Hi all,
I have been trying to make a C++ extension for Pytorch for a project of mine.
I have setup a conda environment and built pytorch and torchvision from source.
Here are the dependencies and packages installed in the conda environment:
# packages in environment at /home/balaji5199/anaconda3/envs/cuda_en:
#
# Name Version Build Channel
_libgcc_mutex 0.1 main
blas 1.0 mkl
brotlipy 0.7.0 py38h27cfd23_1003
bzip2 1.0.8 h7b6447c_0
ca-certificates 2021.1.19 h06a4308_1
certifi 2020.12.5 py38h06a4308_0
cffi 1.14.5 py38h261ae71_0
chardet 4.0.0 py38h06a4308_1003
cmake 3.19.6 h973ab73_0
cryptography 3.3.1 py38h3c74f83_1
cudatoolkit 10.2.89 hfd86e86_1
cudnn 7.6.5 cuda10.2_0
dataclasses 0.8 pyh6d0b6a4_7
expat 2.2.10 he6710b0_2
future 0.18.2 py38_1
idna 2.10 pyhd3eb1b0_0
intel-openmp 2020.2 254
krb5 1.18.2 h173b8e3_0
ld_impl_linux-64 2.33.1 h53a641e_7
libcurl 7.71.1 h20c2e04_1
libedit 3.1.20191231 h14c3975_1
libffi 3.3 he6710b0_2
libgcc-ng 9.1.0 hdf63c60_0
libssh2 1.9.0 h1ba5d50_1
libstdcxx-ng 9.1.0 hdf63c60_0
libuv 1.40.0 h7b6447c_0
lz4-c 1.9.3 h2531618_0
magma-cuda102 2.5.2 1 pytorch
mkl 2020.2 256
mkl-include 2020.2 256
mkl-service 2.3.0 py38he904b0f_0
mkl_fft 1.3.0 py38h54f3939_0
mkl_random 1.1.1 py38h0573a6f_0
ncurses 6.2 he6710b0_1
ninja 1.10.2 py38hff7bd54_0
numpy 1.19.2 py38h54aff64_0
numpy-base 1.19.2 py38hfa32c7d_0
nvcc_linux-64 10.2 h1a5f58c_12 conda-forge
openssl 1.1.1j h27cfd23_0
pip 21.0.1 py38h06a4308_0
pycparser 2.20 py_2
pyopenssl 20.0.1 pyhd3eb1b0_1
pysocks 1.7.1 py38h06a4308_0
python 3.8.8 hdb3f193_4
pyyaml 5.4.1 py38h27cfd23_1
readline 8.1 h27cfd23_0
requests 2.25.1 pyhd3eb1b0_0
rhash 1.4.1 h3c74f83_1
sed 4.8 he412f7d_0 conda-forge
setuptools 52.0.0 py38h06a4308_0
six 1.15.0 py38h06a4308_0
sqlite 3.33.0 h62c20be_0
tk 8.6.10 hbc83047_0
typing_extensions 3.7.4.3 pyha847dfd_0
urllib3 1.26.3 pyhd3eb1b0_0
wheel 0.36.2 pyhd3eb1b0_0
xz 5.2.5 h7b6447c_0
yaml 0.2.5 h7b6447c_0
zlib 1.2.11 h7b6447c_3
zstd 1.4.5 h9ceee32_0
When I try to run a code where I use PYBIND11 to extend a c++ code to pytorch I get the following error:
/usr/include/vector_functions.h:172:10: fatal error: vector_functions.hpp: No such file or directory
#include "vector_functions.hpp"
^~~~~~~~~~~~~~~~~~~~~~
When I give the following command to check if the file is actually present I get the following result:
(cuda_en) balaji5199@balaji5199-Lenovo-Legion-Y730-15ICH:~/pytorch$ sudo find / -iname "vector_functions.hpp"
/usr/local/cuda-10.2/targets/x86_64-linux/include/vector_functions.hpp
find: ‘/run/user/1000/gvfs’: Permission denied
Here are all the environment variable values that have been set:
$PATH=/home/balaji5199/Desktop/Xilinx_vivado_2016_version/DocNav:/home/balaji5199/Desktop/Xilinx_vivado_2016_version/Vivado/2016.4/bin:/home/balaji5199/Desktop/Xilinx_vivado_2016_version/Vivado_HLS/2016.4/bin:/home/balaji5199/anaconda3/envs/cuda_en/bin:/home/balaji5199/anaconda3/condabin:/usr/local/cuda-10.2/bin:/usr/local/cuda-10.2/bin:/home/balaji5199/Desktop/Xilinx_vivado_2016_version/DocNav:/home/balaji5199/Desktop/Xilinx_vivado_2016_version/Vivado/2016.4/bin:/home/balaji5199/Desktop/Xilinx_vivado_2016_version/Vivado_HLS/2016.4/bin:/home/balaji5199/miniconda3/envs/new_cuda/bin:/home/balaji5199/miniconda3/condabin:/home/balaji5199/.local/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/games:/usr/local/games:/snap/bin:/usr/local/cuda/bin: No such file or directory
$LD_LIBRARY_PATH=/usr/local/cuda-10.2/targets/x86_64-linux/include
Can someone tell me where the problem is?