Owen-Liuyuxuan / visualDet3D

Official Repo for Ground-aware Monocular 3D Object Detection for Autonomous Driving / YOLOStereo3D: A Step Back to 2D for Efficient Stereo 3D Detection
https://owen-liuyuxuan.github.io/papers_reading_sharing.github.io/3dDetection/GroundAwareConvultion/
Apache License 2.0
361 stars 76 forks source link

Errors: 'KittiData' object has no attribute 'label' #57

Closed BitandPoly closed 2 years ago

BitandPoly commented 2 years ago

Hi,

Thanks very much for your wonderful contributions. It went smoothly when training the model, I faced some problems when testing with the checkpoint.

  1. The file './visualDet3D/workdirs/Stereo3D/output/validation/test/imdb.pkl' cannot be found. But I have one imdb.pkl in './visualDet3D/workdirs/Stereo3D/output/test/imdb.pkl'. Is the one in './test' the same as the one should be in './validation/test/imdb.pkl'? Thanks.
  2. I tried to copy the imdb.pkl into './validation/test', and it returned another error. The logs can be seen as follows (I have followed the instructions to set file paths): ./launchers/eval.sh ./config/Stereo.py 0 /home/qw/visualDet3D/workdirs/Stereo3D/checkpoint/Stereo3D_19.pth validation/test CUDA available: True Found evaluate function clean up the recorder directory of /home/qw/visualDet3D/workdirs/Stereo3D/output/validation/test/data rebuild /home/qw/visualDet3D/workdirs/Stereo3D/output/validation/test/data 0%| | 0/7518 [00:00<?, ?it/s]kitti.data in stereo_dataset.py is: <visualDet3D.data.kitti.kittidata.KittiData object at 0x7fa1279c37b8> 0%| | 0/7518 [00:00<?, ?it/s] Traceback (most recent call last): File "scripts/eval.py", line 55, in fire.Fire(main) File "/home/qw/anaconda3/envs/VD3D/lib/python3.6/site-packages/fire/core.py", line 141, in Fire component_trace = _Fire(component, args, parsed_flag_args, context, name) File "/home/qw/anaconda3/envs/VD3D/lib/python3.6/site-packages/fire/core.py", line 471, in _Fire target=component.name) File "/home/qw/anaconda3/envs/VD3D/lib/python3.6/site-packages/fire/core.py", line 681, in _CallAndUpdateTrace component = fn(*varargs, *kwargs) File "scripts/eval.py", line 52, in main evaluate_detection(cfg, detector, dataset, None, 0, result_path_split=split_to_test) File "/home/qw/anaconda3/envs/VD3D/lib/python3.6/site-packages/torch/autograd/grad_mode.py", line 26, in decorate_context return func(args, **kwargs) File "/home/qw/visualDet3D/visualDet3D/networks/pipelines/evaluators.py", line 85, in evaluate_kitti_obj test_one(cfg, index, dataset_val, model, test_func, backprojector, projector, result_path) File "/home/qw/visualDet3D/visualDet3D/networks/pipelines/evaluators.py", line 102, in test_one data = dataset[index] File "/home/qw/visualDet3D/visualDet3D/data/kitti/dataset/stereo_dataset.py", line 105, in getitem for obj in kitti_data.label: AttributeError: 'KittiData' object has no attribute 'label'

Looking forward to hearing from you. Any suggestions will be helpful.

Sincerely!