When importing caffe2.pyton.onnx.backend with import torch,
I get a import error as below.
This is for pytorch-cpu install.
Problem description
Python 3.6.7 | packaged by conda-forge | (default, Nov 21 2018, 03:09:43)
[GCC 7.3.0] on linux
>>> import caffe2.python.onnx.backend as backend
WARNING:root:This caffe2 python run does not have GPU support. Will run in CPU only mode.
>>> import torch
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
File "/home/xxxx/.local/lib/python3.6/site-packages/torch/__init__.py", line 80, in <module>
from torch._C import *
ImportError: /home/xxxx/.local/lib/python3.6/site-packages/torch/lib/libcaffe2_gpu.so: undefined symbol: _ZTIN2at10TensorImplE
>>>
Environment information
Collecting environment information…
PyTorch version: 0.4.1
Is debug build: No
CUDA used to build PyTorch: 9.0.176
OS: Debian GNU/Linux buster/sid
GCC version: (Debian 8.2.0-15) 8.2.0
CMake version: version 3.13.2
Python version: 3.6
Is CUDA available: Yes
CUDA runtime version: 9.2.148
GPU models and configuration: GPU 0: GeForce 940M
Nvidia driver version: 390.87
cuDNN version: Could not collect
Versions of relevant libraries:
[pip] Could not collect
[conda] blas 1.0 mkl
[conda] mkl 2018.0.3 1
[conda] pytorch-nightly-cpu 1.0.0.dev20190128 py3.6_cpu_0 pytorch
library deps
:$~/python_tests$ ldd /home/xxxx/.local/lib/python3.6/site-packages/torch/lib/libcaffe2_gpu.so
linux-vdso.so.1 (0x00007fffafbf9000)
libdl.so.2 => /lib/x86_64-linux-gnu/libdl.so.2 (0x00007f2001c2e000)
librt.so.1 => /lib/x86_64-linux-gnu/librt.so.1 (0x00007f2001c24000)
libpthread.so.0 => /lib/x86_64-linux-gnu/libpthread.so.0 (0x00007f2001c03000)
libm.so.6 => /lib/x86_64-linux-gnu/libm.so.6 (0x00007f2001a80000)
libcaffe2.so => /home/xxxx/.local/lib/python3.6/site-packages/torch/lib/libcaffe2.so (0x00007f1ffa0b0000)
libstdc++.so.6 => /usr/lib/x86_64-linux-gnu/libstdc++.so.6 (0x00007f1ff9f2d000)
libgomp-7bcb08ae.so.1 => /home/xxxx/.local/lib/python3.6/site-packages/torch/lib/libgomp-7bcb08ae.so.1 (0x00007f1ff9d01000)
libgcc_s.so.1 => /lib/x86_64-linux-gnu/libgcc_s.so.1 (0x00007f1ff9ce7000)
libc.so.6 => /lib/x86_64-linux-gnu/libc.so.6 (0x00007f1ff9b26000)
/lib64/ld-linux-x86-64.so.2 (0x00007f2025559000)
#running python -v
running python -v
extension module ‘torch._nvrtc’ executed from ‘/home/xxxx/.local/lib/python3.6/site-packages/torch/_nvrtc.cpython-36m-x86_64-linux-gnu.so’
import ‘torch._nvrtc’ # <_frozen_importlib_external.ExtensionFileLoader object at 0x7f88675c1a20>
Traceback (most recent call last):
File “import1.py”, line 2, in
import torch
File “”, line 971, in _find_and_load
File “”, line 955, in _find_and_load_unlocked
File “”, line 665, in _load_unlocked
File “”, line 678, in exec_module
File “”, line 219, in _call_with_frames_removed
File “/home/lcp/.local/lib/python3.6/site-packages/torch/init.py”, line 80, in
from torch._C import *
File “”, line 971, in _find_and_load
File “”, line 955, in _find_and_load_unlocked
File “”, line 658, in _load_unlocked
File “”, line 571, in module_from_spec
File “”, line 922, in create_module
File “”, line 219, in _call_with_frames_removed
ImportError: /home/xxxx/.local/lib/python3.6/site-packages/torch/lib/libcaffe2_gpu.so: undefined symbol: _ZTIN2at10TensorImplE