finetune-vgg.ipynb

{
  "cells": [
    {
      "cell_type": "code",
      "execution_count": 1,
      "metadata": {
        "id": "ly7jYi2e-l2n"
      },
      "outputs": [],
      "source": [
        "import os\n",
        "import cv2\n",
        "import numpy as np\n",
        "from torch.utils.data import Dataset, DataLoader\n",
        "import torchvision.transforms as transforms\n",
        "import torch\n",
        "import tqdm\n",
        "import torch.nn as nn"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 55,
      "metadata": {
        "id": "Z2VVPnty-l2o"
      },
      "outputs": [],
      "source": [
        "transform = transforms.Compose([\n",
        "    transforms.ToPILImage(),\n",
        "    transforms.Resize((224, 224)),\n",
        "    transforms.ToTensor(),\n",
        "    transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]),\n",
        "])"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 56,
      "metadata": {
        "id": "pYca-Cyy-l2o"
      },
      "outputs": [],
      "source": [
        "# check device\n",
        "device = 'cuda' if torch.cuda.is_available() else 'cpu'"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 57,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "jRAFAJCV-l2o",
        "outputId": "15a4c0d7-0c53-41f8-eaee-5183774e2a2d"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "torch.Size([999, 3, 224, 224]) torch.Size([999])\n"
          ]
        }
      ],
      "source": [
        "classes = os.listdir('Data/images_original')\n",
        "classes.sort()\n",
        "class_to_idx = {cls_name: i for i, cls_name in enumerate(classes)}\n",
        "idx_to_class = {i: cls_name for i, cls_name in enumerate(classes)}\n",
        "\n",
        "images = []\n",
        "labels = []\n",
        "for i, cls_name in enumerate(classes):\n",
        "    path = os.path.join('Data/images_original', cls_name)\n",
        "    for img_file in os.listdir(path):\n",
        "        img_path = os.path.join(path, img_file)\n",
        "        img = cv2.imread(img_path)\n",
        "        img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)\n",
        "        img = transform(img)\n",
        "        images.append(img)\n",
        "        labels.append(i)\n",
        "images = torch.stack(images).to(device)\n",
        "labels = torch.tensor(labels).to(device)\n",
        "print(images.shape, labels.shape)"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "torch.manual_seed(42)\n",
        "if device == 'cuda':\n",
        "    torch.cuda.manual_seed_all(42)"
      ],
      "metadata": {
        "id": "f7YQsj_qEH1e"
      },
      "execution_count": 58,
      "outputs": []
    },
    {
      "cell_type": "code",
      "execution_count": 59,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "89U9163y-l2p",
        "outputId": "12d282d8-5ecd-4191-c4d5-fcfd5c756199"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "torch.Size([799, 3, 224, 224]) torch.Size([200, 3, 224, 224]) torch.Size([799]) torch.Size([200])\n"
          ]
        }
      ],
      "source": [
        "from sklearn.model_selection import train_test_split\n",
        "\n",
        "train_images, test_images, train_labels, test_labels = train_test_split(images, labels, shuffle=True, test_size=0.2, random_state=42)\n",
        "print(train_images.shape, test_images.shape, train_labels.shape, test_labels.shape)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 60,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 507
        },
        "id": "5jyvFSli-l2p",
        "outputId": "fa679e98-588b-4664-ae94-0b314a45bbb5"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "WARNING:matplotlib.image:Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Text(0.5, 1.0, 'jazz')"
            ]
          },
          "metadata": {},
          "execution_count": 60
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ],
      "source": [
        "# print a image from the dataset\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "plt.imshow(train_images[110].cpu().numpy().transpose(1, 2, 0))\n",
        "plt.title(idx_to_class[train_labels[110].item()])"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 61,
      "metadata": {
        "id": "yi6aLK5M-l2p"
      },
      "outputs": [],
      "source": [
        "\n",
        "class MusicDataset(Dataset):\n",
        "    def __init__(self, images, labels):\n",
        "        self.images = images\n",
        "        self.labels = labels\n",
        "\n",
        "    def __len__(self):\n",
        "        return len(self.images)\n",
        "\n",
        "    def __getitem__(self, idx):\n",
        "        return self.images[idx], self.labels[idx]"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 62,
      "metadata": {
        "id": "TEwExQEc-l2p"
      },
      "outputs": [],
      "source": [
        "dataset_train = MusicDataset(train_images, train_labels)\n",
        "DataLoader_train = DataLoader(dataset_train, batch_size=50, shuffle=True)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 63,
      "metadata": {
        "id": "sr4SU1JM-l2p"
      },
      "outputs": [],
      "source": [
        "import torch.nn as nn\n",
        "\n",
        "class Identity(nn.Module):\n",
        "    def __init__(self):\n",
        "        super(Identity, self).__init__()\n",
        "\n",
        "    def forward(self, x):\n",
        "        return x"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 67,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "fp_Kq3My-l2p",
        "outputId": "24885634-535f-4361-b013-900fd0328e88"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "VGG(\n",
              "  (features): Sequential(\n",
              "    (0): Conv2d(3, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
              "    (1): ReLU(inplace=True)\n",
              "    (2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
              "    (3): ReLU(inplace=True)\n",
              "    (4): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)\n",
              "    (5): Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
              "    (6): ReLU(inplace=True)\n",
              "    (7): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
              "    (8): ReLU(inplace=True)\n",
              "    (9): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)\n",
              "    (10): Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
              "    (11): ReLU(inplace=True)\n",
              "    (12): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
              "    (13): ReLU(inplace=True)\n",
              "    (14): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
              "    (15): ReLU(inplace=True)\n",
              "    (16): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
              "    (17): ReLU(inplace=True)\n",
              "    (18): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)\n",
              "    (19): Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
              "    (20): ReLU(inplace=True)\n",
              "    (21): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
              "    (22): ReLU(inplace=True)\n",
              "    (23): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
              "    (24): ReLU(inplace=True)\n",
              "    (25): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
              "    (26): ReLU(inplace=True)\n",
              "    (27): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)\n",
              "    (28): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
              "    (29): ReLU(inplace=True)\n",
              "    (30): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
              "    (31): ReLU(inplace=True)\n",
              "    (32): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
              "    (33): ReLU(inplace=True)\n",
              "    (34): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
              "    (35): ReLU(inplace=True)\n",
              "    (36): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)\n",
              "  )\n",
              "  (avgpool): Identity()\n",
              "  (classifier): Sequential(\n",
              "    (0): Linear(in_features=25088, out_features=4096, bias=True)\n",
              "    (1): ReLU()\n",
              "    (2): Dropout(p=0.5, inplace=False)\n",
              "    (3): Linear(in_features=4096, out_features=512, bias=True)\n",
              "    (4): ReLU()\n",
              "    (5): Dropout(p=0.5, inplace=False)\n",
              "    (6): Linear(in_features=512, out_features=256, bias=True)\n",
              "    (7): ReLU()\n",
              "    (8): Dropout(p=0.5, inplace=False)\n",
              "    (9): Linear(in_features=256, out_features=10, bias=True)\n",
              "    (10): LogSoftmax(dim=1)\n",
              "  )\n",
              ")"
            ]
          },
          "metadata": {},
          "execution_count": 67
        }
      ],
      "source": [
        "from torchvision.models import vgg19\n",
        "model = vgg19(pretrained=True)\n",
        "\n",
        "\n",
        "for param in model.parameters():\n",
        "    param.requires_grad = False\n",
        "\n",
        "model.avgpool = Identity()\n",
        "\n",
        "model.classifier = nn.Sequential(\n",
        "    nn.Linear(25088, 4096),\n",
        "    nn.ReLU(),\n",
        "    nn.Dropout(0.5),\n",
        "    nn.Linear(4096, 512),\n",
        "    nn.ReLU(),\n",
        "    nn.Dropout(0.5),\n",
        "    nn.Linear(512, 256),\n",
        "    nn.ReLU(),\n",
        "    nn.Dropout(0.5),\n",
        "    nn.Linear(256, 10),\n",
        "    nn.LogSoftmax(dim=1)\n",
        ")\n",
        "\n",
        "model = model.to(device)\n",
        "model"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 68,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Ffrtx1T8-l2p",
        "outputId": "c822c9e1-822b-4670-a8fe-61b3d448a07f"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "100%|██████████| 16/16 [00:02<00:00,  5.72it/s]\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 0, loss_train: 0.14147788286209106, loss_val: 2.0461575984954834, accuracy_val: 0.275\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "100%|██████████| 16/16 [00:02<00:00,  5.73it/s]\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1, loss_train: 0.11256866157054901, loss_val: 1.7315908670425415, accuracy_val: 0.44\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "100%|██████████| 16/16 [00:02<00:00,  5.72it/s]\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 2, loss_train: 0.11232107877731323, loss_val: 1.4877068996429443, accuracy_val: 0.465\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "100%|██████████| 16/16 [00:02<00:00,  5.71it/s]\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 3, loss_train: 0.09857981652021408, loss_val: 1.3799045085906982, accuracy_val: 0.515\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "100%|██████████| 16/16 [00:02<00:00,  5.70it/s]\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 4, loss_train: 0.08827432245016098, loss_val: 1.294062614440918, accuracy_val: 0.535\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "100%|██████████| 16/16 [00:02<00:00,  5.70it/s]\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 5, loss_train: 0.0543968491256237, loss_val: 1.1753191947937012, accuracy_val: 0.575\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "100%|██████████| 16/16 [00:02<00:00,  5.68it/s]\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 6, loss_train: 0.052727289497852325, loss_val: 1.1255176067352295, accuracy_val: 0.62\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "100%|██████████| 16/16 [00:02<00:00,  5.68it/s]\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 7, loss_train: 0.06571488827466965, loss_val: 1.0616832971572876, accuracy_val: 0.64\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "100%|██████████| 16/16 [00:02<00:00,  5.68it/s]\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 8, loss_train: 0.03920059651136398, loss_val: 1.0677241086959839, accuracy_val: 0.64\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "100%|██████████| 16/16 [00:02<00:00,  5.64it/s]\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 9, loss_train: 0.04016546532511711, loss_val: 1.1211111545562744, accuracy_val: 0.625\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "100%|██████████| 16/16 [00:02<00:00,  5.62it/s]\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 10, loss_train: 0.04252101480960846, loss_val: 1.0286717414855957, accuracy_val: 0.67\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "100%|██████████| 16/16 [00:02<00:00,  5.62it/s]\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 11, loss_train: 0.03570150211453438, loss_val: 1.1004862785339355, accuracy_val: 0.655\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "100%|██████████| 16/16 [00:02<00:00,  5.61it/s]\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 12, loss_train: 0.02411029487848282, loss_val: 1.2193114757537842, accuracy_val: 0.65\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "100%|██████████| 16/16 [00:02<00:00,  5.61it/s]\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 13, loss_train: 0.027886103838682175, loss_val: 1.303288221359253, accuracy_val: 0.605\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "100%|██████████| 16/16 [00:02<00:00,  5.61it/s]\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 14, loss_train: 0.03480253741145134, loss_val: 1.1574209928512573, accuracy_val: 0.635\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "100%|██████████| 16/16 [00:02<00:00,  5.61it/s]\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 15, loss_train: 0.027502307668328285, loss_val: 1.3478784561157227, accuracy_val: 0.615\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "100%|██████████| 16/16 [00:02<00:00,  5.61it/s]\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 16, loss_train: 0.02177617698907852, loss_val: 1.2270087003707886, accuracy_val: 0.66\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "100%|██████████| 16/16 [00:02<00:00,  5.62it/s]\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 17, loss_train: 0.01861315779387951, loss_val: 1.2387034893035889, accuracy_val: 0.63\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "100%|██████████| 16/16 [00:02<00:00,  5.62it/s]\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 18, loss_train: 0.008181529119610786, loss_val: 1.4332902431488037, accuracy_val: 0.65\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "100%|██████████| 16/16 [00:02<00:00,  5.63it/s]\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 19, loss_train: 0.013150743208825588, loss_val: 1.3508830070495605, accuracy_val: 0.65\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "100%|██████████| 16/16 [00:02<00:00,  5.63it/s]\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 20, loss_train: 0.018280500546097755, loss_val: 1.526210069656372, accuracy_val: 0.655\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "100%|██████████| 16/16 [00:02<00:00,  5.64it/s]\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 21, loss_train: 0.02146727964282036, loss_val: 1.4533003568649292, accuracy_val: 0.675\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "100%|██████████| 16/16 [00:02<00:00,  5.64it/s]\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 22, loss_train: 0.007510902360081673, loss_val: 1.5504333972930908, accuracy_val: 0.655\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "100%|██████████| 16/16 [00:02<00:00,  5.65it/s]\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 23, loss_train: 0.011682013049721718, loss_val: 1.4852566719055176, accuracy_val: 0.635\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "100%|██████████| 16/16 [00:02<00:00,  5.66it/s]\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 24, loss_train: 0.010403943248093128, loss_val: 1.3541765213012695, accuracy_val: 0.68\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "100%|██████████| 16/16 [00:02<00:00,  5.66it/s]\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 25, loss_train: 0.002472061663866043, loss_val: 1.4994065761566162, accuracy_val: 0.645\n",
            "Early stopping at epoch 25\n"
          ]
        }
      ],
      "source": [
        "model = model.to(device)\n",
        "\n",
        "# 定义损失函数和优化器\n",
        "criterion = nn.CrossEntropyLoss()\n",
        "optimizer = torch.optim.Adam(model.parameters())\n",
        "\n",
        "# 训练模型\n",
        "num_epochs = 300\n",
        "patience = 15  # Number of epochs to wait for improvement before stopping\n",
        "best_val_loss = float('inf')  # Initialize best validation loss\n",
        "patience_counter = 0  # Counter to keep track of the number of epochs without improvement\n",
        "log = {}\n",
        "\n",
        "\n",
        "for epoch in range(num_epochs):\n",
        "    model.train()\n",
        "    run_loss = 0.0\n",
        "    for images, labels in tqdm.tqdm(DataLoader_train):\n",
        "        optimizer.zero_grad()\n",
        "        outputs = model(images)\n",
        "        loss = criterion(outputs, labels)\n",
        "        loss.backward()\n",
        "        optimizer.step()\n",
        "        run_loss += loss.item()\n",
        "\n",
        "    model.eval()\n",
        "    with torch.no_grad():\n",
        "        test_outputs = model(test_images)\n",
        "        test_loss = criterion(test_outputs, test_labels)\n",
        "        pred = torch.argmax(test_outputs, dim=1)\n",
        "        acc = (pred == test_labels).sum().item() / len(test_labels)\n",
        "\n",
        "        # save the loss and accuracy\n",
        "        log[epoch] = {'loss_train': loss.item(), 'loss_val': test_loss.item(), 'accuracy_val': acc}\n",
        "        print(f'Epoch {epoch}, loss_train: {loss.item() / len(DataLoader_train)}, loss_val: {test_loss.item()}, accuracy_val: {acc}')\n",
        "\n",
        "    if test_loss < best_val_loss:\n",
        "        best_val_loss = test_loss\n",
        "        patience_counter = 0\n",
        "        torch.save(model.state_dict(), 'find-tuned-vgg.pth')\n",
        "    else:\n",
        "        patience_counter += 1\n",
        "        if patience_counter >= patience:\n",
        "            print(f'Early stopping at epoch {epoch}')\n",
        "            break"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 69,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 430
        },
        "id": "Ws8_5U3h-l2q",
        "outputId": "c57df28c-52ae-4170-b66f-5e570f4e44e6"
      },
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ],
      "source": [
        "# draw the loss and accuracy\n",
        "import matplotlib.pyplot as plt\n",
        "plt.figure()\n",
        "plt.plot([log[k]['loss_train'] for k in log], label='train loss')\n",
        "plt.plot([log[k]['loss_val'] for k in log], label='val loss')\n",
        "plt.legend()\n",
        "plt.show()\n"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# load the model\n",
        "model.load_state_dict(torch.load('find-tuned-vgg.pth'))\n",
        "\n",
        "# test the model\n",
        "model.eval()\n",
        "from sklearn.metrics import classification_report\n",
        "\n",
        "print(classification_report(test_labels.detach().cpu().numpy(), torch.argmax(model(test_images), dim=1).detach().cpu().numpy()))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "G9F2qUZzB4Gx",
        "outputId": "779b06c6-f318-4f10-93c8-bb1bc34293c9"
      },
      "execution_count": 70,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "              precision    recall  f1-score   support\n",
            "\n",
            "           0       0.63      0.57      0.60        21\n",
            "           1       0.83      0.83      0.83        12\n",
            "           2       0.59      0.54      0.57        24\n",
            "           3       0.55      0.55      0.55        22\n",
            "           4       0.80      0.80      0.80        15\n",
            "           5       0.77      0.89      0.83        27\n",
            "           6       1.00      0.83      0.91        18\n",
            "           7       0.43      0.84      0.57        19\n",
            "           8       0.68      0.68      0.68        22\n",
            "           9       1.00      0.25      0.40        20\n",
            "\n",
            "    accuracy                           0.67       200\n",
            "   macro avg       0.73      0.68      0.67       200\n",
            "weighted avg       0.72      0.67      0.66       200\n",
            "\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [],
      "metadata": {
        "id": "1qcY4anlMxzF"
      },
      "execution_count": null,
      "outputs": []
    }
  ],
  "metadata": {
    "kernelspec": {
      "display_name": "Python 3",
      "language": "python",
      "name": "python3"
    },
    "language_info": {
      "codemirror_mode": {
        "name": "ipython",
        "version": 3
      },
      "file_extension": ".py",
      "mimetype": "text/x-python",
      "name": "python",
      "nbconvert_exporter": "python",
      "pygments_lexer": "ipython3",
      "version": "3.8.18"
    },
    "colab": {
      "provenance": []
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}