Cmake not able to find OpenMP while installing LGBM on PopOS 22 Jammy #5349
Comments
|
Thanks very much for your interest in LightGBM! I've re-pasted the command from the description here with one argument per line, so it's easier to see what's been provided: sudo cmake \
-DUSE_GPU=1 \
-DOpenCL_LIBRARY=/usr/lib/x86_64-linux-gnu/libOpenCL.so \
-DOpenCL_INCLUDE_DIR=/usr/lib/nvidia-cuda-toolkit/bin/include/ \
-DOpenMP_C_FLAGS=-fopenmp=lomp \
-DOpenMP_C_LIB_NAMES="libomp" \
-DOpenMP_libomp_LIBRARY=/usr/lib/llvm-14/lib/libomp.so \
..I have two suggestions to try:
|
|
Hey, thank you very much for your response. First suggestion didn't help, but the second one worked. It started compiling, but after that i started to get these messages during compilation: It finished the compilation successfully, but then if i go to Python and try to import the library i would get: Any suggestions? Thank you very much again |
Yes, we've seen that problem before in the situation "building LightGBM from source inside a See "Workarounds with no changes to LightGBM" in #5106 (comment) for some suggested workarounds. For your case, I recommend following the suggestion Specifically, install C and C++ compilers from conda install -y \
-n "${YOUR_CONDA_ENVIRONMENT_NAME}" \
cmake \
gcc_linux-64 \
gxx_linux-64Then activate your conda environment and install |
|
This issue has been automatically closed because it has been awaiting a response for too long. When you have time to to work with the maintainers to resolve this issue, please post a new comment and it will be re-opened. If the issue has been locked for editing by the time you return to it, please open a new issue and reference this one. Thank you for taking the time to improve LightGBM! |
|
This issue has been automatically locked since there has not been any recent activity since it was closed. To start a new related discussion, open a new issue at https://github.com/microsoft/LightGBM/issues including a reference to this. |
Description
Hi,
I'm trying to install the lightgbm GPU package on Pop!_OS 22.04 LTS. I'm following the instructions posted on the official docs. I'm stuck in the "Build LGBM" section, at the step
cmake -DUSE_GPU=1 -DOpenCL_LIBRARY=/usr/local/cuda/lib64/libOpenCL.so -DOpenCL_INCLUDE_DIR=/usr/local/cuda/include/ ..The paths and options of this command specified for my sistem differs from the original posted on the site. This is the cmake call i'm actually using:
sudo cmake -DUSE_GPU=1 -DOpenCL_LIBRARY=/usr/lib/x86_64-linux-gnu/libOpenCL.so -DOpenCL_INCLUDE_DIR=/usr/lib/nvidia-cuda-toolkit/bin/include/ -DOpenMP_C_FLAGS=-fopenmp=lomp -DOpenMP_C_LIB_NAMES="libomp" -DOpenMP_libomp_LIBRARY=/usr/lib/llvm-14/lib/libomp.so ..This is giving the follwing the error:
I assume that cmake is asking me to set up those missing flags but i don't know how to set them. I tried many alleged solutions founf on the web, like this and this among others, but none of them worked for me. I would appreciate your help here. Thank you very much
My system information is this:
Cuda setup information
The text was updated successfully, but these errors were encountered: