train.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "fee99e0b",
   "metadata": {},
   "outputs": [],
   "source": [
    "import time\n",
    "import os\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn.model_selection import train_test_split\n",
    "from PIL import Image\n",
    "from tempfile import TemporaryDirectory\n",
    "\n",
    "import torch\n",
    "import torch.nn as nn\n",
    "import torch.nn.functional as F\n",
    "from torch.optim import lr_scheduler\n",
    "import torchvision\n",
    "from torchvision import datasets, models, transforms"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "0e4b1ecd",
   "metadata": {},
   "outputs": [],
   "source": [
    "LABEL_PATH = './data/classes.xlsx'\n",
    "IMAGE_PATH = './data/'\n",
    "IMAGE_SIZE = 224\n",
    "\n",
    "df = pd.read_excel(LABEL_PATH)\n",
    "\n",
    "df.columns = ['num', 'tirads']\n",
    "\n",
    "df['cross'] = IMAGE_PATH + \\\n",
    "    df['num'].astype(str).values + \\\n",
    "    \"/\" + df['num'].astype(str).values + \"_TIRADS\" + df['tirads'].astype(str).values + \"_cross.tif\"\n",
    "\n",
    "df['long'] = IMAGE_PATH + \\\n",
    "    df['num'].astype(str).values + \\\n",
    "    \"/\" + df['num'].astype(str).values + \"_TIRADS\" + df['tirads'].astype(str).values + \"_long.tif\"\n",
    "\n",
    "# Удаляем несуществующие данные\n",
    "for n in [178, 360, 363, 365, 366, 202, 346]:\n",
    "    df.drop(index=df[df['num'] == n].index, inplace=True)\n",
    "\n",
    "# Определим таргет для бинарной классификации\n",
    "df['target'] = (df['tirads'] == 5).astype(int)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "91823559",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>num</th>\n",
       "      <th>tirads</th>\n",
       "      <th>cross</th>\n",
       "      <th>long</th>\n",
       "      <th>target</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>179</td>\n",
       "      <td>5</td>\n",
       "      <td>./data/179/179_TIRADS5_cross.tif</td>\n",
       "      <td>./data/179/179_TIRADS5_long.tif</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>180</td>\n",
       "      <td>3</td>\n",
       "      <td>./data/180/180_TIRADS3_cross.tif</td>\n",
       "      <td>./data/180/180_TIRADS3_long.tif</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>181</td>\n",
       "      <td>5</td>\n",
       "      <td>./data/181/181_TIRADS5_cross.tif</td>\n",
       "      <td>./data/181/181_TIRADS5_long.tif</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>182</td>\n",
       "      <td>3</td>\n",
       "      <td>./data/182/182_TIRADS3_cross.tif</td>\n",
       "      <td>./data/182/182_TIRADS3_long.tif</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>183</td>\n",
       "      <td>5</td>\n",
       "      <td>./data/183/183_TIRADS5_cross.tif</td>\n",
       "      <td>./data/183/183_TIRADS5_long.tif</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   num  tirads                             cross  \\\n",
       "1  179       5  ./data/179/179_TIRADS5_cross.tif   \n",
       "2  180       3  ./data/180/180_TIRADS3_cross.tif   \n",
       "3  181       5  ./data/181/181_TIRADS5_cross.tif   \n",
       "4  182       3  ./data/182/182_TIRADS3_cross.tif   \n",
       "5  183       5  ./data/183/183_TIRADS5_cross.tif   \n",
       "\n",
       "                              long  target  \n",
       "1  ./data/179/179_TIRADS5_long.tif       1  \n",
       "2  ./data/180/180_TIRADS3_long.tif       0  \n",
       "3  ./data/181/181_TIRADS5_long.tif       1  \n",
       "4  ./data/182/182_TIRADS3_long.tif       0  \n",
       "5  ./data/183/183_TIRADS5_long.tif       1  "
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "31ac6745",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    21\n",
       "1    20\n",
       "Name: target, dtype: int64"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['target'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "a3f5ff64",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Добавим аугментацию к тренировочному датасету\n",
    "train_transform = transforms.Compose([\n",
    "    transforms.Resize((IMAGE_SIZE, IMAGE_SIZE)),\n",
    "    transforms.RandomAutocontrast(p=0.5),\n",
    "    transforms.RandomHorizontalFlip(p=0.5),\n",
    "    transforms.RandomAdjustSharpness(sharpness_factor=2, p=0.5),\n",
    "    transforms.RandomRotation(30),\n",
    "    transforms.ToTensor(),\n",
    "    transforms.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225])\n",
    "])\n",
    "\n",
    "val_transform = transforms.Compose([\n",
    "    transforms.Resize((IMAGE_SIZE, IMAGE_SIZE)),\n",
    "    transforms.ToTensor(),\n",
    "    transforms.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225])\n",
    "])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "be576f5f",
   "metadata": {},
   "outputs": [],
   "source": [
    "class TiradDataset:\n",
    "\n",
    "    def __init__(self, image_paths, targets, transform=None):\n",
    "        self.image_paths = image_paths\n",
    "        self.targets = targets\n",
    "        self.transform = transform\n",
    "\n",
    "    def __getitem__(self, idx):\n",
    "        image = Image.open(self.image_paths[idx])\n",
    "        image = image.convert(\"RGB\")\n",
    "        label = self.targets[idx]\n",
    "        if self.transform:\n",
    "            image = self.transform(image)\n",
    "        return image, torch.tensor(label, dtype=torch.float32)\n",
    "\n",
    "    def __len__(self):\n",
    "        return len(self.image_paths) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "f124585a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(36, 5)"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cross_images = df['cross'].values\n",
    "long_images = df['long'].values\n",
    "targets = df['target'].values\n",
    "\n",
    "train_X, val_X, train_y, val_y = train_test_split(\n",
    "    long_images, targets, train_size=0.9, random_state=777, shuffle=True\n",
    ")\n",
    "\n",
    "train_dataset = TiradDataset(\n",
    "    image_paths=train_X, \n",
    "    targets=train_y, \n",
    "    transform=train_transform\n",
    ")\n",
    "\n",
    "val_dataset = TiradDataset(\n",
    "    image_paths=val_X, \n",
    "    targets=val_y, \n",
    "    transform=val_transform\n",
    ")\n",
    "\n",
    "train_loader = torch.utils.data.DataLoader(\n",
    "    train_dataset,\n",
    "    batch_size=4,\n",
    "    shuffle=True,\n",
    "    num_workers=4,\n",
    ")\n",
    "\n",
    "val_loader = torch.utils.data.DataLoader(\n",
    "    val_dataset,\n",
    "    batch_size=4,\n",
    "    num_workers=4,\n",
    ")\n",
    "\n",
    "len(train_dataset), len(val_dataset)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "7580bd4e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "def imshow(inp, title=None):\n",
    "    \"\"\"Display image for Tensor.\"\"\"\n",
    "    inp = inp.numpy().transpose((1, 2, 0))\n",
    "    mean = np.array([0.485, 0.456, 0.406])\n",
    "    std = np.array([0.229, 0.224, 0.225])\n",
    "    inp = std * inp + mean\n",
    "    inp = np.clip(inp, 0, 1)\n",
    "    plt.imshow(inp)\n",
    "    if title is not None:\n",
    "        plt.title(title)\n",
    "    plt.pause(0.001) # pause a bit so that plots are updated\n",
    "\n",
    "\n",
    "# Get a batch of training data\n",
    "inputs, classes = next(iter(train_loader))\n",
    "\n",
    "# Make a grid from batch\n",
    "out = torchvision.utils.make_grid(inputs)\n",
    "\n",
    "imshow(out, title=classes)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "61e8434e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "torch.Size([4, 3, 224, 224])"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "inputs.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "2c30a2f7",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Выходной слой классификации\n",
    "class Classifier(nn.Module):\n",
    "    def __init__(self):\n",
    "        super().__init__()\n",
    "        self.fc1 = nn.Linear(512,120)\n",
    "        self.fc2 = nn.Linear(120,2)\n",
    "    \n",
    "    def forward(self,x):\n",
    "        x = F.relu(self.fc1(x))\n",
    "        x = self.fc2(x)\n",
    "        return x"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "43d1afab",
   "metadata": {},
   "outputs": [],
   "source": [
    "device = torch.device(\"cuda:0\" if torch.cuda.is_available() else \"cpu\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "id": "b3169fb5",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ResNet(\n",
      "  (conv1): Conv2d(3, 64, kernel_size=(7, 7), stride=(2, 2), padding=(3, 3), bias=False)\n",
      "  (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "  (relu): ReLU(inplace=True)\n",
      "  (maxpool): MaxPool2d(kernel_size=3, stride=2, padding=1, dilation=1, ceil_mode=False)\n",
      "  (layer1): Sequential(\n",
      "    (0): BasicBlock(\n",
      "      (conv1): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "      (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "      (relu): ReLU(inplace=True)\n",
      "      (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "      (bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "    )\n",
      "    (1): BasicBlock(\n",
      "      (conv1): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "      (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "      (relu): ReLU(inplace=True)\n",
      "      (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "      (bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "    )\n",
      "  )\n",
      "  (layer2): Sequential(\n",
      "    (0): BasicBlock(\n",
      "      (conv1): Conv2d(64, 128, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)\n",
      "      (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "      (relu): ReLU(inplace=True)\n",
      "      (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "      (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "      (downsample): Sequential(\n",
      "        (0): Conv2d(64, 128, kernel_size=(1, 1), stride=(2, 2), bias=False)\n",
      "        (1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "      )\n",
      "    )\n",
      "    (1): BasicBlock(\n",
      "      (conv1): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "      (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "      (relu): ReLU(inplace=True)\n",
      "      (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "      (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "    )\n",
      "  )\n",
      "  (layer3): Sequential(\n",
      "    (0): BasicBlock(\n",
      "      (conv1): Conv2d(128, 256, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)\n",
      "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "      (relu): ReLU(inplace=True)\n",
      "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "      (downsample): Sequential(\n",
      "        (0): Conv2d(128, 256, kernel_size=(1, 1), stride=(2, 2), bias=False)\n",
      "        (1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "      )\n",
      "    )\n",
      "    (1): BasicBlock(\n",
      "      (conv1): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "      (relu): ReLU(inplace=True)\n",
      "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "    )\n",
      "  )\n",
      "  (layer4): Sequential(\n",
      "    (0): BasicBlock(\n",
      "      (conv1): Conv2d(256, 512, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)\n",
      "      (bn1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "      (relu): ReLU(inplace=True)\n",
      "      (conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "      (bn2): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "      (downsample): Sequential(\n",
      "        (0): Conv2d(256, 512, kernel_size=(1, 1), stride=(2, 2), bias=False)\n",
      "        (1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "      )\n",
      "    )\n",
      "    (1): BasicBlock(\n",
      "      (conv1): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "      (bn1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "      (relu): ReLU(inplace=True)\n",
      "      (conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "      (bn2): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "    )\n",
      "  )\n",
      "  (avgpool): AdaptiveAvgPool2d(output_size=(1, 1))\n",
      "  (fc): Classifier(\n",
      "    (fc1): Linear(in_features=512, out_features=120, bias=True)\n",
      "    (fc2): Linear(in_features=120, out_features=2, bias=True)\n",
      "  )\n",
      ")\n"
     ]
    }
   ],
   "source": [
    "model =  models.resnet18(pretrained=True).to(device)\n",
    "\n",
    "model.fc = Classifier().to(device)\n",
    "print(model)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 93,
   "id": "7e981d6f",
   "metadata": {},
   "outputs": [],
   "source": [
    "for param in model.parameters():\n",
    "        param.requires_grad = False\n",
    "for param in model.fc.parameters():\n",
    "        param.requires_grad = True"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "id": "805db406",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Лосс и оптимизатор\n",
    "\n",
    "criterion = nn.CrossEntropyLoss()\n",
    "optimizer = torch.optim.Adam(model.parameters(), lr=0.01)\n",
    "\n",
    "exp_lr_scheduler = lr_scheduler.StepLR(optimizer, step_size=5, gamma=0.5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "id": "f83fa99d",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Обучение\n",
    "\n",
    "def train(model, criterion, optimizer, train_loader, val_loader, exp_lr_scheduler=None):\n",
    "    train_loss = []\n",
    "    val_loss = []\n",
    "    \n",
    "    train_acc = []\n",
    "    val_acc = []\n",
    "\n",
    "    epochs = 24\n",
    "\n",
    "    for epoch in range(epochs):\n",
    "        print(\"epoch {}/{}\".format(epoch+1,epochs))\n",
    "        running_loss = 0.0\n",
    "        running_score = 0.0\n",
    "\n",
    "        for image,label in train_loader:\n",
    "            image = image.to(device)\n",
    "            label = label.type(torch.LongTensor).to(device)\n",
    "\n",
    "            optimizer.zero_grad()\n",
    "            y_pred = model.forward(image)\n",
    "            loss = criterion(y_pred,label)         \n",
    "            loss.backward() #calculate derivatives \n",
    "            optimizer.step() # update parameters\n",
    "            val, index_ = torch.max(y_pred,axis=1)\n",
    "            running_score += torch.sum(index_ == label.data).item()\n",
    "            running_loss += loss.item()\n",
    "\n",
    "        epoch_score = running_score/len(train_loader.dataset)\n",
    "        epoch_loss = running_loss/len(train_loader.dataset)\n",
    "        train_loss.append(epoch_loss)\n",
    "        train_acc.append(epoch_score)\n",
    "        print(\"Training loss: {}, accuracy: {}\".format(epoch_loss,epoch_score))\n",
    "\n",
    "        if  exp_lr_scheduler:\n",
    "            exp_lr_scheduler.step()\n",
    "\n",
    "        with torch.no_grad():\n",
    "            model.eval()\n",
    "            running_loss = 0.0\n",
    "            running_score = 0.0\n",
    "            for image,label in val_loader:\n",
    "                image = image.to(device)\n",
    "                label = label.type(torch.LongTensor).to(device)\n",
    "                optimizer.zero_grad()\n",
    "                y_pred = model.forward(image)\n",
    "                loss = criterion(y_pred,label)\n",
    "                running_loss += loss.item()\n",
    "\n",
    "                val, index_ = torch.max(y_pred,axis=1)\n",
    "                running_score += torch.sum(index_ == label.data).item()\n",
    "\n",
    "            epoch_score = running_score/len(val_loader.dataset)\n",
    "            epoch_loss = running_loss/len(val_loader.dataset)\n",
    "            val_loss.append(epoch_loss)\n",
    "            val_acc.append(epoch_score)\n",
    "            print(\"Validation loss: {}, accuracy: {}\".format(epoch_loss,epoch_score))\n",
    "            \n",
    "    return model, train_loss, val_loss, train_acc, val_acc"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 96,
   "id": "791de3d7",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "epoch 1/24\n",
      "Training loss: 0.7874786108732224, accuracy: 0.3611111111111111\n",
      "Validation loss: 0.24842460155487062, accuracy: 0.6\n",
      "epoch 2/24\n",
      "Training loss: 0.3853792647520701, accuracy: 0.4444444444444444\n",
      "Validation loss: 0.4815326690673828, accuracy: 0.4\n",
      "epoch 3/24\n",
      "Training loss: 0.2207366575797399, accuracy: 0.5833333333333334\n",
      "Validation loss: 0.23601973056793213, accuracy: 0.6\n",
      "epoch 4/24\n",
      "Training loss: 0.21068337559700012, accuracy: 0.4444444444444444\n",
      "Validation loss: 0.3683188557624817, accuracy: 0.4\n",
      "epoch 5/24\n",
      "Training loss: 0.18631751504209307, accuracy: 0.4722222222222222\n",
      "Validation loss: 0.23055303692817689, accuracy: 0.6\n",
      "epoch 6/24\n",
      "Training loss: 0.1756358096996943, accuracy: 0.5\n",
      "Validation loss: 0.2523255467414856, accuracy: 0.6\n",
      "epoch 7/24\n",
      "Training loss: 0.17866393592622545, accuracy: 0.3888888888888889\n",
      "Validation loss: 0.2701582074165344, accuracy: 1.0\n",
      "epoch 8/24\n",
      "Training loss: 0.17235142489274344, accuracy: 0.5555555555555556\n",
      "Validation loss: 0.2697015643119812, accuracy: 1.0\n",
      "epoch 9/24\n",
      "Training loss: 0.1688714689678616, accuracy: 0.6666666666666666\n",
      "Validation loss: 0.2704341173171997, accuracy: 0.8\n",
      "epoch 10/24\n",
      "Training loss: 0.16988179749912685, accuracy: 0.5833333333333334\n",
      "Validation loss: 0.27735152244567873, accuracy: 0.4\n",
      "epoch 11/24\n",
      "Training loss: 0.1700549473365148, accuracy: 0.5833333333333334\n",
      "Validation loss: 0.2802147626876831, accuracy: 0.4\n",
      "epoch 12/24\n",
      "Training loss: 0.16989374491903517, accuracy: 0.6111111111111112\n",
      "Validation loss: 0.2721752405166626, accuracy: 0.8\n",
      "epoch 13/24\n",
      "Training loss: 0.17105560335848066, accuracy: 0.5\n",
      "Validation loss: 0.2674426555633545, accuracy: 1.0\n",
      "epoch 14/24\n",
      "Training loss: 0.1667601035700904, accuracy: 0.7777777777777778\n",
      "Validation loss: 0.2726204514503479, accuracy: 0.6\n",
      "epoch 15/24\n",
      "Training loss: 0.1662994921207428, accuracy: 0.6666666666666666\n",
      "Validation loss: 0.2710054874420166, accuracy: 0.8\n",
      "epoch 16/24\n",
      "Training loss: 0.16924478279219735, accuracy: 0.5555555555555556\n",
      "Validation loss: 0.2720435619354248, accuracy: 0.6\n",
      "epoch 17/24\n",
      "Training loss: 0.16925389733579424, accuracy: 0.6666666666666666\n",
      "Validation loss: 0.26937568187713623, accuracy: 0.8\n",
      "epoch 18/24\n",
      "Training loss: 0.16754216121302712, accuracy: 0.5833333333333334\n",
      "Validation loss: 0.27279233932495117, accuracy: 0.6\n",
      "epoch 19/24\n",
      "Training loss: 0.17391046384970346, accuracy: 0.5277777777777778\n",
      "Validation loss: 0.29060927629470823, accuracy: 0.4\n",
      "epoch 20/24\n",
      "Training loss: 0.16387384798791674, accuracy: 0.6666666666666666\n",
      "Validation loss: 0.2684781074523926, accuracy: 0.8\n",
      "epoch 21/24\n",
      "Training loss: 0.1650868836376402, accuracy: 0.5555555555555556\n",
      "Validation loss: 0.2662047743797302, accuracy: 0.8\n",
      "epoch 22/24\n",
      "Training loss: 0.16734793451097277, accuracy: 0.6388888888888888\n",
      "Validation loss: 0.26701167821884153, accuracy: 0.8\n",
      "epoch 23/24\n",
      "Training loss: 0.16783643099996778, accuracy: 0.6111111111111112\n",
      "Validation loss: 0.2655384302139282, accuracy: 0.8\n",
      "epoch 24/24\n",
      "Training loss: 0.1662920895549986, accuracy: 0.6944444444444444\n",
      "Validation loss: 0.2673328399658203, accuracy: 0.8\n"
     ]
    }
   ],
   "source": [
    "model, train_loss, val_loss, train_acc, val_acc = train(\n",
    "    model, criterion, optimizer, train_loader, val_loader, exp_lr_scheduler\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 97,
   "id": "2263d82f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAiMAAAGdCAYAAADAAnMpAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8pXeV/AAAACXBIWXMAAA9hAAAPYQGoP6dpAABRyklEQVR4nO3deXxU1d0/8M+dPdtMEiYbEAhhR9kDERTEGgXqw6NWKyqVpUp/paC1qVapBdxqWqWUVqk85ZG6tlKtPNqKuEQBUQQkoqgYtkDCkj2Z7DPJzP39cWbLPjOZmZuQz/v1uq+5uXNn5iRDmE/O+Z5zJVmWZRAREREpRKV0A4iIiKh/YxghIiIiRTGMEBERkaIYRoiIiEhRDCNERESkKIYRIiIiUhTDCBERESmKYYSIiIgUpVG6Ab5wOBw4f/48YmJiIEmS0s0hIiIiH8iyjNraWgwcOBAqVef9H30ijJw/fx6pqalKN4OIiIgCUFRUhMGDB3d6f58IIzExMQDEN2M0GhVuDREREfmipqYGqamp7s/xzvSJMOIamjEajQwjREREfUx3JRYsYCUiIiJFMYwQERGRohhGiIiISFEB1Yxs2rQJTz31FIqLizFx4kQ8/fTTmD59eqfnb9y4Ec8++ywKCwthNptx8803IycnBwaDIeCGExHRxUGWZbS0tMButyvdFPKTWq2GRqPp8bIbfoeRbdu2ITs7G5s3b0ZmZiY2btyIuXPnIj8/H4mJie3O//vf/44HH3wQW7duxcyZM3Hs2DEsXboUkiRhw4YNPWo8ERH1bTabDRcuXEBDQ4PSTaEARUZGIiUlBTqdLuDnkGRZlv15QGZmJqZNm4ZnnnkGgFiQLDU1FXfffTcefPDBduevWrUKR48eRW5urvvYL3/5S+zfvx979+716TVrampgMplgsVg4m4aI6CLhcDhw/PhxqNVqJCQkQKfTcWHLPkSWZdhsNpSVlcFut2PkyJHtFjbz9fPbr54Rm82GQ4cOYfXq1e5jKpUKWVlZ2LdvX4ePmTlzJl5++WUcOHAA06dPx6lTp7Bjxw7ccccdnb6O1WqF1Wpt9c0QEdHFxWazuf+gjYyMVLo5FICIiAhotVqcOXMGNpst4PILv8JIeXk57HY7kpKSWh1PSkrCd9991+Fjbr/9dpSXl+OKK65wjwv+9Kc/xa9//etOXycnJwePPPKIP00jIqI+qqtlwqn3C8b7F/J/Abt27cITTzyBv/zlL8jLy8Mbb7yBt99+G4899linj1m9ejUsFot7KyoqCnUziYiISCF+9YyYzWao1WqUlJS0Ol5SUoLk5OQOH7NmzRrccccduOuuuwAA48ePR319PX7yk5/goYce6jBR6fV66PV6f5pGREREfZRfPSM6nQ5Tp05tVYzqcDiQm5uLGTNmdPiYhoaGdoFDrVYDEMUvRERE/V1aWho2btyo+HMoxe+pvdnZ2ViyZAkyMjIwffp0bNy4EfX19Vi2bBkAYPHixRg0aBBycnIAAAsWLMCGDRswefJkZGZm4sSJE1izZg0WLFjgDiVERER9yZw5czBp0qSgffgfPHgQUVFRQXmuvsjvMLJw4UKUlZVh7dq1KC4uxqRJk7Bz5053UWthYWGrnpDf/OY3kCQJv/nNb3Du3DkkJCRgwYIF+O1vfxu87yJAf/ukACdK67Ds8mEYkRitdHOIiOgiIssy7HY7NJruP2oTEhLC0KLeK6AC1lWrVuHMmTOwWq3Yv38/MjMz3fft2rULzz//vPtrjUaDdevW4cSJE2hsbERhYSE2bdqE2NjYnra9x948fB6v7C/EidI6pZtCREQQH+ANthZFNl9LB5YuXYrdu3fjT3/6EyRJgiRJOH36NHbt2gVJkvDOO+9g6tSp0Ov12Lt3L06ePInrr78eSUlJiI6OxrRp0/DBBx+0es62QyySJOF///d/ceONNyIyMhIjR47EW2+95dfPsrCwENdffz2io6NhNBpxyy23tKr5/PLLL3HVVVchJiYGRqMRU6dOxeeffw4AOHPmDBYsWIC4uDhERUXhkksuwY4dO/x6fX8EtBz8xSIhRhTJltdZuzmTiIjCobHZjnFr31Xktb99dC4idd1/LP7pT3/CsWPHcOmll+LRRx8FIHo2Tp8+DQB48MEHsX79eqSnpyMuLg5FRUX4/ve/j9/+9rfQ6/V48cUXsWDBAuTn52PIkCGdvs4jjzyCJ598Ek899RSefvppLFq0CGfOnEF8fHy3bXQ4HO4gsnv3brS0tGDlypVYuHAhdu3aBQBYtGgRJk+ejGeffRZqtRqHDx+GVqsFAKxcuRI2mw179uxBVFQUvv32W0RHh24EoV+HEXM0wwgREfnHZDJBp9MhMjKyw5mkjz76KK655hr31/Hx8Zg4caL768ceewzbt2/HW2+9hVWrVnX6OkuXLsVtt90GAHjiiSfw5z//GQcOHMC8efO6bWNubi6OHDmCgoICpKamAgBefPFFXHLJJTh48CCmTZuGwsJC3H///RgzZgwAYOTIke7HFxYW4qabbsL48eMBAOnp6d2+Zk/06zCSEC3W0WcYISLqHSK0anz76FzFXjsYMjIyWn1dV1eHhx9+GG+//TYuXLiAlpYWd9lCVyZMmODej4qKgtFoRGlpqU9tOHr0KFJTU91BBADGjRuH2NhYHD16FNOmTUN2djbuuusuvPTSS8jKysIPf/hDDB8+HABwzz33YMWKFXjvvfeQlZWFm266qVV7gq1fL3tndg3T1NoUbgkREQGiViJSp1FkC9Z1cdrOirnvvvuwfft2PPHEE/j4449x+PBhjB8/HjZb1589riET75+Nw+EIShsB4OGHH8Y333yD6667Dh9++CHGjRuH7du3AwDuuusunDp1CnfccQeOHDmCjIwMPP3000F77bb6dxjhMA0REQVAp9PBbrf7dO4nn3yCpUuX4sYbb8T48eORnJzsri8JlbFjx6KoqKjVCubffvstqqurMW7cOPexUaNG4Re/+AXee+89/OAHP8Df/vY3932pqan46U9/ijfeeAO//OUvsWXLlpC1l2EEQBnDCBER+SEtLQ379+/H6dOnUV5e3mWPxciRI/HGG2/g8OHD+PLLL3H77bcHtYejI1lZWRg/fjwWLVqEvLw8HDhwAIsXL8aVV16JjIwMNDY2YtWqVdi1axfOnDmDTz75BAcPHsTYsWMBAPfeey/effddFBQUIC8vDx999JH7vlDo12HEPZumlmGEiIh8d99990GtVmPcuHFISEjosv5jw4YNiIuLw8yZM7FgwQLMnTsXU6ZMCWn7JEnCm2++ibi4OMyePRtZWVlIT0/Htm3bAIiV0CsqKrB48WKMGjUKt9xyC+bPn+++SK3dbsfKlSsxduxYzJs3D6NGjcJf/vKX0LVX7gNrstfU1MBkMsFiscBoNAbteWubmjH+4fcAAEcfnYcIHVeEJSIKl6amJhQUFGDYsGEBX3qelNfV++jr53e/7hmJ1mug14gfAetGiIiIlNGvw4gkSawbISIiUli/DiOA9/RehhEiIiIl9Psw4lr4jD0jREREymAY4cJnREREiur3YYQLnxERESmLYYRhhIiISFEMIwwjREREimIYcV+5lzUjREQUPmlpadi4cWOn9y9duhQ33HBD2NqjJIYRZwFrGaf2EhERKaLfhxHXbJo6awuamn27AiMREREFT78PIzF6DXTOJeHZO0JERN3561//ioEDB7a78u7111+PH//4xwCAkydP4vrrr0dSUhKio6Mxbdo0fPDBBz16XavVinvuuQeJiYkwGAy44oorcPDgQff9VVVVWLRoERISEhAREYGRI0fib3/7GwDAZrNh1apVSElJgcFgwNChQ5GTk9Oj9gRTvw8jkiQhgUWsRES9gywDtnplNh+vG/vDH/4QFRUV+Oijj9zHKisrsXPnTixatAgAUFdXh+9///vIzc3FF198gXnz5mHBggVdXt23O7/61a/wr3/9Cy+88ALy8vIwYsQIzJ07F5WVlQCANWvW4Ntvv8U777yDo0eP4tlnn4XZbAYA/PnPf8Zbb72Ff/7zn8jPz8crr7yCtLS0gNsSbBqlG9AbmKN1OFfdyCJWIiKlNTcATwxU5rV/fR7QRXV7WlxcHObPn4+///3vuPrqqwEAr7/+OsxmM6666ioAwMSJEzFx4kT3Yx577DFs374db731FlatWuV30+rr6/Hss8/i+eefx/z58wEAW7Zswfvvv4/nnnsO999/PwoLCzF58mRkZGQAQKuwUVhYiJEjR+KKK66AJEkYOnSo320IpX7fMwJwei8REfln0aJF+Ne//gWrVXxuvPLKK7j11luhUomP1bq6Otx3330YO3YsYmNjER0djaNHjwbcM3Ly5Ek0Nzfj8ssvdx/TarWYPn06jh49CgBYsWIFXn31VUyaNAm/+tWv8Omnn7rPXbp0KQ4fPozRo0fjnnvuwXvvvRfotx4S7BmBJ4ywZoSISGHaSNFDodRr+2jBggWQZRlvv/02pk2bho8//hh//OMf3fffd999eP/997F+/XqMGDECERERuPnmm2Gzha4Hfv78+Thz5gx27NiB999/H1dffTVWrlyJ9evXY8qUKSgoKMA777yDDz74ALfccguysrLw+uuvh6w9/mAYgdf1adgzQkSkLEnyaahEaQaDAT/4wQ/wyiuv4MSJExg9ejSmTJnivv+TTz7B0qVLceONNwIQPSWnT58O+PWGDx8OnU6HTz75xD3E0tzcjIMHD+Lee+91n5eQkIAlS5ZgyZIlmDVrFu6//36sX78eAGA0GrFw4UIsXLgQN998M+bNm4fKykrEx8cH3K5gYRiB98JnDCNEROSbRYsW4b/+67/wzTff4Ec/+lGr+0aOHIk33ngDCxYsgCRJWLNmTbvZN/6IiorCihUrcP/99yM+Ph5DhgzBk08+iYaGBtx5550AgLVr12Lq1Km45JJLYLVa8Z///Adjx44FAGzYsAEpKSmYPHkyVCoVXnvtNSQnJyM2NjbgNgUTwwg8C5/xyr1EROSr733ve4iPj0d+fj5uv/32Vvdt2LABP/7xjzFz5kyYzWY88MADqKmp6dHr/e53v4PD4cAdd9yB2tpaZGRk4N1330VcXBwAQKfTYfXq1Th9+jQiIiIwa9YsvPrqqwCAmJgYPPnkkzh+/DjUajWmTZuGHTt2uGtclCbJso9zmRRUU1MDk8kEi8UCo9EY9Of/7FQFbv3rZ0g3R+HD++YE/fmJiKi9pqYmFBQUYNiwYTAYDEo3hwLU1fvo6+d374hECnMXsHKYhoiIKOwYRgD3ome1TVwSnoiIKNwYRgAYIzTQqcWPgkWsRERE4cUwArEkvGdGDYtYiYiIwolhxMkzo4Y9I0REROHEMOLEJeGJiJTRByZ1UheC8f4xjDhx4TMiovDSarUAgIaGBoVbQj3hev9c72cguOiZE69PQ0QUXmq1GrGxsSgtLQUAREZGQpIkhVtFvpJlGQ0NDSgtLUVsbCzUanXAz8Uw4uS5Pg0LWImIwiU5ORkA3IGE+p7Y2Fj3+xgohhEnLnxGRBR+kiQhJSUFiYmJaG5uVro55CetVtujHhEXhhEnFrASESlHrVYH5UON+iYWsDolxDgLWFkzQkREFFYMI06unpGaphZYW7gkPBERUbgwjDiZIrTQqkUVN4tYiYiIwiegMLJp0yakpaXBYDAgMzMTBw4c6PTcOXPmQJKkdtt1110XcKNDQSwJz1VYiYiIws3vMLJt2zZkZ2dj3bp1yMvLw8SJEzF37txOp2W98cYbuHDhgnv7+uuvoVar8cMf/rDHjQ82FrESERGFn99hZMOGDVi+fDmWLVuGcePGYfPmzYiMjMTWrVs7PD8+Ph7Jycnu7f3330dkZGQvDSNchZWIiCjc/AojNpsNhw4dQlZWlucJVCpkZWVh3759Pj3Hc889h1tvvRVRUVH+tTQMPD0jrBkhIiIKF7/WGSkvL4fdbkdSUlKr40lJSfjuu++6ffyBAwfw9ddf47nnnuvyPKvVCqvV0ztRU1PjTzMD5rpyL5eEJyIiCp+wzqZ57rnnMH78eEyfPr3L83JycmAymdxbampqWNrHmhEiIqLw8yuMmM1mqNVqlJSUtDpeUlLS7br09fX1ePXVV3HnnXd2+zqrV6+GxWJxb0VFRf40M2AJ7BkhIiIKO7/CiE6nw9SpU5Gbm+s+5nA4kJubixkzZnT52Ndeew1WqxU/+tGPun0dvV4Po9HYagsHFrASERGFn9/XpsnOzsaSJUuQkZGB6dOnY+PGjaivr8eyZcsAAIsXL8agQYOQk5PT6nHPPfccbrjhBgwYMCA4LQ+BBBawEhERhZ3fYWThwoUoKyvD2rVrUVxcjEmTJmHnzp3uotbCwkKoVK07XPLz87F371689957wWl1iLhqRiyNzbC1OKDTcIFaIiKiUJNkWZaVbkR3ampqYDKZYLFYQjpk43DIGPWbd9DikLFv9feQYooI2WsRERFd7Hz9/Oaf/l5UKgkDXHUjtRyqISIiCgeGkTbcM2rqmhRuCRERUf/AMNKG52J57BkhIiIKB4aRNlxhpIzTe4mIiMKCYaQNrsJKREQUXgwjbXgWPuMwDRERUTgwjLThKmAt55LwREREYcEw0kYCa0aIiIjCimGkDXMMa0aIiIjCiWGkDVcBa3VDM5rtDoVbQ0REdPFjGGkjNkILtUoCAFSwiJWIiCjkGEbaUKkkDIhyzajhUA0REVGoMYx0gAufERERhQ/DSAfc16fh9F4iIqKQYxjpAFdhJSIiCh+GkQ6YY5w1I7xYHhERUcgxjHQggT0jREREYcMw0gEO0xAREYUPw0gHGEaIiIjCh2GkA5xNQ0REFD4MIx0wR4sC1iouCU9ERBRyDCMdiIvUuZeEr6znjBoiIqJQYhjpgEolId65JDyHaoiIiEKLYaQTLGIlIiIKD4aRTrjqRsp55V4iIqKQYhjpBGfUEBERhQfDSCe4CisREVF4MIx0gjUjRERE4cEw0gn3xfIYRoiIiEKKYaQT7p4RXrmXiIgopBhGOsFhGiIiovBgGOmEazZNZYMNLVwSnoiIKGQYRjoRF6mDSgJkmUvCExERhRLDSCfUKgnxUc61RjhUQ0REFDIMI13gKqxEREShxzDSBVfdSDlXYSUiIgoZhpEucEYNERFR6DGMdIHXpyEiIgo9hpEueGpGGEaIiIhChWGkC55hGhawEhERhQrDSBdYM0JERBR6DCNdYBghIiIKPYaRLriu3FtZb4PdISvcGiIiootTQGFk06ZNSEtLg8FgQGZmJg4cONDl+dXV1Vi5ciVSUlKg1+sxatQo7NixI6AGh9OAKD1UEuCQgYp69o4QERGFgt9hZNu2bcjOzsa6deuQl5eHiRMnYu7cuSgtLe3wfJvNhmuuuQanT5/G66+/jvz8fGzZsgWDBg3qceNDTSwJ75xRU8siViIiolDQ+PuADRs2YPny5Vi2bBkAYPPmzXj77bexdetWPPjgg+3O37p1KyorK/Hpp59Cq9UCANLS0nrW6jAyR+tRXmdj3QgREVGI+NUzYrPZcOjQIWRlZXmeQKVCVlYW9u3b1+Fj3nrrLcyYMQMrV65EUlISLr30UjzxxBOw2+2dvo7VakVNTU2rTSksYiUiIgotv8JIeXk57HY7kpKSWh1PSkpCcXFxh485deoUXn/9ddjtduzYsQNr1qzBH/7wBzz++OOdvk5OTg5MJpN7S01N9aeZQcWFz4iIiEIr5LNpHA4HEhMT8de//hVTp07FwoUL8dBDD2Hz5s2dPmb16tWwWCzuraioKNTN7BQXPiMiIgotv2pGzGYz1Go1SkpKWh0vKSlBcnJyh49JSUmBVquFWq12Hxs7diyKi4ths9mg0+naPUav10Ov1/vTtJDh9WmIiIhCy6+eEZ1Oh6lTpyI3N9d9zOFwIDc3FzNmzOjwMZdffjlOnDgBh8PhPnbs2DGkpKR0GER6G9aMEBERhZbfwzTZ2dnYsmULXnjhBRw9ehQrVqxAfX29e3bN4sWLsXr1avf5K1asQGVlJX7+85/j2LFjePvtt/HEE09g5cqVwfsuQsjMnhEiIqKQ8ntq78KFC1FWVoa1a9eiuLgYkyZNws6dO91FrYWFhVCpPBknNTUV7777Ln7xi19gwoQJGDRoEH7+85/jgQceCN53EUKeAlbWjBAREYWCJMtyr1/nvKamBiaTCRaLBUajMayvXVrThOlP5EIlAcd/+32oVVJYX5+IiKiv8vXzm9em6UZ8lA6Sc0n4qgb2jhAREQUbw0g3NGoV4iO51ggREVGoMIz4wDWjhkWsREREwccw4gNzDHtGiIiIQoVhxAfutUZ45V4iIqKgYxjxARc+IyIiCh2GER+4a0YYRoiIiIKOYcQHruvTcOEzIiKi4GMY8YFrFVbOpiEiIgo+hhEfsGaEiIgodBhGfOAapqmst8Hh6PWr5xMREfUpDCM+iI8SwzR2h8wl4YmIiIKMYcQHWrUKcZFaACxiJSIiCjaGER95ZtSwboSIiCiYGEZ8xOvTEBERhQbDiI84o4aIiCg0GEZ8xFVYiYiIQoNhxEfuK/fyYnlERERBxTDiIw7TEBERhQbDiI84m4aIiCg0GEZ8lMDZNERERCHBMOIj1zBNBZeEJyIiCiqGER8NiPYsCV/d2Kxwa4iIiC4eDCM+0qpViHUvCc+hGiIiomBhGPGDe0YN60aIiIiChmHEDwlc+IyIiCjoGEb8YI7hjBoiIqJgYxjxg9lZxFpex1VYiYiIgoVhxA9chZWIiCj4GEb8kMAwQkREFHQMI35wXyyPYYSIiChoGEb8kBBtAMAr9xIREQUTw4gfvHtGuCQ8ERFRcDCM+GFAlKgZaXHIsHBJeCIioqBgGPGDTqOCKYJLwhMREQUTw4ifXGuNcBVWIiKi4GAY8ZNnrREWsRIREQUDw4ifEmJ4sTwiIqJgYhjxk5kXyyMiIgoqhhE/sWeEiIgouBhG/OS5WB7DCBERUTAwjPiJBaxERETBxTDiJ165l4iIKLgCCiObNm1CWloaDAYDMjMzceDAgU7Pff755yFJUqvNYDAE3GCluWtG6qyQZS4JT0RE1FN+h5Ft27YhOzsb69atQ15eHiZOnIi5c+eitLS008cYjUZcuHDBvZ05c6ZHjVbSAGfNSLOdS8ITEREFg99hZMOGDVi+fDmWLVuGcePGYfPmzYiMjMTWrVs7fYwkSUhOTnZvSUlJPWq0kvQaNYwGDQAO1RAREQWDX2HEZrPh0KFDyMrK8jyBSoWsrCzs27ev08fV1dVh6NChSE1NxfXXX49vvvmmy9exWq2oqalptfUmZudQTVkti1iJiIh6yq8wUl5eDrvd3q5nIykpCcXFxR0+ZvTo0di6dSvefPNNvPzyy3A4HJg5cybOnj3b6evk5OTAZDK5t9TUVH+aGXIsYiUiIgqekM+mmTFjBhYvXoxJkybhyiuvxBtvvIGEhAT8z//8T6ePWb16NSwWi3srKioKdTP9ksAwQkREFDQaf042m81Qq9UoKSlpdbykpATJyck+PYdWq8XkyZNx4sSJTs/R6/XQ6/X+NC2svGfUEBERUc/41TOi0+kwdepU5Obmuo85HA7k5uZixowZPj2H3W7HkSNHkJKS4l9LexHXKqxlXBKeiIiox/zqGQGA7OxsLFmyBBkZGZg+fTo2btyI+vp6LFu2DACwePFiDBo0CDk5OQCARx99FJdddhlGjBiB6upqPPXUUzhz5gzuuuuu4H4nYcRVWImIiILH7zCycOFClJWVYe3atSguLsakSZOwc+dOd1FrYWEhVCpPh0tVVRWWL1+O4uJixMXFYerUqfj0008xbty44H0XYcYCViIiouCR5D6wjGhNTQ1MJhMsFguMRqPSzcHhomrcsOkTDDQZ8Onqq5VuDhERUa/k6+c3r00TAM+Ve21cEp6IiKiHGEYC4BqmsdkdqGlqUbg1REREfRvDSAAMWjVinEvCu2fUOBzAuw8BX7ysYMuIiIj6HoaRALVb+OzCF8C+Z4B3HgQ4dENEROQzhpEAtZtRU3ZM3NpqgbqSTh5FREREbTGMBMgc4yxidQ3TlB/z3FnR+eqyRERE1BrDSIDaLXzGMEJERBQQhpEAtasZKT/uuZNhhIiIyGcMIwEyOy+WV1ZrBezNQOUpz50VJxVqFRERUd/DMBKgVgWsVWcAR7PnTvaMEBER+YxhJEDeq7C660UizeK2sgCwczE0IiIiXzCMBMjVM1JWZ4XsCiPDZgMag+glsRQq2DoiIqK+g2EkQAnOmhFbiwPNJfnOg2OA+OFin3UjREREPmEYCZBBq0aMXiwJ7yhzhhHzSGAAwwgREZE/GEZ6QMyokaGpdBasmkcCA0aIfRaxEhER+YRhpAfM0ToMQA00NgsASQzRMIwQERH5RaN0A/oyc7QeknRefBGbCugivcIIh2mIiIh8wTDSA+ZoPWJVzjBiHiVuXWHEUgQ0NwLaCGUaR0RE1EdwmKYHzNF6pEsXnF84w0hkPGCIBSCL9UaIiIioSwwjPZAQo8dw1zCNeaS4lSSvGTWsGyEiIuoOw0gPmKN1XmFklOcOFrESERH5jGGkBxIiZKRKZeKLDsMIi1iJiIi6wzDSAykt56GSZFjkKMiu69IAHKYhIiLyA8NID8Q1ngYAnJRTUGeze+7gMA0REZHPGEZ6QF8thmFOOgaKq/e6uK5P01AONFYp0DIiIqK+g2GkJ5xX6z0pD0R5ndVzXB8NxKSI/YpTCjSMiIio72AY6QnvMFJrbX0fh2qIiIh8wjASKIcDKD8OQISRsrq2YYRFrERERL5gGAlU7XmguQF2qFEoJ7JnhIiIKEAMI4FyDtFYIlLRAg3KvAtYAU8YqeRaI0RERF1hGAmUc4imLiZdfNlumMZr4TNZDmfLiIiI+hSGkUA5e0Za4kXoaBdGYocCkhqw1QF1JeFuHRERUZ/BMBIoZxiRnMvAtwsjGh0QN1Tss26EiIioUwwjgXIO00SkjAUAlNVaIbcdjmERKxERUbcYRgLRVAPUXgAAGFNFGGlqdqDee0l4wLMSK8MIERFRpxhGAlEhekUQnYRI4wBE6tQA0MH0XlcY4YwaIiKizjCMBMI5RANnvYg5Wi8Odzqjhj0jREREnWEYCYSzeBXmkQCAhJhuwkhlAWBvCVfriIiI+hSGkUC4w4irZ0QHAO0XPjMOAjQGwNEMWArD2UIiIqI+g2EkEK5hmgGiZ8Q1TFPWtmZEpfIqYmXdCBERUUcYRvxlb/EEC3PrMNJumAbgBfOIiIi6wTDir+ozYthFYwBMqQAAs6tmpG3PCMAiViIiom4wjPjLVS8yYKQYhgGQ4KwZ6bhnhGGEiIioKwGFkU2bNiEtLQ0GgwGZmZk4cOCAT4979dVXIUkSbrjhhkBetndwT+sd6T7kmU1ja3++9wXziIiIqB2/w8i2bduQnZ2NdevWIS8vDxMnTsTcuXNRWlra5eNOnz6N++67D7NmzQq4sb1Cm5k0QHc1I84wYikCmhtD3ToiIqI+x+8wsmHDBixfvhzLli3DuHHjsHnzZkRGRmLr1q2dPsZut2PRokV45JFHkJ6e3qMGK66DnhFXGGmw2VFvbbOeSGQ8YIgV+5UFYWggERFR3+JXGLHZbDh06BCysrI8T6BSISsrC/v27ev0cY8++igSExNx5513+vQ6VqsVNTU1rbZeo4OekSi9BhFa55LwbXtHJIl1I0RERF3wK4yUl5fDbrcjKSmp1fGkpCQUFxd3+Ji9e/fiueeew5YtW3x+nZycHJhMJveWmprqTzNDp74CaKwU+66A4WSOYRErERFRIEI6m6a2thZ33HEHtmzZArPZ7PPjVq9eDYvF4t6KiopC2Eo/uHpFTEMAXWSruzwLn7GIlYiIyB8af042m81Qq9UoKSlpdbykpATJycntzj958iROnz6NBQsWuI85HA7xwhoN8vPzMXz48HaP0+v10Ov1/jQtPNpck8ZbQpdFrM46GfaMEBERteNXz4hOp8PUqVORm5vrPuZwOJCbm4sZM2a0O3/MmDE4cuQIDh8+7N7++7//G1dddRUOHz7ce4ZffNVBvYiLubOL5QEcpiEiIuqCXz0jAJCdnY0lS5YgIyMD06dPx8aNG1FfX49ly5YBABYvXoxBgwYhJycHBoMBl156aavHx8bGAkC7431CBzNpXDq9Pg3guT5NQznQWAVExIWqhURERH2O32Fk4cKFKCsrw9q1a1FcXIxJkyZh586d7qLWwsJCqFQX6cKuXfSMdLkKqz4aiEkBai8AFaeAwVND2UoiIqI+xe8wAgCrVq3CqlWrOrxv165dXT72+eefD+QlldfcJK5LA3Q8TBPdxSqsgBiqqb0ghmoYRoiIiNwu0i6MEKg8BcgOQG8CohPb3d1lzQjAq/cSERF1gmHEV94zaSSp3d2D4yIAAOeqGtFos7d/PItYiYiIOsQw4it38Wr7IRoASDYakGTUo8Uh48g5S/sTGEaIiIg6xDDiqy7WGAEASZIwZYiYJZNXWNX+BO+Fz2Q5FC0kIiLqkxhGfNXFTBqXyUNiAQB5ZzoII7FDAUkNNNcDtR0vnU9ERNQfMYz4Qpa7HaYB4O4Z+aKoGnLb3g+NDogbKvY5VENEROTGMOKLmvOiR0OlAeKHdXrapYNM0KgklNVacbaqsf0JrqGaSl6jhoiIyIVhxBeuIZq4YYBa2+lpBq0alww0AhC9I+2wiJWIiKgdhhFf+DBE4zLZVcTaUd2Ie60R9owQERG5MIz4opuZNN5cRay9rmck70Vg95OcyUNERL1OQMvB9zs+zKRxcRWxfnvegqZmOwxatedO1wXzKgsAewugDtOPv7EK+PfPxQqyI64GBnE5eiIi6j3YM+KLLq7W29bguAiYo/Votsv4uu3iZ8ZBgMYAOJoBS2EIGtqJgj0iiADAqd3he10iIiIfMIx0x1oL1J4X+65hli6Ixc9iAQBfFFa3vlOl8vSOhLNu5OSHnv1Tu8L3ukRERD5gGOmOq74jKgGIjPfpIZO7XIk1zBfMk2XghFcYKfwMaO5g2jEREZFCGEa648dMGpdOe0aA8BexVpwUQ0JqHRCVCNitQNH+8Lw2ERGRDxhGuuPHTBqX8YNNUKskFNc04Xx1m16IcIeRk7nidshlongV4FANERH1Kgwj3fFjJo1LpE6DsSkxADoYqvG+YF44uOpFhn8PGHal2GcRKxER9SIMI90JYJgG8LpOTduhGlcYsRSFvnajxQYUfCz2h18NpDvDyPkvxHRfIiKiXoBhpCsOu2c4xY9hGsDrCr5te0Yi4wGDuA+Vp3rWvu4U7RfX1IlKAJIuBYwDAfNoALInpBARESmMYaQr1WcAu02sDWJK9euhrp6Rb87VwNpi99whSeGrG/EeolE532pX70gBh2qIiKh3YBjpimuIZsAIQKXu+tw2hsRHIj5KB5vdgW/O17S+M2xhxFm8Ovx7nmPpc8Qti1iJiKiXYBjpSgAzaVy6XPwsHEWs9eXAhS/FfvpVnuNDLwcklQhClrOhe30iIiIfMYx0JYCZNN46XfwsHFfvPfmRuE0aD8QkeY5HxAIDp4h9zqohIqJegGGkKwHOpHFxX8H3TGfTe0M4TOOuF7mq/X0cqiEiol6EYaQrPRimAYCJg2OhkoDzliYUW5o8d8Sni9uG8tBMsZVlTxhxLXTmzbuIVZaD//pERER+YBjpTH0F0FAh9n24QF5HovQajE42AgC+8B6q0UcDMSlivyIE03tLvwXqigFNBJB6Wfv7B08X99WVAGXfBf/1iYiI/MAw0pkK5xCNKRXQRQX8NO4i1qLq1neEcqjG1SuSdgWgNbS/X2sAhs4Q+xyqISIihTGMdKaHQzQu7iLWdnUjIbx674kOpvS2xaXhiYiol2AY6UwPZ9K4uHpGjpyzwNbi8NwRqp6R5kbgzKdiv6sw4ipiPb0XsLcEtw1ERER+YBjpjHsmTc96RoaZoxAbqYW1xYGjF7wWPwtVGDnzCWC3AsZBQMLozs9LngBExAG2WuB8XnDbQERE5AeGkc4EqWdEkiRMTo0F0Ga9Ee+Fz4I5o8W1vsjwq8TS851RqYBhs8U+60aIiEhBDCMdabECVafFfg/DCNDJFXxjhwKSWlzIrra4x6/h5l5fpIMpvW1xvREiIuoFGEY6UnkKkB2A3ghEJ3V/fjc6XIlVowPihor9YA3V1JwX03oheYJGV1xFrEUHAFt9cNpARETkJ4aRjnjPpOlqqMNHE1NNkCTgbFUjSmu9Fj8Ldt2Ia4hm0BQgMr778+PTAdMQwNEMnNkXnDYQERH5iWGkI0GqF3GJMWgxKjEGQJuhmqCHER+m9HqTJCDdWTdSsCs4bSAiIvITw0hHgjSTxtuUobEA2haxBvGCeQ6HV/Gqj2EE8FzRl3UjRESkEIaRjrh6RgYEL4xM7qiINZg9I8VfAo2VgC4GGDzN98e5ZtQUHwHqy3veDiIiIj8xjLQlyz2+Wm9HXIuffXW2Gs125+JnrjBSVdDzhcdcq64Omw2otb4/LjoRSLxE7Bfs6VkbiIiIAsAw0lZtMWCrE9Nu44cF7WnTzdEwGjRoanYgv7hWHIwZKC5Y52gBLIU9ewHXEM0IP4Zo3I2bI245VENERApgGGnLNUQTlwZo9EF7WpVKwqS2U3xVquDUjVhrgaLPxL4/9SIu6c4pvgW8Tg0REYUfw0hbQZ5J4819BV/vupH4dHHbk7qR03tF70pcmuf5/DF0JqDSiIXeKgsCbwcREVEAGEbaCsFMGpcOFz8LRhGrP6uudkTvVfTK3hEiIgozhpG2QtgzMsl5jZozFQ0or7OKg8EIIyf8XF+kI67VWE8xjBARUXgFFEY2bdqEtLQ0GAwGZGZm4sCBA52e+8YbbyAjIwOxsbGIiorCpEmT8NJLLwXc4JALwUwaF1OEFiMTowEAh11DNd4XzAtE1Wmg8qQouHVN0w2Eq4i1YLdYs4SIiChM/A4j27ZtQ3Z2NtatW4e8vDxMnDgRc+fORWlpaYfnx8fH46GHHsK+ffvw1VdfYdmyZVi2bBnefffdHjc+6Kx1QM1ZsR+CYRoAmOysG3EP1bjCiKUIaG70/wldQzSp0wGDMfCGDZoKaKOAhgqg9JvAn4eIiMhPfoeRDRs2YPny5Vi2bBnGjRuHzZs3IzIyElu3bu3w/Dlz5uDGG2/E2LFjMXz4cPz85z/HhAkTsHfv3h43PuhcQyWRZt+u7RKAdlfwjYwHDLFiv/KU/0/orhfpwRANIC7cl3a52OcUXyIiCiO/wojNZsOhQ4eQlZXleQKVCllZWdi3r/sLrcmyjNzcXOTn52P27M6HFKxWK2pqalptYRHCIRoXVxHrl2er0WJ3iOvDBFo3Ym8BTjkXKgu0eNUb1xshIiIF+BVGysvLYbfbkZSU1Op4UlISiouLO32cxWJBdHQ0dDodrrvuOjz99NO45pprOj0/JycHJpPJvaWmpvrTzMB5X603REYmRiNGr0GDzY78EufiZ4GGkXOHAKsFiIgDBk7qeeNcRaxnPgVabD1/PiIiIh+EZTZNTEwMDh8+jIMHD+K3v/0tsrOzsWvXrk7PX716NSwWi3srKioKRzNDOpPGRSx+FgvAa6gm0CJW1xBN+hxApe554xLHAVEJQHMDcPZgz5+PiIjIB36FEbPZDLVajZKSklbHS0pKkJyc3PmLqFQYMWIEJk2ahF/+8pe4+eabkZOT0+n5er0eRqOx1RYWYRimAYDJzim+niJW1yqsfvaMnAzClF5vKpXXFN9dwXlOIiKibvgVRnQ6HaZOnYrc3Fz3MYfDgdzcXMyYMcPn53E4HLBarf68dOg57J4wEMJhGgCYPFTUjbSf3utHGGmsEsM0QPDCCMCl4YmIKOw0/j4gOzsbS5YsQUZGBqZPn46NGzeivr4ey5YtAwAsXrwYgwYNcvd85OTkICMjA8OHD4fVasWOHTvw0ksv4dlnnw3ud9JT1YWA3Qqo9UDskJC+lKtn5FR5ParqbYhzLeHeUAE0VPo2k+fUbkB2AObRgGlw8BrnKmI9+znQVNOz6cJERJ0pPQrs/aO4OOn31gCp05RuESnI7zCycOFClJWVYe3atSguLsakSZOwc+dOd1FrYWEhVCpPh0t9fT1+9rOf4ezZs4iIiMCYMWPw8ssvY+HChcH7LoLBNUQzYERw6i+6EBupQ3pCFE6V1eNwUTWuGpMoruBbe15M7/UljARrSm+7xg0B4oYBVQWikHX0vOA+PxH1bxe+AvY8BRx9y3Psud3AlCVA1sMhW1aBeje/wwgArFq1CqtWrerwvraFqY8//jgef/zxQF4mvMIwk8bb5NQ4nCqrR15hlQgjA4aLMFJxEhic0fWDZRk4+ZHYHxGEKb1tpc8BDhWIuhGGESIKhnN5IoTk7/AcG/vfgC4K+PIfQN4LwNF/A9c8Akz6kahho36D77ZLGGbSeJsyNBZAByux+lI3UnECsBQCap244m6wcb0RIgqWogPAyzcDW65yBhEJuPRm4GefAQtfAm7cDCx7R8zma6wE3rob2DoXKD6idMspjALqGbkohWkmjYtrJdYviyywO2So/ZlR4xqiGTJD/FURbMNmA5CAsqNAbQkQk9TtQ4iIWjm9F9j9pKcYXlIDE24BZv2yfQ/00JnA/9sD7N8M7PodcPYA8D+zgen/D7jq16xd6wfYM+IS5mGaUUkxiNKpUWdtwfHSWv96RoJxld6uRMYDKRPEPmfVXPyaGwHLWeDCl2KGViDXSCICPEPIf/s+8Px14v8PlQaYfAdw9+eiF6Sz/2PVWmDm3cDKA8C4G0SB/v5ngWemAUdeF89NFy32jABiBktDudh3hYIQU6skTEyNxacnK/BFYTXGpHstfCbLYpn4jrRYgdMfi/1QhRFArDdy4Usxa2fCLaF7HQoue4vo6m6oaL/Vtz3mPK+5vvVzqDRA8nhg8HRg8DQxyyF2aOf/JolkGTjxgegJOeu8irtaJ0LIFff6N0PRNAi45QXxfDvuF0X9/7oTyHsRuO4PYfuDkcKLYQTwDNEYBwP66LC97OQhIozknanCbVPGim7M5nox1c2Y0vGDivaLFVKjEoGkS0PXuPQ5wKd/FnUjXYUjCi9ZFr0YxV+JMfWSr8VQmitgNFUH9rwqDRA5QPw1Wl8GnP9CbAf+R9wflSiCyeAMcYXogZNDM0RIfYssA/nvAHueFP9eAEBjAKYuBWbeI4JFoEZkASv2AZ/8Cfj4D6KX5S8zgMvvAWbdB+gig/ItUO/AMAKEfYjGxX0F36JqcdXcuKHir4CKE52HEfeU3qtCW20+ZIb4y6bmrOitMYenx4i82FuAiuMidFz40hNAGqu6f2xEnPPq0wOcW7xnP6qD43qjCJyyDFiKRNHh2c/FX7kXvgLqS4H8t8UGiOCcdIkIJoOniS0+naG1v3A4xNTcPeuBEmehqTYSyPixCCHBqjPTGoA5DwATfgi88wBw/D0RTL56Dfj+k8Do+cF5HVIcwwgQ9pk0LpOci5+dKK2DpaEZpgEjPGFk2KyOH+QOIyGY0utNFwmkZoohoYJdDCOhZmsASr9tHTpKvgFamtqfq9IACWOA5AlA8qWiC9wdLgYAhlhAHeCvtiSJ54sdAoy/WRxrbhLtOntAXLOo6KCYhl78ldgO/q84L3KAp/dk8HRg0BRAHxNYO0JJlsUQVVO1+ADVRYktxOsL9WmyDDRZgLpS0QOy94+iwB0AdNHA9J8AM1aKoBsK8enA7f8EvvsP8M6DYjbhP24FRs0H5v9e/CFHfRrDCOA1kya8PSMDovVIGxCJ0xUN+KKoCnMGjBDJv7Mi1roy8aEAeKbfhlL6lSKMnNoFTLsr9K/XXzRUtg4dF74SPSCyo/25umgxHJcyQdRxJE8AEscCGn342qs1AEMyxeZiOecMJ5+LXpQLh8Uw0bGdYgMASQXEpYktdqj4wHDvp4nem1D1pMiyaE/FSRHwK08Blc79ilPiatdtqfWeYOIdUtz7kYA2qoN95zmyQ4RHu03Udtlt4usWm1jducW52a2dH3M9XnYABhMQES96ryLinL1d8W2OOfe1EYH9nJobRcCoKwXqSpxbace39jaX8NCbgMt+CmT+NDwLlUkSMHaBqJXb/SSw7xng2Dvi/6fZ94ni13D+XlBQMYwA4oMAUKQwasqQOBFGCqsxxz29t5Or97rW/UgaH57ptsPmAHgcKPhYXLunt//lKMviZ7TvGfFhqVKLD0SVxrmvFrfe+74ck1SAo0Vs9mbA0Sxu3fstnmOur+22Du5zPr7tf+ouUYmtQ0fKRLEabm9c/Mk0CDDdCFxyo/i6xSqCVZGz9+TsQTHc4woCHdEbOwgprv0h3X/AyjJQX+4JGZWnWocPa03Xj9dFi/orVwi0W4FGqygA7ms0EV5hJa59WLE3dxwyOgplXTGYgJgUsU5I5k/E1+GmixILo028Ddhxn/iD6cPHgC9fBa5bH54/1CjoGEZabEBlgdgP8zANIIpY3/jinFj8bHg303tdV+kdEcJZNN4GThYfGE3V4i/5QVPC87r+kmXg+PuiiO7sQaVb45v49NahI3k8ENP5la97PY3eOTzjtXpwbbHodaw+A1SdBqrOePbrSkRYKDniqTloKzqpdUiJSQZqLniFj4JuAockrtsUPwyIHy5+5gOct3FpIuzIsghStnpRPG5raLNfJwJLp/v1zvMbRHDV6EWtlcYg6sDUenHMfVwv7nPv69ucoxePgyR+7xoqRThqrAIaqpwzpbyPVQKyHWhpBGobxfCZv9R68bOOTvTcxiR7fe08FpUoesl6i8QxwJJ/A0deA959SPxR+eL1gHGQGKo0mPzfgvEHlyyLP97sNucfJS2efUktetG0EWILd42VLIveN1s9YK0V/45t9YC1TuynzwEiYsPbJieGkaoC8cusixaJP8wmO4tYDxdVwxE3Tiz8UlUg/qL2HveX5dBdj6Yzag2QNksULZ7a1fvCiCyLFR13PymGCQBPJf+oeeIvXtnh7NWwi/fZYe/kmPO2o2Oyw9lTohVrIah1ordFrXUe0zhvdV77XdwXEds7aymCLSbZGbA6qH9qbhQXp6xyhpO2gcVa4xk2KNrfxYtIgClVBA5X0Ij3DhzdfHhKkjhHawAwINDvVDmyLH5WDc5w0ljpCS2usNJYKf69egcL732Dqe8WHkuSWHpg5LXAR78V9Us158QWCF1M63Cii2rTI2pz9m7aPL2j7l7QZk/o8JU20muLcA7/ucKK9/Go9se0EZ5A7AoTtjrnfj1gq/Xa97pPtnfenuUfAoOmBvaz6yGGEe+ZNAr8Qo5JjkGEVo3aphactBoxUhMh/sqpPiP+c3Up/Vb8x6yJEDNdwiX9ShFGCnYDs7LD97pdCVclP4WONgJIGC22tmRZfJC2DSi1xSLcePdyxA7tXX+th5skeT44MUzp1ignIhb4/lPA7PvF1Pcmi3+ba60dW63Yas4Gt30q5x8isqP1MG1zg9iUoI0Uf4TrosSSFrpo8UeTQhhGFJpJ46JRqzBhsAn7CyqRV2TByAHDxdoRFSdbhxHXqqtpV4S3SMs1/lr4mZhVoeR//A478M12cbGtsu/EMV00MH05MGNV6Cr5KbwkyTnlOF4MFRL5KjpRbP6yNwNNNWJozDuk2OqdvaFarx5Orx7SVsc1zltXz6nOc473H7r2FvEHZ3Ojc4iv0RNKuj3WKIKTa18bIf4PdIWJVuEipnXQaHVe75s9xjCi0Ewab5OHxGF/QSW+KKzGQncYOQHgWs9JriGaUFyltyvmUWL4qvaC6C5PvzK8rw+IX94jrwEfr/fU0+hNQOb/Ay5bwUuOE1HPqLVA1ACxhfy1NIA6pn8M1fqBYUThnhEAmDIkFoDzCr4TOihitTUAZz4V++GqF3GRJLE0/FevirqRcIaRFpt43Y//ILrrATFD4LKVojdEoUIrIiIKrv4dRmQ57Ffr7YiriPV4aR0aY9IQAYgZAy6Fn4pxRuMgZdqZPscTRrAu9K/XYgW+eAnYu1FMDwXEaqIzV4n1TvgXBRHRRaV/hxHX9EJJJQriFJIQo0dqfASKKhuR35KESUDrtUZOeM2iUaLq3dUbcuGwKCyMiAvN6zQ3AodeENeicE1RjE4SRakZy3gtFCKii1T/DiOuIZq4NMVX7psyJA5FlY04UBMvwoilyFOgFO4pvW0ZB4oemfJjwOm9YhXEYLI1AJ9vFSGkvlQcixkorvY5ZXHgq0sSEVGf0AuXdgyjXlAv4jLZeZ2aT887xII9gFjYqea88xoQkrIrC7pe27UKbDDIMvDN/wHPTAPee0gEEdMQ4LoNwM8PiwJVBhEiooteP+8ZUX4mjcuUoa4r+FogDxwB6dznoojVWitOGDRF2Vkjw64EDvwVOLU7OM9Xdgx4535PuDENAa78FTDxVlHZTkRE/UY/DyO9p2dkbIoReo0KlsZm1EUPRQycYaTkG3GCUkM0LmlXiNqaiuPiui+mQYE9j7VOLNu+7y9i1UK1XgzHXPEL9oIQEfVT/XuYpsUmPmB7QRjROhc/A4AzGCgOlh0DTn4k9oeHeX2RtiJiPQtQFQTQOyLLwNdviCGZT/4kgsioecDKz4Crfs0gQkTUj/XvnpFlb4tppFLvWIluypA4HDxdhSNNZlwKiMtjN1nESnreFyBTSvoc4NwhMbQy6XbfH1eWD+y43xNiYocC838PjJ4filYSEVEf0797RgDnlTJ7Ryab7Fz87OMKcYsm5+W906/sHXUU3kWsstz9+dZa4L3fAM/OFEFEYwDmrAZW7mcQISIit97xKUwAPIuf7amIAbxnGg+/SpkGtTV4uggUdSWityNxTMfnyTLw9b9EEKm9II6Nmg/MyxFXVyUiIvLCnpFeJMlowKDYCNTJBlgjkz13KF0v4qI1eK4Y3NkU39KjwAsLgH/dKYJIXBpw+z+B219lECEiog4xjPQyrqGaMu1gcSBuWO/6EHetxtq2iNVaC7z7ELD5CuD0x84hmV8DP9sPjJob/nYSEVGfwTDSy0xxDtXkO5xTZ8N9ld7uuOpGCj4WV9OVZeCr14CnM4B9zwCOFmD0daIuZM4DojeFiIioC6wZ6WVcPSNP1X8f35uZBunye5RtUFvJE8S1aRqrgMOvAF/9EzizV9wXnw7MfxIYeY2ybSQioj6FYaSXuWSgCTqNCt81xODMlAeQFtXLLg6nUgNps4CjbwH/dgYlTQQw+5fAjLvZE0JERH7jME0vo9OocOlAIwAgr7BK4dZ0wns12DH/Baw6AMy+n0GEiIgCwp6RXmjKkDjkFVYjr7AKP5gyWOnmtDf5DsBWByRdovwy9URE1OcxjPRCYr2RAnxRWK10Uzqm1gAz71a6FUREdJHgME0vNGVoLADgu+JaNNhalG0MERFRiDGM9EIppgikmAywO2R8ddaidHOIiIhCimGkl3JN8V3+wuf4yYuf4+XPzqCwokHZRhEREYUAa0Z6qUWZQ7H/VCUq6m1479sSvPdtCQAgbUAkZo9KwOyRCbhs+ABE6/kWEhFR3ybJsi+XX1VWTU0NTCYTLBYLjEaj0s0JG4dDxtfnLdhzrAx7jpcj70wVWhyet0urljBlSBxmj0rAlaMSMC7FCJVKUrDFREREHr5+fjOM9CG1Tc3Yd7ICe46X4ePj5TjTZthmQJQOV4w0Y/bIBMwaZUZiDNf9ICIi5TCM9ANnKuqx51gZdh8rx76T5ai32VvdPyY5BleOSsDsUQnISIuDXqNWqKVERNQfMYz0M812B/LOVLl7TY6cs8D7nTVoVbh6bBJ+c91YpJgilGsoERH1Gwwj/VxFnRV7T5Rjz7FyfHy8DKW1VgBAjEGDNf81Dj+cOhiSxPoSIiIKHYYRcpNlGUfOWbD2zW9wuKgaAHDV6ATk/GACkk2sKyEiotDw9fM7oHVGNm3ahLS0NBgMBmRmZuLAgQOdnrtlyxbMmjULcXFxiIuLQ1ZWVpfnU/BJkoQJg2PxrxUz8eD8MdBpVPgovwzX/HE3Xj90Fn0gjxIR0UXM7zCybds2ZGdnY926dcjLy8PEiRMxd+5clJaWdnj+rl27cNttt+Gjjz7Cvn37kJqaimuvvRbnzp3rcePJP2qVhJ9eORxv330FJg42obapBfe99iXueuFzlNQ0Kd08IiLqp/wepsnMzMS0adPwzDPPAAAcDgdSU1Nx991348EHH+z28Xa7HXFxcXjmmWewePFin16TwzTB12J34K8fn8LG94/DZnfAaNDg4f++BDdOHsRaEiIiCoqQDNPYbDYcOnQIWVlZnidQqZCVlYV9+/b59BwNDQ1obm5GfHx8p+dYrVbU1NS02ii4NGoVfjZnBP599xUYP8iEmqYWZP/zSyx/8RBKa9lLQkRE4eNXGCkvL4fdbkdSUlKr40lJSSguLvbpOR544AEMHDiwVaBpKycnByaTyb2lpqb600zyw+jkGGz/2UzcP3c0tGoJHxwtwTUb9uD/vjjHWhIiIgqLsF4o73e/+x1effVVbN++HQZD57M4Vq9eDYvF4t6KiorC2Mr+R6NWYeVVopfk0kFGWBqbce+2w/h/Lx1CmXNKMBERUaj4FUbMZjPUajVKSkpaHS8pKUFycnKXj12/fj1+97vf4b333sOECRO6PFev18NoNLbaKPTGJBux/WeXI/uaUdCqJbz3bQmu/eNuvPXlefaSEBFRyPgVRnQ6HaZOnYrc3Fz3MYfDgdzcXMyYMaPTxz355JN47LHHsHPnTmRkZATeWgo5rVqFe64eibdWXYFxKUZUNTTjnn98gRUv56G8jr0kREQUfH4P02RnZ2PLli144YUXcPToUaxYsQL19fVYtmwZAGDx4sVYvXq1+/zf//73WLNmDbZu3Yq0tDQUFxejuLgYdXV1wfsuKOjGphjx5qrLcW/WSGhUEnZ+U4xr/7gH//nqvNJNIyKii4zfYWThwoVYv3491q5di0mTJuHw4cPYuXOnu6i1sLAQFy5ccJ//7LPPwmaz4eabb0ZKSop7W79+ffC+CwoJrVqFe7NG4c1Vl2NMcgwq621Y9fcvsPKVPFSwl4SIiIKEy8GTT2wtDjzz4XFs2nUSdoeMAVE63DlrGK4ek4RRSdFcm4SIiNrhtWkoJL4+Z8F9r32J74pr3ccGxUbgqjEJ+N6YRMwcboZBq1awhURE1FswjFDIWFvseP3QWXzwbQk+PVkBa4vDfZ9Bq8LM4WZcNSYR3xuTiEGxEQq2lIiIlMQwQmHRaLPj05Pl+PC7Unz0XSnOW1qv3jo6KcYdTKYMiYVGHdalbYiISEEMIxR2siwjv6QWH35Xig+PliKvsAoOr39dpggtrhwlhnOuHJWAuChdj16vxe5AVUMzKuttqKi3oqLOhsp6scUYNBidHIPRSTFIiNGzpoWISAEMI6S4qnob9hwvw4fflWL3sTJUNzS771NJwOQhcfies9dkTHIM7A4ZlQ3OQFFnQ0W9DRV1VmfYsLnDRkW9OFbd2Axf/vXGRWoxKikGo5NjMCopBmOSYzAyKQamCG0Iv3siImIYoV6lxe7AF0XV7uEc7wJYAIjUqdFgs/v9vJIExEXqEB8ltgFROsRF6VBVb0N+SS1Ol9e36p3xlmIyuHtPXEFlRGI0C3CJiIKEYYR6tXPVjfjou1J8+F0pPj1ZjqZmUQTbNlyYo137eq99HczResRH6RAXqYNa1fkQTFOzHSdK63CspBb5xbXIL6nFseLadrUtLioJSDNHYXRS616UFJMBUXpNSH4WRP2BrcWBU+V1yC+uxXfFzt/H4lq0OByYkT4As0YmYNYoMxJjOr9uGfU9DCPUZzQ123G2qhGxkdpuw0WwWBqbcbxE/KfoHVS8h5LaitKpkRCjR2KMAQkxeveW6L41INGoR3ykDqowfA99lSzLaHHIaGq2w9riEJvXvvu49/0tdrTYZSTE6DEwNgKDYiNgjtaxFqgXcjhknKtudP9OieBRg1Nl9WjprJvSy9gUI2aPNGP2qARMHRrHnso+jmGEyE+yLKOs1or8Es9fbcdKanGitA71fgwhqVUSzNE6T3CJ1iPR6AkukToNbC0O2OwOcdvigNVrX9xnh7W5u3PErd0hQ3a2X5YBGa5b72POfQAO1zHZ8307nI9zUUkSJMD9YS9Jzg2S81acA+e+5D7f6xxJgt3hQFOzCBMiYIh9Hz6TuqXTqDDQZHCHk1a3cRFIMRl6/QeZwyGjpqkZFc7C64o6G6oabKhpbEYw/2PWqlWI0qkRqdeIW50GUfo2tzq137Pdqupt7rDhCh7Hims7/X2J0TsLy11bUgwcMvDx8TJ8fLwcR85ZWp1v0KpwWfoAzB6ZgNmjzBiewAUW+xqGEaIgqrO2oKzWitKaJpTVWVFaY211W1ZrRVltEyrqbT4V1ZKHTqOCXqOCXqMWt1oVDBo19NrWx9UqCaW1VpyrakRJbZNPP2dztA4DYyMw0CQCiggsIsAMiNZDLUlQq5ybJEGlQquv1SrJrw8/W4sDVQ2eUFFRb0OlswjbVZztuq+y3oaqhmbYg5HMgkSnUbULK9F6DSJ1akQ5bzUqCafK6/FdcS3Kaju+LIRWLWF4QjTGJMdgdLIRo5OjMTrZiIEmQ5c/z4o6K/aeKMeeY+X4+HgZSts8/0CTAbNGJmD2qARcPmIAYiN7NiMv2GRZhrXFgZrGZtQ0NcPS2IKapmbn1y2wNtthd4ieQYfrVpZhd3g217FW5zhv7bIMu13cyrIMvUaNCJ0akTrnrVaDCJ0KEToNIrVex3Uar301IrRiX6dWhTzcMYwQKaDF7kBFvU0El9omZ4BpHVzqrS3Qa1TQuTa1yvmBrG51TO+9r3Wd1/4clUpy91S4ei3g7KFQOXsoJO8ejDa9Ga7HAaJnw/U/gqODXpXWPS2te1vgfRyit0WrkpyhQg2D1itwOMOGTq0KaEir2e5AsaUJ56sbcd7SiHNVjThX7fy6uhHnqhsDKojuiCQBGpUEleQVUtSu8OK8lYBaawtqm1oCeo0Yg6ZVEbbRoA3qUJ+txYEGWwvqrXZxa7OjwSpu660tPg2fdCY1PgKjk4zO4CG2YeYoaHu4ppBrqYA9x0Svyf6CSti8FlhUScCEwbGYPSoBs0eaMSk1sHWMZFlGs11GU4vojXQNEzY129HUbEdtU+tA4QoaNW2CRq3zmM3u6P5Fewm1SkKk1hNS/nzbZEwYHBvU12AYIaJ+S5ZlWBqbca66EeedIeWcc3MFlqqGZvdfnMGkkuAuro6P0mGAq/DaVZgdrRezviLFfXGROug0yi4G6A4rXiHFfesdYqx2WFvsGBIfidHO4u7oMBV2N9rs2F9QgY+Pl2PPsTIcL2195fcYvQYzRwzAgGi9CBXOcGF1hYwuboP9KaiSAGOEFkaDFsYIDYwGLWIMGhi0aqhVEjQqqU0PnApqFVrfShI0ahGCNSoRfL1vJQDWFgcabHY02lrQYLOjodmORpvYxH6L8367uN/Wgibn8G9H3lp1OcNIVxhGiCiUHK4u8DZd5HbncYcDaHE44HDAfZ5rc3WzR+k1GBClgykiuL0a1LELlkZ8fKwcu4+X4ZMT5V0Wn/tKkgC9RgWDVvTgRWjV7lARY9C0ChfGiNb7nvu1iNKpe3VtS7Pdgcbm1iGl0WbH2BRj0GcNMowQEVG/YHfIOHLOgn0nK2BtsbvDhEHrGR7s7Fav9YSPcNRQ9De+fn5z4QQiIurT1CoJk1JjMSk1VummUIB41TIiIiJSFMMIERERKYphhIiIiBTFMEJERESKYhghIiIiRTGMEBERkaIYRoiIiEhRDCNERESkKIYRIiIiUhTDCBERESmKYYSIiIgUxTBCREREimIYISIiIkX1iav2yrIMQFyKmIiIiPoG1+e263O8M30ijNTW1gIAUlNTFW4JERER+au2thYmk6nT+yW5u7jSCzgcDpw/fx4xMTGQJCloz1tTU4PU1FQUFRXBaDQG7XnJP3wfege+D70D34fege9DcMiyjNraWgwcOBAqVeeVIX2iZ0SlUmHw4MEhe36j0ch/bL0A34fege9D78D3oXfg+9BzXfWIuLCAlYiIiBTFMEJERESK6tdhRK/XY926ddDr9Uo3pV/j+9A78H3oHfg+9A58H8KrTxSwEhER0cWrX/eMEBERkfIYRoiIiEhRDCNERESkKIYRIiIiUlS/DiObNm1CWloaDAYDMjMzceDAAaWb1K88/PDDkCSp1TZmzBilm3XR27NnDxYsWICBAwdCkiT83//9X6v7ZVnG2rVrkZKSgoiICGRlZeH48ePKNPYi1t37sHTp0na/H/PmzVOmsRexnJwcTJs2DTExMUhMTMQNN9yA/Pz8Vuc0NTVh5cqVGDBgAKKjo3HTTTehpKREoRZfnPptGNm2bRuys7Oxbt065OXlYeLEiZg7dy5KS0uVblq/cskll+DChQvube/evUo36aJXX1+PiRMnYtOmTR3e/+STT+LPf/4zNm/ejP379yMqKgpz585FU1NTmFt6cevufQCAefPmtfr9+Mc//hHGFvYPu3fvxsqVK/HZZ5/h/fffR3NzM6699lrU19e7z/nFL36Bf//733jttdewe/dunD9/Hj/4wQ8UbPVFSO6npk+fLq9cudL9td1ulwcOHCjn5OQo2Kr+Zd26dfLEiROVbka/BkDevn27+2uHwyEnJyfLTz31lPtYdXW1rNfr5X/84x8KtLB/aPs+yLIsL1myRL7++usVaU9/VlpaKgOQd+/eLcuy+Pev1Wrl1157zX3O0aNHZQDyvn37lGrmRadf9ozYbDYcOnQIWVlZ7mMqlQpZWVnYt2+fgi3rf44fP46BAwciPT0dixYtQmFhodJN6tcKCgpQXFzc6nfDZDIhMzOTvxsK2LVrFxITEzF69GisWLECFRUVSjfpomexWAAA8fHxAIBDhw6hubm51e/EmDFjMGTIEP5OBFG/DCPl5eWw2+1ISkpqdTwpKQnFxcUKtar/yczMxPPPP4+dO3fi2WefRUFBAWbNmoXa2lqlm9Zvuf7983dDefPmzcOLL76I3Nxc/P73v8fu3bsxf/582O12pZt20XI4HLj33ntx+eWX49JLLwUgfid0Oh1iY2NbncvfieDqE1ftpYvT/Pnz3fsTJkxAZmYmhg4din/+85+48847FWwZkfJuvfVW9/748eMxYcIEDB8+HLt27cLVV1+tYMsuXitXrsTXX3/N2jUF9MueEbPZDLVa3a4auqSkBMnJyQq1imJjYzFq1CicOHFC6ab0W65///zd6H3S09NhNpv5+xEiq1atwn/+8x989NFHGDx4sPt4cnIybDYbqqurW53P34ng6pdhRKfTYerUqcjNzXUfczgcyM3NxYwZMxRsWf9WV1eHkydPIiUlRemm9FvDhg1DcnJyq9+Nmpoa7N+/n78bCjt79iwqKir4+xFksixj1apV2L59Oz788EMMGzas1f1Tp06FVqtt9TuRn5+PwsJC/k4EUb8dpsnOzsaSJUuQkZGB6dOnY+PGjaivr8eyZcuUblq/cd9992HBggUYOnQozp8/j3Xr1kGtVuO2225TumkXtbq6ulZ/XRcUFODw4cOIj4/HkCFDcO+99+Lxxx/HyJEjMWzYMKxZswYDBw7EDTfcoFyjL0JdvQ/x8fF45JFHcNNNNyE5ORknT57Er371K4wYMQJz585VsNUXn5UrV+Lvf/873nzzTcTExLjrQEwmEyIiImAymXDnnXciOzsb8fHxMBqNuPvuuzFjxgxcdtllCrf+IqL0dB4lPf300/KQIUNknU4nT58+Xf7ss8+UblK/snDhQjklJUXW6XTyoEGD5IULF8onTpxQulkXvY8++kgG0G5bsmSJLMtieu+aNWvkpKQkWa/Xy1dffbWcn5+vbKMvQl29Dw0NDfK1114rJyQkyFqtVh46dKi8fPlyubi4WOlmX3Q6eg8AyH/729/c5zQ2Nso/+9nP5Li4ODkyMlK+8cYb5QsXLijX6IuQJMuyHP4IRERERCT0y5oRIiIi6j0YRoiIiEhRDCNERESkKIYRIiIiUhTDCBERESmKYYSIiIgUxTBCREREimIYISIiIkUxjBAREZGiGEaIiIhIUQwjREREpCiGESIiIlLU/we43bSEVQZ4bwAAAABJRU5ErkJggg==",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(train_loss,label='train loss')\n",
    "plt.plot(val_loss,label='val loss')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 98,
   "id": "6772110b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(train_acc,label='train accuracy')\n",
    "plt.plot(val_acc,label='val accuracy')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "594726be",
   "metadata": {},
   "source": [
    "### Посмотрим, что будет  с моделью для поперечных картинок"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 99,
   "id": "f5b03f07",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(36, 5)"
      ]
     },
     "execution_count": 99,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_X, val_X, train_y, val_y = train_test_split(\n",
    "    cross_images, targets, train_size=0.9, random_state=777, shuffle=True\n",
    ")\n",
    "\n",
    "train_dataset = TiradDataset(\n",
    "    image_paths=train_X, \n",
    "    targets=train_y, \n",
    "    transform=train_transform\n",
    ")\n",
    "\n",
    "val_dataset = TiradDataset(\n",
    "    image_paths=val_X, \n",
    "    targets=val_y, \n",
    "    transform=val_transform\n",
    ")\n",
    "\n",
    "train_loader = torch.utils.data.DataLoader(\n",
    "    train_dataset,\n",
    "    batch_size=4,\n",
    "    shuffle=True,\n",
    "    num_workers=4,\n",
    ")\n",
    "\n",
    "val_loader = torch.utils.data.DataLoader(\n",
    "    val_dataset,\n",
    "    batch_size=4,\n",
    "    num_workers=4,\n",
    ")\n",
    "\n",
    "len(train_dataset), len(val_dataset)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 100,
   "id": "86d60f94",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/era/.local/lib/python3.8/site-packages/torchvision/models/_utils.py:208: UserWarning: The parameter 'pretrained' is deprecated since 0.13 and may be removed in the future, please use 'weights' instead.\n",
      "  warnings.warn(\n",
      "/home/era/.local/lib/python3.8/site-packages/torchvision/models/_utils.py:223: UserWarning: Arguments other than a weight enum or `None` for 'weights' are deprecated since 0.13 and may be removed in the future. The current behavior is equivalent to passing `weights=ResNet18_Weights.IMAGENET1K_V1`. You can also use `weights=ResNet18_Weights.DEFAULT` to get the most up-to-date weights.\n",
      "  warnings.warn(msg)\n"
     ]
    }
   ],
   "source": [
    "model_cross =  models.resnet18(pretrained=True).to(device)\n",
    "\n",
    "model_cross.fc = Classifier().to(device)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 101,
   "id": "b6e7858e",
   "metadata": {},
   "outputs": [],
   "source": [
    "for param in model_cross.parameters():\n",
    "        param.requires_grad = False\n",
    "for param in model_cross.fc.parameters():\n",
    "        param.requires_grad = True\n",
    "        \n",
    "# Лосс и оптимизатор\n",
    "\n",
    "criterion = nn.CrossEntropyLoss()\n",
    "optimizer = torch.optim.Adam(model_cross.parameters(), lr=0.01)\n",
    "\n",
    "exp_lr_scheduler = lr_scheduler.StepLR(optimizer, step_size=5, gamma=0.5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 102,
   "id": "f04f0a07",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "epoch 1/24\n",
      "Training loss: 0.45461061431301963, accuracy: 0.5\n",
      "Validation loss: 0.2696269154548645, accuracy: 0.6\n",
      "epoch 2/24\n",
      "Training loss: 0.22462438543637595, accuracy: 0.4722222222222222\n",
      "Validation loss: 0.3524322032928467, accuracy: 0.4\n",
      "epoch 3/24\n",
      "Training loss: 0.17377252711190116, accuracy: 0.6111111111111112\n",
      "Validation loss: 0.2804721236228943, accuracy: 0.6\n",
      "epoch 4/24\n",
      "Training loss: 0.1823015825615989, accuracy: 0.3611111111111111\n",
      "Validation loss: 0.32607672214508054, accuracy: 0.4\n",
      "epoch 5/24\n",
      "Training loss: 0.1882212973303265, accuracy: 0.5\n",
      "Validation loss: 0.28749220371246337, accuracy: 0.4\n",
      "epoch 6/24\n",
      "Training loss: 0.18035699923833212, accuracy: 0.4722222222222222\n",
      "Validation loss: 0.26046645641326904, accuracy: 0.6\n",
      "epoch 7/24\n",
      "Training loss: 0.1714530752764808, accuracy: 0.5277777777777778\n",
      "Validation loss: 0.2824313521385193, accuracy: 0.4\n",
      "epoch 8/24\n",
      "Training loss: 0.17081819143560198, accuracy: 0.5277777777777778\n",
      "Validation loss: 0.28988187313079833, accuracy: 0.4\n",
      "epoch 9/24\n",
      "Training loss: 0.17249366641044617, accuracy: 0.5833333333333334\n",
      "Validation loss: 0.2601309657096863, accuracy: 0.6\n",
      "epoch 10/24\n",
      "Training loss: 0.1707163022624122, accuracy: 0.5\n",
      "Validation loss: 0.2835571527481079, accuracy: 0.4\n",
      "epoch 11/24\n",
      "Training loss: 0.1696427199575636, accuracy: 0.6111111111111112\n",
      "Validation loss: 0.27763454914093016, accuracy: 0.4\n",
      "epoch 12/24\n",
      "Training loss: 0.16571411656008828, accuracy: 0.6111111111111112\n",
      "Validation loss: 0.2815889358520508, accuracy: 0.4\n",
      "epoch 13/24\n",
      "Training loss: 0.1668321109480328, accuracy: 0.6666666666666666\n",
      "Validation loss: 0.27326891422271726, accuracy: 0.6\n",
      "epoch 14/24\n",
      "Training loss: 0.1697307957543267, accuracy: 0.5833333333333334\n",
      "Validation loss: 0.30625813007354735, accuracy: 0.4\n",
      "epoch 15/24\n",
      "Training loss: 0.16978678438398573, accuracy: 0.5555555555555556\n",
      "Validation loss: 0.2772204756736755, accuracy: 0.6\n",
      "epoch 16/24\n",
      "Training loss: 0.16975187758604685, accuracy: 0.5277777777777778\n",
      "Validation loss: 0.2871042013168335, accuracy: 0.2\n",
      "epoch 17/24\n",
      "Training loss: 0.16728397376007503, accuracy: 0.6666666666666666\n",
      "Validation loss: 0.29613802433013914, accuracy: 0.4\n",
      "epoch 18/24\n",
      "Training loss: 0.1620223025480906, accuracy: 0.7222222222222222\n",
      "Validation loss: 0.28208882808685304, accuracy: 0.4\n",
      "epoch 19/24\n",
      "Training loss: 0.1655242062277264, accuracy: 0.6111111111111112\n",
      "Validation loss: 0.27816071510314944, accuracy: 0.4\n",
      "epoch 20/24\n",
      "Training loss: 0.16520163747999403, accuracy: 0.6388888888888888\n",
      "Validation loss: 0.31513544321060183, accuracy: 0.4\n",
      "epoch 21/24\n",
      "Training loss: 0.15843589107195535, accuracy: 0.75\n",
      "Validation loss: 0.3082041501998901, accuracy: 0.4\n",
      "epoch 22/24\n",
      "Training loss: 0.15866277118523917, accuracy: 0.7222222222222222\n",
      "Validation loss: 0.3080842733383179, accuracy: 0.4\n",
      "epoch 23/24\n",
      "Training loss: 0.1543238643142912, accuracy: 0.7777777777777778\n",
      "Validation loss: 0.30954153537750245, accuracy: 0.4\n",
      "epoch 24/24\n",
      "Training loss: 0.15962363117271, accuracy: 0.6944444444444444\n",
      "Validation loss: 0.30625473260879515, accuracy: 0.4\n"
     ]
    }
   ],
   "source": [
    "model_cross, train_loss, val_loss, train_acc, val_acc = train(\n",
    "    model_cross, criterion, optimizer, train_loader, val_loader, exp_lr_scheduler\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 103,
   "id": "bc78851d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(train_acc,label='train accuracy')\n",
    "plt.plot(val_acc,label='val accuracy')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ded2d995",
   "metadata": {},
   "source": [
    "### Возможно tirad == 5 не очень видны с поперечных снимков"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.15"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}