diff --git a/examples/kaggle_chest_x-ray_images.ipynb b/examples/kaggle_chest_x-ray_images.ipynb
new file mode 100644
index 000000000..a02d545f3
--- /dev/null
+++ b/examples/kaggle_chest_x-ray_images.ipynb
@@ -0,0 +1 @@
+{"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"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"},"kaggle":{"accelerator":"nvidiaTeslaT4","dataSources":[{"sourceId":23812,"sourceType":"datasetVersion","datasetId":17810}],"dockerImageVersionId":31192,"isInternetEnabled":true,"language":"python","sourceType":"notebook","isGpuEnabled":true}},"nbformat_minor":4,"nbformat":4,"cells":[{"cell_type":"markdown","source":"## Pyhealth Kaggle Example\n\nThis notebook demonstrates how to use PyHealth to load and train a model on a public dataset from Kaggle.\nIn particular, we will use the Chest X-Ray Pneumonia dataset: https://www.kaggle.com/datasets/paultimothymooney/chest-xray-pneumonia\n\n\n### Dependencies\n1. download PyHealth\n2. downgrade protobuf to avoid depdendecieis conflict with Kaggle env\n```bash\npip install git+https://github.com/rz-coder/PyHealth.git\npip install \"protobuf<=3.20.3\"\n```","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\n# for 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":"2025-12-02T22:37:08.125167Z","iopub.execute_input":"2025-12-02T22:37:08.125897Z","iopub.status.idle":"2025-12-02T22:37:08.129774Z","shell.execute_reply.started":"2025-12-02T22:37:08.125869Z","shell.execute_reply":"2025-12-02T22:37:08.128920Z"}},"outputs":[],"execution_count":22},{"cell_type":"code","source":"\n# This will be the input dir path\nkaggle_root = \"/kaggle/input/chest-xray-pneumonia/chest_xray\"","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-12-02T22:08:15.271176Z","iopub.execute_input":"2025-12-02T22:08:15.271408Z","iopub.status.idle":"2025-12-02T22:08:15.275071Z","shell.execute_reply.started":"2025-12-02T22:08:15.271390Z","shell.execute_reply":"2025-12-02T22:08:15.274441Z"}},"outputs":[],"execution_count":3},{"cell_type":"code","source":"import os\nimport pandas as pd\nimport yaml\nfrom pathlib import Path\nfrom typing import Dict, List, Optional\nimport shutil\nfrom PIL import Image\nimport numpy as np\nimport torch","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-12-02T22:08:15.275767Z","iopub.execute_input":"2025-12-02T22:08:15.275957Z","iopub.status.idle":"2025-12-02T22:08:15.290900Z","shell.execute_reply.started":"2025-12-02T22:08:15.275941Z","shell.execute_reply":"2025-12-02T22:08:15.290257Z"}},"outputs":[],"execution_count":4},{"cell_type":"code","source":"import torchvision.transforms as T\nimport matplotlib.pyplot as plt\n\npath_pneumonia = \"/kaggle/input/chest-xray-pneumonia/chest_xray/chest_xray/val/PNEUMONIA/person1950_bacteria_4881.jpeg\"\npath_normal = \"/kaggle/input/chest-xray-pneumonia/chest_xray/chest_xray/val/NORMAL/NORMAL2-IM-1431-0001.jpeg\"\n\n\ntransform = T.Compose([\n    T.Resize((224, 224)),\n    T.ToTensor(),\n])\n\n# Load images\nimg_pneumonia = Image.open(path_pneumonia).convert(\"L\")\nimg_normal = Image.open(path_normal).convert(\"L\")\n\ntensor_pneumonia = transform(img_pneumonia)\ntensor_normal = transform(img_normal)\n\nfig, axs = plt.subplots(1, 2, figsize=(8,4))\n\naxs[0].imshow(img_pneumonia, cmap=\"gray\")\naxs[0].set_title(\"PNEUMONIA\")\naxs[0].axis(\"off\")\n\naxs[1].imshow(img_normal, cmap=\"gray\")\naxs[1].set_title(\"NORMAL\")\naxs[1].axis(\"off\")\n\nplt.show()","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-12-02T22:38:41.904765Z","iopub.execute_input":"2025-12-02T22:38:41.905409Z","iopub.status.idle":"2025-12-02T22:38:42.327806Z","shell.execute_reply.started":"2025-12-02T22:38:41.905383Z","shell.execute_reply":"2025-12-02T22:38:42.327111Z"}},"outputs":[{"output_type":"display_data","data":{"text/plain":"<Figure size 800x400 with 2 Axes>","image/png":"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\n"},"metadata":{}}],"execution_count":25},{"cell_type":"code","source":"config_data = {\n    \"version\": \"1.0\",\n    \"tables\": {\n        \"pneumonia\": {\n            \"file_path\": \"metadata.csv\",\n            \"patient_id\": \"patient_id\",\n            \"attributes\": [\"path\", \"label\", \"split\"]\n        }\n    }\n}\n\n# Save the yaml config file\nwork_dir = \".\"\nconfig_path = os.path.join(work_dir, \"pneumonia.yaml\")\nwith open(config_path, \"w\") as f:\n    yaml.dump(config_data, f)\n\nprint(f\"Created config file at {config_path}\")","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-12-02T22:38:47.059261Z","iopub.execute_input":"2025-12-02T22:38:47.059840Z","iopub.status.idle":"2025-12-02T22:38:47.065798Z","shell.execute_reply.started":"2025-12-02T22:38:47.059815Z","shell.execute_reply":"2025-12-02T22:38:47.065070Z"}},"outputs":[{"name":"stdout","text":"Created config file at ./pneumonia.yaml\n","output_type":"stream"}],"execution_count":26},{"cell_type":"code","source":"from pyhealth.datasets import BaseDataset\nfrom pyhealth.tasks import BaseTask\nfrom pyhealth.data import Event, Patient\nfrom pyhealth.datasets import get_dataloader, split_by_sample\nfrom pyhealth.models import TorchvisionModel\nfrom pyhealth.trainer import Trainer","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-12-02T22:08:15.307084Z","iopub.execute_input":"2025-12-02T22:08:15.307288Z","iopub.status.idle":"2025-12-02T22:08:15.320485Z","shell.execute_reply.started":"2025-12-02T22:08:15.307265Z","shell.execute_reply":"2025-12-02T22:08:15.319953Z"}},"outputs":[],"execution_count":6},{"cell_type":"code","source":"class PneumoniaDataset(BaseDataset):\n    def __init__(self, root, config_path=None, dev=False):\n        self._index_data(root)\n        \n        # Initialize BaseDataset\n        if config_path is None:\n            config_path = \"./pneumonia.yaml\"\n            \n        super().__init__(\n            root=\".\",\n            tables=[\"pneumonia\"],\n            dataset_name=\"PneumoniaDataset\",\n            config_path=config_path,\n            dev=dev\n        )\n\n    def _index_data(self, root):\n        data = []\n        splits = [\"train\", \"test\", \"val\"]\n        classes = {\"NORMAL\": 0, \"PNEUMONIA\": 1}\n        \n        print(f\"Scanning dataset at {root}...\")\n        for split in splits:\n            for cls_name, cls_label in classes.items():\n                dir_path = os.path.join(root, split, cls_name)\n                if not os.path.exists(dir_path):\n                    print(f\"Warning: Directory not found: {dir_path}\")\n                    continue\n                    \n                for img_name in os.listdir(dir_path):\n                    if not img_name.lower().endswith(('.jpeg', '.jpg', '.png')):\n                        continue\n                        \n                    img_path = os.path.join(root, split, cls_name, img_name)\n                    # Use filename as patient_id for simplicity\n                    patient_id = f\"{split}_{cls_name}_{img_name.split('.')[0]}\"\n                    \n                    data.append({\n                        \"patient_id\": patient_id,\n                        \"path\": img_path,\n                        \"label\": cls_label,\n                        \"split\": split\n                    })\n        \n        df = pd.DataFrame(data)\n        df.to_csv(\"./metadata.csv\", index=False)\n        print(f\"Indexed {len(df)} images to ./metadata.csv\")\n\n    @property\n    def default_task(self):\n        return PneumoniaClassificationTask()","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-12-02T22:08:15.321289Z","iopub.execute_input":"2025-12-02T22:08:15.321509Z","iopub.status.idle":"2025-12-02T22:08:15.339968Z","shell.execute_reply.started":"2025-12-02T22:08:15.321486Z","shell.execute_reply":"2025-12-02T22:08:15.339218Z"}},"outputs":[],"execution_count":7},{"cell_type":"code","source":"class PneumoniaClassificationTask(BaseTask):\n    task_name = \"PneumoniaClassification\"\n    input_schema = {\"image\": \"image\"}\n    output_schema = {\"label\": \"binary\"}\n\n    def __call__(self, patient: Patient) -> List[Dict]:\n        samples = []\n        # Retrieve events from the 'pneumonia' table\n        events = patient.get_events(event_type=\"pneumonia\")\n        \n        for event in events:\n            samples.append({\n                \"image\": event.attr_dict[\"path\"],\n                \"label\": int(event.attr_dict[\"label\"])\n            })\n        return samples","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-12-02T22:08:15.340721Z","iopub.execute_input":"2025-12-02T22:08:15.340966Z","iopub.status.idle":"2025-12-02T22:08:15.356842Z","shell.execute_reply.started":"2025-12-02T22:08:15.340942Z","shell.execute_reply":"2025-12-02T22:08:15.356221Z"}},"outputs":[],"execution_count":8},{"cell_type":"code","source":"dataset = PneumoniaDataset(root=kaggle_root)\nsample_dataset = dataset.set_task()","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-12-02T22:08:15.357642Z","iopub.execute_input":"2025-12-02T22:08:15.357865Z","iopub.status.idle":"2025-12-02T22:09:45.108127Z","shell.execute_reply.started":"2025-12-02T22:08:15.357849Z","shell.execute_reply":"2025-12-02T22:09:45.107510Z"}},"outputs":[{"name":"stdout","text":"Scanning dataset at /kaggle/input/chest-xray-pneumonia/chest_xray...\nIndexed 5856 images to ./metadata.csv\nInitializing PneumoniaDataset dataset from . (dev mode: False)\n","output_type":"stream"},{"name":"stderr","text":"INFO:pyhealth.datasets.base_dataset:Initializing PneumoniaDataset dataset from . (dev mode: False)\n","output_type":"stream"},{"name":"stdout","text":"Scanning table: pneumonia from /kaggle/working/metadata.csv\n","output_type":"stream"},{"name":"stderr","text":"INFO:pyhealth.datasets.base_dataset:Scanning table: pneumonia from /kaggle/working/metadata.csv\n","output_type":"stream"},{"name":"stdout","text":"Setting task PneumoniaClassification for PneumoniaDataset base dataset...\n","output_type":"stream"},{"name":"stderr","text":"INFO:pyhealth.datasets.base_dataset:Setting task PneumoniaClassification for PneumoniaDataset base dataset...\n","output_type":"stream"},{"name":"stdout","text":"Generating samples with 1 worker(s)...\n","output_type":"stream"},{"name":"stderr","text":"INFO:pyhealth.datasets.base_dataset:Generating samples with 1 worker(s)...\n","output_type":"stream"},{"name":"stdout","text":"Collecting global event dataframe...\n","output_type":"stream"},{"name":"stderr","text":"INFO:pyhealth.datasets.base_dataset:Collecting global event dataframe...\n","output_type":"stream"},{"name":"stdout","text":"Collected dataframe with shape: (5856, 6)\n","output_type":"stream"},{"name":"stderr","text":"INFO:pyhealth.datasets.base_dataset:Collected dataframe with shape: (5856, 6)\nGenerating samples for PneumoniaClassification with 1 worker: 100%|██████████| 5856/5856 [00:02<00:00, 2326.97it/s]","output_type":"stream"},{"name":"stdout","text":"Label label vocab: {0: 0, 1: 1}\n","output_type":"stream"},{"name":"stderr","text":"\nINFO:pyhealth.processors.label_processor:Label label vocab: {0: 0, 1: 1}\nProcessing samples: 100%|██████████| 5856/5856 [01:27<00:00, 67.21it/s]","output_type":"stream"},{"name":"stdout","text":"Generated 5856 samples for task PneumoniaClassification\n","output_type":"stream"},{"name":"stderr","text":"\nINFO:pyhealth.datasets.base_dataset:Generated 5856 samples for task PneumoniaClassification\n","output_type":"stream"}],"execution_count":9},{"cell_type":"code","source":"sample_dataset[0]","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-12-02T22:09:45.109691Z","iopub.execute_input":"2025-12-02T22:09:45.109899Z","iopub.status.idle":"2025-12-02T22:09:45.118681Z","shell.execute_reply.started":"2025-12-02T22:09:45.109882Z","shell.execute_reply":"2025-12-02T22:09:45.117926Z"}},"outputs":[{"execution_count":10,"output_type":"execute_result","data":{"text/plain":"{'image': tensor([[[0.0000, 0.0000, 0.0000,  ..., 0.4275, 0.4510, 0.4588],\n          [0.0000, 0.0000, 0.0000,  ..., 0.4431, 0.4627, 0.4784],\n          [0.0000, 0.0000, 0.0000,  ..., 0.4588, 0.4784, 0.5020],\n          ...,\n          [0.0000, 0.0000, 0.0000,  ..., 0.0000, 0.0000, 0.0000],\n          [0.0000, 0.0000, 0.0000,  ..., 0.0000, 0.0000, 0.0000],\n          [0.0000, 0.0000, 0.0000,  ..., 0.0000, 0.0000, 0.0000]]]),\n 'label': tensor([0.])}"},"metadata":{}}],"execution_count":10},{"cell_type":"code","source":"train_ds, val_ds, test_ds = split_by_sample(sample_dataset, [0.8, 0.1, 0.1])\n","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-12-02T22:09:45.119427Z","iopub.execute_input":"2025-12-02T22:09:45.119766Z","iopub.status.idle":"2025-12-02T22:09:45.130078Z","shell.execute_reply.started":"2025-12-02T22:09:45.119740Z","shell.execute_reply":"2025-12-02T22:09:45.129569Z"}},"outputs":[],"execution_count":11},{"cell_type":"code","source":"train_loader = get_dataloader(train_ds, batch_size=32, shuffle=True)\nval_loader = get_dataloader(val_ds, batch_size=32, shuffle=False)\ntest_loader = get_dataloader(test_ds, batch_size=32, shuffle=False)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-12-02T22:09:45.130774Z","iopub.execute_input":"2025-12-02T22:09:45.130968Z","iopub.status.idle":"2025-12-02T22:09:45.144650Z","shell.execute_reply.started":"2025-12-02T22:09:45.130931Z","shell.execute_reply":"2025-12-02T22:09:45.143913Z"}},"outputs":[],"execution_count":12},{"cell_type":"code","source":"print(f\"Train size: {len(train_ds)}, Val size: {len(val_ds)}, Test size: {len(test_ds)}\")","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-12-02T22:09:45.145944Z","iopub.execute_input":"2025-12-02T22:09:45.146491Z","iopub.status.idle":"2025-12-02T22:09:45.159243Z","shell.execute_reply.started":"2025-12-02T22:09:45.146475Z","shell.execute_reply":"2025-12-02T22:09:45.158558Z"}},"outputs":[{"name":"stdout","text":"Train size: 4684, Val size: 586, Test size: 586\n","output_type":"stream"}],"execution_count":13},{"cell_type":"markdown","source":"## Trainning \nKaggle provide GPU option. If your kernel has GPU, you can use it to train your data. I am usigng Tesla T4 in this notebook","metadata":{}},{"cell_type":"code","source":"print(\"CUDA Available:\", torch.cuda.is_available())\nprint(\"GPU Name:\", torch.cuda.get_device_name(0) if torch.cuda.is_available() else \"No GPU\")","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-12-02T22:09:45.160461Z","iopub.execute_input":"2025-12-02T22:09:45.161018Z","iopub.status.idle":"2025-12-02T22:09:45.172562Z","shell.execute_reply.started":"2025-12-02T22:09:45.161000Z","shell.execute_reply":"2025-12-02T22:09:45.171855Z"}},"outputs":[{"name":"stdout","text":"CUDA Available: True\nGPU Name: Tesla T4\n","output_type":"stream"}],"execution_count":14},{"cell_type":"code","source":"model = TorchvisionModel(\n    dataset=sample_dataset,\n    model_name=\"resnet18\",\n    model_config={\"weights\": \"DEFAULT\"} # Use pretrained weights\n)\n\ntrainer = Trainer(model=model,\n                  device=\"cuda\" if torch.cuda.is_available() else \"cpu\", \n                  metrics=[\"accuracy\", \"f1\"])\ntrainer.train(\n    train_dataloader=train_loader,\n    val_dataloader=val_loader,\n    epochs=5,\n    monitor=\"accuracy\",\n)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-12-02T22:11:23.048502Z","iopub.execute_input":"2025-12-02T22:11:23.048829Z","iopub.status.idle":"2025-12-02T22:12:37.728962Z","shell.execute_reply.started":"2025-12-02T22:11:23.048789Z","shell.execute_reply":"2025-12-02T22:12:37.728312Z"}},"outputs":[{"name":"stdout","text":"Downloading: \"https://download.pytorch.org/models/resnet18-f37072fd.pth\" to /root/.cache/torch/hub/checkpoints/resnet18-f37072fd.pth\n","output_type":"stream"},{"name":"stderr","text":"100%|██████████| 44.7M/44.7M [00:00<00:00, 176MB/s]\n","output_type":"stream"},{"name":"stdout","text":"TorchvisionModel(\n  (model): ResNet(\n    (conv1): Conv2d(3, 64, kernel_size=(7, 7), stride=(2, 2), padding=(3, 3), bias=False)\n    (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n    (relu): ReLU(inplace=True)\n    (maxpool): MaxPool2d(kernel_size=3, stride=2, padding=1, dilation=1, ceil_mode=False)\n    (layer1): Sequential(\n      (0): BasicBlock(\n        (conv1): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n        (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n        (relu): ReLU(inplace=True)\n        (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n        (bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n      )\n      (1): BasicBlock(\n        (conv1): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n        (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n        (relu): ReLU(inplace=True)\n        (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n        (bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n      )\n    )\n    (layer2): Sequential(\n      (0): BasicBlock(\n        (conv1): Conv2d(64, 128, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)\n        (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n        (relu): ReLU(inplace=True)\n        (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n        (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n        (downsample): Sequential(\n          (0): Conv2d(64, 128, kernel_size=(1, 1), stride=(2, 2), bias=False)\n          (1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n        )\n      )\n      (1): BasicBlock(\n        (conv1): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n        (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n        (relu): ReLU(inplace=True)\n        (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n        (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n      )\n    )\n    (layer3): Sequential(\n      (0): BasicBlock(\n        (conv1): Conv2d(128, 256, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)\n        (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n        (relu): ReLU(inplace=True)\n        (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n        (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n        (downsample): Sequential(\n          (0): Conv2d(128, 256, kernel_size=(1, 1), stride=(2, 2), bias=False)\n          (1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n        )\n      )\n      (1): BasicBlock(\n        (conv1): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n        (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n        (relu): ReLU(inplace=True)\n        (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n        (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n      )\n    )\n    (layer4): Sequential(\n      (0): BasicBlock(\n        (conv1): Conv2d(256, 512, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)\n        (bn1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n        (relu): ReLU(inplace=True)\n        (conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n        (bn2): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n        (downsample): Sequential(\n          (0): Conv2d(256, 512, kernel_size=(1, 1), stride=(2, 2), bias=False)\n          (1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n        )\n      )\n      (1): BasicBlock(\n        (conv1): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n        (bn1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n        (relu): ReLU(inplace=True)\n        (conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n        (bn2): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n      )\n    )\n    (avgpool): AdaptiveAvgPool2d(output_size=(1, 1))\n    (fc): Linear(in_features=512, out_features=1, bias=True)\n  )\n)\n","output_type":"stream"},{"name":"stderr","text":"INFO:pyhealth.trainer:TorchvisionModel(\n  (model): ResNet(\n    (conv1): Conv2d(3, 64, kernel_size=(7, 7), stride=(2, 2), padding=(3, 3), bias=False)\n    (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n    (relu): ReLU(inplace=True)\n    (maxpool): MaxPool2d(kernel_size=3, stride=2, padding=1, dilation=1, ceil_mode=False)\n    (layer1): Sequential(\n      (0): BasicBlock(\n        (conv1): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n        (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n        (relu): ReLU(inplace=True)\n        (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n        (bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n      )\n      (1): BasicBlock(\n        (conv1): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n        (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n        (relu): ReLU(inplace=True)\n        (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n        (bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n      )\n    )\n    (layer2): Sequential(\n      (0): BasicBlock(\n        (conv1): Conv2d(64, 128, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)\n        (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n        (relu): ReLU(inplace=True)\n        (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n        (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n        (downsample): Sequential(\n          (0): Conv2d(64, 128, kernel_size=(1, 1), stride=(2, 2), bias=False)\n          (1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n        )\n      )\n      (1): BasicBlock(\n        (conv1): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n        (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n        (relu): ReLU(inplace=True)\n        (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n        (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n      )\n    )\n    (layer3): Sequential(\n      (0): BasicBlock(\n        (conv1): Conv2d(128, 256, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)\n        (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n        (relu): ReLU(inplace=True)\n        (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n        (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n        (downsample): Sequential(\n          (0): Conv2d(128, 256, kernel_size=(1, 1), stride=(2, 2), bias=False)\n          (1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n        )\n      )\n      (1): BasicBlock(\n        (conv1): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n        (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n        (relu): ReLU(inplace=True)\n        (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n        (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n      )\n    )\n    (layer4): Sequential(\n      (0): BasicBlock(\n        (conv1): Conv2d(256, 512, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)\n        (bn1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n        (relu): ReLU(inplace=True)\n        (conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n        (bn2): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n        (downsample): Sequential(\n          (0): Conv2d(256, 512, kernel_size=(1, 1), stride=(2, 2), bias=False)\n          (1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n        )\n      )\n      (1): BasicBlock(\n        (conv1): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n        (bn1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n        (relu): ReLU(inplace=True)\n        (conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n        (bn2): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n      )\n    )\n    (avgpool): AdaptiveAvgPool2d(output_size=(1, 1))\n    (fc): Linear(in_features=512, out_features=1, bias=True)\n  )\n)\n","output_type":"stream"},{"name":"stdout","text":"Metrics: ['accuracy', 'f1']\n","output_type":"stream"},{"name":"stderr","text":"INFO:pyhealth.trainer:Metrics: ['accuracy', 'f1']\n","output_type":"stream"},{"name":"stdout","text":"Device: cuda\n","output_type":"stream"},{"name":"stderr","text":"INFO:pyhealth.trainer:Device: cuda\n","output_type":"stream"},{"name":"stdout","text":"\n","output_type":"stream"},{"name":"stderr","text":"INFO:pyhealth.trainer:\n","output_type":"stream"},{"name":"stdout","text":"Training:\n","output_type":"stream"},{"name":"stderr","text":"INFO:pyhealth.trainer:Training:\n","output_type":"stream"},{"name":"stdout","text":"Batch size: 32\n","output_type":"stream"},{"name":"stderr","text":"INFO:pyhealth.trainer:Batch size: 32\n","output_type":"stream"},{"name":"stdout","text":"Optimizer: <class 'torch.optim.adam.Adam'>\n","output_type":"stream"},{"name":"stderr","text":"INFO:pyhealth.trainer:Optimizer: <class 'torch.optim.adam.Adam'>\n","output_type":"stream"},{"name":"stdout","text":"Optimizer params: {'lr': 0.001}\n","output_type":"stream"},{"name":"stderr","text":"INFO:pyhealth.trainer:Optimizer params: {'lr': 0.001}\n","output_type":"stream"},{"name":"stdout","text":"Weight decay: 0.0\n","output_type":"stream"},{"name":"stderr","text":"INFO:pyhealth.trainer:Weight decay: 0.0\n","output_type":"stream"},{"name":"stdout","text":"Max grad norm: None\n","output_type":"stream"},{"name":"stderr","text":"INFO:pyhealth.trainer:Max grad norm: None\n","output_type":"stream"},{"name":"stdout","text":"Val dataloader: <torch.utils.data.dataloader.DataLoader object at 0x7bccf8939f90>\n","output_type":"stream"},{"name":"stderr","text":"INFO:pyhealth.trainer:Val dataloader: <torch.utils.data.dataloader.DataLoader object at 0x7bccf8939f90>\n","output_type":"stream"},{"name":"stdout","text":"Monitor: accuracy\n","output_type":"stream"},{"name":"stderr","text":"INFO:pyhealth.trainer:Monitor: accuracy\n","output_type":"stream"},{"name":"stdout","text":"Monitor criterion: max\n","output_type":"stream"},{"name":"stderr","text":"INFO:pyhealth.trainer:Monitor criterion: max\n","output_type":"stream"},{"name":"stdout","text":"Epochs: 5\n","output_type":"stream"},{"name":"stderr","text":"INFO:pyhealth.trainer:Epochs: 5\n","output_type":"stream"},{"name":"stdout","text":"Patience: None\n","output_type":"stream"},{"name":"stderr","text":"INFO:pyhealth.trainer:Patience: None\n","output_type":"stream"},{"name":"stdout","text":"\n","output_type":"stream"},{"name":"stderr","text":"INFO:pyhealth.trainer:\n","output_type":"stream"},{"output_type":"display_data","data":{"text/plain":"Epoch 0 / 5:   0%|          | 0/147 [00:00<?, ?it/s]","application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":"c99ec32d3f2b47cb971b77f5d203d9a7"}},"metadata":{}},{"name":"stdout","text":"--- Train epoch-0, step-147 ---\n","output_type":"stream"},{"name":"stderr","text":"INFO:pyhealth.trainer:--- Train epoch-0, step-147 ---\n","output_type":"stream"},{"name":"stdout","text":"loss: 0.1750\n","output_type":"stream"},{"name":"stderr","text":"INFO:pyhealth.trainer:loss: 0.1750\nEvaluation: 100%|██████████| 19/19 [00:00<00:00, 30.35it/s]","output_type":"stream"},{"name":"stdout","text":"--- Eval epoch-0, step-147 ---\n","output_type":"stream"},{"name":"stderr","text":"\nINFO:pyhealth.trainer:--- Eval epoch-0, step-147 ---\n","output_type":"stream"},{"name":"stdout","text":"accuracy: 0.9096\n","output_type":"stream"},{"name":"stderr","text":"INFO:pyhealth.trainer:accuracy: 0.9096\n","output_type":"stream"},{"name":"stdout","text":"f1: 0.9426\n","output_type":"stream"},{"name":"stderr","text":"INFO:pyhealth.trainer:f1: 0.9426\n","output_type":"stream"},{"name":"stdout","text":"loss: 0.4396\n","output_type":"stream"},{"name":"stderr","text":"INFO:pyhealth.trainer:loss: 0.4396\n","output_type":"stream"},{"name":"stdout","text":"New best accuracy score (0.9096) at epoch-0, step-147\n","output_type":"stream"},{"name":"stderr","text":"INFO:pyhealth.trainer:New best accuracy score (0.9096) at epoch-0, step-147\n","output_type":"stream"},{"name":"stdout","text":"\n","output_type":"stream"},{"name":"stderr","text":"INFO:pyhealth.trainer:\n","output_type":"stream"},{"output_type":"display_data","data":{"text/plain":"Epoch 1 / 5:   0%|          | 0/147 [00:00<?, ?it/s]","application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":"3a9c6c1f6529435890012406d0905aee"}},"metadata":{}},{"name":"stdout","text":"--- Train epoch-1, step-294 ---\n","output_type":"stream"},{"name":"stderr","text":"INFO:pyhealth.trainer:--- Train epoch-1, step-294 ---\n","output_type":"stream"},{"name":"stdout","text":"loss: 0.1154\n","output_type":"stream"},{"name":"stderr","text":"INFO:pyhealth.trainer:loss: 0.1154\nEvaluation: 100%|██████████| 19/19 [00:00<00:00, 33.40it/s]","output_type":"stream"},{"name":"stdout","text":"--- Eval epoch-1, step-294 ---\n","output_type":"stream"},{"name":"stderr","text":"\nINFO:pyhealth.trainer:--- Eval epoch-1, step-294 ---\n","output_type":"stream"},{"name":"stdout","text":"accuracy: 0.7918\n","output_type":"stream"},{"name":"stderr","text":"INFO:pyhealth.trainer:accuracy: 0.7918\n","output_type":"stream"},{"name":"stdout","text":"f1: 0.8799\n","output_type":"stream"},{"name":"stderr","text":"INFO:pyhealth.trainer:f1: 0.8799\n","output_type":"stream"},{"name":"stdout","text":"loss: 0.6418\n","output_type":"stream"},{"name":"stderr","text":"INFO:pyhealth.trainer:loss: 0.6418\n","output_type":"stream"},{"name":"stdout","text":"\n","output_type":"stream"},{"name":"stderr","text":"INFO:pyhealth.trainer:\n","output_type":"stream"},{"output_type":"display_data","data":{"text/plain":"Epoch 2 / 5:   0%|          | 0/147 [00:00<?, ?it/s]","application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":"1b3a776f4062464fa9429c9fad7fff4d"}},"metadata":{}},{"name":"stdout","text":"--- Train epoch-2, step-441 ---\n","output_type":"stream"},{"name":"stderr","text":"INFO:pyhealth.trainer:--- Train epoch-2, step-441 ---\n","output_type":"stream"},{"name":"stdout","text":"loss: 0.0900\n","output_type":"stream"},{"name":"stderr","text":"INFO:pyhealth.trainer:loss: 0.0900\nEvaluation: 100%|██████████| 19/19 [00:00<00:00, 33.17it/s]\n","output_type":"stream"},{"name":"stdout","text":"--- Eval epoch-2, step-441 ---\n","output_type":"stream"},{"name":"stderr","text":"INFO:pyhealth.trainer:--- Eval epoch-2, step-441 ---\n","output_type":"stream"},{"name":"stdout","text":"accuracy: 0.7730\n","output_type":"stream"},{"name":"stderr","text":"INFO:pyhealth.trainer:accuracy: 0.7730\n","output_type":"stream"},{"name":"stdout","text":"f1: 0.8705\n","output_type":"stream"},{"name":"stderr","text":"INFO:pyhealth.trainer:f1: 0.8705\n","output_type":"stream"},{"name":"stdout","text":"loss: 0.9299\n","output_type":"stream"},{"name":"stderr","text":"INFO:pyhealth.trainer:loss: 0.9299\n","output_type":"stream"},{"name":"stdout","text":"\n","output_type":"stream"},{"name":"stderr","text":"INFO:pyhealth.trainer:\n","output_type":"stream"},{"output_type":"display_data","data":{"text/plain":"Epoch 3 / 5:   0%|          | 0/147 [00:00<?, ?it/s]","application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":"cd55bf1d5be0418ca31f6b88a24253ed"}},"metadata":{}},{"name":"stdout","text":"--- Train epoch-3, step-588 ---\n","output_type":"stream"},{"name":"stderr","text":"INFO:pyhealth.trainer:--- Train epoch-3, step-588 ---\n","output_type":"stream"},{"name":"stdout","text":"loss: 0.0712\n","output_type":"stream"},{"name":"stderr","text":"INFO:pyhealth.trainer:loss: 0.0712\nEvaluation: 100%|██████████| 19/19 [00:00<00:00, 32.74it/s]\n","output_type":"stream"},{"name":"stdout","text":"--- Eval epoch-3, step-588 ---\n","output_type":"stream"},{"name":"stderr","text":"INFO:pyhealth.trainer:--- Eval epoch-3, step-588 ---\n","output_type":"stream"},{"name":"stdout","text":"accuracy: 0.9659\n","output_type":"stream"},{"name":"stderr","text":"INFO:pyhealth.trainer:accuracy: 0.9659\n","output_type":"stream"},{"name":"stdout","text":"f1: 0.9774\n","output_type":"stream"},{"name":"stderr","text":"INFO:pyhealth.trainer:f1: 0.9774\n","output_type":"stream"},{"name":"stdout","text":"loss: 0.1214\n","output_type":"stream"},{"name":"stderr","text":"INFO:pyhealth.trainer:loss: 0.1214\n","output_type":"stream"},{"name":"stdout","text":"New best accuracy score (0.9659) at epoch-3, step-588\n","output_type":"stream"},{"name":"stderr","text":"INFO:pyhealth.trainer:New best accuracy score (0.9659) at epoch-3, step-588\n","output_type":"stream"},{"name":"stdout","text":"\n","output_type":"stream"},{"name":"stderr","text":"INFO:pyhealth.trainer:\n","output_type":"stream"},{"output_type":"display_data","data":{"text/plain":"Epoch 4 / 5:   0%|          | 0/147 [00:00<?, ?it/s]","application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":"f81875652b174bcbbb73af7a9eabf776"}},"metadata":{}},{"name":"stdout","text":"--- Train epoch-4, step-735 ---\n","output_type":"stream"},{"name":"stderr","text":"INFO:pyhealth.trainer:--- Train epoch-4, step-735 ---\n","output_type":"stream"},{"name":"stdout","text":"loss: 0.0584\n","output_type":"stream"},{"name":"stderr","text":"INFO:pyhealth.trainer:loss: 0.0584\nEvaluation: 100%|██████████| 19/19 [00:00<00:00, 32.28it/s]","output_type":"stream"},{"name":"stdout","text":"--- Eval epoch-4, step-735 ---\n","output_type":"stream"},{"name":"stderr","text":"\nINFO:pyhealth.trainer:--- Eval epoch-4, step-735 ---\n","output_type":"stream"},{"name":"stdout","text":"accuracy: 0.9625\n","output_type":"stream"},{"name":"stderr","text":"INFO:pyhealth.trainer:accuracy: 0.9625\n","output_type":"stream"},{"name":"stdout","text":"f1: 0.9759\n","output_type":"stream"},{"name":"stderr","text":"INFO:pyhealth.trainer:f1: 0.9759\n","output_type":"stream"},{"name":"stdout","text":"loss: 0.0988\n","output_type":"stream"},{"name":"stderr","text":"INFO:pyhealth.trainer:loss: 0.0988\n","output_type":"stream"},{"name":"stdout","text":"Loaded best model\n","output_type":"stream"},{"name":"stderr","text":"INFO:pyhealth.trainer:Loaded best model\n","output_type":"stream"}],"execution_count":16},{"cell_type":"markdown","source":"### Evulation\n\nWe can also use Pyhealth trainer to evulate our accuracy.\n","metadata":{}},{"cell_type":"code","source":"print(\"=== Evaluation ===\")\ntrainer.evaluate(test_loader)\n","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-12-02T22:43:24.431616Z","iopub.execute_input":"2025-12-02T22:43:24.431882Z","iopub.status.idle":"2025-12-02T22:43:25.108973Z","shell.execute_reply.started":"2025-12-02T22:43:24.431864Z","shell.execute_reply":"2025-12-02T22:43:25.108367Z"}},"outputs":[{"name":"stdout","text":"=== Evaluation ===\n","output_type":"stream"},{"name":"stderr","text":"Evaluation: 100%|██████████| 19/19 [00:00<00:00, 28.52it/s]\n","output_type":"stream"},{"execution_count":31,"output_type":"execute_result","data":{"text/plain":"{'accuracy': 0.9385665529010239,\n 'f1': 0.958997722095672,\n 'loss': 0.18863907215295753}"},"metadata":{}}],"execution_count":31},{"cell_type":"code","source":"trainer.evaluate(val_loader)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-12-02T22:43:15.832305Z","iopub.execute_input":"2025-12-02T22:43:15.833126Z","iopub.status.idle":"2025-12-02T22:43:16.515309Z","shell.execute_reply.started":"2025-12-02T22:43:15.833098Z","shell.execute_reply":"2025-12-02T22:43:16.514621Z"}},"outputs":[{"name":"stderr","text":"Evaluation: 100%|██████████| 19/19 [00:00<00:00, 28.33it/s]\n","output_type":"stream"},{"execution_count":30,"output_type":"execute_result","data":{"text/plain":"{'accuracy': 0.9658703071672355,\n 'f1': 0.9773755656108597,\n 'loss': 0.12135807142936085}"},"metadata":{}}],"execution_count":30},{"cell_type":"code","source":"","metadata":{"trusted":true},"outputs":[],"execution_count":null}]}
\ No newline at end of file