def postprocessing_pipeline_simplified(cache_dirpath, loader_mode):
if loader_mode == 'resize':
size_adjustment_function = partial(resize_image, target_size=ORIGINAL_SIZE)
else:
raise NotImplementedError
mask_resize = Step(name='mask_resize',
transformer=make_apply_transformer(size_adjustment_function,
output_name='resized_images',
apply_on=['images'],
n_threads=NUM_THREADS),
input_data=['network_output'],
adapter=Adapter({'images': E('network_output', 'mask_prediction'),
}))
binarizer = Step(name='binarizer',
transformer=make_apply_transformer(
partial(binarize, threshold=THRESHOLD),
output_name='binarized_images',
apply_on=['images'],
n_threads=NUM_THREADS),
input_steps=[mask_resize],
adapter=Adapter({'images': E(mask_resize.name, 'resized_images'),
}))
labeler = Step(name='labeler',
transformer=make_apply_transformer(
label,
output_name='labeled_images',
apply_on=['images'],
n_threads=NUM_THREADS),
input_steps=[binarizer],
adapter=Adapter({'images': E(binarizer.name, 'binarized_images'),
}))
labeler.set_mode_inference()
labeler.set_parameters_upstream({'experiment_directory': cache_dirpath,
'is_fittable': False
})
return labeler
def inference_segmentation_pipeline(config):
if config.general.loader_mode == 'resize_and_pad':
size_adjustment_function = partial(
postprocessing.crop_image,
target_size=config.general.original_size)
elif config.general.loader_mode == 'resize' or config.general.loader_mode == 'stacking':
size_adjustment_function = partial(
postprocessing.resize_image,
target_size=config.general.original_size)
else:
raise NotImplementedError
if USE_TTA:
preprocessing, tta_generator = pipelines.preprocessing_inference_tta(
config, model_name='segmentation_network')
segmentation_network = Step(
name='segmentation_network',
transformer=models.SegmentationModel(
**config.model['segmentation_network']),
input_steps=[preprocessing])
tta_aggregator = pipelines.aggregator('tta_aggregator',
segmentation_network,
tta_generator=tta_generator,
config=config.tta_aggregator)
prediction_renamed = Step(name='prediction_renamed',
transformer=IdentityOperation(),
input_steps=[tta_aggregator],
adapter=Adapter({
'mask_prediction':
E(tta_aggregator.name,
'aggregated_prediction')
}))
mask_resize = Step(name='mask_resize',
transformer=misc.make_apply_transformer(
size_adjustment_function,
output_name='resized_images',
apply_on=['images'],
n_threads=config.execution.num_threads,
),
input_steps=[prediction_renamed],
adapter=Adapter({
'images':
E(prediction_renamed.name, 'mask_prediction'),
}))
else:
preprocessing = pipelines.preprocessing_inference(
config, model_name='segmentation_network')
segmentation_network = misc.FineTuneStep(
name='segmentation_network',
transformer=models.SegmentationModel(
**config.model['segmentation_network']),
input_steps=[preprocessing],
)
mask_resize = Step(
name='mask_resize',
transformer=misc.make_apply_transformer(
size_adjustment_function,
output_name='resized_images',
apply_on=['images'],
n_threads=config.execution.num_threads,
),
input_steps=[segmentation_network],
adapter=Adapter({
'images':
E(segmentation_network.name, 'mask_prediction'),
}),
)
binarizer = Step(name='binarizer',
transformer=misc.make_apply_transformer(
partial(postprocessing.binarize,
threshold=config.thresholder.threshold_masks),
output_name='binarized_images',
apply_on=['images'],
n_threads=config.execution.num_threads),
input_steps=[mask_resize],
adapter=Adapter({
'images':
E(mask_resize.name, 'resized_images'),
}))
labeler = Step(name='labeler',
transformer=misc.make_apply_transformer(
postprocessing.label,
output_name='labeled_images',
apply_on=['images'],
n_threads=config.execution.num_threads,
),
input_steps=[binarizer],
adapter=Adapter({
'images':
E(binarizer.name, 'binarized_images'),
}))
mask_postprocessing = Step(name='mask_postprocessing',
transformer=misc.make_apply_transformer(
postprocessing.mask_postprocessing,
output_name='labeled_images',
apply_on=['images'],
n_threads=config.execution.num_threads,
),
input_steps=[labeler],
adapter=Adapter({
'images':
E(labeler.name, 'labeled_images'),
}))
mask_postprocessing.set_mode_inference()
mask_postprocessing.set_parameters_upstream({
'experiment_directory':
config.execution.experiment_dir,
'is_fittable':
False
})
segmentation_network.is_fittable = True
return mask_postprocessing