problem encountered with torch_tensorrt

[arielsolomon]

Hi,
I am trying to use torch_tensorrt in docker container (nvcr.io/nvidia/tensorrt:22.05-py3 and also in nvcr.io/nvidia/tensorrt:22.04-py3) with error repeating regardless the docker image I used or even different gpu models (rtx3090, rtx2060) when trying to import torch_tensorrt:

import torch_tensorrt
Traceback (most recent call last):
File "", line 1, in
File "/usr/local/lib/python3.8/dist-packages/torch_tensorrt/init.py", line 11, in
from torch_tensorrt._compile import *
File "/usr/local/lib/python3.8/dist-packages/torch_tensorrt/_compile.py", line 2, in
from torch_tensorrt import _enums
File "/usr/local/lib/python3.8/dist-packages/torch_tensorrt/_enums.py", line 1, in
from torch_tensorrt._C import dtype, DeviceType, EngineCapability, TensorFormat
ImportError: libtorch_cuda_cu.so: cannot open shared object file: No such file or directory
We tried different ways to install torch_tensorrt and its dependencies with no luck.
Would appreciate some help.

[narendasan]

Can you provide the exact versions of the libraries you have installed in the container and how you installed them? Specifically PyTorch, Torch-TensorRT and TensorRT?

Also we have a dockerfile based on ngc/tensorrt available in //docker which you could use as well

[arielsolomon]

Thanks for your reply.
Problem solved after uninstalling and reinstalling the compatible torch packages to cuda driver