{"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"name":"python","version":"3.12.12","mimetype":"text/x-python","codemirror_mode":{"name":"ipython","version":3},"pygments_lexer":"ipython3","nbconvert_exporter":"python","file_extension":".py"},"kaggle":{"accelerator":"nvidiaTeslaT4","dataSources":[{"sourceType":"competition","sourceId":3649,"databundleVersionId":46718}],"dockerImageVersionId":31328,"isInternetEnabled":true,"language":"python","sourceType":"notebook","isGpuEnabled":true}},"nbformat_minor":4,"nbformat":4,"cells":[{"cell_type":"code","source":"import numpy as np \nimport pandas as pd \nimport os\n","metadata":{"_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","trusted":true,"execution":{"iopub.status.busy":"2026-04-19T15:57:07.729231Z","iopub.execute_input":"2026-04-19T15:57:07.729498Z","iopub.status.idle":"2026-04-19T15:57:08.81265Z","shell.execute_reply.started":"2026-04-19T15:57:07.729461Z","shell.execute_reply":"2026-04-19T15:57:08.81191Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# Visualization\nimport matplotlib.pyplot as plt \nimport seaborn as sns\n# Deep Learning\nimport torch\nimport torch.nn as nn\nimport torch.nn.functional as F\nfrom torch.utils.data import DataLoader, Dataset, random_split\nimport torchvision.transforms as T\n#Transfer Learning\nfrom torchvision.models import resnet18, resnet34, resnet50\n#Loading Image\nfrom PIL import Image","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-04-19T15:57:08.814694Z","iopub.execute_input":"2026-04-19T15:57:08.815299Z","iopub.status.idle":"2026-04-19T15:57:18.837781Z","shell.execute_reply.started":"2026-04-19T15:57:08.815263Z","shell.execute_reply":"2026-04-19T15:57:18.837199Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"device = (\"cuda\" if torch.cuda.is_available() else \"cpu\")\nprint(device)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-04-19T15:57:18.838583Z","iopub.execute_input":"2026-04-19T15:57:18.839063Z","iopub.status.idle":"2026-04-19T15:57:19.102562Z","shell.execute_reply.started":"2026-04-19T15:57:18.839021Z","shell.execute_reply":"2026-04-19T15:57:19.101833Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"!pip install py7zr","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-04-19T15:57:19.104016Z","iopub.execute_input":"2026-04-19T15:57:19.104383Z","iopub.status.idle":"2026-04-19T15:57:24.770905Z","shell.execute_reply.started":"2026-04-19T15:57:19.104342Z","shell.execute_reply":"2026-04-19T15:57:24.769898Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"import py7zr\n\npath = \"/kaggle/input/competitions/cifar-10\"\nsample_sub = pd.read_csv(os.path.join(path, \"sampleSubmission.csv\"))\ntrain_labels = pd.read_csv(os.path.join(path, \"trainLabels.csv\"))\n\nwith py7zr.SevenZipFile(os.path.join(path, \"train.7z\" ), mode='r') as archive:\n    archive.extractall(path='/kaggle/working')\n    \nwith py7zr.SevenZipFile(os.path.join(path, \"test.7z\" ), mode='r') as archive:\n    archive.extractall(path='/kaggle/working')","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-04-19T15:57:24.772393Z","iopub.execute_input":"2026-04-19T15:57:24.772753Z","iopub.status.idle":"2026-04-19T16:00:02.481124Z","shell.execute_reply.started":"2026-04-19T15:57:24.772713Z","shell.execute_reply":"2026-04-19T16:00:02.480224Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"# Prerpocessing Pipeline\n\n![image.png](attachment:4db2809a-5f79-41ed-b020-ad21fb2b558c.png)","metadata":{},"attachments":{"4db2809a-5f79-41ed-b020-ad21fb2b558c.png":{"image/png":"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"}}},{"cell_type":"code","source":"##Visualizing some images from the train set with labels\ntrain_path = \"/kaggle/working/train\"\n\nfig, axes = plt.subplots(5, 5, figsize=(25, 25))\naxes = axes.flatten()\n\nimage_files = os.listdir(train_path)\n\nfor i, img_name in enumerate(image_files[:25]):\n    img_path = os.path.join(train_path, img_name)\n    match = train_labels[train_labels[\"id\"] == int(img_name.split(\".\")[0])][\"label\"].values[0] \n    img = Image.open(img_path).convert(\"RGB\")\n\n    axes[i].imshow(img)\n    axes[i].set_title(match)\n    axes[i].axis(\"off\")\n\nplt.tight_layout()\nplt.show()","metadata":{"trusted":true,"_kg_hide-output":true,"execution":{"iopub.status.busy":"2026-04-19T16:00:02.48238Z","iopub.execute_input":"2026-04-19T16:00:02.482877Z","iopub.status.idle":"2026-04-19T16:00:03.874444Z","shell.execute_reply.started":"2026-04-19T16:00:02.482849Z","shell.execute_reply":"2026-04-19T16:00:03.873706Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"train_labels[\"label\"].value_counts(normalize = True)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-04-19T16:00:03.876636Z","iopub.execute_input":"2026-04-19T16:00:03.876927Z","iopub.status.idle":"2026-04-19T16:00:03.894762Z","shell.execute_reply.started":"2026-04-19T16:00:03.876904Z","shell.execute_reply":"2026-04-19T16:00:03.894042Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"##Visualizing some images from the test set\ntest_path = \"/kaggle/working/test\"\n\nfig, axes = plt.subplots(5, 5, figsize=(25, 25))\naxes = axes.flatten()\n\nimage_files = os.listdir(test_path)\n\nfor i, img_name in enumerate(image_files[:25]):\n    img_path = os.path.join(test_path, img_name)\n    img = Image.open(img_path).convert(\"RGB\")\n    axes[i].imshow(img)\n    axes[i].axis(\"off\")\n\nplt.tight_layout()\nplt.show()","metadata":{"trusted":true,"scrolled":true,"execution":{"iopub.status.busy":"2026-04-19T16:00:03.895636Z","iopub.execute_input":"2026-04-19T16:00:03.896002Z","iopub.status.idle":"2026-04-19T16:00:04.677812Z","shell.execute_reply.started":"2026-04-19T16:00:03.895977Z","shell.execute_reply":"2026-04-19T16:00:04.677206Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"train_labels","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-04-19T16:00:04.678813Z","iopub.execute_input":"2026-04-19T16:00:04.679159Z","iopub.status.idle":"2026-04-19T16:00:04.698602Z","shell.execute_reply.started":"2026-04-19T16:00:04.679103Z","shell.execute_reply":"2026-04-19T16:00:04.697713Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"from sklearn.preprocessing import LabelEncoder\nencoder =LabelEncoder().fit(train_labels[\"label\"])\ntrain_labels[\"label\"] = encoder.transform(train_labels[\"label\"])","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-04-19T16:00:04.699427Z","iopub.execute_input":"2026-04-19T16:00:04.699724Z","iopub.status.idle":"2026-04-19T16:00:04.770892Z","shell.execute_reply.started":"2026-04-19T16:00:04.699699Z","shell.execute_reply":"2026-04-19T16:00:04.770379Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"len(train_labels)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-04-19T16:00:04.771756Z","iopub.execute_input":"2026-04-19T16:00:04.772218Z","iopub.status.idle":"2026-04-19T16:00:04.776649Z","shell.execute_reply.started":"2026-04-19T16:00:04.772194Z","shell.execute_reply":"2026-04-19T16:00:04.775893Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"# The Dataset\nA pytorch dataset is a special type of class that navigates pytorch through our data\n\nA custom pytorch dataset has two essential methods:\n1. > __len__ method (tells pytorch the total number of examples in our dataset)\n2. > __getitem__ method (navigates pytorch to our dataset)\n   \nReasons of Inhereting the Dataset attribute from Pytorch for making a Custom Dataset:\n1. Works with DataLoaders Automatically\n2. Can be used for shuffling, parrallel loading and batching your dataset with no pain\n3. Loads data faster\n","metadata":{}},{"cell_type":"code","source":"path = \"/kaggle/working/train\"\ntest_path  = os.path.join(\"/kaggle/working\", \"test\")\nimg_name = \"1.jpg\"\nimg = os.path.join(path, '12653.png')\nImage.open(img).convert(\"RGB\")","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-04-19T16:00:04.778244Z","iopub.execute_input":"2026-04-19T16:00:04.778525Z","iopub.status.idle":"2026-04-19T16:00:04.793606Z","shell.execute_reply.started":"2026-04-19T16:00:04.778502Z","shell.execute_reply":"2026-04-19T16:00:04.792918Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"class df(Dataset):\n    def __init__(self, path, df=None, transform=None, is_test=False):\n        self.path = path\n        self.df = df\n        self.transform =transform\n        self.is_test = is_test\n        self.images = (os.listdir(self.path))\n\n        if self.df is not None:\n            self.df = self.df.reset_index(drop=True)\n\n    def __len__(self):\n        return len(self.images)\n\n    def __getitem__(self, i):\n        img_name = self.images[i]\n        img_path = os.path.join(self.path, img_name)\n\n        image = Image.open(img_path).convert(\"RGB\")\n\n        if self.transform:\n            image = self.transform(image)\n\n        if self.is_test:\n            return image\n\n        name = os.path.splitext(img_name)[0]\n\n        label= self.df[self.df[\"id\"] == int(name)][\"label\"].values[0]\n\n        if label is None:\n            print(f\"[WARNING] Missing label for: {name}\")\n            return None\n\n        label = torch.tensor(label, dtype=torch.long)\n\n        return image, label","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-04-19T16:00:04.794486Z","iopub.execute_input":"2026-04-19T16:00:04.794974Z","iopub.status.idle":"2026-04-19T16:00:04.800946Z","shell.execute_reply.started":"2026-04-19T16:00:04.794907Z","shell.execute_reply":"2026-04-19T16:00:04.800433Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"# AUGMENTATION","metadata":{}},{"cell_type":"code","source":"import torchvision.transforms as T","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-04-19T16:00:04.801793Z","iopub.execute_input":"2026-04-19T16:00:04.802122Z","iopub.status.idle":"2026-04-19T16:00:04.8137Z","shell.execute_reply.started":"2026-04-19T16:00:04.802088Z","shell.execute_reply":"2026-04-19T16:00:04.812846Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"train_path = \"/kaggle/working/train\"\ntest_path = \"/kaggle/working/test\"\ntrain_trans =  T.Compose([\n            T.Resize((224, 224)),\n            T.RandomHorizontalFlip(p = 0.5),\n            T.ToTensor(),\n            T.Normalize(mean = [0.485, 0.456, 0.406], std = [0.229, 0.224, 0.225])\n        ])\ntest_trans =  T.Compose([\n            T.Resize((224, 224)),\n            T.ToTensor(),\n            T.Normalize(mean = [0.485, 0.456, 0.406], std = [0.229, 0.224, 0.225])\n        ])\nfull_df= df(train_path, train_labels, train_trans, is_test = False)\ntest_df = df(test_path, None, test_trans, is_test = True)\n\ng = torch.Generator(device=\"cpu\")\ng.manual_seed(42)\n\ntrain_size = int(0.8 * len(full_df))\nval_size = len(full_df) - train_size\n\ntrain_df, val_df = random_split(full_df, [train_size, val_size], generator=g)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-04-19T16:03:20.99276Z","iopub.execute_input":"2026-04-19T16:03:20.993518Z","iopub.status.idle":"2026-04-19T16:03:21.194396Z","shell.execute_reply.started":"2026-04-19T16:03:20.993484Z","shell.execute_reply":"2026-04-19T16:03:21.193412Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"# Pytorch DataLoader\n\nDataLaoders are tools for loading and working with datas in pytorch effiecienly. You can still do most of the pytorch stuffs without using a pytorch dataloader . Then Why use a pytorch dataLoader. \n1. Loading massive datasets effieciently\n2. Batching the dataset efficiently (esp for mini-batch learning techniques)\n3. Shuffling and preocssing datas\n4. Converting datas into gpus\n\nFundemental Parameters of a Pytorch DataLoader:\n1. Dataset --> The pytorch dataframe to load data from (the dataframes we created earlier)\n2. batch_size --> int tha specify the toatal number of batches to split the data on. (typically 2^x where x is a int from starting from 1)\n3. shuffle --> bool. Shuffle to remove ordered patters, generalize better and reduce the chance of overfitting. btw don't shuffle time-series data cause sequence is important there and also don't shuffle test and val/dev datasets\n4. number of workers(num_workers) --> int. number of subprocess to load data in parallel. helps loading data using cpu parallely while training using the gpu.\n5. pin_memory --> bool. it makes the data transfer from cpu to gpu faster. only can use it when using gpu for training\n","metadata":{}},{"cell_type":"code","source":"full_dl = DataLoader(full_df, batch_size =32, shuffle = True, num_workers = 4, pin_memory= True)\ntrain_dl = DataLoader(train_df, batch_size = 32, shuffle =True, num_workers = 4 , pin_memory = True)\nval_dl = DataLoader(val_df, batch_size = 32, shuffle =False)\ntest_dl = DataLoader(test_df, batch_size = 32, shuffle =False)\n\n\nfor batch in test_dl:\n    x= batch\n    print(x.shape)\n    break","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-04-19T16:03:22.461009Z","iopub.execute_input":"2026-04-19T16:03:22.461958Z","iopub.status.idle":"2026-04-19T16:03:22.58174Z","shell.execute_reply.started":"2026-04-19T16:03:22.461922Z","shell.execute_reply":"2026-04-19T16:03:22.580857Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"# MODELING\nThe model will consist of convolutional and linear layers","metadata":{}},{"cell_type":"code","source":"# model class\n\nclass MODEL(nn.Module):\n    def __init__(self):\n        super().__init__()\n        self.model = nn.Sequential(\n            nn.Conv2d(3, 32, kernel_size=3, padding=1),\n            nn.ReLU(),\n            nn.MaxPool2d(2),                          \n            nn.Conv2d(32, 64, kernel_size=3, padding=1),\n            nn.ReLU(),\n            nn.MaxPool2d(2),     \n            nn.Conv2d(64, 128, kernel_size=3, padding=1),\n            nn.ReLU(),\n            nn.AdaptiveAvgPool2d(1),                          \n            nn.Flatten(),\n            nn.Linear(128, 256),\n            nn.ReLU(),\n            nn.Dropout(0.5),\n            nn.Linear(256, 10)                          \n        )\n\n    def forward(self, x):\n        return self.model(x)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-04-19T16:03:24.357635Z","iopub.execute_input":"2026-04-19T16:03:24.358261Z","iopub.status.idle":"2026-04-19T16:03:24.363852Z","shell.execute_reply.started":"2026-04-19T16:03:24.35823Z","shell.execute_reply":"2026-04-19T16:03:24.36311Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"model = MODEL().to(device)\nmodel = nn.DataParallel(model)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-04-19T16:03:52.750823Z","iopub.execute_input":"2026-04-19T16:03:52.751342Z","iopub.status.idle":"2026-04-19T16:03:52.77228Z","shell.execute_reply.started":"2026-04-19T16:03:52.751309Z","shell.execute_reply":"2026-04-19T16:03:52.77147Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"model","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-04-19T16:03:55.397738Z","iopub.execute_input":"2026-04-19T16:03:55.398302Z","iopub.status.idle":"2026-04-19T16:03:55.40405Z","shell.execute_reply.started":"2026-04-19T16:03:55.398271Z","shell.execute_reply":"2026-04-19T16:03:55.403215Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"#getting the total parameters\nparams = sum(p.numel() for p in model.parameters() if p.requires_grad)\nprint(params)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-04-19T16:05:27.339937Z","iopub.execute_input":"2026-04-19T16:05:27.340706Z","iopub.status.idle":"2026-04-19T16:05:27.345653Z","shell.execute_reply.started":"2026-04-19T16:05:27.340671Z","shell.execute_reply":"2026-04-19T16:05:27.344676Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"# training the model\n\nThe training will have two steps and will consist of training and validation\nThe steps are:\n1. Forward Step\n2. Backward Step\n\nIn the Evaluation we'll evaluate the model using:\n1. Accuracy Score\n2. Precision\n3. Recall\n4. F1-SCORE","metadata":{}},{"cell_type":"code","source":"## EVALUATION METRICS\nfrom sklearn.metrics import precision_score, recall_score, f1_score","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-04-19T12:03:18.721976Z","iopub.execute_input":"2026-04-19T12:03:18.722192Z","iopub.status.idle":"2026-04-19T12:03:18.72587Z","shell.execute_reply.started":"2026-04-19T12:03:18.722171Z","shell.execute_reply":"2026-04-19T12:03:18.725155Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"EPOCHS = 5\noptimizer = torch.optim.Adam(model.parameters(), lr=1e-3)\nloss_fn = nn.CrossEntropyLoss()\n\nfor epoch in range(EPOCHS):\n    model.train()\n    train_loss = 0\n    correct = 0\n    total = 0\n\n    for x, y in train_dl:\n        x, y = x.to(device), y.to(device)\n\n        optimizer.zero_grad()\n        out = model(x)\n        loss = loss_fn(out, y)\n        loss.backward()\n        optimizer.step()\n\n        train_loss += loss.item()\n\n        pred = out.argmax(dim=1)\n        correct += (pred == y).sum().item()\n        total += y.size(0)\n\n    train_acc = correct / total\n    model.eval()\n    val_loss = 0\n    correct = 0\n    total = 0\n\n    all_preds = []\n    all_labels = []\n\n    with torch.no_grad():\n        for x, y in val_dl:\n            x, y = x.to(device), y.to(device)\n\n            out = model(x)\n            loss = loss_fn(out, y)\n\n            val_loss += loss.item()\n\n            pred = out.argmax(dim=1)\n            correct += (pred == y).sum().item()\n            total += y.size(0)\n\n            all_preds.append(pred.cpu())\n            all_labels.append(y.cpu())\n\n    val_acc = correct / total\n\n    all_preds = torch.cat(all_preds)\n    all_labels = torch.cat(all_labels)\n\n    val_prec = precision_score(all_labels, all_preds, average='macro')\n    val_rec  = recall_score(all_labels, all_preds, average='macro')\n    val_f1   = f1_score(all_labels, all_preds, average='macro')\n    print(\n        f\"Epoch {epoch+1}/{EPOCHS} | \"\n        f\"train_loss: {train_loss/len(train_dl):.4f}  train_acc: {train_acc:.4f} | \"\n        f\"val_loss: {val_loss/len(val_dl):.4f}  val_acc: {val_acc:.4f} | \"\n        f\"val_prec: {val_prec:.4f}  val_rec: {val_rec:.4f}  val_f1: {val_f1:.4f}\"\n    )","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-04-19T12:03:18.88442Z","iopub.execute_input":"2026-04-19T12:03:18.884645Z","iopub.status.idle":"2026-04-19T12:11:16.778938Z","shell.execute_reply.started":"2026-04-19T12:03:18.884623Z","shell.execute_reply":"2026-04-19T12:11:16.777868Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"import matplotlib.pyplot as plt\nimport random\n\nmodel.eval()\n\nall_imgs, all_labels, all_preds = [], [], []\n\nwith torch.no_grad():\n    for x, y in val_dl:\n        x, y = x.to(device), y.to(device)\n        out  = model(x)\n        pred = out.argmax(dim=1)\n\n        all_imgs.append(x.cpu())\n        all_labels.append(y.cpu())\n        all_preds.append(pred.cpu())\n\n        if sum(len(b) for b in all_imgs) >= 20:\n            break\n\nall_imgs   = torch.cat(all_imgs)\nall_labels = torch.cat(all_labels)\nall_preds  = torch.cat(all_preds)\n\nindices = random.sample(range(len(all_imgs)), 20)\n\nfig, axes = plt.subplots(4, 5, figsize=(15, 12))\naxes = axes.flatten()\n\nfor i, idx in enumerate(indices):\n    img = all_imgs[idx]\n    img = img.permute(1, 2, 0)  # (C, H, W) → (H, W, C)\n    img = img*torch.tensor([0.485, 0.456, 0.406]) + torch.tensor( [0.229, 0.224, 0.225])\n    img = img.clamp(0, 1)\n    axes[i].imshow(img)\n    actual    = all_labels[idx].item()\n    predicted = all_preds[idx].item()\n    axes[i].set_title(\n        f\"Actual: {actual}\\nPred: {predicted}\",\n        color=\"green\" if actual == predicted else \"red\",\n        fontsize=10\n    )\n    axes[i].axis(\"off\")\n\nplt.suptitle(\"Validation Samples — Green: Correct  |  Red: Wrong\", fontsize=13, y=1.01)\nplt.tight_layout()\nplt.show()","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-04-19T12:11:32.580872Z","iopub.execute_input":"2026-04-19T12:11:32.581184Z","iopub.status.idle":"2026-04-19T12:11:33.870329Z","shell.execute_reply.started":"2026-04-19T12:11:32.581152Z","shell.execute_reply":"2026-04-19T12:11:33.869408Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"# making predictions in the test set","metadata":{}},{"cell_type":"code","source":"test_preds = []\n\nmodel.eval()\nwith torch.no_grad():\n    prds = []\n    for batch in test_dl:\n        x = batch\n        x = x.to(device)\n\n        out = model(x)\n        pred = out.argmax(dim=1)\n\n        prds.append(pred.cpu())\n\ntest_preds = torch.cat(prds, dim=0)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-04-19T12:15:37.070165Z","iopub.execute_input":"2026-04-19T12:15:37.070917Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"sample_sub= pd.read_csv(os.path.join(path, \"sampleSubmission.csv\"))\nsample_sub[\"label\"] = test_preds\nsample_sub.to_csv(\"sub1.csv\",index = False)","metadata":{"trusted":true},"outputs":[],"execution_count":null},{"cell_type":"code","source":"sample_sub","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-04-19T16:15:55.984761Z","iopub.execute_input":"2026-04-19T16:15:55.985235Z","iopub.status.idle":"2026-04-19T16:15:55.994257Z","shell.execute_reply.started":"2026-04-19T16:15:55.985203Z","shell.execute_reply":"2026-04-19T16:15:55.993433Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"","metadata":{"trusted":true},"outputs":[],"execution_count":null}]}