{"metadata":{"kernelspec":{"name":"python3","display_name":"Python 3","language":"python"},"language_info":{"name":"python","version":"3.10.12","mimetype":"text/x-python","codemirror_mode":{"name":"ipython","version":3},"pygments_lexer":"ipython3","nbconvert_exporter":"python","file_extension":".py"},"kaggle":{"accelerator":"gpu","dataSources":[{"sourceId":5441,"databundleVersionId":38425,"sourceType":"competition"}],"dockerImageVersionId":30626,"isInternetEnabled":true,"language":"python","sourceType":"notebook","isGpuEnabled":true}},"nbformat_minor":4,"nbformat":4,"cells":[{"cell_type":"markdown","source":"Kaggle link: https://www.kaggle.com/c/dogs-vs-cats-redux-kernels-edition","metadata":{}},{"cell_type":"code","source":"# This Python 3 environment comes with many helpful analytics libraries installed\n# It is defined by the kaggle/python Docker image: https://github.com/kaggle/docker-python\n# For example, here's several helpful packages to load\n\nimport numpy as np # linear algebra\nimport pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)\n\n# Input data files are available in the read-only \"../input/\" directory\n# For example, running this (by clicking run or pressing Shift+Enter) will list all files under the input directory\n\nimport os\nfor dirname, _, filenames in os.walk('/kaggle/input'):\n    for filename in filenames:\n        print(os.path.join(dirname, filename))\n\n# You can write up to 20GB to the current directory (/kaggle/working/) that gets preserved as output when you create a version using \"Save & Run All\" \n# You can also write temporary files to /kaggle/temp/, but they won't be saved outside of the current session","metadata":{"_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","execution":{"iopub.status.busy":"2024-01-11T21:08:21.060258Z","iopub.execute_input":"2024-01-11T21:08:21.061126Z","iopub.status.idle":"2024-01-11T21:08:21.069254Z","shell.execute_reply.started":"2024-01-11T21:08:21.061085Z","shell.execute_reply":"2024-01-11T21:08:21.068270Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## Import everything needed","metadata":{}},{"cell_type":"code","source":"import zipfile\nimport glob\nfrom PIL import Image\nimport matplotlib.pyplot as plt\n\nfrom sklearn.model_selection import train_test_split\nimport torch\nimport torch.nn as nn\nimport torch.nn.functional as F\nimport torch.optim as optim\nfrom torch.utils.data import DataLoader, Dataset\nfrom torchvision import datasets, transforms\nnp.random.seed(0)\ntorch.manual_seed(0)\ntorch.cuda.manual_seed(0)","metadata":{"execution":{"iopub.status.busy":"2024-01-11T21:08:21.070838Z","iopub.execute_input":"2024-01-11T21:08:21.071152Z","iopub.status.idle":"2024-01-11T21:08:21.083247Z","shell.execute_reply.started":"2024-01-11T21:08:21.071120Z","shell.execute_reply":"2024-01-11T21:08:21.082309Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"device = \"cuda\" if torch.cuda.is_available() else \"cpu\"\ndevice","metadata":{"execution":{"iopub.status.busy":"2024-01-11T21:08:21.084816Z","iopub.execute_input":"2024-01-11T21:08:21.085111Z","iopub.status.idle":"2024-01-11T21:08:21.095910Z","shell.execute_reply.started":"2024-01-11T21:08:21.085069Z","shell.execute_reply":"2024-01-11T21:08:21.095044Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## Unzip datasets","metadata":{}},{"cell_type":"code","source":"train_dir = 'train'\ntest_dir = 'test'\nwith zipfile.ZipFile('/kaggle/input/dogs-vs-cats-redux-kernels-edition/train.zip') as train_zip:\n    train_zip.extractall('')\n    \nwith zipfile.ZipFile('/kaggle/input/dogs-vs-cats-redux-kernels-edition/test.zip') as test_zip:\n    test_zip.extractall('')\ntrain_list = glob.glob(os.path.join(train_dir,'*.jpg'))\ntest_list = glob.glob(os.path.join(test_dir, '*.jpg'))\n\nprint(f\"Train Data: {len(train_list)}\")\nprint(f\"Test Data: {len(test_list)}\")","metadata":{"execution":{"iopub.status.busy":"2024-01-11T21:08:21.098134Z","iopub.execute_input":"2024-01-11T21:08:21.098433Z","iopub.status.idle":"2024-01-11T21:08:32.256264Z","shell.execute_reply.started":"2024-01-11T21:08:21.098404Z","shell.execute_reply":"2024-01-11T21:08:32.255235Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"#train_list = train_list[:2500]\nprint(f\"Train Data: {len(train_list)}\")\nprint(f\"Test Data: {len(test_list)}\")\ntrain_list[0]","metadata":{"execution":{"iopub.status.busy":"2024-01-11T21:08:32.257399Z","iopub.execute_input":"2024-01-11T21:08:32.257679Z","iopub.status.idle":"2024-01-11T21:08:32.265123Z","shell.execute_reply.started":"2024-01-11T21:08:32.257656Z","shell.execute_reply":"2024-01-11T21:08:32.264095Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"labels = [path.split('/')[-1].split('.')[0] for path in train_list]\nlen(labels)","metadata":{"execution":{"iopub.status.busy":"2024-01-11T21:08:32.266341Z","iopub.execute_input":"2024-01-11T21:08:32.266650Z","iopub.status.idle":"2024-01-11T21:08:32.859218Z","shell.execute_reply.started":"2024-01-11T21:08:32.266626Z","shell.execute_reply":"2024-01-11T21:08:32.858253Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## Plot random image with their label","metadata":{}},{"cell_type":"code","source":"random_idx = np.random.randint(1, len(train_list), size=9)\nfig, axes = plt.subplots(3, 3, figsize=(16, 12))\n\nfor idx, ax in zip(random_idx, axes.ravel()):  # fixed usage of random_idx\n    img = Image.open(train_list[idx])\n    ax.set_title(labels[idx])\n    ax.imshow(img)","metadata":{"execution":{"iopub.status.busy":"2024-01-11T21:08:32.860409Z","iopub.execute_input":"2024-01-11T21:08:32.861190Z","iopub.status.idle":"2024-01-11T21:08:35.535556Z","shell.execute_reply.started":"2024-01-11T21:08:32.861163Z","shell.execute_reply":"2024-01-11T21:08:35.534099Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## Use Sklearn to split the training data","metadata":{}},{"cell_type":"code","source":"train_list, valid_list = train_test_split(train_list,\n                                          test_size=0.2,\n                                          stratify=labels,\n                                          random_state=0)","metadata":{"execution":{"iopub.status.busy":"2024-01-11T21:08:35.536952Z","iopub.execute_input":"2024-01-11T21:08:35.537316Z","iopub.status.idle":"2024-01-11T21:08:35.575778Z","shell.execute_reply.started":"2024-01-11T21:08:35.537286Z","shell.execute_reply":"2024-01-11T21:08:35.574797Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"print(f\"Train Data: {len(train_list)}\")\nprint(f\"Validation Data: {len(valid_list)}\")\nprint(f\"Test Data: {len(test_list)}\")","metadata":{"execution":{"iopub.status.busy":"2024-01-11T21:08:35.579671Z","iopub.execute_input":"2024-01-11T21:08:35.580462Z","iopub.status.idle":"2024-01-11T21:08:35.585644Z","shell.execute_reply.started":"2024-01-11T21:08:35.580428Z","shell.execute_reply":"2024-01-11T21:08:35.584632Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"We will discuss this in more detail in a near future...","metadata":{}},{"cell_type":"code","source":"SIZE = 224\ntrain_transforms = transforms.Compose([\n    transforms.Resize((SIZE, SIZE)),\n    transforms.TrivialAugmentWide(),\n    transforms.RandomHorizontalFlip(p=0.5),\n    transforms.ToTensor(),\n])\n\ntest_transforms = transforms.Compose([\n    transforms.Resize((SIZE, SIZE)),\n    transforms.ToTensor(),\n    #transforms.Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5))\n])","metadata":{"execution":{"iopub.status.busy":"2024-01-11T21:08:35.587375Z","iopub.execute_input":"2024-01-11T21:08:35.587762Z","iopub.status.idle":"2024-01-11T21:08:35.598064Z","shell.execute_reply.started":"2024-01-11T21:08:35.587728Z","shell.execute_reply":"2024-01-11T21:08:35.596954Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"Define the dataset using PIL to read image","metadata":{}},{"cell_type":"code","source":"class CatsDogsDataset(Dataset):\n    def __init__(self, file_list, transform=None):\n        self.file_list = file_list\n        self.transform = transform\n        self.filelength = len(file_list)\n\n    def __len__(self):\n        return self.filelength\n\n    def __getitem__(self, idx):\n        img_path = self.file_list[idx]\n        img = Image.open(img_path)\n        img_transformed = self.transform(img)\n        label = img_path.split(\"/\")[-1].split(\".\")[0]\n        label = 1 if label == \"dog\" else 0\n        return img_transformed, label","metadata":{"execution":{"iopub.status.busy":"2024-01-11T21:08:35.599319Z","iopub.execute_input":"2024-01-11T21:08:35.599634Z","iopub.status.idle":"2024-01-11T21:08:35.610289Z","shell.execute_reply.started":"2024-01-11T21:08:35.599611Z","shell.execute_reply":"2024-01-11T21:08:35.609220Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"train_data = CatsDogsDataset(train_list, transform=train_transforms)\nvalid_data = CatsDogsDataset(valid_list, transform=test_transforms)\ntest_data = CatsDogsDataset(test_list, transform=test_transforms)","metadata":{"execution":{"iopub.status.busy":"2024-01-11T21:08:35.611734Z","iopub.execute_input":"2024-01-11T21:08:35.612191Z","iopub.status.idle":"2024-01-11T21:08:35.620418Z","shell.execute_reply.started":"2024-01-11T21:08:35.612156Z","shell.execute_reply":"2024-01-11T21:08:35.619464Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"train_data[0][0].shape\nlen(train_data)","metadata":{"execution":{"iopub.status.busy":"2024-01-11T21:08:35.621614Z","iopub.execute_input":"2024-01-11T21:08:35.621919Z","iopub.status.idle":"2024-01-11T21:08:35.636655Z","shell.execute_reply.started":"2024-01-11T21:08:35.621886Z","shell.execute_reply":"2024-01-11T21:08:35.635679Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"Create dataloader, you can modify the batch size if needed","metadata":{}},{"cell_type":"code","source":"NUM_WORKERS = os.cpu_count()\nNUM_WORKERS","metadata":{"execution":{"iopub.status.busy":"2024-01-11T21:08:35.638224Z","iopub.execute_input":"2024-01-11T21:08:35.638603Z","iopub.status.idle":"2024-01-11T21:08:35.647734Z","shell.execute_reply.started":"2024-01-11T21:08:35.638568Z","shell.execute_reply":"2024-01-11T21:08:35.646707Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"batch_size = 64\ntrain_loader = DataLoader(dataset=train_data, batch_size=batch_size, num_workers=NUM_WORKERS, shuffle=True)\nvalid_loader = DataLoader(dataset=valid_data, batch_size=batch_size, num_workers=NUM_WORKERS, shuffle=False)\ntest_loader = DataLoader(dataset=test_data, batch_size=batch_size, num_workers=NUM_WORKERS, shuffle=False)","metadata":{"execution":{"iopub.status.busy":"2024-01-11T21:08:35.649028Z","iopub.execute_input":"2024-01-11T21:08:35.649327Z","iopub.status.idle":"2024-01-11T21:08:35.656882Z","shell.execute_reply.started":"2024-01-11T21:08:35.649302Z","shell.execute_reply":"2024-01-11T21:08:35.655787Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"import torch.nn as nn\nimport torch.nn.functional as F\n\n\nclass AlexNet(nn.Module):\n    \n    def __init__(self):\n        super(AlexNet, self).__init__()\n        self.conv1 = nn.Conv2d(3, 96, 11, stride=4)\n        self.batch1 = nn.BatchNorm2d(96)\n        self.maxPool = nn.MaxPool2d(3, stride=2)\n        self.conv2 = nn.Conv2d(96, 256, 5, padding=2)\n        self.batch2 = nn.BatchNorm2d(256)\n        self.conv3 = nn.Conv2d(256, 384, 3, padding=1)\n        self.batch3 = nn.BatchNorm2d(384)\n        self.conv4 = nn.Conv2d(384, 384, 3, padding=1)\n        self.batch4 = nn.BatchNorm2d(384)\n        self.conv5 = nn.Conv2d(384, 256, 3, padding=1)\n        self.batch5 = nn.BatchNorm2d(256)\n\n        self.fc1 = nn.Linear(5 * 5 * 256, 4096)\n        self.fc2 = nn.Linear(4096, 4096)\n        self.fc3 = nn.Linear(4096, 1000)\n        self.fc4 = nn.Linear(1000, 256)\n        self.fc5 = nn.Linear(256, 2)\n        self.relu = nn.ReLU()\n        self.dropout = nn.Dropout(p=0.5)\n    \n    def forward(self, x):\n        x = self.maxPool(self.relu(self.batch1(self.conv1(x))))\n        x = self.maxPool(self.relu(self.batch2(self.conv2(x))))\n        x = self.dropout(self.relu(self.batch3(self.conv3(x))))\n        x = self.dropout(self.relu(self.batch4(self.conv4(x))))\n        x = self.dropout(self.relu(self.batch5(self.conv5(x))))\n        x = self.maxPool(x)\n        x = x.reshape(x.size(0), -1)\n        x = self.dropout(self.relu(self.fc1(x)))\n        x = self.dropout(self.relu(self.fc2(x)))\n        x = self.dropout(self.relu(self.fc3(x)))\n        x = self.dropout(self.relu(self.fc4(x)))\n        return self.fc5(x)\n\nnet = AlexNet().to(device)","metadata":{"execution":{"iopub.status.busy":"2024-01-11T21:08:35.658204Z","iopub.execute_input":"2024-01-11T21:08:35.658957Z","iopub.status.idle":"2024-01-11T21:08:36.137467Z","shell.execute_reply.started":"2024-01-11T21:08:35.658917Z","shell.execute_reply":"2024-01-11T21:08:36.136628Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"import torch.optim as optim\nlearning_rate = 0.003\nweight_decay = 0.00001\nmomentum = 0.9\ncriterion = nn.CrossEntropyLoss()\noptimizer = optim.SGD(net.parameters(), lr=learning_rate, weight_decay=weight_decay, momentum=momentum)","metadata":{"execution":{"iopub.status.busy":"2024-01-11T21:08:36.138710Z","iopub.execute_input":"2024-01-11T21:08:36.139021Z","iopub.status.idle":"2024-01-11T21:08:36.144566Z","shell.execute_reply.started":"2024-01-11T21:08:36.138996Z","shell.execute_reply":"2024-01-11T21:08:36.143582Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"import wandb\n\nepochs = 10\n\nwandb.init(\n    project='CATS_VS_DOGS',\n    save_code=True,\n    config={\n        \"learning_rate\": learning_rate,\n        \"epochs\": epochs,\n        \"batch_size\": batch_size,\n        \"weight_decay\": weight_decay,\n        \"num_training_samples\": len(train_data),\n        \"momentum\": momentum,\n        \"optimizer\": type(optimizer)\n    },\n    mode=\"disabled\"\n)\n\n\ndef train_loop(dataloader, model, loss_fn, optimizer):\n    num_batches = len(dataloader)\n    model.train()\n    train_loss = 0\n    for batch, (X, y) in enumerate(dataloader):\n        # Compute prediction and loss\n        X, y = X.to(device), y.to(device)\n        pred = model(X)\n        loss = loss_fn(pred, y)\n\n        # Backpropagation\n        loss.backward()\n        optimizer.step()\n        optimizer.zero_grad()\n        train_loss += loss.item()\n    print({\"train_loss\": train_loss / num_batches})\n    wandb.log({\"train_loss\": train_loss / num_batches})\n\n\ndef test_loop(dataloader, model, loss_fn):\n    model.eval()\n    size = len(dataloader.dataset)\n    num_batches = len(dataloader)\n    test_loss, correct = 0, 0\n\n    with torch.no_grad():\n        for X, y in dataloader:\n            pred = model(X.to(device))\n            test_loss += loss_fn(pred, y.to(device)).item()\n            correct += (pred.argmax(1) == y.to(device)).type(torch.float).sum().item() # correct\n\n    test_loss /= num_batches\n    correct /= size\n    wandb.log({\"test_loss\": test_loss, \"accuracy\": correct})\n    print(f\"Test Error: \\n Accuracy: {(100*correct):>0.1f}%, Avg loss: {test_loss:>8f} \\n\")\n\nfor t in range(epochs):\n    print(f\"Epoch {t+1}\\n-------------------------------\")\n    train_loop(train_loader, net, criterion, optimizer)\n    test_loop(valid_loader, net, criterion)","metadata":{"execution":{"iopub.status.busy":"2024-01-11T21:08:36.145671Z","iopub.execute_input":"2024-01-11T21:08:36.145956Z","iopub.status.idle":"2024-01-11T21:16:13.254681Z","shell.execute_reply.started":"2024-01-11T21:08:36.145933Z","shell.execute_reply":"2024-01-11T21:16:13.253352Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"with torch.no_grad():\n    net.eval()\n    test_pred = torch.LongTensor()\n    for i, data in enumerate(test_loader):\n        output = F.softmax(net(data[0].to(device)), dim=1)[..., 1]\n        predicted = output.cpu()\n        test_pred = torch.cat((test_pred, predicted), dim=0)\n    out_df = pd.DataFrame(np.c_[np.arange(1, len(test_data)+1)[:,None], test_pred.numpy()], columns=['id', 'label']).astype({'id':'int'})\n    out_df.to_csv('submission.csv', index=False)\n","metadata":{"execution":{"iopub.status.busy":"2024-01-11T21:16:13.256221Z","iopub.execute_input":"2024-01-11T21:16:13.256604Z","iopub.status.idle":"2024-01-11T21:16:32.697529Z","shell.execute_reply.started":"2024-01-11T21:16:13.256570Z","shell.execute_reply":"2024-01-11T21:16:32.696212Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"import matplotlib.pyplot as plt\nimport numpy\nfrom sklearn import metrics\nvalid_labels = [sample[1] for sample in valid_data]\n\nwith torch.no_grad():\n    net.eval()\n    val_pred = torch.LongTensor()\n    for i, data in enumerate(valid_loader):\n        output = F.softmax(net(data[0].to(device)), dim=1).argmax(1)\n        predicted = output.cpu()\n        val_pred = torch.cat((val_pred, predicted), dim=0)\n\nconfusion_matrix = metrics.confusion_matrix(valid_labels, val_pred)\ncm_display = metrics.ConfusionMatrixDisplay(confusion_matrix = confusion_matrix, display_labels = [\"cat\", \"dog\"])\n\ncm_display.plot()\nplt.show() ","metadata":{"execution":{"iopub.status.busy":"2024-01-11T21:16:32.699279Z","iopub.execute_input":"2024-01-11T21:16:32.699694Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"metrics.accuracy_score(valid_labels, val_pred)","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"","metadata":{},"execution_count":null,"outputs":[]}]}