Closed coper16 closed 3 years ago
Thank you, it may be because your GPU is an old one? https://github.com/pytorch/pytorch/issues/31285 Maybe, you can try re-compile the hard exampling mining module.
Thank you for your quick reply!
I see. I come up with anything that might have caused it. I set up these code in local PC and I moved it to external server because l want to use many gpu. But, GPU name is not same.
So, I compiled extend utilities in external server. I have this error. Do you have any idea why this happens?
I think
cuda_include=f'/usr/local/cuda-10.1/include'
cuda_library=f'/usr/local/cuda-10.1/lib64'
is wrong on my code because I use cudatoolkit by conda(I need to use conda environment).
but, I cannot find same .h file( in /usr/local/cuda-10.1/include or usr/local/cuda-10.1/lib64) in anaconda directory...
` Cannot determine CUDA_HOME: cuda-gdb not in PATH /home/nshimobayashi/anaconda3/envs/GIFT_2/bin/nvcc: line 2: /bin/nvcc: No such file or directory x86_64-conda_cos6-linux-gnu-gcc: error: src/nearest_neighborhood.cu.o: No such file or directory Traceback (most recent call last): File "/home/nshimobayashi/anaconda3/envs/GIFT_2/lib/python3.7/distutils/unixccompiler.py", line 205, in link self.spawn(linker + ld_args) File "/home/nshimobayashi/anaconda3/envs/GIFT_2/lib/python3.7/distutils/ccompiler.py", line 910, in spawn spawn(cmd, dry_run=self.dry_run) File "/home/nshimobayashi/anaconda3/envs/GIFT_2/lib/python3.7/distutils/spawn.py", line 36, in spawn _spawn_posix(cmd, search_path, dry_run=dry_run) File "/home/nshimobayashi/anaconda3/envs/GIFT_2/lib/python3.7/distutils/spawn.py", line 159, in _spawn_posix % (cmd, exit_status)) distutils.errors.DistutilsExecError: command 'x86_64-conda_cos6-linux-gnu-gcc' failed with exit status 1
During handling of the above exception, another exception occurred:
Traceback (most recent call last): File "/home/nshimobayashi/anaconda3/envs/GIFT_2/lib/python3.7/site-packages/cffi/ffiplatform.py", line 51, in _build dist.run_command('build_ext') File "/home/nshimobayashi/anaconda3/envs/GIFT_2/lib/python3.7/distutils/dist.py", line 985, in run_command cmd_obj.run() File "/home/nshimobayashi/anaconda3/envs/GIFT_2/lib/python3.7/distutils/command/build_ext.py", line 340, in run self.build_extensions() File "/home/nshimobayashi/anaconda3/envs/GIFT_2/lib/python3.7/distutils/command/build_ext.py", line 449, in build_extensions self._build_extensions_serial() File "/home/nshimobayashi/anaconda3/envs/GIFT_2/lib/python3.7/distutils/command/build_ext.py", line 474, in _build_extensions_serial self.build_extension(ext) File "/home/nshimobayashi/anaconda3/envs/GIFT_2/lib/python3.7/distutils/command/build_ext.py", line 559, in build_extension target_lang=language) File "/home/nshimobayashi/anaconda3/envs/GIFT_2/lib/python3.7/distutils/ccompiler.py", line 717, in link_shared_object extra_preargs, extra_postargs, build_temp, target_lang) File "/home/nshimobayashi/anaconda3/envs/GIFT_2/lib/python3.7/distutils/unixccompiler.py", line 207, in link raise LinkError(msg) distutils.errors.LinkError: command 'x86_64-conda_cos6-linux-gnu-gcc' failed with exit status 1
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "build_extend_utils_cffi.py", line 30, in
You may need to find a path "cuda_include" which contains "cuda_runtime.h" and a path "cuda_library" which contains "libcudart.so". It depends how you install the cuda library. In default, these two paths are "/usr/local/cuda/include" and "/usr/local/cuda/lib64".
I see... Thanks! I'll try it!!
Hi Thank you for your great work!
I try to train GIFT-stage1. And, I had a error. Do you have any idea why this happens?