Encountered cuda 710 error when optimizing BERT with nebullvm

[valeriosofi]

When optimizing BERT with nebullvm, the python process is killed with an error of PyTorch-TensorRT that causes a segmentation fault error:

CUDA initialization failure with error: 710. Please check your CUDA installation: http://docs.nvidia.com/cuda/cuda-installation-guide-linux/index.html
Segmentation fault (core dumped)

For now the problem can be avoided by setting ignore_compilers=["tensor RT"] as parameter in the optimize_model, function.
Must wait for the fix from Pytorch-TensorRT, the issue is reported here: pytorch/TensorRT#1418