxNBnrrhpIlvEMoWs6ziUG8kGZv/jZWo5SyDmc7knmxBVADiXHB5AY7iku9DTTbhlmrmanU5Sl0VbSFfod5cjKon3lpJ9eV+znG8JNFbBxMnwMvX6/vibUgSN0FJfcB1xwO4VQeKmc2hvnglFB9Fkuh9n6Isa5AvoaC6XWEijEb+GerDA9BggO6rCUwUwHQ0eH+AIPoAHAbA+SWy1ycyjqII+vEjcBQCxiC8mEHNE+DoAOZqLM/q5ZtFhGsNeDqVHI/c+zRxUi0nX3qDN0PXcaUj/HziyprUiYtYVVWyUvJf6g3mlkyAnaCwMVwnyrGEPF2zhYfr72bcUIRi2+u4OWQ0QnZ5VZM3aGnembkVymReEGRTtUb5kVCtm0bOGgui3JYC7VsG/egRKOqIyF+cNN/kEqGubtbd0Rqr6gCQko7JRr5vOjhzdTLWeuqJpANv4NEvTZy+KdjOT4qkyniW2c1ZvQCAS+bffnOmFgmbgCSZ1j3+MBSDqxMCDRPd3caPE/jilQ43Tkuw91WVFO3vfaPNWSD5JEqD0hQ0n8s9KSUbMWaQsepObBO/lRVb2EDohZzwSE82mBTH9nRmtUuQ9BsKpQA6TWqkgmvqyxQ/4ypaSRmc2EADiNR0s9PeJKjTEcpV08W3SdL5LmGytzsK6XJC0lT6chQBV2P/DKnbBfciqKs51CyF6QQw3Q7Q78B0Q4TzeJE/pGXjBiP0NTVNgDSTTcx4At7wWZUaXGw2e7031R+0lnXAqUcqAjK9cJaUjUmS+cfn4XzZcVcGo63r5JwkFAS1ohxLP3X1pNrN2TKs2BvN5jdqE5DWMAd7UEkCGk8rryU1wjqAXXN314CKOrxaLcRrijUYm2BTcbGGY0cc0PhK0WmbgAY9cC8CTafWVm/yo0LOaNkht0lrJKLXAMYA3qw7tkWLW8BDtH2xxf1B2x9b3Be0fbHFfULbH1vcF+yyL36dmR+VfXGrmzMAIKJ/XMWxbNHiNtH2xRb3CW1/bHFf0PbFFvcJbX9scV9wW32xPq7WokWLFi1atGjRokWLFi1uBe3mrEWLFi1atGjRokWLFi3uAe5ic/bn7uCaLVqUoe2LLe4T2v7Y4r6g7Yst7hPa/tjivuBW+uKt55y1aNGiRYsWLVq0aNGiRYtVtLTGFi1atGjRokWLFi1atLgHuLXNGRH9PiL6dSL6ARH9/G1dt8VXF0T054noFRH9Su6zYyL6W0T0m/a/R7nv/oTtn79ORL/3blrd4ssIInqfiP4uEX2fiH6ViP64/bztjy1uFUTUJaJ/RET/3PbF/6P9vO2LLe4MRKSJ6J8S0X9u/277Y4tbBxF9RES/TET/jIj+sf3s1vvirWzOiEgD+L8B+DcA/BSAP0pEP3Ub127xlcb/E8DvK3z28wB+gZm/DeAX7N+w/fGPAPhp+5v/0PbbFi12gRTAv8fM3wXwswD+Xdvn2v7Y4rYxB/C7mflfBPDbAPw+IvpZtH2xxd3ijwP4fu7vtj+2uCv8a8z823KS+bfeF28rcvbbAfyAmX/EzDGAvwTgD93StVt8RcHM/z8Ap4WP/xCAv2D//RcA/E9zn/8lZp4z848B/ADSb1u0uDaY+Qtm/iX77xHECHkXbX9scctgwZX9M7T/Y7R9scUdgYjeA/BvAvi/5z5u+2OL+4Jb74u3tTl7F8Cnub8/s5+1aHHbeMLMXwBiMAN4bD9v+2iLWwERfQDgZwD8Q7T9scUdwFLI/hmAVwD+FjO3fbHFXeLPAvj3AZjcZ21/bHEXYAD/FRH9EyL6OfvZrffFYBcnaQAq+ayViWxxn9D20RY3DiLaA/CfAvjfMvMlUVm3k0NLPmv7Y4udgJkzAL+NiA4B/GdE9FtqDm/7YosbAxH9AQCvmPmfENHvavKTks/a/thiV/idzPw5ET0G8LeI6Ndqjr2xvnhbkbPPALyf+/s9AJ/f0rVbtMjjJRE9AwD731f287aPtrhREFEI2Zj9RWb+q/bjtj+2uDMw8zmAvwfJl2j7You7wO8E8AeJ6CNIysvvJqL/GG1/bHEHYObP7X9fAfjPIDTFW++Lt7U5+0UA3yaiD4moA0mg++u3dO0WLfL46wD+mP33HwPw13Kf/xEiiojoQwDfBvCP7qB9Lb6EIAmR/UcAvs/Mfyb3VdsfW9wqiOiRjZiBiHoAfg+AX0PbF1vcAZj5TzDze8z8AcQ2/DvM/G+j7Y8tbhlENCCiofs3gH8dwK/gDvrirdAamTklov8NgP8SgAbw55n5V2/j2i2+uiCi/w+A3wXgIRF9BuD/AOBPA/jLRPS/AvAJgD8MAMz8q0T0lwF8D6Ks9+9a6k+LFrvA7wTw7wD4ZZvrAwB/Em1/bHH7eAbgL1hVMQXgLzPzf05E/wBtX2xxf9DOjS1uG08gNG9A9kf/b2b+m0T0i7jlvkjMLVW3RYsWLVq0aNGiRYsWLe4at1aEukWLFi1atGjRokWLFi1aVKPdnLVo0aJFixYtWrRo0aLFPUC7OWvRokWLFi1atGjRokWLe4B2c9aiRYsWLVq0aNGiRYsW9wDt5qxFixYtWrRo0aJFixYt7gHazVmLFi1atGjRokWLFi1a3AO0m7MWLVq0aNGiRYsWLVq0uAdoN2ctWrRo0aJFixYtWrRocQ/w/wfD7Eo7RK9A6wAAAABJRU5ErkJggg==",
      "text/plain": [
       "<Figure size 1080x216 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "res = vgg.features(sample_image).cpu()\n",
    "plt.figure(figsize=(15,3))\n",
    "plt.imshow(res.detach().view(512,-1).T)\n",
    "print(res.size())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The dimension of feature tensor is 512x7x7, but in order to visualize it we had to reshape it to 2D form.\n",
    "\n",
    "Now let's try to see if those features can be used to classify images. Let's manually take some portion of images (800 in our case), and pre-compute their feature vectors. We will store the result in one big tensor called `feature_tensor`, and also labels into `label_tensor`:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "...................................................................................................."
     ]
    }
   ],
   "source": [
    "bs = 8\n",
    "dl = torch.utils.data.DataLoader(dataset,batch_size=bs,shuffle=True)\n",
    "num = bs*100\n",
    "feature_tensor = torch.zeros(num,512*7*7).to(device)\n",
    "label_tensor = torch.zeros(num).to(device)\n",
    "i = 0\n",
    "for x,l in dl:\n",
    "    with torch.no_grad():\n",
    "        f = vgg.features(x.to(device))\n",
    "        feature_tensor[i:i+bs] = f.view(bs,-1)\n",
    "        label_tensor[i:i+bs] = l\n",
    "        i+=bs\n",
    "        print('.',end='')\n",
    "        if i>=num:\n",
    "            break\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now we can define `vgg_dataset` that takes data from this tensor, split it into training and test sets using `random_split` function, and train a small one-layer dense classifier network on top of extracted features:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/anaconda/envs/py38_pytorch/lib/python3.8/site-packages/torch/nn/modules/container.py:119: UserWarning: Implicit dimension choice for log_softmax has been deprecated. Change the call to include dim=X as an argument.\n",
      "  input = module(input)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch  0, Train acc=0.879, Val acc=0.990, Train loss=0.110, Val loss=0.007\n",
      "Epoch  1, Train acc=0.981, Val acc=0.980, Train loss=0.015, Val loss=0.021\n",
      "Epoch  2, Train acc=0.999, Val acc=0.990, Train loss=0.001, Val loss=0.002\n",
      "Epoch  3, Train acc=1.000, Val acc=0.980, Train loss=0.000, Val loss=0.002\n",
      "Epoch  4, Train acc=1.000, Val acc=0.980, Train loss=0.000, Val loss=0.002\n",
      "Epoch  5, Train acc=1.000, Val acc=0.980, Train loss=0.000, Val loss=0.002\n",
      "Epoch  6, Train acc=1.000, Val acc=0.980, Train loss=0.000, Val loss=0.002\n",
      "Epoch  7, Train acc=1.000, Val acc=0.980, Train loss=0.000, Val loss=0.002\n",
      "Epoch  8, Train acc=1.000, Val acc=0.980, Train loss=0.000, Val loss=0.002\n",
      "Epoch  9, Train acc=1.000, Val acc=0.980, Train loss=0.000, Val loss=0.002\n"
     ]
    }
   ],
   "source": [
    "vgg_dataset = torch.utils.data.TensorDataset(feature_tensor,label_tensor.to(torch.long))\n",
    "train_ds, test_ds = torch.utils.data.random_split(vgg_dataset,[700,100])\n",
    "\n",
    "train_loader = torch.utils.data.DataLoader(train_ds,batch_size=32)\n",
    "test_loader = torch.utils.data.DataLoader(test_ds,batch_size=32)\n",
    "\n",
    "net = torch.nn.Sequential(torch.nn.Linear(512*7*7,2),torch.nn.LogSoftmax()).to(device)\n",
    "\n",
    "history = train(net,train_loader,test_loader)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The result is great, we can distinguish between a cat and a dog with almost 98% probability! However, we have only tested this approach on a small subset of all images, because manual feature extraction seems to take a lot of time.\n",
    "\n",
    "## Transfer learning using one VGG network\n",
    "\n",
    "We can also avoid manually pre-computing the features by using the original VGG-16 network as a whole during training. Let's look at the VGG-16 object structure:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "VGG(\n",
      "  (features): Sequential(\n",
      "    (0): Conv2d(3, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
      "    (1): ReLU(inplace=True)\n",
      "    (2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
      "    (3): ReLU(inplace=True)\n",
      "    (4): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)\n",
      "    (5): Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
      "    (6): ReLU(inplace=True)\n",
      "    (7): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
      "    (8): ReLU(inplace=True)\n",
      "    (9): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)\n",
      "    (10): Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
      "    (11): ReLU(inplace=True)\n",
      "    (12): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
      "    (13): ReLU(inplace=True)\n",
      "    (14): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
      "    (15): ReLU(inplace=True)\n",
      "    (16): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)\n",
      "    (17): Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
      "    (18): ReLU(inplace=True)\n",
      "    (19): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
      "    (20): ReLU(inplace=True)\n",
      "    (21): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
      "    (22): ReLU(inplace=True)\n",
      "    (23): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)\n",
      "    (24): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
      "    (25): ReLU(inplace=True)\n",
      "    (26): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
      "    (27): ReLU(inplace=True)\n",
      "    (28): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
      "    (29): ReLU(inplace=True)\n",
      "    (30): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)\n",
      "  )\n",
      "  (avgpool): AdaptiveAvgPool2d(output_size=(7, 7))\n",
      "  (classifier): Sequential(\n",
      "    (0): Linear(in_features=25088, out_features=4096, bias=True)\n",
      "    (1): ReLU(inplace=True)\n",
      "    (2): Dropout(p=0.5, inplace=False)\n",
      "    (3): Linear(in_features=4096, out_features=4096, bias=True)\n",
      "    (4): ReLU(inplace=True)\n",
      "    (5): Dropout(p=0.5, inplace=False)\n",
      "    (6): Linear(in_features=4096, out_features=1000, bias=True)\n",
      "  )\n",
      ")\n"
     ]
    }
   ],
   "source": [
    "print(vgg)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "You can see that the network contains:\n",
    "* feature extractor (`features`), comprised of a number of convolutional and pooling layers\n",
    "* average pooling layer (`avgpool`)\n",
    "* final `classifier`, consisting of several dense layers, which turns 25088 input features into 1000 classes (which is the number of classes in ImageNet)\n",
    "\n",
    "To train the end-to-end model that will classify our dataset, we need to:\n",
    "* **replace the final classifier** with the one that will produce required number of classes. In our case, we can use one `Linear` layer with 25088 inputs and 2 output neurons.\n",
    "* **freeze weights of convolutional feature extractor**, so that they are not trained. It is recommended to initially do this freezing, because otherwise untrained classifier layer can destroy the original pre-trained weights of convolutional extractor. Freezing weights can be accomplished by setting `requires_grad` property of all parameters to `False`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "==========================================================================================\n",
       "Layer (type:depth-idx)                   Output Shape              Param #\n",
       "==========================================================================================\n",
       "VGG                                      --                        --\n",
       "├─Sequential: 1-1                        [1, 512, 7, 7]            --\n",
       "│    └─Conv2d: 2-1                       [1, 64, 244, 244]         (1,792)\n",
       "│    └─ReLU: 2-2                         [1, 64, 244, 244]         --\n",
       "│    └─Conv2d: 2-3                       [1, 64, 244, 244]         (36,928)\n",
       "│    └─ReLU: 2-4                         [1, 64, 244, 244]         --\n",
       "│    └─MaxPool2d: 2-5                    [1, 64, 122, 122]         --\n",
       "│    └─Conv2d: 2-6                       [1, 128, 122, 122]        (73,856)\n",
       "│    └─ReLU: 2-7                         [1, 128, 122, 122]        --\n",
       "│    └─Conv2d: 2-8                       [1, 128, 122, 122]        (147,584)\n",
       "│    └─ReLU: 2-9                         [1, 128, 122, 122]        --\n",
       "│    └─MaxPool2d: 2-10                   [1, 128, 61, 61]          --\n",
       "│    └─Conv2d: 2-11                      [1, 256, 61, 61]          (295,168)\n",
       "│    └─ReLU: 2-12                        [1, 256, 61, 61]          --\n",
       "│    └─Conv2d: 2-13                      [1, 256, 61, 61]          (590,080)\n",
       "│    └─ReLU: 2-14                        [1, 256, 61, 61]          --\n",
       "│    └─Conv2d: 2-15                      [1, 256, 61, 61]          (590,080)\n",
       "│    └─ReLU: 2-16                        [1, 256, 61, 61]          --\n",
       "│    └─MaxPool2d: 2-17                   [1, 256, 30, 30]          --\n",
       "│    └─Conv2d: 2-18                      [1, 512, 30, 30]          (1,180,160)\n",
       "│    └─ReLU: 2-19                        [1, 512, 30, 30]          --\n",
       "│    └─Conv2d: 2-20                      [1, 512, 30, 30]          (2,359,808)\n",
       "│    └─ReLU: 2-21                        [1, 512, 30, 30]          --\n",
       "│    └─Conv2d: 2-22                      [1, 512, 30, 30]          (2,359,808)\n",
       "│    └─ReLU: 2-23                        [1, 512, 30, 30]          --\n",
       "│    └─MaxPool2d: 2-24                   [1, 512, 15, 15]          --\n",
       "│    └─Conv2d: 2-25                      [1, 512, 15, 15]          (2,359,808)\n",
       "│    └─ReLU: 2-26                        [1, 512, 15, 15]          --\n",
       "│    └─Conv2d: 2-27                      [1, 512, 15, 15]          (2,359,808)\n",
       "│    └─ReLU: 2-28                        [1, 512, 15, 15]          --\n",
       "│    └─Conv2d: 2-29                      [1, 512, 15, 15]          (2,359,808)\n",
       "│    └─ReLU: 2-30                        [1, 512, 15, 15]          --\n",
       "│    └─MaxPool2d: 2-31                   [1, 512, 7, 7]            --\n",
       "├─AdaptiveAvgPool2d: 1-2                 [1, 512, 7, 7]            --\n",
       "├─Linear: 1-3                            [1, 2]                    50,178\n",
       "==========================================================================================\n",
       "Total params: 14,764,866\n",
       "Trainable params: 50,178\n",
       "Non-trainable params: 14,714,688\n",
       "Total mult-adds (G): 17.99\n",
       "==========================================================================================\n",
       "Input size (MB): 0.71\n",
       "Forward/backward pass size (MB): 128.13\n",
       "Params size (MB): 59.06\n",
       "Estimated Total Size (MB): 187.91\n",
       "=========================================================================================="
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "vgg.classifier = torch.nn.Linear(25088,2).to(device)\n",
    "\n",
    "for x in vgg.features.parameters():\n",
    "    x.requires_grad = False\n",
    "\n",
    "summary(vgg,(1, 3,244,244))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As you can see from the summary, this model contain around 15 million total parameters, but only 50k of them are trainable - those are the weights of classification layer. That is good, because we are able to fine-tune smaller number of parameters with smaller number of examples.\n",
    "\n",
    "Now let's train the model using our original dataset. This process will take a long time, so we will use `train_long` function that will print some intermediate results without waiting for the end of epoch. It is highly recommended to run this training on GPU-enabled compute!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 0, minibatch 0: train acc = 0.5, train loss = 0.0431101992726326\n",
      "Epoch 0, minibatch 90: train acc = 0.9539835164835165, train loss = 0.0960497070144821\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/anaconda/envs/py38_pytorch/lib/python3.8/site-packages/PIL/TiffImagePlugin.py:793: UserWarning: Truncated File Read\n",
      "  warnings.warn(str(msg))\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 0, minibatch 180: train acc = 0.9582182320441989, train loss = 0.12481052967724879\n",
      "Epoch 0, minibatch 270: train acc = 0.9587177121771218, train loss = 0.14185787918822793\n",
      "Epoch 0, minibatch 360: train acc = 0.9634695290858726, train loss = 0.14566257719848294\n",
      "Epoch 0, minibatch 450: train acc = 0.966879157427938, train loss = 0.13402751914149114\n",
      "Epoch 0, minibatch 540: train acc = 0.9686922365988909, train loss = 0.13931148902766144\n",
      "Epoch 0, minibatch 630: train acc = 0.9694928684627575, train loss = 0.1386710044510202\n",
      "Epoch 0, minibatch 720: train acc = 0.970613730929265, train loss = 0.13363790313678375\n",
      "Epoch 0, minibatch 810: train acc = 0.9709463625154131, train loss = 0.1342217084364885\n",
      "Epoch 0, minibatch 900: train acc = 0.9721143174250833, train loss = 0.13233261023721474\n",
      "Epoch 0, minibatch 990: train acc = 0.9726286579212916, train loss = 0.1334670727957871\n",
      "Epoch 0, minibatch 1080: train acc = 0.9733464384828863, train loss = 0.13777193110039893\n",
      "Epoch 0, minibatch 1170: train acc = 0.9734735269000854, train loss = 0.14239378162778207\n",
      "Epoch 0 done, validation acc = 0.9671868747499, validation loss = 0.25287964306816474\n"
     ]
    }
   ],
   "source": [
    "trainset, testset = torch.utils.data.random_split(dataset,[20000,len(dataset)-20000])\n",
    "train_loader = torch.utils.data.DataLoader(trainset,batch_size=16)\n",
    "test_loader = torch.utils.data.DataLoader(testset,batch_size=16)\n",
    "\n",
    "train_long(vgg,train_loader,test_loader,loss_fn=torch.nn.CrossEntropyLoss(),epochs=1,print_freq=90)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "It looks like we have obtained reasonably accurate cats vs. dogs classifier! Let's save it for future use!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "torch.save(vgg,'data/cats_dogs.pth')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can then load the model from file at any time. You may find it useful in case the next experiment destroys the model - you would not have to re-start from scratch."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "vgg = torch.load('data/cats_dogs.pth')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Fine-tuning transfer learning\n",
    "\n",
    "In the previous section, we have trained the final classifier layer to classify images in our own dataset. However, we did not re-train the feature extractor, and our model relied on the features that the model has learned on ImageNet data. If your objects visually differ from ordinary ImageNet images, this combination of features might not work best. Thus it makes sense to start training convolutional layers as well.\n",
    "\n",
    "To do that, we can unfreeze the convolutional filter parameters that we have previously frozen. \n",
    "\n",
    "> **Note:** It is important that you freeze parameters first and perform several epochs of training in order to stabilize weights in the classification layer. If you immediately start training end-to-end network with unfrozen parameters, large errors are likely to destroy the pre-trained weights in the convolutional layers."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "for x in vgg.features.parameters():\n",
    "    x.requires_grad = True"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "After unfreezing, we can do a few more epochs of training. You can also select lower learning rate, in order to minimize the impact on the pre-trained weights. However, even with low learning rate, you can expect the accuracy to drop in the beginning of the training, until finally reaching slightly higher level than in the case of fixed weights.\n",
    "\n",
    "> **Note:** This training happens much slower, because we need to propagate gradients back through many layers of the network! You may want to watch the first few minibatches to see the tendency, and then stop the computation."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 0, minibatch 0: train acc = 1.0, train loss = 0.0\n",
      "Epoch 0, minibatch 90: train acc = 0.8990384615384616, train loss = 0.2978392171335744\n",
      "Epoch 0, minibatch 180: train acc = 0.9060773480662984, train loss = 0.1658294214069514\n",
      "Epoch 0, minibatch 270: train acc = 0.9102859778597786, train loss = 0.11819224340009514\n",
      "Epoch 0, minibatch 360: train acc = 0.9191481994459834, train loss = 0.09244130522920814\n",
      "Epoch 0, minibatch 450: train acc = 0.9261363636363636, train loss = 0.07583886292451236\n",
      "Epoch 0, minibatch 540: train acc = 0.928373382624769, train loss = 0.06537413817456822\n",
      "Epoch 0, minibatch 630: train acc = 0.9318541996830428, train loss = 0.057419379426257924\n",
      "Epoch 0, minibatch 720: train acc = 0.9361130374479889, train loss = 0.05114534460059813\n",
      "Epoch 0, minibatch 810: train acc = 0.938347718865598, train loss = 0.04657612246737968\n",
      "Epoch 0, minibatch 900: train acc = 0.9407602663706992, train loss = 0.04258851655712403\n",
      "Epoch 0, minibatch 990: train acc = 0.9431130171543896, train loss = 0.03927870595491257\n",
      "Epoch 0, minibatch 1080: train acc = 0.945536540240518, train loss = 0.03652716609309053\n",
      "Epoch 0, minibatch 1170: train acc = 0.9463065755764304, train loss = 0.03445258006186286\n",
      "Epoch 0 done, validation acc = 0.974389755902361, validation loss = 0.005457923144233279\n"
     ]
    }
   ],
   "source": [
    "train_long(vgg,train_loader,test_loader,loss_fn=torch.nn.CrossEntropyLoss(),epochs=1,print_freq=90,lr=0.0001)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Other computer vision models\n",
    "\n",
    "VGG-16 is one of the simplest computer vision architectures. `torchvision` package provides many more pre-trained networks. The most frequently used ones among those are **ResNet** architectures, developed by Microsoft, and **Inception** by Google. For example, let's explore the architecture of the simplest ResNet-18 model (ResNet is a family of models with different depth, you can try experimenting with ResNet-151 if you want to see what a really deep model looks like):"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ResNet(\n",
      "  (conv1): Conv2d(3, 64, kernel_size=(7, 7), stride=(2, 2), padding=(3, 3), bias=False)\n",
      "  (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "  (relu): ReLU(inplace=True)\n",
      "  (maxpool): MaxPool2d(kernel_size=3, stride=2, padding=1, dilation=1, ceil_mode=False)\n",
      "  (layer1): Sequential(\n",
      "    (0): BasicBlock(\n",
      "      (conv1): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "      (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "      (relu): ReLU(inplace=True)\n",
      "      (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "      (bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "    )\n",
      "    (1): BasicBlock(\n",
      "      (conv1): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "      (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "      (relu): ReLU(inplace=True)\n",
      "      (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "      (bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "    )\n",
      "  )\n",
      "  (layer2): Sequential(\n",
      "    (0): BasicBlock(\n",
      "      (conv1): Conv2d(64, 128, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)\n",
      "      (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "      (relu): ReLU(inplace=True)\n",
      "      (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "      (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "      (downsample): Sequential(\n",
      "        (0): Conv2d(64, 128, kernel_size=(1, 1), stride=(2, 2), bias=False)\n",
      "        (1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "      )\n",
      "    )\n",
      "    (1): BasicBlock(\n",
      "      (conv1): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "      (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "      (relu): ReLU(inplace=True)\n",
      "      (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "      (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "    )\n",
      "  )\n",
      "  (layer3): Sequential(\n",
      "    (0): BasicBlock(\n",
      "      (conv1): Conv2d(128, 256, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)\n",
      "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "      (relu): ReLU(inplace=True)\n",
      "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "      (downsample): Sequential(\n",
      "        (0): Conv2d(128, 256, kernel_size=(1, 1), stride=(2, 2), bias=False)\n",
      "        (1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "      )\n",
      "    )\n",
      "    (1): BasicBlock(\n",
      "      (conv1): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "      (relu): ReLU(inplace=True)\n",
      "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "    )\n",
      "  )\n",
      "  (layer4): Sequential(\n",
      "    (0): BasicBlock(\n",
      "      (conv1): Conv2d(256, 512, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)\n",
      "      (bn1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "      (relu): ReLU(inplace=True)\n",
      "      (conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "      (bn2): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "      (downsample): Sequential(\n",
      "        (0): Conv2d(256, 512, kernel_size=(1, 1), stride=(2, 2), bias=False)\n",
      "        (1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "      )\n",
      "    )\n",
      "    (1): BasicBlock(\n",
      "      (conv1): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "      (bn1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "      (relu): ReLU(inplace=True)\n",
      "      (conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "      (bn2): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "    )\n",
      "  )\n",
      "  (avgpool): AdaptiveAvgPool2d(output_size=(1, 1))\n",
      "  (fc): Linear(in_features=512, out_features=1000, bias=True)\n",
      ")\n"
     ]
    }
   ],
   "source": [
    "resnet = torchvision.models.resnet18()\n",
    "print(resnet)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As you can see, the model contains the same building blocks: feature extractor and final classifier (`fc`). This allows us to use this model in exactly the same manner as we have been using VGG-16 for transfer learning. You can try experimenting with the code above, using different ResNet models as the base model, and see how accuracy changes.\n",
    "\n",
    "## Batch Normalization\n",
    "\n",
    "This network contains yet another type of layer: **Batch Normalization**. The idea of batch normalization is to bring values that flow through the neural network to right interval. Usually neural networks work best when all values are in the range of [-1,1] or [0,1], and that is the reason that we scale/normalize our input data accordingly. However, during training of a deep network, it can happen that values get significantly out of this range, which makes training problematic. Batch normalization layer computes average and standard deviation for all values of the current minibatch, and uses them to normalize the signal before passing it through a neural network layer. This significantly improves the stability of deep networks.\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Takeaway\n",
    "\n",
    "Using transfer learning, we were able to quickly put together a classifier for our custom object classification task, and achieve high accuracy. However, this example was not completely fair, because original VGG-16 network was pre-trained to recognize cats and dogs, and thus we were just reusing most of the patterns that were already present in the network. You can expect lower accuracy on more exotic domain-specific objects, such as details on production line in a plant, or different tree leaves.\n",
    "\n",
    "You can see that more complex tasks that we are solving now require higher computational power, and cannot be easily solved on the CPU. In the next unit, we will try to use more lightweight implementation to train the same model using lower compute resources, which results in just slightly lower accuracy. "
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "py37_pytorch",
   "language": "python",
   "name": "conda-env-py37_pytorch-py"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}