KPConv-PyTorch/INSTALL.md
2020-04-27 18:43:55 -04:00

55 lines
1.9 KiB
Markdown

# Installation instructions
## Ubuntu 18.04
* Make sure <a href="https://docs.nvidia.com/cuda/cuda-installation-guide-linux/index.html">CUDA</a> and <a href="https://docs.nvidia.com/deeplearning/sdk/cudnn-install/index.html">cuDNN</a> are installed. One configuration has been tested:
- PyTorch 1.4.0, CUDA 10.1 and cuDNN 7.6
* Ensure all python packages are installed :
sudo apt update
sudo apt install python3-dev python3-pip python3-tk
* Follow <a href="https://pytorch.org/get-started/locally/">PyTorch installation procedure</a>.
* Install the other dependencies with pip:
- numpy
- scikit-learn
- PyYAML
- matplotlib (for visualization)
- mayavi (for visualization)
- PyQt5 (for visualization)
* Compile the C++ extension modules for python located in `cpp_wrappers`. Open a terminal in this folder, and run:
sh compile_wrappers.sh
You should now be able to train Kernel-Point Convolution models
## Windows 10
* Make sure <a href="https://docs.nvidia.com/cuda/cuda-installation-guide-linux/index.html">CUDA</a> and <a href="https://docs.nvidia.com/deeplearning/sdk/cudnn-install/index.html">cuDNN</a> are installed. One configuration has been tested:
- PyTorch 1.4.0, CUDA 10.1 and cuDNN 7.5
* Follow <a href="https://pytorch.org/get-started/locally/">PyTorch installation procedure</a>.
* We used the PyCharm IDE to pip install all python dependencies (including PyTorch):
- numpy
- scikit-learn
- PyYAML
- matplotlib (for visualization)
- mayavi (for visualization)
- PyQt5 (for visualization)
* Compile the C++ extension modules for python located in `cpp_wrappers`. You just have to execute two .bat files:
cpp_wrappers/cpp_neighbors/build.bat
and
cpp_wrappers/cpp_subsampling/build.bat
You should now be able to train Kernel-Point Convolution models