Dataset

Download WAYMO dataset and organize it as follows:

└── WAYMO_DATASET_PATH 
       ├── tfrecord_training       
       ├── tfrecord_validation   
       ├── tfrecord_testing 

Remember to change the path in [start.sh] to the WAYMO_DATASET_PATH path above.

export PYTHONPATH="${PYTHONPATH}:/home/trainer/rvcdet/RVCDet"
export PYTHONPATH="${PYTHONPATH}:/home/trainer/rvcdet/nuscenes-devkit/python-sdk"

Getting Started

Prepare train set

CUDA_VISIBLE_DEVICES=-1 python det3d/datasets/waymo/waymo_converter.py \
        --record_path '/home/trainer/rvcdet/RVCDet/data/Waymo/training/*.tfrecord' \  
        --root_path '/home/trainer/rvcdet/RVCDet/data/Waymo/train/'

Prepare validation set

CUDA_VISIBLE_DEVICES=-1 python det3d/datasets/waymo/waymo_converter.py \
        --record_path '/home/trainer/rvcdet/RVCDet/data/Waymo/tfrecord_validation/*.tfrecord' \  
        --root_path '/home/trainer/rvcdet/RVCDet/data/Waymo/val/'

Prepare testing set

CUDA_VISIBLE_DEVICES=-1 python det3d/datasets/waymo/waymo_converter.py \
        --record_path '/home/trainer/rvcdet/RVCDet/data/Waymo/tfrecord_testing/*.tfrecord' \ 
        --root_path '/home/trainer/rvcdet/RVCDet/data/Waymo/test/'

Create info files

Three Sweep Infos

# for train split
python tools/create_data.py waymo_data_prep \
        --root_path=data/Waymo \
        --split train \
        --nsweeps=3  

# for val split
python tools/create_data.py waymo_data_prep \
        --root_path=data/Waymo \
        --split val \
        --nsweeps=3 

# for test split
python tools/create_data.py waymo_data_prep \
        --root_path=data/Waymo \
        --split test \
        --nsweeps=3 

Training on WAYMO dataset:

For distributed training use:

python -m torch.distributed.launch --nproc_per_node=2 ./tools/train.py \
/home/josh/workspace/RVCDet/configs/waymo/pp/waymo_rvdet_pp_two_pfn_stride1_3x.py \
        --work_dir waymo_exp/rvdet-pp-dyn

For single device training use:

CUDA_VISIBLE_DEVICES=1 python tools/train.py \
/home/josh/workspace/RVCDet/configs/waymo/pp/waymo_rvdet_pp_two_pfn_stride1_3x.py \
        --work_dir waymo_exp/rvdet-pp-dyn

Validation on WAYMO dataset:

python tools/dist_test.py \
/home/josh/workspace/RVCDet/configs/waymo/pp/waymo_rvdet_pp_two_pfn_stride1_3x.py \
        --work_dir waymo_exp/rvdet-pp-dyn \
        --checkpoint waymo_exp/rvdet-pp-dyn/latset.pth  \
        --speed_test \
        --gpus 1

Testing on WAYMO dataset:

python tools/dist_test.py \
/home/josh/workspace/RVCDet/configs/waymo/pp/waymo_rvdet_pp_two_pfn_stride1_3x.py \
        --work_dir waymo_exp/rvdet-pp-dyn \
        --checkpoint waymo_exp/rvdet-pp-dyn/latset.pth  \
        --speed_test \
        --testset --gpus 1

Turn on classification module on WAYMO dataset :

To turn on classification module add classification flag.

Example for validation on WAYMO:

python tools/dist_test.py \
/home/josh/workspace/RVCDet/configs/waymo/pp/waymo_rvdet_pp_two_pfn_stride1_3x.py \
        --work_dir waymo_exp/rvdet-pp-dyn \
        --checkpoint waymo_exp/rvdet-pp-dyn/latset.pth  \
        --speed_test \
        --gpus 1 --classification

Example for testing on WAYMO:

python tools/dist_test.py \
/home/josh/workspace/RVCDet/configs/waymo/pp/waymo_rvdet_pp_two_pfn_stride1_3x.py \
        --work_dir waymo_exp/rvdet-pp-dyn \
        --checkpoint waymo_exp/rvdet-pp-dyn/latset.pth  \
        --speed_test \
        --testset --gpus 1 --classification