def get_segmentation_transform(self) -> ModelTransformsPerExecutionMode:
if self.imaging_feature_type in [ImagingFeatureType.Segmentation, ImagingFeatureType.ImageAndSegmentation]:
return ModelTransformsPerExecutionMode(
train=ImageTransformationPipeline(
transforms=[RandomAffine(10), ColorJitter(0.2)],
use_different_transformation_per_channel=True))
return ModelTransformsPerExecutionMode()
def get_image_sample_transforms(self) -> ModelTransformsPerExecutionMode:
if self.use_combined_model:
return ModelTransformsPerExecutionMode(
train=ScalarItemAugmentation(
transform=RandAugmentSlice(use_joint_channel_transformation=False,
is_transformation_for_segmentation_maps=True)))
else:
return ModelTransformsPerExecutionMode()
def get_image_transform(self) -> ModelTransformsPerExecutionMode:
if self.use_combined_model:
return ModelTransformsPerExecutionMode(
train=ImageTransformationPipeline(
transforms=[RandomAffine(degrees=30, translate=(0.1, 0.1), shear=15),
ColorJitter(brightness=0.2)]))
else:
return ModelTransformsPerExecutionMode()
def get_image_transform(self) -> ModelTransformsPerExecutionMode:
"""
Get transforms to perform on image samples for each model execution mode.
"""
if self.imaging_feature_type in [ImagingFeatureType.Image, ImagingFeatureType.ImageAndSegmentation]:
return ModelTransformsPerExecutionMode(
train=ImageTransformationPipeline(
transforms=[RandomAffine(10), ColorJitter(0.2)],
use_different_transformation_per_channel=True))
return ModelTransformsPerExecutionMode()
def get_cropped_image_sample_transforms(
self) -> ModelTransformsPerExecutionMode:
"""
Get transforms to perform on cropped samples for each model execution mode.
By default no transformation is performed.
"""
return ModelTransformsPerExecutionMode()
def get_image_sample_transforms(self) -> ModelTransformsPerExecutionMode:
"""
Get transforms to perform on samples for each model execution mode.
By default only no transformation is performed.
For data augmentation, specify a Compose3D for the training execution mode.
"""
return ModelTransformsPerExecutionMode()
def get_image_sample_transforms(self) -> ModelTransformsPerExecutionMode:
"""
Get transforms to perform on image samples for each model execution mode.
"""
return ModelTransformsPerExecutionMode(
train=ScalarItemAugmentation(
RandAugmentSlice(is_transformation_for_segmentation_maps=(
self.imaging_feature_type == ImagingFeatureType.Segmentation
or self.imaging_feature_type == ImagingFeatureType.ImageAndSegmentation))))
def get_image_transform(self) -> ModelTransformsPerExecutionMode:
config = load_yaml_augmentation_config(path_linear_head_augmentation_cxr)
train_transforms = Compose(
[DicomPreparation(), create_transforms_from_config(config, apply_augmentations=True)])
val_transforms = Compose(
[DicomPreparation(), create_transforms_from_config(config, apply_augmentations=False)])
return ModelTransformsPerExecutionMode(train=train_transforms,
val=val_transforms,
test=val_transforms)
def get_scalar_item_transform(self) -> ModelTransformsPerExecutionMode:
from InnerEye.ML.dataset.scalar_dataset import ScalarItemAugmentation
image_transform = self.get_image_transform()
segmentation_transform = self.get_segmentation_transform()
return ModelTransformsPerExecutionMode(
train=ScalarItemAugmentation(image_transform.train,
segmentation_transform.train),
val=ScalarItemAugmentation(image_transform.val,
segmentation_transform.val),
test=ScalarItemAugmentation(image_transform.test,
segmentation_transform.test))
def get_full_image_sample_transforms(self) -> ModelTransformsPerExecutionMode:
"""
Get transforms to perform on full image samples for each model execution mode.
By default only PhotometricNormalization is performed.
"""
from InnerEye.ML.utils.transforms import Compose3D
from InnerEye.ML.photometric_normalization import PhotometricNormalization
photometric_transformation = Compose3D(transforms=[PhotometricNormalization(self, use_gpu=False)])
return ModelTransformsPerExecutionMode(train=photometric_transformation,
val=photometric_transformation,
test=photometric_transformation)
def get_segmentation_transform(self) -> ModelTransformsPerExecutionMode:
"""
Get transforms to apply on segmentations maps inputs for each model execution mode.
By default only no transformation is performed.
"""
return ModelTransformsPerExecutionMode()
def get_image_transform(self) -> ModelTransformsPerExecutionMode:
"""
Get transforms to apply to images for each model execution mode.
By default only no transformation is performed.
"""
return ModelTransformsPerExecutionMode()
