{"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"name":"python","version":"3.11.11","mimetype":"text/x-python","codemirror_mode":{"name":"ipython","version":3},"pygments_lexer":"ipython3","nbconvert_exporter":"python","file_extension":".py"},"kaggle":{"accelerator":"nvidiaTeslaT4","dataSources":[{"sourceId":101330,"databundleVersionId":12348894,"sourceType":"competition"},{"sourceId":4684556,"sourceType":"datasetVersion","datasetId":2713968}],"dockerImageVersionId":31041,"isInternetEnabled":false,"language":"python","sourceType":"notebook","isGpuEnabled":true}},"nbformat_minor":4,"nbformat":4,"cells":[{"cell_type":"markdown","source":"Welcome to the ACM AI MiniHack Challenge! To help get you started, here's a notebook that helps you load the dataset, import useful libraries, and eventually submit your predictions!\n\nIn this challenge, you are tasked with building a model that can classify Minecraft biomes. The training dataset features 2416 images from 10 different biomes.\n\nLet's start by importing some helpful libraries. ","metadata":{}},{"cell_type":"code","source":"import numpy as np \nimport pandas as pd \nfrom PIL import Image\nfrom torch.utils.data import Dataset\nfrom torchvision import transforms\nimport torchvision.models as models\nimport torch.nn as nn\nimport torch.optim as optim\nimport torch.nn.functional as F\nimport os\n\n\n\n\n","metadata":{"_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","trusted":true,"execution":{"iopub.status.busy":"2025-05-17T23:39:29.242034Z","iopub.execute_input":"2025-05-17T23:39:29.242197Z","iopub.status.idle":"2025-05-17T23:39:32.264024Z","shell.execute_reply.started":"2025-05-17T23:39:29.242183Z","shell.execute_reply":"2025-05-17T23:39:32.263224Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"The ImageCSV Dataset is a custom PyTorch Dataset that takes the train.csv file and path to the training image folder, which can be found under the ACM AI Mini ML Challenge folder. Additionally, we pass in the a transformation to normalize the dataset, a label map, and a test boolean to indicate whether the test set is being loaded.\n\nEach row in the train.csv file contains the relative path to the image and its corresponding label, as well as an image ID. For the training data, all we care about is the relative path and the corresponding label.\n\nThe test.csv file, meanwhile, contains images IDs and the relative path to these images. In your submission, you are expected to predict the label for each image ID, and turn in a csv file with a column of ID numbers and a column of predictions.\n\nAdditionally, we define a label map to map the label strings to integers.\n\nThe __getitem__ method loads an image with the appropriate transformations and its corresponding label for a given index for the training dataset. For the test dataset, it returns the image and the corresponding image id. \n\n\nTake a moment to inspect train.csv and test.csv, which can be found under the \"Input\" tab of the notebook. \n\nNOTE: DO NOT CHANGE ANY OF THE TRANSFORMATIONS APPLIED TO THE DATASET. YOU MUST TRAIN AND TEST ON BLACK AND WHITE IMAGES. \n","metadata":{},"attachments":{"5a6dc9ef-b104-4048-9968-77cefdf22dfb.png":{"image/png":"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"},"2de756a5-b7d8-4edd-b71a-70e9577507a3.png":{"image/png":"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"}}},{"cell_type":"code","source":"class ImageCSVDataset(Dataset):\n    def __init__(self, csv_file, root_dir, transform=None, label_map=None, is_test=False):\n        self.df = pd.read_csv(csv_file)\n        self.root_dir = root_dir  \n        self.transform = transform\n        self.label_map = label_map\n        self.is_test = is_test\n\n    def __len__(self):\n        return len(self.df)\n\n    def __getitem__(self, idx):\n        row = self.df.iloc[idx]\n        rel_path = self.df.iloc[idx]['path']  \n        img_path = os.path.join(self.root_dir, rel_path)\n        \n        image = Image.open(img_path).convert('RGB')\n        \n        \n        if self.transform:\n            image = self.transform(image)\n\n        if self.is_test:\n            image_id = int(row['ID'])\n            print(image_id)\n            return image, image_id  # useful for mapping back to submission\n        else:\n            label = self.label_map.get(row.get(\"biome_type\", -1), -1)\n            return image, label if label != -1 else row['path']\n\n\n\ntrain_df = pd.read_csv(\"/kaggle/input/acm-ai-hack/train.csv\")\n\n\ntransform = transforms.Compose([\n    transforms.Resize((224, 224)),\n    transforms.ToTensor(),\n    transforms.Normalize(mean=[0.485, 0.456, 0.406],\n                         std=[0.229, 0.224, 0.225]),\n    transforms.Grayscale(num_output_channels=1)\n])\n\nclass_names = sorted(train_df['biome_type'].unique())\nprint(f\"Number of classes: {len(class_names)}\")\nlabel2idx = {label: idx for idx, label in enumerate(class_names)}\n\ntrain_dataset = ImageCSVDataset(\n    csv_file=\"/kaggle/input/acm-ai-hack/train.csv\",\n    root_dir=\"/kaggle/input/acm-ai-hack/train/\",\n    transform=transform,\n    label_map=label2idx\n   \n)\n\n\n","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-05-17T23:39:32.264781Z","iopub.execute_input":"2025-05-17T23:39:32.265054Z","iopub.status.idle":"2025-05-17T23:39:32.285075Z","shell.execute_reply.started":"2025-05-17T23:39:32.265036Z","shell.execute_reply":"2025-05-17T23:39:32.284397Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"Let's load the data using our custom PyTorch Dataset:","metadata":{}},{"cell_type":"code","source":"train_dataset = ImageCSVDataset(\n    csv_file=\"/kaggle/input/acm-ai-hack/train.csv\",\n    root_dir=\"/kaggle/input/acm-ai-hack/train/\",\n    transform=transform,\n    label_map=label2idx\n   \n)\n\n\n","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-05-17T23:39:32.285930Z","iopub.execute_input":"2025-05-17T23:39:32.286125Z","iopub.status.idle":"2025-05-17T23:39:32.293231Z","shell.execute_reply.started":"2025-05-17T23:39:32.286108Z","shell.execute_reply":"2025-05-17T23:39:32.292425Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"Here's an example of an image and its shape and label. Note that the images are black at white, so there are only one channel of pixel values. ","metadata":{}},{"cell_type":"code","source":"import matplotlib.pyplot as plt\nimport torchvision.transforms.functional as F\nexample_img, example_label = train_dataset[0]\n\n# Print out the shape and label\nprint(f\"Image shape: {example_img.shape}\")  # Should be [1, 224, 224] if grayscale\nprint(f\"Label: {example_label}\")\n# Convert back to PIL image for display\nimg_pil = F.to_pil_image(example_img)\n\nplt.imshow(img_pil, cmap='gray')  # use cmap='gray' since image is grayscale\n# Create a reverse map: label index to class name\nidx2label = {v: k for k, v in label2idx.items()}\n\n# Get the original label name\nlabel_name = idx2label.get(example_label, \"Unknown\")\nplt.title(f\"Label: {example_label}, {label_name}\")\n\nplt.axis('off')\nplt.show()","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-05-17T23:39:32.293960Z","iopub.execute_input":"2025-05-17T23:39:32.294151Z","iopub.status.idle":"2025-05-17T23:39:32.461573Z","shell.execute_reply.started":"2025-05-17T23:39:32.294136Z","shell.execute_reply":"2025-05-17T23:39:32.460852Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"Now it's your turn to make a model! We've helped define a very simple CNN for you. \n\n\nHere are some resources to help improve this CNN! Spend the challenge studying these resources and these different techniques \n\n\n* **Improve generalization**\t Key terms: Dropout, BatchNorm, Data Augmentation\t\n* **Handle overfitting**\tKey terms: Early stopping, weight decay\n* **Tune model size**\tKey terms: depth vs width in CNNs, Conv2d params\n\t\n* **Improve training**\tKey terms: Learning rate scheduler, Adam vs SGD\t\n* **Better evaluation**\tKey terms: Confusion matrix, Per-class accuracy\t\n\nAnd here are some resources: \n[PyTorch: How To Train A Classifier](https://docs.pytorch.org/tutorials/beginner/blitz/cifar10_tutorial.html)","metadata":{}},{"cell_type":"code","source":"import torch.nn as nn\n\nclass SELayer(nn.Module):\n    def __init__(self, channel, reduction=4):\n        super().__init__()\n        self.avg_pool = nn.AdaptiveAvgPool2d(1)\n        self.fc = nn.Sequential(\n            nn.Conv2d(channel, channel//reduction, 1, bias=False),\n            nn.ReLU(inplace=True),\n            nn.Conv2d(channel//reduction, channel, 1, bias=False),\n            nn.Sigmoid()\n        )\n    def forward(self, x):\n        return x * self.fc(self.avg_pool(x))\n\n\nclass ResidualBlock(nn.Module):\n    def __init__(self, in_ch, out_ch, stride=1):\n        super().__init__()\n        self.conv1 = nn.Conv2d(in_ch, out_ch, 3, stride, padding=1, bias=False)\n        self.bn1   = nn.BatchNorm2d(out_ch)\n        self.relu  = nn.ReLU(inplace=True)\n        self.conv2 = nn.Conv2d(out_ch, out_ch, 3, padding=1, bias=False)\n        self.bn2   = nn.BatchNorm2d(out_ch)\n        self.se    = SELayer(out_ch)\n        if stride != 1 or in_ch != out_ch:\n            self.downsample = nn.Sequential(\n                nn.Conv2d(in_ch, out_ch, 1, stride, bias=False),\n                nn.BatchNorm2d(out_ch)\n            )\n        else:\n            self.downsample = nn.Identity()\n\n    def forward(self, x):\n        identity = self.downsample(x)\n        out = self.relu(self.bn1(self.conv1(x)))\n        out = self.bn2(self.conv2(out))\n        out = self.se(out)\n        out = self.relu(out + identity)\n        return out\n\n\nclass SimpleCNN(nn.Module):\n    \"\"\"\n    ResNet-style small net (~800K params) with SE blocks.\n    - Input: 1×224×224\n    - Stages: 16→16, 16→32, 32→64, 64→128\n    - Global avg-pool + MLP head\n    \"\"\"\n    def __init__(self, num_classes: int):\n        super().__init__()\n        self.stem = nn.Sequential(\n            nn.Conv2d(1, 16, 3, padding=1, bias=False),\n            nn.BatchNorm2d(16),\n            nn.ReLU(inplace=True)\n        )\n        self.layer1 = nn.Sequential(\n            ResidualBlock(16, 16),\n            ResidualBlock(16, 16)\n        )                            # 224 → 224\n        self.layer2 = nn.Sequential(\n            ResidualBlock(16, 32, stride=2),\n            ResidualBlock(32, 32)\n        )                            # 224 → 112\n        self.layer3 = nn.Sequential(\n            ResidualBlock(32, 64, stride=2),\n            ResidualBlock(64, 64)\n        )                            # 112 → 56\n        self.layer4 = nn.Sequential(\n            ResidualBlock(64, 128, stride=2),\n            ResidualBlock(128, 128)\n        )                            # 56 → 28\n\n        self.avgpool   = nn.AdaptiveAvgPool2d(1)  # → 128×1×1\n        self.classifier = nn.Sequential(\n            nn.Dropout(0.5),\n            nn.Linear(128, 64, bias=False),\n            nn.BatchNorm1d(64),\n            nn.ReLU(inplace=True),\n            nn.Dropout(0.5),\n            nn.Linear(64, num_classes)\n        )\n\n        # Kaiming init\n        for m in self.modules():\n            if isinstance(m, nn.Conv2d):\n                nn.init.kaiming_normal_(m.weight, nonlinearity='relu')\n            elif isinstance(m, nn.Linear):\n                nn.init.kaiming_normal_(m.weight)\n                if m.bias is not None:\n                    nn.init.zeros_(m.bias)\n\n    def forward(self, x):\n        x = self.stem(x)\n        x = self.layer1(x)\n        x = self.layer2(x)\n        x = self.layer3(x)\n        x = self.layer4(x)\n        x = self.avgpool(x).view(x.size(0), -1)\n        return self.classifier(x)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-05-17T23:39:32.462178Z","iopub.execute_input":"2025-05-17T23:39:32.462347Z","iopub.status.idle":"2025-05-17T23:39:32.474552Z","shell.execute_reply.started":"2025-05-17T23:39:32.462332Z","shell.execute_reply":"2025-05-17T23:39:32.473878Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"import torch\nfrom torch.utils.data import random_split\nimport torch\n\ndevice = torch.device(\"cuda\" if torch.cuda.is_available() else \"cpu\")\n\nnum_classes = len(label2idx)\nmodel = SimpleCNN(num_classes).to(device)\n\ncriterion = nn.CrossEntropyLoss()\noptimizer = optim.Adam(model.parameters(), lr=0.001)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-05-17T23:39:32.476234Z","iopub.execute_input":"2025-05-17T23:39:32.476425Z","iopub.status.idle":"2025-05-17T23:39:32.711629Z","shell.execute_reply.started":"2025-05-17T23:39:32.476410Z","shell.execute_reply":"2025-05-17T23:39:32.710551Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"Here's an evaluate and train function. Once you define the model, you can pass it in along with a dataloader to start training the model.","metadata":{}},{"cell_type":"code","source":"train_size = int(0.8 * len(train_dataset))\nval_size = len(train_dataset) - train_size\n\ntrain_dataset, val_dataset = random_split(train_dataset, [train_size, val_size])\n\ndef train_one_epoch(model, dataloader):\n    model.train()\n    total_loss = 0\n\n    for batch in dataloader:\n        imgs, labels = batch  \n        imgs = imgs.to(device)\n        labels = labels.to(device)\n\n        optimizer.zero_grad()\n        outputs = model(imgs)\n        loss = criterion(outputs, labels)\n        loss.backward()\n        optimizer.step()\n\n        total_loss += loss.item()\n\n    return total_loss / len(dataloader)\n\n\ndef evaluate(model, dataloader):\n    model.eval()\n    correct, total = 0, 0\n\n    with torch.no_grad():\n        for batch in dataloader:\n            imgs, labels = batch  \n            imgs = imgs.to(device)\n            labels = labels.to(device)\n            outputs = model(imgs)\n            _, preds = torch.max(outputs, 1)\n            correct += (preds == labels).sum().item()\n            total += labels.size(0)\n\n    return correct/total","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-05-17T23:39:32.712824Z","iopub.execute_input":"2025-05-17T23:39:32.713142Z","iopub.status.idle":"2025-05-17T23:39:32.723475Z","shell.execute_reply.started":"2025-05-17T23:39:32.713109Z","shell.execute_reply":"2025-05-17T23:39:32.722732Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"from torch.utils.data import DataLoader\n\ntrain_loader = DataLoader(train_dataset, batch_size=32, shuffle=True, num_workers=2)\nval_loader = DataLoader(val_dataset, batch_size=32, shuffle=False, num_workers=2)\n\nEPOCHS = 5\n\nfor epoch in range(EPOCHS):\n    train_loss = train_one_epoch(model, train_loader)\n    val_acc = evaluate(model, val_loader)\n    print(f\"Epoch {epoch+1}: Train Loss = {train_loss:.4f}, Val Acc = {val_acc:.2%}\")","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-05-17T23:39:32.724730Z","iopub.execute_input":"2025-05-17T23:39:32.725556Z","iopub.status.idle":"2025-05-17T23:41:05.059578Z","shell.execute_reply.started":"2025-05-17T23:39:32.725492Z","shell.execute_reply":"2025-05-17T23:41:05.058676Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"Once you've defined your model, you can load the test dataset, and generate the predictions. You'll need the label map to map the integer predictions back to the biome names.","metadata":{}},{"cell_type":"code","source":"test_df = pd.read_csv(\"/kaggle/input/acm-ai-hack/test.csv\")\n\ntransform = transforms.Compose([\n    transforms.Resize((224, 224)),\n    transforms.ToTensor(),\n    transforms.Normalize(\n        mean=[0.485, 0.456, 0.406],\n        std=[0.229, 0.224, 0.225]\n    ),\n    transforms.Grayscale(num_output_channels=1),\n])\n\ntest_dataset = ImageCSVDataset(\n    csv_file=\"/kaggle/input/acm-ai-hack/test.csv\",\n    root_dir=\"/kaggle/input/acm-ai-hack/test\",\n    transform=transform,\n    is_test=True\n)\ntest_loader = DataLoader(\n    test_dataset,\n    batch_size=32,\n    shuffle=False,\n    num_workers=2\n)\n\nidx2label = {v: k for k, v in label2idx.items()}\n\nmodel.eval()\npreds = []\nids = []\n\nwith torch.no_grad():\n    for images, image_id in test_loader:\n        images = images.to(device)\n        outputs = model(images)\n        _, predicted = torch.max(outputs, 1)\n\n        preds.extend(predicted.cpu().numpy())\n        ids.extend(\n            id.item() if isinstance(id, torch.Tensor) else id\n            for id in image_id\n        )\n\npred_labels = [idx2label[p] for p in preds]\n\nsubmission = pd.DataFrame({\n    \"ID\": ids,\n    \"target\": pred_labels\n})","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-05-17T23:41:05.060673Z","iopub.execute_input":"2025-05-17T23:41:05.060928Z","iopub.status.idle":"2025-05-17T23:41:08.925431Z","shell.execute_reply.started":"2025-05-17T23:41:05.060903Z","shell.execute_reply":"2025-05-17T23:41:08.924380Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"submission.to_csv(\"/kaggle/working/submission.csv\", index=False)\n","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-05-17T23:41:08.926649Z","iopub.execute_input":"2025-05-17T23:41:08.926917Z","iopub.status.idle":"2025-05-17T23:41:08.934125Z","shell.execute_reply.started":"2025-05-17T23:41:08.926891Z","shell.execute_reply":"2025-05-17T23:41:08.933464Z"}},"outputs":[],"execution_count":null}]}