- Python >= 3.6
- PyTorch >= 1.5
- GCC >= 4.9
- cython:
pip install -U cython
- simplejson:
pip install simplejson
- PyAV:
conda install av -c conda-forge
- iopath:
pip install -U iopath
orconda install -c iopath iopath
- psutil:
pip install psutil
- OpenCV:
pip install opencv-python
- tensorboard:
pip install tensorboard
- moviepy: (optional, for visualizing video on tensorboard)
conda install -c conda-forge moviepy
orpip install moviepy
git clone https://github.com/facebookresearch/detectron2.git
python -m pip install -e detectron2
If have to reinstall, first:
rm -rf build/ **/*.so
git clone https://github.com/NVIDIA/apex
cd apex
pip install -v --disable-pip-version-check --no-cache-dir --global-option="--cpp_ext" --global-option="--cuda_ext" ./
For old env, (e.g. cuda9.0, torch.15), you should rollback apex to earlier version
# https://github.com/NVIDIA/apex/issues/802#issuecomment-618699214
git checkout f3a960f80244cf9e80558ab30f7f7e8cbf03c0a0
After having the above dependencies, run:
git clone https://github.com/ShoufaChen/WOO.git
cd WOO
python setup.py build develop
Now the installation is finished.