LeNet-5_sam_optim.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# LeNet-5 with Sharpness-aware minimization"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "cuda\n"
     ]
    }
   ],
   "source": [
    "# 우선, MNIST dataset에 적용할 transformation 객체를 생성합니다.\n",
    "from torchvision import transforms\n",
    "import torch\n",
    "import copy\n",
    "\n",
    "device = 'cuda' if torch.cuda.is_available() else 'cpu'\n",
    "print(device)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 1. Training & Validation 데이터셋 준비"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "변형 없는 transformation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [],
   "source": [
    "# transformation 정의하기\n",
    "data_transform = transforms.Compose([\n",
    "            #transforms.Resize((32, 32)),\n",
    "            transforms.ToTensor(),\n",
    "])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "get_train_valid_loader 함수는 MNIST Training 데이터셋을 split하여 Training dataset과 Validation dataset을 리턴함."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import torch\n",
    "from torchvision import datasets, transforms\n",
    "from torch.utils.data.sampler import SubsetRandomSampler\n",
    "\n",
    "\n",
    "def get_train_valid_loader(data_dir,\n",
    "                           batch_size,\n",
    "                           random_seed,\n",
    "                           augment=False,\n",
    "                           valid_size=0.2,\n",
    "                           shuffle=True,\n",
    "                           num_workers=1,\n",
    "                           pin_memory=True):\n",
    "    \"\"\"\n",
    "    Utility function for loading and returning train and valid \n",
    "    multi-process iterators over the MNIST dataset. A sample \n",
    "    9x9 grid of the images can be optionally displayed.\n",
    "    If using CUDA, num_workers should be set to 1 and pin_memory to True.\n",
    "    Params\n",
    "    ------\n",
    "    - data_dir: path directory to the dataset.\n",
    "    - batch_size: how many samples per batch to load.\n",
    "    - augment: whether to apply the data augmentation scheme\n",
    "      mentioned in the paper. Only applied on the train split.\n",
    "    - random_seed: fix seed for reproducibility.\n",
    "    - valid_size: percentage split of the training set used for\n",
    "      the validation set. Should be a float in the range [0, 1].\n",
    "    - shuffle: whether to shuffle the train/validation indices.\n",
    "    - show_sample: plot 9x9 sample grid of the dataset.\n",
    "    - num_workers: number of subprocesses to use when loading the dataset.\n",
    "    - pin_memory: whether to copy tensors into CUDA pinned memory. Set it to\n",
    "      True if using GPU.\n",
    "    Returns\n",
    "    -------\n",
    "    - train_loader: training set iterator.\n",
    "    - valid_loader: validation set iterator.\n",
    "    \"\"\"\n",
    "    error_msg = \"[!] valid_size should be in the range [0, 1].\"\n",
    "    assert ((valid_size >= 0) and (valid_size <= 1)), error_msg\n",
    "\n",
    "    normalize = transforms.Normalize((0.1307,), (0.3081,))  # MNIST\n",
    "\n",
    "    '''\n",
    "    # define transforms\n",
    "    valid_transform = transforms.Compose([\n",
    "            transforms.ToTensor(),\n",
    "            normalize\n",
    "        ])\n",
    "    if augment:\n",
    "        train_transform = transforms.Compose([\n",
    "            transforms.RandomCrop(32, padding=4),\n",
    "            transforms.RandomHorizontalFlip(),\n",
    "            transforms.ToTensor(),\n",
    "            normalize\n",
    "        ])\n",
    "    else:\n",
    "        train_transform = transforms.Compose([\n",
    "            transforms.ToTensor(),\n",
    "            normalize\n",
    "        ])\n",
    "    '''\n",
    "\n",
    "    # load the dataset\n",
    "    train_dataset = datasets.MNIST(root=data_dir, train=True, \n",
    "                download=True, transform=data_transform)\n",
    "\n",
    "    valid_dataset = datasets.MNIST(root=data_dir, train=True, \n",
    "                download=True, transform=data_transform)\n",
    "\n",
    "    num_train = len(train_dataset)\n",
    "    indices = list(range(num_train))\n",
    "    split = int(np.floor(valid_size * num_train))\n",
    "\n",
    "    if shuffle == True:\n",
    "        np.random.seed(random_seed)\n",
    "        np.random.shuffle(indices)\n",
    "\n",
    "    train_idx, valid_idx = indices[split:], indices[:split]\n",
    "\n",
    "    train_sampler = SubsetRandomSampler(train_idx)\n",
    "    valid_sampler = SubsetRandomSampler(valid_idx)\n",
    "\n",
    "    train_loader = torch.utils.data.DataLoader(train_dataset, \n",
    "                    batch_size=batch_size, sampler=train_sampler, \n",
    "                    num_workers=num_workers, pin_memory=pin_memory)\n",
    "\n",
    "    valid_loader = torch.utils.data.DataLoader(valid_dataset, \n",
    "                    batch_size=batch_size, sampler=valid_sampler, \n",
    "                    num_workers=num_workers, pin_memory=pin_memory)\n",
    "\n",
    "    return (train_dataset, train_loader, valid_dataset, valid_loader)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [],
   "source": [
    "# MNIST training dataset 불러오기\n",
    "from torchvision import datasets\n",
    "\n",
    "# 데이터를 저장할 경로 설정\n",
    "path2data = './data'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [],
   "source": [
    "# data loader 를 생성합니다.\n",
    "from torch.utils.data import DataLoader\n",
    "\n",
    "train_data, train_dl, val_data, val_dl = get_train_valid_loader(path2data, 128, random_seed=np.random.seed(100), augment=False, valid_size=0.2, shuffle=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Validation dataset의 비율을 0.2로 설정하였으므로  \n",
    "Training dataset과 Validation dataset 비율은 8:2.  \n",
    "\n",
    "Training dataset : 48000  \n",
    "Validation dataset : 12000  "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Sample 이미지 확인"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# sample images를 확인합니다.\n",
    "from torchvision import utils\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "%matplotlib inline\n",
    "\n",
    "# training data를 추출합니다.\n",
    "x_train, y_train = train_data.data, train_data.targets\n",
    "\n",
    "# val data를 추출합니다.\n",
    "x_val, y_val = val_data.data, val_data.targets\n",
    "\n",
    "# 차원을 추가하여 B*C*H*W 가 되도록 합니다.\n",
    "if len(x_train.shape) == 3:\n",
    "    x_train = x_train.unsqueeze(1)\n",
    "\n",
    "if len(x_val.shape) == 3:\n",
    "    x_val = x_val.unsqueeze(1)\n",
    "\n",
    "# tensor를 image로 변경하는 함수를 정의합니다.\n",
    "def show(img):\n",
    "    # tensor를 numpy array로 변경합니다.\n",
    "    npimg = img.numpy()\n",
    "    # C*H*W를 H*W*C로 변경합니다.\n",
    "    npimg_tr = npimg.transpose((1,2,0))\n",
    "    plt.imshow(npimg_tr, interpolation='nearest')\n",
    "\n",
    "# images grid를 생성하고 출력합니다.\n",
    "# 총 40개 이미지, 행당 8개 이미지를 출력합니다.\n",
    "x_grid = utils.make_grid(x_train[:40], nrow=8, padding=2)\n",
    "\n",
    "show(x_grid)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 2. LeNet-5 모델 정의"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [],
   "source": [
    "from torch import nn\n",
    "import torch.nn.functional as F\n",
    "\n",
    "class LeNet_5(nn.Module):\n",
    "    def __init__(self):\n",
    "        super(LeNet_5,self).__init__()\n",
    "        self.conv1 = nn.Conv2d(1, 6, kernel_size=5, stride=1, padding=2)\n",
    "        self.conv2 = nn.Conv2d(6, 16, kernel_size=5, stride=1)\n",
    "        self.conv3 = nn.Conv2d(16, 120, kernel_size=5, stride=1)\n",
    "        self.fc1 = nn.Linear(120, 84)\n",
    "        self.fc2 = nn.Linear(84, 10)\n",
    "\n",
    "    def forward(self, x):\n",
    "        x = F.tanh(self.conv1(x))\n",
    "        x = F.avg_pool2d(x, 2, 2)\n",
    "        x = F.tanh(self.conv2(x))\n",
    "        x = F.avg_pool2d(x, 2, 2)\n",
    "        x = F.tanh(self.conv3(x))\n",
    "        x = x.view(-1, 120)\n",
    "        x = F.tanh(self.fc1(x))\n",
    "        x = self.fc2(x)\n",
    "        return F.log_softmax(x, dim=1)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 3. 하이퍼파라미터 설정 및 학습"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Loss function 정의"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [],
   "source": [
    "def smooth_crossentropy(pred, gold, smoothing=0.1):\n",
    "    n_class = pred.size(1)\n",
    "\n",
    "    one_hot = torch.full_like(pred, fill_value=smoothing / (n_class - 1))\n",
    "    one_hot.scatter_(dim=1, index=gold.unsqueeze(1), value=1.0 - smoothing)\n",
    "    log_prob = F.log_softmax(pred, dim=1)\n",
    "\n",
    "    return F.kl_div(input=log_prob, target=one_hot, reduction='none').sum(-1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "SAM + SGD + momentum 조합으로 학습시킴.  \n",
    "LR 함수로는 LR이 계단형으로 증가하는 StepLR 사용함.  \n",
    "  \n",
    "SAM Optimizer는 첫 번째 Step과 두 번째 Step이 있음.  \n",
    "첫 번째 Step : 로컬 `rho`-neighbor에서 손실이 가장 큰 가중치를 찾는 첫 번째 최적화 단계를 수행.  \n",
    "두 번째 Step : Loss landscape에서 (로컬적으로) 가장 높은 지점의 기울기로 원래 가중치를 업데이트하는 두 번째 최적화 단계를 수행.  \n",
    "  \n",
    "배치 크기 : 128  \n",
    "에폭 수 : 80  \n",
    "초기 LR : 0.1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "LeNet_5(\n",
      "  (conv1): Conv2d(1, 6, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2))\n",
      "  (conv2): Conv2d(6, 16, kernel_size=(5, 5), stride=(1, 1))\n",
      "  (conv3): Conv2d(16, 120, kernel_size=(5, 5), stride=(1, 1))\n",
      "  (fc1): Linear(in_features=120, out_features=84, bias=True)\n",
      "  (fc2): Linear(in_features=84, out_features=10, bias=True)\n",
      ")\n",
      "┏━━━━━━━━━━━━━━┳━━━━━━━╸T╺╸R╺╸A╺╸I╺╸N╺━━━━━━━┳━━━━━━━╸S╺╸T╺╸A╺╸T╺╸S╺━━━━━━━┳━━━━━━━╸L-O-A-D-I-N-G╺━━━━━━━|━━━━━━━╸V╺╸A╺╸L╺╸I╺╸D╺━━━━━━━┓\n",
      "┃              ┃              ╷              ┃              ╷              ┃                             ┃              ╷              ┃\n",
      "┃       epoch  ┃        loss  │    accuracy  ┃        l.r.  │     elapsed  ┃             bar             ┃        loss  │    accuracy  ┃\n",
      "┠──────────────╂──────────────┼──────────────╂──────────────┼──────────────╂─────────────────────────────┃──────────────┼──────────────┨\n",
      "┃           0  ┃      0.3110  │     88.52 %  ┃   1.000e-01  │   00:09 min  ┃┈██████████████████████████▒┈┨      0.1110  │     96.80 %  ┃\n",
      "┃           1  ┃      0.0919  │     97.12 %  ┃   1.000e-01  │   00:09 min  ┃┈██████████████████████████▒┈┨      0.0749  │     97.77 %  ┃\n",
      "┃           2  ┃      0.0677  │     97.97 %  ┃   1.000e-01  │   00:09 min  ┃┈██████████████████████████▒┈┨      0.0700  │     97.98 %  ┃\n",
      "┃           3  ┃      0.0594  │     98.23 %  ┃   1.000e-01  │   00:09 min  ┃┈██████████████████████████▒┈┨      0.0546  │     98.41 %  ┃\n",
      "┃           4  ┃      0.0544  │     98.40 %  ┃   1.000e-01  │   00:09 min  ┃┈██████████████████████████▒┈┨      0.0584  │     98.19 %  ┃\n",
      "┃           5  ┃      0.0487  │     98.60 %  ┃   1.000e-01  │   00:09 min  ┃┈██████████████████████████▒┈┨      0.0523  │     98.43 %  ┃\n",
      "┃           6  ┃      0.0473  │     98.65 %  ┃   1.000e-01  │   00:09 min  ┃┈██████████████████████████▒┈┨      0.0502  │     98.50 %  ┃\n",
      "┃           7  ┃      0.0438  │     98.80 %  ┃   1.000e-01  │   00:09 min  ┃┈██████████████████████████▒┈┨      0.0483  │     98.59 %  ┃\n",
      "┃           8  ┃      0.0425  │     98.84 %  ┃   1.000e-01  │   00:09 min  ┃┈██████████████████████████▒┈┨      0.0522  │     98.52 %  ┃\n",
      "┃           9  ┃      0.0419  │     98.85 %  ┃   1.000e-01  │   00:09 min  ┃┈██████████████████████████▒┈┨      0.0487  │     98.43 %  ┃\n",
      "┃          10  ┃      0.0410  │     98.87 %  ┃   1.000e-01  │   00:09 min  ┃┈██████████████████████████▒┈┨      0.0451  │     98.67 %  ┃\n",
      "┃          11  ┃      0.0397  │     98.90 %  ┃   1.000e-01  │   00:09 min  ┃┈██████████████████████████▒┈┨      0.0505  │     98.38 %  ┃\n",
      "┃          12  ┃      0.0379  │     98.99 %  ┃   1.000e-01  │   00:09 min  ┃┈██████████████████████████▒┈┨      0.0424  │     98.68 %  ┃\n",
      "┃          13  ┃      0.0384  │     98.90 %  ┃   1.000e-01  │   00:09 min  ┃┈██████████████████████████▒┈┨      0.0442  │     98.65 %  ┃\n",
      "┃          14  ┃      0.0375  │     98.97 %  ┃   1.000e-01  │   00:09 min  ┃┈██████████████████████████▒┈┨      0.0438  │     98.66 %  ┃\n",
      "┃          15  ┃      0.0361  │     99.03 %  ┃   1.000e-01  │   00:09 min  ┃┈██████████████████████████▒┈┨      0.0474  │     98.42 %  ┃\n",
      "┃          16  ┃      0.0366  │     99.04 %  ┃   1.000e-01  │   00:09 min  ┃┈██████████████████████████▒┈┨      0.0461  │     98.54 %  ┃\n",
      "┃          17  ┃      0.0360  │     99.01 %  ┃   1.000e-01  │   00:09 min  ┃┈██████████████████████████▒┈┨      0.0419  │     98.64 %  ┃\n",
      "┃          18  ┃      0.0354  │     99.01 %  ┃   1.000e-01  │   00:09 min  ┃┈██████████████████████████▒┈┨      0.0393  │     98.79 %  ┃\n",
      "┃          19  ┃      0.0349  │     99.06 %  ┃   1.000e-01  │   00:09 min  ┃┈██████████████████████████▒┈┨      0.0493  │     98.52 %  ┃\n",
      "┃          20  ┃      0.0348  │     99.04 %  ┃   1.000e-01  │   00:09 min  ┃┈██████████████████████████▒┈┨      0.0381  │     98.78 %  ┃\n",
      "┃          21  ┃      0.0339  │     99.08 %  ┃   1.000e-01  │   00:09 min  ┃┈██████████████████████████▒┈┨      0.0397  │     98.73 %  ┃\n",
      "┃          22  ┃      0.0341  │     99.07 %  ┃   1.000e-01  │   00:09 min  ┃┈██████████████████████████▒┈┨      0.0424  │     98.63 %  ┃\n",
      "┃          23  ┃      0.0347  │     99.08 %  ┃   1.000e-01  │   00:09 min  ┃┈██████████████████████████▒┈┨      0.0400  │     98.72 %  ┃\n",
      "┃          24  ┃      0.0246  │     99.38 %  ┃   2.000e-02  │   00:09 min  ┃┈██████████████████████████▒┈┨      0.0329  │     98.93 %  ┃\n",
      "┃          25  ┃      0.0233  │     99.42 %  ┃   2.000e-02  │   00:09 min  ┃┈██████████████████████████▒┈┨      0.0330  │     98.92 %  ┃\n",
      "┃          26  ┃      0.0231  │     99.45 %  ┃   2.000e-02  │   00:09 min  ┃┈██████████████████████████▒┈┨      0.0322  │     99.01 %  ┃\n",
      "┃          27  ┃      0.0231  │     99.45 %  ┃   2.000e-02  │   00:09 min  ┃┈██████████████████████████▒┈┨      0.0335  │     98.91 %  ┃\n",
      "┃          28  ┃      0.0229  │     99.47 %  ┃   2.000e-02  │   00:09 min  ┃┈██████████████████████████▒┈┨      0.0319  │     98.95 %  ┃\n",
      "┃          29  ┃      0.0230  │     99.47 %  ┃   2.000e-02  │   00:09 min  ┃┈██████████████████████████▒┈┨      0.0319  │     99.02 %  ┃\n",
      "┃          30  ┃      0.0230  │     99.48 %  ┃   2.000e-02  │   00:09 min  ┃┈██████████████████████████▒┈┨      0.0330  │     99.05 %  ┃\n",
      "┃          31  ┃      0.0228  │     99.50 %  ┃   2.000e-02  │   00:09 min  ┃┈██████████████████████████▒┈┨      0.0320  │     99.03 %  ┃\n",
      "┃          32  ┃      0.0231  │     99.44 %  ┃   2.000e-02  │   00:08 min  ┃┈██████████████████████████▒┈┨      0.0327  │     98.96 %  ┃\n",
      "┃          33  ┃      0.0229  │     99.49 %  ┃   2.000e-02  │   00:09 min  ┃┈██████████████████████████▒┈┨      0.0322  │     99.05 %  ┃\n",
      "┃          34  ┃      0.0229  │     99.51 %  ┃   2.000e-02  │   00:09 min  ┃┈██████████████████████████▒┈┨      0.0325  │     98.99 %  ┃\n",
      "┃          35  ┃      0.0232  │     99.48 %  ┃   2.000e-02  │   00:09 min  ┃┈██████████████████████████▒┈┨      0.0328  │     99.00 %  ┃\n",
      "┃          36  ┃      0.0231  │     99.50 %  ┃   2.000e-02  │   00:09 min  ┃┈██████████████████████████▒┈┨      0.0323  │     99.08 %  ┃\n",
      "┃          37  ┃      0.0230  │     99.50 %  ┃   2.000e-02  │   00:09 min  ┃┈██████████████████████████▒┈┨      0.0337  │     98.96 %  ┃\n",
      "┃          38  ┃      0.0230  │     99.48 %  ┃   2.000e-02  │   00:09 min  ┃┈██████████████████████████▒┈┨      0.0323  │     99.02 %  ┃\n",
      "┃          39  ┃      0.0226  │     99.52 %  ┃   2.000e-02  │   00:09 min  ┃┈██████████████████████████▒┈┨      0.0330  │     99.00 %  ┃\n",
      "┃          40  ┃      0.0229  │     99.49 %  ┃   2.000e-02  │   00:09 min  ┃┈██████████████████████████▒┈┨      0.0327  │     98.92 %  ┃\n",
      "┃          41  ┃      0.0230  │     99.48 %  ┃   2.000e-02  │   00:09 min  ┃┈██████████████████████████▒┈┨      0.0353  │     98.92 %  ┃\n",
      "┃          42  ┃      0.0228  │     99.49 %  ┃   2.000e-02  │   00:09 min  ┃┈██████████████████████████▒┈┨      0.0332  │     98.92 %  ┃\n",
      "┃          43  ┃      0.0226  │     99.51 %  ┃   2.000e-02  │   00:09 min  ┃┈██████████████████████████▒┈┨      0.0326  │     98.94 %  ┃\n",
      "┃          44  ┃      0.0226  │     99.51 %  ┃   2.000e-02  │   00:09 min  ┃┈██████████████████████████▒┈┨      0.0331  │     98.96 %  ┃\n",
      "┃          45  ┃      0.0225  │     99.49 %  ┃   2.000e-02  │   00:09 min  ┃┈██████████████████████████▒┈┨      0.0316  │     99.04 %  ┃\n",
      "┃          46  ┃      0.0224  │     99.50 %  ┃   2.000e-02  │   00:09 min  ┃┈██████████████████████████▒┈┨      0.0331  │     98.97 %  ┃\n",
      "┃          47  ┃      0.0224  │     99.54 %  ┃   2.000e-02  │   00:09 min  ┃┈██████████████████████████▒┈┨      0.0323  │     99.00 %  ┃\n",
      "┃          48  ┃      0.0204  │     99.57 %  ┃   4.000e-03  │   00:09 min  ┃┈██████████████████████████▒┈┨      0.0307  │     99.09 %  ┃\n",
      "┃          49  ┃      0.0202  │     99.59 %  ┃   4.000e-03  │   00:10 min  ┃┈██████████████████████████▒┈┨      0.0308  │     99.07 %  ┃\n",
      "┃          50  ┃      0.0201  │     99.61 %  ┃   4.000e-03  │   00:09 min  ┃┈██████████████████████████▒┈┨      0.0309  │     99.03 %  ┃\n",
      "┃          51  ┃      0.0201  │     99.61 %  ┃   4.000e-03  │   00:09 min  ┃┈██████████████████████████▒┈┨      0.0306  │     99.07 %  ┃\n",
      "┃          52  ┃      0.0200  │     99.61 %  ┃   4.000e-03  │   00:09 min  ┃┈██████████████████████████▒┈┨      0.0305  │     99.06 %  ┃\n",
      "┃          53  ┃      0.0201  │     99.61 %  ┃   4.000e-03  │   00:09 min  ┃┈██████████████████████████▒┈┨      0.0306  │     99.06 %  ┃\n",
      "┃          54  ┃      0.0200  │     99.59 %  ┃   4.000e-03  │   00:09 min  ┃┈██████████████████████████▒┈┨      0.0303  │     99.10 %  ┃\n",
      "┃          55  ┃      0.0199  │     99.62 %  ┃   4.000e-03  │   00:09 min  ┃┈██████████████████████████▒┈┨      0.0304  │     99.11 %  ┃\n",
      "┃          56  ┃      0.0200  │     99.61 %  ┃   4.000e-03  │   00:09 min  ┃┈██████████████████████████▒┈┨      0.0304  │     99.08 %  ┃\n",
      "┃          57  ┃      0.0201  │     99.59 %  ┃   4.000e-03  │   00:09 min  ┃┈██████████████████████████▒┈┨      0.0306  │     99.09 %  ┃\n",
      "┃          58  ┃      0.0200  │     99.62 %  ┃   4.000e-03  │   00:09 min  ┃┈██████████████████████████▒┈┨      0.0306  │     99.06 %  ┃\n",
      "┃          59  ┃      0.0199  │     99.62 %  ┃   4.000e-03  │   00:09 min  ┃┈██████████████████████████▒┈┨      0.0308  │     99.05 %  ┃\n",
      "┃          60  ┃      0.0200  │     99.60 %  ┃   4.000e-03  │   00:09 min  ┃┈██████████████████████████▒┈┨      0.0307  │     99.08 %  ┃\n",
      "┃          61  ┃      0.0200  │     99.62 %  ┃   4.000e-03  │   00:09 min  ┃┈██████████████████████████▒┈┨      0.0306  │     99.05 %  ┃\n",
      "┃          62  ┃      0.0199  │     99.61 %  ┃   4.000e-03  │   00:09 min  ┃┈██████████████████████████▒┈┨      0.0311  │     99.02 %  ┃\n",
      "┃          63  ┃      0.0200  │     99.62 %  ┃   4.000e-03  │   00:09 min  ┃┈██████████████████████████▒┈┨      0.0307  │     99.04 %  ┃\n",
      "┃          64  ┃      0.0196  │     99.62 %  ┃   8.000e-04  │   00:09 min  ┃┈██████████████████████████▒┈┨      0.0302  │     99.08 %  ┃\n",
      "┃          65  ┃      0.0195  │     99.63 %  ┃   8.000e-04  │   00:09 min  ┃┈██████████████████████████▒┈┨      0.0302  │     99.09 %  ┃\n",
      "┃          66  ┃      0.0195  │     99.62 %  ┃   8.000e-04  │   00:09 min  ┃┈██████████████████████████▒┈┨      0.0302  │     99.08 %  ┃\n",
      "┃          67  ┃      0.0195  │     99.63 %  ┃   8.000e-04  │   00:09 min  ┃┈██████████████████████████▒┈┨      0.0301  │     99.07 %  ┃\n",
      "┃          68  ┃      0.0195  │     99.64 %  ┃   8.000e-04  │   00:09 min  ┃┈██████████████████████████▒┈┨      0.0302  │     99.07 %  ┃\n",
      "┃          69  ┃      0.0194  │     99.63 %  ┃   8.000e-04  │   00:09 min  ┃┈██████████████████████████▒┈┨      0.0302  │     99.11 %  ┃\n",
      "┃          70  ┃      0.0195  │     99.64 %  ┃   8.000e-04  │   00:09 min  ┃┈██████████████████████████▒┈┨      0.0302  │     99.07 %  ┃\n",
      "┃          71  ┃      0.0195  │     99.63 %  ┃   8.000e-04  │   00:09 min  ┃┈██████████████████████████▒┈┨      0.0301  │     99.08 %  ┃\n",
      "┃          72  ┃      0.0195  │     99.63 %  ┃   8.000e-04  │   00:09 min  ┃┈██████████████████████████▒┈┨      0.0301  │     99.09 %  ┃\n",
      "┃          73  ┃      0.0194  │     99.64 %  ┃   8.000e-04  │   00:09 min  ┃┈██████████████████████████▒┈┨      0.0301  │     99.08 %  ┃\n",
      "┃          74  ┃      0.0194  │     99.63 %  ┃   8.000e-04  │   00:09 min  ┃┈██████████████████████████▒┈┨      0.0301  │     99.07 %  ┃\n",
      "┃          75  ┃      0.0194  │     99.64 %  ┃   8.000e-04  │   00:09 min  ┃┈██████████████████████████▒┈┨      0.0302  │     99.09 %  ┃\n",
      "┃          76  ┃      0.0194  │     99.64 %  ┃   8.000e-04  │   00:09 min  ┃┈██████████████████████████▒┈┨      0.0301  │     99.11 %  ┃\n",
      "┃          77  ┃      0.0194  │     99.64 %  ┃   8.000e-04  │   00:09 min  ┃┈██████████████████████████▒┈┨      0.0302  │     99.09 %  ┃\n",
      "┃          78  ┃      0.0194  │     99.64 %  ┃   8.000e-04  │   00:09 min  ┃┈██████████████████████████▒┈┨      0.0301  │     99.07 %  ┃\n",
      "┃          79  ┃      0.0194  │     99.63 %  ┃   8.000e-04  │   00:09 min  ┃┈██████████████████████████▒┈┨      0.0301  │     99.07 %  ┃\n"
     ]
    }
   ],
   "source": [
    "import torch\n",
    "\n",
    "import os\n",
    "import sys\n",
    "module_path = os.path.abspath(os.path.join('..'))\n",
    "if module_path not in sys.path:\n",
    "    sys.path.append(module_path)    \n",
    "from utility.log import Log\n",
    "from utility.initialize import initialize\n",
    "from utility.step_lr import StepLR\n",
    "from utility.bypass_bn import enable_running_stats, disable_running_stats\n",
    "\n",
    "import sys; sys.path.append(\"..\")\n",
    "from sam import SAM\n",
    "\n",
    "\n",
    "args = dict()\n",
    "args['adaptive'] = True\n",
    "args['batch_size'] = 128\n",
    "args['epochs'] = 80\n",
    "# 라벨 스무딩, 레이블을 그대로 사용하는 것이 아니라 조금 smooth하게 만들어서 정규화를 시키는 것\n",
    "args['label_smoothing'] = 0.1\n",
    "# 초기 학습률\n",
    "args['learning_rate'] = 0.1\n",
    "# Momentum은 Local Minimum에 빠지는 경우를 대처하기 위해 적용.\n",
    "args['momentum'] = 0.9\n",
    "# SAM의 gradient 전개식 계산에 쓰일 파라미터, 논문에 따르면 경험적으로 ρ = 2로 하여 L2-norm 계산한 것이 최적의 결과를 냄.\n",
    "args['rho'] = 2.0\n",
    "args['weight_decay'] = 0.0005\n",
    "\n",
    "initialize(args, seed=42)\n",
    "\n",
    "log = Log(log_each=10)\n",
    "model = LeNet_5().to(device)\n",
    "print(model)\n",
    "\n",
    "base_optimizer = torch.optim.SGD\n",
    "optimizer = SAM(model.parameters(), base_optimizer, rho=args['rho'], adaptive=args['adaptive'], lr=args['learning_rate'], momentum=args['momentum'], weight_decay=args['weight_decay'])\n",
    "scheduler = StepLR(optimizer, args['learning_rate'], args['epochs'])\n",
    "\n",
    "epoch_loss = {\n",
    "    'train':[],\n",
    "    'valid':[]\n",
    "}\n",
    "\n",
    "for epoch in range(args['epochs']):\n",
    "    model.train()\n",
    "    log.train(len_dataset=len(train_dl))\n",
    "\n",
    "    for batch in train_dl:\n",
    "        inputs, targets = (b.to(device) for b in batch)\n",
    "\n",
    "        # first forward-backward step\n",
    "        enable_running_stats(model)\n",
    "        predictions = model(inputs)\n",
    "        loss = smooth_crossentropy(predictions, targets, smoothing=args['label_smoothing'])\n",
    "        loss.mean().backward()\n",
    "        optimizer.first_step(zero_grad=True)\n",
    "\n",
    "        # second forward-backward step\n",
    "        disable_running_stats(model)\n",
    "        smooth_crossentropy(model(inputs), targets, smoothing=args['label_smoothing']).mean().backward()\n",
    "        optimizer.second_step(zero_grad=True)\n",
    "\n",
    "        with torch.no_grad():\n",
    "            correct = torch.argmax(predictions.data, 1) == targets\n",
    "            log(model, loss.cpu(), correct.cpu(), scheduler.lr())\n",
    "            scheduler(epoch)\n",
    "\n",
    "    model.eval()\n",
    "    log.eval(len_dataset=len(val_dl))\n",
    "\n",
    "    with torch.no_grad():\n",
    "        for batch in val_dl:\n",
    "            inputs, targets = (b.to(device) for b in batch)\n",
    "\n",
    "            predictions = model(inputs)\n",
    "            loss = smooth_crossentropy(predictions, targets)\n",
    "            correct = torch.argmax(predictions, 1) == targets\n",
    "\n",
    "            log(model, loss.cpu(), correct.cpu())\n",
    "\n",
    "log.flush()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0, 0.5, 'Loss')"
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "loss_history = log.get_epoch_loss_history()\n",
    "\n",
    "epochs = range(0, args['epochs'])\n",
    "plt.plot(list(loss_history['train']), label='Training Loss')\n",
    "#plt.plot(epochs, loss_history['valid'], label='Validation Loss')\n",
    "\n",
    "plt.title('Training Loss')\n",
    "plt.xlabel('Epochs')\n",
    "plt.ylabel('Loss')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "위에서 학습시킨 모델 저장"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Cannot save into file. The file already exists..\n"
     ]
    }
   ],
   "source": [
    "import os.path\n",
    "\n",
    "number_of_epochs = args['epochs']\n",
    "batch_size = args['batch_size']\n",
    "\n",
    "if os.path.isfile(f'./model/epochs-{number_of_epochs}/LeNet-5_SAM_b{batch_size}.pth') == False:\n",
    "    torch.save(model, f'./model/epochs-{number_of_epochs}/LeNet-5_SAM_b{batch_size}.pth')\n",
    "else:\n",
    "    print('Cannot save into file. The file already exists..')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 4. Test 데이터셋 준비 및 테스트"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "저장된 모델 불러오기"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "LeNet_5(\n",
       "  (conv1): Conv2d(1, 6, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2))\n",
       "  (conv2): Conv2d(6, 16, kernel_size=(5, 5), stride=(1, 1))\n",
       "  (conv3): Conv2d(16, 120, kernel_size=(5, 5), stride=(1, 1))\n",
       "  (fc1): Linear(in_features=120, out_features=84, bias=True)\n",
       "  (fc2): Linear(in_features=84, out_features=10, bias=True)\n",
       ")"
      ]
     },
     "execution_count": 55,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import torch\n",
    "\n",
    "model = torch.load('./model/epochs-80/LeNet-5_SAM_b128.pth')\n",
    "model.to(device)\n",
    "model.eval()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "정확도 측정 함수 정의"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [],
   "source": [
    "def check_accuracy(loader, model):\n",
    "    \n",
    "    correct_list=[]\n",
    "    wrong_list=[]\n",
    "\n",
    "    total = 0\n",
    "    correct = 0  # 정답 개수를 기록하기 위한 변수\n",
    "\n",
    "    model.eval()\n",
    "    with torch.no_grad():\n",
    "        for inputs, labels in loader:\n",
    "            inputs, labels = inputs.to(device), labels.to(device)\n",
    "\n",
    "            infer = model(inputs)\n",
    "            predicted = infer.argmax(dim=1)\n",
    "            total += labels.size(0)\n",
    "            correct += (predicted == labels).sum().item()\n",
    "\n",
    "            # 올바르게 분류된 샘플과 잘못 분류된 샘플을 구분하여 저장\n",
    "            for i in range(len(predicted)):\n",
    "                if predicted[i] == labels[i]:\n",
    "                    correct_list.append((inputs[i], labels[i], predicted[i]))\n",
    "                else:\n",
    "                    wrong_list.append((inputs[i], labels[i], predicted[i]))\n",
    "\n",
    "    # 정확도 출력\n",
    "    accuracy = 100 * correct / total\n",
    "    print(f'Accuracy on the test set: {accuracy:.3f}%')\n",
    "\n",
    "    # 옳게 분류된 샘플들과 잘못된 분류된 샘플들을 반환\n",
    "    return correct_list, wrong_list"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### TEST1 : 변형이 없는 테스트 데이터셋을 사용하여 테스트"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [],
   "source": [
    "from torchvision import datasets\n",
    "from torchvision import datasets, transforms\n",
    "from torch.utils.data import DataLoader\n",
    "\n",
    "path2data = './data'\n",
    "data_transform = transforms.Compose([\n",
    "            #transforms.Resize((32, 32)),\n",
    "            transforms.ToTensor(),\n",
    "])\n",
    "\n",
    "# MNIST test dataset 불러오기\n",
    "test_data = datasets.MNIST(path2data, train=False, download=True, transform=data_transform)\n",
    "test_dl = DataLoader(test_data, batch_size=128)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "정확도 출력"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy on the test set: 99.000%\n"
     ]
    }
   ],
   "source": [
    "correct_samples, wrong_samples = check_accuracy(test_dl, model)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "틀리게 분류된 샘플 랜덤 선택 출력"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Wrongly classified samples:\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x400 with 10 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import random\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "# 잘못 분류된 샘플 중 10개를 랜덤하게 선택하여 출력\n",
    "print(\"Wrongly classified samples:\")\n",
    "plt.figure(figsize=(10, 4))  # 전체 그림의 크기 설정\n",
    "for i in range(10):\n",
    "    sample = random.choice(wrong_samples)\n",
    "    input_img, true_label, predicted_label = sample\n",
    "    plt.subplot(2, 5, i + 1)  # 2행 5열의 subplot 중 i+1 번째에 그림을 출력\n",
    "    plt.imshow(input_img.squeeze().cpu().numpy(), cmap='gray')\n",
    "    plt.title(f'True: {true_label}, Predicted: {predicted_label}')\n",
    "    plt.axis('off')  # 축 제거\n",
    "plt.tight_layout()  # subplot 간격 자동 조절\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "맞게 분류된 샘플 랜덤 선택 출력"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Correctly classified samples:\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x400 with 10 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# 정확하게 분류된 샘플 중 10개를 랜덤하게 선택하여 출력\n",
    "print(\"Correctly classified samples:\")\n",
    "plt.figure(figsize=(10, 4))  # 전체 그림의 크기 설정\n",
    "for i in range(10):\n",
    "    sample = random.choice(correct_samples)\n",
    "    input_img, true_label, predicted_label = sample\n",
    "    plt.subplot(2, 5, i + 1)  # 2행 5열의 subplot 중 i+1 번째에 그림을 출력\n",
    "    plt.imshow(input_img.squeeze().cpu().numpy(), cmap='gray')\n",
    "    plt.title(f'True: {true_label}, Predicted: {predicted_label}')\n",
    "    plt.axis('off')  # 축 제거\n",
    "plt.tight_layout()  # subplot 간격 자동 조절\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### TEST2 : 표준편차 0.4 가우시안 노이즈를 적용한 테스트 데이터셋을 사용하여 테스트"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "가우시안 노이즈를 적용하기 위한 클래스를 정의 (랜덤 시드를 지정함)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 가우시안 노이즈를 적용하기 위한 클래스를 정의 (랜덤 시드를 지정함)\n",
    "class AddGaussianNoise(object):\n",
    "    def __init__(self, mean=0., std=1.):\n",
    "        self.std = std\n",
    "        self.mean = mean\n",
    "        \n",
    "    def __call__(self, tensor):\n",
    "        torch.manual_seed(0)\n",
    "        return tensor + torch.randn(tensor.size()) * self.std + self.mean\n",
    "    \n",
    "    def __repr__(self):\n",
    "        return self.__class__.__name__ + '(mean={0}, std={1})'.format(self.mean, self.std)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "표준편차 0.4 가우시안 노이즈를 추가하고 노이즈를 추가한 Sample 이미지 확인"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from torchvision import utils\n",
    "\n",
    "noisy_transform=transforms.Compose([\n",
    "    #transforms.Resize((32, 32)),\n",
    "    transforms.ToTensor(),\n",
    "    transforms.Normalize((0.1307,), (0.3081,)),\n",
    "    AddGaussianNoise(mean=0., std=0.4),\n",
    "])\n",
    "\n",
    "# MNIST test dataset 불러오기\n",
    "gau_04_test_data = datasets.MNIST(path2data, train=False, download=True, transform=noisy_transform)\n",
    "gau_04_test_dl = DataLoader(gau_04_test_data, batch_size=128)\n",
    "# test data를 추출합니다.\n",
    "x_test, y_test = gau_04_test_data.data, gau_04_test_data.targets\n",
    "\n",
    "if len(x_test.shape) == 3:\n",
    "    x_test = x_test.unsqueeze(1)\n",
    "\n",
    "# tensor를 image로 변경하는 함수를 정의합니다.\n",
    "def show(img):\n",
    "    # tensor를 numpy array로 변경합니다.\n",
    "    npimg = img.numpy()\n",
    "    # C*H*W를 H*W*C로 변경합니다.\n",
    "    npimg_tr = npimg.transpose((1,2,0))\n",
    "    plt.imshow(npimg_tr, interpolation='nearest')\n",
    "\n",
    "for batch in gau_04_test_dl:\n",
    "    inputs, targets = (b.to(device) for b in batch)\n",
    "    x_test_grid = utils.make_grid(inputs, nrow=8, padding=2)\n",
    "    show(x_test_grid.to('cpu'))\n",
    "    break"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "정확도 출력 (표준편차 0.4 가우시안 노이즈 적용)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy on the test set: 98.600%\n"
     ]
    }
   ],
   "source": [
    "correct_samples, wrong_samples = check_accuracy(gau_04_test_dl, model)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "틀리게 분류된 샘플 랜덤 선택 출력"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Wrongly classified samples:\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x400 with 10 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import random\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "# 잘못 분류된 샘플 중 10개를 랜덤하게 선택하여 출력\n",
    "print(\"Wrongly classified samples:\")\n",
    "plt.figure(figsize=(10, 4))  # 전체 그림의 크기 설정\n",
    "for i in range(10):\n",
    "    sample = random.choice(wrong_samples)\n",
    "    input_img, true_label, predicted_label = sample\n",
    "    plt.subplot(2, 5, i + 1)  # 2행 5열의 subplot 중 i+1 번째에 그림을 출력\n",
    "    plt.imshow(input_img.squeeze().cpu().numpy(), cmap='gray')\n",
    "    plt.title(f'True: {true_label}, Predicted: {predicted_label}')\n",
    "    plt.axis('off')  # 축 제거\n",
    "plt.tight_layout()  # subplot 간격 자동 조절\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "맞게 분류된 샘플 랜덤 선택 출력"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Correctly classified samples:\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x400 with 10 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# 정확하게 분류된 샘플 중 10개를 랜덤하게 선택하여 출력\n",
    "print(\"Correctly classified samples:\")\n",
    "plt.figure(figsize=(10, 4))  # 전체 그림의 크기 설정\n",
    "for i in range(10):\n",
    "    sample = random.choice(correct_samples)\n",
    "    input_img, true_label, predicted_label = sample\n",
    "    plt.subplot(2, 5, i + 1)  # 2행 5열의 subplot 중 i+1 번째에 그림을 출력\n",
    "    plt.imshow(input_img.squeeze().cpu().numpy(), cmap='gray')\n",
    "    plt.title(f'True: {true_label}, Predicted: {predicted_label}')\n",
    "    plt.axis('off')  # 축 제거\n",
    "plt.tight_layout()  # subplot 간격 자동 조절\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### TEST3 : 표준편차 0.6 가우시안 노이즈를 적용한 테스트 데이터셋을 사용하여 테스트"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "표준편차 0.6 가우시안 노이즈를 추가하고 노이즈를 추가한 Sample 이미지 확인"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from torchvision import utils\n",
    "\n",
    "noisy_transform=transforms.Compose([\n",
    "    #transforms.Resize((32, 32)),\n",
    "    transforms.ToTensor(),\n",
    "    transforms.Normalize((0.1307,), (0.3081,)),\n",
    "    AddGaussianNoise(mean=0., std=0.6),\n",
    "])\n",
    "\n",
    "# MNIST test dataset 불러오기\n",
    "gau_06_test_data = datasets.MNIST(path2data, train=False, download=True, transform=noisy_transform)\n",
    "gau_06_test_dl = DataLoader(gau_06_test_data, batch_size=128)\n",
    "# test data를 추출합니다.\n",
    "x_test, y_test = gau_06_test_data.data, gau_06_test_data.targets\n",
    "\n",
    "if len(x_test.shape) == 3:\n",
    "    x_test = x_test.unsqueeze(1)\n",
    "\n",
    "# tensor를 image로 변경하는 함수를 정의합니다.\n",
    "def show(img):\n",
    "    # tensor를 numpy array로 변경합니다.\n",
    "    npimg = img.numpy()\n",
    "    # C*H*W를 H*W*C로 변경합니다.\n",
    "    npimg_tr = npimg.transpose((1,2,0))\n",
    "    plt.imshow(npimg_tr, interpolation='nearest')\n",
    "\n",
    "for batch in gau_06_test_dl:\n",
    "    inputs, targets = (b.to(device) for b in batch)\n",
    "    x_test_grid = utils.make_grid(inputs, nrow=8, padding=2)\n",
    "    show(x_test_grid.to('cpu'))\n",
    "    break\n",
    "\n",
    "\n",
    "# 아래처럼 x_test로 접근하면 transform을 거치기 전 원본 데이터가 시각화 되므로 위의 코드처럼 순환하여 시각화하는 방식을 사용\n",
    "#x_test_grid = utils.make_grid(x_test[:40], nrow=8, padding=2)\n",
    "#show(x_test_grid)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "정확도 출력 (표준편차 0.6 가우시안 노이즈 적용)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy on the test set: 98.110%\n"
     ]
    }
   ],
   "source": [
    "correct_samples, wrong_samples = check_accuracy(gau_06_test_dl, model)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "틀리게 분류된 샘플 랜덤 선택 출력"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Wrongly classified samples:\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x400 with 10 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import random\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "# 잘못 분류된 샘플 중 10개를 랜덤하게 선택하여 출력\n",
    "print(\"Wrongly classified samples:\")\n",
    "plt.figure(figsize=(10, 4))  # 전체 그림의 크기 설정\n",
    "for i in range(10):\n",
    "    sample = random.choice(wrong_samples)\n",
    "    input_img, true_label, predicted_label = sample\n",
    "    plt.subplot(2, 5, i + 1)  # 2행 5열의 subplot 중 i+1 번째에 그림을 출력\n",
    "    plt.imshow(input_img.squeeze().cpu().numpy(), cmap='gray')\n",
    "    plt.title(f'True: {true_label}, Predicted: {predicted_label}')\n",
    "    plt.axis('off')  # 축 제거\n",
    "plt.tight_layout()  # subplot 간격 자동 조절\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "맞게 분류된 샘플 랜덤 선택 출력"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Correctly classified samples:\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x400 with 10 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# 정확하게 분류된 샘플 중 10개를 랜덤하게 선택하여 출력\n",
    "print(\"Correctly classified samples:\")\n",
    "plt.figure(figsize=(10, 4))  # 전체 그림의 크기 설정\n",
    "for i in range(10):\n",
    "    sample = random.choice(correct_samples)\n",
    "    input_img, true_label, predicted_label = sample\n",
    "    plt.subplot(2, 5, i + 1)  # 2행 5열의 subplot 중 i+1 번째에 그림을 출력\n",
    "    plt.imshow(input_img.squeeze().cpu().numpy(), cmap='gray')\n",
    "    plt.title(f'True: {true_label}, Predicted: {predicted_label}')\n",
    "    plt.axis('off')  # 축 제거\n",
    "plt.tight_layout()  # subplot 간격 자동 조절\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### TEST4 : 표준편차 0.8 가우시안 노이즈를 적용한 테스트 데이터셋을 사용하여 테스트"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "표준편차 0.8 가우시안 노이즈를 추가하고 노이즈를 추가한 Sample 이미지 확인"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from torchvision import utils\n",
    "\n",
    "noisy_transform=transforms.Compose([\n",
    "    #transforms.Resize((32, 32)),\n",
    "    transforms.ToTensor(),\n",
    "    transforms.Normalize((0.1307,), (0.3081,)),\n",
    "    AddGaussianNoise(mean=0., std=0.8),\n",
    "])\n",
    "\n",
    "# MNIST test dataset 불러오기\n",
    "gau_08_test_data = datasets.MNIST(path2data, train=False, download=True, transform=noisy_transform)\n",
    "gau_08_test_dl = DataLoader(gau_08_test_data, batch_size=128)\n",
    "# test data를 추출합니다.\n",
    "x_test, y_test = gau_08_test_data.data, gau_08_test_data.targets\n",
    "\n",
    "if len(x_test.shape) == 3:\n",
    "    x_test = x_test.unsqueeze(1)\n",
    "\n",
    "# tensor를 image로 변경하는 함수를 정의합니다.\n",
    "def show(img):\n",
    "    # tensor를 numpy array로 변경합니다.\n",
    "    npimg = img.numpy()\n",
    "    # C*H*W를 H*W*C로 변경합니다.\n",
    "    npimg_tr = npimg.transpose((1,2,0))\n",
    "    plt.imshow(npimg_tr, interpolation='nearest')\n",
    "\n",
    "for batch in gau_06_test_dl:\n",
    "    inputs, targets = (b.to(device) for b in batch)\n",
    "    x_test_grid = utils.make_grid(inputs, nrow=8, padding=2)\n",
    "    show(x_test_grid.to('cpu'))\n",
    "    break"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "정확도 출력 (표준편차 0.8 가우시안 노이즈 적용)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy on the test set: 97.230%\n"
     ]
    }
   ],
   "source": [
    "correct_samples, wrong_samples = check_accuracy(gau_08_test_dl, model)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "틀리게 분류된 샘플 랜덤 선택 출력"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Wrongly classified samples:\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA88AAAGJCAYAAAC0FQGbAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8pXeV/AAAACXBIWXMAAA9hAAAPYQGoP6dpAACXfklEQVR4nO3dd3hVZbo28CcESEgChJCEToAktNA7CITelerIzAjIoI5HHIqOWEaPOCLiWEYdlQFHERgLSBMYpAkISO+9E0IndJJA6vv94SGf8b0fVlbYITvO/buuc11nbl82K3uvZ79rkeTePsYYI0RERERERESkKpTfB0BERERERETk7XjzTEREREREROSAN89EREREREREDnjzTEREREREROSAN89EREREREREDnjzTEREREREROSAN89EREREREREDnjzTEREREREROSAN89EREREREREDnjzXIB9/vnn4uPjI3FxcVlZu3btpF27dvl2TL+EjpEor3AmiLJbtWqV+Pj4yKpVq7KyRx55RKpUqZJvx/RL6BiJ8gr3CaLsuE+4k+ObZx8fnxz9nzd8UUiVKlXg8T7xxBMee8zw8HBp06aNzJ0714NHnveSk5Nl7NixXvva7d+/X7p16yZBQUESEhIigwYNkoSEhPw+rAI/E4mJiTJq1CipWLGi+Pn5Sa1atWTixIl39Zicibz3ySefSGxsrJQpU0b8/PykatWqMnToUK+4yCroMzF69Ghp1KiRhISESEBAgNSqVUvGjh0riYmJuX7Mdu3aZfvaQ0JCpGnTpvLZZ59JZmamB48+740fP17mzZuX34cBLV++XNq3by+hoaESHBwszZo1k+nTp+f3YRX4meA+ofPWfSIzM1M+//xzeeCBB6RSpUoSGBgoderUkXHjxsmtW7fy+/AK/Exwn7gzb90n5syZIw899JBUq1ZNAgICpEaNGvLMM8/I1atXXT9W4Zwu/OUmNG3aNFm2bJmV16pVy/VB3CsNGjSQZ555JltWvXp1jz3mmTNnZNKkSdKvXz+ZOHHiXd2Y59bSpUtd/5nk5GR59dVXRUS86l9eRUROnTolbdu2lZIlS8r48eMlMTFR3n77bdm9e7ds2rRJihYtmm/HVpBnIiMjQ7p27SpbtmyR4cOHS3R0tCxZskSefPJJuXLlirz44ou5fmzORN7avn27VK1aVR544AEpVaqUHD9+XD755BNZuHCh7Ny5U8qXL59vx1aQZ0JEZPPmzdKmTRsZOnSo+Pv7y/bt22XChAmyfPlyWb16tRQqlLsf1qpYsaK88cYbIiKSkJAg06ZNk2HDhsmhQ4dkwoQJnvwScuSTTz7J1QXZ+PHjZcCAAdKnTx/PH9RdmD9/vvTp00datmwpY8eOFR8fH5k5c6YMHjxYLl68KKNHj863YyvIM8F94s68dZ9ITk6WoUOHSosWLeSJJ56Q8PBwWb9+vbzyyivy/fffy4oVK8THxyffjq8gz4QI9wkn3rpPPP7441K+fHl5+OGHpXLlyrJ792758MMPZdGiRbJt2zYpVqxYzh/M5NLw4cNNTv54UlJSbv8Kj4qIiDA9e/bM88c8e/asCQwMNNWrV1f/XFpamklJSbnrv3/KlClGRMzx48fv6nESEhKMiJhXXnnlro/pl+72GP/nf/7HFCtWzJw4cSIrW7ZsmRERM2nSJA8dpWcUpJmYOXOmERHz6aefZsv79+9v/P39zfnz53P1uJwJZ546xp/bsmWLERHzxhtveOwxPaEgzYTm7bffNiJi1q9fn6s/Hxsba2JiYrJlSUlJpmLFiiYwMNCkpqbCP5eRkWFu3ryZq7/z51auXGlExKxcufKuHyswMNAMGTLkrh/nl+72GDt37mzKly9vbt26lZWlpaWZyMhIU69ePQ8dpWcUpJngPnFn3rpPpKSkmB9//NHKX331VSMiZtmyZR44Qs8pSDOh4T7x/3nrPoH+3NSpU42ImE8++cTVY3n0d57btWsnderUka1bt0rbtm0lICAg618mfXx8ZOzYsdafqVKlijzyyCPZsqtXr8qoUaOkUqVK4ufnJ1FRUfLmm29a/wJy9uxZOXDggKSlpeX4GFNTUyUpKcn115ZTZcuWlVq1asnx48dFRCQuLk58fHzk7bfflvfee08iIyPFz89P9u3bJyIiBw4ckAEDBkhISIj4+/tLkyZNZP78+dbj7t27Vzp06CDFihWTihUryrhx4+C/CKHf27l165aMHTtWqlevLv7+/lKuXDnp16+fHD16VOLi4iQsLExERF599dWsHxn5+Wvl6WO8du2aHDhwQK5du+b4fM6ePVt69eollStXzso6deok1atXl5kzZzr++fzmrTOxZs0aEREZOHBgtnzgwIFy69Yt+fbbb11+pTrOhGdnArn9e0m5+fGje81bZ0KTF89tQECAtGjRQpKSkrJ+BcXHx0eeeuop+eKLLyQmJkb8/Pxk8eLFIiJy+vRp+cMf/pD1o/oxMTHy2WefWY976tQp6dOnjwQGBkp4eLiMHj1aUlJSrHXod9kyMzPl/fffl7p164q/v7+EhYVJt27dZMuWLVnHl5SUJFOnTs2aiZ+/Jp4+xuTkZDlw4IBcvHjR8fm8fv26lCpVSvz8/LKywoULS2hoqLvvJuQTb50J7hMFc58oWrSotGrVysr79u0rIj/9Kpy389aZ0HCf8P59Av10SG5nIsc/tp1Tly5dku7du8vAgQPl4YcfljJlyrj688nJyRIbGyunT5+WP/7xj1K5cmVZt26dvPDCC3L27Fl57733sta+8MILMnXqVDl+/HiOfql9xYoVEhAQIBkZGRIRESGjR4+WkSNHuvwK7ywtLU1OnjwppUuXzpZPmTJFbt26JY8//rj4+flJSEiI7N27V+677z6pUKGCPP/88xIYGCgzZ86UPn36yOzZs7Ne1HPnzkn79u0lPT09a93kyZNzdFGQkZEhvXr1ku+//14GDhwoI0eOlBs3bsiyZctkz5490qlTJ5k4caL8z//8j/Tt21f69esnIiL16tUTEcmTY5w7d64MHTpUpkyZYr3R/dzp06flwoUL0qRJE+u/NWvWTBYtWuT49XsDb5yJlJQU8fX1tX7sPSAgQEREtm7dKo899pir49RwJjw3Ez936dIlycjIkPj4ePnrX/8qIiIdO3bM0Z/Nb944E7elp6fL1atXJTU1Vfbs2SMvvfSSFC9eXJo1a+byq7yzY8eOia+vrwQHB2dlK1askJkzZ8pTTz0loaGhUqVKFTl//ry0aNEi66IpLCxMvvvuOxk2bJhcv35dRo0aJSIiN2/elI4dO0p8fLyMGDFCypcvL9OnT5cVK1bk6HiGDRsmn3/+uXTv3l0effRRSU9PlzVr1siGDRukSZMmMn36dHn00UelWbNm8vjjj4uISGRkpIhInhzjpk2bpH379vLKK6/AC+Wfa9eunbz55pvy8ssvy5AhQ8THx0e+/PJL2bJlS4H4R1YR75wJ7hMFf5/4uXPnzomISGhoqOs/mx+8cSZu4z5R8PYJJNczkavvfRv8YxaxsbFGRMw///lPa70oP9oSERGR7dv7r732mgkMDDSHDh3Ktu755583vr6+Jj4+PisbMmRIjn+s5f777zdvvvmmmTdvnvn0009NmzZtjIiYMWPGOP5ZTUREhOnSpYtJSEgwCQkJZufOnWbgwIFGRMyf/vQnY4wxx48fNyJiSpQoYS5cuJDtz3fs2NHUrVs324+aZWZmmlatWpno6OisbNSoUUZEzMaNG7OyCxcumJIlS1pff2xsrImNjc3635999pkREfPuu+9ax5+ZmWmMufOPHuXFMd7+caQpU6ZYf9/Pbd682YiImTZtmvXfnn32WSMi2Y4rvxWkmXjnnXeMiJg1a9ZYjykiplevXnf88xrORN7OxM/5+fkZETEiYkqXLm0++OCDHP/Ze6UgzcRt69evz3peRcTUqFHjrn6ULTY21tSsWTNrJvbv329GjBhhRMTcf//9WetExBQqVMjs3bs3258fNmyYKVeunLl48WK2fODAgaZkyZImOTnZGGPMe++9Z0TEzJw5M2tNUlKSiYqKsn7UbciQISYiIiLrf69YscKIiBkxYoR1/Ldnwhj9x/Hy4hhv/4heTn4kNjEx0fzmN78xPj4+Wa9bQECAmTdvnuOfvdcK0kxwnyj4+8TPderUyZQoUcJcuXIlV38+rxSkmbiN+0R2BWGfQIYNG2Z8fX2tc8SJx2+e/fz84O+k5PRkr1evnunWrVvWCXT7/5YvX25ExPz73//O7SFnk5mZabp27WoKFy5sTp48mavHiIiIyDY8ImJ8fX3NoEGDsk6C2xvA0KFDs/3ZS5cuGR8fH/Paa69ZX+vt30s5deqUMcaY6tWrmxYtWlh//5NPPum4AfTs2dOEhoaatLQ09evQNoC8OsacWr16tRERM2PGDOu/vfzyy0ZEvGoTKEgzcfbsWVOyZEkTHR1tli5dao4fP24mTZpkSpQoYUTEdOzY0fVj3j52zkTezcTPrVixwixatMi88847pmHDhl73+87GFKyZuO3atWtm2bJlZt68eWbMmDGmUaNGZsGCBbl+vNsXgT//Px8fH9OzZ0+TkJCQtU5ETPv27bP92czMTBMcHGwef/xx62u9fSG9du1aY4wxXbp0MeXKlct2EWOMMX/7298cL4qGDx9ufHx8zKVLl+74taCLorw6RjfS0tLMSy+9ZB588EHz1VdfmX//+9+mbdu2JigoKNe/g5hXCtJMcJ8o+PvEba+//roREfPxxx975PE8qSDNxG3cJ3Teuk/80hdffGFEcvdNVI//2HaFChXuqgH58OHDsmvXrqzfJfmlCxcu5Pqxf87Hx0dGjx4tS5YskVWrVsnDDz+cq8dp3ry5jBs3Tnx8fLIq63/+4xW3Va1aNdv/PnLkiBhj5OWXX5aXX34ZPvaFCxekQoUKcuLECWnevLn132vUqOF4fEePHpUaNWpI4cLuX+p7dYya2z+2hH7P4fbHLRSE32fzxpkoW7aszJ8/XwYNGiRdunQREZESJUrIP/7xDxkyZIgEBQXl+ng5E3k3Ez/Xvn17ERHp3r279O7dW+rUqSNBQUHy1FNPeeTx85I3zsRtJUqUkE6dOomISO/eveXLL7+U3r17y7Zt26R+/fq5eswqVarIJ598Ij4+PuLv7y/R0dESHh5urfvlTCQkJMjVq1dl8uTJMnnyZPjYt7/WEydOSFRUlNWim9OZKF++vISEhOT0S7rnx3gnTz31lGzYsEG2bduW1XT7m9/8RmJiYmTkyJGycePGu3r8e8EbZ4L7RMHfJ0REZsyYIS+99JIMGzZM/ud//sdjj5vXvHEmbuM+4Y437BM/t2bNGhk2bJh07dpVXn/9ddd/3uM3z25vZjIyMrL978zMTOncubOMGTMGrr/bj5b6uUqVKomIyOXLl3P9GKGhoVkDdCe/fF5ulxX8+c9/lq5du8I/ExUVlevj8oT8PsZy5cqJyE9FDr909uxZCQkJyVYQ4628dSbatm0rx44dk927d0tSUpLUr19fzpw5c1ePKcKZyI9jjIyMlIYNG8oXX3xRIG6evXUmkH79+smgQYPk66+/zvVFUWBg4F3NxMMPPyxDhgyBf+b271jml/w+xtTUVPn0009lzJgx2T4ipkiRItK9e3f58MMPJTU1NV8/1jAnvHUmuE+4503HuGzZMhk8eLD07NlT/vnPf96Tv9NTvHUmEO4Td+ZNx7hz50554IEHpE6dOjJr1qxc/QOZx2+eNaVKlbJa6FJTU60bo8jISElMTMzRCXS3jh07JiKi/qtUXqpWrZqI/LTBO32tERERcvjwYSs/ePCg498TGRkpGzdulLS0NClSpAhco33e3706Rk2FChUkLCwsq8Xv5zZt2iQNGjTI9WN7A2+YCV9f32zP4/Lly0VE7sn8/RJn4u7cvHkT/pRGQeINM/FLKSkpkpmZmesm9LsRFhYmxYsXl4yMjBydb3v27BFjTLbzN6czsWTJErl8+fIdv6uA5uJeHaPm0qVLkp6ebl04i/xUQpWZmQn/W0HhDTPBfaJg7hMbN26Uvn37SpMmTWTmzJm5uknwRt4wE7/EfeL/88Z94rajR49Kt27dJDw8XBYtWpTrn57x6EdV3UlkZKSsXr06WzZ58mRrU/vNb34j69evlyVLlliPcfXqVUlPT8/63zmtlr98+bL196SlpcmECROkaNGiWT/+eC+Fh4dLu3btZNKkSfA7q7dr6UVEevToIRs2bJBNmzZl++9ffPGF49/Tv39/uXjxonz44YfWfzPGiMj/b8785ZtRXh2jm4/l6d+/vyxcuFBOnjyZlX3//fdy6NAhefDBBx3/vDfLz5lAEhIS5M0335R69erly0URZ8J5JtLT0+XKlStWvmnTJtm9ezdspi9I8nMmrl69Ctf861//EhHJl+fW19dX+vfvL7Nnz5Y9e/ZY//2X59uZM2dk1qxZWVlycrL6I3I/179/fzHGyKuvvmr9t9szIfLTd0Z+ORN5dYw5/QiS8PBwCQ4Olrlz50pqampWnpiYKAsWLJCaNWsWiF/v0XCfyI77RM6unfbv3y89e/aUKlWqyMKFCwv0DPwS94nsuE/k7KOqzp07J126dJFChQrJkiVL7uobp/fsn6EeffRReeKJJ6R///7SuXNn2blzpyxZssSqB3/22Wdl/vz50qtXL3nkkUekcePGkpSUJLt375ZZs2ZJXFxc1p/JabX8/PnzZdy4cTJgwACpWrWqXL58Wb788kvZs2ePjB8/XsqWLZu1Ni4uTqpWrSpDhgyRzz//PC+eiiwfffSRtG7dWurWrSuPPfaYVKtWTc6fPy/r16+XU6dOyc6dO0VEZMyYMTJ9+nTp1q2bjBw5MuujDCIiImTXrl13/DsGDx4s06ZNk6efflo2bdokbdq0kaSkJFm+fLk8+eST0rt3bylWrJjUrl1bZsyYIdWrV5eQkBCpU6eO1KlTJ0+O0c3HLbz44ovyzTffSPv27WXkyJGSmJgob731ltStW1eGDh2a+yffC+TnTIiIxMbGSsuWLSUqKkrOnTsnkydPlsTERFm4cGG2H3/kTHjPTCQmJkqlSpXkoYcekpiYGAkMDJTdu3fLlClTpGTJkurv1xUU+TkTq1atkhEjRsiAAQMkOjpaUlNTZc2aNTJnzhxp0qSJ1Yvh4+MjsbGxsmrVKk8/DdlMmDBBVq5cKc2bN5fHHntMateuLZcvX5Zt27bJ8uXLs37t6LHHHpMPP/xQBg8eLFu3bpVy5crJ9OnTsy7w76R9+/YyaNAg+eCDD+Tw4cPSrVs3yczMlDVr1kj79u2zfhWgcePGsnz5cnn33XelfPnyUrVqVWnevHmeHGNOP4LE19dX/vznP8tLL70kLVq0kMGDB0tGRoZ8+umncurUKfn3v/+di2fde3CfsHGfuPM+cePGDenatatcuXJFnn32WfnPf/6T7b9HRkZKy5Ytc/HMewfuEzbuE84fVdWtWzc5duyYjBkzRtauXStr167N+m9lypSRzp075/TpFo+3bcfExMD1GRkZ5rnnnjOhoaEmICDAdO3a1Rw5csRqxzPGmBs3bpgXXnjBREVFmaJFi5rQ0FDTqlUr8/bbb5vU1NSsdTmtlt+yZYu5//77TYUKFUzRokVNUFCQad26dbYa9Nt2795tRMQ8//zzjs9BRESE6dmz5x3X3G6MfOutt+B/P3r0qBk8eLApW7asKVKkiKlQoYLp1auXmTVrVrZ1u3btMrGxscbf399UqFDBvPbaa+bTTz91bIw0xpjk5GTzl7/8xVStWtUUKVLElC1b1gwYMMAcPXo0a826detM48aNTdGiRa0mQ08fo9uPW9izZ4/p0qWLCQgIMMHBweb3v/+9OXfuXI7+7L1UkGbCGGNGjx5tqlWrZvz8/ExYWJj53e9+l+2cuI0z4T0zkZKSYkaOHGnq1atnSpQoYYoUKWIiIiLMsGHDPNbK6kkFaSaOHDliBg8ebKpVq2aKFStm/P39TUxMjHnllVdMYmKi9XeLiBk4cKDjc3Cnr/fnRMQMHz4c/rfz58+b4cOHm0qVKmWdrx07djSTJ0/Otu7EiRPmgQceMAEBASY0NNSMHDnSLF682LFF1Rhj0tPTzVtvvWVq1qxpihYtasLCwkz37t3N1q1bs9YcOHDAtG3b1hQrVsyISLbXxNPH6PYjSL744gvTrFkzExwcbIoVK2aaN29uzaM3KEgzYQz3iYK4T9x+7rT/Qx8jlJ8K0kxwnyi4+8SdZuKXs+/E5/8ekP7Pxx9/LGPGjJGjR4+6/kB2ol8jzgRRdosWLZJevXrJzp07pW7duvl9OET5jvsEUXbcJ3697tnvPBcUK1eulBEjRvDNn+j/cCaIslu5cqUMHDiQF0RE/4f7BFF23Cd+vfidZyIiIiIiIiIH/M4zERERERERkQPePBMRERERERE54M0zERERERERkQPePBMRERERERE54M0zERERERERkYPCuf2DTZs2zfHasmXLwjwtLQ3m1apVg3l6erqVnTp1Cq7dtWsXzNu1a2dle/fuhWvr1KkDc62gHH09lSpVgmszMjJgfuTIESs7efIkXHvz5k2Y33fffTD38/OzspSUFLh248aNMI+IiLCyVq1awbXacR89ehTmgYGBVhYQEADXfvXVVzDPT94yE2fOnIFrd+zYAXM0E/v374dra9euDXM3M1G5cmW4Fn0tIvd+JrTXYP369TB3MxPa+xX6GkU4EyLcJ5C8nImiRYtaWWpqKlzrLftEsWLF4Nqvv/4a5vmpefPmMM/MzLSy8uXLw7WemInTp0/Dtdu3b4d5hw4drCwvZ0LbJ7SZOHz4sJXFx8fDtdpMtGnTBuZFihSxMm2/WrduHcyrVKliZdoMajPhZp/QZuLLL7+EeX5q2LBhjteGh4fDXJsJ9LyL4Nfv7NmzcK12nqPX78CBA3BtzZo1Ya7NBDo+7f0AvXeIiBw/ftzKtLm/desWzJs1awZzN/vE1q1bYY5mXLtm0I47Li4O5ug6yd/fH66dO3cuzO+E33kmIiIiIiIicsCbZyIiIiIiIiIHvHkmIiIiIiIicsCbZyIiIiIiIiIHvHkmIiIiIiIicuBjtKo3B88//zzMUQPblStX4Fqt/Rc1G4qIlC5d2squX7+e47XaY2stcyVKlIC51iiHjkVrkNXaFFFDYlBQEFxbv359mH/66acwR+195cqVg2u1ZkD0+p47dw6uTU5OhrnWioiORWs//O6772Cen1544QWYo3PO7UygVmgRkVKlSlmZ25lAr6nWRqrNhPaaXrt2zco8MRPFixeHa7WZ+Ne//gXzvJqJ8+fPw7VJSUk5Pg4R/FwVpJnQ9gk3M7Fz584cP4aISEhIiJXduHEDrnWzT3hqJtB8lilTBq7VWsK9fZ9Az5+nZqKg7xMvvvgizD1x7YQeQ8TdPhEaGgpzNzNRsmRJmGvXTm72CTcz4XafmDx5MsxRm7E2szExMTBHz5927aTNhPaeglqYtU/dWLRoEczz08iRI2Hu6+trZdp5qzViFy6MP1QIzYS2T6C1InjetPsJ7Vx0s09oTePaa43apbV9QvsklS+++ALm6P1Zm9no6GiYo5m4cOECXOv2fgIdizZvK1asgPmd8DvPRERERERERA5480xERERERETkgDfPRERERERERA5480xERERERETkAP8mfQ5ohRURERFWFhYWBteOHj0a5i+//DLMUalE37594doffvgB5kiNGjVgrhXSaPm2bdusrEuXLnDtsWPHYI6KCTZs2ADXasUZqDxCRAR1w2nlC1pB1fHjx61MK5NJSUmBuVYqsH//fivTSjm80fbt22FepUoVK9NKWdzOBCqU8aaZ2Lp1q5V17doVrnUzE+vXr4drtdIxbSYQ9D4j4pmZ0MpEfq0z4Yl9YtSoUTD/3//9X5ij169Pnz5w7erVq2GOuJ0JrajGzT6Bzi0RkeDgYCvLy30iL2dC2ye09QV9JrR9omrVqlam7RNPP/00zP/yl7/AHO0T/fr1g2u1fQKdF1phnJuSJhGRLVu2WFn37t3hWm0m0GP/+OOPcK22T1SsWBHmnpgJtL9p7/vaPuFmJrRyKW+0e/dumFeqVMnKtJLHP/7xjzCfMGECzNG1b48ePeBa7XoDnReRkZFwrVbmp80KKsps3749XHvixAmYo+K+zZs3w7Va6Zh2jiLa/YT2tcfHx1uZ9l6u7RPa+sOHD1uZ9n6aG/zOMxEREREREZED3jwTEREREREROeDNMxEREREREZED3jwTEREREREROeDNMxEREREREZEDH4Pq4nLgySefhDlqQqxevTpce/nyZZhfvXoV5h07drSy999/H65t2LAhzFGjnI+PD1yrNS9qjZGoCfHbb7+Fa+vWrQtz1Lw3Z84cuFZrwdNaYYOCgmCOpKenwxw1Sfr6+ub4cUVEjh49CvOYmBgr01oYtRbN/JQfM9GpUycre++99+BaNzOhvaZaQ29ISAjM0TnqdiZQm/Hs2bNz/PeJFNyZqFOnjpXt2rULruVM/ORe7xMVKlSAuTYT/w37BGpX1WZCe15/rfvEU089BXPUxqy9Rm5nonPnzlb297//Ha7VZuL06dNW5naf0JqS0UzMnTsXrq1Xrx7M0bFo+4TWWq093yVKlLAy7dI5L/cJ1CAsgt8ntH1i7dq1rv7Oe2Ho0KEwRy3SUVFRcK127mut6G3atLGyTz75BK7V3ofPnTtnZehcERFp3Lixq8du0qSJlS1evBiuDQwMhDlqbV+4cCFcqzVlo092EBEpXry4lWVmZsK1GRkZMHezT2ji4uJgjj4JYO/evXCt1kB+J/zOMxEREREREZED3jwTEREREREROeDNMxEREREREZED3jwTEREREREROeDNMxEREREREZEDu7Izh06dOgXzTZs2WVmxYsVcPXaVKlVgvnXrVivTWh01aH2XLl3gWq018T//+Q/MUatfhw4d4NrU1FSYo6ZTrY0Ptd2JiFy5cgXmoaGhVlaoEP73E+1rR22UWlOs9tpo+ZEjR6wMtQV6K20mUJOfp2Ziy5YtVqa1/2oCAgKsDLUgiog0atQI5qtXr4Y5asB0OxPoPNcaV7WmYK1x0xMzgVod582bB9e6nQnUrvprmAlv3yfQTKDWdxF9JtasWQNz7hPZac3H/237BJoJdB6K6A3l2kygPcjtTKBzDrW+i+jNwj/88APM0T6BPklCRCQlJSXHx6ftE9pMaPtEeHi4lWmvgZt9QmsU12ZC+wSYgr5PnD17FuY7duywMn9/f7hWe4+qVKkSzHfu3Gll2icTaKpVq2ZlPXv2hGtbt24N8/Xr18McvaaoIVzE3Uxoz592vmhN1Oj8d9tAj5rTv/vuO7i2TJkyMNdeM/SJHtp+mhv8zjMRERERERGRA948ExERERERETngzTMRERERERGRA948ExERERERETngzTMRERERERGRg1y3bfv6+sI8OjrayrQWt+DgYJjHxcXBHDVPRkREwLVaAxtqdYyMjIRrr1+/DnOtvQ81Mm7btg2u3b9/P8ybNWtmZVrLa+nSpWFesmRJmKPGO20tanHWHsNtM+TBgwdhjpr3tJY+b+RmJrTWvxIlSsDczUxUrlwZrtVm4sKFC1amNVdu2LAB5l27doV5kyZNrEw79z/44AOY165d28q0lldPzIT2vqS1TnImdAV1nzh37pyVaeecNhN+fn4wDwsLs7K0tDS49uLFizB/8MEHrUw751CrrIhIcnIyzD2xT6D3N+34tP30wIEDMC/oM6G9t6KZ0Fp0tdcDNcyKiAQGBlqZNhPa64RmQjuOhIQEmHfu3BnmLVq0sDKtgfnEiRMwb968uZU1bdoUrtXeaxYsWABzNJ95uU9oe7g2E+jc+TXMBGqz1t4rtQb1kydPwhx9uoP2SSVa0zNqZ9fOC+29Utsn0HOivZdrDf4NGza0Mm0fQ+eQiEjVqlVhju6PVq1aBddqx43OUe251l4b9OkLIviTBzIyMuDa3OB3nomIiIiIiIgc8OaZiIiIiIiIyAFvnomIiIiIiIgc8OaZiIiIiIiIyEGuC8NQAYUI/uVyrZxBK5PRfqkbFcRoZV/ff/89zNH6lStXwrXly5eHec+ePWFeq1YtK9N++b1ly5YwR7+EX7NmTbj2xo0bMPf394c5ev6010ajFVYgiYmJMNcKYrQCm4LCzUzcvHkTrtVmwhgD87yaCe280MpaUIGRCD7/MzMz4dpvvvkG5p9++qmVaWVRp0+fhrk2b94+E+g80c4Fb5SXM3Gv9wntvNBej9///vcwr1evnpVduXIFrq1evTrMUSGTVlSjFb5oz7cnZuLQoUM5Xvvftk8EBQXB3BPXTtq5iIrEUPGaiMiyZctgjtZ37NgRrtXOLe1cRPuENt9t2rSBOSrXO3bsGFyr7UGtWrWC+caNG61M+xo1WlEmos2EViRW0GdC2yfQ16vNhFaEpb3WqEhMK9FbvXo1zNH6FStWwLVa+R8quBTB83b+/Hm4VtvfLl++bGXaXqPRSvfQtWBISAhcqxXCaWVfSFJSEsy1IjG3e5Zb/M4zERERERERkQPePBMRERERERE54M0zERERERERkQPePBMRERERERE54M0zERERERERkYNct22np6fDHLWtxcfH53itiN6qhmgNe40bN4Y5aqnUHkNrXtQah1GjXN26deFarREbPa9am29YWBjMP/74Y5hfuHDBylq3bg3Xao2WW7ZssbIqVarAtVrb3dWrV2F+5swZKytXrhxc643czARqVdfWinhmJrTWRB8fHyvbvn07XKu1VmtfO5oJrYlXa9zs3bu3lW3evBmu1eZt69atML/XM5GamgpzzoT+6QGlSpWC+b3eJy5dugTXPv300zBHjdgiuF0VtaKK6M2oaGa151rbx9555x2Ynz171sq0mdByNG/cJ36ivU6lS5e2Mu1aQ9sntBlC3O4T6FNGtH3svvvug7l2Lai1hyPXrl2DOdprtOdaey/XGrHROae9/7jZJ7T2cW0mtK/91KlTVvZrmAn03o++VhH90wa05nLUwq3NRP369WGOztsffvgBrq1Tpw7Mtet71OaOzkMR/VNGUGP9vn374Nq///3vMN+xYwfM0X7YqVMnuFb7FIANGzZYmfYpC9q1k/YehPYx7dOPcoPfeSYiIiIiIiJywJtnIiIiIiIiIge8eSYiIiIiIiJywJtnIiIiIiIiIge8eSYiIiIiIiJykOu2ba3hLDo62sq0xlAtRy14IrhJ99y5c3CtlsfExFiZ1gKqNUOiVkwRkU2bNllZvXr14FqtbXvBggVW5ufnB9du27YN5iVLloQ5ai8tXrw4XKu1TqLnSmuh1Ro0tTa98uXLW9nhw4fhWm+kzUT16tWtzFMzgc4v1DIoojcyRkREWJnW5lqxYkWYN2nSBObffPONlWltmVo7Mfp6unbtCtfOmjUL5pGRkTDnTOQtT+wTWuuyJ/YJbSZ+//vfW5n2GmnnLfoaRfA+obXEFylSBOZLliyxMm1P2bVrF8xr1qwJc3TuBgUFwbVaU2xezkTZsmWt7OjRo3CtN3KzT2jPmTYrGtQWrO0TJ0+ehDlq4UbvTyJ6Szj6GkVw667WjLt8+XKYo7lq3rw5XLto0SKYN2jQAOZFixa1MtTuLaK3GaNm7YsXL8K1bmcCNWsXpH1Ca8RGz5m2H2i51vSM7gXQp2+I6LOCXg+t9Vu77qlWrRrM0fV97dq14Vptn/j++++trESJEnDt/PnzYd6iRQuYo/cx7XXcu3cvzNHz5+YTJkT0a9Lw8HArO378OFybG/zOMxEREREREZED3jwTEREREREROeDNMxEREREREZED3jwTEREREREROch1YdiNGzdgfujQIStDZRUietmIVsSAChC0kpTKlSvDfMSIEVamFapopQ1a2cR9991nZVp5l1Yyg4qNtOdDK8zRCsZQqdPatWvhWlTKIiKSnp5uZYGBgXDtmjVrYO6mUOHmzZtwrTdyMxNakZzbmUCPrc0EKhURweVgWpGSdtxamRY6Pq0IRiv9Qedcly5d4Nq4uDiYa8U2qARHO2+15y8tLc3KOBM/yY994uDBg1ZWrFgxuPbBBx+Eefv27a1M2yfcFsSEhYVZmbZPoBIkEVzMohWUnTlzBuZPPPEEzNF7v9uZyMt9AhVa3bp1C671RteuXYM5Om+1Aivt/dbNTGj7RIUKFWC+efNmK9MKwBo3bgxzbSbQnqC9/ui9QwSf5+i9WUQvwNOK+1Dx2OrVq+FareQNzYR2HD/88APM3cxEcnIyXOuNtH3i2LFjVqaVZh05cgTm2nU8ut7QZkK7ZkGvh1Yupl1Ta+/PqJhYK/vSytK6detmZdpsakVdNWrUgDl6b1q8eDFci/Y8ETyf2l69ceNGmGszgZ5XT+4T/M4zERERERERkQPePBMRERERERE54M0zERERERERkQPePBMRERERERE54M0zERERERERkYNct21r7WmoqVFrS9Ua4rS2OvR3VqtWDa7dvXs3zI0xVqa1z2nHoTWjopZFrTVRa7REx4eaVUVEYmNjYf7999/D/OLFi1ZWp04duHbDhg0wR+u1tkCt/RUdh4iIj4+PlcXExMC13kibCdT0rM2E1sSqPceoBbJq1apwrfaaohZI1Mwugs9PEZHz58/DHDVDr1q1Cq6tVKkSzNF5ceHCBbj2d7/7HcxfffVVmKOvU5sJrfkYtX9q7x2emAnt+LxRXu4T2vsiaszt1asXXItaakXc7RPabF6/fh3mhQvb2+4HH3wA12qt9+j4tHOrZcuWMN+xYwfMUcur1kKcH/sEapQuSPuE1rpbsWJFK9NatbWZ0M4X9Hdq107aa4qa1dG5LKJ/Eog2E6VKlbIy7f1AO5/37dtnZdqniYwdOxbm2t7kZp9YunQpzFNTU61Me720hmM3+4Q2s94oNDQU5uic0z6VQzu3tGsT9HdGRETAtVu3boV5eHi4lWmfGnHu3DmYa03j6D5Dm/vOnTvDHM2h1jiNGuVF9PcD9HW2atUKrtU+XQhdx+3duxeu1d6vLl++DHM0EzVr1oRrc4PfeSYiIiIiIiJywJtnIiIiIiIiIge8eSYiIiIiIiJywJtnIiIiIiIiIge8eSYiIiIiIiJykOu27SJFisD8xx9/tDLUNiyiN7Bpjb6o2W79+vVwbdGiRWGOjkVrtETtiCJ623ZISIiV+fv7w7WHDx+GOWqrQy2cInp7X7du3WCOmiSnTJkC12pN3qdOnbIy7fXVmgG112zo0KFWpjVXeiNtJtatW2dlqEVSRG+R1mYCtRmjv09EJCMjA+ao+VhrPNSa3LVzrlixYjl+DO35Q8cXFRUF12qtnYMGDYL5jBkzrEx7X+rYsSPM0Uxor6/WCIreN0UK/kxobbzo69WeM+310FrlBw8ebGXa+axB72mozVdE5MqVKzDfvHkzzO+77z4rQ43tIiLNmjWDOTr/tTZSbZ+IjIyE+YQJE6zs2LFjcG2HDh1gjlqEtddX+4QBN/vEkiVL4Fpv5OvrC3PUDK09Z1qDsAbtE9p7jrZPoJnQ9hptZmvVqgXz4OBgK9O+9qCgIJhPmjTJyrQ2fW2v0Rqq0ae3aK+B1sq/YMECK9OunbSZ0NrDH330UStbvHgxXOuNtH0CfcIF+oQREb2JXLt2QtcK2nu2NhOoxf7QoUNwrXY/oZ3P6FzU9oPAwECYo3sS9GkKIvq1p9ZQjfZD7TXo0aMHzNH9h/ZpBForufYpKOiTV1asWAHX5ga/80xERERERETkgDfPRERERERERA5480xERERERETkgDfPRERERERERA5480xERERERETkINdt2yVKlID5uHHjrOyjjz6Ca7W2VK15DzV4aq3a7du3hzlqbNOaRNHfJyLi5+cH80aNGlnZtGnT4NpKlSrB/IknnrCygQMHwrWdO3eGuY+PD8y7dOliZcnJyXCt1pSNnm+t4VhrGm/dujXML1++bGVaK6Y38sRMaA24hQrhf+dCTc/aTGgN6gEBAVZ2/PhxuFZr+9SaiMeMGWNlbdq0gWu7d+8Oc9RSqrVLlipVCuZ16tSBOTru+Ph4uBY1h4t4Zia054Qz4X6fOHDggJVpM9GpUyeYu5kJbf/Q1KhRw8q0T3xA79kiInPnzrWyzMxMuFZr9E1LS4N5y5YtrUx7/3EzE6jx+U6P4Waf0F5fb6R9Wsf48eOt7B//+Adcq33agNbkja5ltOsYbZ9A77nauf/DDz/AXHsf/uc//2llWkN+06ZNYT5nzpwc/33a+5LW6Isep3jx4nCt9mkSnpiJtm3bwvzSpUs5+vu8lfZcvvDCC1b22WefwbVVqlSBufbedfbsWSvTnjP0niiC94kTJ07k+O8T0ecQffKO1sLesGFDmP/nP/+xMq09W2u3T0pKgjl6HLf3TKg9XGs8166dWrRoAXP0KRienAl+55mIiIiIiIjIAW+eiYiIiIiIiBzw5pmIiIiIiIjIAW+eiYiIiIiIiBz4GGNMbv6gVuahFTEg4eHh+KCUwitUopCYmAjXVqxYEeaoPKVChQpwrfa1aH/n9OnTrUz7BX+txODChQtWVrlyZbhWK0jQ1qMyqtKlS8O1y5cvhzkqGLt582aO14qIbNy4Eeao9EArhpo0aRLM85M2E8HBwVamlfvkx0ygIgut+CEoKAjm2nmEHhuVbIjopRIJCQlWphXuPfDAAzDXypG++eYbK9NKd7RSE/Sc3Lp1C67VyjfczIT2GqDSnfzmiX1CK9XR3kN79uxpZdq5r+W7du2ysi1btsC19erVg3lERATM0UykpKTAtdrXiGZCe98PCQmBOSoZEhE5fPiwlWklelrBEjpHuU/8RCsHRPuEdnmm7RMa9Pxor5323or2BG2f0OZbOxfPnDljZdp7uVZsdPHiRSvTvhat0BE9hgh+HbSvccmSJTBH7/3aTGj7xIYNG2COymq1mZg4cSLM81Pz5s1h7maf0K5BPHHtpJUunj9/3sq0YjBt39ZKTs+dOwdzBJVQiuD3eO2eBL3/iODiLRGRfv36WZl2L6V97Wju3dx7iIhs27YN5nXr1rUy7dpz6tSpML8TfueZiIiIiIiIyAFvnomIiIiIiIgc8OaZiIiIiIiIyAFvnomIiIiIiIgc8OaZiIiIiIiIyEHh3P7Btm3bwrxLly5W9u2338K1ixcvhnnZsmVhXrVqVSvTWie1VkLUchwbGwvXokZTEZG5c+fCfMiQIVamNcG1bNkS5qipEbWLiogsXLgQ5lpzK2o01Br9tEZL1HinNUNu374d5h07doT5pk2brKxHjx5wrTdyMxPz58+Ha7/77juYu5mJMmXKwLVuZkJrl9RawrVz8ZFHHrEytzNx5MiRHK/dvHkzzB988EGYt2/f3sq++uoruFZrxYyOjrYybSZ27twJczczgdqkvVV+7BPLli2zMq1VWzsXUQuu1hRcrVo1mC9YsADmbmaiRYsWMEf7RIMGDVwdR5MmTWAeFRVlZevXr4drly5dCnM3M6HtEx06dIB5QZ+Jdu3awbxr165WNm/ePLhW2ye09370SRvaWu21zsjIsDKtXVfbJ7R9b+jQoVamtVa/9tprML9+/bqVaddO2kw0bdoU5qgtWGvfRw3MIp6ZiU6dOsEcNdMXpJnQ9nM0K9p+sHLlSphrn9aAPp1Aa2HfunUrzNPT061Mu3bWmvO199CBAwdaGfoUCBG9RRpd82nvq9q8devWDebo2vPEiRNwrdbkjRrStddg7969MNc+vQDtqZ07d4Zrc4PfeSYiIiIiIiJywJtnIiIiIiIiIge8eSYiIiIiIiJywJtnIiIiIiIiIge8eSYiIiIiIiJykOu2ba31DeVaI6PWSqe1Rd+6dcvKrl27BtdqjXKoMVVrhnTTECeC2yvr1asH12qNljVq1LAy1PwqIhITEwPzY8eOwfzjjz+2sieffBKuDQwMhPns2bOtDLUhi+iNzcePH4c5asLT2pO90e7du3Oce9NMXL582cq0+a5VqxbM3cxE3bp14VpPzESjRo1grs3E4cOHrUx7DVDbrAhuRj158iRc63YmUCs1ahv2Vnm5T2jnwI0bN6wsPj4erm3cuDHM3cwEauIV0du53cwEag4XwU3jM2fOhGu19l/UTiyC28OrVKkC1/r5+cHcEzMRFxcHczQTBWmf0Br3Ua61xGszgfYDLdf2Ca35/8qVK1a2Y8cOuFbbJ7QmXXQ+a++3o0ePhjlq9teeD23etH2iWLFiVqa9X2mN3StWrLAy7X1Je2xtn0BN7QVpJvbt25fjXDv3tU/e0c6B1NRUK0N7h4je6Hz16lUr01qhUdu6iP4pHqglXJsr7bVGn5ygPR81a9aEuXZ/hPZIrcFfm2X0qQGoOf5Oj63NEPoUJa3FPjf4nWciIiIiIiIiB7x5JiIiIiIiInLAm2ciIiIiIiIiB7x5JiIiIiIiInKQ68Kwc+fOwbxy5cpWppVeaMUZzz77LMz/+c9/WtmlS5fg2kqVKsEc/fK7VoZSuDB+enr27AlzVOSDfmFfRMTX1xfmqDjj9OnTcO2WLVtgrpVNBAQEWFmRIkXgWh8fH5gjJUuWhLlWoKUVaixevNjKtNfGG509exbmbmZCK2AZM2YMzCdOnGhl2kxof+fRo0etzN/fH65FxSkiIr169YI5Kn/QijO0mUAFIadOnYJr09PTYd6iRQuYFy9e3MqKFi0K12oFO2i925lo3rw5zFGhRkGaCW/fJ7S/E5UgaueF9j6sFbDMmzfPyrT32/nz58MclXp5ap94+OGHrQztSyL4dRQROXjwoJXl5T4REREB13ojbZ9Ac62dn1rxzfPPPw/zjz76yMq091vt7zx06JCVaYVx2j5x//33w3zDhg1WVr16dbi2SZMmMO/evbuVafvE1q1bYa6V16HyprS0NLgWFVGJiBhjrMztTDRr1gzm//nPf6ysatWqcK03unDhAszRuai9Rnv27IH5U089BfPPP//cyrTzOTIyEubJyclWhs4VEX3f1oolUVma9hjaLKOSLVT8J6KXBLspidWuv7S9KSEhwcrQNZmIyP79+2HesGFDmK9cudLKtPvC3OB3nomIiIiIiIgc8OaZiIiIiIiIyAFvnomIiIiIiIgc8OaZiIiIiIiIyAFvnomIiIiIiIgc5LptW2vMPXnypJVp7YOoiVcEt6WK4Ea5QoXw/f+MGTNg/uijj1pZ7dq14Vqtbfvdd9+FeVBQkJVpzZrBwcEwR43dSUlJcO1DDz0E8+XLl8McNc2h9nERdy2V2teCGmFF9NY89Ppqj+2NtHbQ+Ph4K0tJSYFrtZlArdoi+BzVmnu1mUBtitp8ay3c33zzTY4f2+1MoOdEm4k//elPMNfaVePi4qxMmwmtmR61fGpfi9bauXfvXpij17cgzYR2HqGZyI99YubMmTCvUaOGlWnzrc2E1pSNGkm1xlVtD0LPCWp+FXG/T5w5c8bKtObwNm3awBw1fF+/fh2u9cQ+UapUKbjWG7m5dtL2CdROLYJbtUXweaTNxNdffw1z9F7udiZQK7SISMeOHa2sf//+cK12np84ccLKEhMT4drf/va3MF+6dCnM0eN8++23OT4OEdxAr5232kygBmYR/PoWpJnQvl7U0qzNxLZt22A+d+5cmKPr/rp168K1N27cgPkDDzxgZdonEGgzsWDBApijT+DQmqhLlCgBc3Q/0aBBA7j2wQcfhPnq1athjtrjtQZ6bQ9Hny6jNdBrTeOHDx+GOdontMfODX7nmYiIiIiIiMgBb56JiIiIiIiIHPDmmYiIiIiIiMgBb56JiIiIiIiIHPDmmYiIiIiIiMhBrtu2tcbDmJgYK5s2bRpcW7p0aZiHhobCHLXp9e7dG649duwYzHfu3GllWjue5u2334Y5arZbt24dXIvaXEVwE7HW0ocaa0VEbt68CXPUth0QEADXRkREwLxMmTJWVrVqVbgWtRCLiERFRcEcNbTOnj0brv3kk09gnp+0BmhPzERYWBjMUYt0nz594FptJq5du2ZlWnOlpnPnzjBPSEiwsqtXr8K1derUgTmaCa15UWts1t6vUHO1thad+yK4RVhrD929ezfM3czEnDlz4FpvnImCuk+gdlVtJrT289dffx3mixYtsrL169fDtVr7K2rE9tQ+cfHixRyv1fYJ1M6ttW3/t+0T2kyg97+pU6fCtdq572Ym+vbtC9cePXoU5uj10z45oX79+jD/+9//DnPUjOt2n0DNzG5n4scff4Q5unbSzmdtJtA+q336gtt9Al17zpo1C66dNGkSzPOTNhPoOln7hAS0p4iIDBkyBOaozb9FixZw7eLFi2GO9gnttdPExsbC/NKlS1amXWNq13zoulH7NBG0p4jo7/3ocbRPNdHmEB2fNj8HDhyAuXb/sX37ditbuHAhXJsb/M4zERERERERkQPePBMRERERERE54M0zERERERERkQPePBMRERERERE54M0zERERERERkYNct20XLVoU5idPnrQyrQVPa2ArVaoUzFErXfPmzeHa+fPnw9zHx8fKUAuiiEitWrVgHhgYCPNOnTpZmdbeh9oRRUQaNmxoZVqjpdbQOnz4cJij1trz58/Dtai5UkSkUCH731tQ46CISHR0NMyrV68Oc/Q6aK+BN8rLmdAafd3MxLfffgvzihUrWhlqyRbRZ7NwYfxWglpKtWZNrZketT22bt0arl26dCnMtZlFzbKo+VVEb3XkTOiKFCkC81OnTllZ7dq14dpixYrBPC/3CTczUb58eZhrX3uXLl2sTJsJ7T0eta4GBQXBtVrDaLt27WCO3oNQA7eISOPGjWGOWsK1xlptJrS8oM+Em31CmwntUzJCQkJgjl5r7dpEe63RdYU2E9p+pX1iAXoP1WZCa8RGM5GYmAjXai3s2nvKiRMnrEzb87QGbbSvaI3d2rmvfUoLmgmtldwbaTOBrn1Rk7+I3jgdHh4OczQrjRo1gmu19y703q/NT8mSJWGu7RMVKlSwMm0mtPsJN/vEkiVLYK4938uWLbMy1GAuoj+vaCa0mdWuv7R5Q7OlzU9u8DvPRERERERERA5480xERERERETkgDfPRERERERERA5480xERERERETkwMcYY3LzBx966CGYZ2RkWFlYWBhci36ZXUT/hXtUgLBmzRq4FhVvieACAlRqJCLSuXNnmMfFxcEcFXA0aNAArt2wYQPML126ZGWXL1+Ga7XjRuUWIrhsABXjiIjMmTMH5levXrUyrdQEFaCI4NI2EXzcO3fuhGu1UoH85ImZ0AohtGIWVMyzevVquNbNTGglen5+fjD3xExs3LgR5qjsq0OHDnCtds7NmDED5qj0RXuetPcrNBNagREqyrqT/6aZCA0NhWu1mbjX+4Q2E1ohTatWrWDesmVLK6tfvz5c+/nnn8N81qxZVoaK9UT0kiatiLJfv35WhsowRfSZQO8HO3bsgGv/2/aJ3/72tzDPy5lA1zJaWZU2s+i9S3sMrQBK2yfQTGizqZXoTZ8+Pcd/n1aadOzYMZijfUIrQdLOObRPaMWh2jWcVmZ5/PhxK9PmzRtnQiv7QjOBim9FRKKiomCuXePWq1fPyrRrEHSdJSJy7tw5K3M7E9r7X5MmTaxMK4HbunUrzNH9hPaerR3HihUrYJ6enp7j49Per9CeqpV6uSkxFsEztG/fPrhWe07uhN95JiIiIiIiInLAm2ciIiIiIiIiB7x5JiIiIiIiInLAm2ciIiIiIiIiB7x5JiIiIiIiInJgVwjmkNYAjdp4tbUBAQEwDw4OhjlqTNUaGbVGPtSOpzUyao2HVatWhTlqlNOa4LSmbNSEqDX3ooZKEZExY8bAfNy4cVY2f/58uPbo0aMwRw2Au3btgmuLFCkCc639MDU11cpef/11uNYbaee5v7+/lV25cgWuLVasGMy1mUDN2p6YCXTMIiIlS5aEeWRkJMxRA7DWGPrAAw/AHJ1H2nsHaucUETl06BDM33jjDStbtWoVXKu1PTZt2tTKPDUTqNFy/PjxcK03crNPaDORH/sEaqLWZkI7vlKlSsHczUx06dIF5mj/0P4+1HgvIvLcc8/BfPTo0Vb2ww8/wLVae3x0dLSVTZs2Da79b9snUAOuiGf2Ce1aATWud+/eHa7VZuLMmTM5Po4SJUrAPCUlBebo/fLWrVtwrXbNgs5F7VMt7rvvPpg/88wzMP/b3/5mZStXroRr3ewTWks8avcWcbdPTJgwAa71Rtp5jvYJ1FouInL48GGYa4316BNvUAO3iL7XuLl2Kl68OMy19360T2j3E1qT94EDB6xMa75G79kiInv37oX5Sy+9ZGU//vgjXIueJxH8ySva36ftE+XKlYM5mokXXngBrs0NfueZiIiIiIiIyAFvnomIiIiIiIgc8OaZiIiIiIiIyAFvnomIiIiIiIgc8OaZiIiIiIiIyEGu27ZR46UIboPT2vFQm5yISLt27WCOWuK2b98O10ZERMActVkbY+Dar776CuZaI1+fPn2sTGut1trx0LHEx8fDtdevX4e51nS5b98+K9PaibVGcdSuiZppRUQaN24Mc62JOCoqysqWLFkC144cORLm+SktLQ3maCa0+cnLmahSpQrM3czEyZMnYa61g6LmSa2xu379+jBHMzR9+nS4NikpCeZa0+Xu3butTJtNbSbQ1+ipmahWrZqVLV68GK4dMWIEzPOTm31Ca1zNj30CnUfaTBw8eBDmqJ1YRGTo0KFW5nafQO/b2qdDXLt2DebaTKBzUWs61d5T0GvAfeIn2j6BWnfdzoTWoI4aoLds2QLXau9zN27csDLtNV23bh3MtbZtlGvnFmpg1mgz4fbaCZ2L2nG42Se09wj0qSbacYi42yeeeuopmOcnN/uE9n6mtbNrn/iAzmftvVz7tAZ0fNo+cezYMZhr5yi6TkKfriKiX8ej5mr0dYvon0gSFBQEc3Q/oc2E9ulCaP2FCxfgWu2+S2vnRnOofZIKag53wu88ExERERERETngzTMRERERERGRA948ExERERERETngzTMRERERERGRg1wXhkVGRsK8fPnyVnbu3Dm4VitL2LFjR44fWytaQWtFcDlSp06d4FqtUAMVloiIvPnmm1b22GOPwbVa8VK/fv2sDJUaiehf47Bhw2D+4osvWpnbQhJUkBAQEADXauUB2npEK5fyRqi0Q0SkQoUKVqYVrbidiYoVK1qZNhPlypWDOSqQ6Ny5M1yrlWGgr1FEZOfOnVY2aNAguPabb76BOSrYSUhIgGujo6Nh3qpVK5ivXLnSynx8fODazMxMmKOin7ycieDg4ByvzW9u9gnt3Nf2D2/fJ7SSmYkTJ1qZVk6Jzk8RkaZNm1rZnj174Frta9QK5rRSFUQr70F7ltuZQOWUmoK0T2gzkZf7BDoXtZnQnsvPPvvMyrQSH+29snnz5jD/97//bWVdu3aFa2fMmAHzRo0aWZl27aTthaNGjYK5NoeItkeia6fAwEC4VisKdLNPaOVn3kgrWStbtqyVaYVS58+fhzm6BhHBz/369evhWq3k78iRI1bWtm1bV8dRt25dmM+ePdvKfv/738O18+fPz/Fj79+/H65Fz7WIfg/z448/wtwNdO2kve9720zwO89EREREREREDnjzTEREREREROSAN89EREREREREDnjzTEREREREROSAN89EREREREREDnLdtq21BKIW3I0bN8K1WlOn1lKK2tZSU1PhWq0tOiMjw8pOnz4N1xYpUgTmWqNlcnKylX3wwQdwbePGjWH++uuvW1nx4sXhWq0BUmsMRC2a2ut48eJFmMfExFgZamu+03HExsbCfNOmTVZWunRpuNYbac8laof01Eygc1SbiVu3bsEcNaNqM6G1lIaFhcEcPScff/yxq8detGiRldWqVQuu1cyaNQvmbmbi1KlTMEczcf36dbh2w4YNMNdmAp0nv4aZQPvE5s2b4dpf2z6B5hB9yoKIyCuvvAJz9CkJNWvWhGs12kygNvdt27bBtd9//z3MW7ZsaWXae4TbfQLNRGhoKFzrjdzsE9pzo82E1pSekpKSo0wEt/yK4PNfey8qXBhfWm7fvh3maIamT58O16JWbRGRy5cvW5nbfWLmzJkwL1WqlJVpLb/aPlGnTh0r0/YJ7XVv3759jtdr8+aNtOfy0qVLVrZ161a4VnsutUZ4dM5prdraPoGunbRPh9BmQmuER8enfSKJ1th99epVK9M+kUSzcOFCmKPmaq0pW/vUALRnafcT2uuO9hoRvGeFhITAtbnB7zwTEREREREROeDNMxEREREREZED3jwTEREREREROeDNMxEREREREZED3jwTEREREREROfAxxpjc/MFx48bB/P3337ey7t27w7VaY6TW4Ihap/38/ODaBQsWwDw8PNzKEhMT4dr69evDXGvN0x4H0dr7UGverl274Nrjx4/DXGvnRu14FStWhGtRc6WIyPnz561Ma+/TWidR066ISM+ePa0sPj4erp02bRrM85ObmejRowdci9oRRfSZQK8pahsW0WeiTJkyVqady1pzZV7OBPo7tZk4duwYzN3MRKVKleBabSZQu6anZqJXr15Wps3E1KlTYZ6ftJl47733rEybCU/sE9pMaE2i3Ceyy499ArVPi7ibCW/cJ9544w2Yv/vuu1aG9kQRfSa01nE3M/Htt9/CvGzZslamncsNGjSAua+vL8zR4/j4+MC1SUlJMEdzuHPnTrjW7T6BPpXB7UygxuEaNWrAtSdPnoS5NhMPPPCAlZ04cQKu/fzzz2Gen1588UWYT5o0yco6deoE12pt26gpXUQkKCjIyrSZWLJkCczRvKFP3RHBn8oh4m6fKFQIf79T+ztr165tZfv27YNrtfMFPU8ieJ/QPjFFu65F53O1atXg2jNnzsAcNbKLiHTu3NnKtE/MmDFjBszvhN95JiIiIiIiInLAm2ciIiIiIiIiB7x5JiIiIiIiInLAm2ciIiIiIiIiB7x5JiIiIiIiInKAa95yALXgiYjcf//9OX6M1NRUmGdmZsL80KFDVtaxY0e4Vmv2RG2PVatWhWu1FrwrV67A/MaNG1amfY3t2rWDOWr1y8jIgGsjIiJgrjXKVa9e3cq0hsIKFSrAvEqVKlYWFxcH12qNlh06dIA5agxEDZ/eavLkyTBHM6G1Jnr7TBQpUgTmWsNoQZwJrcm2fPnyOf47PTUTqG25IM2Em31Ca9fVzhftHPg17ROxsbEwX7p0qZVpz0flypVhrrVcc5/IWxMnToR57969rUzbJ1JSUmCunQMHDhywMtRGK6I3QKNjcTsTbvaJW7duwbXaebF48WIr89Q+gZ4TtzOBnivt3Ndy7X0M7RNa87E30j4pomvXrlbmdia0a6cjR45YmfZ+GxkZCXN0LOi9T0SfCa2JGrVtp6WlwbWtWrWC+YoVK6xMez609nj0aSIi+DlBcyyivz+jv1NrmtfawNu0aQNz9P6BPkUjt/idZyIiIiIiIiIHvHkmIiIiIiIicsCbZyIiIiIiIiIHvHkmIiIiIiIicpDrwrC6devCPCQkxMq2bdsG12q/WH/w4EGYo8KRsLAwuFb7xXpU8lCvXj24Nj4+HuZaMQEqaNDWamUY1apVy1Emov9i/f79+2GOnj/0eonozx/6O1G5johI27ZtYb5mzRqYo6KN4sWLw7XeqE6dOjC/1zMRGhoK17qZifr168O13j4TWuGLJ2ZCK59Bf6dWfvXfNhPaPlGqVCkr27FjB1yrlfugYjAR79kntPPZzUxoOTr/tfImrRgMlUiJ4OcvODgYruU+4Z52Hnlin3DzmmrlOVp5HTqfGzRokOO/T3sMEVwopJ1b2kygAiNtn3A7E6isSJsJbZ9AJWBa+ZVWnulmJoKCguBab1S7dm2Yo+d49+7dcG2lSpVgfvjwYZijc0C7dtJmAp2LMTExcK12baKdz2XKlMnxWu340J6gFUieOXMG5qhYTQS/x2szkZ6eDvOzZ89amXbt1LJlS5hv3LgR5ug9JTAwEK7NDX7nmYiIiIiIiMgBb56JiIiIiIiIHPDmmYiIiIiIiMgBb56JiIiIiIiIHPDmmYiIiIiIiMiBjzHG5PdBEBEREREREXkzfueZiIiIiIiIyAFvnomIiIiIiIgc8OaZiIiIiIiIyAFvnomIiIiIiIgc8OaZiIiIiIiIyAFvnomIiIiIiIgc8OaZiIiIiIiIyAFvnomIiIiIiIgc8Oa5AFu1apX4+PjIqlWrsrJHHnlEqlSpkm/H9EvoGInyCmeCKDvOBFF2nAmi7D7//HPx8fGRuLi4rKxdu3bSrl27fDumX0LHmF9yfPPs4+OTo//z1kGfMWOGPPzwwxIdHS0+Pj4eOSHatWuX7WsPCQmRpk2bymeffSaZmZl3f9D30Pjx42XevHn5fRiqGTNmSMuWLSUwMFCCg4OlVatWsmLFinw9poI8E5cuXZK33npL2rZtK2FhYRIcHCwtWrSQGTNm3NXjciby3sGDB2X06NHSqlUr8ff395rNRKRgz4SIyOjRo6VRo0YSEhIiAQEBUqtWLRk7dqwkJibm+jE5E3lv7Nix8Dzz9/fP70PjTACcibw3Z84ceeihh6RatWoSEBAgNWrUkGeeeUauXr2a34dW4GciMTFRRo0aJRUrVhQ/Pz+pVauWTJw48a4es0qVKtm+9vDwcGnTpo3MnTvXQ0d9byQnJ8vYsWO99rW7LS0tTWrXri0+Pj7y9ttvu/7zhXO6cPr06dn+97Rp02TZsmVWXqtWLdcHcS9MnDhRtm7dKk2bNpVLly557HErVqwob7zxhoiIJCQkyLRp02TYsGFy6NAhmTBhgsf+npz65JNPcrX5jB8/XgYMGCB9+vTx/EHdpbFjx8pf//pXGTBggDzyyCOSlpYme/bskdOnT+frcRXkmVi/fr385S9/kR49eshLL70khQsXltmzZ8vAgQNl37598uqrr+b6sTkTeWv9+vXywQcfSO3ataVWrVqyY8eO/D6kLAV5JkRENm/eLG3atJGhQ4eKv7+/bN++XSZMmCDLly+X1atXS6FCufthLc7EvTFx4kQJCgrK+t++vr75eDQ/4UxgnIm89fjjj0v58uXl4YcflsqVK8vu3bvlww8/lEWLFsm2bdukWLFi+XZsBXkmMjIypGvXrrJlyxYZPny4REdHy5IlS+TJJ5+UK1euyIsvvpjrx27QoIE888wzIiJy5swZmTRpkvTr108mTpwoTzzxhKe+hBxbunSp6z+TnJycdf3oTd+1/qV//OMfEh8fn/sHMLk0fPhwk5M/npSUlNu/wqPi4+NNRkaGMcaYmJgYExsbe9ePGRsba2JiYrJlSUlJpmLFiiYwMNCkpqbCP5eRkWFu3rx513//ypUrjYiYlStX3vVjBQYGmiFDhtz14/zS3R7j+vXrjY+Pj3n33Xc9e2B5oCDNxLFjx0xcXFy2LDMz03To0MH4+fmZxMTEXD0uZ8LZ3R7jpUuXzPXr140xxrz11ltGRMzx48c9d4AeVJBmQvP2228bETHr16/P1Z/nTDi722N85ZVXjIiYhIQEzx5YHuBMcCZy4m6PEf25qVOnGhExn3zyyd0dnIcVpJmYOXOmERHz6aefZsv79+9v/P39zfnz53P1uBEREaZnz57ZsrNnz5rAwEBTvXp19c+lpaWZlJSUXP2dPzdlyhSPXEskJCQYETGvvPLKXR/TL3nqGM+fP29Klixp/vrXvxoRMW+99Zbrx/Do7zy3a9dO6tSpI1u3bpW2bdtKQEBA1r/C+Pj4yNixY60/U6VKFXnkkUeyZVevXpVRo0ZJpUqVxM/PT6KiouTNN9+0/lXw7NmzcuDAAUlLS3M8tkqVKuX6X0jdCAgIkBYtWkhSUpIkJCSIyE9f+1NPPSVffPGFxMTEiJ+fnyxevFhERE6fPi1/+MMfpEyZMuLn5ycxMTHy2WefWY976tQp6dOnjwQGBkp4eLiMHj1aUlJSrHXo93YyMzPl/fffl7p164q/v7+EhYVJt27dZMuWLVnHl5SUJFOnTs36kZGfvyaePsbk5GQ5cOCAXLx40fH5fO+996Rs2bIycuRIMcbc1Y+K5QdvnYmqVatKREREtszHx0f69OkjKSkpcuzYMfdfrIIz4dmZCAkJkeLFizuu81beOhOa2+eOJ3/ckTPh2Zm4zRgj169fF2NMjv+MN+BMcCY8PRPou359+/YVEZH9+/c7/vn85q0zsWbNGhERGThwYLZ84MCBcuvWLfn2229dfqW6smXLSq1ateT48eMiIhIXF5f1Y8bvvfeeREZGip+fn+zbt09ERA4cOCADBgyQkJAQ8ff3lyZNmsj8+fOtx927d6906NBBihUrJhUrVpRx48bBn7pAv/N869YtGTt2rFSvXl38/f2lXLly0q9fPzl69KjExcVJWFiYiIi8+uqrWTPx89fK08d47do1OXDggFy7di3Hz+vzzz8vNWrUkIcffjjHf+aXcvxj2zl16dIl6d69uwwcOFAefvhhKVOmjKs/n5ycLLGxsXL69Gn54x//KJUrV5Z169bJCy+8IGfPnpX33nsva+0LL7wgU6dOlePHj3tV0cOxY8fE19dXgoODs7IVK1bIzJkz5amnnpLQ0FCpUqWKnD9/Xlq0aJG1QYSFhcl3330nw4YNk+vXr8uoUaNEROTmzZvSsWNHiY+PlxEjRkj58uVl+vTpOf6d32HDhsnnn38u3bt3l0cffVTS09NlzZo1smHDBmnSpIlMnz5dHn30UWnWrJk8/vjjIiISGRkpIpInx7hp0yZp3769vPLKK/AN8Oe+//57adWqlXzwwQcybtw4uXTpkpQtW1b+8pe/yFNPPZWjrz+/FaSZOHfunIiIhIaGuv6zd8KZ8NxM/Bp480ykp6fL1atXJTU1Vfbs2SMvvfSSFC9eXJo1a+byq7wzzoTnZ6JatWqSmJgogYGB0qdPH3nnnXdcn1v5hTPBmcjrfSKv9ve84o0zkZKSIr6+vlK0aNFseUBAgIiIbN26VR577DFXx6lJS0uTkydPSunSpbPlU6ZMkVu3bsnjjz8ufn5+EhISInv37pX77rtPKlSoIM8//7wEBgbKzJkzpU+fPjJ79uysfzg5d+6ctG/fXtLT07PWTZ48OUc/xp+RkSG9evWS77//XgYOHCgjR46UGzduyLJly2TPnj3SqVMnmThxovzP//yP9O3bV/r16yciIvXq1RMRyZNjnDt3rgwdOlSmTJli/cMJsmnTJpk6daqsXbtWfHx8HNercvttb/RjFrGxsUZEzD//+U9rvSjfxo+IiMj2Iy+vvfaaCQwMNIcOHcq27vnnnze+vr4mPj4+KxsyZEiuvoXvyR/brlmzpklISDAJCQlm//79ZsSIEUZEzP3335+1TkRMoUKFzN69e7P9+WHDhply5cqZixcvZssHDhxoSpYsaZKTk40xxrz33ntGRMzMmTOz1iQlJZmoqCjrx3qGDBliIiIisv73ihUrjIiYESNGWMefmZmZ9f9rP3qUF8d4+8eRnH6s4/Lly0ZETOnSpU1QUJB56623zIwZM0y3bt3U8yw/FeSZMOanHwcODw83bdq0cf1nb+NM5O1M/FJB/LFtb5+J9evXGxHJ+r8aNWrc1Y93cibyfibee+8989RTT5kvvvjCzJo1y4wcOdIULlzYREdHm2vXrjn++XuJM8GZuNf7xM+PydfX1zpH8ltBmol33nnHiIhZs2aN9ZgiYnr16nXHP6+JiIgwXbp0yZqJnTt3moEDBxoRMX/605+MMcYcP37ciIgpUaKEuXDhQrY/37FjR1O3bl1z69atrCwzM9O0atXKREdHZ2WjRo0yImI2btyYlV24cMGULFnS+vpjY2Oz3St99tlnRkTgr1Henok7/dh2Xhzj7R/lnjJlivX3oWNs1qyZ+e1vf2uM+f/PZ25+bNvjN89+fn7w5+9zerLXq1fPdOvWLesEuv1/y5cvNyJi/v3vf+f2kLN48ub55xuKiBgfHx/Ts2fPbL97JSKmffv22f5sZmamCQ4ONo8//rj1td4+GdauXWuMMaZLly6mXLly2d6wjTHmb3/7m+MGMHz4cOPj42MuXbp0x68FbQB5dYw5FR8fn/W8fv3111l5RkaGqV27tqlYsaLrx8xLBXkmMjIyTLdu3UzRokXNjh07cv04nIm8nYlfKqg3z948E9euXTPLli0z8+bNM2PGjDGNGjUyCxYsyPXjcSbu7Uzc9sUXXxgRMW+88YbHHtMTOBOcifyYidvzMGbMGI88nicVpJk4e/asKVmypImOjjZLly41x48fN5MmTTIlSpQwImI6duzo+jFvH/svZ8LX19cMGjQo6x9abt/sDR06NNufvXTpkvHx8TGvvfaa9bW++uqrRkTMqVOnjDHGVK9e3bRo0cL6+5988knHm+eePXua0NBQk5aWpn4d2s1zXh2jG5999pkpVqxY1j+a3M3Ns8d/bLtChQrWjzO4cfjwYdm1a1fWz83/0oULF3L92HmhSpUq8sknn2R9LEZ0dLSEh4db66pWrZrtfyckJMjVq1dl8uTJMnnyZPjYt7/WEydOSFRUlPUjBjVq1HA8vqNHj0r58uUlJCQkp1/SPT9Gze0f0ShSpIgMGDAgKy9UqJA89NBD8sorr0h8fLxUrlw513/HvVAQZuJPf/qTLF68WKZNmyb169e/q8fiTOTdTPxaePNMlChRQjp16iQiIr1795Yvv/xSevfuLdu2bcv1bHAm7v1M/O53v5NnnnlGli9fLs8//7zHH9/TOBOcCbfHmFNr1qyRYcOGSdeuXeX111/32OPmNW+cibJly8r8+fNl0KBB0qVLFxH5aT7+8Y9/yJAhQ7K1/bvVvHlzGTdunPj4+GR9LNzPf4Xhtl/OxJEjR8QYIy+//LK8/PLL8LEvXLggFSpUkBMnTkjz5s2t/57TmahRo4YULuz+1vFeHaPm+vXr8sILL8izzz4rlSpVyvXj3Obxm2e39fcZGRnZ/ndmZqZ07txZxowZA9dXr14918eWFwIDA7M2lTv55fNy+xffH374YRkyZAj8M7d/TyC/5Pcx3i4UCA4Otj5y5PYme+XKFa+/efb2mXj11Vfl448/lgkTJsigQYPu6rFEOBP5fYwFgbfPxM/169dPBg0aJF9//XWubxQ4E/lzjJUqVZLLly/ny9/tFmcC40zcnZ07d8oDDzwgderUkVmzZuXqxie/eOtMtG3bVo4dOya7d++WpKQkqV+/vpw5c+auHlPkp99Fv5uZ+POf/yxdu3aFfyYqKirXx+UJ+X2Mb7/9tqSmpspDDz0kcXFxIvJTUZ/IT/cRcXFxUr58+Rz/Y809m6JSpUpZzYypqaly9uzZbFlkZKQkJibm6AQqyMLCwqR48eKSkZHh+LVGRETInj17xBiT7V8nDx486Pj3REZGypIlS+Ty5ct3/BdU9Ivz9+oYNYUKFZIGDRrI5s2bJTU1NdtJffuNSvsXxYLAG2bio48+krFjx8qoUaPkueee8/jju8GZIG+YiV9KSUmRzMxMV22ensKZyD1jjMTFxUnDhg09/tj3EmciO85Ezh09elS6desm4eHhsmjRorv6rqg38YaZ8PX1lQYNGmT97+XLl4uI5Mu9S7Vq1UTkp5/SzMn5dvjwYSvP6Uxs3LhR0tLSpEiRInCNVsJ1r45REx8fL1euXJGYmBjrv40fP17Gjx8v27dvz/aa3knef3bT/4mMjJTVq1dnyyZPnmz9S9FvfvMbWb9+vSxZssR6jKtXr0p6enrW/77bj1vIT76+vtK/f3+ZPXu27Nmzx/rvtz+qQUSkR48ecubMGZk1a1ZWlpycrP440M/1799fjDFZH1r+c+ZnH+cRGBhovRnl1TG6+biFhx56SDIyMmTq1KlZ2a1bt+SLL76Q2rVrS/ny5R0fw1vl90zMmDFDRowYIb///e/l3XffzeVX4TmcCXcfy/NrlJ8zcfXqVbjmX//6l4iINGnSJMdfh6dwJnI2Ez//O26bOHGiJCQkSLdu3Rz/vDfjTGTHmcjZTJw7d066dOkihQoVkiVLlhTobzT8Un5fO/1SQkKCvPnmm1KvXr18uXkODw+Xdu3ayaRJk6x/QLh9fLf16NFDNmzYIJs2bcr237/44gvHv6d///5y8eJF+fDDD63/dnsmbreO/3Im8uoYc/pRVSNGjJC5c+dm+79JkyaJyE8fUzd37lzrx+Hv5J595/nRRx+VJ554Qvr37y+dO3eWnTt3ypIlS6zK/GeffVbmz58vvXr1kkceeUQaN24sSUlJsnv3bpk1a5bExcVl/Rk3H7ewevXqrGFLSEiQpKQkGTdunIj89CMYbdu2zVrr4+MjsbGxsmrVKs89AcCECRNk5cqV0rx5c3nsscekdu3acvnyZdm2bZssX74868fNHnvsMfnwww9l8ODBsnXrVilXrpxMnz496yS9k/bt28ugQYPkgw8+kMOHD0u3bt0kMzNT1qxZI+3bt8/6uKfGjRvL8uXL5d1335Xy5ctL1apVpXnz5nlyjG4+buGPf/yj/Otf/5Lhw4fLoUOHpHLlyjJ9+nQ5ceKELFiwwOUz7l3ycyY2bdokgwcPltKlS0vHjh2tN6VWrVpl/UuhCGfCm2bi2rVr8o9//ENERH788UcREfnwww8lODhYgoODC8xHuCH5OROrVq2SESNGyIABAyQ6OlpSU1NlzZo1MmfOHGnSpIn1mZCcCe+ZiYiICHnooYeyPo937dq18vXXX0uDBg3kj3/8o8tn3LtwJmycCeeZ6Natmxw7dkzGjBkja9eulbVr12b9tzJlykjnzp1z+nR7nfy+n4iNjZWWLVtKVFSUnDt3TiZPniyJiYmycOFCKVTo/39PMi4uTqpWrSpDhgyRzz//3NNPQzYfffSRtG7dWurWrSuPPfaYVKtWTc6fPy/r16+XU6dOyc6dO0VEZMyYMTJ9+nTp1q2bjBw5MutjoCIiImTXrl13/DsGDx4s06ZNk6efflo2bdokbdq0kaSkJFm+fLk8+eST0rt3bylWrJjUrl1bZsyYIdWrV5eQkBCpU6eO1KlTJ0+OMacfVdWoUSNp1KhRtuz2j2/HxMRInz593D3huaosM3o7XkxMDFyfkZFhnnvuORMaGmoCAgJM165dzZEjR6x2PGOMuXHjhnnhhRdMVFSUKVq0qAkNDTWtWrUyb7/9tklNTc1a5+bjFl555RWrye72//28Fe7GjRtGRMzAgQMdH/NOX+/PiYgZPnw4/G/nz583w4cPN5UqVTJFihQxZcuWNR07djSTJ0/Otu7EiRPmgQceMAEBASY0NNSMHDnSLF682LEx0hhj0tPTzVtvvWVq1qxpihYtasLCwkz37t3N1q1bs9YcOHDAtG3b1hQrVsyISLbXxNPH6PbjFs6fP2+GDBliQkJCjJ+fn2nevLlZvHhxjv7svVSQZuJ246f2fz+v/edMeNdM3G6IRP/3y68zvxWkmThy5IgZPHiwqVatmilWrJjx9/c3MTEx5pVXXjGJiYnW382Z8J6ZePTRR03t2rVN8eLFTZEiRUxUVJR57rnnzPXr1x3/7L3GmeBM3IuZuNP+7olPm/GkgjQTxhgzevRoU61aNePn52fCwsLM7373O3P06FFr3e7du42ImOeff97xMSMiIkzPnj3vuMapHfro0aNm8ODBpmzZsqZIkSKmQoUKplevXmbWrFnZ1u3atcvExsYaf39/U6FCBfPaa6+ZTz/91LFt2xhjkpOTzV/+8hdTtWrVrHN6wIAB2b7+devWmcaNG5uiRYta56unj9HNR1X90t20bfsY87OfPyFZtGiR9OrVS3bu3Cl169bN78MhynecCaLsOBNE2XEmiLL7+OOPZcyYMXL06FEpU6ZMfh8OedA9+53ngmLlypUycOBAvvkT/R/OBFF2nAmi7DgTRNmtXLlSRowYwRvnXyF+55mIiIiIiIjIAb/zTEREREREROSAN89EREREREREDnjzTEREREREROSAN89EREREREREDnjzTEREREREROSgcG7/YIsWLWCOyrvLli0L16alpcE8MjIS5unp6VZ2+vRpuHb79u0w79Chg5Xt3bsXrq1Tpw7MtYLy1NRUK4uIiIBrMzIyYH7o0CEri4+Ph2tv3rwJ87Zt28K8aNGiVoaOWURk3bp1MK9ataqVtW7dGq49ceIEzI8cOQLzoKAgK/P394drv/zyS5jnp6ZNm+Z4rduZqFatGszRTJw6dQqu3bVrF8zbtWtnZZ6aCfT1VKpUCa7VZgKdLydPnoRrtZm47777YO7n52dlKSkpcO3GjRthjma8VatWcK123EePHoV5YGCglQUEBMC1X331FczzE2fC5u0z4Waf8JaZKFasGFz79ddfwzw/NW/ePMdr8/LaSZuJHTt2wNzNtVNMTAzM3cxE5cqV4VptJg4fPmxlbq+d2rRpA/MiRYpYGXpORdxdO2kzqB23du3kZia88dpJex4yMzOtrEKFCnCtNhPR0dE5Xq+9F23ZsgXmXbt2tTJtT6lXrx7M3cxElSpVcrxWBN9PxMXFwbVJSUkw79ixI8zRPqEdx+rVq2GO9nC094rox33w4EGYo/sJNCciIlOmTIH5nfA7z0REREREREQOePNMRERERERE5IA3z0REREREREQOePNMRERERERE5IA3z0REREREREQOfIxW9ebgpZdegjlqJbxy5QpcqzViowZcEZFSpUpZ2fXr1+Ha0NDQHB+f1rxYsmRJmGvNo1evXrWy8uXLw7VamyJq0kWtcSIiDRo0gPmkSZNgjpoLy5UrB9fWrl0b5oUL2wXt586dg2uTk5NhrrVUomPR2tQXLVoE8/z0/PPPwxy1EmozoTWdovNWRKR06dJWps0EWqs99q1bt+DaEiVKwFybCXQsWoOs1nSJWkO1mahfvz7MP/30U5ijc1GbiZo1a8Icvb5uZ0JrqUTHcvbsWbj2u+++g3l+0mbCzT6xc+fOHD+GiEhISIiV3bhxA651MxPaPqHNhPaaopkoU6YMXKs1Inv7TKDn7/z583Ct1vLqZp8oSDORl9dO6L1IBM9EXl47uZ0JdO2k7RNuZqJ48eJwbUG9dtL2WdRAXZCunV577TWYo/P58uXLcO3mzZthrn1iC5qJa9euwbVhYWE5Pr78uJ/QmqhRu7Q2E02aNIH5P/7xD5ij92etCV37RAr0/J05cwau1fYJ7T0FPVfaY8+bNw/md8LvPBMRERERERE54M0zERERERERkQPePBMRERERERE54M0zERERERERkQO7wSCHtm3bBvMqVapYWXh4OFz75z//GeYvvvgizFF5Rv/+/eHalStXwhxxU3pypxwVFvTo0QOuPXr0KMxRWdqPP/4I12oFSxUrVoQ5gkoJtOMQwcetlRhoRQha0ca+ffusTCvS8UZa2VdERISVaQUUo0ePhvnLL78Mc1Rw0bdvX7j2hx9+gDlSo0YNmLudCfQ+0aVLF7j22LFjMEdFgRs2bIBrtTIZ7RxFfYlakY42E8ePH7cy7RxPSUmBuVaOs3//fiv7b5uJUaNGwfx///d/YY5mok+fPnDt6tWrYY64nQlUECTibibQuSUiEhwcbGV5ORNakU5ezoS2vqDPhJtrJ20mnnnmGZh74tpp1apVMEfy8tqpe/fucK02E2if8NS1k5uZ0Erb0P6mzaBW2IneN0UK/rWTVvYVGRlpZdr9hFbEp11ToeKxgQMHwrVLly6FOaKVY2n7gXa+bNy40cp69eoF1x45cgTmqCxN2/O00jFP3E9oXyO6n9BKx7R9Qpuh3bt3W5l2nZUb/M4zERERERERkQPePBMRERERERE54M0zERERERERkQPePBMRERERERE54M0zERERERERkQMfg2oEc+BPf/oTzFF7mtbIiNruRPTGtk6dOlnZ3//+d7i2UaNGMEfNo76+vnCt1vpWunRpmKN20Dlz5sC1jRs3hvn9999vZVu2bIFrz549C/OLFy/CvESJElamvfxpaWkwR+2q2vOn0ZrGY2JirAw15omIrFmzxtXfeS88+eSTMEftoNWrV4dr3c5Ex44drez999+Haxs2bAjzM2fOWJmPjw9cqzUvoqZTETwT3377LVxbt25dmKOWSm2utDZFrSk5KCgI5kh6ejrMvWUmtGbZ/PTfMBPaPhESEgLz/4aZQO2q2kxoz+uvdSby8toJtWqLiHTu3NnK8uPaSZsJ1Jirnc/16tWDOWry/uabb+Barclde7695drp8OHDMEfvEwXp2kn75B3UIl2rVi249tKlSzDXZgV9Es4bb7wB1zZt2hTmqLVde021a6fQ0FCYoxmaMWMGXKvtY2gmvvzyS7hWmwmtPbx48eIwR/Ly2unQoUMwb9CggZVt374drnXz6Uy38TvPRERERERERA5480xERERERETkgDfPRERERERERA5480xERERERETkgDfPRERERERERA7sys4cQi1zIiKbNm2ysoCAALhWa9msUqUKzFHrNGppvBPUJIoa6UT0RuxVq1bBHLW/opZLEb2NskWLFlY2d+5cuLZQIfxvH1oLbXh4uJVpr4HWGImaDrVWTO1r1F4z1Kx48+ZNuNYboTZSETwTxYoVc/XY2kxs3brVytzOBJpPbSa0VketwfPatWtW1rt3b7gWtSSL4K+9S5cucK3WrpqQkABz1HSpzZXWGIlai7X2ZO21cTMTt27dgmu90X/DTGjtxG5mokOHDnBtamoqzNE+pjVza62oWjPzvZ4JreX11zoTeXntVLVqVZhv3rzZyrTnV2uRRuccan0X8cy1E/p0FRF9JtB5rjVzazOhXTuFhYVZmfYaaO8Tnrh20hqbC/q104kTJ2CO2vK1fUJ7PSIjI2G+fv16K9OeX42b+4nmzZvD/Pvvv4c5en/u3r07XJuSkgJzdJ5r+5XbfaJMmTJWpr0G2j5Ru3ZtK5s5cyZcq702lSpVgvmBAwesLDk5Ga7NDX7nmYiIiIiIiMgBb56JiIiIiIiIHPDmmYiIiIiIiMgBb56JiIiIiIiIHPDmmYiIiIiIiMhBrtu2tfbN6OhoK9PaEUuWLAnz48ePwxw1T2qNq1qD57lz56zM19cXrkVtfCIi/v7+MEdQy6WISGJiIsyPHTtmZVrjZlJSEsy1Fk3UoB0cHAzXouZPEdyapz3XblrwRESioqKsLCMjA671Rtp5hGZCa4fVXo+4uDiYo3MjIiICri1btizM0UxoLaobNmyAudbUWKdOHSvTGmG1cw61v2rH16ZNG5hv3LgR5uh81t6XUNu/9hjaTGiNkQcPHoQ5mgmtudIbectMaI2rPXv2hPn58+etzO2nGwwdOhTm6Nw4e/YsXBsfHw/zpk2bWpn2NZ4+fRrm2vuwJ2YC7fl5uU8UpJlwc+2ktehqM+GJaydtn3AzE+vWrYO5du2EGoq1946iRYvCPCQkxMoCAwPhWtQoL6KfR26unbRrPk9cO+3fvx/mv9Zrp5o1a1qZ25k4evQozNG5oV2baO3n6H1bmwnt0xe0mUCPo11/aXthq1atrExrA0eN8iIipUqVgjl6j9fWaseN5kp7rrXr2r1798IcfeKDJ/cJfueZiIiIiIiIyAFvnomIiIiIiIgc8OaZiIiIiIiIyAFvnomIiIiIiIgc5LowTCsIQsUUycnJcK1WEoAKgkTwL8WjogQRkaVLl8IcrdeKarRfwtdKBapVq2ZlZcqUgWtfeuklmCckJFgZ+qV6EZGTJ0/CvHXr1jBHX8/NmzfhWo1WWIFohWaVK1eGudtj8TZaOQk6X7RzTpsJrfwDvaZacdD3338Pc7ReOz6tkKZLly4wf+ihh6xMK0fSCiFQWZE2m9o5h2ZTexzta9doxUaIVhSoFcRo750FhZuZ0ObfEzPRtWtXuFYr+6pdu7aVud0ntKJHtCdkZmbCtVoBHtontCIlrZAmL2fi0KFDOV773zYTQUFBMEfvw26vnbTzyM0+sWzZMph74tqpRYsWMP/DH/5gZT4+PnAtKgISEblw4YKVaQVGK1eudPXYR44csTK3M+GJaydtJgr6tZM2E+ic075WrURUmwlUJFa9enW4dtGiRTBH67XzAp1DIvo9DHp/1sq+/vWvf8EcvU+8+OKLcK12PxEbGwtzT8yEVvaFXL9+HeZa8WFe7xP8zjMRERERERGRA948ExERERERETngzTMRERERERGRA948ExERERERETngzTMRERERERGRg1y3bWsN0KGhoVamtaSFhITAXGvfRLSGvWbNmsG8UCH73wu2bt0K1953330w15okUbNdgwYN4Fqt5RUd3/bt2+FarYl18+bNMEdtlNpjtG3bNsePrbW2as172td+6tQpKytXrhxc643S09Nhjs7z+Pj4HK8V0ds3EW0mGjduDHN03mrn3MCBA2Fep06dHB6dyJIlS2CuzT1q8tbef7TW75dffhnmaCa0tnptVrZs2WJlWgOk25k4c+aMlf1aZ+LGjRtwrdaY62YmypYtC/MmTZrA3M37cKtWrWB+4sSJHD+29gkTbvYJ7bnW9rG//e1vMEfvw9pMaDnaUzkTP9Fep7y8dkLnl9bc64lrJ20mtPZ91CKttf/OmTMH5uhaCx2ziMjYsWNh/v7778N83759VuaJayftk1tSUlJgfu3aNZj/Wq+dwsLCrAy1ZIuIlC5dGubavoJoM9GyZUuYo/NLu/5u164dzLVPMEGf1jB48GC4VvvkCdQ4rb1na5+Cos0hao/Xzv327dvDfMOGDVamtY+73SfQcVeoUAGuzQ1+55mIiIiIiIjIAW+eiYiIiIiIiBzw5pmIiIiIiIjIAW+eiYiIiIiIiBzw5pmIiIiIiIjIQa7btrUGuxo1aljZxYsX4dorV67APDMzE+b169e3Mq0hTmszRu2GWkuf1paKvkYR3AI4bNgwuLZo0aIw//zzz61Ma15cuHAhzLVWZdRKV7x4cbh27969MEfN2trrqzVdVq5cGebly5e3skOHDsG13khrRo2OjrYy7dx3OxN169a1snPnzsG1p0+fhnlERISVaW2uWuOhNhOzZs2yMq0dMTw8HOarV6+2Mq25991334V5yZIlYe5mJlC7pHYsly5dgmtRs7kIbpsVwTNx+PBhuNYbeWImtPPFzUxoDe+opVZEpGnTplamvWcfO3YM5jVr1oT5pk2brKxevXpwrdb+unjxYivz9/eHa7/66iuYFytWDObo6wwKCoJrDxw4APO8nAnUnK418Hoj7dqpevXqVqbtrZcvX4a5J66dtHZddO2k7RPa9VfXrl1hvmDBAivTGoS1x/7oo4+s7Mknn4Rr//Of/8BcayJGM+uJa6eEhAS4VpsJ7doJNWsXpH1CaxGvVauWlWkzob2/aDPRqFEjK9P2A20m0Guq3U9ordraPoHe+7XrntmzZ8Mc3U9o9yTnz5+H+RNPPAHzSZMmWZk2E3v27IE5atZGn4Aios+E9pygZu2DBw/CtbnB7zwTEREREREROeDNMxEREREREZED3jwTEREREREROeDNMxEREREREZGDXBeGaUUw6BeyGzRoANdqBTzaL4ajchKt9AT9sriISGhoqJXt3LnT1WNo5QGtWrWyMq2oSPuleFR6gAqdRPQCqDp16sC8WbNmVobKmERwAYUILvHQymRWrVoF81KlSsEcPa83b96Ea72RVgSDSs9QgYuIXsCjla+hUhCtOEh7TVHpizbfDRs2hLlWtIHKHFDhjwguzBHBM6G9RwQEBMBcK1pp0qSJla1duxau1Y47PT3dygIDA+HaNWvWwDw4OBjm6HnlTPxEmwm0B2n7xI4dO2COSve0IiCtgEWbibCwMCsrUaIEXKsVCqGiGm0mkpOTYd6pUyeYowK0H374Aa7V3lPycibQPqEVGXojrQDPEzOhnQNuZsLNtZN2DaI9hna9gY5bK1LSyqUiIyOtTHvfP3PmDMz9/Pxg/uCDD1pZXl47afP2a50J7TXdt2+flWmFuNr7szYT6LG1mahYsSLM0Xu5VminnYva69S+fXsr27hxI1x75MgRmKPj1t6HY2NjYa69B7Vp08bK5syZA9dq7wdon9BmYsWKFTDX7ifQ66DthbnB7zwTEREREREROeDNMxEREREREZED3jwTEREREREROeDNMxEREREREZED3jwTEREREREROch123aZMmVgjtrdtFZtrXVSa0JEfydqHRURWb9+PcxRE2LTpk3hWq1JUmsGrFu3rpVpjdNa+9zZs2etLDExEa5t3bo1zLWmRtQsXK9ePbhWe/5Qk7fWKlujRg2Yaw2yPj4+VoaeU2+FmhdF8GuttaVqM4HOC+3v1GZCa2pELdKoOV5Eb/LW2rnR8Z0/fx6uTUtLgzmizUSLFi1g/vXXX8MczYTWVr9hwwaYo/Xae1h0dDTML168CHM0EzExMXCtN/o1zcTvf/97uFb7NARtn0DtoFpzr7ZPINpMdOnSBeao3VkEt8Jqjavr1q2DeV7OBGpZL0gzobX2o2sntzOhPcfh4eFWps2E9j6Hrp205mNtNrWZQJ/4sHLlSri2UqVKMEeN9dpMaPvbW2+9BXP0npIf107/bTOB3ou0Vu0rV67AXPt0HHQ+R0VFwbXaJwKgx2jZsiVciz7BQUSkc+fOMEcN2v/617/gWjf7hLbXjB8/Hub//ve/YX78+HEr0/aJH3/8EeZovfZ61apVC+Zu7ie0T37KDX7nmYiIiIiIiMgBb56JiIiIiIiIHPDmmYiIiIiIiMgBb56JiIiIiIiIHPDmmYiIiIiIiMhBrtu2fX19YY5a6VAjnYjIhQsXYJ6ZmQnz0NBQK9Na3NLT02GOWum09r6UlBSY165dG+ZFixa1Mu1r19SsWdPKUCOwiN5oqTVU79y508q010BrAETtkFrTn9asqTUXPvbYY1b23XffwbXeqEiRIjBH52j58uXhWq050BgDc9SiqrV9ZmRkwBwdy759++DaEydOwFxrrEezfOzYMbhWa/JGjYxai/Nf//pXmGsN2rt27bIy7TXo2LEjzNFMaK+v1pKrvWZDhw61sqVLl8K13qhwYbzFoJnQ3ivzcia0fQK9ftu2bYNrtfe/4sWLwxwdt9Y2q0F7kDYTWsur1saLWri1BlStbRl9UoX2+mr7hJuZWLJkCVzrjbSZcHPtpL2m2rUTOje0pvS8vHbSZgIdt/a1a+32f/jDH6wMtTWL6NeNpUuXhnleXTtp+4T26RW/1msnbSZQ4zpqpRfRr4c16NN7tCZqbSbQ+bVnzx64tnv37jBH+5WIyPTp061M2ycCAgJgjmalX79+cC26fxHR73e++eYbK9M+SaVbt24wj4+PtzJtP9WunbRPFxo+fLiVzZ8/H67NDX7nmYiIiIiIiMgBb56JiIiIiIiIHPDmmYiIiIiIiMgBb56JiIiIiIiIHPDmmYiIiIiIiMhBrtu2S5YsCfMJEyZY2fvvvw/XRkZGwlxr8kYNnlpDnNZ4GBgYaGVa+6/WMBoVFQVzN817NWrUgDlqDdXas7VWusTERJjXq1fPyrT25GLFisEcNUprLa/aY8TGxsL84sWLVubn5wfXeqMSJUrAfNy4cVb20UcfwbVVq1aFudZG6WYm2rZtC3M0E8ePH8/x3ycismrVKpj37t3byrSG1kcffRTmqF1aa20NCgqCudZeimYLNUCK6G3g6PlGnwxwp8do3bo1zC9fvmxlWqu7N/L2mejUqRPM0Uygxl0R/Xxu06YNzDt06GBlc+bMgWubNWsG80mTJlmZdg6hGRQRuXnzJsxRM702E1rjMHq+3e4TbmZCe329kTYTb7zxhpV98MEHcG10dDTMCxXC3w/xlmsn7f2vb9++VpaWlgbX/uUvf4E5atLdv38/XKtdl2l/572+dtLak3+t107BwcEwf++996zs7bffhmu1a2ptJtDrpz1nPXr0gDmaCe0c15qov/rqK5ij9+GjR4/Cta1atYI5asR+4IEH4Frtnu7MmTMwb9CggZXFxcXBtdr5jJ5v1IJ+p8dA+6lI3s8Ev/NMRERERERE5IA3z0REREREREQOePNMRERERERE5IA3z0REREREREQOcl0YppWkvPPOO1ZmjMF/uVL4ov2CPyoEuHHjhqvHRgUSWmlDtWrVYN6nTx+Yo5Il7TimT58O8/T0dCtDv/h+J9p69DqEhITAtVrpAfpl/uTkZLhWK6TRyjoaN25sZaiow1tp5SR///vfc/wYWnGGj48PzFFxjFYYpxXxHTx40Mq0mdDKm0qXLg1zVCijzcTmzZthjmYClYiJ6IUvly5dgnlmZqaVaWVkCQkJMEfPlTYTt27dgvm+fftg3rBhQysrSDNx4MABmLuZCa3IRNsn8momtBI47XzR9qaPP/7YyrTzVivoQ0Vd2vut9jxphZ1oJrSCqkOHDsGcM6HTvi507aTR3kM16NpJK1F0c+2kzYR27aSdR+XKlbOyKlWqwLVr166FOSpT0q41tH3WzbWTVvymleiFh4dbmaeunRo1amRlBWkmtK9r/PjxOX4Mt/cTMTExVqbNhPYYaJa1YlZtJk6dOgXzI0eOWJm2X6GiYY12Lakdh1ZShr52ba8+d+4czNH9RFJSElyrzcquXbtg3rRpUytDc5Jb/M4zERERERERkQPePBMRERERERE54M0zERERERERkQPePBMRERERERE54M0zERERERERkYNct223b98e5t26dbOyuXPnwrXfffcdzFEroYhIVFSUlaG2NhGR9evXwxw191asWBGu1dr7tOP+85//bGUrV66Ea7V24iJFilgZaqEWEfn2229hjlrmRHAr7MmTJ+HaUqVKwRy1cGvteNu3b4d5586dYb5hwwYru//+++Fab6S1LHbp0sXKtNdu8eLFMC9btizMUfu1Nj/o+RXB7brly5fP8VoRkYULF8J8yJAhVqY19bdq1QrmX3/9tZVprd+zZs2CeZMmTWAeEBBgZdp5q7UCowZZtzPRsWNHmG/atMnKevToAdd6o4I6ExkZGVZWoUIFuNYTM7Ft2za4tmXLljBH7aqohVpbKyLyyiuvwBw1pmrHpzXQoz3I7Ux06NAB5mgmevbsCdd6o3bt2sG8a9euVjZv3jy4VrsG0a6HULO6J66dtJnQ2qy1JuqRI0damfbpC+gaSURk2bJlVoauR0XcXzsFBQVZ2ZYtW+Ba7RNM0Huh9ikAO3bsgHmnTp1gvnHjRisrSNdO2qyjr+Gbb76Ba7XXVLuWQfu2tqf8+OOPMEfnBWq2FxFZtGgRzLWv54knnrAy9N4nIhIbGwtz1EStneOzZ8+Gufa+XatWLStD56GIPhPoMbSZ0N4PunfvDnP0mvXt2xeuzQ1+55mIiIiIiIjIAW+eiYiIiIiIiBzw5pmIiIiIiIjIAW+eiYiIiIiIiBzw5pmIiIiIiIjIQa7btnfu3JnjXGtkLFeuHMxv3rwJc9R2e/XqVbhWa6+9cuWKlbltwNWaqFF7Ze3ateHaqVOnwhy1zyUnJ8O1devWhfmxY8dgXqxYMStr0aIFXKs1WqIGu/j4eLhWaznUjg81Y2rtgt4INRtquduZSElJgTmaiWvXrsG1WqPl5cuXrUz7WtD5KaK3x6OZaNasGVy7bt06mKelpVkZmmMRkZiYGJi7mQntNUANzCL4/UNrsddm4vjx4zBHzfRa66Q3ysuZcLNPePtM1KtXD67Vmo+rV69uZdrzoTUf7969G+aocVZ7DbR9whMzERcXB3PU1F6QZsLNtZP2SSDa66F9IoCbmdCue9BMaK8pagoW0c9nNBOpqalw7UcffQRzNBN5ee2kvV+hVnIR3Fjv9tpJ2yfQtZPWfOyNtDZ/lEdERMC12qxo74vo3PD19YVrtdb2KlWqWNnhw4fhWm02w8LCYI5mvFGjRnDtDz/8APMBAwZYmfZ8oE9ZENFn1hMzgd63tfd97fU9cuQIzHv16mVl2jXm008/DfM74XeeiYiIiIiIiBzw5pmIiIiIiIjIAW+eiYiIiIiIiBzw5pmIiIiIiIjIQa4Lw86cOQPzqlWrWlmlSpXgWq2o68UXX4T5hx9+aGWFCuH7/8qVK8P80KFDVubv7w/XVqtWDeZt2rSB+YYNG6xMK43RiglQUcSpU6fgWq0kRful/fDwcCtDhQciItevX4e5McbKtKIBrZCmefPmMF+0aFGOj88bnTt3DuboXNTKD7QymWeffRbm//znP63s0qVLcK02h6hwoWjRonAtKokQEenZsyfM9+/fb2V9+vSBa/fu3QvzwMBAK9Pef7Zs2QJzbSZQWQcqKBPRizZQaZLbmWjatCnMFy9ebGWciZ94YibclJB4aiZQCWJUVBRcq+1vqEzm9OnTcK02E+3atYN5aGiolWmFL5GRkTBHpTl5ORNaiZA3Onv2LMzRXLu9dnrhhRdgjkq2tHNLmwl07dS3b1+4NiQkBOaoxEdEZPbs2Va2b98+uNbNtZNWaKYVkWrPN9ontJlISkqCOXr/4LXTT7SZQPcT2rW99j43duxYmL/zzjtWdvHiRbhWe385cOCAlWmFwgEBATDv168fzFE5b82aNV09dvfu3a3s6NGjcC16PkREfHx8YI5K/txeO/n5+VmZVnCpvefdd999MJ83b56VaftVbvA7z0REREREREQOePNMRERERERE5IA3z0REREREREQOePNMRERERERE5IA3z0REREREREQOct22XaNGDZifOHHCymJiYuBa1E4tglu1RXDLota8+OWXX8K8du3aVqZ9LVqr39y5c2GOvk6tFVhro0TPidbe+Nvf/hbmS5YsgTlqys7MzIRrUYOwiMiePXusTGuM1JrttBZN1P6pNRd6o+joaJijxs/U1FS4duPGjTBHDcIi+PzXWlRnzJgBc3T+V69eHa7Vmum//fZbmA8ePNjKrly5AtcWLozfjtBxazPx0EMPwXz58uUwR+e/25lALeFaY6TW4I9ayUXw66s9tjfSZiI+Pt7K8mMm1qxZA/PHHnvMylDzqwj+WkREXn75ZZjXqlXLyrS2We39Dz0nycnJcK02E7du3YI5os2Etr/d65koSPuEm2unlJQUuFa7dkKt2iLurp2+/vprmKNrJ5SJ6PuEtgeh9nitWVg77rVr11rZ5cuX4drf/e53MF+6dCnMEW0mtFZ+N9dO2ky4uXYqSPsEek8UETl+/LiVaTOBXn8RvUUa7efauTV16lSYP/LII1ZWp04duFa77pk+fTrM69WrZ2XaJ+/Ur18f5rt27bIy7b2jY8eOMEdN7iKeuZ/YsWOHlWn3Rto1qdZMj64ntcfODX7nmYiIiIiIiMgBb56JiIiIiIiIHPDmmYiIiIiIiMgBb56JiIiIiIiIHPDmmYiIiIiIiMhBrtu2tbbbunXrWtmUKVPg2tDQUJiHhYXBfMuWLVbWr18/uPbIkSMwv3btmpXt3LkTrtUaD7t27QrzhIQEKztw4ABcqzXHoTY9rblSa3m9efMmzFGzY2BgIFxbtmxZmJcpU8bKtOcJtUuKiERFRcEcNch+8803cK3WtJuftLZb1MI+bdo0uLZ06dIw12Zl27ZtVta7d2+49tixYzC/ceNGjh5XBM+3iMjkyZNhjhqUr169CteidkkRkU6dOlmZNlenT5+GuTYTqJFUe2/TZuL69etWpjUzu50J9J43e/ZsuPaTTz6BeX7ylpnQ9on09HSYo8Zc1JovoreoovNWROTixYtWps2ENm+o5VrbJ7Smca1ZFu0T2kxoezXaJzgTP9GeS9TS+/nnn8O1nrh26tu3L1zr5tppxYoVcK02K126dIE5mom0tDS4Vrv+Ql87+rrvdHxurp08sU/k5bXTrFmz4NpJkybBPD8lJibCHLVIa8evnfva67F+/XorGzhwIFx78OBBmKP37QsXLsC127dvh3nPnj1hjq7LGjRoANeiTzUREVm1apWVXbp0Ca7V5k07z91cO6E2fRH8nqJ9Sg9qDhfRP70Avb7apzB98MEHML8TfueZiIiIiIiIyAFvnomIiIiIiIgc8OaZiIiIiIiIyAFvnomIiIiIiIgc8OaZiIiIiIiIyEGu27ZRG6kIbjFELZIiIsWKFYN5qVKlYN6+fXsra9GiBVw7d+5cmFeqVMnKUNPjnY6jSJEiOX7sHTt2wLU//vgjzFHDHmp6FBH59ttvYd6hQweYoxZNralOa4FEDa3omEVEoqOjYa79nah5Tzt3vJGbmUBtwyJ6Y652LsbGxlpZ8+bN4dr58+fDHL1OqC1XRKRjx44w1xqRCxe232K0Zs3NmzfDHLVlaufWsmXLYI5aO0XwTKBjFtFnArUZu50JrX0fzUStWrXgWm+kvVeiTxWoXbs2XOt2n0Az0bRpU7hWawuuWLGilaFPUxARqVKlCszR1yiCvx5tJtatWwdz1GoaFBQE12pz9cwzz8AczYTW1F+iRAmYe2ImtLygz4S2T6DzRdv/3O4T7dq1s7KWLVvCtfPmzYM5ur7RrmO0BnrUEi8i0rhxYyvLyMiAa9euXQtz9Jx07twZrn399ddhrl2bHD582Mrc7hNurp20Vm3t+FDr86/h2unEiRNWpu3lAQEBMA8JCYE5an5v3bo1XDtz5kyYo4Zq7RMmnn32WZhnZmbCHL2fa3vQZ599BnN0Xaa1bS9duhTmWkM+aiDX9nutQdvNTGjXSNp1A3oP0j69Ijf4nWciIiIiIiIiB7x5JiIiIiIiInLAm2ciIiIiIiIiB7x5JiIiIiIiInKQ68Iw9IveIrgMJTw8HK5FpSciIsePH4d5vXr1rOxvf/sbXKsVRaBSjvT0dLjWz88P5gsXLoQ5KuDQymS0giVjjJVpzxMqxhHB5RYi+Jf5tddm165dMEdFMBUqVIBr4+PjXT02Km7bvn07XOuN0HMjgosswsLC4FrttY6Li4M5KgV5++234dpu3brBvEmTJlamFdJoRTDa64RKHrRCs8uXL8McFWrcvHkTrh04cCDMUbmFiMjWrVutTCuC2b17N8zR616+fHm4FpXHiegzgco9du7cCdd6I22fQDMRGhoK12oFLG5m4q233oJrR40aBXNUwKMdh7ZP+Pj4wBwVwWjvoeg4RPA+oZVm3X///TBHxWAieJ9AxTgi7vYJtzOhzVtBnwltn0DXTtpMaO9/2rUTKsrxxLWTNhPaeaGVnN53331W1qBBA7hWKx1r1aqVlWnP0+jRo2H+wQcfwBzNRMOGDeFa7bxF7wd5uU8UpGsnN/cTqEBURL92Onr0KMzR+fXaa6/BtT169IA5KlLUSoK1wqsLFy7AHF2XaYWA2l5TvHhxK9OKLLWiYe3aCc2EVszp5v0ZlXWK6Pv9tm3bYI7uJ7Zs2ZLj43DC7zwTEREREREROeDNMxEREREREZED3jwTEREREREROeDNMxEREREREZED3jwTEREREREROch12zZq9xPBLb1aiy5qXBURCQkJgfnq1autrFGjRnCt1maNmhq1prpr167BPCoqCubXr1+3Mq39snLlyjC/ceOGlWntgqihUkTk6aefhjlqYUaNdCIiP/zwA8yrVatmZVq7IGrjExGpVKkSzFGjq9YI6o208xy18bqdieDgYJivWbPGyrQW0IiICJiXKVPGylJSUuDakiVL5vgxRERmzJhhZYcOHYJrtUbfxMREK0PNqiJ6Y+TixYthPnnyZCvTGje1x0CtmForqjYTWsNkamqqlb3++utwrTdyMxNak3tezoT2/oz2BO04tJlo1qwZzFGDrNZS2qVLF5jv27fPyrQZbN26Ncyfe+45mI8ZM8bK/vSnP8G12vPHmdBpM4GunbS2de1c1Np40bVT48aN4Vrt+gtdO2mfylCiRAmYR0dHwxy1dmsNzBr06R7aJztozcfDhw+HOWrrX7FiBVx7+vRpmKMZ11qIPXHtNGHCBLjWGyUkJMAcnV/avUdgYCDMtZlAr5/2nq3dT6Dmau3TF7T7iVu3bsEcvV+2b98ertWu7dCxaPcvbdu2hfmTTz4J8w8//NDKli5dCtdq7fEtWrSwMq09W5sJ7V4K7RPvv/8+XJsb/M4zERERERERkQPePBMRERERERE54M0zERERERERkQPePBMRERERERE54M0zERERERERkYNct22jJjMRkUKF7Ptxbe3NmzdhrjXKodbdrVu3wrVa+xx6DGMMXDtnzhyY9+/fH+aoHfLIkSNwbdGiRWGOjuXEiRNwLWr3FtGbLlGzo9asqUHth2fPnoVrUeOqiN7OjVoAv/vuO7hWawDMT25mQmtR1WaiXbt2MEctpdu3b4drtdcjKSnJyrSZ2L17t6scNSgfO3YMrtXaFFGz8IYNG+Daw4cPwxzNvQhutNSamQ8ePAhz1J6szYTWcKs1EaOZWLJkCVw7cuRImOcnNzOhtW3n5Uxozcdo/cCBA+Fa7f1Ze49H57l2vqBzS8u1ll/tfClevDjM3XzCQdWqVWHuZp/gTPwkL6+d0Hv85s2b4VrtNXVz7aS162rv8YUL25eiR48ehWu1pnHUuqvNpvbYbq6dtKZx7flzs0+4vXZCn4KifTrEU089BfP8pH3SBjovtPdsbSY6d+4MczQT2nVFZGQkzNFMZGRkwLVxcXEw12YCNWVr1yDaJ/Kg+w/tExK0a1Lt0yRQK7Y2E9rz52Yf05rQtXZu1Ki/YMECuPYPf/gDzO+E33kmIiIiIiIicsCbZyIiIiIiIiIHvHkmIiIiIiIicsCbZyIiIiIiIiIHuS4MQ6UdIiIVKlSwMq0U4cyZMzDXyl0qVqxoZVrxA1orgn/Bv0uXLnCtVqhx6dIlmE+fPt3KHn/8cbj266+/hjkqT9GKU8qXLw/zp59+Gubff/+9lWklBloRCBIYGAhzVHhwp/WIVlbgjbRSBPQ6nTt3Dq7VZkIrCkGPrc2EVg6Hio2Cg4Ph2v/85z8wr1WrFsw3bdpkZVqBxzfffANzVOClFZRps4IKVURExo8fb2UpKSlw7a1bt2COin60UhttJrT1yK91JrRzX5sVT8yE9h566NAhK9OKasqUKQPzjh07wvzLL7+0sk6dOsG1s2bNgrmbmdBKx2rXrg3zTz75xMq0/UDLPTET2nEjBWkm3Fw7aTOhXVN5+7VTo0aNYI6unbp16wbXakWujz76qJVp+4Fm9OjRMF+xYkWOH0ObCVQo6/baSZshNG/aHu6NoqOjYY5K4LRCKS3fsmVLjh8bve9ra0VwaW+fPn3gWq2MrGnTpjD/9NNPraxHjx5w7bp162COCgS1gjJt7p977jmYL1u2DOaIJ2ZCKyNzc+3kyZngd56JiIiIiIiIHPDmmYiIiIiIiMgBb56JiIiIiIiIHPDmmYiIiIiIiMgBb56JiIiIiIiIHOS6bVtrRLtw4YKVaS1zV69ehbnWyIhagbVm3OTkZJhnZmZamdbSp7XSaY2WaP20adPgWtSqLSJy5coVK9NaUTUzZ86EOWok1V7HU6dOwbxu3bpWhhoHRUR+/PFHmKMGQBGR9evXW1lYWBhc64205zIhIcHKNm7cCNdeu3YN5qhdVwTPRGpqKly7ZMkSmKPG9Tp16sC1GRkZMNeaj9FMoGZVEXczobV7azwxExcvXoR5TEyMld24cQOuRee4iEhsbCzMUVt56dKl4Vpv5GYmtIbevJyJmzdvwhyd5wcOHIBrtfe5gwcPwhy9p7399ttwrdZWvnTpUisrXrw4XKu916BPXxDxzD6BZkLbJ9zOBPp6QkND4Vpv5ObaSXvt8uPaKT093crcXjtt27Ytx+u1a6fXXnsN5qgNX/uUBa2x+9tvv4V5qVKlrEx7HU+ePAlzXjvptPcu9EkLa9euhWvRdYKISLNmzWDuZp9ATfMi+H5Ce/21mdD2PXR8qIFbRGT48OEwR+/lrVq1gmu1r/GLL76AOZqJoKAguDY+Ph7m9evXtzJtv1+zZg3MtU9vQes9ORP8zjMRERERERGRA948ExERERERETngzTMRERERERGRA948ExERERERETngzTMRERERERGRAx9jjMnNH5wwYQLM33nnHSvr1asXXKu1qmnNmajJzc/PD66dN28ezMuWLWtlWsuc1ubq6+sLc/Q4Pj4+cK3WaFmvXj0r05qMjx07BnOtuTA4ONjKKlasCNei5koR3H5YvXp1uFZrHTx//jzMe/fubWUnTpyAa6dMmQLz/DRu3DiYv//++1bWvXt3uFabCa1hGb3W2kwsWLAA5uHh4VamzQRqRxQRKVwYF/drj4NozceopXTXrl1w7fHjx2GuzUSJEiWszO1MoPM5OjoartXaiVH7tIhIz549rUxrrtTaafOTNhPvvfeelfXo0QOu9cRMoOZSEZGFCxfCnDORXX7MBGqfFsHXEwVpJjxx7aS1bWttsvf62qlBgwYw98RMJCUl5fjv5LWTzRuvndA1kojI66+/bmV9+/aFa7W2bfReLoLf57R9Yvbs2TBHM6F90kaTJk1grt1PaI/jxquvvmplf/vb3+DadevWwRw9TyJ4JiIiIuBa7ZNKzpw5Y2U1a9aEa7X3eG0m+vfvb2XaXjhp0iSY3wm/80xERERERETkgDfPRERERERERA5480xERERERETkgDfPRERERERERA5480xERERERETkAFcf5sDHH38M8z59+lhZoUL4Hv3WrVswT09Ph/mhQ4esrFOnTnCt1tiGmu2qVq2a47UiepsiasdLTU2Fa9u1awfzJUuWWJn2fFSpUgXmWnspek60JtsKFSrAHD1XWnPl0aNHYa69ZqhZtly5cnCtN9Ia++6///4cP4Z2vmRmZsIczUTHjh3hWq3t1s1MaG2pWtNlXs1ERkYGXKu1PZ4+fRrmqO30+vXrcK02E2gO4+Li4FptVjp06ABz1MqPGj69lZuZ0D6ZQDtftHPg1zQTsbGxMF+6dKmVac9H5cqVYa7tE5yJvKVdO6HGZO3aKSUlBebatcKBAwesrEuXLnBtjRo1YI6Oxe1M5OW10+LFi61Mm4n8uHZCf6fW/uv22gnNREG6dvr73/8O8wEDBliZdl2unS/aTOzbt8/KtE9BcXM/ERUVBddqM3Hp0iWYo/dcbe47d+4M80GDBlmZNhPVqlWDudZyXatWLSvTPgWgUqVKMI+MjLSyw4cPw7Va3q1bN5ijmShfvjxcmxv8zjMRERERERGRA948ExERERERETngzTMRERERERGRA948ExERERERETnIdWFY/fr1YR4SEmJl27Ztg2u10ob9+/fDHBWOhIWFwbXaL9ajkjLtazlx4oSrx0a/jK6VGGg5+gV67Rf5tXILN89fqVKl4FqtVACVu2hFP+3bt4f56tWrYY6KNoKCguBab1S3bl2Ye2ImDh48CHM3M5GWlgZzVExRr149uFYrj9BmApWWuJlNEXz+azNx8uRJmLuZCfR6iejPH/o7tVKTtm3bwnzNmjUwRzNRvHhxuNYbaTOB3nd27NgB12olcKgYTMR7ZkI7n93MhJaj818rb9L2CVQiJYKfv+DgYLiWM+Gedh6VLl3ayrZu3QrXavuEm9c0PDwcrtWuTdD53LBhQ7jW7bUTKnzTzi3t+FBRU35cO2kFVejv5LXTT7TzKDQ01Mo2b94M12rvf6gYTARfy2ozoZ2LqDSucePGcK1WDqftE+g11dZqc4XKH7VCM20f27NnD8zR1+N2JtD7hLZPaKVoK1asgDkqKStRogRcmxv8zjMRERERERGRA948ExERERERETngzTMRERERERGRA948ExERERERETngzTMRERERERGRAx9jjMnvgyAiIiIiIiLyZvzOMxEREREREZED3jwTEREREREROeDNMxEREREREZED3jwTEREREREROeDNMxEREREREZED3jwTEREREREROeDNMxEREREREZED3jwTEREREREROeDNMxEREREREZGD/wf0uJ4KaELZ0AAAAABJRU5ErkJggg==",
      "text/plain": [
       "<Figure size 1000x400 with 10 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import random\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "# 잘못 분류된 샘플 중 10개를 랜덤하게 선택하여 출력\n",
    "print(\"Wrongly classified samples:\")\n",
    "plt.figure(figsize=(10, 4))  # 전체 그림의 크기 설정\n",
    "for i in range(10):\n",
    "    sample = random.choice(wrong_samples)\n",
    "    input_img, true_label, predicted_label = sample\n",
    "    plt.subplot(2, 5, i + 1)  # 2행 5열의 subplot 중 i+1 번째에 그림을 출력\n",
    "    plt.imshow(input_img.squeeze().cpu().numpy(), cmap='gray')\n",
    "    plt.title(f'True: {true_label}, Predicted: {predicted_label}')\n",
    "    plt.axis('off')  # 축 제거\n",
    "plt.tight_layout()  # subplot 간격 자동 조절\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "맞게 분류된 샘플 랜덤 선택 출력"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Correctly classified samples:\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x400 with 10 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# 정확하게 분류된 샘플 중 10개를 랜덤하게 선택하여 출력\n",
    "print(\"Correctly classified samples:\")\n",
    "plt.figure(figsize=(10, 4))  # 전체 그림의 크기 설정\n",
    "for i in range(10):\n",
    "    sample = random.choice(correct_samples)\n",
    "    input_img, true_label, predicted_label = sample\n",
    "    plt.subplot(2, 5, i + 1)  # 2행 5열의 subplot 중 i+1 번째에 그림을 출력\n",
    "    plt.imshow(input_img.squeeze().cpu().numpy(), cmap='gray')\n",
    "    plt.title(f'True: {true_label}, Predicted: {predicted_label}')\n",
    "    plt.axis('off')  # 축 제거\n",
    "plt.tight_layout()  # subplot 간격 자동 조절\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 5. 인용\n",
    "\n",
    "~~~\n",
    "@inproceedings{foret2021sharpnessaware,\n",
    "  title={Sharpness-aware Minimization for Efficiently Improving Generalization},\n",
    "  author={Pierre Foret and Ariel Kleiner and Hossein Mobahi and Behnam Neyshabur},\n",
    "  booktitle={International Conference on Learning Representations},\n",
    "  year={2021},\n",
    "  url={https://openreview.net/forum?id=6Tm1mposlrM}\n",
    "}\n",
    "~~~\n",
    "\n",
    "~~~\n",
    "@inproceesings{pmlr-v139-kwon21b,\n",
    "  title={ASAM: Adaptive Sharpness-Aware Minimization for Scale-Invariant Learning of Deep Neural Networks},\n",
    "  author={Kwon, Jungmin and Kim, Jeongseop and Park, Hyunseo and Choi, In Kwon},\n",
    "  booktitle ={Proceedings of the 38th International Conference on Machine Learning},\n",
    "  pages={5905--5914},\n",
    "  year={2021},\n",
    "  editor={Meila, Marina and Zhang, Tong},\n",
    "  volume={139},\n",
    "  series={Proceedings of Machine Learning Research},\n",
    "  month={18--24 Jul},\n",
    "  publisher ={PMLR},\n",
    "  pdf={http://proceedings.mlr.press/v139/kwon21b/kwon21b.pdf},\n",
    "  url={https://proceedings.mlr.press/v139/kwon21b.html},\n",
    "  abstract={Recently, learning algorithms motivated from sharpness of loss surface as an effective measure of generalization gap have shown state-of-the-art performances. Nevertheless, sharpness defined in a rigid region with a fixed radius, has a drawback in sensitivity to parameter re-scaling which leaves the loss unaffected, leading to weakening of the connection between sharpness and generalization gap. In this paper, we introduce the concept of adaptive sharpness which is scale-invariant and propose the corresponding generalization bound. We suggest a novel learning method, adaptive sharpness-aware minimization (ASAM), utilizing the proposed generalization bound. Experimental results in various benchmark datasets show that ASAM contributes to significant improvement of model generalization performance.}\n",
    "}\n",
    "~~~"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}