Assignment_1_XHD.ipynb

{
 "cells": [
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Assignment 1: Multi-Layer Perceptron with MNIST Dataset"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In this assignment, you are required to train two MLPs to classify images from the [MNIST database](http://yann.lecun.com/exdb/mnist/) hand-written digit database by using PyTorch.\n",
    "\n",
    "The process will be broken down into the following steps:\n",
    ">1. Load and visualize the data.\n",
    "2. Define a neural network. (30 marks)\n",
    "3. Train the models. (30 marks)\n",
    "4. Evaluate the performance of our trained models on the test dataset. (20 marks)\n",
    "5. Analysis your results. (20 marks)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 236,
   "metadata": {},
   "outputs": [],
   "source": [
    "import torch\n",
    "from torch import nn\n",
    "import numpy as np\n",
    "import logging\n",
    "import sys\n",
    "\n",
    "# set log\n",
    "logging.basicConfig(level=logging.INFO,\n",
    "                    format='%(asctime)s %(levelname)s: %(message)s',\n",
    "                     datefmt='%Y-%m-%d %H:%M:%S',)"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Get the version information of your package"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 237,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2023-10-13 17:31:12 INFO: The version information:\n",
      "2023-10-13 17:31:12 INFO: Python: 3.9.18 (main, Sep 11 2023, 14:09:26) [MSC v.1916 64 bit (AMD64)]\n",
      "2023-10-13 17:31:12 INFO: PyTorch: 1.12.0\n"
     ]
    }
   ],
   "source": [
    "logging.info('The version information:')\n",
    "logging.info(f'Python: {sys.version}')\n",
    "logging.info(f'PyTorch: {torch.__version__}')\n",
    "assert torch.cuda.is_available() == True, 'Please finish your GPU develop environment'"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---\n",
    "## Load and Visualize the Data\n",
    "\n",
    "Downloading may take a few moments, and you should see your progress as the data is loading. You may also choose to change the `batch_size` if you want to load more data at a time.\n",
    "\n",
    "This cell will create DataLoaders for each of our datasets."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 238,
   "metadata": {},
   "outputs": [],
   "source": [
    "from torchvision import datasets\n",
    "import torchvision.transforms as transforms\n",
    "from torch.utils.data.dataset import Dataset\n",
    "\n",
    "# number of subprocesses to use for data loading\n",
    "num_workers = 0\n",
    "# how many samples per batch to load\n",
    "batch_size = 20\n",
    "\n",
    "# convert data to torch.FloatTensor\n",
    "transform = transforms.ToTensor()\n",
    "\n",
    "# choose the training and test datasets\n",
    "train_data = datasets.MNIST(root='data', train=True,\n",
    "                                   download=True, transform=transform)\n",
    "test_data = datasets.MNIST(root='data', train=False,\n",
    "                                  download=True, transform=transform)\n",
    "\n",
    "# prepare data loaders\n",
    "def classify_label(dataset, num_classes):\n",
    "    list_index = [[] for _ in range(num_classes)]\n",
    "    for idx, datum in enumerate(dataset):\n",
    "        list_index[datum[1]].append(idx)\n",
    "    return list_index\n",
    "\n",
    "def partition_train(list_label2indices: list, num_per_class: int):\n",
    "    random_state = np.random.RandomState(0)\n",
    "    list_label2indices_train = []\n",
    "    for indices in list_label2indices:\n",
    "        random_state.shuffle(indices)\n",
    "        list_label2indices_train.extend(indices[:num_per_class])\n",
    "    return list_label2indices_train\n",
    "\n",
    "class Indices2Dataset(Dataset):\n",
    "    def __init__(self, dataset):\n",
    "        self.dataset = dataset\n",
    "        self.indices = None\n",
    "\n",
    "    def load(self, indices: list):\n",
    "        self.indices = indices\n",
    "\n",
    "    def __getitem__(self, idx):\n",
    "        idx = self.indices[idx]\n",
    "        image, label = self.dataset[idx]\n",
    "        return image, label\n",
    "\n",
    "    def __len__(self):\n",
    "        return len(self.indices)\n",
    "\n",
    "#  sort train data by label\n",
    "list_label2indices = classify_label(dataset=train_data, num_classes=10)\n",
    "\n",
    "# how many samples per class to train\n",
    "list_train = partition_train(list_label2indices, 500)\n",
    "\n",
    "# prepare data loaders  \n",
    "indices2data = Indices2Dataset(train_data)\n",
    "indices2data.load(list_train)\n",
    "train_loader = torch.utils.data.DataLoader(indices2data, batch_size=batch_size, num_workers=num_workers, shuffle=True)\n",
    "test_loader = torch.utils.data.DataLoader(test_data, batch_size=batch_size, num_workers=num_workers, shuffle=True)"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Visualize a Batch of Training Data\n",
    "\n",
    "The first step in a classification task is to take a look at the data, make sure it is loaded in correctly, then make any initial observations about patterns in that data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 239,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 2500x400 with 20 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "    \n",
    "# obtain one batch of training images\n",
    "dataiter = iter(train_loader)\n",
    "images, labels = dataiter.next()\n",
    "images = images.numpy()\n",
    "\n",
    "# plot the images in the batch, along with the corresponding labels\n",
    "fig = plt.figure(figsize=(25, 4))\n",
    "for idx in np.arange(20):\n",
    "    ax = fig.add_subplot(2, 20//2, idx+1, xticks=[], yticks=[])\n",
    "    ax.imshow(np.squeeze(images[idx]), cmap='gray')\n",
    "    # print out the correct label for each image\n",
    "    # .item() gets the value contained in a Tensor\n",
    "    ax.set_title(str(labels[idx].item()))"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### View an Image in More Detail"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 240,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x1200 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "img = np.squeeze(images[1])\n",
    "\n",
    "fig = plt.figure(figsize = (12,12)) \n",
    "ax = fig.add_subplot(111)\n",
    "ax.imshow(img, cmap='gray')\n",
    "width, height = img.shape\n",
    "thresh = img.max()/2.5\n",
    "for x in range(width):\n",
    "    for y in range(height):\n",
    "        val = round(img[x][y],2) if img[x][y] !=0 else 0\n",
    "        ax.annotate(str(val), xy=(y,x),\n",
    "                    horizontalalignment='center',\n",
    "                    verticalalignment='center',\n",
    "                    color='white' if img[x][y]<thresh else 'black')"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---\n",
    "## Set random seed\n",
    "A random seed is used to ensure that results are reproducible. In other words, using this parameter makes sure that anyone who re-runs your code will get the exact same outputs. Reproducibility is an extremely important concept in data science and other fields. More details to read: [How to Use Random Seeds Effectively](https://towardsdatascience.com/how-to-use-random-seeds-effectively-54a4cd855a79)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 241,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2023-10-13 17:31:21 INFO: tha value of the random seed: 2023\n"
     ]
    }
   ],
   "source": [
    "import random\n",
    "import os\n",
    "\n",
    "## give the number you like such as 2023\n",
    "seed_value = 2023\n",
    "\n",
    "np.random.seed(seed_value)\n",
    "random.seed(seed_value)\n",
    "os.environ['PYTHONHASHSEED'] = str(seed_value)\n",
    "\n",
    "torch.manual_seed(seed_value)     \n",
    "torch.cuda.manual_seed(seed_value)     \n",
    "torch.cuda.manual_seed_all(seed_value)   \n",
    "torch.backends.cudnn.deterministic = True\n",
    "logging.info(f\"tha value of the random seed: {seed_value}\")"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---\n",
    "## Define the Network Architecture (30 marks)\n",
    "\n",
    "* Input: a 784-dim Tensor of pixel values for each image.\n",
    "* Output: a 10-dim Tensor of number of classes that indicates the class scores for an input image. \n",
    "\n",
    "You need to implement three models:\n",
    "1. a vanilla multi-layer perceptron. (10 marks)\n",
    "2. a multi-layer perceptron with regularization (dropout or L2 or both). (10 marks)\n",
    "3. the corresponding loss functions and optimizers. (10 marks)"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Build model_1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "ename": "NameError",
     "evalue": "name 'nn' is not defined",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mNameError\u001b[0m                                 Traceback (most recent call last)",
      "\u001b[1;32mf:\\OneDrive\\课\\深度学习\\Assignment 1\\Assignment_1_XHD.ipynb 单元格 16\u001b[0m line \u001b[0;36m2\n\u001b[0;32m      <a href='vscode-notebook-cell:/f%3A/OneDrive/%E8%AF%BE/%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0/Assignment%201/Assignment_1_XHD.ipynb#X21sZmlsZQ%3D%3D?line=0'>1</a>\u001b[0m \u001b[39m## Define the MLP architecture\u001b[39;00m\n\u001b[1;32m----> <a href='vscode-notebook-cell:/f%3A/OneDrive/%E8%AF%BE/%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0/Assignment%201/Assignment_1_XHD.ipynb#X21sZmlsZQ%3D%3D?line=1'>2</a>\u001b[0m \u001b[39mclass\u001b[39;00m \u001b[39mVanillaMLP\u001b[39;00m(nn\u001b[39m.\u001b[39mModule):\n\u001b[0;32m      <a href='vscode-notebook-cell:/f%3A/OneDrive/%E8%AF%BE/%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0/Assignment%201/Assignment_1_XHD.ipynb#X21sZmlsZQ%3D%3D?line=2'>3</a>\u001b[0m     \u001b[39mdef\u001b[39;00m \u001b[39m__init__\u001b[39m(\u001b[39mself\u001b[39m):\n\u001b[0;32m      <a href='vscode-notebook-cell:/f%3A/OneDrive/%E8%AF%BE/%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0/Assignment%201/Assignment_1_XHD.ipynb#X21sZmlsZQ%3D%3D?line=3'>4</a>\u001b[0m         \u001b[39msuper\u001b[39m(VanillaMLP, \u001b[39mself\u001b[39m)\u001b[39m.\u001b[39m\u001b[39m__init__\u001b[39m()\n",
      "\u001b[1;31mNameError\u001b[0m: name 'nn' is not defined"
     ]
    }
   ],
   "source": [
    "## Define the MLP architecture\n",
    "class VanillaMLP(nn.Module):\n",
    "    def __init__(self):\n",
    "        super(VanillaMLP, self).__init__()\n",
    "        \n",
    "        # implement your codes here\n",
    "        self.linear1 = torch.nn.Linear(784,100)\n",
    "        self.linear2 = torch.nn.Linear(100,10)\n",
    "        \n",
    "\n",
    "    def forward(self, x):\n",
    "        # flatten image input\n",
    "        x = x.view(-1, 28 * 28)#view ()相当于reshape、resize，重新调整Tensor的形状,-1表示不确定的数\n",
    "\n",
    "        # implement your codes here\n",
    "        h_relu = self.linear1(x).clamp(min=0)\n",
    "        x = self.linear2(h_relu)\n",
    "        \n",
    "        \n",
    "        return x\n",
    "\n",
    "# initialize the MLP\n",
    "model_1 = VanillaMLP()\n",
    "\n",
    "# specify loss function\n",
    "# implement your codes here\n",
    "loss_model_1 = torch.nn.CrossEntropyLoss() #十分类任务\n",
    "\n",
    "# specify your optimizer\n",
    "# implement your codes here\n",
    "optimizer_model_1 = torch.optim.SGD(model_1.parameters(),lr=0.05)"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Build model_2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 243,
   "metadata": {},
   "outputs": [],
   "source": [
    "## Define the MLP architecture\n",
    "class RegularizedMLP(nn.Module):\n",
    "    def __init__(self):\n",
    "        super(RegularizedMLP, self).__init__()\n",
    "        \n",
    "        # implement your codes here\n",
    "        self.layer1 = torch.nn.Sequential(\n",
    "            \n",
    "            nn.Linear(784,100),\n",
    "            nn.ReLU(),\n",
    "            nn.Dropout(0.2),\n",
    "            )\n",
    "        self.layer2 =  torch.nn.Sequential(\n",
    "            nn.Linear(100,10),\n",
    "            nn.ReLU(), \n",
    "            nn.Dropout(0.2)\n",
    "            #nn.BatchNorm2d(100)\n",
    "            )\n",
    "        \n",
    "\n",
    "    def forward(self, x):\n",
    "        # flatten image input\n",
    "        x = x.view(-1, 28 * 28)\n",
    "\n",
    "        # implement your codes here\n",
    "        x = self.layer1(x)\n",
    "        x = self.layer2(x)\n",
    "        \n",
    "        return x\n",
    "\n",
    "\n",
    "# initialize the MLP\n",
    "model_2 = RegularizedMLP()\n",
    "\n",
    "# specify loss function\n",
    "# implement your codes here\n",
    "loss_model_2 = torch.nn.CrossEntropyLoss()\n",
    "\n",
    "# specify your optimizer\n",
    "# implement your codes here\n",
    "optimizer_model_2 = torch.optim.SGD(model_2.parameters(),lr=0.05)"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---\n",
    "## Train the Network (30 marks)\n",
    "\n",
    "Train your models in the following two cells.\n",
    "\n",
    "The following loop trains for 30 epochs; feel free to change this number. For now, we suggest somewhere between 20-50 epochs. As you train, take a look at how the values for the training loss decrease over time. We want it to decrease while also avoiding overfitting the training data. \n",
    "\n",
    "We will introduce some metrics of classification tasks and you will learn how implement these metrics with scikit-learn.\n",
    "\n",
    "There are supply some references for you to learn: [evaluation_metrics_spring2020](https://cs229.stanford.edu/section/evaluation_metrics_spring2020.pdf).\n",
    "\n",
    "In training processing, we will use accuracy,  Area Under ROC and top k accuracy.\n",
    "\n",
    "**The key parts in the training process are left for you to implement.**"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Train model_1\n",
    "#### Train model_1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 244,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2023-10-13 17:31:21 INFO: GPU is available: True\n",
      "2023-10-13 17:31:21 INFO: Train model on 1 GPUs:\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\t GPU 0.: NVIDIA GeForce GTX 1650\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2023-10-13 17:31:22 INFO: Epoch: 1 \tTraining Loss: 0.055782 \tTraining Acc: 72.66% \t top 3 Acc: 0.91% \t AUC Score: 0.8848\n",
      "2023-10-13 17:31:24 INFO: Epoch: 2 \tTraining Loss: 0.022577 \tTraining Acc: 87.36% \t top 3 Acc: 0.97% \t AUC Score: 0.9625\n",
      "2023-10-13 17:31:25 INFO: Epoch: 3 \tTraining Loss: 0.018006 \tTraining Acc: 89.66% \t top 3 Acc: 0.98% \t AUC Score: 0.9695\n",
      "2023-10-13 17:31:26 INFO: Epoch: 4 \tTraining Loss: 0.015784 \tTraining Acc: 91.10% \t top 3 Acc: 0.98% \t AUC Score: 0.9727\n",
      "2023-10-13 17:31:28 INFO: Epoch: 5 \tTraining Loss: 0.014200 \tTraining Acc: 91.80% \t top 3 Acc: 0.99% \t AUC Score: 0.9754\n",
      "2023-10-13 17:31:29 INFO: Epoch: 6 \tTraining Loss: 0.012953 \tTraining Acc: 92.36% \t top 3 Acc: 0.99% \t AUC Score: 0.9777\n",
      "2023-10-13 17:31:31 INFO: Epoch: 7 \tTraining Loss: 0.011576 \tTraining Acc: 92.94% \t top 3 Acc: 0.99% \t AUC Score: 0.9797\n",
      "2023-10-13 17:31:32 INFO: Epoch: 8 \tTraining Loss: 0.010691 \tTraining Acc: 94.10% \t top 3 Acc: 0.99% \t AUC Score: 0.9814\n",
      "2023-10-13 17:31:33 INFO: Epoch: 9 \tTraining Loss: 0.009657 \tTraining Acc: 94.70% \t top 3 Acc: 0.99% \t AUC Score: 0.9829\n",
      "2023-10-13 17:31:34 INFO: Epoch: 10 \tTraining Loss: 0.008905 \tTraining Acc: 94.68% \t top 3 Acc: 0.99% \t AUC Score: 0.9842\n",
      "2023-10-13 17:31:35 INFO: Epoch: 11 \tTraining Loss: 0.008193 \tTraining Acc: 95.46% \t top 3 Acc: 0.99% \t AUC Score: 0.9851\n",
      "2023-10-13 17:31:37 INFO: Epoch: 12 \tTraining Loss: 0.007532 \tTraining Acc: 95.86% \t top 3 Acc: 0.99% \t AUC Score: 0.9860\n",
      "2023-10-13 17:31:38 INFO: Epoch: 13 \tTraining Loss: 0.006931 \tTraining Acc: 96.18% \t top 3 Acc: 0.99% \t AUC Score: 0.9871\n",
      "2023-10-13 17:31:39 INFO: Epoch: 14 \tTraining Loss: 0.006398 \tTraining Acc: 96.74% \t top 3 Acc: 0.99% \t AUC Score: 0.9879\n",
      "2023-10-13 17:31:40 INFO: Epoch: 15 \tTraining Loss: 0.005822 \tTraining Acc: 97.06% \t top 3 Acc: 1.00% \t AUC Score: 0.9889\n",
      "2023-10-13 17:31:42 INFO: Epoch: 16 \tTraining Loss: 0.005261 \tTraining Acc: 97.48% \t top 3 Acc: 1.00% \t AUC Score: 0.9894\n",
      "2023-10-13 17:31:43 INFO: Epoch: 17 \tTraining Loss: 0.004857 \tTraining Acc: 97.76% \t top 3 Acc: 1.00% \t AUC Score: 0.9900\n",
      "2023-10-13 17:31:44 INFO: Epoch: 18 \tTraining Loss: 0.004407 \tTraining Acc: 98.10% \t top 3 Acc: 1.00% \t AUC Score: 0.9906\n",
      "2023-10-13 17:31:45 INFO: Epoch: 19 \tTraining Loss: 0.004135 \tTraining Acc: 98.26% \t top 3 Acc: 1.00% \t AUC Score: 0.9908\n",
      "2023-10-13 17:31:47 INFO: Epoch: 20 \tTraining Loss: 0.003704 \tTraining Acc: 98.46% \t top 3 Acc: 1.00% \t AUC Score: 0.9914\n"
     ]
    }
   ],
   "source": [
    "# import scikit-learn packages\n",
    "# please use the function imported from scikit-learn to metric the process of training of the model\n",
    "from sklearn.metrics import accuracy_score,roc_auc_score, top_k_accuracy_score\n",
    "# number of epochs to train the model\n",
    "n_epochs = 20  # suggest training between 20-50 epochs\n",
    "\n",
    "model_1.train() # prep model for training\n",
    "\n",
    "train_loss_list = []\n",
    "train_acc_list = []\n",
    "train_auc_list = []\n",
    "train_top_k_acc_list = []\n",
    "\n",
    "\n",
    "# GPU check\n",
    "logging.info(f'GPU is available: {torch.cuda.is_available()}')\n",
    "if torch.cuda.is_available():\n",
    "    gpu_num = torch.cuda.device_count()\n",
    "    logging.info(f\"Train model on {gpu_num} GPUs:\")\n",
    "    for i in range(gpu_num):\n",
    "        print('\\t GPU {}.: {}'.format(i,torch.cuda.get_device_name(i)))\n",
    "    model_1 = model_1.cuda()\n",
    "\n",
    "for epoch in range(n_epochs):\n",
    "    # monitor training loss\n",
    "    train_loss = 0.0\n",
    "    pred_array = None\n",
    "    label_array =  None\n",
    "    \n",
    "    one_hot_label_matrix = None\n",
    "    pred_matrix = None\n",
    "    \n",
    "\n",
    "    for data, label in train_loader:\n",
    "        data = data.cuda()\n",
    "        label = label.cuda()\n",
    "        # implement your code here\n",
    "        \n",
    "        optimizer_model_1.zero_grad()\n",
    "\n",
    "        \n",
    "        pred = model_1(data)\n",
    "        #torch.Tensor.cpu(pred)\n",
    "        loss_model_1(pred,label).backward()\n",
    "        \n",
    "        optimizer_model_1.step()\n",
    "        train_loss += loss_model_1(pred,label)\n",
    "\n",
    "        #label=label.cpu().numpy()\n",
    "        one_hot_matrix = np.eye(10)[label.cpu().numpy()]\n",
    "        #print(one_hot_matrix)\n",
    "       \n",
    "        # finish the the computation of variables of metric\n",
    "        # implement your codes here\n",
    "        if pred_matrix is None:\n",
    "            pred_matrix = pred.cpu().detach().numpy()     #torch.Tensor.__array__(pred)#y_score  #pred [20,10]\n",
    "            #print(pred_matrix)\n",
    "        else:\n",
    "            pred_matrix = np.concatenate((pred_matrix,pred.cpu().detach().numpy()))\n",
    "\n",
    "        if one_hot_label_matrix is None:\n",
    "            one_hot_label_matrix = one_hot_matrix#torch.nn.functional.one_hot(label) #y_true 也应该[20,10]\n",
    "        else:\n",
    "            one_hot_label_matrix = np.concatenate((one_hot_label_matrix,one_hot_matrix)) #one_hot_label_matrix.cpu().detach().numpy()\n",
    "\n",
    "        pred = torch.argmax(pred, axis=1)\n",
    "        if pred_array is None:\n",
    "            pred_array = pred.cpu().detach().numpy() #1*20 or 20*1\n",
    "        else:\n",
    "            pred_array = np.concatenate((pred_array,pred.cpu().detach().numpy())) #pred_array\n",
    "\n",
    "        if label_array is None:\n",
    "            label_array = label.cpu().numpy() #y_true  20*1 没问题\n",
    "        else:\n",
    "            label_array = np.concatenate((label_array,label.cpu().numpy()))\n",
    "\n",
    "        \n",
    "\n",
    "\n",
    "        \n",
    "        \n",
    "    # print training statistics \n",
    "    # read the API document at https://scikit-learn.org/stable/modules/classes.html#module-sklearn.metrics to finish your code\n",
    "    # don't craft your own code\n",
    "    # calculate average loss and accuracy over an epoch\n",
    "    \n",
    "    top_k = 3\n",
    "    train_loss = train_loss / len(train_loader.dataset)\n",
    "    train_acc = 100. * accuracy_score(label_array, pred_array)\n",
    "    train_auc = roc_auc_score(one_hot_label_matrix, pred_matrix , multi_class='ovo')\n",
    "    top_k_acc = top_k_accuracy_score(label_array, pred_matrix , k=top_k,)\n",
    "    # append the value of the metric to the list\n",
    "    train_loss_list.append(train_loss.cpu().detach().numpy())\n",
    "    train_acc_list.append(train_acc)\n",
    "    train_auc_list.append(train_auc)\n",
    "    train_top_k_acc_list.append(top_k_acc)\n",
    "    \n",
    "    logging.info('Epoch: {} \\tTraining Loss: {:.6f} \\tTraining Acc: {:.2f}% \\t top {} Acc: {:.2f}% \\t AUC Score: {:.4f}'.format(\n",
    "        epoch+1, \n",
    "        train_loss,\n",
    "        train_acc,\n",
    "        top_k,\n",
    "        top_k_acc,\n",
    "        train_auc,\n",
    "        ))"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Visualize the training process of the model_1\n",
    "Please read the [document](https://matplotlib.org/stable/tutorials/introductory/pyplot.html#sphx-glr-tutorials-introductory-pyplot-py) to finish the training process visualization.\n",
    "For more information, please refer to the [document](https://matplotlib.org/stable/tutorials/index.html)\n",
    "##### Plot the change of the loss of model_1 during training"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 245,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 2000x800 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "epochs_list = list(range(1,n_epochs+1))\n",
    "plt.figure(figsize=(20, 8))\n",
    "plt.plot(epochs_list, train_loss_list)\n",
    "plt.title('Model_1 loss')\n",
    "plt.ylabel('Loss')\n",
    "plt.xlabel('Epoch')\n",
    "plt.legend(['Train'], loc='upper right')\n",
    "plt.show()"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Plot the change of the accuracy of model_1 during training"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 246,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABlEAAAK7CAYAAABieqm4AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8pXeV/AAAACXBIWXMAAA9hAAAPYQGoP6dpAAB/yklEQVR4nOz9eZyddX03/r/O7JPZkpA9JBOWALIvoShqAS2CgrXu1VZR7vbnXWyV232pSl1A6Y211d7l7vdWVFDEfaG1VFGsKNok7IsQkISE7OtMMpn9/P6YyZAhE0gmkzkzk+fz8TiPOdd1Ptc17zOEkJwX78+7UCwWiwEAAAAAAGCQslIXAAAAAAAAMBYJUQAAAAAAAIYgRAEAAAAAABiCEAUAAAAAAGAIQhQAAAAAAIAhCFEAAAAAAACGIEQBAAAAAAAYghAFAAAAAABgCEIUAAAAAACAIQhRAABgAvjyl7+cQqGQQqGQ2267bY/Xi8Vijj766BQKhZx77rkj+r0LhUKuuOKK/b5u+fLlKRQK+fKXv7xf1331q1/Nn/7pn+bYY49NWVlZFixYsN/fGwAAYF8IUQAAYAJpaGjIF7/4xT3O/+IXv8hjjz2WhoaGElQ1sq6//vo88MAD+YM/+IMcddRRpS4HAACYwIQoAAAwgbz+9a/Pd77znbS0tAw6/8UvfjHPe97zMn/+/BJVNnJuueWW3Hfffbn++utz7LHHlrqcg6qrqyvd3d2lLgMAAA5ZQhQAAJhA3vCGNyRJbrzxxoFz27Zty3e+851ceumlQ16zefPmXHbZZZk7d26qqqpy5JFH5sMf/nA6OjoGrWtpaclf/uVf5rDDDkt9fX0uvPDCPPLII0Pec9myZXnjG9+YGTNmpLq6Os95znPyz//8zyPyHsvKhv/XmPb29rz73e/OqaeemqampkydOjXPe97z8oMf/GCPtb29vfn85z+fU089NbW1tZk8eXKe+9zn5oc//OGgdV//+tfzvOc9L/X19amvr8+pp546qBtowYIFectb3rLH/c8999xBW6vddtttKRQKuf766/Pud787c+fOTXV1dR599NFs2LAhl112WY4//vjU19dnxowZedGLXpRf/vKXe9y3o6MjH//4x/Oc5zwnNTU1Oeyww3Leeefl17/+dZLkxS9+cY477rgUi8VB1+3a8u2iiy7anx8pAABMaBWlLgAAABg5jY2Nec1rXpMvfelLedvb3pakL1ApKyvL61//+nzuc58btL69vT3nnXdeHnvssfzd3/1dTj755Pzyl7/MVVddlbvvvjv/9m//lqTvA/Y/+ZM/ya9//et89KMfzZlnnplf/epXeelLX7pHDQ8++GDOPvvszJ8/P9dcc01mzZqVW265Je94xzuycePGfOxjHzvoP4e96ejoyObNm/Oe97wnc+fOTWdnZ37605/mVa96Va677rq8+c1vHlj7lre8JTfccEP+x//4H/n4xz+eqqqq3HnnnVm+fPnAmo9+9KP5xCc+kVe96lV597vfnaamptx///1ZsWLFsGv84Ac/mOc973m59tprU1ZWlhkzZmTDhg1Jko997GOZNWtWtm/fnu9973s599xzc+uttw6EMd3d3XnpS1+aX/7yl7n88svzohe9KN3d3fnNb36TJ554ImeffXbe+c535hWveEVuvfXW/NEf/dHA9/3xj3+cxx57LP/0T/807NoBAGCiEaIAAMAEc+mll+a8887LAw88kBNOOCFf+tKX8trXvnbIeShf+cpXcu+99+ab3/xmXvva1yZJzj///NTX1+f9739/fvKTn+T888/PLbfckp///Of5x3/8x7zjHe8YWFdVVZUPf/jDg+75rne9Kw0NDbn99tvT2Ng4sLajoyOf/vSn8453vCNTpkw5yD+FoTU1NeW6664bOO7p6cmLX/zibNmyJZ/73OcGQpRf/vKXuf766/PhD384n/zkJwfWX3jhhQPPH3/88Vx55ZX5sz/7s9xwww0D588///wDqvGoo47Kt771rUHnpk6dmv/zf/7PoLovuOCCLF++PP/0T/80EKLceOON+fnPf57/7//7//IXf/EXA+tf/vKXDzy/+OKLc+SRR+YLX/jCoBDlC1/4Qo466qghgzEAADhU2c4LAAAmmHPOOSdHHXVUvvSlL+W+++7L4sWL97qV189+9rPU1dXlNa95zaDzu7afuvXWW5MkP//5z5Mkf/ZnfzZo3Rvf+MZBx+3t7bn11lvzyle+MpMmTUp3d/fA42Uve1na29vzm9/8ZiTe5rB961vfyvOf//zU19enoqIilZWV+eIXv5iHHnpoYM2Pf/zjJMnb3/72vd7nJz/5SXp6ep5xzXC8+tWvHvL8tddem9NPPz01NTUDdd9666171F1TU7PXf95J33Zof/3Xf52bb745TzzxRJLksccey3/8x3/ksssuS6FQGNH3AwAA45kQBQAAJphCoZC3vvWtueGGG3LttdfmmGOOyQtf+MIh127atCmzZs3a44PzGTNmpKKiIps2bRpYV1FRkcMOO2zQulmzZu1xv+7u7nz+859PZWXloMfLXvayJMnGjRtH6q3ut+9+97t53etel7lz5+aGG27IHXfcMRAytbe3D6zbsGFDysvL93h/u9u1xdbhhx8+ojXOnj17j3Of/exn81d/9Vc566yz8p3vfCe/+c1vsnjx4lx44YXZuXPnoJrmzJnzrHNjLr300tTW1ubaa69NkvzzP/9zamtrnzF8AQCAQ5HtvAAAYAJ6y1veko9+9KO59tpr86lPfWqv6w477LD89re/TbFYHBSkrF+/Pt3d3Zk2bdrAuu7u7mzatGlQkLJ27dpB95syZUrKy8vzpje9aa8dGkccccSBvLUDcsMNN+SII47ITTfdNOj9dnR0DFo3ffr09PT0ZO3atUOGGrvWJMmqVasyb968vX7PmpqaPe6f9IVJu36+uxuqE+SGG27Iueeem3/5l38ZdL61tXWPmm6//fb09vY+Y5DS1NSUSy65JP/v//2/vOc978l1112XN77xjZk8efJerwEAgEORThQAAJiA5s6dm/e+9715+ctfnksuuWSv61784hdn+/bt+f73vz/o/Fe/+tWB15PkvPPOS5J87WtfG7Tu61//+qDjSZMm5bzzzstdd92Vk08+OYsWLdrj8fRultFUKBRSVVU1KKhYu3ZtfvCDHwxat2suyNNDi9295CUvSXl5+TOuSZIFCxbk3nvvHXTukUceycMPP7xfdVdXVw86d++99+aOO+7Yo+729vZ8+ctfftZ7vuMd78jGjRvzmte8Jlu3bs1f//Vf73M9AABwqNCJAgAAE9SnP/3pZ13z5je/Of/8z/+cSy65JMuXL89JJ52U22+/PVdeeWVe9rKXDQwef8lLXpI//MM/zPve977s2LEjixYtyq9+9atcf/31e9zzH//xH/OCF7wgL3zhC/NXf/VXWbBgQVpbW/Poo4/mRz/6UX72s58d0Pt68MEH8+CDDybpC0Da2try7W9/O0ly/PHH5/jjj9/rtRdffHG++93v5rLLLstrXvOarFy5Mp/4xCcye/bsLFu2bGDdC1/4wrzpTW/KJz/5yaxbty4XX3xxqqurc9ddd2XSpEn5m7/5myxYsCAf+tCH8olPfCI7d+7MG97whjQ1NeXBBx/Mxo0b83d/93dJkje96U358z//81x22WV59atfnRUrVuTqq68e6GTZFxdffHE+8YlP5GMf+1jOOeecPPzww/n4xz+eI444It3d3QPr3vCGN+S6667L//yf/zMPP/xwzjvvvPT29ua3v/1tnvOc5+RP//RPB9Yec8wxufDCC/PjH/84L3jBC3LKKafscz0AAHCoEKIAAMAhrKamJj//+c/z4Q9/OH//93+fDRs2ZO7cuXnPe96Tj33sYwPrysrK8sMf/jDvete7cvXVV6ezszPPf/7z8+///u857rjjBt3z+OOPz5133plPfOIT+du//dusX78+kydPzsKFCwfmohyIb37zmwMBxS6vfe1rkyQf+9jHcsUVV+z12re+9a1Zv359rr322nzpS1/KkUcemQ984ANZtWrVHvf88pe/nNNPPz1f/OIX8+Uvfzm1tbU5/vjj86EPfWhgzcc//vEsXLgwn//85/Nnf/ZnqaioyMKFC/OOd7xjYM0b3/jGrF69Otdee22uu+66nHjiifmXf/mXPb7fM/nwhz+ctra2fPGLX8zVV1+d448/Ptdee22+973v5bbbbhtYV1FRkX//93/PVVddlRtvvDGf+9zn0tDQkFNOOSUXXnjhHvd9/etfnx//+Me6UAAAYC8KxWKxWOoiAAAAGH2vfvWr85vf/CbLly9PZWVlqcsBAIAxRycKAADAIaSjoyN33nln/vu//zvf+9738tnPflaAAgAAe6ETBQAAGBN2n+0xlLKyspSVlY1SNRPX8uXLc8QRR6SxsTFvfOMb84UvfCHl5eWlLgsAAMYkIQoAAFByuz7YfybPNu8EAABgpNnOCwAAKLk5c+Zk8eLFz7oGAABgNOlEAQAAAAAAGIINhQEAAAAAAIYw4bfz6u3tzerVq9PQ0JBCoVDqcgAAAAAAgBIqFotpbW3NnDlzUlb2zL0mEz5EWb16debNm1fqMgAAAAAAgDFk5cqVOfzww59xzYQPURoaGpL0/TAaGxtLXA0AAAAAAFBKLS0tmTdv3kB+8EwmfIiyawuvxsZGIQoAAAAAAJAk+zQCxGB5AAAAAACAIQhRAAAAAAAAhiBEAQAAAAAAGMKEn4kCAAAAAADjRbFYTHd3d3p6ekpdyrhVXl6eioqKfZp58myEKAAAAAAAMAZ0dnZmzZo1aWtrK3Up496kSZMye/bsVFVVHdB9hCgAAAAAAFBivb29efzxx1NeXp45c+akqqpqRDopDjXFYjGdnZ3ZsGFDHn/88SxcuDBlZcOfbCJEAQAAAACAEuvs7Exvb2/mzZuXSZMmlbqcca22tjaVlZVZsWJFOjs7U1NTM+x7GSwPAAAAAABjxIF0TfCUkfo5+qcBAAAAAAAwBCEKAAAAAADAEIQoAAAAAADAmHLuuefm8ssvL3UZBssDAAAAAADDUygUnvH1Sy65JF/+8pf3+77f/e53U1lZOcyqRo4QBQAAAAAAGJY1a9YMPL/pppvy0Y9+NA8//PDAudra2kHru7q69ikcmTp16sgVeQBs5wUAAAAAAGNQsVhMW2d3SR7FYnGfapw1a9bAo6mpKYVCYeC4vb09kydPzje/+c2ce+65qampyQ033JBNmzblDW94Qw4//PBMmjQpJ510Um688cZB9336dl4LFizIlVdemUsvvTQNDQ2ZP39+/vVf/3Ukf9xD0okCAAAAAABj0M6unhz/0VtK8r0f/PgFmVQ1MhHC+9///lxzzTW57rrrUl1dnfb29pxxxhl5//vfn8bGxvzbv/1b3vSmN+XII4/MWWedtdf7XHPNNfnEJz6RD33oQ/n2t7+dv/qrv8of/uEf5rjjjhuROociRAEAAAAAAA6ayy+/PK961asGnXvPe94z8Pxv/uZv8h//8R/51re+9Ywhyste9rJcdtllSfqCmX/4h3/IbbfdJkQBAAAAAIBDTW1leR78+AUl+94jZdGiRYOOe3p68ulPfzo33XRTnnzyyXR0dKSjoyN1dXXPeJ+TTz554PmubcPWr18/YnUORYgCAAAAAABjUKFQGLEttUrp6eHINddck3/4h3/I5z73uZx00kmpq6vL5Zdfns7Ozme8z9MH0hcKhfT29o54vbsb/z99AAAAAABg3PjlL3+ZV7ziFfnzP//zJElvb2+WLVuW5zznOSWubE9lpS4AAAAAAAA4dBx99NH5yU9+kl//+td56KGH8ra3vS1r164tdVlDEqIAAAAAAACj5iMf+UhOP/30XHDBBTn33HMza9as/Mmf/EmpyxpSoVgsFktdxMHU0tKSpqambNu2LY2NjaUuBwAAAAAA9tDe3p7HH388RxxxRGpqakpdzrj3TD/P/ckNdKIAAAAAAAAMwWB5AAAAAAAY53qLxfT0FtPbW0xP8amvleVlmVQlChguPzkAAAAAACiRYrGYju7ebNrRka6e3uzs7E5XuvpDkPQFI/0ByaDnxWJ6ezPwfG+TO6bWVQlRDoCfHAAAAAAADEOxWEx7V29a27vS0t6d1vautLZ39z+6Br62PP1cx+B1XT3FzG0ozxXnzUhxc1sKFd3DrqmsUEh5WSHlhULKygqpKjfV40AIUQAAAAAAOOQUi8Xs7OrZe9CxWyDSMsS5Xc+7e4fuANlfvcWkUEgqygqprCwfCEHKC4WUl2W354XBz3e93n9cKBRGpJ7xbm+dOftLiAIAAAAAwLhSLBbT1vn0AGToLpC+EGTPLpDtIxiAFApJfXVFGmsq01BT0f8Y6nllGoc411BTkdryQh59dFlm1JXlsMMaRqSuQ1lbW1uSpLKy8oDuI0QBAAAAAOCg6e0tpq2rJ20d3dnR2ZMdHd1p6+xJW2ff113HOzq709bxtK97e72zJz0jFICU9Qcgu8KMoYOQp841DnGurqoiZWUH3gEyefLkrF+/PkkyadIkXSXDUCwW09bWlvXr12fy5MkpLy8/oPsJUQAAAAAASJJ09fTuFlR0Z8e+hhpDvL7r+p1dPQet3vKyQhpqKp4Wguy9C2TP1ytTV1U+ZsKKWbNmJclAkMLwTZ48eeDneSCEKAAAAAAA40yxWExHd++gQGNHx1PBRVtnX9fHru6Pga9DvL57INLZ03vQai4UkrqqikyqKk9ddf/XqopMqi7f8/w+vN5QU5HayrETgIyEQqGQ2bNnZ8aMGenq6ip1OeNWZWXlAXeg7CJEAQAAAAAogbbO7qxv6ci6lvasb+37uqWtc+gQZIiujxHazWpIVeVlg8KLSdUVqasqz6SqitRV93/d/fxur/eFHLvWPBWC1FSWTajA42AqLy8fsRCAAyNEAQAAAAAYQTs6ugdCkfWtHVm/W0iyvqUj61rbs6GlI60d3SPy/Wornwot9trFMVQ3x27ndw89aqvKU1VRNiK1wXgnRAEAAAAA2AfbO7oHgpD1re17dJFs6P+6o3PfZ4DUVpZnZmN1ZjTWZEZDdabVV6e+es/Qo7Zq8PGu7o/ayvKUj8BAc2BoQhQAAAAA4JBVLBb7w5GngpH1re39x4PDkbb9CEfqqsozs7Em0xuqM7M/IJnRWL3HufrqCltcwRgmRAEAAAAAJpxisZiW9u5sGAhE+r8OEZbs7Nr3cKS+uiIzGqszY7cgZM+wpCb11T56hYnAv8kAAAAAwLhRLBbTsrN7t26R9r10kbSnvat3n+/bUFPRF4A01AzaXmtGY01mNjx1XCccgUOKf+MBAAAAgJIrFovZtrNrr8HIroHs61s60tG97+FIY01FXxDS2BeQzGjcLSjZ7WttVflBfHfAeCVEAQAAAAAOmp7eYra0dWbj9o5Bg9jX7zaQfX1r3/yRzv0IR5pqKwdtqTVjt621ZjRWZ2Z/YFJTKRwBhk+IAgAAAADss87u3mxp68ym7Z3ZvKMzm3Z0ZPOOXc87s/lp57fu7EqxuO/3nzypciAA2dU5sms7rV1dI9MbhCPA6BCiAAAAAMAhrL2rZyD8GCoQ2bSjM5v7z2/a0ZnW9u5hfZ8pkyr3GMD+9C4S4Qgw1ghRAAAAAGCCKBaL2d7RPRCEDIQhu55vHxyIbN7RmbbOnv3+PmWFZGpd1cDjsLrqp57X73l+yqTKVJSXHYR3DHBwCVEAAAAAYIzq7S2mpb1rIPDYtYXW5h0dQ4Qjndnc1rlfc0V2qSwv9Acf1TlsUDhSlan1/V/7A5HD6qrSVFuZsrLCQXjHAGOLEAUAAAAARsmuIet7C0QGzxTpzJa2zvT07sdAkX41lWWDu0N2BSNDBCJT66vSUF2RQkEoAvB0QhQAAAAAGKaunt5BgchID1nfpaG6IlPrBwciU+qeCkQGdY/UV2VSlY/9AEaC300BAAAAYC/au3qyeuvOPLl1Z1Zt2Zknt+x63pYnt+zM2pb2DKNRJJMnVQ7uENktCHn6TJEpdZWprjBsHaAUhCgAAAAAHLLaOrvz5JadWbVbSLJqS9tAaLKhteNZ72HIOsDEJUQBAAAAYMJqae/qD0Z25sktbX1fd+ss2byj81nvMamqPHMn12bulNocPqU2cydP6vvafzytrtqQdYAJSogCAAAAwLhULBazta1rYHutVbvCkq1PhSYt7d3Pep+G6or+QKQvHOkLSvqO506pzZRJlYauAxyihCgAAAAAjEnFYjEbt3cO2l7r6TNJdnT2POt9pkyq7AtEJtcOdJDsHpI01VaOwrsBYDwSogAAAABQEr29xaxv7RgUkqzabSbJk1t2pqO791nvM62+etD2WofvFpDMnVybumofgQEwPP4LAgAAAMBB0d3TmzXb2p/WRfLUXJLVW3emq6f4jPcoFJJZjTWDukh27yqZM7k2NZXlo/SOADjUCFEAAAAAGJaO7p6s2do+aHutVVt2ZlV/F8nalvb09D5zSFJeVsjsppqBge1P7yaZ1VSTqoqyUXpHADCYEAUAAACAIbV39eyxvdaq3WaSrG/tSPGZM5JUlZdlzuSaPWaS7Npua2ZDdSrKhSQAjE1CFAAAAIBDSLFYzLadXdnQ2tH32N7x1PPdjte3dmTzjs5nvV9NZdleh7YfPqU20+urU1ZWGIV3BgAjT4gCAAAAMAG0dXZnY2tnNmxvHzIU2f342eaQ7K6+uqJ/q62hZ5JMratKoSAkAWBiEqIAAAAAjFFdPb3ZtL2zP/xo7w9Jhu4c2d7RvV/3njypMtPrqzO9of+x+/OG6kyrr86cpto01lYISQA4ZAlRAAAAAEZRsVjM1rauvYYhux/vy3Zau6upLMuMhpohQ5Hdjw+rr0p1RflBeocAMHEIUQAAAABGQFtn97OGIhtaO7JxP7fTKi8rZFp91Z7BSH11pu8KTPofdVXlukYAYAQJUQAAAAD24unbaT1TSLKjs2e/7v1s22ntOjdlUpXB7ABQIkIUAAAA4JByMLfTqq0sz4zGwYHItHrbaQHAeCVEAQAAACakzu7erNi0I8vWb8+ydduzbH1rHl2/Pb/fsCOdPb37fJ+KssJTQcgzdIxMb6hOXbWPWgBgIvFfdgAAAGBca+/qye837BgISZat255HN2zP8o070t2799kjUyZV7j0YqX9q1sjk2krbaQHAIUqIAgAAAIwLbZ3deWx9X1iyq7vk0fWteWJzW/aWldRXV+ToGfVZOKM+C2fWZ+GMhhw1vT6zmmpSVVE2um8AABh3hCgAAADAmNLa3tXXUbJ+e39nSV9osmrLzr1e01RbORCUHD2jYeD5rMaaFAq6SACA4RGiAAAAACWxta1zj3klj67fnjXb2vd6zbT6qv7OkoaBDpOjZ9Znen21sAQAGHFCFAAAAOCgKRaL2bSjc2DrradCk+3ZuL1jr9fNbKx+KiiZ+VRoMrWuahSrBwAOdUIUAAAA4IAVi8Wsb+0Y6CpZtn57Hu1/vqWta6/XzZ1cO2hmydH9YUlTbeUoVg8AMDQhCgAAALDPenuLWb1t50BI0je7pC80aW3vHvKaQiGZN2XSwNZbC/tnlhw1oz711T6aAADGLn9SAQAAAPbQ01vMqi1tA1tv7T6zpK2zZ8hryssKaZ46aY8tuI6aXp/aqvJRfgcAAAdOiAIAAACHsO6e3qzY3LbHzJLHNmxPR3fvkNdUlhdyxLS6PWaWLJg2KdUVwhIAYOIQogAAAMAhoKO7J8s3tg10lOzajuv3G7enq6c45DVVFWU5anr/vJLdZpY0HzYpleVlo/wOAABGnxAFAAAAJpD2rp48tqF/VsluQ95XbGpLT+/QYUltZfnAcPfdZ5bMmzop5WWFUX4HAABjhxAFAAAAxoGe3mI27ejIhtbdHtsHH6/Z1p6VW9pSHDorSX11xUBYsvvMkrmTa1MmLAEA2IMQBQAAAEqkWCymZWd3Nmxvz/pnCEc2bu/Iph2dew1Hnq6ptjLH9G+9tXBG/cDcklmNNSkUhCUAAPtKiAIAAAAjrL2rJxtaO54KRrbv2UGysf95Z8/Qw9uHUigkh9VVZ3pD/6N+t+cN1ZnRUJ2jptdnWn2VsAQAYAQIUQAAAGAfdPf0ZvOOzr5gZIhQZEPrU8FIa0f3ft27oaZiyFDk6cdTJ1WlwkB3AIBRI0QBAADgkHWwttNKkqqKsszYSxiy+/G0+urUVJYfvDcJAMCwCVEAAACYcHZ29mTj9pHfTquskBxW/+wdI9MbqtNQXWFLLQCAcU6IAgAAwLjQ3dObTTs6h9xG60C302qqrcy0+qr+AKRmryHJ1LqqlJcJRgAADhVCFAAAAEqut7eYDds78uTWnVk98GgfOF67rT2b2/ZvO63qirLMaHxal0h9zR4dI4fVVdlOCwCAIQlRAAAAOOh2dHRn9dad/aFI+0BQ8uTWnVm9rS8k6ep59oSkvKyQw+qq9r6N1m7H9bbTAgDgAAlRAAAAOCA9vcWsb23vD0Xad+skeep4286uZ71PeVkhsxprMmdyTeZMrh14zJ1ck1mNtZnRWJ0pk2ynBQDA6BGiAAAA8Ixa2rv2CEVWb92ZNf3bba1taU9P77N3kTTWVPSHIrW7hSQ1A8czGqpTUV42Cu8IAAD2jRAFAADgENbV05t1Le0DW2w9fSbJ6q0792lIe0VZIbMn12RO09AhyezJtamv9ldQAADGF3+CBQAAmKCKxWJadnY/FYxs23MmybqW9uxDE0mmTKrcbXut2kFbbs2dXJtp9dW22QIAYMIRogAAAIxTnd19XSRPDjGDZNdjR2fPs96nqrxsoItk1wySOU/rJplU5a+PAAAcevwpGAAAYAwqFovZ0tY15BZbT27dmTXbdmZ9a0eK+9BFclhd1UAYMtRMkml11SnTRQIAAHsQogAAAJTI+tb2PLpu++AttrY9FZq0d/U+6z2qK8p2C0VqMnvQTJK+0KSmsnwU3g0AAEw8QhQAAIBR0NNbzCPrWrN0xZaBxxOb2571uukN1U9tsdVUu8dckql1VSkUdJEAAMDBIEQBAAA4CLZ3dOfuJ7ZmyYrNWbpiS+5+YmtaO7oHrSkUkiMOq8vcKXtusTV3cm1mNdWkukIXCQAAlIoQBQAA4AAVi8Ws2rIzdz6xJUuW93WZ/G5tS3qfNq+krqo8p82fkjOa+x6nzp+cxprK0hQNAAA8KyEKAADAfurs7s2Da1qyZPnm3PlEX2iyrqVjj3WHT6nNGc1Tsqh5Sk5vnpLjZjWm3AB3AAAYN4QoAAAAz2LLjs6+OSb9gck9K7emo3vw0PeKskJOmNuURc1PdZrMbKwpUcUAAMBIEKIAAADsplgs5rENO7K0f5bJkhVb8vsNO/ZYN3lSZc6YPyVnLJiSM+ZPySnzJqem0vwSAACYSIQoAADAIW1nZ0/uWbW1r9NkxZbc+cSWbG3r2mPdUdPrsqh5al+XyYIpOXJaXQoFW3MBAMBEVtIQpbW1NR/5yEfyve99L+vXr89pp52Wf/zHf8yZZ56ZJHnLW96Sr3zlK4OuOeuss/Kb3/ymFOUCAAATwLqW9oHh70tXbM4Dq1vS/bQJ8NUVZTll3uSBrblOnz8lU+qqSlQxAABQKiUNUf7iL/4i999/f66//vrMmTMnN9xwQ/7oj/4oDz74YObOnZskufDCC3PdddcNXFNV5S8uAADAvunu6c3v1rbmzie2DAQnT27duce6mY3VWdQ8Naf3D4F/zuzGVFWUlaBiAABgLClZiLJz58585zvfyQ9+8IP84R/+YZLkiiuuyPe///38y7/8Sz75yU8mSaqrqzNr1qxSlQkAAIwjLe1dueuJrQNdJnc/sTU7OnsGrSkrJMfNasyiBU8NgJ87udbWXAAAwB5KFqJ0d3enp6cnNTU1g87X1tbm9ttvHzi+7bbbMmPGjEyePDnnnHNOPvWpT2XGjBl7vW9HR0c6OjoGjltaWka+eAAAoOSKxWKe2Nw2MPz9zhVb8vC61hQH78yVhuqKnNbcN/x90YK+AfD11cZDAgAAz65QLD79rxij5+yzz05VVVW+/vWvZ+bMmbnxxhvz5je/OQsXLszDDz+cm266KfX19Wlubs7jjz+ej3zkI+nu7s7SpUtTXV095D2vuOKK/N3f/d0e57dt25bGxsaD/ZYAAICDpKO7J/c/2ZI7V2zJkhWbs3TF1mzc3rHHuvlTJ2VR85S+rbkWTMnCGQ0pL9NlAgAA9GlpaUlTU9M+5QYlDVEee+yxXHrppfmv//qvlJeX5/TTT88xxxyTO++8Mw8++OAe69esWZPm5uZ84xvfyKte9aoh7zlUJ8q8efOEKAAAMM5s3N6RO1fsGgC/Jfc+uS2d3b2D1lSWF3LS3Kb+bbmm5vTmyZnRULOXOwIAAOxfiFLSHvajjjoqv/jFL7Jjx460tLRk9uzZef3rX58jjjhiyPWzZ89Oc3Nzli1bttd7VldX77VLBQAAGJt6e4t5dMP2geHvdz6xJY9v3LHHusPqqnJ6/xyTRc1TcuLcptRUlpegYgAA4FAwJjYCrqurS11dXbZs2ZJbbrklV1999ZDrNm3alJUrV2b27NmjXCEAADCS2jq7c/fKrVm6fEuWPtE3z6SlvXuPdcfMrB/oMjmjeUoWHDbJAHgAAGDUlDREueWWW1IsFnPsscfm0UcfzXvf+94ce+yxeetb35rt27fniiuuyKtf/erMnj07y5cvz4c+9KFMmzYtr3zlK0tZNgAAsJ9Wb905MPx96YoteXBNS3p6B+8sXFtZnlPnTe4LTRZMyenzpqRpUmWJKgYAAChxiLJt27Z88IMfzKpVqzJ16tS8+tWvzqc+9alUVlamu7s79913X7761a9m69atmT17ds4777zcdNNNaWhoKGXZAAAcQnp6i1mxaUceXtua361tzbL1rdnZ2ZOyQiGFQlIoFFJIBo7LCoWk/2vf+f41haSQQv9xnvn6QcfPcH2eem3g+5UNcX2e+l59x33rhrz+6fWXDXX9U3Xs7f0Xkjy+ccfAPJM129r3+NnOaarpG/7e32ly3OyGVJaXHfx/qAAAAPuopIPlR8P+DIgBAODQVSwWs6G1I79b2zoQmDyyri80ae/qffYb8IzKywo5fnZj/9ZcfY85k2tLXRYAAHAIGjeD5QEAoBR2dHTn4XV9YUlfYNKSh9e2Zktb15DrayrLcszMhhw7syHHzmpIY21lUkx6i8UU0/+12BfEFNM3JL3vfP+5YlJMsf+4b30GrXva9f3Hvf3X7Xptj+sH1vVdV9z9eG/X91+X4rNcP1T9vU+9j2d8//3H0xuqc8b8vq25Tjl8cuqq/fUDAAAYX/wtBgCACau7pzePb9wxqLvk4XUtWbl555DrywrJgml1OW5WQ46d2ZhjZzXkuFkNmTd1UsrLDDMHAAA41AhRAAAY94rFYtZsa39ad0lrHlu/PZ09Q2/FNaOheiAkOXZWY46b1ZCjZ9SnprJ8lKsHAABgrBKiAAAwrrS0dz3VVdK/DdfDa1vT0t495Pq6qvIcsyssmflUYDKlrmqUKwcAAGC8EaIAADAmdXb35rEN2/cITFZvax9yfXlZIUdOq9uju2Tu5NqU2YoLAACAYRCiAABQUsViMau27OzrKFn3VGDy+w070t1bHPKaOU01OXZWw24dJo05akZdqitsxQUAAMDIEaIAADBqtuzoHJhbsisseWTd9mzvGHorroaaiv6ukqc6S46Z2ZCm2spRrhwAAIBDkRAFAIAR197Vk0fXbx8ISn7XP7dkfWvHkOsryws5anr9oG24jp3VkNlNNSkUbMUFAABAaQhRAAAYtt7eYp7Y3DYQkjy8ri8wWb5xR/ayE1cOn1K7R3fJEdPqUlleNrrFAwAAwLMQogAAsE82bu/YY8j7I+u2Z2dXz5DrJ0+qzLEznxryfmx/cFJf7Y+gAAAAjA/+BgsAwCBtnd15ZN32/qBkex5e1xeYbNzeOeT66oqyLJxZn2NnNubYWfUD3SUzGqptxQUAAMC4JkQBADjEtbR35ScPrMtPH1qXB9e05InNbSkOsRVXoZA0T500aBuuY2c1ZMFhdSkvE5YAAAAw8QhRAAAOQTs6uvPTh9bl5nvX5BcPb0hnT++g16fVV/WFJTOfCksWzqzPpCp/fAQAAODQ4W/BAACHiJ2dPfn5w+tz872r87PfrU9711PByVHT63LRyXNy1hFTc+yshkyrry5hpQAAADA2CFEAACaw9q6e/OKRDbn53jW59aF1aet8agj8gsMm5eKT5+TiU2bn2JkN5pcAAADA0whRAAAmmM7u3tz+6IbcfM+a/OTBdWnt6B54be7k2lx8yuy8/OQ5OWFOo+AEAAAAnoEQBQBgAuju6c2vH9uUm+9dnVseWJdtO7sGXpvVWJOLTp6di0+enVPnTRacAAAAwD4SogAAjFM9vcX89vFNufneNfmP+9dm847OgdemN1TnZSfOysWnzMkZ86ekrExwAgAAAPtLiAIAMI709hazZMWW3Hzv6vz7fWuzcXvHwGtT66ry0hNn5eKT5+QPjpiacsEJAAAAHBAhCgDAGFcsFnPXyq25+Z41+ff71mRtS/vAa021lbnwhFm5+JTZed6Rh6WivKyElQIAAMDEIkQBABiDisVi7n+yJTffuzo337smT27dOfBaQ3VFzj9hZl5+8pw8/+hpqaoQnAAAAMDBIEQBABgjisVifre2dSA4WbGpbeC1SVXl+aPnzMzFJ8/OHx4zPTWV5SWsFAAAAA4NQhQAgBJbtq41P7p3Tf7t3tV5bMOOgfM1lWV58XF9wcm5x85IbZXgBAAAAEaTEAUAoAQe37gjN9/T13Hy8LrWgfNVFWU595jpufiUOXnxcTNSV+2PawAAAFAq/lYOADBKVm5uy833rsnN967OA6tbBs5XlhfywoXTc/HJs3P+8TPTUFNZwioBAACAXYQoAAAH0eqtO/Pv963Jj+5dk3tWbh04X15WyNlHHZaXnzwnF5wwK02TBCcAAAAw1ghRAABG2PqW9vzbfWty871rsnTFloHzZYXkuUcelotPnpMLTpiZw+qrS1glAAAA8GyEKAAAI2Dj9o78+P61ufme1fnv5ZtTLPadLxSSM5un5uJTZufCE2dlRkNNaQsFAAAA9pkQBQBgmLa2deY/7l+bm+9dk18/tjG9xadeO23+5Fx88pxcdNLszGoSnAAAAMB4JEQBANgPLe1d+c8H1uXme1fn9mUb071bcnLS3KZcfPLsXHTy7Bw+ZVIJqwQAAABGghAFAOBZbO/ozk8f7AtO/uuRjens6R147TmzG/uCk5NmZ8G0uhJWCQAAAIw0IQoAwBDaOrvzs9+tz833rMnPH16fju6ngpOFM+r7tuo6eXaOnlFfwioBAACAg0mIAgDQr72rJ7c9vCE337s6tz60Pju7egZeO2JaXS4+eXYuPnlOjp3VUMIqAQAAgNEiRAEADmmd3b355bINufneNfnJg+uyvaN74LV5U2sHhsOfMKcxhUKhhJUCAAAAo02IAgAccrp6evOrRzfm3+5dk1seWJuW9qeCkzlNNbmov+Pk5MObBCcAAABwCBOiAACHhJ7eYn7z+025+d7V+Y/712ZLW9fAazMaqvOyk2bn5afMzmnzpqSsTHACAAAACFEAgAmqvasnj67fnmXrW7N0xZb8x/1rs3F758Drh9VV5aUnzcrFJ8/JmQumplxwAgAAADyNEAUAGNc6u3vz+MYdeXhda5ata80j61rzyLrtWbFpR3qLg9dOnlSZl544KxedNCfPPXJqKsrLSlM0AAAAMC4IUQCAcaG7pzcrNrflkbV9Ickj/YHJ4xt3pPvpaUm/yZMqc8zMhhw3qyEvOm5Gnn/0tFQKTgAAAIB9JEQBAMaU3t5iVm5pGxSUPLy2Nb/fsCOdPb1DXtNQXZGFM+tz7KyGLJzR0Pd1Zn2m11cbDA8AAAAMmxAFACiJYrGY1dva+4KS3bpLHl2/PTu7eoa8prayPMfMrM/CmQ05ZmZ9jpnZkGNmNmR2U42wBAAAABhxQhQA4KAqFovZ0NqRh/tnlTyytjWPrG/NsnXbs72je8hrqirKcvT0+r6gZFZDjunvLpk7uTZlBsADAAAAo0SIAgCMmE3bO/LIuu1Ztr5vC65l67bn4XWt2baza8j1FWWFHDm9LgtnNuTY3bpL5k+dZOg7AAAAUHJCFABgv23b2ZVl61rz8Lr+oGRta5atb83G7Z1Dri8rJAsOq+vffqu/u2RmQxYcVpeqCmEJAAAAMDYJUQCAvdre0Z1l/UHJI7uFJmtb2vd6zfypkwbNK1k4sz5HTa9PTWX5KFYOAAAAcOCEKABA2rt68uj6wUHJw2tb8+TWnXu9Zk5TTd82XLMasnBGfY6d1ZCjZ9RnUpU/XgAAAAATg085AOAQ0tndm99v3D4wr+SRda15ZF1rVmxuS7E49DXTG6pzbH9HSd/XvueNNZWjWzwAAADAKBOiAMAE1N3Tm+Wb2gZCkr7H9jy+cUd6eodOS6ZMqhzYguuYWQ05ZkbfllxT6qpGuXoAAACAsUGIAgDjWE9vMSs394Uly9b3dZg8sq41v9+wI509vUNe01BT8VRYslt3ybT6qhQKhVF+BwAAAABjlxAFAMaB7p7ePLG5LY+u355l67f3f23No+u3p71r6LBkUlV5Fs6oH9xdMrM+sxprhCUAAAAA+0CIAgBjSGd3b1Zs2pFl67dn2bqngpJn6iypqigbHJbM7Hs+d3JtysqEJQAAAADDJUQBgBJo7+rJ4xv7wpJH+7fiWrZ+e5Zv3JHuvcwsqaksy9Ez6rNwRkP/1/ocPaM+zYfVpVxYAgAAADDihCgAcBC1dXbn9xt2ZNn61v7Okr6tuFZs2pG9ZCWpqyrP0TMbsrA/KFk4sy840VkCAAAAMLqEKAAwAlrbuwbPK+nvLlm1Zeder2nsH/C+cGZ9jp7RMNBZMrvJzBIAAACAsUCIAgD7YVtbV19XydNmlqzZ1r7Xaw6rq+rbfqu/o2ThjPocPbM+0+urhSUAAAAAY5gQBQCGsGl7x8Cckt1nlmxo7djrNTMaqgeCkt1nlhxWXz2KlQMAAAAwUoQoAByyisVi1rd2DOoo2bUd1+YdnXu9bk5TzR4zS46e3pCmSZWjWD0AAAAAB5sQBYAJr1gsZvW29ixb1x+U9Icmy9ZvT2t795DXFArJvCmTBrbe2rUN11Ez6lNf7T+fAAAAAIcCnwIBMGH09hazasvOQTNLHu3vMNnR2TPkNWWFZMFhdYNmlhw9oz5HTa9PbVX5KL8DAAAAAMYSIQoA4053T29WbG7LsnXb89iG7VnWP7PksQ3b097VO+Q1FWWFHDGtrm/rrf6ukoUz63PEtLpUVwhLAAAAANiTEAWAMauzuzfLN+0YtP3Wo+u25/GNO9LZM3RYUlVRlqOm1w8Mdt8VljQfVpfK8rJRfgcAAAAAjGdCFADGhO0d3blv1bbc9+TW3LtqWx5a05Llm9rS01sccn1tZflAULL7zJJ5UyelvKwwytUDAAAAMBEJUQAYde1dPXlgdUvuXbU1963alntWbc3vN+5IcYi8pL664qmukt1mlsydXJsyYQkAAAAAB5EQBYCDqrO7Nw+vbc09/YHJvU9uyyPrWofsMJnTVJOTD5+ckw5vyolzm3LMzPrMaqxJoSAsAQAAAGD0CVEAGDHdPb15dMP23Ltq20CXyUNrWoecXzKtvjqnHN6Ukw5vyimHT86Jc5syvaG6BFUDAAAAwNCEKAAMS29vMY9v2tHXXdIfmjywuiU7u3r2WNtUW5mTD2/KyYc35aS5k3PKvCYdJgAAAACMeUIUAJ5VsVjMqi07+8KSJ7fm3pXbcv+T29La0b3H2rqq8pw4tymnzJuck+b2dZnMm1orMAEAAABg3BGiALCHdS3tuWfl1tz3ZF+XyX1PbsvmHZ17rKuuKMsJcxpz8uGTBzpNjpxWb+A7AAAAABOCEAXgELd5R2fuXbW1f0uuvm251rd27LGusryQ42Y19s8w6duW65iZ9akoLytB1QAAAABw8AlRAA4hLe1duX/Vttz75LaB4GTVlp17rCsrJMfMbMhJc5v6O0wm59hZDampLC9B1QAAAABQGkIUgAmqrbM7D6xuGeguuW/Vtvx+444h1x45ra5v6Pvhk3PK4U05fk5jJlX5TwQAAAAAhzafkAFMAB3dPfndmtZB23ItW9+a3uKeaw+fUptTDp+ckw5vyslzm3Li4U1prKkc/aIBAAAAYIwTogCMM109vVm2bntfYPLktty3alt+t7YlXT17JiYzG6tz0ty+7pKT+rflmlpXVYKqAQAAAGD8EaIAjGG9vcX8fuP2QUPfH1jdko7u3j3WTplUmZMPnzwww+Tkw5sys7GmBFUDAAAAwMQgRAEYI4rFYp7Y3DYQlty7alvuf3JbdnT27LG2oboiJ/V3l5xy+OScNLcph0+pTaFQKEHlAAAAADAxCVEASqBYLGZtS3vuWbkt9z351ByTbTu79lhbW1meE+c29m3LNa8pJ81tyoLD6lJWJjABAAAAgINJiAIwSrbs6Mx37lyVOx7blHtWbcvG7R17rKkqL8tzZjfk5P7B76ccPjlHTa9LRXlZCSoGAAAAgEObEAXgILv/yW35yq+X54f3rB40y6S8rJBjZjY8NfR97uQcO6shVRUCEwAAAAAYC4QoAAdBZ3dvfnz/mnzl18tz5xNbB86fOLcxrzzt8Jw6b3JOmNOYmsry0hUJAAAAADwjIQrACFq7rT1f/+2KfP2/Vw5s11VZXshFJ83Om89ekNPmTTb8HQAAAADGCSEKwAEqFov57eOb89U7lueWB9alp7eYJJnVWJM/O2t+/vQP5md6Q3WJqwQAAAAA9pcQBWCYdnR05/t3P5mv/npFHl7XOnD+rCOm5pKzF+T842em0kB4AAAAABi3hCgA++nxjTty/R0r8q2lK9Pa3p0kqa0szytPn5s3P685x81qLHGFAAAAAMBIEKIA7IOe3mJue3h9vnLHivzXIxsGzh8xrS5//tzmvOaMw9NUW1nCCgEAAACAkSZEAXgGW9s6880lK3P9b1Zk5eadSZJCIXnRsTPy5rMX5IVHT0tZmUHxAAAAADARCVEAhnD/k9ty/R0r8v27n0xHd2+SpKm2Mq8/c17+/KzmzD9sUokrBAAAAAAONiEKQL/O7t78+P41uf6OFVmyYsvA+eNnN+aSs5vzx6fMTW1VeQkrBAAAAABGkxAFOOSta2nP1377RL7+2yeycXtHkqSirJCXnjQ7lzyvOWc0T0mhYMsuAAAAADjUCFGAQ1KxWMzi5VvylTuW55b716a7t5gkmdFQnT87qzlv+IN5mdFYU+IqAQAAAIBSEqIAh5S2zu784O7V+cqvl+d3a1sHzv/Bgql589nNueCEWaksLythhQAAAADAWCFEAQ4JyzfuyA2/WZFvLlmZlvbuJElNZVleedrcvOm5C3L8nMYSVwgAAAAAjDVCFGDC6u0t5hePbMhX7lieXzyyIcW+Hbsyf+qkvPl5zXntGfPSNKmytEUCAAAAAGOWEAWYcLa1deVbS1fm+t+syIpNbQPnzzt2et589oKcs3B6ysoMigcAAAAAnpkQBZgwHlzdkut/szzfu+vJtHf1JkkaayryukXz8ufPbc6CaXUlrhAAAAAAGE+EKMC41tXTm/+4f22+esfyLF6+ZeD8cbMacsnZC/KKU+dkUpXf6gAAAACA/eeTRWBcWt/Snq//9xP5+m+fyPrWjiRJRVkhF5w4K5c8b0HOXDAlhYItuwAAAACA4ROiAONGsVjM0hVb8pU7VuTH961Jd2/fpPjpDdV54x/MzxvPmp+ZjTUlrhIAAAAAmCjKSvnNW1tbc/nll6e5uTm1tbU5++yzs3jx4oHXi8VirrjiisyZMye1tbU599xz88ADD5SwYqAUdnb25Bv//UQu+qfb85pr78iP7lmd7t5iFjVPyT+94bT86v0vyv86/xgBCgAAAAAwokraifIXf/EXuf/++3P99ddnzpw5ueGGG/JHf/RHefDBBzN37txcffXV+exnP5svf/nLOeaYY/LJT34y559/fh5++OE0NDSUsnRgFDyxqS3X/2Z5vrlkVbbt7EqSVFeU5U9OnZs3Pa85J85tKnGFAAAAAMBEVigWi8VSfOOdO3emoaEhP/jBD3LRRRcNnD/11FNz8cUX5xOf+ETmzJmTyy+/PO9///uTJB0dHZk5c2Y+85nP5G1ve9s+fZ+WlpY0NTVl27ZtaWxsPCjvBRg5vb3F/NeyDfnqHSvy84fXZ9fvUPOm1uZNz23O6xbNy+RJVaUtEgAAAAAYt/YnNyhZJ0p3d3d6enpSUzN4+53a2trcfvvtefzxx7N27dq85CUvGXituro655xzTn7961/vNUTp6OhIR0fHwHFLS8vBeQPAiNq2syvfXroq19+xPMs3tQ2cP+eY6bnk7Oacc8yMlJcZFA8AAAAAjJ6ShSgNDQ153vOel0984hN5znOek5kzZ+bGG2/Mb3/72yxcuDBr165NksycOXPQdTNnzsyKFSv2et+rrroqf/d3f3dQawdGzkNrWvLVO1bk+3c9mZ1dPUmShpqKvPaMefnz587PkdPrS1whAAAAAHCoKulMlOuvvz6XXnpp5s6dm/Ly8px++ul54xvfmDvvvHNgTaEw+P88LxaLe5zb3Qc/+MG8613vGjhuaWnJvHnzRr54YNi6enrznw+sy1fuWJ7/fnzzwPljZzbkzWc3509OnZu66pL+9gQAAAAAUNoQ5aijjsovfvGL7NixIy0tLZk9e3Ze//rX54gjjsisWbOSJGvXrs3s2bMHrlm/fv0e3Sm7q66uTnV19UGvHdh/61vb843/Xpmv/XZF1rX0bbtXXlbIBSfMzJuftyBnHTH1GUNSAAAAAIDRNCb+V++6urrU1dVly5YtueWWW3L11VcPBCk/+clPctpppyVJOjs784tf/CKf+cxnSlwxsK+KxWLufGJrvnrH8vz7fWvS1dM3KX5afVXe8Afz88az5md2U22JqwQAAAAA2FNJQ5RbbrklxWIxxx57bB599NG8973vzbHHHpu3vvWtKRQKufzyy3PllVdm4cKFWbhwYa688spMmjQpb3zjG0tZNrAP2rt68sO7V+crdyzPA6tbBs6fPn9yLjl7QS48cVaqK8pLWCEAAAAAwDMraYiybdu2fPCDH8yqVasyderUvPrVr86nPvWpVFZWJkne9773ZefOnbnsssuyZcuWnHXWWfnP//zPNDQ0lLJs4Bms3NyWG36zIjctWZmtbV1JkqqKsrzilDl58/MW5KTDm0pcIQAAAADAvikUi8ViqYs4mFpaWtLU1JRt27alsbGx1OXAhPWrRzfmul89nlt/tz67fleZO7k2b3pec16/aF6m1FWVtkAAAAAAgOxfbjAmZqIA49eWHZ352A8fyA/vWT1w7oULp+XNz1uQFx03I+VlBsUDAAAAAOOTEAUYtp88uC4f+t592dDakbJC8mdnNectz1+Qo6bXl7o0AAAAAIADJkQB9tu2tq783c0P5Lt3PpkkOXpGfa557Sk5Zd7k0hYGAAAAADCChCjAfvn5w+vzge/cm3UtHSkUkv/fC4/M/zr/mNRUlpe6NAAAAACAESVEAfZJS3tXPnXzQ7lpycokyRHT6vK/X3tyzmieWuLKAAAAAAAODiEK8KxuX7Yx7/v2PVm9rT2FQvLWs4/Iey84NrVVuk8AAAAAgIlLiALs1Y6O7lz57w/la799Ikkyf+qk/P1rTs5ZRx5W4soAAAAAAA4+IQowpDse25T3fvuerNqyM0lyyfOa8/6XHpdJVX7bAAAAAAAODT4NBQZp6+zO1f/xcL786+VJkrmTa/P3rzk5Zx89rbSFAQAAAACMMiEKMGDx8s1577fuyfJNbUmSN/zB/Hz4ouekvtpvFQAAAADAoccno0Dau3ryv295OF/81eMpFpPZTTX5zKtPzh8eM73UpQEAAAAAlIwQBQ5xdz6xJe/51j35/YYdSZLXLTo8f3vx8WmsqSxxZQAAAAAApSVEgUNUe1dPPvfTZfnX/3osvcVkRkN1Pv3qk/Ki42aWujQAAAAAgDFBiAKHoHtXbc27v3lPlq3fniR55Wlzc8XLT0jTJN0nAAAAAAC7CFHgENLZ3ZvP/2xZ/s9tj6Wnt5hp9VX51CtPygUnzCp1aQAAAAAAY44QBQ4RD6zelnd/8578bm1rkuTlp8zJ3/3xCZlaV1XiygAAAAAAxiYhCkxwXT29+T8/fyyf/9mydPcWM7WuKp94xYm56OTZpS4NAAAAAGBME6LABPbw2ta8+1t35/4nW5IkF54wK5985YmZVl9d4soAAAAAAMY+IQpMQN09vfm///X7fO6nj6Srp5im2sp8/BUn5I9PmZNCoVDq8gAAAAAAxgUhCkwwj65vzbu/dW/uWbk1SfJHz5mRK195UmY01pS2MAAAAACAcUaIAhNET28xX7z99/nf//lIOrt701BTkStefkJedfpc3ScAAAAAAMMgRIEJ4Pcbtuc937ondz6xNUly7rHT8+lXnZxZTbpPAAAAAACGS4gC41hvbzFf/vXyXH3L79Le1Zv66op85OLn5HWL5uk+AQAAAAA4QEIUGKdWbNqR93773vz345uTJC84elo+85qTM3dybYkrAwAAAACYGIQoMM709hbztd+uyJX//rvs7OrJpKryfOhlz8mfnTVf9wkAAAAAwAgSosA4snJzW97/nXvz68c2JUmee+TU/P1rTsm8qZNKXBkAAAAAwMQjRIFxoFgs5huLV+aTNz+YHZ09qaksywcuPC5vft6ClJXpPgEAAAAAOBiEKDDGrd66Mx/47n35r0c2JEkWNU/J/37tKVkwra7ElQEAAAAATGxCFBijisVivr10VT7+owfT2tGd6oqyvPeCY/PW5x+Rct0nAAAAAAAHnRAFxqB1Le350Hfvy62/W58kOXXe5Pzv156So2fUl7gyAAAAAIBDhxAFxpBisZgf3L06H/vhA9m2sytV5WV510uOyV+84IhUlJeVujwAAAAAgEOKEAXGiA2tHfnw9+7Lfz64Lkly0tymXPO6U3LMzIYSVwYAAAAAcGgSosAYcPO9q/OR79+fLW1dqSwv5B0vWpj/ee5RqdR9AgAAAABQMkIUKKFN2zvy0R88kH+7b02S5DmzG3PNa0/J8XMaS1wZAAAAAABCFCiR/7h/TT78vfuzaUdnyssKeft5R+evzzs6VRW6TwAAAAAAxgIhCoyyrW2d+dgPH8gP7l6dJDl2ZkOued0pOXFuU4krAwAAAABgd0IUGEU/fXBdPvi9+7KhtSNlheR/nnNU3vlHC1NdUV7q0gAAAAAAeBohCoyCbTu78vEfPZjv3LkqSXLU9Lpc87pTc+q8yaUtDAAAAACAvRKiwEF228Pr84Hv3Je1Le0pFJK/fOGRedf5x6SmUvcJAAAAAMBYJkSBg6S1vSuf+reH8o3FK5MkR0yry/9+7ck5o3lqiSsDAAAAAGBfCFHgILh92ca879v3ZPW2vu6Tt559RN57wbGprdJ9AgAAAAAwXghRYATt6OjOVT9+KDf85okkyfypk/L3rzk5Zx15WIkrAwAAAABgfwlRYITc8dimvO8792Tl5p1Jkjc9tzkfeOlxqav2rxkAAAAAwHjk0104QG2d3bn6Px7Ol3+9PEkyd3Jtrn7NyXn+0dNKWxgAAAAAAAdEiAIHYMnyzXnPt+7J8k1tSZI3/MH8fOhlx6WhprLElQEAAAAAcKCEKDAM7V09ueY/H87/u/3xFIvJrMaafOY1J+ecY6aXujQAAAAAAEaIEAX2011PbMm7v3VPfr9hR5LkNWccno9cfHyaanWfAAAAAABMJEIU2Ecd3T353E+X5f/+4rH0FpMZDdW56lUn5cXPmVnq0gAAAAAAOAiEKLAP7lu1Le/+1t15ZN32JMkrT5ubj738+EyeVFXiygAAAAAAOFiEKPAMenuL+dxPH8k/3/ZYenqLmVZflU/+yUm58MRZpS4NAAAAAICDTIgCz+BH967OP/3s0STJxSfPzsdfcWKm1uk+AQAAAAA4FAhR4Bn81yMbkyRvOXtBrvjjE0pcDQAAAAAAo6ms1AXAWLZkxeYkyTnHTi9xJQAAAAAAjDYhCuzF+tb2rNjUlkIhOX3+lFKXAwAAAADAKBOiwF4sXb4lSXLszIY01VaWuBoAAAAAAEabEAX2YsmKvhBl0QJdKAAAAAAAhyIhCuzFkuV981AWNU8tcSUAAAAAAJSCEAWG0NbZnQdWtyTRiQIAAAAAcKgSosAQ7l65Nd29xcxqrMncybWlLgcAAAAAgBIQosAQdg2VX7RgSgqFQomrAQAAAACgFIQoMITFu4bKN9vKCwAAAADgUDWsEOW2224b4TJg7OjpLeauXSHKAkPlAQAAAAAOVcMKUS688MIcddRR+eQnP5mVK1eOdE1QUg+vbU1rR3fqqspz3KyGUpcDAAAAAECJDCtEWb16dd75znfmu9/9bo444ohccMEF+eY3v5nOzs6Rrg9G3dIVm5MkpzdPSUW5He8AAAAAAA5Vw/qEeOrUqXnHO96RO++8M0uWLMmxxx6bt7/97Zk9e3be8Y535J577hnpOmHULO4fKn+GeSgAAAAAAIe0A/7f7E899dR84AMfyNvf/vbs2LEjX/rSl3LGGWfkhS98YR544IGRqBFG1dL+eShnmocCAAAAAHBIG3aI0tXVlW9/+9t52ctelubm5txyyy35whe+kHXr1uXxxx/PvHnz8trXvnYka4WDbvXWnXly686UlxVy6rzJpS4HAAAAAIASqhjORX/zN3+TG2+8MUny53/+57n66qtz4oknDrxeV1eXT3/601mwYMGIFAmjZUl/F8rxsxtTVz2sfz0AAAAAAJgghvUp8YMPPpjPf/7zefWrX52qqqoh18yZMyc///nPD6g4GG1LlvcNlTcPBQAAAACAYYUot95667PfuKIi55xzznBuDyWzZLl5KAAAAAAA9BnWTJSrrroqX/rSl/Y4/6UvfSmf+cxnDrgoKIXW9q78bm1LkmTRAp0oAAAAAACHumGFKP/3//7fHHfccXucP+GEE3LttdcecFFQCnc9sTW9xWTe1NrMbKwpdTkAAAAAAJTYsEKUtWvXZvbs2Xucnz59etasWXPARUEp7JqHsqjZVl4AAAAAAAwzRJk3b15+9atf7XH+V7/6VebMmXPARUEpLFnRNw/FVl4AAAAAACTDHCz/F3/xF7n88svT1dWVF73oRUn6hs2/733vy7vf/e4RLRBGQ1dPb+56YmsSnSgAAAAAAPQZVojyvve9L5s3b85ll12Wzs7OJElNTU3e//7354Mf/OCIFgij4aE1LdnZ1ZPGmoosnFFf6nIAAAAAABgDhhWiFAqFfOYzn8lHPvKRPPTQQ6mtrc3ChQtTXV090vXBqFi8vG8rrzOap6SsrFDiagAAAAAAGAuGFaLsUl9fnzPPPHOkaoGSWbqif6j8Alt5AQAAAADQZ9ghyuLFi/Otb30rTzzxxMCWXrt897vfPeDCYLQUi8WBTpRFzYbKAwAAAADQp2w4F33jG9/I85///Dz44IP53ve+l66urjz44IP52c9+lqamppGuEQ6qlZt3ZkNrRyrLCzll3uRSlwMAAAAAwBgxrBDlyiuvzD/8wz/k5ptvTlVVVf7xH/8xDz30UF73utdl/vz5I10jHFSLl/dt5XXi3KbUVJaXuBoAAAAAAMaKYYUojz32WC666KIkSXV1dXbs2JFCoZD/9b/+V/71X/91RAuEg23Jir6tvM40DwUAAAAAgN0MK0SZOnVqWltbkyRz587N/fffnyTZunVr2traRq46GAVL+jtRzjAPBQAAAACA3QxrsPwLX/jC/OQnP8lJJ52U173udXnnO9+Zn/3sZ/nJT36SF7/4xSNdIxw0W9s6s2z99iSGygMAAAAAMNiwQpQvfOELaW9vT5J88IMfTGVlZW6//fa86lWvykc+8pERLRAOpqX9W3kdOa0uh9VXl7gaAAAAAADGkv0OUbq7u/OjH/0oF1xwQZKkrKws73vf+/K+971vxIuDg23XPJRFC3ShAAAAAAAw2H7PRKmoqMhf/dVfpaOj42DUA6Nq1zyURc2GygMAAAAAMNiwBsufddZZueuuu0a6FhhVHd09uWfVtiQ6UQAAAAAA2NOwZqJcdtllefe7351Vq1bljDPOSF1d3aDXTz755BEpDg6m+5/cls7u3hxWV5UjptU9+wUAAAAAABxShhWivP71r0+SvOMd7xg4VygUUiwWUygU0tPTMzLVwUG0ZHnfPJQzmqekUCiUuBoAAAAAAMaaYYUojz/++EjXAaNu8XJD5QEAAAAA2LthhSjNzc0jXQeMqmKxmKUr+ofKLzBUHgAAAACAPQ0rRPnqV7/6jK+/+c1vHlYxMFoe27AjW9q6Ul1RlhPnNJW6HAAAAAAAxqBhhSjvfOc7Bx13dXWlra0tVVVVmTRpkhCFMW9XF8op8yanqqKsxNUAAAAAADAWDevT4y1btgx6bN++PQ8//HBe8IIX5MYbbxzpGmHEDcxDaTYPBQAAAACAoY3Y/4K/cOHCfPrTn96jSwXGoqUr+kKUM81DAQAAAABgL0Z0H6Py8vKsXr16JG8JI25Da0ce37gjSXL6fJ0oAAAAAAAMbVgzUX74wx8OOi4Wi1mzZk2+8IUv5PnPf/6IFAYHy64ulGNnNqRpUmWJqwEAAAAAYKwaVojyJ3/yJ4OOC4VCpk+fnhe96EW55ppr9vk+3d3dueKKK/K1r30ta9euzezZs/OWt7wlf/u3f5uysr4mmbe85S35yle+Mui6s846K7/5zW+GUzpkyfK+ofJnLNCFAgAAAADA3g0rROnt7R2Rb/6Zz3wm1157bb7yla/khBNOyJIlS/LWt741TU1Ng2arXHjhhbnuuusGjquqqkbk+3NoWjIwD0WIAgAAAADA3g0rRBkpd9xxR17xilfkoosuSpIsWLAgN954Y5YsWTJoXXV1dWbNmlWKEplgdnb25P4ntyVJFjUbKg8AAAAAwN4Na7D8a17zmnz605/e4/zf//3f57Wvfe0+3+cFL3hBbr311jzyyCNJknvuuSe33357Xvaylw1ad9ttt2XGjBk55phj8pd/+ZdZv379Xu/Z0dGRlpaWQQ/Y5Z5VW9PdW8zMxuocPqW21OUAAAAAADCGDStE+cUvfjHQPbK7Cy+8MP/1X/+1z/d5//vfnze84Q057rjjUllZmdNOOy2XX3553vCGNwyseelLX5qvfe1r+dnPfpZrrrkmixcvzote9KJ0dHQMec+rrroqTU1NA4958+bt/xtkwto1D2VR89QUCoUSVwMAAAAAwFg2rO28tm/fPuRcksrKyv3q/Ljppptyww035Otf/3pOOOGE3H333bn88sszZ86cXHLJJUmS17/+9QPrTzzxxCxatCjNzc35t3/7t7zqVa/a454f/OAH8653vWvguKWlRZDCgF3zUBaZhwIAAAAAwLMYVohy4okn5qabbspHP/rRQee/8Y1v5Pjjj9/n+7z3ve/NBz7wgfzpn/5pkuSkk07KihUrctVVVw2EKE83e/bsNDc3Z9myZUO+Xl1dnerq6n2ugUNHb28xS3eFKOahAAAAAADwLIYVonzkIx/Jq1/96jz22GN50YtelCS59dZbc+ONN+Zb3/rWPt+nra0tZWWDdxQrLy9Pb2/vXq/ZtGlTVq5cmdmzZw+ndA5hj6xvTWt7dyZVlec5sxtKXQ4AAAAAAGPcsEKUP/7jP873v//9XHnllfn2t7+d2tranHzyyfnpT3+ac845Z5/v8/KXvzyf+tSnMn/+/Jxwwgm566678tnPfjaXXnppkr5tw6644oq8+tWvzuzZs7N8+fJ86EMfyrRp0/LKV75yOKVzCFu8vK8L5bT5k1NRPqxxQAAAAAAAHEKGFaIkyUUXXTTkcPn98fnPfz4f+chHctlll2X9+vWZM2dO3va2tw1sE1ZeXp777rsvX/3qV7N169bMnj075513Xm666aY0NOgkYP8s3W2oPAAAAAAAPJtCsVgs7u9FixcvTm9vb84666xB53/729+mvLw8ixYtGrECD1RLS0uampqybdu2NDY2lrocSuj5n/5Znty6M9f/jz/ICxdOL3U5AAAAAACUwP7kBsPa0+jtb397Vq5cucf5J598Mm9/+9uHc0s4qNZs25knt+5MWSE5bf6UUpcDAAAAAMA4MKwQ5cEHH8zpp5++x/nTTjstDz744AEXBSNtSf88lOfMbkx99bB3sQMAAAAA4BAyrBCluro669at2+P8mjVrUlHhA2rGnqUr+kKUMxeYhwIAAAAAwL4ZVohy/vnn54Mf/GC2bds2cG7r1q350Ic+lPPPP3/EioORsrh/qPwZzbbyAgAAAABg3wyrbeSaa67JH/7hH6a5uTmnnXZakuTuu+/OzJkzc/31149ogXCgtnd056E1LUmSRQuEKAAAAAAA7JthhShz587Nvffem6997Wu55557Ultbm7e+9a15wxvekMrKypGuEQ7IXU9sSW8xmTu5NrObaktdDgAAAAAA48SwB5jU1dXlBS94QebPn5/Ozs4kyY9//OMkyR//8R+PTHUwAnYNlT9TFwoAAAAAAPthWCHK73//+7zyla/Mfffdl0KhkGKxmEKhMPB6T0/PiBUIB2rJiv55KIbKAwAAAACwH4Y1WP6d73xnjjjiiKxbty6TJk3K/fffn1/84hdZtGhRbrvtthEuEYavu6c3dz2xNYlOFAAAAAAA9s+wOlHuuOOO/OxnP8v06dNTVlaW8vLyvOAFL8hVV12Vd7zjHbnrrrtGuk4YlofWtKatsycNNRU5ZkZDqcsBAAAAAGAcGVYnSk9PT+rr65Mk06ZNy+rVq5Mkzc3Nefjhh0euOjhAA1t5NU9JWVnhWVYDAAAAAMBThtWJcuKJJ+bee+/NkUcembPOOitXX311qqqq8q//+q858sgjR7pGGLZdQ+UXNdvKCwAAAACA/TOsEOVv//Zvs2PHjiTJJz/5yVx88cV54QtfmMMOOyw33XTTiBYIw1UsFgc6URYZKg8AAAAAwH4aVohywQUXDDw/8sgj8+CDD2bz5s2ZMmVKCgVbJjE2rNqyM+taOlJRVsgph08udTkAAAAAAIwzwwpRhjJ1qv/Tn7FlVxfKiXObUltVXuJqAAAAAAAYb4Y1WB7Gg8XmoQAAAAAAcACEKExYS3eFKOahAAAAAAAwDEIUJqRtbV15eF1rkuQMnSgAAAAAAAyDEIUJ6c4n+rpQjphWl+kN1SWuBgAAAACA8UiIwoS0eHnfUHnzUAAAAAAAGC4hChPSkhW75qEIUQAAAAAAGB4hChNOZ3dv7lm5NYmh8gAAAAAADJ8QhQnn/tXb0tHdm6l1VTlyWl2pywEAAAAAYJwSojDhLOmfh3JG85QUCoUSVwMAAAAAwHglRGHCWbK8fx6KofIAAAAAABwAIQoTSrFYzNKBofLmoQAAAAAAMHxCFCaUxzfuyKYdnamqKMuJcxtLXQ4AAAAAAOOYEIUJZddWXqcePjnVFeUlrgYAAAAAgPFMiMKEsmRF/1D5BeahAAAAAABwYIQoTCi7OlHOFKIAAAAAAHCAhChMGJu2d+T3G3ckSU6fL0QBAAAAAODACFGYMJas6OtCOWZmfSZPqipxNQAAAAAAjHdCFCaMpf0hyhnNU0tcCQAAAAAAE4EQhQlj8fK+ofLmoQAAAAAAMBKEKEwI7V09uf/JbUmSRTpRAAAAAAAYAUIUJoR7Vm5NV08xMxqqM29qbanLAQAAAABgAhCiMCHsGiq/aMGUFAqFElcDAAAAAMBEIERhQljSPw/FVl4AAAAAAIwUIQrjXm9vMUt360QBAAAAAICRIERh3Fu2fnta2rszqao8x89uLHU5AAAAAABMEEIUxr0lK/q28jp13uRUlPslDQAAAADAyPCJM+PekuW7tvIyDwUAAAAAgJEjRGHc29WJsqjZPBQAAAAAAEaOEIVxbV1Le1Zu3pmyQnLa/MmlLgcAAAAAgAlEiMK4tmsrr+NmNaahprLE1QAAAAAAMJEIURjXFi/v28rrzAW28gIAAAAAYGQJURjXlq7o60Q5w1B5AAAAAABGmBCFcWt7R3ceWL0tiU4UAAAAAABGnhCFcevuJ7amt5jMnVyb2U21pS4HAAAAAIAJRojCuLVkRd88lEW6UAAAAAAAOAiEKIxbS5b3zUNZ1CxEAQAAAABg5AlRGJe6e3pz1xP9IYqh8gAAAAAAHARCFMal361tzY7OnjRUV+SYmQ2lLgcAAAAAgAlIiMK4tGR53zyU05unpLysUOJqAAAAAACYiIQojEuLV5iHAgAAAADAwSVEYdwpFosDnSjmoQAAAAAAcLAIURh3Vm3ZmXUtHakoK+TUeZNLXQ4AAAAAABOUEIVxZ2n/Vl4nzG1KbVV5iasBAAAAAGCiEqIw7izetZWXeSgAAAAAABxEQhTGnV2dKGcuEKIAAAAAAHDwCFEYV7bt7MrD61qTJGc0GyoPAAAAAMDBI0RhXLnziS0pFpMFh03K9IbqUpcDAAAAAMAEJkRhXFnSPw9FFwoAAAAAAAebEIVxZcly81AAAAAAABgdQhTGjc7u3ty9cmuSZJEQBQAAAACAg0yIwrjxwOpt6ejuzZRJlTlqen2pywEAAAAAYIITojBu7NrK64zmKSkUCiWuBgAAAACAiU6IwrixZEXfUPlFCwyVBwAAAADg4BOiMC4Ui8WBTpRFzeahAAAAAABw8AlRGBeWb2rLph2dqaooy0mHN5W6HAAAAAAADgFCFMaFxcv7tvI6eW5TqivKS1wNAAAAAACHAiEK48LSXVt5mYcCAAAAAMAoEaIwLizeNVTePBQAAAAAAEaJEIUxb9P2jvx+w44kyRlCFAAAAAAARokQhTFv6Yq+rbyOnlGfKXVVJa4GAAAAAIBDhRCFMW9XiHLmAl0oAAAAAACMHiEKY97i5X3zUM5oNlQeAAAAAIDRI0RhTGvv6sl9T25LohMFAAAAAIDRJURhTLt31bZ09RQzrb4686dOKnU5AAAAAAAcQoQojGlLVvRt5XXmgikpFAolrgYAAAAAgEOJEIUxbcnyvqHyZzTbygsAAAAAgNElRGHM6u0tZumKvhDlzAWGygMAAAAAMLqEKIxZj27Ynm07u1JbWZ7j5zSWuhwAAAAAAA4xQhTGrF1beZ06b3Iqy/1SBQAAAABgdPlkmjFryfK+ofKLFpiHAgAAAADA6BOiMGYt6Z+Hssg8FAAAAAAASkCIwpi0vqU9T2xuS6GQnDZ/cqnLAQAAAADgECREYUza1YVy3KzGNNZUlrgaAAAAAAAORUIUxqTFu+ahNJuHAgAAAABAaQhRGJOWDsxDEaIAAAAAAFAaQhTGnB0d3XlgdUsSQ+UBAAAAACgdIQpjzj0rt6ant5g5TTWZO7m21OUAAAAAAHCIEqIw5ixe3reV1xm6UAAAAAAAKCEhCmPOkhV9Q+XPNA8FAAAAAIASEqIwpvT0FnPXE1uTJGc0C1EAAAAAACgdIQpjyu/WtmR7R3fqqyty3KzGUpcDAAAAAMAhTIjCmLKkfx7KafMnp7ysUOJqAAAAAAA4lAlRGFOWrOgLUc40VB4AAAAAgBIraYjS3d2dv/3bv80RRxyR2traHHnkkfn4xz+e3t7egTXFYjFXXHFF5syZk9ra2px77rl54IEHSlg1B9OS5X1D5ReZhwIAAAAAQImVNET5zGc+k2uvvTZf+MIX8tBDD+Xqq6/O3//93+fzn//8wJqrr746n/3sZ/OFL3whixcvzqxZs3L++eentbW1hJVzMDy5dWfWbGtPeVkhp86fXOpyAAAAAAA4xFWU8pvfcccdecUrXpGLLrooSbJgwYLceOONWbJkSZK+LpTPfe5z+fCHP5xXvepVSZKvfOUrmTlzZr7+9a/nbW972x737OjoSEdHx8BxS0vLKLwTRsKuLpQT5jRmUlVJf2kCAAAAAEBpO1Fe8IIX5NZbb80jjzySJLnnnnty++2352Uve1mS5PHHH8/atWvzkpe8ZOCa6urqnHPOOfn1r3895D2vuuqqNDU1DTzmzZt38N8II2LXUPlFzeahAAAAAABQeiX93/3f//73Z9u2bTnuuONSXl6enp6efOpTn8ob3vCGJMnatWuTJDNnzhx03cyZM7NixYoh7/nBD34w73rXuwaOW1paBCnjxOJd81AWmIcCAAAAAEDplTREuemmm3LDDTfk61//ek444YTcfffdufzyyzNnzpxccsklA+sKhcKg64rF4h7ndqmurk51dfVBrZuR19LelYfX9c25MVQeAAAAAICxoKQhynvf+9584AMfyJ/+6Z8mSU466aSsWLEiV111VS655JLMmjUrSV9HyuzZsweuW79+/R7dKYxvd67YkmIxmT91UmY01pS6HAAAAAAAKO1MlLa2tpSVDS6hvLw8vb29SZIjjjgis2bNyk9+8pOB1zs7O/OLX/wiZ5999qjWysG1dEX/PBRbeQEAAAAAMEaUtBPl5S9/eT71qU9l/vz5OeGEE3LXXXfls5/9bC699NIkfdt4XX755bnyyiuzcOHCLFy4MFdeeWUmTZqUN77xjaUsnRE2MA/FUHkAAAAAAMaIkoYon//85/ORj3wkl112WdavX585c+bkbW97Wz760Y8OrHnf+96XnTt35rLLLsuWLVty1lln5T//8z/T0NBQwsoZSV09vbl75dYkyZk6UQAAAAAAGCMKxWKxWOoiDqaWlpY0NTVl27ZtaWxsLHU5DOHulVvzJ//8qzTVVuauj5yfsrJCqUsCAAAAAGCC2p/coKQzUSBJlgxs5TVFgAIAAAAAwJghRKHklizvGyp/hq28AAAAAAAYQ4QolFSxWMySFX0hypkLDJUHAAAAAGDsEKJQUis2tWXj9o5UlZflpLlNpS4HAAAAAAAGCFEoqV1dKCcd3pSayvISVwMAAAAAAE8RolBSuw+VBwAAAACAsUSIQknt6kRZZB4KAAAAAABjjBCFktmyozOPrt+eJDlDJwoAAAAAAGOMEIWSWdrfhXLU9LpMrasqcTUAAAAAADCYEIWSWbxi1zwUW3kBAAAAADD2CFEomaXLd81DsZUXAAAAAABjjxCFkmjv6sm9q7YlMVQeAAAAAICxSYhCSdz/5LZ09vRmWn1VFhw2qdTlAAAAAADAHoQolMTi/q28zmiekkKhUOJqAAAAAABgT0IUSmJp/1D5M23lBQAAAADAGCVEYdT19hazZMVTnSgAAAAAADAWCVEYdb/fuD1b27pSU1mWE+Y0lbocAAAAAAAYkhCFUbdrHsoph09OVYVfggAAAAAAjE0+wWbULekPUcxDAQAAAABgLBOiMOqW9A+VP2OBeSgAAAAAAIxdQhRG1frW9qzY1JZCITl9vhAFAAAAAICxS4jCqFrav5XXsTMb0lRbWeJqAAAAAABg74QojKolK/pClEW28gIAAAAAYIwTojCqlizvm4eyqNlQeQAAAAAAxjYhCqOmrbM7D6xuSaITBQAAAACAsU+Iwqi5e+XWdPcWM6uxJnMn15a6HAAAAAAAeEZCFEbNrqHyixZMSaFQKHE1AAAAAADwzIQojJrFu4bKN9vKCwAAAACAsU+Iwqjo6S3mrl0hygJD5QEAAAAAGPuEKIyKh9e2prWjO3VV5TluVkOpywEAAAAAgGclRGFULF2xOUlyevOUVJT7ZQcAAAAAwNjn02xGxeL+ofJnmIcCAAAAAMA4IURhVCztn4dypnkoAAAAAACME0IUDrrVW3fmya07U15WyKnzJpe6HAAAAAAA2CdCFA66Jf1dKMfPbkxddUWJqwEAAAAAgH0jROGgW7K8b6i8eSgAAAAAAIwnQhQOuiXLzUMBAAAAAGD8EaJwULW2d+V3a1uSJIsW6EQBAAAAAGD8EKJwUN31xNb0FpN5U2szs7Gm1OUAAAAAAMA+E6JwUO2ah7Ko2VZeAAAAAACML0IUDqolK/rmodjKCwAAAACA8UaIwkHT1dObu57YmkQnCgAAAAAA448QhYPmoTUt2dnVk8aaiiycUV/qcgAAAAAAYL8IUThoFi/v28rrjOYpKSsrlLgaAAAAAADYP0IUDpqlK/qHyi+wlRcAAAAAAOOPEIWDolgsDnSiLGo2VB4AAAAAgPFHiMJBsXLzzmxo7UhleSGnzJtc6nIAAAAAAGC/CVE4KBYv79vK68S5TampLC9xNQAAAAAAsP+EKBwUS1b0beV1pnkoAAAAAACMU0IUDool/Z0oZ5iHAgAAAADAOCVEYcRtbevMsvXbkxgqDwAAAADA+CVEYcQt7d/K68hpdTmsvrrE1QAAAAAAwPAIURhxu+ahLFqgCwUAAAAAgPFLiMKI2zUPZVGzofIAAAAAAIxfQhRGVEd3T+5ZtS2JThQAAAAAAMY3IQoj6v4nt6WzuzeH1VXliGl1pS4HAAAAAACGTYjCiFqyvG8eyhnNU1IoFEpcDQAAAAAADJ8QhRG1eLmh8gAAAAAATAxCFEZMsVjM0hX9Q+UXGCoPAAAAAMD4JkRhxDy2YUe2tHWluqIsJ85pKnU5AAAAAABwQIQojJhdXSinzJucqgq/tAAAAAAAGN980s2IGZiH0mweCgAAAAAA458QhRGzdEVfiHKmeSgAAAAAAEwAQhRGxIbWjjy+cUeS5PT5OlEAAAAAABj/hCiMiF1dKMfObEjTpMoSVwMAAAAAAAdOiMKIWLK8b6j8GQt0oQAAAAAAMDEIURgRSwbmoQhRAAAAAACYGIQoHLCdnT25/8ltSZJFzYbKAwAAAAAwMQhROGD3rNqa7t5iZjZW5/AptaUuBwAAAAAARoQQhQO2ax7KouapKRQKJa4GAAAAAABGhhCFA7ZrHsoi81AAAAAAAJhAhCgckN7eYpbuClHMQwEAAAAAYAIRonBAHlnfmtb27kyqKs9zZjeUuhwAAAAAABgxQhQOyOLlfV0op82fnIpyv5wAAAAAAJg4fOrNAVm621B5AAAAAACYSIQoHJBdnSiGygMAAAAAMNEIURi2Ndt25smtO1NWSE6bL0QBAAAAAGBiEaIwbEv6u1CeM7sx9dUVJa4GAAAAAABGlhCFYVu6oi9EOXOBeSgAAAAAAEw8QhSGbXH/UPkzmm3lBQAAAADAxCNEYVi2d3TnoTUtSQyVBwAAAABgYhKiMCx3PbElvcVk7uTazG6qLXU5AAAAAAAw4oQoDMvi5bvmoehCAQAAAABgYhKiMCxLV/TPQzFUHgAAAACACUqIwn7r7unNXU9sTaITBQAAAACAiUuIwn57aE1r2jp70lBTkWNmNJS6HAAAAAAAOCiEKOy3xcv7t/JqnpKyskKJqwEAAAAAgINDiMJ+W7qib6j8omZbeQEAAAAAMHEJUdgvxWJxoBNlkaHyAAAAAABMYEIU9suqLTuzvrUjFWWFnHL45FKXAwAAAAAAB40Qhf2yqwvlxLlNqa0qL3E1AAAAAABw8AhR2C9LzEMBAAAAAOAQIURhvywxDwUAAAAAgEOEEIV9tq2tK4+s254kOUMnCgAAAAAAE5wQhX229Im+LpQjptVlekN1iasBAAAAAICDS4jCPluyvG8eii4UAAAAAAAOBUIU9tmuEOXMBUIUAAAAAAAmvpKGKAsWLEihUNjj8fa3vz1J8pa3vGWP15773OeWsuRDVkd3T+5ZtTVJckazofIAAAAAAEx8FaX85osXL05PT8/A8f3335/zzz8/r33tawfOXXjhhbnuuusGjquqqka1Rvrc/2RLOrp7M2VSZY6aXlfqcgAAAAAA4KAraYgyffr0Qcef/vSnc9RRR+Wcc84ZOFddXZ1Zs2aNdmk8zdIVfUPlz2iemkKhUOJqAAAAAADg4BszM1E6Oztzww035NJLLx30If1tt92WGTNm5Jhjjslf/uVfZv369c94n46OjrS0tAx6cOAWm4cCAAAAAMAhZsyEKN///vezdevWvOUtbxk499KXvjRf+9rX8rOf/SzXXHNNFi9enBe96EXp6OjY632uuuqqNDU1DTzmzZs3CtVPbMViMUtX9IUoi4QoAAAAAAAcIgrFYrFY6iKS5IILLkhVVVV+9KMf7XXNmjVr0tzcnG984xt51ateNeSajo6OQSFLS0tL5s2bl23btqWxsXHE6z4UPLZhe158zS9SVVGW+654SaoryktdEgAAAAAADEtLS0uampr2KTco6UyUXVasWJGf/vSn+e53v/uM62bPnp3m5uYsW7Zsr2uqq6tTXV090iUe0pb2b+V16uGTBSgAAAAAABwyxsR2Xtddd11mzJiRiy666BnXbdq0KStXrszs2bNHqTKSZPHy/qHytvICAAAAAOAQUvIQpbe3N9ddd10uueSSVFQ81Rizffv2vOc978kdd9yR5cuX57bbbsvLX/7yTJs2La985StLWPGhZ9c8FEPlAQAAAAA4lJR8O6+f/vSneeKJJ3LppZcOOl9eXp777rsvX/3qV7N169bMnj075513Xm666aY0NDSUqNpDz8btHfn9xh1JktPnC1EAAAAAADh0lDxEeclLXpKhZtvX1tbmlltuKUFF7G5XF8oxM+szeVJViasBAAAAAIDRU/LtvBjbluyah9I8tcSVAAAAAADA6BKi8IyWmIcCAAAAAMAhSojCXrV39eT+J7clSRbpRAEAAAAA4BAjRGGv7lm5NV09xcxoqM68qbWlLgcAAAAAAEaVEIW92rWV16IFU1IoFEpcDQAAAAAAjC4hCnu1a6i8rbwAAAAAADgUCVEYUm9vMUt360QBAAAAAIBDjRCFIS1bvz0t7d2ZVFWe42c3lrqc/3979x9kZVn3D/x9CFh+PAuCCrsr8iNDHNR8VChQ09SJwLJQTExT6IfGhCbjOEM5OeLk6PiU1phJ5SDqaDbDoMioZaCApViYooxfJJpQeUJCTVjERIz7+4eyTytnd4Fczi77es3szO4513Xv55655jP3nPe57wsAAAAAAPY6IQplLXv/UV7/ffB+6fwRywQAAAAAgI7Hp+OU9X+P8rIfCgAAAAAAHZMQhbKWNWwqbz8UAAAAAAA6JiEKO1m/6e387xv/TKdScvTA/SpdDgAAAAAAVIQQhZ089dJ7d6EcVtMr1d26VLgaAAAAAACoDCEKO3nqxff2Qxk52KO8AAAAAADouIQo7GTHnSjH2lQeAAAAAIAOTIhCI29ufTf/b119EneiAAAAAADQsQlRaGT5yxuzvUgO2q97ant3r3Q5AAAAAABQMUIUGtnxKK8R7kIBAAAAAKCDE6LQyI5N5UcMEqIAAAAAANCxCVFo8O6/tueZl98PUWwqDwAAAABABydEocEL6zdnyzv/SnVV5xzav7rS5QAAAAAAQEUJUWjw1Ivv7YdyzKA++UinUoWrAQAAAACAyhKi0GDZS/ZDAQAAAACAHYQoJEmKomi4E8V+KAAAAAAAIEThff/7xj/z9/qt6dyplP8+eL9KlwMAAAAAABUnRCFJ8qf3H+V1+EG9073rRypcDQAAAAAAVJ4QhSTJsh2P8rIfCgAAAAAAJBGi8L4dd6KMHCxEAQAAAACARIhCkk3/3JZVf9+cJDl2kE3lAQAAAAAgEaKQ5OmX30hRJIP375EDq6sqXQ4AAAAAALQJQhTy1Pv7obgLBQAAAAAA/o8QhTz1ov1QAAAAAADgg4QoHdw7727P8rUbkyQjhCgAAAAAANBAiNLBPb9uU7a+uz19enTJIQf+V6XLAQAAAACANkOI0sHteJTXsYP6pFQqVbgaAAAAAABoO4QoHdxTL723qfyIwTaVBwAAAACAfydE6cCKomi4E2XEIPuhAAAAAADAvxOidGAvvv5WXt/yTrp27pQjB/SudDkAAAAAANCmCFE6sGUvvvcor48f1DtVnT9S4WoAAAAAAKBtEaJ0YH/a8Sgv+6EAAAAAAMBOhCgd2LIdm8rbDwUAAAAAAHYiROmgXn9za/766pYkybFCFAAAAAAA2IkQpYP600vvPcrrY/3+K316dq1wNQAAAAAA0PYIUTqoHSHKyMHuQgEAAAAAgHKEKB1Ube9u+fiA3vnkkP0rXQoAAAAAALRJpaIoikoX0Zrq6+vTu3fvbNq0Kb169ap0OQAAAAAAQAXtTm7gThQAAAAAAIAyhCgAAAAAAABlCFEAAAAAAADKEKIAAAAAAACUIUQBAAAAAAAoQ4gCAAAAAABQhhAFAAAAAACgDCEKAAAAAABAGUIUAAAAAACAMoQoAAAAAAAAZQhRAAAAAAAAyhCiAAAAAAAAlCFEAQAAAAAAKEOIAgAAAAAAUIYQBQAAAAAAoAwhCgAAAAAAQBlCFAAAAAAAgDKEKAAAAAAAAGUIUQAAAAAAAMoQogAAAAAAAJQhRAEAAAAAAChDiAIAAAAAAFCGEAUAAAAAAKAMIQoAAAAAAEAZQhQAAAAAAIAyOle6gNZWFEWSpL6+vsKVAAAAAAAAlbYjL9iRHzRnnw9RNm/enCQ5+OCDK1wJAAAAAADQVmzevDm9e/dudkyp2JWopR3bvn171q1bl+rq6pRKpUqXAx+K+vr6HHzwwVm7dm169epV6XLgQ2eN0xFY5+zrrHH2ddY4HYF1zr7OGqcjsM7LK4oimzdvTl1dXTp1an7Xk33+TpROnTplwIABlS4DWkWvXr00P/Zp1jgdgXXOvs4aZ19njdMRWOfs66xxOgLrfGct3YGyg43lAQAAAAAAyhCiAAAAAAAAlCFEgXaoqqoqV111VaqqqipdCrQKa5yOwDpnX2eNs6+zxukIrHP2ddY4HYF1/p/b5zeWBwAAAAAA2BPuRAEAAAAAAChDiAIAAAAAAFCGEAUAAAAAAKAMIQoAAAAAAEAZQhRoY6677rqMHDky1dXV6devX8aPH59Vq1Y1O2fx4sUplUo7/bzwwgt7qWrYdTNmzNhprdbU1DQ7Z8mSJTn22GPTrVu3fPSjH83PfvazvVQt7JnBgweX7ctTp04tO14fp6177LHHcvrpp6euri6lUinz5s1r9H5RFJkxY0bq6urSvXv3fPrTn87zzz/f4nHnzp2b4cOHp6qqKsOHD899993XSmcAzWtujW/bti3Tp0/PkUcemZ49e6auri4XXHBB1q1b1+wxb7/99rK9/e23327ls4HyWurlkydP3mm9jho1qsXj6uW0FS2t8XI9uVQq5Qc/+EGTx9TLaUt25TND1+WtQ4gCbcySJUsyderUPPnkk1mwYEHefffdjBkzJlu2bGlx7qpVq/LKK680/AwdOnQvVAy77/DDD2+0VlesWNHk2DVr1uS0007Lpz71qTzzzDO54oor8u1vfztz587dixXD7lm2bFmjNb5gwYIkyZe+9KVm5+njtFVbtmzJUUcdlZtvvrns+//zP/+TG2+8MTfffHOWLVuWmpqafOYzn8nmzZubPObSpUszceLEnH/++Xn22Wdz/vnn5+yzz84f/vCH1joNaFJza/ytt97K008/nSuvvDJPP/107r333vz5z3/OF77whRaP26tXr0Z9/ZVXXkm3bt1a4xSgRS318iQZO3Zso/X60EMPNXtMvZy2pKU1/sF+fNttt6VUKmXChAnNHlcvp63Ylc8MXZe3jlJRFEWliwCa9uqrr6Zfv35ZsmRJTjzxxLJjFi9enJNPPjlvvPFG9ttvv71bIOymGTNmZN68eVm+fPkujZ8+fXrmz5+flStXNrw2ZcqUPPvss1m6dGkrVQkfrmnTpuWBBx7I6tWrUyqVdnpfH6c9KZVKue+++zJ+/Pgk733bra6uLtOmTcv06dOTJFu3bk3//v1z/fXX55vf/GbZ40ycODH19fX59a9/3fDa2LFj06dPn9xzzz2tfh7QlA+u8XKWLVuWT3ziE3nppZcycODAsmNuv/32TJs2LRs3bmydQuE/UG6dT548ORs3btzp2/vN0ctpq3all48fPz6bN2/OI4880uQYvZy27IOfGboubz3uRIE2btOmTUmSvn37tjj26KOPTm1tbU499dQsWrSotUuDPbZ69erU1dVlyJAhOeecc/LXv/61ybFLly7NmDFjGr322c9+Nk899VS2bdvW2qXCf+ydd97JXXfdla997WtlA5R/p4/THq1Zsybr169v1Kurqqpy0kkn5YknnmhyXlP9vbk50FZs2rQppVKpxeD7zTffzKBBgzJgwIB8/vOfzzPPPLN3CoQ9tHjx4vTr1y+HHnpoLrzwwmzYsKHZ8Xo57dXf//73PPjgg/n617/e4li9nLbqg58Zui5vPUIUaMOKoshll12WE044IUcccUST42pra/OLX/wic+fOzb333pthw4bl1FNPzWOPPbYXq4Vd88lPfjJ33nlnHn744dx6661Zv359jjvuuLz++utlx69fvz79+/dv9Fr//v3z7rvv5rXXXtsbJcN/ZN68edm4cWMmT57c5Bh9nPZs/fr1SVK2V+94r6l5uzsH2oK333473/nOd3LuueemV69eTY477LDDcvvtt2f+/Pm555570q1btxx//PFZvXr1XqwWdt24ceNy991359FHH80NN9yQZcuW5ZRTTsnWrVubnKOX017dcccdqa6uzplnntnsOL2ctqrcZ4auy1tP50oXADTt4osvznPPPZff//73zY4bNmxYhg0b1vD36NGjs3bt2vzwhz9s8hFgUCnjxo1r+P3II4/M6NGjc8ghh+SOO+7IZZddVnbOB7+9v+NJlC19qx/aglmzZmXcuHGpq6trcow+zr6gXK9uqU/vyRyopG3btuWcc87J9u3bc8sttzQ7dtSoUY025T7++ONzzDHH5Cc/+Uluuumm1i4VdtvEiRMbfj/iiCMyYsSIDBo0KA8++GCzHzTr5bRHt912W84777wW9zbRy2mrmvvM0HX5h8+dKNBGXXLJJZk/f34WLVqUAQMG7Pb8UaNG+WYE7ULPnj1z5JFHNrlea2pqdvr2w4YNG9K5c+fsv//+e6NE2GMvvfRSFi5cmG984xu7PVcfp72oqalJkrK9+oPfaPvgvN2dA5W0bdu2nH322VmzZk0WLFjQ7F0o5XTq1CkjR47U22k3amtrM2jQoGbXrF5Oe/S73/0uq1at2qNrdL2ctqCpzwxdl7ceIQq0MUVR5OKLL869996bRx99NEOGDNmj4zzzzDOpra39kKuDD9/WrVuzcuXKJtfr6NGjs2DBgkav/fa3v82IESPSpUuXvVEi7LHZs2enX79++dznPrfbc/Vx2oshQ4akpqamUa9+5513smTJkhx33HFNzmuqvzc3ByplR4CyevXqLFy4cI++yFEURZYvX6630268/vrrWbt2bbNrVi+nPZo1a1aOPfbYHHXUUbs9Vy+nklr6zNB1eevxOC9oY6ZOnZpf/vKXuf/++1NdXd2QBPfu3Tvdu3dPknz3u9/N3/72t9x5551Jkh//+McZPHhwDj/88IYNjOfOnZu5c+dW7DygKZdffnlOP/30DBw4MBs2bMg111yT+vr6TJo0KcnO63vKlCm5+eabc9lll+XCCy/M0qVLM2vWrNxzzz2VPA1o0fbt2zN79uxMmjQpnTs3vuTSx2lv3nzzzfzlL39p+HvNmjVZvnx5+vbtm4EDB2batGm59tprM3To0AwdOjTXXnttevTokXPPPbdhzgUXXJCDDjoo1113XZLk0ksvzYknnpjrr78+X/ziF3P//fdn4cKFLT7GFFpDc2u8rq4uZ511Vp5++uk88MAD+de//tVwjd63b9907do1yc5r/Oqrr86oUaMydOjQ1NfX56abbsry5cvz05/+dO+fIKT5dd63b9/MmDEjEyZMSG1tbV588cVcccUVOeCAA3LGGWc0zNHLactaul5Jkvr6+syZMyc33HBD2WPo5bRlLX1mWCqVXJe3lgJoU5KU/Zk9e3bDmEmTJhUnnXRSw9/XX399ccghhxTdunUr+vTpU5xwwgnFgw8+uPeLh10wceLEora2tujSpUtRV1dXnHnmmcXzzz/f8P4H13dRFMXixYuLo48+uujatWsxePDgYubMmXu5ath9Dz/8cJGkWLVq1U7v6eO0N4sWLSp7fTJp0qSiKIpi+/btxVVXXVXU1NQUVVVVxYknnlisWLGi0TFOOumkhvE7zJkzpxg2bFjRpUuX4rDDDivmzp27l84IGmtuja9Zs6bJa/RFixY1HOODa3zatGnFwIEDi65duxYHHnhgMWbMmOKJJ57Y+ycH72tunb/11lvFmDFjigMPPLDo0qVLMXDgwGLSpEnFyy+/3OgYejltWUvXK0VRFD//+c+L7t27Fxs3bix7DL2ctmxXPjN0Xd46SkXx/u68AAAAAAAANLAnCgAAAAAAQBlCFAAAAAAAgDKEKAAAAAAAAGUIUQAAAAAAAMoQogAAAAAAAJQhRAEAAAAAAChDiAIAAAAAAFCGEAUAAAAAAKAMIQoAAEAzSqVS5s2bV+kyAACAChCiAAAAbdbkyZNTKpV2+hk7dmylSwMAADqAzpUuAAAAoDljx47N7NmzG71WVVVVoWoAAICOxJ0oAABAm1ZVVZWamppGP3369Eny3qO2Zs6cmXHjxqV79+4ZMmRI5syZ02j+ihUrcsopp6R79+7Zf//9c9FFF+XNN99sNOa2227L4YcfnqqqqtTW1ubiiy9u9P5rr72WM844Iz169MjQoUMzf/781j1pAACgTRCiAAAA7dqVV16ZCRMm5Nlnn81XvvKVfPnLX87KlSuTJG+99VbGjh2bPn36ZNmyZZkzZ04WLlzYKCSZOXNmpk6dmosuuigrVqzI/Pnz87GPfazR/7j66qtz9tln57nnnstpp52W8847L//4xz/26nkCAAB7X6koiqLSRQAAAJQzefLk3HXXXenWrVuj16dPn54rr7wypVIpU6ZMycyZMxveGzVqVI455pjccsstufXWWzN9+vSsXbs2PXv2TJI89NBDOf3007Nu3br0798/Bx10UL761a/mmmuuKVtDqVTK9773vXz/+99PkmzZsiXV1dV56KGH7M0CAAD7OHuiAAAAbdrJJ5/cKCRJkr59+zb8Pnr06EbvjR49OsuXL0+SrFy5MkcddVRDgJIkxx9/fLZv355Vq1alVCpl3bp1OfXUU5ut4eMf/3jD7z179kx1dXU2bNiwp6cEAAC0E0IUAACgTevZs+dOj9dqSalUSpIURdHwe7kx3bt336XjdenSZae527dv362aAACA9seeKAAAQLv25JNP7vT3YYcdliQZPnx4li9fni1btjS8//jjj6dTp0459NBDU11dncGDB+eRRx7ZqzUDAADtgztRAACANm3r1q1Zv359o9c6d+6cAw44IEkyZ86cjBgxIieccELuvvvu/PGPf8ysWbOSJOedd16uuuqqTJo0KTNmzMirr76aSy65JOeff3769++fJJkxY0amTJmSfv36Zdy4cdm8eXMef/zxXHLJJXv3RAEAgDZHiAIAALRpv/nNb1JbW9votWHDhuWFF15Iklx99dX51a9+lW9961upqanJ3XffneHDhydJevTokYcffjiXXnppRo4cmR49emTChAm58cYbG441adKkvP322/nRj36Uyy+/PAcccEDOOuusvXeCAABAm1UqiqKodBEAAAB7olQq5b777sv48eMrXQoAALAPsicKAAAAAABAGUIUAAAAAACAMuyJAgAAtFueTgwAALQmd6IAAAAAAACUIUQBAAAAAAAoQ4gCAAAAAABQhhAFAAAAAACgDCEKAAAAAABAGUIUAAAAAACAMoQoAAAAAAAAZQhRAAAAAAAAyvj/pcW4TpMthaAAAAAASUVORK5CYII=",
      "text/plain": [
       "<Figure size 2000x800 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(20, 8))\n",
    "plt.plot(epochs_list, train_acc_list)\n",
    "plt.title('Model_1 accuracy')\n",
    "plt.ylabel('accuracy')\n",
    "plt.xlabel('Epoch')\n",
    "plt.legend(['Train'], loc='upper right')\n",
    "plt.show()"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Plot the change of the AUC Score of model_1 during training\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 247,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 2000x800 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(20, 8))\n",
    "plt.plot(epochs_list, train_auc_list)\n",
    "plt.title('Model_1 auc')\n",
    "plt.ylabel('Auc')\n",
    "plt.xlabel('Epoch')\n",
    "plt.legend(['Train'], loc='upper right')\n",
    "plt.show()"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Train model_2\n",
    "\n",
    "#### Train model_2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 248,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2023-10-13 17:31:47 INFO: GPU is available: True\n",
      "2023-10-13 17:31:47 INFO: Train model on 1 GPUs:\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\t GPU 0.: NVIDIA GeForce GTX 1650\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2023-10-13 17:31:49 INFO: Epoch: 1 \tTraining Loss: 0.076815 \tTraining Acc: 55.92% \t top 3 Acc: 0.71% \t AUC Score: 0.7593\n",
      "2023-10-13 17:31:50 INFO: Epoch: 2 \tTraining Loss: 0.049890 \tTraining Acc: 70.00% \t top 3 Acc: 0.81% \t AUC Score: 0.8553\n",
      "2023-10-13 17:31:52 INFO: Epoch: 3 \tTraining Loss: 0.045188 \tTraining Acc: 71.40% \t top 3 Acc: 0.82% \t AUC Score: 0.8643\n",
      "2023-10-13 17:31:53 INFO: Epoch: 4 \tTraining Loss: 0.041770 \tTraining Acc: 73.18% \t top 3 Acc: 0.83% \t AUC Score: 0.8714\n",
      "2023-10-13 17:31:54 INFO: Epoch: 5 \tTraining Loss: 0.040614 \tTraining Acc: 73.68% \t top 3 Acc: 0.83% \t AUC Score: 0.8720\n",
      "2023-10-13 17:31:55 INFO: Epoch: 6 \tTraining Loss: 0.038043 \tTraining Acc: 75.52% \t top 3 Acc: 0.84% \t AUC Score: 0.8797\n",
      "2023-10-13 17:31:57 INFO: Epoch: 7 \tTraining Loss: 0.037246 \tTraining Acc: 75.86% \t top 3 Acc: 0.84% \t AUC Score: 0.8813\n",
      "2023-10-13 17:31:58 INFO: Epoch: 8 \tTraining Loss: 0.037259 \tTraining Acc: 75.44% \t top 3 Acc: 0.83% \t AUC Score: 0.8767\n",
      "2023-10-13 17:31:59 INFO: Epoch: 9 \tTraining Loss: 0.037120 \tTraining Acc: 75.24% \t top 3 Acc: 0.84% \t AUC Score: 0.8756\n",
      "2023-10-13 17:32:00 INFO: Epoch: 10 \tTraining Loss: 0.034924 \tTraining Acc: 76.72% \t top 3 Acc: 0.84% \t AUC Score: 0.8828\n",
      "2023-10-13 17:32:02 INFO: Epoch: 11 \tTraining Loss: 0.035898 \tTraining Acc: 76.08% \t top 3 Acc: 0.83% \t AUC Score: 0.8779\n",
      "2023-10-13 17:32:03 INFO: Epoch: 12 \tTraining Loss: 0.033167 \tTraining Acc: 77.74% \t top 3 Acc: 0.84% \t AUC Score: 0.8862\n",
      "2023-10-13 17:32:04 INFO: Epoch: 13 \tTraining Loss: 0.033950 \tTraining Acc: 76.88% \t top 3 Acc: 0.83% \t AUC Score: 0.8805\n",
      "2023-10-13 17:32:05 INFO: Epoch: 14 \tTraining Loss: 0.033503 \tTraining Acc: 76.92% \t top 3 Acc: 0.84% \t AUC Score: 0.8803\n",
      "2023-10-13 17:32:07 INFO: Epoch: 15 \tTraining Loss: 0.032901 \tTraining Acc: 77.48% \t top 3 Acc: 0.84% \t AUC Score: 0.8818\n",
      "2023-10-13 17:32:08 INFO: Epoch: 16 \tTraining Loss: 0.031697 \tTraining Acc: 78.30% \t top 3 Acc: 0.84% \t AUC Score: 0.8867\n",
      "2023-10-13 17:32:09 INFO: Epoch: 17 \tTraining Loss: 0.030984 \tTraining Acc: 78.50% \t top 3 Acc: 0.84% \t AUC Score: 0.8892\n",
      "2023-10-13 17:32:10 INFO: Epoch: 18 \tTraining Loss: 0.030959 \tTraining Acc: 78.62% \t top 3 Acc: 0.85% \t AUC Score: 0.8878\n",
      "2023-10-13 17:32:12 INFO: Epoch: 19 \tTraining Loss: 0.031368 \tTraining Acc: 78.12% \t top 3 Acc: 0.84% \t AUC Score: 0.8843\n",
      "2023-10-13 17:32:13 INFO: Epoch: 20 \tTraining Loss: 0.029563 \tTraining Acc: 79.52% \t top 3 Acc: 0.85% \t AUC Score: 0.8917\n"
     ]
    }
   ],
   "source": [
    "# import scikit-learn packages\n",
    "# please use the function imported from scikit-learn to metric the process of training of the model\n",
    "from sklearn.metrics import accuracy_score,roc_auc_score, top_k_accuracy_score\n",
    "# number of epochs to train the model\n",
    "n_epochs = 20  # suggest training between 20-50 epochs\n",
    "\n",
    "model_2.train() # prep model for training\n",
    "\n",
    "train_loss_list = []\n",
    "train_acc_list = []\n",
    "train_auc_list = []\n",
    "train_top_k_acc_list = []\n",
    "\n",
    "\n",
    "# GPU check\n",
    "logging.info(f'GPU is available: {torch.cuda.is_available()}')\n",
    "if torch.cuda.is_available():\n",
    "    gpu_num = torch.cuda.device_count()\n",
    "    logging.info(f\"Train model on {gpu_num} GPUs:\")\n",
    "    for i in range(gpu_num):\n",
    "        print('\\t GPU {}.: {}'.format(i,torch.cuda.get_device_name(i)))\n",
    "    model_2 = model_2.cuda()\n",
    "\n",
    "for epoch in range(n_epochs):\n",
    "    # monitor training loss\n",
    "    train_loss = 0.0\n",
    "    pred_array = None\n",
    "    label_array =  None\n",
    "    \n",
    "    one_hot_label_matrix = None\n",
    "    pred_matrix = None\n",
    "    \n",
    "\n",
    "    for data, label in train_loader:\n",
    "        data = data.cuda()\n",
    "        label = label.cuda()\n",
    "        # implement your code here\n",
    "        optimizer_model_2.zero_grad()\n",
    "        pred = model_2(data)\n",
    "\n",
    "        loss_model_2(pred,label).backward()\n",
    "        optimizer_model_2.step()\n",
    "        train_loss += loss_model_2(pred,label)\n",
    "        #train_loss+=loss.item()*data.size(0)\n",
    "        optimizer_model_2.zero_grad()\n",
    "\n",
    "        # finish the the computation of variables of metric\n",
    "        # implement your codes here\n",
    "        if pred_matrix is None:\n",
    "            pred_matrix = pred.cpu().detach().numpy()\n",
    "        else:\n",
    "            pred_matrix = np.concatenate((pred_matrix,pred.cpu().detach().numpy()),axis=0)\n",
    "\n",
    "        if one_hot_label_matrix is None:\n",
    "            one_hot_label_matrix = np.eye(10)[label.cpu().detach().numpy()]\n",
    "        else:\n",
    "            one_hot_label_matrix = np.concatenate((one_hot_label_matrix,np.eye(10)[label.cpu().detach().numpy()]))\n",
    "\n",
    "        pred = torch.argmax(pred, axis=1)\n",
    "        if pred_array is None:\n",
    "            pred_array = pred.cpu().detach().numpy()\n",
    "        else:\n",
    "            pred_array = np.concatenate((pred_array,pred.cpu().detach().numpy()))\n",
    "\n",
    "        if label_array is None:\n",
    "            label_array = label.cpu().numpy()\n",
    "        else:\n",
    "            label_array = np.concatenate((label_array,label.cpu().numpy()))\n",
    "\n",
    "        \n",
    "\n",
    "\n",
    "        \n",
    "        \n",
    "    # print training statistics \n",
    "    # read the API document at https://scikit-learn.org/stable/modules/classes.html#module-sklearn.metrics to finish your code\n",
    "    # don't craft your own code\n",
    "    # calculate average loss and accuracy over an epoch\n",
    "    \n",
    "    top_k = 3\n",
    "    train_loss = train_loss / len(train_loader.dataset)\n",
    "    train_acc = 100. * accuracy_score(label_array, pred_array)\n",
    "    train_auc = roc_auc_score(one_hot_label_matrix, pred_matrix , multi_class='ovo')\n",
    "    top_k_acc = top_k_accuracy_score(label_array, pred_matrix , k=top_k,)\n",
    "    # append the value of the metric to the list\n",
    "    train_loss_list.append(train_loss.cpu().detach().numpy())\n",
    "    train_acc_list.append(train_acc)\n",
    "    train_auc_list.append(train_auc)\n",
    "    train_top_k_acc_list.append(top_k_acc)\n",
    "    \n",
    "    logging.info('Epoch: {} \\tTraining Loss: {:.6f} \\tTraining Acc: {:.2f}% \\t top {} Acc: {:.2f}% \\t AUC Score: {:.4f}'.format(\n",
    "        epoch+1, \n",
    "        train_loss,\n",
    "        train_acc,\n",
    "        top_k,\n",
    "        top_k_acc,\n",
    "        train_auc,\n",
    "        ))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Visualize the training process of the model_2\n",
    "Please read the [document](https://matplotlib.org/stable/tutorials/introductory/pyplot.html#sphx-glr-tutorials-introductory-pyplot-py) to finish the training process visualization.\n",
    "For more information, please refer to the [document](https://matplotlib.org/stable/tutorials/index.html)\n",
    "##### Plot the change of the loss of model_2 during training"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 249,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 2000x800 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "epochs_list = list(range(1,n_epochs+1))\n",
    "plt.figure(figsize=(20, 8))\n",
    "plt.plot(epochs_list, train_loss_list)\n",
    "plt.title('Model_2 loss')\n",
    "plt.ylabel('Loss')\n",
    "plt.xlabel('Epoch')\n",
    "plt.legend(['Train'], loc='upper right')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Plot the change of the accuracy of model_2 during training"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 250,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 2000x800 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(20, 8))\n",
    "plt.plot(epochs_list, train_acc_list)\n",
    "plt.title('Model_2 accuracy')\n",
    "plt.ylabel('accuracy')\n",
    "plt.xlabel('Epoch')\n",
    "plt.legend(['Train'], loc='upper right')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Plot the change of the AUC Score of model_2 during training\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 251,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 2000x800 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(20, 8))\n",
    "plt.plot(epochs_list, train_auc_list)\n",
    "plt.title('Model_2 auc')\n",
    "plt.ylabel('Auc')\n",
    "plt.xlabel('Epoch')\n",
    "plt.legend(['Train'], loc='upper right')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---\n",
    "## Test the Trained Network (20 marks)\n",
    "\n",
    "Test the performance of trained models on test data. Except the total test accuracy, you should calculate the accuracy for each class.\n",
    "\n",
    "About metrics, in test processing, we will use accuracy, top k accuracy, precision, recall, f1-score and confusion matrix. \n",
    "\n",
    "Besides, we will visualize the confusion matrix.\n",
    "\n",
    "Last but not least, we will compare your implementation of function to compute accuracy with the implementation of scikit-learn."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 252,
   "metadata": {},
   "outputs": [],
   "source": [
    "## define your implementation of function to compute accuracy\n",
    "def accuracy_score_manual(label_array, pred_array):\n",
    "    # implement your codes here\n",
    "    \n",
    "    if len(pred_array) != len(label_array):\n",
    "        raise ValueError(\"预测值和标签的数量不一致\")\n",
    "\n",
    "    correct_count = 0\n",
    "    total_count = len(pred_array)\n",
    "\n",
    "    for i in range(total_count):\n",
    "        if pred_array[i] == label_array[i]:\n",
    "            correct_count += 1\n",
    "\n",
    "    accuracy = correct_count / total_count\n",
    "    return accuracy"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Test model_1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 253,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2023-10-13 17:32:16 INFO: Test Loss: 0.011951\n",
      "2023-10-13 17:32:16 INFO: Test Accuracy: 0.930600\n",
      "2023-10-13 17:32:16 INFO: Test top 3 Accuracy: 0.987700\n",
      "2023-10-13 17:32:16 INFO: Test top 5 Accuracy: 0.996600\n",
      "2023-10-13 17:32:16 INFO: The classification report of test for model_1\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.94      0.97      0.96       980\n",
      "           1       0.96      0.99      0.97      1135\n",
      "           2       0.93      0.93      0.93      1032\n",
      "           3       0.94      0.90      0.92      1010\n",
      "           4       0.92      0.93      0.93       982\n",
      "           5       0.90      0.92      0.91       892\n",
      "           6       0.93      0.94      0.93       958\n",
      "           7       0.92      0.95      0.93      1028\n",
      "           8       0.93      0.90      0.91       974\n",
      "           9       0.95      0.87      0.91      1009\n",
      "\n",
      "    accuracy                           0.93     10000\n",
      "   macro avg       0.93      0.93      0.93     10000\n",
      "weighted avg       0.93      0.93      0.93     10000\n",
      "\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import classification_report,ConfusionMatrixDisplay\n",
    "# initialize lists to monitor test loss and accuracy\n",
    "test_loss = 0.0\n",
    "\n",
    "pred_array = None\n",
    "label_array =  None\n",
    "\n",
    "one_hot_label_matrix = None\n",
    "pred_matrix = None\n",
    "\n",
    "model_1.eval() # prep model for *evaluation*\n",
    "\n",
    "for data, label in test_loader:\n",
    "    data = data.cuda()\n",
    "    label = label.cuda()\n",
    "    # implement your code here\n",
    "    pred = model_1(data)\n",
    "    test_loss += loss_model_1(pred,label)\n",
    "    \n",
    "    if pred_matrix is None:\n",
    "        pred_matrix = pred.cpu().detach().numpy()\n",
    "    else:\n",
    "        pred_matrix =  np.concatenate((pred_matrix, pred.cpu().detach().numpy()))\n",
    "\n",
    "    if one_hot_label_matrix is None:\n",
    "        one_hot_label_matrix = np.eye(10)[label.cpu().detach().numpy()]\n",
    "    else:\n",
    "        one_hot_label_matrix = np.concatenate((one_hot_label_matrix,np.eye(10)[label.cpu().detach().numpy()]))\n",
    "    pred = torch.argmax(pred, axis=1)\n",
    "    \n",
    "    if pred_array is None:\n",
    "        pred_array = pred.cpu().detach().numpy()\n",
    "    else:\n",
    "        pred_array = np.concatenate((pred_array,pred.cpu().detach().numpy()))\n",
    "\n",
    "    if label_array is None:\n",
    "        label_array = label.cpu().numpy()\n",
    "    else:\n",
    "        label_array = np.concatenate((label_array,label.cpu().numpy()))\n",
    "# calculate and print avg test loss\n",
    "test_loss = test_loss / len(test_loader.dataset)\n",
    "test_acc = accuracy_score(label_array, pred_array)\n",
    "test_auc = roc_auc_score(one_hot_label_matrix, pred_matrix , multi_class='ovo')\n",
    "test_top_k3_acc = top_k_accuracy_score(label_array, pred_matrix , k=3)\n",
    "test_top_k5_acc = top_k_accuracy_score(label_array, pred_matrix , k=5)\n",
    "logging.info('Test Loss: {:.6f}'.format(test_loss))\n",
    "logging.info('Test Accuracy: {:.6f}'.format(test_acc))\n",
    "logging.info('Test top 3 Accuracy: {:.6f}'.format(test_top_k3_acc ))\n",
    "logging.info('Test top 5 Accuracy: {:.6f}'.format(test_top_k5_acc ))\n",
    "logging.info('The classification report of test for model_1')\n",
    "print(classification_report(label_array, pred_array))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 254,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "ConfusionMatrixDisplay.from_predictions(label_array,pred_array)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 255,
   "metadata": {},
   "outputs": [],
   "source": [
    "## compare your implementation of function to compute accuracy with the implementation of scikit-learn.\n",
    "your_test_acc = accuracy_score_manual(label_array, pred_array)\n",
    "assert abs(your_test_acc - test_acc) < 1e-5 , 'Please check your implementation of function to compute accuracy'"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Test model_2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 256,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2023-10-13 17:32:19 INFO: Test Loss: 0.013490\n",
      "2023-10-13 17:32:19 INFO: Test Accuracy: 0.934200\n",
      "2023-10-13 17:32:19 INFO: Test top 3 Accuracy: 0.984200\n",
      "2023-10-13 17:32:19 INFO: Test top 5 Accuracy: 0.988400\n",
      "2023-10-13 17:32:19 INFO: The classification report of test for model_2\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.95      0.98      0.96       980\n",
      "           1       0.98      0.98      0.98      1135\n",
      "           2       0.95      0.90      0.92      1032\n",
      "           3       0.93      0.92      0.93      1010\n",
      "           4       0.91      0.94      0.93       982\n",
      "           5       0.95      0.90      0.93       892\n",
      "           6       0.95      0.94      0.94       958\n",
      "           7       0.90      0.96      0.93      1028\n",
      "           8       0.89      0.93      0.91       974\n",
      "           9       0.94      0.89      0.91      1009\n",
      "\n",
      "    accuracy                           0.93     10000\n",
      "   macro avg       0.93      0.93      0.93     10000\n",
      "weighted avg       0.93      0.93      0.93     10000\n",
      "\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import classification_report,ConfusionMatrixDisplay\n",
    "# initialize lists to monitor test loss and accuracy\n",
    "test_loss = 0.0\n",
    "\n",
    "pred_array = None\n",
    "label_array =  None\n",
    "\n",
    "one_hot_label_matrix = None\n",
    "pred_matrix = None\n",
    "\n",
    "model_2.eval() # prep model for *evaluation*\n",
    "\n",
    "for data, label in test_loader:\n",
    "    data = data.cuda()\n",
    "    label = label.cuda()\n",
    "    # implement your code here\n",
    "    pred = model_2(data)\n",
    "    test_loss += loss_model_2(pred,label)\n",
    "    \n",
    "    if pred_matrix is None:\n",
    "        pred_matrix = pred.cpu().detach().numpy()\n",
    "    else:\n",
    "        pred_matrix =  np.concatenate((pred_matrix, pred.cpu().detach().numpy()))\n",
    "\n",
    "    if one_hot_label_matrix is None:\n",
    "        one_hot_label_matrix = np.eye(10)[label.cpu().detach().numpy()]\n",
    "    else:\n",
    "        one_hot_label_matrix = np.concatenate((one_hot_label_matrix,np.eye(10)[label.cpu().detach().numpy()]))\n",
    "    pred = torch.argmax(pred, axis=1)\n",
    "    \n",
    "    if pred_array is None:\n",
    "        pred_array = pred.cpu().detach().numpy()\n",
    "    else:\n",
    "        pred_array = np.concatenate((pred_array,pred.cpu().detach().numpy()))\n",
    "\n",
    "    if label_array is None:\n",
    "        label_array = label.cpu().numpy()\n",
    "    else:\n",
    "        label_array = np.concatenate((label_array,label.cpu().numpy()))\n",
    "# calculate and print avg test loss\n",
    "test_loss = test_loss / len(test_loader.dataset)\n",
    "test_acc = accuracy_score(label_array, pred_array)\n",
    "test_auc = roc_auc_score(one_hot_label_matrix, pred_matrix , multi_class='ovo')\n",
    "test_top_k3_acc = top_k_accuracy_score(label_array, pred_matrix , k=3)\n",
    "test_top_k5_acc = top_k_accuracy_score(label_array, pred_matrix , k=5)\n",
    "\n",
    "logging.info('Test Loss: {:.6f}'.format(test_loss))\n",
    "logging.info('Test Accuracy: {:.6f}'.format(test_acc))\n",
    "logging.info('Test top 3 Accuracy: {:.6f}'.format(test_top_k3_acc ))\n",
    "logging.info('Test top 5 Accuracy: {:.6f}'.format(test_top_k5_acc ))\n",
    "logging.info('The classification report of test for model_2')\n",
    "print(classification_report(label_array, pred_array))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 257,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAgQAAAGwCAYAAADWsX1oAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8pXeV/AAAACXBIWXMAAA9hAAAPYQGoP6dpAACfSUlEQVR4nOzdd3QU9drA8e+27GbTe0gIJEDoAREVwQI2lKKorwVB7JWiCKKComAhiAqoqKiXC1wRAbteFQEvIChIEaRJh5BCGult67x/RAJLQkhIdmchz+ecPSc7O7PzZHd25plf1SiKoiCEEEKIJk2rdgBCCCGEUJ8kBEIIIYSQhEAIIYQQkhAIIYQQAkkIhBBCCIEkBEIIIYRAEgIhhBBCAHq1A2gIp9NJRkYGAQEBaDQatcMRQghRT4qiUFxcTExMDFqt++5RKyoqsFqtDX4fHx8fTCZTI0Tkfc7phCAjI4O4uDi1wxBCCNFAqampNG/e3C3vXVFRQUJLfzKzHQ1+r+joaA4dOnReJgXndEIQEBAAwDu/dsXXX6dyNCd8eqEkKeI8JyVydSMDwZ6RHRtr+bHqfO4OVquVzGwHKZvjCQw4+1KIomInLbsfxmq1SkLgbY5XE/j66zB7UUKg1xjUDkEI95KEoI4kITijfz4iT1T7+gdo8A84+/04Ob+P+3M6IRBCCCHqyqE4cTQgR3MozsYLxgtJQiCEEKJJcKLgbECpTUO2PRdIt0MhhBBCSAmBEEKIpsGJk4YU+jdsa+8nCYEQQogmwaEoOBrQ86Mh254LpMpACCGEEFJCIIQQommQRoW1k4RACCFEk+BEwSEJwWlJlYEQQgghpIRACCFE0yBVBrU7rxMCW4mGP98OJmWFmYpjWkI7WukxIZ+ILpUzXq15Loz9X/u7bBPR1cLAJZkAFKfp+OKamifb6DMzh4R+ZW6LfeC9udz+eA6hkTZS9pqY/WIMOzb4n3lDN7hzZBaX9S8kro0Fa4WWXZvMzHmtGWkH1B3Lu3OPEm4fnkNiUhlh0XYmPRDPuqVBEpOXx6TVKQwbm8nVt+QTEmEjL9vA8iWhLHw7CkVRb2jYsGgrD044ysVXF+FjcpJ+0Mj0sS3Yv92sWkwnu3NkFg9MyOTrj8OZ/VKsqrF40/mpPqSXQe3O64Rg7QthFOwzcOW0XMyRDg5858fP90dxy48Z+EVVznoVe0U5lyfnVm2jO2kaAr9mDu5cm+rynnsXB7B9TiDNryx3W9y9b8rnsckZzJoQy84NfgwYdoxXPz3Ew33akZPu47b9nk6XnqV8Py+cvVvN6PQK9z17lCmfHeTh3u2wlKs3h4TJ7OTgThPLFoXw4pwU1eI4mcR0ZneOyGLAsFzeHN2ClD0mEruWM3b6EUqLdXwzJ0KVmPyD7Ez/Zh/bfg/ghbtbUZCrp1m8ldIi75gjpW3XMvrfncfBnepPqONt5yfReFRvQ/D++++TkJCAyWSie/furFmzplHe116hIWWZmYvGFRB9sYXAlna6jSokoLmd3QtPzKql81EwRzirHsbgEwNPaHW4vGaOcJKywkxCvzIMfu7LFG99JJefPwtl6cIwUvebmP1SLDkZBgbec8xt+6zN80NbsXxJKCl7TRzc5ctbT7UgqrmNxC7uS4rqYtPKQOZPa8ZvPwWrGsfJJKYz69C9jHU/B7HhlyCy0oys/SGYP1cHkNjVfSVuZ3LH8GxyM3x4a0wL9mz1IyvNyNa1ARxNMaoW03Ems4NnZ6Uwc1xzigvVT1C87fxUH85GeJzPVE0IFi9ezOjRo3n++efZsmULV1xxBf369ePIkSMNfm/FDopDg87oeuHWmRSy/zzxI8/cYOKzns358voYfnshlPJjp/9Icnf4kPe3D4m3lTQ4vtPRG5wkdilj82rXqUA3rw6g40WlbttvffgFVpauFBeof3IS554dG/y44PJiYltVANCqYzmdLill4y+BqsV0ad9C9m4z8/yHh1j81w7e+3kP/YZ4xwVu5JR0NvwSyJY17pseuK7OhfNTbRz/9DJoyON8pmqVwfTp03nwwQd56KGHAJg5cyY///wzH3zwAcnJyQ16b4O/QkS3Cv56P4jgVjZM4Q4O/dePnL98CGxpByD2ynLibyjDP8ZOcZqeLW8Hs/TeKG766ii6Gkq+9n3hT1BrK1EXWhoUW20CQx3o9FCQ6/rVFOToCYm0u22/dafwyKQMdvzhR8oeX7WDEeegJe9F4hfg4F+rd+N0VJbEzXu9Gau+DVEtpmYtrAwclstXH0ew6J0o2nUr4/GX07BZNaz4IlS1uHoPyqdNUjmj+ieqFsPJvP/8VDuHQgNnO2y8WLyRagmB1Wpl8+bNPPfccy7L+/bty++//17jNhaLBYvlxMW4qKio1n1cOe0YayeEsfjK5mh0CmEdrbQaWMqxXZVX+1b9TxRRhrS1Ed7ZyudXx5K6ypf4vq7F4fYKDQf/60fX4QX1+TfP2qltVzQavGJq9RFT0knoUM7Ym9uoHYo4R/W+qYBr/i+fqSNakrLXROtO5Tw2OZ1jWQZWfK7OxVejhX3bfJk7NQaAAzvNtGxbwYB7clVLCCJirDz+cgYT7mqFzaJ67a4Lbz0/iYZRLSHIzc3F4XAQFRXlsjwqKorMzMwat0lOTmby5Ml13kdgCzv9F2RhK9NgK9FijnSwcnQ4Ac1rzmTNkQ78YuwUHTYArgnB4aVm7BUa2tzs3mKxojwdDjuERLjGGBRuJz9H3Tagw19No2ffIsbe0prco9J4SJydhydmsHhWJKu/qywROLzbl8jmVgaPzFItIcjL1pOy17XBXup+E5f3L1QlHoA2XcoJibAza+neqmU6PSRdWspN9+cyML4LTqdne2V48/mpLhraDkDaELiZRuN6QCuKUm3ZcePHj6ewsLDqkZqaWuN6pzKYFcyRDiyFWjLW+tLimpobw1Xkayk7qscc6aj22r4v/Ym7ugxTqHsPCbtNy75tZi68sthl+YVXFrNrk59b9316CiNeS+OyfoU8c3trslLVb2glzl1GX2e17oVOhwaNimejXRv9iGvtWhUY28pCdrrhNFu439Y1/jxyVVsev+7EY89WX/73VQiPX9fW48kAeOv5qe6caHA04OFEvW6xnqBaShceHo5Op6tWGpCdnV2t1OA4o9GI0Vj3i1H6GhOKAkEJdoqO6Nk0LYTABBuJt5ZgK9WwZVYQ8X3L8I1wUJKuZ/OMYIwhDlpe69rauShFT+ZGI9d9lF3/f/QsfPVROOPeSWXvNl/+3uRH/7uPERlr44f/hHlk/6caOSWdq27JZ9L9CZSXaAmJsAFQWqzDWqHeWdxkdhCTYK16Hh1npVWncooLdKp1f5KYzmz98kAGP5FFdrqBlD0mWncu59ZHslm2SJ3jG+CrjyOZ8e1eBo/K4tfvg2l3QRn9hx5j5jM1j0PiCeWlumrtdCrKtBTnV1/uSd52fhKNR7WEwMfHh+7du7N8+XJuueWWquXLly9n0KBBjbIPa7GWzdODKc3UYwx20LJvGd2fKkBrAKcD8vf6cOAbf6zFWnwjHDTrUUGfGbkY/F0rw/Z96Y85ykHs5RWNEteZrP4uhIAQB0OfyiI00k7KHhMv3J1AtkoXlBvvq2xt/eZXB1yWvzk6juVL1Gtw1bZrOW98eSKmxyZnALBscQhvPdVCYvLSmN5/oTn3PnOUkVPSCA6zcyzLwI8Lwvl0Rs03Ap6w9y8zLz+UwP3PHWXo6EwyU32Y/VIsK79W7/j2Vt52fqoPp1L5aMj25zONoqg39NLixYsZNmwYs2fPpmfPnnz00Ud8/PHH7Ny5k5YtW55x+6KiIoKCgvj4zwsx+3tPF7i57c4cuxDntNNU64lTnOcj2zUGu2JjFd9SWFhIYKB7up4ev1b8sTMa/4CzL9UsKXbSo1OmW2NVk6qtQO68806OHTvGyy+/zNGjR+ncuTM//vhjnZIBIYQQQjQe1ZuFDh8+nOHDh6sdhhBCiPPc8caBDdn+fKZ6QiCEEEJ4glPR4GzABFoN2fZcoHq3QyGEEEKoT0oIhBBCNAlSZVA7SQiEEEI0CQ60OBpQMF59yLrziyQEQgghmgSlgW0ITh1h83wjbQiEEEIIIQmBEEKIpqEh8xicTfuDX3/9lRtvvJGYmBg0Gg3ffPONy+uKojBp0iRiYmLw9fWlT58+7Ny502Udi8XCqFGjCA8Px8/Pj5tuuom0tDSXdfLz8xk2bBhBQUEEBQUxbNgwCgoK6v35SEIghBCiSXAo2gY/6qO0tJSuXbsya9asGl+fNm0a06dPZ9asWWzcuJHo6Giuu+46iotPTB41evRovv76axYtWsTatWspKSlh4MCBOBwnWjQMGTKErVu3snTpUpYuXcrWrVsZNmxYvT8faUMghBBCuEG/fv3o169fja8pisLMmTN5/vnnufXWWwGYP38+UVFRLFy4kEcffZTCwkLmzJnDJ598wrXXXgvAggULiIuLY8WKFVx//fX8/fffLF26lPXr19OjRw8APv74Y3r27MmePXto165dneOVEgIhhBBNghMNTrQNeFRWGRQVFbk8LBbLGfZc3aFDh8jMzKRv375Vy4xGI7179+b3338HYPPmzdhsNpd1YmJi6Ny5c9U669atIygoqCoZALj00ksJCgqqWqeuJCEQQgjRJDRWG4K4uLiq+vqgoCCSk5PrHUtmZiYAUVGus3xGRUVVvZaZmYmPjw8hISG1rhMZGVnt/SMjI6vWqavzosrg0wvj0GsMaodR5eeMrWqHUM31MReoHUJ13jhjnsxOVzfyOdWJRu99p1jFblc7hHNeamqqy2yHRqPxrN9Lc8p5UFGUastOdeo6Na1fl/c5lZQQCCGEaBIaq1FhYGCgy+NsEoLo6GiAanfx2dnZVaUG0dHRWK1W8vPza10nKyur2vvn5ORUK304E0kIhBBCNAmVbQga9mgsCQkJREdHs3z58qplVquV1atX06tXLwC6d++OwWBwWefo0aPs2LGjap2ePXtSWFjIhg0bqtb5448/KCwsrFqnrryvPEsIIYQ4D5SUlLB///6q54cOHWLr1q2EhobSokULRo8ezZQpU0hMTCQxMZEpU6ZgNpsZMmQIAEFBQTz44IOMHTuWsLAwQkNDefrpp0lKSqrqddChQwduuOEGHn74YT788EMAHnnkEQYOHFivHgYgCYEQQogmwtnAuQyc1K/tzKZNm7jqqquqno8ZMwaAe++9l3nz5vHMM89QXl7O8OHDyc/Pp0ePHixbtoyAgICqbWbMmIFer+eOO+6gvLyca665hnnz5qHT6arW+fTTT3niiSeqeiPcdNNNpx37oDYaRTl3WwcVFRURFBREHwZJo8IzkEaFdXTu/hyEF5JGhWdmV2ys4lsKCwtdGuo1puPXikVbO2IO0J15g9MoK3Yw+IJdbo1VTd53tAohhBBucHw8gbPf/vy+YZBGhUIIIYSQEgIhhBBNg0PR4GjAFMYN2fZcIAmBEEKIJsHRwEaFDqkyEEIIIcT5TkoIhBBCNAlORYuznlMYu25/fpcQSEIghBCiSZAqg9pJQgAMvDeX2x/PITTSRspeE7NfjGHHBv8Gv+/29X58/n4k+7abycsy8NKcQ/TqV1j1+tofg/jxkzD2bTNTlK/n/WV7aN253OU9rBYNH78cw6pvQrBUaOh2eQkjk9OIiLEB8Nfv/jxzW5sa9//Oj3tod0F5ja/Vx50js7isfyFxbSxYK7Ts2mRmzmvNSDtgavB7n627xxxl2FjX8bvzsvXc1a2zShFV6tyjhNuH55CYVEZYtJ1JD8SzbmmQqjGB+47xs4rlnlwG3HOMqDgrACl7THw6I4pNK9Xr1+0NMXW+pJjbHsuqPHaibEx+qDXrlgXXuO4TySn0H5rL7MnN+WZO/carbwzedDyJxtPk2xD0vimfxyZn8Nk7kQzv25Ydf/jx6qeHiIi1Nvi9K8q0tOpUzojX0k77eseLS3lgQsZp32P2S7H8vjSI8R8cZvo3+ykv0/LiPa1wOCpf73hRKZ9t3eHyuGHIMaLiLLTt2vBkAKBLz1K+nxfO6IGJjB/cCp1OYcpnBzH6Ohrl/c/W4d0mBl/Qqerx2DXtVY0HwGR2cnCnifeej1U7lCruPMbPRs5RA/+e0oxR/doyql9b/vrNn0lzD9OybYUq8XhLTCazk0O7fHl/Ylyt6/XsW0C7C0rJzVRnMDZvO57qw8mJngZn83Cq/Q+4maoJwa+//sqNN95ITEwMGo2Gb775xuMx3PpILj9/FsrShWGk7jcx+6VYcjIMDLznWIPf++Kri7nv2Uwu719Y4+vX3pbP3WOy6HZlSY2vlxZp+fmzUB5+MYMLryyhTVI5z76bwuHdJrasqRza0uCjEBppr3oEhthZvyyQ6wfnNdpAgM8PbcXyJaGk7DVxcJcvbz3VgqjmNhK7NE7CcbYcDsjPMVQ9CvPUL/DatDKQ+dOa8dtPwWqHUsWdx/jZ+GN5EBv/F0j6QSPpB43Me70ZFaVa2ncvVSUeb4lp06og5r8Zy29LQ067TliUleGvHGHakwk4bOp0gfO246k+jg9M1JDH+UzV/660tJSuXbue1ZjLjUFvcJLYpYzNqwNclm9eHUDHi9Q7OR23b5sZu01L997FVcvCou20bF/Bro1+NW6zblkQRXl6rrsjz21x+QVWlgwUF5z9EKCNITbBysLNO5i/bhfj3z9MdAuLqvF4I28/xrVahd6D8jGanfy9qeZj2tO8MSYAjUZh3MzDfPFhFCl7fVWJwduPJ9Ewqt5S9evXj379+qm2/8BQBzo9FOS6fgwFOXpCItUf7zsvW4/Bx0lAsGvRfEi4jfycmr+6nz8Lo3ufYiJjbW6KSuGRSRns+MOPlD3qnJQAdm/x440nfUk7aCQkws5dT2Qy49t9PHJ1e4rz1S8p8BbeeozHty9n5vf78TE6KS/V8vKD8RzZp16bFG+N6WR3DM/E4YBv/x2pWgzeejzVlUPR4mhAL4OGbHsuOKfOnBaLBYvlxF1gUVFRo7zvqT1JNBrw5sakiqKhpmm5czIMbF4VwIQPD7tt3yOmpJPQoZyxN9fckNFTTm7sdXg37NpkZt7vf3Pd7Xl89ZF6J0xv5W3HeNoBI8Ova4tfoIPLBxTy9NtHGHdrG1UvwN4Y03FtkkoZdH82Iwd0oMYfv4d52/FUV040OBvw+TVk23PBOZXuJCcnExQUVPWIi6u98c2ZFOXpcNghJMI1sw0Kt5/2DtyTQiPt2KzaakXzBcf0hIRXz8aXLQ4lIMROz741t1loqOGvptGzbxHP3Naa3KM+btnH2bKU6zi820RsglQbnMxbj3G7TUvGYSP7tpmZm9yMQ7t8ufmhHNXi8daYjut8SQnB4XY+WbedHw5u5oeDm4mKs/LwC2nM/227x+Lw1uOpro6XEDTkcT47p/678ePHU1hYWPVITU1t0PvZbVr2bTNz4ZXFLssvvLKYXV5Qd5jYpQy9wcmfv56orzuWpSdlt4mOF7vW1ylKZUJw7W356Bu98bHCiNfSuKxfIc/c3pqsVGNj76DBDD5O4hIt5GV5zzTY3sDbj/GTGXy87xbTW2L65cswHu/bkeE3nHjkZhr44sMonh+W6LE4zqXjSdSf96d0JzEajRiNjXsx+uqjcMa9k8rebb78vcmP/ncfIzLWxg//CWvwe5eXask4dCLezFQfDuzwJSDYTmRzG0X5OnLSfTiWVfk1pB6oXDck0kZopB2/QCfX35XHR5NjCAyxExDs4ONXYohvX0G3K1x/kFvX+pN5xMgNQxq/pe/IKelcdUs+k+5PoLxES0hEZfuE0mId1gp1csqHJ6azfnkQ2ekGgsPtDHkyC7O/g+Wfh6oSz3Ems4OYhBPdr6LjrLTqVE5xQeV3rQZ3HuNn4/7njrLxfwHkZPjg6++gz6ACuvQq4YWhrVSJx1tiMpkdxMSfKOGKjrPQqmMZxQV6cjJ8KC5wPV07bBrycwykHfRslYa3HU/10fCBic6pe+h6O6cSAndY/V0IASEOhj6VRWiknZQ9Jl64O4HsRjh57/3L7DJo0IeTKvumX3dHHk/PPML6ZUG89VSLqteTH48H4O4xmQx7OhOAxyalo9MpvPZYPNZyLRdcXszk+QfRndLAf+lnYXS8qIQWiY1fZH7jfZVJxptfHXBZ/uboOJYvUecCHN7Mxvj3DhMY6qDwmJ7df5oZfWPbRvneGqJt13Le+PLE5/TY5MoxJpYtDnH5rj3Jncf42QiOsDPu3SOERtopK9Zx6G8TLwxt5VIS1hRjatuljGlL9lY9f/SlyvFLln8exltj4z0Wx5l42/FUH05Fg7MBMxY2ZNtzgUZR1BucuaSkhP379wPQrVs3pk+fzlVXXUVoaCgtWpz55FlUVERQUBB9GIRe4z1FxT9nbFU7hGquj7lA7RCqa6yBEhrTeT5WufAsjd777rkUu3f1BrArNlbxLYWFhQQGumdkyOPXimkbr8DX/+y/k/ISO89cvMatsapJ1aN106ZNXHXVVVXPx4wZA8C9997LvHnzVIpKCCHE+cjZwCqD831gIlUTgj59+qBiAYUQQogmpOGzHZ7fCcH5/d8JIYQQok68r4JLCCGEcAMHGhwNGFyoIdueCyQhEEII0SRIlUHtzu//TgghhBB1IiUEQgghmgQHDSv2d5x5lXOaJARCCCGaBKkyqJ0kBEIIIZoEmf64duf3fyeEEEKIOpESAiGEEE2CggZnA9oQKNLtUAghhDj3SZVB7c7v/04IIYQQdXJelBBo9Ho0Gu/5V7xxZsGBO/PVDqGa/yaFqx1Cdd5YIuj0ws5OXjhTpUbvPTOeHqcxeM956TjF4W3HkwY8NKWNTH9cO+87WoUQQgg3cDRwtsOGbHsuOL//OyGEEELUiZQQCCGEaBKkyqB2khAIIYRoEpxocTagYLwh254Lzu//TgghhBB1IiUEQgghmgSHosHRgGL/hmx7LpCEQAghRJMgbQhqJwmBEEKIJkFp4GyHioxUKIQQQojznZQQCCGEaBIcaHA0YDjShmx7LpCEQAghRJPgVBrWDsDpoSGW1SJVBkIIIYRoWiUEnS8p5rbHskhMKiMsysbkh1qzbllwjes+kZxC/6G5zJ7cnG/mRHk20JPcOTKLByZk8vXH4cx+KdYt+7CXwp53fMn8xYAlT0tQBwednisjOMmB01b5WvYaA2VpWvT+CuE9bXR4qhxTZPV0WVFgw2P+5Kw1cNE7JURfY2uUGDv3KOb2x7JITConLNrGpAdbse7n4JP3zN1jjtJ/yDH8g+3s3uLHe8/HkbLXt1H2X1e+fg7uHZdBrxsKCQ63cWCHmQ9eas7ev/w8Gsdxd47M4rL+hcS1sWCt0LJrk5k5rzUj7YBJlXgAtDqFYWMzufqWfEIibORlG1i+JJSFb0eheKAV953DM7jshnyat66o/Ew2+/Pvqc1JO3jysaJw9+gM+g3JwT/Izp4t/rw3sSUp+9x3PA0YksmAIVlENbcAkLLPl4XvNmfTryEADH0ild4DcoloZsVm07B/hz/zp8ex568At8V0KrW/u4ZyNrBRYUO2PRec3//dKUxmJ4d2+fL+xLha1+vZt4B2F5SSm6nu7Gltu5bR/+48Du5078n7rxf9yFln4IKpZfT+uojwXjbWPxRAeZYGRwUU/q0j8bFyrvi8iIveLqH0sI6NI/1rfK9D/zG6ZcZAk9nJwV1m3pvYvMbX7xiexa0PZ/PexOaMGtCe/GwDyQv34+vn2ZndnnojhQuvKGbaky157NoObP41gKmf7SMs2urROI7r0rOU7+eFM3pgIuMHt0KnU5jy2UGMvurNeHfniCwGDMvlvRdiebhPe/71Wgy3PZ7NoAdyPbL/pB7FfP+fKJ66uSPj726HTq/w2id7XT6T2x/L5JaHMnn/xRY8cWNH8nIMTPl0j1uPp9xMH+a+0YInbk7iiZuT+GtdEC/O3kOLxDIA0g+ZeH9yAo8P6MrTgzuTlW7ktXl/ExTaOEl3Xaj93TWUE02DH+czVROC5ORkLr74YgICAoiMjOTmm29mz549btvfplVBzH8zlt+Whpx2nbAoK8NfOcK0JxNw2NT78k1mB8/OSmHmuOYUF+rcth9HBWQuN9BhbBlhF9nxa+mk3YgKzLEOUhYZMQTApf8qIeYGG/4JTkK6Oug8oYzCnXrKM1w/n6LdOg7+x0TXV0obPc5NK4OY/0YMv/1U03encPOD2Sx6N5rffgohZY8vbz7VEqOvk6tuzmv0WE7Hx+Tk8v4F/Ou1WHb8EUDGYRMLpseQmWpk4DB1TpjPD23F8iWhpOw1cXCXL2891YKo5jYSu5SrEg9Ah+5lrPs5iA2/BJGVZmTtD8H8uTqAxK5lHtn/C/e2Y/kX4aTs8+XQ32amP51AVHMriUnH969wy4NZLJoVw29LQ0nZa+atsQkYTU6uGnTMbXH98b9QNq4OIf2wL+mHfZk/vQUVZVraX1AMwKrvI9j6ezCZqSaO7DPz8ZSW+AU4SGjnmc8N1P/uhHupmhCsXr2aESNGsH79epYvX47dbqdv376Uljb+BaUuNBqFcTMP88WHUR4vaj7VyCnpbPglkC1r3FscqDhAcWjQGV2Xa02Qt6XmGiVbiQY0CvrAE1UGjnL4c5wfnZ8vwxTh2ZY30S2shEXZ2bw68ESMVi3b1/vT8SLPHUs6nYJOD1aLa6JkqdDS6ZISj8VRG7/Ayjvc4gL3JZlnsmODHxdcXkxsqwoAWnUsp9MlpWz8JfAMW7qHOcD1M4mOsxAaaePPNaccT38E0KG7Z75HrVah94BcTGYnu7dUPwfoDU763ZlNSZGOg7vNHokJvO+7q6/jIxU25HE+U7UNwdKlS12ez507l8jISDZv3syVV17p8XjuGJ6JwwHf/jvS4/s+We9B+bRJKmdU/0S370vvByEX2Nk724R/q1KMYQrpP/pQsE2HX0tntfUdFtg9w5fYAVYMJ9Ua7HzdTEg3O9FXe6748rjQiMp95ue6Hs75uXoiYz1XVF9eqmPXJj+GjM7kyH4TBTkG+tycR/tupaQfMp75DdxO4ZFJGez4w4+UPeolvEvei8QvwMG/Vu/G6QCtDua93oxV356+5M59FB6dmMqODf6k7K28sIZE/nM85bhWGebnGoiKtbg1mvi2pUz/fAc+RiflZTpeebwdR/afuOBfclU+z83ci9HXSV62gefv7UhRvueqNr3ru6s/aUNQO69qVFhYWAhAaGhoja9bLBYslhM/yKKiokbbd5ukUgbdn83IAR1wSyV4HUXEWHn85Qwm3NUKm8UzB98FyaX8NdHMiquC0egUAjs4iB1gpXCX6+HhtMGfT/uhOKHzxBNFhJn/M5D7h54rv2i87+OsnJK9azTg6e9y2pPxjHkrhc8278Bhh/07zKz8JoQ2ndUroj9uxJR0EjqUM/bmNqrG0fumAq75v3ymjmhJyl4TrTuV89jkdI5lGVjxec2/fXcZ8coREtqXMfa2DmdcV6NRUNxc+JV2yJcRN3XBP8DBZTccY+wb+3lmSKeqpOCv9YGMuKkLQSF2brgzi/Hv7GX0/yVRmOeZpMCbvjvR+LwmIVAUhTFjxnD55ZfTuXPnGtdJTk5m8uTJbtl/50tKCA6388m67VXLdHp4+IU0bnkgm3svS3LLfk/Vpks5IRF2Zi3d6xJH0qWl3HR/LgPju+B0Nu5Fzq+Fk17zS7CXgb1UgylCYfNYP8zNT5QQOG2weawfZWlaes4tcSkdOPaHnrJULT/3DHZ5302j/QjtbqfXPPcWs+b9cyd3vNXzccFhdvJzPHuIH00xMu62thh9HfgFVN7FTXj/IJmpPh6N41TDX02jZ98ixt7Smtyj6sby8MQMFs+KZPV3lXeVh3f7EtncyuCRWR69qDw+OYVLr83n6Ts6kJt54jPJzz75eDqxPDjMTn6uey+8dpuWoymVpTf7dvjTNqmUQfce5d2JrQGwlOs4muLL0RTYvTWAf63YwvV3ZLNktnt6IJ3KW767s+WkgXMZnOeNCr0mIRg5ciTbtm1j7dq1p11n/PjxjBkzpup5UVERcXG19xioq1++DGPLGtd6sNcW7OOXr0JZviS8UfZRF1vX+PPIVW1dlo2dkUrqfhNL3oto9GTgZHoz6M0K1kINOb/p6TCm8q62KhlI0XHp3GJ8gl1vk1o/VEHcba5Fqb/eHESnZ8uJ6uP+KoTMIz4cy9Jz4ZVFHNhZeSelNzhJurSEOVNi3L7/mljKdVjKdfgH2eneu5h/TfHMCbs6hRGvpdPrhkLG3daGrFT1qy6Mvs5qXdScDg0aj5XGKgx/+Qi9rs/nmTvbV/tMMlON5GUb6HZ5EQd2VnYX1RucJPUo5t9Ta+7l4i4ajYLB5/TFEpWvV6/acxf1v7uGURrYU0CRhMD9Ro0axXfffcevv/5K8+an/8EZjUaMxrM/oZnMDmLiT1y4ouMstOpYRnGBnpwMH4oLXD8Oh01Dfo6BtIOe67NdXqqrVr9bUaalOL/68saSvVYPCvgnOCk9ouXvN33xj3cSd4sVpx02P+VH4d96LnmvBMUBFTmVPwqfIAWtD5gilBobEvo2c7qUMjTEmb67b+ZEMnhkFumHTKQfMnLXqEws5VpWfuPZu5buvYvQaBRSD5iIjbfw0AvppB00smxxmEfjOG7klHSuuiWfSfcnUF6iJeSf9halxTqsFeqcxdcvD2TwE1lkpxtI2WOidedybn0km2WLPPMZjXg1hatuymPyw20oL9Wd+EyKdFgtWkDD13OiGDziKBmHK4+nwSOPYqnQsvJb98V479gjbFodTM5RH8x+DnoPPEZSjyImPtABo6+DwcPT+eOXEPKyfQgIsTFwaBbh0VbW/OS5Y0vt766hZLbD2qmaECiKwqhRo/j6669ZtWoVCQkJbt1f2y5lTFtyoij+0ZfSAFj+eRhvjY136769mb1Ew+6ZvlRkajEEKURfZ6X9k+VoDVCWriVrZWWx6a//51qCcuncYsIvsXskxrZdy3jj831Vzx+blA7AsiWhvDUmniXvR+FjcjLytSMEBDnYvdWP8UMrT/ie5Bfg4P7n0glvZqO4QMdvP4Uw9/UYHHZ1TiQ33lfZTe7Nrw64LH9zdBzLl6hTxPv+C82595mjjJySRnCYnWNZBn5cEM6nMzwzANiNw3IAeGOJaxfnt8YmsPyLytLAz2dHYzQ5GflqCv6BdnZv9WfC3W3dejyFhFsZ9+Z+QiOtlBbrOLTbj4kPdGDLb8EYfJzEtSrn2luyCQq1U5SvZ+92f8YN7syRfZ7rZaD2dyfcS6Mo7m4mc3rDhw9n4cKFfPvtt7Rr165qeVBQEL6+Z74bLioqIigoiKv0/4deo+4gQidT7J65SNbHwJ35aodQzX+TPFcVc05zqjeI0GlpvO9OSaP3nnPAcRqDVxTCunCWq9/A9WR2xcYq5RsKCwsJDHRP98Xj14pblt+Pwe/s29DYSq18fd1ct8aqJlWP1g8++ACAPn36uCyfO3cu9913n+cDEkIIcd6SKoPaqdoURFGUGh+SDAghhDjX2e12XnjhBRISEvD19aVVq1a8/PLLOJ0n2lYpisKkSZOIiYnB19eXPn36sHPnTpf3sVgsjBo1ivDwcPz8/LjppptIS0tr9HjPkbahQgghRMN4ei6D119/ndmzZzNr1iz+/vtvpk2bxhtvvMG7775btc60adOYPn06s2bNYuPGjURHR3PddddRXFxctc7o0aP5+uuvWbRoEWvXrqWkpISBAwficDRudaL3VXAJIYQQbuDpKoN169YxaNAgBgwYAEB8fDyfffYZmzZtAipLB2bOnMnzzz/PrbfeCsD8+fOJiopi4cKFPProoxQWFjJnzhw++eQTrr32WgAWLFhAXFwcK1as4Prrrz/r/+dUUkIghBBC1ENRUZHL4+QRdE92+eWX88svv7B3b2Xvtr/++ou1a9fSv39/AA4dOkRmZiZ9+/at2sZoNNK7d29+//13ADZv3ozNZnNZJyYmhs6dO1et01ikhEAIIUST0FglBKcOiPfSSy8xadKkaus/++yzFBYW0r59e3Q6HQ6Hg9dee4277roLgMzMTACioly7bUZFRZGSklK1jo+PDyEhIdXWOb59Y5GEQAghRJPQWAlBamqqS7fD0w2Yt3jxYhYsWMDChQvp1KkTW7duZfTo0cTExHDvvfdWrac5pRuvoijVlp2qLuvUlyQEQgghRD0EBgbWaRyCcePG8dxzzzF48GAAkpKSSElJITk5mXvvvZfo6GigshSgWbNmVdtlZ2dXlRpER0djtVrJz893KSXIzs6mV69ejflvSRsCIYQQTcPxEoKGPOqjrKwMrdb1MqvT6aq6HSYkJBAdHc3y5curXrdaraxevbrqYt+9e3cMBoPLOkePHmXHjh2NnhBICYEQQogmQaFhMxbWd1jfG2+8kddee40WLVrQqVMntmzZwvTp03nggQeAyqqC0aNHM2XKFBITE0lMTGTKlCmYzWaGDBkCVI7c++CDDzJ27FjCwsIIDQ3l6aefJikpqarXQWORhEAIIUST4Oluh++++y4TJ05k+PDhZGdnExMTw6OPPsqLL75Ytc4zzzxDeXk5w4cPJz8/nx49erBs2TICAgKq1pkxYwZ6vZ477riD8vJyrrnmGubNm4dO17hza6g6l0FDyVwGdSdzGZzDZC6DOpG5DOqmKc9lcPUPj6H3O/sZc+2lFv43YLbMZSCEEEKcy2Qug9qdFwmBYrejeOEdizfxxrvxW3Y0bh/axvB1xwi1Q6hGo/e+n6k3loJ5I8Umn5M3kYSgdtLLQAghhBDnRwmBEEIIcSZSQlA7SQiEEEI0CYqiQWnARb0h254LpMpACCGEEFJCIIQQomlwomnQwEQN2fZcIAmBEEKIJkHaENROqgyEEEIIISUEQgghmgZpVFg7SQiEEEI0CVJlUDtJCIQQQjQJUkJQO2lDIIQQQggpIRBCCNE0KA2sMjjfSwiafELQuUcJtw/PITGpjLBoO5MeiGfd0iC1w2Lgvbnc/ngOoZE2UvaamP1iDDs2+Htk3517FHP7Y1kkJpUTFm1j0oOtWPdzMAA6vcJ9z2Rw8dWFNGthpbRIx5a1AcxJjiEvy6fRYrCVavj7HTMZK4xY8rQEd7DTZXwJIUmVk8X8PctM2k9GyjN1aA0KwR3tdHyylNCula9bCzT8PcuP7N8NlGfq8Al20uwaCx2fKMMQ4L4Zv9U+njpfUsxtj2VV7j/KxuSHWrNuWXDV63c/lUHvG/OIiLFhs2nYv93MvGmx7Nnq57EYB96Ty4B7jhEVZwUgZY+JT2dEsWmlZ6aTvXN4BpfdkE/z1hVYK7Ts2uzPv6c2J+2gb9U6l92QR/8hObRJKiMo1M7wfp04uMvcpGI6nbBoKw9OOMrFVxfhY3KSftDI9LEt2L/d87HUlwIoDfj5u+/M4R2afJWByezk4E4T7z0fq3YoVXrflM9jkzP47J1Ihvdty44//Hj100NExFo9sn+T2cnBXWbem9i82mtGXydtOpexcGYzRtzQnpcfaUVsqwom//tgo8awZaI/2b/7cNHrxVzzTR6RvaysfTCI8qzKQ9Y/3kHX50u45ps8rvykAHOsg98eDsKSV5nBV+RoqcjR0nlcKVd/k0f3KcVkrfXhz4kBjRrnqdQ+nkxmJ4d2+fL+xLgaX087aOL9F1vwWN+OPP1/7chK9WHKgr0Ehdo8FmPOUQP/ntKMUf3aMqpfW/76zZ9Jcw/Tsm2FR/af1KOY7/8TxVM3d2T83e3Q6RVe+2QvRl9H1TomXyc7N/kz9/Xqv4GmElNN/IPsTP9mHw67hhfubsUjfdrz0cuxlBbpVItJNB5VSwg++OADPvjgAw4fPgxAp06dePHFF+nXr5/HYti0MvCkO5MUj+23Nrc+ksvPn4WydGEYALNfiqV7n2IG3nOMucnN3L7/TSuD2LTy+F3tIZfXyop1jB+S6LLs/YlxvPvDHiJirORkNLyUwFEBGcuNXDqriPCLKi9UHUaWcfQXI4cWmej4ZBlxAy0u2yQ9W0rKl74U7tET2dNGYKKDHm8XVb3u38JJpydL2fRsIE47aN105Kt9PG1aFcSmVacvkVj1bajL849eieOGu46R0KGcrb8Z3B0eAH8sd41v3uvNGHjPMdp3LyVlr8nt+3/h3nYuz6c/ncDiLVtJTCpjx4bKhPGXryunC49qbqm2fVOJqSZ3DM8mN8OHt8a0qFqWlWZULZ76cqJBIyMVnpaqJQTNmzdn6tSpbNq0iU2bNnH11VczaNAgdu7cqWZYqtIbnCR2KWPzatc72c2rA+h4UalKUdXOL8CB00mj3SU4HRoUhwatj2sBndakcOzP6hctpxUOLzFhCHAS1P7088/bSrTo/RW3JQPnGr3BSb8hOZQU6lQpegbQahV6D8rHaHby9ybPVVuczBxQeRdeXOA9d7neGBPApX0L2bvNzPMfHmLxXzt47+c99BtyTO2w6ux4L4OGPM5nqp4ab7zxRpfnr732Gh988AHr16+nU6dO1da3WCxYLCey46KiomrrnOsCQx3o9FCQ6/rVFOToCYk8/cVOLQajkwfGZ7DymxDKShrn5GXwUwi9wMae2WYCWhdjCnOS+oOR/G16/FueKEI9usqHjWMDcVSAKcLJZf8qxBhScy2fpUDD7g/MJNxR3igxnssuuaaA8bMOYfR1kpdtYMLQRIryPXsqiG9fzszv9+NjdFJequXlB+M5ss/9pQPVKTw6MZUdG/xJ2estdeDeGFOlZi2sDByWy1cfR7DonSjadSvj8ZfTsFk1rPgi9MxvILya17QhcDgcLFq0iNLSUnr27FnjOsnJyQQFBVU94uJqric9H5za8EWjwetatOj0ChPeO4RGqzBrQoszb1AP3acWoSiwtE8Y314QzsFPfYkbYEFz0hEbcYmVq7/Ko/fCAqIut7JhTCCWY9UzeFuJhnWPBRHY2kH74WWNGue56K/fAxh+QwfG3NKOzasCmfD+QYLCPNeGACDtgJHh17XlyYGJ/Pc/4Tz99hFaJHqmDcHJRrxyhIT2ZUwd1drj+z4db4zpOI0W9u/wZe7UGA7sNPPjgnB+WhjGgHty1Q6tTo4PTNSQx/lM9YRg+/bt+Pv7YzQaeeyxx/j666/p2LFjjeuOHz+ewsLCqkdqaqqHo3W/ojwdDjuERLiWBgSF28nP8Z6ybp1e4fnZB4luYWX8XYmNVjpwnH8LJ1f+p5AbN+Vww//y6LO4AKcdzM1PlBDozeDf0kloVzsXvlqCRqdw+EvXu0xbqYbfHwlCb1bo8W4hWs9Uk3s1S7mOoykmdm/xZ8Yz8TgcGm4Y7NkTut2mJeOwkX3bzMxNbsahXb7c/FCOR2N4fHIKl16bzzN3tSc3s/F6yDSEN8Z0srxsfbV2Hqn7TUTGeDahPFuK0vDH+Uz1hKBdu3Zs3bqV9evX8/jjj3Pvvfeya9euGtc1Go0EBga6PM43dpuWfdvMXHhlscvyC68sZpdKdaynOp4MxMZbeG5wG4oL3Jeo6M2V1QHWQg3Zv/nQ7OpaelooGpzWExm8rUTDbw8FoTUoXPpeIbpzp+2TR2k0YPBR/0znuRgUhr+cwmU35PPsXe3JSvWGA8MbY6pu10Y/4lq7NmqMbWUhO10y7fOB6recPj4+tGnTBoCLLrqIjRs38vbbb/Phhx96ZP8ms4OYhBMXmeg4K606lVNcoCMnXZ0M/auPwhn3Tip7t/ny9yY/+t99jMhYGz/8J8wj+zeZHcTEn/jRR8dZaNWxjOICPceyDEz88CBtksp48d7WaHUQElF5d1BcoMNua5wcM2utARTwT3BQekTHjjf88Y930PKWCuxlsOdDP5pdbcEU7sRaqOXgZybKs7TEXl8Zt620MhlwVGi46PUi7CUa7CWV720MVdC4qa2W2sdTbd9dUb6Ou0Zlsn55EHnZBgJDHAwclk14tJU1P4S4Pbbj7n/uKBv/F0BOhg++/g76DCqgS68SXhjayiP7H/FqClfdlMfkh9tQXqqrOn5Li3RYLf90aw2yExlrJSyq8rts3qqy7Ul+joH8nMa/+HljTDX56uNIZny7l8Gjsvj1+2DaXVBG/6HHmPmMel0h60OGLq6dRlG8qxDkmmuuIS4ujnnz5p1x3aKiIoKCgujDIPSas/tBdOlZwhtfHqi2fNniEN56qnHrxetj4L253D48m9BIOyl7TMx+KYYdfzRgYCJt3a+AXXoW88bn+6otX7YklAXTm/Gf9TX3Ahl3eyLb1tW9n/8tOzJP+1raT0Z2zfSjPFOLIchJbF8rHZ8sxRCg4LDAxnGB5G/TY83X4hPsJLiznfaPlVUNXJSzwcDa+4JrfO++y4/hF+us8bWvO0bUOf6auON40ujrnrd3ubSYaUv2Vlu+/PMw3pnQgufeOUS7bqUEhtgpLtCz9y8zn73TjL3b6lf6pNjPvoHrU2+lcsHlxYRG2ikr1nHobxNL3ovkz18bNkaExlC3hGtpysYal781NoHlX1R27bvutlzGvnWo2joLZsSwYGbjjzHhyZgUe8OK93tcW8j9zx0lNsFCZqoPX30UyU8Lz/5mxa7YWKV8Q2FhodtKfY9fKzp89iw689mXvjjKLPx91+tujVVNqiYEEyZMoF+/fsTFxVFcXMyiRYuYOnUqS5cu5brrrjvj9o2REDQZ9UgIPKW2hEAtDU0I3KE+CYGnNCQhcJe6JgRNXUMTgsbmyYSg3cLnGpwQ7Bky9bxNCFQ902RlZTFs2DCOHj1KUFAQXbp0qXMyIIQQQojGo2pCMGfOHDV3L4QQoglpaE8B76pgb3zeVxYphBBCuEFlQtCQRoWNGIwXUr3boRBCCCHUJyUEQgghmgTpdlg7SQiEEEI0CQoNGwH+PK8xkCoDIYQQQkgJgRBCiCZCqgxqJwmBEEKIpkHqDGolCYEQQoimoYElBJznJQTShkAIIYQQUkIghBCiaZCRCmsnCYEQQogmQRoV1k4SgqbC6VA7gmq8cWbBgTvz1Q6hmv92ClE7hOo03ndiVGxWtUM4J2j96jfVtbtpFSuUqh2FAEkIhBBCNBWKpmENA6WEQAghhDj3SRuC2kkvAyGEEEJICYEQQogmQgYmqpUkBEIIIZoE6WVQuzolBO+8806d3/CJJ54462CEEEIIoY46JQQzZsyo05tpNBpJCIQQQniv87zYvyHqlBAcOnTI3XEIIYQQbiVVBrU7614GVquVPXv2YLfbGzMeIYQQwj2URnicx+qdEJSVlfHggw9iNpvp1KkTR44cASrbDkydOrXRAxRCCCGE+9U7IRg/fjx//fUXq1atwmQyVS2/9tprWbx4caMGJ4QQQjQeTSM8zl/17nb4zTffsHjxYi699FI0J41n3rFjRw4cONCowQkhhBCNRsYhqFW9E4KcnBwiIyOrLS8tLXVJEM4FA+/JZcA9x4iKq5wUJWWPiU9nRLFpZaBqMd05MovL+hcS18aCtULLrk1m5rzWjLQDpjNv3IRiAujco4Tbh+eQmFRGWLSdSQ/Es25pkNv2Zy+FPe/4kvmLAUuelqAODjo9V0ZwkgOnrfK17DUGytK06P0Vwnva6PBUOabI6mcRRYENj/mTs9bARe+UEH2NzW1xAwy8N5fbH88hNNJGyl4Ts1+MYccGf7fu83Tmr99JdFz1//e7eeG893xzFSLyznPBqe4cmcUDEzL5+uNwZr8U6/b9DRiSyYC7MolqbgEgZZ8vC2fFsenXE5NtxbUu44FxKSRdUoRGo3Bkv5kpT7Qj56jR7fGJxlfvhODiiy/mhx9+YNSoUQBVScDHH39Mz549Gzc6N8s5auDfU5qRcbjy4L3u9jwmzT3MiL5tSdmrzsWuS89Svp8Xzt6tZnR6hfuePcqUzw7ycO92WMp1EtNJTGYnB3eaWLYohBfnpLh9f3+96EfxPh0XTC3DFOEk7b8+rH8ogN7fFaI3KxT+rSPxsXIC2zmwFWnYOdXMxpH+XLGkuNp7HfqP0WOlj71vyuexyRnMmhDLzg1+DBh2jFc/PcTDfdqRk+7jmSBO8kT/dmh1J5Kk+PYVTF10gDX/dV8ydybeeC44WduuZfS/O4+DOz0XS26mD3PfbElGSuU+r70lmxc/2M3IQV05st9MsxYVvPnZDn7+IpIF78RRWqwnrnUZVosX3xhKCUGt6p0QJCcnc8MNN7Br1y7sdjtvv/02O3fuZN26daxevfqsA0lOTmbChAk8+eSTzJw586zfpz7+WO56Apr3ejMG3nOM9t1LVTsJPD+0lcvzt55qwZIdO0nsUs6OP9S5o/PGmAA2rQw86Q7OvQmBowIylxu46N0Swi6q7FnTbkQFWb8YSFlkpP2TFVz6rxKXbTpPKGPt4EDKMzT4xpw4kxTt1nHwPyYuX1TEij7Bbo0b4NZHcvn5s1CWLgwDYPZLsXTvU8zAe44xN7mZ2/d/qsI819POnSOzyDjkw7Z16h1L3nguOM5kdvDsrBRmjmvOXU9meWy/f/wv1OX5/BktGTAki/YXFHNkv5l7n0ph4+oQ/j0tvmqdzFT1k6dayWyHtap3o8JevXrx22+/UVZWRuvWrVm2bBlRUVGsW7eO7t27n1UQGzdu5KOPPqJLly5ntX1j0GoVeg/Kx2h28vcm75kv3C/QAUBxgXp34qfyxpjcTXGA4tCgO6UkVGuCvC0159W2Eg1oFPSBJ5IBRzn8Oc6Pzs+XYYpw/+2G3uAksUsZm1cHuCzfvDqAjhepPwm93uDk6lvz+XlxGN7SYMvbzgUjp6Sz4ZdAtqwJOPPKbqLVKvQekIvJ7GD31gA0GoWL++STftjEq//exWfrNzDji230vPaYajGKhjuruQySkpKYP39+owRQUlLC0KFD+fjjj3n11VdrXddisWCxWKqeFxUVNXj/8e3Lmfn9fnyMTspLtbz8YDxH9nlLlqvwyKQMdvzhR8oeX7WD+Yc3xuR+ej8IucDO3tkm/FuVYgxTSP/Rh4JtOvxaOqut77DA7hm+xA6wYjjpxnfn62ZCutmJvtq9bQaOCwx1oNNDQa7rT70gR09IpPpjiPS6oRD/QAfLloSeeWU388ZzQe9B+bRJKmdU/0RV9h/ftpTpS7ZXfiZlOl4Z3p4j+82EhFsx+zu545F05s9owb/faEn3K/J54b09PDesE9s3qFf9UxuZ/rh2ZzUwkcPh4IsvvuCVV17h1Vdf5csvvzzrAYpGjBjBgAEDuPbaa8+4bnJyMkFBQVWPuLi4s9rnydIOGBl+XVueHJjIf/8TztNvH6FFYkWD37cxjJiSTkKHcpKHt1A7lCreGJOnXJBcCgqsuCqYH7sFc2iBkdgBVjSn/IqcNvjzaT8UJ3SeWFa1PPN/BnL/0NPp2TI87dQTmUaDV9SHXj84j40rA8nLMqgditedCyJirDz+cgbTRrXAZlFnpvq0Q76MuKkrT93ehR8WRjN22j5atCmrOubX/RLKN/NiOPi3H59/1JwNK0Pof5fnqjXqTYWBidLT07n77rsJCwvDbDZzwQUXsHnz5hMhKQqTJk0iJiYGX19f+vTpw86dO13ew2KxMGrUKMLDw/Hz8+Omm24iLS2t/sGcQb1LCHbs2MGgQYPIzMykXbt2AOzdu5eIiAi+++47kpKS6vxeixYt4s8//2Tjxo11Wn/8+PGMGTOm6nlRUVGDkwK7TVvVkGjfNjPtLijj5odyeOfZhicbDTH81TR69i1i7C2tyT3q+YZfNfHGmDzJr4WTXvNLsJeBvVSDKUJh81g/zM1PlBA4bbB5rB9laVp6zi1xKR049oeeslQtP/cMdnnfTaP9CO1up9c81zYIjaEoT4fDDiERrgl7ULid/Bx1JzuNjLXS7YpiXnkoQdU4jvO2c0GbLuWERNiZtXRv1TKdHpIuLeWm+3MZGN8Fp9O91Sx2m5ajRypLAvft8KdtUgmD7j3KBy8nYLdpOLLftZQw9YAvHbtXb0TbVOXn53PZZZdx1VVX8dNPPxEZGcmBAwcIDg6uWmfatGlMnz6defPm0bZtW1599VWuu+469uzZQ0BAZTXR6NGj+f7771m0aBFhYWGMHTuWgQMHsnnzZnS6xqu6rfcZ4aGHHqJTp05s2rSJkJDK7if5+fncd999PPLII6xbt65O75OamsqTTz7JsmXLXAY4qo3RaMRodH93FoOPmrdOCiNeS6fXDYWMu60NWane0H3HG2NSj94MerOCtVBDzm96OowpB05KBlJ0XDq3GJ9g1+Oo9UMVxN1mcVn2681BdHq2nKg+7qlCsNu07Ntm5sIri/n9pC6ZF15ZzLqf1S3W7XvnMQpy9fzxi/d07TuVmueCrWv8eeSqti7Lxs5IJXW/iSXvRbg9GaiJRgMGHyd2m5a92/1pnuBaghIbX0F2hhefHxqpUeGp1dWnuza9/vrrxMXFMXfu3Kpl8fHxJ95OUZg5cybPP/88t956KwDz588nKiqKhQsX8uijj1JYWMicOXP45JNPqkrSFyxYQFxcHCtWrOD6668/+//nFPUuh/rrr79ITk6uSgYAQkJCeO2119i6dWud32fz5s1kZ2fTvXt39Ho9er2e1atX884776DX63E4HPUNrd7uf+4onS8pIaq5lfj25dz37FG69Cph5dchZ97YTUZOSefqW/OZOqIl5SVaQiJshETY8DFVr6duyjFBZevrVp3KadWp8oIcHWelVadyImKtbtlf9lo92Wv0lKVpyfldz/r7/fGPdxJ3ixWnHTY/5UfhTj3dXi9FcUBFjoaKHA3Of8IxRSgEJjpdHgC+zZwupQyN7auPwrlhSB59Bx8jrk0Fj05KJzLWxg//CXPbPs9Eo1Hoe2ceKz4PxelQvzGhN54Lykt1pOzxdXlUlGkpztd5pP3OvWNS6HRREZGxFcS3LeXep1JI6lHIyu8iAPjyXzFc2T+XG+7IolmLcm68+yg9rs7jh0+j3R7b2dIoDX8AxMXFuVRfJycn17i/7777josuuojbb7+dyMhIunXrxscff1z1+qFDh8jMzKRv375Vy4xGI7179+b3338HKq+VNpvNZZ2YmBg6d+5ctU5jqXcJQbt27cjKyqJTp04uy7Ozs2nTpk2d3+eaa65h+/btLsvuv/9+2rdvz7PPPtuoxSCnExxhZ9y7RwiNtFNWrOPQ3yZeGNqKP39VrzXvjfdVttJ98yvXUR/fHB3HcpUaXnljTABtu5bzxpcnYnpscgYAyxaH8NZTjd/GwV6iYfdMXyoytRiCFKKvs9L+yXK0BihL15K1srIa5df/c73jvXRuMeGXqNeAb/V3IQSEOBj6VBahkXZS9ph44e4EslUYg+C4blcUE9Xcxs+L1W9MCN55LlBbSLiNcW/sIzTSSmmxjkO7/Zj4YEe2/BYMwO/Lw5j1UivueDSdxyYeIu2QiVdHtmfnZu8t8WmscQhSU1MJDDzxf56u5PrgwYN88MEHjBkzhgkTJrBhwwaeeOIJjEYj99xzD5mZmQBERUW5bBcVFUVKSmVX6szMTHx8fFxuwo+vc3z7xlKnhODk4pEpU6bwxBNPMGnSJC699FIA1q9fz8svv8zrr79e5x0HBATQuXNnl2V+fn6EhYVVW+4uM8aq206gJtfHdFU7hGq8MSaAbev8PRpbzA02Ym6ouWjfHOtk4M78er/n2WxzNv47P5z/zg/3yL7q4s9fA7k+9gK1w6jijeeCmjxzW91vuhpq5oQz72vZF1Es+yLqjOudbwIDA10SgtNxOp1cdNFFTJkyBYBu3bqxc+dOPvjgA+65556q9U4d5VdRlDOO/FuXdeqrTglBcHCwy44VReGOO+6oWqb804T5xhtv9EhRvxBCCFFvHh6YqFmzZnTs2NFlWYcOHfjyyy8BiI6urF7JzMykWbMTg4RlZ2dXlRpER0djtVrJz893KSXIzs6mV69eZ/VvnE6dEoKVK1c26k5PZ9WqVR7ZjxBCiCbIw0MXX3bZZezZs8dl2d69e2nZsiUACQkJREdHs3z5crp16waA1Wpl9erVVSXu3bt3x2AwsHz5cu644w4Ajh49yo4dO5g2bVoD/pnq6pQQ9O7du1F3KoQQQpzvnnrqKXr16sWUKVO444472LBhAx999BEfffQRUFlVMHr0aKZMmUJiYiKJiYlMmTIFs9nMkCFDAAgKCuLBBx9k7NixhIWFERoaytNPP01SUlKdxu+pj7PuiFxWVsaRI0ewWl1bdKs5/LAQQghxWh4uIbj44ov5+uuvGT9+PC+//DIJCQnMnDmToUOHVq3zzDPPUF5ezvDhw8nPz6dHjx4sW7asagwCgBkzZqDX67njjjsoLy/nmmuuYd68eY3e+F6jKPUbjDEnJ4f777+fn376qcbXPdmGoKioiKCgIPowCL1G/ZHOxLnPU4386uO/ndTr+nZa3jjV+fk+rmwj0fqpPz/DyeyKlf+VfkZhYWGdGuqdjePXirg3X0Hre/bDUTvLK0h9eqJbY1VTvcchGD16NPn5+axfvx5fX1+WLl3K/PnzSUxM5LvvvnNHjEIIIYRws3pXGfzvf//j22+/5eKLL0ar1dKyZUuuu+46AgMDSU5OZsCAAe6IUwghhGgYmf64VvUuISgtLSUyMhKA0NBQcnJygMoZEP/888/GjU4IIYRoJI01UuH5qt4JQbt27aq6UVxwwQV8+OGHpKenM3v2bJd+lEIIIYQ4d9S7ymD06NEcPXoUgJdeeonrr7+eTz/9FB8fH+bNm9fY8QkhhBCNw8O9DM419U4ITu4u0a1bNw4fPszu3btp0aIF4eHeMzSqEEIIIequwROim81mLrzwwsaIRQghhHAbDQ1rB3B+NymsY0IwZsyYOr/h9OnTzzoYIYQQQqijTgnBli1b6vRmjT3zUp1pdaBx/3TJdeaUCZ7qQnOaKUPV9N8k76v28vvVO6YIPlnZ1d43gJPihROraTwwjXt9OUtL1Q7BhVOpeQZRt5Buh7XyqsmNhBBCCLeRRoW1qne3QyGEEEKcfxrcqFAIIYQ4J0gJQa0kIRBCCNEkNHS0QRmpUAghhBDnPSkhEEII0TRIlUGtzqqE4JNPPuGyyy4jJiaGlJQUAGbOnMm3337bqMEJIYQQjUZphMd5rN4JwQcffMCYMWPo378/BQUFOP7p+xscHMzMmTMbOz4hhBBCeEC9E4J3332Xjz/+mOeffx7dSYNuXHTRRWzfvr1RgxNCCCEai0x/XLt6tyE4dOgQ3bp1q7bcaDRS6mUjYAkhhBBVZKTCWtW7hCAhIYGtW7dWW/7TTz/RsWPHxohJCCGEaHzShqBW9S4hGDduHCNGjKCiogJFUdiwYQOfffYZycnJ/Otf/3JHjEIIIYRws3onBPfffz92u51nnnmGsrIyhgwZQmxsLG+//TaDBw92R4yNpnOPYm5/LIvEpHLCom1MerAV634Ornr9sn759B+aS2KXMoJCHTzetz0Hd5k9GuOdI7O4rH8hcW0sWCu07NpkZs5rzUg7YPJoHLW5c2QWD0zI5OuPw5n9Uqxn9vl4Bpddn0/z1uWVn8uf/vz79TjSDvpWrTP2jYNcd1uuy3Z/b/HjqVs7uSWmMx1PoHD3mKP0H3IM/2A7u7f48d7zcaTs9T3dW9aLYlewzS3DvrwCJc+JJkyLvp8Jwz1mNNrKok1F+Wed7ytQip1oOxowPuWPNqHyp68UObH+uxTHRhtKtgNNkBbdFUZ8HjSj8W+cYUo6X1LMbY9lkZhURliUjckPtWbdsuCq1+9+KoPeN+YREWPDZtOwf7uZedNi2bPVr1H2Xxfz1+8kOq76JDvfzQvnveebeySGM31OJ3siOYX+Q3OZPbk538yJ8kh8NVHjXNAQMjBR7c5qHIKHH36Yhx9+mNzcXJxOJ5GRkY0dl1uYzE4O7jKzbEkYL358qMbXd23yZ80PITz1xhEVIoQuPUv5fl44e7ea0ekV7nv2KFM+O8jDvdthKVd/5rS2Xcvof3ceB3d6NkFJ6lHM959EsnebH1o93Dc2ldf+s4dHrkty+Vw2rgpi+riEquc2m/vG3jrT8XTH8CxufTibt8a0JO2giSFPZJK8cD8P9u5IeWnDv0vbwjJs35VjnBCANl6Pc48dS3IxGj8NhtvN/6xTjm1JOcbxAWjjdFj/U0bFmEJ8Pw1BY9ai5DpRcp34DPdDG69HyXRgeasES64D0ytBDY4RKj+nQ7t8Wb4kjIkfHaz2etpBE++/2IKjR4wYTU5ueTCLKQv28sCVnSnMMzRKDGfyRP92aHUnzvbx7SuYuugAa/7bOJ9BXZzpczquZ98C2l1QSm6mZz6b01HrXNAgMg5BrRo0MFF4eMOmip00aRKTJ092WRYVFUVmZmaD3vd0Nq0MYtPK4z/w6ifwX74Mq4yhucUt+6+L54e2cnn+1lMtWLJjJ4ldytnxh79KUVUymR08OyuFmeOac9eTWR7d9wv3tXN5Pv2ZVizevIXEpFJ2bAisWm6zasjP9fFITLUfTwo3P5jNonej+e2nEADefKoli7Zs56qb8/jx04gG79+5047+MiP6npXTSGub6bCvqMCxx46BytIB++flGIaZ0feuXMc4IYCym49hX27BMMgXbSs9pldPuujF6vB52A/Lq0UodgWNvuGNqDatCmLTqtNfWFd96zq980evxHHDXcdI6FDO1t88c9ErzHM9Fd45MouMQz5sW+e539yZPieAsCgrw185wgvDEnl57n4PRVadmucC4T71TggSEhLQaE5/kjh48PSZbU06derEihUrqp7rvHD+cDX5BVaO81BcoP7nMnJKOht+CWTLmgDVTwLmgOOfi+sh3OXSYhZt/JOSYh3b/whk3pvNKTzm+Tup6BZWwqLsbF59crKiZft6fzpeVNooCYE2yYD9u3KcqXa0cXoc++04ttswjqq8iClHnSh5TnQXn/j/NT4adF0NOHfYYFDNVRdKqRPMmkZJBupLb3DSb0gOJYU6j1fXnRzD1bfm89VHkYD3tCrXaBTGzTzMFx9GNVq109nypnNBvTS066CUELgaPXq0y3ObzcaWLVtYunQp48aNq38Aej3R0dF1WtdisWCxnLh7Lyoqqvf+zi0Kj0zKYMcffqTsUfcE0HtQPm2SyhnVP1HVOCopPPrCEXZs9Cdl74mLxsZVQaz5MZSsdB+i4yzcMyad1z/dzaibOmGzenbajtCIyvro/FzXn1h+rp7IWGuj7MMw1BdKnZTfnV/ZX8gJhof90F9bWYSrHHMCoAl1/d81oVqUTEeN76kUOrHNL8Nwk2ePt0uuKWD8rEMYfZ3kZRuYMDSRonx1RlbvdUMh/oEOli0JPfPKHnTH8EwcDvj23+pW0XrXuaCepMqgVvX+xT355JM1Ln/vvffYtGlTvQPYt28fMTExGI1GevTowZQpU2jVqlWN6yYnJ1erYjifjZiSTkKHcsbe3EbVOCJirDz+cgYT7mqFzaL+fFgjXk4hoX0ZY2937eb66w9hVX+n7DWzb5sf89f+xSVXFfDbzyqd3E/pt1xZuNY4d52O/1mwL7NgfLGyDYFjvx3ruyVowrQY+tVSr6vUHIJS6qTi2UK08ToM93v27vyv3wMYfkMHgkLt9LsrlwnvH+TJQe1VKd25fnAeG1cGkpelbh39ydoklTLo/mxGDuiAmqUW3nYuEI2r0b7Rfv368eWXX9Zrmx49evCf//yHn3/+mY8//pjMzEx69erFsWPHalx//PjxFBYWVj1SU1MbI3SvNPzVNHr2LeKZ21qTe9QzdeKn06ZLOSERdmYt3cuPR/7ixyN/0bVXKYMezOXHI3+h1XoubX580mEuvaaAZ+7qQG5m7Z9LXo4P2ek+xMRXeCi6k/ddeTEJiXBtuR4cZic/p3HufK3vl2IYakZ/jQltaz2G600YbvfF9mkZAJqwyp+3kud02U7Jd6IJcf3pK2VOKp4uBF8NxleDPF5dYCnXcTTFxO4t/sx4Jh6HQ8MNg3PPvGEji4y10u2KYpYuDDvzyh7U+ZISgsPtfLJuOz8c3MwPBzcTFWfl4RfSmP+b50aI9aZzwVmRcQhq1Whlcl988QWhofW7C+vXr1/V30lJSfTs2ZPWrVszf/58xowZU219o9GI0WhscKzeTWHEa+n0uqGQcbe1IStV/f936xp/HrmqrcuysTNSSd1vYsl7ETidnrh4KAyfnEKvvvk8c1cHstLO/LkEBNuIiLGSl+P5hCrziA/HsvRceGURB3ZW3m3rDU6SLi1hzpSYRtmHYlGqp/Q6Dfxz/dc006IJ1eLYZEPXtjJBUWwKjr9s+Dx6okufUvpPMmDQYEoOQmNUv95cowGDj+fPvn3vPEZBrp4/fgk888oe9MuXYWxZ4xrTawv28ctXoSxf0rDG3fXhHeeCsyfdDmtX74SgW7duLo0KFUUhMzOTnJwc3n///QYF4+fnR1JSEvv27WvQ+5yOyewgJv5EG4ToOAutOpZRXKAnJ8OHgGA7ETFWwqIr7+riWlfeWebnGMjP8Uzx4cgp6Vx1Sz6T7k+gvERbdYdZWqzDWqFOEV15qa5aG4aKMi3F+dWXu8uIl1O4atAxJj+SWPm5hFfWw5cW67FatJjMDu4enc5vP4WQl+1DVHML941LozBPz+8/h7glpjMdT9/MiWTwyCzSD5lIP2TkrlGZWMq1rPymcaov9L18sH1ShiZKW9ntcJ8d2+IyDP0rqws0Gg36232xLShD21yHtrkO64IyNEYN+usqEyqlzEnF2EKUCgXTC4EopQpKaeVZTxOsQaNr+Am+ts+pKF/HXaMyWb88iLxsA4EhDgYOyyY82sqaH9zzvZ2ORqPQ9848VnweitPh+QvbmY6nUxvQOmwa8nMMpB30XLc/bzgXCPepd0Jw8803uzzXarVERETQp08f2rdv36BgLBYLf//9N1dccUWD3ud02nYt443PTyQbj01KB2DZklDeGhPPpdcV8vSMlKrXJ3xwGIBPpkezYHrj3NWdyY33VVaXvPnVAZflb46OY7mXNXLypBuHZQPwxqLdLsvfejqB5V9G4HRoSGhXxrW35OIX6CAvx8C2dYFMGdW6Ufr81+RMx9OS96PwMTkZ+doRAoIc7N7qx/ihbRotHp/R/lj/VYZ1ekllNUC4FsNNvhjuO1H/bxjiCxalcp0SJ9oOBkxvBaExVyaXzj12nLvsAJTflefy/r6LQ9E0a3isbbuUMW3J3qrnj76UBsDyz8N4Z0IL4lpXcO1txwgMsVNcoGfvX2aevq2dx1vSd7uimKjmNn5erM7vrLbP6a2x8arEJJoWjaIodS4EsdvtfPrpp1x//fV17hlQm6effpobb7yRFi1akJ2dzauvvsrq1avZvn07LVu2POP2RUVFBAUF0Ud7K3qN9zQAwllzC27hSuOF1T+Kza52CNX4rfK+RLDs6ny1Q6hGcXjf707jhd2oFbt3HeN2xcYqvqWwsJDAQPdU1Ry/VrQePwWd6exLVBwVFRxInuDWWNVUrxICvV7P448/zt9//90oO09LS+Ouu+4iNzeXiIgILr30UtavX1+nZEAIIYSoD2lDULt6Vxn06NGDLVu2NMpFe9GiRQ1+DyGEEEI0XL0TguHDhzN27FjS0tLo3r07fn6uE5B06dKl0YITQgghGtV5fpffEHVOCB544AFmzpzJnXfeCcATTzxR9ZpGo0FRFDQaDQ4vrMcTQgghZKTC2tU5IZg/fz5Tp07l0KHqkwIJIYQQ4txW54TgeGcEafAnhBDiXCSNCmtXrzYEtc1yKIQQQng1qTKoVb0SgrZt254xKcjLy6v1dSGEEEJ4n3olBJMnTyYoKMhdsQghhBBuI1UGtatXQjB48GAiI9Wdi1sIIYQ4K1JlUKs6z5Yj7QeEEEKI81e9exkIIYQQ5yQpIahVnRMCp9PpzjiEEEIIt5I2BLWr99DFQjQWxWI580oepjH4qB1CNaV9vK/nTu+tJWqHUM3qLp6dLrkuvHEGxiZNSghqVec2BEIIIYQ4f0kJgRBCiKZBSghqJQmBEEKIJkHaENROqgyEEEIIISUEQgghmgipMqiVJARCCCGaBKkyqJ1UGQghhBBCSgiEEEI0EVJlUCtJCIQQQjQNkhDUSqoMhBBCCCEJgRBCiKZB0wiPs5WcnIxGo2H06NFVyxRFYdKkScTExODr60ufPn3YuXOny3YWi4VRo0YRHh6On58fN910E2lpaQ2I5PQkIRBCCNE0KI3wOAsbN27ko48+okuXLi7Lp02bxvTp05k1axYbN24kOjqa6667juLi4qp1Ro8ezddff82iRYtYu3YtJSUlDBw4EIcb5sloUm0IOvco5vbHskhMKics2sakB1ux7ufgk9ZQuHvMUfoPOYZ/sJ3dW/x47/k4UvaqN2nKnSOzeGBCJl9/HM7sl2JViWHgPbkMuOcYUXFWAFL2mPh0RhSbVgaqEs9xnXuUcPvwHBKTygiLtjPpgXjWLQ3y2P7vHJ7BZTfk07x1BdYKLbs2+/Pvqc1JO3jy8aJw9+gM+g3JwT/Izp4t/rw3sSUp+zx3TPn6Obh3XAa9bigkONzGgR1mPnipOXv/8nPL/uylcHiWgdz/abHlafBv76T1szYCO1eeTRUFUj7Qc/RLPfYiCEhykjjBhl+bE2fbjC90ZP+oo+RvLY5SDZetLUfvxsPtzpFZXNa/kLg2lsrvcpOZOa81I+2AyX07PQOtTmHY2EyuviWfkAgbedkGli8JZeHbUShKQ+5VG27gvbnc/ngOoZE2UvaamP1iDDs2+KsaU100VrfDoqIil+VGoxGj0VjjNiUlJQwdOpSPP/6YV199tWq5oijMnDmT559/nltvvRWA+fPnExUVxcKFC3n00UcpLCxkzpw5fPLJJ1x77bUALFiwgLi4OFasWMH1119/9v9MDZpUCYHJ7OTgLjPvTWxe4+t3DM/i1oezeW9ic0YNaE9+toHkhfvx9VNnxrK2Xcvof3ceB3eqd1ICyDlq4N9TmjGqX1tG9WvLX7/5M2nuYVq2rVA1LpPZycGdJt57Xp1EKalHMd//J4qnbu7I+LvbodMrvPbJXoy+J46X2x/L5JaHMnn/xRY8cWNH8nIMTPl0j0ePqafeSOHCK4qZ9mRLHru2A5t/DWDqZ/sIi7a6ZX97JxnIX6+l/Ws2LvrSQkhPJ9seMWLJqnw9da6etE/0tBlv5cKFFnzCFbY9asReeuI9nOUQepmTFg/Z3RLjqbr0LOX7eeGMHpjI+MGt0OkUpnx20OW79LQ7R2QxYFgu770Qy8N92vOv12K47fFsBj2Qq1pMAL1vyuexyRl89k4kw/u2Zccffrz66SEiYt1zPHmjuLg4goKCqh7JycmnXXfEiBEMGDCg6oJ+3KFDh8jMzKRv375Vy4xGI7179+b3338HYPPmzdhsNpd1YmJi6Ny5c9U6jUn1hCA9PZ27776bsLAwzGYzF1xwAZs3b3bLvjatDGL+GzH89lNIDa8q3PxgNovejea3n0JI2ePLm0+1xOjr5KqbPT/9rMns4NlZKcwc15ziQp3H93+yP5YHsfF/gaQfNJJ+0Mi815tRUaqlfffSM2/sRptWBjJ/WjN++ylYlf2/cG87ln8RTso+Xw79bWb60wlENbeSmFT2zxoKtzyYxaJZMfy2NJSUvWbeGpuA0eTkqkHHPBKjj8nJ5f0L+Ndrsez4I4CMwyYWTI8hM9XIwGGNf2FxVEDOCh2tnrIRfJET3xYK8cPtmGIVMpboURRIX6CnxcN2Iq514peo0P5VG44KyP7xxHHefJiDFg/aCezibPQYa/L80FYsXxJKyl4TB3f58tZTLYhqbiOxS7lH9l+TDt3LWPdzEBt+CSIrzcjaH4L5c3UAiV3LzryxG936SC4/fxbK0oVhpO43MfulWHIyDAy8xzPHdIM0UpVBamoqhYWFVY/x48fXuLtFixbx559/1pgwZGZmAhAVFeWyPCoqquq1zMxMfHx8CAkJOe06jUnVhCA/P5/LLrsMg8HATz/9xK5du3jrrbcIDg72eCzRLayERdnZvPpEuaTNqmX7en86XuT5C9/IKels+CWQLWsCPL7v2mi1Cr0H5WM0O/l7k3uKnM9V5oDKu8nigsoLW3SchdBIG3+uOeWY+iOADt1LPBKTTqeg04PV4lrEbKnQ0umSxo9BcQAODVof1+VaIxRu0VKRrsGaqyGk54k7b60PBHd3UrRV9fuTKn6Brt+lGnZs8OOCy4uJbVVZEteqYzmdLill4y/qVdXpDU4Su5SxebXreWnz6gBVzpNnpRHaDwQGBro8aqouSE1N5cknn2TBggWYTKcv5dVoXH+biqJUW1btX6jDOmdD1TYEr7/+OnFxccydO7dqWXx8/GnXt1gsWCyWquen1uM0RGiEDYD8XNePJD9XT6SHi8J6D8qnTVI5o/onenS/tYlvX87M7/fjY3RSXqrl5QfjObJP3aoM76Lw6MRUdmzwJ2WvGYCQyH+OqRyDy5r5uQaiYi3V3sEdykt17Nrkx5DRmRzZb6Igx0Cfm/No362U9EM113k2hN4PArs6SPlIj7mVFZ8wyP5JR/F2Db4twJpbeRLzCXOtyPUJU6g4qm69+AkKj0zKYMcffqTsUa/90JL3IvELcPCv1btxOkCrg3mvN2PVtzWVcHpGYKgDnR4KTjlPFuToCYn0TPXOuWLz5s1kZ2fTvXv3qmUOh4Nff/2VWbNmsWfPHqCyFKBZs2ZV62RnZ1eVGkRHR2O1WsnPz3cpJcjOzqZXr16NHrOqKfl3333HRRddxO23305kZCTdunXj448/Pu36ycnJLvU2cXFxjR/UKY11KpMwz52oImKsPP5yBtNGtcBm8Z47prQDRoZf15YnByby3/+E8/TbR2iRqG4bAm8y4pUjJLQvY+qo1mdcV6NRUDw4wMm0J+PRaOCzzTv478Et3PxADiu/CcHpcM9x3X6KDRRYf60vv15kIn2hnsj+DjQn32yfsmtPfh5nMmJKOgkdykke3kLVOHrfVMA1/5fP1BEtGXFDO94c3YLbHsvm2ts9X4V5qlO/L42Gc2LQnuONChvyqKtrrrmG7du3s3Xr1qrHRRddxNChQ9m6dSutWrUiOjqa5cuXV21jtVpZvXp11cW+e/fuGAwGl3WOHj3Kjh073JIQqFpCcPDgQT744APGjBnDhAkT2LBhA0888QRGo5F77rmn2vrjx49nzJgxVc+LiooaLSnI++cu7nhr3uOCw+zk53juY2rTpZyQCDuzlu6tWqbTQ9Klpdx0fy4D47vgdHr+Tspu05JxuPKOct82M+0uKOPmh3J451k3JGXnmMcnp3Dptfk8fUcHcjNPlJXnZ598TJ1YHhxmJz/XUO193OVoipFxt7XF6OvAL8BJXraBCe8fJDPV58wbnwXfOIUL5lpxlFX2ODBGwK5xBkyxCj7hlWdUa64GY8SJs6stT1Ot1EANw19No2ffIsbe0prco+75fOrq4YkZLJ4VyervKu8MD+/2JbK5lcEjs1jxeagqMRXl6XDYISTCtTQgKNyz58mz5sGRCgMCAujcubPLMj8/P8LCwqqWjx49milTppCYmEhiYiJTpkzBbDYzZMgQAIKCgnjwwQcZO3YsYWFhhIaG8vTTT5OUlFStkWJjUPUbdDqdXHTRRUyZMgWAbt26sXPnTj744IMaE4LaunY0VOYRH45l6bnwyiIO7Kws8tUbnCRdWsKcKTFu2WdNtq7x55Gr2rosGzsjldT9Jpa8F6FKMnA6Bh/1T+DqUhj+8hF6XZ/PM3e2JyvV9djMTDWSl22g2+VFHNhZ2d5Cb3CS1KOYf0+tuaeLO1nKdVjKdfgH2eneu5h/TXFv7wydufJhK4K83ysbGh5PCvLX6QjoUHlRcdqgYLOWVqNtbo2ndgojXkun1w2FjLutTbXvUg1GX2e17oVOhwaNigWHdpuWfdvMXHhlMb+f1MX3wiuLWfez57r8ni+eeeYZysvLGT58OPn5+fTo0YNly5YREHCijcaMGTPQ6/XccccdlJeXc8011zBv3jx0usZv36JqQtCsWTM6duzosqxDhw58+eWXbtmfyewgJv5E3W10nIVWHcsoLtCTk+HDN3MiGTwyi/RDJtIPGblrVCaWci0rv/FcNl5eqqtWb1lRpqU4v/pyT7n/uaNs/F8AORk++Po76DOogC69SnhhaCtV4jnOZHYQk3CifUd0nJVWncopLtCRk+7+u7sRr6Zw1U15TH64DeWlOkL+aYdSWqTDatECGr6eE8XgEUfJOFx5TA0eeRRLhZaV34a5Pb7juvcuQqNRSD1gIjbewkMvpJN20Miyxe6JIe83LSjgG69Qnqrh4HQD5pYK0YMcaDQQe7edI3P0mFtW9kI48i89OhNE9j/R0NCaW1mKUH6k8oJYsk+L3k/B2EzB4Ibrzsgp6Vx1Sz6T7k+gvER74rss1mGtUOcKvH55IIOfyCI73UDKHhOtO5dz6yPZLFvkuWOnJl99FM64d1LZu82Xvzf50f/uY0TG2vjhP+rGVRdqT3+8atUq1/fTaJg0aRKTJk067TYmk4l3332Xd999t2E7rwNVE4LLLrusqmHFcXv37qVly5Zu2V/brmW88fm+quePTUoHYNmSUN4aE8+S96PwMTkZ+doRAoIc7N7qx/ihlSf7piw4ws64d48QGmmnrFjHob9NvDC0FX/+qm4PiLZdy3njywNVzx+bnAHAssUhvPWU++t/bxyWA8AbS1yP4bfGJrD8i3AAPp8djdHkZOSrKfgH2tm91Z8Jd7f16DHlF+Dg/ufSCW9mo7hAx28/hTD39RgcdveUNtlLNBx6W48lS4MhCMKvdZAwyob2n1qSuPvtOCtg32s+2IogMMlJl9kW9Cd1WslYoidl9olqlb/ur7xjb/eKlehBjT82wI33VXaZe/OrAy7L3xwdx/Il6hTPv/9Cc+595igjp6QRHGbnWJaBHxeE8+mMqDNv7EarvwshIMTB0KeyCI20k7LHxAt3J5DtgSS8wWRyo1ppFEW95jwbN26kV69eTJ48mTvuuIMNGzbw8MMP89FHHzF06NAzbl9UVERQUBB9tLei13iuTvaMnOoNZiIaRmPwvpOa4oYhShuq91bPdJusj9Vd1OsRcFpu6BrWYN7UghOwKzZW8S2FhYUEBrqnS+Xxa0XSg1PQ+Zx97yiHtYLtcya4NVY1qdqM/eKLL+brr7/ms88+o3PnzrzyyivMnDmzTsmAEEIIUR+e7GVwLlK9WejAgQMZOHCg2mEIIYQ430mVQa1UTwiEEEIIj5CEoFbeM/KNEEIIIVQjJQRCCCGaBLW7HXo7SQiEEEI0DVJlUCupMhBCCCGElBAIIYRoGjSKgqYB4zA0ZNtzgSQEQgghmgapMqiVVBkIIYQQQkoIhBBCNA3Sy6B2khAIIYRoGqTKoFZSZSCEEEKI86SEwOkAjRflNl44w5lG531TOCt2u9ohVKPYbWqHUJ0Xtmz2xpkFx+7fqXYI1bzVppPaIVTndecnjcfuvKXKoHbnR0IghBBCnIlUGdRKEgIhhBBNgpQQ1M6LytmFEEIIoRYpIRBCCNE0SJVBrSQhEEII0WSc78X+DSFVBkIIIYSQEgIhhBBNhKI0rBuvF3YBbkySEAghhGgSpJdB7aTKQAghhBBSQiCEEKKJkF4GtZKEQAghRJOgcVY+GrL9+UyqDIQQQgghJQSde5Rw+/AcEpPKCIu2M+mBeNYtDVItHq1OYdjYTK6+JZ+QCBt52QaWLwll4dtRKIpnJiXpfEkxtz2WVfmZRNmY/FBr1i0LrnHdJ5JT6D80l9mTm/PNnCiPxFeTO0dm8cCETL7+OJzZL8WqEsPdY44ybGyWy7K8bD13deusSjwnG3hvLrc/nkNopI2UvSZmvxjDjg3+TSYma4mW32ZGsm9ZAOXH9ER0rODqiUeJ7lJR+XqpljVvRLJ/eSAVBToCm9vods8xLhiaX/UeBSkGVk+NJn2TGYdVQ/yVJVz90lH8wh1uifk4b/ru5q/fSXRc9QnAvpsXznvPN1chonqSKoNaNfmEwGR2cnCniWWLQnhxTora4XDniCwGDMvlzdEtSNljIrFrOWOnH6G0WMc3cyI8EoPJ7OTQLl+WLwlj4kcHT7tez74FtLuglNxMg0fiOp22Xcvof3ceB3eaVI0D4PBuE88Nbl313OlQf2a53jfl89jkDGZNiGXnBj8GDDvGq58e4uE+7chJ92kSMf08IYZje430fzMdv0g7f38bxOf3xHPf0v0ERNtZ9Vo0qevN9H8rjcDmNlLW+rPipWb4R9ppc10xtjINX9wXT0SHCm5fcBiA32ZE8s0jLRjyxSG3Tbbqbd/dE/3bodWduCrGt69g6qIDrPmvejdR9SG9DGqnapVBfHw8Go2m2mPEiBEei2HTykDmT2vGbz8Fe2yftenQvYx1Pwex4ZcgstKMrP0hmD9XB5DYtcxjMWxaFcT8N2P5bWnIadcJi7Iy/JUjTHsyAYdNvYueyezg2VkpzBzXnOJC9ad4djggP8dQ9SjMUz/nvvWRXH7+LJSlC8NI3W9i9kux5GQYGHjPsSYRk61Cw76fA7ny2SyaX1JGSLyVXk/mEBRn5a+FoQBkbPGl462FxF1aRlBzG10G5xPRvoKsHZXTPKdvNlOUbuCG19OJaGchop2FG15PJ3ObmSPr/Bo95uO87bsrzNO7HN89ri0k45AP29apW9pUZ8fHIWjI4zymakKwceNGjh49WvVYvnw5ALfffruaYalqxwY/Lri8mNhWlUWZrTqW0+mSUjb+EqhyZCdoNArjZh7miw+jSNnrq2osI6eks+GXQLasCVA1juNiE6ws3LyD+et2Mf79w0S3sKgaj97gJLFLGZtXu34+m1cH0PGi0iYRk2LXoDg06IyuJ3O9USF9kxmA2IvKOPBLAMWZehQFjqwzk3/Yh5ZXlADgsGpAAzqfE++hMypotCfeo7F543d3Mr3BydW35vPz4jBA/ZIw0XCq3r5ERLgWgU+dOpXWrVvTu3fvGte3WCxYLCdOsEVFRW6NTw1L3ovEL8DBv1bvxukArQ7mvd6MVd+e/m7d0+4YnonDAd/+O1LVOHoPyqdNUjmj+ieqGsdxu7f48caTvqQdNBISYeeuJzKZ8e0+Hrm6PcX56vzUAkMd6PRQkOu6/4IcPSGR9iYRk4+/k2bdylg/K4Kw1hbM4XZ2fx/E0b98CYm3AnD1xEyWPR/DR5e3Q6tX0GgU+k7JoPlFlSVzzS4ox+DrZM0bUVw+NgsU+HVaFIpTQ2mOe75bb/zuTtbrhkL8Ax0sWxKqdih1JlUGtVO/PPMfVquVBQsWMGbMGDSamrPN5ORkJk+e7OHIPKv3TQVc83/5TB3RkpS9Jlp3KuexyekcyzKw4nP1f3htkkoZdH82Iwd0QM27gogYK4+/nMGEu1phs3hHZ5lNK0+U4hzeDbs2mZn3+99cd3seX32kbvJ0akmnRoPqDaQ8GVP/N9P4+blYPrysHRqdQlSncjrcWEjWP+1O/vxPKEe3+nLzhykExtpI2+DHiknN8Iu00/KyUsxhDm58N5UVL8bw5/xQNFpoP7CQyE7lbms/cJw3fncA1w/OY+PKQPKy1G1DVC/SqLBWXpMQfPPNNxQUFHDfffeddp3x48czZsyYqudFRUXExcV5IDrPeXhiBotnRbL6u8oSgcO7fYlsbmXwyCyvSAg6X1JCcLidT9Ztr1qm08PDL6RxywPZ3HtZkkfiaNOlnJAIO7OW7nWJI+nSUm66P5eB8V1wOtUtxrSU6zi820RsgnrVBkV5Ohx2CIlwvaMMCreT76Y7W2+MKbiljTs/O4ytTIOlRId/pJ3vn2hOUJwNW4WGtW9FMuj9VFpdVVlFENHeQvbfJjb9K4yWl1UWz8dfUcpDK/dRlqdDq1cwBTr54NK2BMVZ3RKzN353x0XGWul2RTGvPJSgahyicXlNQjBnzhz69etHTEzMadcxGo0YjUYPRuV5Rl9nte6FTofG7XchdfXLl2FsWePanuG1Bfv45atQli8J91gcW9f488hVbV2WjZ2RSup+E0vei1A9GQAw+DiJS7Sw4w/1GlzZbVr2bTNz4ZXF/H5Sd9oLryxm3c/qtAxXMyaDWcFgtlNRqCVljT9XPpuF06bBadNW+41pdQpKDceRObSym+GRdX6UHdPT+ppit8Tqjd/dcX3vPEZBrp4/vKhtU11IlUHtvCIhSElJYcWKFXz11Vce37fJ7CAm4USGHx1npVWncooLdKp061m/PJDBT2SRnW4gZY+J1p3LufWRbJYtCvNYDCazg5j4E3e10XEWWnUso7hAT06GD8UFroeNw6YhP8dA2kHPdfsrL9WRsse1QWNFmZbi/OrLPeXhiemsXx5EdrqB4HA7Q57MwuzvYLnKJTtffRTOuHdS2bvNl783+dH/7mNExtr44T+eO6bUjunwr34oiobQVhbyU3z49fUoQlpZ6PR/+egM0PySUlZPjUJvdBIYayN1gx+7vg6m94TMqvfY8UUwoa0tmEPtZGwxs/LVaLrff4zQVu4pIQDv/O40GoW+d+ax4vNQr+hWWy8y22GtvCIhmDt3LpGRkQwYMMDj+27btZw3vjxQ9fyxyRkALFscwltPtfB4PO+/0Jx7nznKyClpBIfZOZZl4McF4Xw6w3OD/rTtUsa0JSeK4h99KQ2A5Z+H8dbYeI/Fca4Jb2Zj/HuHCQx1UHhMz+4/zYy+sS3ZKvX1P271dyEEhDgY+lQWoZF2UvaYeOHuBFXj8nRMlmIda96MoiRTjynYQeL1RVw+NhvdP9XfA99OY82bkfw4tjkVBToCYm1cNiabrkNODEyUd9CHNW9GUlGoIyjWRo/Hc+n+gHu7/3njd9ftimKimtv4ebH6VZiicWkURd2Ux+l0kpCQwF133cXUqVPrtW1RURFBQUH0YRB6jRc1bDlNo0g1aXTq99E/lWJXv6V0NV743Z3vdyWNZez+nWqHUM1bbTqpHUJ1XnaM2xUbq5RvKCwsJDDQPVUQx68VPfu9jN5w9iWZdlsF63560a2xqkn1EoIVK1Zw5MgRHnjgAbVDEUIIcT6TXga1Uj0h6Nu3LyoXUgghhBBNnuoJgRBCCOEJ0sugdpIQCCGEaBqcSuWjIdufxyQhEEII0TRIG4JaeclwN0IIIYRQk5QQCCGEaBI0NLANQaNF4p0kIRBCCNE0yEiFtZIqAyGEEEJICYEQQoimQbod1k4SAiGEEE2D9DKolVQZCCGEEEJKCIQQQjQNGkVB04CGgQ3Z9lxwfiQEGo13zeDlhQeN4nCoHUI1Gr33HX7e+Dl51bHtxaa376p2CNVcuyP/zCt52IrOAWqH4MqT50vnP4+GbH8ekyoDIYQQQpwnJQRCCCHEGUiVQe2khEAIIUTToDTCox6Sk5O5+OKLCQgIIDIykptvvpk9e/a4hqQoTJo0iZiYGHx9fenTpw87d+50WcdisTBq1CjCw8Px8/PjpptuIi0trb7//RlJQiCEEKJpOD5SYUMe9bB69WpGjBjB+vXrWb58OXa7nb59+1JaWlq1zrRp05g+fTqzZs1i48aNREdHc91111FcXFy1zujRo/n6669ZtGgRa9eupaSkhIEDB+Jo5DZPUmUghBBCuMHSpUtdns+dO5fIyEg2b97MlVdeiaIozJw5k+eff55bb70VgPnz5xMVFcXChQt59NFHKSwsZM6cOXzyySdce+21ACxYsIC4uDhWrFjB9ddf32jxSgmBEEKIJuH4SIUNeQAUFRW5PCwWS532X1hYCEBoaCgAhw4dIjMzk759+1atYzQa6d27N7///jsAmzdvxmazuawTExND586dq9ZpLJIQCCGEaBoaqcogLi6OoKCgqkdycnIddq0wZswYLr/8cjp37gxAZmYmAFFRUS7rRkVFVb2WmZmJj48PISEhp12nsUiVgRBCCFEPqampBAYGVj03Go1n3GbkyJFs27aNtWvXVntNc8pYI4qiVFt2qrqsU19SQiCEEKJJ0Dgb/gAIDAx0eZwpIRg1ahTfffcdK1eupHnz5lXLo6OjAard6WdnZ1eVGkRHR2O1WsnPzz/tOo1FEgIhhBBNg4d7GSiKwsiRI/nqq6/43//+R0JCgsvrCQkJREdHs3z58qplVquV1atX06tXLwC6d++OwWBwWefo0aPs2LGjap3GIlUGQgghhBuMGDGChQsX8u233xIQEFBVEhAUFISvry8ajYbRo0czZcoUEhMTSUxMZMqUKZjNZoYMGVK17oMPPsjYsWMJCwsjNDSUp59+mqSkpKpeB41FEgIhhBBNg4enP/7ggw8A6NOnj8vyuXPnct999wHwzDPPUF5ezvDhw8nPz6dHjx4sW7aMgIATc07MmDEDvV7PHXfcQXl5Oddccw3z5s1Dp9M14J+pTqMo5+5YjEVFRQQFBdFHczN6jaHe29895ijDxma5LMvL1nNXt84NC6wBH+mdI7O4rH8hcW0sWCu07NpkZs5rzUg7YGpYTA1ofOKuz0lTj4O58yXF3PZYFolJZYRF2Zj8UGvWLQuucd0nklPoPzSX2ZOb882c+tWxNWRyo/nrdxIdZ6u2/Lt54bz3fPMatnA/b4xJq1MYNjaTq2/JJyTCRl62geVLQln4dhSKcvbHaWMeT2PfOsx1tx9z2ebvP/146ub29Yrpmq2nn9zIXgoH3jWS84sea56GgPZO2j5XQVBSZUV19nI9aZ8bKN6lxVagpccXpQS0d51dx2mFvW8ayfpRj8OiIbSHg/YvVGCKPv05qCGTG3XuUcLtw3MqP7doO5MeiGfd0qCzfj8Au2JjFd9SWFjo0lCvMR2/Vlx10QT0+rM/l9rtFazcNMWtsaqpyZcQHN5t4rnBraueOx3qzizXpWcp388LZ+9WMzq9wn3PHmXKZwd5uHc7LOWNmw3Wh9qfk8ns5NAuX5YvCWPiRwdPu17PvgW0u6CU3Mz6J4gN9UT/dmh1J07E8e0rmLroAGv+27AT5vkW050jshgwLJc3R7cgZY+JxK7ljJ1+hNJiHd/MifBIDHU5njauDGT60/FVz23Wxj3m/37RRMl+LZ2SKzBGOjn6vYE/HzbT89tSTFEKjnII7uYgqq+dvyfVfBHbM9VI7mo9nd+owBCssO8NI1tH+NJjSRkaN5wuTGYnB3eaWLYohBfnpDT+DoSqVE0I7HY7kyZN4tNPPyUzM5NmzZpx33338cILL6DVeqa9o8MB+Tmev3iczvNDW7k8f+upFizZsZPELuXs+MNfpajU/5w2rQpi06raL2JhUVaGv3KEF4Yl8vLc/R6K7ITCPNef050js8g45MO2dep9b94YU4fuZaz7OYgNv1R+n1lpRq4alE9i1zKPxVCX48lm1bjtmHdUQPYKPV3fKSfkospSqdYjrOT8T0/aYgNtnrDS7CY7AOXpNSci9mLI+MpAp+QKwnpWvkfnqRWsudaPvPU6wi5r/Km8N60MZNPK43fG52BCcBYNA6ttfx5TNSF4/fXXmT17NvPnz6dTp05s2rSJ+++/n6CgIJ588kmPxBCbYGXh5h3YrFp2bzEzd2ozMo+cuU+pp/gFVv6oiwvUKx0A7/+cNBqFcTMP88WHUaTs9VU7HPQGJ1ffms9XH0UC6pY6HectMe3Y4MeAYbnEtqog/aCJVh3L6XRJKbNfilUtppp0ubSERX/+RUmRju1/BDBvWgyFxxonQVAcoDg0aE/5CelMUPBn3X7rRbt0KHYNYb3sVcuMkQr+bZwUbHFPQnDOUwDnGdeqffvzmKoJwbp16xg0aBADBgwAID4+ns8++4xNmzbVuL7FYnEZIrKoqKhB+9+9xY83nvQl7aCRkAg7dz2RyYxv9/HI1e0pzveG2hSFRyZlsOMPP1L2qHeR8/7PCe4YnonDAd/+O1LtUADodUMh/oEOli0JVTuUKt4S05L3IvELcPCv1btxOkCrg3mvN2PVtyFn3thDNq4KZM0PIWSl+RDdwsI9YzN4fdFeRg3ogM3a8NJLvR8EdXVwcLYPfq0q8AlTyPxRT+E2LeaWdbvqWHM1aAwKhlMKOnzCFKzHvCMJ9TYy/XHtVB2H4PLLL+eXX35h7969APz111+sXbuW/v3717h+cnKyy3CRcXFxDdr/ppWBrP0xmMO7fdmyJoCJ91QW1193e16D3rexjJiSTkKHcpKHt1A1Dm//nNoklTLo/mzeGhuPt9yNXz84j40rA8nL8p7qKG+JqfdNBVzzf/lMHdGSETe0483RLbjtsWyu9ZLjCeDX70PZ8L8gUvb68seKYCbem0hsgoVLri5stH10Si4HYM3V/vzvQn9SP/Uhur8djbaBF53z+5ol3EjV27tnn32WwsJC2rdvj06nw+Fw8Nprr3HXXXfVuP748eMZM2ZM1fOioqIGJwUns5TrOLzbRGxC3SaqcKfhr6bRs28RY29pTe5RH7XDceFNnxNA50tKCA6388m67VXLdHp4+IU0bnkgm3svS/JoPJGxVrpdUcwrDyWceWUP8aaYHp6YweJZkaz+rrJE4PBuXyKbWxk8MosVn3tPicrJ8rINZKf7ENOIx7y5hcJF88pxlIG9VIMxQmH7WBOm2Lpd0X3CFRSbBlshLqUE1jwNQRdIVlAjhQa2IWi0SLySqgnB4sWLWbBgAQsXLqRTp05s3bqV0aNHExMTw7333lttfaPRWKcxo8+WwcdJXKJF1cZ7oDDitXR63VDIuNvakJXqPfX0x3nH53TCL1+GsWWNaxeg1xbs45evQlm+JNzj8fS98xgFuXr++MV7uiV5U0xGX2e17oVOhwaNF4+bGhBsJ6KZlbzsxi9d0ZlBZ1awFcKx3/W0GVO3pCOwowONXiFvnZ6oGyrbEVhyNJTs15I4VtoP1EgaFdZK1YRg3LhxPPfccwwePBiApKQkUlJSSE5OrjEhaGwPT0xn/fIgstMNBIfbGfJkFmZ/B8tVvEsZOSWdq27JZ9L9CZSXaAmJqOxDXlqsw1qhzhnTGz4nk9lBTPyJE2V0nIVWHcsoLtCTk+FDcYHroeywVbYQTzvYwPEb6kmjUeh7Zx4rPg9VvQvrcd4W0/rlgQx+IovsdAMpe0y07lzOrY9ks2xRmMdiqO14Ki7QcfdTR/ntp2Dysg1ENbdy3zPpFObr+X1pcKPFcOw3HYoCfvFOyo5o2feWEXO8k5ibK3/ztkKoOKrFkl35nZUeqvz9+4QrGMMV9AEQc6uNvW8YMQQr6IMU9r1pxD/RSeil7kkITGYHMQnWqufRcVZadSqnuEBHTrp3lWSK+lM1ISgrK6vWvVCn0+F0NqQZaN2FN7Mx/r3DBIY6KDymZ/efZkbf2JZsFQ/sG++rHAzlza8OuCx/c3Qcy1VqDOYNn1PbLmVMW7K36vmjL6UBsPzzsH/aDniHblcUE9Xcxs+Lvafo29tiev+F5tz7zFFGTkkjOMzOsSwDPy4I59MZjTtRS21qO57endCChPblXPt/x/ALdJCXbWDbugCmjGhFeWnj9faxF2vYP9NIRZYGQ5BC5HV22jxhQftPIUTOSj27XjjRmHjHuMq/Ex630HpE5UW57bMWNHrYPtYXhwVCezjoNKvcLWMQALTtWs4bX544Nz02OQOAZYtDeOspdds61YmThjUz8sylSTWqjlR43333sWLFCj788EM6derEli1beOSRR3jggQd4/fXXz7h9Q0cqdBtvLFZq5GkyG0N9RpbzlIaMVCjU5Y3HU20jFaqlISMVuoMnRyq8pvMz6HVnXw1rd1j4Zcc0GanQHd59910mTpzI8OHDyc7OJiYmhkcffZQXX3xRzbCEEEKIJkfVhCAgIICZM2cyc+ZMNcMQQgjRFEijwlp5x6gyQgghhLtJQlArL+7oI4QQQghPkRICIYQQTYOUENRKEgIhhBBNg3Q7rJUkBEIIIZoEmdyodtKGQAghhBBSQiCEEKKJkDYEtZKEQAghRNPgVEDTgIu68/xOCKTKQAghhBBSQiCEEKKJkCqDWklCIIQQooloYEKAJATeT1Hwqi/KC2cW9MbM1htnFtSazWqHUI2zrEztEKrzwuNJYzz7Wezc5ZcLQtQOoZortpWoHYKLihIbq3qqHYWA8yUhEEIIIc5EqgxqJQmBEEKIpsHZwNJk6WUghBBCiPOdlBAIIYRoGhRn5aMh25/HJCEQQgjRNEgbglpJQiCEEKJpkDYEtZI2BEIIIYSQEgIhhBBNhFQZ1EoSAiGEEE2DQgMTgkaLxCtJlYEQQgghpIRACCFEEyFVBrWShEAIIUTT4HQCDRhLwCnjEDQZd47M4oEJmXz9cTizX4pVLY6waCsPTjjKxVcX4WNykn7QyPSxLdi/Xd2Jdwbem8vtj+cQGmkjZa+J2S/GsGODvyqxzF+/k+g4W7Xl380L573nm3skhgFDMhlwVyZRzS0ApOzzZeGsODb9WjmhzZjX93HdrTku2+ze6s9Tt3fxSHzgHZ9TTTr3KOH24TkkJpURFm1n0gPxrFsa5LH9n+m7M5kd3P90Cr2uyyMg2E5WupHv/tOMHxZGuy2mzpcUc9tjWZWfSZSNyQ+1Zt2y4KrXx751mOtuP+ayzd9/+vHUze0bZf+KHVI+0JP9gxbbMQ0+4QqRgxy0eMSB5p/KZesxODTDQME6LfZiCLrQSevxdnxbut45F/2l4fA7eoq3a9EYwL+dQqf3rehMjRKqcBNJCP7RtmsZ/e/O4+BOdY9Y/yA707/Zx7bfA3jh7lYU5OppFm+ltEinaly9b8rnsckZzJoQy84NfgwYdoxXPz3Ew33akZPu4/F4nujfDq3uxEkovn0FUxcdYM1/PXdRyc30Ye6bLclIqTxmrr0lmxc/2M3IQV05sr8yedu4OpgZz7Wp2sZm8+xMmN7wOdXEZHZycKeJZYtCeHFOisf3f6bv7pEJh+h6aRHTxiaSlW6k++UFjJh0kGNZPqz/JdQtMZnMTg7t8mX5kjAmfnSwxnU2rgxk+tPxVc9t1sY7nlL/rePo5zravWrD3FqheKeGfS8a0PtD7N0OFAV2PemDRg8d37ai84P0T3Rsf8SH7l9b0P1zv1L0l4Ydj/sQ96Cd1uPtaA1QskdTlVSoSqoMaqVqQlBcXMzEiRP5+uuvyc7Oplu3brz99ttcfPHFHo3DZHbw7KwUZo5rzl1PZnl036e6Y3g2uRk+vDWmRdWyrDT1p3W99ZFcfv4slKULwwCY/VIs3fsUM/CeY8xNbubxeArzXA/dO0dmkXHIh23rPFdi8cf/XC8M82e0ZMCQLNpfUFyVENisWvJzPZ8wHecNn1NNNq0MZNPKwH+eeT4hONN316FbMSu+jmD7hsrE6afF0fQbnEViUonbEoJNq4LYtKr2RM1m1ZCfY3DL/ou3aQm7ykHolZXF4qZYhZyfnJTs0gIOylM0FG/TcuFXFvzaVF4Y2zxvZ30fHTk/6Yj+v8rpzA9OMxAzxEHcgyemNz+1BEE1khDUStWc7aGHHmL58uV88sknbN++nb59+3LttdeSnp7u0ThGTklnwy+BbFkT4NH91uTSvoXs3Wbm+Q8PsfivHbz38x76DTl25g3dSG9wktiljM2rXT+fzasD6HhRqUpRnaA3OLn61nx+XhwGePYO/DitVqH3gFxMZge7t574nLr0KOSz9Rv4eNmfPPHqfoJCrarEB97xOXmjmr67nZsDufTqPMKiLIBClx6FxMaX8+eaYFVj7XJpCYv+/It/rdrBk6+nEBRWvTrobAV2c1Lwh46yw5XHRskeDUVbtIRcXnlhV/45dLUn3Z9odKAxQOGWykuJ9RgUb9diCFXYOsyH9X2M/HW/D4V/yvF2LlCthKC8vJwvv/ySb7/9liuvvBKASZMm8c033/DBBx/w6quvVtvGYrFgsViqnhcVFTU4jt6D8mmTVM6o/okNfq/G0KyFlYHDcvnq4wgWvRNFu25lPP5yGjarhhVfuOfO5EwCQx3o9FCQ63q4FOToCYm0qxLTyXrdUIh/oINlSzz/+cS3LWX6ku34GJ2Ul+l4ZXj7qtKBTatDWPNTGNnpRqLjLAwbfYSpn+zkiVu6YrN6PhdX83PyRrV9d7NfSeDJVw+wYO1m7DYNigIzJ7Rm5+bAM7yr+2xcFciaH0LISvMhuoWFe8Zm8PqivYwa0KFRjqfmDziwl2jYPMgHjQ4UB8SPshPZv7LEwDdBwRijcPhtPW1etKHzhfT/6LDlarDmVr5HRVrlhf/IB3oSxtrxb+ck63sd2x/2oftXVvVLCmTo4lqplhDY7XYcDgcmk2udva+vL2vXrq1xm+TkZCZPntxoMUTEWHn85Qwm3NUKm8UbKrhAo4V923yZOzUGgAM7zbRsW8GAe3JVSwiOO7W0TKPBKwbquH5wHhtXBpKX5Z6i1NqkHfJlxE1d8Q90cNn1xxg7bR/PDO3Mkf1mfv0xvGq9lH1+7N3uz/xVm7m4Tz6/LwvzeKxqfk7eqLbvbtA9R2l/QTGTHm1PVrqRpIuLGDHpIHk5Pmz9PViVeH/9/sTvP2WvL/u2+TH/9+1ccnUhvy0NafD75yzVkv1fHe2m2vBrrVCyR8PBaQZ8IhSiBjnRGqDDdCv7XjKw/nIT6BRCejirShCAqvNBs9scRN9cudy/g52CP7RkfqMj4Ul1byAUxYnSgBkLG7LtuUC1hCAgIICePXvyyiuv0KFDB6Kiovjss8/4448/SEys+W59/PjxjBkzpup5UVERcXFxZx1Dmy7lhETYmbV0b9UynR6SLi3lpvtzGRjfBafTs0Vdedl6Uva6Jkmp+01c3r/Qo3GcrChPh8MOIRGuP+agcDv5Oeq2S42MtdLtimJeeShBlf3bbVqOHvEFYN8Of9omlTDo3qO8O7F1tXXzc3zIzjASG1/u6TBV/5y80em+uw9fjefeMUd4ZUQ7Nq6qvAgf3uNHqw6l/N+DGaolBKfKy/7/9u49Lqo6/+P4a4BhBka8QIKggIAKeElF0lCrLV13zXzgz/0prrrhgrvLrqZm3oq8lKJRK5W6kdouupar7sPLpmsRmWnWegElXeDnJZXINCwVcJDLMN/fH8TkBBbG5YzyeT4e83gw55w58+YMzPmc7/me89VTeMEVv6DyH1+4Hs6l6PGPt+A9vHqnZ+qmKL9ooeCvLvhEV58v8OiuiPhnBZYSsFaCqydkj3elVY/q17h+WwO7h9jvON2DFeUXHeC0gVINO8q/y/sQaPptvmHDBuLi4ujYsSPOzs5EREQwfvx4jh49WufyBoMBg6HxOthlf9SK3z/czW7aUy8XUHDGyJa/tG/2YgAg94gJ/xD7f/COweUUXtDuqM5S6cTp4+5EPFjCJzddGhbxYAn/Sde2t/qwmG+49rULh/Zo15R7M50O9K51H0V4tK2kvW85Vwqbv5Oho20nR1Tz2bnoFXpXhfre/7/VqsPJyXF2CB5tLbT3reBKYeN8N1jLqNW1ROdEna2ALt92k7mRr6MkV0fg1Oq/eUNHhau3ovS8Ezdf738jX4fnoLv76PpuoGlBEBISwr59+zCbzRQXF+Pr60tMTAxBQc1zFHPD7Ez+STe7aWWlTpRcrT29uWxb683L/zrFuCe+Yv/OtoT2KeXRCd/wyhztrhkH2LbmHmavKODUcTfyMk08OvEbvDtW8u+/N3/Tdw2dTjEs5grv/9MTa1XzF2+xM/PJ3N+OyxddcTdV8dCIr+k1oIj58d0xulcx8YkCDqR7ceWyHp+O5Ux66nOKr+r5JKN5t5nW26kuRvcq/IK+62DZwb+C4B43KLnm3CyXsf7QZ1d63YXjh1oTP/c85WVOFH5poFf/YoaMuszaZZ2bLJPRvQq/zt8dDHTwLye4eykl11wouebMxCcv8vE7bblSqMenUwWT5lyg6KoLn7zbtlHe3/MhKwVrXTD6KtxDFNf/T8cXG1xsTf8Al99zQt8ODL6K0tM6PkvW4/WwlXYDq3f2Oh10irWQn+qCqZuVVmGKr9525sY5HT7Lq2711s1HNbAPgbQQND2TyYTJZOLq1aukp6fz4osvah1JM6c+def5yUH8dt5FJsy4xKUCV15f2JG927XtP7Dv7XZ4tKtiwpNf4eltIf+kkWcnBlGowT0IavR9oASfTpWkb9Zm27S7p5LZL53G07sCc4kz5/7PxPz47hz7uC2uhio6h5Yy5H8KMXlUceWynuOH2rBsejdumJv3nhJab6e6dOt9g5e2fmZ7nvDclwC8t7kdy58MuNXLGs0PfXYAL8zoxqRZ+cxZfhqPthYKLxhYnxLAvzf6NFmmbveW8uKW705f/mHhFwBk/NOLlc8EEBR2g6G/+gZT6yquFOo5/h8Plk4JbrS/p5CnK8lf5cKZJD2VV8C1vcL3f6sISPjuVGHFZR1nX3Kh8htwbQ/eI6sI+IP9qcSOv6nCWgFnX9JjKQJTqKLn6grc/B1gZ2q1gq4BLRV3eR8CnVLalTzp6ekopQgNDeXMmTPMnj0bg8HAgQMH0Ot/vBmsuLiYNm3a8DOicdE5UEcpnWMchdlxxMrWAbeTk7u2d4Osi7W0VOsItTng35OTyaR1hFpUeeOc329Mg49e1zqCnbLrlSyLSqeoqIjWrZvmlFbNvmKIxwRcdD/9IMaiKthT8laTZtWSpi0ERUVFPP3003zxxRd4enryq1/9iqSkpHoVA0IIIcRtkVMGP0jTgmDs2LGMHTtWywhCCCFaCGW1ohpwyuBuv+zQMS6+F0IIIYSmHKJToRBCCNHk5JTBD5KCQAghRMtgVaCTguBW5JSBEEIIIaSFQAghRAuhFDffQfGnvf7uJQWBEEKIFkFZFaoBpww0vG1Ps5CCQAghRMugrDSshUAuOxRCCCHET/Taa68RFBSE0WikX79+fPTRR1pHqpMUBEIIIVoEZVUNftyuzZs3M2PGDBITEzl27BgPPPAAw4cP5/PPP2+C37BhpCAQQgjRMihrwx+3KSUlhfj4eCZPnkx4eDivvPIK/v7+pKamNsEv2DB3dB+Cmg4eFiobdK+Jxud4g/Y4Zu9Yx9tOTqrixxdqZlZVqXWE2hzw78kRPzvlgJ9d2XXHylRurh4tsTk67DV0X2GhetsVFxfbTTcYDBgMhlrLV1RUkJWVxbx58+ymDxs2jE8++eSnB2kid3RBUFJSAsABdmuc5Hsc77vSMTnidjJrHUD8ZPLZ1cveKK0T1K2kpIQ2bdo0ybpdXV3p0KEDBy41fF/RqlUr/P397aYtXLiQRYsW1Vr266+/pqqqCh8f+2GzfXx8uHTpUoOzNLY7uiDw8/OjoKAADw8PdA0cSre4uBh/f38KCgocZlhLyVQ/jpbJ0fKAZKovyVQ/jZlJKUVJSQl+fn6NlK42o9HIuXPnqKhoeCuSUqrW/qau1oGbfX/5utbhCO7ogsDJyYlOnTo16jpbt27tMP90NSRT/ThaJkfLA5KpviRT/TRWpqZqGbiZ0WjEaDQ2+fvc7J577sHZ2blWa0BhYWGtVgNHIJ0KhRBCiCbg6upKv379yMjIsJuekZHBwIEDNUp1a3d0C4EQQgjhyGbOnMlvfvMbIiMjiYqKYs2aNXz++eckJCRoHa0WKQi+ZTAYWLhw4Y+eC2pOkql+HC2To+UByVRfkql+HDGTo4qJieGbb77h+eef5+LFi/Ts2ZPdu3cTGBiodbRadOpuvzmzEEIIIX6U9CEQQgghhBQEQgghhJCCQAghhBBIQSCEEEIIpCAAHG9oyv379zNy5Ej8/PzQ6XTs2LFD0zzLli3jvvvuw8PDA29vb0aNGsXJkyc1zZSamsq9995ruzFKVFQU77zzjqaZvm/ZsmXodDpmzJihWYZFixah0+nsHh06dNAsT40LFy4wceJEvLy8cHd3p0+fPmRlZWmWp3PnzrW2k06nY8qUKZplslgsPPvsswQFBeHm5kZwcDDPP/88VuvtD7DTmEpKSpgxYwaBgYG4ubkxcOBAjhw5omkm0ThafEHgiENTms1mevfuzapVqzTLcLN9+/YxZcoUDh48SEZGBhaLhWHDhmE2a3fz+E6dOvHCCy+QmZlJZmYmjzzyCNHR0eTk5GiW6WZHjhxhzZo13HvvvVpHoUePHly8eNH2OHHihKZ5rl69yqBBg9Dr9bzzzjvk5uayfPly2rZtq1mmI0eO2G2jmhvJjBkzRrNMycnJvP7666xatYq8vDxefPFFXnrpJVauXKlZJoDJkyeTkZHBhg0bOHHiBMOGDWPo0KFcuHBB01yiEagWrn///iohIcFuWlhYmJo3b55GiewBavv27VrHsFNYWKgAtW/fPq2j2GnXrp164403tI6hSkpKVNeuXVVGRoZ66KGH1PTp0zXLsnDhQtW7d2/N3r8uc+fOVYMHD9Y6xg+aPn26CgkJUVarVbMMI0aMUHFxcXbTRo8erSZOnKhRIqVKS0uVs7Oz2rVrl9303r17q8TERI1SicbSolsIaoamHDZsmN10Rx2a0lEUFRUB4OnpqXGSalVVVWzatAmz2UxUlPZDuU2ZMoURI0YwdOhQraMAcPr0afz8/AgKCmLcuHGcPXtW0zxvv/02kZGRjBkzBm9vb/r27cvatWs1zXSziooK3nzzTeLi4jQdgGbw4MHs2bOHU6dOAfDpp59y4MABHn30Uc0yWSwWqqqqao0J4ObmxoEDBzRKJRpLi75T4Z02NKUjUEoxc+ZMBg8eTM+ePTXNcuLECaKioigrK6NVq1Zs376d7t27a5pp06ZNHD161GHOqQ4YMIC///3vdOvWja+++oolS5YwcOBAcnJy8PLy0iTT2bNnSU1NZebMmTzzzDMcPnyYadOmYTAYePzxxzXJdLMdO3Zw7do1Jk2apGmOuXPnUlRURFhYGM7OzlRVVZGUlMSvf/1rzTJ5eHgQFRXF4sWLCQ8Px8fHh3/84x8cOnSIrl27apZLNI4WXRDUuFOGpnQEU6dO5fjx4w5xNBAaGkp2djbXrl1j69atxMbGsm/fPs2KgoKCAqZPn857773X7KOq3crw4cNtP/fq1YuoqChCQkJYv349M2fO1CST1WolMjKSpUuXAtC3b19ycnJITU11iILgr3/9K8OHD2/S4XjrY/Pmzbz55pts3LiRHj16kJ2dzYwZM/Dz8yM2NlazXBs2bCAuLo6OHTvi7OxMREQE48eP5+jRo5plEo2jRRcEd9rQlFp74oknePvtt9m/f3+jDzv9U7i6utKlSxcAIiMjOXLkCK+++iqrV6/WJE9WVhaFhYX069fPNq2qqor9+/ezatUqysvLcXZ21iRbDZPJRK9evTh9+rRmGXx9fWsVbeHh4WzdulWjRN/Jz8/n/fffZ9u2bVpHYfbs2cybN49x48YB1QVdfn4+y5Yt07QgCAkJYd++fZjNZoqLi/H19SUmJoagoCDNMonG0aL7ENxpQ1NqRSnF1KlT2bZtGx988IHD/uMrpSgvL9fs/YcMGcKJEyfIzs62PSIjI5kwYQLZ2dmaFwMA5eXl5OXl4evrq1mGQYMG1bps9dSpUw4x2EtaWhre3t6MGDFC6yiUlpbi5GT/Fe3s7Kz5ZYc1TCYTvr6+XL16lfT0dKKjo7WOJBqoRbcQgGMOTXn9+nXOnDlje37u3Dmys7Px9PQkICCg2fNMmTKFjRs38q9//QsPDw9bi0qbNm1wc3Nr9jwAzzzzDMOHD8ff35+SkhI2bdrEhx9+yLvvvqtJHqg+v/r9fhUmkwkvLy/N+lvMmjWLkSNHEhAQQGFhIUuWLKG4uFjTI8wnn3ySgQMHsnTpUsaOHcvhw4dZs2YNa9as0SwTVJ/KSEtLIzY2FhcX7b8aR44cSVJSEgEBAfTo0YNjx46RkpJCXFycprnS09NRShEaGsqZM2eYPXs2oaGh/Pa3v9U0l2gEml7j4CD+8pe/qMDAQOXq6qoiIiI0v5xu7969Cqj1iI2N1SRPXVkAlZaWpkkepZSKi4uzfWbt27dXQ4YMUe+9955meW5F68sOY2JilK+vr9Lr9crPz0+NHj1a5eTkaJanxs6dO1XPnj2VwWBQYWFhas2aNVpHUunp6QpQJ0+e1DqKUkqp4uJiNX36dBUQEKCMRqMKDg5WiYmJqry8XNNcmzdvVsHBwcrV1VV16NBBTZkyRV27dk3TTKJxyPDHQgghhGjZfQiEEEIIUU0KAiGEEEJIQSCEEEIIKQiEEEIIgRQEQgghhEAKAiGEEEIgBYEQQgghkIJACCGEEEhBIESDLVq0iD59+tieT5o0iVGjRjV7jvPnz6PT6cjOzr7lMp07d+aVV16p9zrXrVtH27ZtG5xNp9OxY8eOBq9HCNF0pCAQd6VJkyah0+nQ6XTo9XqCg4OZNWsWZrO5yd/71VdfZd26dfVatj47cSGEaA7aj+AhRBP55S9/SVpaGpWVlXz00UdMnjwZs9lMampqrWUrKyvR6/WN8r5t2rRplPUIIURzkhYCcdcyGAx06NABf39/xo8fz4QJE2zN1jXN/H/7298IDg7GYDCglKKoqIjf//73eHt707p1ax555BE+/fRTu/W+8MIL+Pj44OHhQXx8PGVlZXbzv3/KwGq1kpycTJcuXTAYDAQEBJCUlARgG0q6b9++6HQ6fvazn9lel5aWRnh4OEajkbCwMF577TW79zl8+DB9+/bFaDQSGRnJsWPHbnsbpaSk0KtXL0wmE/7+/vzpT3/i+vXrtZbbsWMH3bp1w2g08vOf/5yCggK7+Tt37qRfv34YjUaCg4N57rnnsFgst51HCKEdKQhEi+Hm5kZlZaXt+ZkzZ9iyZQtbt261NdmPGDGCS5cusXv3brKysoiIiGDIkCFcuXIFgC1btrBw4UKSkpLIzMzE19e31o76+55++mmSk5OZP38+ubm5bNy4ER8fH6B6pw7w/vvvc/HiRbZt2wbA2rVrSUxMJCkpiby8PJYuXcr8+fNZv349AGazmccee4zQ0FCysrJYtGgRs2bNuu1t4uTkxIoVK/jvf//L+vXr+eCDD5gzZ47dMqWlpSQlJbF+/Xo+/vhjiouLGTdunG1+eno6EydOZNq0aeTm5rJ69WrWrVtnK3qEEHcIjUdbFKJJxMbGqujoaNvzQ4cOKS8vLzV27FillFILFy5Uer1eFRYW2pbZs2ePat26tSorK7NbV0hIiFq9erVSSqmoqCiVkJBgN3/AgAGqd+/edb53cXGxMhgMau3atXXmPHfunALUsWPH7Kb7+/urjRs32k1bvHixioqKUkoptXr1auXp6anMZrNtfmpqap3rullgYKB6+eWXbzl/y5YtysvLy/Y8LS1NAergwYO2aXl5eQpQhw4dUkop9cADD6ilS5farWfDhg3K19fX9hxQ27dvv+X7CiG0J30IxF1r165dtGrVCovFQmVlJdHR0axcudI2PzAwkPbt29ueZ2Vlcf36dby8vOzWc+PGDT777DMA8vLySEhIsJsfFRXF3r1768yQl5dHeXk5Q4YMqXfuy5cvU1BQQHx8PL/73e9s0y0Wi61/Ql5eHr1798bd3d0ux+3au3cvS5cuJTc3l+LiYiwWC2VlZZjNZkwmEwAuLi5ERkbaXhMWFkbbtm3Jy8ujf//+ZGVlceTIEbsWgaqqKsrKyigtLbXLKIRwXFIQiLvWww8/TGpqKnq9Hj8/v1qdBmt2eDWsViu+vr58+OGHtdb1Uy+9c3Nzu+3XWK1WoPq0wYABA+zmOTs7A6CU+kl5bpafn8+jjz5KQkICixcvxtPTkwMHDhAfH293agWqLxv8vpppVquV5557jtGjR9daxmg0NjinEKJ5SEEg7lomk4kuXbrUe/mIiAguXbqEi4sLnTt3rnOZ8PBwDh48yOOPP26bdvDgwVuus2vXrri5ubFnzx4mT55ca76rqytQfURdw8fHh44dO3L27FkmTJhQ53q7d+/Ohg0buHHjhq3o+KEcdcnMzMRisbB8+XKcnKq7E23ZsqXWchaLhczMTPr37w/AyZMnuXbtGmFhYUD1djt58uRtbWshhOORgkCIbw0dOpSoqChGjRpFcnIyoaGhfPnll+zevZtRo0YRGRnJ9OnTiY2NJTIyksGDB/PWW2+Rk5NDcHBwnes0Go3MnTuXOXPm4OrqyqBBg7h8+TI5OTnEx8fj7e2Nm5sb7777Lp06dcJoNNKmTRsWLVrEtGnTaN26NcOHD6e8vJzMzEyuXr3KzJkzGT9+PImJicTHx/Pss89y/vx5/vznP9/W7xsSEoLFYmHlypWMHDmSjz/+mNdff73Wcnq9nieeeIIVK1ag1+uZOnUq999/v61AWLBgAY899hj+/v6MGTMGJycnjh8/zokTJ1iyZMntfxBCCE3IVQZCfEun07F7924efPBB4uLi6NatG+PGjeP8+fO2qwJiYmJYsGABc+fOpV+/fuTn5/PHP/7xB9c7f/58nnrqKRYsWEB4eDgxMTEUFhYC1efnV6xYwerVq/Hz8yM6OhqAyZMn88Ybb7Bu3Tp69erFQw89xLp162yXKbZq1YqdO3eSm5tL3759SUxMJDk5+bZ+3z59+pCSkkJycjI9e/bkrbfeYtmyZbWWc3d3Z+7cuYwfP56oqCjc3NzYtGmTbf4vfvELdu3aRUZGBvfddx/3338/KSkpBAYG3lYeIYS2dKoxTkYKIYQQ4o4mLQRCCCGEkIJACCGEEFIQCCGEEAIpCIQQQgiBFARCCCGEQAoCIYQQQiAFgRBCCCGQgkAIIYQQSEEghBBCCKQgEEIIIQRSEAghhBAC+H8AWm1TUMO+uAAAAABJRU5ErkJggg==",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "ConfusionMatrixDisplay.from_predictions(label_array,pred_array)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 258,
   "metadata": {},
   "outputs": [],
   "source": [
    "## compare your implementation of function to compute accuracy with the implementation of scikit-learn.\n",
    "your_test_acc = accuracy_score_manual(label_array, pred_array)\n",
    "assert abs(your_test_acc - test_acc) < 1e-5 , 'Please check your implementation of function to compute accuracy'"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---\n",
    "## Analyze Your Result (20 marks)\n",
    "Compare the performance of your models with the following analysis. Both English and Chinese answers are acceptable.\n",
    "1. Does your vanilla MLP overfit to the training data? (5 marks)\n",
    "\n",
    "Answer:是的，vanilla MLP存在过拟合\n",
    "\n",
    "2. If yes, how do you observe it? If no, why? (5 marks)\n",
    "\n",
    "Answer:模型vanilla MLP在训练结束时候training loss还在继续下降但是test_loss在某个点开始不再下降反而开始上升了\n",
    "\n",
    "3. Is regularized model help prevent overfitting? (5 marks)\n",
    "\n",
    "Answer:是的，在模型RegularizedMLP中加入了dropout的正则化项，一定程度上解决了过拟合的问题，虽然在训练集上的效果有一点下降，但测试集上的效果有所提升。\n",
    "\n",
    "4. Generally compare the performance of two models. (5 marks)\n",
    "\n",
    "Answer:vanillaMLP模型存在过拟合的问题，在测试集上的表现与在训练集上的表现差距较大，RegularizedMLP通过dropout正则化的方法优化了vanillaMLP模型，具有更好的鲁棒性。\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "anaconda-cloud": {},
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.18"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}