nnUNet config

nnUNet configuration template

A JSON template used to set the configuration parameters for experiments based on nnUNet framework.
properties
Experiment Name	String describing the experiment, usually related to the selected DL framework/architecture. Example: “nnUNet_3D_fullres”
	type	string
Seed	Random seed number used when randomizing events and actions. Example: 12345
	type	integer
label_suffix	File suffix (including file extension) of the files containing the label map. Example: “_mask.nii.gz”
	type	string
Modalities	Dictionary listing the input modalities. The key-value pair contains the file suffix as key and the modality name as value. Example: “_image.nii.gz”: “CT”
	type	object
label_dict	Dictionary describing the labels (excluding the background). The key-value pair contains the label description as key and the label value as value (starting from 0). Example: “Background”:0
	type	object
n_folds	Number of folds to run cross-validation. Example: 5.
	type	integer
FileExtension	String for the file extension of the dataset. “Example: .nii.gz”
	type	string
RegionClassOrder	List of the classes in the order they should be used in the region-based training. Example: [ 1, 2, 3]
	type	array

{
      "Seed": 12345,
      "label_suffix": "-seg.nii.gz",
      "Modalities": {
        "-t1c.nii.gz": "MRI",
        "-t1n.nii.gz": "MRI",
        "-t2f.nii.gz": "MRI",
        "-t2w.nii.gz": "MRI"
      },
      "label_dict": {
        "background": 0,
        "NCR": 1,
        "ED": 2,
        "ET": 3
      },
      "n_folds": 5,
      "Experiment Name": "BraTS_nnUNet_3D_fullres",
      "FileExtension": ".nii.gz",
}