{"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"name":"python","version":"3.10.14","mimetype":"text/x-python","codemirror_mode":{"name":"ipython","version":3},"pygments_lexer":"ipython3","nbconvert_exporter":"python","file_extension":".py"},"kaggle":{"accelerator":"gpu","dataSources":[{"sourceId":3649,"databundleVersionId":46718,"sourceType":"competition"}],"dockerImageVersionId":30786,"isInternetEnabled":true,"language":"python","sourceType":"notebook","isGpuEnabled":true}},"nbformat_minor":4,"nbformat":4,"cells":[{"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","trusted":true,"execution":{"iopub.status.busy":"2024-11-04T18:19:32.046261Z","iopub.execute_input":"2024-11-04T18:19:32.047024Z","iopub.status.idle":"2024-11-04T18:19:32.080564Z","shell.execute_reply.started":"2024-11-04T18:19:32.046982Z","shell.execute_reply":"2024-11-04T18:19:32.079591Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"import torch\nimport torch.nn as nn\nimport torch.optim as optim\nimport torch.nn.functional as F\nimport torch.backends.cudnn as cudnn\nfrom torch.utils.data import DataLoader\n\nimport torchvision\nimport torchvision.transforms as transforms\n\n!pip install py7zr\nimport py7zr\nfrom PIL import Image\nfrom io import BytesIO\nfrom tqdm import tqdm\nfrom copy import deepcopy\nimport os\n\nimport matplotlib.pyplot as plt\nimport pandas as pd","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-11-04T18:19:32.082144Z","iopub.execute_input":"2024-11-04T18:19:32.082496Z","iopub.status.idle":"2024-11-04T18:19:43.946095Z","shell.execute_reply.started":"2024-11-04T18:19:32.082462Z","shell.execute_reply":"2024-11-04T18:19:43.944909Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')\ndevice","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-11-04T18:19:43.948234Z","iopub.execute_input":"2024-11-04T18:19:43.948681Z","iopub.status.idle":"2024-11-04T18:19:43.957793Z","shell.execute_reply.started":"2024-11-04T18:19:43.948629Z","shell.execute_reply":"2024-11-04T18:19:43.956943Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"transform_train = transforms.Compose([\n    transforms.Resize(64),\n    transforms.RandomAffine(degrees=10, translate=(0.1, 0.1), scale=(0.95, 1.05)),\n    transforms.RandomHorizontalFlip(0.1),\n    transforms.ToTensor(),\n    transforms.Normalize((0.5,0.5,0.5),(0.5,0.5,0.5)),\n])\n\ntransform_test = transforms.Compose([\n    transforms.Resize(64),\n    transforms.ToTensor(),\n    transforms.Normalize((0.5,0.5,0.5),(0.5,0.5,0.5)),\n])\n\n\ntrain_dataset = torchvision.datasets.CIFAR10(root='./data', train=True, download=True, transform=transform_train)\ntest_dataset = torchvision.datasets.CIFAR10(root='./data', train=False, download=True, transform=transform_test)\n\n\n\ntrain_dataset, valid_dataset = torch.utils.data.random_split(train_dataset, [45000, 5000])\n\ntrain_loader = torch.utils.data.DataLoader(train_dataset, batch_size=64, shuffle=True)\nvalid_loader = torch.utils.data.DataLoader(valid_dataset, batch_size=64, shuffle=True)\ntest_loader = torch.utils.data.DataLoader(test_dataset, batch_size=64, shuffle=False)\n\n\nclasses = ['airplane', 'automobile', 'bird', 'cat', 'deer',\n           'dog', 'frog', 'horse', 'ship', 'truck']","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-11-04T18:19:43.959122Z","iopub.execute_input":"2024-11-04T18:19:43.959507Z","iopub.status.idle":"2024-11-04T18:19:45.582105Z","shell.execute_reply.started":"2024-11-04T18:19:43.959474Z","shell.execute_reply":"2024-11-04T18:19:45.581111Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"for images, labels in train_loader:\n    fig, ax = plt.subplots(figsize=(12, 6))\n    ax.set_xticks([])\n    ax.set_yticks([])\n    ax.imshow(torchvision.utils.make_grid(images, nrow=20).permute(1, 2, 0))\n    break\n","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-11-04T18:19:45.584417Z","iopub.execute_input":"2024-11-04T18:19:45.584732Z","iopub.status.idle":"2024-11-04T18:19:45.921519Z","shell.execute_reply.started":"2024-11-04T18:19:45.584700Z","shell.execute_reply":"2024-11-04T18:19:45.920556Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"for i in range(0, 12):\n    img, label = train_dataset[i]\n    plt.subplot(3,4,i+1)\n    plt.title(classes[label])\n    plt.axis('off')\n    plt.imshow(img.permute(1, 2, 0))","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-11-04T18:19:45.922815Z","iopub.execute_input":"2024-11-04T18:19:45.923167Z","iopub.status.idle":"2024-11-04T18:19:46.750239Z","shell.execute_reply.started":"2024-11-04T18:19:45.923122Z","shell.execute_reply":"2024-11-04T18:19:46.749348Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"class Cifar10Classifier(nn.Module):\n    def __init__(self):\n        super().__init__()\n        self.network = nn.Sequential(\n            nn.Conv2d(in_channels=3, out_channels=16, kernel_size=3, padding=1, stride=1),\n            nn.ReLU(),\n            nn.Conv2d(in_channels=16, out_channels=32, kernel_size=3, padding=1, stride=1),\n            nn.ReLU(),\n            nn.MaxPool2d(2, 2),\n            nn.Dropout2d(p=0.3),\n\n            nn.Conv2d(in_channels=32, out_channels=64, kernel_size=3, padding=1, stride=1),\n            nn.ReLU(),\n            nn.Conv2d(in_channels=64, out_channels=128, kernel_size=3, padding=1, stride=1),\n            nn.ReLU(),\n            nn.MaxPool2d(2, 2),\n            nn.Dropout2d(p=0.3),\n\n            nn.Conv2d(in_channels=128, out_channels=256, kernel_size=3, padding=1, stride=1),\n            nn.ReLU(),\n            nn.Conv2d(in_channels=256, out_channels=256, kernel_size=3, padding=1, stride=1),\n            nn.ReLU(),\n            nn.MaxPool2d(2, 2),\n            nn.Dropout2d(p=0.3),\n\n            # (W - F + 2P)/S + 1\n            nn.Flatten(),\n            nn.Linear(256 * 8 * 8, 1024),\n            nn.ReLU(),\n            nn.Dropout(),\n            nn.Linear(1024, 512),\n            nn.ReLU(),\n            nn.Dropout(),\n            nn.Linear(512, 10)\n        )\n\n    def accuracy(self, outputs, labels):\n        _, preds = torch.max(outputs, dim=1)\n        return torch.tensor(torch.sum(preds == labels).item() / len(preds))\n\n    def forward(self, x):\n        return self.network(x)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-11-04T18:19:46.751518Z","iopub.execute_input":"2024-11-04T18:19:46.751889Z","iopub.status.idle":"2024-11-04T18:19:46.764740Z","shell.execute_reply.started":"2024-11-04T18:19:46.751844Z","shell.execute_reply":"2024-11-04T18:19:46.763913Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"@torch.no_grad()\ndef evaluate_model(model, validation_loader):\n    model.eval()\n    outputs = []\n    for batch in validation_loader:\n        images, labels = batch\n        images = images.to(device)\n        labels = labels.to(device)\n        out = model(images)\n        loss = F.cross_entropy(out, labels)\n        accuracy = model.accuracy(out, labels)\n        outputs.append({\"loss\": loss, \"accuracy\": accuracy})\n        \n\n    batchLosses = [row[\"loss\"] for row in outputs]\n    epochLosses = torch.stack(batchLosses).mean()\n    batchAcc = [row[\"accuracy\"] for row in outputs]\n    epochAcc = torch.stack(batchAcc).mean()\n    return {\"loss\": epochLosses.item(), \"accuracy\": epochAcc.item()}","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-11-04T18:19:46.766056Z","iopub.execute_input":"2024-11-04T18:19:46.766400Z","iopub.status.idle":"2024-11-04T18:19:46.778798Z","shell.execute_reply.started":"2024-11-04T18:19:46.766321Z","shell.execute_reply":"2024-11-04T18:19:46.778074Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"def train_model(epochs, lr, model, train_loader, validation_loader, optimization_func=torch.optim.Adam):\n    optimizer = optimization_func(model.parameters(), lr)\n\n    for epoch in range(epochs):\n        print(f\"Epoch: {epoch}\")\n        model.train()\n        trainingLosses = []\n\n        for batch in train_loader:\n            images, labels = batch\n            images = images.to(device)\n            labels = labels.to(device)\n            out = model(images)\n            loss = F.cross_entropy(out, labels)\n        \n            trainingLosses.append(loss)\n            loss.backward()\n            optimizer.step()\n            optimizer.zero_grad()\n\n        print(f\"{evaluate_model(model, validation_loader)}\\n\")","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-11-04T18:22:07.069117Z","iopub.execute_input":"2024-11-04T18:22:07.070098Z","iopub.status.idle":"2024-11-04T18:22:07.080120Z","shell.execute_reply.started":"2024-11-04T18:22:07.070045Z","shell.execute_reply":"2024-11-04T18:22:07.078961Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"model = Cifar10Classifier()\nmodel.to(device)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-11-04T18:22:08.203570Z","iopub.execute_input":"2024-11-04T18:22:08.203961Z","iopub.status.idle":"2024-11-04T18:22:08.387832Z","shell.execute_reply.started":"2024-11-04T18:22:08.203923Z","shell.execute_reply":"2024-11-04T18:22:08.386889Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"train_model(60, 0.001, model, train_loader=train_loader, validation_loader=valid_loader, optimization_func=torch.optim.Adam)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-11-04T18:22:08.492383Z","iopub.execute_input":"2024-11-04T18:22:08.493501Z","iopub.status.idle":"2024-11-04T19:03:08.322404Z","shell.execute_reply.started":"2024-11-04T18:22:08.493443Z","shell.execute_reply":"2024-11-04T19:03:08.321336Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"evaluate_model(model, valid_loader)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-11-04T19:03:08.324283Z","iopub.execute_input":"2024-11-04T19:03:08.324696Z","iopub.status.idle":"2024-11-04T19:03:11.606482Z","shell.execute_reply.started":"2024-11-04T19:03:08.324662Z","shell.execute_reply":"2024-11-04T19:03:11.605566Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"evaluate_model(model, test_loader)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-11-04T19:03:11.607575Z","iopub.execute_input":"2024-11-04T19:03:11.607852Z","iopub.status.idle":"2024-11-04T19:03:16.427584Z","shell.execute_reply.started":"2024-11-04T19:03:11.607822Z","shell.execute_reply":"2024-11-04T19:03:16.426690Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"","metadata":{"trusted":true},"outputs":[],"execution_count":null}]}