PizzaSorterSearcher.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Категория данных: Количество файлов\n",
      "------------------------------\n",
      "not_pizza: 983 файлов\n",
      "pizza: 983 файлов\n"
     ]
    }
   ],
   "source": [
    "import os\n",
    "from pathlib import Path\n",
    "\n",
    "DIR = os.path.join(os.getcwd(), 'pizza_not_pizza')\n",
    "\n",
    "subject = {\n",
    "    class_: len(os.listdir(os.path.join(DIR, class_)))\n",
    "    for class_ in (x.name for x in Path(DIR).iterdir() if x.is_dir())\n",
    "}\n",
    "\n",
    "print(\"Категория данных: Количество файлов\")\n",
    "print(\"-\" * 30)  \n",
    "for class_, count in subject.items():\n",
    "    print(f\"{class_}: {count} файлов\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "import torch\n",
    "from torch import nn\n",
    "import torch.nn.functional as F\n",
    "from torchvision import datasets, transforms\n",
    "from torchvision.models import efficientnet_b0\n",
    "from torch.utils.data import DataLoader, random_split\n",
    "import matplotlib.pyplot as plt\n",
    "from pathlib import Path\n",
    "from os import listdir\n",
    "import util  # Importing STUDY util module\n",
    "\n",
    "\n",
    "# Без нормализации 2.3.1\n",
    "transform = transforms.Compose([\n",
    "    transforms.Resize((256, 256)),\n",
    "    transforms.ToTensor(),\n",
    "])\n",
    "dataset = datasets.ImageFolder(root=DIR, transform=transform)\n",
    "\n",
    "# Разделение на тренировочный и валидационный наборы\n",
    "train_size = int(0.7 * len(dataset))  # 70% для тренировки\n",
    "val_size = len(dataset) - train_size\n",
    "train_dataset, val_dataset = random_split(dataset, [train_size, val_size])\n",
    "\n",
    "# Создание DataLoader\n",
    "train_loader = DataLoader(train_dataset, batch_size=32, shuffle=True)\n",
    "val_loader = DataLoader(val_dataset, batch_size=64)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "class ImprovedModel(nn.Module):\n",
    "    def __init__(self):\n",
    "        super(ImprovedModel, self).__init__()\n",
    "\n",
    "        self.conv_layers = nn.Sequential(\n",
    "            nn.Conv2d(3, 24, kernel_size=3, stride=1, padding=1),\n",
    "            nn.BatchNorm2d(24),\n",
    "            nn.ReLU(),\n",
    "            nn.MaxPool2d(kernel_size=2, stride=2),\n",
    "            \n",
    "            nn.Conv2d(24, 40, kernel_size=3, stride=2, padding=1),\n",
    "            nn.BatchNorm2d(40),\n",
    "            nn.ReLU(),\n",
    "            nn.MaxPool2d(kernel_size=2, stride=2),\n",
    "            \n",
    "            nn.Conv2d(40, 80, kernel_size=3, stride=2, padding=1),\n",
    "            nn.BatchNorm2d(80),\n",
    "            nn.ReLU(),\n",
    "            \n",
    "            nn.Conv2d(80, 160, kernel_size=3, stride=2, padding=1),\n",
    "            nn.BatchNorm2d(160),\n",
    "            nn.ReLU(),\n",
    "            \n",
    "            nn.Conv2d(160, 160, kernel_size=3, stride=2, padding=1),\n",
    "            nn.BatchNorm2d(160),\n",
    "            nn.ReLU(),\n",
    "            \n",
    "            nn.Conv2d(160, 320, kernel_size=3, stride=2, padding=1),\n",
    "            nn.BatchNorm2d(320),\n",
    "            nn.ReLU(),\n",
    "            \n",
    "            nn.Conv2d(320, 320, kernel_size=3, stride=1, padding=1),\n",
    "            nn.BatchNorm2d(320),\n",
    "            nn.ReLU(),\n",
    "            nn.AdaptiveAvgPool2d((1, 1))\n",
    "        )\n",
    "\n",
    "        self.fc_layers = nn.Sequential(\n",
    "            nn.Flatten(),\n",
    "            nn.Linear(320, 128),\n",
    "            nn.ReLU(),\n",
    "            nn.Linear(128, 42),\n",
    "            nn.ReLU(),\n",
    "            nn.Linear(42, 1),\n",
    "            nn.Sigmoid()  # Sigmoid for the final layer\n",
    "        )\n",
    "\n",
    "    def forward(self, x):\n",
    "        x = self.conv_layers(x)\n",
    "        x = self.fc_layers(x)\n",
    "        return x\n",
    "    \n",
    "#CNN 好\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "def train_and_evaluate_model(train_loader, val_loader, epo, rate=1e-5, mdl=None): # rate was taken empirically. This value is optimal in the base\n",
    "    print(f\"Train size:\\t\\t{len(train_loader.dataset)}\\n\"\n",
    "          f\"Validation size:\\t{len(val_loader.dataset)}\\n\")\n",
    "\n",
    "\n",
    "    model = mdl if mdl is not None else ImprovedModel()\n",
    "    loss_fn = torch.nn.BCELoss()\n",
    "    opt = torch.optim.Adam(model.parameters(), lr=rate)       \n",
    "\n",
    "    history = lab4_util.train(\n",
    "        model,\n",
    "        loss_fn,\n",
    "        opt,\n",
    "        train_loader,\n",
    "        val_loader,\n",
    "        max_epochs=epo\n",
    "    )\n",
    "\n",
    "    return lab4_util.accuracy(model, val_loader), history\n",
    "\n",
    "def outputs(model_type, text, plot):\n",
    "    print(f\"{model_type} Accuracy: {text:.3f}\\n\")\n",
    "    lab4_util.plot_history(plot, \"loss\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train size:\t\t1376\n",
      "Validation size:\t590\n",
      "\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "  0%|          | 0/43 [00:00<?, ?it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:07<00:00,  6.05it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 0: train loss 0.69316\n",
      "Epoch 0: val loss 0.69618\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:08<00:00,  4.90it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1: train loss 0.68432\n",
      "Epoch 1: val loss 0.68872\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:08<00:00,  4.92it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 2: train loss 0.67407\n",
      "Epoch 2: val loss 0.67931\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:08<00:00,  5.03it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 3: train loss 0.65855\n",
      "Epoch 3: val loss 0.66453\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:08<00:00,  4.99it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 4: train loss 0.63674\n",
      "Epoch 4: val loss 0.64455\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:08<00:00,  4.89it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 5: train loss 0.60745\n",
      "Epoch 5: val loss 0.61818\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:08<00:00,  4.84it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 6: train loss 0.57270\n",
      "Epoch 6: val loss 0.59074\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:08<00:00,  4.97it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 7: train loss 0.53603\n",
      "Epoch 7: val loss 0.56653\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:08<00:00,  4.91it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 8: train loss 0.49743\n",
      "Epoch 8: val loss 0.54459\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:08<00:00,  4.86it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 9: train loss 0.45412\n",
      "Epoch 9: val loss 0.52557\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:08<00:00,  5.04it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 10: train loss 0.41293\n",
      "Epoch 10: val loss 0.51104\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:08<00:00,  4.99it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 11: train loss 0.37143\n",
      "Epoch 11: val loss 0.49959\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:08<00:00,  5.17it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 12: train loss 0.33210\n",
      "Epoch 12: val loss 0.49021\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:08<00:00,  5.09it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 13: train loss 0.28941\n",
      "Epoch 13: val loss 0.48246\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:08<00:00,  4.95it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 14: train loss 0.24131\n",
      "Epoch 14: val loss 0.48318\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:08<00:00,  4.99it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 15: train loss 0.21469\n",
      "Epoch 15: val loss 0.49782\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:08<00:00,  5.06it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 16: train loss 0.18227\n",
      "Epoch 16: val loss 0.48619\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:08<00:00,  5.05it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 17: train loss 0.15478\n",
      "Epoch 17: val loss 0.48187\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:08<00:00,  5.04it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 18: train loss 0.12832\n",
      "Epoch 18: val loss 0.49249\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:08<00:00,  4.96it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 19: train loss 0.10261\n",
      "Epoch 19: val loss 0.48045\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:08<00:00,  4.98it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 20: train loss 0.08994\n",
      "Epoch 20: val loss 0.48233\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:08<00:00,  4.93it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 21: train loss 0.08278\n",
      "Epoch 21: val loss 0.48173\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:08<00:00,  4.96it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 22: train loss 0.07349\n",
      "Epoch 22: val loss 0.49590\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:08<00:00,  4.95it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 23: train loss 0.06201\n",
      "Epoch 23: val loss 0.50875\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:08<00:00,  4.85it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 24: train loss 0.05514\n",
      "Epoch 24: val loss 0.49632\n"
     ]
    }
   ],
   "source": [
    "accuracy, history = train_and_evaluate_model(train_loader, val_loader, epo=25)  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Standart Model Accuracy: 0.760\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "outputs(\"Standart Model\", accuracy, history)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Mean: [0.55433035 0.4268966  0.3238646 ]\n",
      "Std: [0.26205462 0.26139945 0.2647449 ]\n"
     ]
    }
   ],
   "source": [
    "# Нашли среднее значение и стадартное отклонение\n",
    "import numpy as np\n",
    "import torchvision\n",
    "\n",
    "def collate_fn(batch):\n",
    "    return tuple(zip(*batch))\n",
    "\n",
    "\n",
    "def calculate_mean_std(loader):\n",
    "    channels_sum, channels_squared_sum, num_batches = 0, 0, 0\n",
    "\n",
    "    for batch in loader:\n",
    "        for image in batch[0]:\n",
    "            image = np.array(image, dtype=np.float32) / 255\n",
    "            channels_sum += np.mean(image, axis=(0, 1))\n",
    "            channels_squared_sum += np.mean(image**2, axis=(0, 1))\n",
    "            num_batches += 1\n",
    "\n",
    "    mean = channels_sum / num_batches\n",
    "    std = np.sqrt(channels_squared_sum / num_batches - mean ** 2)\n",
    "\n",
    "    return mean, std\n",
    "\n",
    "preprocess = torchvision.transforms.Resize((256, 256))\n",
    "pre_dataset = torchvision.datasets.ImageFolder(root=DIR, transform=preprocess)\n",
    "pre_v_dataset = torchvision.datasets.ImageFolder(root=DIR, transform=preprocess)\n",
    "\n",
    "pre_combined_dataset = torch.utils.data.ConcatDataset([pre_dataset, pre_v_dataset])\n",
    "pre_combined_loader = torch.utils.data.DataLoader(pre_combined_dataset, batch_size=32, shuffle=False, collate_fn=collate_fn)\n",
    "\n",
    "mean, std = calculate_mean_std(pre_combined_loader)\n",
    "print(f'Mean: {mean}\\nStd: {std}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "# C нормализацией \n",
    "dataset1 = datasets.ImageFolder(DIR, transform=transforms.Compose([\n",
    "                                transforms.Resize((256, 256)), \n",
    "                                transforms.ToTensor(),\n",
    "                                transforms.Normalize(mean=mean, std=std), ## Normalize the input values\n",
    "                            ]))\n",
    "\n",
    "# Разделение на тренировочный и валидационный наборы\n",
    "train_size1 = int(0.7 * len(dataset1))  # 70% для тренировки\n",
    "val_size1 = len(dataset1) - train_size1\n",
    "train_dataset1, val_dataset1 = random_split(dataset1, [train_size1, val_size1])\n",
    "\n",
    "\n",
    "train_loader1 = DataLoader(train_dataset1, batch_size=32, shuffle=True)\n",
    "val_loader1 = DataLoader(val_dataset1, batch_size=64)"
   ]
  },
  {
   "attachments": {
    "image.png": {
     "image/png": "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"
    }
   },
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Я бы с радостью сделал бы так \n",
    "```py\n",
    "dataset1 = datasets.ImageFolder(DIR, transform=transforms.Compose([\n",
    "                                transforms.Resize((512, 512)),\n",
    "                                transforms.Normalize(mean=mean, std=std), ## Normalize the input values\n",
    "                                transforms.ToTensor()\n",
    "                            ]))\n",
    "```\n",
    "\n",
    "но вылезает ругань на lab4_util.py которую пофиксить, мне увы, не получилось. Разумно.\n",
    "\n",
    "![image.png](attachment:image.png)\n",
    "\n",
    "Поэтому делаю нормализацию после преобразования ToTensor"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train size:\t\t1376\n",
      "Validation size:\t590\n",
      "\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:06<00:00,  6.98it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 0: train loss 0.69146\n",
      "Epoch 0: val loss 0.69346\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:06<00:00,  6.65it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1: train loss 0.68318\n",
      "Epoch 1: val loss 0.68458\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:06<00:00,  6.66it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 2: train loss 0.67246\n",
      "Epoch 2: val loss 0.67292\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:06<00:00,  6.60it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 3: train loss 0.65616\n",
      "Epoch 3: val loss 0.65684\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:06<00:00,  6.61it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 4: train loss 0.63538\n",
      "Epoch 4: val loss 0.63685\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:06<00:00,  6.59it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 5: train loss 0.61176\n",
      "Epoch 5: val loss 0.61238\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:06<00:00,  6.57it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 6: train loss 0.57810\n",
      "Epoch 6: val loss 0.58769\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:06<00:00,  6.62it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 7: train loss 0.54265\n",
      "Epoch 7: val loss 0.55653\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:06<00:00,  6.59it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 8: train loss 0.50454\n",
      "Epoch 8: val loss 0.53167\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:06<00:00,  6.71it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 9: train loss 0.47481\n",
      "Epoch 9: val loss 0.52176\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:06<00:00,  6.61it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 10: train loss 0.43582\n",
      "Epoch 10: val loss 0.49989\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:06<00:00,  6.46it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 11: train loss 0.40518\n",
      "Epoch 11: val loss 0.48986\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:06<00:00,  6.34it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 12: train loss 0.36545\n",
      "Epoch 12: val loss 0.47905\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:06<00:00,  6.56it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 13: train loss 0.32273\n",
      "Epoch 13: val loss 0.46967\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:06<00:00,  6.51it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 14: train loss 0.28145\n",
      "Epoch 14: val loss 0.46129\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:06<00:00,  6.46it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 15: train loss 0.24522\n",
      "Epoch 15: val loss 0.47267\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:06<00:00,  6.49it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 16: train loss 0.21457\n",
      "Epoch 16: val loss 0.45553\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:06<00:00,  6.54it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 17: train loss 0.18103\n",
      "Epoch 17: val loss 0.44687\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:06<00:00,  6.70it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 18: train loss 0.15736\n",
      "Epoch 18: val loss 0.45371\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:06<00:00,  6.68it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 19: train loss 0.13663\n",
      "Epoch 19: val loss 0.47456\n"
     ]
    }
   ],
   "source": [
    "acc_std_norm, hist__std_norm = train_and_evaluate_model(train_loader1, val_loader1, epo=20)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Standart Normalizeted Model Accuracy: 0.775\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "outputs(\"Standart Normalizeted Model\", acc_std_norm, hist__std_norm)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'torch.Tensor'> <class 'torch.Tensor'>\n"
     ]
    }
   ],
   "source": [
    "import warnings\n",
    "\n",
    "warnings.filterwarnings('ignore')\n",
    "data_iter = iter(train_loader)\n",
    "images, labels = next(data_iter)\n",
    "print(type(images), type(labels)) ## Доказательсвто, что я пытался отладить прикол с нормализацией входных значений"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [],
   "source": [
    "# C нормализацией и разные скорости обучения с оптимальным epoch count\n",
    "dataset2 = datasets.ImageFolder(DIR, transform=transforms.Compose([\n",
    "                                transforms.Resize((256, 256)), \n",
    "                                transforms.ToTensor(),\n",
    "                                transforms.Normalize(mean=mean, std=std), ## Normalize the input values\n",
    "                            ]))\n",
    "\n",
    "# Разделение на тренировочный и валидационный наборы\n",
    "train_size2 = int(0.7 * len(dataset2))  # 70% для тренировки\n",
    "val_size2 = len(dataset2) - train_size2\n",
    "train_dataset2, val_dataset2 = random_split(dataset2, [train_size2, val_size2])\n",
    "\n",
    "# Создание DataLoader\n",
    "train_loader2 = DataLoader(train_dataset2, batch_size=32, shuffle=True)\n",
    "val_loader2 = DataLoader(val_dataset2, batch_size=64)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train size:\t\t1376\n",
      "Validation size:\t590\n",
      "\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "  0%|          | 0/43 [00:00<?, ?it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:08<00:00,  4.93it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 0: train loss 0.69344\n",
      "Epoch 0: val loss 0.69398\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:08<00:00,  5.22it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1: train loss 0.69318\n",
      "Epoch 1: val loss 0.69419\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:09<00:00,  4.55it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 2: train loss 0.69288\n",
      "Epoch 2: val loss 0.69411\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:09<00:00,  4.59it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 3: train loss 0.69302\n",
      "Epoch 3: val loss 0.69408\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:09<00:00,  4.59it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 4: train loss 0.69259\n",
      "Epoch 4: val loss 0.69407\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:09<00:00,  4.57it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 5: train loss 0.69254\n",
      "Epoch 5: val loss 0.69403\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:09<00:00,  4.56it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 6: train loss 0.69250\n",
      "Epoch 6: val loss 0.69398\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:09<00:00,  4.58it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 7: train loss 0.69241\n",
      "Epoch 7: val loss 0.69389\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:09<00:00,  4.59it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 8: train loss 0.69213\n",
      "Epoch 8: val loss 0.69391\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:09<00:00,  4.58it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 9: train loss 0.69187\n",
      "Epoch 9: val loss 0.69377\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:09<00:00,  4.56it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 10: train loss 0.69168\n",
      "Epoch 10: val loss 0.69377\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:09<00:00,  4.59it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 11: train loss 0.69171\n",
      "Epoch 11: val loss 0.69375\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:09<00:00,  4.57it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 12: train loss 0.69163\n",
      "Epoch 12: val loss 0.69367\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:09<00:00,  4.55it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 13: train loss 0.69162\n",
      "Epoch 13: val loss 0.69364\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:09<00:00,  4.45it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 14: train loss 0.69127\n",
      "Epoch 14: val loss 0.69362\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:09<00:00,  4.32it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 15: train loss 0.69123\n",
      "Epoch 15: val loss 0.69359\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:08<00:00,  4.81it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 16: train loss 0.69092\n",
      "Epoch 16: val loss 0.69347\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:08<00:00,  4.81it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 17: train loss 0.69096\n",
      "Epoch 17: val loss 0.69350\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:09<00:00,  4.53it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 18: train loss 0.69067\n",
      "Epoch 18: val loss 0.69342\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:09<00:00,  4.55it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 19: train loss 0.69094\n",
      "Epoch 19: val loss 0.69336\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:09<00:00,  4.52it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 20: train loss 0.69064\n",
      "Epoch 20: val loss 0.69332\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:09<00:00,  4.56it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 21: train loss 0.69062\n",
      "Epoch 21: val loss 0.69335\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:09<00:00,  4.54it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 22: train loss 0.69029\n",
      "Epoch 22: val loss 0.69335\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:09<00:00,  4.60it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 23: train loss 0.69032\n",
      "Epoch 23: val loss 0.69329\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:09<00:00,  4.57it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 24: train loss 0.69012\n",
      "Epoch 24: val loss 0.69321\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:09<00:00,  4.57it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 25: train loss 0.68999\n",
      "Epoch 25: val loss 0.69316\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:09<00:00,  4.59it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 26: train loss 0.68988\n",
      "Epoch 26: val loss 0.69314\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:09<00:00,  4.50it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 27: train loss 0.68979\n",
      "Epoch 27: val loss 0.69310\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:09<00:00,  4.64it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 28: train loss 0.68956\n",
      "Epoch 28: val loss 0.69305\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:09<00:00,  4.59it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 29: train loss 0.68923\n",
      "Epoch 29: val loss 0.69288\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:09<00:00,  4.46it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 30: train loss 0.68948\n",
      "Epoch 30: val loss 0.69288\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:09<00:00,  4.58it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 31: train loss 0.68913\n",
      "Epoch 31: val loss 0.69283\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:09<00:00,  4.47it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 32: train loss 0.68895\n",
      "Epoch 32: val loss 0.69287\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:09<00:00,  4.58it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 33: train loss 0.68896\n",
      "Epoch 33: val loss 0.69284\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:09<00:00,  4.52it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 34: train loss 0.68874\n",
      "Epoch 34: val loss 0.69268\n"
     ]
    }
   ],
   "source": [
    "acc_std_norm2, hist__std_norm2 = train_and_evaluate_model(train_loader2, val_loader2, epo=35, rate = 1e-7)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Standart Normalizeted Model2 Accuracy: 0.484\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "outputs(\"Standart Normalizeted Model2\", acc_std_norm2, hist__std_norm2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "# C нормализацией и разные скорости обучения с оптимальным epoch count\n",
    "dataset3 = datasets.ImageFolder(DIR, transform=transforms.Compose([\n",
    "                                transforms.Resize((256, 256)), \n",
    "                                transforms.ToTensor(),\n",
    "                                transforms.Normalize(mean=mean, std=std), ## Normalize the input values\n",
    "                            ]))\n",
    "\n",
    "# Разделение на тренировочный и валидационный наборы\n",
    "train_size3 = int(0.7 * len(dataset3))  # 70% для тренировки\n",
    "val_size3 = len(dataset3) - train_size3\n",
    "train_dataset3, val_dataset3 = random_split(dataset3, [train_size3, val_size3])\n",
    "\n",
    "# Создание DataLoader\n",
    "train_loader3 = DataLoader(train_dataset3, batch_size=32, shuffle=True)\n",
    "val_loader3 = DataLoader(val_dataset3, batch_size=64)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train size:\t\t1376\n",
      "Validation size:\t590\n",
      "\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:07<00:00,  6.01it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 0: train loss 0.57493\n",
      "Epoch 0: val loss 0.61281\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:08<00:00,  4.83it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1: train loss 0.46597\n",
      "Epoch 1: val loss 0.52346\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:08<00:00,  4.84it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 2: train loss 0.43223\n",
      "Epoch 2: val loss 0.48513\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:08<00:00,  5.21it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 3: train loss 0.38314\n",
      "Epoch 3: val loss 0.67779\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:09<00:00,  4.52it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 4: train loss 0.35595\n",
      "Epoch 4: val loss 0.52568\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:09<00:00,  4.61it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 5: train loss 0.27888\n",
      "Epoch 5: val loss 0.50741\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:09<00:00,  4.70it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 6: train loss 0.23075\n",
      "Epoch 6: val loss 0.80986\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:09<00:00,  4.69it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 7: train loss 0.16576\n",
      "Epoch 7: val loss 0.63151\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:09<00:00,  4.75it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 8: train loss 0.13344\n",
      "Epoch 8: val loss 0.81759\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:08<00:00,  4.87it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 9: train loss 0.11412\n",
      "Epoch 9: val loss 1.15976\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:09<00:00,  4.77it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 10: train loss 0.10859\n",
      "Epoch 10: val loss 0.87864\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:09<00:00,  4.73it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 11: train loss 0.07609\n",
      "Epoch 11: val loss 1.18420\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:09<00:00,  4.58it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 12: train loss 0.09318\n",
      "Epoch 12: val loss 0.91947\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:09<00:00,  4.58it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 13: train loss 0.04647\n",
      "Epoch 13: val loss 0.78409\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:09<00:00,  4.68it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 14: train loss 0.02908\n",
      "Epoch 14: val loss 0.90953\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:09<00:00,  4.62it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 15: train loss 0.04598\n",
      "Epoch 15: val loss 1.09513\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:09<00:00,  4.62it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 16: train loss 0.06223\n",
      "Epoch 16: val loss 0.81803\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:09<00:00,  4.77it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 17: train loss 0.01448\n",
      "Epoch 17: val loss 1.07794\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:09<00:00,  4.67it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 18: train loss 0.01752\n",
      "Epoch 18: val loss 1.47661\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:09<00:00,  4.52it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 19: train loss 0.06653\n",
      "Epoch 19: val loss 0.87886\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:09<00:00,  4.68it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 20: train loss 0.02642\n",
      "Epoch 20: val loss 0.83791\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:09<00:00,  4.59it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 21: train loss 0.00435\n",
      "Epoch 21: val loss 1.00930\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:09<00:00,  4.68it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 22: train loss 0.00113\n",
      "Epoch 22: val loss 1.15695\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:08<00:00,  4.94it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 23: train loss 0.00418\n",
      "Epoch 23: val loss 1.65093\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:09<00:00,  4.65it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 24: train loss 0.04366\n",
      "Epoch 24: val loss 1.30637\n"
     ]
    }
   ],
   "source": [
    "acc_std_norm3, hist__std_norm3 = train_and_evaluate_model(train_loader3, val_loader3, epo=25, rate = 1e-3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Standart Normalizeted Model3 Accuracy: 0.764\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "outputs(\"Standart Normalizeted Model3\", acc_std_norm3, hist__std_norm3)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 3 Добавить аугментации\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Каждая аугментация подобрана с точки зрения вариативности и разнообразия данных для обучаемой модели. Также хотел сделать ее более устойчевой, к изменения формата, света, насыщенности изображения."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "# C нормализацией и разные скорости обучения с оптимальным epoch count\n",
    "datasetAUG = datasets.ImageFolder(DIR, transform=transforms.Compose([\n",
    "                                transforms.Resize((256, 256)),  \n",
    "                                transforms.RandomHorizontalFlip(),  \n",
    "                                transforms.RandomRotation(15),      \n",
    "                                transforms.ColorJitter(brightness=0.2, contrast=0.2),  \n",
    "                                transforms.ToTensor(),\n",
    "                            ]))\n",
    "\n",
    "# Разделение на тренировочный и валидационный наборы\n",
    "train_sizeAUG = int(0.7 * len(datasetAUG))  # 70% для тренировки\n",
    "val_sizeAUG = len(datasetAUG) - train_sizeAUG\n",
    "train_datasetAUG, val_datasetAUG = random_split(datasetAUG, [train_sizeAUG, val_sizeAUG])\n",
    "\n",
    "# Создание DataLoader\n",
    "train_loaderAUG = DataLoader(train_datasetAUG, batch_size=32, shuffle=True)\n",
    "val_loaderAUG = DataLoader(val_datasetAUG, batch_size=64)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train size:\t\t1376\n",
      "Validation size:\t590\n",
      "\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:07<00:00,  5.44it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 0: train loss 0.59074\n",
      "Epoch 0: val loss 0.87799\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:09<00:00,  4.77it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1: train loss 0.49421\n",
      "Epoch 1: val loss 0.53027\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:08<00:00,  4.78it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 2: train loss 0.48307\n",
      "Epoch 2: val loss 0.50693\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:08<00:00,  4.81it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 3: train loss 0.45776\n",
      "Epoch 3: val loss 0.47579\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:09<00:00,  4.77it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 4: train loss 0.44203\n",
      "Epoch 4: val loss 0.73336\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:08<00:00,  4.81it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 5: train loss 0.41964\n",
      "Epoch 5: val loss 0.40765\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:08<00:00,  4.78it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 6: train loss 0.41196\n",
      "Epoch 6: val loss 0.51044\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:08<00:00,  4.79it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 7: train loss 0.37292\n",
      "Epoch 7: val loss 0.51975\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:08<00:00,  4.81it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 8: train loss 0.39095\n",
      "Epoch 8: val loss 0.42850\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:08<00:00,  4.84it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 9: train loss 0.36717\n",
      "Epoch 9: val loss 0.47812\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:08<00:00,  4.80it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 10: train loss 0.34255\n",
      "Epoch 10: val loss 0.46496\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:08<00:00,  4.79it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 11: train loss 0.34717\n",
      "Epoch 11: val loss 0.45177\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:09<00:00,  4.72it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 12: train loss 0.29218\n",
      "Epoch 12: val loss 0.50671\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:08<00:00,  4.82it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 13: train loss 0.28387\n",
      "Epoch 13: val loss 0.68889\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:08<00:00,  4.84it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 14: train loss 0.30564\n",
      "Epoch 14: val loss 0.56409\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:08<00:00,  4.87it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 15: train loss 0.31179\n",
      "Epoch 15: val loss 0.62967\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:08<00:00,  4.88it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 16: train loss 0.26454\n",
      "Epoch 16: val loss 0.91004\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:08<00:00,  4.84it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 17: train loss 0.25296\n",
      "Epoch 17: val loss 0.59131\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:08<00:00,  4.84it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 18: train loss 0.23985\n",
      "Epoch 18: val loss 0.57001\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:08<00:00,  4.83it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 19: train loss 0.24171\n",
      "Epoch 19: val loss 0.54038\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:08<00:00,  4.82it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 20: train loss 0.20352\n",
      "Epoch 20: val loss 0.62308\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:08<00:00,  4.85it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 21: train loss 0.23038\n",
      "Epoch 21: val loss 0.59239\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:08<00:00,  4.88it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 22: train loss 0.19811\n",
      "Epoch 22: val loss 0.52918\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:08<00:00,  4.83it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 23: train loss 0.16061\n",
      "Epoch 23: val loss 0.54361\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:08<00:00,  4.82it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 24: train loss 0.15908\n",
      "Epoch 24: val loss 0.70401\n"
     ]
    }
   ],
   "source": [
    "acc_AUG, hist__AUG = train_and_evaluate_model(train_loaderAUG, val_loaderAUG, epo=25, rate = 1e-3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Standart Normalizeted Model3 Accuracy: 0.783\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "outputs(\"Model with AUG\", acc_AUG, hist__AUG)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "# C нормализацией и разные скорости обучения с оптимальным epoch count\n",
    "datasetAUG1 = datasets.ImageFolder(DIR, transform=transforms.Compose([\n",
    "                                transforms.Resize((256, 256)),  # Resize all images to the same scale\n",
    "                                transforms.RandomHorizontalFlip(),  # Randomly flip images horizontally\n",
    "                                transforms.RandomRotation(10),      # Randomly rotate images by up to 10 degrees\n",
    "                                transforms.ColorJitter(brightness=0.1, contrast=0.2, saturation=0.2, hue=0.1),  # Subtle color adjustments\n",
    "                                transforms.ToTensor(),\n",
    "                                transforms.Normalize(mean=mean, std=std)\n",
    "                            ]))\n",
    "\n",
    "# Разделение на тренировочный и валидационный наборы\n",
    "train_sizeAUG1 = int(0.7 * len(datasetAUG1))  # 70% для тренировки\n",
    "val_sizeAUG1 = len(datasetAUG1) - train_sizeAUG1\n",
    "train_datasetAUG1, val_datasetAUG1 = random_split(datasetAUG1, [train_sizeAUG1, val_sizeAUG1])\n",
    "\n",
    "# Создание DataLoader\n",
    "train_loaderAUG1 = DataLoader(train_datasetAUG1, batch_size=32, shuffle=True)\n",
    "val_loaderAUG1 = DataLoader(val_datasetAUG1, batch_size=64)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train size:\t\t1376\n",
      "Validation size:\t590\n",
      "\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:15<00:00,  2.74it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 0: train loss 0.61479\n",
      "Epoch 0: val loss 0.77420\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:15<00:00,  2.75it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1: train loss 0.54723\n",
      "Epoch 1: val loss 0.60660\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:15<00:00,  2.72it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 2: train loss 0.50080\n",
      "Epoch 2: val loss 0.50105\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:15<00:00,  2.73it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 3: train loss 0.47831\n",
      "Epoch 3: val loss 0.49189\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:16<00:00,  2.69it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 4: train loss 0.47220\n",
      "Epoch 4: val loss 0.53849\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:15<00:00,  2.75it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 5: train loss 0.46242\n",
      "Epoch 5: val loss 0.47927\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:16<00:00,  2.68it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 6: train loss 0.46529\n",
      "Epoch 6: val loss 0.53238\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:15<00:00,  2.72it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 7: train loss 0.43350\n",
      "Epoch 7: val loss 0.51069\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:15<00:00,  2.72it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 8: train loss 0.41304\n",
      "Epoch 8: val loss 0.45981\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:15<00:00,  2.73it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 9: train loss 0.38977\n",
      "Epoch 9: val loss 0.56294\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:15<00:00,  2.70it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 10: train loss 0.39479\n",
      "Epoch 10: val loss 0.61037\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:15<00:00,  2.78it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 11: train loss 0.38158\n",
      "Epoch 11: val loss 0.50319\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:15<00:00,  2.76it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 12: train loss 0.38398\n",
      "Epoch 12: val loss 0.45891\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:15<00:00,  2.76it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 13: train loss 0.33731\n",
      "Epoch 13: val loss 0.57182\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:15<00:00,  2.79it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 14: train loss 0.34757\n",
      "Epoch 14: val loss 0.54025\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:15<00:00,  2.78it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 15: train loss 0.33172\n",
      "Epoch 15: val loss 0.54386\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:15<00:00,  2.77it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 16: train loss 0.31185\n",
      "Epoch 16: val loss 0.60108\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:15<00:00,  2.77it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 17: train loss 0.29238\n",
      "Epoch 17: val loss 0.45691\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:15<00:00,  2.75it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 18: train loss 0.29671\n",
      "Epoch 18: val loss 0.54508\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:15<00:00,  2.72it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 19: train loss 0.28229\n",
      "Epoch 19: val loss 0.59758\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:15<00:00,  2.74it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 20: train loss 0.24194\n",
      "Epoch 20: val loss 0.54569\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:15<00:00,  2.77it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 21: train loss 0.23925\n",
      "Epoch 21: val loss 0.62140\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:15<00:00,  2.77it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 22: train loss 0.25438\n",
      "Epoch 22: val loss 0.50074\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:15<00:00,  2.70it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 23: train loss 0.22932\n",
      "Epoch 23: val loss 0.74007\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:16<00:00,  2.65it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 24: train loss 0.20325\n",
      "Epoch 24: val loss 0.66886\n"
     ]
    }
   ],
   "source": [
    "acc_AUG1, hist__AUG1 = train_and_evaluate_model(train_loaderAUG1, val_loaderAUG1, epo=25, rate = 1e-3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model with AUG1 Accuracy: 0.807\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "outputs(\"Model with AUG1\", acc_AUG1, hist__AUG1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Потери на валидации растут. Далее пробуем исправить. Такое, не очень интересно оставлять, не смотря на результат."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "# C нормализацией и разные скорости обучения с оптимальным epoch count\n",
    "datasetAUG2 = datasets.ImageFolder(DIR, transform=transforms.Compose([\n",
    "                                transforms.Resize((356, 356)),  # Resize all images to the same scale\n",
    "                                transforms.RandomHorizontalFlip(),  # Randomly flip images horizontally\n",
    "                                transforms.RandomRotation(10),      # Randomly rotate images by up to 10 degrees\n",
    "                                transforms.ColorJitter(brightness=0.1, contrast=0.2, saturation=0.3, hue=0.1),  # Subtle color adjustments\n",
    "                                transforms.ToTensor(),\n",
    "                                transforms.Normalize(mean=mean, std=std)\n",
    "                            ]))\n",
    "\n",
    "# Разделение на тренировочный и валидационный наборы\n",
    "train_sizeAUG2 = int(0.7 * len(datasetAUG2))  # 70% для тренировки\n",
    "val_sizeAUG2 = len(datasetAUG2) - train_sizeAUG2\n",
    "train_datasetAUG2, val_datasetAUG2 = random_split(datasetAUG2, [train_sizeAUG2, val_sizeAUG2])\n",
    "\n",
    "# Создание DataLoader\n",
    "train_loaderAUG2 = DataLoader(train_datasetAUG2, batch_size=32, shuffle=True)\n",
    "val_loaderAUG2 = DataLoader(val_datasetAUG2, batch_size=64, shuffle=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train size:\t\t1376\n",
      "Validation size:\t590\n",
      "\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:23<00:00,  1.86it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 0: train loss 0.68987\n",
      "Epoch 0: val loss 0.68855\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:22<00:00,  1.87it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1: train loss 0.68273\n",
      "Epoch 1: val loss 0.68025\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:23<00:00,  1.86it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 2: train loss 0.67430\n",
      "Epoch 2: val loss 0.66408\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:23<00:00,  1.86it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 3: train loss 0.66233\n",
      "Epoch 3: val loss 0.65133\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:22<00:00,  1.88it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 4: train loss 0.64790\n",
      "Epoch 4: val loss 0.64041\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:22<00:00,  1.88it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 5: train loss 0.62632\n",
      "Epoch 5: val loss 0.61782\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:22<00:00,  1.88it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 6: train loss 0.60769\n",
      "Epoch 6: val loss 0.60050\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:23<00:00,  1.85it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 7: train loss 0.58429\n",
      "Epoch 7: val loss 0.58067\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:23<00:00,  1.87it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 8: train loss 0.56678\n",
      "Epoch 8: val loss 0.56644\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:23<00:00,  1.87it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 9: train loss 0.54901\n",
      "Epoch 9: val loss 0.56177\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:23<00:00,  1.87it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 10: train loss 0.52871\n",
      "Epoch 10: val loss 0.54951\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:23<00:00,  1.87it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 11: train loss 0.51647\n",
      "Epoch 11: val loss 0.54967\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:23<00:00,  1.83it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 12: train loss 0.51942\n",
      "Epoch 12: val loss 0.53849\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:23<00:00,  1.80it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 13: train loss 0.50047\n",
      "Epoch 13: val loss 0.53568\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:23<00:00,  1.84it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 14: train loss 0.48507\n",
      "Epoch 14: val loss 0.52156\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:23<00:00,  1.84it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 15: train loss 0.47291\n",
      "Epoch 15: val loss 0.51228\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:23<00:00,  1.82it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 16: train loss 0.47320\n",
      "Epoch 16: val loss 0.51125\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:23<00:00,  1.86it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 17: train loss 0.44979\n",
      "Epoch 17: val loss 0.52334\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:22<00:00,  1.88it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 18: train loss 0.43807\n",
      "Epoch 18: val loss 0.51191\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:22<00:00,  1.87it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 19: train loss 0.42986\n",
      "Epoch 19: val loss 0.49895\n"
     ]
    }
   ],
   "source": [
    "acc_AUG2, hist__AUG2 = train_and_evaluate_model(train_loaderAUG2, val_loaderAUG2, epo=20)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model with AUG2 Accuracy: 0.732\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "outputs(\"Model with AUG2\", acc_AUG2, hist__AUG2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "# C нормализацией и разные скорости обучения с оптимальным epoch count\n",
    "datasetAUG3 = datasets.ImageFolder(DIR, transform=transforms.Compose([\n",
    "                                transforms.Resize((356, 356)),  # Resize all images to the same scale\n",
    "                                transforms.RandomHorizontalFlip(),  # Randomly flip images horizontally\n",
    "                                transforms.RandomRotation(10),      # Randomly rotate images by up to 10 degrees\n",
    "                                transforms.ColorJitter(brightness=0.1, contrast=0.2, saturation=0.3),  # Subtle color adjustments\n",
    "                                transforms.ToTensor(),\n",
    "                                transforms.Normalize(mean=mean, std=std)\n",
    "                            ]))\n",
    "\n",
    "# Разделение на тренировочный и валидационный наборы\n",
    "train_sizeAUG3 = int(0.7 * len(datasetAUG3))  # 70% для тренировки\n",
    "val_sizeAUG3 = len(datasetAUG3) - train_sizeAUG3\n",
    "train_datasetAUG3, val_datasetAUG3 = random_split(datasetAUG3, [train_sizeAUG3, val_sizeAUG3])\n",
    "\n",
    "# Создание DataLoader\n",
    "train_loaderAUG3 = DataLoader(train_datasetAUG3, batch_size=32, shuffle=True)\n",
    "val_loaderAUG3 = DataLoader(val_datasetAUG3, batch_size=64, shuffle=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train size:\t\t1376\n",
      "Validation size:\t590\n",
      "\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:12<00:00,  3.51it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 0: train loss 0.69167\n",
      "Epoch 0: val loss 0.69114\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:12<00:00,  3.53it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1: train loss 0.68243\n",
      "Epoch 1: val loss 0.67691\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:12<00:00,  3.51it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 2: train loss 0.66861\n",
      "Epoch 2: val loss 0.65810\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:12<00:00,  3.53it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 3: train loss 0.64032\n",
      "Epoch 3: val loss 0.63501\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:12<00:00,  3.51it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 4: train loss 0.61024\n",
      "Epoch 4: val loss 0.59291\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:12<00:00,  3.52it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 5: train loss 0.58325\n",
      "Epoch 5: val loss 0.57009\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:12<00:00,  3.51it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 6: train loss 0.55627\n",
      "Epoch 6: val loss 0.55233\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:12<00:00,  3.49it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 7: train loss 0.53479\n",
      "Epoch 7: val loss 0.52612\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:12<00:00,  3.51it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 8: train loss 0.51757\n",
      "Epoch 8: val loss 0.51831\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:12<00:00,  3.51it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 9: train loss 0.50509\n",
      "Epoch 9: val loss 0.52084\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:12<00:00,  3.51it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 10: train loss 0.49575\n",
      "Epoch 10: val loss 0.50186\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:12<00:00,  3.50it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 11: train loss 0.47969\n",
      "Epoch 11: val loss 0.49020\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:12<00:00,  3.52it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 12: train loss 0.46749\n",
      "Epoch 12: val loss 0.47629\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:12<00:00,  3.51it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 13: train loss 0.45741\n",
      "Epoch 13: val loss 0.45425\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:12<00:00,  3.54it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 14: train loss 0.45194\n",
      "Epoch 14: val loss 0.47067\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:12<00:00,  3.53it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 15: train loss 0.43453\n",
      "Epoch 15: val loss 0.43831\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:12<00:00,  3.54it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 16: train loss 0.43695\n",
      "Epoch 16: val loss 0.44475\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:12<00:00,  3.51it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 17: train loss 0.42549\n",
      "Epoch 17: val loss 0.43942\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:12<00:00,  3.51it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 18: train loss 0.39915\n",
      "Epoch 18: val loss 0.42605\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:12<00:00,  3.52it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 19: train loss 0.41131\n",
      "Epoch 19: val loss 0.43919\n"
     ]
    }
   ],
   "source": [
    "acc_AUG3, hist__AUG3 = train_and_evaluate_model(train_loaderAUG3, val_loaderAUG3, epo=20)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model with AUG3 Accuracy: 0.800\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "outputs(\"Model with AUG3\", acc_AUG3, hist__AUG3)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    " >80% с аугментацией есть на \"Model with AUG3\", и на \"Model with AUG1\"\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 4 Сопоставить качество модели"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Сопоставляю самые удачные модели из группы Standart VS Model with AUG"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Standart Normalizeted Model| Accuracy: 0.775\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model with AUG| Accuracy: 0.800\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "outputs(\"Standart Normalizeted Model|\", acc_std_norm, hist__std_norm)\n",
    "outputs(\"Model with AUG|\", acc_AUG3, hist__AUG3)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Модель, обученная с помощью дополнения данных, превзошла стандартную нормализованную модель с точки зрения точности.\n",
    "\n",
    "Увеличение данных, по-видимому, помогает предотвратить переобучение, о чем свидетельствуют тенденции потерь при проверке, следующие за потерями при обучении, что особенно заметно на графике дополненной модели.\n",
    "\n",
    "Потери на обучение расширенной модели неуклонно уменьшаются, что указывает на то, что модель эффективно обучается на расширенном наборе данных.\n",
    "\n",
    "Кривые обучения предполагают, что расширенная модель может продолжать получать выгоду от дальнейшего обучения, поскольку потери при проверке все еще снижаются в конце наблюдаемых эпох.\n",
    "\n",
    "*Включение увеличения данных в процесс обучения явно повысило точность прогнозирования модели, что свидетельствует о ее эффективности в улучшении обобщения модели для этой конкретной задачи*"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 5 Усовершенствовать модель"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "class MODImprovedModel(nn.Module):\n",
    "    def __init__(self):\n",
    "        super(MODImprovedModel, self).__init__()\n",
    "\n",
    "        self.conv_layers = nn.Sequential(\n",
    "            nn.Conv2d(3, 24, kernel_size=3, stride=1, padding=1),\n",
    "            nn.BatchNorm2d(24),\n",
    "            nn.ReLU(),\n",
    "            nn.MaxPool2d(kernel_size=2, stride=2),\n",
    "            \n",
    "            nn.Conv2d(24, 40, kernel_size=3, stride=2, padding=1),\n",
    "            nn.BatchNorm2d(40),\n",
    "            nn.ReLU(),\n",
    "            nn.MaxPool2d(kernel_size=2, stride=2),\n",
    "            \n",
    "            nn.Conv2d(40, 80, kernel_size=3, stride=2, padding=1),\n",
    "            nn.BatchNorm2d(80),\n",
    "            nn.ReLU(),\n",
    "            \n",
    "            nn.Conv2d(80, 160, kernel_size=3, stride=2, padding=1),\n",
    "            nn.BatchNorm2d(160),\n",
    "            nn.ReLU(),\n",
    "            \n",
    "            nn.Conv2d(160, 160, kernel_size=3, stride=2, padding=1),\n",
    "            nn.BatchNorm2d(160),\n",
    "            nn.ReLU(),\n",
    "            \n",
    "            nn.Conv2d(160, 320, kernel_size=3, stride=2, padding=1),\n",
    "            nn.BatchNorm2d(320),\n",
    "            nn.ReLU(),\n",
    "            \n",
    "            nn.Conv2d(320, 320, kernel_size=3, stride=1, padding=1),\n",
    "            nn.BatchNorm2d(320),\n",
    "            nn.ReLU(),\n",
    "            nn.Dropout(0.3),\n",
    "            nn.AdaptiveAvgPool2d((1, 1))\n",
    "        )\n",
    "\n",
    "        self.fc_layers = nn.Sequential(\n",
    "            nn.Flatten(),\n",
    "            nn.Linear(320, 128),\n",
    "            nn.ReLU(),\n",
    "            nn.Dropout(0.4),  # Adding dropout layer here\n",
    "            nn.Linear(128, 42),\n",
    "            nn.ReLU(),\n",
    "            nn.Dropout(0.4),  # Adding dropout layer here\n",
    "            nn.Linear(42, 1),\n",
    "            nn.Sigmoid()  # Sigmoid for the final layer\n",
    "        )\n",
    "\n",
    "    def forward(self, x):\n",
    "        x = self.conv_layers(x)\n",
    "        x = self.fc_layers(x)\n",
    "        return x"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "def MODtrain_and_evaluate_model(train_loader, val_loader, epo, rate=1e-5, mdl=None): # rate was taken empirically. This value is optimal in the base\n",
    "    print(f\"Train size:\\t\\t{len(train_loader.dataset)}\\n\"\n",
    "          f\"Validation size:\\t{len(val_loader.dataset)}\\n\")\n",
    "\n",
    "\n",
    "    model = mdl if mdl is not None else MODImprovedModel()\n",
    "    loss_fn = torch.nn.BCELoss()\n",
    "    opt = torch.optim.Adam(model.parameters(), lr=rate)       \n",
    "\n",
    "    history = lab4_util.train(\n",
    "        model,\n",
    "        loss_fn,\n",
    "        opt,\n",
    "        train_loader,\n",
    "        val_loader,\n",
    "        max_epochs=epo\n",
    "    )\n",
    "\n",
    "    return lab4_util.accuracy(model, val_loader), history"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train size:\t\t1376\n",
      "Validation size:\t590\n",
      "\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:11<00:00,  3.70it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 0: train loss 0.69499\n",
      "Epoch 0: val loss 0.69174\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:12<00:00,  3.48it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1: train loss 0.69117\n",
      "Epoch 1: val loss 0.68560\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:12<00:00,  3.49it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 2: train loss 0.68402\n",
      "Epoch 2: val loss 0.68115\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:12<00:00,  3.47it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 3: train loss 0.67839\n",
      "Epoch 3: val loss 0.67551\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:12<00:00,  3.48it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 4: train loss 0.67034\n",
      "Epoch 4: val loss 0.66765\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:12<00:00,  3.49it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 5: train loss 0.66171\n",
      "Epoch 5: val loss 0.65509\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:12<00:00,  3.50it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 6: train loss 0.64470\n",
      "Epoch 6: val loss 0.64255\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:12<00:00,  3.50it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 7: train loss 0.63450\n",
      "Epoch 7: val loss 0.62222\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:12<00:00,  3.50it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 8: train loss 0.61689\n",
      "Epoch 8: val loss 0.60241\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:12<00:00,  3.48it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 9: train loss 0.60211\n",
      "Epoch 9: val loss 0.59272\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:12<00:00,  3.51it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 10: train loss 0.59191\n",
      "Epoch 10: val loss 0.57558\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:12<00:00,  3.44it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 11: train loss 0.58306\n",
      "Epoch 11: val loss 0.56599\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:12<00:00,  3.47it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 12: train loss 0.57260\n",
      "Epoch 12: val loss 0.55446\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:12<00:00,  3.52it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 13: train loss 0.56198\n",
      "Epoch 13: val loss 0.54470\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:12<00:00,  3.53it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 14: train loss 0.54504\n",
      "Epoch 14: val loss 0.52761\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:12<00:00,  3.53it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 15: train loss 0.53517\n",
      "Epoch 15: val loss 0.52979\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:12<00:00,  3.52it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 16: train loss 0.52238\n",
      "Epoch 16: val loss 0.51631\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:12<00:00,  3.53it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 17: train loss 0.50740\n",
      "Epoch 17: val loss 0.50653\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:12<00:00,  3.52it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 18: train loss 0.49455\n",
      "Epoch 18: val loss 0.49823\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:12<00:00,  3.55it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 19: train loss 0.48895\n",
      "Epoch 19: val loss 0.49758\n"
     ]
    }
   ],
   "source": [
    "acc_AUG3, hist__AUG3 = MODtrain_and_evaluate_model(train_loaderAUG3, val_loaderAUG3, epo=20)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model with AUG3 Accuracy: 0.774\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAkAAAAGwCAYAAABB4NqyAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8pXeV/AAAACXBIWXMAAA9hAAAPYQGoP6dpAABtnklEQVR4nO3dd3gUVd/G8e9uekISSiAFQu8tQIBQBSUSigiI0ntRERVEH5XXh2LlsSOKoFSxUFVAaUIEpHeQ3gktAQKkQgrZef9YiUZCCSmbcn+uay93Z8+c+Q1DzM3MmTkmwzAMRERERAoQs60LEBEREclpCkAiIiJS4CgAiYiISIGjACQiIiIFjgKQiIiIFDgKQCIiIlLgKACJiIhIgWNv6wJyI4vFwoULF3B3d8dkMtm6HBEREbkPhmEQGxuLn58fZvPdz/EoAKXjwoUL+Pv727oMEREReQBnz56lVKlSd22jAJQOd3d3wPoH6OHhYeNqRERE5H7ExMTg7++f+nv8bhSA0nHrspeHh4cCkIiISB5zP8NXNAhaREREChwFIBERESlwckUAmjRpEmXLlsXZ2ZmgoCC2bdt2x7YtW7bEZDLd9mrfvn1qG8MwGDNmDL6+vri4uBAcHMyxY8dyYldEREQkD7D5GKB58+YxcuRIpkyZQlBQEBMmTCAkJIQjR45QokSJ29r/9NNPJCUlpX6+cuUKAQEBPPXUU6nLPvjgAyZOnMg333xDuXLlGD16NCEhIRw8eBBnZ+cc2S8REckdUlJSSE5OtnUZkgUcHByws7PLkr5MhmEYWdLTAwoKCqJBgwZ88cUXgPUZPP7+/rzwwgu8/vrr91x/woQJjBkzhvDwcNzc3DAMAz8/P15++WVeeeUVAKKjo/H29mbWrFl07979nn3GxMTg6elJdHS0BkGLiORRhmEQERFBVFSUrUuRLFS4cGF8fHzSHeickd/fNj0DlJSUxM6dOxk1alTqMrPZTHBwMJs3b76vPqZPn0737t1xc3MD4NSpU0RERBAcHJzaxtPTk6CgIDZv3pxuAEpMTCQxMTH1c0xMzIPukoiI5BK3wk+JEiVwdXXVg23zOMMwuH79OpcuXQLA19c3U/3ZNABFRkaSkpKCt7d3muXe3t4cPnz4nutv27aN/fv3M3369NRlERERqX38u89b3/3b+PHjefPNNzNavoiI5FIpKSmp4adYsWK2LkeyiIuLCwCXLl2iRIkSmboclisGQT+o6dOnU6tWLRo2bJipfkaNGkV0dHTq6+zZs1lUoYiI2MKtMT+urq42rkSy2q1jmtlxXTYNQF5eXtjZ2XHx4sU0yy9evIiPj89d142Pj2fu3LkMGjQozfJb62WkTycnp9SHHurhhyIi+Ycue+U/WXVMbRqAHB0dCQwMJDQ0NHWZxWIhNDSUxo0b33XdBQsWkJiYSO/evdMsL1euHD4+Pmn6jImJYevWrffsU0RERAoGm98GP3LkSPr160f9+vVp2LAhEyZMID4+ngEDBgDQt29fSpYsyfjx49OsN336dDp16nTbtV2TycSIESN45513qFSpUupt8H5+fnTq1CmndktERERyMZsHoG7dunH58mXGjBlDREQEderUYcWKFamDmM+cOXPblPZHjhxhw4YN/Pbbb+n2+eqrrxIfH8/TTz9NVFQUzZo1Y8WKFXoGkIiIFEhly5ZlxIgRjBgxwtal5Bo2fw5QbpRdzwG6Fp/ExdgEqni767q0iEg2SkhI4NSpU5QrVy5P/eP3Xr8bxo4dy7hx4zLc7+XLl3Fzc8sXg8LvdmzzzHOACprl+yP4v5/34VXIkcYVvGhWsRhNKnjhXzTv/4UUEZHMCw8PT30/b948xowZw5EjR1KXFSpUKPW9YRikpKRgb3/vX+XFixfP2kLzgTx9G3xec+16Ei4OdkTGJfHL3gu89uM+mn+whoc+WMOon/7kl70XuBKXeO+OREQkwwzD4HrSTZu87vdii4+PT+rL09MTk8mU+vnw4cO4u7uzfPlyAgMDcXJyYsOGDZw4cYKOHTvi7e1NoUKFaNCgAatXr07Tb9myZZkwYULqZ5PJxLRp0+jcuTOurq5UqlSJJUuWZOUfd66nM0A5aNjDFRnSvDy7z1xj44krbDoeyZ6zUZy5ep0z264zZ5v1+UNVfdxpWtGLphWL0bBcMQo56TCJiGTWjeQUqo9ZaZNtH3wrBFfHrPl/+euvv85HH31E+fLlKVKkCGfPnqVdu3a8++67ODk5MXv2bDp06MCRI0coXbr0Hft58803+eCDD/jwww/5/PPP6dWrF2FhYRQtWjRL6szt9Js1hznamwkqX4yg8sUY+Whl4hJvsv3UVTYej2TjiSscCo/hcEQshyNimb7hFPZmEwH+ha2BqEIx6pYugqO9TtyJiBRUb731Fo8++mjq56JFixIQEJD6+e233+bnn39myZIlPP/883fsp3///vTo0QOA9957j4kTJ7Jt2zbatGmTfcXnIgpAOckwICUJ7J1SFxVysufhqiV4uGoJACLjEtl84gqbTkSy8fgVzly9zs6wa+wMu8bE0GO4ONjRoFxRmlYoRtOKXlT39cBs1oBqEZF7cXGw4+BbITbbdlapX79+ms9xcXGMGzeOpUuXEh4ezs2bN7lx4wZnzpy5az+1a9dOfe/m5oaHh0fqPFsFgQJQTjq9HhYMgMB+UH8geJa6rYlXISc6BPjRIcAPgLNXr6eGoU0nIomMS+KPo5f54+hlAAq7OtC4fDGa/HWGqJyXm+4wExFJh8lkyrLLULZ0a/LvW1555RVWrVrFRx99RMWKFXFxceHJJ58kKSnprv04ODik+WwymbBYLFleb26V9/8m5CV/zofrkbD+Y9jwKVRtDw2fhrLN4Q6hxb+oK92KlqZbg9IYhsGRi7HWMHQ8kq2nrhJ1PZnl+yNYvt860WupIi4MbFqOHg1L4+KYdf/iEBGR3Gnjxo3079+fzp07A9YzQqdPn7ZtUXmAAlBOemwCVGoN2762ng069Iv1VbwqNBwCtbuDU6E7rm4ymajq40FVHw8GNStHcoqFP89Fs+l4JBtPRLIrLIpz127w1q8H+XLtCZ5+qBy9G5XJF//iERGR9FWqVImffvqJDh06YDKZGD16dIE6k/OgNJo2J9nZQ/XHof+v8NwWqD8IHNzg8mFY+jJ8Ug2WvwaRx+6rOwc7M4FlivBCq0rMfboxe8e25r3OtShZ2IXIuETeW3aYZu+v4cu1x4lLvJnNOyciIrbwySefUKRIEZo0aUKHDh0ICQmhXr16ti4r19OToNORXU+CTldCNOyZYz0rdPXE38srPGK9PFapNZgzdikrOcXCz7vPM2nNccKuXAesY4UGNS1Hv6Zl8XB2uEcPIiJ5W159ErTcW1Y9CVoBKB05GoBusVjg5BrYNhWOrgD+OiyFS0ODwVC3D7hm7NkMN1MsLNl7gS9+P87JyHgA3J3tGdC0HAOblqWwq2MW74SISO6gAJR/KQBlI5sEoH+6dhq2T4ddsyEhyrrM3hlqPWk9K+QbcLe1b5NiMfj1T2sQOnYpDrDeft+3cRkGNy9PUTcFIRHJXxSA8i8FoGxk8wB0S9J12P8jbPsKIvb9vdw/yBqEqj0O9vcfXiwWgxUHIpgYeozDEbEAuDra0aeRNQgVd3e6Rw8iInmDAlD+pQCUjXJNALrFMODsNus4oYOLwPLXgGa3ElB/AAQOAA/f++7OYjFYdegin/9+jP3nYwBwdjDTs2EZnmlRHm8P/c9CRPI2BaD8SwEoG+W6APRPsRGw8xvYMQPirM/+wWwP1TpYzwqVbnzHZwr9m2EYrDlyic9Cj7P3bBRgnaqjRwN/nm1ZAV9Pl2zaCRGR7KUAlH8pAGWjXB2AbklJtj5DaNtUOLPp7+UlakD1jlClLfjUuq8wZBgG649F8lnoMXaGXQPA0c7Mk/VLMbRFBfyLumbXXoiIZAsFoPxLASgb5YkA9E8R+2H7VNg7D27e+Hu5p781CFVpB2Wa3nO8kGEYbD5xhc9Cj7H11FUA7M0mutQrxXMPV6BMMbe7ri8iklsoAOVfCkDZKM8FoFtuXINDv8KR5XDi97RhyMkDKgZbp9+oGAwuhe/a1daTV/j89+NsOB4JgJ3ZRMc6fjzXsiIVS9z5adUiIrmBAlD+lVUBSE+Czk9cikC9PtDjB3j1JPSYC/X6WgdLJ8bAgZ/gx0HwYQX4pgNsmWK95T4dQeWL8d3gIH4c2pgWlYuTYjH4add5gj9ZR9evNvPjznPcSErJ2f0TEZF7atmyJSNGjEj9XLZsWSZMmHDXdUwmE4sWLcr0trOqn5ygSaLyK0fXvy5/tbU+ZPH8TjiyzPq6fBhO/WF9rXjNOm6oajtrW9+6YP47FweWKco3Axuy92wUn/9+nN8PX2TbqatsO3WVcUsO0KGOH90b+FOrpKdmoRcRyaQOHTqQnJzMihUrbvtu/fr1PPTQQ+zdu5fatWvfd5/bt2+/bQb5zBo3bhyLFi1iz549aZaHh4dTpEiRLN1WdlEAKgjMZvBvYH0Fj4UrJ6xPmz68zDqA+tIB6+uPD8HdFyq3sY4bKvcQOFhPLwb4F2Zav/qER99g4Y5zzN95lrNXb/DD1jP8sPUMVX3c6dbAn851S+oJ0yIiD2jQoEF06dKFc+fOUapUqTTfzZw5k/r162co/AAUL148K0u8Kx8fnxzbVmbpElhBVKwCNB4GA5bCf05A56+td445FoLYcNg5E354Cj4oD/N6w54fIP4KAL6eLrzQqhLrXnmYHwYH0bGOH472Zg5HxPLmLwdp+G4oL8zZzYZjkVgsGl4mIpIRjz32GMWLF2fWrFlplsfFxbFgwQI6depEjx49KFmyJK6urtSqVYs5c+bctc9/XwI7duwYDz30EM7OzlSvXp1Vq1bdts5rr71G5cqVcXV1pXz58owePZrk5GQAZs2axZtvvsnevXsxmUyYTKbUev99CWzfvn088sgjuLi4UKxYMZ5++mni4uJSv+/fvz+dOnXio48+wtfXl2LFijFs2LDUbWUnnQEq6FyLQkA36+tmIpxebz0zdGQ5xF6w3mp/6BcwmcG/kfV5Q7W7YnbzoklF6+vN60ks3nOBedvPcjA8hl/2XuCXvRcoVcSFpwL9eap+KfwK65lCImJjhgHJ122zbQfX+3osib29PX379mXWrFm88cYbqUMLFixYQEpKCr1792bBggW89tpreHh4sHTpUvr06UOFChVo2LDhPfu3WCw88cQTeHt7s3XrVqKjo9OMF7rF3d2dWbNm4efnx759+xgyZAju7u68+uqrdOvWjf3797NixQpWr14NgKen5219xMfHExISQuPGjdm+fTuXLl1i8ODBPP/882kC3po1a/D19WXNmjUcP36cbt26UadOHYYMGXLP/ckMBSD5m72T9Q6xisHQ/mMI3/v3uKGIfdbLZWc2waox1vFC9fpChUco7OpIvyZl6dekLPvPRzNv+1kW7TnPuWs3+HT1USaEHqV5peJ0b+BPcDVvHO114lFEbCD5OrznZ5tt/98FcLy/cTgDBw7kww8/ZN26dbRs2RKwXv7q0qULZcqU4ZVXXklt+8ILL7By5Urmz59/XwFo9erVHD58mJUrV+LnZ/2zeO+992jbtm2adv/9739T35ctW5ZXXnmFuXPn8uqrr+Li4kKhQoWwt7e/6yWvH374gYSEBGbPnp06BumLL76gQ4cOvP/++3h7ewNQpEgRvvjiC+zs7KhatSrt27cnNDRUAUhsxGQCvzrW18P/B1FnrUFo7xy4sBsOLbG+PEpCnZ5QtzcUKUvNkp7ULOnJG+2rsXx/OPO2n2XLyav8cfQyfxy9TFE3RzrXLUm3Bv5U9na39V6KiOQ6VatWpUmTJsyYMYOWLVty/Phx1q9fz1tvvUVKSgrvvfce8+fP5/z58yQlJZGYmIir6/09sPbQoUP4+/unhh+Axo0b39Zu3rx5TJw4kRMnThAXF8fNmzcz/FiYQ4cOERAQkGYAdtOmTbFYLBw5ciQ1ANWoUQM7O7vUNr6+vuzbt++2/rKaApDcn8L+EPSM9RWxH3Z/C3/Og5jz1sHTf3wI5VpYzwpVfQxnB2c61y1F57qlOB0Zz/wdZ1m48xyXYhOZvuEU0zecom7pwnSr789jAX4UctJfRRHJZg6u1jMxttp2BgwaNIgXXniBSZMmMXPmTCpUqECLFi14//33+eyzz5gwYQK1atXCzc2NESNGkJSUlGWlbt68mV69evHmm28SEhKCp6cnc+fO5eOPP86ybfyTg4NDms8mkwmLxZIt2/on/daRjPOpCW3fh+A34chS2PUtnFwLp9ZZX86FoXZXqNsHfGtT1suNV9tUZeSjlVl39DLztp/l98OX2H0mit1nonjr14O0r+VLtwb+BJYpotvpRSR7mEz3fRnK1rp27crw4cP54YcfmD17NkOHDsVkMrFx40Y6duxI7969AeuYnqNHj1K9evX76rdatWqcPXuW8PBwfH2tk2hv2bIlTZtNmzZRpkwZ3njjjdRlYWFhado4OjqSknL3Z8FVq1aNWbNmER8fn3oWaOPGjZjNZqpUqXJf9WYnDcaQB+fgDDW7QN9FMHwvtHgNPEpBQpR15vqvmsNXD1nnK7sRhb2dmVbVvPm6b302j2rFqLZVKV/cjetJKSzYeY4np2ym06SNbDl5xdZ7JiJiU4UKFaJbt26MGjWK8PBw+vfvD0ClSpVYtWoVmzZt4tChQzzzzDNcvHjxvvsNDg6mcuXK9OvXj71797J+/fo0QefWNs6cOcPcuXM5ceIEEydO5Oeff07TpmzZspw6dYo9e/YQGRlJYmLibdvq1asXzs7O9OvXj/3797NmzRpeeOEF+vTpk3r5y5YUgCRrFCljHSs04k/o/SNU7wRmB+tA6mWvwMdV4Mch1ocvWiwUd3fimRYVCB3ZggXPNubJwFK4ONix91w03b/ewuBvdnD8Utw9Nysikl8NGjSIa9euERISkjpm57///S/16tUjJCSEli1b4uPjQ6dOne67T7PZzM8//8yNGzdo2LAhgwcP5t13303T5vHHH+ell17i+eefp06dOmzatInRo0enadOlSxfatGnDww8/TPHixdO9Fd/V1ZWVK1dy9epVGjRowJNPPkmrVq344osvMv6HkQ00F1g68uxcYLlN/BXrOKHd38Klg38vL1LWOmi6Ti/w+Hsg3uXYRCasPsrc7WdJsRjYmU30bFia4cGV8CrklPP1i0iepbnA8i9NhpqNFICymGHA+V2wezbs+xGSYq3LTWbrLfd1+1ifPv3XbPXHL8Xyv+WHWX3oEgCFnOwZ2rICA5uWw8XR7k5bERFJpQCUfykAZSMFoGyUFA8HF1sHTp/Z9PdyVy8I6A4NBkPRcgBsPnGF95YdYt/5aAB8PJx5JaQKneuWxM6sgdIicmcKQPmXAlA2UgDKIZHHrJfH9syBeOvZHswOUH8gPPQfKFQci8Xglz8v8MGKI5yPugFANV8P3mhXjWaVvGxYvIjkZgpA+ZcCUDZSAMphKclw7DfrnWMn11qXORaCJi9a5yxzKkRCcgrfbDrNF2uOE5twE4AWlYszql1VqvroGIlIWgpA+VdWBSDdBSa2Z+cAVdtD38XWl28dSIqDte/BxLqwfRrOZgvPtKjAH/95mAFNy+JgZ2Ld0cu0+2w9ry38k4sxCbbeCxHJhfRv/Pwnq46pApDkLuVbwpA18OQM691i8Zdg6cswKQgOLKKIqwNjO9Rg1UstaFfLB4sB83acpeWHa/lk1VHiE2/aeg9EJBe49XTh69dtNPmpZJtbx/TfT5DOKF0CS4cugeUSN5Ng5yxY9z5cj7QuKxlofQJ1ueYA7Ay7yrtLD7HrTBQAXoWcGPloZbrWL4W9nfK9SEEWHh5OVFQUJUqUwNXVVU+Zz+MMw+D69etcunSJwoULpz7J+p80BiiTFIBymcRY2PQ5bPoCkuOtyyq1hlZjwacmhmGwYn8E7684zOkr1n8ZVCpRiFHtqvJwlRL6n55IAWUYBhEREURFRdm6FMlChQsXxsfHJ93/tysAZZICUC4VexH++MB6VshyEzBBQA/rE6gL+5N008L3W8OYGHqMa9eTAWhSoRj/164aNUt62rR0EbGdlJQUkpOTbV2GZAEHB4c0M8f/mwJQJikA5XJXTkDoW3BwkfWznRMEPQ3NRoJrUaJvJPPl2uPM3HiapJvWGYU71y3JKyFVKFnYxXZ1i4hItlIAyiQFoDzi3E5YPRZOr7d+dvKE5i9B0LPg4MK5a9f5aOURFu25AICjvZlnHirPcy0r6onSIiL5kAJQJikA5SGGAcdXw6qxcOmAdZm7n/WyWJ2eYLZj37lo3l12kC0nrwJQsrALox+rRkiN9K8hi4hI3qQAlEkKQHmQJQX+nA9r3oXos9ZlxatC8Dio3AYDWHngIm//ejD1idLNKnox7vHqVCzhbrOyRUQk6ygAZZICUB6WnADbp8H6j+DGNeuy0o3h0bfAvyE3klKYvO4EU9adIOmmBXuziYHNyvFiq0oUcrK3be0iIpIpCkCZpACUD9yIgo0TYMtkuPnXU6KrPgZt3wfPUpy5cp23fj3I6kMXASjh7sT/tatGxzp+uiwmIpJHKQBlkgJQPhJzAdaOh93fgWGBQj7Qcy741QVgzeFLvPnLgdTnBzUsW5Rxj9egup+Ou4hIXqMAlEkKQPnQpcOwcABcOggOrtapNqq0BSDxZgrT1p/ii9+PcyM5BbMJ+jQqw8hHq+DpmrlHrYuISM5RAMokBaB8KiEa5veDk2vAZIY2/4OgZ1K/vhB1g3eXHWLpn+EAFHVz5NWQKnSt74/ZrMtiIiK5nQJQJikA5WMpybB0JOyabf3c6Dlo/Q6Y/34u0KYTkYxdfIBjl+IACCjlyZsda1LHv7ANChYRkfulAJRJCkD5nGHAhk+sT5MGqNIeukwFR7fUJskpFmZvDmPCqqPE/jXDfLf6/rzapgrFCjnZomoREbkHBaBMUgAqIPb/CD8PhZRE66DoHvPA3TtNk0uxCby//Ag/7joHgIezPS+3rkKvoNKabV5EJJdRAMokBaAC5MwWmNMDblwFT3/otQBKVLut2c6wq4xZfIADF2IAqOrjzlsda9KwXNGcrlhERO5AASiTFIAKmCsn4Pun4OoJcPKArrOhwsO3NUuxGMzZdoYPVx4h+oZ1ZulOdfwY1a4a3h7OOV21iIj8iwJQJikAFUDXr8LcXnBmE5jt4bEJUK9Puk2vxifx0W9HmLPtDIYBbo52DA+uRP8m5XC012UxERFbUQDKJAWgAupmIiweBvsWWD83fxke/i+Y0w81+85FM2bJfnafiQKgYolCzB7YEL/CLjlUsIiI/FNGfn/rn6sit9g7wRNT4aH/WD+v/xh+GmydXywdtUp58uOzTfjwydp4FXLk+KU4Bs7aTtxfd42JiEjupQAk8k8mEzzyX+g4yXopbP+P8G0niL+SbnOz2cRT9f1Z/Hwzirs7cTgilud/2MXNFEvO1i0iIhmiACSSnrq9ofeP4OQJZzbD9GDrYOk7KFnYhen96uPsYGbtkcu89etBdHVZRCT3UgASuZPyLWHQb+BZGq6ehGnBELb5js1rlyrMhG51MZlg9uYwZm48nWOliohIxigAidxNiaoweDX41bM+K2j247Bv4R2bt6npw6i2VQF4e+lBVh+8mFOViohIBigAidyLuzf0XwpVH4OUJPhxEPzxkXVKjXQMaV6eHg1LYxjw4tzd7D8fncMFi4jIvSgAidwPR1frAxIbDbN+/v1tWPKCdXLVfzGZTLzVsQbNK3lxPSmFQd9sJzz6Rg4XLCIid6MAJHK/zHbQ5j1o9xGYzLD7W/j+SUi4/QyPg52ZSb3qUdm7EBdjEhk0awfxuj1eRCTXUAASyaiGQ6DHXHBwg5NrYXoIRJ25rZmHswPT+zXAq5AjB8NjeHHOblIsujNMRCQ3UAASeRCVQ2DgcijkA5cPWe8QO7/rtmb+RV2Z2rc+TvZmQg9f4u1fD9qgWBER+TcFIJEH5RsAQ0KhRA2Iuwiz2sPhpbc1q1u6CJ92qwPArE2nmbXxVA4XKiIi/6YAJJIZnqVg4AqoGAzJ160Tqm6ZfNsdYu1q+fJaG+vt8W/9epDfD+v2eBERW1IAEsksZw/oMQ8CBwAGrHgdlr8KKWkHPT/bojzd6vtjMeD5H3Zz4IJujxcRsRUFIJGsYGcPj30Kj75t/bzta5jbExLjUpuYTCbe6VyTphWLWW+Pn7WDiOj0J1oVEZHspQAkklVMJmj6ovV5QfbOcGwlzGwLMeGpTRzszHzZK5BKJQoREZPAoG+26/Z4EREbUAASyWrVO0K/X8HVCyL+hGmtIGJf6teeLg7M6N+AYm6OHLgQw/C5uj1eRCSn2TwATZo0ibJly+Ls7ExQUBDbtm27a/uoqCiGDRuGr68vTk5OVK5cmWXLlqV+P27cOEwmU5pX1apVs3s3RNLyb2C9Q8yrMsSchxlt4Njqv78u6srUftbb41cfusS7Sw/ZsFgRkYLHpgFo3rx5jBw5krFjx7Jr1y4CAgIICQnh0qVL6bZPSkri0Ucf5fTp0yxcuJAjR44wdepUSpYsmaZdjRo1CA8PT31t2LAhJ3ZHJK0iZa2zyZdtDklx8ENX2DEj9et6pYvwSdc6AMzYeIpvN5+2SZkiIgWRvS03/sknnzBkyBAGDBgAwJQpU1i6dCkzZszg9ddfv639jBkzuHr1Kps2bcLBwQGAsmXL3tbO3t4eHx+f+64jMTGRxMTE1M8xMTEZ3BORO3ApAr1/gl+Gw94f4NeX4OpJCH4LzGba1/bl9JUqfLjyCGOXHKBUUVcerlLC1lWLiOR7NjsDlJSUxM6dOwkODv67GLOZ4OBgNm/enO46S5YsoXHjxgwbNgxvb29q1qzJe++9R0pKSpp2x44dw8/Pj/Lly9OrVy/OnLl9moJ/Gj9+PJ6enqkvf3//zO+gyC32jtDpS3j4DevnTZ/Dgr6QdB2A51pWoGv9Utbb47/fxaFwBXARkexmswAUGRlJSkoK3t7eaZZ7e3sTERGR7jonT55k4cKFpKSksGzZMkaPHs3HH3/MO++8k9omKCiIWbNmsWLFCiZPnsypU6do3rw5sbGxd6xl1KhRREdHp77Onj2bNTspcovJBC1ehSemgp0jHPoFvnkM4i5Zb4/vVIvG5YsRn5TCoFnbuRSj2+NFRLKTzQdBZ4TFYqFEiRJ8/fXXBAYG0q1bN9544w2mTJmS2qZt27Y89dRT1K5dm5CQEJYtW0ZUVBTz58+/Y79OTk54eHikeYlki9pdoe9i66Wx8zutd4hdOoyjvZkpvQOpUNyNC9EJDPpmB9eTdHu8iEh2sVkA8vLyws7OjosX004JcPHixTuO3/H19aVy5crY2dmlLqtWrRoREREkJSWlu07hwoWpXLkyx48fz7riRTKjTBMYtBqKlrfOIj+9NZxch6erAzP7N6SomyP7zkczfO4e3R4vIpJNbBaAHB0dCQwMJDQ0NHWZxWIhNDSUxo0bp7tO06ZNOX78OBaLJXXZ0aNH8fX1xdHRMd114uLiOHHiBL6+vlm7AyKZ4VXRGoL8G0FiNHz3BOz+ntLFXJnaNxBHezOrDl5k/DLdHi8ikh1segls5MiRTJ06lW+++YZDhw4xdOhQ4uPjU+8K69u3L6NGjUptP3ToUK5evcrw4cM5evQoS5cu5b333mPYsGGpbV555RXWrVvH6dOn2bRpE507d8bOzo4ePXrk+P6J3JVbMevlsJpdwHITFj8Hv79DYOkifPRUAADTNpziuy1hNi5URCT/selt8N26dePy5cuMGTOGiIgI6tSpw4oVK1IHRp85cwaz+e+M5u/vz8qVK3nppZeoXbs2JUuWZPjw4bz22mupbc6dO0ePHj24cuUKxYsXp1mzZmzZsoXixYvn+P6J3JODMzwxzfrMoPUfwx8fwrXTPN5xEmdaV+aj344ydskB/Iu60qKy/g6LiGQVk2EYGmTwLzExMXh6ehIdHa0B0ZJzdn0Lv46wng0q3Rij23e8svQ8P+46RyEnexYObUxVH/19FBG5k4z8/s5Td4GJ5Gv1+kCvheDkAWc2Y5r+KONbutKofFHiEm/Se9o21h29bOsqRUTyBQUgkdykwsPW6TM8S8PVkzjOfJSpLZOp6uNOZFwi/WZsY9ySAyQkp9y7LxERuSMFIJHcpkQ1GLwa/OrBjWu4z+vCkpYR9G9SFoBZm07z2Ocb2H8+2rZ1iojkYQpAIrmRuzf0XwpVH4OUJBwXDWFc0d/4ZmBDirs7cfxSHJ2/3MiXa4/rWUEiIg9AAUgkt3J0ha6zofHz1s+rx9HC2MnKEQ/RpoYPySkGH6w4Qo+vt3D26nXb1ioikscoAInkZmY7CHkXgoZaPy9+jqKWa0zuXY8PnqyNm6Md205fpe1n6/lp1zl0U6eIyP1RABLJC4LHgXdNuH4FFj2LyTDoWt+f5cMfIrBMEeISbzJy/l6e/2E3UdfTnxZGRET+pgAkkhc4OEOX6WDvDCd+h62TAShdzJV5TzfildaVsTebWLovnJAJf7DhWKSNCxYRyd0UgETyihJVrZfDAFaPg/A/AbC3M/P8I5X46bkmlC/uxsWYRHpP38qbv+h2eRGRO1EAEslL6g+CKu0gJQl+HAxJfw9+rl2qMEtfaE6fRmUAmLnxNI9/sYGDF2JsVa2ISK6lACSSl5hM8PjnUMgbIo/Ab/9N87WLox1vd6rJzP4N8CrkxNGLcXSctIGv1p3Q7fIiIv+gACSS17h5Qecp1vc7psPhZbc1ebhqCVaOaM6j1b1JTjEYv/wwPadu4XzUjRwuVkQkd1IAEsmLKjzy9/OBFg+DmPDbmhQr5MTXfQJ5v0stXB3t2HrqKm0m/MHiPedzuFgRkdxHAUgkr2o1BnxqwY2rsOhZsFhua2IymejWoDTLXmxO3dKFiU24yfC5e3hhzm6iryfboGgRkdxBAUgkr7J3gi4zwN4FTq6FLZPu2LSslxsLnmnMS8GVsTOb+GXvBdp89gebjut2eREpmBSARPKy4pWhzXjr+9VvQvjeOza1tzMzPLgSC59tTDkvN8KjE+g5bSvv/HpQt8uLSIGjACSS1wX2t06aakm+7db49NQtXYSlLzajZ1BpAKZtOEWnSRs5djE2B4oVEckdFIBE8rpbt8a7+0LkUVj5f/dcxdXRnvc612Ja3/oUc3PkcEQsvaZt5WJMQg4ULCJiewpAIvmBa9G/bo03wc6ZcOiX+1otuLo3K0Y8RGXvQlyKTeTp2Tt0OUxECgQFIJH8onxLaPKC9f2SFyDmwn2tVtzdiWl9G1DY1YG956J5deGfmlVeRPI9BSCR/OSR0eAbADeuwc/p3xqfntLFXPmyVz3szSaW7L3Al2tPZHOhIiK2pQAkkp/YO1pnjXdwhVPrYPPn971qkwpejHu8BgAfrjzCbwcisqtKERGbUwASyW+8KkGb/1nfh74NF3bf96q9G5Whb2PrZKoj5u3hULgmUhWR/EkBSCQ/qtcXqnX4x63x8fe96ujHqtOkQjGuJ6Uw+JsdXIlLzMZCRURsQwFIJD8ymaDDRHD3gyvHYcXr972qg52ZL3vVo0wxV85H3WDod7tIunl/Y4lERPIKBSCR/Mq1KDzxNWCCXbPh4OL7XrWwqyPT+9XH3cmebaevMmbxft0ZJiL5igKQSH5Wrjk0G2F9v+RFiL7/meArlnBnYs+6mE0wd/tZZm06nS0liojYggKQSH7X8v/Ary4kRMHPz4Dl/h90+HCVEoxqWw2At389yB9HL2dTkSIiOUsBSCS/S7013g1Or4eNn2Vo9cHNy/FkYCksBgz7YRcnLsdlU6EiIjlHAUikIChWAdp9YH2/5l04v/O+VzWZTLzbuSaBZYoQm3CTId/sIPp6cjYVKiKSMxSARAqKOr2geiew3IQfh0Di/Z/JcbK3Y0rvQPw8nTkZGc/zc3ZxM0V3holI3qUAJFJQmEzQYQJ4lIKrJ2DFaxlavbi7E1P71cfFwY71xyJ5d9mh7KlTRCQHKACJFCQuReCJrwAT7P4ODvycodVr+HnySdcAAGZuPM3cbWeyoUgRkeynACRS0JRtBs1ftr7/ZThEnc3Q6m1r+fJScGUARi/ez7ZTV7O6QhGRbKcAJFIQtXwdStaHhOgM3xoP8GKrirSv7UtyisGz3+3k7NXr2VSoiEj2UAASKYjsHKDLVHAsBGEbYcOnGVrdZDLx0ZMB1CzpwdX4JIbM3kFc4s1sKlZEJOspAIkUVEXLQ7sPre/XvAfndmRodRdHO6b2rU9xdycOR8Ty0rw9WCyaLkNE8gYFIJGCLKAH1OwCRgr8OMh6SSwDfD1d+KpPII72ZlYdvMjHq45kU6EiIllLAUikIDOZoP0nULg0XDsNv4yADE56Wq90Ef73RC0AJq05weI99z/fmIiIrSgAiRR0LoWhywww28OBn2D3txnu4ol6pXimRXkAXl34J3vPRmVtjSIiWUwBSETAvwE8Mtr6ftmrcOlwhrt4NaQqraqWIPGmhSGzdxARnZDFRYqIZB0FIBGxavIiVHgEbt6AhQMg+UaGVrczm5jQvQ6VvQtxKTaRp7/dQUJyxm6vFxHJKQpAImJlNkPnr8CtBFw6CCv/L8NduDs7MK1vA4q4OvDnuWheXfgnRgbHFImI5AQFIBH5W6ES8MTXgAl2zIADizLcRelirnzZKxB7s4kley/w5doTWV6miEhmKQCJSFoVHoZmL1nfL3kRroVluIvGFYrxZscaAHy48gi/HYjIygpFRDJNAUhEbvfw/0GphpAYbX0+UEpyhrvoFVSGvo3LADBi3h72n8/YM4ZERLKTApCI3M7OAbpMA2dPOLcd1rz7QN2Mfqw6TSsW43pSCn2mb+VQeEwWFyoi8mAUgEQkfUXKwOOfW99v+BSOh2a4Cwc7M5N7BxLgX5hr15PpNW0rRyJis7hQEZGMUwASkTur3hHqD7K+//kZiL2Y4S48nB2YPbAhtUp6cjU+iV7TtnDsokKQiNiWApCI3F3Iu1CiBsRftoYgiyXDXXi6OPDtoIbU8PMgMi6JHlO3cvxSXDYUKyJyfxSAROTuHFzgqZng4Aon18Cmzx6om8Kujnw3KIiqPu5ExiXSc+oWTkXGZ3GxIiL3RwFIRO6teBVo+4H1fejbcHbbA3VTxM2R7wcHUcXbnUuxifT4egthVxSCRCTnKQCJyP2p2xtqPglGCiwcBDeuPVA3xQo58f2QICqVKERETAI9vt7C2avXs7hYEZG7UwASkftjMsFjn0KRshB9xvqQxAec5sKrkBM/DGlEheJuXIhOoLtCkIjkMAUgEbl/zh7w5EwwO8ChJdbpMh5QcXcn5gxpRHkvN85H3aDntC2cj8rYBKwiIg9KAUhEMqZkPQgeZ32/YhRE7H/grkp4OPPDkEaUKebK2as36Dl1C+HRCkEikv0UgEQk4xo9B5VaQ0oiLBwISQ8+kNnH05k5QxrhX9SFsCvX6Tl1KxdjErKwWBGR2ykAiUjGmc3QaTIU8oHII7D8tUx151fYhTlDGlGqiAunIuPpMXULl2IVgkQk+ygAiciDcfOCLlMBE+z+FvYtzFR3pYq4MmdII0oWduHk5Xh6Tt3K5djErKlVRORfFIBE5MGVewge+o/1/S8j4OrJTHXnX9SVH4YE4evpzPFLcfSatoUrcQpBIpL1FIBEJHNavAalG0NSrHU80M2kTHVXppgbPwxphLeHE0cvxtFr2lauxmeuTxGRf1MAEpHMsbOHLtPAuTBc2A2hb2a6y3Je1hBU3N2JwxGx9J62lajrCkEiknUUgEQk8zxLWQdFA2z+Ao7+lukuKxQvxJwhQXgVcuRgeAy9p28l+npypvsVEQEFIBHJKlXbQdCz1veLnoWY8Ex3WbGEOz8MaUQxN0f2n4+h74ytRN9QCBKRzFMAEpGs8+hb4FMbrl+Bn4aAJSXTXVb2duf7IUEUcXVg77lo+s3YRmyCQpCIZI4CkIhkHXsn61QZDm5wej2s/zhLuq3q48H3gxtR2NWBPWej6D9zO3GJN7OkbxEpmBSARCRreVWExz6xvl87HsI2ZUm31f08+G5QEB7O9uwMu8aAmduIVwgSkQekACQiWS+gOwT0AMMCPw6G61ezpNuaJT35bnAQ7s72bD99jYGztnM9SSFIRDJOAUhEske7j6BoBYg5D4uHgWFkSbe1SxVm9sCGuDvZs/XUVQZ/s4MbSZkfayQiBYsCkIhkD6dC8NRMsHOEI8usg6Kz6ExQ3dJFmDWwIW6Odmw6cYXBs7frOUEikiE2D0CTJk2ibNmyODs7ExQUxLZt2+7aPioqimHDhuHr64uTkxOVK1dm2bJlmepTRLKJb4D1TJDJDPsWwJeN4PCye693HwLLWEOQq6MdG49foc2E9Ww4FpklfYtI/mfTADRv3jxGjhzJ2LFj2bVrFwEBAYSEhHDp0qV02yclJfHoo49y+vRpFi5cyJEjR5g6dSolS5Z84D5FJJsF9oNBq8CrMsRdhLk94MesORvUoGxR5j3dmPJebkTEJNB7+lbe/vUgCcm6JCYid2cyjCy6MP8AgoKCaNCgAV988QUAFosFf39/XnjhBV5//fXb2k+ZMoUPP/yQw4cP4+DgkCV9picmJgZPT0+io6Px8PB4wL0TkTSSE2Dte7Dpc+vgaLcS0GECVG2f6a6vJ93k3aWH+H7rGQCqeLvzabc6VPfTz69IQZKR3982OwOUlJTEzp07CQ4O/rsYs5ng4GA2b96c7jpLliyhcePGDBs2DG9vb2rWrMl7771HSkrKA/cJkJiYSExMTJqXiGQxB2frgxIHrQKvKhB/Ceb2zJK7xFwd7Xm3cy1m9K+PVyFHjlyMpdOkjXz9xwksFpv9G09EcjGbBaDIyEhSUlLw9vZOs9zb25uIiIh01zl58iQLFy4kJSWFZcuWMXr0aD7++GPeeeedB+4TYPz48Xh6eqa+/P39M7l3InJHperDM39A0xF/jw2aFASHfs10149U9WbFiIcIruZNUoqF95Ydpue0LZyPupH5ukUkX7H5IOiMsFgslChRgq+//prAwEC6devGG2+8wZQpUzLV76hRo4iOjk59nT17NosqFpF0OTjDo2/CoNV/nw2a1ytLzgZ5FXJiat9Axj9RCxcHO7acvEqbCX+weM/5LCpeRPKDBwpAZ8+e5dy5c6mft23bxogRI/j666/vuw8vLy/s7Oy4ePFimuUXL17Ex8cn3XV8fX2pXLkydnZ2qcuqVatGREQESUlJD9QngJOTEx4eHmleIpIDSgVazwY1eylLzwaZTCZ6NCzNsuHNqeNfmNiEmwyfu4cX5+zWjPIiAjxgAOrZsydr1qwBICIigkcffZRt27bxxhtv8NZbb91XH46OjgQGBhIaGpq6zGKxEBoaSuPGjdNdp2nTphw/fhyLxZK67OjRo/j6+uLo6PhAfYqIjTk4Q/A469mg4lX/Phu0cFCmzwaV83Jj4bONGRFcCTuziSV7L9Dmsz/YdEK3y4sUdA8UgPbv30/Dhg0BmD9/PjVr1mTTpk18//33zJo16777GTlyJFOnTuWbb77h0KFDDB06lPj4eAYMGABA3759GTVqVGr7oUOHcvXqVYYPH87Ro0dZunQp7733HsOGDbvvPkUklyoVCE+v+/ts0P6FMKkhHPolU93a25kZEVyZhc82pmwxV8KjE+g1bSvvLj1I4k3dLi9SUNk/yErJyck4OTkBsHr1ah5//HEAqlatSnh4+H33061bNy5fvsyYMWOIiIigTp06rFixInUQ85kzZzCb/85o/v7+rFy5kpdeeonatWtTsmRJhg8fzmuvvXbffYpILnbrbFC1DrDoObh8GOb1hppdoO2H4FbsgbuuW7oIS19szjtLDzFn2xmmrj/F+mORTOheh6o+uuwtUtA80HOAgoKCePjhh2nfvj2tW7dmy5YtBAQEsGXLFp588sk044PyIj0HSCQXuJkIa/8HGyf89dyg4tD+E6j+eKa7XnXwIq//+CdX4pNwtDPzapsqDGxaDrPZlPm6RcRmsv05QO+//z5fffUVLVu2pEePHgQEBADW5/TcujQmIpIp9k4QPBYGr4bi1SD+MszvAwsHQvyVTHX9aHXr7fKPVC1BUoqFd5Yeovf0rYRH63Z5kYLigZ8EnZKSQkxMDEWKFElddvr0aVxdXSlRokSWFWgLOgMkksvcTIR178OGCWCkZNnZIMMw+GHbGd759RA3klPwcLY+ULFDgF/W1C0iOSojv78fKADduHEDwzBwdXUFICwsjJ9//plq1aoREhLyYFXnIgpAIrnU+V1/jQ06ZP1c4wnrZKuZGBsEcPJyHC/N28Pec9EAdKrjx5sda+Lpkv6UOyKSO2X7JbCOHTsye/ZswDo7e1BQEB9//DGdOnVi8uTJD9KliMi9lawHz6yD5i+DyQ4O/GS9U+zgkkx1W754IRYObcKLrSphNsGiPRdoO+EPNp/I3KU2Ecm9HigA7dq1i+bNmwOwcOFCvL29CQsLY/bs2UycODFLCxQRScPeCVqNsY4NKlEdrkdaxwbtnZupbh3szIx8tDILnm1C6aKuXIhOoOe0LYxffki3y4vkQw8UgK5fv467uzsAv/32G0888QRms5lGjRoRFhaWpQWKiKSrZD14ei3UH2T9vOQFCLvzpMf3K7BMEZYNb063+v4YBny17iSPf76R7acz91BGEcldHigAVaxYkUWLFnH27FlWrlxJ69atAbh06ZLGzIhIzrF3so4BqtYBUpKss8tfPZnpbgs52fP+k7X5qk8gRd2ss8s/NWUzryzYS2RcYhYULiK29kABaMyYMbzyyiuULVuWhg0bpk4z8dtvv1G3bt0sLVBE5K7MZuj8NfjVhRtX4fuucONalnQdUsOH0JEt6NHQH4CFO8/xyEdr+W5LGCmWB7qBVkRyiQe+DT4iIoLw8HACAgJSn9a8bds2PDw8qFq1apYWmdN0F5hIHhQbAVMfgZjzUO4h6P0T2GXdXVy7zlzjvz/v52B4DAABpTx5p1MtapXyzLJtiEjmZPtt8P9066nPpUqVykw3uYoCkEgeFbEPpodAcjzU6wsdJoIp657ufDPFwndbwvj4t6PEJt7EZILeQWV4pXUVPF11y7yIrWX7bfAWi4W33noLT09PypQpQ5kyZShcuDBvv/12mpnaRURylE8teHKGdTLVXbNh0+dZ2r29nZn+TcsR+nILOtXxwzDg2y1htPpkLT/uPEcm/z0pIjnogQLQG2+8wRdffMH//vc/du/eze7du3nvvff4/PPPGT16dFbXKCJy/6q0gdbvWt+vGgOHfs3yTZTwcGZC97r8MCSICsXdiIxL4uUFe+n29RaORMRm+fZEJOs90CUwPz8/pkyZkjoL/C2LFy/mueee4/z581lWoC3oEphIHmcYsPRl2DEdHFxhwHLwq5Mtm0q6aWH6hlNMDD3GjeQU7MwmBjUrx/BWlXBzss+WbYpI+rL9EtjVq1fTHehctWpVrl7VszJExMZMJmj7AVR4BJKvw5zuEHMhWzblaG9maMsKrH65BSE1vEmxGHz9x0lafbyOZfvCdVlMJJd6oAAUEBDAF198cdvyL774gtq1a2e6KBGRTLOzh6dmQfGqEBsOP3SDxLhs21zJwi581ac+M/rXx7+oCxExCTz3/S76zdzOqcj4bNuuiDyYB7oEtm7dOtq3b0/p0qVTnwG0efNmzp49y7Jly1KnycirdAlMJB+5dhqmtrJOmVGlHXT7Dsx22brJhOQUvlx7gilrT5CUYsHRzsyzLSvwXMsKODtk77ZFCrJsvwTWokULjh49SufOnYmKiiIqKoonnniCAwcO8O233z5Q0SIi2aJIWegxB+yc4Mgy68DobObsYMfIRyuz8qWHaF7Ji6QUCxNDj9H60z9Yc/hStm9fRO4t088B+qe9e/dSr149UlLy9sSBOgMkkg/tWwg//jVvWIfPILB/jmzWMAyW74/grV8OEhGTAEBIDW/GdKhBycIuOVKDSEGR7WeARETynFpPQsv/s75f+jKcWJMjmzWZTLSr5cvql1swpHk57MwmVh64SPDH65i89gRJN/XsNBFbUAASkYKjxatQuxtYbsL8fnD5SI5tupCTPW+0r87SF5vRoGwRbiSn8P6Kw7SbuJ6dYbp7ViSnKQCJSMFhMsHjn4N/I0iMhh+6QvyVHC2hqo8H859pzEdPBVDMzZHjl+J4aspmPlt9jJspOhskklMyNAboiSeeuOv3UVFRrFu3TmOARCR3i4+Eaa2sd4j5N4J+S8DeKcfLiL6ezLhfDvDzbuvDYxuULcKn3epQqohrjtcikh9k2xggT0/Pu77KlClD3759M1W8iEi2c/OCnvPByRPOboElL1ifHp3DPF0d+LRbHT7tFkAhJ3u2n75G28/W8+uf2fPQRhH5W5beBZZf6AyQSAFxYg181wWMFHj4DesYIRs5c+U6L87dzZ6zUQA8FViKcY/X0HQaIhmgu8BERO5HhYfhsU+s79e8a71V3kZKF3NlwbONeeGRiphMsGDnOR77fAP7zkXbrCaR/EwBSEQKtsD+0Ph56/tFz8HZbTYrxcHOzMutqzBnSCN8PZ05FRnPE5M38tW6E1gsOlkvkpUUgEREHn0LqrSHlESY0wOuhdm0nEbli7F8eHPa1PAhOcVg/PLD9JmxlYt/PUhRRDJPAUhExGwHXaaCT23rnGE/dIME2156KuzqyOTe9fjfE7VwcbBj4/ErtJnwB6sOXrRpXSL5hQKQiAiAoxv0nAfuvnD5ECzoDyk3bVqSyWSie8PS/PJCM6r7enDtejJDZu9g9KL9JCTn7ceNiNiaApCIyC0eftBjLji4wonfYfmrNrk9/t8qlijEz8OaMKR5OQC+3RLG419s4HBEjI0rE8m7FIBERP7Jrw50mQaYYMd02DrF1hUB4GRvxxvtq/PNwIZ4FXLi6MU4Hv9iI99sOo2eZiKScQpAIiL/VrU9tH7b+n7l/8GR5bat5x9aVC7OihHNebhKcZJuWhi75ACDv9nBlbhEW5cmkqcoAImIpKfx81CvHxgWmNsTfhsNyTdsXRUAXoWcmNG/AeM6VMfR3kzo4Uu0+Ww9649dtnVpInmGApCISHpMJmj/MQT0tIagTRNhSjM4s8XWlQHWAdL9m5Zj8bCmVCpRiMuxifSZvo33lh0i6aYmVRW5FwUgEZE7sXOAzpOtA6PdfeHKcZjRBpa/Dknxtq4OgGq+Hix5vhm9G5UG4Os/TvLE5I2cuBxn48pEcjcFIBGRe6nSFp7bAnV6AwZsnQyTm8Cp9bauDAAXRzve6VSLr/sEUtjVgf3nY3hs4gbmbT+jAdIid6DJUNOhyVBF5I6Or4YlwyHmnPVz/UHw6Jvg5G7buv4SEZ3AyPl72HTiCgDta/nyXudaeLo62LgykeynyVBFRLJLxWB4bjPUH2j9vGM6fNkYjofatq6/+Hg6892gIF5vWxV7s4ml+8JpN3E9O05ftXVpIrmKzgClQ2eAROS+nFwHS16AqL/mDqvbB1q/Ay6FbVrWLXvPRvHi3N2EXbmOndnEiFaVeO7hitiZTbYuTSRb6AyQiEhOKN8Chm6CoGcBE+z+1no26OhKW1cGQIB/YX59oRmd6viRYjH4eNVRek/bSkS0JlUV0RmgdOgMkIhkWNhmWDwMrp6wfq7dDdr8D1yL2rYuwDAMftx1njGL93M9KYUirg589FQArap527o0kSylM0AiIjmtTGMYuhGavAAmM/w5DyYFwaFfbF0ZJpOJJwNL8esLzajhZ51UddA3Oxi35ACJNzWpqhRMOgOUDp0BEpFMObcDFj0HkUesn2t0hnYfgZuXbesCEm+m8P7yI8zYeAqA6r4efN6zLhWKF7JxZSKZpzNAIiK2VKo+PLsemr8MJjs48DNMagj7f7T57PJO9naM6VCdGf3rU9TNkYPhMXT4fAMLdpzVM4OkQNEZoHToDJCIZJkLu2Hx83Bxv/Vz1ceg/SfgbvvxNxdjEnhp3t/PDHo8wI93O9fE3VnPDJK8SWeARERyC7+6MGQNtBwFZns4/Kv1bNDeuTY/G+Tt4cy3g4L4T0gV7Mwmluy9QPuJG9hzNsqmdYnkBJ0BSofOAIlItojYD4ufg/C91s+VQuCxT8GzpG3rAnaGXePFObs5H3UDe7OJ/4RUYUjz8pj1zCDJQ3QGSEQkN/KpCYN/h1ZjwM4Rjq2ELxvBnwtsXRmBZYqwbHhz2tfy5abFYPzyw/SftZ3LsYm2Lk0kWygAiYjkJDt76+DoZ9ZDyfqQGAM/DbbOMJ+SbNPSPF0c+KJnXcY/UQtnBzN/HL1M28/+4I+jl21al0h2UAASEbGFElVh0G/w0H+sn7dOhtmdIO6STcsymUz0aFiaX55vRlUfdyLjkug7Yxvjlx0i6abFprWJZCUFIBERWzHbwSP/hW7fg6M7hG2Ar1pYnyNkY5W83Vk0rCl9GpUB4Ks/TvLUlE2EXYm3cWUiWUMBSETE1qo9BkN+B6/KEHsBZraFnbNsXRXODna83akmX/UJxNPFgb3nomk/cQOL95y3dWkimaYAJCKSGxSvDINDrc8JSkmCX4bDkhfhpu0HIYfU8GH58OY0LFuUuMSbDJ+7h1cW7CU+8aatSxN5YApAIiK5hbMHdP0WHhkNmGDXNzCzHUTb/oyLX2EXfhgSxIjgSphNsHDnOTp8voH956NtXZrIA1EAEhHJTcxmeOgV6LUQnAvD+R3wdQs4vdHWlWFvZ2ZEcGXmDGmEr6czJyPj6TJ5E7vOXLN1aSIZpgAkIpIbVQqGp9eCd02IvwyzH4ctU2z+9GiAoPLFWPZic5pX8iLxpoUh3+zg7NXrti5LJEMUgEREcqui5ay3ytd8Eiw3YcVr8PMzkGT7sFHEzZEpvQOp4efBlfgk+s/cRvR12z7HSCQjFIBERHIzRzfoMg1Cxltnlv9zHsxoDddO27oy3JzsmdG/Ab6ezpy4HM/Q73fqWUGSZygAiYjkdiYTNH4O+i4GVy+I2Gd9XtDxUFtXhreHM9P7NcDN0Y5NJ67w30X70BSTkhcoAImI5BXlmsMzf0DJQEiIgu+6wPqPbT4uqLqfB1/0rIfZBPN3nOPLtSdsWo/I/VAAEhHJSzxLQv9lUK8vYEDoWzC/DyTG2rSsh6uW4M3HawDw4coj/LL3gk3rEbkXBSARkbzGwRke/xwemwBmBzj0C0xtBZHHbFpWn8ZlGdSsHAAvL9jLzrCrNq1H5G4UgERE8qr6A2DAcnD3hcgj8PXDcHipTUv6v3bVeLS6N0k3LQyZvVNzh0mupQAkIpKX+TeAp9dB6SaQFAtze8Lv74AlxSbl2JlNfNa9DrVKenI1PokBs7YTdT3JJrWI3I0CkIhIXufuDf2WQMNnrJ//+BB+6AY3bPOEZldHe6b3q4+fpzMnL8fzzLe6PV5yHwUgEZH8wM4B2n0Anb8Ce2c4vsp6SeziAZuUU8LDmen9G1DIyZ6tp67y+k9/6vZ4yVUUgERE8pOA7jBwJXiWhmunYFowbJ9uk1vlq/l6MKlXPezMJn7adZ7Pfz+e4zWI3IkCkIhIfuNXB55ZB+UfhuTrsHQkfNsJos7keCktKhfnrY7W2+M/WXWUxXtsP7O9CCgAiYjkT65FofeP1ik07J3h5Fr4sgns/CbHzwb1CirD0w+VB+A/C/5k+2ndHi+2pwAkIpJfme2sU2g8uxFKNbTeJfbLi9YnSEfn7JmY19tUpU0NH5JSLDw9ewenInV7vNiWApCISH7nVREGroBH3wY7JzgRCl82gt3f5djZILPZxKfd6hBQypNr15MZOGs71+J1e7zYTq4IQJMmTaJs2bI4OzsTFBTEtm3b7th21qxZmEymNC9nZ+c0bfr3739bmzZt2mT3boiI5F5mO2j6Ijy7wTqXWGIMLB4GP3SFmJyZtsLF0Y6p/epTsrALpyKtt8cn3rTN84pEbB6A5s2bx8iRIxk7diy7du0iICCAkJAQLl26dMd1PDw8CA8PT32FhYXd1qZNmzZp2syZMyc7d0NEJG8oXhkG/gbB48DOEY79Zj0btHdujpwNKuHuzIz+DXB3smfb6au8tlC3x4tt2DwAffLJJwwZMoQBAwZQvXp1pkyZgqurKzNmzLjjOiaTCR8fn9SXt7f3bW2cnJzStClSpMgd+0tMTCQmJibNS0Qk37Kzh2YvWWeW96sLCdHw8zMwpwfERmT75qv4uPNlb+vt8Yv2XGDCatvOYSYFk00DUFJSEjt37iQ4ODh1mdlsJjg4mM2bN99xvbi4OMqUKYO/vz8dO3bkwIHbH/S1du1aSpQoQZUqVRg6dChXrly5Y3/jx4/H09Mz9eXv75+5HRMRyQtKVINBq+GR0dZJVY8uh0lB8OeCbD8b1LxScd7pVBOAz0KP8ePOc9m6PZF/s2kAioyMJCUl5bYzON7e3kREpP+vkCpVqjBjxgwWL17Md999h8VioUmTJpw79/cPT5s2bZg9ezahoaG8//77rFu3jrZt25KSkv615lGjRhEdHZ36Onv2bNbtpIhIbmZnDw+9Yn1ukG8AJETBT4NhXm+Iu/NQhKzQo2Fpnm1RAYDXf/qTLSfv/A9VkaxmMmx48fXChQuULFmSTZs20bhx49Tlr776KuvWrWPr1q337CM5OZlq1arRo0cP3n777XTbnDx5kgoVKrB69WpatWp1zz5jYmLw9PQkOjoaDw+P+98hEZG8LCUZNnwK694Hy01wKQrtP4aaT2TbJi0Wg+fn7GLZvgg8XRz46bkmVCheKNu2J/lbRn5/2/QMkJeXF3Z2dly8eDHN8osXL+Lj43NffTg4OFC3bl2OH7/zI9bLly+Pl5fXXduIiBR4dg7Q4lUYsga8a8GNq7BwAMzvB/GR2bJJs9nEJ13rUMe/MNE3rLfHX9Xt8ZIDbBqAHB0dCQwMJDQ0NHWZxWIhNDQ0zRmhu0lJSWHfvn34+vresc25c+e4cuXKXduIiMhffGvDkN+hxWtgsoODi6xjgw4uzpbNOTvYMbVvfUoVcSHsynWenr2DhGTdHi/Zy+Z3gY0cOZKpU6fyzTffcOjQIYYOHUp8fDwDBgwAoG/fvowaNSq1/VtvvcVvv/3GyZMn2bVrF7179yYsLIzBgwcD1gHS//nPf9iyZQunT58mNDSUjh07UrFiRUJCQmyyjyIieY69Izz8f9YgVKIGXI+E+X1h4UC4nvVTWRR3d2Jm/wa4O9uzI+wa/1n4JxaLbo+X7GPzANStWzc++ugjxowZQ506ddizZw8rVqxIHRh95swZwsPDU9tfu3aNIUOGUK1aNdq1a0dMTAybNm2ievXqANjZ2fHnn3/y+OOPU7lyZQYNGkRgYCDr16/HycnJJvsoIpJn+dWBp9dA81esZ4P2/2g9G3R4aZZvqpK3O1N6B2JvNvHL3gt8uvpolm9D5BabDoLOrTQIWkQkHed3wqLn4PJh6+daXaHt+9aJV7PQ/O1nefXHPwF4u1NN+jQqk6X9S/6VZwZBi4hIHlIyEJ5eZ32IoskM++bDlGYQdSZLN9O1gT/DHrbeHj960X6Gfb9LA6MlyykAiYjI/XNwtk6jMWgVFC0PMefhh+6QkLVP0H/50SqMCK6EndnE0n3htP50Hb8dyP6nVEvBoQAkIiIZV6o+9PsFCnnDpQPWwdEpN7Ose7PZxIjgyix6rimVvQsRGZfE09/uZOT8PUTfSM6y7UjBpQAkIiIPxrMU9JgL9i5wfBWsHHXvdTKoVilPljzfjGdalMdkgp92nSfk0z9Yd/Rylm9LChYFIBEReXAl60GXqYAJtn0NW7/K8k04O9gxqm01Fj7bmLLFXImISaDfjG2M+mkfcYlZd9ZJChYFIBERyZxqHeDRN63vV7wOR1dmy2YCyxRl+fCH6N+kLABztp2hzYQ/2HxCc4hJxikAiYhI5jV5Eer1BcNiHQ8UsS9bNuPiaMe4x2vww5AgShZ24dy1G/SYuoU3fznAjSQ9PVrunwKQiIhknskE7T+Bcg9BUhz80A1is++urSYVvFj50kP0aOgPwMyNp2k3cT07w65l2zYlf1EAEhGRrGHnAF1ng1flv26P7wZJ8dm2uUJO9ox/ojazBjTA28OJU5HxPDVlE/9bfpjEmzobJHenACQiIlnHpQj0nAcuRSF8D/z0NFgs2brJllVK8NuIFjxRtyQWA6asO8Hjn29k//nobN2u5G0KQCIikrWKlofuP4CdIxz+FULHZfsmPV0d+KRbHb7qE4hXIUeOXIyl06SNTFh9lOSU7A1gkjcpAImISNYr0xg6TrK+3/gZ7PwmRzYbUsOHlSMeol0tH25aDCasPkbnLzdyJCI2R7YveYcCkIiIZI/aXaHF69b3S0fCybU5stlihZyY1LMeE3vUpbCrA/vPx9Dh8w1MXnuCFIvm/xYrBSAREck+LV+Hmk+C5SbM6wuXj+TIZk0mE48H+PHbiIdoVbUESSkW3l9xmKembOLk5bgcqUFyNwUgERHJPiaT9VKYfxAkRsMPXSE+Msc2X8LDmWn96vPBk7Vxd7Jn15ko2k1cz4wNp7DobFCBpgAkIiLZy8HZOii6cBm4dhrm9oLkhBzbvMlkomt9f1a89BDNKnqRkGzhrV8P0nPaFi5E3cixOiR3UQASEZHs5+YFvRaAkyec3QJLngcjZ8/AlCzswreDGvJ2p5q4Otqx5eRVOk3ayIELul2+IFIAEhGRnFG8CnSbDWZ72LcA1r2f4yWYTCb6NCrD8uHNqeLtzqXYRLpO2azZ5QsgBSAREck55VtC+4+t79eOhz/n26SMMsXcmP9sYxqXL0Z8UgoDZ21n/vazNqlFbEMBSEREclZgf+vkqQCLh8GZLTYpw9PFgW8GNqRTHT9SLAav/vgnn646ipHDl+bENhSAREQk5wW/CVUfg5QkmNsTrp60SRmO9mY+7VaHYQ9XAOCz0GO8uvBPPT26AFAAEhGRnGc2wxNfg28duH7FOnHqDdvM5G4ymfhPSFXe61wLswkW7DzHwFnbiU1Itkk9kjMUgERExDYc3aDHXPAoCZFHYX5fSLFd6OgZVJpp/erj4mDH+mORdP1qCxdjcu52fclZCkAiImI7Hr7W2eMdC8GpP+DXl3L89vh/eqSqN/OeaYRXIUcOhcfQedJGjl7UPGL5kQKQiIjYlk8teHIGmMyw+1vYNDHrt2FJgaun4Nhq2DIZlr4MCwdB2KbbmtYuVZifn2tK+eJuXIhOoMvkTWw6kXNPr5acYTI03P02MTExeHp6Eh0djYeHh63LEREpGLZMgRWvASboOhuqP56x9Q3DOp7oynHrK/LY3++vnrQOuP43O0dr+KrW4bavrsUnMWT2DnaEXcPBzsRHTwXQsU7JB9s3yREZ+f2tAJQOBSARERswDFj2H9g+FexdYMAyKFnv9nbJN+DKib/CzTHr+1thJyHqzv3bOULRClCsAnhVgosH4dhK65mnDhOhXp/bVklITmHk/D0s2xcBwKttqjC0RQVMJlMW7bRkJQWgTFIAEhGxkZSbMKcbHF8NhbytD02MuZD2bE70PR5Y6OlvDTnFKkGxitaXV0XrcrNd2m39Ohx2f2f93PodaPLCbd1ZLAbvLjvE9A2nAOgVVJo3H6+BvZ1GkeQ2CkCZpAAkImJDCTEwIwQuHbxzG2fPvwOO118hp1glKFoeHF3vf1uGAatGw6bPrZ+bjYRWY6yz2P/LjA2neHvpQQwDWlUtwec96+LqaJ/BnZPspACUSQpAIiI2FnXGOmv8zUTr5ap/ntHxqgSuxdINKQ9sw6ewepz1feAA65mnf54t+suK/eEMn7uHxJsWAkp5Mq1fA4q7O2VdHZIpCkCZpAAkIlIA7ZhpvQ0fA2p0hs5fg73jbc12hl1l8Dc7uHY9Gf+iLswa0JAKxQvlfL1ym4z8/tYFTBEREYD6A6x3hJkd4MDPMKc7JMXf1iywTFF+HNqE0kVdOXv1Bl0mb2LH6as2KFgyQwFIRETklppPQM+54OAKJ0Jhdie4fnu4KV+8ED8914SAUp5EXU+m57StLN8XnvP1ygNTABIREfmnisHQdzE4F4Zz22BWe4iNuK2ZVyEn5jzdiOBqJUi6aeG5H3al3ikmuZ8CkIiIyL/5N4QBy6GQj/VutOmt052x3tXRnq/61Kd3o9IYBrz960He+uUgFouG1+Z2CkAiIiLp8a4OA1dAkbIQFQYz2sDFA7c1szObeLtjTV5rUxWAGRtPMeyHXSQkp+RwwZIRCkAiIiJ3UrQcDFwJJWpA3EWY2RbObL2tmclkYmjLCnzWvQ6OdmaW74+g17StRN+w3ez2cncKQCIiInfj7gMDloJ/ECREw7edrE+qTkfHOiWZPaghHs727Ay7Rt8Z24hNUAjKjRSARERE7sWlCPT52TpAOvk6/NAd9v+YbtNG5Ysx75nGFHZ1YO/ZKAbM3E584s0cLljuRQFIRETkfji6Qfc5UOMJsCTDwkGwY0a6Tav5evDdoCDcne3ZEXaNwd/s4EaSxgTlJgpAIiIi98veEbpMg/oDAcP65Oj1H1vnFPuXmiU9mT2wIYWc7Nl88gpPf7tDA6NzEQUgERGRjDDbQftPoPkr1s+hb8Fv/003BNUtXYSZAxrg4mDH+mORDPt+F0k3LTlcsKRHAUhERCSjTCZoNRpC3rN+3vwFLH4eUm4f69OgbFGm96uPk72Z0MOXeHHObpJTFIJsTQFIRETkQTUeBh2/BJMZ9nwHC/pBcsJtzZpU9OLrvvVxtDOz4kAEI+fvJUUPS7QpBSAREZHMqNsLun4Ldo5w+Ff44SlIjL2tWYvKxZncux72ZhO/7L3Afxbu1ROjbUgBSEREJLOqPQa9fwTHQnDqD/imA8Rfua1Zq2refNGzLnZmEz/tOs8bi/YpBNmIApCIiEhWKPcQ9PsFXIrChd0wsw1cOnxbszY1ffm0Wx3MJpiz7SzjfjmAkc4AasleCkAiIiJZpWQ96/xhHiUh8ih89RBs/AwsaW9/fzzAjw+fDMBkgtmbw3h36SGFoBymACQiIpKVileBwaFQqTWkJMKqMTAjBCKPpWnWJbAU73WuBcC0Daf46LcjCkE5SAFIREQkq3n4Qs/50HESOHnAue0wpRlsnpTmbFCPhqV5q2MNACatOcHE0OO2qrjAUQASERHJDiYT1O0Nz22GCo/AzQRY+X8wqz1cOZHarG/jsvy3fTUAPl19lMlrT9ypR8lCCkAiIiLZybMU9P4JHptgvUvszGaY3BS2fgUW6wMRBzcvz39CqgDw/orDTN9wyoYFFwwKQCIiItnNZIL6A2DoJuvdYjdvwPJXrbfLX7WGnWEPV+TFVpUAePvXg3y7JcyWFed7CkAiIiI5pUgZ6LMY2n0EDq4QtsF6Nmj7dDAMXgquxLMtKgAwetF+5m8/a+OC8y8FIBERkZxkNkPDIdazQWWaQnI8LB0J33bCFH2W19pUYUDTsgC89tOf/Lz7nG3rzacUgERERGyhaDno9yu0eR/sXeDkWviyCaZdsxnTvhq9G5XGMODl+Xv59c8Ltq4231EAEhERsRWzGRo9C0M3gn8QJMXCLy9i+v5J3mpZhK71S2ExYPjcPaw8EGHravMVBSARERFbK1YBBiyH1u+CnROcCMU8uQn/K7+PznX8SLEYPP/DLtYcvmTrSvMNBSAREZHcwGwHTZ6HZzdAyfqQGI15yTA+ThlPz2oOJKcYPPPdTtYfu2zrSvMFBSAREZHcpHhlGLgSgt8EO0fMx1bybvgQ/uu/j6SbKQyZvYMtJ2+faV4yRgFIREQkt7Gzh2Yj4Jk/wK8upoQoBl8ez49FvqRQ8jUGztrOzrCrtq4yT1MAEhERya1KVINBq+GR/4LZgcAbG1nj+hqP3NxA3+nb+OOoLoc9KAUgERGR3MzOHh76Dzy9Fnxq4W6J4QvHz/nI+IjRs5axcKeeE/QgFIBERETyAp+aMGQNtByFYbanrd12fnN4mYif/4+vftuDYRi2rjBPUQASERHJK+wcoOXrmJ75A6PsQziZknnefjGdN3bkx+nvc/PmTVtXmGcoAImIiOQ13jUw9VsC3ecQ41qaEqYonjw3nvMfBJFw/A9bV5cnKACJiIjkRSYTVG2Hx8idHA54nRjDlTJJx3H+rgOJ3/dKnWVe0qcAJCIikpfZO1K18yhO9VzPPEJIMUw4HfsV44uGsGosJMTYusJcSQFIREQkHwioUpEGw2YwwOVT/kiphcmSBBsnwOf1YOc3YEmxdYm5isnQsPHbxMTE4OnpSXR0NB4eHrYuR0RE5L5djk1k0KxtFAtfxxiH7yhnCrd+4V0L2rwH5R6ybYHZKCO/v3PFGaBJkyZRtmxZnJ2dCQoKYtu2bXdsO2vWLEwmU5qXs7NzmjaGYTBmzBh8fX1xcXEhODiYY8eOZfduiIiI2FxxdyfmPtMYU+UQWie+z1s3+5Bo7w4X98E3HWBuL7hywtZl2pzNA9C8efMYOXIkY8eOZdeuXQQEBBASEsKlS3ee8dbDw4Pw8PDUV1hYWJrvP/jgAyZOnMiUKVPYunUrbm5uhISEkJCQkN27IyIiYnOujvZ83SeQJxuWY8bNtgTFfcRO76cwTHZw+FeYFAS//RcSom1dqs3Y/BJYUFAQDRo04IsvvgDAYrHg7+/PCy+8wOuvv35b+1mzZjFixAiioqLS7c8wDPz8/Hj55Zd55ZVXAIiOjsbb25tZs2bRvXv3e9akS2AiIpIfGIbBpDXH+ei3owA8XS2J183fYj4Ram3g6gWPvAF1+1qfOJ3H5ZlLYElJSezcuZPg4ODUZWazmeDgYDZv3nzH9eLi4ihTpgz+/v507NiRAwcOpH536tQpIiIi0vTp6elJUFDQHftMTEwkJiYmzUtERCSvM5lMPP9IJT56KgB7s4mvDznS68arxD85F7wqw/VI+PUl+OohOLHG1uXmKJsGoMjISFJSUvD29k6z3Nvbm4iIiHTXqVKlCjNmzGDx4sV89913WCwWmjRpwrlz1rlQbq2XkT7Hjx+Pp6dn6svf3z+zuyYiIpJrPBlYihn9G+DmaMfmk1d4YpUbF3qEQtsPwaUIXDoA33aCH7pD5HFbl5sjbD4GKKMaN25M3759qVOnDi1atOCnn36iePHifPXVVw/c56hRo4iOjk59nT17NgsrFhERsb2HKhdn/rONKeHuxJGLsTzx1XYOl+kOL+yCoKFgtoejy+HLIFjxf3BmK0Tsh6snIe4SJMaBxWLr3cgyNr3g5+XlhZ2dHRcvXkyz/OLFi/j4+NxXHw4ODtStW5fjx62J9dZ6Fy9exNfXN02fderUSbcPJycnnJycHmAPRERE8o4afp789FwT+s/czvFLcTw1eTNf9QmkSdv/Qf2B1oHRx1bClknWV3ocXMHR7a//FgLHW5/drP91/Gv5rXa3Xv9u7+4LhUrk7B/AP9g0ADk6OhIYGEhoaCidOnUCrIOgQ0NDef755++rj5SUFPbt20e7du0AKFeuHD4+PoSGhqYGnpiYGLZu3crQoUOzYzdERETyjFJFXPnx2SYM+XYH205dpd/MbXz0VAAd61SGXvPheCis/wRizkHSdUiKh+T4vztIvm59ZVaTF6H125nv5wHZfMj3yJEj6devH/Xr16dhw4ZMmDCB+Ph4BgwYAEDfvn0pWbIk48ePB+Ctt96iUaNGVKxYkaioKD788EPCwsIYPHgwYB3wNWLECN555x0qVapEuXLlGD16NH5+fqkhS0REpCDzdHVg9sCGvLxgL0v/DGf43D1ciErg2RblMVVsBRVbpV3BYoGbN/4KRHHWAJQU//cr+a/l/wxMSfHptP9HG9eittn5v9g8AHXr1o3Lly8zZswYIiIiqFOnDitWrEgdxHzmzBnM5r+HKl27do0hQ4YQERFBkSJFCAwMZNOmTVSvXj21zauvvkp8fDxPP/00UVFRNGvWjBUrVtz2wEQREZGCytnBjs+718XXw5lpG07x/orDXIi6wbjHa2BnNqVtbDb/fSmL4japN6vZ/DlAuZGeAyQiIgXJ9A2neGfpQQwDHq3uzcTudXFxtLN1WRmWZ54DJCIiIrY3qFk5JvWsh6O9mVUHL9Jz2hYuxybauqxspQAkIiIitKvly/eDg/B0cWD3mSiavv87L87ZzcbjkVgs+e9ikS6BpUOXwEREpKA6fimO4XN3c+DC37MilCriwlOB/jxZvxQlC7vYsLq7y8jvbwWgdCgAiYhIQWYYBvvORzN/x1kW77lAbMJNAEwmaFbRi671/Wldwxsn+9w1TkgBKJMUgERERKwSklNYsT+C+TvOsunEldTlhV0d6FSnJE/VL0UNP08bVvg3BaBMUgASERG53Zkr11m48ywLdp4jPDohdXnNkh50re9Px4CSeLo62Kw+BaBMUgASERG5sxSLwYbjkczffpbfDkaQnGKNEo72ZtrU8KFrfX+aVCiG+d/PE8pmCkCZpAAkIiJyf67GJ7F4z3nmbT/L4YjY1OUlC7vwVP1SPBlYilJFXHOkFgWgTFIAEhERyRjDMNh/PoZ5O86kO3D6qfr+tK7ujbND9g2cVgDKJAUgERGRB3engdOeLg50quNH1wb+2TJwWgEokxSAREREssadBk53b+DP/7rUztJtZeT3t80nQxUREZH8q3QxV0a2rsLw4MrWgdM7zrLqwEUCyxSxaV0KQCIiIpLt7MwmWlQuTovKxbkWn2TzyVYVgERERCRHFXFztHUJmgxVRERECh4FIBERESlwFIBERESkwFEAEhERkQJHAUhEREQKHAUgERERKXAUgERERKTAUQASERGRAkcBSERERAocBSAREREpcBSAREREpMBRABIREZECRwFIREREChzNBp8OwzAAiImJsXElIiIicr9u/d6+9Xv8bhSA0hEbGwuAv7+/jSsRERGRjIqNjcXT0/OubUzG/cSkAsZisXDhwgXc3d0xmUxZ2ndMTAz+/v6cPXsWDw+PLO07t9G+5l8FaX+1r/lXQdrfgrKvhmEQGxuLn58fZvPdR/noDFA6zGYzpUqVytZteHh45Ou/hP+kfc2/CtL+al/zr4K0vwVhX+915ucWDYIWERGRAkcBSERERAocBaAc5uTkxNixY3FycrJ1KdlO+5p/FaT91b7mXwVpfwvSvt4vDYIWERGRAkdngERERKTAUQASERGRAkcBSERERAocBSAREREpcBSAssGkSZMoW7Yszs7OBAUFsW3btru2X7BgAVWrVsXZ2ZlatWqxbNmyHKr0wY0fP54GDRrg7u5OiRIl6NSpE0eOHLnrOrNmzcJkMqV5OTs751DFD27cuHG31V21atW7rpMXj+ktZcuWvW1/TSYTw4YNS7d9Xjquf/zxBx06dMDPzw+TycSiRYvSfG8YBmPGjMHX1xcXFxeCg4M5duzYPfvN6M98TrjbviYnJ/Paa69Rq1Yt3Nzc8PPzo2/fvly4cOGufT7Iz0JOudex7d+//221t2nT5p795rVjC6T782symfjwww/v2GduPrbZRQEoi82bN4+RI0cyduxYdu3aRUBAACEhIVy6dCnd9ps2baJHjx4MGjSI3bt306lTJzp16sT+/ftzuPKMWbduHcOGDWPLli2sWrWK5ORkWrduTXx8/F3X8/DwIDw8PPUVFhaWQxVnTo0aNdLUvWHDhju2zavH9Jbt27en2ddVq1YB8NRTT91xnbxyXOPj4wkICGDSpEnpfv/BBx8wceJEpkyZwtatW3FzcyMkJISEhIQ79pnRn/mccrd9vX79Ort27WL06NHs2rWLn376iSNHjvD444/fs9+M/CzkpHsdW4A2bdqkqX3OnDl37TMvHlsgzT6Gh4czY8YMTCYTXbp0uWu/ufXYZhtDslTDhg2NYcOGpX5OSUkx/Pz8jPHjx6fbvmvXrkb79u3TLAsKCjKeeeaZbK0zq126dMkAjHXr1t2xzcyZMw1PT8+cKyqLjB071ggICLjv9vnlmN4yfPhwo0KFCobFYkn3+7x6XAHj559/Tv1ssVgMHx8f48MPP0xdFhUVZTg5ORlz5sy5Yz8Z/Zm3hX/va3q2bdtmAEZYWNgd22T0Z8FW0tvffv36GR07dsxQP/nl2Hbs2NF45JFH7tomrxzbrKQzQFkoKSmJnTt3EhwcnLrMbDYTHBzM5s2b011n8+bNadoDhISE3LF9bhUdHQ1A0aJF79ouLi6OMmXK4O/vT8eOHTlw4EBOlJdpx44dw8/Pj/Lly9OrVy/OnDlzx7b55ZiC9e/0d999x8CBA+86MXBePa7/dOrUKSIiItIcO09PT4KCgu547B7kZz63io6OxmQyUbhw4bu2y8jPQm6zdu1aSpQoQZUqVRg6dChXrly5Y9v8cmwvXrzI0qVLGTRo0D3b5uVj+yAUgLJQZGQkKSkpeHt7p1nu7e1NREREuutERERkqH1uZLFYGDFiBE2bNqVmzZp3bFelShVmzJjB4sWL+e6777BYLDRp0oRz587lYLUZFxQUxKxZs1ixYgWTJ0/m1KlTNG/enNjY2HTb54djesuiRYuIioqif//+d2yTV4/rv906Phk5dg/yM58bJSQk8Nprr9GjR4+7TpSZ0Z+F3KRNmzbMnj2b0NBQ3n//fdatW0fbtm1JSUlJt31+ObbffPMN7u7uPPHEE3dtl5eP7YPSbPCSacOGDWP//v33vF7cuHFjGjdunPq5SZMmVKtWja+++oq33347u8t8YG3btk19X7t2bYKCgihTpgzz58+/r39V5WXTp0+nbdu2+Pn53bFNXj2uYpWcnEzXrl0xDIPJkyfftW1e/lno3r176vtatWpRu3ZtKlSowNq1a2nVqpUNK8teM2bMoFevXve8MSEvH9sHpTNAWcjLyws7OzsuXryYZvnFixfx8fFJdx0fH58Mtc9tnn/+eX799VfWrFlDqVKlMrSug4MDdevW5fjx49lUXfYoXLgwlStXvmPdef2Y3hIWFsbq1asZPHhwhtbLq8f11vHJyLF7kJ/53ORW+AkLC2PVqlV3PfuTnnv9LORm5cuXx8vL64615/VjC7B+/XqOHDmS4Z9hyNvH9n4pAGUhR0dHAgMDCQ0NTV1msVgIDQ1N8y/kf2rcuHGa9gCrVq26Y/vcwjAMnn/+eX7++Wd+//13ypUrl+E+UlJS2LdvH76+vtlQYfaJi4vjxIkTd6w7rx7Tf5s5cyYlSpSgffv2GVovrx7XcuXK4ePjk+bYxcTEsHXr1jseuwf5mc8tboWfY8eOsXr1aooVK5bhPu71s5CbnTt3jitXrtyx9rx8bG+ZPn06gYGBBAQEZHjdvHxs75utR2HnN3PnzjWcnJyMWbNmGQcPHjSefvppo3DhwkZERIRhGIbRp08f4/XXX09tv3HjRsPe3t746KOPjEOHDhljx441HBwcjH379tlqF+7L0KFDDU9PT2Pt2rVGeHh46uv69eupbf69r2+++aaxcuVK48SJE8bOnTuN7t27G87OzsaBAwdssQv37eWXXzbWrl1rnDp1yti4caMRHBxseHl5GZcuXTIMI/8c039KSUkxSpcubbz22mu3fZeXj2tsbKyxe/duY/fu3QZgfPLJJ8bu3btT73z63//+ZxQuXNhYvHix8eeffxodO3Y0ypUrZ9y4cSO1j0ceecT4/PPPUz/f62feVu62r0lJScbjjz9ulCpVytizZ0+an+HExMTUPv69r/f6WbClu+1vbGys8corrxibN282Tp06ZaxevdqoV6+eUalSJSMhISG1j/xwbG+Jjo42XF1djcmTJ6fbR146ttlFASgbfP7550bp0qUNR0dHo2HDhsaWLVtSv2vRooXRr1+/NO3nz59vVK5c2XB0dDRq1KhhLF26NIcrzjgg3dfMmTNT2/x7X0eMGJH65+Lt7W20a9fO2LVrV84Xn0HdunUzfH19DUdHR6NkyZJGt27djOPHj6d+n1+O6T+tXLnSAIwjR47c9l1ePq5r1qxJ9+/trf2xWCzG6NGjDW9vb8PJyclo1arVbX8GZcqUMcaOHZtm2d1+5m3lbvt66tSpO/4Mr1mzJrWPf+/rvX4WbOlu+3v9+nWjdevWRvHixQ0HBwejTJkyxpAhQ24LMvnh2N7y1VdfGS4uLkZUVFS6feSlY5tdTIZhGNl6iklEREQkl9EYIBERESlwFIBERESkwFEAEhERkQJHAUhEREQKHAUgERERKXAUgERERKTAUQASERGRAkcBSERERAocBSARkTswmUwsWrTI1mWISDZQABKRXKl///6YTKbbXm3atLF1aSKSD9jbugARkTtp06YNM2fOTLPMycnJRtWISH6iM0Aikms5OTnh4+OT5lWkSBHAenlq8uTJtG3bFhcXF8qXL8/ChQvTrL9v3z4eeeQRXFxcKFasGE8//TRxcXFp2syYMYMaNWrg5OSEr68vzz//fJrvIyMj6dy5M66urlSqVIklS5akfnft2jV69epF8eLFcXFxoVKlSrcFNhHJnRSARCTPGj16NF26dGHv3r306tWL7t27c+jQIQDi4+MJCQmhSJEibN++nQULFrB69eo0AWfy5MkMGzaMp59+mn379rFkyRIqVqyYZhtvvvkmXbt25c8//6Rdu3b06tWLq1evpm7/4MGDLF++nEOHDjF58mS8vLxy7g9ARB6craejFxFJT79+/Qw7OzvDzc0tzevdd981DMMwAOPZZ59Ns05QUJAxdOhQwzAM4+uvvzaKFClixMXFpX6/dOlSw2w2GxEREYZhGIafn5/xxhtv3LEGwPjvf/+b+jkuLs4AjOXLlxuGYRgdOnQwBgwYkDU7LCI5SmOARCTXevjhh5k8eXKaZUWLFk1937hx4zTfNW7cmD179gBw6NAhAgICcHNzS/2+adOmWCwWjhw5gslk4sKFC7Rq1equNdSuXTv1vZubGx4eHly6dAmAoUOH0qVLF3bt2kXr1q3p1KkTTZo0eaB9FZGcpQAkIrmWm5vbbZeksoqLi8t9tXNwcEjz2WQyYbFYAGjbti1hYWEsW7aMVatW0apVK4YNG8ZHH32U5fWKSNbSGCARybO2bNly2+dq1aoBUK1aNfbu3Ut8fHzq9xs3bsRsNlOlShXc3d0pW7YsoaGhmaqhePHi9OvXj++++44JEybw9ddfZ6o/EckZOgMkIrlWYmIiERERaZbZ29unDjResGAB9evXp1mzZnz//fds27aN6dOnA9CrVy/Gjh1Lv379GDduHJcvX+aFF16gT58+eHt7AzBu3DieffZZSpQoQdu2bYmNjWXjxo288MIL91XfmDFjCAwMpEaNGiQmJvLrr7+mBjARyd0UgEQk11qxYgW+vr5pllWpUoXDhw8D1ju05s6dy3PPPYevry9z5syhevXqALi6urJy5UqGDx9OgwYNcHV1pUuXLnzyySepffXr14+EhAQ+/fRTXnnlFby8vHjyySfvuz5HR0dGjRrF6dOncXFxoXnz5sydOzcL9lxEspvJMAzD1kWIiGSUyWTi559/plOnTrYuRUTyII0BEhERkQJHAUhEREQKHI0BEpE8SVfvRSQzdAZIREREChwFIBERESlwFIBERESkwFEAEhERkQJHAUhEREQKHAUgERERKXAUgERERKTAUQASERGRAuf/AUaHGmRvDdYbAAAAAElFTkSuQmCC",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "outputs(\"Model with AUG3\", acc_AUG3, hist__AUG3)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    " Не особо получилось улучшить MODImprovedModel(Model with AUG3){77,4%} ===*VS*=== ImprovedModel(Model with AUG3){80%}\n",
    "\n",
    " Усовершенствовать получилось, далее опишу. \n",
    " В процессе улучшения модели нейронной сети были реализованы две ключевые модификации: интеграция слоев Dropout. Эти модификации основаны на устоявшихся практиках машинного обучения.\n",
    " Слои исключения(Dropout) были введены в стратегические точки архитектуры модели. В частности, эти слои были добавлены после функций активации как в сверточном, так и в полносвязном слоях.\n",
    " Теоретические преимущества:\n",
    " 1. Dropout случайным образом деактивируют подмножество нейронов во время обучения. Это предотвращает слишком большую совместную адаптацию нейронов и зависимость от присутствия других конкретных нейронов, побуждая модель изучать более надежные функции, которые лучше обобщаются на невидимые данные.\n",
    " 2. Обнуляя определенные функции, исключение гарантирует, что сеть не станет чрезмерно зависеть от какой-либо отдельной функции, что приведет к более распределенному и надежному представлению функций\n",
    "\n",
    " Наблюдая за работой модели пока составлялась лаба, видимо, не совсем корректные выводы сделал, чтоб улучшить модель.\n",
    "\n",
    " Опишу следующую стратегию усовершенствования модели.\n",
    " 1. Увеличение глубины или ширины. Можно рассмотреть возможность увеличения глубины (больше слоев) или ширины (больше нейронов) сети.\n",
    " 2. Global Average Pooling. Вместо использования AdaptiveAvgPool2d можно использовать пул глобальных средних значений, поскольку это может помочь уменьшить общее количество параметров и тем самым снизить риск переобучения.(Особенно критично) \n",
    "Замена адаптивного пула средних на пул глобальных средних в модели машинного обучения может привести к снижению её способности изучать сложные функции. Однако, этот подход часто обеспечивает преимущества в виде уменьшения переобучения и улучшения обобщающей способности на данных, которые не участвовали в обучении модели. \n",
    "Говорю об переобучении, вспоминая первую (Standart Model)... Там явно около этого дело. \n",
    "\n",
    "Реализую, но не в этот раз. ПОНЯТЬ ПРОСТИТЬ\n",
    "\n",
    "В прочем, 80% уже есть\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 6 Адаптировать предобученную сеть EfficientNet-B0 для решения поставленной задачи"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "from torchvision.models import efficientnet_b0\n",
    "import torchvision\n",
    "\n",
    "transform = transforms.Compose([\n",
    "    transforms.Resize((256, 256)),\n",
    "    transforms.ToTensor(),\n",
    "    transforms.Normalize(mean=mean, std=std)\n",
    "])\n",
    "\n",
    "dataset_b0 = datasets.ImageFolder(root=DIR, transform=transform)\n",
    "\n",
    "train_size_b0 = int(0.7 * len(dataset_b0))\n",
    "val_size_b0 = len(dataset_b0) - train_size_b0\n",
    "train_dataset_b0, val_dataset_b0 = random_split(dataset_b0, [train_size_b0, val_size_b0])\n",
    "\n",
    "train_loader_b0 = DataLoader(train_dataset_b0, batch_size=32, shuffle=True)\n",
    "val_loader_b0 = DataLoader(val_dataset_b0, batch_size=64, shuffle=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:06<00:00,  6.66it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 0: train loss 0.53071\n",
      "Epoch 0: val loss 0.39328\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:06<00:00,  6.74it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1: train loss 0.36825\n",
      "Epoch 1: val loss 0.30827\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:06<00:00,  6.86it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 2: train loss 0.30314\n",
      "Epoch 2: val loss 0.26827\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:06<00:00,  6.86it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 3: train loss 0.27093\n",
      "Epoch 3: val loss 0.24576\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:06<00:00,  6.69it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 4: train loss 0.24410\n",
      "Epoch 4: val loss 0.22539\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:06<00:00,  6.81it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 5: train loss 0.22973\n",
      "Epoch 5: val loss 0.22251\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:06<00:00,  6.82it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 6: train loss 0.21809\n",
      "Epoch 6: val loss 0.20701\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:06<00:00,  6.77it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 7: train loss 0.20899\n",
      "Epoch 7: val loss 0.19586\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:06<00:00,  6.90it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 8: train loss 0.22661\n",
      "Epoch 8: val loss 0.19310\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:06<00:00,  6.92it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 9: train loss 0.18570\n",
      "Epoch 9: val loss 0.19039\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:06<00:00,  6.88it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 10: train loss 0.19806\n",
      "Epoch 10: val loss 0.18668\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:06<00:00,  6.58it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 11: train loss 0.19484\n",
      "Epoch 11: val loss 0.18390\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:06<00:00,  6.65it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 12: train loss 0.18671\n",
      "Epoch 12: val loss 0.17967\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:06<00:00,  6.73it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 13: train loss 0.18305\n",
      "Epoch 13: val loss 0.18253\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:06<00:00,  6.88it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 14: train loss 0.16824\n",
      "Epoch 14: val loss 0.17614\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:06<00:00,  6.65it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 15: train loss 0.16468\n",
      "Epoch 15: val loss 0.17624\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:06<00:00,  6.85it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 16: train loss 0.17788\n",
      "Epoch 16: val loss 0.17485\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:06<00:00,  6.80it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 17: train loss 0.17376\n",
      "Epoch 17: val loss 0.17668\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:06<00:00,  6.84it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 18: train loss 0.16687\n",
      "Epoch 18: val loss 0.17366\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:06<00:00,  6.83it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 19: train loss 0.16734\n",
      "Epoch 19: val loss 0.17430\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:06<00:00,  6.84it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 20: train loss 0.15698\n",
      "Epoch 20: val loss 0.17006\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:06<00:00,  6.92it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 21: train loss 0.15547\n",
      "Epoch 21: val loss 0.17862\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:06<00:00,  6.90it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 22: train loss 0.15978\n",
      "Epoch 22: val loss 0.16723\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:06<00:00,  6.78it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 23: train loss 0.16109\n",
      "Epoch 23: val loss 0.17118\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43/43 [00:06<00:00,  6.86it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 24: train loss 0.16203\n",
      "Epoch 24: val loss 0.17158\n"
     ]
    }
   ],
   "source": [
    "model_b0 = torchvision.models.efficientnet_b0(pretrained=True)\n",
    "\n",
    "for param in model_b0.parameters():\n",
    "    param.requires_grad = False\n",
    "\n",
    "num_ftrs = model_b0.classifier[1].in_features\n",
    "model_b0.classifier[1] = torch.nn.Sequential(\n",
    "    torch.nn.Linear(num_ftrs, 1),\n",
    "    torch.nn.Sigmoid()\n",
    ")\n",
    "\n",
    "loss_fn = torch.nn.BCELoss()\n",
    "optimizer = torch.optim.Adam(model_b0.classifier[1].parameters())\n",
    "history = lab4_util.train(model_b0, loss_fn, optimizer, train_loader_b0, val_loader_b0, max_epochs=25)  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "0.9265625"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lab4_util.plot_history(history, 'loss', ylim=(0, 1.0))\n",
    "lab4_util.accuracy(model_b0, val_loader_b0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 50/50 [00:14<00:00,  3.49it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 0: train loss 0.53163\n",
      "Epoch 0: val loss 0.39736\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 50/50 [00:13<00:00,  3.62it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1: train loss 0.36433\n",
      "Epoch 1: val loss 0.31282\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 50/50 [00:13<00:00,  3.63it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 2: train loss 0.31897\n",
      "Epoch 2: val loss 0.26753\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 50/50 [00:13<00:00,  3.64it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 3: train loss 0.27108\n",
      "Epoch 3: val loss 0.24376\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 50/50 [00:13<00:00,  3.64it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 4: train loss 0.24536\n",
      "Epoch 4: val loss 0.22265\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 50/50 [00:13<00:00,  3.63it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 5: train loss 0.23477\n",
      "Epoch 5: val loss 0.20478\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 50/50 [00:13<00:00,  3.62it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 6: train loss 0.22718\n",
      "Epoch 6: val loss 0.20539\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 50/50 [00:13<00:00,  3.62it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 7: train loss 0.21400\n",
      "Epoch 7: val loss 0.18480\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 50/50 [00:13<00:00,  3.61it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 8: train loss 0.21371\n",
      "Epoch 8: val loss 0.18593\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 50/50 [00:13<00:00,  3.60it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 9: train loss 0.21808\n",
      "Epoch 9: val loss 0.17270\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 50/50 [00:14<00:00,  3.51it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 10: train loss 0.22774\n",
      "Epoch 10: val loss 0.18313\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 50/50 [00:13<00:00,  3.61it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 11: train loss 0.21904\n",
      "Epoch 11: val loss 0.17854\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 50/50 [00:14<00:00,  3.55it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 12: train loss 0.19738\n",
      "Epoch 12: val loss 0.18444\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 50/50 [00:14<00:00,  3.50it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 13: train loss 0.19995\n",
      "Epoch 13: val loss 0.16591\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 50/50 [00:13<00:00,  3.66it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 14: train loss 0.20071\n",
      "Epoch 14: val loss 0.18261\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 50/50 [00:13<00:00,  3.66it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 15: train loss 0.17440\n",
      "Epoch 15: val loss 0.17443\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 50/50 [00:13<00:00,  3.66it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 16: train loss 0.18643\n",
      "Epoch 16: val loss 0.17989\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 50/50 [00:13<00:00,  3.61it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 17: train loss 0.17148\n",
      "Epoch 17: val loss 0.16635\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 50/50 [00:14<00:00,  3.55it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 18: train loss 0.18159\n",
      "Epoch 18: val loss 0.16616\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 50/50 [00:14<00:00,  3.46it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 19: train loss 0.19030\n",
      "Epoch 19: val loss 0.18969\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 50/50 [00:14<00:00,  3.52it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 20: train loss 0.17429\n",
      "Epoch 20: val loss 0.16185\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 50/50 [00:14<00:00,  3.54it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 21: train loss 0.18083\n",
      "Epoch 21: val loss 0.16138\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 50/50 [00:13<00:00,  3.67it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 22: train loss 0.16372\n",
      "Epoch 22: val loss 0.15153\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 50/50 [00:13<00:00,  3.66it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 23: train loss 0.15709\n",
      "Epoch 23: val loss 0.14386\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 50/50 [00:14<00:00,  3.54it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 24: train loss 0.18028\n",
      "Epoch 24: val loss 0.15737\n"
     ]
    }
   ],
   "source": [
    "from torchvision.models import efficientnet_b0\n",
    "from torchvision.models.efficientnet import efficientnet_b0, EfficientNet_B0_Weights\n",
    "import torchvision\n",
    "\n",
    "transformAUG = transforms.Compose([\n",
    "    transforms.Resize((356, 356)),  # Resize all images to the same scale\n",
    "    transforms.RandomHorizontalFlip(),  \n",
    "    transforms.RandomRotation(15),      \n",
    "    transforms.ColorJitter(brightness=0.1, contrast=0.2, saturation=0.3),  \n",
    "    transforms.ToTensor(),\n",
    "    transforms.Normalize(mean=mean, std=std)\n",
    "])\n",
    "\n",
    "dataset_b0AUG = datasets.ImageFolder(root=DIR, transform=transformAUG)\n",
    "\n",
    "train_size_b0AUG = int(0.8 * len(dataset_b0AUG))\n",
    "val_size_b0AUG = len(dataset_b0AUG) - train_size_b0AUG\n",
    "train_dataset_b0AUG, val_dataset_b0AUG = random_split(dataset_b0AUG, [train_size_b0AUG, val_size_b0AUG])\n",
    "\n",
    "train_loader_b0AUG = DataLoader(train_dataset_b0AUG, batch_size=32, shuffle=True)\n",
    "val_loader_b0AUG = DataLoader(val_dataset_b0AUG, batch_size=64, shuffle=False)\n",
    "\n",
    "model_b0_AUG = torchvision.models.efficientnet_b0(pretrained=True)\n",
    "\n",
    "for param in model_b0_AUG.parameters():\n",
    "    param.requires_grad = False\n",
    "\n",
    "num_ftrs = model_b0_AUG.classifier[1].in_features\n",
    "model_b0_AUG.classifier[1] = torch.nn.Sequential(\n",
    "    torch.nn.Linear(num_ftrs, 1),\n",
    "    torch.nn.Sigmoid()\n",
    ")\n",
    "\n",
    "loss_fn = torch.nn.BCELoss()\n",
    "optimizer = torch.optim.Adam(model_b0_AUG.classifier[1].parameters())\n",
    "history = lab4_util.train(model_b0_AUG, loss_fn, optimizer, train_loader_b0AUG, val_loader_b0AUG, max_epochs=25)  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAjwAAAG2CAYAAAB7zFy5AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8pXeV/AAAACXBIWXMAAA9hAAAPYQGoP6dpAABPfElEQVR4nO3deXhU5d3/8fdM9p2QPSEQ9p2wx4ALCAouVBQrKhW0KlWRR+Xyp1IFxFapa3kUKmpVtBVFrFifSkVAXEERkE0h7HsWQiAr2WbO74+TTAhrEiYZ5vB5XddcM3PmLN85TJuP97nPfdsMwzAQERERsTC7pwsQERERaWwKPCIiImJ5CjwiIiJieQo8IiIiYnkKPCIiImJ5CjwiIiJieQo8IiIiYnkKPCIiImJ5CjwiIiJieQo8IiIiYnkeDTzffPMNI0aMIDExEZvNxieffHLWbb766it69+5NQEAA7dq1Y+7cuY1ep4iIiHg3jwae4uJiUlNTmT17dp3W37VrF9dccw2DBw9m3bp1PPjgg9x1110sXry4kSsVERERb2Y7XyYPtdlsLFy4kJEjR552nUcffZTPPvuMTZs2uZbdfPPNHD16lM8//7wJqhQRERFv5OvpAupj5cqVDB06tNayYcOG8eCDD552m7KyMsrKylzvnU4neXl5REVFYbPZGqtUERERcSPDMCgsLCQxMRG7vf4XqLwq8GRlZREXF1drWVxcHAUFBRw7doygoKCTtpkxYwbTp09vqhJFRESkEe3bt48WLVrUezuvCjwNMXnyZCZNmuR6n5+fT8uWLdm3bx/h4eEerExERETqqqCggOTkZMLCwhq0vVcFnvj4eLKzs2sty87OJjw8/JStOwABAQEEBASctDw8PFyBR0RExMs0tDuKV43Dk56ezrJly2otW7JkCenp6R6qSERERLyBRwNPUVER69atY926dYB52/m6devYu3cvYF6OGjt2rGv9e+65h507d/LII4+wZcsW/va3v/Hhhx/y0EMPeaJ8ERER8RIeDTyrV6+mV69e9OrVC4BJkybRq1cvpk6dCkBmZqYr/AC0bt2azz77jCVLlpCamsqLL77I3//+d4YNG+aR+kVERMQ7nDfj8DSVgoICIiIiyM/PVx8eEREv53A4qKio8HQZ4ib+/v6nveX8XP9+e1WnZRERETDHZMnKyuLo0aOeLkXcyG6307p1a/z9/d2+bwUeERHxOtVhJzY2luDgYA0kawFOp5ODBw+SmZlJy5Yt3f5vqsAjIiJexeFwuMJOVFSUp8sRN4qJieHgwYNUVlbi5+fn1n171W3pIiIi1X12goODPVyJuFv1pSyHw+H2fSvwiIiIV9JlLOtpzH9TBR4RERGxPAUeERERL5aSksLMmTM9XcZ5T4FHRESkCdhstjM+nnzyyQbt96effmL8+PHuLdaCdJeWiIhIE8jMzHS9nj9/PlOnTiUjI8O1LDQ01PXaMAwcDge+vmf/Mx0TE+PeQi1KLTwiIiJNID4+3vWIiIjAZrO53m/ZsoWwsDD++9//0qdPHwICAvjuu+/YsWMH1113HXFxcYSGhtKvXz+WLl1aa78nXtKy2Wz8/e9/5/rrryc4OJj27dvz6aefNvG3Pf8o8IiIiNczDIOS8kqPPNw5Q9Njjz3GX/7yFzZv3kyPHj0oKiri6quvZtmyZfz8888MHz6cESNG1Jpn8lSmT5/OTTfdxIYNG7j66qsZM2YMeXl5bqvTG+mSloiIeL1jFQ66TF3skWP/+tQwgv3d8+f0qaee4oorrnC9b968Oampqa73f/rTn1i4cCGffvop999//2n3c/vtt3PLLbcA8Mwzz/Dyyy+zatUqhg8f7pY6vZFaeERERM4Tffv2rfW+qKiIhx9+mM6dO9OsWTNCQ0PZvHnzWVt4evTo4XodEhJCeHg4OTk5jVKzt1ALj4iIeL0gPx9+fWqYx47tLiEhIbXeP/zwwyxZsoQXXniBdu3aERQUxI033kh5efkZ93PitAw2mw2n0+m2Or2RAo+IiHg9m83mtstK55Pvv/+e22+/neuvvx4wW3x2797t2aK8lC5piYiInKfat2/Pxx9/zLp161i/fj233nrrBd9S01AKPCIiIuepl156icjISAYMGMCIESMYNmwYvXv39nRZXslmuPN+Oi9QUFBAREQE+fn5hIeHe7ocERGpp9LSUnbt2kXr1q0JDAz0dDniRmf6tz3Xv99q4RERERHLU+ARERERy1PgEREREctT4BERERHLU+ARERERy1PgEREREctT4BERERHLU+ARERERy1PgEREREctT4BEREfESgwYN4sEHH3S9T0lJYebMmWfcxmaz8cknn5zzsd21H09R4BEREWkCI0aMYPjw4af87Ntvv8Vms7Fhw4Z67fOnn35i/Pjx7ijP5cknn6Rnz54nLc/MzOSqq65y67GakgKPiIhIE7jzzjtZsmQJ+/fvP+mzt99+m759+9KjR4967TMmJobg4GB3lXhG8fHxBAQENMmxGoMCj4iISBO49tpriYmJYe7cubWWFxUVsWDBAkaOHMktt9xCUlISwcHBdO/enffff/+M+zzxkta2bdu49NJLCQwMpEuXLixZsuSkbR599FE6dOhAcHAwbdq0YcqUKVRUVAAwd+5cpk+fzvr167HZbNhsNle9J17S2rhxI5dffjlBQUFERUUxfvx4ioqKXJ/ffvvtjBw5khdeeIGEhASioqKYMGGC61hNzdcjRxUREXEnw4CKEs8c2y8YbLazrubr68vYsWOZO3cujz/+OLaqbRYsWIDD4eB3v/sdCxYs4NFHHyU8PJzPPvuM2267jbZt29K/f/+z7t/pdHLDDTcQFxfHjz/+SH5+fq3+PtXCwsKYO3cuiYmJbNy4kbvvvpuwsDAeeeQRRo8ezaZNm/j8889ZunQpABERESfto7i4mGHDhpGens5PP/1ETk4Od911F/fff3+tQLd8+XISEhJYvnw527dvZ/To0fTs2ZO77777rN/H3RR4RETE+1WUwDOJnjn2Hw+Cf0idVv3973/P888/z9dff82gQYMA83LWqFGjaNWqFQ8//LBr3YkTJ7J48WI+/PDDOgWepUuXsmXLFhYvXkxionkunnnmmZP63TzxxBOu1ykpKTz88MN88MEHPPLIIwQFBREaGoqvry/x8fGnPda8efMoLS3l3XffJSTE/O6zZs1ixIgRPPvss8TFxQEQGRnJrFmz8PHxoVOnTlxzzTUsW7bMI4FHl7RERESaSKdOnRgwYABvvfUWANu3b+fbb7/lzjvvxOFw8Kc//Ynu3bvTvHlzQkNDWbx4MXv37q3Tvjdv3kxycrIr7ACkp6eftN78+fMZOHAg8fHxhIaG8sQTT9T5GMcfKzU11RV2AAYOHIjT6SQjI8O1rGvXrvj4+LjeJyQkkJOTU69juYtaeERExPv5BZstLZ46dj3ceeedTJw4kdmzZ/P222/Ttm1bLrvsMp599ln+93//l5kzZ9K9e3dCQkJ48MEHKS8vd1upK1euZMyYMUyfPp1hw4YRERHBBx98wIsvvui2YxzPz8+v1nubzYbT6WyUY52NAo+IiHg/m63Ol5U87aabbuKBBx5g3rx5vPvuu9x7773YbDa+//57rrvuOn73u98BZp+crVu30qVLlzrtt3Pnzuzbt4/MzEwSEhIA+OGHH2qts2LFClq1asXjjz/uWrZnz55a6/j7++NwOM56rLlz51JcXOxq5fn++++x2+107NixTvU2NV3SEhERaUKhoaGMHj2ayZMnk5mZye233w5A+/btWbJkCStWrGDz5s384Q9/IDs7u877HTp0KB06dGDcuHGsX7+eb7/9tlawqT7G3r17+eCDD9ixYwcvv/wyCxcurLVOSkoKu3btYt26deTm5lJWVnbSscaMGUNgYCDjxo1j06ZNLF++nIkTJ3Lbbbe5+u+cbxR4REREmtidd97JkSNHGDZsmKvPzRNPPEHv3r0ZNmwYgwYNIj4+npEjR9Z5n3a7nYULF3Ls2DH69+/PXXfdxdNPP11rnd/85jc89NBD3H///fTs2ZMVK1YwZcqUWuuMGjWK4cOHM3jwYGJiYk55a3xwcDCLFy8mLy+Pfv36ceONNzJkyBBmzZpV/5PRRGyGYRieLqIpFRQUEBERQX5+PuHh4Z4uR0RE6qm0tJRdu3bRunVrAgMDPV2OuNGZ/m3P9e+3WnhERETE8hR4RERExPIUeERERMTyFHhERETE8hR4RETEK11g99xcEBrz31SBR0REvEr16L0lJR6aLFQaTfWo0sdPR+EuGmlZRES8io+PD82aNXPNyRQcHOyaeVy8l9Pp5NChQwQHB+Pr6/54osAjIiJep3omb09NRCmNw26307Jly0YJsAo8IiLidWw2GwkJCcTGxlJRUeHpcsRN/P39sdsbp7eNAo+IiHgtHx+fRunvIdajTssiIiJieQo8IiIiYnkKPCIiImJ5CjwiIiJieQo8IiIiYnkKPCIiImJ5CjwiIiJieQo8IiIiYnkKPCIiImJ5CjwiIiJieQo8IiIiYnkKPCIiImJ5CjwiIiJieQo8IiIiYnkeDzyzZ88mJSWFwMBA0tLSWLVq1RnXnzlzJh07diQoKIjk5GQeeughSktLm6haERER8UYeDTzz589n0qRJTJs2jbVr15KamsqwYcPIyck55frz5s3jscceY9q0aWzevJk333yT+fPn88c//rGJKxcRERFv4tHA89JLL3H33Xdzxx130KVLF+bMmUNwcDBvvfXWKddfsWIFAwcO5NZbbyUlJYUrr7ySW2655aytQiIiInJh81jgKS8vZ82aNQwdOrSmGLudoUOHsnLlylNuM2DAANasWeMKODt37mTRokVcffXVpz1OWVkZBQUFtR4iIiJyYfH11IFzc3NxOBzExcXVWh4XF8eWLVtOuc2tt95Kbm4uF198MYZhUFlZyT333HPGS1ozZsxg+vTpbq1dREREvIvHOy3Xx1dffcUzzzzD3/72N9auXcvHH3/MZ599xp/+9KfTbjN58mTy8/Ndj3379jVhxSIiInI+8FgLT3R0ND4+PmRnZ9danp2dTXx8/Cm3mTJlCrfddht33XUXAN27d6e4uJjx48fz+OOPY7efnN8CAgIICAhw/xcQERERr+GxFh5/f3/69OnDsmXLXMucTifLli0jPT39lNuUlJScFGp8fHwAMAyj8YoVERERr+axFh6ASZMmMW7cOPr27Uv//v2ZOXMmxcXF3HHHHQCMHTuWpKQkZsyYAcCIESN46aWX6NWrF2lpaWzfvp0pU6YwYsQIV/AREREROZFHA8/o0aM5dOgQU6dOJSsri549e/L555+7OjLv3bu3VovOE088gc1m44knnuDAgQPExMQwYsQInn76aU99BREREfECNuMCuxZUUFBAREQE+fn5hIeHe7ocERERqYNz/fvtVXdpiYiIiDSEAo+IiIhYngKPiIiIWJ4Cj4iIiFieAo+IiIhYngKPiIiIWJ4Cj4iIiFieAo+IiIhYngKPiIiIWJ4Cj4iIiFieAo+IiIhYngKPiIiIWJ4Cj4iIiFieAo+IiIhYngKPiIiIWJ4Cj4iIiFieAo+IiIhYngKPiIiIWJ4Cj4iIiFieAo+IiIhYngKPiIiIWJ4Cj4iIiFieAo+IiIhYngKPiIiIWJ4Cj4iIiFieAo+IiIhYngKPiIiIWJ4Cj4iIiFieAo+IiIhYngKPiIiIWJ4Cj4iIiFieAo+IiIhYngKPiIiIWJ4Cj4iIiFieAo+IiIhYngKPiIiIWJ4Cj4iIiFieAo+IiIhYngKPiIiIWJ4Cj4iIiFieAo+IiIhYngKPiIiIWJ4Cj4iIiFieAo+IiIhYngKPiIiIWJ4Cj4iIiFieAo+IiIhYngKPiIiIWJ4Cj4iIiFieAo+IiIhYngKPiIiIWJ4Cj4iIiFieAo+IiIhYngKPiIiIWJ4Cj4iIiFieAo+IiIhYngKPiIiIWJ4Cj4iIiFieAo+IiIhYngKPiIiIWJ4Cj4iIiFieAo+IiIhYngKPiIiIWJ7HA8/s2bNJSUkhMDCQtLQ0Vq1adcb1jx49yoQJE0hISCAgIIAOHTqwaNGiJqpWREREvJGvJw8+f/58Jk2axJw5c0hLS2PmzJkMGzaMjIwMYmNjT1q/vLycK664gtjYWD766COSkpLYs2cPzZo1a/riRURExGvYDMMwPHXwtLQ0+vXrx6xZswBwOp0kJyczceJEHnvssZPWnzNnDs8//zxbtmzBz8+vQccsKCggIiKC/Px8wsPDz6l+ERERaRrn+vfbY5e0ysvLWbNmDUOHDq0pxm5n6NChrFy58pTbfPrpp6SnpzNhwgTi4uLo1q0bzzzzDA6H47THKSsro6CgoNZDRERELiweCzy5ubk4HA7i4uJqLY+LiyMrK+uU2+zcuZOPPvoIh8PBokWLmDJlCi+++CJ//vOfT3ucGTNmEBER4XokJye79XuIiIjI+c/jnZbrw+l0Ehsby+uvv06fPn0YPXo0jz/+OHPmzDntNpMnTyY/P9/12LdvXxNWLCIiIucDj3Vajo6OxsfHh+zs7FrLs7OziY+PP+U2CQkJ+Pn54ePj41rWuXNnsrKyKC8vx9/f/6RtAgICCAgIcG/xIiIi4lU81sLj7+9Pnz59WLZsmWuZ0+lk2bJlpKenn3KbgQMHsn37dpxOp2vZ1q1bSUhIOGXYEREREQEPX9KaNGkSb7zxBu+88w6bN2/m3nvvpbi4mDvuuAOAsWPHMnnyZNf69957L3l5eTzwwANs3bqVzz77jGeeeYYJEyZ46iuIiIiIF/DoODyjR4/m0KFDTJ06laysLHr27Mnnn3/u6si8d+9e7PaaTJacnMzixYt56KGH6NGjB0lJSTzwwAM8+uijnvoKIiIi4gU8Og6PJ2gcHhEREe/jtePwiIiIiDQVBR4RERGxPAUeERERsTwFHhEREbE8BR4RERGxPAUeERERsTwFHhEREbE8BR4RERGxPAUeERERsTwFHhEREbE8BR4RERGxPAUeERERsTwFHhEREbE8BR4RERGxPAUeERERsTwFHhEREbE8BR43czgNT5cgIiIiJ1DgcZM1e/L47ZwVTHhvradLERERkRP4NmSjffv2YbPZaNGiBQCrVq1i3rx5dOnShfHjx7u1QG8R6OfDT7uPEOTnQ2mFg0A/H0+XJCIiIlUa1MJz6623snz5cgCysrK44oorWLVqFY8//jhPPfWUWwv0Fl0SwkmMCORYhYMVO3I9XY6IiIgcp0GBZ9OmTfTv3x+ADz/8kG7durFixQree+895s6d6876vIbNZmNolzgAlvya4+FqRERE5HgNCjwVFRUEBAQAsHTpUn7zm98A0KlTJzIzM91XnZcZ0tkMPMs2Z+NU52UREZHzRoMCT9euXZkzZw7ffvstS5YsYfjw4QAcPHiQqKgotxboTS5q05wQfx9yCsvYeCDf0+WIiIhIlQYFnmeffZbXXnuNQYMGccstt5CamgrAp59+6rrUdSEK8PXhso4xgNnKIyIiIueHBt2lNWjQIHJzcykoKCAyMtK1fPz48QQHB7utOG80tHMcizZmsWRzDpOu7OjpckRERIQGtvAcO3aMsrIyV9jZs2cPM2fOJCMjg9jYWLcW6G0Gd4zFboPNmQXsP1Li6XJERESEBgae6667jnfffReAo0ePkpaWxosvvsjIkSN59dVX3Vqgt4kM8advq+YALNusu7VERETOBw0KPGvXruWSSy4B4KOPPiIuLo49e/bw7rvv8vLLL7u1QG80tIvZyrVU/XhERETOCw0KPCUlJYSFhQHwxRdfcMMNN2C327nooovYs2ePWwv0RkOrbk//YedhCksrPFyNiIiINCjwtGvXjk8++YR9+/axePFirrzySgBycnIIDw93a4HeqE1MKG1iQqhwGHyzVaMui4iIeFqDAs/UqVN5+OGHSUlJoX///qSnpwNma0+vXr3cWqC3qm7l0WUtERERz2tQ4LnxxhvZu3cvq1evZvHixa7lQ4YM4a9//avbivNm1YHnyy05VDqcHq5GRETkwtagcXgA4uPjiY+PZ//+/QC0aNHigh508ES9WzYjMtiPIyUVrN5zhIvaXLgjUIuIiHhag1p4nE4nTz31FBEREbRq1YpWrVrRrFkz/vSnP+F0qjUDwNfHzuBO5t1aGnVZRETEsxoUeB5//HFmzZrFX/7yF37++Wd+/vlnnnnmGV555RWmTJni7hq91hWdq2dPz8YwNJmoiIiIpzToktY777zD3//+d9cs6QA9evQgKSmJ++67j6efftptBXqzSzrE4O9jZ/fhEnYcKqZdbKinSxIREbkgNaiFJy8vj06dOp20vFOnTuTl5Z1zUVYRGuDLRW3Nvju6W0tERMRzGhR4UlNTmTVr1knLZ82aRY8ePc65KCu5onPVqMu/KvCIiIh4SoMuaT333HNcc801LF261DUGz8qVK9m3bx+LFi1ya4HebkjnOKb8+xfW7j3C4aIyokIDPF2SiIjIBadBLTyXXXYZW7du5frrr+fo0aMcPXqUG264gV9++YV//OMf7q7RqyU2C6JLQjhOA5ZnHPJ0OSIiIhckm+HG24fWr19P7969cTgc7tql2xUUFBAREUF+fn6TTYPx0pKtvLxsG8O7xjPntj5NckwRERErOde/3w1q4ZH6qb49/ZtthyitOH/DoIiIiFUp8DSBbknhxIUHUFLu4Iedhz1djoiIyAVHgacJ2Gw2TSYqIiLiQfW6S+uGG2444+dHjx49l1osbWjnON77cS9Lf83hT9cZ2Gw2T5ckIiJywahX4ImIiDjr52PHjj2ngqwqvW0UQX4+ZBWU8svBArolnflcioiIiPvUK/C8/fbbjVWH5QX6+XBph2gW/5LNkl+zFXhERESakPrwNKHqfjzLtqgfj4iISFNS4GlCgzvFYrPBpgMFZOYf83Q5IiIiFwwFniYUHRpA75aRACzdnOPhakRERC4cCjxNzHV7uiYTFRERaTIKPE3sii7m7OkrdxymqKzSw9WIiIhcGBR4mljbmFBSooIpdzj5bpsmExUREWkKCjxNzGazMaTqstaSX9WPR0REpCko8HhAdT+eL7dk43C6bbJ6EREROQ0FHg/omxJJRJAfR0oqWLv3iKfLERERsTwFHg/w87EzuGMMoLu1REREmoICj4cM7aLZ00VERJqKAo+HXNohBl+7jR2Hitl5qMjT5YiIiFiaAo+HhAf6cVGbKACWadRlERGRRqXA40FDO5uDEC7RZS0REZFGpcDjQdXj8azenceR4nIPVyMiImJdCjwelNw8mE7xYTgN+GqrLmuJiIg0lvMi8MyePZuUlBQCAwNJS0tj1apVddrugw8+wGazMXLkyMYtsBHVTCaqwCMiItJYPB545s+fz6RJk5g2bRpr164lNTWVYcOGkZNz5gCwe/duHn74YS655JImqrRxVN+e/vXWQ5RVOjxcjYiIiDV5PPC89NJL3H333dxxxx106dKFOXPmEBwczFtvvXXabRwOB2PGjGH69Om0adOmCat1vx5JEcSEBVBUVsmPO/M8XY6IiIgleTTwlJeXs2bNGoYOHepaZrfbGTp0KCtXrjztdk899RSxsbHceeedZz1GWVkZBQUFtR6NJmsTZK6v1yZ2u811t5YGIRQREWkcHg08ubm5OBwO4uLiai2Pi4sjKyvrlNt89913vPnmm7zxxht1OsaMGTOIiIhwPZKTk8+57lPa9C947VL49wRwVNZr0yGdzO+/bHMOhqHJREVERNzN45e06qOwsJDbbruNN954g+jo6DptM3nyZPLz812Pffv2NU5xrS+DgDDI2girT3857lQGtosm0M/OgaPH2JxZ2Dj1iYiIXMB8PXnw6OhofHx8yM6ufSknOzub+Pj4k9bfsWMHu3fvZsSIEa5lTqcTAF9fXzIyMmjbtm2tbQICAggICGiE6k8QEg1DpsJnk+DLP0PXkRAaW6dNg/x9uLhdDEs3Z7N0czZdEsMbt1YREZELjEdbePz9/enTpw/Lli1zLXM6nSxbtoz09PST1u/UqRMbN25k3bp1rsdvfvMbBg8ezLp16xrvclVd9bkdEnpCWT4smVavTa/oon48IiIijcWjLTwAkyZNYty4cfTt25f+/fszc+ZMiouLueOOOwAYO3YsSUlJzJgxg8DAQLp161Zr+2bNmgGctNwj7D5wzUvw9yGwfh70HgutTg5upzK4kxl4NuzPJ7uglLjwwMasVERE5ILi8T48o0eP5oUXXmDq1Kn07NmTdevW8fnnn7s6Mu/du5fMzEwPV1kPLfqYQQdg0cN17sAcGxZIz+RmgCYTFRERcTebcYHdFlRQUEBERAT5+fmEhzdSX5niwzCrDxw7AsOfhYvuqdNms5dv5/nFGVzeKZa3bu/XOLWJiIh4oXP9++3xFh5LCokyOzADLH8aCuvWL6d6monvtudSUl6/W9tFRETk9BR4GkvvcZDYC8oKYMnUOm3SIS6U5OZBlFc6+XZbbiMXKCIicuFQ4Gksdh+45kXABhs+gN3fn3UTm8123CCEultLRETEXRR4GlNSH+gzzny96GFwVJx1kyu61Iy67HBeUN2rREREGo0CT2MbMg2CIiHnV1h19ukw+rduTligL4eLy1m372jj1yciInIBUOBpbMHNYeiT5uvlz0DhqecIq+bnY2dQR3NMnvd+2INTrTwiIiLnTIGnKfQaa17eKi+sUwfmUb2TAPj45wNM+nAd5ZXOxq5QRETE0hR4moLdDle/gNmBef5ZOzAP6hjLC79Nxddu45N1B/n93J8oLD17/x8RERE5NQWeppLUG/qa02XUpQPzjX1a8PdxfQn29+G77bmMfu0HcgpKm6BQERER61HgaUqXT4Gg5lUdmF8/6+qDOsYyf3w60aH+/JpZwA2vrmDHoaImKFRERMRaFHiaUnBzuGK6+Xr5DCg4+xxh3VtE8PG9A0mJCmb/kWOMenUFa/YcaeRCRURErEWBp6n1/B0k9a3qwDylTpu0jArmX/cOIDW5GUdLKrj1jR9Y8qsGJhQREakrBZ6mZrfXjMC8cQHs+rZOm0WFBvD+3WkM7hhDWaWTP/xjNe/9uKdxaxUREbEIBR5PSOwJ/e40X9dxBGaAYH9f3hjbl5v6tsBpwOMLN/HSFxlcYBPei4iI1JsCj6dc/gQER8GhLfDjnDpv5utj59lRPfifIe0BePnL7Tz2r41UOjRWj4iIyOko8HhKUCQMrerA/NVfoOBgnTe12WxMuqIDz1zfHbsN5q/ex93vrqakvLKRihUREfFuCjye1HMMtOgH5UXwxRP13vzWtJa8dltfAv3sLM84xC2v/8DhorJGKFRERMS7KfB4UvUIzDY7bPoX7Py63ru4oksc7911EZHBfqzfn8+oV1ew53BxIxQrIiLivRR4PC2xJ/St7sD8/6CyvN676NMqko/uHUCLyCB2Hy5h1Ksr2LD/qFvLFBER8WYKPOeDyx+H4GjIzYAfX23QLtrGhPLxvQPokhBOblE5N7/+A19l5Li5UBEREe+kwHM+CIqEK54yX3/1LOQfaNBuYsMDmf+Hi7i4XTQl5Q7uemc1H63Z78ZCRUREvJMCz/ki9RZIToOKYvji8QbvJizQj7du78fInolUOg0eXrCe2cu3a6weERG5oCnwnC+O78D8y0LY+VWDd+Xva+elm3ryh8vaAPD84gye+s+vOJ0KPSIicmFS4DmfJPSAfnebrxvYgbma3W5j8lWdmXptFwDe/n43j/5rAw6FHhERuQAp8JxvBv8RQmIgdyssmw7neCnq9xe35sXfpmK3wYI1+5n4/lrKKzUqs4iIXFgUeM43Qc1g2Azz9cpZ8OlEcJzbCMqj+rTgb2N64+djY9HGLO5+dzXHyh3nXquIiIiXUOA5H/X4LVz7V7M/z8//gA9ugfJzG0xweLcE3hzXj0A/O19vPcS4t1ZRWFq3SUtFRES8nQLP+arv72H0P8E3ELZ9AXOvheLcc9rlpR1i+OedaYQF+LJqdx63vvEjecUN7yckIiLiLRR4zmedroFx/2eO03NwLbx5BeTtPKdd9k1pzvvjL6J5iD8bD+Qz+rWVZOWXuqlgERGR85MCz/kuuT/cuQSatTTDzptXwoG157TLbkkRfPiHdOLDA9mWU8RvX1vB3sMlbipYRETk/KPA4w2i25uhJ747FB8yL29tW3JOu2wXG8qCe9JpFRXMvrxj/Pa1FWzLLnRTwSIiIucXBR5vERYPty+CNoPM0ZjnjYaf3zunXSY3D2bBH9LpEBdKdkEZN722ko37891Tr4iIyHlEgcebBIbDrQugx2gwHPDv++Cb589prJ7Y8EDmj08ntUUER0oquOWNH/hx52E3Fi0iIuJ5Cjzextcfrn8NLn7IfP/ln+GzSeBs+Lg6kSH+vHf3RaS1bk5RWSVj31qlmdZFRMRSFHi8kc0GQ5+Eq54HbLD6LZh/G5Q3vONxaIAv7/y+P5d3iqWs0snd765m0cZMt5UsIiLiSQo83ixtPNz0DvgEQMZn8O51UJLX4N0F+vkw53d9uLZHAhUOg/vnreXD1fvcWLCIiIhnKPB4uy7Xwdh/Q2AE7F9l3rZ+ZE+Dd+fva+d/b+7Fzf2ScRrwyEcbeOu7XW4sWEREpOkp8FhBq3T4/RcQ3gIObzMHKMxc3+Dd+dhtzLihO3df0hqAp/7zKy8v24ZxjhOZioiIeIoCj1XEdoK7lkBcNyjKhrevgR3LG7w7m83GH6/uzKQrOgDw0pKtPLNos0KPiIh4JQUeKwlPhDsWQcolUF4I790IGz5s8O5sNhv/M6Q9U6/tAsAb3+7i0X9toKjs3GZvFxERaWoKPFYTGAG/+xd0GwXOSvj4bvhu5jmN1fP7i1vz3Kge2G3w4er9XPLsl/ztq+0UK/iIiIiXsBkX2DWKgoICIiIiyM/PJzw83NPlNB6nE5ZMgZWzzPeJvcyxezpdC3afBu1y+ZYc/vSfX9mZWwxA8xB//nBpG25Lb0Wwv6+7KhcRETnJuf79VuCxuh/mwNInofKY+T6qHQz4H0i9GXwD6r27SoeTT9cf5H+XbWNP1YSj0aH+3HNZW8aktSLIv2FhSkRE5EwUeOrpggs8AMW58ONrsOp1KD1qLguNh/T7oM8d5pQV9VTpcLLw5wO8/OU29uWZYSomLIB7L2vLrWktCfRT8BEREfdR4KmnCzLwVCsrgrXvwIpZUHjQXBYQAf3vgrR7IDS23ruscDj5eO1+XvlyO/uPmMEnLjyACYPbMbpfMgG+Cj4iInLuFHjq6YIOPNUqy2HjAvh+JuRuNZf5BkLPMTBgIjRvXe9dllc6+WjNfmZ9uY2D+aUAJEQEMmFwO27qm4y/r/rHi4hIwynw1JMCz3GcTshYBN/9FQ6sNpfZ7ND1Brj4QYjvXu9dllU6+HD1fmZ/uZ2sAjP4JDUL4v7L23Fjnxb4+Sj4iIhI/Snw1JMCzykYBuz53gw+25fWLG831Lyzq9VAc8LSeiitcDD/p33MXr6dnMIyAFpEBjHx8nbc0FvBR0RE6keBp54UeM4ic4N5qeuXhWA4zWUt+sHAB6Hj1WCvX1AprXAw78e9/O2rHeQWmcGnZfNgJl7ejut7JeGr4CMiInWgwFNPCjx1lLfT7Nz88z/BYQYVojuYt7R3vxH8guq1u2PlDt77cQ+vfrWDw8XlAEQG+9GjRTNSk5uR2iKCHi2aERNW/1vlRUTE+hR46kmBp56KcuDHObDq71CWby4LioRet0Hf39e7g3NJeSX/WLmH177ZSV5V8DleUrMgelSFn9TkCLonRRAW6OeObyIiIl5MgaeeFHgaqLQA1rwNq96A/H1VC21mP5/+d5vP9RjBubzSyebMAtbvP8r6ffms33+UHYeKTpoBw2aDtjGh9GgRQWpVa1DnhDDd7i4icoFR4KknBZ5z5HTAti/gp7/X7uDcrKXZ4tPrNgiJbtCuC0sr2HTADEEbqoLQgaPHTlrPz8dGp/hwUpOrWoJaNKNtTIj6A4mIWJgCTz0p8LjR4R2w+i2zn0/1CM4+/uZt7f3ughZ9631314kOFZaZ4Wd/flUIOsqRkoqT1vO122gZFUyb6BBaR4fQJibUfI4OISYsANs51iEiIp6lwFNPCjyNoOIYbPqXebkrc13N8vge5uWubjeCf7BbDmUYBvuPHGPdvqOuILTpQD4l5Y7TbhPi70PrmBDaRFeFoBgzFLWODlH/IBERL6HAU08KPI3swBr46U3Y+FHN3V2BEeYozn3vhOh2bj+k02mQWVDKrkPF7MotYmduMbtyi9l5qJj9R0pwnuEXHhMW4GoJqm4Z6t2yGVGhulvsdHIKS/nlYAHbs4uIDPGnY1wY7WJDNXGsiDQqBZ56UuBpIiV55qWu1W/Ckd01y9sMNi93dRgOPr6NXkZZpYN9eSXsPGSGoOogtDO32DUu0Kl0ig9jQNtoBrSNIq1N8wuyJcjpNNiTV8IvB/P59WABv1Q9TnXebDZzfKUOcWF0jAujQ3wYHeJCaRMdqmlFRMQtFHjqSYGniTmdsGOZ2cl562Kg6ucW3gJ63wbdfwtRbT1SWkFpBbuPC0G7covJyCokI7uw1no+dhvdkyIY2C6KAW2j6dMq0nKzwZdVOtiWXVQr3GzOLKD4FJcKbTZoEx1Cp/hw8orLycguPOUQA2D2rWodHUKH+KogFBdKh7gwWkWF4GNXvyoRqTsFnnpS4PGgI3uqOjn/A0oO1yxP7G0Gn243QFi85+qrkltUxg87D7Nix2FWbM9l9+GSWp/7+9rp0zKSAW2jGNAuih4tmnnVVBkFpRWuUGM+57M9p4jKU1z7C/C10yk+jC6JEXRNDKdLYjid4sMI9q/dOpdbVMbWrEK2ZheSkV3E1uxCtmYVUlhWecoaAnzttIsNpWNcGO3jwuiUEMaAtlEabkBETkuBp54UeM4DFaWw+VPYMB92LAejqhXBZoeUS8zw03kEBDXzaJnVDhw9xortuazccZjvd+SSXVD7kk6Ivw/9WzdnYLto0ttG0Tk+HHsTt16UVjg4XFzO4aIyDheVk1tUVvt91evcorKT6q8WEeRH18TwqkcEXRLDaRPd8Nv9DcMgq6CUjOoglFXEthzzdWmF86T1o0L8uaV/S8Zc1JKEiPqN5C0i1qfAU08KPOeZokPw6yewcQHs+7FmuY8/tL8SetxkPtdzKovGYhgGO3OLXa0/K3ce5ugJt8lHBvuR3jaK3i0jCfDzwQbYbTZsNrDbwIb52mazme9t5ufUWs+GDXMdmw3KKp2u8HK4uIzcoqowU1zO4aJyik7TknI6Sc2C6HJCuEmMCGyS2/cdToP9R0rIyCpkW04RGVmFrNqVR1ZBKWBeQhzeNZ5xA1LolxKpIQVEBFDgqTcFnvPYkd3m3V0bF8ChLTXLA8LNFp/uN0LKpU3S2bmunE6DzVkFrNh+mBU7clm1K++U/V6agr+PnahQf/MREkBUqD/RoQFEhfgTFVr1PiSA5OZBNAv290iNp1PpcLLk12zeXrGbVbvyXMs7J4Rz+4BWXNczyXL9pkSkfhR46kmBxwsYBmT/YgafTf86bioLICTW7OvT/beQ1OecBzZ0twqHkw37j7Ji+2G2ZBXiNAychoFhUHV7vIHTMFuKnIbZhdtwfX7c8wnL/XzsRIcFEH1ceIkKCSA6tOZ9WICvJVpDNmcW8M6K3Xyy7oDr0lezYD9u7teS313UkhaR7hnTSUS8iyUCz+zZs3n++efJysoiNTWVV155hf79+59y3TfeeIN3332XTZs2AdCnTx+eeeaZ065/IgUeL+N0mpe6Ni6AXxbCsZr/+icypaqz843mnV5Oh9kfyPXsrHquPOEz5ynWrXrYfcwBE33PrxaQC9HRknLm/7SPf/ywh/1HzClG7Da4oksc4wakkN4myhIBT0TqxusDz/z58xk7dixz5swhLS2NmTNnsmDBAjIyMoiNjT1p/TFjxjBw4EAGDBhAYGAgzz77LAsXLuSXX34hKSnprMdT4PFijgqzk/PGBbDlM6gobpzjhMRAr99Bn9vNUCUe5XAaLNuczTsrd/P99pq7+zrGhTF2QCuu75V00l1jImI9Xh940tLS6NevH7NmzQLA6XSSnJzMxIkTeeyxx866vcPhIDIyklmzZjF27Nizrq/AYxHlxZDxX7PPz/YlZivO6dh8zJYbmw/YfcFuP2FZ9bPdnBXe1Ypkg3ZDoM8dTTZQopzZtuxC3lm5m4/XHnBNJxIe6MtNfZMZm55Cy6iGX+4yDINjFQ6KyiopLnNQXFZJWaWDSoeBw2lQ6Tz+2el6f9JnjuM+MwwcDnN5z+RmDOoYo1YpkQby6sBTXl5OcHAwH330ESNHjnQtHzduHEePHuXf//73WfdRWFhIbGwsCxYs4Nprrz3p87KyMsrKam7DLSgoIDk5WYHHSspLoLLUDC5235NDTH04KmDr57D6bXPAxGphidB7rPmIOHtLojSu/GMVfLRmP++u3M2eqnGSbDa4vGMsI3uZ/z7FZZU14aW8+nVNmDlpWXkljf3/hj2Tm/HwlR0Z2E6X40Tqy6sDz8GDB0lKSmLFihWkp6e7lj/yyCN8/fXX/Pjjj2fY2nTfffexePFifvnlFwIDA0/6/Mknn2T69OknLVfgkbPK2wlr3jGnyCjJNZfZ7GZrT9/fQ9vLzWAlHuN0Gny1NYe5K/bwzdZDbtmnzQYh/r6EBPgQ6OeDr92Gr92O3W7D127D5/hnHxt2W/V7u/nsU/O5j81cp6zSyaKNma5O2Gmtm/P/hnWkb0pzt9QsciG4oAPPX/7yF5577jm++uorevToccp11MIj56yyDDb/H6yZC7u/rVnerCX0Hge9boOwOI+VJ6Ydh4r4x8o9/Lz3CIF+PoQG+BJS9QgN8DFf+1cvq/m8Zj1zWZCfT6O0vuQUlvK35TuY9+Neyh1m8LmsQwwPX9mR7i0i3H48Eavx6sBzLpe0XnjhBf785z+zdOlS+vbtW+djqg+PnJNDW2HN27BuHpQeNZfZfaHTtdD3DnOcoPpeRpMLysGjx3jly218uHo/jqrpPIZ1jWPSFR3pGB/m4eq8m2EYZBeUkZFdyI4DuQTu+hxbdAf6pF1K+9hQXUb0cl4deMDstNy/f39eeeUVwOy03LJlS+6///7Tdlp+7rnnePrpp1m8eDEXXXRRvY6nwCNuUXEMfvnEnBts/6qa5c3bmsEn9VYIiTr34zid4Kyo6myty2dWsju3mP9dto1P1h3AMMxLab9JTeTBoR1oHR3S6McvLK1g/5FjtIsN9aq54KrlFZfXTFuSXci27EIysgqh9Ci/81nKHb6fE2MrAGChYyDzQ8eS2j2VK7vG0yu5WZNP/yLnzusDz/z58xk3bhyvvfYa/fv3Z+bMmXz44Yds2bKFuLg4xo4dS1JSEjNmzADg2WefZerUqcybN4+BAwe69hMaGkpoaOhZj6fAI26XtdHs5LzhQyivmmndxx/aXWFOieEoNztDOytqXjuOf11e9Vn1ssqq5/Kaecb8QqBVOrS+FFpfBvHdFYCqleZD/n7IP2AOUul0QMeroFmypyurk23Zhfx16VYWbcwCzKk1buzdgv8Z2p6kZu6bUiWnsJTVu4+walceP+3OY3NmAU4DwgJ9ubR9DIM7xTKoYwzRoQFuO6Y7FJZWsLVqQlpzOhJzXrbcotpzwiVwmN/7/pdbfL4k1GZOU1LiG0lw5REAygxf3nEMY3bldfiHRXFFlziu7BLHgLbR+Pt6X+C7EHl94AGYNWuWa+DBnj178vLLL5OWlgbAoEGDSElJYe7cuQCkpKSwZ8+ek/Yxbdo0nnzyybMeS4FHGk1ZEWz6CH56E7I2NO6xAptB60vM8NP6Mohu37ijTjsdkLsVDqyFgz/DwbXmaNg+/ua4RaFxEBpjjoQdGlu1LLbmfWhsw+ZDqyyHwoNVgeYUj4IDUFZw6m1TLoEeo6HLdRDoof+tF2Sa/b+2/td8H90RYjpUPXeEkGjXqpsO5PPiFxkszzA7X/v72LmlfzITBrcjNvzkGzLOxDAMdh8u4addeazancfq3Xnsrrqb7XiBfvZaE7nabJDaohmXd4rl8k6xdE0Mb7LLQA6nwa7cIjYeyGdLptlyszW7iANHj512m5bNgxkUeZibyj+mS+5i7EbV8BSxXWHgA+ao7NmbqFw8Bd89Zv+7fCOEVypH8g/HFZRhjlA+uFMsV3aNY1DHWEIDGnf4icLSCorLHMRH1O/f1JuUVjgoKXfQPMS9A7haIvA0JQUeaXSGYQaCPStrbpX38a96+FU9ql7b/Y777Lj1am3ja/5x3/k17PoGdn9X05JULSyhqvWnqgXoXFo3DMO8Q+3gzzUBJ3P9uQ/06B92XCg6IRwFRkDxoZMDTVE25gQcZxHU3BwuICLZHEtpz3c1n/kGQqdroMfN5p11jT2eUv5++PVT+PXfsO+HM68bHHVCCOrAhrI4ZnxXxMqqOcUC/eyMS0/hnsvaEnmaPyCVDiebMwv5aXde1eNIVQuIQTOKiLcdIcF+mNTwY/QIL6FtYD5xtjwCKkvIiejK945uzMtuwerM2uNZxYUHMLhjLIM7xXJxu2hC3BQGHE6DHYeK2Lg/n00H89l0IJ9fDha4xlY6UXx4IB3iw+gYF0qHuKrnsk0ErpplDiNRLeUSM+i0G1r7PwAMA7YvhSVTIedXAI74xfOSczT/LO6HgdnC4+9jZ2C7KIZ1jWdol7gGtXaVVjg4cPQY+/JK2HfkGPvzSth3pIR9ecfYd6TENdlw+9hQrumRwLU9EmgXa42+W3nF5by7cjfvrtzDlV3i+MuoU99M1FAKPPWkwCNez1EJmetg51ew62vY+yM4ajfv07xNTfhpfWmtloRaDMP8A31wbU3AyVxnXiY6kV8IJKRCUm9I7GW+NgwozoGiHDOwFGUf9/q45xPrqw+fAIhocfIjvCrgRCSB/wl9Xo7ug40fwvr5kJtRszwkxpyOpMdos353tV4c2V0Tcg6srv1Zchp0/g0EhJmtZIcyzEf+3tPvzy+EorDW/FQSy0+FMWw3Ejno14ohA9O489L2+NntrNuTy+Zt29i7eztHsvYQ6cgl3naEeNthM+DY8kiwHcGf8rp9B7sv5fF92BLSh0XFnZi3P4qC8prz4+9jJ61Nc1frT6uouvUzqnQ42XGomI0HzGCz8UA+vx4s4FjFyeEmyM+HronhdEkMp2N8GB3iwugQG0ZEsJ+5gtMJGYvg+/89ru+czZxceOAD0OIsN7A4HbD+ffjyz1CYCUBxVDf+HXMPr+9rUasVzGaDvq0iGdY1niu7xLsGtax0OMnML2XfkRL2V4WY6nCzL6+EnMKz/9bttuq59Uwd48Jc4adNzNm7ZpxvducW8/fvdvLRmv2uFsM2MSEsfvBSt/YPU+CpJwUesZyKUnO+sV3fmAHowNqavj/V4rrVBCCoHXCqxxg6nk+A2U8osVdNwInu0LB+Q4ZhXnYqqg5GOVB0qHZQOnbUbO05KdS0MMNaQ4OJYZjfc8N8c1Tu479rTCdIvRm639SwwSQP7zADzq//NkOiiw1aDTAvpXUeAeGJp96+vBhyt9WEoNyqIJS387Qjh5cbPhwgliBKieEoPrY6/t93cJQ5eGZ4IoQn1Lz2DYA935tTthyt3VXA8A8lL6Y/q2w9mJfblm+PNje/W5W2MSFc3sls/emX0hw/HzuVDifbcopc4WbTgXx+zSyoddnMVZK/GW66JUXQverRJiYUn1N1Jq4sM/8Nv38ZDm8zl/kEQM9bIH0iRLer23moVl4CP/wNvpvpai012l3Bnt6P8H+ZkXzxazYbD9QO/W1jQih3OMk8Wkql88znPcTfh+TmwbSIDCa5eRDJkcEkNzdft4gMxmkYLP01m882ZPLNtkNUOGr21zkhnGt7JHBN9wRSmqDz+rn4ee8RXv9mJ5//kuUatLN7UgTjL23DVd3i8XVzZ3gFnnpS4BHLKy2APSvM8LPrG8jedOb17b4Q2xkSe9cEnJjO1ptA1VEB25fBhg9gy6LjWp1sZn+o1FvMgBJwhssLudvg10/MkJO1sWa5zQ6tBtaEnLD4c6szb+dxIWgrRm4GzpwMfByltVfFTmlADLaIJIKiWmALTzIDTXiSeZkzPNF89qtDf5G8XWarYXXL4bEjtT6uDI5jZ3hflpR25r2c1hx0Rro+CwvwJSU6hK3ZhZRVnhxuQvx96JoYYYabFuF0T4qgdfRpws3xSvPNGwJ+eBWKzE7dBERAvzsh7Z5zH/+qOBe+fg5Wv1kVMm3QcwwM/iMHjOZ88UsWX/ySzardea4hBMBs7UqKDKJFZJAZZE4INpHBfnXu+5R/rIIvfsnis42ZfLctt1aY6poYbrb8dE88p2lT3MnpNFi2JYc3vtnJqt01kzkP6hjD+EvbNOqkvgo89aTAIxecokOw+5uq/j/fm600ib1qAk58t4Z1KPZmpflmaFn/gdnCUc03CDpfa7b8tB5knqtDW2pacqr6fwDm1CWtLzVDTqdrzX5JjcnpxHF0LzsyNhLRLIrYpBRsoXGNc7ee02l2vN/5FexcbvZHO+GyZGFYO372S+Xjo+1ZWtKOIsw/yKEBvnRNNENN9xZmyGkdFVK/28ALMuHHV82wU90pPSwR0u8zB/t0dyf0wztg2VNmmAXzd5B+n3mZLDCCI8XlrN5zhIggP5KbBxEXFtgot7UfLSnni1+y+c/GTL7fnlsrZPVoEcE13RO4unsCyc2bPvyUVjj45OcDvPHtTnYcMvvz+fnYuK5nEndf0qZJxpBS4KknBR4RqeXInqr+Ph/A4e01y0PjzM7UuVtrltn9oM2gqpBzDQRfIFNDVBwzL5tWtwAdXMfxnckNmw9FYW3w9w/A388HG7aqy5D1fca8DLl/lTksA5iXHgf8j9n3qrFbHff9BEumwN6V5vvgKLjsUXMC4SZu8cwrLueLX7L4z4ZMVuzIrdXnJzW5GSN6JHBV9wS3Dl1wKvklFfzzxz28/f1u11AAYYG+jElrxe0DUpr0bjMFnnpS4BGRUzIMs0/Thg/M/j7Hqprrffyh7RAz5HQcDkGRZ97PhaAkz2wxrA5AR3a5/xgt080WlvbDmnb0csMwO0YvmVbTXyiyNQyZYraK2uzmw+5T89rmYwa2Uy6vXr/h3+FwURmf/5LFZxsy+WHn4Vrhp3NCOK2jg4+7tBZMy+bBJDULOqfxhfbllfDmd7v4cPU+191ziRGB/P7i1ozul0xYoF+D991QCjz1pMAjImdVWW7+Ia8oMW9l99Q4Pt7iyG7I3Q4YZmA41TOc/rNaz5h3GSb1bvKvUYujEn5+F5bPMDvYu0N1+PELNvvNxfcw7xZMSDVbsurQinSosIzPN2Xynw2ZrNqdx+n+gttskBAeSIuqINSyqtO0+RxMTGjAKS/LbTqQz2vf7GTRxkzXJbXOCeGMv7Q11/ZI9Oio3Ao89aTAIyIidVZWBCtnmdPIlBWB4ax6OGpeu4OP/wkhqCfEdQX/0/fXySkoZf3+fPbllbA3r4T9VeP97M0rOeVt/8fz97WTfFyn68RmQXyz9RArdx52rXNJ+2jGX9qGi9tF160j8vF9v0Kiodfv6vjl60aBp54UeERExK2czuOC0AlhyOkwW66OX16aD1mbzOEMsjaYA3ueauwrm90cDiIhtSYIxXeHoGZnLMcwDA4Xl7M3r2qMoLyaILTvSAmZ+aW1OkQfz9duY0RqIndd0pquiRFn/+6nu7svsReM/+rs29eDAk89KfCIiMh5xTDMcZAyq8JP9eN0l9IiU2oHoKh20KxVnUcRr6gaT2jfkRJXKNp/5BiJzYIYm96KxDN1hC4+bAab6pBzwvhN+IdBysXmpeD+d7t1yhsFnnpS4BEREa9QmHVcCFpntgYdPc0I3XY/s+9TVDtzIMao9lWv25t3mzU0eJSXmHetVQecE+cJtPtCi/7m3YttBpl9r3wap0Pzuf79buRJZURERKRBwuLNR4cra5aV5FVdBqsKQjmbIW8HVJaaA1XmZkDGCfsJbFYTfqLa1bxu3ubkMbicjpqpa3YsN4cjcJwwPUls15qA02oABHjHdBhq4REREfFmTicU7DdHAj+83XxUv87fd4YNbeZ8dNFVIagw0xxu4MT+ROFJ0GawGXBaX3ruI1w3kC5p1ZMCj4iIXDAqjpkjSR+uCkC5VYHo8LZTd5QGc/qO1pfUtOJEtXNrX5yG0iUtEREROTW/IHP6mPhutZcbhjmXWHX4ObzdnEeuzWDzlvg6doD2Jtb7RiIiInJmNps5/1toDLRK93Q1TcJzQyaKiIiINBEFHhEREbE8BR4RERGxPAUeERERsTwFHhEREbE8BR4RERGxPAUeERERsTwFHhEREbE8BR4RERGxPAUeERERsTwFHhEREbE8BR4RERGxPAUeERERsTwFHhEREbE8BR4RERGxPAUeERERsTwFHhEREbE8BR4RERGxPAUeERERsTwFHhEREbE8BR4RERGxPAUeERERsTwFHhEREbE8BR4RERGxPAUeERERsTwFHhEREbE8BR4RERGxPAUeERERsTwFHhEREbE8BR4RERGxPAUeERERsTwFHhEREbE8BR4RERGxPAUeERERsTwFHhEREbE8BR4RERGxPAUeERERsTwFHhEREbE8BR4RERGxPAUeERERsTwFHhEREbE8BR4RERGxPAUeERERsTwFHhEREbE8BR4RERGxPAUeERERsTwFHhEREbG88yLwzJ49m5SUFAIDA0lLS2PVqlVnXH/BggV06tSJwMBAunfvzqJFi5qoUhEREfFGHg888+fPZ9KkSUybNo21a9eSmprKsGHDyMnJOeX6K1as4JZbbuHOO+/k559/ZuTIkYwcOZJNmzY1ceUiIiLiLWyGYRieLCAtLY1+/foxa9YsAJxOJ8nJyUycOJHHHnvspPVHjx5NcXEx//nPf1zLLrroInr27MmcOXPOeryCggIiIiLIz88nPDzcfV9EREREGs25/v32bYSa6qy8vJw1a9YwefJk1zK73c7QoUNZuXLlKbdZuXIlkyZNqrVs2LBhfPLJJ6dcv6ysjLKyMtf7/Px8wDxxIiIi4h2q/243tJ3Go4EnNzcXh8NBXFxcreVxcXFs2bLllNtkZWWdcv2srKxTrj9jxgymT59+0vLk5OQGVi0iIiKeUlhYSERERL2382jgaQqTJ0+u1SLkdDrJy8sjKioKm83m1mMVFBSQnJzMvn37dLmsCem8e4bOu2fovHuGzrtnHH/ew8LCKCwsJDExsUH78mjgiY6OxsfHh+zs7FrLs7OziY+PP+U28fHx9Vo/ICCAgICAWsuaNWvW8KLrIDw8XP+D8ACdd8/QefcMnXfP0Hn3jOrz3pCWnWoevUvL39+fPn36sGzZMtcyp9PJsmXLSE9PP+U26enptdYHWLJkyWnXFxEREfH4Ja1JkyYxbtw4+vbtS//+/Zk5cybFxcXccccdAIwdO5akpCRmzJgBwAMPPMBll13Giy++yDXXXMMHH3zA6tWref311z35NUREROQ85vHAM3r0aA4dOsTUqVPJysqiZ8+efP75566OyXv37sVur2mIGjBgAPPmzeOJJ57gj3/8I+3bt+eTTz6hW7dunvoKLgEBAUybNu2kS2jSuHTePUPn3TN03j1D590z3HnePT4Oj4iIiEhj8/hIyyIiIiKNTYFHRERELE+BR0RERCxPgUdEREQsT4HHTWbPnk1KSgqBgYGkpaWxatUqT5dkaU8++SQ2m63Wo1OnTp4uy3K++eYbRowYQWJiIjab7aQ56wzDYOrUqSQkJBAUFMTQoUPZtm2bZ4q1kLOd99tvv/2k3//w4cM9U6yFzJgxg379+hEWFkZsbCwjR44kIyOj1jqlpaVMmDCBqKgoQkNDGTVq1EmD4Ur91OW8Dxo06KTf/D333FOv4yjwuMH8+fOZNGkS06ZNY+3ataSmpjJs2DBycnI8XZqlde3alczMTNfju+++83RJllNcXExqaiqzZ88+5efPPfccL7/8MnPmzOHHH38kJCSEYcOGUVpa2sSVWsvZzjvA8OHDa/3+33///Sas0Jq+/vprJkyYwA8//MCSJUuoqKjgyiuvpLi42LXOQw89xP/93/+xYMECvv76aw4ePMgNN9zgwaq9X13OO8Ddd99d6zf/3HPP1e9Ahpyz/v37GxMmTHC9dzgcRmJiojFjxgwPVmVt06ZNM1JTUz1dxgUFMBYuXOh673Q6jfj4eOP55593LTt69KgREBBgvP/++x6o0JpOPO+GYRjjxo0zrrvuOo/UcyHJyckxAOPrr782DMP8ffv5+RkLFixwrbN582YDMFauXOmpMi3nxPNuGIZx2WWXGQ888MA57VctPOeovLycNWvWMHToUNcyu93O0KFDWblypQcrs75t27aRmJhImzZtGDNmDHv37vV0SReUXbt2kZWVVeu3HxERQVpamn77TeCrr74iNjaWjh07cu+993L48GFPl2Q5+fn5ADRv3hyANWvWUFFRUes336lTJ1q2bKnfvBudeN6rvffee0RHR9OtWzcmT55MSUlJvfbr8ZGWvV1ubi4Oh8M1MnS1uLg4tmzZ4qGqrC8tLY25c+fSsWNHMjMzmT59OpdccgmbNm0iLCzM0+VdELKysgBO+duv/kwax/Dhw7nhhhto3bo1O3bs4I9//CNXXXUVK1euxMfHx9PlWYLT6eTBBx9k4MCBrpH8s7Ky8Pf3P2kCav3m3edU5x3g1ltvpVWrViQmJrJhwwYeffRRMjIy+Pjjj+u8bwUe8UpXXXWV63WPHj1IS0ujVatWfPjhh9x5550erEyk8d18882u1927d6dHjx60bduWr776iiFDhniwMuuYMGECmzZtUt/AJna68z5+/HjX6+7du5OQkMCQIUPYsWMHbdu2rdO+dUnrHEVHR+Pj43NSL/3s7Gzi4+M9VNWFp1mzZnTo0IHt27d7upQLRvXvW799z2vTpg3R0dH6/bvJ/fffz3/+8x+WL19OixYtXMvj4+MpLy/n6NGjtdbXb949TnfeTyUtLQ2gXr95BZ5z5O/vT58+fVi2bJlrmdPpZNmyZaSnp3uwsgtLUVERO3bsICEhwdOlXDBat25NfHx8rd9+QUEBP/74o377TWz//v0cPnxYv/9zZBgG999/PwsXLuTLL7+kdevWtT7v06cPfn5+tX7zGRkZ7N27V7/5c3C2834q69atA6jXb16XtNxg0qRJjBs3jr59+9K/f39mzpxJcXExd9xxh6dLs6yHH36YESNG0KpVKw4ePMi0adPw8fHhlltu8XRpllJUVFTrv6B27drFunXraN68OS1btuTBBx/kz3/+M+3bt6d169ZMmTKFxMRERo4c6bmiLeBM57158+ZMnz6dUaNGER8fz44dO3jkkUdo164dw4YN82DV3m/ChAnMmzePf//734SFhbn65URERBAUFERERAR33nknkyZNonnz5oSHhzNx4kTS09O56KKLPFy99zrbed+xYwfz5s3j6quvJioqig0bNvDQQw9x6aWX0qNHj7of6Jzu8RKXV155xWjZsqXh7+9v9O/f3/jhhx88XZKljR492khISDD8/f2NpKQkY/To0cb27ds9XZblLF++3ABOeowbN84wDPPW9ClTphhxcXFGQECAMWTIECMjI8OzRVvAmc57SUmJceWVVxoxMTGGn5+f0apVK+Puu+82srKyPF221zvVOQeMt99+27XOsWPHjPvuu8+IjIw0goODjeuvv97IzMz0XNEWcLbzvnfvXuPSSy81mjdvbgQEBBjt2rUz/t//+39Gfn5+vY5jqzqYiIiIiGWpD4+IiIhYngKPiIiIWJ4Cj4iIiFieAo+IiIhYngKPiIiIWJ4Cj4iIiFieAo+IiIhYngKPiFyQbDYbn3zyiafLEJEmosAjIk3u9ttvx2aznfQYPny4p0sTEYvSXFoi4hHDhw/n7bffrrUsICDAQ9WIiNWphUdEPCIgIID4+Phaj8jISMC83PTqq69y1VVXERQURJs2bfjoo49qbb9x40Yuv/xygoKCiIqKYvz48RQVFdVa56233qJr164EBASQkJDA/fffX+vz3Nxcrr/+eoKDg2nfvj2ffvqp67MjR44wZswYYmJiCAoKon379icFNBHxHgo8InJemjJlCqNGjWL9+vWMGTOGm2++mc2bNwNQXFzMsGHDiIyM5KeffmLBggUsXbq0VqB59dVXmTBhAuPHj2fjxo18+umntGvXrtYxpk+fzk033cSGDRu4+uqrGTNmDHl5ea7j//rrr/z3v/9l8+bNvPrqq0RHRzfdCRAR93L7tKciImcxbtw4w8fHxwgJCan1ePrppw3DMGdPvueee2ptk5aWZtx7772GYRjG66+/bkRGRhpFRUWuzz/77DPDbre7Zg1PTEw0Hn/88dPWABhPPPGE631RUZEBGP/9738NwzCMESNGGHfccYd7vrCIeJz68IiIRwwePJhXX3211rLmzZu7Xqenp9f6LD09nXXr1gGwefNmUlNTCQkJcX0+cOBAnE4nGRkZ2Gw2Dh48yJAhQ85YQ48ePVyvQ0JCCA8PJycnB4B7772XUaNGsXbtWq688kpGjhzJgAEDGvRdRcTzFHhExCNCQkJOusTkLkFBQXVaz8/Pr9Z7m82G0+kE4KqrrmLPnj0sWrSIJUuWMGTIECZMmMALL7zg9npFpPGpD4+InJd++OGHk9537twZgM6dO7N+/XqKi4tdn3///ffY7XY6duxIWFgYKSkpLFu27JxqiImJYdy4cfzzn/9k5syZvP766+e0PxHxHLXwiIhHlJWVkZWVVWuZr6+vq2PwggUL6Nu3LxdffDHvvfceq1at4s033wRgzJgxTJs2jXHjxvHkk09y6NAhJk6cyG233UZcXBwATz75JPfccw+xsbFcddVVFBYW8v333zNx4sQ61Td16lT69OlD165dKSsr4z//+Y8rcImI91HgERGP+Pzzz0lISKi1rGPHjmzZsgUw76D64IMPuO+++0hISOD999+nS5cuAAQHB7N48WIeeOAB+vXrR3BwMKNGjeKll15y7WvcuHGUlpby17/+lYcffpjo6GhuvPHGOtfn7+/P5MmT2b17N0FBQVxyySV88MEHbvjmIuIJNsMwDE8XISJyPJvNxsKFCxk5cqSnSxERi1AfHhEREbE8BR4RERGxPPXhEZHzjq60i4i7qYVHRERELE+BR0RERCxPgUdEREQsT4FHRERELE+BR0RERCxPgUdEREQsT4FHRERELE+BR0RERCxPgUdEREQs7/8D3Jh7Ytzwj2cAAAAASUVORK5CYII=",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "0.9397321428571429"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lab4_util.plot_history(history, 'loss', ylim=(0, 1.0))\n",
    "lab4_util.accuracy(model_b0_AUG, val_loader_b0AUG)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**94%** c моими аугментациями - not bad EfficientNet-B0"
   ]
  }
 ],
 "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": "3.11.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}