Modified from det3d's original document.
- Both Spconv 1.x and 2.x work.
- Recent pytorch/spconv/cuda version will be faster and consume less memory.
- If you have problem installing apex, you can change the apex syncbn to torch's native sync bn at https://github.com/tianweiy/CenterPoint/blob/3fd0b8745b77575cb9810035aafc76796613f942/det3d/torchie/apis/train.py#L268.
we have tested the following versions of OS and softwares:
- OS: Ubuntu 16.04/18.04
- Python: 3.6.5/3.7.10
- PyTorch: 1.1/1.9/1.10.1
- spconv: 1.0/1.2.1/master
- CUDA: 10.0/11.1
# basic python libraries
conda create --name centerpoint python=3.6
conda activate centerpoint
conda install pytorch==1.1.0 torchvision==0.3.0 cudatoolkit=10.0 -c pytorch
git clone https://github.com/tianweiy/CenterPoint.git
cd CenterPoint
pip install -r requirements.txt
# add CenterPoint to PYTHONPATH by adding the following line to ~/.bashrc (change the path accordingly)
export PYTHONPATH="${PYTHONPATH}:PATH_TO_CENTERPOINT"git clone https://github.com/tianweiy/nuscenes-devkit
# add the following line to ~/.bashrc and reactivate bash (remember to change the PATH_TO_NUSCENES_DEVKIT value)
export PYTHONPATH="${PYTHONPATH}:PATH_TO_NUSCENES_DEVKIT/python-sdk"# set the cuda path(change the path to your own cuda location)
export PATH=/usr/local/cuda-10.0/bin:$PATH
export CUDA_PATH=/usr/local/cuda-10.0
export CUDA_HOME=/usr/local/cuda-10.0
export LD_LIBRARY_PATH=/usr/local/cuda-10.0/lib64:$LD_LIBRARY_PATH
# Rotated NMS
cd ROOT_DIR/det3d/ops/iou3d_nms
python setup.py build_ext --inplace
# Deformable Convolution (Optional and only works with old torch versions e.g. 1.1)
cd ROOT_DIR/det3d/ops/dcn
python setup.py build_ext --inplacegit clone https://github.com/NVIDIA/apex
cd apex
git checkout 5633f6 # recent commit doesn't build in our system
pip install -v --no-cache-dir --global-option="--cpp_ext" --global-option="--cuda_ext" ./sudo apt-get install libboost-all-dev
git clone https://github.com/traveller59/spconv.git --recursive
cd spconv && git checkout 7342772
python setup.py bdist_wheel
cd ./dist && pip install *