How to use mini dataset

[PranavMandke0103]

Hey, I have been trying to run evaluation on the mini dataset for a while now. I could generate the pickle files by:

python tools/data_converter/vad_nuscenes_converter.py nuscenes --root-path ./data/nuscenes --out-dir ./data/nuscenes --extra-tag vad_nuscenes --version v1.0 --canbus ./data

And I ran the evaluation using:

CUDA_VISIBLE_DEVICES=0 python tools/test.py
/VAD/projects/configs/VAD/VAD_base_stage_2.py
/VAD/ckpts/vad_base.pth
--eval bbox \

But the L2 values and col values I get are way off from what is mentioned on the paper. Am I doing something wrong?

[csguanhu]

--version v1.0-mini

[MargeryLab]

--version v1.0-mini

But I still get an error when performing inference. Are there any other configurations that need to be modified?

[csguanhu]

filenotfoundError: boston-seaport.json
you'd better download map-file...