- Linux
- Python 3.6+
- PyTorch 1.3+
- CUDA 9.2+ (If you build PyTorch from source, CUDA 9.0 is also compatible)
- NCCL 2+
- GCC 5+
- mmcv
- mmdetection
- BDD100K api
a. Create a conda virtual environment and activate it.
conda create -n pcan python=3.7 -y
conda activate pcan
b. Install PyTorch and torchvision following the official instructions, e.g.,
conda install pytorch torchvision -c pytorch
Note: Make sure that your compilation CUDA version and runtime CUDA version match. You can check the supported CUDA version for precompiled packages on the PyTorch website.
c. Install mmcv and mmdetection.
pip install mmcv-full==1.2.7
pip install mmdet==2.10.0
You can also refer to the offical instructions.
Note that mmdetection uses their forked version of pycocotools via the github repo instead of pypi for better compatibility. If you meet issues, you may need to re-install the cocoapi through
pip uninstall pycocotools
pip install mmpycocotools
d. Install mot metrics
pip install motmetrics
e. Install PCAN
python setup.py develop