Gabriel_Sansigolo_Trabalho_de_DL_2.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "tags": []
   },
   "source": [
    "# **Atividade 2 - Uso de Redes Neurais Profundas**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "tags": []
   },
   "source": [
    "## **Disciplina:** CAP-421-3 Aprendizado Profundo\n",
    "\n",
    "**Discintes:**\n",
    "\n",
    "Gabriel Sansigolo"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "tags": []
   },
   "source": [
    "### **Importando bibliotecas**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "import cv2\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import torch\n",
    "import torch.nn as nn\n",
    "import torch.optim as optim\n",
    "import torch.nn.functional as F\n",
    "import torchvision.transforms as transforms\n",
    "\n",
    "from tqdm import tqdm\n",
    "from torchinfo import summary\n",
    "from torch.utils.data import DataLoader"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### **Utiliza GPU (cuda), caso contrário, utiliza CPU**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "device = 'cuda' if torch.cuda.is_available() else 'cpu'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'NVIDIA GeForce RTX 4080'"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cuda_id = torch.cuda.current_device()\n",
    "torch.cuda.get_device_name(cuda_id)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "tags": []
   },
   "source": [
    "### **1 - Pré-processamento da Imagens**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "path_dir = os.path.dirname(\"\")\n",
    "\n",
    "def load_images(path, pista = False):\n",
    "\n",
    "    images = []\n",
    "    filenames = os.listdir(path)\n",
    "    \n",
    "    for filename in tqdm(filenames):\n",
    "        if filename == '_DS_Store':\n",
    "            continue\n",
    "        image = cv2.imread(os.path.join(path, filename))\n",
    "        image = cv2.resize(image, dsize=(100,100))\n",
    "        image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)\n",
    "\n",
    "        images.append(image)\n",
    "        \n",
    "        if pista == True:\n",
    "            \n",
    "            #apply gama 0.5\n",
    "            invGamma = 1.0 / 0.5\n",
    "            table = np.array([((i / 255.0) ** invGamma) * 255\n",
    "            for i in np.arange(0, 256)]).astype(\"uint8\")\n",
    "            new_image = cv2.LUT(image, table)\n",
    "                        \n",
    "            images.append(new_image)\n",
    "\n",
    "\n",
    "\n",
    "    return np.array(images)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### **2 - Dividindo os Dados (Treinamento e Teste)**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 3228/3228 [00:07<00:00, 456.08it/s]\n",
      "100%|██████████| 16394/16394 [00:30<00:00, 532.12it/s]\n",
      "100%|██████████| 1935/1935 [00:03<00:00, 535.06it/s]\n",
      "100%|██████████| 530/530 [00:01<00:00, 529.03it/s]\n"
     ]
    }
   ],
   "source": [
    "mapeado_train = load_images(os.path.join(path_dir,'Treinamento/Pistas/'), True)\n",
    "nao_mapeado_train = load_images(os.path.join(path_dir,'Treinamento/NaoPistas/'), False)\n",
    "\n",
    "nao_mapeado_train = nao_mapeado_train[:16140]\n",
    "\n",
    "mapeado_test = load_images(os.path.join(path_dir,'Teste/NovasNaoPistas/'), False)\n",
    "nao_mapeado_test = load_images(os.path.join(path_dir,'Teste/NovasPistas/'), False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### **3 - Verificação das Dimensões dos Dados de Treinamento e Teste**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(6456, 100, 100, 3)\n",
      "(16140, 100, 100, 3)\n",
      "(1935, 100, 100, 3)\n",
      "(530, 100, 100, 3)\n"
     ]
    }
   ],
   "source": [
    "print(mapeado_train.shape)\n",
    "print(nao_mapeado_train.shape)\n",
    "print(mapeado_test.shape)\n",
    "print(nao_mapeado_test.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(22596, 100, 100, 3)\n",
      "(2365, 100, 100, 3)\n",
      "(100, 100, 100, 3)\n"
     ]
    }
   ],
   "source": [
    "X_train = np.append(mapeado_train, nao_mapeado_train, axis=0)\n",
    "X_test  = np.append(mapeado_test, nao_mapeado_test, axis=0)\n",
    "\n",
    "X_test = X_test[100:]\n",
    "X_val = X_test[:100]\n",
    "\n",
    "print(X_train.shape)\n",
    "print(X_test.shape)\n",
    "print(X_val.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(22596,)\n",
      "(2365,)\n",
      "(100,)\n"
     ]
    }
   ],
   "source": [
    "y_train = np.array([0] * len(mapeado_train) + [1] * len(nao_mapeado_train))\n",
    "y_test = np.array([0] * len(mapeado_test) + [1] * len(nao_mapeado_test))\n",
    "\n",
    "y_test = y_test[100:]\n",
    "y_val = y_test[:100] \n",
    "\n",
    "print(y_train.shape)\n",
    "print(y_test.shape)\n",
    "print(y_val.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_train = torch.from_numpy(y_train.reshape(len(y_train),1))\n",
    "y_test = torch.from_numpy(y_test.reshape(len(y_test),1))\n",
    "y_val = torch.from_numpy(y_val.reshape(len(y_val),1))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### **4 - Execução de Transformações Necessárias para Dados de Treinamento e Teste**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "transforms_train = transforms.Compose([transforms.ToTensor(), \n",
    "                                       transforms.Normalize(mean=[0.5,0.5,0.5], std=[0.5,0.5,0.5]) # squeeze to -1 and 1\n",
    "                                      ])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "transforms_test = transforms.Compose([transforms.ToTensor(), \n",
    "                                     transforms.Normalize([0.5,0.5,0.5], [0.5,0.5,0.5])])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "transforms_val = transforms.Compose([transforms.ToTensor(), \n",
    "                                     transforms.Normalize([0.5,0.5,0.5], [0.5,0.5,0.5])])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### **5 - Definição das Classes**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "class Mapeado_Nao_Mapeado():\n",
    "    def __init__(self, images, labels, transform=None):\n",
    "        self.images = images\n",
    "        self.labels = labels\n",
    "        self.transform = transform\n",
    "        \n",
    "    def __len__(self):\n",
    "        return len(self.images)\n",
    "    \n",
    "    def __getitem__(self, index):\n",
    "        image = self.images[index]\n",
    "        label = self.labels[index]\n",
    "        \n",
    "        if self.transform:\n",
    "            image = self.transform(image)\n",
    "        \n",
    "        return (image, label)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### **6 - Criação dos DataLoaders para os Dados de Treinamento e Teste**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_dataset = Mapeado_Nao_Mapeado(images=X_train, labels=y_train, transform=transforms_train)\n",
    "test_dataset  = Mapeado_Nao_Mapeado(images=X_test, labels=y_test, transform=transforms_test)\n",
    "val_dataset  = Mapeado_Nao_Mapeado(images=X_val, labels=y_val, transform=transforms_val)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_loader = DataLoader(train_dataset, batch_size=32, shuffle=True, drop_last=True)\n",
    "test_loader  = DataLoader(test_dataset, batch_size=32, shuffle=True, drop_last=True)\n",
    "val_loader  = DataLoader(val_dataset, batch_size=32, shuffle=True, drop_last=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([32, 3, 100, 100])\n"
     ]
    }
   ],
   "source": [
    "iterator = iter(train_loader)\n",
    "image_batch, label_batch = next(iterator)\n",
    "\n",
    "print(image_batch.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([32, 100, 100, 3])\n"
     ]
    }
   ],
   "source": [
    "image_batch_permuted = image_batch.permute(0, 2, 3, 1)\n",
    "\n",
    "print(image_batch_permuted.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### **7 - Definindo a Rede Neural Convolucional (CNN) para Classificação**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "class CNN(nn.Module):\n",
    "    def __init__(self):\n",
    "        super().__init__()\n",
    "        self.conv0 = nn.Conv2d(in_channels=3, out_channels=16, kernel_size=(3,3), stride=(1,1), padding=(1,1), bias=False)\n",
    "        self.bn0 = nn.BatchNorm2d(num_features=16)\n",
    "        self.maxpool = nn.MaxPool2d(kernel_size=(2,2), stride=(2,2))\n",
    "        \n",
    "        self.conv1 = nn.Conv2d(in_channels=16, out_channels=32, kernel_size=(3,3), stride=(1,1), padding=(1,1), bias=False)\n",
    "        self.bn1 = nn.BatchNorm2d(num_features=32)\n",
    "        # self.maxpool\n",
    "        \n",
    "        self.conv2 = nn.Conv2d(in_channels=32, out_channels=64, kernel_size=(3,3), stride=(1,1), padding=(1,1), bias=False)\n",
    "        self.bn2 = nn.BatchNorm2d(num_features=64)\n",
    "        # self.maxpool\n",
    "        \n",
    "        self.conv3 = nn.Conv2d(in_channels=64, out_channels=128, kernel_size=(3,3), stride=(1,1), padding=(1,1), bias=False)\n",
    "        self.bn3 = nn.BatchNorm2d(num_features=128)\n",
    "        # self.maxpool\n",
    "        \n",
    "        self.dropout = nn.Dropout(p=0.5)\n",
    "        self.fc0 = nn.Linear(in_features=128*6*6, out_features=64)\n",
    "        self.fc1 = nn.Linear(in_features=64, out_features=32)\n",
    "        self.fc2 = nn.Linear(in_features=32, out_features=1)\n",
    "        \n",
    "    def forward(self, x):\n",
    "        x = F.relu(self.bn0(self.conv0(x)))\n",
    "        x = self.maxpool(x)\n",
    "        \n",
    "        x = F.relu(self.bn1(self.conv1(x)))\n",
    "        x = self.maxpool(x)\n",
    "        \n",
    "        x = F.relu(self.bn2(self.conv2(x)))\n",
    "        x = self.maxpool(x)\n",
    "        \n",
    "        x = F.relu(self.bn3(self.conv3(x)))\n",
    "        x = self.maxpool(x)\n",
    "\n",
    "        x = x.reshape(x.shape[0], -1)\n",
    "        \n",
    "        x = self.dropout(x)\n",
    "        x = F.relu(self.fc0(x))\n",
    "        x = F.relu(self.fc1(x))\n",
    "        x = F.sigmoid(self.fc2(x))\n",
    "        \n",
    "        return x"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "model = CNN().to(device)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "==========================================================================================\n",
       "Layer (type:depth-idx)                   Output Shape              Param #\n",
       "==========================================================================================\n",
       "CNN                                      [4, 1]                    --\n",
       "├─Conv2d: 1-1                            [4, 16, 100, 100]         432\n",
       "├─BatchNorm2d: 1-2                       [4, 16, 100, 100]         32\n",
       "├─MaxPool2d: 1-3                         [4, 16, 50, 50]           --\n",
       "├─Conv2d: 1-4                            [4, 32, 50, 50]           4,608\n",
       "├─BatchNorm2d: 1-5                       [4, 32, 50, 50]           64\n",
       "├─MaxPool2d: 1-6                         [4, 32, 25, 25]           --\n",
       "├─Conv2d: 1-7                            [4, 64, 25, 25]           18,432\n",
       "├─BatchNorm2d: 1-8                       [4, 64, 25, 25]           128\n",
       "├─MaxPool2d: 1-9                         [4, 64, 12, 12]           --\n",
       "├─Conv2d: 1-10                           [4, 128, 12, 12]          73,728\n",
       "├─BatchNorm2d: 1-11                      [4, 128, 12, 12]          256\n",
       "├─MaxPool2d: 1-12                        [4, 128, 6, 6]            --\n",
       "├─Dropout: 1-13                          [4, 4608]                 --\n",
       "├─Linear: 1-14                           [4, 64]                   294,976\n",
       "├─Linear: 1-15                           [4, 32]                   2,080\n",
       "├─Linear: 1-16                           [4, 1]                    33\n",
       "==========================================================================================\n",
       "Total params: 394,769\n",
       "Trainable params: 394,769\n",
       "Non-trainable params: 0\n",
       "Total mult-adds (Units.MEGABYTES): 153.10\n",
       "==========================================================================================\n",
       "Input size (MB): 0.48\n",
       "Forward/backward pass size (MB): 19.10\n",
       "Params size (MB): 1.58\n",
       "Estimated Total Size (MB): 21.16\n",
       "=========================================================================================="
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "summary(model, input_size=(4,3,100,100))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### **8 - Definição da Função de Perda e do Otimizador**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "loss_function = nn.BCELoss()\n",
    "optimizer = optim.Adam(model.parameters(), lr=0.001)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### **9 - Definição da Função de Predição de Dados de Teste**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "def predict_test_data(model, test_loader):\n",
    "    \n",
    "    num_correct = 0\n",
    "    num_samples = 0\n",
    "    \n",
    "    model.eval()\n",
    "    \n",
    "    with torch.no_grad():\n",
    "        for batch, (X_test, y_test) in enumerate(test_loader):\n",
    "            X_test = X_test.float().to(device)\n",
    "            y_test = y_test.float().to(device)\n",
    "\n",
    "            test_preds = model(X_test)\n",
    "            test_loss  = loss_function(test_preds, y_test)\n",
    "            \n",
    "            rounded_test_preds = torch.round(test_preds)\n",
    "            num_correct += torch.sum(rounded_test_preds == y_test)\n",
    "            num_samples += len(y_test)\n",
    "        \n",
    "    model.train()\n",
    "    \n",
    "    test_acc = num_correct/num_samples\n",
    "    \n",
    "    return test_loss, test_acc"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### **10 - Treinamento do Modelo CNN e Avaliação do Desempenho**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:05<00:00, 130.30it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 0 \t| Train loss: 0.605 \t| Test loss: 2.22 \t| Train acc: 0.85 \t| Test acc: 0.22\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:05<00:00, 135.12it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 1 \t| Train loss: 0.359 \t| Test loss: 1.859 \t| Train acc: 0.85 \t| Test acc: 0.22\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:05<00:00, 135.83it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 2 \t| Train loss: 0.235 \t| Test loss: 1.918 \t| Train acc: 0.86 \t| Test acc: 0.25\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:05<00:00, 135.94it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 3 \t| Train loss: 0.359 \t| Test loss: 2.055 \t| Train acc: 0.86 \t| Test acc: 0.22\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:05<00:00, 135.93it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 4 \t| Train loss: 0.486 \t| Test loss: 2.21 \t| Train acc: 0.86 \t| Test acc: 0.21\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:05<00:00, 137.40it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 5 \t| Train loss: 0.265 \t| Test loss: 2.831 \t| Train acc: 0.86 \t| Test acc: 0.2\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:05<00:00, 133.10it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 6 \t| Train loss: 0.34 \t| Test loss: 1.908 \t| Train acc: 0.86 \t| Test acc: 0.22\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:05<00:00, 137.18it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 7 \t| Train loss: 0.46 \t| Test loss: 2.148 \t| Train acc: 0.87 \t| Test acc: 0.18\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:05<00:00, 136.46it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 8 \t| Train loss: 0.482 \t| Test loss: 2.009 \t| Train acc: 0.87 \t| Test acc: 0.18\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:05<00:00, 136.70it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 9 \t| Train loss: 0.322 \t| Test loss: 2.299 \t| Train acc: 0.87 \t| Test acc: 0.19\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:05<00:00, 137.13it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 10 \t| Train loss: 0.375 \t| Test loss: 2.986 \t| Train acc: 0.87 \t| Test acc: 0.21\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:05<00:00, 140.23it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 11 \t| Train loss: 0.21 \t| Test loss: 2.371 \t| Train acc: 0.87 \t| Test acc: 0.21\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:05<00:00, 137.45it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 12 \t| Train loss: 0.355 \t| Test loss: 1.981 \t| Train acc: 0.88 \t| Test acc: 0.21\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:05<00:00, 137.87it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 13 \t| Train loss: 0.256 \t| Test loss: 2.247 \t| Train acc: 0.88 \t| Test acc: 0.26\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:05<00:00, 138.58it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 14 \t| Train loss: 0.305 \t| Test loss: 2.259 \t| Train acc: 0.88 \t| Test acc: 0.21\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:05<00:00, 136.28it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 15 \t| Train loss: 0.194 \t| Test loss: 2.58 \t| Train acc: 0.88 \t| Test acc: 0.25\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:05<00:00, 138.28it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 16 \t| Train loss: 0.218 \t| Test loss: 2.462 \t| Train acc: 0.88 \t| Test acc: 0.21\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:05<00:00, 135.53it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 17 \t| Train loss: 0.291 \t| Test loss: 3.006 \t| Train acc: 0.88 \t| Test acc: 0.21\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:05<00:00, 138.06it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 18 \t| Train loss: 0.262 \t| Test loss: 2.399 \t| Train acc: 0.89 \t| Test acc: 0.21\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:05<00:00, 135.37it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 19 \t| Train loss: 0.303 \t| Test loss: 3.129 \t| Train acc: 0.89 \t| Test acc: 0.19\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:04<00:00, 141.71it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 20 \t| Train loss: 0.277 \t| Test loss: 2.783 \t| Train acc: 0.89 \t| Test acc: 0.21\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:05<00:00, 139.53it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 21 \t| Train loss: 0.121 \t| Test loss: 3.155 \t| Train acc: 0.89 \t| Test acc: 0.21\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:05<00:00, 138.67it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 22 \t| Train loss: 0.238 \t| Test loss: 2.86 \t| Train acc: 0.89 \t| Test acc: 0.21\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:05<00:00, 139.83it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 23 \t| Train loss: 0.287 \t| Test loss: 3.333 \t| Train acc: 0.9 \t| Test acc: 0.21\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:04<00:00, 144.41it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 24 \t| Train loss: 0.159 \t| Test loss: 3.487 \t| Train acc: 0.9 \t| Test acc: 0.22\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:04<00:00, 142.64it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 25 \t| Train loss: 0.303 \t| Test loss: 3.12 \t| Train acc: 0.91 \t| Test acc: 0.21\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:05<00:00, 140.47it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 26 \t| Train loss: 0.114 \t| Test loss: 3.382 \t| Train acc: 0.91 \t| Test acc: 0.21\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:04<00:00, 144.90it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 27 \t| Train loss: 0.132 \t| Test loss: 2.533 \t| Train acc: 0.91 \t| Test acc: 0.2\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:04<00:00, 141.72it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 28 \t| Train loss: 0.109 \t| Test loss: 3.749 \t| Train acc: 0.91 \t| Test acc: 0.21\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:05<00:00, 140.92it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 29 \t| Train loss: 0.278 \t| Test loss: 4.387 \t| Train acc: 0.92 \t| Test acc: 0.2\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:05<00:00, 136.92it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 30 \t| Train loss: 0.117 \t| Test loss: 3.476 \t| Train acc: 0.92 \t| Test acc: 0.2\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:05<00:00, 138.22it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 31 \t| Train loss: 0.123 \t| Test loss: 2.486 \t| Train acc: 0.92 \t| Test acc: 0.21\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:05<00:00, 136.06it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 32 \t| Train loss: 0.206 \t| Test loss: 3.94 \t| Train acc: 0.92 \t| Test acc: 0.2\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:05<00:00, 138.06it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 33 \t| Train loss: 0.214 \t| Test loss: 3.699 \t| Train acc: 0.93 \t| Test acc: 0.2\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:05<00:00, 136.89it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 34 \t| Train loss: 0.251 \t| Test loss: 8.208 \t| Train acc: 0.93 \t| Test acc: 0.2\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:05<00:00, 136.19it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 35 \t| Train loss: 0.127 \t| Test loss: 4.715 \t| Train acc: 0.93 \t| Test acc: 0.21\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:05<00:00, 140.55it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 36 \t| Train loss: 0.145 \t| Test loss: 3.971 \t| Train acc: 0.93 \t| Test acc: 0.2\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:04<00:00, 141.59it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 37 \t| Train loss: 0.124 \t| Test loss: 5.376 \t| Train acc: 0.94 \t| Test acc: 0.2\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:04<00:00, 142.94it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 38 \t| Train loss: 0.224 \t| Test loss: 5.223 \t| Train acc: 0.94 \t| Test acc: 0.2\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:05<00:00, 137.66it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 39 \t| Train loss: 0.191 \t| Test loss: 4.913 \t| Train acc: 0.94 \t| Test acc: 0.19\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:04<00:00, 142.46it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 40 \t| Train loss: 0.065 \t| Test loss: 4.118 \t| Train acc: 0.94 \t| Test acc: 0.2\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:05<00:00, 134.85it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 41 \t| Train loss: 0.205 \t| Test loss: 5.85 \t| Train acc: 0.94 \t| Test acc: 0.2\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:06<00:00, 109.80it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 42 \t| Train loss: 0.281 \t| Test loss: 6.072 \t| Train acc: 0.95 \t| Test acc: 0.2\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:11<00:00, 60.66it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 43 \t| Train loss: 0.189 \t| Test loss: 4.187 \t| Train acc: 0.95 \t| Test acc: 0.21\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:11<00:00, 61.16it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 44 \t| Train loss: 0.013 \t| Test loss: 6.402 \t| Train acc: 0.95 \t| Test acc: 0.2\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:11<00:00, 61.01it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 45 \t| Train loss: 0.354 \t| Test loss: 6.1 \t| Train acc: 0.96 \t| Test acc: 0.2\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:07<00:00, 92.72it/s] \n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 46 \t| Train loss: 0.031 \t| Test loss: 9.052 \t| Train acc: 0.96 \t| Test acc: 0.19\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:05<00:00, 123.40it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 47 \t| Train loss: 0.065 \t| Test loss: 5.107 \t| Train acc: 0.96 \t| Test acc: 0.2\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:05<00:00, 130.43it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 48 \t| Train loss: 0.077 \t| Test loss: 11.651 \t| Train acc: 0.96 \t| Test acc: 0.21\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:05<00:00, 134.42it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 49 \t| Train loss: 0.206 \t| Test loss: 5.338 \t| Train acc: 0.96 \t| Test acc: 0.2\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:05<00:00, 136.12it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 50 \t| Train loss: 0.033 \t| Test loss: 8.613 \t| Train acc: 0.96 \t| Test acc: 0.2\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:05<00:00, 133.45it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 51 \t| Train loss: 0.139 \t| Test loss: 9.03 \t| Train acc: 0.96 \t| Test acc: 0.21\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:05<00:00, 138.67it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 52 \t| Train loss: 0.071 \t| Test loss: 5.33 \t| Train acc: 0.97 \t| Test acc: 0.2\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:05<00:00, 137.28it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 53 \t| Train loss: 0.035 \t| Test loss: 10.377 \t| Train acc: 0.97 \t| Test acc: 0.2\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:05<00:00, 139.06it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 54 \t| Train loss: 0.054 \t| Test loss: 5.443 \t| Train acc: 0.97 \t| Test acc: 0.21\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:05<00:00, 136.72it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 55 \t| Train loss: 0.093 \t| Test loss: 5.543 \t| Train acc: 0.97 \t| Test acc: 0.22\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:05<00:00, 129.76it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 56 \t| Train loss: 0.108 \t| Test loss: 6.856 \t| Train acc: 0.97 \t| Test acc: 0.2\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:05<00:00, 126.66it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 57 \t| Train loss: 0.148 \t| Test loss: 8.52 \t| Train acc: 0.97 \t| Test acc: 0.2\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:05<00:00, 129.51it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 58 \t| Train loss: 0.043 \t| Test loss: 13.962 \t| Train acc: 0.97 \t| Test acc: 0.19\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:05<00:00, 122.90it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 59 \t| Train loss: 0.023 \t| Test loss: 8.573 \t| Train acc: 0.97 \t| Test acc: 0.21\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:05<00:00, 132.80it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 60 \t| Train loss: 0.027 \t| Test loss: 12.199 \t| Train acc: 0.97 \t| Test acc: 0.21\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:05<00:00, 139.64it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 61 \t| Train loss: 0.067 \t| Test loss: 13.52 \t| Train acc: 0.97 \t| Test acc: 0.19\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:05<00:00, 136.57it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 62 \t| Train loss: 0.024 \t| Test loss: 11.297 \t| Train acc: 0.97 \t| Test acc: 0.2\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:05<00:00, 137.99it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 63 \t| Train loss: 0.051 \t| Test loss: 11.443 \t| Train acc: 0.97 \t| Test acc: 0.19\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:05<00:00, 135.77it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 64 \t| Train loss: 0.018 \t| Test loss: 8.427 \t| Train acc: 0.98 \t| Test acc: 0.19\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:05<00:00, 140.44it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 65 \t| Train loss: 0.073 \t| Test loss: 12.677 \t| Train acc: 0.98 \t| Test acc: 0.2\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:05<00:00, 137.30it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 66 \t| Train loss: 0.07 \t| Test loss: 13.214 \t| Train acc: 0.98 \t| Test acc: 0.19\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:05<00:00, 130.18it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 67 \t| Train loss: 0.006 \t| Test loss: 25.297 \t| Train acc: 0.98 \t| Test acc: 0.2\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:05<00:00, 135.29it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 68 \t| Train loss: 0.067 \t| Test loss: 13.874 \t| Train acc: 0.98 \t| Test acc: 0.19\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:05<00:00, 122.75it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 69 \t| Train loss: 0.04 \t| Test loss: 12.945 \t| Train acc: 0.98 \t| Test acc: 0.19\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:05<00:00, 132.70it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 70 \t| Train loss: 0.054 \t| Test loss: 7.181 \t| Train acc: 0.98 \t| Test acc: 0.2\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:05<00:00, 132.79it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 71 \t| Train loss: 0.13 \t| Test loss: 7.766 \t| Train acc: 0.98 \t| Test acc: 0.22\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:05<00:00, 127.92it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 72 \t| Train loss: 0.186 \t| Test loss: 12.498 \t| Train acc: 0.98 \t| Test acc: 0.2\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:05<00:00, 133.52it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 73 \t| Train loss: 0.027 \t| Test loss: 7.992 \t| Train acc: 0.98 \t| Test acc: 0.2\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:05<00:00, 131.75it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 74 \t| Train loss: 0.005 \t| Test loss: 15.769 \t| Train acc: 0.98 \t| Test acc: 0.19\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:05<00:00, 128.55it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 75 \t| Train loss: 0.094 \t| Test loss: 12.649 \t| Train acc: 0.98 \t| Test acc: 0.18\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:05<00:00, 121.93it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 76 \t| Train loss: 0.083 \t| Test loss: 9.69 \t| Train acc: 0.98 \t| Test acc: 0.2\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:10<00:00, 67.05it/s] \n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 77 \t| Train loss: 0.033 \t| Test loss: 5.808 \t| Train acc: 0.98 \t| Test acc: 0.2\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:11<00:00, 60.51it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 78 \t| Train loss: 0.019 \t| Test loss: 10.048 \t| Train acc: 0.98 \t| Test acc: 0.22\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:06<00:00, 112.68it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 79 \t| Train loss: 0.079 \t| Test loss: 15.474 \t| Train acc: 0.98 \t| Test acc: 0.19\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:05<00:00, 135.92it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 80 \t| Train loss: 0.133 \t| Test loss: 19.062 \t| Train acc: 0.98 \t| Test acc: 0.2\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:05<00:00, 126.22it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 81 \t| Train loss: 0.09 \t| Test loss: 5.277 \t| Train acc: 0.98 \t| Test acc: 0.22\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:11<00:00, 63.36it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 82 \t| Train loss: 0.028 \t| Test loss: 4.403 \t| Train acc: 0.98 \t| Test acc: 0.2\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:11<00:00, 61.23it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 83 \t| Train loss: 0.012 \t| Test loss: 5.895 \t| Train acc: 0.98 \t| Test acc: 0.2\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:11<00:00, 60.24it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 84 \t| Train loss: 0.009 \t| Test loss: 20.094 \t| Train acc: 0.98 \t| Test acc: 0.19\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:11<00:00, 59.85it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 85 \t| Train loss: 0.049 \t| Test loss: 10.891 \t| Train acc: 0.98 \t| Test acc: 0.2\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:11<00:00, 59.98it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 86 \t| Train loss: 0.002 \t| Test loss: 18.51 \t| Train acc: 0.98 \t| Test acc: 0.18\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:11<00:00, 62.07it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 87 \t| Train loss: 0.019 \t| Test loss: 9.245 \t| Train acc: 0.98 \t| Test acc: 0.2\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:11<00:00, 60.42it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 88 \t| Train loss: 0.022 \t| Test loss: 18.334 \t| Train acc: 0.98 \t| Test acc: 0.19\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:11<00:00, 60.58it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 89 \t| Train loss: 0.047 \t| Test loss: 12.551 \t| Train acc: 0.98 \t| Test acc: 0.2\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:11<00:00, 60.69it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 90 \t| Train loss: 0.013 \t| Test loss: 5.499 \t| Train acc: 0.98 \t| Test acc: 0.2\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:11<00:00, 61.73it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 91 \t| Train loss: 0.006 \t| Test loss: 17.048 \t| Train acc: 0.98 \t| Test acc: 0.18\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:11<00:00, 60.43it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 92 \t| Train loss: 0.012 \t| Test loss: 18.638 \t| Train acc: 0.98 \t| Test acc: 0.2\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:11<00:00, 60.61it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 93 \t| Train loss: 0.038 \t| Test loss: 16.815 \t| Train acc: 0.99 \t| Test acc: 0.19\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:11<00:00, 61.22it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 94 \t| Train loss: 0.189 \t| Test loss: 10.312 \t| Train acc: 0.98 \t| Test acc: 0.2\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:11<00:00, 59.55it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 95 \t| Train loss: 0.01 \t| Test loss: 6.886 \t| Train acc: 0.99 \t| Test acc: 0.2\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:11<00:00, 60.39it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 96 \t| Train loss: 0.014 \t| Test loss: 25.587 \t| Train acc: 0.99 \t| Test acc: 0.19\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:11<00:00, 61.43it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 97 \t| Train loss: 0.076 \t| Test loss: 14.665 \t| Train acc: 0.99 \t| Test acc: 0.2\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:11<00:00, 60.36it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 98 \t| Train loss: 0.178 \t| Test loss: 24.483 \t| Train acc: 0.99 \t| Test acc: 0.19\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 706/706 [00:11<00:00, 59.45it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 99 \t| Train loss: 0.06 \t| Test loss: 6.555 \t| Train acc: 0.99 \t| Test acc: 0.2\n"
     ]
    }
   ],
   "source": [
    "train_losses = []    \n",
    "train_accs = []\n",
    "test_losses = []\n",
    "test_accs  = []\n",
    "\n",
    "for epoch in range(100):\n",
    "    \n",
    "    num_correct_train = 0\n",
    "    num_samples_train = 0\n",
    "    for batch, (X_train, y_train) in tqdm(enumerate(train_loader), total=len(train_loader)):\n",
    "        X_train = X_train.float().to(device)\n",
    "        y_train = y_train.float().to(device)\n",
    "        \n",
    "        train_preds = model(X_train)\n",
    "        train_loss = loss_function(train_preds, y_train)\n",
    "        \n",
    "        with torch.no_grad():\n",
    "            rounded_train_preds = torch.round(train_preds)\n",
    "            num_correct_train += torch.sum(rounded_train_preds == y_train)\n",
    "            num_samples_train += len(y_train)\n",
    "            \n",
    "        optimizer.zero_grad()\n",
    "        train_loss.backward()\n",
    "        \n",
    "        optimizer.step()\n",
    "    \n",
    "    train_acc = num_correct_train/num_samples_train\n",
    "    test_loss, test_acc = predict_test_data(model, test_loader)\n",
    "    \n",
    "    train_losses.append(train_loss.item())\n",
    "    test_losses.append(test_loss.item())\n",
    "    train_accs.append(train_acc.item())\n",
    "    test_accs.append(test_acc.item())\n",
    "        \n",
    "    print(f'Epoch: {epoch} \\t|' \\\n",
    "            f' Train loss: {np.round(train_loss.item(),3)} \\t|' \\\n",
    "            f' Test loss: {np.round(test_loss.item(),3)} \\t|' \\\n",
    "            f' Train acc: {np.round(train_acc.item(),2)} \\t|' \\\n",
    "            f' Test acc: {np.round(test_acc.item(),2)}')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### **11 - Visualização das Curvas de Treinamento e Teste**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(range(100), train_losses , label='train_losses')\n",
    "plt.plot(range(100), train_accs, label='train_accs')\n",
    "plt.title('Curva de Treinamento')\n",
    "plt.xlabel('Época')\n",
    "plt.ylabel('Loss')   \n",
    "plt.legend(bbox_to_anchor=(1.1, 1.05))\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(range(100), test_losses , label='test_losses')\n",
    "plt.plot(range(100), test_accs, label='test_accs')\n",
    "plt.title('Curva de Teste')\n",
    "plt.xlabel('Época')\n",
    "plt.ylabel('Loss')  \n",
    "plt.legend(bbox_to_anchor=(1.1, 1.05))\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAugAAAHJCAYAAADXQsyeAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjkuMCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy80BEi2AAAACXBIWXMAAA9hAAAPYQGoP6dpAABuO0lEQVR4nO3deVxUVeMG8GdmYIZ9ANkRAZfcBRUlXEqTRFPTtFLzTbTU1zRfi1+ZZrlVki2mpmmZZpnlUrZqbmiaRuKSu+IuKLsIwzoDM/f3x5FBBBQQmAGe7+czH4Y759575rLMc88951yZJEkSiIiIiIjILMhNXQEiIiIiIirGgE5EREREZEYY0ImIiIiIzAgDOhERERGRGWFAJyIiIiIyIwzoRERERERmhAGdiIiIiMiMMKATEREREZkRBnQiIiIiIjPCgE5E9dKaNWsgk8lw9epVU1eFiIioUhjQiRqYS5cu4b///S+aNm0KKysrODg4oHv37li8eDHy8vJMXb06YevWrZDJZPDy8oLBYDB1dYiIqJ6xMHUFiKj2bNmyBc888wxUKhVGjx6Ndu3aQafTYf/+/Xj99ddx+vRpfPHFF6auptlbt24d/Pz8cPXqVezevRuhoaGmrhIREdUjDOhEDcSVK1cwYsQI+Pr6Yvfu3fD09DS+NnnyZFy8eBFbtmypln3l5OTA1ta2WrZlbnJycvDLL78gMjISX331FdatW8eATkRE1YpdXIgaiA8++ADZ2dlYtWpViXBepHnz5pg6dSoA4OrVq5DJZFizZk2pcjKZDHPmzDF+P2fOHMhkMpw5cwbPPfccnJyc0KNHD3z00UeQyWS4du1aqW3MmDEDSqUSt27dAgD89ddfeOaZZ9CkSROoVCr4+Pjg1VdfrXCXm9OnT+Oxxx6DtbU1GjdujHfffbfcrid//PEHevbsCVtbW9jb22PAgAE4ffp0hfYDAD/99BPy8vLwzDPPYMSIEdi8eTPy8/NLlcvPz8ecOXPw0EMPwcrKCp6enhg6dCguXbpkLGMwGLB48WK0b98eVlZWcHV1Rb9+/XD48GFjmZ07d6JHjx5wdHSEnZ0dWrZsiTfffLPC9SUiorqHLehEDcRvv/2Gpk2bolu3bjWy/WeeeQYtWrTA/PnzIUkSBg4ciGnTpmHjxo14/fXXS5TduHEj+vbtCycnJwDApk2bkJubi5deegmNGjVCTEwMPv30U1y/fh2bNm26536TkpLQu3dvFBYWYvr06bC1tcUXX3wBa2vrUmXXrl2L8PBwhIWFYcGCBcjNzcXy5cvRo0cP/Pvvv/Dz87vv+1y3bh169+4NDw8PjBgxAtOnT8dvv/2GZ555xlhGr9dj4MCBiIqKwogRIzB16lRkZWVh586dOHXqFJo1awYAePHFF7FmzRr0798f48aNQ2FhIf766y/8888/CAoKwunTpzFw4EB06NAB8+bNg0qlwsWLF3HgwIH71pOIiOowiYjqvczMTAmANHjw4AqVv3LligRA+uqrr0q9BkCaPXu28fvZs2dLAKSRI0eWKhsSEiJ17ty5xLKYmBgJgPTNN98Yl+Xm5pZaNzIyUpLJZNK1a9fuWddXXnlFAiAdPHjQuCwlJUVSq9USAOnKlSuSJElSVlaW5OjoKI0fP77E+klJSZJarS61vCzJycmShYWFtHLlSuOybt26lTquq1evlgBICxcuLLUNg8EgSZIk7d69WwIg/e9//yu3zCeffCIBkFJTU+9bNyIiqj/YxYWoAdBoNAAAe3v7GtvHxIkTSy0bPnw4jhw5UqJbx4YNG6BSqTB48GDjsjtbu3NycpCWloZu3bpBkiT8+++/99zv1q1b8fDDD6Nr167GZa6urhg1alSJcjt37kRGRgZGjhyJtLQ040OhUCA4OBh79uy573tcv3495HI5hg0bZlw2cuRI/PHHH8buOgDw448/wsXFBVOmTCm1DZlMZiwjk8kwe/bscss4OjoCAH755RfOFkNE1IAwoBM1AA4ODgCArKysGtuHv79/qWXPPPMM5HI5NmzYAACQJAmbNm1C//79jXUCgLi4OIwZMwbOzs6ws7ODq6srHn30UQBAZmbmPfd77do1tGjRotTyli1blvj+woULAIDHHnsMrq6uJR47duxASkrKfd/jt99+i65du+LmzZu4ePEiLl68iI4dO0Kn05XoinPp0iW0bNkSFhbl9yK8dOkSvLy84OzsXG6Z4cOHo3v37hg3bhzc3d0xYsQIbNy4kWGdiKieYx90ogbAwcEBXl5eOHXqVIXKF7Xg3k2v15e7Tll9vr28vNCzZ09s3LgRb775Jv755x/ExcVhwYIFJbb5+OOPIz09HW+88QZatWoFW1tb3LhxA2PGjKm2MFq0nbVr18LDw6PU6/cK04AI+IcOHQKAMk8I1q1bhwkTJlRDTYtZW1tj37592LNnD7Zs2YJt27Zhw4YNeOyxx7Bjxw4oFIpq3R8REZkHBnSiBmLgwIH44osvEB0djZCQkHuWLRq8mZGRUWJ5WTOy3M/w4cMxadIkxMbGYsOGDbCxscGgQYOMr588eRLnz5/H119/jdGjRxuX79y5s0Lb9/X1NbaO3yk2NrbE90UDM93c3Ko0LeK6detgaWmJtWvXlgrG+/fvx5IlSxAXF4cmTZqgWbNmOHjwIAoKCmBpaVnm9po1a4bt27cjPT39nq3ocrkcffr0QZ8+fbBw4ULMnz8fM2fOxJ49ezi9I1E9JUkSCgsL79koQnWPpaVlhRtWGNCJGohp06Zh3bp1GDduHHbv3g13d/cSr1+6dAm///47pk6dCgcHB7i4uGDfvn145ZVXjGU+++yzSu932LBhmDJlCr7//nts2rQJAwcOLDFHetE/K0mSjMskScLixYsrtP0nnngCixYtQkxMjLEfempqKtatW1eiXFhYGBwcHDB//nz07t27VHBOTU2Fq6truftZt24devbsieHDh5d6LSQkBEuWLMH333+PN954A8OGDcOWLVuwdOlSvPrqqyXKSpIEmUyGYcOGYdmyZZg7d26p91pUpqzwHhgYCADQarX3PjBEVCfpdDokJiYiNzfX1FWhaiaTydC4cWPY2dndtywDOlED0axZM3z33XcYPnw4WrduXeJOon///Tc2bdqEMWPGGMuPGzcO77//PsaNG4egoCDs27cP58+fr/R+3dzc0Lt3byxcuBBZWVmlAm6rVq3QrFkzvPbaa7hx4wYcHBzw448/lhh0eS/Tpk3D2rVr0a9fP0ydOtU4zaKvry9OnDhhLOfg4IDly5fj+eefR6dOnTBixAi4uroiLi4OW7ZsQffu3bF06dIy93Hw4EFcvHgRL7/8cpmve3t7o1OnTli3bh3eeOMNjB49Gt988w0iIiIQExODnj17IicnB7t27cKkSZMwePBg9O7dG88//zyWLFmCCxcuoF+/fjAYDPjrr7/Qu3dvvPzyy5g3bx727duHAQMGwNfXFykpKfjss8/QuHFj9OjRo4I/ASKqKwwGA65cuQKFQgEvLy8olcpyuxxS3SJJElJTU3H9+nW0aNHi/i3ppptAhohM4fz589L48eMlPz8/SalUSvb29lL37t2lTz/9VMrPzzeWy83NlV588UVJrVZL9vb20rPPPiulpKSUO83ivaYCXLlypQRAsre3l/Ly8kq9fubMGSk0NFSys7OTXFxcpPHjx0vHjx8vd6rHu504cUJ69NFHJSsrK8nb21t65513pFWrVpWYZrHInj17pLCwMEmtVktWVlZSs2bNpDFjxkiHDx8ud/tTpkyRAEiXLl0qt8ycOXMkANLx48clSRLHb+bMmZK/v79kaWkpeXh4SE8//XSJbRQWFkoffvih1KpVK0mpVEqurq5S//79pSNHjkiSJElRUVHS4MGDJS8vL0mpVEpeXl7SyJEjpfPnz9/3mBBR3ZOXlyedOXNGysnJMXVVqAbk5uZKZ86cKfNz8G4ySbrjujIRERERmUR+fj6uXLkCf39/WFlZmbo6VM0q8/PlNItERERERGaEAZ2IiIiIzIafnx8WLVpk6mqYFAeJEhEREdED6dWrFwIDA6slWB86dKjEbF8NEQM6EREREdUoSZKg1+vve1M4APec8rahYBcXIiIiIqqyMWPGYO/evVi8eDFkMhlkMhnWrFkDmUyGP/74A507d4ZKpcL+/ftx6dIlDB48GO7u7rCzs0OXLl2wa9euEtu7u4uLTCbDl19+iaeeego2NjZo0aIFfv311wrVTa/X48UXX4S/vz+sra3RsmXLMu+zsXr1arRt2xYqlQqenp4lptXNyMjAf//7X7i7u8PKygrt2rXD77//DkDcwG/QoEFwcnKCra0t2rZti61bt1bhKJbEFnQiIiIiMyRJEvIKTHM3UWtLRYXnYF+8eDHOnz+Pdu3aYd68eQCA06dPAwCmT5+Ojz76CE2bNoWTkxPi4+PxxBNP4L333oNKpcI333yDQYMGITY2Fk2aNCl3H3PnzsUHH3yADz/8EJ9++ilGjRqFa9eu3fNOzICYW75x48bYtGkTGjVqhL///hsTJkyAp6cnnn32WQDA8uXLERERgffffx/9+/dHZmYmDhw4YFy/f//+yMrKwrfffotmzZrhzJkzxnnMJ0+eDJ1Oh3379sHW1hZnzpyp0I2I7qfBBXSDwYCEhATY29tz8n8iIqI6QpIkZGVlwcvLC3J5w+gAkFegR5tZ202y7zPzwmCjrFhMVKvVUCqVsLGxgYeHBwDg3LlzAIB58+bh8ccfN5Z1dnZGQECA8ft33nkHP/30E3799ddybwYHiFb6kSNHAgDmz5+PJUuWICYmBv369btn3SwtLTF37lzj9/7+/oiOjsbGjRuNAf3dd9/F//3f/2Hq1KnGcl26dAEA7Nq1CzExMTh79iweeughAEDTpk2N5eLi4jBs2DC0b9++1GsPosEF9ISEBPj4+Ji6GkRERFQF8fHxaNy4samrQRUUFBRU4vvs7GzMmTMHW7ZsQWJiIgoLC5GXl4e4uLh7bqdDhw7G57a2tnBwcEBKSkqF6rBs2TKsXr0acXFxyMvLg06nQ2BgIAAgJSUFCQkJ6NOnT5nrHjt2DI0bNzaG87v973//w0svvYQdO3YgNDQUw4YNK1HXqmpwAd3e3h6A+AN3cHAwcW2IiIioIjQaDXx8fIyf4w2BtaUCZ+aFmWzf1eHu2Vhee+017Ny5Ex999BGaN28Oa2trPP3009DpdPfcjqWlZYnvZTIZDAbDffe/fv16vPbaa/j4448REhICe3t7fPjhhzh48CAAwNra+p7r3+/1cePGISwsDFu2bMGOHTsQGRmJjz/+GFOmTLlv3e6lwQX0om4tDg4ODOhERER1TEPqniqTySrczcTUlEol9Pr795c/cOAAxowZg6eeegqAaFG/evVqjdXrwIED6NatGyZNmmRcdunSJeNze3t7+Pn5ISoqCr179y61focOHXD9+nWcP3++3FZ0Hx8fTJw4ERMnTsSMGTOwcuVKBnQiIiIiMi0/Pz8cPHgQV69ehZ2dXbmt2y1atMDmzZsxaNAgyGQyvP322xVqCa+qFi1a4JtvvsH27dvh7++PtWvX4tChQ/D39zeWmTNnDiZOnAg3NzfjgNADBw5gypQpePTRR/HII49g2LBhWLhwIZo3b45z585BJpOhX79+eOWVV9C/f3889NBDuHXrFvbs2YPWrVs/cL0bxigLIiIiIqoxr732GhQKBdq0aQNXV9dy+5QvXLgQTk5O6NatGwYNGoSwsDB06tSpxur13//+F0OHDsXw4cMRHByMmzdvlmhNB4Dw8HAsWrQIn332Gdq2bYuBAwfiwoULxtd//PFHdOnSBSNHjkSbNm0wbdo049UCvV6PyZMno3Xr1ujXrx8eeughfPbZZw9cb5kkSdIDb6UO0Wg0UKvVyMzMZBcXIiKiOqIhfH7n5+fjypUr8Pf3h5WVlamrQ9WsMj9ftqATEREREZkRkwb0ffv2YdCgQfDy8oJMJsPPP/9833X+/PNPdOrUCSqVCs2bN8eaNWtqvJ5EREREZH4mTpwIOzu7Mh8TJ040dfWqzKSDRHNychAQEIAXXngBQ4cOvW/5K1euYMCAAZg4cSLWrVuHqKgojBs3Dp6enggLM800RERERERkGvPmzcNrr71W5mt1uSuUSQN6//790b9//wqXX7FiBfz9/fHxxx8DAFq3bo39+/fjk08+YUAnIiIiamDc3Nzg5uZm6mpUuzrVBz06OhqhoaElloWFhSE6OtpENSIiIiIiql51ah70pKQkuLu7l1jm7u4OjUaDvLy8Mu/2pNVqodVqjd9rNJoarycRERERUVXVqRb0qoiMjIRarTY+fHx8TF0lIiIiIqJy1amA7uHhgeTk5BLLkpOT4eDgUGbrOQDMmDEDmZmZxkd8fHxtVJWIiIiIqErqVBeXkJAQbN26tcSynTt3IiQkpNx1VCoVVCpVTVeNiIiIiKhamLQFPTs7G8eOHcOxY8cAiGkUjx07Zrw97IwZMzB69Ghj+YkTJ+Ly5cuYNm0azp07h88++wwbN27Eq6++aorqExER1QuSJOFWjg4GQ4O6uTiR2TJpC/rhw4fRu3dv4/cREREAgPDwcKxZswaJiYnGsA4A/v7+2LJlC1599VUsXrwYjRs3xpdffskpFomIqN7RFRqQlV8AG6UFrCzlkMlkFV43V1eIq2m5uHozB1fScnArR4eWHvbo5OuEpi62xm1dSM7CbycSseVEAi6l5sBWqUBLD3u09nRAGy8HNLJV4lZuAW7l6pCRW4DM3AJYKxVQW1tCbW0JRxtLAMCNW3m4kXH7cSsPrT0dsGxUpxo5LkQNgUkDeq9evSBJ5Z+tl3WX0F69euHff/+twVoREVFDU6A3IFenR66uEDla8TVbWwhdoQEqCwWslQpYWcqhslAgMTMP5xKzEJuUhXNJGtzIyEcjWyXc1VbwcFDBw8EKVkoFtAUG6PQGaAsMMEgSvByt4NfIFv4utmjSyAaWcjnSc3VIysxHUmY+EjPzcCUtF1fSsnElLQfxt/Kgv92iLZMB1pYK2Cgt0MhWCTcHFdzsreDuoIKtygIpmnwkZuYj6fbX1Cxtue9VbW2JQB9HJGXmIzY5q8RrOTo9jsZl4GhcxgMdT6VFnRriRtWgV69eCAwMxKJFi6ple2PGjEFGRkaF7jJfH9WpPuhERFQ3FQXIjLwCZOYVIDNXB22hAa72Imi6Oajg7mAFlYUceoOEQoMEvV6CXpIglwEymQxyGSCXyWCpkMNSISvRopxfoMfVmzm4nJqDy6nZSM3SIlenR16BHvkFRV8N0BbqoS0wQFtoEMtvlyl8wK4dadnaUmH3XmQywFIuh05vqFB5ScLtEwh9hfflaGNpPCFQW1vidEImTlzPRGZeAfaeTwUAWCpk6NnCFQM7eOKxVm5IzdLiTKIGZxI1OJuYhaz8AjjbKOFkq4STjWg1zyvQIzOvQLSo5xWIkw+1NbydrOHtKB4+zjYVPhZEVBoDOhERVZokSThy7RZ++vcG/rqQBt9GNnislRv6tHJHk0YinCVr8rHlRCJ+PZ6AY/EZ1bp/hVwGa0sFrCwVsJDLkJyVj3tckK3Udm2VCtipLGCjsoBSIYe2UIT7/Nthv5GdSnQD8bBHSw8HNHG2wa1cHZI0+Ui+3YpdoDdAaSGHUqEwtiZfvyW6nFxNyxWt83oDZDLAxU60unuoreDrbIOmrnbwd7FFU1dbuNipkF+gL9G6fzNHi2SNFsmafKRo8pGt1cPdQQVPtRU81NbwcLCCj7M1HG2Upd6frtCAc0kaHIvPgK3SAqGt3aG+3U0FABxtlGjhbo/Bgd4PfjDpwUkSUJBrmn1b2ogzyQoYM2YM9u7di71792Lx4sUAxLjC7OxsvP766/jrr79ga2uLvn374pNPPoGLiwsA4IcffsDcuXNx8eJF2NjYoGPHjvjll1/w4Ycf4uuvvwYA44n4nj170KtXr3vW44033sBPP/2E69evw8PDA6NGjcKsWbNgaVn8O/7bb79h3rx5OHnyJOzs7NCzZ0/89NNPAMS9c2bNmoXvvvsOKSkp8PHxwYwZM/Diiy/i1q1bePnll7Fjxw5kZ2ejcePGePPNNzF27NhKHdaKYkAnIqpH9AYJSZp83MzWGlubLRQyWMrlyNIWICVLi1SNFqnZWtzK0cHSQg6Vhei6obKQQyGXQZIkGCTAIEmQJNFdQWUhh5WlKHM6QYOfj93A9Vt5xv3GpefirwtpmPvbGbRws4OTrRKHrqYbQ7NcBniqreFgbQnH2/2XLS3kSMvSIiUrHykaLbK0hZV6n9la0Q2liIOVBZq62qGpqy28Ha1hrVTA2vL2Q6kQ79Gy+L0oFXLYKEW3EWtLBayUcigVlevrXRWSJCEtWwdtoR5u9lb37Q5iq7KArcoCwIPPSKa0kKNDY0d0aOz4wNuiWlCQC8z3Ms2+30wAlLYVKrp48WKcP38e7dq1w7x58wAAlpaW6Nq1K8aNG4dPPvkEeXl5eOONN/Dss89i9+7dSExMxMiRI/HBBx/gqaeeQlZWFv766y9IkoTXXnsNZ8+ehUajwVdffQUAcHZ2vm897O3tsWbNGnh5eeHkyZMYP3487O3tMW3aNADAli1b8NRTT2HmzJn45ptvoNPpSswOOHr0aERHR2PJkiUICAjAlStXkJaWBgB4++23cebMGfzxxx9wcXHBxYsXkZeXV2Y9qgMDOhFRDZMkCdGXbyJXq0cXf2eorS3vv9I96A0SEjLycDE1G5dSsnEpNQfx6bmIv5WLhIw8FOhrZyYOO5UF+rXzQL+2Hrh6MwdRZ1MQczUdF1KyjWU6+zphUAdPPNHBE272VvfcXq6uEAWFEhQKGSzk4iGXySBBnCwYJAkGA1BgMIiuKbe7p+gKDfB2skYjW2WNh+vqIJPJ4GrP6X+p/lCr1VAqlbCxsYGHhwcA4N1330XHjh0xf/58Y7nVq1fDx8cH58+fR3Z2NgoLCzF06FD4+voCANq3b28sa21tDa1Wa9xeRbz11lvG535+fnjttdewfv16Y0B/7733MGLECMydO9dYLiAgAABw/vx5bNy4ETt37kRoaCgAoGnTpsZycXFx6NixI4KCgozbr0kM6ERENURXaMCvxxPwxb5LOJ8sQqtcBrT3VqNbcxcE+TohM68AV2/m4trNHFy9mQttgR7uDla3uytYwdVehYzcAuPsGDcy8hCfngttYfl9ly3kMrjYqWCQRF/uAr0BhXoJtioLuNmrbvf7VsHJVolCvVTchaNQD0jiqrb8dp9vmUwGXaHBWEZbqIezrRIDO3ghtLU7rJUK437H9WyKzLwC7Dufilu5OvRu6Vapvsg2SgugdK8MAIACxcHbGgo4WD3YSQ5RnWBpI1qyTbXvB3D8+HHs2bMHdnZ2pV67dOkS+vbtiz59+qB9+/YICwtD37598fTTT8PJyanK+9ywYQOWLFmCS5cuGU8AHBwcjK8fO3YM48ePL3PdY8eOQaFQ4NFHHy3z9ZdeegnDhg3D0aNH0bdvXwwZMgTdunWrcl3vhwGdiOge9AYJukIDCg0G4+DFolArgquYqaOg0FAiDF9MzcaaA1eRpMkHANgqFXB3sMLltBwcv56J49czy93nuaT7DwBUKuTwd7FFMzdbNHWxg28jG/g4i4eHgxUUctO0JKutLTEowESX5InqG5mswt1MzE12djYGDRqEBQsWlHrN09MTCoUCO3fuxN9//40dO3bg008/xcyZM3Hw4EH4+/tXen/R0dEYNWoU5s6di7CwMKjVaqxfvx4ff/yxsUx5d52/32sA0L9/f1y7dg1bt27Fzp070adPH0yePBkfffRRpetaEQzoRNTgSJKEAr2EvAI9EjLycO1mLuLSRQt2UmY+0nN0yMjVIT1HB01+xftFl8XVXoUXuvvjueAmUFtbIjEzD39fvIkDl9Jw8nomXOxU8HOxgW8jW/g1soG10gLJmUVT5uUhNUsLB2tLNHYsmiXDBo2drNHYyRoWCk5lR0TmQalUQq/XG7/v1KkTfvzxR/j5+cHCouy4KZPJ0L17d3Tv3h2zZs2Cr68vfvrpJ0RERJTa3v38/fff8PX1xcyZM43Lrl27VqJMhw4dEBUVVebAzvbt28NgMGDv3r3GLi53c3V1RXh4OMLDw9GzZ0+8/vrrDOhERJVVqDfg3/gM7I1Nxd7zqbialgOt3gDdPbqHVISFXGYcOKm8/bBUyGEpFwMyLRRy2KkUGBzgjcEdvaCyKO4G4qm2xrDOjTGsc+MHfXtERGbDz88PBw8exNWrV2FnZ4fJkydj5cqVGDlyJKZNmwZnZ2dcvHgR69evx5dffonDhw8jKioKffv2hZubGw4ePIjU1FS0bt3auL3t27cjNjYWjRo1glqtLjEby91atGiBuLg4rF+/Hl26dMGWLVuMs7MUmT17Nvr06YNmzZphxIgRKCwsxNatW/HGG2/Az88P4eHheOGFF4yDRK9du4aUlBQ8++yzmDVrFjp37oy2bdtCq9Xi999/N9a1JjCgE1G9kafT43RCJo7FZ+Dw1Vs4cCkNWfdpAXe0sYSvsw2aNLKFr7MNvJ2s4WyrhPPteZ8dbZSwtlRAIRczohT1yyYiomKvvfYawsPD0aZNG+Tl5eHKlSs4cOAA3njjDfTt2xdarRa+vr7o168f5HI5HBwcsG/fPixatAgajQa+vr74+OOP0b9/fwDA+PHj8eeffyIoKAjZ2dn3nWbxySefxKuvvoqXX34ZWq0WAwYMwNtvv405c+YYy/Tq1QubNm3CO++8g/fffx8ODg545JFHjK8vX74cb775JiZNmoSbN2+iSZMmePPNNwGIKwQzZszA1atXYW1tjZ49e2L9+vU1ciwBQCbd61ae9ZBGo4FarUZmZmaJgQNEZN5SsvJx8HI6Dl65ibOJWZDLcPuGNaIFOyEjD+eSsox3XiziZGOJni1c8ehDrgjwcYS1Ukyvd+fUgURk/hrC53d+fj6uXLkCf39/WFnde9Yjqnsq8/NlCzoRmVyOthCJmXlIyMhHSpYWGbk6aG7fcfJWbgFO3cjE5bScCm3L1V6FQB9HBPo4ontzF7T3VptswCQREVFVMKATUa3SFRpw+Fo69p1Pw9+X0nA1LadCAzFlMqC1hwOCmzoj0McRSoW4Tbru9iwqzjZKBPg4wlNtxS4oRET1zPz580vMqX6nnj174o8//qjlGtUsBnQiqnaZeQW4mJKF1CwdMvN0uJVbgFu5OlxMzhY37NGVHplvr7KAp6MV3B2s4GSjhNraEo424o6Tfo1s0cXPucQtyYmIqOGYOHEinn322TJfu98UiXURAzoRPRBJkhB96Sb+upiGc4kaxCZlISEz/57ruNgp0bOFKx55yAVtvdTwVFvBnjeeISKicjg7O8PZ2dnU1ag1DOhEVCWFegO2nkrC53sv4XSCptTrRXfCdLJRwtHGEo7WSniqrdCteSO09nCAnP3CiYjK1MDm72gwKvNzZUAnontK0eQj/pa4tXzRnTPj03PxdfRVxKfnAQCsLOUY1MELHXwc0crDHg+520NtzRZxIqLKKJrnOzc3t15222jodDodAEChuP/sYQzoRFSCJEm4lJqN7aeTseN00j1vSe9sq0R4iB+eD/GFs62yFmtJRFT/KBQKODo6IiUlBQBgY2PDQe/1hMFgQGpqKmxsbMq9s+qdGNCJGqjEzDzsO5+K+PQ8ZOTpkJlXiIxcHeLTc3H1Zq6xnEwGNHayhpWFwnjXTBulAmFtPfBMZx9YKzmPOBFRdfHw8AAAY0in+kMul6NJkyYVOuliQCdqIAr0Bpy4noHd51Kw+1wqziaW7jdeRKmQo3vzRujb1gN9WrvBzZ43zCAiqg0ymQyenp5wc3NDQUGBqatD1UipVEIul1eoLAM6UT0kSRJSs7U4eT0TR+Nu4ci1Wzgen4m8guLpDWUyINDHEe291XC0toT69tSGjeyUCPJ14qwqREQmpFAoKtRXmeonBnSiOi4rvwBbTiQi5ko6EjLzkJiZj8TMfOgKDaXKqq0t0bOFCx5r5YZHH3JFIzuVCWpMRERE98KATlQHSZKEmCvp2Hj4OraeTCzRMl5EJgP8XWwR5OuETk2c0NnXCc1c7Ti9IRERkZljQCeqI/QGCf/G3cLOM8n441QS4tKLB3I2dbXFoA5e8HexhafaCl6O1nB3sILSomJ93YiIiMh8MKATmalcXSEupeTgYmoW/rmUjl1nk3EzR2d83VapwKAALzwT5INOTRw5FRcREVE9wYBOZEKJmXk4n5yNpNt9x5My83EjIw+XU3NwIyOvVHl7Kwv0aeWGx9t4oHcrV9go+SdMRERU3/DTnagWFegNOHLtFvbEpuDPc6mITc66Z/lGtko0c7NDWy8HhLZ2R1d/Z1gq2G2FiIioPmNAJ6pBBoOE8ylZiL50E39fuol/Lt1ElrbQ+LpcBjR3s4OXozU81VbwcLCGh1qFpq52aO5qByfenZOIiKjBYUAnqmYGg4R/rtzEpsPXse98aol+4wDgbKvEow+5oldLVzzSwpUhnIiIiEpgQCeqJilZ+fjhyHVsOBSPazeLZ1ixspSji58zujVzQUizRmjvrYaCUx0SERFRORjQiapAb5BwKTUbJ65n4sT1DBy/nonTNzJRaJAAAHYqCzwZ6IUnA7zQqYkTpzskIiKiCmNAJ6ogg0HCwSvp+OXYDfxxKgmZeQWlynRq4ogRXZpgQAdP2Kr450VERESVZ/IEsWzZMnz44YdISkpCQEAAPv30U3Tt2rXMsgUFBYiMjMTXX3+NGzduoGXLlliwYAH69etXy7Wm+kpXaMDltGxo8gqRqytEfoEeeQV6nE3Mwq/HEpCkyTeWtVEq0M5LjQ6N1WjfWI2OPk5o0sjGhLUnIiKi+sCkAX3Dhg2IiIjAihUrEBwcjEWLFiEsLAyxsbFwc3MrVf6tt97Ct99+i5UrV6JVq1bYvn07nnrqKfz999/o2LGjCd4B1WXpOTrEJmXhXJIGpxM0OJOgwYWULBTopXLXcbCywBPtPfFkoBeC/RuxLzkRERFVO5kkSeWnkRoWHByMLl26YOnSpQAAg8EAHx8fTJkyBdOnTy9V3svLCzNnzsTkyZONy4YNGwZra2t8++23FdqnRqOBWq1GZmYmHBwcqueNUJ0Qn56Ltf9cw5kEDWKTs5CapS2znL2VBVztVLBWKmBtqYC1UgEXOxX6tfNAr5auUFkoarnmRETEz29qSEzWgq7T6XDkyBHMmDHDuEwulyM0NBTR0dFlrqPVamFlZVVimbW1Nfbv31/ufrRaLbTa4iCm0WgesOZU1+TqCrH8z0v4fN9l6AoNJV5r4myDh9zt0dbLAW29HNDGywHejtaQydgyTkRERKZhsoCelpYGvV4Pd3f3Esvd3d1x7ty5MtcJCwvDwoUL8cgjj6BZs2aIiorC5s2bodfry91PZGQk5s6dW611p7pBkiT8ejwB7/9xDomZou94t2aNMCTQGw952KOFmx0HchIREZHZqVPpZPHixRg/fjxatWoFmUyGZs2aYezYsVi9enW568yYMQMRERHG7zUaDXx8fGqjulRDziVpsOVEIq6k5SAzrwC3cnXIyC1AVr64Q6dcBshlMuglCRm5YqaVxk7WeGtAa4S19WDrOBEREZk1kwV0FxcXKBQKJCcnl1ienJwMDw+PMtdxdXXFzz//jPz8fNy8eRNeXl6YPn06mjZtWu5+VCoVVCpVtdadal98ei5+PZ6AX47dwPnk7AqvZ2Upx6RezTHhkaawsmTfcSIiIjJ/JgvoSqUSnTt3RlRUFIYMGQJADBKNiorCyy+/fM91rays4O3tjYKCAvz444949tlna6HGVJskScLpBA12nU3GrrPJOHWjeOyAUiFHr5au6OrvDGdbJRxtLOFoo4SDlQVkMhkkSYJBAgySBE8Ha6htLE34ToiIiIgqx6RdXCIiIhAeHo6goCB07doVixYtQk5ODsaOHQsAGD16NLy9vREZGQkAOHjwIG7cuIHAwEDcuHEDc+bMgcFgwLRp00z5NqgaaAv1OJ+UjVMJmThxPRN7Y1OQkFk857hcBoQ0a4TBAd4Ia+cBtTVDNxEREdVPJg3ow4cPR2pqKmbNmoWkpCQEBgZi27ZtxoGjcXFxkMuLb5Gen5+Pt956C5cvX4adnR2eeOIJrF27Fo6OjiZ6B3Q/kiTh+q08nLqRiRM3MnEuUYP8gpIzqWTkFeBCchYKDSVn/LSylKNnC1c83sYdj7Vyg4sduyoRERFR/WfSedBNgfOo1jy9QcL+i2nYdDgeBy6m4dbtgZr342RjibZearT1ckAXP2f0aOHCfuNERASAn9/UsNSpWVzIvF27mYMfjlzHD0euG6c1BABLhQwtPezR3luNtl7qUt1TrC0VaO3lAC+1FWdYISIiogaPAZ0qTJIk6A0S9Le/XruZi8PXbuHotVs4fC0d8el5xrJqa0sMCfTCk4HeaOftwLtvEhEREVUQAzqVq0BvwJ5zKdh4+Dr2nU+FTm+4Z3m5DOjRwhXPBjVGaGt3dk8hIiIiqgIGdDIyGCSk5WgRn56H7aeTsPnodaRl68otb6eyQMcmjujUxAlBfk4I9HGEvRVnVyEiIiJ6EAzoDYAkSUjL1iEuPRfx6blIy9YiI/f2HTjzCnAzW4vEzHwkZuSXaiV3sVNhaCdvPNXRG272KijkMsjlMihkMlhbKiCXs884ERERUXViQK9n9AYJsUlZOBJ3C0eupuNsYhbi0nORV6Cv0PpyGeBmb4UOjdV4JsgHvVq6wlIhv/+KRERERFQtGNDruFxdIY7FZeDQVTFQ81hcBrK0haXKyWSAl9oajZ2s4e5gBScbS6htlHCysYSzrRKeamt4OVrB3cGKgZyIiIjIhBjQzVzczVzsPZ+CmKu3oCss2QqemJmP0wka6O+6wY+tUoGOTZzQ2Vf0C/dzsYW3ozWUFgzeREREROaOAd3MaAv1+OdyOvacS8G+86m4nJZz33U81VYI8nNGkK8YrNnKwwEK9g0nIiIiqpMY0M1AZl4B/oxNwY4zydgbm4rsO7qoWMhl6OTrhJ7NXeBkqyyxnoO1JTr7OsHb0bq2q0xERERENYQBvRbkF+hxITkbsclZOJ+chYSMPKTn6HAzW4ebOVqk5+hwZy8VN3sV+rR2w6MPuaFb80Zw4NSFRERERA0GA3oNuZmtxVcHrmLryURcuZkDSbp3+RZudni8jTv6tvVAB281py8kIiIiaqAY0KtZQkYevth3GesPxSG/oHhOcScbS7T0sEdLd3v4ONvA1V4FZ1slGtmq4GovHkREREREDOjVJO5mLpbuuYCf/r2BAr1oLm/vrcaER5ri4aaN4GKnhEzGVnEiIiIiujcG9GpyOS0bGw9fBwA83NQZk3s3R4/mLgzlRERERFQpDOjV5NGHXPFCd38M6OCJzr5Opq4OEREREdVRDOjVRCaTYdagNqauBhERERHVcby1JBERERGRGWFAJyIiIiIyIwzoRERERERmhAGdiIiIiMiMMKATEREREZkRBnQiIiIiIjPCgE5EREREZEYY0ImIiIiIzAgDOhERERGRGWFAJyIiIiIyIwzoRERERERmhAGdiIiIiMiMMKATEREREZkRkwf0ZcuWwc/PD1ZWVggODkZMTMw9yy9atAgtW7aEtbU1fHx88OqrryI/P7+WaktEREREVLNMGtA3bNiAiIgIzJ49G0ePHkVAQADCwsKQkpJSZvnvvvsO06dPx+zZs3H27FmsWrUKGzZswJtvvlnLNSciIiIiqhkmDegLFy7E+PHjMXbsWLRp0wYrVqyAjY0NVq9eXWb5v//+G927d8dzzz0HPz8/9O3bFyNHjrxvqzsRERERUV1hsoCu0+lw5MgRhIaGFldGLkdoaCiio6PLXKdbt244cuSIMZBfvnwZW7duxRNPPFHufrRaLTQaTYkHEREREZG5sjDVjtPS0qDX6+Hu7l5iubu7O86dO1fmOs899xzS0tLQo0cPSJKEwsJCTJw48Z5dXCIjIzF37txqrTsRERERUU0x+SDRyvjzzz8xf/58fPbZZzh69Cg2b96MLVu24J133il3nRkzZiAzM9P4iI+Pr8UaExERERFVjsla0F1cXKBQKJCcnFxieXJyMjw8PMpc5+2338bzzz+PcePGAQDat2+PnJwcTJgwATNnzoRcXvp8Q6VSQaVSVf8bICIiIiKqASZrQVcqlejcuTOioqKMywwGA6KiohASElLmOrm5uaVCuEKhAABIklRzlSUiIiIiqiUma0EHgIiICISHhyMoKAhdu3bFokWLkJOTg7FjxwIARo8eDW9vb0RGRgIABg0ahIULF6Jjx44IDg7GxYsX8fbbb2PQoEHGoE5EREREVJeZNKAPHz4cqampmDVrFpKSkhAYGIht27YZB47GxcWVaDF/6623IJPJ8NZbb+HGjRtwdXXFoEGD8N5775nqLRARERERVSuZ1MD6hmg0GqjVamRmZsLBwcHU1SEiIqIK4Oc3NSR1ahYXIiIiIqL6jgGdiIiIiMiMMKATEREREZkRBnQiIiIiIjPCgE5EREREZEYY0ImIiIiIzAgDOhERERGRGWFAJyIiIiIyIwzoRERERERmhAGdiIiIiMiMMKATEREREZkRBnQiIiIiIjPCgE5EREREZEYY0ImIiIiIzAgDOhERERGRGWFAJyIiIiIyIwzoRERERERmhAGdiIiIiMiMMKATEREREZkRBnQiIiIiIjPCgE5EREREZEYY0ImIiIiIzAgDOhERERGRGWFAJyIiIiIyIwzoRERERERmhAGdiIiIiMiMMKATEREREZkRBnQiIiIiIjPCgE5EREREZEYY0ImIiIiIzIhZBPRly5bBz88PVlZWCA4ORkxMTLlle/XqBZlMVuoxYMCAWqwxEREREVHNMHlA37BhAyIiIjB79mwcPXoUAQEBCAsLQ0pKSpnlN2/ejMTEROPj1KlTUCgUeOaZZ2q55kRERERE1c/kAX3hwoUYP348xo4dizZt2mDFihWwsbHB6tWryyzv7OwMDw8P42Pnzp2wsbFhQCciIiKiesGkAV2n0+HIkSMIDQ01LpPL5QgNDUV0dHSFtrFq1SqMGDECtra2NVVNIiIiIqJaY2HKnaelpUGv18Pd3b3Ecnd3d5w7d+6+68fExODUqVNYtWpVuWW0Wi20Wq3xe41GU/UKExERERHVMJN3cXkQq1atQvv27dG1a9dyy0RGRkKtVhsfPj4+tVhDIiIiIqLKMWlAd3FxgUKhQHJyconlycnJ8PDwuOe6OTk5WL9+PV588cV7lpsxYwYyMzONj/j4+AeuNxERERFRTTFpQFcqlejcuTOioqKMywwGA6KiohASEnLPdTdt2gStVov//Oc/9yynUqng4OBQ4kFEREREZK5M2gcdACIiIhAeHo6goCB07doVixYtQk5ODsaOHQsAGD16NLy9vREZGVlivVWrVmHIkCFo1KiRKapNRERERFQjTB7Qhw8fjtTUVMyaNQtJSUkIDAzEtm3bjANH4+LiIJeXbOiPjY3F/v37sWPHDlNUmYiIiIioxsgkSZJMXYnapNFooFarkZmZye4uREREdQQ/v6khqdOzuBARERER1TcM6EREREREZoQBnYiIiIjIjDCgExERERGZEQZ0IiIiIiIzwoBORERERGRGGNCJiIiIiMwIAzoRERERkRlhQCciIiIiMiMM6EREREREZoQBnYiIiIjIjDCgExERERGZEQZ0IiIiIiIzwoBORERERGRGGNCJiIiIiMwIAzoRERERkRlhQCciIiIiMiMM6EREREREZoQBnYiIiIjIjDCgExERERGZEQZ0IiIiIiIzwoBORERERGRGGNCJiIiIiMwIAzoRERERkRlhQCciIiIiMiMM6EREREREZoQBnYiIiIjIjDCgExERERGZEQZ0IiIiIiIzwoBORERERGRGGNCJiIiIiMxIlQJ6fHw8rl+/bvw+JiYGr7zyCr744otKb2vZsmXw8/ODlZUVgoODERMTc8/yGRkZmDx5Mjw9PaFSqfDQQw9h69atld4vEREREZE5qlJAf+6557Bnzx4AQFJSEh5//HHExMRg5syZmDdvXoW3s2HDBkRERGD27Nk4evQoAgICEBYWhpSUlDLL63Q6PP7447h69Sp++OEHxMbGYuXKlfD29q7K2yAiIiIiMjsySZKkyq7k5OSEf/75By1btsSSJUuwYcMGHDhwADt27MDEiRNx+fLlCm0nODgYXbp0wdKlSwEABoMBPj4+mDJlCqZPn16q/IoVK/Dhhx/i3LlzsLS0rGy1AQAajQZqtRqZmZlwcHCo0jaIiIiodvHzmxqSKrWgFxQUQKVSAQB27dqFJ598EgDQqlUrJCYmVmgbOp0OR44cQWhoaHFl5HKEhoYiOjq6zHV+/fVXhISEYPLkyXB3d0e7du0wf/586PX6cvej1Wqh0WhKPIiIiIiIzFWVAnrbtm2xYsUK/PXXX9i5cyf69esHAEhISECjRo0qtI20tDTo9Xq4u7uXWO7u7o6kpKQy17l8+TJ++OEH6PV6bN26FW+//TY+/vhjvPvuu+XuJzIyEmq12vjw8fGp4LskIiIiIqp9VQroCxYswOeff45evXph5MiRCAgIACBauLt27VqtFbyTwWCAm5sbvvjiC3Tu3BnDhw/HzJkzsWLFinLXmTFjBjIzM42P+Pj4GqsfEREREdGDsqjKSr169UJaWho0Gg2cnJyMyydMmAAbG5sKbcPFxQUKhQLJyckllicnJ8PDw6PMdTw9PWFpaQmFQmFc1rp1ayQlJUGn00GpVJZaR6VSGbvjEBERERGZuyq1oOfl5UGr1RrD+bVr17Bo0SLExsbCzc2tQttQKpXo3LkzoqKijMsMBgOioqIQEhJS5jrdu3fHxYsXYTAYjMvOnz8PT0/PMsM5EREREVFdU6WAPnjwYHzzzTcAxLzkwcHB+PjjjzFkyBAsX768wtuJiIjAypUr8fXXX+Ps2bN46aWXkJOTg7FjxwIARo8ejRkzZhjLv/TSS0hPT8fUqVNx/vx5bNmyBfPnz8fkyZOr8jaIiIiIiMxOlQL60aNH0bNnTwDADz/8AHd3d1y7dg3ffPMNlixZUuHtDB8+HB999BFmzZqFwMBAHDt2DNu2bTMOHI2LiysxK4yPjw+2b9+OQ4cOoUOHDvjf//6HqVOnljklIxERERFRXVSledBtbGxw7tw5NGnSBM8++yzatm2L2bNnIz4+Hi1btkRubm5N1LVacB5VIiKiuoef39SQVKkFvXnz5vj5558RHx+P7du3o2/fvgCAlJQU/tEQERERET2AKgX0WbNm4bXXXoOfnx+6du1qHNS5Y8cOdOzYsVorSERERETUkFSpiwsAJCUlITExEQEBAZDLRc6PiYmBg4MDWrVqVa2VrE68REZERFT38PObGpIqzYMOAB4eHvDw8MD169cBAI0bN67RmxQRERERETUEVeriYjAYMG/ePKjVavj6+sLX1xeOjo545513SsxRTkRERERElVOlFvSZM2di1apVeP/999G9e3cAwP79+zFnzhzk5+fjvffeq9ZKEhERERE1FFXqg+7l5YUVK1bgySefLLH8l19+waRJk3Djxo1qq2B1Yx82IiKiuoef39SQVKmLS3p6epkDQVu1aoX09PQHrhQRERERUUNVpYAeEBCApUuXllq+dOlSdOjQ4YErRURERETUUFWpD/oHH3yAAQMGYNeuXcY50KOjoxEfH4+tW7dWawWJiIiIiBqSKrWgP/roozh//jyeeuopZGRkICMjA0OHDsXp06exdu3a6q4jEREREVGDUeUbFZXl+PHj6NSpE/R6fXVtstpxkAkREVHdw89vakiq1IJOREREREQ1gwGdiIiIiMiMMKATEREREZmRSs3iMnTo0Hu+npGR8SB1ISIiIiJq8CoV0NVq9X1fHz169ANViIiIiIioIatUQP/qq69qqh5ERERERAT2QSciIiIiMisM6EREREREZoQBnYiIiIjIjDCgExERERGZEQZ0IiIiIiIzwoBORERERGRGGNCJiIiIiMwIAzoRERERkRlhQCciIiIiMiMM6EREREREZoQBnYiIiIjIjDCgExERERGZEbMI6MuWLYOfnx+srKwQHByMmJiYcsuuWbMGMpmsxMPKyqoWa0tEREREVHNMHtA3bNiAiIgIzJ49G0ePHkVAQADCwsKQkpJS7joODg5ITEw0Pq5du1aLNSYiIiIiqjkmD+gLFy7E+PHjMXbsWLRp0wYrVqyAjY0NVq9eXe46MpkMHh4exoe7u3st1piIiIiIqOaYNKDrdDocOXIEoaGhxmVyuRyhoaGIjo4ud73s7Gz4+vrCx8cHgwcPxunTp8stq9VqodFoSjyIiIiIiMyVSQN6Wloa9Hp9qRZwd3d3JCUllblOy5YtsXr1avzyyy/49ttvYTAY0K1bN1y/fr3M8pGRkVCr1caHj49Ptb8PIiIiIqLqYvIuLpUVEhKC0aNHIzAwEI8++ig2b94MV1dXfP7552WWnzFjBjIzM42P+Pj4Wq4xEREREVHFWZhy5y4uLlAoFEhOTi6xPDk5GR4eHhXahqWlJTp27IiLFy+W+bpKpYJKpXrguhIRERER1QaTtqArlUp07twZUVFRxmUGgwFRUVEICQmp0Db0ej1OnjwJT0/PmqomEREREVGtMWkLOgBEREQgPDwcQUFB6Nq1KxYtWoScnByMHTsWADB69Gh4e3sjMjISADBv3jw8/PDDaN68OTIyMvDhhx/i2rVrGDdunCnfBhERERFRtTB5QB8+fDhSU1Mxa9YsJCUlITAwENu2bTMOHI2Li4NcXtzQf+vWLYwfPx5JSUlwcnJC586d8ffff6NNmzamegtERERERNVGJkmSZOpK1CaNRgO1Wo3MzEw4ODiYujpERERUAfz8poakzs3iQkRERERUnzGgExERERGZEQZ0IiIiIiIzwoBORERERGRGGNCJiIiIiMwIAzoRERERkRlhQCciIiIiMiMM6EREREREZoQBnYiIiIjIjDCgExERERGZEQZ0IiIiIiIzwoBORERERGRGGNCJiIiIiMwIAzoRERERkRlhQCciIiIiMiMM6EREREREZoQBnYiIiIjIjDCgExERERGZEQZ0IiIiIiIzwoBORERERGRGGNCJiIiIiMwIAzoRERERkRlhQCciIiIiMiMM6EREREREZoQBnYiIiIjIjDCgExERERGZEQZ0IiIiIiIzwoBORERERGRGGNCJiIiIiMwIAzoRERERkRlhQCciIiIiMiNmEdCXLVsGPz8/WFlZITg4GDExMRVab/369ZDJZBgyZEjNVpCIiIiIqJaYPKBv2LABERERmD17No4ePYqAgACEhYUhJSXlnutdvXoVr732Gnr27FlLNSUiIiIiqnkmD+gLFy7E+PHjMXbsWLRp0wYrVqyAjY0NVq9eXe46er0eo0aNwty5c9G0adNarC0RERERUc0yaUDX6XQ4cuQIQkNDjcvkcjlCQ0MRHR1d7nrz5s2Dm5sbXnzxxfvuQ6vVQqPRlHgQEREREZkrkwb0tLQ06PV6uLu7l1ju7u6OpKSkMtfZv38/Vq1ahZUrV1ZoH5GRkVCr1caHj4/PA9ebiIiIiKimmLyLS2VkZWXh+eefx8qVK+Hi4lKhdWbMmIHMzEzjIz4+voZrSURERERUdRam3LmLiwsUCgWSk5NLLE9OToaHh0ep8pcuXcLVq1cxaNAg4zKDwQAAsLCwQGxsLJo1a1ZiHZVKBZVKVQO1JyIiIiKqfiZtQVcqlejcuTOioqKMywwGA6KiohASElKqfKtWrXDy5EkcO3bM+HjyySfRu3dvHDt2jN1XiIiIiKjOM2kLOgBEREQgPDwcQUFB6Nq1KxYtWoScnByMHTsWADB69Gh4e3sjMjISVlZWaNeuXYn1HR0dAaDUciIiIiKiusjkAX348OFITU3FrFmzkJSUhMDAQGzbts04cDQuLg5yeZ3qKk9EREREVGUySZIkU1eiNmk0GqjVamRmZsLBwcHU1SEiIqIK4Oc3NSRsmiYiIiIiMiMM6EREREREZoQBnYiIiIjIjDCgExERERGZEQZ0IiIiIiIzwoBORERERGRGGNCJiIiIiMwIAzoRERERkRlhQCciIiIiMiMM6EREREREZoQBnYiIiIjIjDCgExERERGZEQZ0IiIiIiIzwoBORERERGRGGNCJiIiIiMwIAzoRERERkRlhQCciIiIiMiMM6EREREREZoQBnYiIiIjIjDCgExERERGZEQZ0IiIiIiIzwoBORERERGRGGNCJiIiIiMwIAzoRERERkRlhQCciIiIiMiMM6EREREREZoQBnYiIiIjIjDCgExERERGZEQZ0IiIiIiIzwoBORERERGRGzCKgL1u2DH5+frCyskJwcDBiYmLKLbt582YEBQXB0dERtra2CAwMxNq1a2uxtkRERERENcfkAX3Dhg2IiIjA7NmzcfToUQQEBCAsLAwpKSlllnd2dsbMmTMRHR2NEydOYOzYsRg7diy2b99eyzUnIiIiIqp+MkmSJFNWIDg4GF26dMHSpUsBAAaDAT4+PpgyZQqmT59eoW106tQJAwYMwDvvvHPfshqNBmq1GpmZmXBwcHiguhMREVHt4Oc3NSQmbUHX6XQ4cuQIQkNDjcvkcjlCQ0MRHR193/UlSUJUVBRiY2PxyCOPlFlGq9VCo9GUeBARERERmSuTBvS0tDTo9Xq4u7uXWO7u7o6kpKRy18vMzISdnR2USiUGDBiATz/9FI8//niZZSMjI6FWq40PHx+fan0PRERERETVyeR90KvC3t4ex44dw6FDh/Dee+8hIiICf/75Z5llZ8yYgczMTOMjPj6+ditLRERERFQJFqbcuYuLCxQKBZKTk0ssT05OhoeHR7nryeVyNG/eHAAQGBiIs2fPIjIyEr169SpVVqVSQaVSVWu9iYiIiIhqiklb0JVKJTp37oyoqCjjMoPBgKioKISEhFR4OwaDAVqttiaqSERERERUq0zagg4AERERCA8PR1BQELp27YpFixYhJycHY8eOBQCMHj0a3t7eiIyMBCD6lAcFBaFZs2bQarXYunUr1q5di+XLl5vybRARERERVQuTB/Thw4cjNTUVs2bNQlJSEgIDA7Ft2zbjwNG4uDjI5cUN/Tk5OZg0aRKuX78Oa2trtGrVCt9++y2GDx9uqrdARERERFRtTD4Pem3jPKpERER1Dz+/qSGpk7O4EBERERHVVwzoRERERERmhAGdiIiIiMiMMKATEREREZkRBnQiIiIiIjPCgE5EREREZEYY0ImIiIiIzAgDOhERERGRGWFAp6rR5QDabFPXgoiIiKjeYUCnyjPogS96A5+FALpcU9eGiIiIqF5hQKfKSzoBpMUCmXFAwlFT14aIiIioXmFAp8q7eqD4eXxM+eX2fgCsHwUU5Nd8nYiIiIjqCQb02qYvEF1E6rJrfxc/v36o7DK6HGDvAuDc78DlP2ulWkRERET1AQN6bcrLAJZ0Ar4MBQwGU9emagwGIO6OgB4fA0hS6XJx/wCGQvH86l+1UzciIiKieoABvTYdW1fcbzv+oKlrUzUpZ4C8W4ClLaBQArlpQPrl0uWu7i/7ORERERHdEwN6bTEYgJiVxd+f+sF0dXkQRd1bmgQDnoHieVndXO5sNU86Ia4eEBEREdF9MaDXlktRwK0rAGTi+9M/A/pCU9aoaq7dbg337Qb4dBXP7x4oqs0Cbtye3cXaGZAMossLEREREd0XA3ptiflCfO06HrBpJLqGXPnTpFWqNEkqbkH37QE07iKeX78roMcdBCQ94OgLtB4klrEfOhEREVGFMKDXhpuXgAs7xfPgiUCbIeL5yR9NVqUqSTsP5KQCFlaAd6fiFvTk06LVvMjVfeKrX0/xANgPnYiIiKiCGNBrw+HVACSg+eNAo2ZA+6fF8nO/1605wq/dnv+8cRfAQgU4eAFqH9GF5cYdNywqCuP+PQG/7uI5+6ETERERVQgDek3T5QD/rhXPu04QX30eBhy8Aa0GuLDDdHWrrKIbFPl2L152dzeXfA2QcEw89+shQrxzM/ZDJyIiIqogBvSadnITkJ8JOPkBzUPFMrkcaDdUPK8rs7lIUnELut8dAd04UPT2TC5x/4j+507+gLrx7fI9xFf2QyciIiK6Lwb0miRJQMyX4nmX8SKYF2l3u5vL+e2i1dnc3boCZCUCcsviVnMAaHw7oF+/fcMiY//zHsVl2A+diIiIqMIY0KubvgDITgFSzgLHvgOSTwIW1kDHUSXLeQYAjZoDhflA7Naq7adQVz11roii7i3enQFL6+LlHu3FoNG8W8DNi8CV263k/o8Ul2E/dKpPbl0Fjqypm9OkEhFRnWBh6grUG9eige+Hi+4sd+vwLGDtVHKZTCZa0fe+D5z8AQgYUfF9SRLw7VAg6SQw+RBg5/pgda+Isrq3AICFEvDqCMRFAxd3iRAOlGxBL+qHnn5JdIFp2a/m60tUE9IvA6v7AdnJ4gQ5eIKpa2S+JEn8nzPFfn8cJ35GozaVbFAgIqoj2IJeXZQ2d4RzmbhBT6MWQNPeQM//K3udotlcLu8Bcm5WfF+X9wBX9olW64u7HqjaFXatjAGiRYq6vEQvE4NBnZuJUH4n9kOnui4rCVj7lAh+AHD0axEGqbQfxwMfNgPSLtb+vs/8LMb2XP0LOPt77e+fiKgaMKBXF5eWojX79cvArJvAG1eAKYeB0T8DTr7lrNMC8OgAGAqB3/4nWucq4sDi4ue10a87Ix7IiANkiuJBoXcqWpYZL7769yxdxlz7oeemi3nqie4lLwP4dpjo3uLoCyhUQPIpIOFfU9fM/CSfBk5uBHJvAjvfrt19F+qAXXOLvz/+Xe3un4iomjCgVxdLK8D1IcC2ESBXVHy9blPE13O/A58GAb9MFiGgPAnHgMt/Fn//oC3S+gLRX768/rQGvfiwBQCvQEBlX7pM47tCu19ZAb2a+6Eb9IAuFzAYqr6NlHPA0iDg087AiY0PXqeqkCS2wpq7gjzg+5EikNu6iZPuNk+K145+82Dbro8/++jPip/Hbi0el1IbjnwlBrQXdSm8tAfIvFGxdbOSgJ8nAad/rrHqERFVFAO6qXV4FhgXJaZglPTAv9+KwPjHdBFC7/b3p+LrQ/1Fi3bGNdG6XVW//g/47GHgo+bATy+JS8K6XHGSsPtd4JN2QNQ8UfbOgZ93sncXrYpF7ux/XqQ650PPzwSWBALzPYF5TsC7HsACP2BRe2DD8yIg3DgiTj7Kk34Z+GawaOWDBPw0sfYvh2uzgRU9gGVdyx67UBkGAwfg1oRCHbBpLBD3N6BSA89vBpybAp1Gi9dP/iDudVAVsduASB/xN6jLrb46m1J2SvEJfVF3uB0zH+xEuqLyM4G9C8TzPrOAJt0ASMCJDfdft+gKybF1wObxQGpsTdaUiOi+GNDNQeMg4D8/Ai/uFH3WDYXAweXFwbjIrWvA6Z/E894zAO9O4nnRDCuVdeNI8SXgvFvi+YZRIuwuDgD2fQhkJYjWqIcnAd2nlr+tom4ujVoA9h5llykK7n99DOx+D/hnBXBikwjslfkAP/ZdyZOSwjxR/4w44OyvwPYZwMrHgPebAN8/V/IupwCQeR34ejCQnQS4tRGDdSU98MPY0n36NYniZGnds/duiTv5gwj857dX/H3seU+0yqadB3bNqfh6dyvIB74eCHzcCrhQS2MS6iKDASjUVrx8brroc37+DzFT0XPrxaxFAODbQ8z1r8uqWourJImfuS5L9GVf2RtIPlP57ZibQ18Cep24qvbsN4DKAUg8XrGQ/KAOLBYn3I1aAB1HA4HPieXHv7/3lQpdLvD9CPG3CIj6/zqldk4qiIjKYRYBfdmyZfDz84OVlRWCg4MRExNTbtmVK1eiZ8+ecHJygpOTE0JDQ+9Zvk7x6Sounw9ZIb4/sAg4vr749X8+E0GyaW8xTaNx4GUV+nVLErBjlnje/llgzBYRwh2bAHotABnQ7DHg6a+A/4sF+kWWnonmTi2fEF/bDim/TLPe4uv1GGDfB8C2N4DN44DVYeKE4M8FIjzfi8EAxKwUz/stAF6/BLxyEpgcI95D6BzgoX6AlSNQkAvEbhHh57sRontQdooI0plxoiX0+Z+Bpz4HWj8pPpjX/0ec8GQli2C+OECcLF3YLmbp0WaXrtPFXcDmCaLr0XfPilZ8TcK938f1w8A/y4u/P7wauPb3vdcpiyQBW/9PDOItzAN+fFGcyNVnBgOQfqVy3UOykoHPHwEWtgGuH7l/+ZuXgFWPA9f2A0p7YOT3gG+34tflcqDjf8TzojsFV8bFKCD1LKC0A+w8gNRz4vf0yJq62+2lIE8EdAAImQTYuhQPkI+aV7NXCTJviEHqAPD4XEBhAbQZLKa4TTsvGiPKoi8QJ+Zx0eIKycgN4mcSfxA4vKrm6ktEdB8ySTLtp8GGDRswevRorFixAsHBwVi0aBE2bdqE2NhYuLm5lSo/atQodO/eHd26dYOVlRUWLFiAn376CadPn4a3t/d996fRaKBWq5GZmQkHB4eaeEvVI2qeaGlWKIExW4FGzYBP2orQ+fxPIjxf3CUuyzr6Aq+cqNz2z28XYVKhAqYcARx9xHJJAtIuAEpbQH3/41lC2gXRqqgoZ/ZOg0G0pN26Ilq6ih4JxwHtHTPgNO8D9HoTaNy59DaK3rPKAYg4C6jsyt9XymnxoX1ig+haAwA2jcQ+1T7A2D+K33ehDlj/HHBxJ2BpK8oX5onXfB4WU0TmpIrwP+K74nEGyaeBVWGiJdQzAEg6JU6ilPZAn7eBLuNKj0ko1AFfPAqknAE6DBc/43/XinnxJx4Q4xkqKmYlsPU1QCYXxz79EuAZCLywvXLbqW656eK4N+0NuLWqvu1qs4CN4cClKCBwFDBoSfm/b0WyU4A1A4G0290WVGpxIlx0Bepu16LF70Jeuvg9eW4D4N62dDlNIvBJG/G7MvmQGINSUV8/CVzZC4S8DHR/Bfjpv+I9AeKE+akVlRvLYg6OfC0Gu6ubAP/7V/xcCvKBpV3ECXHvmcCj02pm3z9PBo59CzQJEX/XRdM7bp4gfg+DXgQGLiy5jsEA/DJJtLBbWIn/q77div+mlHbApH+K/0dUhjZLdEFU2jz4e6stkiT+x9mV/tw1F3Xm85uoGpi8BX3hwoUYP348xo4dizZt2mDFihWwsbHB6tWryyy/bt06TJo0CYGBgWjVqhW+/PJLGAwGREVF1XLNa1jvt4BWA2+36j4H7H5HhHOP9iL0ACI4VqUfur4Q2Hm79fzhiSU/gGQyETQqG84BMSvNvcKSXA4EjgR6vwkM+Bh4Zg0Q/hvwWiwwdOXtwaXS7RD+lAhWdytqPQ8cVX44L9qXR3sRdCYfEqEHMhHO7dyB0b+UfN8WSmD4WlGHghwRzht3FR/aL2wDRq4XH+LntwHb3xTrZCWJri+6LNHl4cVdwH/3At5BYtkf04AvQ8Ug3DsdWCzCuU0jICwS6PuOqNPNi6JbUUVdPQBsmy6eh84V78naGUg8BvzxesW3cy/6QjHuYfd74mSkIuVjVgJLOoq6rXxM9LWuDkVBuyjIHlsHbAq/d7eV7FTg60EinDt4i5+pNhNYO0RcUblTQZ7odvXNkyKce3US40PKCucA4OAJtAgTz/+9a7Dotb+BHW+XfSUl8YQI5zIFEDxR3Mdg1A/A4/MAuYXow31gUQUOiBmRJHGFDwCC/1v8f8DSCgidLZ7vXyT+Zqrb5b3idwEAHn+n5NzrASPF11M/iJOFIgaD+P08/r34OTzzdfEVkqAXAZ9gQJcNbImo/BWN9Mvi9/+zh0VQL4tBD3w3HFje4/5X26ri1rWKj0nR5YorN8u7Ax+1eLDudkRUbUwa0HU6HY4cOYLQ0FDjMrlcjtDQUERHR1doG7m5uSgoKICzs3NNVdM05HLR9cK9HZCTIrpAAEC3qcUfQCq7qvVDP/6duKRu7QT0iKjeeleFpbUYLDvmd2DKURGs8zOLg3CR9CvFfby7jKv49l2aA8NWim4wvd4ULWyNmpVdj5HrRTeZ//wIvLhDXKmQycQ4gac+F+UOrgD2fyI+YDXXRcv38LUi5Hu0F2MJBiwULbUJR0XXin0fisvpqbGiew8guujYNhI/hyc+EssOLBKt8PeTeR3YOFqMV2j3tJgNyNEHeHoVAJmYXaRohhGDAYg7KE7Kvh4kQu6djx1vATlpZe/j69uv7/sAWN4N+CwE2PeRCCF399G9sk+8162vAfkZ4ipHQQ6wfqQIvg+iqMtJ4jFxYhM6V1x5OPc7sO6Zsrse5aSJsJ16DrD3EieDz28WJ7b5maKrU+JxEWT++lgMMt72hjgpbj1IdJmyd793vYoGix5fL66M5KaL1tyv+gN/LxFXqe7u2hG9VHxt+1TxSaJcLsZ4PHl7EPie+aJudcWlKHGclXZAp+dLvtZumDhpLcgRv3/bZ4oB2WX9zlVG2kVgw3/EzxiS6NLi06VkGf9HxIlZfqYYSwCIn9Pm8UDM7b/nwctK3jxNLr99ZUYJXNgBnPqx5DbLGrxfRJstxr3kpIqGk/2flF3u37XiZD/5pBjnkJteqbd+T3H/iIkGlgTee0xKRpw4iVzYGvhtqrjiCIg6n9j0YHWQJHHCS0RVZtIuLgkJCfD29sbff/+NkJAQ4/Jp06Zh7969OHjw4H23MWnSJGzfvh2nT5+GlVXpS/parRZabXELm0ajgY+PT925RJYRB3zRG8hNu33p+CigsCx+fedsEeoC/wMMWXb/7elyxD/vrEQgbD4QMrnGql5lCf+KllfJIIJy89sncDveEq25zfqIoGUK+z8p2cJk7QyMjxL92e+mSQB+f1V8EAOiC4xMLt5fi77AcxtLtvatHyUCp1cn0WpfqBUfcgW5oiXO2C0oXbTYJp0UJwQv7Ch5KX3fR+KKi0IlwtHFXeIk716U9kCPV8Q4BKUNEPsH8PNLYvCt0l60Ll7eI8LrnSysxImNhZX4nQLECcdjb4nfyT9eLz5R6DpBXDEoal0t1ImButbO974aEn9I9P/PvQk4+QH/2SxOsC7vFdMfFuSIAPjsN6LVMzNe9EmO+UIM/LPzEGHbpbnYXr5G3In3+iHASi1ONHS3WzrVPiIoB70ogtr96AtF17PsJKBTuPj55d6+6Zilrahbu6eBYV+Kn3XmDWDx7XsfTPhT3IX3TpIkTrzO/gq4thJl7rwTpiSJmZ7i/xF/B81DAatq+j+mSRBXfRRKccJoU4lGj7VPAZd2i9+ffpGlX79+WITzgrtOVtzaiP9DReNTKiI7Bfjz/dv99fXibypwFND3XcDasXT5ou6CLcLEz2HDf8QVDLkFMPgzIGB42fv5cwHw53xxou3SXPzd5aaLKzBNugFDvyh5Fe7On53STvwuKlTAy4dK3gsjLwP4tJP4PVEobw+q7SKugClty66LNluMM7m0R/zs2w4Fuv+vdLlCrZgZKu387QUyoNd04JFpxb/POWnA3g9EH3vD7el1nfyALuPF33D0UvH3/OIO8T/rTilngcNfiVm5Wg0QV03vdPOSGDB/cpMYIzVoUdnvp4rYxYUakjod0N9//3188MEH+PPPP9GhQ4cyy8yZMwdz584ttbxO/YHHHxKXY3u8CrQeWPK1C7uAdZXoh77vQzF9oqOv+OCwUNVMnR/Uthnikrmjr+gHCoiWnvwMMZDrzhav2iRJYoaHf9eKD9fRvwK+Ifcuf2KjCD75GWJZeX1bNYnAsuA7+uPfh7WzCHB33wjLYBDdoopaDAERMlo8LoKQ5R1hviBPXBFIuv27Y+8pPlhP3m5B8wwEnvlKnIDkZQBnfxOvXdkH4K5/HTI5EPSC6GtcFO4kSbQkF3Wp8gwA5JaA5sbt7g4SAJn4oPcMEA97TxEEkk6IVuSiO3d6BoiuIHf2kb1+RPz+590q+xjZuolwfnf/8PxMESqLBg+6tRHBvN2wkifAFbFrLrD/jv7Nbm2AgYtE+PnmSfH18XdEoNrxtjgefj3FFaOy5NwEloeI931n4NXliBO+O2dEUShFK3GrAeJmaZbW4udbFOo1CeJKSGa8eO7SouzuYed3AD9PLD65cPIXA2PdWpf/vrOSRGvttb9Fa7RMLvqeO/mVXV6TIOZEj4sW66Xe0fUr+CXRFebOk5Hy9rmiZ/EJ50P9xBWve9Uz7YK434FMAbi2FN3LlHbihK55n/LXu3OsSFmsnUXgL9pG0Ymx3BIYu1U8v7IPaDMEePbr4vX+mC4Gnru0FOt/PUj8b2geCoz4XlyJKzpepzaLE/y4fwDDXdPGPv0V0G5oyWW73xNXu2zdxP/IopPj5o8DgxaLq6f7FxefkPo/In7HWvQVYx6Kut5c3ClOVif8KQb7GgziZ7xz9u1JBG5zeUj87tk0Elca7rxxl70X8Orpip3oVhADOjUkJg3oOp0ONjY2+OGHHzBkyBDj8vDwcGRkZOCXX34pd92PPvoI7777Lnbt2oWgoKByy9X5FvT70WYB7/uKlqRXTopZWMqTnSr6RuqygGGrgPZP1149K0ubLcKq5ro4MXFuKoKxo68IAaYcQKcvEF2OPAOBJsEVWycrWfRnjf1DfFDe3Q2gyPH1YsCgkUwELqWt+KC0dhbh185NtPK6tyl7O3kZ4qROaSs+QH17FH/w381gEB+uu+eVHMvw8CQRfso6iSvIF4GxIKe4ld/WrfyxC2d+EQP2CvNLLpdblg4ed5PJxXiMIZ+VfaOs5DNimryMa+JERN1Y1MPJD3j4pbKvbgDiGMWsBDw7iIBy59WMyki/Irr/SBLQ6w0x8LMo5N85iPeZNcAvLwNazf1PMi/sBNbd/vsc/YsISxueF90QZArxt3v9sBgUXFlWjuJqRvB/RTekqLnF3W6KupdlxIkQO3Ql0Or2DE0GgwjXp34QLbm3rpTcbocRwNDPK16P3HQx1WjRzC+urcT+PMtubIEkiZ/z+W2iW9mgxWXfc6EsX4aKKyYAYHu7z79X4P3Xy0oSVwas1CKE2jQSf/8/vyS6W0EmTkg92omrOZBE95jO4aKr2uc9xZXAsX+Iq1ApZ0Vfb0lfPNg/PkZ0tyrIFSeI/o+IVuir+1HiJNixiShfkA+cWC/+L7y4o3jqz+TToouZoVD0qW87BDj2PfD7K6X/7jwDxJiHpr1Kv+e8DHEFM/2SOJEc8pmYq//yHvG6/6Pi6sOVfaX/dmUK0QjQ/hkxs1d1Xd25jQGdGhKTz+ISHByMrl274tNPRd9Lg8GAJk2a4OWXX8b06dPLXOeDDz7Ae++9h+3bt+Phhx+u1P7q5R/4yj7AjcNiesbAkeWX++Vl0fLr1REYt7taWzZqxLmtov+y3EJcUs2IK26JrKsKtfe/apF3S4SRoq4jVQ2OVanboS/FSUTIZKBl/+rdftoFMRWlnfvtEO0jTjpyUsXAycRjosU8K0nM/OLRQZwEube9/2wYBoPoUlDNgaDCMuJEYLJ1KblckoBfXxbdUoo0aiHGQ9zv7+/3V8WJoJ27OAnSakS4fGaNCKaSJLoynPtdXEnLSS0+WSrIEyHQwUscZ3Vjse65LcWh3sJaXMUp6g7R9b9iwLI2Wwy+vfoXABnQM0J0wzi1WVz5MJKJMTJNHhaPNkPuP6NOWc7vEHdQzkkRJ2x93xEnVnf7d52YdUWhBCbsLf/ktCz/fiv24dxUdJFy9q98Pe9UkC+uih0tahmXAZDEFaSBd/Q7/+0VcXdTzwBg/J9icPKVvUDLAcDI74rLXdglunEVdTkp4vOwaCVvHirqLpOJVu51T4sTB0df0cptpQZW9RWfA60GAsO/Lf6/kXRSdOu5dVWE/MdmiROBe/3+pZwDvuwj/qbkFqJeFtbiZ9NlnNh2fqY4kYzdKv5ntXxCjKu4+2+gGtXLz2+icpg8oG/YsAHh4eH4/PPP0bVrVyxatAgbN27EuXPn4O7ujtGjR8Pb2xuRkeIy74IFCzBr1ix899136N69u3E7dnZ2sLO7Rz/W2+rlH3hF+qHfOCpaRSCJPssVbfk1tQ3/Ed0qABFWI85Wrm8skakV5ANrnijuTjNoMdB5zP3X0+WI/sTpl8X3Pg+LcO7gWfW6GPTi7+nAouLuCFaOYqDknd3n9AXiCkxR63YRlVoMoG3zpAjlVuqq1+VOOWlioOK5291+ur8irt4UhczM62KAslYjlvd4tXLblyQRjL06Vl+dARH8t/yfaKH2eVgMRL7zSlV2quhvrtWIE5gzP4t+6ZMPlj5JOPkD8PMk0Q2p/dMiRJd3RTQ3Xcybf+uqaAVvHirG6KgcxLYdvEqWz9eI+0/49ax4t8azv4sb1wHiuA1dWbrPeS2rl5/fROUweUAHgKVLl+LDDz9EUlISAgMDsWTJEgQHiwDZq1cv+Pn5Yc2aNQAAPz8/XLtW+kYss2fPxpw5c+67r3r5B36/fugGg7gZ0PUYMe/20C9qv45VpUkAlnYV3XI6Pg8MXmrqGhFVniZBzJevsAReOnD/vtZFEv4Vs8I0fwzoM7vy/ePLUxRYr/wlThbKm+v78FdigKV3ZxEamz9ec/Pr3z1eofMYMRuSTC7GC1zeIwZTvrDdvOaITzolrjoFvSBmZbrbgSXAzreLv+/5mrhHQlkMhopf2Uw+DXz5uOhmVmTgIiBobIWrfl9nfhVjITqPqb7fvQdQLz+/icphFgG9NtXLP/D79UMv6tdsaStuSvQgLXCmcPY30Zf3yU9LD4gkqisKtSJsmkHQMWtHvhat6ZDEbCVNHhbdSSysgIn7Td6KW2mFOuCzYHElxN4LmHK4/NlaKuvML2LmGADw7Q6E/27+XRcfQL38/CYqR/39S25IVPbF07XdPR+6Nqu4RerR1+teOAfEJfXwXxnOqW6zUDGcV0TncDFzkNwSOL1ZhHNAXEGoa+EcEF1enlwqBnMOXlp94RwQc7/3fVdMzTp4ab0O50QNTRVG9JBZ8ushBggdXCFm+GjaW/yz3vehuETp3FTMykFEZO7aPiXm39/wH3FXX9/u4q6rdZVfd9H6XxO6TREPIqpXGNDri5b9xS3kE4+Jm7Com4jR/9G3b78dFmm+c54TEd2tRaiYT/zUjyKAsnWYiBoQ9kGvT5JOiv6bJzeKKbCKNA8V8/7W1nR9RERE1axef34T3YUBvT4qyBMDK49+I7q3jFwvbo1ORERURzWIz2+i29jFpT6ytAY6PCseRERERFSnsFMfEREREZEZYUAnIiIiIjIjDOhERERERGaEAZ2IiIiIyIwwoBMRERERmREGdCIiIiIiM8KATkRERERkRhjQiYiIiIjMCAM6EREREZEZYUAnIiIiIjIjDOhERERERGaEAZ2IiIiIyIwwoBMRERERmREGdCIiIiIiM2Jh6grUNkmSAAAajcbENSEiIqKKKvrcLvocJ6rPGlxAz8rKAgD4+PiYuCZERERUWVlZWVCr1aauBlGNkkkN7FTUYDAgISEB9vb2kMlk1bptjUYDHx8fxMfHw8HBoVq3TSXxWNceHuvaw2Nde3isa091HWtJkpCVlQUvLy/I5eyhS/Vbg2tBl8vlaNy4cY3uw8HBgf/wawmPde3hsa49PNa1h8e69lTHsWbLOTUUPAUlIiIiIjIjDOhERERERGaEAb0aqVQqzJ49GyqVytRVqfd4rGsPj3Xt4bGuPTzWtYfHmqjyGtwgUSIiIiIic8YWdCIiIiIiM8KATkRERERkRhjQiYiIiIjMCAM6EREREZEZYUCvJsuWLYOfnx+srKwQHByMmJgYU1epzouMjESXLl1gb28PNzc3DBkyBLGxsSXK5OfnY/LkyWjUqBHs7OwwbNgwJCcnm6jG9cf7778PmUyGV155xbiMx7r63LhxA//5z3/QqFEjWFtbo3379jh8+LDxdUmSMGvWLHh6esLa2hqhoaG4cOGCCWtcN+n1erz99tvw9/eHtbU1mjVrhnfeeQd3zo3AY111+/btw6BBg+Dl5QWZTIaff/65xOsVObbp6ekYNWoUHBwc4OjoiBdffBHZ2dm1+C6IzBMDejXYsGEDIiIiMHv2bBw9ehQBAQEICwtDSkqKqatWp+3duxeTJ0/GP//8g507d6KgoAB9+/ZFTk6Oscyrr76K3377DZs2bcLevXuRkJCAoUOHmrDWdd+hQ4fw+eefo0OHDiWW81hXj1u3bqF79+6wtLTEH3/8gTNnzuDjjz+Gk5OTscwHH3yAJUuWYMWKFTh48CBsbW0RFhaG/Px8E9a87lmwYAGWL1+OpUuX4uzZs1iwYAE++OADfPrpp8YyPNZVl5OTg4CAACxbtqzM1ytybEeNGoXTp09j586d+P3337Fv3z5MmDChtt4CkfmS6IF17dpVmjx5svF7vV4veXl5SZGRkSasVf2TkpIiAZD27t0rSZIkZWRkSJaWltKmTZuMZc6ePSsBkKKjo01VzTotKytLatGihbRz507p0UcflaZOnSpJEo91dXrjjTekHj16lPu6wWCQPDw8pA8//NC4LCMjQ1KpVNL3339fG1WsNwYMGCC98MILJZYNHTpUGjVqlCRJPNbVCYD0008/Gb+vyLE9c+aMBEA6dOiQscwff/whyWQy6caNG7VWdyJzxBb0B6TT6XDkyBGEhoYal8nlcoSGhiI6OtqENat/MjMzAQDOzs4AgCNHjqCgoKDEsW/VqhWaNGnCY19FkydPxoABA0ocU4DHujr9+uuvCAoKwjPPPAM3Nzd07NgRK1euNL5+5coVJCUllTjWarUawcHBPNaV1K1bN0RFReH8+fMAgOPHj2P//v3o378/AB7rmlSRYxsdHQ1HR0cEBQUZy4SGhkIul+PgwYO1Xmcic2Jh6grUdWlpadDr9XB3dy+x3N3dHefOnTNRreofg8GAV155Bd27d0e7du0AAElJSVAqlXB0dCxR1t3dHUlJSSaoZd22fv16HD16FIcOHSr1Go919bl8+TKWL1+OiIgIvPnmmzh06BD+97//QalUIjw83Hg8y/qfwmNdOdOnT4dGo0GrVq2gUCig1+vx3nvvYdSoUQDAY12DKnJsk5KS4ObmVuJ1CwsLODs78/hTg8eATnXC5MmTcerUKezfv9/UVamX4uPjMXXqVOzcuRNWVlamrk69ZjAYEBQUhPnz5wMAOnbsiFOnTmHFihUIDw83ce3ql40bN2LdunX47rvv0LZtWxw7dgyvvPIKvLy8eKyJyKyxi8sDcnFxgUKhKDWbRXJyMjw8PExUq/rl5Zdfxu+//449e/agcePGxuUeHh7Q6XTIyMgoUZ7HvvKOHDmClJQUdOrUCRYWFrCwsMDevXuxZMkSWFhYwN3dnce6mnh6eqJNmzYllrVu3RpxcXEAYDye/J/y4F5//XVMnz4dI0aMQPv27fH888/j1VdfRWRkJAAe65pUkWPr4eFRajKFwsJCpKen8/hTg8eA/oCUSiU6d+6MqKgo4zKDwYCoqCiEhISYsGZ1nyRJePnll/HTTz9h9+7d8Pf3L/F6586dYWlpWeLYx8bGIi4ujse+kvr06YOTJ0/i2LFjxkdQUBBGjRplfM5jXT26d+9earrQ8+fPw9fXFwDg7+8PDw+PEsdao9Hg4MGDPNaVlJubC7m85MecQqGAwWAAwGNdkypybENCQpCRkYEjR44Yy+zevRsGgwHBwcG1Xmcis2LqUar1wfr16yWVSiWtWbNGOnPmjDRhwgTJ0dFRSkpKMnXV6rSXXnpJUqvV0p9//iklJiYaH7m5ucYyEydOlJo0aSLt3r1bOnz4sBQSEiKFhISYsNb1x52zuEgSj3V1iYmJkSwsLKT33ntPunDhgrRu3TrJxsZG+vbbb41l3n//fcnR0VH65ZdfpBMnTkiDBw+W/P39pby8PBPWvO4JDw+XvL29pd9//126cuWKtHnzZsnFxUWaNm2asQyPddVlZWVJ//77r/Tvv/9KAKSFCxdK//77r3Tt2jVJkip2bPv16yd17NhROnjwoLR//36pRYsW0siRI031lojMBgN6Nfn000+lJk2aSEqlUuratav0zz//mLpKdR6AMh9fffWVsUxeXp40adIkycnJSbKxsZGeeuopKTEx0XSVrkfuDug81tXnt99+k9q1ayepVCqpVatW0hdffFHidYPBIL399tuSu7u7pFKppD59+kixsbEmqm3dpdFopKlTp0pNmjSRrKyspKZNm0ozZ86UtFqtsQyPddXt2bOnzP/R4eHhkiRV7NjevHlTGjlypGRnZyc5ODhIY8eOlbKyskzwbojMi0yS7rilGhERERERmRT7oBMRERERmREGdCIiIiIiM8KATkRERERkRhjQiYiIiIjMCAM6EREREZEZYUAnIiIiIjIjDOhERERERGaEAZ2IzMbUqVMxYcIE463YiYiIGiIGdCIyC/Hx8WjZsiU+//xzyOX810RERA0X7yRKRERERGRG2ExFRCY1ZswYyGSyUo9+/fqZumpEREQmYWHqChAR9evXD1999VWJZSqVykS1ISIiMi22oBORyalUKnh4eJR4ODk5AQBkMhmWL1+O/v37w9raGk2bNsUPP/xQYv2TJ0/iscceg7W1NRo1aoQJEyYgOzu7RJnVq1ejbdu2UKlU8PT0xMsvv2x8beHChWjfvj1sbW3h4+ODSZMmlVqfiIiotjCgE5HZe/vttzFs2DAcP34co0aNwogRI3D27FkAQE5ODsLCwuDk5IRDhw5h06ZN2LVrV4kAvnz5ckyePBkTJkzAyZMn8euvv6J58+bG1+VyOZYsWYLTp0/j66+/xu7duzFt2rRaf59EREQAB4kSkYmNGTMG3377LaysrEosf/PNN/Hmm29CJpNh4sSJWL58ufG1hx9+GJ06dcJnn32GlStX4o033kB8fDxsbW0BAFu3bsWgQYOQkJAAd3d3eHt7Y+zYsXj33XcrVKcffvgBEydORFpaWvW9USIiogpiH3QiMrnevXuXCOAA4OzsbHweEhJS4rWQkBAcO3YMAHD27FkEBAQYwzkAdO/eHQaDAbGxsZDJZEhISECfPn3K3f+uXbsQGRmJc+fOQaPRoLCwEPn5+cjNzYWNjU01vEMiIqKKYxcXIjI5W1tbNG/evMTjzoD+IKytre/5+tWrVzFw4EB06NABP/74I44cOYJly5YBAHQ6XbXUgYiIqDIY0InI7P3zzz+lvm/dujUAoHXr1jh+/DhycnKMrx84cAByuRwtW7aEvb09/Pz8EBUVVea2jxw5AoPBgI8//hgPP/wwHnroISQkJNTcmyEiIroPdnEhIpPTarVISkoqsczCwgIuLi4AgE2bNiEoKAg9evTAunXrEBMTg1WrVgEARo0ahdmzZyM8PBxz5sxBamoqpkyZgueffx7u7u4AgDlz5mDixIlwc3ND//79kZWVhQMHDmDKlClo3rw5CgoK8Omnn2LQoEE4cOAAVqxYUbsHgIiI6A5sQScik9u2bRs8PT1LPHr06GF8fe7cuVi/fj06dOiAb775Bt9//z3atGkDALCxscH27duRnp6OLl264Omnn0afPn2wdOlS4/rh4eFYtGgRPvvsM7Rt2xYDBw7EhQsXAAABAQFYuHAhFixYgHbt2mHdunWIjIys3QNARER0B87iQkRmTSaT4aeffsKQIUNMXRUiIqJawRZ0IiIiIiIzwoBORERERGRGOEiUiMwae+EREVFDwxZ0IiIiIiIzwoBORERERGRGGNCJiIiIiMwIAzoRERERkRlhQCciIiIiMiMM6EREREREZoQBnYiIiIjIjDCgExERERGZEQZ0IiIiIiIz8v8LGgMpZHUNdwAAAABJRU5ErkJggg==",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(range(100), train_accs, label='train_accs')\n",
    "plt.plot(range(100), test_accs, label='test_accs')\n",
    "plt.title('Curva de Accs')\n",
    "plt.xlabel('Época')\n",
    "plt.ylabel('Loss')  \n",
    "plt.legend(bbox_to_anchor=(1.1, 1.05))\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(range(100), test_losses , label='test_losses')\n",
    "plt.plot(range(100), train_losses , label='train_losses')\n",
    "plt.title('Curva de Loss')\n",
    "plt.xlabel('Época')\n",
    "plt.ylabel('Loss')   \n",
    "plt.legend(bbox_to_anchor=(1.1, 1.05))\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### **11.5 - Aplicação do Modelo nos Dados Validados para Score**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 3/3 [00:00<00:00, 74.72it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "71.875%\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    }
   ],
   "source": [
    "with torch.no_grad():\n",
    "    listas = []\n",
    "    for batch, (X_val, y_val) in tqdm(enumerate(val_loader), total=len(val_loader)):\n",
    "        num_correct_train = 0\n",
    "        num_samples_train = 0\n",
    "        X_val = X_val.float().to(device)\n",
    "        y_val = y_val.float().to(device)\n",
    "        val_preds = model(X_val)\n",
    "        rounded_validation_preds = torch.round(val_preds)\n",
    "        num_correct_train += torch.sum(rounded_validation_preds == y_train)\n",
    "        num_samples_train += len(y_train)\n",
    "\n",
    "    acc = num_correct_train/num_samples_train\n",
    "\n",
    "acc = acc.item()*100\n",
    "print(str(acc) + '%')\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### **12 - Aplicação do Modelo nos Dados Não Validados**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 44033/44033 [02:56<00:00, 249.67it/s]\n"
     ]
    }
   ],
   "source": [
    "todos = load_images(os.path.join(path_dir,'Todas/'), False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(44033, 100, 100, 3)\n"
     ]
    }
   ],
   "source": [
    "print(todos.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "class Validacao():\n",
    "    def __init__(self, images, transform=None):\n",
    "        self.images = images\n",
    "        self.transform = transform\n",
    "        \n",
    "    def __len__(self):\n",
    "        return len(self.images)\n",
    "    \n",
    "    def __getitem__(self, index):\n",
    "        image = self.images[index]\n",
    "        \n",
    "        if self.transform:\n",
    "            image = self.transform(image)\n",
    "        \n",
    "        return (image)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [],
   "source": [
    "transforms_todos = transforms.Compose([transforms.ToTensor(), \n",
    "                                       transforms.Normalize(mean=[0.5,0.5,0.5], std=[0.5,0.5,0.5]) # squeeze to -1 and 1\n",
    "                                      ])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [],
   "source": [
    "todos_dataset  = Validacao(images=todos, transform=transforms_todos)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [],
   "source": [
    "todos_loader  = DataLoader(todos_dataset, batch_size=32, shuffle=True, drop_last=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [],
   "source": [
    "with torch.no_grad():\n",
    "\n",
    "    n_pista = 0\n",
    "    n_nao_pista = 0\n",
    "    listas = []\n",
    "\n",
    "    for batch, (val_todo) in enumerate(todos_loader):\n",
    "        val_todo = val_todo.to(device)\n",
    "        validation_preds = model(val_todo)\n",
    "        rounded_validation_preds = torch.round(validation_preds)\n",
    "        list = [int(a[0]) for a in rounded_validation_preds]\n",
    "        listas.append(list)\n",
    "        n_nao_pista += list.count(1) \n",
    "        n_pista += list.count(0) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 1, 0, 0, 1, 1, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0]\n",
      "[1, 1, 1, 1, 0, 1, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 1, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]\n",
      "[0, 1, 1, 1, 1, 1, 1, 0, 1, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 1, 1, 1, 0, 1, 1, 1, 1, 1, 1, 1, 1]\n",
      "[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 1, 0, 1, 1, 1, 1, 0, 1, 1, 1, 1, 0, 1, 1, 1, 0]\n",
      "[1, 1, 1, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 1, 0]\n",
      "[0, 1, 1, 1, 1, 1, 1, 0, 1, 1, 0, 1, 0, 1, 0, 1, 1, 1, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 1, 1]\n",
      "[0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 1, 1, 1, 1, 1, 1]\n",
      "[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 1, 1, 1, 1, 1, 0, 1, 1, 1, 1, 1]\n",
      "[1, 1, 1, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 1, 1, 1, 1, 1, 1, 1, 1, 0, 1, 1, 1, 0, 1, 1, 1, 1, 0]\n",
      "[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 1, 1, 1, 1, 0, 1, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]\n"
     ]
    }
   ],
   "source": [
    "for lista in listas[:10]:\n",
    "    print(lista)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "6082"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "n_pista"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "37950"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "n_nao_pista"
   ]
  }
 ],
 "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": "3.12.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}