Skip to content

Latest commit

 

History

History
59 lines (48 loc) · 1.34 KB

README.md

File metadata and controls

59 lines (48 loc) · 1.34 KB

Hand-by-hand guidance for Setup MinkowskiEngine 0.4.3 Env

Note ME 0.5.4 is much faster than ME 0.4.3, recommend to use ME 0.5.4

About us

We provide our environment info as follows:

  • Hardware: Nvidia GeForce RTX TITAN XP (12 GB) * 8
  • CUDA: 10.2
  • PyTorch: 1.6
  • Python: 3.7
  • GCC/G++: 7.5.0

Install basic python and pytorch environment

conda create -n cpcm_me043 python=3.7 -y
conda activate cpcm_me043
conda install pytorch==1.6.0 torchvision==0.7.0 cudatoolkit=10.2 -c pytorch -y

Setup CUDA env variable

May replace to your cuda path

export CUDA_HOME=/usr/local/cuda-10.2/
export PATH=${CUDA_HOME}/bin${PATH:+:${PATH}}
export CMAKE_CUDA_COMPILER=${CUDA_HOME}/bin/nvcc

Install other libraries, which is needed for MinkowskiEngine 0.4.3

conda install mkl mkl-include tbb -c intel -y

Install MinkowskiEngine 0.4.3

Get minkowski engine from official website and unzip it

mkdir env
cd env
wget https://github.com/NVIDIA/MinkowskiEngine/archive/refs/tags/v0.4.3.zip
unzip v0.4.3.zip
cd MinkowskiEngine-0.4.3

Install the MinkowskiEngine 0.4.3

python setup.py install

Go back to the root directory of CPCM

cd ../..

Install requirements.txt

Finally, install the requirements.txt

pip install -r requirements.txt -i https://pypi.tuna.tsinghua.edu.cn/simple