AgML is a centralized framework for agricultural machine learning. AgML provides access to public agricultural datasets for common agricultural deep learning tasks, with standard benchmarks and pretrained models, as well the ability to generate synthetic data and annotations.
Changed boxes.py because _resolve_coco_annotations can return a dictionary with key 'bboxes' or 'bbox'.
When running the bounding box visualizer, the fixed line gave a KeyError for datasets like mango_detection_australia, plant_doc_detection, and my new new dataset because it assumed that _resolve_coco_annotations returned a dictionary with key 'bboxes' while it really returned a dictionary with key 'bbox.'
The consistent use of 'bbox' and 'bboxes' by annotation files and functions would make more sense but that would require more extensive changes. Maybe this is just a temporary fix.
Changed boxes.py because
_resolve_coco_annotations
can return a dictionary with key 'bboxes' or 'bbox'.When running the bounding box visualizer, the fixed line gave a KeyError for datasets like mango_detection_australia, plant_doc_detection, and my new new dataset because it assumed that
_resolve_coco_annotations
returned a dictionary with key 'bboxes' while it really returned a dictionary with key 'bbox.'The consistent use of 'bbox' and 'bboxes' by annotation files and functions would make more sense but that would require more extensive changes. Maybe this is just a temporary fix.