Performance Summary

base_report()

It summarizes all the performance scores of the model at all levels of granularity: it provides the overall scores, the per-category scores and the per-property scores.

Parameters

metrics

list, optional

List of evaluation metrics to be included in the analysis. If not specified, all the evaluation metrics are included.
(default is None)

categories

list, optional

List of categories to be included in the analysis. If not specified, all the categories are included.
(default is None)

properties

bool, optional

List of properties to be included in the analysis. If not specified, all the properties are included.
(default is None)

show_categories

bool, optional

Indicates whether the categories should be included in the report.
(default is True)

show_properties

bool, optional

Indicates whether the properties should be included in the report.
(default is True)

include_reliability

bool, optional

Indicates whether the 'ece' and 'mce' should be included in the report.
(default is True)

Example

Classification

from odin.classes import AnalyzerClassification

my_analyzer = AnalyzerClassification("my_classifier_name", my_classification_dataset)
my_analyzer.base_report()

Localization

from odin.classes import AnalyzerLocalization

my_analyzer = AnalyzerLocalization("my_detector_name", my_localization_dataset)
my_analyzer.base_report()

N.B. For localization models, the Accuracy and Error Rate evaluation metrics are not supported.

Models comparison

Example

Classification

from odin.classes import ComparatorClassification

my_comparator = ComparatorClassification(dataset_gt_param, classification_type, models_proposals)
my_comparator.base_report()

Localization

from odin.classes import ComparatorLocalization

my_comparator = ComparatorLocalization(dataset_gt_param, task_type, models_proposals)
my_comparator.base_report()

N.B. For localization models, the Accuracy and Error Rate evaluation metrics are not supported.

Tasks supported