Skip to content

An error while running the install.sh script #294

Description

@user67979868

I encountered an error while running the install.sh script. The error message is as follows:

/usr/local/cuda/bin/nvcc -I/root/miniconda3/envs/myenv37/lib/python3.7/site-packages/torch/lib/include -I/root/miniconda3/envs/myenv37/lib/python3.7/site-packages/torch/lib/include/TH -I/root/miniconda3/envs/myenv37/lib/python3.7/site-packages/torch/lib/include/THC -I/usr/local/cuda/include -I/root/miniconda3/envs/myenv37/include/python3.7m -c channelnorm_kernel.cu -o build/temp.linux-x86_64-cpython-37/channelnorm_kernel.o -gencode arch=compute_52,code=sm_52 -gencode arch=compute_60,code=sm_60 -gencode arch=compute_61,code=sm_61 -gencode arch=compute_70,code=sm_70 -gencode arch=compute_70,code=compute_70 -DTORCH_EXTENSION_NAME=channelnorm_cuda -D_GLIBCXX_USE_CXX11_ABI=0 --compiler-options '-fPIC' -std=c++11
channelnorm_kernel.cu:3:35: fatal error: ATen/cuda/CUDAContext.h: No such file or directory
compilation terminated.
error: command '/usr/local/cuda/bin/nvcc' failed with exit code 1

The error indicates that the compiler cannot find the file ATen/cuda/CUDAContext.h. Indeed, I have been unable to locate this file in my environment.

My current setup is: Ubuntu 16.04, PyTorch version 0.4.1, CUDA 9.2, and Python 3.7.

I would greatly appreciate any assistance you can provide in resolving this issue.

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions