DROID-SLAM icon indicating copy to clipboard operation
DROID-SLAM copied to clipboard

setup.py problem

Open xhangHU opened this issue 3 years ago • 8 comments

src/droid_kernels.cu:15:10: fatal error: Eigen/Sparse: No such file or directory #include <Eigen/Sparse> ^~~~~~~~~~~~~~ compilation terminated. error: command '/usr/local/cuda-11/bin/nvcc' failed with exit code 1

when i run python setup.py install, i got some problem, can you help me~

xhangHU avatar Dec 20 '21 07:12 xhangHU

src/droid_kernels.cu:15:10: fatal error: Eigen/Sparse: No such file or directory #include <Eigen/Sparse> ^~~~~~~~~~~~~~ compilation terminated. error: command '/usr/local/cuda-11/bin/nvcc' failed with exit code 1

when i run python setup.py install, i got some problem, can you help me~

It is possible that your third party (./thirdparty/eigen) dose not be downloaded

Riser6 avatar Dec 20 '21 12:12 Riser6

I installed this, but this error is still reported..

xhangHU avatar Dec 21 '21 08:12 xhangHU

I had a similar problem that I solved because I did not clone with --recursive. Even though the Eigen folder was there, the submodule was not cloned.

jasonyzhang avatar Feb 23 '22 21:02 jasonyzhang

Is Eigen needed download by ourself? i cannot find download part in code

lockingball avatar Apr 19 '22 08:04 lockingball

I solved this by using symbolic links to eigen3 in /usr/include

rmadan2 avatar May 25 '22 21:05 rmadan2

(DRD) ita:~/DROID-SLAM$ sudo python3 setup.py install Traceback (most recent call last): File "setup.py", line 2, in from torch.utils.cpp_extension import BuildExtension, CUDAExtension ModuleNotFoundError: No module named 'torch' But torch in installed , we can import torch

Another Error: (DRD) ita:~/DROID-SLAM$ python3 setup.py install nvcc fatal : Unsupported gpu architecture 'compute_80' error: command '/bin/nvcc' failed with exit code 1

PLz help me

pradyumnakulkarni87 avatar Jun 07 '22 07:06 pradyumnakulkarni87

For your 2nd error, after that it will work I think:

@pradyumnakulkarni87 what is the output of torch.cuda.get_arch_list()? after importing torch. you need to comment some lines in setup.py: if your GPU doesn't support compute_86 and compute_80 (which seems to be your case), comment out line no: 25, 26, 55, 56. I found that in one of the issues earlier.

surajiitd avatar Jun 07 '22 12:06 surajiitd

For your 2nd error, after that it will work I think:

@pradyumnakulkarni87 what is the output of torch.cuda.get_arch_list()? after importing torch. you need to comment some lines in setup.py: if your GPU doesn't support compute_86 and compute_80 (which seems to be your case), comment out line no: 25, 26, 55, 56. I found that in one of the issues earlier.

Thank you ! It works after Commenting ....Thank you

pradyumnakulkarni87 avatar Jun 08 '22 10:06 pradyumnakulkarni87