otx.algorithms.action.utils.data#

Collection of utils for data in Action Task.

Functions

`find_label_by_name`(labels, name, domain)	Return label from name.
`load_cls_annotations`(ann_file, data_root)	Load annotation file to get video information.
`load_cls_dataset`(ann_file_path, ...[, ...])	Convert video annotation information into DatasetItemEntity.
`load_det_annotations`(ann_file, data_root)	Load AVA annotations.
`load_det_dataset`(ann_file_path, ...[, ...])	Convert video annotation information into DatasetItemEntity.
`parse_img_record`(img_records)	Accumulate and colligate bbox annotation info.

otx.algorithms.action.utils.data.find_label_by_name(labels: List[LabelEntity], name: str, domain: Domain)[source]#: Return label from name.

otx.algorithms.action.utils.data.load_cls_annotations(ann_file, data_root)[source]#: Load annotation file to get video information.

otx.algorithms.action.utils.data.load_cls_dataset(ann_file_path: str, data_root_dir: str, domain: Domain, subset: Subset = Subset.NONE, labels_list: List[LabelEntity] | None = None)[source]#: Convert video annotation information into DatasetItemEntity.

otx.algorithms.action.utils.data.load_det_annotations(ann_file, data_root)[source]#: Load AVA annotations.

otx.algorithms.action.utils.data.load_det_dataset(ann_file_path: str, data_root_dir: str, domain: Domain, subset: Subset = Subset.NONE, labels_list: List[LabelEntity] | None = None)[source]#: Convert video annotation information into DatasetItemEntity.

otx.algorithms.action.utils.data.parse_img_record(img_records)[source]#: Accumulate and colligate bbox annotation info.