Open ArseniuML opened 6 months ago
Hello, thanks for your issue but it is your misunderstanding. We convert Argoverse 2 data into KITTI format. In KITTI format, the 'training' folder actually contains both the train split and the validation split. So, when you test FSDv2 in Argoverse 2 validation split, the loader will load data from 'training/velodyne'.
I run dist_test.sh with Argo2 config and obtained the same metrics that were reported in article (AP and others). However, it seems that test.py loads files from train split:
why training/velodyne/1002000.bin is being loaded? Was it used for training?