CIFAR10 ResNet18 from Scratch.ipynb

{
  "cells": [
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "M3AHb_X8eiPi"
      },
      "outputs": [],
      "source": [
        "import torch\n",
        "import torch.nn as nn\n",
        "import torch.nn.functional as F\n",
        "import torch.backends.cudnn as cudnn\n",
        "import torch.optim as optim\n",
        "import os\n",
        "import torchvision\n",
        "import torchvision.transforms as transforms"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "WJGgKUlLekNo"
      },
      "outputs": [],
      "source": [
        "class BasicBlock(nn.Module):\n",
        "    def __init__(self, in_planes, planes, stride = 1):\n",
        "        super(BasicBlock, self).__init__()\n",
        "\n",
        "        self.conv1 = nn.Conv2d(in_planes, planes, kernel_size = 3, stride = stride, padding = 1, bias = False)\n",
        "        self.bn1 = nn.BatchNorm2d(planes)\n",
        "\n",
        "        self.conv2 = nn.Conv2d(planes,planes,kernel_size = 3, stride = 1, padding = 1, bias = False)\n",
        "        self.bn2 = nn.BatchNorm2d(planes)\n",
        "\n",
        "        self.shortcut = nn.Identity()\n",
        "        if stride != 1:\n",
        "            self.shortcut = nn.Sequential(\n",
        "                            nn.Conv2d(in_planes, planes, kernel_size = 1, stride = stride, bias = False),\n",
        "                            nn.BatchNorm2d(planes)\n",
        "                            )\n",
        "\n",
        "    def forward(self, x):\n",
        "        out = F.relu(self.bn1(self.conv1(x)))\n",
        "        out = self.bn2(self.conv2(out))\n",
        "        out += self.shortcut(x)\n",
        "        out = F.relu(out)\n",
        "\n",
        "        return out"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "AVQ_s39hemd0"
      },
      "outputs": [],
      "source": [
        "class ResNet(nn.Module):\n",
        "    def __init__(self, block, num_blocks, num_classes = 10):\n",
        "        super(ResNet, self).__init__()\n",
        "        self.in_planes = 64\n",
        "\n",
        "        self.conv1 = nn.Conv2d(3,64,kernel_size = 3, stride = 1, padding = 1, bias = False)\n",
        "        self.bn1 = nn.BatchNorm2d(64)\n",
        "        self.layer1 = self._make_layer(block, 64, num_blocks[0], stride = 1)\n",
        "        self.layer2 = self._make_layer(block, 128, num_blocks[1], stride = 2)\n",
        "        self.layer3 = self._make_layer(block, 256, num_blocks[2], stride = 2)\n",
        "        self.layer4 = self._make_layer(block, 512, num_blocks[3], stride = 2)\n",
        "        self.linear = nn.Linear(512, num_classes)\n",
        "\n",
        "    def _make_layer(self, block, planes, num_blocks, stride):\n",
        "        strides = [stride] + [1] * (num_blocks - 1)\n",
        "        layers = []\n",
        "\n",
        "        for stride in strides:\n",
        "            layers.append(block(self.in_planes, planes, stride))\n",
        "            self.in_planes = planes\n",
        "\n",
        "        return nn.Sequential(*layers)\n",
        "\n",
        "    def forward(self, x):\n",
        "        out = F.relu(self.bn1(self.conv1(x)))\n",
        "        out = self.layer1(out)\n",
        "        out = self.layer2(out)\n",
        "        out = self.layer3(out)\n",
        "        out = self.layer4(out)\n",
        "        out = F.avg_pool2d(out, 4)\n",
        "        out = out.view(out.size(0), -1) # out.size(0) = out.reshape(0)\n",
        "        out = self.linear(out)\n",
        "\n",
        "        return out"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "LRP4B2Tieqh9",
        "outputId": "5e82c0bd-f3e4-4462-8b94-b0a5c497de78"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Files already downloaded and verified\n",
            "Files already downloaded and verified\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.10/dist-packages/torch/utils/data/dataloader.py:557: UserWarning: This DataLoader will create 4 worker processes in total. Our suggested max number of worker in current system is 2, which is smaller than what this DataLoader is going to create. Please be aware that excessive worker creation might get DataLoader running slow or even freeze, lower the worker number to avoid potential slowness/freeze if necessary.\n",
            "  warnings.warn(_create_warning_msg(\n"
          ]
        }
      ],
      "source": [
        "train = transforms.Compose([\n",
        "    transforms.RandomCrop(32, padding = 4),\n",
        "    transforms.RandomHorizontalFlip(),\n",
        "    transforms.ToTensor(),\n",
        "])\n",
        "\n",
        "test = transforms.Compose([\n",
        "    transforms.ToTensor(),\n",
        "])\n",
        "\n",
        "train_dataset = torchvision.datasets.CIFAR10(root= './data', train = True, download = True, transform = train)\n",
        "test_dataset = torchvision.datasets.CIFAR10(root= './data', train = False, download = True, transform = test)\n",
        "\n",
        "train_loader = torch.utils.data.DataLoader(train_dataset, batch_size = 128, shuffle = True, num_workers = 4)\n",
        "test_loader = torch.utils.data.DataLoader(test_dataset, batch_size = 100, shuffle = False, num_workers = 4)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "6qLQhkZzev6C"
      },
      "outputs": [],
      "source": [
        "def ResNet18():\n",
        "    return ResNet(BasicBlock, [2,2,2,2])"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "on-PcEaSezd4"
      },
      "outputs": [],
      "source": [
        "device = 'cuda'\n",
        "cudnn.benchmark = True\n",
        "\n",
        "net = ResNet18()\n",
        "net = net.to(device)\n",
        "net = torch.nn.DataParallel(net)\n",
        "\n",
        "learning_rate = 0.1\n",
        "file_name = 'resnet18_cifar10.pth'\n",
        "\n",
        "criterion = nn.CrossEntropyLoss()\n",
        "optimizer = optim.SGD(net.parameters(), lr = learning_rate, momentum = 0.9, weight_decay = 0.0002)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "BgkRkOmse0uv"
      },
      "outputs": [],
      "source": [
        "def train(epoch):\n",
        "    print('\\n[ Train epoch: %d ]' % epoch)\n",
        "    net.train()\n",
        "    train_loss = 0\n",
        "    correct = 0\n",
        "    total = 0\n",
        "\n",
        "    for batch_idx, (inputs, targets) in  enumerate(train_loader):\n",
        "        inputs= inputs.to(device)\n",
        "        targets = targets.to(device)\n",
        "\n",
        "        outputs = net(inputs)\n",
        "\n",
        "        optimizer.zero_grad()\n",
        "        loss = criterion(outputs, targets)\n",
        "        loss.backward()\n",
        "        optimizer.step()\n",
        "\n",
        "        train_loss += loss.item()\n",
        "        _, predicted = outputs.max(1)\n",
        "\n",
        "        total += targets.size(0)\n",
        "        correct += predicted.eq(targets).sum().item()\n",
        "\n",
        "        if batch_idx % 100 == 0:\n",
        "            print('\\nCurrent batch:', str(batch_idx))\n",
        "            print('Current benign train accuracy:', str(predicted.eq(targets).sum().item() / targets.size(0)))\n",
        "            print('Current benign train loss:', loss.item())\n",
        "\n",
        "    print('\\nTotal benign train accuarcy:', 100. * correct / total)\n",
        "    print('Total benign train loss:', train_loss)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "gJnaOGe2e4dm"
      },
      "outputs": [],
      "source": [
        "def test(epoch):\n",
        "    print('\\n[ Test epoch: %d ]' % epoch)\n",
        "    net.eval()\n",
        "    loss = 0\n",
        "    correct = 0\n",
        "    total = 0\n",
        "\n",
        "    for batch_idx, (inputs, targets) in enumerate(test_loader):\n",
        "        inputs, targets = inputs.to(device), targets.to(device)\n",
        "        total += targets.size(0)\n",
        "\n",
        "        outputs = net(inputs)\n",
        "        loss += criterion(outputs, targets).item()\n",
        "\n",
        "        _, predicted = outputs.max(1)\n",
        "        correct += predicted.eq(targets).sum().item()\n",
        "\n",
        "    print('\\nTest accuarcy:', 100. * correct / total)\n",
        "    print('Test average loss:', loss / total)\n",
        "\n",
        "    state = {\n",
        "        'net': net.state_dict()\n",
        "    }\n",
        "    if not os.path.isdir('checkpoint'):\n",
        "        os.mkdir('checkpoint')\n",
        "    torch.save(state, './checkpoint/' + file_name)\n",
        "    print('Model Saved!')"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "8am8QwKde7Jz"
      },
      "outputs": [],
      "source": [
        "def adjust_learning_rate(optimizer, epoch):\n",
        "    lr = learning_rate\n",
        "    if epoch >= 100:\n",
        "        lr /= 10\n",
        "    if epoch >= 150:\n",
        "        lr /= 10\n",
        "    for param_group in optimizer.param_groups:\n",
        "        param_group['lr'] = lr"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "exGqrpMGe-hC",
        "outputId": "3edf791f-768f-403f-ad25-f613826ceae1"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "\n",
            "[ Train epoch: 0 ]\n",
            "\n",
            "Current batch: 0\n",
            "Current benign train accuracy: 0.078125\n",
            "Current benign train loss: 2.4161665439605713\n",
            "\n",
            "Current batch: 100\n",
            "Current benign train accuracy: 0.2578125\n",
            "Current benign train loss: 2.0340278148651123\n",
            "\n",
            "Current batch: 200\n",
            "Current benign train accuracy: 0.296875\n",
            "Current benign train loss: 1.8409966230392456\n",
            "\n",
            "Current batch: 300\n",
            "Current benign train accuracy: 0.34375\n",
            "Current benign train loss: 1.7640700340270996\n",
            "\n",
            "Total benign train accuarcy: 28.848\n",
            "Total benign train loss: 790.1268329620361\n",
            "\n",
            "[ Test epoch: 0 ]\n",
            "\n",
            "Test accuarcy: 38.67\n",
            "Test average loss: 0.017548961484432222\n",
            "Model Saved!\n",
            "\n",
            "[ Train epoch: 1 ]\n",
            "\n",
            "Current batch: 0\n",
            "Current benign train accuracy: 0.4609375\n",
            "Current benign train loss: 1.4852935075759888\n",
            "\n",
            "Current batch: 100\n",
            "Current benign train accuracy: 0.5\n",
            "Current benign train loss: 1.4296295642852783\n",
            "\n",
            "Current batch: 200\n",
            "Current benign train accuracy: 0.4453125\n",
            "Current benign train loss: 1.4241976737976074\n",
            "\n",
            "Current batch: 300\n",
            "Current benign train accuracy: 0.5\n",
            "Current benign train loss: 1.4308282136917114\n",
            "\n",
            "Total benign train accuarcy: 45.584\n",
            "Total benign train loss: 578.5272495746613\n",
            "\n",
            "[ Test epoch: 1 ]\n",
            "\n",
            "Test accuarcy: 50.64\n",
            "Test average loss: 0.013403666651248931\n",
            "Model Saved!\n",
            "\n",
            "[ Train epoch: 2 ]\n",
            "\n",
            "Current batch: 0\n",
            "Current benign train accuracy: 0.5390625\n",
            "Current benign train loss: 1.1776936054229736\n",
            "\n",
            "Current batch: 100\n",
            "Current benign train accuracy: 0.515625\n",
            "Current benign train loss: 1.283424973487854\n",
            "\n",
            "Current batch: 200\n",
            "Current benign train accuracy: 0.515625\n",
            "Current benign train loss: 1.3411551713943481\n",
            "\n",
            "Current batch: 300\n",
            "Current benign train accuracy: 0.609375\n",
            "Current benign train loss: 1.1945905685424805\n",
            "\n",
            "Total benign train accuarcy: 55.348\n",
            "Total benign train loss: 485.00874215364456\n",
            "\n",
            "[ Test epoch: 2 ]\n",
            "\n",
            "Test accuarcy: 55.91\n",
            "Test average loss: 0.0126853373169899\n",
            "Model Saved!\n",
            "\n",
            "[ Train epoch: 3 ]\n",
            "\n",
            "Current batch: 0\n",
            "Current benign train accuracy: 0.5234375\n",
            "Current benign train loss: 1.3490324020385742\n",
            "\n",
            "Current batch: 100\n",
            "Current benign train accuracy: 0.6171875\n",
            "Current benign train loss: 0.9896416664123535\n",
            "\n",
            "Current batch: 200\n",
            "Current benign train accuracy: 0.65625\n",
            "Current benign train loss: 1.0637246370315552\n",
            "\n",
            "Current batch: 300\n",
            "Current benign train accuracy: 0.625\n",
            "Current benign train loss: 1.1303431987762451\n",
            "\n",
            "Total benign train accuarcy: 63.5\n",
            "Total benign train loss: 402.7446832060814\n",
            "\n",
            "[ Test epoch: 3 ]\n",
            "\n",
            "Test accuarcy: 49.12\n",
            "Test average loss: 0.014184522724151612\n",
            "Model Saved!\n",
            "\n",
            "[ Train epoch: 4 ]\n",
            "\n",
            "Current batch: 0\n",
            "Current benign train accuracy: 0.6171875\n",
            "Current benign train loss: 1.0414137840270996\n",
            "\n",
            "Current batch: 100\n",
            "Current benign train accuracy: 0.6171875\n",
            "Current benign train loss: 0.9517923593521118\n",
            "\n",
            "Current batch: 200\n",
            "Current benign train accuracy: 0.6953125\n",
            "Current benign train loss: 0.7726293206214905\n",
            "\n",
            "Current batch: 300\n",
            "Current benign train accuracy: 0.703125\n",
            "Current benign train loss: 0.8462729454040527\n",
            "\n",
            "Total benign train accuarcy: 68.49\n",
            "Total benign train loss: 348.587971329689\n",
            "\n",
            "[ Test epoch: 4 ]\n",
            "\n",
            "Test accuarcy: 64.41\n",
            "Test average loss: 0.010101207888126374\n",
            "Model Saved!\n",
            "\n",
            "[ Train epoch: 5 ]\n",
            "\n",
            "Current batch: 0\n",
            "Current benign train accuracy: 0.71875\n",
            "Current benign train loss: 0.8299735188484192\n",
            "\n",
            "Current batch: 100\n",
            "Current benign train accuracy: 0.671875\n",
            "Current benign train loss: 0.8214394450187683\n",
            "\n",
            "Current batch: 200\n",
            "Current benign train accuracy: 0.703125\n",
            "Current benign train loss: 0.7295781970024109\n",
            "\n",
            "Current batch: 300\n",
            "Current benign train accuracy: 0.65625\n",
            "Current benign train loss: 0.8390317559242249\n",
            "\n",
            "Total benign train accuarcy: 73.022\n",
            "Total benign train loss: 299.4558882713318\n",
            "\n",
            "[ Test epoch: 5 ]\n",
            "\n",
            "Test accuarcy: 73.85\n",
            "Test average loss: 0.007425785893201828\n",
            "Model Saved!\n",
            "\n",
            "[ Train epoch: 6 ]\n",
            "\n",
            "Current batch: 0\n",
            "Current benign train accuracy: 0.765625\n",
            "Current benign train loss: 0.6538844108581543\n",
            "\n",
            "Current batch: 100\n",
            "Current benign train accuracy: 0.7734375\n",
            "Current benign train loss: 0.650529682636261\n",
            "\n",
            "Current batch: 200\n",
            "Current benign train accuracy: 0.765625\n",
            "Current benign train loss: 0.6631922125816345\n",
            "\n",
            "Current batch: 300\n",
            "Current benign train accuracy: 0.8671875\n",
            "Current benign train loss: 0.46979719400405884\n",
            "\n",
            "Total benign train accuarcy: 77.01\n",
            "Total benign train loss: 257.70417031645775\n",
            "\n",
            "[ Test epoch: 6 ]\n",
            "\n",
            "Test accuarcy: 73.96\n",
            "Test average loss: 0.007432034111022949\n",
            "Model Saved!\n",
            "\n",
            "[ Train epoch: 7 ]\n",
            "\n",
            "Current batch: 0\n",
            "Current benign train accuracy: 0.8046875\n",
            "Current benign train loss: 0.6408601403236389\n",
            "\n",
            "Current batch: 100\n",
            "Current benign train accuracy: 0.6953125\n",
            "Current benign train loss: 0.7675060629844666\n",
            "\n",
            "Current batch: 200\n",
            "Current benign train accuracy: 0.8359375\n",
            "Current benign train loss: 0.48999226093292236\n",
            "\n",
            "Current batch: 300\n",
            "Current benign train accuracy: 0.7890625\n",
            "Current benign train loss: 0.6103171706199646\n",
            "\n",
            "Total benign train accuarcy: 79.56\n",
            "Total benign train loss: 229.30814635753632\n",
            "\n",
            "[ Test epoch: 7 ]\n",
            "\n",
            "Test accuarcy: 77.03\n",
            "Test average loss: 0.00664035165309906\n",
            "Model Saved!\n",
            "\n",
            "[ Train epoch: 8 ]\n",
            "\n",
            "Current batch: 0\n",
            "Current benign train accuracy: 0.8046875\n",
            "Current benign train loss: 0.48851802945137024\n",
            "\n",
            "Current batch: 100\n",
            "Current benign train accuracy: 0.78125\n",
            "Current benign train loss: 0.5729619264602661\n",
            "\n",
            "Current batch: 200\n",
            "Current benign train accuracy: 0.828125\n",
            "Current benign train loss: 0.5399771928787231\n",
            "\n",
            "Current batch: 300\n",
            "Current benign train accuracy: 0.8203125\n",
            "Current benign train loss: 0.5680356025695801\n",
            "\n",
            "Total benign train accuarcy: 81.906\n",
            "Total benign train loss: 204.43982297182083\n",
            "\n",
            "[ Test epoch: 8 ]\n",
            "\n",
            "Test accuarcy: 77.71\n",
            "Test average loss: 0.006526682442426682\n",
            "Model Saved!\n",
            "\n",
            "[ Train epoch: 9 ]\n",
            "\n",
            "Current batch: 0\n",
            "Current benign train accuracy: 0.8203125\n",
            "Current benign train loss: 0.5025578737258911\n",
            "\n",
            "Current batch: 100\n",
            "Current benign train accuracy: 0.8515625\n",
            "Current benign train loss: 0.3654932975769043\n",
            "\n",
            "Current batch: 200\n",
            "Current benign train accuracy: 0.8359375\n",
            "Current benign train loss: 0.5519003868103027\n",
            "\n",
            "Current batch: 300\n",
            "Current benign train accuracy: 0.796875\n",
            "Current benign train loss: 0.592422366142273\n",
            "\n",
            "Total benign train accuarcy: 83.248\n",
            "Total benign train loss: 186.08199056982994\n",
            "\n",
            "[ Test epoch: 9 ]\n",
            "\n",
            "Test accuarcy: 80.7\n",
            "Test average loss: 0.005701816670596599\n",
            "Model Saved!\n",
            "\n",
            "[ Train epoch: 10 ]\n",
            "\n",
            "Current batch: 0\n",
            "Current benign train accuracy: 0.828125\n",
            "Current benign train loss: 0.5796314477920532\n",
            "\n",
            "Current batch: 100\n",
            "Current benign train accuracy: 0.8515625\n",
            "Current benign train loss: 0.43663060665130615\n",
            "\n",
            "Current batch: 200\n",
            "Current benign train accuracy: 0.859375\n",
            "Current benign train loss: 0.4631894826889038\n",
            "\n",
            "Current batch: 300\n",
            "Current benign train accuracy: 0.828125\n",
            "Current benign train loss: 0.5068852305412292\n",
            "\n",
            "Total benign train accuarcy: 84.776\n",
            "Total benign train loss: 170.25200901925564\n",
            "\n",
            "[ Test epoch: 10 ]\n",
            "\n",
            "Test accuarcy: 82.44\n",
            "Test average loss: 0.0052765056848526\n",
            "Model Saved!\n",
            "\n",
            "[ Train epoch: 11 ]\n",
            "\n",
            "Current batch: 0\n",
            "Current benign train accuracy: 0.8984375\n",
            "Current benign train loss: 0.38049623370170593\n",
            "\n",
            "Current batch: 100\n",
            "Current benign train accuracy: 0.8828125\n",
            "Current benign train loss: 0.3019861578941345\n",
            "\n",
            "Current batch: 200\n",
            "Current benign train accuracy: 0.84375\n",
            "Current benign train loss: 0.3967040777206421\n",
            "\n",
            "Current batch: 300\n",
            "Current benign train accuracy: 0.90625\n",
            "Current benign train loss: 0.3099786639213562\n",
            "\n",
            "Total benign train accuarcy: 85.662\n",
            "Total benign train loss: 161.7069485038519\n",
            "\n",
            "[ Test epoch: 11 ]\n",
            "\n",
            "Test accuarcy: 82.83\n",
            "Test average loss: 0.005101148867607117\n",
            "Model Saved!\n",
            "\n",
            "[ Train epoch: 12 ]\n",
            "\n",
            "Current batch: 0\n",
            "Current benign train accuracy: 0.9140625\n",
            "Current benign train loss: 0.2779412865638733\n",
            "\n",
            "Current batch: 100\n",
            "Current benign train accuracy: 0.859375\n",
            "Current benign train loss: 0.41547247767448425\n",
            "\n",
            "Current batch: 200\n",
            "Current benign train accuracy: 0.8515625\n",
            "Current benign train loss: 0.359947144985199\n",
            "\n",
            "Current batch: 300\n",
            "Current benign train accuracy: 0.84375\n",
            "Current benign train loss: 0.45178675651550293\n",
            "\n",
            "Total benign train accuarcy: 86.514\n",
            "Total benign train loss: 151.14924056828022\n",
            "\n",
            "[ Test epoch: 12 ]\n",
            "\n",
            "Test accuarcy: 83.44\n",
            "Test average loss: 0.005004959163069725\n",
            "Model Saved!\n",
            "\n",
            "[ Train epoch: 13 ]\n",
            "\n",
            "Current batch: 0\n",
            "Current benign train accuracy: 0.8125\n",
            "Current benign train loss: 0.5018453598022461\n",
            "\n",
            "Current batch: 100\n",
            "Current benign train accuracy: 0.8515625\n",
            "Current benign train loss: 0.4503873288631439\n",
            "\n",
            "Current batch: 200\n",
            "Current benign train accuracy: 0.84375\n",
            "Current benign train loss: 0.3644731342792511\n",
            "\n",
            "Current batch: 300\n",
            "Current benign train accuracy: 0.8984375\n",
            "Current benign train loss: 0.24315732717514038\n",
            "\n",
            "Total benign train accuarcy: 87.468\n",
            "Total benign train loss: 142.2157876342535\n",
            "\n",
            "[ Test epoch: 13 ]\n",
            "\n",
            "Test accuarcy: 79.48\n",
            "Test average loss: 0.00644746113717556\n",
            "Model Saved!\n",
            "\n",
            "[ Train epoch: 14 ]\n",
            "\n",
            "Current batch: 0\n",
            "Current benign train accuracy: 0.796875\n",
            "Current benign train loss: 0.4142252206802368\n",
            "\n",
            "Current batch: 100\n",
            "Current benign train accuracy: 0.8671875\n",
            "Current benign train loss: 0.4072858691215515\n",
            "\n",
            "Current batch: 200\n",
            "Current benign train accuracy: 0.875\n",
            "Current benign train loss: 0.337582528591156\n",
            "\n",
            "Current batch: 300\n",
            "Current benign train accuracy: 0.8828125\n",
            "Current benign train loss: 0.3314664959907532\n",
            "\n",
            "Total benign train accuarcy: 87.906\n",
            "Total benign train loss: 136.05463761091232\n",
            "\n",
            "[ Test epoch: 14 ]\n",
            "\n",
            "Test accuarcy: 84.79\n",
            "Test average loss: 0.004545100612938404\n",
            "Model Saved!\n",
            "\n",
            "[ Train epoch: 15 ]\n",
            "\n",
            "Current batch: 0\n",
            "Current benign train accuracy: 0.875\n",
            "Current benign train loss: 0.4226185381412506\n",
            "\n",
            "Current batch: 100\n",
            "Current benign train accuracy: 0.890625\n",
            "Current benign train loss: 0.3299214541912079\n",
            "\n",
            "Current batch: 200\n",
            "Current benign train accuracy: 0.8515625\n",
            "Current benign train loss: 0.318938285112381\n",
            "\n",
            "Current batch: 300\n",
            "Current benign train accuracy: 0.890625\n",
            "Current benign train loss: 0.27762100100517273\n",
            "\n",
            "Total benign train accuarcy: 88.39\n",
            "Total benign train loss: 129.87384817004204\n",
            "\n",
            "[ Test epoch: 15 ]\n",
            "\n",
            "Test accuarcy: 82.48\n",
            "Test average loss: 0.00555184853374958\n",
            "Model Saved!\n",
            "\n",
            "[ Train epoch: 16 ]\n",
            "\n",
            "Current batch: 0\n",
            "Current benign train accuracy: 0.8125\n",
            "Current benign train loss: 0.5524894595146179\n",
            "\n",
            "Current batch: 100\n",
            "Current benign train accuracy: 0.8671875\n",
            "Current benign train loss: 0.31404855847358704\n",
            "\n",
            "Current batch: 200\n",
            "Current benign train accuracy: 0.859375\n",
            "Current benign train loss: 0.38494202494621277\n",
            "\n",
            "Current batch: 300\n",
            "Current benign train accuracy: 0.859375\n",
            "Current benign train loss: 0.3902370035648346\n",
            "\n",
            "Total benign train accuarcy: 88.798\n",
            "Total benign train loss: 126.07847663760185\n",
            "\n",
            "[ Test epoch: 16 ]\n",
            "\n",
            "Test accuarcy: 83.94\n",
            "Test average loss: 0.005049480494856835\n",
            "Model Saved!\n",
            "\n",
            "[ Train epoch: 17 ]\n",
            "\n",
            "Current batch: 0\n",
            "Current benign train accuracy: 0.8828125\n",
            "Current benign train loss: 0.27409806847572327\n",
            "\n",
            "Current batch: 100\n",
            "Current benign train accuracy: 0.90625\n",
            "Current benign train loss: 0.34949636459350586\n",
            "\n",
            "Current batch: 200\n",
            "Current benign train accuracy: 0.8828125\n",
            "Current benign train loss: 0.3057290315628052\n",
            "\n",
            "Current batch: 300\n",
            "Current benign train accuracy: 0.8671875\n",
            "Current benign train loss: 0.3326595723628998\n",
            "\n",
            "Total benign train accuarcy: 89.532\n",
            "Total benign train loss: 119.49821038544178\n",
            "\n",
            "[ Test epoch: 17 ]\n",
            "\n",
            "Test accuarcy: 85.77\n",
            "Test average loss: 0.004204181465506554\n",
            "Model Saved!\n",
            "\n",
            "[ Train epoch: 18 ]\n",
            "\n",
            "Current batch: 0\n",
            "Current benign train accuracy: 0.921875\n",
            "Current benign train loss: 0.23483701050281525\n",
            "\n",
            "Current batch: 100\n",
            "Current benign train accuracy: 0.9375\n",
            "Current benign train loss: 0.1940385401248932\n",
            "\n",
            "Current batch: 200\n",
            "Current benign train accuracy: 0.8984375\n",
            "Current benign train loss: 0.23528273403644562\n",
            "\n",
            "Current batch: 300\n",
            "Current benign train accuracy: 0.9140625\n",
            "Current benign train loss: 0.22861240804195404\n",
            "\n",
            "Total benign train accuarcy: 89.644\n",
            "Total benign train loss: 116.27699079364538\n",
            "\n",
            "[ Test epoch: 18 ]\n",
            "\n",
            "Test accuarcy: 84.75\n",
            "Test average loss: 0.004356096994876862\n",
            "Model Saved!\n",
            "\n",
            "[ Train epoch: 19 ]\n",
            "\n",
            "Current batch: 0\n",
            "Current benign train accuracy: 0.9375\n",
            "Current benign train loss: 0.22542455792427063\n",
            "\n",
            "Current batch: 100\n",
            "Current benign train accuracy: 0.875\n",
            "Current benign train loss: 0.3500388562679291\n",
            "\n",
            "Current batch: 200\n",
            "Current benign train accuracy: 0.9296875\n",
            "Current benign train loss: 0.1808854639530182\n",
            "\n",
            "Current batch: 300\n",
            "Current benign train accuracy: 0.8203125\n",
            "Current benign train loss: 0.5538946390151978\n",
            "\n",
            "Total benign train accuarcy: 90.03\n",
            "Total benign train loss: 112.42608098685741\n",
            "\n",
            "[ Test epoch: 19 ]\n",
            "\n",
            "Test accuarcy: 85.04\n",
            "Test average loss: 0.004640522158145905\n",
            "Model Saved!\n"
          ]
        }
      ],
      "source": [
        "for epoch in range(0, 20):\n",
        "    adjust_learning_rate(optimizer, epoch)\n",
        "    train(epoch)\n",
        "    test(epoch)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "MJQRMDj2i8oD",
        "outputId": "e2975bcd-ffcd-47b4-ffee-2c2e0cd7d3e4"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Model weights loaded successfully!\n",
            "Accuracy on test set: 85.04%\n"
          ]
        }
      ],
      "source": [
        "def test_model_on_test_set(model, test_loader):\n",
        "    model.eval()\n",
        "\n",
        "    correct = 0\n",
        "    total = 0\n",
        "\n",
        "    with torch.no_grad():\n",
        "        for images, labels in test_loader:\n",
        "            images = images.to(device)\n",
        "            labels = labels.to(device)\n",
        "\n",
        "            outputs = model(images)\n",
        "            _, predicted = torch.max(outputs, 1)\n",
        "\n",
        "            total += labels.size(0)\n",
        "            correct += (predicted == labels).sum().item()\n",
        "\n",
        "    accuracy = 100 * correct / total\n",
        "    print(f'Accuracy on test set: {accuracy:.2f}%')\n",
        "\n",
        "def load_model_with_prefix(model, checkpoint_path):\n",
        "    checkpoint = torch.load(checkpoint_path)\n",
        "    model.load_state_dict({k.replace('module.', ''): v for k, v in checkpoint['net'].items()})\n",
        "\n",
        "device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')\n",
        "model = ResNet18()\n",
        "model = model.to(device)\n",
        "\n",
        "checkpoint_path = './checkpoint/' + file_name\n",
        "if os.path.exists(checkpoint_path):\n",
        "    load_model_with_prefix(model, checkpoint_path)\n",
        "    print('Model weights loaded successfully!')\n",
        "else:\n",
        "    print('Model weights not found.')\n",
        "\n",
        "test_model_on_test_set(model, test_loader)"
      ]
    }
  ],
  "metadata": {
    "accelerator": "GPU",
    "colab": {
      "gpuType": "T4",
      "provenance": []
    },
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}