{"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"pygments_lexer":"ipython3","nbconvert_exporter":"python","version":"3.6.4","file_extension":".py","codemirror_mode":{"name":"ipython","version":3},"name":"python","mimetype":"text/x-python"}},"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":{"_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","execution":{"iopub.status.busy":"2023-01-05T08:27:17.815121Z","iopub.execute_input":"2023-01-05T08:27:17.816191Z","iopub.status.idle":"2023-01-05T08:27:17.846346Z","shell.execute_reply.started":"2023-01-05T08:27:17.816080Z","shell.execute_reply":"2023-01-05T08:27:17.845380Z"},"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\nimport wandb\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":"2023-01-05T08:27:17.849327Z","iopub.execute_input":"2023-01-05T08:27:17.850028Z","iopub.status.idle":"2023-01-05T08:27:22.280783Z","shell.execute_reply.started":"2023-01-05T08:27:17.849993Z","shell.execute_reply":"2023-01-05T08:27:22.279417Z"},"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'))\nprint(f\"Train Data: {len(train_list)}\")\nprint(f\"Test Data: {len(test_list)}\")","metadata":{"execution":{"iopub.status.busy":"2023-01-05T08:27:22.286470Z","iopub.execute_input":"2023-01-05T08:27:22.289132Z","iopub.status.idle":"2023-01-05T08:27:43.985512Z","shell.execute_reply.started":"2023-01-05T08:27:22.289096Z","shell.execute_reply":"2023-01-05T08:27:43.984392Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"labels = [path.split('/')[-1].split('.')[0] for path in train_list]","metadata":{"execution":{"iopub.status.busy":"2023-01-05T08:27:43.988824Z","iopub.execute_input":"2023-01-05T08:27:43.989140Z","iopub.status.idle":"2023-01-05T08:27:44.006511Z","shell.execute_reply.started":"2023-01-05T08:27:43.989112Z","shell.execute_reply":"2023-01-05T08:27:44.005382Z"},"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 enumerate(axes.ravel()):\n    img = Image.open(train_list[idx])\n    ax.set_title(labels[idx])\n    ax.imshow(img)","metadata":{"execution":{"iopub.status.busy":"2023-01-05T08:27:44.008462Z","iopub.execute_input":"2023-01-05T08:27:44.009234Z","iopub.status.idle":"2023-01-05T08:27:45.843107Z","shell.execute_reply.started":"2023-01-05T08:27:44.009184Z","shell.execute_reply":"2023-01-05T08:27:45.841907Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## Use Sklearn to split 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)\nprint(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":"2023-01-05T08:27:45.844148Z","iopub.execute_input":"2023-01-05T08:27:45.845015Z","iopub.status.idle":"2023-01-05T08:27:45.881216Z","shell.execute_reply.started":"2023-01-05T08:27:45.844984Z","shell.execute_reply":"2023-01-05T08:27:45.880087Z"},"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":"train_transforms = transforms.Compose([\n        transforms.Resize(128), # makes it easier for the GPU\n        transforms.RandomResizedCrop(112),\n        transforms.RandomHorizontalFlip(),\n        transforms.ToTensor()])\n\nval_transforms = transforms.Compose([\n        transforms.Resize(128),\n        transforms.CenterCrop(112),\n        transforms.ToTensor()])\n\n\ntest_transforms = transforms.Compose([\n        transforms.Resize(128),\n        transforms.CenterCrop(112),\n        transforms.ToTensor()])","metadata":{"execution":{"iopub.status.busy":"2023-01-05T08:27:45.883005Z","iopub.execute_input":"2023-01-05T08:27:45.883373Z","iopub.status.idle":"2023-01-05T08:27:45.889856Z","shell.execute_reply.started":"2023-01-05T08:27:45.883337Z","shell.execute_reply":"2023-01-05T08:27:45.888781Z"},"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, use_gpu=torch.cuda.is_available(), test=False):\n        self.file_list = file_list\n        self.transform = transform\n        self.filelength = len(file_list)\n        self.use_gpu = use_gpu\n        self.test = test\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        if not self.test:\n            label = 1 if label == \"dog\" else 0\n        else:\n            label = int(label)\n        label = torch.tensor(label)\n        if self.use_gpu:\n            img_transformed = img_transformed.cuda()\n            label = label.cuda()\n        return img_transformed, label","metadata":{"execution":{"iopub.status.busy":"2023-01-05T08:27:45.891248Z","iopub.execute_input":"2023-01-05T08:27:45.892982Z","iopub.status.idle":"2023-01-05T08:27:45.963161Z","shell.execute_reply.started":"2023-01-05T08:27:45.892955Z","shell.execute_reply":"2023-01-05T08:27:45.962038Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"train_data = CatsDogsDataset(train_list, transform=train_transforms)\nvalid_data = CatsDogsDataset(valid_list, transform=val_transforms)\ntest_data = CatsDogsDataset(test_list, transform=test_transforms, test=True)","metadata":{"execution":{"iopub.status.busy":"2023-01-05T08:27:45.965938Z","iopub.execute_input":"2023-01-05T08:27:45.967157Z","iopub.status.idle":"2023-01-05T08:27:45.973319Z","shell.execute_reply.started":"2023-01-05T08:27:45.967084Z","shell.execute_reply":"2023-01-05T08:27:45.972557Z"},"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":"batch_size = 32\ntrain_loader = DataLoader(dataset=train_data, batch_size=batch_size, shuffle=True)\nvalid_loader = DataLoader(dataset=valid_data, batch_size=batch_size, shuffle=False)\ntest_loader = DataLoader(dataset=test_data, batch_size=batch_size, shuffle=False)","metadata":{"execution":{"iopub.status.busy":"2023-01-05T08:27:45.977097Z","iopub.execute_input":"2023-01-05T08:27:45.977707Z","iopub.status.idle":"2023-01-05T08:27:45.987059Z","shell.execute_reply.started":"2023-01-05T08:27:45.977671Z","shell.execute_reply":"2023-01-05T08:27:45.986058Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"class Residual(nn.Module):\n    def __init__(self, input_channels, num_channels, use_1x1conv=False, strides=1):\n        super().__init__()\n        self.conv1 = nn.Conv2d(input_channels, num_channels, kernel_size=3,\n                               padding=1, stride=strides)\n        self.conv2 = nn.Conv2d(num_channels, num_channels, kernel_size=3,\n                               padding=1)\n        if use_1x1conv:\n            self.conv3 = nn.Conv2d(input_channels, num_channels,\n                                   kernel_size=1, stride=strides)\n        else:\n            self.conv3 = None\n        self.bn1 = nn.BatchNorm2d(num_channels)\n        self.bn2 = nn.BatchNorm2d(num_channels)\n\n    def forward(self, X):\n        Y = F.relu(self.bn1(self.conv1(X)))\n        Y = self.bn2(self.conv2(Y))\n        if self.conv3:\n            X = self.conv3(X)\n        Y += X\n        return F.relu(Y)\n    \ndef resnet_block(input_channels, num_channels, num_residuals):\n    blk = []\n    for i in range(num_residuals):\n        if i == 0 and input_channels != num_channels: # 1x1 applied only once per block if needed\n            # Convolution is needed to transform the input of the block to the good number of channels.\n            blk.append(\n                Residual(input_channels, num_channels, use_1x1conv=True,\n                         strides=2))\n        else:\n            blk.append(Residual(num_channels, num_channels))\n    return blk","metadata":{"execution":{"iopub.status.busy":"2023-01-05T08:27:45.988111Z","iopub.execute_input":"2023-01-05T08:27:45.988386Z","iopub.status.idle":"2023-01-05T08:27:46.148098Z","shell.execute_reply.started":"2023-01-05T08:27:45.988361Z","shell.execute_reply":"2023-01-05T08:27:46.146972Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"b1 = nn.Sequential(nn.Conv2d(3, 64, kernel_size=7, stride=2, padding=3),\n                              nn.BatchNorm2d(64), nn.ReLU(),\n                              nn.MaxPool2d(kernel_size=3, stride=2, padding=1))\nb2 = nn.Sequential(*resnet_block(64, 64, 2))\nb3 = nn.Sequential(*resnet_block(64, 128, 2))\nb4 = nn.Sequential(*resnet_block(128, 256, 2))\nb5 = nn.Sequential(*resnet_block(256, 512, 2))\n\nnet = nn.Sequential(b1, b2, b3, b4, b5, nn.AdaptiveAvgPool2d((1, 1)),\n                    nn.Flatten(), nn.Linear(512, 1), nn.Sigmoid())","metadata":{"execution":{"iopub.status.busy":"2023-01-05T08:39:07.202364Z","iopub.execute_input":"2023-01-05T08:39:07.202760Z","iopub.status.idle":"2023-01-05T08:39:07.332310Z","shell.execute_reply.started":"2023-01-05T08:39:07.202700Z","shell.execute_reply":"2023-01-05T08:39:07.331336Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# Hyper parameters\nmomentum = 0.9\nlr = 3e-3\nnum_epochs = 30","metadata":{"execution":{"iopub.status.busy":"2023-01-05T08:27:46.283802Z","iopub.execute_input":"2023-01-05T08:27:46.284301Z","iopub.status.idle":"2023-01-05T08:27:46.290572Z","shell.execute_reply.started":"2023-01-05T08:27:46.284262Z","shell.execute_reply":"2023-01-05T08:27:46.289483Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"loss = nn.BCELoss()\noptimizer = torch.optim.SGD(net.parameters(), lr=lr, momentum=momentum)\n\nwandb.init(project='Dogs_vs_Cats', save_code=True, config={\"Epochs\": num_epochs, \"lr\": lr, \"momentum\": momentum})\nwandb.watch(net, log=\"all\", criterion=loss, log_freq=1000,  log_graph=(True))\n\nif torch.cuda.is_available():\n    device = \"cuda\"\n    net.to(\"cuda\")\n    loss.to(\"cuda\")\nelse:\n    device = \"cpu\"\n    \nfor i in range(num_epochs):\n    net.train()\n    epoch_t_loss = 0\n    epoch_v_loss = 0\n    nb_found = 0\n    nb_found_val = 0\n    \n    for X, y in train_loader:\n        X = X.to(device)\n        y = y.to(device)\n        \n        y_hat = net(X)\n        \n        training_loss = loss(y_hat.reshape(y.shape).float(), y.float())\n        epoch_t_loss += training_loss\n        \n        optimizer.zero_grad()\n        training_loss.backward()\n        optimizer.step()\n        \n        preds = torch.ge(y_hat.flatten(), 0.5)\n        nb_found += len(preds[preds==y])\n        \n    with torch.no_grad():\n        net.eval()\n        for X, y in valid_loader:\n            X = X.to(device)\n            y = y.to(device)\n            \n            y_hat = net(X)\n            validation_loss = loss(y_hat.reshape(y.shape).float(), y.float())\n            epoch_v_loss += validation_loss\n            \n            preds = torch.ge(y_hat.flatten(), 0.5)\n            nb_found_val += len(preds[preds==y])\n    \n    epoch_t_acc = nb_found/len(train_list)\n    epoch_v_acc = nb_found_val/len(valid_list)\n    print(f\"Epoch {i+1},   Training: loss={epoch_t_loss}, accuracy={epoch_t_acc},     Validation: loss={epoch_v_loss}, accuracy={epoch_v_acc}\")\n    \n    wandb.log({\"Validation/Loss\": epoch_v_loss.item(), 'Training/Loss': epoch_t_loss.item(), \"Training/Accuracy\": epoch_t_acc, \"Validation/Accuracy\": epoch_v_acc})\n\ntorch.save(net.state_dict(), \"nn.pt\")\n        ","metadata":{"execution":{"iopub.status.busy":"2023-01-05T08:39:08.707558Z","iopub.execute_input":"2023-01-05T08:39:08.708311Z","iopub.status.idle":"2023-01-05T09:43:41.475436Z","shell.execute_reply.started":"2023-01-05T08:39:08.708274Z","shell.execute_reply":"2023-01-05T09:43:41.474301Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"with torch.no_grad():\n    net.eval()\n    test_pred = torch.LongTensor()\n    result_dict = []\n    for i, data in enumerate(test_loader):\n        x, y = data\n        output = net(x)\n        predicted = output.cpu()\n        predicted = torch.where(predicted > 0.5, torch.ones_like(predicted), torch.zeros_like(predicted))\n        for i in range(len(y)):\n            result_dict.append({'id': y[i].cpu().item(), 'label': predicted[i].item()})\n    out_df = pd.DataFrame(result_dict)\n    out_df.to_csv('submission.csv', index=False)\n    print(out_df)","metadata":{"execution":{"iopub.status.busy":"2023-01-05T10:41:18.328494Z","iopub.execute_input":"2023-01-05T10:41:18.328926Z","iopub.status.idle":"2023-01-05T10:42:13.985372Z","shell.execute_reply.started":"2023-01-05T10:41:18.328882Z","shell.execute_reply":"2023-01-05T10:42:13.984322Z"},"trusted":true},"execution_count":null,"outputs":[]}]}