niftynet.evaluation.evaluation_application_driver module

This module defines a general procedure for running evaluations Example usage:

app_driver = EvaluationApplicationDriver() app_driver.initialise_application(system_param, input_data_param) app_driver.run_application()

system_param and input_data_param should be generated using:

class EvaluationApplicationDriver[source]

Bases: object

This class represents the application logic for evaluating a set of results inferred within NiftyNet (or externally generated)

initialise_application(workflow_param, data_param)[source]

This function receives all parameters from user config file, create an instance of application. :param workflow_param: a dictionary of user parameters, keys correspond to sections in the config file :param data_param: a dictionary of input image parameters, keys correspond to data properties to be used by image_reader :return:


This is the main application logic for evaluation. Computation of all metrics for all subjects is delegated to an Evaluator objects owned by the application object. The resulting metrics are aggregated as defined by the evaluation classes and output to one or more csv files (based on their ‘group_by’ headings). For example, per-subject metrics will be in one file, per-label-class metrics will be in another and per-subject-per-class will be in a third. :return: