SDCoT Installation (CUDA 10.2 Python 3.8)

1. Clone the repository, git clone https://github.com/Na-Z/SDCoT.git.
2. Go to the SDCoT folder, cd SDCoT.
3. Replace the following in the pointnet2 folder to use with CUDA 10.2, Pytorch 1.9.1 or higher. You can use the search tab on the right of VSCode for searching and replacing.

• AT_CHECK -> TORCH_CHECK.
• type().is_cuda() -> is_cuda().
• .data<float>() -> .data_ptr<float>(), and .data<int>() -> .data_ptr<int>(). Beware of the . at front.
• Use absolute path for #include ... .h if necessary.

4. Create the virtual environment, python3 -m venv venv.
5. Activate the virtual environment, source venv/bin/activate.
6. Update packages, pip3 install --upgrade pip setuptools wheel
7. Install PyTorch, pip3 install torch==1.9.1+cu102 torchvision==0.10.1+cu102 torchaudio==0.9.1 -f https://download.pytorch.org/whl/torch_stable.html
8. Install Pointnet2, cd pointnet2/ and python3 setup.py install.
9. Install other packages, pip3 install -r requirements_2.txt, which removes package versions and correct the name for scikit-learn.
10. Debug setuptools version, pip3 uninstall setuptools and pip3 install setuptools==59.5.0. Referring to link.