monailabel.tasks.infer.bundle module¶

class monailabel.tasks.infer.bundle.BundleConstants[source]¶

Bases: object

configs()[source]¶

Return type: Sequence[str]

key_bundle_root()[source]¶

Return type: str

key_detector()[source]¶

Return type: Sequence[str]

key_detector_ops()[source]¶

Return type: Sequence[str]

key_device()[source]¶

Return type: str

key_displayable_configs()[source]¶

Return type: Sequence[str]

key_inferer()[source]¶

Return type: Sequence[str]

key_network_def()[source]¶

Return type: str

key_postprocessing()[source]¶

Return type: Sequence[str]

key_preprocessing()[source]¶

Return type: Sequence[str]

metadata_json()[source]¶

Return type: str

model_pytorch()[source]¶

Return type: str

model_torchscript()[source]¶

Return type: str

class monailabel.tasks.infer.bundle.BundleInferTask(path, conf, const=None, type='', pre_filter=None, post_filter=[<class 'monai.transforms.io.dictionary.SaveImaged'>], extend_load_image=True, add_post_restore=True, dropout=0.0, load_strict=False, **kwargs)[source]¶

Bases: monailabel.tasks.infer.basic_infer.BasicInferTask

This provides Inference Engine for Monai Bundle.

Parameters

path (str) – Model File Path. Supports multiple paths to support versions (Last item will be picked as latest)
network – Model Network (e.g. monai.networks.xyz). None in case if you use TorchScript (torch.jit).
type (Union[str, InferType]) – Type of Infer (segmentation, deepgrow etc..)
labels – Labels associated to this Infer
dimension – Input dimension
description – Description
model_state_dict – Key for loading the model state from checkpoint
input_key – Input key for running inference
output_label_key – Output key for storing result/label of inference
output_json_key – Output key for storing result/label of inference
config – K,V pairs to be part of user config
load_strict – Load model in strict mode
roi_size – ROI size for scanning window inference
preload – Preload model/network on all available GPU devices
train_mode – Run in Train mode instead of eval (when network has dropouts)
skip_writer – Skip Writer and return data dictionary

detector(data=None)[source]¶

Return type: Optional[Callable, None]

inferer(data=None)[source]¶

Return type: Inferer

info()[source]¶

Return type: Dict[str, Any]

is_valid()[source]¶

Return type: bool

post_transforms(data=None)[source]¶

Provide List of post-transforms

Parameters

data –

current data dictionary/request which can be helpful to define the transforms per-request basis

For Example:

return [
    monai.transforms.EnsureChannelFirstd(keys='pred', channel_dim='no_channel'),
    monai.transforms.Activationsd(keys='pred', softmax=True),
    monai.transforms.AsDiscreted(keys='pred', argmax=True),
    monai.transforms.SqueezeDimd(keys='pred', dim=0),
    monai.transforms.ToNumpyd(keys='pred'),
    monailabel.interface.utils.Restored(keys='pred', ref_image='image'),
    monailabel.interface.utils.ExtremePointsd(keys='pred', result='result', points='points'),
    monailabel.interface.utils.BoundingBoxd(keys='pred', result='result', bbox='bbox'),
]

Return type

Sequence[Callable]

pre_transforms(data=None)[source]¶

Provide List of pre-transforms

Parameters

data –

current data dictionary/request which can be helpful to define the transforms per-request basis

For Example:

return [
    monai.transforms.LoadImaged(keys='image'),
    monai.transforms.EnsureChannelFirstd(keys='image', channel_dim='no_channel'),
    monai.transforms.Spacingd(keys='image', pixdim=[1.0, 1.0, 1.0]),
    monai.transforms.ScaleIntensityRanged(keys='image',
        a_min=-57, a_max=164, b_min=0.0, b_max=1.0, clip=True),
]

Return type

Sequence[Callable]

monailabel.tasks.infer.basic_infer module

monailabel.tasks.scoring package