{
    "imports": [
        "$import glob",
        "$import os",
        "$import torch",
        "$import numpy as np"
    ],
    "bundle_root": "/workspace/bundle/endoscopic_tool_segmentation",
    "ckpt_dir": "$@bundle_root + '/models'",
    "output_dir": "$@bundle_root + '/eval'",
    "dataset_dir": "/workspace/data/endoscopic_tool_dataset",
    "images": "$list(sorted(glob.glob(os.path.join(@dataset_dir,'train', '*[!seg].jpg'))))",
    "labels": "$[x.replace('.jpg', '_seg.jpg') for x in @images]",
    "val_images": "$list(sorted(glob.glob(os.path.join(@dataset_dir,'valid', '*[!seg].jpg'))))",
    "val_labels": "$[x.replace('.jpg', '_seg.jpg') for x in @val_images]",
    "val_interval": 1,
    "device": "$torch.device('cuda:0' if torch.cuda.is_available() else 'cpu')",
    "network_def": {
        "_target_": "FlexibleUNet",
        "in_channels": 3,
        "out_channels": 2,
        "backbone": "efficientnet-b0",
        "spatial_dims": 2,
        "pretrained": true,
        "is_pad": false
    },
    "network": "$@network_def.to(@device)",
    "loss": {
        "_target_": "DiceLoss",
        "include_background": false,
        "to_onehot_y": true,
        "softmax": true
    },
    "optimizer": {
        "_target_": "torch.optim.Adam",
        "params": "$@network.parameters()",
        "lr": 0.0001
    },
    "train": {
        "deterministic_transforms": [
            {
                "_target_": "LoadImaged",
                "keys": [
                    "image",
                    "label"
                ]
            },
            {
                "_target_": "ToTensord",
                "keys": [
                    "image",
                    "label"
                ]
            },
            {
                "_target_": "AsChannelFirstd",
                "keys": "image"
            },
            {
                "_target_": "AddChanneld",
                "keys": "label"
            },
            {
                "_target_": "Resized",
                "keys": [
                    "image",
                    "label"
                ],
                "spatial_size": [
                    736,
                    480
                ],
                "mode": [
                    "bilinear",
                    "nearest"
                ]
            },
            {
                "_target_": "ScaleIntensityd",
                "keys": [
                    "image",
                    "label"
                ]
            }
        ],
        "random_transforms": [
            {
                "_target_": "RandRotated",
                "keys": [
                    "image",
                    "label"
                ],
                "range_x": "$np.pi",
                "prob": 0.8,
                "mode": [
                    "bilinear",
                    "nearest"
                ]
            },
            {
                "_target_": "RandZoomd",
                "keys": [
                    "image",
                    "label"
                ],
                "min_zoom": 0.8,
                "max_zoom": 1.2,
                "prob": 0.2,
                "mode": [
                    "bilinear",
                    "nearest"
                ]
            }
        ],
        "preprocessing": {
            "_target_": "Compose",
            "transforms": "$@train#deterministic_transforms + @train#random_transforms"
        },
        "dataset": {
            "_target_": "CacheDataset",
            "data": "$[{'image': i, 'label': l} for i, l in zip(@images, @labels)]",
            "transform": "@train#preprocessing",
            "cache_rate": 0.1,
            "num_workers": 4
        },
        "dataloader": {
            "_target_": "DataLoader",
            "dataset": "@train#dataset",
            "batch_size": 8,
            "shuffle": true,
            "num_workers": 4
        },
        "inferer": {
            "_target_": "SimpleInferer"
        },
        "handlers": [
            {
                "_target_": "ValidationHandler",
                "validator": "@validate#evaluator",
                "epoch_level": true,
                "interval": "@val_interval"
            },
            {
                "_target_": "StatsHandler",
                "tag_name": "train_loss",
                "output_transform": "$monai.handlers.from_engine(['loss'], first=True)"
            },
            {
                "_target_": "TensorBoardStatsHandler",
                "log_dir": "@output_dir",
                "tag_name": "train_loss",
                "output_transform": "$monai.handlers.from_engine(['loss'], first=True)"
            }
        ],
        "key_metric": {
            "train_iou": {
                "_target_": "MeanIoUHandler",
                "include_background": false,
                "reduction": "mean",
                "output_transform": "$monai.handlers.from_engine(['pred', 'label'])"
            }
        },
        "postprocessing": {
            "_target_": "Compose",
            "transforms": [
                {
                    "_target_": "AsDiscreted",
                    "argmax": [
                        true,
                        false
                    ],
                    "to_onehot": [
                        2,
                        2
                    ],
                    "keys": [
                        "pred",
                        "label"
                    ]
                }
            ]
        },
        "trainer": {
            "_target_": "SupervisedTrainer",
            "max_epochs": 60,
            "device": "@device",
            "train_data_loader": "@train#dataloader",
            "network": "@network",
            "loss_function": "@loss",
            "optimizer": "@optimizer",
            "inferer": "@train#inferer",
            "postprocessing": "@train#postprocessing",
            "key_train_metric": "@train#key_metric",
            "train_handlers": "@train#handlers"
        }
    },
    "validate": {
        "preprocessing": {
            "_target_": "Compose",
            "transforms": "%train#deterministic_transforms"
        },
        "dataset": {
            "_target_": "CacheDataset",
            "data": "$[{'image': i, 'label': l} for i, l in zip(@val_images, @val_labels)]",
            "transform": "@validate#preprocessing",
            "cache_rate": 0.1
        },
        "dataloader": {
            "_target_": "DataLoader",
            "dataset": "@validate#dataset",
            "batch_size": 8,
            "shuffle": false,
            "num_workers": 4
        },
        "inferer": {
            "_target_": "SimpleInferer"
        },
        "postprocessing": {
            "_target_": "Compose",
            "transforms": "%train#postprocessing"
        },
        "handlers": [
            {
                "_target_": "StatsHandler",
                "iteration_log": false
            },
            {
                "_target_": "TensorBoardStatsHandler",
                "log_dir": "@output_dir",
                "iteration_log": false
            },
            {
                "_target_": "CheckpointSaver",
                "save_dir": "@ckpt_dir",
                "save_dict": {
                    "model": "@network"
                },
                "save_key_metric": true,
                "key_metric_filename": "model.pt"
            }
        ],
        "additional_metrics": {
            "val_mean_dice": {
                "_target_": "MeanDice",
                "include_background": false,
                "reduction": "mean",
                "output_transform": "$monai.handlers.from_engine(['pred', 'label'])"
            }
        },
        "key_metric": {
            "val_iou": {
                "_target_": "MeanIoUHandler",
                "include_background": false,
                "reduction": "mean",
                "output_transform": "$monai.handlers.from_engine(['pred', 'label'])"
            }
        },
        "evaluator": {
            "_target_": "SupervisedEvaluator",
            "device": "@device",
            "val_data_loader": "@validate#dataloader",
            "network": "@network",
            "inferer": "@validate#inferer",
            "postprocessing": "@validate#postprocessing",
            "key_val_metric": "@validate#key_metric",
            "additional_metrics": "@validate#additional_metrics",
            "val_handlers": "@validate#handlers"
        }
    },
    "training": [
        "$monai.utils.set_determinism(seed=0)",
        "$setattr(torch.backends.cudnn, 'benchmark', True)",
        "$@train#trainer.run()"
    ]
}