Dockerfile or docker image will be much help

[project-splinter]

Dockerfile or docker image will be much help

[tianyilt]

Dockerimage:pytorch/pytorch:1.4-cuda10.1-cudnn7-runtime
download and run https://developer.nvidia.com/compute/cuda/10.1/Prod/local_installers/cuda_10.1.105_418.39_linux.run

export PATH=/usr/local/cuda-10.1/bin:$PATH
export CPATH=/usr/local/cuda-10.1/include:$CPATH
export LD_LIBRARY_PATH=/usr/local/cuda-10.1/lib64:$LD_LIBRARY_PATH

pip install -r requirement.txt
pip install PyOpenGL PyOpenGL_accelerate
apt install libgl1-mesa-glx

[YuliangXiu]

I tested demo using GCC 5.4.0, you can firstly try to downgrade the GCC version and then reinstall the ImplicitSegCUDA. Moreover, current ImplicitSegCUDA only supports pytorch==1.4.0, compile it with 1.5 or 1.6 will fail