niftynet.engine.windows_aggregator_classifier module

windows aggregator resize each item in a batch output and save as an image

class ClassifierSamplesAggregator(image_reader, name='image', output_path='./output', postfix='_niftynet_out')[source]

Bases: niftynet.engine.windows_aggregator_base.ImageWindowsAggregator

This class decodes each item in a batch by saving classification labels to a new image volume.

decode_batch(window, location)[source]

window holds the classifier labels location is a holdover from segmentation and may be removed in a later refactoring, but currently hold info about the stopping signal from the sampler