avalanche.evaluation.metrics.detection.get_detection_api_from_dataset(dataset, supported_types: ~typing.Sequence[~typing.Tuple[str, ~typing.Union[~typing.Tuple[~typing.Type], ~typing.Type]]] = (('coco', <class 'pycocotools.coco.COCO'>), ('lvis_api', <class 'lvis.lvis.LVIS'>)), default_to_coco: bool = True, none_if_not_found=False)[source]

Adapted from: https://github.com/pytorch/vision/blob/main/references/detection/engine.py

  • dataset – The test dataset.

  • supported_types – The supported API types

  • default_to_coco – If True, if no API object can be found, the dataset will be converted to COCO.

  • none_if_not_found – If True, it will return None if no valid detection API object is found. Else, it will consider default_to_coco or will raise an error.


The detection object.