Closed rockywind closed 9 months ago
Sorry for the late reply. Have you solved this problem? And have you made any modifications to the code?
Hi, I have not solve it. I follow the guide from the link.
The data was download from the google drive you provide.
I am confuse that what's the cls function.
This is the fsd_base_vehicle.yaml config.
bin_path: /SHFP12/xiaoquan.wang/01_bev/SST/01_work_dir/fsd6f6e_vehicle_full_train.bin
val_bin_path: /SHFP12/xiaoquan.wang/01_bev/SST/01_work_dir/fsd_base_vehicle_val.bin
data_root: /SHFP12/xiaoquan.wang/01_bev/SST/data/public_datasets/waymo_bev_ctrl/
exist_ok: False
# change the split for your need
# split: training
split: val
# split: test
type: [1,] # 1:vehicle 2:pedestrian 4: cyclist
box:
extra_width: 1
selection:
mode: random
size: 1.0
candidate:
affinity_thresh: 0.5
This is my generate_candidates.py config.
parser = argparse.ArgumentParser()
parser.add_argument('--config', default="/SHFP12/xiaoquan.wang/01_bev/SST/tools/ctrl/data_configs/fsd_base_vehicle_my.yaml",type=str)
parser.add_argument('--gt-bin-path', type=str, default='/SHFP12/xiaoquan.wang/01_bev/SphereFormer/data/public_datasets/waymo/waymo_format/train_gt.bin')
parser.add_argument('--process', type=int, default=1)
args = parser.parse_args()
The cls
is LiDARTracklet class itself. So the input arguments should be consistent with its __init__
func.
Could you please show what arguments are passed into the init?
感谢您的帮助,我重新走一遍流程,现在解决了。
Hi, thank you for sharing the great work! I met the error below when I run the script![image](https://github.com/tusen-ai/SST/assets/14838781/41da35d1-01c3-4875-9f42-9a326b4510dc)
python ./tools/ctrl/generate_candidates.py ./tools/ctrl/data_configs/fsd_base_vehicle.yaml --process 8