You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I saw the KITTI eval metric. In eval.py there is a function called do_eval. The eval_class returns a dictionary, which contains precision, recall, and aos. The problem is that after that, instead of calculating the area under the P_R curve, the function only passes ret["precision"] for the calculation final result. Is this Wrong? It does not count any of the FNs in this way. It does not use the recall values!!!
Here is the code I'm talking about:
Hi,
I saw the KITTI eval metric. In eval.py there is a function called do_eval. The eval_class returns a dictionary, which contains precision, recall, and aos. The problem is that after that, instead of calculating the area under the P_R curve, the function only passes ret["precision"] for the calculation final result. Is this Wrong? It does not count any of the FNs in this way. It does not use the recall values!!!
Here is the code I'm talking about:
https://github.com/open-mmlab/OpenPCDet/blob/master/pcdet/datasets/kitti/kitti_object_eval_python/eval.py#L590
The text was updated successfully, but these errors were encountered: