{"metadata":{"kernelspec":{"name":"python3","display_name":"Python 3","language":"python"},"language_info":{"name":"python","version":"3.11.13","mimetype":"text/x-python","codemirror_mode":{"name":"ipython","version":3},"pygments_lexer":"ipython3","nbconvert_exporter":"python","file_extension":".py"},"colab":{"machine_shape":"hm","gpuType":"T4"},"accelerator":"GPU","kaggle":{"accelerator":"nvidiaTeslaT4","dataSources":[{"sourceId":20270,"databundleVersionId":1222630,"isSourceIdPinned":false,"sourceType":"competition"},{"sourceId":9797223,"sourceType":"datasetVersion","datasetId":982170},{"sourceId":80230206,"sourceType":"kernelVersion"},{"sourceId":265822687,"sourceType":"kernelVersion"},{"sourceId":267285227,"sourceType":"kernelVersion"}],"dockerImageVersionId":31090,"isInternetEnabled":true,"language":"python","sourceType":"notebook","isGpuEnabled":true}},"nbformat_minor":4,"nbformat":4,"cells":[{"cell_type":"code","source":"import sys\nsys.path.append(\"../input/tez-lib/\")","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-10-11T14:01:49.163583Z","iopub.execute_input":"2025-10-11T14:01:49.163847Z","iopub.status.idle":"2025-10-11T14:01:49.171470Z","shell.execute_reply.started":"2025-10-11T14:01:49.163819Z","shell.execute_reply":"2025-10-11T14:01:49.170764Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"import random\nimport numpy as np\nimport torch\nimport os\n\ndef seed_everything(seed):\n    random.seed(seed)\n    os.environ['PYTHONHASHSEED'] = str(seed)\n    np.random.seed(seed)\n    torch.manual_seed(seed)\n    torch.cuda.manual_seed(seed)\n    torch.backends.cudnn.deterministic = True\n    \nseed_everything(69)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-10-11T14:01:49.172160Z","iopub.execute_input":"2025-10-11T14:01:49.172383Z","iopub.status.idle":"2025-10-11T14:01:52.735741Z","shell.execute_reply.started":"2025-10-11T14:01:49.172363Z","shell.execute_reply":"2025-10-11T14:01:52.735206Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"import torch.nn as nn\n\nfrom tez import Tez, TezConfig\nimport tez\nimport albumentations\nimport pandas as pd\nimport cv2\nimport numpy as np\nimport timm\nimport torch.nn as nn\nfrom sklearn import metrics\nimport torch\nfrom tez.callbacks import EarlyStopping\nfrom tqdm import tqdm","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-10-11T14:01:52.737296Z","iopub.execute_input":"2025-10-11T14:01:52.737561Z","iopub.status.idle":"2025-10-11T14:02:33.905669Z","shell.execute_reply.started":"2025-10-11T14:01:52.737544Z","shell.execute_reply":"2025-10-11T14:02:33.904817Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"class args:\n    batch_size = 8\n    image_size = 384\n    epochs = 2\n    fold = 0","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-10-11T14:02:33.906510Z","iopub.execute_input":"2025-10-11T14:02:33.906893Z","iopub.status.idle":"2025-10-11T14:02:33.910658Z","shell.execute_reply.started":"2025-10-11T14:02:33.906866Z","shell.execute_reply":"2025-10-11T14:02:33.910105Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"class CustomDataset:\n    def __init__(self, image_paths, dense_features, targets, augmentations):\n        self.image_paths = image_paths\n        self.dense_features = dense_features\n        self.targets = targets\n        self.augmentations = augmentations\n        \n    def __len__(self):\n        return len(self.image_paths)\n    \n    def __getitem__(self, item):\n        image = cv2.imread(self.image_paths[item])\n        image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)\n        \n        if self.augmentations is not None:\n            augmented = self.augmentations(image=image)\n            image = augmented[\"image\"]\n            \n        image = np.transpose(image, (2, 0, 1)).astype(np.float32)\n        \n        features = self.dense_features[item, :]\n        targets = self.targets[item]\n        \n        return {\n            \"image\": torch.tensor(image, dtype=torch.float),\n            \"features\": torch.tensor(features, dtype=torch.float),\n            \"targets\": torch.tensor(targets, dtype=torch.float),\n        }","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-10-11T14:02:33.911391Z","iopub.execute_input":"2025-10-11T14:02:33.911686Z","iopub.status.idle":"2025-10-11T14:02:33.929943Z","shell.execute_reply.started":"2025-10-11T14:02:33.911663Z","shell.execute_reply":"2025-10-11T14:02:33.929339Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"class CustomModel(nn.Module):\n    def __init__(self):\n        super().__init__()        \n        self.model = timm.create_model(\"resnet50\", pretrained=True, in_chans=3)\n        self.dropout = nn.Dropout(0.5)\n        self.out = nn.Linear(1000, 1)\n        self.step_scheduler_after = \"epoch\"\n\n\n    def monitor_metrics(self, outputs, targets, loss):\n        valid_binaryloss = loss\n        if str(valid_binaryloss) == 'nan':\n            valid_binaryloss = float('inf')\n        return {\"binaryloss\": valid_binaryloss}\n\n    def optimizer_scheduler(self):\n        opt = torch.optim.AdamW(self.parameters(), lr=2.5e-05, weight_decay=0.01)\n        sch = torch.optim.lr_scheduler.CosineAnnealingWarmRestarts(\n            opt, T_0=10, T_mult=1, eta_min=1e-6, last_epoch=-1\n        )\n        return opt,sch\n\n    def forward(self, image, features, targets=None):\n\n        x = self.model(image)\n        x = self.dropout(x)\n        x = self.out(x)\n\n        if targets is not None:\n            loss = nn.BCEWithLogitsLoss()(x, targets.view(-1, 1).type_as(x))\n            metrics = self.monitor_metrics(x, targets, loss)\n            return x, loss, metrics\n        return x, 0, {}\n\n","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-10-11T14:02:33.930722Z","iopub.execute_input":"2025-10-11T14:02:33.930908Z","iopub.status.idle":"2025-10-11T14:02:33.941861Z","shell.execute_reply.started":"2025-10-11T14:02:33.930894Z","shell.execute_reply":"2025-10-11T14:02:33.941240Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"train_aug = albumentations.Compose([\n    albumentations.LongestMaxSize(args.image_size, p=1),\n    albumentations.PadIfNeeded(args.image_size, args.image_size, p=1, border_mode=cv2.BORDER_REFLECT_101),\n\n    albumentations.HorizontalFlip(p=0.5),\n    albumentations.VerticalFlip(p=0.1),\n    albumentations.RandomRotate90(p=0.2),  \n    albumentations.ShiftScaleRotate(\n        shift_limit=0.1, scale_limit=0.15, rotate_limit=30, border_mode=cv2.BORDER_REFLECT_101, p=0.5\n    ),\n\n    albumentations.OneOf([\n        albumentations.HueSaturationValue(20, 30, 20, p=0.7),\n        albumentations.RGBShift(10, 10, 10, p=0.3),\n    ], p=0.5),\n\n    albumentations.RandomBrightnessContrast(0.15, 0.15, p=0.5),\n    albumentations.CLAHE(clip_limit=2.0, p=0.2),  \n\n    albumentations.OneOf([\n        albumentations.GaussNoise(var_limit=(10.0, 50.0), p=0.3),\n        albumentations.MotionBlur(blur_limit=3, p=0.2),\n    ], p=0.2),\n\n    albumentations.Normalize(\n        mean=[0.485, 0.456, 0.406],\n        std=[0.229, 0.224, 0.225],\n        max_pixel_value=255.0,\n    ),\n], p=1.0)\n\nvalid_aug = albumentations.Compose([\n    albumentations.LongestMaxSize(args.image_size, p=1),\n    albumentations.PadIfNeeded(args.image_size, args.image_size, p=1, border_mode=cv2.BORDER_REFLECT_101),\n    albumentations.Normalize(\n        mean=[0.485, 0.456, 0.406],\n        std=[0.229, 0.224, 0.225],\n        max_pixel_value=255.0,\n    ),\n], p=1.0)\n","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-10-11T14:02:33.942555Z","iopub.execute_input":"2025-10-11T14:02:33.942778Z","iopub.status.idle":"2025-10-11T14:02:33.966712Z","shell.execute_reply.started":"2025-10-11T14:02:33.942759Z","shell.execute_reply":"2025-10-11T14:02:33.965978Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"df = pd.read_csv(\"/kaggle/input/23bcs10181-abhayraj-kfold-melanoma/train_5folds.csv\")\ndf.head()","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-10-11T14:02:33.967395Z","iopub.execute_input":"2025-10-11T14:02:33.967583Z","iopub.status.idle":"2025-10-11T14:02:34.069076Z","shell.execute_reply.started":"2025-10-11T14:02:33.967567Z","shell.execute_reply":"2025-10-11T14:02:34.068534Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"i=0\nprint(f'training fold: {i} start')\nargs.fold = 0\ndf_train = df[df.kfold != args.fold].reset_index(drop=True)\ndf_valid = df[df.kfold == args.fold].reset_index(drop=True)\ndense_features = [\n    'age_approx'\n]\ntrain_img_paths = [f\"/kaggle/input/siim-isic-melanoma-classification/jpeg/train/{x}.jpg\" for x in df_train[\"image_name\"].values]\nvalid_img_paths = [f\"/kaggle/input/siim-isic-melanoma-classification/jpeg/train/{x}.jpg\" for x in df_valid[\"image_name\"].values]\ntrain_dataset = CustomDataset(\n    image_paths=train_img_paths,\n    dense_features=df_train[dense_features].values,\n    targets=df_train.target.values,\n    augmentations=train_aug,\n)\n\nvalid_dataset = CustomDataset(\n    image_paths=valid_img_paths,\n    dense_features=df_valid[dense_features].values,\n    targets=df_valid.target.values,\n    augmentations=valid_aug,\n)\n\nmodel = CustomModel()\nmodel = Tez(model)\nconfig = TezConfig(\n    training_batch_size=args.batch_size,\n    validation_batch_size=2 * args.batch_size,\n    epochs=args.epochs,\n    step_scheduler_after=\"epoch\",\n    step_scheduler_metric=\"valid_binaryloss\",\n    fp16=True,\n    # fp16=False,\n    val_strategy=\"batch\",\n    val_steps=900,\n)\n\nes = EarlyStopping(\n    monitor=\"valid_binaryloss\",\n    model_path=f\"model_f{args.fold}.bin\",\n    patience=4,#3,\n    mode=\"min\",\n    save_weights_only=True,\n)\n\nmodel.fit(\n    train_dataset,\n    valid_dataset=valid_dataset,\n    callbacks=[es],\n    config=config,\n)\nprint(f'training fold: {i} complete')","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-10-11T14:04:03.551539Z","iopub.execute_input":"2025-10-11T14:04:03.551802Z","iopub.status.idle":"2025-10-11T15:24:27.068121Z","shell.execute_reply.started":"2025-10-11T14:04:03.551782Z","shell.execute_reply":"2025-10-11T15:24:27.067463Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"","metadata":{"trusted":true},"outputs":[],"execution_count":null}]}