Is anyone evaluate successfully on your custom dataset?

[KevinDu1]

I evaluate my custom dataset using python ./pytorch/train.py evaluate --config_path=./configs/car.fhd.config --model_dir=/path/to/model_dir --measure_time=True --batch_size=1 After running, I visualize the results but there are little cars are detected, the 3D bboxes are not in the correct situation in the pointcloud. My own dataset don't have 3D labels. Does anyone know how I can get the correct results using the pretrained model?

[simaker]

which resolution has the used lidar?

[shiqi1994]

May I ask how did you evaluate your custom dataset? Thank you very much!