Detection evaluation was updated so that eval_key + "_iou" no longer exists on false negatives which was causing a bug when computing detection mistakenness. This PR resolves that issue.
Previously this would fail:
import fiftyone.brain as fob
import fiftyone.zoo as foz
dataset = foz.load_zoo_dataset("quickstart")
fob.compute_mistakenness(dataset, "predictions", label_field="ground_truth")
Detection evaluation was updated so that
eval_key + "_iou"no longer exists on false negatives which was causing a bug when computing detection mistakenness. This PR resolves that issue.Previously this would fail: