niftynet.application.classification_application module

This module defines an image-level classification application that maps from images to scalar, multi-class labels.

This class is instantiated and initalized by the application_driver.

class ClassificationApplication(net_param, action_param, action)

Bases: niftynet.application.base_application.BaseApplication

This class defines an application for image-level classification problems mapping from images to scalar labels.

This is the application class to be instantiated by the driver and referred to in configuration files.

Although structurally similar to segmentation, this application supports different samplers/aggregators (because patch-based processing is not appropriate), and monitoring metrics.

REQUIRED_CONFIG_SECTION = 'CLASSIFICATION'

initialise_dataset_loader(data_param=None, task_param=None, data_partitioner=None)

this function initialise self.readers

Parameters:

• data_param – input modality specifications
• task_param – contains task keywords for grouping data_param
• data_partitioner – specifies train/valid/infer splitting if needed

Returns:

initialise_resize_sampler()

initialise_aggregator()

initialise_sampler()

Samplers take self.reader as input and generates sequences of ImageWindow that will be fed to the networks

This function sets self.sampler.

initialise_network()

This function create an instance of network and sets self.net

Returns:None

add_confusion_matrix_summaries_(outputs_collector, net_out, data_dict)

This method defines several monitoring metrics that are derived from the confusion matrix

connect_data_and_network(outputs_collector=None, gradients_collector=None)

Adding sampler output tensor and network tensors to the graph.

Parameters:

• outputs_collector –
• gradients_collector –

Returns:

interpret_output(batch_output)

Implement output interpretations, e.g., save to hard drive cache output windows.

Parameters:batch_output – outputs by running the tf graph Returns:True indicates the driver should continue the loop False indicates the drive should stop

initialise_evaluator(eval_param)

add_inferred_output(data_param, task_param)