After you configure the environment according to the steps, running
python train.py -s <path to scene> -m <path to save outputs> -r 8 --eval
it will prompt the following errors:
ModuleNotFoundError: No module named 'cv2'
ModuleNotFoundError: No module named 'scipy'
ModuleNotFoundError: No module named 'imageio'
ModuleNotFoundError: No module named 'skimage'
ModuleNotFoundError: No module named 'pytorch3d'
Just need to install these packages. This is not an issue. Note that compiling and installing pytorch3d will consume a large amount of memory. Please use
export MAX_JOBS=6
pip install "git+https://github.com/facebookresearch/pytorch3d.git@stable"
unset MAX_JOBS